Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
11 REVIEW OF POSTOPERATIVE WOUND INFECTIONS W. A. Altemeier* Infection is one of the salient features of human life, and many infections in patients are of surgical significance. An understanding of this aspect of medicine is, therefore, an essential part of clinical surgery and the supporting professional scientific disciplines which help to provide the surgeon with a safe environment within which to operate and work (l-5). The ubiquity of infectious agents in man's environment, their propensity for in- vading the tissues of the body, their potential for producing significant pathophysiologic effects on the various bodily functions, their remarkable adaptability to circumstance and newer forms of treatment, and the necessity for excluding their presence or controlling their growth sufficiently to permit surgical treatment and effective wound healing, all emphasize the sig- nificance of infection in surgical practice (1,3). Moreover, the expanding horizons of surgery have often been dependent upon the development and appli- cation of special methods of overcoming the hazards of postoperative infec- tions (1,8). It should be remembered also that every operation is an experi- ment in applied and practical bacteriology. History has shown that postoperative and hospital-acquired infections have been problems for as long as there have been hospitals, and attempts to prevent their incidence and spread began hundreds of years ago when separate hospitals were built for patients with communicable diseases. Fever hospitals, smallpox hospitals, tuberculosis sanitoria, and "Pest Houses" were established in an effort to separate infected patients from other patients and the com- munity. It is generally recognized that the introduction of modern antibiotic therapy has had a revolutionary effect on the treatment of many established infections, but clinical experience and experimental bacteriolog- ical studies have shown that its general use for approximately one third of a *Chairman, Department of Surgery, Cincinnati General Hospital. Work reported supported by U.S.P.H.S. Grant 5-P01-GM-15428-08 and U.S. Army Contract DAMD 17-74-C-5018.
12 W. A. Altemeier century has failed to decrease the overall incidence of postoperative surgical infections (1,2,3,7,8,10). In addition, the widespread use and misuse of antibiotic therapy has undoubtedly increased the problems concerned with their prevention (3,5,8,9,12,15,17,18). Too often its widespread use prophylactical- ly in surgical patients has contributed to the development of an unwarranted overdependence on its effectiveness, a de-emphasis or discredit of the impor- tance of the established surgical principles, a relaxation of the "surgical conscience," a breakdown of isolation procedures, and the establishment of a reservoir of virulent and antibiotic-resistant bacteria concentrated in the hospital environment (1,4,8). These trends have been accentuated by the complexities of modern surgical practice with the concentration of large numbers of patients with established infections in hospitals, the extension of prolonged surgical opera- tions and supportive procedures to a rapidly increasing number of high-risk patients, the increase of the number of individuals with severe trauma, and the growing use of drugs which decrease bodily resistance to infection. Thus, it must be kept in mind that the modern general hospital has become a complex community in which the opportunities for infection are numerous and ever pres- ent. Current hospital practice has resulted in the concentration of many people admitted with a large variety of infections as well as many others who are unusually susceptible to secondary or hospital-acquired infections. The latter have included patients admitted for treatment of recent trauma, persons debilitated by chronic disease, individuals undergoing surgical operations and a multitude of other technical diagnostic procedures, people requiring steroid therapy, and those under treatment for neoplastic diseases. The extension of surgical treatment to aged and debilitated patients, and the widespread use of complicated diagnostic, therapeutic, and anesthetic procedures also have favored the development of hospital-acquired infections. In addition, the urgency associated with the treatment of many patients with life-threatening disease or extensive injuries, the demands of large numbers of patients in a short period of time, and shortages of skilled nursing and other personnel have tended to produce compromises and administrative trends not necessarily in the best interests of infection control (l-3).
13 Postoperative Wound Infections A particularly important effect of general and indiscriminate antibiotic therapy in hospitals has been the progressive development of re- sistance to penicillin and many of the other antibiotic agents which was acquired by a large variety of important bacteria concentrated in the hospital environment. These virulent microorganisms, both Gram-positive and Gram-nega- tive, have the potential of becoming pathogenic in patients weakened by dis- ease, injury, metabolic conditions, and other debilitating factors as mentioned above (8,10,12,17,18). There are a number of other important factors which have influenced the incidence of postoperative infections at the present time. In a 2 1/2- year collaborative investigation of 15,613 consecutive operative procedures done in five American university centers, an overall infection rate of 7.4 percent for all types of operations was determined under close, energetic, and continuing surveillance. These centers included the University of Pennsyl- vania, Hahnemann Medical Schoo1, the University of California, Los Angeles, George Washington University, and the University of Cincinnati. This project was carried out under the support of the U.S. Public Health Service and the aegis of the National Research Council (13). Many national and international conferences have been held in an effort to bring the various aspects of this situation into proper perspective, to disseminate existing knowledge between countries and scientific disciplines, and to stimulate research for the solution of related problems. One of the first and highly significant meetings of this type was the National Conference on Hospital-Acquired Staphylococcal Disease, held on September 15-17, 1958, in Atlanta, Georgia under the joint sponsorship of the United States Public Health Service and the National Research Council. An International Seminar on Hospital Infections was also held by the Council for International Organiza- tions of Medical Sciences in London, September 24-28, 1962. Delegates from 26 countries and the World Health Organization participated in the discussion, representing a considerable body of resource knowledge and research potential. There were 111 delegates in attendance, each being an expert on some aspect of hospital-acquired infection. Another important example was the International
A. W. Altemeier Conference on Surgical Infections held by the International Society of Surgery in Moscow on August 2l-28, 1971 (8). These and many other conferences have been helpful, but the fact remains that the problems of postoperative wound and hospital-acquired infec- tions persist, and that considerable confusion exists in the minds of surgeons, physicians, hospital administrators, lay hospital boards, bacteriologists, epidemiologists, sanitary engineers, architectural engineers, attorneys, judges, and others regarding the significant factors actually responsible for these infections. In the meantime, the controversy over the relative importance of the aerial spread and contact spread of infection has continued from the last half of the 19th century to the present (10). Following the discovery and intro- duction of the surgical antisepsis by Lister, Von Bruns and Von Bergmann be- came convinced that contact was the principal method for spread of infection and by their evidence persuaded Lord Lister to abandon his carbolic acid spray of the operating room aerial environment. The use of ultraviolet light, lithium oxide, electrical precipitation of bacterial-laden dust particles, and various types of laminar flow have each been developed and used in an effort to reduce the incidence of postoperative wound infections. Moreover, the increasing costs associated with the care of surgical patients with infections, the progressive threat of malpractice claims, and the growing demands of an informed and often "oversold" public have become important related matters of concern. THE INCIDENCE OF INFECTIONS Currently there is still considerable confusion about the incidence, nature, and causes of infections in surgical practice. There is also much confusion and great divergence of thought regarding the best methods of pre- venting and controlling surgical infections. Consequently, there are no accurate figures in the United States to indicate the incidence of the many types of infection acquired by hospitalized surgical patients, but there is evidence that it is much greater than generally suspected.
15 V, Postoperative Wound Infections It is generally recognized, however, that from 40 to 65 percent of the patients in general hospitals are usually surgica1, and that approximately 30 percent of hospitalized surgical patients either have established infections at the time of admission or develop some type of infection during their hospi- tal stay. In 1963 it was estimated that approximately 25,000,000 patients had been admitted to hospitals in the United States and that over 1,000,000 of them developed postoperative or hospital-acquired infections. More recent figures made available for the year 1967 have been reported by Altemeier as follows (6,8,10): Table 1. Hospital Infections USA - 1967 Estimated Incidence: Hospital admissions 31,600,000 Surgical operations performed 18,800,000 in the operating room Estimated number of postoperative 1,391,200 wound infections for all types of operations (7.4% of operations) Estimated number of hospital-acquired 2,101,037 infections These estimates are based upon data obtained from 1,118 hospitals in the United States; this number being approximately one-fourth of the total number of hospitals under surveillance of the Joint Commission on Accredita- tion of Hospitals and Blue Cross-Blue Shield. The overall wound infection rate of 7.4 percent used in these estimations was the incidence for all types of operative wounds as determined during a 2 1/2-year collaborative ultraviolet study of 15,613 consecutive operative procedures done in five American university centers under the support of the U.S. Public Health Service and the aegis of the National Research Council. The operative wounds thus studied were composed of clean, clean- contaminated, contaminated, and dirty types. Elective and emergency operations were both included in Tafcle 2 (13).
16 W. A. Altemeier Table 2. Type of Operative Wound Clean Clean-contaminated Contaminated and dirty Not reported Incidence of Infection No. 11,690 2,589 1,262 72 No. Percent 594 5.1 280 10.8 277 21.9 6 8.3 In the 11,690 clean elective operations in this series, the average wound infection rate was 5.1 percent, but the average rate for all types of operations was 7.4 percent. It also became apparent that the incidence of in- fection varied a great deal with the type of operation (Table .3) ⢠Table 3. Incidence of Infection Following Selected Commonly Performed Operative Procedures (Adapted from Ultraviolet Study) Number of Incidence of times infection performed percent 1,312 1.9 406 2.2 628 6.1 756 6.9 220 10.0 288 10.1 551 11.4 127 17.3 227 18.9 Operative Procedure Herniorrhaphy* Thyrmidectomy Hysterectomy Cholecystectomy Partial Colectomy Subtotal Gastrectomy Appendectomy Nephrectomy Radical Mastectomy *Including inguina1, femoral, and epigastric; excluding incisional and ventral. Noteworthy too was the fact that the incidence of postoperative wound infection was increased three to five fold or more whenever one of the major tracts was transected or resected during a planned elective operation, as compared to elective operations without such procedures.
17 Postoperative Wound Infections For example, the incidence of infection following hernioplasty was 1.9 percent; whereas it was 6.1 percent for hysterectomy, 6.9 percent for chole- cystectomy, 10.0 percent for partial colectomy, and 10.1 percent for subtotal gastrectomy. It can be seen from the above discussion that hospital-acquired infection is also a problem of considerable monetary magnitude. The average cost per infection has been reported by Swartz at the University of Virginia to vary between $6,000 and $9,000. Using an estimated postoperative wound in- fection cost of approximately $7,000 per patient, on the basis of 1,400,000 postoperative wound infections (as shown in Table 1), the estimated overall cost to society has been estimated to be $9.8 billion (6,8,11). Specific programs are matters for physicians to consider and correct. There appears to be some consensus that a broad-based control program centered around isolation procedures, combined with both formal and informal education, would help to decrease the incidence and cost of infection. Government subsidies would help to defray the cost of infection control and help to insure an optimal output of this public good. Another avenue of possible effective action under consideration is Professional Services Review Organization (PSRO). A review of postoperative wound infections should include a consideration of their classification on the basis of the source or base of in- fection. In this regard, there are three general types: community or home-based infections, operating room-based infections, and hospital-based (other than operating room-based) infections. COMMUNITY OR HOME-BASED INFECTIONS These are infections which develop spontaneously or otherwise in the home or community, being unrelated to the hospital and the hospital reservoir of microorganisms. As many as 30 or 40 percent of patients admitted to a busy surgical service in a general hospital may have infections of this type.
18 W. A. Altemeier Examples are acute appendicitis, acute cholecystitis, acute diverticulitis with perforation and peritonitis, acute perforated peptic ulcer with peritonitis, etc. Infections in this category are generally not caused by antibiotic resis- tant bacteria, and they do not present at this time new or special problems re- lated to the primary disease. OPERATING ROOM-BASED INFECTIONS These are hospital-based infections which consist primarily of operative wound infections developing postoperatively in relation to surgical procedures performed in the operating room. Because of their importance and the problems which they currently present, they are considered separately from other hospital-acquired infections for purposes of special discussion. The sources of these infections are both endogenous and exogenous. HOSPITAL-BASED INFECTIONS (OTHER THAN OPERATING ROOM) Infections which occur in patients during their hospitalizations are designated as hospital-acquired or nosocomial infections. They are the re- sult of microbial invasion, most often by antibiotic-resistant and virulent microorganisms of the hospital environment. Invasion may follow diagnostic and various therapeutic procedures such as lower urinary tract catheterization or instrumentation, intravenous therapy (especially continuous), tracheostomy, or arteriography. Infection may also be related to the emergence of virulent bacteria during antibiotic therapy and other medications, the decreased host re- sistance by immunosuppression, the impairment of other host defenses, the replace- ment of the normal flora of organisms in the major tracts of the body by virulent or antibiotic-resistant bacteria of the hospital reservoir during prolonged pre- operative hospitalization periods, and the progressive extension of surgical practices in the high-risk, aged, or otherwise debilitated patients. In a review of surgical postoperative infections, their classification by anatomic location and pathologic changes is of interest (Table 4) (11).
19 Postoperative Wound Infections Table 4. Classification of Surgical Infections By Anatomic Location and Pathologic Changes I. Local Wound Infection A. Cellulitis B. Suppuration or septic liquefaction of tissues C. Abscess D. Septic necrosis E. Local wound, septic thrombophlebitis II. Regional Extension A. Adjacent tissues by direct extension B. Lymphangitis and lymphadenitis C. Thrombophlebitis D. Peritonitis E. Central nervous system infection, including meningitis and brain abscess F. Mediastinitis G. Retroperitoneal cellulitis III. Organ or Visceral Infection IV. Systemic Invasion Based upon clinical and laboratory research on the Surgical Services and the Surgical Research Bacteriology Laboratory of the University of Cincinnati during the past 32 years and a review of pertinent literature, an etiologic classification of surgical infections has been developed (Table 5) (1,11). This emphasizes the large numbers and varieties of bacteria which may produce wound infections.
20 W. A. Altemeir Table 5. Etiologic Classification of Surgical Wound Infections I. Aerobic Bacterial A. Gram-positive cocci 1. Staphylococcus 2. Streptococcus 3. Pneumococcus B. Gram-negative cocci 1. Neisseria catarrhalis 2. Neisseria gonorrhoeae C. Gram-negative bacillary 1. Escherichia coli 2. Aerobacter aerogenes 3. Klebsiella 4. Pseudomonas aeruginosa 5. Proteus 6. Serratia 7. Paracolgon 8. Alcaligenes faecalis 9. Salmonella 10. Haemophilus influenzae D. Gram-positive bacterial 1. Bacillus anthracis 2. Corynebacterium 3. Diphtheroid 4. Bacillus species 5. Mycobacterium a. Tuberculosis b. Balnei II. Microaerophilic Bacteria A. Gram-positive cocci 1. Streptococcus a. Hemolyticus b. Non-hemolyticus III. Mixed Infections A. Aerobic and Anaerobic B. Gram-positive and Gram-negative C. Synergistic
21 Postoperative Wound Infections IV. Anaerobic Bacterial A. Gram-positive cocci 1. Peptococcus 2. Peptostreptococcus B. Gram-positive Bacilli 1. Cl. perfringens 2. Cl. novyi 3. Cl. septicum 4. Cl. histolyticum 5. Cl. tetani C. Gram-negative bacilli 1. B. fragilis 2. Bacteroides species 3. B. melaninogenicum V. Non-bacterial Infections A. Fungal 1. Candidiasis (Candida albicans) 2. Actinomycosis (Actinomyces Israelii) 3. Nocardiosis (Nocardia asteroides) 4. Blastomycosis (Blastomvces dermatitides) 5. Coccidioidomycosis (Coccidoides immit.is) 6. Histoplasmosis (Histoplasma capsulatum) 7. Sporotrichosis (Sporotrichum schenckii) 8. Phycomycosis (Mucor sp.) 9. Aspergillosis (Aspergillus niger) B. Viral 1. Herpesvirus 2. Poxvirus (vaccinia) 3. Varicella (Herpes Zoster virus) 4. Cytomegaloviruses 5. Mumps virus (parotitis and pancreatitis) 6. Polovirus 7. Hepatitus virus (infectious and serum) 8. Rabies virus Finally, I would like to point out two other important considerations: first, that there is considerable confusion and controversy as to the definition of a surgical postoperative infection and, secondly, the classification of infec- tions based upon the recognition of different degrees of wound contamination (11).
22 W. A. Altemeier There are significant differences in the minds of many surgeons as to what actually constitutes a postoperative wound infection and particularly a sig- nificant infection. This has emphasized the importance of establishing certain criteria in order to provide objective definitions of surgical infections which would permit consistent evaluation and comparable reports. Laboratory microbial culture studies alone are not completely reliable for a number of reasons which include technical limitations of many hospital laboratories, inadequate wound sampling and testing methods, and the report of secondary microbial invaders or wound contaminants which are incorrectly considered to be of primary etiologic importance. Cultures obtained from studies of purulent discharges not infre- quently are negative following culture by the usual standard technics whereas special bacteriologic procedures may reveal the true pathogen. Moreover, it is not unusual for microorganisms to be recovered from wounds that are healing with- out any evidence of infection. Individual clinical judgment thus comes into play, and this admittedly may often be subjective and biased. When the five-university collaborative study on the evaluation of ultraviolet irradiation was undertaken, these factors were taken under consideration and uniform criteria were establish- ed to promote understanding and agreement among surgeons and other observers con- cerned with the problem of wound infections. During the past ten years, these criteria have become relatively standard and have also been incorporated in a use- ful classification of operative wounds in relation to their contamination and the increasing risk of infections. CURRENT DEFINITION OF POSTOPERATIVE WOUND INFECTION Postoperative wounds are considered to be definitely infected if there is a purulent discharge, even if microorganisms are not cultured from the puru- lent material. Wounds are considered to be uninfected if they heal per primam without discharge. Minute stich abscesses are excluded if the inflammation and discharge are minimal and confined to points of suture penetration or if the in- cision itself heals per primam without drainage. In the belief that clinical judgment is most important in the diagnosis of wound infections, however, wounds may be judged infected or uninfected by the ,
23 Postoperative Wound Infections responsible surgeon who should be free to diagnose infection that doesn't meet the above criteria, but not allowed to judge a wound "uninfected" that does meet the above criteria for infection. OPERATIVE WOUND CLASSIFICIATION CONTAMINATION-INFECTION RISK A wound classification developed on the basis of a clinical estima- tion of bacterial density, contamination, and risk of subsequent infection is contained in the following outline which has now become widely accepted as a standard classification of operative wounds (Table 6). Table 6. Classification of Operative Wounds in Relation to Contamination and Increasing Risk of Infection I. Clean A. Non-traumatic B. No inflammation encountered C. No break in technic D. Respiratory, alimentary, genitourinary tracts not entered II. Clean-Contaminated A. Gastrointestinal or respiratory tracts entered without significant spillage B. Gastrointestinal and repiratory tracts not entered 1. Transection of appendix or cholecystic duct is considered clean-contaminated in the absence of acute inflammation 2. Entrance into genitourinary or biliary tracts is clean-contaminated in absence of infected urine or bile C. Minor break in technic
24 W. A. Altemeier III. Contaminated A. Major break in technic B. Acute bacterial inflammation encountered, without pus C. Gross spillage from gastrointestinal tract D. Traumatic wound, fresh, from relatively clean source E. Entrance of genitourinary or biliary tracts in presence of infected urine or bile IV. Dirty A. Pus encountered B. Perforated viscus encountered C. Old or traumatic wound, or from dirty source A clean wound is considered to be a non-traumatic, uninfected operative wound in which neither the respiratory, alimentary, or genitourinary tracts or the oro-pharyngeal cavities are entered. Clean wounds are elective, primarily closed, and undrained wounds. Clean-contaminated wounds are operative wounds in which the respiratory, alimentary, or genitourinary tract is entered without unusual contamination, or wounds which are mechanically drained. Contaminated wounds include open, fresh traumatic wounds, operations with a major break in sterile technic (e.g., open cardiac massage), and incisions encountering acute, non-purulent inflammation. Dirty wounds include old tramatic wounds and those involving abscesses or perforated viscera. The very definition of this classification suggests that the organisms causing postoperative infection are present in the operative field before operation. SUMMARY A review of the subject of postoperative wound infection leads to the conclusion that this type of surgical sepsis is a complex phenomenon in clinical practice whose dimensions are real, continuing, significant, and changing. The general use of antibiotics during the past one-third of a century has had many
25 Postoperative Wound Infections revolutionary effects on the treatment of established infections, but infections nevertheless continue to be serious problems of world-wide scope. It has been disappointing to note that antibiotic therapy has not decreased the overall incidence of postoperative infections. The estimated cost of postoperative wound infections in the United States during 1967 was approxi- mately $9.8 billion. This does not include community or home-based and nosocomial infections other than wound infections. While we have been successfully prevent- ing some types of infection, others have replaced them. There are differences of opinion in the minds of many surgeons, microbiologists, and others as to what a postoperative wound infection is. Needed are criteria to provide objective definitions and the mechanism of consistent evaluation and comparable reporting. A classification of operative wounds in re- lation to the degree of contamination and increasing risk of infection has been recommended for these purposes. The controversy over the relative importance of aerial microbial contamination in the incidence of postoperative infection has continued since the last part of the nineteenth century to the present. Experimental evidence and clinical experience have emphasized the greater importance of contamination from endogenous and exogenous sources through contact. It is generally agreed, how- ever, that there have been proven examples of infection from aerial contamination in hospital practice, but that additional research is needed to further define its importance and to develop appropriate methods of control.
26 REFERENCES 1. Altemeier, W. A., Control of wound infections. J. Roy. Coll. Surg., Edinburgh, 11:271, 1966. 2. Altemeier, W. A., J. C. Todd, and W. W. Inge, Gram-negative septicemia: A growing threat. Ann. Surg. 166:530, 1967. 3. Altemeier, W. A., Bodily response to infectious agents. JAMA 202:1085, 1967. 4. Altemeier, W. A., W. R. Culbertson, and R. P. Hummel, Surgical con- siderations of endogenous infectionsâSources, types, and methods of control. Surg. Clin. No. Am. 48:227, 1968. 5. Altemeier, W. A., W. R. Culbertson, W. D. Fullen, and J. J. McDonough, Serratia marcescens septicemia: A new threat in surgery. Arch. Surg. 99:232, 1969. 6. Altemeier, W. A. and S. Levenson, Trauma workshop report: Infections, immunology, and gnotobiosis. J. Trauma 10:1084, 1970. 7. Altemeier, W. A., B. J. Barnes, E. J. Pulaski, W. R. Sandusky, J. F. Burke, and G. A. Clowes, Jr., Infections: Prophylaxis and managementâ A symposium. Surg. 67:369, 1970. 8. Altemeier, W. A., Bacteriology of surgical infections. Clinical and experimental considerations. Vingt-quatrieme Congres de la Societe Internationale de Chirurgie, Moscow, pp. 53-70, 1971. 9. Altemeier, W. A. and E. J. Berkich, Wound Sepsis and Dehiscence. Chapt. 8 in Critical Surgical Illness, James Hardy, Ed. W. B. Saunders Co., Philadelphia, Pa., 1971. 10. Altemeier, W. A., The significance of infection in trauma. Bull. Am. Coll. Surg. 7:2, 1972. 11. Altemeier, W. A., J. F. Burke, B. Pruitt, and W. R. Sandusky, Manual on control of surgical infections. To be Published. 12. Finland, M., W. F. Jones, and M. W. Barnes, Occurrence of serious bacterial infections since introduction of antibacterial agents. JAMA 170:2188, 1969. 13. Howard, J. M., W. F. Barker, W. R. Culbertson, P. J. Grotzinger, V. M. lovine, R. J. Keehn, and R. G. Ravdin, Postoperative wound infections: The influence of ultraviolet irradiation of the operating room and of various other factors. Ann. Surg., Suppl. 160:1, 1964.
27 14. Meleny, F. L., The study of the prevention of infection in contaminated accidental wounds, compound fractures, and burns. Ann. Surg. 118:171, 1943. 15. Meleny, F. L., and A. 0. Whipple, A statistical analysis of a study of the prevention of infection in soft part wounds, compound fractures, and burns with special reference to the sulfonamides. Surg. Gynecol. Obstet. 80:263, 1945. 16. Pulaski, E. J., Surgical infections: Prophylaxis, Treatment, Anti- biotic Therapy. Charles C Thomas, Springfield, 111., 1953. 17. Rogers, D. E., A changing pattern of life-threatening microbial disease. N. Engl. J. Med. 261:677, 1959. 18. Williams, R.E.O. and R. A. Shooter, Ed., Infection in Hospitals, Epidemiology and Control. F. A. Davis Co., Philadelphia, Pa., 1963.
28 OPERATING ROOM ENVIRONMENT AND DEEP WOUND SEPSIS FOLLOWING JOINT REPLACEMENT AT UCLA Harlan C. Amstutz* INTRODUCTION Deep infection following total-joint replacement usually requires removal of the prosthetic components; therefore, studies concerning the preven- tion and control of wound sepsis in total-joint replacement procedures have high priority at UCLA. The main modes of surgical infection, direct, airborne, and endogenous, were considered with relevance to existing preventive measures and controls. Use of prophylactic antibiotics was found to provide coverage at sur- gery and during the immediate postoperative period. Whether the so-called "clean air laminar flow" room can be used safely for total-hip joint replacement (THR) procedures without the use of antibiotics was also considered. This author participated in one of the earliest total-hip replacement (THR) studies (8), in which 98 percent of these procedures were performed with- out the use of prophylactic antibiotics in conventional operating rooms and with- out high velocity directed horizontal filtered airflow systems (HVDHFAFS). The deep sepsis rate from this series is now 12 percent in the first 100 procedures with a four- to six-year follow-up. Subsequently, studies by investigators such as Stinchfield (7), Coventry (5), and Wilson (8) indicate that low infection rates were achieved using prophylactic antibiotics without "clean air laminar flow." MATERIALS AND METHODS Control of sepsis at UCLA includes: 1) preoperative evaluation to exclude all direct sources of infection; preparation of the operative site; 2) the use of antistaphylococcal prophylactic antibiotics; 3) evaluation of methods of reducing or eliminating airborne bacteria at surgery; 4) stringent control of urological procedures and control of other potential sources of bacteremias and septicemias; and 5) early diagnosis and treatment of subcutaneous infection to prevent the involvement of deep space (2). PATIENT MANAGEMENT AND SURGICAL FACILITIES Since October 1970, all patients who received total-joint replacements at UCLA have had standard preoperative preparation and postoperative evaluation *Professor and Chief, Division of Orthopaedic Surgery, UCLA School of Medicine, Los Angeles, California
29 Harlan C. Amstutz and care. Special efforts to avoid intraoperative contamination have included the wearing of double gloves, use of impermeable disposable drapes, gowns, and headwear which completely covered all hair. Stringent limitation of operating room traffic was attempted. Clinical and microbiological evaluation of 274 THR procedures was performed to determine whether there was a correlation between postoperative in- fections and airborne organisms present at surgery (6). After 167 of these pro- cedures had been performed in the modern (1953) "conventional" operating room which has 10 to 12 air changes per hour, we attempted to reduce airborne contami- nation further by installing an HVDHFAFS. This system, installed in an operating enclosure 3.05 m by 3.65 m by 2.74 m has an air velocity of 26.5 m per minute with 400 air changes per hour. Initially, the ancillary use of aspirator systems was also planned; however, because of the inconvenience and difficulty in communica- tion during surgery caused by this equipment and the overall good results using the HVDHFAFS with prophylactic antibiotics, this additional change has not been made. Within this series of 274 THRs, a study was made of 16 total-hip replacements which included detailed bacteriological sampling and speciation of operating room air obtained during surgery and microorganisms cultured from pa- tient wounds. Bacteriological studies for operating theater air were performed during eight procedures performed in a HVDHFAFS room. All patients received systemic prophylactic antibiotics and anticoag- ulants. One and a half grams of methycillin was given intramuscularly every six hours for three doses preoperatively, followed by one gram intravenously during surgery. At surgery, each patient received 750 to 1,000 ml of antibiotic irri- gation solution consisting of 10 g of neomycin, 100 mg of polymyxin B and 100,000 units of bacitracin diluted to dne liter of 0.90 percent sterile saline. After surgery, each patient received one gram of this agent intrave- nously every six hours until the intravenous infusion was discontinued; one half gram of cloxacillin was given every six hours by mouth thereafter until the sixth postoperative day.
30 Operating Room Environment Patients who were allergic to penicillin received 600 mg of lincomycin or clindamycin phosphate intramuscularly every eight hours for three doses befon surgery, 600 mg intravenously during surgery, and 600 mg intravenously every eig: hours until the intravenous infusion was discontinued. Then, 300 mg of clindamy- cin hydrochloride or 500 mg of lincomycin were given every six hours orally until the sixth postoperative day. Warfarin was given the night of surgery and subse- quent daily dosages were adjusted to maintain the prothrombin time at approxi- mately 40 to 60 percent of control values. RESULTS Analyses of all 274 THR procedures in this series were made in con- junction with microbiological studies of air sampled at the wound site in both rooms. Bacterial counts revealed a constant five- to eight-fold decrease in the directed flow room (1). However, during these studies, the microorganisms cultured from patients with deep sepsis were usually gram negative and from the patient's genitourinary tract. Deep sepsis occurred in nine patients (two percent), all of whom had been considered high-risk patients for special complications. Seven patients had previous surgery, three of these had previous hip sepsis. Gram-negative bacteria were cultured from the wounds of eight. One patient with negative cul- tures had histological evidence of acute inflammation at the time of debride- ment three months postoperatively. Two patients had systemic arthritis and associated steroid therapy. Three had problems of obesity and one had minimal subcutaneous tissue. During surgery, four patients had excessive blood loss and two received bilateral replacements. To date, a total of 430 THR procedures with a follow-up range of six months to four years have been performed at UCLA. Since completion of the 274 THR series, stringent methods of reducing urinary tract infections have been in- stituted, and no sepsis has occurred in the 156 THR procedures subsequent to the initial series. We have tried to avoid catherization; but when this is necessary, patients receive gram-negative suppressives (Gantrisin, Furadantin,
31 Harlan C. Amstutz or Hyprex), and adequate antibiotic treatment when the catheter is removed. The importance and rationale of prevention has been summarized in previous articles (2,3). Initially, there were slightly higher deep sepsis rates in total-knee joint replacements. All of these procedures were performed in the HVDHFAFS room. In the first 100 procedures, the deep sepsis rate was four percent. In the next series of 125 patients followed for six or more months, the infection rate de- clined to one percent. Factors that seem to correlate with poor wound healing include the superficial nature of the joint and possibly the close positioning of the surgeon's skin to the operative wound site. This was evidenced by the pres- ence of staphylococcus epidermidis in two deep infections. The decline in THR sepsis reflects the increased care in ensuring wound healing during the postoper- ative period. The question remains unanswered as to whether surgery can safely be performed in "clean air laminar flow" rooms with or without isolation systems, without the use of prophylactic antibiotics to provide coverage at surgery and during the immediate postoperative period. If antibiotics prevent replication of infectious organisms which directly contaminate the wound intraoperatively, then we must consider whether the use of HVDHFAFS is needed. However, the air- borne bacteria should be periodically cultured and speciated to determine whether gram-negative organisms are present. We also need to determine whether the prophy- lactic antistaphylococcal antibiotics increase wound susceptibility to infection from gram-negative organisms. The high-risk patient may require the methods of "clean air", plus isolation systems, plus prophylactic antibiotics to prevent infection. Certainly, if "clean air" systems are to be used, they should be used properly, e.g., the operating table should be placed close to the filtered air inlet and oriented in a way to maximize unobstructed passage of air through the region of the surgical site (4).
32 Operating Room Environment REFERENCES 1. Amstutz, H. C., High velocity directional air flow systems (HVDAFS). Med., in press, 1974. Calif, 2. Amstutz, H. C., Prevention of operative infections. Proceedings of the Clean Air Symposium, Cleve. Clin. Q. 40:125-131, 1973. 3. Amstutz, H. C., Treatment of Sepsis on Total Hip Replacement. AAOS Instruc- tional Course Lectures, 23, C. V. Mosby, St. Louis, 1974. 4. Buchberg, H., H. C. Amstutz, J. P. Wright, and R. M. Lodwig, Evaluation and optimum use of directed horizontal sterile air flow for surgeries. Clin. Orthop.,in press, 1974. 5. Coventry, M. B., D. R. Nolan, R. D. Bechkenbauch, and D. M. Ilstrup, 2,012 total hip arthroplasties: A study of postoperative course and early compli- cations. J. Bone Joint Surg. 56A:273-284, 1974. 6. Irvine, R., B. L. Johnson, and H. C. Amstutz, The relationship of genitouri- nary tract procedures to deep sepsis in total hip replacements. An evaluation of 274 procedures including microbiological studies of air sampled from con- ventional and clean air operating rooms. Surg. Gynecol. Obstet. 139:70l-706, 1974. 7. Stinchfield, F. E., Eric S. White, N. S. Eftekhar, and K. M. Kurokawa, Low friction arthroplasty. Surg. Gynecol. Obstet. 135:l-10, 1972. 8. Wilson, P. D., H. C. Amstutz, A. Czerniecki, E. A. Salvati, and D. E. Mendes, Total hip replacement with fixation by acrylic cement. A preliminary study of 100 consecutive McKee-Farrar prosthetic replacements. J. Bone Joint Surg. 54A:207-236, 1972.
33 LOCAL AIRFLOW PROTECTION OF SURGICAL WOUNDS TO PREVENT AIRBORNE CONTAMINATION William E. Anspach, Jr.* M. BakeIs It is well established that wounds can be contaminated by airborne bacteria. That these airborne bacteria are a significant cause of postopera- tive wound infection, however, has not been proven. If one reviews the avail- able literature on wound infection, one must reasonably consider that the air is one of many possible modes of surgical contamination. Charnley's infection rate in total-hip replacement surgery has de- creased dramatically in recent years (2). He attributes this to his clean room facility. One cannot help but feel that the same effect may have come as a result of his increased experience with the procedure and, in part, to his elaborate preoperative preparation of the patient. Series of cases presently being gathered in the United States (4) already show that a super-clean ordi- nary operating room in which the surgeons follow a rigid aseptic technique, and in which modifications to present air filtration systems are made, can re- sult in an infection rate for total-hip replacement as satisfactory as Charnley's (13). These are relatively early studies, and with the well known delayed onset of infection in total-hip replacement surgery, the figures may not stand the test of time. If one approaches the problems of surgical wound contamination with these thoughts in mind, and realizes that the air is only one mode of bacterial transport, then a rational program to eliminate these factors can be considered. Bacteria range in size from 0.3y to over lOy. Bacteria in air will usually not appear singly, but rather in droplets, in dried droplet nuclei, at- tached to dust, or on epithelial cells. The particles which carry these micro- organisms are usually 8y to 14y in diameter (9,10,11). It has been appreciated for some time that the highest concentrations of bacteria in an operating room are found within the circle of the surgical team, directly over the wound (3,8). It is ironic that such is the case, but its cause can be readily understood when one considers that the wound is "where the action is." This is where the surgeon's breath forms a droplet aerosol, and his face and neck shed bacteria- laden particles of skin and sweat. * North Palm Beach, Florida; Member of the Committee on Operating Room Environment and Surgical Asepsis of the American Academy of Orthopaedic Surgeons.
34 William E. Anspach, Jr. M. Bakels To become infected, a wound must not only have a low resistance, but the virulence and dose of bacteria must be substantial. In the presence of foreign material, the infective dose of bacteria is substantially less than in its absence (3,5). It is in these wounds that small quantities of airborne bacteria previously considered insignificant may play a major role in the onset of a subsequent infection. It is generally agreed that the operating room personnel and patient are the largest source of bacteria (6,11). It is therefore only reasonable that attempts be made to isolate these reservoirs from the surgical environ- ment despite the type of air filtration system used. Unidirectional airflow systems for removing bacteria from the air in operating rooms are generally quite expensive, hot, and frequently noisy. To eliminate the heat generated by the electric motors and compression of air be- hind filter banks, major modifications to existing air conditioning systems may be required. Major changes in electrical wiring and in the construction of the operating suite itself may also be required. Horizontal laminar-flow wall systems have the disadvantage that they create laminar flow for only a distance of approximately three feet from the filter bank (8). From this point the air becomes turbulent, and this effect is accentuated by the presence of any obstruction in the airflow, i.e., instrument tables, the operating table, and the surgical team. Because of this turbulence, it is conceivable that eddy currents set up about the operative site could actually bring air from the surgeon's head and neck into contact with the wound. Because of the aforementioned problems concerning turbulence, cost, heat, and noise generation in existing clean air systems, a new method was de- veloped. It was felt that placing a source of unidirectional flow bacteria- free air quite close to the wound could eliminate much of the undesirable tur- bulence at the operative site. By making the air filtration units small, it was hoped that heat, noise generation, and cost could be kept to a minimum. In early 1971, after much trial and error, such a system was devised. It was not until a year later that Dr. Beck's (1) work on a similar system was
35 Local Airflow Protection brought to our attention. A prospective controlled study was designed to evaluate its effect on the number of bacteria and particles at the wound and over the instrument tables during surgery, and on the particulate content of the operating room outside the direct flow of air. All studies were conducted in a small community hospital on clean orthopaedic and general surgical cases. A filter system was devised which projects essentially sterile air at a desired tangential angle across the surgical wound. The air is first passed through a prefilter which removes large particles of lint and dust, then it is forced through a high efficiency particulate air (HEPA) filter which removes 99.9 percent of all particles 0.3y and larger. The exhausted air obtains its identity and flows in smooth, streamline, or laminar-flow profiles having a Reynolds number less than 2,000. This air is then passed through an exit nozzle which was made removable for sterilization so that it could be moved into close proximity to the surgical wound. The entire unit was mounted on an adjustable base, so that the height and the potential flow core angle could be changed at will (Fig. 1). The maximum design capacity of the unit was 2,000 cubic feet per minute; however, the blower speed was variable so that flow rates could be ad- justed. In an average operating theater, the air theoretically could be fil- tered in one minute. Actually, part of the air would be in a recirculating sequence, and the probability would be that in a 2,000-cubic foot operating room, 50 percent would have passed through the system in one minute, and 90 percent would have been exchanged in approximately 10 minutes. AIRFLOW STUDIES Several studies of air jets exhausting into quiescent air were made to achieve a laminar air profile having a constant velocity region as wide as possible from a small source. If air is exhausted through a filter at uneven pressures, the high pressure zones will tend to mix with low pressure areas. Thus, the streamline having the higher pressure will shear the weaker stream, become unstable, and eventually break up to form vortices. These vortices will transport mass in- ertia, temperature, and any other characteristics slowly across the stream.
36 Fig. 1. Filter unit on an adjustable base.
37 Local Airflow Protection Due to the importance of equalizing the pressure over the HEPA filter face, a series of vanes was developed and mounted at the blower exit orifice. The extreme turbulence created by these vanes resulted in a remarkably equal diffusion of pressure behind the HEPA filter. Airflow measurements were made with a hot wire anemometer for low velocity flows and for determining the turbulent regions. A Kiel probe in conjunction with a six-degree inclined micromanometer was used to plot the core flow region. To further extend the laminar airflow region, a low angle nozzle was designed so that the filter exit flow would be compressed and thus maintain its identity for a longer time when exhausted into ambient environment. Entrainment of room air into the clean zone was discouraged by placing sterile drape sheets at a tangential angle along the sides of the exit nozzle. The operating team along the table is usually in quite close contact with these drapes so that a partial side wall along the sterile flow is created. Warren (12), Kranz (7) and others have shown that air streams will remain linear in a constant velocity coneshaped pattern when released from an orifice for a distance equal to approximately three times the orifice diameter. That this distance was extended by the incorporated modifications to the air source is shown in Figure 2, which illustrates the actual measured air patterns. The central triangle of the diagram should be considered a three- dimensional cone. The apex of this constant velocity region where the flow is laminar extends to six feet eight inches from the filter face, after which the velocity drops and gradually becomes zero. As can be seen in Figure 2, the flow becomes less laminar outside the central core. The outer funnel-shaped area is the edge of the flow profile in which maximum mixing with ambient air takes place. It delineates the most peripheral area in which the filtered air has any effect, and where zero velocity was measured with the hot wire anemometer.
38 William E. Anspach, Jr. M. Bakels 5JHDNI Nl ONIOV3JMS 1 CS s
39 Local Airflow Protection PARTICLE COUNTS Particle counts were obtained using a Coulter 550 Airborne Particle Counter which tabulates the number of 0.5y particles per cubic foot of aspirat- ed air. A hose was run from the counter to the area in which air was to be sampled. During surgery, a sterile plastic tube was clamped into the surgical incision so that its orifice aspirated air directly from the wound. It has been our experience that when studies of airborne particles are made in an operating room, they are seldom taken directly from the wound but rather from an adjacent area which would not produce data representative of the actual wound. Control samples were taken with the air unit turned off but with the tube left in its original position. Intermittently, during each surgical case, air samples were taken over the instrument tables. Particle counts were also taken with and without the surgical team's hands and arms in the airflow. Figure 3 compares the number of 0.5y particles counted per cubic foot of air at the wound during surgery within the airflow, and particles counted with the unit turned off. The air unit was always kept within 42 inches of the wound and closer, if possible. The average of 61.3 particles per cubic foot with air unit turned on is considerably less than the 19,541 particles per cubic foot found in the same area with the system off. It was initially felt that the turbulence generated by the surgeon's hands and arms in the airflow would adversely affect particle counts. There is no question that turbulence is generated in this fashion, but we were surprised to find that the particulate content rose only if the surgeon's hands were placed directly upwind and in line with the sampling probe. If the hands were placed even an inch or so above the sampling orifice, no increase in the par- ticle count was recorded. It is hypothesized that the central core of laminar flow air is quite uniform and that transfer of particulate matter from one lamina to another occurs only with some difficulty in this region. Particle counts made over the back instrument tables averaged 1,500 - 2,500 0.5y particles or larger per cubic foot with the unit on, as compared to 15,000 with the unit off.
40 William E. Anspach, Jr. M. Bakels ⢠» FIG 1 ⢠| . i ⢠⢠t J f I , ⢠1 r ⢠⢠⢠â¢* ⢠⢠1 ⢠; J ⢠* 10.000. : ⢠^ ; r * 4 â r- ⢠⢠0 AIR SYSTEM OFF j f T i t 1.000. PARTICLES >0.5p PER CUBIC FOOT OF WOUND A!3 AVG. ⢠19,541 RANGE - 3,iOO ⢠65,000 AIM SYSTEM ON iAVG. ⢠613 RANG* -19. 180 ⢠f I 100 ⢠~ * r 1 * K Jr ⢠⢠r V p * p J tf ⢠⢠⢠t i " f ' M -J> ^_ * * f â¢Â» ⢠9 ⢠I* ! ⢠| ^ r m ⢠I * E j» f 1 . f f ⢠⢠⢠* t f . 'f ⢠⢠â¢Â» t i f » ⢠*,. p * ⢠⢠CAIU 7 1 » 10 11 12 11 14 11 14 lOGAJUTHMfC KAU Fig. 3. Comparison of particles counted during surgery with airflow unit on vs. airflow unit off.
41 Local Airflow Protection BACTERIOLOGIC SAMPLING Viable airborne particles were counted by clamping two sterile plastic hoses with their orifices directly at the wound edge. The hoses were kept as short as possible to minimize entrapment of bacteria along their inner walls, and were connected to either a standard millipore air sampler or a slit sampler. The millipore sampler is, essentially, a sterile chamber containing 30 cc brain-heart infusion broth. Air is projected into the broth, thus trap- ping most particulate matter. This culture media is then passed through a millipore filter where bacteria are deposited on its surface. A pad beneath the filter is kept saturated with a nutrient media to promote bacterial growth. This system, although theoretically quite efficient, has many disadvantages. It is fragile, difficult to set up, samples only one cubic foot of air, and if the slightest error is made when the broth is drawn through the filter, the filter is left too dry and very little growth will result. At the present time all counts are taken with a New Brunswick STA-101 slit sampler which moni- tors two cubic feet per minute. The experimental samples were taken with the air unit on, and the control samples aspirated from the same region with the filter turned off. In ten clean orthopaedic cases an average of .83 viable particles per cubic foot were found with the unit on, as compared to 12.5 with it off. Figure 4 shows similar data except the samples were collected by a slit sampler (New Brunswick STA-101). In this series, a comparison was made to a horizontal air wall sys- tem installed in the same operating room. The types of bacteria isolated on blood agar are shown as approximate percentages in Figure 5. These compare favorably with studies performed by Nelson (8).
42 William E. Anspach, Jr. M. Bakels PRKticiES m cubic FOOT Cleao Or^opedtC Scmvplcr Ranne Insirunenf flverofje R^T o J Nc fl.r Flou/ 1.8 0.7 -S.O 0.9 o.-,0 Local H.r plow/ .1 0-0-4 0-2 .02-/.7 u â¢2 0-4 .1 O-.8 Fig. 4. Viable particles per cubic foot. - Bacillus SubtiliS Tbeudomonas Fig. 5. Types of bacteria.
43 Local Airflow Protection DISCUSSION The present report describes the effectiveness of locally directed clean air for the protection of surgical wounds from airborne contamination. The core area of true laminar flow across the wound was found to form a very effective barrier to prevent particles dropped into it from reaching the wound. Particulate matter generated by the surgical team and entrained in this flow could, however, be carried away and deposited on the open instrument tables. This does not seem to be a significant factor however, as the counts over these tables were satisfactorily low. To minimize this posibility, we are now using aspirator suits for all personnel in the opera- ting room. The surgeon's expelled breath is exhausted via a small vacuum sys- tem into the air unit's HEPA filter where it is isolated from the ambient environment. It may be that these additional steps are unnecessary, as the total room counts are qaite low. On the other hand, it may eventually be proven that isolating the surgical team and the patient from the operating room atmosphere will result in such low air bacterial counts that no special air filtration system may be required. SUMMARY A portable air filtration unit for use in hospital operating rooms has been described. Prospective studies of its effectiveness in controlling airborne (particles and bacteria were presented, and compared to a horizontal wall system. It is concluded that both systems significantly decrease the number of airborne bacteria. REFERENCES 1. Beck, W. C., Laminar air barrier for wound asepsis: A preliminary report. Guthrie Clinic Bull. 35:126-134, 1964. 2. Charnley, J., and N. Eftekhar, Postoperative infection in total pros- thetic replacement arthroplasty of the hip joint. Br. J. Sure. 56:641-649, 1969. 3. Coriell, L. L., Research on airborne contamination. Med. Sure. Rev. 7. 1971. 4. Coventry, M., Personal Communication.
William E. Anspach, Jr. M. Bakels 5. Elek, S. D., Staphylococcus Pyogenes and its Relation to Disease. Williams & Wilkins, Baltimore, Maryland, 1959. 6. Howe, C. W., and A. T. Marston, A study of sources of postoperative staphyloccocal infection. Surg. Gynecol. Obstet. 115:266-275, 1962. 7. Kranz, P., Jet stream ventilation for extreme air cleanliness, ASHRAE J. 4:37-39, August 1962. 8. Nelson, J. P., Use of total horizontal laminar flow systems in surgery Symposium on Clean Room Technology in Surgery Suites, NASA Midwest Re- search Institute, Cape Kennedy, pp. 67-79, May 1971. 9. Noble, W. C., The size distribution of airborne particles carrying micro- organisms. J. Hyg. (Camb) 61:385-391, 1963. 10. Smith, L. R., Eight years' experience using an aseptic air system in sur- gery. Int. Surg. 52:135-140, 1969. 11. Walter, C. W., R. B. Kundsin, and M. M. Brubaker, The incidence of airborne wound infection during operation. JAMA 186:908-913, 1963. 12. Warren, W. R., An analytical and experimental study of compressible free jets. University Microfilms, Ann Arbor, Michigan, 1970. 13. Welch, R., Symposium on Bio-clean Rooms. San Francisco, California, November 1971.
45 THE DEFINING OF BIOLOGIC CLEAN AIR AND ITS CLINICAL SIGNIFICANCE William C. Beck* Every surgeon wants to work in an operating room surrounded by biologically clean air. The Committee on Operating Room Environment of the American College of Surgeons published a position paper in 1972 in which it mentioned clean air emerging from a final bacterial filter. It is generally believed by this committee that the 1972 statement is still valid. There is presently no definition of biologic clean air. If one wants to make any specification in either the design, the function, or the effects of the use of either conventional ventilation or special air systems in operating rooms, one is forced to employ the definitions of Federal Stan- dard 209B (2). This standard makes reference only to the particulate control parameters and not to desirable microbial control conditions. It states that airborne microorganisms are a part of the particulate count of the different air cleanliness classes. An attempt was made in the NASA standard for the microbially con- trolled environment to adapt the clean air of industry by engrafting "bio- clean" concepts upon the particulate ones. In surgery, we have no information at all that nonviable particu- lates have any adverse effects upon the healing processes, at least in the numbers found in our operating rooms (although Warren Cole has suggested that airborne powders from rubber gloves might supply rafts for carrying micro- organisms) (l). Therefore, many of the constraints imposed by either 209B or NHB 5340.2 (4; do not apply to surgical environments. There still remains a need for a definition of biologic clean air. It is not difficult to create such a standard if one uses the guidelines on which the particulate standards were based: the recognition of degrees of cleanliness or, to stay with existing terminology, Biologic Cleanliness Classes. *President, Donald Guthrie Foundation for Medical Research, Sayre, Pennsylvania.
William C. Beck CREATING A DEFINITION Some conference, such as the one in which we are presently participating, could create a committee which would set up three or four levels of air cleanliness. Federal Standard 209B does this (Table 1). Our committee could do it in an identical manner (Table 2). There would also be needed a variety of other definitions which would enable the designer to specify, the manufacturer to provide, and the experimenter to evaluate the effect of clean air systems, both conventional and special. These can parallel 209B. It must be remembered that the min- imum requirements of construction and equipment for hospital and medical facilities of the Health Resources Administration (HEW-74-4000 (3), which are guides to all engineers and architects, define filter efficiencies, air changes, pressure relationships, etc., for all hospital areas. These are the conventional systems which also should be put to test for level of bio- logic cleanliness. Certain decisions will be necessary. The first, of course, would be to accept viable particles rather than bacterial counts. The second would be to accept the use of a specified volumetric air sampling system with certain minimum feet per minute so that statistically valid figures were ob- tained . SIGNIFICANCE From an economic standpoint, it might prove very useful to have a standard of biologic cleanliness and not have to rely upon 209B. Ulrich, Cribbs, and Michaelson (5) have shown that High Efficiency Particulate Air (HEPA) filtration is not necessary in removing viable particulates. Other studies have shown that completely acceptable results are obtainable using 99.0 percent efficient filtration rather than the previously ubiquitous HEPA filter. Possibly even lesser filtration would be even more efficient thus permitting lesser blowers, lowered noise, and diminished cost. Such effect can only be sought if we can define the clean air for surgical environments^ The clinical significance of such a definition will permit us to evaluate whether or not special air systems are worth the added expense, and
47 The defining of Biologic Clean Air and its Clinical Significance might extend the studies of the type which have been reported at the conference with a baseline of a common definition. This will reduce one of the variables in the study of surgical nosocomial infection. Table 1. Air Cleanliness Classes Maximum number of particles per cu ft (per liter)* 0.5 micron & larger Class English System (Metric System)* Maximum number of particles per cu ft (per liter)* 5.0 micron & larger 100 (3.5)* 10,000 (350)* 100,000 (3500)* 100 (3.5)* 10,000 (350)* 100,000 (3500)* 65 (2.3)* 700 (25)* *Metric system Table 2. Air Cleanliness Classes Maximum number of Maximum number of Maximum number of bacterial particles bacterial particles particles in total per cu ft per cu meter sample test of ( ) feet 1 * 35 (30) 30 5 ** 175 (30) 150 20 *** 700 (10) 200 *Class 1 Microbiologic Cleanliness: viable microbiologic airborne particle counts not to exceed one particle per cubic foot of air, with a minimum sample of 30 cubic feet of air. **Class 5 Microbiologic Cleanliness: viable microbiologic airborne particles to average more than one and up to five per cubic foot of air with a minimum sample of 30 cubic feet of air. ***Class 20 Microbiologic Cleanliness: viable microbiologic airborne particles to average more than five, and not to exceed 20 per cubic foot with a minimum sample of 10 cubic feet of air. These classifications are based upon viable microbiologic particle counts (colony forming units). Counts are to be taken during periods of normal work activity at a location which will yield the vi- able particle count of the air as it approaches the location of the actua] site of the work and/or equipment used in the work. This may be at a surgical incision, at an instrument table, etc. The site must be described for each sampling including surface, height from floor, relation to walls, etc. Reliability is achieved by multiple repeti- tive sampling.
48 William C. Beck REFERENCES 1. Cole, W. H., Eliminate powdering hands and packing linens in the operating room. Ann Surg. 153:161, February 1961 2. Department of Health, Education, and Welfare, (HRA), Federal Standard No. 209B, Clean room and work station requirements, controlled environ- ment, April 24, 1973. 3. Department of Health, Education, and Welfare (HRA), Publication No. 74- 4000, Minimum requirements of construction and equipment for hospitals and medical facilities. 4. NASA Handbook, NHB 5340.2, Clean rooms and work stations for the micro- bially controlled environment, August 1967. 5. Ulrich, John, W. Cribbs, and G. S. Michaelson, Recirculation of air in operating rooms. Report under Research Contract PH 108-65-26, Division of Hospital and Medical Facilities, Bureau of Health Services, Public Health Service, Washington, D.C. ASHRAE Journal, p. 56, August 1974.
49 ULTRACLEAN OPERATING THEATRES VERSUS CONVENTIONAL THEATRESâ A CONTROLLED CLINICAL TRIAL George Bentley and Andrew B. Simmonds* INTRODUCTION From early times it was appreciated that contact contamination could be reduced by improved hygiene (Semelweiss 1861). Later Lister improved wound sepsis by antiseptic surgery. The role of airborne bacteria in the aetiology of sepsis was first introduced in relation to burns (Bourdillon and Colebrook 1946). The use of Plenum ventilation gradually replaced exhaust ventilation in the 1950s (Girdlestone et al., 1951) and different types were evaluated by Blowers and Crew (1960). The logical extension of this work was the introduction of clean air downward linear flow enclosures (Charnley 1964). During the 1960s, the American aerospace industry introduced industrial laminar flow clean rooms with High Efficiency Particulate Air (HEPA) filters, and these were later introduced into medical practice (Lidwell and Towers 1969, Whyte and Shaw 1973). Both developments result in substantial reduction in airborne bacteria (Charnley 1972, Lidwell and Towers 1969, Scott 1971) and the relative merits are largely logistic. Improved surgical wound sepsis rates have been claimed by Charnley (1969, 1972) by operating on patients in ultra-clean air, although similar low sepsis rates have been reported by others using systemi- cally administered prophylactic antibiotics in a conventional operating room (Coventry et al., 1974). The need for a controlled clinical trial of these ultra-clean air systems is clear, and so far has not been performed. This paper describes the early results of a trial in which air contamination, wound contamination, and wound infection were compared in two vertical laminar flow enclosures and a conventional operating room. METHOD After six month's practice with body exhaust systems a controlled clinical trial was set up. A physical assessment of the airflow patterns *Nuffield Department of Orthopaedic Surgery, Nuffield Orthopaedic Centre, Oxford, England.
50 Ultra-Clean Operating Theatres within the enclosures and a conventional operating room was made using the Hiamine bubble technique (Carpenter and Moulsley, 1972). All patients having total-joint replacement surgery were randomly allocated to either conventional theatre or enclosure independently for each of five orthopaedic services. Pre- operative urinary, nasal, and perineal bacteriology was performed. During the operation three settle plates were placed as follows: one immediately behind the operator, one on the opposite side of the operating table, and the third at the foot of the table. Volumetric assessment was made by slit sampler studies (700 litres/min; 20 ft /min for three two-minute periods on a limited number from both environments using a sterile wide-bore polythene extension tube aspir- ing air from the immediate vicinity of the wound. Quantitative wound bacteriology was performed by a washout procedure (after Blowers 1973). One hundred fifty millilitres of 25 percent Ringer saline solution were washed into the blood-free wound at the close of surgery. (Fig. 1). QUANTITATIVE WOUND SAMPLING (after Blowers, 1973) TOTAL NUMBER PERFORMED 245 CONVENTIONAL THEATRE 116 ENCLOSURE 129 25 Ringer saline nutrient & agar 5000 rpm. O'C 5 min. ?antibiotic activity Fig. 1. Method of wound washout and quantitative sampling (after Blowers, 1973).
51 George Bentley and Andrew B. Simmonds The fluid was collected in a sterile pot, cooled, centrifuged, and the supernatant filtered through a double millipore chamber (3y and 0.45y). Each filter was washed through with Ringer saline and the filters placed on blood agar plates. The deposit was made into a pour plate after mixing into 20 m1s. of liquid nutrient agar. The filtrate was assessed for antibacterial activity as the patients routinely received systemic cloxacillin prophylaxis with the premedication one hour before induction of general anaesthesia. Postoperatively, wounds were individually examined and discharges were assessed bacteriologically. Routine and independent follow-up has been performed on all patients. RESULTS The Patients Three hundred and sixty six total-joint replacements were performed over the 10-month period studied, 225 in the enclosures and 141 in the conven- tional operating room (Table 1). Table 1. Total-Joint Replacements Performed ENCLOSURE 225 CONVENTIONAL 141 TOTAL 366
52 Ultra-Clean Operating Theatres Staphylococcus aureus was grown from the nares in 27 patients (19.8 percent) and from the perineum in nine patients (7.5 percent). The midstream speci- mens of urine contained coliforms in 18 patients (14.5 percent). Airflow Patterns The assessment of the airflow patterns was a rewarding exercise and revealed satisfactory linear flow rates of 100 ft/min in the central part of the enclosure with slower flow rates of 30 to 90 ft/min at the sides (Fig. 2), Fig. 2. Photograph of air bubble pattern in Charnley-type en- closure demonstrated by Hiamine technique of Carpenter and Mousley (1972).
53 George Bentley and Andrew B. Simmonds The pattern of airflow in the conventional operating theatre was from inlets on the upper wall at one end, across the theatre diagonally to exhaust vents immediately above floor level at the other end. The conventional operating room was found to have more turbulent flow than had been previously thought with variable flow rates [30 to 90 ft/min in the operating area but with considerable mixing and recycling (Fig. 3)]. Fig. 3. Air bubble pattern in conventional theatre at operation site showing satisfactory horizontal flow.
54 Ultra-Clean Operating Theatres Bacterial Colony Counts Settle plate studies revealed a 10-fold difference between the airborne bacteria in the two environments (Table 2). Table 2. Airborne Bacterial Sedimentation Mean Standard col./sq.ft./hour Deviation ENCLOSURE 1 ENCLOSURE 2 CONVENTIONAL THEATRE 14.87 16.28 166.72 11.94 12.07 65.94 i These remained fairly constant throughout the study. Volumetric sampling by slit sampler was more difficult but the results showed a similar 15-fold dif- ference between the two environments. There was, however, an approximate 20 percent tube Toss resulting in lower recovery rates than would have occurred were it possible to sample air without this exentsion tube. The results are shown in Tav,le 3.
55 «⢠George Bentley and Andrew B. Simmonds Table 3. VOLUMETRIC SAMPLiNG AIRBORNE BACTERIA slit samples - 3 samples of 1400 litres (SOcu.ft.) total air volume sampled - 4,200 litres (150cu.ft.) tube loss o, 20 '/. ENCLOSURE CONVENTlONAL THEATRE SLIT SAMPLES BACTERIAL SEDIMENTATION 0.38 c/cf av. 10 col./sq.ft./hour 5.3 c/cf av. 150 col./sq.ft./hour Wound Contamination Results of quantitative wound samples are shown in Figure 4. The majority of patients from both environments had minimal contamination but small numbers from both environments had more than 30 organisms recovered. These small number of "significant" washouts are tabulated in Tables 4 and 5. It can be seen that there is a large number of high wound washouts from the conventional theatre [17 out of 117 (15 percent)] than the enclosures 2 with 9 out of 130 (7 percent). Applying the X test the result is 3.78 2 (X =3.84 indicating significance at five percent level) so these results are of borderline significance.
56 - I â¢5 4 Ultra-Clean Operating Theatres Fig. 4. QUANTITATIVE WOUND SAMPLING total no. washouts performed conventional 116 200 No. of colonies per washout Table 4. QUANTITATIVE WOUND SAMPLING no. of organisms grown : <30 30-100 > 100 CONVENTlONAL THEATRE ENCLOSURE 100 121 8
57 J George Bentley and Andrew B. Simmonds Table 5. QUANTITATIVE WOUND CONTAMINATION no. of colonies: <30 >30 total CONVENTIONAL THEATRE ENCLOSURE Total 100 121 221 17 26 117 130 247 X2= 3.78 ( X2 significant at 57. - 3.84 ) i.e. BORDERLINE SIGNIFICANCE If these results are analysed from a lower level of washout the results become significant. For example, results comparing the two theatres taking washouts over 20 organisms show that the conventional theatre has a 2 higher statistical number (p = 0.0001, X = 7.94). However, we decided to use the former level, since Lowbury and colleagues were able to recover up to 28 organisms regularly in wounds which were not infected using a different method in hip fractures (Lilly et al., 1970). The organisms recovered from washouts were all skin commensal, and were roughly indentified by their colonial appear- ance.
58 Ultra-Clean Operating Theatres The influence of systemic antibiotics on the growth of organisms is interesting. Antibacterial activity was noted in 29.8 percent of the washouts and there was an overall reduction in numbers of organisms recovered from these. However, there were four notable exceptions in which large numbers of organisms grew out in the presence of antibacterial activity in the filtrate (Table 6) . Table 6. Presence of Antiobiotic Activity in Washouts organisms: < 30 30-100 >100 PRESENT ABSENT 62 (31.72) 133 (68.3%) 11 11 total 66 155 Percentage with antibiotic in washout 29.8% CLINICAL FOLLOW-UP The clinical follow-up is, as yet, too short to provide clinical correlation with crude sepsis figures. Only three patients have developed probable or certain sepsis, two from the conventional threatre, and one from the enclosure. Two of the patients had "significant" washouts of more than 30 colonies (one from the conventional theatre and one from the enclosure). At the same time there were eight wounds in the enclosure and 16 wounds in the conventional theatre with more than 30 organisms, but no infection at 6-12 months.
59 George Bentley and Andrew B. Simmonds DISCUSSION It is likely that sepsis in joint replacement surgery arises through wound contamination at the time of surgery. The source of contamination remains uncertain, and modern ultra-clean operating environments can predictably reduce airborne bacteria. Whether this can eliminate or reduce contamination from contact sources is doubtful. In the vertical laminar flow enclosure, the prevention of contact contamination depends on the efficiency of the body exhaust systems. Quantitative wound sampling in our study has demonstrated a trend toward less wound contamination in vertical laminar flow enclosures using body exhaust systems, compared with standard operating dress in a conventional theatre. Th'ere remains, however, a small unpredictable group of cases in which ⢠heavy wound contamination occurs, the source of which is uncertain. SUMMARY 1. Three hundred and sixty six total-joint replacements were randomly allocated to one of two clean air enclosures or a conventional operating room. 2. Physical evaluation of the airflow patterns by the Hiamine bubble technique was instructive and simple. 3. A 10-fold difference in airborne bacteria was noted with the higher levels in the conventional room. A. Quantitative wound sampling has revealed a higher rate of wound contamina- tion in conventional rooms than in the laminar flow enclosures, the signifi- cance of which is unknown. 5. Three wound infections have occurred: two in the conventional theatre and one in the enclosures. Two of these had "significant" wound contamination at the time of surgery but 24 wounds had a "significant" contamination with no infection. 6. The relative importance of impermeable theatre clothing in the different operating environments is not yet evaluated.
60 George Bentley and Andrew B. Simmonds ACKNOWLEDGEMENTS We wish to thank all the consultant staff at the Nuffield Orthopaedic Centre, Oxford for their generous cooperation in this study. Professor R. B. Duthie gave constant encouragement and advice. The bacterio- logical investigations were performed by Dr. R. G. Mitchell, Consultant Bac- teriologist to the United Oxford Hospitals who gave valuable advice throughout the study. The manuscript was typed by Mrs. J. M. Lever and the illustrations were prepared by the Medical Illustration and Photographic Departments of the Nuffield Orthopaedic Centre. I
61 Ultra-Clean Operating Theatres REFERENCES 1. Blowers, R., Personal communication. 1973. 2. Blowers, R., and B. Crew, Ventilation of operating theaters. J. Hyg. Camb. 58:427-448, 1960. 3. Bourdillon, R. B., and L. Colebrook, Lancet i:561-601, 1946. 4. Carpenter, G., and L. H. Moulsley, A visualization technique for studying air movement in large enclosures over a wide range of ventilation rates. Journal of Institute, of Heating and Ventilation Engineers. 39:279-287, 1972. 5. Charnley, J., Sterile-air operating theatre enclosure. Brit. J. Surg. 51:195, 1964. 6. Charnley, J., Post-operative infection after total hip replacement with special reference to air contamination in the operating room. Clin. Orthop. 87:167-187, 1972. 7. Charnley, J., and N. Eftekhar, Post-operative infection in total prosthetic replacement arthroplasty of the hip joint. Brit. J. Surg.56:9:641-649. 8. Coventry, M. B., R. D. Beckenbaugh, D. R. Nolan, and D. M. Illstrup, 2012 total hip arthroplasties. A study of post-operative course and early com- plications. J. Bone Joint. Surg. 56A:2:273-284, 1974. 9. Girdlestone, G.R.G., R. B. Bourdillon, and A. M. McFarlan, Infection of "clean" surgical wounds by the surgeon and from the air. Lancet i:597, March 17, 1951. 10. Lidwell, 0. M., and A. G. Towers, Protection from microbial contamination in a room ventilated by uni-directional air flow. J. Hyg. Camb. 67:95-106, 1969. 11. Lidwell, 0. M. and A. G. Towers, Unidirectional flow ventilated for patient isolation. Lancet, i, 7746, 1972. 12. Lilly, H. A., P. S. London, E. J. L. Lowburv, and M. F. Porter, Effects of adhesive drapes on contamination of operation wounds. Lancet i:43l-432, 1970. 13. Semelweis, I. P., Die Aetiologie der Bengriff und die Prophylaxis des Kindbettfiebers. Pest, Wien and Leipzig. 1861. 14. Scott, J.H.S., Laminar flow theatre ventilation. Orthopaedics Oxford. 4:2:91-97, 1971. 15. Whyte, W., and B. H. Shaw, A bacteriological evaluation of laminar flow systems for orthopaedic surgery. J. Hyg. Camb. 71:1-6, 1973.
62 LOW VELOCITY UNIDIRECTIONAL ULTRAFILTERED AIR SYSTEMS Lewis L. Coriell* Laminar airflow ventilation is widely accepted as being the most efficient method for removing particles and microorganisms from air. The two principles involved, ultrafiltration through High Efficiency Particulate Air (HEPA) filters and rapid piston-like displacement of air assures almost complete freedom from viable airborne microorganisms. HEPA filters were first developed by the Army Chemical Corps for use in respirators and sampling devices (17) and long experience has confirmed that when properly installed they are durable and efficient for removal of airborne bacteria and viruses. The value of the laminar flow principle described by Whitfield (45) has been confirmed in many studies, even though true laminar flow is never achieved in practice. The term should be replaced by a more appropriate descriptor. "Laminar flow" has come to mean ven- tilation velocity of 100 ft/min throughout an enclosure as recommended by Whitfield and reinforced by issuance of US Air Force Technical Order No. 00-25- 203 (40) and by Federal Standard No. 209A (13). This defines laminar airflow as "airflow in which the entire body of air within a confined area moves with uniform velocity along parallel flow lines." Many installations have employed lower flow rates successfully, and it is the purpose of this report to review the use of lower flow rates in ultrafiltered unidirectional air systems for control of airborne contamination. FILTER EFFICIENCY VS AIR VELOCITY Harsted et al. (19) have shown that the efficiency of modern ultra- high-efficiency filter papers for retention of Tl bacteriophage aerosols is greatest when air passes through at low velocity, and penetration increases pro- gressively as the velocity increases. Data in Table 1 extracted from the Harsted report (19) show that this is true for filter paper from each of three different manufacturers. The column in Table 1, labeled Filter Face Velocity in Feet/Minute is the speed at which air passes through the filter paper, and the second column is a 20-fold multiple of the first column listing the approximate flow rate across the face of a commercial filter where the filter paper surface area is increased 20-fold by means of accord!an folds to conserve space. The performance of the filter paper for neutralized phage particles observed by Harsted is close to the *Director, Institute for Medical Research, Camden, New Jersey
63 Lewis L. Coriell theoretical values derived by Whitby (43). Both theoretical calculations and direct measurements with a small viable particle indicate that it is not neces- %, sary to operate HEPA filters at a flow rate of 100 ft/min and, in fact, they are more efficient at low flow rates. Table 1. Effect of Velocity on Penetration of Neutralized Submicron Phage Aerosols Through Ultra-High-Efficiency Filter Papers* Filter Face Approximate Geometric Mean % Penetration Velocity Flow Rate Ft./Min. Ft./Min. Manuf. #1 Manuf. #2 Manuf. #3 1.1 22 0.00027 0.000021 0.000002 3.7 75 0.0091 0.0014 0.000083 17. 340 0.099 0.043 0.0027 68. 1400 0.34 0.19 0.017 TYPE OF VENTILATION Early studies by Bourdillon and associates (8,9) showed the preva- lence of airborne bacteria and a high incidence of surgical wound infection in surgical theaters ventilated by exhaust fans where dirty air was drawn into the room from the halls. Progressive improvement was obtained by the addition of positive pressure ventilation, filtration through cloth filters, plenum venti- lation, and exclusion of ward blankets and clothing. Many others added refine- ments to these observations (1,4,24,33,39) and by 1968 the British Medical Re- search Council considered plenum ventilation of operating suites an established practice like disinfection of the skin of patients (31). Laminar flow ventilation (45), more appropriately called unidirectional flow, is the successor to plenum ventilation and involves continuous introduction of filtered air at a uniform velocity across the whole ceiling or wall of a room, and its containment by four walls until it is exhausted through the floor or op- posite wall. The two major advantages of such ventilation are 1) all areas of the enclosure are ventilated by air free from microorganisms, and 2) any micro- organisms aerosolized within the enclosure from personnel or equipment are removed in seconds. Other described advantages (36) include: 3) scant opportunity for airborne particles to settle onto a wound or clean surface since they tend to move with the moving airstream; 4) reduced housekeeping and maintenance because less dust settles on exposed surfaces; 5) improved control of humidity and temperature *Extracted from Harsted et al Table 5, 1967 (Reference 5).
64 Low Velocity Unidirectional Ultrafiltered Air Systems through recirculation of already conditioned air; 6) freedom of movement of personnel within the conditioned space in downflow rooms; and 7) elimination of air locks, air showers, and extra garmenting in electronic assembly rooms. Some disadvantages of high velocity unidirectional ventilation systems are: cost of installation; increased volume of air moved, which re- quires larger motors, fans, and ducts; increased heat generation; higher noise level; increased space for ceiling or wall and floor plenums. In spite of their efficienty in removing particles and microorganisms, these disadvan- tages, plus the lack of proof of cost effectiveness in reduction of infections, have prevented widespread adoption of unidirectional filtered air systems in operating theaters. There have been many attempts to minimize the disadvan- tages by modifying some portion of the basic design without loss of effective- ness. One of the most promising modifications is to reduce the velocity of the unidirectional airflow because this modulates every one of the disadvan- tages listed above. VELOCITY One of the early publications by Whitfield (46) listed results of dust sampling at different flow rates in vented hoods. There were an average 3 of 213 particles/ft with a flow rate of 100 ft/min (600 air changes per hour) whereas a flow rate of 50 ft/min (300 air changes per hour) showed 450 parti- 3 cles/ft , an insignificant difference of 0.02 percent considering the sampling rate of dust samples (0.1 ft /min), the relative lack of sensitivity of samples in 1963, and the control of values of 10,000 particles/ft . In another report Whitfield (47) states that "Downflow rooms fulfill class 100 standards (less than 100 particles >0.5u in diameter) for the entire room, and crossflow rooms will maintain class 10,000 standards. Downflow rooms show good control at 60-100 ft/min air velocity, and for best control, cross- flow rooms should operate from 100 to 130 ft/min." These data indicate that lower velocities are more effective in vertical flow installations than in horizontal flow installations. Ir each of three recent controlled studies in operating rooms by Scott e^ al. in 1971 (38), Whyte £t al. in 1973 (48), and Wardle in 1973 (42), 3 the numbers of bacteria per ft /air pointed to vertical flow as the optimal
65 \'J Lewis L. Coriell airflow configuration for application in the operating room. The well controlled studies by Whyte et al. (48) showed that little improvement in airborne bacterial counts was obtained with air velocities over 60 ft/min (.3 m/sec). In a simulated nursing unit in a horizontal flow industrial clean room, Lidwell and Towers (24) found little movement of particles against the direction of airflow or lateral movement of particles if the linear velocity of air did not fall below 35 ft/min. We observed similar results in a ver- tical flow room with lower velocities of 25 ft/min using test suspensions of latex particles, T3 bacteriophage and Serratia marcescens. A low-velocity system was developed by Horneff (14) to reduce the disadvantages of high-velocity laminar flow. It utilized remotely located HEPA filters with the filtered air ducted to a perforated ceiling and verti- cal velocity of 15 ft/min, and gave good protection from bacterial contamina- tion as measured by settling plates and sieve sampler. Modifications of this system with flow rates of 19 to 25 ft/min have been used to prevent airborne contamination of cell cultures during transfer (29), spread of airborne in- fection in laboratory animals (3), and removal of viable airborne bacteria from an operating room during surgical procedures (15). Pelosi (34) com- pared particulate counts in a downflow clean room with flexible side walls and a plenum ceiling operated at velocities of 20, 50, and 80 ft/min. He observed no significant difference at the three velocities and therefore rec- ommended velocities of 20 ft/min. Patients with reduced resistance to infection have been studied in laminar flow nursing units with horizontal flow rates at 35 ft/min (24), 90 ft/min (6,7), and at 90 and 60 ft/min (35,37). In the latter study, 3 mean bacterial counts/ft of air before, during, and after bedmaking at air velocities of 90 and 60 ft/min were not significantly different, varying be- tween 0.04 and 0.06/ft . The mean was 170 times less than in a conventional isolation room and 370 times less than in a conventional hospital room. Lam- inar flow units have been devised with various air velocities for isolation of a single bed (10,23,26), and high-velocity flow air jets (2,49) directed
66 Low Velocity Unidirectional Ultrafiltered Air Systems at the operative site have been tested in surgical theaters. The air jets entrain dirty air, do not protect instrument tables and other equipment out- side the air jet, nor do they isolate observers and other supportive personnel in the room. An unrestrained filtered air jet draws contaminated air around its entire periphery which results in a mixture of clean and contaminated air. To counteract this, partial walls have been placed around the jet source by some investigators to contain the jet stream until it is almost delivered to the site to be protected. However, this also leads to contamination by creep of dirty air up the inner side of the partial retaining walls as observed by Whyte (4). We have also observed this phenomenon in early studies with an over-bed unit equipped with a ceiling plenum and flexible walls that reached to the surface of the bed. With vertical flow rates of - 25 ft/min and the bed in place, bacterial counts remained essentially zero within the enclosure while the room air was heavily contaminated. When the bed was removed smoke tests showed that room air crept up the inside of the flexible walls and con- taminated the whole enclosure. The basic principles of unidirectional airflow are as illustrated in Figure 1: a) entrainment quickly contaminates an unprotected jet stream; b) partial walls around a jet stream lead to contamination by aspiration of dirty air up the inner side of the partial walls; and unidirectional flow can be maintained only by c) complete containment of air by four walls until ex- hausted at the opposite wall or floor or until exhausted around the periphery at the floor through a restricted opening that increases the speed of the es- caping air to prevent penetration of contaminated air. Figure 2 shows appli- cation of these principles in a temporary installation with flexible side walls which is similar to the first OR built at the Lovelace Clinic. Figure 3 shows the same principles applied to a permanent hard-wall room that provides more usable space.
67 Lewis L. Coriell a ' n-t Fig. 1. Basic principle of unidirectional airflow; a) entrain- ment of dirty air by an unprotected jet of clean air; b) contamination by entrainment and aspiration in a jet with partial walls but no floor; and c) complete containment of the clear air jet by four walls and a restricted escape opening at two sides.
68 Low Velocity Unidirectional Ultrafiltered Air Systems Fig. 2. A flexible side-wall temporary downflow clean room that includes all the basic principles for delivering clean air to all critical areas of the room.
69 Lewis L. Coriell Fig. 3. A permanent hard-wall installation that protects all critical areas of the room, including more space for personne1, observers and equipment.
70 Low Velocity Unidirectional Ultrafiltered Air Systems INFECTION CONTROL Proof of efficiency of unidirectional ultrafiltered air in prevention of airborne infection in the operating room is still open to debate and is not the subject of this report. However, in testing the efficacy of low flow rates we have gone to controlled studies of airborne spread of infec- tion in laboratory animals. McGarrity et_ .al. (27,28,29) have shown effective control of airborne spread of infection of laboratory mice in cages placed six inches apart by vertical velocities of 35 ft/min. Rodent studies were per- formed using cross infection with Serratia sp., Proteus sp. and polyoma and Reo-3 viruses. Based in part on these studies more than 50 commercial organi- zations have installed low velocity (15-25 ft/min) vertical unidirectional ultrafiltered air systems to protect experimental and stock animals from air- borne infection (22). We cannot escape the conclusion that the air in animal rooms so equipped is cleaner than the air in the conventional operating room. DATA FROM THE OPERATING ROOM In 1971 Scott et^ a^. (38) observed after comparative studies in three operating rooms and three industrial clean rooms-- "operating rooms with their elaborate rituals are not as clean as some electronic workshops which admit and disgorge several shifts of workers each 24 hours" (Table 2). Whitcomb (44) was unable to draw firm conclusions about the beneficial effects of laminar airflow in the operating room because infection rates in his hos- pital were very low without the use of laminar flow ventilation- This situation occurs in other hospitals when aseptic procedures and Halsted's principles are faithfully followed (16). Morris and Burke (32) noted the common error of determining infection rates over too short a time period. They urged long-term studies, especially where infection rates are low to exclude the possibility that random variations based on chance alone could be responsible for an appar- ent alteration of infection rate. Lidwell (25), using multiple regression analy- sis in a study of postoperative wound infections in 12 hospitals, showed that differences in infection rates could be explained almost entirely by differences in types of patients and operative procedures. The infection rate of clean wound* in the average conventional operating room is in the range of three to five per- cent (3,5,18,20,21,41). , *
71 Lewis L. Coriell I CU <r CN M CN l CO C ^ CJN o o â¢H n 1 1 ⢠M OS O 0 0 O> rH U * M C cn -<r ^ 0 o (d ⢠⢠» d ^J- CO rH ^ CN i 0 0 O O 0 0 O O O O 0 0 o o m o 0 H o o o o o o 0 0 oo d * 0 0 ON ON tN rH l CO M ⢠oc O ON 0 0 m o CN CO CO* o a) d co cu 4J C r^ vO o m ON CN u a) a til CO CM CO CO rH C â¢H rH CO OS A A A CU rl U iH rH CO <f rH rl 4J â¢H TJ 0) i KH CO 4-1 rl 3. 3 cd m O O o O 0 0 O 0 C PH ⢠C O 0 0 o s»' rH O (0 M o 0 vO ON CU A OJ vO 0 o CO â¢a I** DO â¢H § ^ rn M 1 | â¢â¢ I 00 cS â¢o 0) tJ C M 3 DO c o 3 cd SB o c^- <>⢠0 0 o ⢠iH 4J 6 r, rH CN CN o C ,-! o cd 0) co H u 0) â¢H CU o. <3^ TJ 0 C cd *W *> >» cn O § O g â¢H a 4J 4J 4-1 4-1 Cd rH rH rH ^ g c g g 4-1 UH 4J U 4J CU 0) a u co Cd 4-> rH CO H | O H N CO â¢H rl o u â¢H â¢H ^i 4-J CO cn r< o fl o â¢H C cd 1 4J C/3 rl rl -H § o a > a 5 0) 4-1 4-1 U U J3 cd > rH rH M MH T3 ⢠⢠0) CM â¢o 4J ⢠< « u U <U ^^ ^_^ | O *-\ cd rH TJ ^ 0) H 01 CU 8 § g rH O s~> H Sco co 43 i "cd o o O 0 OS 0 u cd OS O ^-^ cd â¢H â¢H â¢H o o ^i^V â¢X rH CO 0 iH C ⢠cd O C « cd o C 0 0) â¢H â¢H â¢H cd o m CO ca 0) rH 0) rH 0) rH ^-J ON rH CO CM CO co co rH rH rH < 0 CM =«= cd "" rH =**= cd U CO o co U CO 4J rH rH co CO co ⢠U ⢠U ⢠CJ ⢠cd ⢠cd CO ⢠0) 4-1 SI . § O ^ â¢0 rH Tj rH TJ rH C U H cn 0 ^ C U C o M v-' M x-'
72 Low Velocity Unidirectional Ultrafiltered Air Systems On the other hand Charnley (H>12) makes an excellent case for âreduction of postoperative infection after total-hip replacement by control of the airborne route of infection. He stresses that evaluation for the statisti- cal significance of any new procedure in total-hip replacement requires at least 800 cases and a postoperative observation period of two years. Under the protection of clean air in the operating room he records the decline of both early and late infections including blood-borne infections, and attributes most of the decline to elimination of the airborne route of infection at the time of surgery. A summary of Charnley's data is shown in Table 3 to illustrate the steps by which he arrived at recommendations for a flow rate of 300 air changes per hour, entire ceiling distribution of filtered air to eliminate turbulence in the entire operating room, and air filters for removal of particles of 1-2 microns in size. Table 3. Ventilation for Control of Infection in Total Hip Replacementl Period Ventilation Filter Infection I Jan. 1959-Nov. 1961 80-90 0 exhaust 7.0 Nov. 1961-Jun. 1962 25 10 pos. press. precipiton 3.7 Jan. 1962-Mar. 1966 plenum +*** 2.2 peripheral turbulance Jun. 1966-Sep. 1967 0.0 300 unidirectional +*** 1.3 .7 1968-1970 0.0 300 unidirectional +*** apron, wound closure * c.p.h. - colonies per settling plate per hour ** a.c.h. -- air changes per hour *** filter efficiency 1-2 micron COSTS In a metered 6 x 10 ft downflow room with variable flow rates and 90 percent recirculation of air, we observed that power consumption and cost of operation increased in proportion to the ventilation rate. Costs doubled for each 40 feet of increased velocity as shown in Figure 4. Extracted from Charnley et al (Reference 11,12)
73 Lewis L. Coriell 70 60 â¢Â£ 40 S 30 20 0 . 10 20 30 40 50 60 FT/min. 70 80 90 Fig. 4. Costs per day in Kw-hrs, and in cents to operate a unidi- rectional downflow room when operated at different airflow rates. The effect of velocity of air movement on temperature buildup and noise levels is shown in Figure 5. The noise level increased 0.4 decibels on the A scale for each foot per minute of increased air velocity, or 20 dB-A for an increase of 50 ft/min. The temperature elevation was more rapid at low air velocities but over the range from 20 to 70 ft/min it averaged 0.2 deg C for each ft/min of increased air velocity, or 10 deg C for an increase of 50 ft/min. Cost of original installation for a low velocity (15-25 ft/min) vertical unidirectional soft-wall room is currently $75/ft2 of room area, and free-standing, hard-wall installations are about $100/ft2 (22). Recent quo- tations on high velocity installations are at least double the above figure.
74 Low Velocity Unidirectional Ultrafiltered Air Systems dB-A *F 75 104 70 â¢â¢ 88 84 80 89 88 84 80 48 74 10 20 30 40 50 FT/min. 60 70 80 90 Fig. 5. Decibels of noise on the A scale, and ambient room temperature without refrigeration in a unidirectional downflow room when operated at different airflow rates. Discussion The purpose of this review is to examine the merits of high velocity vs low velocity unidirectional airflow room ventilation in control of airborne particulates and microorganisms, not to justify the use of laminar flow ventila- tion in operating rooms. The impetus for this reexaminat ion of the merits of low velocity airflow springs from a desire to obtain optimal cost effectiveness. Decisions made in the early days of laminar flow research in medical environ- ments utilized flow rates of 100 feet per minute, and have through repetition ⢠and regulations, assumed more stature than they deserve. For example, a review of the literature shows the HEPA filter media are more efficient at lower flow rates and progressively less efficient at
75 Lewis L. Coriell higher flow rates. It also shows that achievement of unidirectional ventilation with the filtered air is essential to achieve maximal freedom from airborne particles. The mechanisms for achieving this are now well understood. The basic principles are: 1) entry of the filtered air across the whole ceiling at uniform pressure to prevent recycling through areas of lower pressure; 2) con- tainment of the air stream by walls until; 3) its escape at floor level through lateral openings sized to increase the escape velocity. When all these condi- tions are observed the performances at flow rates of 35 ft/min are not signi- ficantly different from those observed at 100 ft/min. Our own experimental studies on containment of particles, bacteria, and viruses in experimental rooms and operating rooms, and in prevention of airborne spread of infection in animal rooms, lead us to conclude that vertical flow rates of 25 to 35 ft/min and 150 to 200 air changes per hour will provide most of the benefits, and mini- mize the disadvantages that can be expected to accrue from application of HEPA- filtered unidirectional airflow room ventilation. REFERENCES 1. Ayliffe, G.A.J., B. J. Collins, E.J.L. Lowbury, J. R. Babb, and H. A. Lilly, Ward floors and other surfaces as reservoirs of hospital infection. J. Hyg. Camb. 65:515-36, 1967. 2. Beck, W. C., Control of airborne microbiological operating room contamina- tion. Guthrie Clinic Bulletin 35:126-34, 1966. 3. Bernard, H. R., W. R. Cole, and D. L. Cravens, Reduction of iatrogenic bac- terial contamination in operating rooms. Ann. Surg. 165:609-13, 1967. 4. Bernard, H. R., R. Speers, Jr., F. O'Grady, and R. A. Shooter, Reduction of dissemenation of skin bacteria by modification of operating room cloth- ing and by ultraviolet light. Lancet ii:458-61, 1965. 5. Blakemore, W. S., G. J. McGarrity, R. S. Thurer, H. W. Wallace, H. MacVaugh, and L. L. Coriell, Infection by airborne bacteria with cardiopulmonary by- pass. Surgery 70:830-37, 1971. 6- Bodey, G. P., Em. Freireich, and E. Frei, III., Studies of patients in a laminar airflow unit. Cancer 24:972-80, 1969. ?⢠Bodey, G. P., and D. Johnston, Microbial evaluation of protected environ- ments during patient occupancy. Applied Microbiol. 22:828-36, 1971. 8. Bourdillon, R. B., A. M. McFarlan, and J. C. Thomas, Airborne bacteria in operating theatres, studies in air hygiene. Spec. Report Service No. 262, Medical Research Council, Carshalton, Surrey, England pp. 241-253, 1948.
76 Low Velocity Unidirectional Ultrafiltered Air Systems 9. Bourdillon, R. B., and L. Colebrook, Air hygiene in dressing rooms for burns or major wounds. Lancet 1:561-65, April 27, 1946. 10. Burke, J. F., A bacteria controlled nursing unit and individual patient isolation facility. Proc. Int. Conference on Nosocomial Infections, CDC, pp. 220-24, Aug. 1970. Charnley, J., Postoperative infection after total-hip replacement with special reference to air contamination in the operating room. Clinical orthopaedics and related research No. 87:167-87, Sept. 1972. 12. Charnley, J., and N. Eftekhar, Postoperative infection in total prosthetic replacement arthroplasty of the hip joint. With special reference to the bacterial content of the air of the operating room. Brit. J. Surg. 56: 641-49, 1969. 13. Clean Room and Work Station Requirements, Controlled Environments, Federal Standard No. 209; General Services Administration, Business Service Center, Washington, B.C. December 1963. 14. Coriell, L., G. J. McGarrity, and J. Horneff, Medical applications of dust free rooms I. Elimination of airborne bacteria in a research laboratory. Amer. J. Public Health 57:1824-36, 1967. 15. Coriell, L., W. S. Blakemore, and G. J. McGarrity, Medical applications of dust free rooms II. Elimination of airborne bacteria from an operating theater. J. Amer. Med. Ass. 203:1038-46, 1968. 16. Cruse, P.J.E., Surgical wound sepsis. CMAJ, 102:251-58, 1970. 17. Decker, H. M., L. M. Buchanan, L. B. Hall, and R. R. Goddard, Air filtra- tion of microbial aerosols. Public Health Service Monograph No. 953, pp. 1-43, June 1962. 18. Dineen, P., and L. Drisin, Epidemics of wound infections associated with hair carriers. Lancet 24:1157-59, 1973. 19. Harsted, J. B., H. M. Decker, M. E. Filler, and C. R. Phillips, Evaluation of air filters with submicron viral aerosols and bacterial aerosols. Inter- agency Service Agreement MIPR 6.0037, Dept. of the Army, Ft. Detrick, June 1967. 20. Hart, D., R. W. Postlethwait, Ivan Brown, W. W. Smith, and P. A. Johnson, Postoperative wound infections: A further report on ultraviolet irradiation with comments on recent (1964) National Research Council Cooperative Study Report. Ann. Surg. 167:728-43, 1968. *
77 Lewis L. Coriell 21- Henderson, E. D., and S. S. Kornblum, Studies on the epidemiology of staphylococcal wound infections in previously clean surgical cases on an orthopaedic service. Instructional Course Lectures. American Acad- emy of Orthopaedic Surgeons 18:282-7, 1961. 22. Horneff, J., (personal communication). 23. Huszar, R. J., Air curtains for patient isolation. JAMA 207:549-51, 1969. 24. Lidwell, 0. M., and A. G. Towers, Protection from microbial contamination in a room ventilated by uni-directional airflow. J. Hyg. Camb. 67:95-106, 1969. 25. Lidwell, 0. M., Sepsis in surgical wounds. Multiple regression analysis applied to records of postoperative hospital sepsis. J. Hyg. Camb. 59:259, 1961. 26. Lowbury, E. J. L., Evaluation of patient isolators. Proc. Int. Conference on Nosocomial Infections. CDC, Aug. 1970. 27. McGarrity, G. J., and L. L. Coriell, Mass airflow cabinet for control of airborne infection of laboratory rodents. Appl. Microbiol. 26:167-172, 1974. 28. McGarrity, G. J., L. L. Coriell, R. W. Schaedler, R. J. Mandle, and A. E. Greene, Studies on airborne infection in an animal care laboratory. In Developments in Industrial Microbiology (C. Corum, ed.) American Inst. Biol. Sciences, Washington, D.C., 1970. 29. McGarrity, G. J., and L. Coriell, Procedures to reduce contamination of cell cultures. In Vitro 6:257-65, 1971. 30. McGarrity, G. J., L. Coriell, R. W. Schaedler, R.. J. Mandle, and A. E. Greene, Medical applications of dust free rooms III. Use in an animal laboratory. Appl. Microbiol. 18:142-2, 1969. 31. Medical Research Council. Aseptic methods in the operating suite. Lancet i:705-09, 764-68, 831-39, 1968. 32. Morris, P., and J. Burke, Surgical wound sepsis. Lancet 1:46, 1967. 33. Noble, W. C., 0. M. Lidwell, and D. Kingston, The size distribution of airborne particles carrying microorganisms. J. Hyg. Camb. 61:385-391 1963. 34. Pelos:', M. H. , Jr., The rationale for low velocity downflow clean rooms. Proc. 8th Tech. Meeting and Exhibit, Am. Assoc. Contam. Control, 6 Beacon St., Boston, Mass. P- 67, May 1969.
*.L 78 Low Velocity Unidirectional Ultrafiltered Air Systems 35. Penland, W. Z., Jr., and S. Perry, Portable laminar airflow isolator. Lancet 1:174-76, Jan. 24, 1970. 36. Phillips, G. Briggs, and Robert S. Runkle, Biomedical applications of laminar airflow. Chapter 2:25-28, CRC Press, Chemical Rubber Co., Cleveland, 1973. 37. Salbert, C., J. Matsen, D. Vesley, D. Wheeler, R. Good, and H. Meuwissen, Laminar airflow protection in bone marrow transplantation. Appl. Microâ biol. 21:209-16, 1971. 38. Scott, C. C., J. T. Sanderson, and T. D. Guthrie, Choice of ventilation system for operating theatres. Comparison of turbulent versus laminar/ linear flow systems in operating rooms and industrial clean rooms. Lancet 1:1288-91, June 19, 1971. 39. Speers, R., Jr., and R. A. Shooter, Shedding of bacteria to the air from contaminated towels in paper sacks. Lancet ii:301-2, 1967. 40. Technical Order Standards and Guidelines for the Design and Operation of Clean Room and Clean Work Stations. US Air Force. T.0. 00-25-203, Office of Technical Services, Dept. of Commerce, Washington, D.C., July 1963. 41. Walter, C. W., R. B. Kundsin, and M. M. Brubaker, The incidence of airborne wound infection during operation. J. Amer. Med. Ass. 186:908-13, 1963. 42. Wardle, M. D., Microbiological and particulate monitoring of a vertical down- flow clean air operating room. Document 1200-111, Jet Propulsion Laboratory, Pasadena, Calif. (JPL internal document), 1973. 43. Whitby, R. T., and D. A. Lundgren, Mechanics of air cleaning. Transcript of the Amer. Soc. Agricultural Engineering 8, No. 3:342-344, 351-352, 1965. 44. Whitcomb, J. G., and W. E. Clapper, Ultraclean operating room. Am. J. Surg. 112:681-685, 1966. 45. Whitfield, W. J., A new approach to clean room design. Sandia Corporation Report SC-4673 (RR) pp. 1-28, 1962. 46. Whitfield, W. J., The design of a dust-controlled vented hood utilizing laminar airflow. Sandia Corp., Publication SC-4905 (RR) pp. 1-20, 1963. 47. Whitfield, W. J., Principles of laminar airflow. 67th Annual Meeting Am. Soc. Microbiology, April 1967. Sandia Corp. Publication SC-R-67-1182, pp. 1-13, 1967. 48. Whyte, W., B. H. Shaw, and R. Barnes, A bacteriological evaluation of laminar-flow systems for orthopaedic surgery. J. Hyg. Camb. 71:559-564, 1973- 49. Whyte, W., B. H. Shaw, and M.A.R. Freeman, An evaluation of a partial walled laminar flow operating room. J. Hyg. Camb. 73:61-74, 1974.
79 THE ORTHOPEDIST'S SPECIAL INTEREST IN SURGICAL INFECTION Mark B. Coventry* A scant five and one-half years ago the first joint implant fixed with methyl methacrylate was authorized in America by the Food and Drug Admin- istration. After a year's trial period as an investigational new drug, it was released, the "bomb burst," and hundreds of thousands of hip, knee, finger joint, elbow, shoulder, and ankle implants have since been inserted. What has this to do with our symposium? While surgical infection is a fear which surgeons (and patients, too, if they are intelligent and informed) have constantly lived with in the past, the catastrophy to a patient's well-being produced by an infected major joint implant now rests solidly on the shoulders of virtually every practicing orthopedist. Why are we in orthopedics so specially interested and involved? Certainly a surgical infection in other fields can be, at times, even more de- vastating. Is it the seriousness of the infection--or the sheer quantity of po- tential problems--or the obviousness of a crippled patient which has absorbed the interest of the orthopedist? It is all of these, plus the nagging conscience of an honest surgeon that somehow he is responsible for this complication. The abdominal surgeon, the thoracic surgeon, the urologist--all work constantly in contaminated areas, and the potential for infection is always there. True, the neurosurgeon is usually operating in a sterile field, as is the cardiovascular surgeon. But the orthopedist almost uniquely violates a microbially sterile field when he makes his incision to replace a joint. Thus it is hard to "blame" the patient for a poor result--a rationalization that we are all guilty of at times. So we look to other sources of infecting organisms besides the patientâ the surgical team, the surgical instruments, the air of the surgery. And this is what we will be discussing at this workshop. Again, so why does the orthopedist have such a special interest in this field besides his peculiar feeling of having no one to blame but himself? His sur- gical infection rates are not higher than in other surgical specialities; indeed, they may be lower. At Mayo Clinic in 1973, orthopedic surgical infections totaled *Professor of Orthopedic Surgery, Mayo Medical School, Rochester, Minnesota.
80 Mark B. Coventry 1.33 percent, both deep and superficial, including trauma as well as elective surgery. But the devastating effects of sepsis in any one patient, and the actual numbers of infections occurring are now coupled with the relative newness of the challenge to determine the cause of the infection, and the means to alter these causes. Careful focus on the problem shows that "accepted rates" such as I have quoted simply aren't good enough today. Means must be found to lower these rates. John Charnley of Wrightington deserves credit for generating impetus in this direction. But all of us have taken up the challenge and have by many and diverse means sought to decrease the rate of infection. Research in immu- nology and stress response is ongoing. Practical measures have included skin preparation studies regarding the effectiveness of different lotions; methods and materials in draping the patient using specially woven cloth or paper im- pervious to microparticles; improvements in gowning the surgeon and masking his exposed skin, including such methods as exhaust systems; antibiotics, administered both generally and locally, for prophylaxis; and, very importantly, wound-handling techniques which avoid tissue trauma and pocketing. But above all is the study of the air which contacts the operative wound. This is the subject of intense exam- ination. In our own institution it was the orthopedist who said, "I am now involved, and I am going to do something about the problem." He has been ag- gressively pursuing the problem (some of his colleagues think too aggressively). Studies of particle collection in the operating rooms began. An entire review of the engineering of the operating theater was undertaken to determine airflow, volume patterns, filtration, etc. Better reporting and classification of sur- gical infections was insisted upon. Changes were made in operating room pro- cedure, including traffic control, draping and gowning. Other surgical teams in the hospital were often openly resistent. But slowly they have tended to follow. Other surgeons are becoming aware. All surgeons are now virtually forced to culture the drainage from their postoperative wounds in spite of the frequent comment: "Wliy culture? This is just drainage - stitch reaction. Drainage
81 Orthopedist's Interest in Surgical Infection doesn't mean infection." We orthopedists have now been joined by other involved surgeons, and we must continue to obtain every bit of important information and evaluate it. This quest for information that can be applied to the safety of our patients continues. Perhaps we will know relatively soon whether our filter systems are proper, whether high volume air exchange is important, and the de- tails as to its direction and its mode of administration. If we find these answers, the orthopedists of this country can take much of the credit. But, candidly, we must admit we were forced to take action by the magnitude of total- joint replacement, and the dire consequences of an infected implant to the pa- tient, the surgeon, and society. I cannot close without paying tribute to at least three of the great men who gave us the knowledge and methods which we continue to refine: first of all to Anton van Leeuwenhoek, without whose microscope organisms could not be seen; then to Louis Pasteur, without whose methods organisms could not be grown; and perhaps most of all, to Lord Joseph Lister, who applied the know- ledge gained up to his time in the surgery of the operating theater.
82 IMPROVED OPERATING ROOM BARRIER MATERIALS Peter Dineen* INTRODUCTION Over the past several years there has been a widespread utilization of disposable operating room materials which include gowns. These materials have two purported advantages. The first is that they reduce bacterial contamination of the environment, and the second is that they are more economical. While the latter consideration is open to debate, there does seem little doubt that the various barrier materials do reduce the number of organisms in the atmosphere. Naturally there has been a great deal of consideration as to whether the degree of microbial contamination is important. It is often stated that pa- tients have been operated upon for a good many decades with reasonably good re- sults by simply using sterile reusable cloth and muslin drapes and gowns. There has not been any pandemic of infections, and the question is reasonably raised as to why something new should be added at the present time. The answer lies in the type of patient upon whom surgery is being performed in the modern era. In recent years the number of individuals who are immunosuppressed, or who are having open cardiac or vascular surgery, or joint and/or organ replacement has significantly increased. Also, the patients upon whom we are operating are older and more de- bilitated in many instances. These are individuals who in years gone by were considered too risky. Therefore, the host is at a significantly higher risk than in the past. The possibility of contamination and serious disease caused by bac- teria which previously were considered nonpathogenic is high. The materials that have been studied by the Surgical Bacteriology Research Laboratory at this Center include cloth which was used for drapes and gowns, and nonwoven fibers (paper) materials. In the study of masks, both cloth and various fiber materials were studied. Examination of these drape and gown materials microscopically shows that the cloth is not closely woven and the in- terstices allow the passage of bacteria whereas the nonwoven material has no such interstices to allow bacteria through. This is particularly evident when water is placed on either of these two groups of materials. The cloth allows water or *Professor of Surgery, Cornell Medical Center, New York Hospita1, New York.
83 Peter Dineen any liquid to pass through quite rapidly, whereas the nonwoven materials are water resistant for varying periods of time up to a couple of hours. MATERIALS AND METHODS Two identical operating rooms were selected. These rooms were used for general surgical procedures. The same type of operation was done in both rooms and the personnel were essentially unchanged during the period of the study. In each room during sampling time there was an anesthesiologist, a surgical team of three individuals, a scrub nurse, a circulating nurse, and a technician. Simultaneous air sampling of these two operating rooms was conducted for one-hour periods during morning and afternoon surgical procedures. For two weeks, cloth drapes and gowns (muslin or cotton) were used exclusively in Room 1, and single use, disposable nonwoven drapes and gowns were used exclusively in Room 2. For the next two-week period the materials were reversed. Single use, disposable drapes and gowns were used exclusively in Room 1; cloth material in Room 2. This two-week reversal was continued for the two-month duration of the study. Air sampling was done with a Reynier air sampler with a rotating Petri dish, sampling at the rate of one cfm. This air sampler is constructed in such a way that the time of impingement of the organisms on the plate can be fixed with great accuracy. When the colonies grow out after incubation the plate represents the face of a one-hour clock. For example, the 15-minute point in time of sampling can be separated from the second 15-minute period and so forth. Bacteria which were isolated were randomly sampled and subcultured for diagnostic identification. RESULTS The number of organisms isolated from the environment were significantly lower when disposable, single-use nonwoven materials were used. As can be seen, there is some degree of variation in the amount of bacteria in the air, particu- larly related to activity. The middle periods seem to have the lower counts when there is less movement and traffic in the rooms. The majority of the cases in this study were abdominal procedures, many of which were finished by the end of the hour sampling time. A 90-percent reduction in airborne organisms was pro- vided by these disposable drapes and gowns compared to cloth. Average air
Improved Operating Room Barrier Materials sampling counts for the study were 23.5 for cloth and 2.2 for single-use dispos- ables. In a similar series of studies the data were collected again over a period of 59 days. On different days the assignment of disposable and nondisposable drapes was randomly assigned to the two rooms. No. Days Lower Counts with Disposable Drapes .................. 53 Lower Counts with Nondisposable Drapes ............... 4 Equal Counts Total ................ 59 The average number of organisms isolated from the environment in the last study: Disposable Drapes ...... Daily Average ... 15.05 Organisms Nondisposable Drapes . .Daily Average. .. .36.59 Organisms A third study measured the passage of bacteria through drapes during the operation. A steady increase with time was noted with muslin drapes, while in those operations with paper and disposable drapes the bacterial count re- mained constantly low for the length of study time. DISCUSSION From these studies it is evident that the use of nonwoven material significantly reduces the bacterial contamination of the environment. In these studies reported here only disposable paper drapes were studied. However, there are similar nonwoven materials on the market which are reported to be equally efficient. Recent meetings of the Sub-Committee on Aseptic Barriers of the Committee on Operating Room Environment of the American College of Surgeons has endorsed the use of nonwoven materials. It has been the experience at this Center that the reduction of the number of organisms in the environment has been consistently maintained by the use of nonwoven drapes and gowns. Systematic monitoring has continued during this It is still too early to have statistical significance on the reduction of wound infections by this means. Because of the number of variables and individuals in- £/ volved, it may never be possible to have a definitive answer to this question.
85 -ft. Peter Dineen MASKS An ongoing study of surgical masks in this laboratory has shown, as have many other investigators, that there are wide diversions in mask filtration effi- ciency. It is obviously desirable to have the most efficient mask which is also comfortable and functional. MATERIALS AND METHODS In brief, using a known inoculum aerosolized into a cloud chamber, and passed through an air sampler, it is possible to calculate the efficiency of any particular mask. RESULTS As can be seen from Table 1, there is a significant difference in the efficiency of the various masks. Mask A is a 16-ply cloth mask. The remainder of the masks are all various currently available disposable masks. Mask Table 1. Filtration Efficiency of New Masks No. of Experiments Percent Efficiency A 20 B 10 C 20 D 10 E 20 F 12 6 6 H 10 I 20 J 6 K 20 50.2 1-81 93.0 87-97 91.1 76-98 91.6 86-97 94.0 82-99 86.3 76-94 82.7 77-92 36.8 20-59 86.4 72-98 8.8 2-13 97.1 89-99 DISCUSSION As can be seen from this study, the number of organisms in the en- vironment will vary greatly depending on the type of mask utilized. Under the circumstances of present day technology, it seems unconscionable to use masks with a filtration efficiency of less than 93 to 94 percent. I
86 THE CASE FOR CLEAN AIR John A. Feagin, Jr.* In June of 1972 upon completion of a six-month Registrarship with Professor John Charnley, I had the privilege of discussing clean air with him under "relaxing" circumstances. Scientifically most of the data has now been published, but one phrase stuck with me far longer than the scientific proof. Charnley said, "Once one has enjoyed surgery under clean air conditions, aesthe- tics alone would bar him from ever practicing his craft in a conventional room." This has proved true and although it may be emotiona1, perhaps aesthetics is as rational an approach as statistics at this time. Nevertheless, being an insti- tutional practioner my clean air room required "justification." "Justification" in 197l-1972 was not easy; my institution required scientific proof and reason and historical perspective were suspect. We here, though, have been invited for decisions involving social responsibility. This is a thought process above mere "justification." For that reason, I would like to discuss operating room en- vironment from four perspectives: the Historical perspective, the Charnley perspective, the Letterman Army perspective, and the Responsible perspective. THE HISTORICAL PERSPECTIVE The historical perspective entwines the lives of two great men, Lord Lister and Pasteur, whose careers fortunately overlapped. One said, "Chance favors the prepared minds." The other was a prepared surgeon. The dramatic development of aseptic surgery occurred simultaneously with the rise of bacteriology. The role of Pasteur in this drama is unquestioned. Lister said that sepsis was the principal obstacle to any great advance in sur- gery. Lister discussed his problems in 1865 with Dr. Thomas Anderson, Professor of Chemistry, who called his attention to the work of Louis Pasteur. Pasteur in his well-known experiment demonstrated that an infusion will putrify when boiled in a flask which is left open. However, if the neck of the flask is drawn out to a fine point, the solution will remain pure because the air will drop its dust and germs at the opening of the neck. Here Lister found the *Assistant Chief, Orthopaedic Service, Letterman Army Medical Center Presidio of San Francisco, California
87 John A. Feagin, Jr. solution to his problem. It was not the air, but the germs in the air that produced suppuration. Volkman became Lister's most devoted disciple saying, "England may be proud that it was one of her sons whose name is bound up with the greatest advance of surgery." Lister died a Peer of the Realm honored by the world. It is my conviction that Pasteur and Lister lit the lamp and that Charnley's data are more sophisticated evidence that where surgical infection is concerned, airborne contamination must be considered. Lister and Pasteur marked the way. Now we will consider the Charnley perspective. THE CHARNLEY PERSPECTIVE All of us know John Charnley by name. Most people know him by reputation, and I am here to assure you that he ranks with the Listers and Pas- teurs. Charnley's data are additional sophisticated evidence that where sur- gical infection is concerned airborne contamination must be considered. His surgical unit, published statistics, and trained students bear witness to the efficacy of his beliefs. I do not want to belabor published statistics, but in a unit that is performing 2,000 hips per year and is wholly responsible for its own complications, each complication is critical to the unit. Otherwise, the Centre would be overburdened with the care of its complications. To those of you who have visited Wrightington the Charnley perspective speaks for itself. THE LETTERMAN PERSPECTIVE Letterman is one of five Army Medical Centers. It is relatively new having been constructed in 1969. It has seven operating rooms that accommodate almost every specialty. To justify a clean air room for our implant surgery, we began to monitor airborne contamination in search of the "proof" to justify a clean air enclosure. We embarked on as extensive an airborne monitoring program as we could support. How did it differ from previous studies?
88 The Case for Clean Air 1. We compared our orthopaedic cases in clean air and in a conventional room by sampling with the Matson-Garvin Slit Sampler and the Climet Particle Counter. 2. We monitored all seven operating rooms and were able to construct a profile by specialty of the various operating rooms. What does this prove? No more than any other study except that we now know the conditions to which we subject our patients. This is not only a matter of academic interest but of responsibility. To those of you who have monitored your surgical suite, I have no new data. To those who have not, I share with you our results in the hope that you too will insist on a monitoring program. The results of an airborne monitor- ing system may well lead you from aesthetics to absolute conviction. Unoccupied, our relatively modern operating suites have a generally low order of contamination. These values represent the average of a number of monitorings and as you may note our operating rooms are generally assigned by specialty (Fig. 1). 280 240 200 160 120 80 40 AVERAGE COLONY COUNT FIRST HOUR room 1 gen. surg. 2 ey« 3 4 orf ho. cleon neuro. 5 orfho. conv 6 gyn open heart Fig. 1 f.
89 John A. Feagin, Jr. The darker bar represents the average colony count during a number of monitorings through a variety of surgical cases. It is the quantum jump that I think is significant. The conventional operating room is unable to handle the pollutants of the operating team. Many people deny the value of the particle counter, but particles are the transport mechanism for airborne bacteria and after settling in the wound are a synergistic link toward clinical infection. They serve as adjuncts. Therefore, I feel the particle count is an important variable although it is in no way a parallel function of the colony count (Fig. 2) CONVENTIONAL CLEAN AIR I 10 15 20 25 " 96 c " ⢠48 o 32 16 \ -10 -5 1 5 minutes 10 IS 20 25 Fig. 2
90 The Case for Clean Air Another feature of our study was comparable cases performed under similar circumstances. To emphasize this we selected monisectomy (Fig. 3). Illliilllilliiii room ::;;;;*x::::xS;;;S:SrSyx:S: 1 I:...../:.: 32 ll$|iiiillxi^:^l 1 I COHVENTtONAL -⢠, ,"{" 1 4 aEAN AIR § 24 â¢:â¢'â¢'â¢'- J"-x:x::::::": ⢠S | 0 K%vi!o:â¢iAwrox ;.--..-.-:|-:-;.:.;.; :.;.;.;. o iliHiiSi â¢' !-â¢â¢3 ' 5 i* Pi 0 SiiHviO:X: â¢'.Xv â¢o-:i:S»:ivS:-J:':i: removal ; O .;. ;.;.;.â¢Â«â¢. â¢â¢'â¢'. vXvX i *.. [â¢'â¢>;â¢â¢â¢ ' " ' â¢â¢â¢â¢'.â¢'.â¢'.⢠X;X i e ' .|;i^xx": K5/ ' ' *. "â¢'â¢â¢â¢'".â¢â¢i->;.y â¢" j /* i» â ry*i i 1_ â .â¢.â¢: :⢠â¢â¢'â¢â¢ x^ _> . L. removal L 1 20 30 40 50 60 rninu Fig. 3 Predictably, contamination is greatest early in the case when draping and traffic have stirred up the particles and the coincident lint serves as an airborne vehicle. Our particle counter has verified that the areas of residual particles in the room are the floor, the people coming in, and the drapes. Vacuuming and washing between cases seems only to stir up a host of contaminates. S The most shocking aspect of room discipline is the role of our circulator. While we intently putter away, they attend the business of their own
91 J John A. Feagin, Jr. concern and the average number of door openings is greater than fifty per hour and is relatively constant from room to room, hour to hour, surgery to surgery, case to case; verifying they must have a routine of their own independent of the surgical team and quite inconsiderate. In summary, two years of monitoring at Letterman emphasize the contamination caused by the surgical team and draping. Equipment poorly handles these contaminates. THE RESPONSIBLE PERSPECTIVE John Charnley has not been able to scientifically prove the value of clean air to the American College of Surgeons. I doubt that any of us can pro- vide more convincing statistics. It is unreasonable and futile to demand a pleth- ora of statistics. Therefore, let us look to our own responsibilities to the best of our abilities based on the documentation that is at hand. Just what are our responsibilities? 1. To acknowledge airborne contamination. Historical perspective requires that we acknowledge airborne contamination. "Those who ignore the lessons of history are doomed to repeat them." 2. To define the risk of implant surgery and to acknowledge that we have created a new surgical milieu with new requirements. Therefore, let us not stumble on implant surgery, but use this as a rationale to a higher technological plateau. 3. To reappraise the statement of the American College of Surgeons. In December 1971 this was a laudable statement. We owe the College a reappraisal in 1974. 4. To take a stand regarding new operating room constructions. I personally cannot imagine a new operating room designed for implant surgery with- out including body exhaust, ultraviolet, and clean air as part of the standard equipment. This is not overkill; it is reasonable insurance. 5. To strive for a mental and physical goal of zero operating room contamination. Let us set our goal at no contamination and explain deviations
92 The Case for Clean Air from there, rather than accepting a seven percent infection rate as the starting point. To summarize, we have considered the operating room environment from the historical perspective, the Charnley contribution, from our own in-house monitoring, and as a social responsibility. It is my hope that this1 workshop will accept these challenges and take a strong stand for improved operating room environment.
93 AIRBORNE BACTERIAL CONTAMINATION IN THE CONVENTIONAL OPERATING ROOM Robert H. Fitzgerald, Jr.* Airborne bacteria in a variety of conventional operating rooms have been implicated in postoperative wound sepsis (1,3,4,5). The variation in the design of the operating rooms studied and the wide range of data have clouded the significance of airborne bacterial contamination. The divergence of design and capability of "modern" operating rooms in which orthopaedic surgery is per- formed at the Mayo Clinic stressed the need for additional information on the bacterial environment in the conventional operating room. The bacterial environment of four operating rooms located in the two hospitals used by orthopaedic surgeons of this clinic was evaluated. The level of airborne bacterial contamination was determined with standard techniques em- ploying the Andersen and Case Ha samplers located within a few feet of the oper- ating table. All four operating rooms were sampled on 8 to 16 separate occasions during total-hip arthroplasty--a procedure in which there has been strict adher- ence to a protocol. The rooms chosen for evaluation varied in air-handling capabilities and design. In two of the rooms, with 12 to 14 exchanges of room air per hour, air passed through a roughing and an electronic precipitating fil- ter (high intermediate efficiency filtration) prior to entering the operating room. All air was freshly filtered without an admixture of recirculated air. Both were located in an operating suite with less than ideal traffic control. The two remaining operating rooms with 28 to 32 exchanges of room air per hour were located in an isolated operating suite with strict traffic control. The air delivered to these rooms passed through roughing and bag filters (high in- termediate efficiency filtration) and was recirculated to a variable degree, de- pending on atmospheric conditions. All four rooms studied had a positive pressure relationship with the adjacent corridors, work rooms, and scrub rooms. The mean level of airborne bacterial flora was 0.31 viable particles per cubic foot of air in all four rooms when they were not in use. During sur- gical procedures, the mean levels of contamination were lower in the room with Instructor in Orthopaedics, Mayo Medical School, Mayo Clinic, Rochester, Minnesota
94 Robert H. Fitzgerald, Jr. high rates of room air exchange per hour (Table 1). However, even rooms with high rates of room air exchange were noted to have elevated levels of airborne bacteria. Recirculation of air did not appear to influence the level of con- tamination. Table 1. Airborne Bacterial Contamination of Operating Room Mean viable particles/cu ft of air Room air exchange /hour With room empty With room in use Range 12-14 0.36 6.5 2.9-9.1 0.4 . 7.0 4.0-9.9 28-32 0.3 5.7 3.2-9.5 0.2 5.3 2.9-8.8 The air-sampling data obtained using the Andersen sampler on three occasions demonstrated that the majority of the microorganisms were isolated on the first two stages (Table 2). Table 2. Level of Airborne Bacterial Contamination Using Andersen Sampler Stage and particle size ( um) collected Colony- forming units/stage, Andersen sampler 1 (9.2-larger) 2 (5.5-9.2) 3 (3.3-5.5) 4 (2.0-3.3) 5 (1.0-2.0) 6 (0 -1.0) Average viable particle/cu ft of air per minute during procedure 1.18 0.67 0.38 0.79 0.01 0.02 3.05 1.58 0.82 0.37 0.59 0.10 0.09 3.56 1.82 1.24 0.85 0.87 0.89 0.04 3.57
95 Contamination in the Operating Room Thus, the majority of the microbial particles were 5 ym and larger. Since the air-filtration system removes 97.5 percent of particles of this magnitude, the microorganisms recovered must have been shed by personnel in the room. High levels of airborne bacterial contamination were noted with increased numbers of personnel working in the operating room (Fig. 1). This re- lationship was more readily apparent in operating rooms with lower rates of room room air exchange per hour. Restriction of the number of personnel in the operating room to only those required for care^of patients, even in teaching in- stitutions, is necessary. 10 8 \ 0 0 2 4 6 8 10 12 14 16 Number of personnel in operating room Fig. 1. Relationship of airborne bacterial con- tamination to increased numbers of personnel working in operating room. (Room air exchanges per hour: open circles, 12 to 14/hour; closed circles, 28 to 32/ hour.) Chronologic recording of the level of airborne bacterial contamination during total-hip arthroplasty demonstrated a rather consistent pattern (Fig. 2). Elevated levels of bacterial contamination were noted during those periods when physical activity was prominent, with lower levels in the intervals between these periods. Thus, elevated levels of contamination existed with the arrival of the patient in the operating room and with the increased activity of the anesthesia team.
96 Robert H. Fitzgerald, Jr. 90 120 Time in minutes Fig. 2. Chronologic recording of airborne bacterial contamination during 25 total-hip arthroplasties. The level fell to 4 to 7 viable particles per cubic foot of air per minute at the time the incision was made. Similar increases were noted when the trochanter was osteotomized, the acetabulum was widened and deepened, the femoral canal was prepared, and the trochanter was reattached. The level of airborne bacteria at each of these intervals was higher in rooms with lower rates of room air exchange per hour (Figs. 3 and 4). The mean level of airborne bacterial contamination during the entire procedure was higher (6.2 versus 5.4 viable particles per cubic foot of air per minute) in rooms with low rates of room air exchange per hour. The level of bacterial contamination noted on settling plates on back tables supports the case method of preparation of instruments (Table 3). The difference between rooms of high and low rates of room air exchange per hour was obvious.
97 Contamination in the Operating Room Incision Trochonteric osteotomy Acetobular preparation Femural canal preparation Trochonteric reattachment Mean entire procedure 6 12 16 20 Viable particles / ft3 air 24 Fig. 3. Chronologic recording of mean and range of airborne bacterial contamina- tion during 13 total-hip arthroplasties in operating room with 12 to 14 room air ex- changes per hour. Incision . . Trochonteric osteotomy 'l 1 Acetobulor preparation 1 . ! I ⢠I Femurol canal preparation 1 1 Trochanteric reattachment Mean entire procedure I 1 - 1 +' I i i i i l I 1 i l i i 8 12 16 20 Viable particles / Hs air 24 Fig. 4. Chronologic recording of mean and range of airborne bacterial contamina- tion during 12 total-hip arthroplasties in operating room with 28 to 32 room air ex- changes per hour.
98 Robert H. Fitzgerald, Jr. Table 3. Bacterial Contamination on Back Table Mean no. of colony- forming units/sq ft per hour Room air exchange /hour No. of observation periods 12-14 28-32 121 48 31 8 The level of contamination of the instrument table observed on settling plates was elevated in both types of operating rooms (Table 4). Higher levels of contamination on settling plates placed near the orifice of the wound were noted in rooms with low rates of room air exchange (Table 4). The level of contamination at the orifice of the wound appeared to be related to the difficul- ty of the procedure. High microbial contamination was consistently noted when failed endoprosthetic or cup arthroplasties were converted to total-hip arthro- plasties. Such procedures require greater activity with associated bacterial shedding by the surgical team. The use of an impermeable hood and large mask to cover the entire face except for the area about the eyes by the entire surgical team resulted in a significant decrease of the direct bacterial shedding by the scrub nurse onto the instruments and by the surgeon and his assistants about the orifice of the wound (Table 5). The high levels of airborne bacterial contamination found in the cast rooms and corridors of the operating suite necessitate positive pressure within the operating room (Tables 6 and 7). Periodic checks of the pressure relation- ships and smoke studies to examine their flow patterns are necessary to ensure maintenance of these parameters.
99 Contamination in the Operating Room * T3 c o M-l O 0) o â¢g OJ 1-1 XI CO I- . 01 00 c c o B CO 4J O u I-l 01 4J C J CO CO XI 1 O oo _c iii p* **^ IâI O i-l I-l e o> ⢠^ a o o C M-l 4J ON 00 â¢a 1 M-l 0 -H c C >â¢. 1-1 3 03 c cr O 0) O 00 * O 00 M-l U 4J O â¢H C Ol 3 0 1-1 IM M o C O i-l 00 M-l 4J T3 O oj O vO t*^ ⢠(-1 l-i 1-1 O 01 CU Z en a. O 3 O M-l C O i-l l-l B ai o o C M-l 4-1 1 M-l O co râ 1 C ^* ^* x> co c cr CO Ol O CO H S 1"* ^* O CO 4-) 0 4-1 c 1-1 a> C I-l 3 c M C O â¢H 00 O 03 O > -H co r^ O CU Ol Z to CX XI O 1 M x: ,_j .x^ ^3 CO CO Ol 00 i i g c CN 00 O CO i-l CXI o x: Pd CJ X Ol co H . to C â¢H 6 â¢a co c 4J 3 c o o ^ o â¢rl CU M U <U -H 4-1 M-l U -â¢-I CO U PQ O C T3 O S CO â¢a <u o 1-1 O Xi S3 co H M-l O 4J C 4J <U CJ g ai 3 M-l M M-l 4-1 W CO O1 M H in OJ râI XI 3 H 3 O M-l 00 X! O C ⢠B 0) O M CU c o S3 â¢o M-l 4-1 C C 1 M-l 1- i 3 CO >> O 0) C ET S 0 co M-l O CO O C J 4J Ol C o 3 â¢H i-l O C O â¢H 00 M-l 4-1 *O O CO O ^ i-* O CO O Ol Ol Iâ 1 1â 1 Z eo D-, jQ O h 3 O M-l 00 X! O C i-l I-l ⢠3 CU O )-i CX a o Ol M-l 4J m i-i 1- 1 C 1 MH vO -^- XI cO >. ca o) c cr H £ O CO iâi ^*w 4-1 O CO C CJ 4-1 0) 1-1 B £3 3 3 M 1 1 ca C 1-1 C O M-l -rl CO O 4J 73 CO O ⢠> i-l 1-1 <y\ O l-i M 1-1 Z Oi Oi ca Ou XI o 4-1 09 3 T3 o x: O ft 4-1 O 4-1 1-1 EC â¢H J5
100 Robert H. Fitzgerald, Jr. Table 6. Airborne Bacterial Flora in Corridor Activity level Viable particles/cu ft of air Mild Moderate Max imum 16.3 20.3 24.9 Table 7. Airborne Bacterial Flora in Cast Room Room air exchange/hour Viable particles/cu ft of air Cast removal Room empty 12-14 9.8 28-32 11.8 6.0 6.0 Previously published studies of total-hip arthroplasty with wound cultures at this institution have shown contamination of the operating wound in up to 33 percent of the procedures (2). Differentiation of endogenous contami- nation from exposed skin surfaces and airborne seeding of the operating wound is not entirely possible. When serial cultures of the operative wound were care- fully performed in aerobic and anaerobic fashion during 157 total-hip arthro- plasties, rooms with high and low room air exchange were noted to have roughly the same rates of bacterial colonization of the operating wound (Table 8). However, anaerobic organisms were isolated in almost 50 percent of the wounds with positive cultures (Table 9). These anaerobic organisms probably are not airborne contaminants but most likely represent endogenous contaminants from exposed edg^s of skin. Bacterial colonization in an additional 28 percent of the wounds by Staphylococcus epidermidis could represent either endogenous or exoge- nous contamination. Differentiation is not possible at the time of this writing.
101 I Contamination in the Operating Room Table 8. Positive Wound Cultures* in 157 Total-Hip Arthroplasties Room air Positive cultures exchange/hour No. of procedures Number ) Percent 12-14 42 17 40 28-32 115 39 34 *Tissue culture. Table 9. Organisms Recovered From 57 Positive Wound Cultures*f Propioaibacterium acnes 24 16 11 6 2 2 2 2 1 1 Staphylococcus epidermidis Streptococcus, viridans group Corynebacterium Staphylococcus aureus Peptococcus Neisseria Bacillus Eubacterium quintum Lactobacillus *More than one organism recovered from some positive cultures. â¢fTissue cultures.
102 Robert H. Fitzgerald, Jr. Bacterial colonization of the operative wound has been associated with an increased incidence of deep wound sepsis; however, the organisms recover- ed at the time of sepsis may differ from those recovered during surgical proce- dures (2). Although higher levels of airborne bacterial contamination and operative wound cultures were noted in rooms of low room air exchange, there was no statistical difference in the rate of deep sepsis after total-hip arthroplasty (Table 10). Table 10. Deep Sepsis After Total-Hip Arthroplasty (March 1969 Through February 1972) Re >om air exchanges/hour 15 J-14 28-32 No. of procedures 1,853 1,360 No. of patients with 25 10 deep sepsis % Patients with deep 1.3 0. 7 sepsis » CONCLUSIONS The personnel working in the operating room and the patient are the sources of airborne bacterial contamination within the operating room. The level of airborne bacterial contamination in the conventiona1, modern operating room is variable, depending on the air handling capabilities in the particular operating room. In genera1, rooms with low rates of room air exchange per hour have higher levels of airborne bacterial contamination. The level of airborne bacterial con- tamination in operating rooms with high rates of room air exchange per hour can be elevated if traffic control is not enforced and if increased numbers of per- sonnel are in the operating room during surgical procedures. To maintain a low
103 Contamination in the Operating Room level of airborne bacterial contamination during an operating procedure in rooms with low rates of room air exchange, strict adherence to traffic con- trol and limitation of the number of personnel in the operating room are necessary. A direct relationship between airborne bacterial contamination and wound colonization has not been established. However, bacterial coloni- zation of the operative wound is associated with an increased incidence of deep sepsis after total-hip arthroplasty. REFERENCES 1. Charnley, J., and N. Eftekhar, Postoperative infection in total prosthetic replacement arthroplasty of the hip joint: With special references to the bacterial content of the air of the operating room. Brit. J. Sure 56*64l- 649, 1969. 2. Fitzgerald, R. H., Jr., L.F.A. Peterson, J. A. Washington, II, R. E. Van Scoy, and M. B. Coventry, Bacterial colonization of wounds and sepsis in total hip arthroplasty. J. Bone Joint Surg. 55A:1242-1250, 1973. 3. Goldner, J. L. and B. L. Allen, Jr. Ultraviolet light in orthopaedic operating rooms at Duke University: Thirty-five years' experience, 1937-1973. Clin. Orthop. 96:195-205, 1973. 4. Hart, D., Sterilization of the air in the operating room by special bacteri- cidal radiant energy. J. Thorac. Surg. 6:45-81, 1936. 5. National Research Council Study, Postoperative wound infections: the influ- ence of ultraviolet irradiation of the operating room and of various other factors. Ann. Surg. Suppl. 160:l-192, 1964.
104 ULTRAVIOLET LIGHT IN THE ORTHOPAEDIC OPERATING ROOMS AT DUKE UNIVERSITYâ 37 YEARS' EXPERIENCE, 1937-74 J. Leonard Goldner* Robert W. Gaines* Mary Higgins* The operating room environment at Duke Hospital in 1930 was patterned after standard operating rooms throughout the United States. During the winter months, the incidence of operating room infections caused by hemo- lytic Staphylococcus aureus was approximately ten percent. Deryl Hart attempted many changes in order to decrease the incidence of infection (13). Hart pub- lished an article in 1938 giving the efforts that had been made to reduce the number of bacteria in the operating room air (14). The procedures attempted were: 1. Rigid isolation with no visitors and a minimum number of personnel in the operating room during any procedure. 2. Large heavy masks covered the nose and mouth, both during and between operations. 3. All carriers of Staphylococcus aureus in the nose and throat were excluded. Serious operative procedures were postponed when contamination of the air was found to be high. 4. Meticulous cleanliness was maintained by frequent washing and painting, and by eliminating or decreasing air currents from other parts of the hospital. 5. Large amounts of air were taken from outside the building, where there were few, if any, pathogenic bacteria, filtered and washed and the air conditioning system then used to dilute the air in the operating rooms. 6. Various types of helmets and suction devices were used in an effort to decrease the number of pathogenic bacteria. In spite of these precautions, contamination of the air with pathogenic bacteria rose as the number of carriers during and following epi- demics of infection of the respiratory tract occurred (17). * Duke University Medical Center, Durham, North Carolina.
105 J. Leonard Goldner Robert W. Gaines Mary Higgins Not until ultraviolet (UV) light was introduced to the operating rooms at Duke did the incidence of infection decrease to a level that was acceptable to the surgeons working at Duke in the late 1930s and acceptable when compared with statistics quoted throughout the country at that time. THE INITIAL EXPERIENCE WITH ULTRAVIOLET LIGHT AT DUKE Ultraviolet light eliminates pathogenic bacteria, just as does sunshine. Air particles carrying bacteria could be sterilized by ultraviolet light (13). Experiments demonstrated that Petri dishes of blood agar, when sprayed with hemolytic Staphylococcus aureus and exposed at the operative site, would be sterilized within one to three minutes; and at this point, the number of bacteria dropping out of the air could be reduced by 95 percent to 100 per- cent. The intensity of the radiation will not give the patient an appreciable skin erythema during an exposure of 90 minutes; and the surgeon wearing minimal protection of a paper hood, plastic or plain glasses, a standard green eyeshade and ordinary gown may work for hours in this environment without the eyes or the skin being affected. In 1937, Hart demonstrated that contamination of operating room air was universal. Fifty surgeons in 25 states were asked to make cultures in their operating rooms in order to determine the extent and degree of air contamination. Reports received from 32 surgeons covering 37 operating rooms in 33 hospitals located in 17 states showed Staphylococcus albas on all plates, and over 50 percent of the plates showed Staphylococcus aureus. Hart emphasized during the same period that exclusive of bactericidal irradiation, air conditioning is the most important single means of reducing bacterial count of the air in the occupied operating room (22). Organisms given off in an operating room by the occupants are predominantly gram-positive cocci and gram-negative bacilli, and these are re- moved rapidly by clean circulating air from an efficient ventilating system (19).
106 Ultraviolet Light at Duke All things being equal, the number of pathogenic bacteria in the air of a room increases directly with the number of occupants and the duration of occupancy. Conversely, the bacterial count drops directly in proportion to the length of time the room is unoccupied and closed. The air in an operating room that contains ordinary air conditioning and no special environmental antibacterial devices, and which has been closed between 15 to 20 hours, is almost free of bacteria. A room with many occupants, however, has a much higher bacterial count than an adjoining room with a few occupants. Hart and his co-workers found that four occupants suffering from colds in an operating room resulted in Petri dishes of blood agar exposed for one hour to show an average of 13.8 colonies of streptococci per Petri dish (17,18,19). The total number of colonies of bacteria sedimenting from the air n gives some idea of the degree of contamination of the air, but is less signifi- cant than the number of pathogenic organisms present. In this respect, the air conditioning, by removing most of the nonpathogenic organisms from the incoming air and only diluting the concentrations of the pathogens in the room, gives more apparent than actual improvement in operating room conditions. In recent years, however, the nonpathogens of former times, in other words, Staphylococcus epidermidis. and certain gram-negative cocci and bacilli, are now considered pathogens because of metallic and plastic implants that are being used as sub- stitutes for all or part of an organ system. Hip joints, knee joints, metacar- pal and phalangeal joints, and heart valves include such foreign materials as high density polyethylene, methyl methacrylate, metallic alloys, silicone, and densely woven Dacron, all of which provide mechanical crevices and cracks and prevent the components of the natural immune system and antibiotics from getting to the bacteria. Airborne bacteria thus become an even more serious problem since both pathogens and nonpathogens may affect the mechanical stability and the durability £ of an implant. The combination of ultraviolet light and a superior air condi- tioning system in 1974 has maintained an infection rate for the Orthopaedic
107 J. Leonard Goldner Robert W. Gaines Mary Higgins Service at Duke of less that 0.5 percent. When the ultraviolet lights were Introduced initially, the air conditioning system was difficult to maintain, the filters were not as efficient, and the air changes per hour per room were less frequent than during the past ten years. In spite of the inconsistent airflow and the difficulty in controlling room temperature and the contamination that resulted when individuals with res- piratory infections were in the operating room, once ultraviolet lights were introduced to the operating rooms at Duke (1934), the high infection incidence decreased to a level that was acceptable (14). The percentage of infections has continued to remain at the same low level from that time to the present. The Orthopaedic Service at Duke has utilized ultraviolet light since 1937 for all major surgical cases. The record of wound infection in refined, clean wounds (first operation in a particular region on a particular patient) has been maintained at a very low level. The incidence of orthopaedic wound in- fections since 1940, reviewed on an annual basis, has been less than 0.5 per- cent for all operative cases done. This percentage has not changed even though the staff has increased from 1 full-time to 7 full-time orthopaedic surgeons, from 4 orthopaedic residents to 35 different residents (not all at the parent hospital simultaneously), and from 100 orthopaedic operations per year average to approximately 3,000 orthopaedic procedures each year. CURRENT STUDIES OF OPERATING ROOM ENVIRONMENT AT DUKE Many questions, especially those concerning the relative benefit of ultraviolet radiation versus newer techniques, such as unidirectional airflow in reducing surgical infections, cannot be answered unequivocally. However, one cannot refute the hypothesis that horizontal or vertical unidirectional flow with High Efficiency Particulate Air (HEPA) filters in the system provides the "cleanest" possible air with reference to particles. The point of maximum patient protection by air cleanliness has neither been determined nor set in perspective with contact contamination and endogenous sources of bacteria (8). While the sterilization of exposed surfaces, the germicidal effect on exposed skin, and the continuous radiation of wet areas on drapes and the surgeons'
108 Ultraviolet Light at Duke clothing are benefits of ultraviolet light, the information presented relates to the incidence of operating room infections for the Orthopaedic Service at Duke and the data relative to contamination in the air, both recently and during the past 35 years (10). Total-hip arthroplasty, other heart and joint implants, and organ transplantation have introduced variables into the criteria for operating room cleanliness and stimulated review of techniques, operating room air, and inci- dence of operating room infection. In many hospitals, and in individual operat- ing rooms supervised by particular services, conditions that had been satis- factory for prior major surgery have proven unsatisfactory for these procedures. The relative value of "sterility of the air" must be defined. We believe that clean air decreases the incidence of operating room infection; however, we continue to ask ourselves such questions as: 1. How important is the reduction of atmospheric bacteria from one to three colony-forming units per cubic foot of air to essentially sterile air? ⢠2. What is the relative importance of contact contamination which may involve thousands of bacteria? 3. How important are endogenous bacteria which can never be completely eliminated from the patient's skin or certain foci within the body? 4. How important is the waning immunologic competency of the patient population undergoing the procedure (11)? Furthermore, how many bacteria per gram of tissue must be in a particular kind of wound in certain patients in order to cause a clinically evident wound infection? And, how important is the presence of a prosthesis in living tissue, when considering the effect of small numbers of nonpatho- genic bacteria inoculated in an area where a prosthesis is present, in causing a clinical infection? These questions under consideration and the data pre- sented by many investigators should provide the answers soon, if not at the present time. Ultraviolet light will provide clean air and kill the bacteria on exposed surfaces.
109 J. Leonard Goldner Robert W. Gaines Mary Higgins DUKE OPERATING ROOM AIR CONDITIONING SYSTEM At Duke, the air conditioning system is superior and efficient and even without ultraviolet light, the air is relatively clean, the air exchange averages 35 times per hour, and the particle count is relatively low. Under ordinary circumstances, i.e., limited traffic, minimal turbulence, and a re- fined, clean case without endogenous elements, the incidence of infection on any service is low. A safety factor should exist to protect the patient against a pathologic shedder, a respiratory infection in one of the operating room personnel, or air contamination from some other source. Ultraviolet light will do this in addition to the protection provided by the air conditioning system. In the Duke operating rooms, modifications of the air conditioning systems or additions of new rooms with new systems have been carried out since 1957. In 1969 when total-hip arthroplasty was initiated at Duke, the proce- dures were done under System #4 (Operating Room Suites 16, 17, 18, and 19). The air handling unit is located in the attic of the Davison Building. The fresh air intake and exhaust are in the roof above the air handling unit. Incoming air passes through the fan, medium efficiency filter, a preheat coil, a heat wheel, high efficiency filter, an air washer, and a cooling coil. Next the air enters the individual ducts leading to the operat- ing rooms where it is reheated and humidified before entering the rooms. The exhaust duct has a high efficiency filter just before the heat wheel. The heat wheel is designed to transfer heat from the exhaust air into the supply air for economy. This is the only point of contact between the exhaust air and the supply air, and this is the reason a high efficiency filter is in the exhaust duct. There is no recirculation of air. These rooms are equipped with ultraviolet lights. Many of the first 200 total hips done at Duke were done in this suite. This information is given only to emphasize the importance of investigators knowing as much as possible about the basic airflow design of the room, the number of changes per hour, the pressure readings, the humidity, the temperature, and other details related to flow of particulate matter.
110 Ultraviolet Light at Duke System #2 at Duke, serving Operating Rooms 3 through 10, consists of an air handling unit located on the 5th floor of the 1957 addition. The fresh air intake and exhaust are in the east wall of the mechanical room of the observation suite. The fresh air passes through medium efficiency filters, high efficiency filters, preheat coi1, an air washer, cooling coils, a reheat coi1, and then the fan. After leaving the fan, the air enters individual ducts leading to Rooms 3 through 10 and passes through a final reheat coil. There are no provisions for the air to be recirculated. System #1 serves Operating Rooms 1 and 2, and includes an air handling unit located on the roof of the 1957 addition which is the location of the fresh air intake and exhaust. The fresh air passes through medium efficiency prefilters, preheating coils, an air washer, and cooling coils. It then enters the fan, passes through high efficiency filters, is reheated, humidified, passes through volume control dampers, and then enters the operat- ing room. This system normally uses 100 percent fresh air, but it is designed so that the air can be recirculated. System #3 serves Operating Rooms A, B, C, and D, and includes an air handling unit located in the hospital attic over the supply rooms for the oper- ating suite. The fresh air passes through medium efficiency filters, high ef- feciency filters, preheating coils, an air washer, and cooling coils. The air then enters the fan, passes through a reheat coil and a humidifier, then enters the operating rooms. There is no provision for recirculation of air in this system. The high efficiency filters used are Farr Model HP-200 which remove 95 percent of particles in the five micron range. Continental Models 9C24, 9C12M, and 9C24M, remove 95 percent of particles in the one micron range, or 92 percent in the 0.3 micron range. The medium efficiency filters are Farr Model RP-2A, removing 36.5 percent of particles in the five micron range, and Continental Models CF-445, CF-248, and CF-448, removing 45 percent of particles in the 1.8 micron range. The parti-.le efficiency ratings are based on the NBS tests using atmospheric A, dust. There are 36 air changes per hour in Operating Room 3, and 34 air changes per hour in Operating Room 4.
â¢*?~ Ill J. Leonard Goldner Robert W. Gaines Mary Higgins TOTAL-HIP ARTHROPLASTY AT DUKE Total-hip arthroplasty utilizing methyl mathacrylate was initiated on the Duke Orthopaedic Service in 1969. This operative procedure was carried out in Great Britain, France, Canada, Australia, and New Zealand in large num- bers prior to that time. Initially, the infection rate in certain centers was high (7 percent to 14 percent) (5). With alterations in operating room environ- ment, these centers decreased their infection rate to an acceptable level. At Duke, we initiated the procedure following the same draping and prepping tech- niques as we had done for other hips, and followed essentially the same program as had been followed for any major orthopaedic procedure. The air conditioning system was as before, and the ultraviolet lights were used consistently. During the time that the first 100 total hips were being done (about one year), other modifications in techniques were added to avoid "taking unnecessary chances." These included: ⢠the use of protective booties on the shoes; ⢠use of a throw-away hood to decrease shedding from hair and neck; ⢠occasional use of wrap-around disposable paper gowns; ⢠limitation of traffic; and ⢠initiation of a clean corridor which eliminated all unnecessary traffic. The open "all-day-long" back table technique which had been used for over 20 years was discontinued even though prior bacteriologic studies had shown the UV lights to be effective in maintaining sterility. To the present time, there have been eight infections evidenced by wound suppuration in about 700 total-hip patients. These problems occurred during the first 300 hip operations. No major infections occurred in the subsequent 400 hip procedures. These operations have been done by both senior staff and resident staff, and the operating room techniques and the environ- ment have not been altered except in the way that has been mentioned. The use of prophylactic antibiotics was not consistent, and this aspect of the proce- dure will be discussed later. q
112 Ultraviolet Light at Duke Hips that have been operated upon previously are aspirated and the material cultured prior to operation; swab culture, as well as tissue for cul- ture, is taken at the time of the procedure although the latter has usually occurred after the initial dose of antibiotic has been given. The patients are screened for endogenous sources of bacteria such as urine, respiratory tract, teeth, and historically at least for such conditions as chronic gall- bladder disease, ulcerative colitis, or thrombophlebitis. If the history or findings are positive the operative procedure is delayed. Certain members of the orthopaedic staff have had no operative infections in doing total-hip replacements; whereas, other members who may have been involved in a greater number of procedures, or procedures of a more difficult nature, may have had a higher incidence of infection. Many factors must be considered in assessing the average figure. The overall incidence of sepsis in all total-hip arthroplasties utilizing methyl methacrylate on the Duke Orthopaedic Service and including all patients, whether they have been operated upon previously or not, has been 1.1 percent. These infections oc- curred during the time that the first 300 patients were operated upon. There have been no infections during the last 400 total-hip operations. This em- phasizes the importance of determining statistics on a large number of patients that may include change of seasons as well as change in personnel. We empha- size again that the patients included in these statistics are no longer those which Hart defined as refined, clean cases, but now include patients who have had prior operative procedures including endoprostheses. PROPHYLACTIC ANTIBIOTICS ON THE DUKE ORTHOPAEDIC SERVICE The necessity or the value of prophylactic antibiotics for orthopaedic patients at Duke has not yet been ascertained completely. The infection rate for all major orthopaedic operations during the past 30 years has been 0.5 per- cent or less, reviewed on an annual basis. This excludes the period of time when the total-hip study was initiated. Antibiotics had been used haphazardly by members of the Duke staff. In certain instances they were started preopera- tively. A double-blind control study was initiated in 1969. In this study were included patients chosen at random, all of whom had a major orthopaedic operation such as spine fusion, joint replacement, but excluding all hand cases,
113 J. Leonard Goldner Robert W. Gaines Mary Higgins most foot operations, and most knee operations. Over 200 of the patients in- cluded in the double-blind study were patients undergoing a total-hip arthro- plasty. All of the patients were operated upon under ultraviolet light. There was no significant difference in the infection rate in those patients receiving antibiotics compared with those who did not; all were operated upon under UV light. We can see that in 'view of the prior low infection rate which we believe related to use of UV light, the number of patients involved in a double-blind antibiotic study from which significant convincing figures would evolve may be over 1,000.) Our current practice is to use a prophylactic antibiotic regime in all patients receiving an implant in view of the possibility of undetected en- dogenous sources of bacterial contamination. The first dose is given the night before operation, continued during the procedure, and discontinued 48 hours postoperatively. The side effects have been minimal although some have been recognized occasionally. The alteration of natural occurring bacteria and the alteration of bacterial flora within the hospital and within the individual patient have been noted. Further, the resistance of certain gram-negative and gram-positive bacteria related to the widespread use of antibiotics cannot be ignored. Gastrointestinal alterations, delayed sensitivities, and yeast and fungal infections are all potential dangers of antibiotics. ANALYSIS OF TOTAL-HIP INFECTIONS SINCE 1969 There have been eight deep wound infections in a total of approximately 700 patients who have had total-hip arthroplasty. Three of these infections occurred in previously unoperated hips and would be in the category of refined, clean cases. In one instance, a Clostridium perfringens organism was isolated without evidence of gas gangrene. This developed in the subcutaneous tissues approximately three weeks after the operative procedure. The methacrylate cement cultured at the time of the initial operation was also positive for this organism. Eventually the prosthesis had to be removed. A second patient de- veloped a deep infection eight months after total-hip replacement even though the immediate postoperative course and the subsequent months showed no unusual signs or symptoms. This patient had a urinary tract infection at the time she re-entered the hospital with hip pain, and Escherichia coli was cultured from both the urine and from the hip aspiration. The prosthesis was removed. A
114 Ultraviolet Light at Duke third patient developed pain about the greater trochanter several months after total-hip replacement which was accompanied and followed by a benign course. The pain was thought to be due to trochanteric wire and was removed by a sur- geon where the patient was visiting. The wound had been closed at the time the wire was removed, but when he returned to Duke several weeks later, drainage had occurred and Staphylococcus aureus was cultured. Case #1 was definitely an operating room infection but probably not airborne. Case #2 is in the category of endogenous infection, and Case #3 was probably not an operating room infection at the time of the initial total-hip arthroplasty. The five other patients being considered with deep wound infections had all undergone operative procedures prior to the total-hip replacement. One patient developed a wound hematoma two weeks after the operation while on Coumadin. A gram-negative organism was cultured from the hematoma; the wound was opened, evacuated and dressed daily for about three weeks. The wound sub- sequently healed, and four and a half years later the prosthesis is in place without evidence of loosening and the patient has no pain. This was considered as a secondary infection and not primarily one that occurred in the operating room. A second patient with a painful endoprosthesis of the femur developed wound drainage about two weeks after total-hip replacement was done. Pre- operative aspiration had not been productive, but a positive wound culture was obtained from tissue material at the time that the total replacement was done (unknown until 48 hours after total-hip replacement). In spite of long term antibiotic, the prosthesis loosened and had to be removed. This patient infection was considered as one which was present when the total-hip replacement was done, and was not considered as one that originated from the operating room environment at the time of the procedure. A third patient in this group of five had a primary diagnosis of lupus erythematosus and required steroids for control of vasculitis. She developed a gram-negative septicemia and a draining wound, and the prosthesis had to be removed. The origin of the gram-negative organism was not determined. Two other patients in this group developed wound drainage after total-hip replacement. Both had positive bac- terial cultures from these wounds, although cultures taken at the time of the total-hip operation were reported as negative. We do not have an absolute opinion as to the origin of these two infections.
115 J. Leonard Goldner Robert W. Gaines Mary Higgins If one is willing to accept the thesis that airborne contamination is an important source of wound infection in refined, clean cases or in apparent- ly clean areas that have been operated upon before, there were no proven infec- tions that were airborne in origin in this group of hip patients operated upon at Duke Hospital. The air conditioning system as described, the regulation of the ultraviolet lights as will be described, and the precautions already men- tioned were in effect during this five-year period. AIR SAMPLING IN RADIATED AND NON-RADIATED ROOMS AT DUKEâVOLUMETRIC AIR SAMPLING In March of 1972 another study was initiated on atmospheric conditions in the Duke operating rooms, and this time the volumetric Raynier Air Sampler was used. The collecting unit was placed first above five feet from the operating table where the room traffic and activity were tabulated as the plate was inoculated and the effect of these variables could be documented. Later, a plastic tube to collect air was anchored within six centimeters of the wound for collection of air. The colony-forming groups per cubic foot per 3 3 minute (CFG/ft /min) and per cubic foot per five minutes (CFG/ft /5 min) were gathered and plotted on a graph on several occasions -- none of which initially were hip arthroplasty. The colony-forming groups varied from zero to 3 three CFG/ft /min, averaging one to two in the relatively quiet, unoccupied room. As the number of individuals in the room increased, the level of air contamina- tion increased. Opening and closing doors did not in itself cause a sig- nificant alteration if the individuals left the room. However, if the doors were opened and more individuals entered the room such as a nursei anesthesi- ologist, surgical attendant, physician, or observer, the CFG increased rapid- ly (Table 4-A through 4-C). Details concerning the culture plates within the sampler have been reviewed. All culture material is prepared as needed and refrigerated until used. Refrigeration keeps the surface area moist, and prolongs the "shelf life" of the media. The efficiency of collection is dependent upon the dis- tance of the slit to the media plate. This is adjustable each time the plate is inserted into the sampler. The efficiency is approximately 95 percent. The advantage of this sampling procedure is that the particles are impacted on the plate, and only a small portion of the plate is exposed to the airflow as the plate rotates every thirty minutes. Therefore, drying out the media
116 Ultraviolet Light at Duke and the collected particles is minimized. Once the operative procedure is completed, the plates are taken immediately to the Microbiology Laboratory and incubated. The question always arises as to the effect of ultraviolet light on the culture media in order to ascertain the possibility of the media being affected in such a way that the organisms would be less likely to grow than on the usual media. The media used in settle plates during operative procedures is Trypticase Soy Agar and not Blood Agar. Our observations have been that the culture media, once exposed to ultraviolet light, was not discolored or af- fected. Additional review of the area outside the main operating rooms at Duke showed that the traffic in the hallways was not strictly controlled. In- dividuals passing from one part of the hopsital to another did have access to the hallways, street clothing was permitted in these corridors and referring or supervising members of the staff could walk in the hallways outside of the main operating room without changing shoes or clothes. This is the pattern that had been allowed for many years and presumably because of the ultraviolet lights in the operating room, the infection rate had not been affected adverse- ly. The initial volumetric sampling in the total hip rooms was done during the time the free flow of traffic was allowed, and in spite of this, the level of airborne organisms within the rooms was well below the suggested minimal standard. The fluctuation in bacterial counts when individuals entered the rooms, and the elevation of counts during cases when air currents were increased and between cases when motion and talking increased was evident. The ortho- paedic infection rate has been maintained at the low level mentioned despite relative freedom of traffic throughout the entire operating suite since 1935. With the advent of total-hip arthroplasty and other joint implants in orthopaedics and in keeping with the effort to standardize the operating room environment, a restricted traffic policy and specified requirements of dress for 'hose within the operating room area have been established. Entrance and exit from the orthopaedic operating rooms have been reduced somewhat, but control is not stringent. Members of the nursing service, surgical attendants,
117 J. Leonard Goldner Robert W. Gaines Mary Higgins anesthesiology personne1, orthopaedic residents, and others, continue to enter and leave the operating room on numerous occasions. In spite of this, the air sampling study shows satisfactory control of the bacterial content in the air because of the air conditioning system, and that the number of pathogens reaches a low level if the ultraviolet lights are used. Data concerning comparative studies of air sampling in orthopaedic rooms both with and without ultraviolet lights are presented in a preliminary form in this paper without precise con- clusions being drawn at this time (24). RECONFIRMATION OF THE EFFECTIVENESS OF ULTRAVIOLET LIGHT ON BACTERIA IN THE OPERATING ROOM The purpose of the study was to demonstrate the time-death relation- ship of a known number of Staphylococcus aureus coagulase (+) to ultraviolet light irradiation at the operative site in two orthopaedic operating rooms at Duke University Medical Center. These are the two rooms in which total-hip arthroplas ties are done and in which volumetric air sampling is being done with ultraviolet lights either on or off. The placement and arrangement of the ultraviolet lights in the two rooms was different, and this study allowed com- parison of the placement in a low ceiling room with placement in a high ceiling room. An appropriate Staphylococcus aureus coagulase (+) dilution was made as follows: a lyophilized disk of Staphylococcus aureus coagulase (+) (Stan- dard Disk; Difco) was aseptically placed in Trypticase Soy Broth and incubated at 37° C for 24 hours. After 24 hours of incubation, serial dilutions of the original broth culture were made and spread onto a Blood Agar Petri dish (15 mm x 150 mm). The surface of the Blood Agar plates was allowed to dry at room temperature and then incubated at 37° C for 48 hours. After incubation, the di- lution (10 ) that produced approximately 200 CFG on the Petri dish was chosen. Preparation of a Staphylococcus aureus coagulase (+) 10 dilution was prepared and one milliliter of the 10 dilution was spread onto 16 Blood Agar Petri dishes (15 mm x 150 mm). The surface of the Blood Agar plates was allowed to dry
118 Ultraviolet Light at Duke at room temperature. The Petri dishes were taken to Operating Rooms 3 and 4 and placed at what approximated the operative site. In Operating Rooms 3 and 4, there was a control plate (no exposure to UV irradiation), and the remaining Petri dishes were exposed to ultraviolet irradiation for 5, 10, 15, 20, 30, 6O, and 120 minutes respectively. Petri dishes were then incubated at 37 C for 48 hours and remaining colony-forming groups were counted and recorded. The original number of colony-forming groups in the broth culture is the number of the control plate. Table 1 shows the percentage of reduction of bacteria with the ultraviolet light on during a recorded time interval. Table 1. UV LightâPercent Reduction in Bacteria Room 4 Room 3 # Colonies 170 % Reduct./Time 0 # Colonies 130 % Reduct./Time 0 Control 5 min 29 83% 45 65% 10 min 15 91% 6 95% 15 min 8 95% 5 96% 20 min 0 100% 0 100% 30 min 0 100% 0 100% 1 hr 0 100% 0 100% 2 hr 0 100% 0 100% The time-death relationship of a known number of Staphylococcus aureus coagulase (+) to ultraviolet light was measured in two orthopaedic operating rooms at Duke. The plates with a known Staphylococcus aureus dilution spread on Blood Agar Petri dishes were exposed to ultraviolet light at a point that approximated the operative site. In Room 4, which was equipped with eight ceiling lights and twelve wall lights, there were 170 colonies X 10 as a control. The control plate was not exposed to UV light and at the end of two hours, there was no reduction in the number of colony-forming groups. In the plates exposed to UV light at five min- utes, the number of colonies had decreased to 29 (83 percent reduction in the number of colonies). At ten minutes after UV light exposure, the number of
119 J. Leonard Goldner Robert W. Gaines Mary Higgins colonies was reduced to 15 (91 percent reduction), and at 15 minutes, the reduction was to eight colonies (95 percent reduction). By 20 minutes after exposure to UV light, there were no colonies noted, nor were there any at 30 minutes, one hour, or two hours. The UV light reading in that room was 17 MW/ cm /sec. Operating Room 3 has no overhead lights but there are 12 sidelights. 2 The intensity in this room is usually 12 MW/cm /sec, whereas the readings in 2 Room 4 are usually 17 MW/cm /sec. In Operating Room 3 the ultraviolet lights are on the sidewalls and shining indirectly as ceiling lights. (The ceiling is not of sufficient height to allow lights to be mounted on the ceiling.) The number of colonies on the initial plate was 130 x 10 . In five minutes, the reduction was to 45 colonies (65 percent). After ten minutes of exposure to UV light, the reduction was to six colonies (95 percent). After 15 minutes there were five colonies remaining (96 percent). After 30 minutes of exposure, there were no colonies noted, nor were there any after one hour, or two hours. These observations confirm the efficacy of ultraviolet light even though the intensity of the reading is not as high as that which is usually recommended and even though a high ceiling room was not available. The variation in the amount of time necessary to kill all the colonies is related to the intensity. The original studies of Hart show that within three minutes the bacterial colonies were absent if the intensity is main- o tained at 25-30 MW/cm /sec. PRELIMINARY ANALYSIS OF DATA FROM VOLUMETRIC AIR SAMPLING DURING TOTAL-HIP ARTHROPLASTY WITH ULTRAVIOLET LIGHTS ON AND OFF A random study is currently in progress. The particular operation to be done with ultraviolet lights on or off is determined by the surgeon. A high-risk patient, i.e., a renal transplant patient with avascular necrosis having total-hip replacement or a patient with rheumatoid arthritis who is on continuous steroid intake, should get prophylactic antibiotics based on current antibiotic-controlled studies. As of this time, 35 total-hip arthroplasties have been monitored with the Raynier sampler. Nineteen of these have been in Operating Room 3 and 16 in Room 4. The lights have been off five times in Room 3 (5 of 19) and six times in Room 4 (6 of 16). There have been no clinical
120 Ultraviolet Light at Duke infections in any of these patients. Please note that the study was not done primarily to determine the rate of infection but rather to determine the CFG of bacteria near the wound during the operative procedure. The pattern of the operative procedure has been the same in each instance. However, the operative steps have been divided into 1) incision and exposure, 2) removal of femoral head, 3) preparing and reaming of the acetabulum, 4) mixing cement and place- ment of the acetabular component, 5) preparation of femoral shaft and trial re- duction, 6) mixing and placement of cement and final reduction, and 7) wound closure. Until the total number of cases with lights on and off are monitored, the comparison of the data will not be completed. The Chi Square Test will be. done and other comparisons will be made. Meanwhile, certain observations based on the data accumulated are emphasized: 1. Thirty-six patients in Rooms 3 and 4 have had total-hip arthro- plasty. Personnel has varied from case to case, but the operating rooms have been the same and the operation has been of the same kind, although the natural and expected variations have occurred. The length of the procedure has varied from 56 minutes to four hours depending on the problems encountered. 2. The number of colony-forming groups collected on the Agar plates has increased with the number of people in the room, the amount of movement and motion around the incision, and by the number of persons entering the room. The air conditioning system itself, with 35 air changes per hour, was sufficient, to maintain a low number of colony-forming groups measured as CFG/ft /min, if there was minimal activity in the room. 3. Table 2 shows the .percent of isolates recovered for all total- hip procedures (Room 3, 19 operations; Room 4, 16 operations, UV lights off 11 operations, UV lights on 24 operations). The number of organisms recovered with the lights off is almost as great as the numbers recovered with the lights on even though only half as many operations were done with the lights off.
121 J. Leonard Goldner Robert W. Gaines Mary Higgins Table 2. Percent of Isolates Recovered for all Total-Hip Procedures Total # Organisms All Op. 7049 Rm. 3 Rm. 4 4195 UV On 3876 UV Off 3173 2854 Fungi, yeast mold 8.25 5.83 9.86 9.67 6.52 Bacillus 24.81 37.49 16.18 11.35 41.25 species Micrococcus 33.46 22.84 40.69 40.94 24.33 species Diptheroids 19.53 22.10 17.78 22.72 15.63 S. epidermidis 13.33 11.66 14.46 14.88 11.44 Mima 0.18 0.00 0.31 0.28 0.06 polymorpha Klebsiella 0.01 0.00 0.02 0.03 0.00 species S . aureus 0.39 0.00 0.67 0.10 0.75 Coag. (+) No Staphylococcus aureus organism was recdvered from Room 4. The difference tween UV lights on and UV lights off is six times greater with the lights off. Review of the figures of all 35 cases shows that the number of colony-forming groups of all kinds shows a pattern of considerably higher numbers in individual instances with the lights off, as compared with the general pattern of lower num- bers of colony-forming groups with the lights on. These figures reflect 11 operations with the ultraviolet lights off; 24 opera- tions with the ultraviolet lights on. Note that the number of organisms re- covered in the rooms with the UV lights off is approaching that with the UV lights on yet over twice as many operations were done with the lights on as with the lights off. In the entire 35 patients monitored, there were three instances in which Staphylococcus aureus coagulase (+) organisms were recovered by means of volumetric sampling. All occurred in Room 3, where the intensity readings are adequate but where the tests shown in Table #1 indicate that within the first five minutes of exposure of staphylococci on the agar plate, the reduction was 65 percent of the colonies, as compared with 83 percent in Room 4. This may have some bearing on the recovery of this organism in Room 3 only.
122 Ultraviolet Light at Duke Case #1 in Room 3, with the UV lights off during total-hip arthroplasty, had a total number of 190 CFG recovered of which 19 were Staphylococcus aureus coagulase (+). Twelve of 19 CFG were recovered during preparation and insertion of the glue in the acetabulum, and seven were recovered during exposure prior to removal of the femoral head. Wound cultures for closure showed no growth, and a settle plate on the back table showed four CFG of diphtheroids and five CFG of Staphylococcus epidermidis. The patient was on the operating table a total of 56 minutes and the average bacter plates, not volumetric sampling). 3 56 minutes and the average bacterial count was 0.16 CFG/ft /min (four settle Case #2, with ultraviolet lights off, showed five of 722 CFG recovered to be Staphylococcus aureus coagulase (+). Comparatively speaking, this was the highest total count of any of the 35 patients monitored. One Staphylococcus aureus of five was recovered during step 4, which was preparation and insertion of the methacrylate in the acetabulum, and four Staphylococcus aureus colony- forming groups of five were recovered during the wound closure. Wound culture after irrigation and before closure showed no growth. The settle plate on the back table showed 12 CFG diphtheroids, 14 CFG micrococcus sp., three CFG Staph- ylococcus aureus coagulase (+). Total time on the operating table for the â¢j patient was 95 minutes, and the average CFG was 0.30 CFG/ft /min (four settle plates, not volumetric sampling). Case #3 in Room 3, with the UV lights on, showed Staphylococcus aureus coagulase (+), and the total number of organisms recovered was four of 80. Three Staphylococcus aureus coagulase (+) of four were recovered during step 4 (preparation and insertion of the methacrylate in the acetabulum) and one of four Staphylococcus aureus CFG during step 6 (final reduction). Wound culture after irrigation and before closure was negative and the settle plate showed no growth. Certain trends are implied as the data from these three patients were reviewed. As in the determination of variations in volumetric air sampling in other operating rooms in this same suite, sampling was done throughout the entire procedure for a single bypass coronary artery graft. The total number of organ- isms recovered was 1,020. The specific details concerning organisms and per- centages of the total are noted in Table 2. The total time of the operative
123 J. Leonard Goldner Robert W. Gaines Mary Higgins procedure was two hours and 49 minutes. The total number of door openings allowing entrance of an individual into the room was approximately 64. The patient was on the pump for about one hour and considerably more individuals were in the room than are present during a total-hip arthroplasty. TECHNICAL DETAILS CONCERNING ULTRAVIOLET LIGHT The ultraviolet light tubes are cool, 30 inches (75 cm) long and contain Argon, Neon and Mercury. Over 80 percent of the output of such tubes is at 2,537 angstrom units. Equipping an operating room with ultraviolet lights re- quires consideration of the following items (Appendix A): 1. ultraviolet light fixtures (brackets for lights); 2. sterile lamps for fixtures (30 W); 3. variac control switch with two per room, since the top lights and the side lights are controlled separately; 4. ultraviolet light monitors which should be available for daily use by nursing personnel; 5. wiring and installation; 6. portable ultraviolet lights for room or floor sterilization after dirty case; 7. plastic glasses with side covers; 8. hoods for total head, neck and face cover; 9. green plastic visors; 10. modification of eye cover in the form of protective goggles that take the place of plastic glasses and the green eyeshades; 11. Figure 1 shows the placement of side lights and overhead lights. A high ceiling in the operating room provides more efficient use of the lights without excessive exposure of the surgeon's head. If the ceiling is low, then the light source must be the sidelights and indirect ceiling lights or sidelights only. When sidelights only are used, it is mandatory that glasses with side protectors be used by everyone in the room, and that the patient's skin be covered with anti-bum cream, or with plastic or cloth covering.
124 Ultraviolet Light at Duke DISTRIBUTION OF WALL AND CEILING MOUNTED UNITS 8 9 10 16 l5 12 t Fig. 1. Distribution of wall- and ceiling-mounted ultraviolet bulbs. The lights mounted on the walls are covered by brackets that prevent direct observa- tion of the bulb by the eyes of the personnel in the room. The overhead lights should have a direct ac- cess to the wound and not be blocked by intervening objects. Intensity of the overhead light is inversely proportional to the square of the distance. This arrangement is utilized by all rooms with a high overhead ceiling. If the ceiling is only 10-12 feet, then the sidelights alone are used.
125 J. Leonard Goldner Robert W. Gaines Mary Higgins The ultraviolet light offers the following advantages: 1. high output in the bactericidal range of the spectrum and low output in the erythemic range; 2. negligible production of ozone, which is rarely detectable by odor even with as many as 16 tubes in the operating room; 3. negligible production of heat; 4. low cost of the irradiation unit, which is made of high trans- mission glass instead of quartz and averages about $1,500 per room, including equipment and installation; 5. low operating cost. The estimated life of the tube is between 4,000 and 5,000 hours, and from 10-16 tubes can be operated with as little current as is required to light a 150 W electric bulb. An average-size operating room at Duke is equipped with four overhead brackets with two bulbs in each bracket, and eight sidelights with two on each wall, or a total of 16 bulbs in each room. This has been the usual number of lights used for the past 35 years without harm to personnel. Intensity of irradiation varies approximately inversely as the square of the distance from the source of radiation and may be compensated by variations in the output of the unit. Room temperature and humidity affect the output. Figure 2 (28) shows that the effectiveness of ultraviolet light drops percipitously as the humidity reaches 60 percent. A humidity gauge should be available in each operating room, and when the humidity approaches 55 percent, the engineer should be asked to reduce it. A desirable level is from 40-50 per- cent. This increases the efficiency of the air conditioning, and also assures full strength of the ultraviolet light. The dangers thought to be associated with the low humidity are overcome now by the types of inhalation anesthetics being utilized. The intensity of the ultraviolet light in the Duke operating rooms is sufficient to kill bacteria on the floor, and on objects at levels close to the floor. The intensity is gauged to be safe for exposure of the surgeon who may be standing anywhere from five to seven feet from the floor.
126 Ultraviolet Light at Duke o â¢m> e tr liJ § a. o o cc UJ m l500 l000 500 10 20 30 40 50 60 70 80 90 RELATIVE HUMIDITY % Fig. 2. Graph (from material by Lowell et al.) con- cerning the effect of relative humidity on bacterial power of ultraviolet light. Note that at 60 percent relative humidity there is a precipitous drop in the efficiency of the intensity of UV light.
127 J. Leonard Goldner Robert W. Gaines Mary Higgins Fungi and bacteria will be killed within three minutes at the five foot distance. Aspergillus, a non-pathogen which grows in the sunshiny and produces a large pigment, was the most resistant. It survived an exposure of 20 minutes to one hour. The recent studies in the Duke operating rooms, demonstrated by Table 1, show that within 15 minutes the Staphylococcus aureus coagulase (+) is eradicated from the exposed areas in the operating room. Cold air blowing on the ultraviolet bulbs will also diminish the effectiveness of the ultraviolet ray. Daily recordings of room temperature, humidity, and intensity of the ultraviolet lights are desirable. The rheostat that controls the intensity should be available to select operating room per- sonnel, who should be instructed in monitoring the intensity of the ultraviolet light in each room daily. The use of ultraviolet light in operating rooms at night, when there is no traffic and no activity, is helpful, but after 20-30 minutes, there are probably no viable bacteria in the room. Furthermore, with no convection cur- rents and with no individuals in the room, the likelihood of further contamina- tion is minimal. The use of ultraviolet light between operating cases is not essential, although from a conditioned reflex standpoint it may be desirable. Occasionally, the lights are turned off while the room is being cleaned, and they are not turned on at the proper time for the next case. If the ultraviolet lights are off between operations and personnel are in the operating room without masks, the environment becomes contaminated. Talking, air currents, the moving of * linens and other objects, all increase the bacterial count. However, once the lights are turned on and 15 minutes have elapsed, the bactericidal effect on the exposed organisms in the room and on open surfaces is sufficient to main- tain the bacterial contamination at a very low to zero level. The policy in the orthopaedic operating rooms at Duke is to keep the ultraviolet lights on continuously. The constant flow of traffic from the hall- way to the operating room the conversations among personnel within the operating
128 Ultraviolet Light at Duke room, and the frequent visits by the anesthesiologists and the operating room nurses, all provide airborne bacteria that should be eliminated. With the UV lights on and the air conditioning functioning, the environment is sanitized (Fig. 3). In a room with unidirectional airflow a lax period between cases may not produce a high bacterial count, because rapid turnover of air continues constantly. However, with the room air changing 36 times per hour, as it does in the Duke operating rooms, the air is cleansed reasonably well after several minutes have elapsed. Bacteria that do arise from the areas of shedding, or from the respiratory tract of personnel in the room, or from other sources, may settle out on instruments, prostheses, sponges, drapes, or into the wound, and the flow of air and frequent changes may be inadequate to eliminate bac- teria from these locations. Ultraviolet light will affect settled surface bacteria, just as has been demonstrated on the settle plates. Furthermore, antibiotics irrigated into the wound constantly during an operative procedure may be somewhat effective in decreasing the number of live bacteria. Systemic antibiotics may also affect bacteria that survive, provided that the proper antibiotic has been used, that the concentration is high enough, and that it is administered at the proper time. Ultraviolet light, however, is bactericidal on airborne particles. It does affect bacteria within the wound that are exposed to the UV rays, and it will kill the bacteria that appear suddenly from personnel entering the room, or from those in the room who might be shedding. Also, the pathologic shedder or carrier is really the target of all our efforts, whether this be done with air conditioning and high efficiency filters, unidirectional airflow, antibiotics, or ultraviolet light. Gryska and O'Dea (12) emphasized that even with prophylactic antibiotics, the strep- tococci carried in an anesthesiologist's rectum may cause a severe epidemic. The solution to that epidemic showed that the organism was recovered from the anesthesiologist, from the settle plate within the room, and from the wound, indicating that the bacteria was primarily airborne. The points that have been mentioned emphasize the importance of | correctly positioning ultraviolet light in the operating room so that it is able to reach critical areas. The overhead UV lights should not be obstructed
129 J. Leonard Goldner Robert W. Gaines Mary Higgins by solid ceiling fixtures that interfere with passage of the rays to the wound, to the instrument tables and to the personnel in the room (Fig. 4). Bacteria occurring in the air settle on instruments, sponges, pros- theses, and other articles used during the operative procedure. Such objects require more than unidirectional airflow for decontamination (Figs. 5-A through 5-D). Fig. 3. Photograph taken during an operative case with UV lights on and with personnel wearing usual head and body coverage. With the intensity of 17 MW/cm /sec, the circulating nurse will not re- ceive erythema of the arms. At an intensity of 25 MW/cm /sec, the arms will become erythematous after an exposure for an hour.
130 Ultraviolet Light at Duke 12 ll.l 10 9 44 3- 2- 25/jw/cm2 10 8 4202 Distance from Center in Feet 8 10 Fig. 4. Vertical intensity pattern of ultraviolet ray output (2537 AU); diagram of the intensity of bactericidal ultraviolet light measured around the central vertical axis of the Duke-Hart UV light irradiator. These intensities were measured with a calibrated UV photometer. Units are in microwatts of ultraviolet light per square centimeter of area at the distances indicated. 25 MW/cm^ is adequate for very swift bacterial kill. This represents a floor source used for sterilizing rooms that have been contaminated by known infection or for sterilizing rooms that have been occupied by patients who are known to carry pathogenic bacteria. The same principle is applied to the operating room flooded by overhead ultraviolet light or sidewall sources.
131 J. Leonard Goldner Robert W. Gaines Mary Higgins Fig. 5-A. Orthopaedic Operating Room 4 with the conventional lights off and the ultraviolet lights on. Note the UV light mounted above eye level with a protective shield. There are three lights on each wall of the room and eight lights in the overhead well. Intensity of the UV light is inversely proportional to the square of the distance. Thus, the lights at the top of the well provide adequate intensity for bactericidal effect at the level of the open surface of the operating table. The operating room adjust- able, movable lights must be placed in such a way that they do not obstruct the UV light rays from the overhead well. In spite of these obstacles, however, the intensity in the room is adequate.
132 Ultraviolet Light at Duke Fig. 5-B. Photograph taken from the top of the operating table showing the central disc which provides the flow of incoming air in a vertical direction. The eight ultraviolet lights , unshielded, shine directly into the room. Minor obstructions occur from the overhead spotlights which are not visible in this picture. The presence of the well enhances the intensity in that the lights can be placed far enough away from the operating table to avoid excessive expo- sure of the personnel standing next to the operating table. If a well is not present, then indirect lights or sidelights only are used. f
133 J. Leonard Goldner Robert W. Gaines Mary Higgins Fig. 5-C. Orthopaedic Operating Room 3 where an overhead well was not present. In order to avoid excessive intensity from overhead lights, the ceiling pans were added and the UV lights are installed within the pans. The lights are directed to the ceiling which is covered with aluminum paint and the reflected light then enters the room. Note that there are three sidelights on each sidewall. A total of 20 lights are in the room. In this situation, the overhead operating light does not obstruct the major diffusion of indirect UV light.
134 Ultraviolet Light at Duke Fig. 5-D. The same room with ultraviolet lights turned off and conventional lights turned on. Note the air inflow disc in the ceiling above the wall clock. The upright Duke-Hart UV lights are used to decontaminate the floor, the sidewalls, the air and the entire room after a known contaminated wound, is treated. The in- tensity of the upright lights is about 100 MW/cm^/min at a dis- tance of about nine feet. At a distance of 20 feet, these lights are still effective in killing bacteria in the air and on exposed surfaces.
135 J. Leonard Goldner Robert W. Gaines Mary Higgins COOPERATIVE STUDY OF ULTRAVIOLET LIGHT IN OPERATING ROOMS, SPONSORED BY THE COMMITTEE ON TRAUMA OF THE NATIONAL RESEARCH COUNCIL - 1959 The study outlined by the National Research Council (NRC) was to be a controlled one where several centers in the country were selected, and ultra- violet lights were placed in selected operating rooms. Functioning lights were used in one room in the hospital and sham lights were used in another. The problems related to testing the intensity of ultraviolet light were present at that time and were not monitored closely in each location. Just how well the intensity readings were maintained and how constant the hu- midity and temperature factors were maintained is not known. Certainly, as much difficulty in determining intensity was encountered then as now. Moni- toring is relatively simple and should be the responsibility of supervisors who understand the purpose of the lights and the way they function. The bulbs should be clean, intensity should be logged, and the temperature and humidity should be recorded. In spite of the variations in ceiling height in certain operating rooms at Duke, the intensity of the ultraviolet lights is still adequate to carry out the assigned task. Because Duke has 23 years of experience with ultraviolet irradiation, and since we were convinced of its value for clean operations, we did not par- ticipate in the NRC study because half of the patients undergoing clean opera- tions at Duke would have been operated upon without lights, and we felt that we had proven to ourselves that the risk of infection was greater for that group. In those hospitals around the country selected for the study that had depended on air conditioning and other means of preventing operating room in- fection, sham lights or regular lights did not affect their usual and cus- tomary routine. Duke, however, did have an opportunity to present a parallel study without random selection but with good control at the Veterans Administration (VA) Hospital associated with Duke University in Durham, North Carolina. This hos- pital was not equipped with ultraviolet lights when originally built, but the house staff, the senior staff and the techniques employed in general were the same as those used at Duke where the lights were in action.
136 Ultraviolet Light at Duke The Veterans Administration Hospital operating rooms did not have ultraviolet lights from 1958 through 1962. The infection rate for refined clean wounds was 1.7 percent, whereas the overall rate at Duke was 0.34 percent for 23,000 refined clean operations (a difference of five times). Emphasis should be placed on the following fact: Operating rooms that had reasonably good environments, air condiâ tioning with good airflow, and personnel who practiced careful surgical tech- niques, were associated with a relatively low infection rate. (Under three percent has usually been defined as a reasonable infection rate throughout the country.) The three-percent infection rate in refined clean cases, however, was considered to be excessively high when compared with the VA Hospital infection rate when lights were not used. Furthermore, there was no assurance from day to day that bacteria introduced into the operating room by a carrier or shedder might not be responsible for an epidemic of infection. The VA Hospital infection rate for drained clean wounds was three times higher than the Duke rate; and the rate for contaminated wounds was two times as high as those treated at Duke with ultraviolet irradiation in the room. Therefore, the latter two categories are eliminated from comparison and only refined clean operations should be considered. At the present time, with the wound hemovac being used on practically all large operative cases, another factor has been added that makes refined clean cases more difficult to define because of this special kind of drainage. The chance of infection occurring is greater because of the hemovac tube which leaves a persistent opening in the skin for a few days, but the likelihood of infection diminishes because of decrease in the size of the hematoma. A retrospective study at Duke from 1941 to 1945 showed an unexplained infection rate of 0.34 percent in refined clean cases. Prior studies by Hart and his associates had shown that comparative groups of patients operated upon with and without ultraviolet light in the room showed noticeable differences in the percentage of positive cultures obtained from postoperative drainage of the wounds. When ultraviolet light was not used, the occurrence of a positive culture from the wound drainage was eight times more prevalent than when
i 137 J. Leonard Goldner Robert W. Geines Mary Higgins ultraviolet light had been used. Also, the amount of postoperative temperature elevation, and the duration of the elevation, was significantly less in those patients operated upon under ultraviolet light than those who were not. The NRC cooperative study included all three categories of wounds that have already been defined: 1) defined clean cases; 2) drained clean wounds; 3) contaminated wounds. If these three categories are not separated clearly and statistics determined on the basis of individual groups, then the overall sta- tistics will not be valid, and the conclusion will be weighted less in favor of ultraviolet light. UV light will not decontaminate a contaminated wound. NATIONAL RESEARCH COUNCIL STUDY ANALYSIS In the NRC study, certain conditions, such as patient population, bacterial carrier states, operative techniques or other unknowns, contributed to the wide variations, all of which cannot be determined. Suggestions were made that sources of bacteria other than through the airborne route were res- ponsible for many infections, and ultraviolet irradiation cannot be expected to have demonstrable benefits for these. Each hospital that participated in the NRC study showed a reduction in the contamination rate when comparing non-ultraviolet and ultraviolet light environment. Examples of some of the reductions are: 3.0 percent to 1.2 per- cent; 4.8 to 2.7 percent; 7.0 to 4.8 percent; 11.7 to 8.6 percent and 8.8 to 6.1 percent. These included refined and other clean, combined operations. The hospitals with the lowest infection rate initially prior to the study showed the greatest improvement with ultraviolet light which would indicate that airborne infection is still the most important cause of operating infec- tion in refined, clean wounds. Also, in certain hospitals where the infection rate was low without ultraviolet light, the number of patients included in the overall study must be extremely high to see a noticeable variation in the final statistics. A change from one percent to 0.5 percent would require a minimum of 200 patients. In order to avoid trends and special problems involved, the statistics should probably be based on 1,000 patients, all of whom have had the same kind of operation (total hip, spinal fusion), were operated on in the
138 Ultraviolet Light at Duke same room, and were managed by the same personnel. Studies should have an ade- quate and complete follow-up by a trained individual assigned to the project who has free access to all data at all times. The NCR study and the Duke study agreed on the following: 1. The air in occupied operating rooms without ultraviolet irradia- tion is contaminated with bacteria of varying degrees of pathogenicity, par- ticularly staphylococci, which precipitate continuously on all exposed surfaces. 2. Direct ultraviolet irradiation has a highly efficient bacteri- cidal effect which kills almost all types of organisms and will markedly reduce any airborne bacterial contamination in the operating room. (Ultraviolet light kills all bacteria, fungi, including Anthrax.) 3. With suitable protection, direct ultraviolet irradiation is safe for operating room personnel and patients. 4. With direct ultraviolet irradiation in the operating room of suitable intensity, there is significant reduction in the number of postopera- tive wound infections following refined clean operations. The article (31) written in August, 1964, concerning the NRC study consisted of 192 pages. The only part of the detailed study that has been read by most surgeons is the summary. It presented in three sentences the NRC cooperative study conclusions regarding the value of ultraviolet irradiation. These are contrary to some results given in the body of the report, and are worded in such a way as to disparage the value and discourage the use of ultra- violet irradiation. Hart and his colleagues indicated that they believe the concluding summary deserved further comment "The only category of wounds that benefits significantly from the use of ultraviolet irradiation was the refined clean group in which the average postoperative infection rate was reduced from 3.8 percent to 2.9 percent" (23). An average drop of only 0.9 percent in the infection rate may not impress the casual reader, but it represents an average improvement of 24 percent for the five hospitals with the range of improvement as high as 44 percent obtained by one of the cooperating hospitals. Another W way of interpreting this would mean that there was one less infection for every 100 patients operated upon. This is a significant reduction in the infection
139 J. Leonard Goldner Robert W. Gaines Mary Higgins and morbidity rate, and a great economic saving (23). The NRC cooperative study found this decrease in infections following refined clean operations to be statistically significant. The body of the report states: "On the basis of the observed infection rate in irradiated refined clean wounds, it may be con- cluded that about 30 of the patients would not have had wound infections if ultraviolet irradiation had been used for all of these refined clean cases" (31). EFFICACY OF ULTRAVIOLET LIGHT IN THE OPERATING ROOM If the reader is willing to accept a fair interpretation of the NRC study as it referred to refined, clean cases, then the use of ultraviolet light In the operating room is the simplest, most economic, least involved method of obtaining a relatively clean environment. Hart stated: "The distribution in radion energy of the ultraviolet ir- radiation employed in the operating rooms of Duke Hospital is such that it kills virtually all exposed bacteria that ordinarily precipitate onto the sterile field or instruments from whatever source, be it exhaled breath or floating lint. The viable bacteria count in the air of these irradiated rooms at almost any point at the level of the operating table and above, averages per year less than one colony per cubic foot of air as compared to a yield of 30 to 50 colonies per cubic foot in the same occupied rooms with- out irradiation" (22). Dr. Hart's quotation refers to studies done prior to installation of new air conditioning units that resulted in change of air an average of 36 times per hour. High efficiency filters were also added. The most recent volumetric samplings with and without the ultraviolet lights revealed that the number of colonies per cubic foot without irradiation and with the modifications in air- flow and air conditioning are considerably less than those that were present in the non-radiated rooms when the studies were done many years ago. The re- sults of the volumetric air sampling with and without ultraviolet irradiation during total-hip arthroplasties will be completed by June of 1975. At that time the exact figures as they relate to the many variables mentioned will be available.
140 Ultraviolet Light at Duke OTHER REASONS FOR WOUND INFECTION As we reviewed the present data concerning wound infection and the efforts directed toward reducing air contamination as well as endogenous sources of infection, we concluded that there may be an irreducible minimum. Hart stated: "The elimination of the air as a vector for pathogenic organisms and the provision of more continuously sterile op- erating environmental surfaces through direct bactericidal irradiation does not prove, by any means, that all unexplained infections after clean operations originate from these sources of contamination. Because of inability to prove the exact source and mode of infection after clean operations, the arguments by surgeons for and against the various possible sources and causes will obviously continue for some time to come. Some will feel that most postoperative infections in clean wounds originate from viable organisms left on or washed up from the pores of the patient's skin; others will be sure that they mainly follow breaks in sterile technic such as punctured rubber gloves, or that they are endogenous from some other infected site within the patient. Certainly, all of these sources and others could be mentioned and they undoubtedly do contribute many infections. However, based on our results from 30 years' experience operating under direct ultraviolet irradiation, we feel that 1. most unexplained postoperative infections, which occur after clean operations, have their origin in the operating room while the wound is open; 2. some of the principal sources of the offending bacteria are the mouths, throats, and noses of the operating room occu- pants ; and 3. the organisms from these sources, along with those on lint and other floating particles, reach the sterile field, the '
141 J. Leonard Goldner Robert W. Gaines Mary Higgins instruments and the wound by way of continuous sedimentation through the air. Our studies indicate that ultraviolet ir- radiation will effectively eliminate this air route or spread of infected organisms, and prevent their continuous sedimentary build-up on all exposed surfaces during operations of any duration" (23). Altemeier, in a discussion of the use of UV light in the operating room, mentioned the study done at the institution in which he works and pointed out that the rooms with the UV lights on had an infection rate of 1.1 percent and those with the lights off, an infection rate of 0.3 percent with an average of 0.7 percent (2). This indeed is a low percentage and reemphasizes the fact that careful, atraumatic surgical technic, limited traffic, and other factors can be effective in diminishing the infection rate (2). However, the guarantee for any particular patient, or at any particular time, does not exist unless some other element is constantly functioning. Altemeier found that aspirated saline, which had puddled in the wound, contained bacteria that were endogenous to the patient. His efforts have been on masks, gloves, skin preparation, tissue handling and other technics rather than alteration of the bacteria in the air. Furthermore, Altemeier"s allusion was to all kinds of operative procedures -- clean and contaminated -- while the emphasis at Duke in gathering statistics has been on evaluation of ultraviolet irradiation on refined clean wounds. Ultraviolet light cannot clean a wound that is deeply contaminated. Dunphy (6) has emphasized that with parallel experimental wounds, one being contaminated by bacteria but tissues being handled carefully, and the other being uncontaminated and scrupulously clean but traumatized, the latter showed a greater tendency to infection than did the former . Altemeier emphasized the importance of tissue necrosis and tissue trauma as it relates to the occurrence of wound infection (2). Thus, any particular incidence of infection must be analyzed with consideration of 1) endogenous bacteria from the patient's foci, 2) the operative procedure to be done, 3) the number of operations that have been done in this same anatomic
142 Ultraviolet Light at Duke location in the same patient, 4) that patient's resistance, 5) medications being received by the patient that might affect resistance, 6) the amount of time taken to do the operation, 7) the amount of trauma involved in doing the operation, and 8) the use of prophylactic or actual antibiotics and their effect on particular organisms (proof of effectiveness against all infections is still not available). Various kinds of clean air, i.e., vertical and hori- zontal laminal airflow, have definitely been shown to reduce the infection rate in situations where the basic percentage of infection was extremely high, and the quoted rate in laminal airflow (5) is now at 2.0 to 0.5 percent or less. The latter figures, of more recent origin, compare favorably with the extremely low infection rate in the Duke orthopaedic operating rooms under an ultraviolet light environment during the past 37 years. Tables 3-A through 3-C demonstrate the air quality study data in Orthopaedic Operating Room 4 during total-hip arthroplasty with the lights off. They demonstrate the efficacy of the air conditioning system in this particular operating room suite and show the colony-forming groups collected during various stages of the procedure. The stages are described within the content of the paper. Table A demonstrates the type of colony-forming groups re- covered at the particular time and stage of the operation. The two columns on the right give the results of the Agar plate cultures in colony-forming groups 2 per cubic foot per minute (CFG/ft /min) and the same numbers for magnification purposes in colony-forming groups per five cubic feet per five minutes (CFG/5 3 ft /5 min). The latter can be used for better visualization of a graph, but the actual numbers on the right are significant. Table B shows the number of times the doors were opened and closed. This is more important if there is an indication as to when an individual enters or leaves the room. Table C is a graph showing the number of colony-forming groups as it relates to stage of operation.
143 J. Leonard Goldner Robert W. Gaines Mary Higgins Table 3-A. Type of CFG Recovered at the Particular Time and Stage of the Operation OR ROOM# _ * DATE _ _Z-08-74 UV LIGHTS Off TYPE PROC. Total Hip - Left OPER. STEP. 0CFG/5FT3 /5 MIN. #CFG/FTJ r/MIN. PLATE t TIME RESULTS 8:19- 1CFG Fui-.Rus; 5CFG Bacillus sp . : 1 8:24 6 1.2 '. 8:24- i 8:29 1CFG Fungus; 2CFG Bacillus sn.; 1CFG S. epid. 3CFG DiDhtheroids 7 1.4 I ; 8:29- 1 8:34 2CFG Bacillus so.: 3CFG Diphtheroids 5 1.0 ' 8:34- 8:39 1CFG Bacillus sp.: 1CFG S. epid. 2 0.4 8:39- > 8:44 4CFG Bacillus sp . : 1CFG S. epid.; 5 1.0 8:44- 8:49 4CFG Bacillus sp.: 2CFG S. epid.; 6 .â¢1.2 8:50- 8:55 4CFG Bacillus sp.: 5CFG S. epid.: 2CFG Diphtheroi i s 11 2.2 i 8:55- 9:00 4CFG Bacillus sp.: 1CFG S. epid.: 5 1.0 2 ⢠9:00- ! 9:05 ZCFG Bacillus sp.; 1CFG S. epid.: 3 0.6 9:05- 2CFG Bacillus sp.: 1CFG S. epid. 9:10 3CFG Diph 'leroicU 6 1.2 9:10- 9:15 2CKG Bacillus so.: 1CFG Candida sp . : i 2CFG S. epid.: 2CFG niohtheroids 7 1.4 9:13- 9:20 5CFG Bacillus sp . : 4CFG S. epid.: ' q 1.8 < 9:21- 9:26 1CFG FunRus 1 0.2 9:26- 9:31 1CFG Bacillus sp. : 1 0.2 1 9:31- 9:36 4CFG Fungus: 3CFG Bacillus sp.: 1CFG S. epid. 3 1 9:36- ICF'G Fungus: 1CFG Diphtheroid, 8 1.6 9:41 2 0.4 9:41- 9:46 1CFG Bacillus sp . : 1CFG S.epid.; 1GFG 1CFG Micrococcus sp.: 3 0.6 9:46- 9:51 1CFG Bacillus sp . ; 1 0.2 9:52- 9:57 1CFG Bacillus sp.; 4CFG Diphtheroids 5 1.0 9:57- 0:02 2CFG Diphthsroids: 3CFG S. epid.: 5 1.0 10:02- .0:07 3CFG Bacillus sp. . 4 I 3 0.6 .0:07- 0:12 3CFG Bacillus sp. ; 3 0.6 .0:12- 1CFG Bacillus sp.: 1CFG S. epid.: 0:15 2/3Min. 0.7 TOTAL 113 106 0.9 -'â¢
144 Ultraviolet Light at Duke Table 3-A. Continued OR ROOM f DATE UV LIGHT* 2-08-74 Off TYPE PROC.Total Hip -Left PL ATE if OPER. STEP TIME RESULTS 0CFG/30FT3 /30MIN <?CPG/FT / MIN 2CFG Fungus: 18CFG Bacillus sp. 5CFG S. epid. 1 1-3 8:19- 8:49 6CFG Diphtheroids 31 1.0 1 CFG Candida sp . : 19CFG Bacillus sp.: 2 3-5 8:50- 9:20 14CFG S. epid.: 7CFG Diphtheroids 41 1.4 6CFG Fungus: 6CFG Bacillus sp . : 2CVG S.^pid. . 3 5-7 9:21- 9:51 1CFG Micrococcus sp . : 1CFG Dlphtheroids 16 0.5 8CFG Bacillus sp.: 4CFG S. epid. 4 7 9:52- 10:15 6CFG Diphtheroids 18/23Mln. 0.8 ! TOTAL 113 106 0.9 Table 3^B. Number of Times the Doors were Opened and Closed NUMBER OF TIMES DOORS OPENED OR CLOSED OR DOOR SCRUB ROOM DOOR INDUCTION ROOM DOOR 24 22 15
mf *1tV -uj 145 J. Leonard Goldner Robert W. Gaines Mary Higgins Table 3-C. Graph Showing the Number of CFG as it Relates to Stages of Operation 3 Ul 0 U> 0 ^ ⢠1 1 1 .,....,,..-,.-â¢-,..--( 8:19- 8:24 ! J 8:24- 8:29 8:29- 8:34 8:34- 8:39 8:39- 8:44 ( j I 8:44- 8:49 \ 8:50- 8:55 S V 4 8:55- 9-00 ⢠9:00- 9:05 [ ⢠9:C5- 9:10 I I fc 9:10- 9:15 \ . 9:15- 9:20 9:21- 9:26 / ' 9:26- 9:31 J 9:31- 9:36 >^^ ( â H X* ^9:36- 9:41 \ M s«:41-9:46 9:46- 9:51 * H \ 9:52- 9:57 \ 9:57-10:02 j -X, 1 10:02-10:07 I 10:07-10:12 1 I. I 10:12-10:13 1 ye g s o H o (n «= cc M- d I I O oo
1A6 Ultraviolet Light at Duke SUMMARY In order to provide clean air and sanitized objects in the operating rooms at Duke University, Hart installed ultraviolet lights approximately 40 years ago. This method of diminishing the concentration of airborne bacteria has allowed the Orthopaedic Service at Duke to report an infection rate on refined clean cases of less than one percent per year during a 37-year period. Ultraviolet irradiation of a specified intensity is effective in killing patho- genic bacteria and fungi within 15 minutes of exposure. Various methods of doing air sampling and bacteria fall-out concentrations have been utilized in order to determine the environmental conditions in the Duke operating rooms. These studies show that ultraviolet irradiation will sanitize the air and ob- jects in the operating room. If the patients and the personnel in the operat- ing room are protected by appropriate skin covering and glasses, there is no harmful effect from the ultraviolet light. The tissues exposed in various kinds of wounds are not affected adversely. Airborne bacteria are the most important causes of wound infection, but other conditions such as self- contamination from urinary tract or abscessed teeth, contamination of the operating site from adjacent areas such as the perineum, contamination of hematoma from adjacent skin caused by wound drainage or by openings in the skin, or involvement of the hematoma by bacteremia, all can account for wound in- fection. Also, insufficient air turnover, and inadequate air circulation due to dirty air filters lead to excess sweating of individuals in the room. These conditions are not eliminated by ultraviolet light, nor by any other type of external air flow. The authors reviewed the National Research Council Study and point out that the evidence indicated that whenever ultraviolet light was used, the incidence of infection when refined clean cases were involved was diminished. Experience with total-hip operations is parallel to experience with all other refined clean orthopaedic cases with respect to the incidence of infection. Statistics quoted by those using laminal airflow compare favorably with those from ultraviolet environment.
147 Ultraviolet Light at Duke REFERENCES 1. Allen, B. L., Jr., M. V. Higgins, and J. L Goldner, The current status of ultraviolet radiation in the operating room. In Press. 2. Altemeier, W. A. and S. Levenson, Trauma workshop report: Infections, immunology, and gnotobiosis. J. Trauma 10:1084-1086, 1970. 3. Bond, R. G., M. M. Halbert, H. D. Putnam, 0. R. Ruschmeyer, and D. Vesley, Survey of Microbial Contamination in the Surgical Suites of 23 Hospitals. University Health Service and School of Public Health, University of Minnesota, Minneapolis, Minnesota, 1964. 4. Brown, I. W., Jr., Wound infections and endocarditis following cardiac surgery. Presentation before the Halstead Society, Montreal, September 1967. 5. Charnley, J. and N. Eftekhar, Penetration of gown material by organisms from the surgeon's body. Lancet 1:172-173, 1969. 6. Dunphy, J. E., Derly Hart Lecture -- Wound Healing Duke University Medical Center, Durham, May 1973. 7. Fitzgerald, R. N., L. F. Peterson, J. A. Washing, II, R. E. VanScoy and M. B. Coventry, Bacterial colonization of wounds and sepsis in total hip arthroplasty. JBJS 55-A:1242-1250, 1973. 8. Fox, D. G., A Study of the Application of Laminar Flow Ventilation to Operating Rooms. Public Health Monograph No. 78. US Department of Health, Education, and Welfare, 1969. 9. Gaines, R. W., J. L. Goldner, M. E. Frankel, R. A. Mortenson, and J. A. Aplington, Prophylactic Antibiotic in Clean Orthopaedic SurgeryâDouble Blind Study. In Preparation. 10. Goldner, J. L. and B. L. Allen, Jr., Ultraviolet light in orthopaedic operating rooms at Duke University - Thirty-five years Experience, 1937- 1973. Clinical Orthopaedics No. 96, October, 1973. 11. Goldstein, A. L. and W. White, Role of Thymosin and Other Thymotic Factors in the Development, Maturation, and Functions of Lymphoid Tissue. Current Topics in Experimental Endocrinology 1:12l-149, 1971. 12. Gryska, P. F. and A. E. O'Dea, Postoperative streptococcal wound infectionâ The anatomy of an epidemic, JAMA 213:7-1189, 1970. 13. Hart, D., Sterilization of the air in the operating room by special bac- tericidal radiant energy, J. Thoracic Surg. 6:45, 1936. 14. Hart, D., Pathogenic bacteria in the air of operating roomsâTheir wide- spread distribution and the methods of control. Arch. Surg. 37:52l-530, 1938.
148 J. Leonard Goldner Robert W. Gaines Mary Higgins 15. Hart, D. and P. W. Sanger, Effect on wound healing of bactericidal ultra- violet radiation from a special unit. Experimental study. Arch. Surg. 38:797-805, 1939. 16. Hart, D., J. W. Devine, and D. W. Martin, Bactericidal and fungicidal effect of ultraviolet radiation. Use of a special unit for sterilizing the air in the operating room. Arch. Surg. 38:806-815, 1939. 17. Hart, D. and H. M. Schiebel, Role of the respiratory tract in contamination of air. A comparative study. Arch. Surg. 38:788-796, 1939. 18. Hart, D. and H. M. Schiebel, Sterilization of air in operating room with bactericidal radiation, comparative analysis of 132 individual stages of extrapleural thoracoplasties performed with radiation and 110 stages per- formed without radiation, J. Thoracic Surg. 7:525, 1938. 19. Hart, D., H. M. Schiebe1, and D. G. Sharp, Bactericidal ultraviolet radia- tion in the operating room: Twenty-nine year study for control of infec- tions. JAMA 172:1019, 1960. 20. Hart, D., The importance of air-borne pathogenic bacteria in the operating room. A method of control by sterilization. JAMA 117:1610-1613, 1941. 21. Hart, D., H. M. Schiebel, and D. G. Sharp, Disinfection of the air in the operating room with bactericidal radiant energy. Correlation of the intensity of radiation with its bactericidal effect. Trans, of the Southern Surg. Assoc. 54:347-372, 1941. 22. Hart, D. and J. Nicks, Ultraviolet radiation in the operating roomâIn- tensities used and bactericidal effects. Arch. Surg. 82:449-465, 1961. 23. Hart, D., R. W. Postlethwait, I. W. Brown, Jr., W. W. Smith, and P. A. Johnson, Postoperative wound infections: A further report on ultraviolet irradiation with comments on the recent (1964) National Research Council Cooperative Study Report. Ann. Surg. 167:728-743, 1968. 24. Higgins, M. V. R. W. Gaines, J. L. Goldner, B. L. Allen, Jr., Air Sampling in the Duke operating rooms. In Preparation. 25. Roller, L. R., Ultraviolet Radiation. John Wiley, New York, 1952. 26. Lawrence, C. A. and S. S. Block, Disinfection. Sterilization, and Preserva- tion, Lea & Ferbiger, Philadelphia, pp. 761-773, 1968. 27. Light, S. Therapeutic Electricity and Ultraviolet Radiation. Waverly Press, Baltimore, 1959.
149 Ultraviolet Light at Duke 28. Lowell, J. Drennan, Personal Communication. 29. McCollum, Donald E., Personal Communication. 30. Moody, Walter R., Suggested Protocol for Monitoring, Maintaining, and Insuring Safe Usage of Ultraviolet Germicidal Lamps in the Hospital. MSPH Thesis. University of North Carolina. Chapel Hil1, N. C., 1971. 31. NRC Study, The Influence of Ultraviolet Light on Postoperative Wound Infections - Supplement to Ann. Surg. 160, No. 2, 1964. 32. Reutschler, H. C., R. Nagy, and G. Monromseff, Bactericidal effect of ultra- violet radiation. J. Bacteriology 41:745-774, 1941. 33. Ritter, Merrill A., Personal Communication. 34. Sharp, D. G., A quantitative method of determining the lethal effect of ultraviolet light on bacteria suspended in air. J. Bacteriology 35:589-599, 1938. 35. Sharp, D. G., Lethal action of short ultraviolet rays on several common pathogenic bacteria. J. Bacteriology 34:447-460, 1939. 36. Sharp, D. G., Effect of ultraviolet light on bacteria suspended in air. J. Bacteriology 39:535-547, 1940. 37. Wright, R. L., Septic Complications of Neurosurgical Spinal Procedures. Charles C Thomas, Publishers, Springfield, 111., 1970.
150 J. Leonard Goldner Robert W. Gaines Mary Higgins APPENDIX A EQUIPMENT NECESSARY FOR INSTALLING ULTRAVIOLET LIGHTS IN OPERATING ROOMS AND FOR PROTECTING PERSONNEL WORKING IN THOSE OPERATING ROOMS FROM THE ULTRAVIOLET LIGHT Ultraviolet fixtures (brackets for lights) Model PF511 - Atlantic Ultraviolet Light Co., purchased through Biddle Purchasing, 10 Riverside Plaza Chicago, 111., 60606 (estimated cost $30 each). Strato-Ray Co,. 6500 Walker Street, Minneapolis, Minn., 55426: Manufacturers of Strato-Ray Air Sanitizer. Distributors of Westinghouse Sterilamp Germicidal Ultraviolet Tubes. You may write to this company for information concerning UV light equipment installation. Variac Control Switches - Two for each operating room (one for the top lights and one for the side lights - controlled separately). This regu- lator switch establishes the intensity of the lights. Order from General Radio Co., West Concord, Mass., Box 872, Boston, Mass., 02103. Request a W-5 Variac, catalog #3030-5110 (estimated cost $25 each). In addition to this Variac, a control switch (custom enclosure) is necessary for each specific room design. Monitoring equipment - commercial meter - purchased from Ultraviolet Products, San Gabriel, Calif., 91778, Model #J-225, (estimated cost $125). More efficient monitor and one that can be used by the nurses in the operating room is Germicidal Photometer (IL 254) from International Lights, Newburyport, Mass., 01950, (estimated cost $600). Sterile lamps for the fixtures manufactured by Westinghouse, catalog #WL 782-L30, #5B-30, obtained from MilIpower Inc., Box 299, Charlotte, N.C., (estimated cost $15 each). Portable ultraviolet lights are used at Duke for additional decon- tamination of the operating room following contaminated cases; they are man- ufactured by Ward Laboratories, 727 Markham Avenue, Durham, N.C., estimated cost of $800 each. Also, a different size lamp fixture, one inch shorter than the lamps for the wall units, must be obtained from Millpower Co. Additional miscellaneous information concerns the protective head coverings and glasses for protection of individuals working under the ultra- violet lights. The hoods can be purchased from several different manufacturers including Spartan Health Care in Spartanburg, S.C., 29301, #HOD409, 50 per box, 48 boxes per case at $37 per case or American Hospital Supply (check with your local dealers) converters #4381 (also American Hospital #47325-012), 100 per case at $16.50 per case, 3700 Atmore St., Charlotte, N.C., 28205.
151 Ultraviolet Light at Duke The protective glasses are purchased from Watchemaket Optical Co., Inc., Providence 3, R.I., Catalogue //CSSF-412 Tuck-A-Way Glasses, 1 per box at $3.25 each. The green visors are purchased from Krisloid Plastics, Inc., Corner Eddy & Porter Streets, Providence, R.I., 02905, #LBX Shades, green, 12 per box at $55 per gross.
152 TWO CONSIDERATIONS IN THE APPLICATION OF LAMINAR CLEAN AIR Edward 0. Goodrich, Jr.* In considering possible surgical applications of ultra-refined air systems, I would like to emphasize two frequently overlooked points. These considerations arose during early attempts to apply "laminar clean" air tech- niques to the general surgical environment. We were given a four-foot by four-foot vertically adjustable cross flow High Efficiency Particulate Air (HEPA) filter unit capable of projecting clean air uniformly from the filter face at a linear rate of 100 feet per min- ute or one mile per hour (1600 cfm)(Fig. 1). When running in an empty room, a standing pyramid of clean air (as measured by near zero counts on all chan- nels of a Royco model 200B particle counter from 0.3y to lOu, sampling 300 cc of air per minute) projected outward from the lip of the unit to a distance of approximately four feet at the apex. This clean zone was surrounded by a stand- ing outward radiating zone of turbulent mixed clean and ordinary room air. The standing mixed zone contained much cleaner air than the surrounding room. How- |? ever, when the unit had been operating in the room for an hour or more, the en- tire room became cleaner because the room volume was only 2800 cubic feet with 12 changes per hour, so that a volume of air equal to the total room volume was being completely cleared of measurable particles every two and three-eighths minutes, and by dilution and turbulent mixing alone, a five-fold decrease from 10,000 to 2,000 particles was observed in the ambient room air. Measuring particle counts at the wound when it was possible to posi- tion and drape the patient so that the wound could be approached from the side, and the part to be operated upon could be placed well within the clean pyramid, such as for a hand or foot procedure, it was possible to operate without regis- tering any particles. However, in the ordinary laparotomy configuration, with the unit at the foot of the table blowing towards the head, counts averaging 500/300 cc were obtained (Fig. 2). *Santa Fe, New Mexico. (Chairman, Subcommittee on Air Environment, f⢠Committee on Operating Room environment, American College of Surgeons).
153 Edward 0. Goodrich, Jr. Fig. 1. ⢠HEPA filter unit. Fig. 2. Configuration of HEPA filter unit at foot of table, blowing towards the head.
154 Two Considerations in the Application of Laminar Clean Air In attempting to contain and direct the clean air towards the abdomen, first bottom, then side drapes were added (Fig. 3). This lowered the counts to the 100/300 cc range, but we were still unable to transfer filter face conditions to the sensitive area, and we could not understand why. Only after extensive smoke testing, did we realize that the channeled column of air, which was otherwise maintaining piston characteristics, gradually lifted up- ward from the bottom sheet as it proceeded away from the filter face, causing room air to be aspirated in a reverse flow from the head of the table, for a distance of three feet in a layer approximately one inch thick right over the bottom drape and the wound. To control this aspiration, we added a plastic cover to the side drapes (Fig. 4) and then were able to obtain counts in the 0 - 10/300 cc range just upstream from the wound (Fig. 5). Downstream counts were less reliable, since we would occasionally and inadvertently misplace a gauze sponge, or a starch-bearing gloved hand, adjacent to or over the sampling tube. When not so distorted, the downwind counts were usually in the 10 or less range. In all these configurations, the nurse and the instrument stand were positioned Fig. 3. Unit with bottom and side drapes added.
155 Edward 0. Goodrich, Jr. ill Fig. 4. Unit with plastic cover added to side drapes. Fig. 5. Obtaining count upstream from wound.
156 Two Considerations in the Application of Laminar Clean Air downwind from the wound, and the stand was draped to create the least possible resistance to, and disturbance of, the airflow. Air passage above and below the horizontal portion of the draped Mayo stand was also demonstrated by smoke tests. Using these draping modifications, we were able routinely to perform two- and three-hour abdominal procedures with total particle counts below 1O per 300 cc of air, measuring all particles from 0.3y to lOy in diameter. Since the natural distribution of particles is heavily weighted toward the lowest sizes, one could reasonably expect that the fewer-than-ten particles per 300 cc which we were able to measure (starting at 0.3y instead of 0.5y as is required in Federal Standard 209B) were primarily particles of less than 0.5y in size. When we looked at the sizes of particles using the discretionary mode of the Royco, few were over l.Oy. Yet it has been well es- tablished that bacteria-laden particles are usually over 4.0y in diameter. Therefore, although one could reasonably state, as has been done several times at this workshop, that viable particle counts are not related to total particle counts, on the other hand, one must be prepared to accept very low total parti- cle counts, especially when measuring at 0.3y and up rather than at 0.5y and up, as excellent inferential evidence that this air is sterile, since there exists work which tends to indicate that an average relationship between total parti- cles 0.5y to lOy to viable particles is 1000 to 1. Further, without testing with smoke any ultra-refined room or unit in its full working configuration with machines, people, lights, and drapes, the prediction of the flow of its air currents is nearly impossible. It is only by visualizing the currents that the vagaries of their distribution will be appreciated. f
157 STANDARD TECHNIQUES TO REDUCE OPERATING ROOM BACTERIA Howard P. Hogshead* This paper is devoted to a discussion of those techniques which can be used in the average operating room to reduce bacterial contamination. This implies the use of conventional air handling systems and conventional apparel. Time does not permit a review of the history of aseptic surgery; only a few topics can be discussed in this paper. Based upon some recent research studies some suggestions for control of bacteriological hazards in the average operating room will be offered. Several studies of the total operating room environment have been reported recently. Such studies must be carried out on a grand scale. A mass of data is collected concerning the particles and the bacteria in the air in several locations around the room, the surgical wound, the surgical team and the surgical equipment. Inherent weaknesses of these studies are the vast number of variables. These variables are difficult to control, since they include changing members of the surgical/anesthesia/nursing team, changing surgical patients, breaks in surgical technique, and variance of surgical procedure. Any attempts to relate bacteriological studies to the terminal event - - a clinical wound infec- tion--are fraught with further difficulties. At the current low rate of infec- tion, the study of a vast number of cases is required to yield a statistically significant difference between two differing surgical techniques. Because of the difficulties involved in such studies of the total operating room environ- ment, an alternative method of study was selected. METHOD In order to eliminate variables and to study individual factors in a controlled environment, a small, sealed "clean-air" room was constructed (Fig. 1). Construction of this unit and subsequent studies described herein were the results of the efforts of several people, including Dr. Charles L. Rogers, M.D., Joseph Morris, M.D., and Jeorge Franco, Ph.D. This unit permitted the study of personnel, operating room dress, and operating room activity. *Director of Rehabilitation, Cathedral Health and Rehabilitation Center, Jacksonville, Florida.
158 Fig. 1. The "telephone booth" used to study 18 individuals, three types of apparel, and three types of activity. The fans and HEPA filters are at right. High velocity, sterile air is directed through the duct and down into the chamber to flush extraneous particles and bacteria prior to the ten-minute sampling period. Waist-level ports lead to the slit sampler and particle counter at left. Ports in the walls of the "telephone booth" allowed for sampling of the air surrounding the individual being studied.* Through one port, the air was drawn through a 7/8 in. Tygon tubing 5 ft. in length. A slit sampler capable of pla- ting out airborne bacteria was used to determine the colony-forming units pres- ent in two cubic feet of air aspirated from the booth during a ten-minute sampling period. A clockwork mechanism turned the sheep's blood agar culture r *A grant from the Florida Orthopaedic Society was of assistance in funding the "telephone booth."
159 Techniques to Reduce Operating Room Bacteria plate. The culture plates were read at 48 and 72 hours. A second port in the booth led off to a Climet photoelectric particle counter capable of counting particles larger than 0.5 micra in diameter. The counts were printed on a digi- tal printout tape for each minute of testing. The procedure used for the study required cleaning the walls of the booth and flushing the air with high velocity, High Efficiency Particulate Air (HEPA) filtered air. The experimental subject was then placed in the booth. The sterile air flush continued for an additional two minutes. The airflow was turned off and the subject was evaluated for the first sampling while standing quietly. The booth was then flushed with HEPA filtered air for two minutes and the subject was evaluated for a second ten-minute period while coughing and talking. The air in the booth was again flushed for a two-minute period. During a third ten-minute sampling period, the subject was required to walk in place. Each individual was first studied for three ten-minute sampling periods while wearing an ordinary surgeon's scrub suit without cap or mask. Hands and arms were not covered. During the second series of three ten-minute sampling periods, the experimental subject was dressed in a surgeon's gown with paper hood, paper mask, shoe covers, and surgical gloves. The surgeon's gown was a conventional loose-weave linen fabric. During a third series of three ten-minute sampling periods, an aspiration suit was worn by the subject. In addition to removing air from around the individual's head, neck and face, the hood provided complete coverage of the head, neck, face, shoulders and upper torso. The fabric of this aspiration suit was a loose-weave linen similar to the common surgical gown. Surgical gloves and shoe covers were worn with the aspiration suit. One group of four individuals tested an aspiration suit of similar design but made of tightly woven fabric (Bar Bac). In this fashion three activities and three types of apparel were studied. RESULTS The fourteen members of the Orthopaedic Surgical Department of the University of Florida were studied for their bacterial- and particle-shedding
160 Howard P. Hogshead characteristics. Additionally, four female operating room technicians from another hospital were evaluated. Sampling periods lasted at least two hours. The temperature ranged from 70 to 80°F and the humidity averaged 70 percent. The particle counts yielded no correlation with the bacteriological counts. At best, the particle counts indicated the amount of linen within the booth during the sampling. These data are omitted from the report. Data obtained from the slit sampler agar culture plates on the 18 experimental subjects are shown in (Fig. 2). Standing quietly Illllllllllllllllll Scrub suit â¢â¢â¢â¢ Surgeon's gown '.â¢:â¢>:â¢.â¢:â¢â¢:-::â¢:â¢:â¢â¢â¢:â¢:⢠Aspiration suit I 1 8 10 11 12 13 14 15 16 17 18 UNIVERSITY OF FLORIDA DEPARTMENT OF ORTHOPEDICS 1 - 14 FLORIDA HOSPITAL TECHNICIANS 15 -18 Fig. 2. Standing quietly while dressed in scrub suit, sur- geon's gown, and aspiration suit. Individuals retain the same number throughout the graphs. Watch Number 5 and Number 10. While standing quietly, most subjects lofted only a few viable particles into the air during the ten-minute period. Only small differences were evident be- tween the sampling periods when the individual was dressed in a scrub suit, a r
161 Techniques to Reduce Operating Room Bacteria surgeon's gown or an aspiration suit. It is noteworthy that individuals Number 5 and Number 10 shed bacteria at rates in excess of their colleagues. (The num- ber designation in the graphs remains unchanged throughout these figures.) With a few exceptions the scrub suit colony count is highest, the surgeon's gown colony count is lower, and the aspiration suit bacterial colony count is the lowest of the three. Talking and coughing (Fig. 3) produced a slight increase in colony counts for all individuals. Again individuals Number 5 and Number 10 are noted to shed a greater number of bacteria than their colleagues. Talking and coughing Scrub suit IHBH Surgeon's gown Xv>Xv-H-Xv: Aspiration suit 1 8 10 11 12 13 14 15 16 17 18 UNIVERSITY OF FLORIDA DEPARTMENT OF ORTHOPEDICS 1 -14 FLORIDA HOSPITAL TECHNICIANS 15-18 Fig. 3. Talking and coughing while wearing three different types of apparel. Walking in place (Fig. 4) resulted in dramatic increases in the number of bacteria scattered into the air. Again individuals Number 5 and Number 10 are noted to be extremely productive of active colony-forming units. In both cases a coagulase-positive Staphylococcus aureus accounted for the majority of these organisms.
162 Howard P. Hogshead Walking in place | Illllllllllllllllll Scrub suit â¢â¢â¢â¢ Surgeonâ¢s gown Aspiration suit t 17 18 UNIVERSITY OF FLORIDA DEPARTMENT OF ORTHOPEDICS 1 -14 FLORIDA HOSPITAL TECHNICIANS 15-18 Fig. 4. Walking in place while wearing three different types of apparel. The vertical scale has been broken to accommodate the data. Number 5 and Number 10 are Staphylococcus aureus shedders. To provide further comparisons of the three types of activity, the same 18 individuals are shown in (Fig. 5) while all were attired in the surgeon's gown. The effect of activity in lofting bacteria into the air can be readily appreciated. The data are massed as shown in (Fig. 6). Comparisons were made between the square of the means for the 14 male orthopaedic surgeons in three styles of dress and in three activities. The four Florida Hospital technicians' data are treated separately but relating to the same factors (Fig. 7). There is no statistical difference between the average of the means for either group when standing quietly compared to talking. There is no significant difference between the University of Florida surgeons and the Florida Hospital technicians while f
163 I Techniques to Reduce Operating Room Bacteria Surgeonâ¢s gown Illllllllllllllllll Walking <n place â¢â¢â¢â¢ Standing quietly :â¢>â¢:->.â¢:â¢:â¢:â¢>:â¢:â¢>:â¢> Talking and coughing 8 123456 UNIVERSITY OF FLORIDA DEPARTMENT OF ORTHOPEDICS 1 -14 FLORIDA HOSPITAL TECHNICIANS 15-18 10 11 12 13 14 15 16 17 18 Fig. 5. Three different types of activity while wearing the same garment--a surgeon's gown. The remarkable effect of body movement on lofting airborne bacteria is evident. The great dif- ference between the orthopaedic surgeons and the surgical tech- nicians is not fully explained. standing quietly or while talking. A statistically significant difference at the 0.05 level of significance exists between the first two types of activity and walking in place. There is a statistically significant difference between the aspiration suit worn by the University of Florida surgeons and the suit worn by the Florida Hospital technicians. Unfortunately, the effectiveness of the two suits was not contrasted on the same individuals.
164 Howard P. Hogshead Total Colony Count University of Florida Surgeons (Means + Standard Error of Mean) Standing Quietly Coughing & Talking Walking in Place Scrub Suit X = 13 + 9 X = 30 + 10 X = 496 + 46 Surgeon's Gown X = 10 + 3 X = !5 ± 5 X = 286 + 56 Aspiration Suit X = 8 ± 3 X = 9 + 3 X = 155 + 78 Fig. 6. Comparisons of colony counts for fourteen male University of Florida Surgeons during three types of standard activities while wearing three types of apparel. Differences between standing quietly and talking and coughing are not sig- nificant. Differences between either of those activities and walking in place are significant at p=0.05 level. Colony count decreases as more skin is covered, but this is more evident when activity is greatest. Values in a simple scrub suit are lower when standing quietly than those for the aspiration suit when walking. Total Colony Count Florida Hospital Technicians (Means + Standard Error of Mean) Standing Quietly Coughing & Talking Walking in Place Scrub Suit X = 2 + 2 V â 5 + 4 X = 40 + 7 Surgeon's Gown X = 0 ± ° x = 1 + 0.8 X = 12 + 5 Aspiration Suit X = 1 + 0.7 X = 0 + 0 X = 1 + 0 .5 Fig. 7. Comparisons of colony counts for four female Florida Hospital Operating Room technicians during three types of standard activity while wearing three different types of apparel. Differences between standing quietly and talking and coughing are not statistically significant. Differences be- tween either of these activities and walking in place are highly significant. Colony counts decrease as more skin is covered. A tightly woven fabric was used in the aspiration suit used by these individuals. It appears to be highly effective and pro- bably better than the loosely woven garment used by the men. Strict comparisons are not possible, however, because of dif- ferent shedding characteristics.
165 Techniques to Reduce Operating Room Bacteria DISCUSSION The differences in shed bacteria of the four female operating room technicians is noticeable in contrast to the larger, more vigorous male sur- geons. Reasons for these differences are not clear at the present time. It is observed that the girls were small and marched in place very gently. None were nasopharyngeal carriers of Staphylococcus aureus. It would appear that the more impervious fabric of the aspiration suit worn by these individuals is more effective than the loosely woven aspiration suit worn by the men in this study. Since these four women generated fewer bacteria than the men, a valid comparison of the effectiveness of the suit is unfortunately not possible. The result of activity is more important than the mode of dress. Indeed, if all individuals were to stand quietly aspiration suits would be unnecessary. The presence of a vastly increased rate of shedding bacteria of individuals Number 5 and Number 10 marks them as "shedders". In both cases the predominant organism shed was the pathogenic Staphylococcus aureus. In- dividual Number 10 was noted after completion of this study to have been a long-term sufferer from atopic eczema. Both Number 5 and Number 10 were nasal pharyngeal carriers of Staphylococcus aureus. Number 10 was able to reduce his shedding with the use of a steroid skin cream. Additional studies of the phenomenon of shedding were carried out on several suspect individuals in order to develop a simple method for the detection of the presence of a shedder. In a quiet office, blood agar settle plates were placed at 6, 12, and 18 inches in front of and to the side of the individual's exposed face and neck. Control values in this setting were zero or one colony per ten-minute sampling period. The shedders produced between four and eight Staphylococcal colonies per plate per ten-minute sampling period. Six or eight pathogenic colonies per plate in a ten-minute period may indicate a surgical risk. Using this method, an individual with a nasal furuncle was also found to shed Staphylococci in greatly increased numbers. The bacterial counts from this particular individual returned to two or three colonies per plate per ten- minute sampling after the furuncle was resolved. This lesion should prompt further study and probably should curb participation in surgery having a high risk of infection.
166 Howard P. Hogshead CONCLUSIONS Applying these studies to the average operating room situation, the following suggestions are offered: 1. The degree of physical activity in the operating room is the most important single factor affecting the number of bacteria shed into the air. Movement must be minimized. 2. Talking and coughing cause only a slight increase in airborne bacteria in comparison to quiet standing. 3. The number of airborne bacteria is decreased proportionately to the degree to which the skin and hair is covered by garments. 4. In a limited number, the use of an aspiration suit made of tightly woven fabric appears to be superior to a similar aspiration suit of a loosely woven fabric. 5. Particle counting is irrelevant to the bacterial count in these studies. The particle count is mainly a reflection of the amount and type of linen worn ,as well as the physical activity. Particle counters are not helpful in operating room monitoring. 6. A simple method for detection of shedders is described. Guide- lines for the rational management of shedders have not been developed. How- ever, the shedder is undeniably a threat to the surgical wound. 7. It is suspected that many shedders will be found to have skin diseases or other skin lesions which cause or which magnify their shedding characteristics. Individuals with such conditions should be examined and special precautions taken to protect the surgical wound. 8. Screening of the operating and anesthesia teams may be of particular importance in surgical procedures having a substantial risk of wound infections.
167 OPERATING ROOM AIR AS A SOURCE OF WOUND CONTAMINATION AND INFECTION Ruth B. Kundsin* The number of airborne microorganisms in the operating room during a procedure is the equilibrium reached between the addition of microorganisms and their removal. Organisms shed by occupants, together with those lofted by ac- tivity, are1 the means by which they are added; their removal is accomplished by ventilation or its sanitary equivalent, ultraviolet irradiation. The three essential parameters for evaluating the microbial aerody- namics in the operating room are volumetric air samples, fallout plates, and cultures of microbial accumulation. Volumetric air samples and fallout plates indicate miroorganisms in transit. From these, viable particle size and set- tling velocity can be calculated. Microorganisms on horizontal surfaces repre- sent the accumulated fallout of large viable particles. Techniques for the detection of microorganisms in these states of suspension differ. The standards that our laboratory has set up and has found useful in evaluating the operating room environment are shown in Table 1. During a pro- cedure a consistent rise in all counts of viable particles occurs. Counts are lower than standards at the start but usually increase and surpass these stan- dards at the conclusion of the procedure. During a one-hour procedure involving replacement of a pacemaker, volumetric air counts rose from two to nine per cubic foot. Fallout of organisms increased from two to six per square foot per minute. Surface impression plates taken before and after the procedure showed increases in six of nine surfaces sampled- Table 1. Operating Room Standards for Viable Particles 3 Volumetric Air Samples 5 vp/ft 2 Settling Plates 5 vp/ft /min Horizontal Surfaces <25 vp/Rodac Plate The logical question at this point is how much of a hazard does this present to the patient? Assuming the operating room intake is within our *Director, Surgical Bacteriology Laboratory, Environmental Sepsis Laboratory, Peter Bent Brigham Hospital, Boston, Massachusetts
168 Operating Room Air hygienic limits and all surfaces have been wiped with an effective germicide, definite predictions can be made. The original work from our laboratory on airborne infection during surgery was done in collaboration with Dr. Carl W. Walter (4). It was based on a study of 250 surgical procedures of all types. Nasopharyngeal and skin cultures were taken preoperatively and postoperatively of all patients. Cultures of the microbiology of the operating room were done. Cultures of the wound were taken on opening and closing. The team furnished masks and gloves for culture following each procedure. Postoperative follow-up of patients was done by a nurse member of the team. This comprehensive microbiology made it possible to ascertain the source of infecting organisms. Staphylococcus aureus was recovered on settling plates on the instrument table in 48 percent of procedures, on settling plates on the floor in 58 percent of procedures, and in air samples (15 cubic feet) in 36 percent of procedures. In 69 percent of procedures, £3. aureus was recovered from some environmental culture (Table 2). One technician in the periphery of the operat- ing room was found to be a disseminating carrier and his S_. aureus had a distinc- tive phage pattern. This made tracing of his JJ. aureus possible during the pro- cedures when he was present. Table 2. Staphylococcus Aureus Recovered From Environment in 250 Surgical Procedures OR Sites Cultured Procedures* Instrument table - settling places 48% Floor - settling plates 58% Air samples (15 cubic feet) 36% *In 69% of procedures, S_. aureus was recovered His particular strain was found on settling plates in 33 percent of procedures, on the floor in 25 percent of procedures, and in air samples (15 cu- bic feet) in 11 percent of procedures. In 49 percent of procedures, his S.. aureus was recovered from the environment (Table 3).
169 Ruth B. Kundsin In the 250 procedures studied, 20 postoperative wound infections were found (Table 4). Sixteen of the 250 (6.0 percent) were endogenous and four of the 250 (2 percent) were exogenous. Staphylococcus aureus from the wound could be traced to three individuals in the operating room in the exogenous wound in- fections. Two infections were traced to two surgeon members of the team. Two of the infections were traced to the disseminating carrier in the periphery of the room whose £[. aureus was recovered in the environment (Table 3). This is an infections rate of one percent for the 169 procedures during which he was present. Table 3. Staphylococcus Aureus Carrier (52/52A/79/42D/7/53/54/70/73/80/81/82/47C) Present in 169 Procedures Identical S. aureus found: OR SITES CULTURED Instrument table - settling plates Floor - settling plates Air sample (15 cubic feet) *In 49% of procedures, £. aureus of carrier recovered from the environment. PROCEDURES* 33% 25% 11% Table 4. Postoperative Wound Infections 250 Cases 6.4% 1.6% 8.0% (16/250) ( 4/250) (20/250) Endogenous Exogenous Total A more recent experience with an anesthesiologist in our hospital indicated an incidence of four (7.3 percent) infections in the 55 procedures which he attended. In another hospital, where we were consultants, an incidence of 13 infections in 183 procedures (7.1 percent) were found to be due to a spe- cific T-type beta-hemolytic group A streptococcus from an anesthesiologist with
170 Operating Room Air a perianal lesion from which the same streptotoccus was isolated (1). A summary of these three carriers and their rate of infectivity is shown in Table 5. Based on these observations, the hazard to each patient can be mathematically predicted depending on the number of disseminating carriers in the operating room. Table 5. The Disseminating Carrier in the Operating Room NUMBER OF PROCEDURES NUMBER OF DOCUMENTED POST-OP INFECTIONS ORGANISM CARRIER S. aureus- S. aureus , 8 -Hemolytic Technician Anesthesiologist 169 55 2 (1.2%) 4 (7.3%) Streptococcus Anesthesiologist 183 13 (7.1%) Group A 252/52A/79/42D/7/70/73/80/82/47C 329/52/83A/81/80/47C/82 and 81/80/47C/82 T Type 12 If the probability of not becoming infected for each disseminating carrier is known, the probability that any patient will be infected is equal to p (I) = 1 - q (GI) . q (c2) ... q (cj Where: P (I ) * probability that a patient will be infected. p (c.) = probability that the i carrier is an infector. 1 i_u q (c.) = probability that the i carrier is not an infector. q (Gi) = 1 - p (cj. n. e-
171 Ruth B. Kundsin Assuming all three carriers described in Table 5 were in an operating room together during a procedure, the probability of the patient's being infec- ted, p (I ), can be calculated as follows: p (I ) = 1 - (.93 x .93 x .99) p ( I ) = .144 or 14,4Z The airborne component of postoperative wound infections is not a fixed rate but varies from hospital to hospital, from operating room to operat- ing room, and from surgical team to surgical team. The rate is directly pro- portioned to the number of disseminating carriers in the operating room. The disseminating carrier can be anywhere in the room. He does not necessarily have to be a member of the surgical team. He can be a casual visitor near the operative site, a transient technician in the periphery of the room, but each disseminating carrier in the room contributes to the cumulative risk incurred by the patient. The emphasis on air supply to the surgical suite is important but not the critical factor. When the air supply plays an etiologic role, exogenous microorganisms are involved. The critical factor in our experience is the com- mensal flora of the people who are present. They can be scrubbed, masked, capped, gowned, and appropriately shod, but the disseminating carrier will still have the capability to project his microorganisms into space. Definition of the problem simultaneously suggests the solution. Since the source of airborne postoperative wound infection is the disseminating carrier whether recognized or unrecognized, a technique which can intercept this micro- bial dissemination must be devised. Only those techniques should be considered which can reliably and reproducibly accomplish this with minimum discomfort to personnel. One solution which has had support in the past is ultraviolet irradi- ation (2). Ultraviolet lamps have been installed in four operating rooms at the 2 Peter Bent Brigham Hospital for irradiation during surgery. With 25-30 yw/cm /sec
172 Operating Room Air at the level of the operating table, a significant reduction in airborne microorganisms on surfaces, fallout plates and volumetric air samples can be de- monstrated. Clinical evaluation of results of orthopedic surgery to date with such irradiation are encouraging (3). Because the airborne component of postoperative wound infections varies with each hospital and is usually a small proportion of total infections, proof of successful control of airborne infections will of necessity involve a large number of patients before statistical significance can be attained. However, objective scientific evaluation of the problem can enable those interested to make a good educated guess as to the methods which will lead to successful eradication of aerially transmitted microorganisms. ACKNOWLEDGMENT The assistance of Marjorie Drolette, Ph.D., in statistical evaluation is gratefully acknowledged. REFERENCES 1. Gryska, D.F., and A. E. O'Dea, Postoperative streptococcal wound infections: the anatomy of an epidemic. JAMA 213:1189-1191, 1970. 2. Hart, D., R- W. Postlethwait, I. W. Brown, Jr., W. W. Smith, and P. A Johnson, Postoperative wound infections: a further report on ultraviolet irradiation with comments on the recent (1964) National Research Council cooperative study report. Ann. Surg. 167:5:728-743, 1968. 3. Lowel1, J. D., Ultraviolet light: its beneficial effects on the operating room environment - or - there is another way. Presented American Academy of Orthopaedic Surgeons, January 1974, Dallas, Texas. 4. Walter, C. W., R. B. Kundsin, and M. M. Brubaker, The incidence of airborne wound infection during operation. JAMA 186:903-911, 1963.
173 WHAT IS THE HARD EVIDENCE ON THE ROLE OF AIRBORNE BACTERIA IN WOUND INFECTION? Harold Laufman* The term "hard data" is employed by scientists as jargon for data which are irrefutable in the light of present-day knowledge. Hard proof is certain proof. It implies documentation which is convincing, and describes valid observations from which conclusions can be drawn beyond reasonable doubt. All observations and data require interpretation. The validity of a conclusion is dependent upon the soundness of logic and absence of prejudice at each stage of the three-part deductive process: the premise, the method, and the conclusion. Even when all three are sound, the interpretive logic may be faulty. The same data based on factual material may have no meaning to one observer but significant meaning to another. Or, it may be interpreted accord- ing to an individual observer's special interests, experiential and educational background, and a host of other factors that set one individual apart from another. The clear zone around a penicillin mold growth had been observed by many people, but it took the deductive processes of Alexander Fleming's mind to read practical significance into this observation. In 1882, when Robert Koch cultivated the tubercle bacillus, he presented his results to the Berlin Physiological Society in a paper which in- cluded "Koch's postulates." Koch knew that it was insufficient to assume that simple cause-and-effeet observations were enough to implicate a given micro- organism as cause of a given disease. His fourth postulate states that the organism recovered from the diseased part must be the same as the one intro- duced to cause it. With this much background, let us look at the evidence at hand on the role of airborne bacteria in wound infection. I think everyone will agree with certain facts which have emerged from repetitive observations and hard data. These are: 1) bacteria may be airborne; 2) highspeed, clean directional or moderately turbulent airflows are capable of diminishing the number of airborne bacterial particles in a confined space. *Institute for Surgical Studies, a division of the Department of Surgery, Montefiore Hospital and Medical Center, New York City.
174 Harold Laufman An apriori deduction from these factual observations â that because microorganisms may be carried by airborne particles, and because microorganisms may cause wound infection, and because the particles may be diminished in num- ber significantly by high-speed air movement â that post hoc ergo propter hoc high-speed air blowing should diminish the incidence of wound infection. How- ever, application of elementary logic tells us that this conclusion may be a non sequitur, and not a legitimate conclusion at all. It could be a non sequitur 1) if hard data show that high-speed clean-air movement does not pre- vent wound infection; 2) if airborne bacteria do not match those recovered from an infected wound, or 3) if infection can be prevented by means other than high-speed clean-air movement, such as, for example, containment of dispersal of airborne bioparticles, exclusion of carriers, correction of engineering de- fects in a low-speed, turbulent air system, use of antibiotics, special surgical and hygienic technics, and so on. It would serve little useful purpose at this point to correct an al- most hopelessly bogged-down misuse of terminology, but in the interests of the search for hard facts, it should at least be mentioned here that the term "laminar flow" is constantly misused. Directional airflow, as it is employed in operding rooms is at best less turbulent than nondirectional flow. But be- cause of the inevitability of obstacles and movement in the operative space, the flow is certainly not laminar. Misusage of the term extends to the tendency to refer to all ceiling-vent systems as laminar. Two reports, one by Walter, Kundsin and Brubaker (3) and the other by Gage and associates (4), provide us with data which trace the infection of tissue directly to pathogens circulated into the operating room air. In the case of the Walter-Kundsin report, a carrier-disseminator (shedder) entered the periphery of an operating room. The air distributors were located on the same wall as the exhaust registers. Thus, all incoming air had to sweep the room before being exhausted on the same side as it entered. The shedder was dressed and shod in the conventional way, but wore a cap, not a hood, and no gloves or shoe covers. The bacteria shed by this person matched £ specifically by phage type those found in the wounds of a series of patients.
175 What is the Hard Evidence We do not know whether the bacteria were deposited directly into the wound from the air, or came to rest on instruments, sponges, or gloves before being deposi- ted into the wound, but the source and the final destination of the bacteria are irrefutable. In the case of the Gage report, a contaminant was brought into the operating room by a faulty air handling system which delivered undiluted Asper- gillus spores directly into the room from a poorly-drained moss-covered roof covered with pigeon excreta. A virtually pure culture of Aspergillus was iso- lated from the pigeon droppings, from the air ducts, from the air handling machinery, and from the thrombi which adhered to the implanted intracardiac valves of five patients. Interestingly, the skin wounds of these patients and those of other patients operated upon in the same operating room (cholecystec- tomies, herniorrhapies, etc.) had the same low infection rate as that of pa- tients operated upon in other operating rooms of the same suite. This rate was well below published infection rates in the ultraviolet study published in 1964 (5). Reasons for this phenomenon can only be speculated upon, but one speculation is that the strong suction employed in the area of the open heart was concentrated into the cardiac chambers, acting much like a slit-sampling device in concentrating the airborne bacteria in one place. Other speculations include the possibility that different tissues and anatomical locations have different levels of susceptibitlity to any given pathogen; that the infectious agent was carried into the wound by gloves, instruments, or sponges, while the sides of the wound were protected by pads or towels; or that the foreign body itself (the valve) may have become contaminated before insertion. The Walter-Kundsin report and the Gage report are examples of hard data gathering, one demonstrating the effects of an incompletely covered shedder and the other, the effects of an erroneously installed and defective air handling system, both definitely supporting the premise that airborne bac- teria are responsible for wound or other tissue infection. However, reference to these observations is misapplied by authors who use them correctly to sup- port the thesis that wound infections can be caused by airborne bacteria, but at the same time report cases in which the airborne bacteria do not match (if, indeed, they were cultured) those found in the infected wound. Such reports, in
176 Harold Laufman addition to IBing hard data to support a questionable conclusion are, of course, guilty of not fulfilling Koch's fourth postulate. Reports such as that of Amstutz (6) trace wound infections in hip- replacement operations to the urinary tract, and not to airborne bacteria, either in a conventional operating room where four deep infections occurred, or in a directional airflow room where one deep and six superficial infections occurred. Gram-negative rods are rarely, if ever, transmitted by air. Also, the bacteriology of infected hip-replacement wounds is far from uniform. Reports in which claim is made that the wound bacteria match those of the air, cover a wide spectrum of bacteria, but usually focus upon either Staph. epidermidis or Staph. aureus, both of which are capable of being air- borne. The conclusion, therefore, is reached that the infection reached the wound from the air, either directly or indirectly by way of an instrument, sponge, glove, or some other fomite. But, as we have seen, such a conclusion may be a non sequitur. Although it is clear that bacteria can be airborne, their arrival at the wound site via this route must remain conjectural with- out measurement of cultures from other sources (e.g., the surgical team's hands). When bacterial counts over a wound or instrument table are at vir- tual zero and the patient develops a Staph. epidermidis or Staph. aureus in- fection, it is less likely that the microbe got to the wound via the air than it is that it got there by direct implantation, either via a tear in the sur- geon's glove or from the patient's own skin. Hard proof would require cultures of objects other than the air, such as instruments, gloves, sponges, surgical team's hands, and so on. We lack certain basic pieces of information or hard data which appear to be necessary before specific recommendations can be made responsibly. With respect to the air, Lidwell (7) has asked, "How clean should operating-room air be for surgery?" Paraphrased in the present context, "How clean should operat- ing-room air be to provide 'clean-enough' air for clean-refined surgery?" We have no definite answer based upon hard data for a number of rea- sons. 1) We have not yet defined clean-enough air. Or, as Beck has put it,
177 What is the Hard Evidence we need standards for biologically clean air, since federal standards for in- dustrial clean rooms are all we have, and these are certainly not relevant to operating rooms- 2) We have no standard protocol for investigators to follow in describing the conventional operating rooms which they invariably use for comparison with directional-airflow rooms. 3) We have no standards for the tests used in monitoring the bioparticles in high-flow rooms. As a result, in- vestigators are measuring different parameters by different methods and attempt- ing to compare such findings. 4) There is little unanimity of opinion on the specifications of the filter necessary to provide clean air into an operating room at reasonable cost. 5) We have had little reliable information, until recently, about comparative permeability of various woven and nonwoven materials depended upon to serve as barrier materials during operations. The new information about barrier materials (11) indicates, contrary to promotional claims, that not all nonwoven materials used in the manufacture of surgical gowns and drapes are truly barrier materials under conditions of operating room use. Moreover, not all woven materials are permeable to moist contamination. Strike-through of contamination may be important as a means of disseminating bioparticles into the air with drying, or as a means of delivering the wearer's bacteria to the outside for direct contact implantation into the surgical wound or onto fomites. Therefore, it is of some importance to know which materials can be depended upon as barriers under specific circumstances. Employing a test designed to simulate operating conditions of stress such as stretch, and changes in permeability to Serratia marcescens were mea- sured against time. Under conditions of the test, waterproofed Pima tight- woven cotton cloth was found to be impermeable to moist bacteria through up to 75 washing and sterilizing cycles. On the other hand, untreated Pima cloth and ordinary cotton of the kind used in surgical gowns permitted bac- terial penetration almost immediately, as did the stockinette cufflets. Among the nonwoven disposable gown and drape materials, only spread tow plastic film composite remained impermeable through all tests, while scrim- reinforced tissue, scrim-reinforced embossed tissue, spunbonded polyethylene
178 Harold Laufman nonwoven fabric, and spunlace nonwoven fabric were considered satisfactory, up to a point, but became somewhat permeable within a half hour. Other nonwoven materials allowed wet bacterial strike-through within five minutes. Choice of barrier materials for purposes of containment of the surgical team, is, there- fore of more than academic interest, and certainly may affect results of envi- ronment studies in the operating room. Barrier effects of materials cannot necessarily be depended upon as a constant (Tables 1 and 2). ⢠Table 1. Unopposed Weight-Support Test (2 kg Weight) Bacterial Penetration = + Material Wovens 5 min 15 min 30 min Tight weave Pima, Quarpel 0 0 0 treated 0 0 0 0 0 0 0 0 0 0 0 0 Tight weave Pima, Quarpel 0 0 0 treated, 55 cyclings 0 0 0 Tight weave Pima, Quarpel 0 0 0 treated, 75 cyclings 0 0 + Tight weave Pima, Quarpel 0 0 + treated, 100 cyclings 0 + + Tight weave Pima, untreated + + + Linen, new + + + Linen, cycled twice + + + (From Laufman, H., W. W. Eudy, A. M. Vandernoot, D. Liu, and C. A. Harris, Strike-through of moist contamination by woven and nonwoven surgical materials. To be published). c
179 Table 2. Unopposed Weight-Support Test (2 kg Weight) Bacterial Penetration = + Nonwovens Material 5 min. 15 min. 30 min. Spread tow plastic 0 0 0 film composite 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Wet-laid nonwoven fabric 0 0 0 laminated to polyethylene 0 0 0 film 0 0 0 0 0 0 0 0 0 Scrim-reinforced tissue 0 0 0 0 + + 0 0 0 0 0 0 0 0 + Scrim-reinforced embossed 0 0 0 tissue 0 0 0 0 + + 0 + + 0 0 0 Spunbonded polyethylene 0 0 0 fabric 0 0 0 0 0 0 0 0 + 0 + + Spunlace nonwoven fabric 0 0 0 0 + + 0 0 0 0 0 + 0 0 0 Wet-laid nonwoven fabric 0 + + 0 0 0 0 + + 0 0 + 0 0 + Fiber reinforced tissue * I t 0 0 + (From Laufman, H., W. W. Eudy, A. M. Vandernoot, D. Liu, and C. A. Harris, Strike-through of moist contamination by woven and nonwoven surgical materials. To be published).
180 Harold Laufman Charnley and Eftekhar (10) have shown that waterproofed tight-weave coverall uniforms, used with helmets, did reduce particulate shedding signifi- cantly even in a directional air chamber. The use of vacuum suction of air from under such coverall uniforms adds materially to the comfort of the wearer by enhancing evaporation, and also prevents perspiration from soaking areas of the uniform and possibly striking through to the sterile surface. Bernard and associates (14) recently showed that one-piece shoe-cover pants similar in design to fisherman's waders made of tightly woven cotton worn by the surgical team together with hoods, masks, and gloves effectively packaged the surgical team and produced an airborne bioparticle count of 0 to 2.6 at the wound site in a fully occupied conventionally ventilated room. These levels are equivalent to those found in high-flow directional air units. It has not been settled at what periods overkill sets in through superimposition of one effective controlling system upon another. The same question must be raised relative to the use of prophylactic and intraoperative antibiotics in conjunction with barrier containment of the surgical team, high- flow directional air systems, ultraviolet light, or other control systems. It is assumed that all common aseptic precautions are universally taken in all operations by members of the surgical team including adequate body hygiene, adequate hand scrubbing, and full hood to cover all hair. Special precautions include wearing two pairs of gloves, strict traffic control, controlled ambient conditions such as temperature and humidity, and wearing of special garb. All these factors, as well as others related to the physical environment must be known in order to arrive at a meaningful conclusion from which valid deductions pertinent to prevention and treatment of wound infections can be made. Finally, I wish to stand up â not in defense of a stand, but in rebuttal â to those who would proceed to advise on how to reduce the incidence of wound infection in clean-refined operations without hard evidence on which to base their advice. We skeptics are not against clean air, nor are we against lowering the incidence of wound infection. We submit that we cannot know the degree to which various possible influences participate in the causation of
181 What is the Hard Evidence wound infections until we have more hard data. Until we do, we can only ex- pect to prevent these catastrophic complications either by imprecise, unquanti- fied shotgun methods, or by missing the target altogether through not being able to identify it. REFERENCES 1. Altemeier, W. A., W. R. Culbertson, and R. P. Hummel, Surgical considerations of endogenous infections â sources, types, and methods of control. Surg. Clin. N Amer., 48:227, 1968. 2. Bernard, H., and C. Wirtz, Turbulent air flow, surgical apparel and total hip replacement. Read at Subcommittee on Barrier Materials, Committee on Operating Room Environment, American College of Surgeons, 1974. 3. Bernard, H. R., R. Speers, Jr., F. O'Grady, and R. A. Shooter, Reduction of dissemination of skin bacteria by modification of operating-room clothing and by ultra-violet irradiation. Lancet 2:458, 1965. 4. Blowers, R., and B. Crew, Ventilation of operating theatres. J. Hygiene, 58:427-448, 1960. 5. Burke, J. F., Identification of the sources of staphylococci contaminating the surgical wound during operation. Ann. Surg., 158:898, 1963. 6. Charnley, J., and N. Eftekhar, Penetration of gown material by organisms from the surgeon's body. Lancet 1:172, 1969. 7. Cohen, F. S., F. R. Feketz, Jr., and L. E. Cluff, Studies in the epidemio- logy of staphylococcal infection VI. Infections in the surgical patient. Ann. Surg. 159:321, 1964. 8. Elek, S. D., Staphylococcus Pyogenes and Its Relation to Disease, 1959, Livingston, Ltd., London. 9. Fox, D. G., A study of Application of Laminar Flow Ventilation to Operating Rooms. Public Health Monograph No. 78, US Department of Health, Education and Welfare. Washington, D.C. 10. Gage, A. A., D. C. Dean, G. Schimert, and N. Minsley, Asperigillus infection after cardiac surgery. Arch. Surg., 101:284, 1970. 11. Irvine, R., B. L. Johnson, Jr., and H. Amstutz, The relationship of genito- urinary tract procedures to deep sepsis in total hip replacements. Surg. Gynec. & Obst. (To be published.) * W 12. Laufman, H., W. W. Eudy, A. M. Vandernoot, D. Liu, and C. A. Harris, Strike- through of moist contamination by woven and nonwoven surgical materials. Read before the Subcommittee on Barrier Materials of the Committee on Opera- ting Room Environment, American College of Surgeons, 1974.
i 182 Harold Laufman 13. Lidwell, 0. M., The ventilation of surgical operating rooms: the last 50 years. Chapter in Airborne Transmission and Airborne Infection, edited by Hers and Winkler, 1973, Utrecht, The Netherlands, Oosthoek Publishing Co., pp. 561-568. 14. Moylan, J. A., E.,Balish, and J. Chan, Intraoperative Bacterial Transmission Surgical Forum, American College of Surgeons, 25:170, 1974. 15. Walter, C» W., R. B. Kundsin, and M. M. Brubaker, The incidence of airborne infection during operation. JAMA, 186:908, 1963. 16. Wells, W. F., Airborne Contagion and Air Hygiene. Harvard University Press, Cambridge, 1955.
183 EVALUATION OF METHODS FOR THE CONTROL OF SURGICAL SEPSIS 0. M. Lidwell* Infection of the wound and the development of sepsis following a surgical operation necessarily involves these conditions: 1. The presence of a source of viable microorganisms somewhere within the circle of possible contact. 2. Dispersal of these microorganisms from their point of origin. 3. Transfer of a sufficient number of microorganisms to a sensitive site. 4. Proliferation of the organisms within the wound. Prevention or reduction in the risk of sepsis can be achieved by measures which interfere with any one of the above steps. For example, some sources may be eliminated by the exclusion or treatment of carriers among the staff, or by more effective decontamination of the patient. Dispersal can be reduced by improvements in protective clothing or by skin treatments. Trans- fer to the patient by the airborne route can be minimized by the use of more effective ventilation systems. Reducing the risk of contact transfer involves, among other things, the protection of instruments, swabs, and dressings from contamination before or while they are used. Plastic isolators interpose a barrier to the deposition of bacteria from the environment to the wound. Some form of ultraviolet irradiation may kill organisms which have reached the tis- sue surface. Multiplication within the wound may be inhibited by local ap- plication of antiseptics or by local or systemic antibiotics. There are, therefore, not only many ways by which a dose of infection may build up in a surgical wound, but an even larger variety of procedures which might be applied to reduce it. As an additional complication, it is ra- ther unlikely that the relationship between the size of the bacterial dose re- ceived by the wound and the risk of infection or sepsis will be anything like a proportional one. Very wide variations in the susceptibility of the popula- tion at risk will lead to a situation in which large reductions in dose lead only to a small reduction in the risk. Figure 1 may serve as an illustration * Central Public Health Laboratory, Colindale Avenue, London, England.
184 0. M. Lidwell to make even more apparent a number of obvious consequences. If the dose to the wound via contact and/or autogenous routes is fairly high, e.g., the upper- most horizontal line in Figure 1, then only if the air is very dirty will im- proved ventilation or reduction in dispersal lead to any reduction in the dose to the wound with a possible fall in the risk of sepsis. If the infectious SOURCES OF WOUND INFECTION DOSE TO WOUND AUTOGENOUS II CONTACT II AIR: DIRTY CLEAN ULTRACLEAN Fig. 1.
185 Control of Surgical Sepsis dose is delivered simultaneously by the contact route and an autogenous source, control methods applied to either of these separately will have little effect. If, however, substantial improvements can be made in relation to both these routes, i.e., the dose transmitted falls to the lower two horizontal lines, then measures to provide a cleaner air environment may result in improved clin- ical results. The advantage obtained will, however, be limited by the highest level of dose resulting from any other route. Only if this is reduced still further will the attainment of the cleanest atmospheres bring clinical benefit. A fundamental difficulty in reaching a decision as to which preventive measures should be applied, and which would merit clinical evaluation, lies in our almost complete ignorance as to the relative dose levels arising in the different possible ways. BACTERIOLOGICAL EVALUATION The first step towards evaluating a control procedure is to establish that it is effective in its purpose, e.g., that a suit of protective clothing does indeed prevent or very substantially reduce dispersal of microorganisms into the environment by the wearer. Laboratory tests should be followed by in- use tests both in order to ensure that some circumstance in the operating room situation does not vitiate the method used, and to establish that it is compat- ible with the other requirements of surgical operation procedure. Evaluations of this kind can establish an order of effectiveness among different methods applied to the same point in one of the routes of possible infection, e.g., methods of skin disinfection for the surgeon's hands. Some estimate of the relative reduction in the dose which might reach the wound as a result of replacing one procedure by another can also be made. If this re- duction is only small, then there is little likelihood of any clinical benefit. However, for reasons discussed above, even if the relative reduction bacterio- logically assessed in this way is very large, there is no guarantee that sig- nificant reduction in the rate of wound sepsis would follow. A more direct approach is to study the contamination actually reaching the wound itself. Although this is by no means a simple procedure, it is
186 0. M. Lidwell probably a good deal less exacting than attempting a full clinical assessment of the basis of change in the rates of wound infection and sepsis. There has been a sufficient number of such studies carried out to demonstrate the feasibility of this approach, and to establish that there is a significant correlation between the finding of microorganisms in the wound at the end of an operation and the risk of subsequent development of wound in- fection. Thus, in a study of over 1,000 surgical operations (Table 1), the risk of subsequent wound infection exceeded 25 percent for the wounds from which Staphylococcus aureus was recovered by bacteriological examination of washings taken from the wound at the end of the operation (6). The rate of subsequent wound infection was less than two percent for those wounds from which this organism was not so recovered. If all kinds of microorganisms were considered, then the rate of wound infection was greater than five percent for those wounds yielding washings positive for any bacteria, compared with about one percent for those wounds yielding apparently sterile washings. This study was part of an investigation into the value of ultraviolet irradiation of the air of the oper- ating room. In the hospital (No. 4 in the study) carrying out wound washings, approximately 40 percent of the operation wounds yielded bacteria on culture of the washings when there was no irradiation. The irradiation produced only a modest reduction in this fraction, to just under 30 percent, and no signif- icant difference in the risk of wound infection was observed. Table 1. Infection in Wounds Found Contaminated with Staphylococcus aureus (Number infected/total number) Wound classification Staph. aureus recovered Derived from Postoperative Wound Infections (6). Not recovered clean 2/12 9/812 clean contaminated 2/9 10/289 contaminated 2/4 1/31 dirty 2/4 2/25 total 8/29 22/1157 percentage 27.6 1.9
187 Control of Surgical Sepsis This correlation between bacterial contamination of the wound and subsequent infection did not imply that the infection was necessarily, or even usually, caused by a strain isolated from the washings. The sampling process is necessarily inefficient. If this were not so, it would itself be a poten- tially highly effective measure for control of infection. The possibility that the sample method may itself affect the prognosis needs, however, to be considered seriously. In addition to the removal of bacteria, probably inef- fectively, it may also remove blood clots and small pieces of tissue which might influence the chance of infection developing. In an attempt to assess the effect of moving from an old hospital into new buildings, Davidson, Smith, and Smylie (1) also examined over 1,000 operations bacteriologically by a swab and contact plate technique. Some seven percent of the wounds were found to be contaminated with "pathogenic" species at the end of the operation. This proportion did not differ appreciably in the two environments. Multi-regression analysis by Davidson, Clark, and Smith (2) showed that bacterial contamination of the wound was by far the most important factor in determining the outcome (Table 2). In conforming with the similarity of the rates of contamination Table 2. Coefficients of Regression Equation for Risk of Wound Sepsis Factor Coefficient Standard deviation 1. Positive wound swab 0.364 0.043 2. Dirty operations 0.106 0.026 3. Old hospital, pre Sept. 1966 0.105 0.021 4. Age over 60 0.091 0.020 5. Surgical team A or C 0.089 0.028 6. Duration over 60 min 0.069 0.018 7. Emergency 0.062 0.037 8. Glove puncture 0.058 0.025 9. Patient a nasal carrier of Staph. aureus 0.058 0.031 Data from 1,000 operations performed by three surgical teams, half before and half after moving into a new hospital. After Davidson, Clark, and Smith (2).
188 0. M. Lidwell found at the end of operation, the infection rates for primary infection with Staphylococcus aureus were indistinguishable in the two environments. The difference between the total infection rates was entirely due to secondary (or late) infections presumed to have developed after return to the ward (Davidson et al.) (3). Jepsen (4) has carried out a similar study in Denmark. CLINICAL EVALUATION Demonstration of a reduction in the rate of wound infection is the only fully convincing evidence of the value of any control measure. For assessment of antibiotic therapy, or treatment of the wound with antibacterials, it may be the only approach. If infection rates are very high then the statis- tical problems are not too great. Thus, Lister, when he introduced the use of antiseptic carbolic acid dressings for covering the wound after limb amputation, was able to show a convincing effect after fewer than 50 operations with the new technique (5) (Table 3). Although he himself stated that these numbers were too small for statistical significance they do show a probability of over 98 percent by the XT test. Fortunately for the patients, the infection rates in clean elective operations, which are those for which the provision of air, free from microorganisms, might seem most likely to be of value, are very much Table 3. Deaths Following Limb Amputation (deaths/number of amputations) Year All amputations Upper limb only 1864 7/17 1866 9/18 1864 and 1866 16/35 6/12 1867 0/7 1868 3/17 1869 3/16 1867-1869 6/40 1/12 Antiseptic procedures using carbolic acid were in use from 1867 onwards. After Lister (5).
189 Control of Surgical Sepsis lower. As a consequence, the numbers of operations that must be studied in order to demonstrate a significant difference between two or more procedures are usually very large. From Figure 2 it can be seen that when the control infection rate is about 10 percent, more than 3,000 operations must be studied in each group to determine, at a reasonable level of significance, whether or not there is a reduction of 25 percent in this figure, i.e., to 7.5 percent. A similar number would be required to show in the same way a fall to one 20,000 10.000 5,000 2,000 1,000 500 200 100 50- NUMBBR OF OBSERVATIONS REQUIRED IN EACH GROUP AT THE INDICATED PERCENTAGE REDUCTION Upper lines 0.9 probability a signi Lower lines 0.9 probability a eigni â.^_ (Short lines for a singl< the 99% icance level the 952 Icance level tail test) 1.0 1.5 2.0 3.0 5.0 7.0 10 15 f. INCIDENCE RATE IN THE CONTROL GROUP, J! 20 Fig. 2
190 0. M. Lidwell percent, from an initial infection rate of two percent, i.e., a reduction of 50 percent. A simple rule is that the expected number of infections in the 2 control group must be at least 1,000 (200-a)/a , where a_ is the percentage re- duction chosen as critical. The above numbers are those required in order to have a 90-percent chance of demonstrating at the 95-percent level of signifi- cance that the procedure being investigated is effective. It is clear that in addition to the effort required to collect records from such numbers of operations, there are also likely to be difficulties in ensuring comparable conditions throughout the investigation. Since, (in addi- tion to possible changes in staff and technique), there are undoubtedly secular changes in the bacterial environment in any hospital, it is essential that op- erations are carried out throughout the same period of time under all the various conditions to be compared. Table 2 also shows another obvious requirement. There was a substantial difference in the rates of infection following operations performed by the different surgical teams. Whether this was due to different carriage rates associated with the members of the teams, variation in surgical technique, or to other factors, it is clear that comparisons between environ- mental conditions or procedural differences can only be made if each surgeon operates equally under both control and trial circumstances. PROPOSED INVESTIGATION OF THE VALUE OF ULTRACLEAN AIR IN SURGERY The practical consequences of some of the points discussed above can perhaps be best seen in a brief description of an investigation planned to start in 1975. Any benefits of operating in an ultraclean air environment are most likely to be apparent in those operations for which the risks of autogenous and contact infection are lowest. In view of the considerable interest among ortho- paedic surgeons, and because most of the operations they perform fall into the clean-elective group and include large numbers of similar operations, e.g., total-hip replacement, it was decided to study these last. The term ultraclean air may be considered to imply a level of airborne contamination at least 10 to 20 times less than that attained by conventional
191 Control of Surgical Sepsis positive pressure turbulent ventilation, i.e., below 10 bacteria-carrying 3 particles per cubic meter of air (0.3/ft ) during operations. These levels can be obtained by a variety of installations, and provided a satisfactory level of air contamination is achieved, there seemed to be no reason to specify the type of equipment. A further reduction in the exposure to microorganisms arising from the surgeon and his assistant can be obtained by the use of pro- tective garments of the Charnley body-exhaust type. Three conditions have therefore been chosen for comparison: 1. Conventional positive pressure ventilation with clean air at 3 3 not less than 0.5 m /sec (1,000 ft /min). 2. Ultraclean air systems, e.g., horizontal or vertical uni- directional flow systems or the Charnley enclosure. Both of these are to be used with the surgeon and staff wearing conventional operating room dress. 3. Ultraclean air systems with the surgeon and his assistant(s) wearing a body exhaust system. All surgeons taking part would have to operate under both conven- tional and ultraclean air conditions. The number of observations required depends both on the infection rate in the control group, and on the reduction in this rate which would be considered worthwhile, and therefore must be detectable with reasonable con- fidence by the study. Observations on between 2,000 and 3,000 operations in each group would be needed if reduction of five percent to three percent or two percent to one percent were to be shown. There is no precise way of determin- ing these figures, so it has been decided to plan for approximately 3,000 op- 'erations per year over a study period of two and one-half to three years. Ob- servations will be made of both deep and superficial infection. The latter is perhaps more relevant to the conditions in other types of surgery, especially where there is no foreign body implanted in the tissues. The long delay period sometimes associated with deep infection around the artificial hip joints means that observations must be continued for up to two years after the last operation,
192 0. M. Lidwell but for a three-year study period over 80 percent of the deep infections should be recorded before the end of the fourth year. The numbers of operations performed annually in representative cen- ters ranges from approximately 100 to over 400. To observe 3,000 operations per year requires 12 to 15 participating hospitals. It is thought that suf- ficient bacteriological data on wound contamination at the end of the operation can be obtained from approximately 2,000 observations in all, and these could be obtained from only five to six centers, each of which would observe approx- imately 150 operations per year. This leads to the organizational plan sketched in Figure 3. Members of the working party will be able to exercise a general oversight of the in- vestigation in each area, and in some cases the microbiologists will carry out the detailed bacteriology in a hospital making studies of wound contamination at the end of operation. They, and the other microbiologists engaged in this DEPARTMENT OF HEALTH AND S.S. ORGANIZATION OF TRIAL ' SEPSIS RECORDS AND FULL BACTERIOLOGY MICROBIOLOGIST and RECORDS NURSE SEPSIS RECORDS RECORDS NURSE INITIATION AREA SUPERVISION PARTICIPATING HOSPITALS Fig. 3.
193 Control of Surgical Sepsis work, will also be responsible for establishing that the clean air installations in their own and selected other hospitals are functioning correctly throughout the period of the study. Further, they will examine, bacteriologically, speci- mens from clinically infected wounds in these other hospitals. In their own hospitals, they will also monitor, with respect to Staphylococcus aureus, the preoperative carriage state of the patients, and the carriage state of the op- erating room staff. In order to obtain comprehensive and consistent recording, and the systematic collection of bacteriological specimens, each participating hospital will have a records nurse, usually part-time. All records will be re- turned to the working party for assessment and analysis. CONCLUSIONS The statistical, ethical, and logistical problems associated with establishing the value, or non-value of medical procedures are often very great. They are especially so when the chance of the occurrence of the event to be prevented is small, and the consequences of it to the individual patient are great. The development of scientific medicine has brought us now to the situ- ation where only a proportion of the competing demands for resources can con- ceivably be met. It is, therefore, as important to decide when a possible treatment, though harmless, is ineffectual, as it is to establish the value of one that is effective. This will demand a greater investment of effort than we have, perhaps, previously thought desirable in such laborious evaluative studies as I have discussed. REFERENCES 1. Davidson, A.J.G., G. Smith, and H. G. Smylie, A bacteriological study of the immediate environment of a surgical wound. Brit. J. Surg. 58:326, 1971. 2. Davidson, A.J.G., C. Clark, and G. Smith, Postoperative wound infection: A computer analysis. Brit. J. Surg. 58:333, 1971. 3. Davidson, A.J.G., H. G. Smylie, A. McDonald, and G. Smith, Ward design in relation to postoperative wound infection. Brit. Med. J. i:72, 1971. 4. Jepsen, 0. B., Contamination of the wound in postoperative wound infection. Ann. Surg. 177:178-180, 1973.
194 0. M. Lidwell 5. Lister, J., On the effects of the antiseptic system of treatment on the salubrity of a surgical hospital. Lancet i:4, 1870. 6. Postoperative Wound Infections. The influence of ultraviolet irradiation of the operating rooms and of various other factors. Ann. Surg. 160:1, 1964.
195 PROPOSAL FOR AN INVESTIGATION OF THE EFFECT OF ULTRA-CLEAN AIR IN OPERATING ROOMS ON SURGICAL SEPSIS* 0. M. Lidwell 1. INTRODUCTION Experience has indicated that the introduction of positive-pressure ventilation for operating rooms has led to substantial reductions in the num- bers of airborne microorganisms, and has been accompanied by a reduction in the frequency of wound sepsis. Methods for still further reducing the air contamination by a factor of ten or more ("ultra-clean air") depend on the introduction of sophisticated ventilation systems, and the use of special clothing worn by the surgical team. The claim is made that their use would be accompanied by a further reduction of wound sepsis, at least in some kinds of surgery. However, ultra-clean air systems are expensive, and controlled assessment of their value seems desirable before they are taken into extensive use. 1.1. Orthopaedic Surgery A major study of sepsis after toal-hip replacement, by Professor John Charnley, shows a nearly ten-fold drop in sepsis rates after the intro- duction of his ventilation system. There has, therefore, been a considerable demand for the installation of ultra-clean air ventilation. As is inevitable in such a sequential study, there were changes during the period of study. There are those who challenge the assertion that the improvement was wholly or substantially due to the cleaner air, and be- lieve that a conventional plenum system would have given comparable results. Published series of observations after hip-replacement operations performed during the last ten years, record infection rates ranging from 0 to 10 percent on numbers of operations between 100 and several thousand. In most of these studies, the recording of sepsis was a secondary interest, and the only consistent criterion of sepsis was breakdown of the joint, presumptively related to deep infection. These deep infections often become apparent only after a considerable period. *Prepared by a working party set up by the Medical Research Council, London, England, as a basis for a controlled study. Chairman, Dr. 0. M. Lidwell.
196 Apart from the data given by Charnley, there is no substantial series that gives any direct comparison between the rates of infection in operations conducted in normally ventilated operating rooms and those performed under 'ultra-clean " air conditions, although there have been a few smaller sequential studies with similar conclusions. In particular, there are no reports of studies in which substantial numbers of operations have been carried out over the same period of time in either ultra-clean air conditions or with conventional ventilation, all other circumstances being similar. 1.2. Prerequisites for an investigation Infection of the surgical wound may arise from the patient's own organisms, from contact with contaminated objects - including the members of the operating team - or directly or indirectly from airborne bacteria in the operating room. The provision of cleaner air can directly affect only the last of these. Any study of ultra-clean air should, therefore, be made on operations in which the risks of autogenous and contact infection are as low as possible in relation to the risk of airborne infection, i.e., clean elective operations lasting at least 45 minutes. Because the risks of wound infection vary between different surgical procedures, it is important that the groups be comparable in this respect. In view of the interest of orthopaedic surgeons, and of the fect that most of the operations they perform fall into the clean- elective group and include many similar operations (for example, total-hip re- placement) it seems convenient to study these operations. Because of the many variables that may differ between surgical teams, control observations would have to be made by each team with each individual surgeon operating under control and ultra-clean air conditions. The studies referred to earlier suggest that for hip-replacement operations, the incidence of deep infection, as defined above may be as low as two percent or even less for operations performed with good conventional ventila tion, and that this rate is dependent on several factors, e.g., prophylactic antibiotic therapy, previous operation on the joint, and whether the joint disease is rheumatoid or osteoarthritic.
197 Although deep infection is of particular importance in total-hip replacement, the genesis of "superficial" infection is probably closer to the conditions relevant to infection in other types of surgery, especially those in which there is no foreign body implanted in the tissues. Rates for "super- ficial" infection in substantial series of orthopaedic operations of less than two percent have been reported, but the criteria employed are not well defined. Charnley has estimated that the incidence of superficial sepsis in his series was rather more than five percent, assessed at 14 days. The extent of the reduction in the incidence of wound infection that can be considered useful is a matter for judgement at a particular time, and depends on the consequences of infection. In determining the size of the in- vestigation necessary, reasonable criteria might be those that would give an 80-percent chance of demonstrating at the 99-percent level of significance - or a 90-percent chance at the 95-percent level - a reduction of one percent in the incidence of deep infection, e.g., from two percent to one percent, or a reduc- tion of two percent in the incidence of superficial infection, e.g., from five percent to three percent. At these levels, the number of observations required under the control conditions and under any other circumstances to be compared with this, would be at least 2500 for the "deep" infections and 1500 for the "superficial" infections. Because any one surgical team is unlikely to perform more than about 200 hip replacements in any year with, perhaps, an equal number of other suit- able elective operations, the study will have to be multicentered. In view of the inevitable uncertainties involved in the above estimates of the numbers of operations needed to obtain a reasonably conclusive result, analysis of the records will be made at intervals as the investigation proceeds. It will be possible to obtain much more realistic estimates after a relatively short time, and to modify the study accordingly. 1.3. Wound contamination Wound infection directly derived from air contamination in the operating room must result from microorganisms deposited in the wound during operation. Several studies have shown good correlation between bacteriological
198 contamination of the wound, as determined by samples taken at the end of the operation, and the risk of subsequent sepsis. If ultra-clean air systems are of any value this should be evident in a reduced amount of wound contamination. Experience has shown that samples from operation wounds are highly variable in this respect, but any difference in bacterial contamination levels under ultra-clean air conditions should be apparent in far fewer examinations than those necessary to determine variations in the level of sepsis. 1.4. Sepsis In view of the difficulties involved in setting out a consistent agreed-upon definition of "sepsis," it will be necessary to record objective descriptive criteria for the progress of the wound after operation. 2. THE PROPOSED INVESTIGATION 2.1. Operating Conditions There are a variety of ventilating systems and other circumstances that influence the bacteriological cleanliness of the air over the surgical field. Some of these are: 2.1.1. None or only extract ventilation: air bacteriologically dirty. 2.1.2. The accepted present-day standard of positive pressure ventilation: air fairly clean. 2.1.3. Professor Charnley's method of high rate of ventilation in an enclosure containing only the surgeons and part of the patient: air very clean. 2.1.4. So-called "laminar-flow" ventila- tion systems, whole or part room: air can be very clean. 2.1.5. Special clothing ventilated and impermeable to bacteria- carrying particles; at present associated with Professor Charnley's system. 2.1.6. Enclosure of the surgical field within a ventilated plastic envelope, e.g., the Trexler surgical isolator. Type 2.1.3. and the many variations of 2.1.4. are the utaa- clean air systems. Because it is the effect of the degree of bacterial cleanliness of the air that is the object of the
199 investigation, it should be possible to group all the ultra- clean systems together if they attain comparable standards. It is not necessary to include "dirty air" conditions in the study. We therefore, have the following four sets of conditions: I. "Clean" air - standard positive-pressure ventilation. II. Ultra-clean air - Charnley-Howorth or unidirectional ("laminar") airflow. III. As II but with special clothing. IV. Plastic isolators. The first of these can conveniently be taken as the control condition to be employed in a proportion of the operations performed by every team. It seems improbable that the 1000-2500 operations required for statisti- cally significant evaluation of the rates of sepsis in each group can be ob- tained for the plastic isolators, at least in the near future, so that only bacteriological evaluation may be practicable for this procedure. The use of special clothing in a conventionally ventilated operating room is not included because to produce a major effect on air contamination every person in the room would have to be dressed in this way. In the ultra-clean air system only the surgeon and his assistant intrude into the air mass that comes into contact with the wound. Such clothing might, of course, contribute to reducing the risk of sepsis by reducing the probability of contact transfer of microorganisms but this lies outside the scope of the present study. Therefore, the total number required, divided between conditions I, II, and III, is approximately 7500 observed operations, corresponding to 40 team-years. With 15 teams this would take about 2 1/2 years. 2.2. Selection and allocation of patients The types of operation that will be included in the study are: 2.2.1. Total-hip joint replacement. 2.2.2. Replacement of knee joint. Other clean elective operations may be included if convenient. It is important that patients are distributed between the three test conditions without bias. Strictly random allocation is probably impracticable,
200 and the method to be used will have to be agreed upon locally with the surgeons concerned and the hospital ethical committee. In most places, it will probably be convenient for each surgeon and team to alternate between the test and control circumstances. If this is done, the conditions should be changed at intervals of no more than seven days to minimize the influence of secular changes in the bacteriological environment. Alternatively, where several operating rooms are in use, each for a different condition, the sur- geons and teams may move at similar intervals from one operating room to another. 2.3. Clinical records Clinical and operation details for each patient included in the study will have to be recorded to check the comparability of the test and control series. These data should also provide the opportunity to evaluate the signi- ficance of some of the potentially significant variables. The progress of wound healing and the subsequent experience of the patient will be recorded in terms of objective descriptive criteria. An assessment of the incidence of differing grades of sepsis will then be made on the basis of these observations. 2.4. Bacteriological studies 2.4.1. Airborne bacteria Because the object of this investigation is the influence of clean air on sepsis, it will be important to measure the air contamination for each set of conditions. It will not be necessary to do this for every operation, but only often enough to obtain a good estimate of the mean and variation for each set of circumstances in each place. 2.4.2. Wound contamination Infection derived from the air of the operating room must relate to bacteria already in the wound when it is closed. If the ultra-clean-air systems are superior, there should be less wound contamination at the end of the opera- tion. By washing out the wound at this time, and culturing the wash fluid, an estimate of this can be obtained. As detected by this method, differences between the performances of the various systems may be apparent after many fewer observations than are necessary to establish differences in the rate of f sepsis. Wound-washing observations made after operations performed by no more than six teams should furnish ample data.
201 2.4.3. Wound infection Bacteriological specimens should be taken from all wounds that show any departure from normal at any time they are examined. These observations can then be related to the clinical observations to obtain information on the species of microorganisms responsible, and to throw some light on the disputed role of bacterial infection in delayed breakdown of hips after joint replacement. It is tempting to attempt to relate the strain of organisms found in each infected wound to those recovered from it at the end of operation, and to those carried by patient or staff, or found in the air of the operating room at that time. Reduction in the rate of infection due to contaminated air might be expected to be accompanied by an increasing proportion of infections due to strains carried by the patient before operation. But experience has shown that such a complete bacteriological picture is very difficult to obtain, and even a limited attempt involves a very large amount of bacteriological work. It will be useful, however, to obtain preoperative specimens from patients and from the scrubbed surgical team in those centres where wound washout is to be done, and to compare typable organisms with any found in the wound, either at the end of the operation or from the abnormal wounds. 2.5. Types of centres As has been indicated in the above description, it is proposed that the observations shall be made at two kinds of centres. 2.5.1. Sepsis and Bacteriology (SB) Centres which will: keep clinical and bacteriological records of wounds, record wound contamination (washout) for all wounds, record preoperative bacteriology of nose and perineum for all patients, record nose carriage of swabbed staff at regular intervals, and record physical and bacteriological conditions in their own operating rooms. There should be at least six SB centres, preferably, at least two for each ventilation system, e.g., Charnley or some type of laminar-flow. At each SB centre, the staff will include the local surgical and bacteriological consul- tants who will need to be assisted by: one full-time graduate microbiologist; one part-time laboratory technician; one part-time infection-record nurse.
202 These centres would also make environmental bacteriological observations at the second type of centre and, where practicable, carry out or oversee the bacteriological observations made at those places using the Trexler plastic isolator. 2.5.2. Sepsis Centres (S) which will keep clinical records of all wounds and bacteriological records of septic wounds. There should be at least nine S centres in all for the Charnley and the laminar-flow systems. At each centre, the local surgical and bacteriological consultants will need the assistance of one part-time infection-record nurse, as well as the help provided by the micro - biologist from the associated SB centre. The precise number and distribution of the centres, of both types, and the times at which they can be brought into the investigation must depend on negotiation with the individual hospitals. Preliminary enquiries suggest that it should be possible to find enough centres willing to take part, but that some provision of special equipment, e.g., ventilation and special clothing is likely to be required to set up a well balanced investigation within a rea- sonably short time. 2.6. Outcome This investigation could give one of three kinds of answers: a) The infection rate after operation in conventionally ventilated operating rooms is appreciable, and is significantly reduced when ultra-clean air systems are employed. b) The infection rate is appreciable and is not significantly different if ultra-clean air is provided. c) The infection rate is, in any case, very low under the best circumstances with conventional ventilation so that the effect of ultra-clean air, if any, is too small to be demonstrable. All these outcomes give an answer to the practical question as to whether ultra-clean air systems are of value, i.e., a) yes, b) no, and c) the effect is too small to be demonstrable.
203 EXPERIENCES WITH ULTRAVIOLET LIGHT IN OPERATING ROOMS J. Drennan Lowell The Peter Bent Brigham Hospital is a relative newcomer to the use of ultraviolet light as a method of reducing airborne bacteria in the operating room while a surgical procedure is in progress. We have used ultraviolet light at the Peter Bent Brigham for years as a method of improving the bacteriological isolation of our operating suites from the corridors of the hospital, and we have used similar lights to help isolate patients with known infections from uninfected patients on the wards and from the superimposition of additional infective organisms. We use them in our critical care unit for burn patients, and on occasion for patients on immunosuppressives. Among the problems encountered with installation of the lights in the operating room were those of personnel acceptance, especially by the Depart- ments of Anesthesia and Nursing. A solution comprising some six steps was successfully employed to solve the problem. The first of these was a test of their effectiveness. A non- pathogenic E. coli was nebulized in two of the four operating rooms equipped with ultraviolet light, and air samples were taken with and without lights. When the lights were not illuminated, the concentration of organisms recovered 3 3 was 930/5 ft in one room and 310/5 ft in the other. When the lights were turned on, even though the nebulizer was not turned off, there was essentially instantaneous removal of 100 percent of the organisms in one room and 99.7 percent in the other. Because their effect was so immediate, we felt we could take a second step and modify the Duke protocol, and our lights are now not illumina- ted until the patient is anesthetized, prepped and draped. This made the logistics of their use much simpler and much more acceptable to the Anesthesia Service whose personnel had vastly easier access to the patient during the in- duction period. We turn them off at the completion of wound closure and appli- cation of the final dressing. * Orthopaedic Clinical Services, Peter Bent Brigham Hospital, Boston, Massachusetts.
204 J. Drennan Lowell The third step was installation of an autotransformer inside the operating room, by which the intensity of output from the lights could be reg- ulated. A fourth step was to teach our circulating nurses how to monitor the intensity of the lights, using a germicidal photometer, compact in size and por- table. Once instructed in the technique of monitoring, the circulating nurses can then make adjustments as needed. The technique was easily learned, gave them a significant role in the use of the lights, and had a very positive ef- fect in reducing not only their apprehension but that of the other operating room personnel as well. The lights are tested with each procedure at time of illumination, and are spot-checked at numerous times throughout their period of illumination. Changes in the autotransformer are not often, but occasion- 2 ally, required. The intensity of light recommended is 25-30 microwatts/cm / second and is the intensity we use. The fifth step consisted of monitoring a series of four total hip- replacement procedures with and without lights, then circulating the results accompanied with photographs of the fallout plates, the plates from the volu- metric air samples and the Rodac agri-contact plates. The fallout plates 2 showed a range of 4-8 organisms/ft /minute in each of the rooms before the lights were turned on. In the room without ultraviolet light, this remained at 4 throughout the procedure and after wound closure. In the rooms with ultraviolet light, there was a 20-fold drop in organisms while the lights were on. Even at the end of the procedure, when the drapes were removed and there was a general stirring up of the air in the operating room, fallout never rose to the earlier levels. The accompanying photographs graphically illustrated these findings. One of the patients done in the room without lights had had a renal transplant and was on immunosuppressives. The problem that arises for patients on immunosuppressives is well known to our hospital staff and, I ex- pect, to those in most other teaching hospitals. It was a very happy coinci- dence that one of the patients monitored during a procedure with lights was a member of the anesthesia staff at the Brigham, and it was quite easy for the operating room personnel to identify with this individual. t
205 Experiences with Ultra Violet Light in Operating Rooms The sixth and final step was to develop a protocol for the use of the lights for protection of patient and personnel. The protocol consists of the wearing of goggles, not unlike ski goggles, which are commercially avail- able and regularly used in industry, or the use of a gambler's-type eye shade in association with ordinary eye glasses, temporarily equipped with side shields. The skin is protected either by the use of fire-resistant paper hoods covering the exposed areas of the face and neck, or by the use of a Benzophenone- containing suntan lotion, such as UVAL or SOLBAR, both of which are commercially available in any drug store. Similar precautionary measures may be taken for patients, although usually the patient is shielded from the lights by the surgical drapes held away from the face and anesthetic equipment by conventionally shaped metal supports. Lights are turned on at the time of incision and off at the comple- tion of wound closure. In our review of the National Academy of SciencesâNational Research Council study on the subject of postoperative wound infections (1), we found that no mention was made of the influence of humidity on the role of ultra- violet light effectiveness as a bactericidal agent. The role is a signifi- cant one and must be recognized and appreciated by any who might consider this method of reducing bacteria in the air. Observations relative to this were made by William Wells (2,3), who stated that when the humidity is above the 60 percent level, the bactericidal power of ultraviolet radiation is critically affected and drops off very rapidly. Our experience has paralleled that of the group at Duke University in that we have found that ultraviolet radiation is an extremely effective method of reducing the numbers of viable airborne bacteria in the operating theater. To date, using this technique in conjunction with preoperative and intraoperative antibiotics, we have experienced no infections in patients undergoing total-hip or total-knee replacements at the Peter Bent Brigham Hospital since the ultraviolet lights were installed. Prior to that time, our infection rate was 1.5 percent.
206 J. Drennan Lowell REFERENCES 1. National Academy of SciencesâNational Research Council Study, Post- operative Wound Infections: The Influence of Ultraviolet Irradiation of the Operating Room and of Various Other Factors, August 1964. Wells, William F. Bactericidal irradiation of air, Journal of the Franklin Institute, 229:3:347-372, March 1940. Wells, William F., Airborne Contagion and Air Hygiene, Harvard University Press, Cambridge, 1955. I f
207 ONE CLINICIAN'S THOUGHTS UPON ATTENDING THE WORKSHOP ON CONTROL OF OPERATING ROOM AIRBORNE BACTERIA J. Drennan Lowell Airborne bacteria are part of the wound infection problem. Sources can be identified and include the patient, scrubbed members of the surgical team, unscrubbed members of the surgical team, and other per- sonnel incidentally present in the operating room area. The numbers of bacteria can be quantified in some measure, by a variety of methods which allow assessment of techniques used to reduce airborne bacterial counts within the framework of one institution, but do not allow eas- ily compared conclusions between institutions, and some standardization of these methods between hospitals is critically needed. Methods are available to reduce the numbers of airborne bacteria in the operating room environment. At the very least, it would appear that a relationship can be shown between the quantity of bacteria in the operating room air and the incidence of postoperative infection in clean surgery. The quantitative or qualitative relationship between organisms found in the air to those found in the operative wound and subsequently found in in- fected wounds is inadequately defined. At least a part of this problem lies in the deficiency which exists in our being able to culture organisms in the operative wound with uniform success in all investigating centers. There is one report (Burke) in which wound culturing was 100 percent successful in re- covering organisms, whereas most other reports given at the meeting or cited elsewhere report only a 20 percent success. Until such a time as wound culturing can be done more successfully, any conclusion as to the existence of an incontrovertible relationship between observed airborne organisms and observed infection must be inferential, even though this inference may be remarkably strong. Until it is incontrovertible, our recommendations must be accordingly tempered, particularly with reference to the employment of air handling tech- niques and cannot be in the form of a mandate. Orthopaedic Clinical Services, Peter Bent Brigham Hospital, Boston, Massachusetts.
208 J. Drennan Lowell Although all of us urge continued effort along any reasonable line to solve the problem of wound infection, I would specifically recommend that one area of endeavor should be along the lines of developing a more success- ful method of intraoperative wound culturing.
209 UNIDIRECTIONAL AIRFLOW IN HOSPITAL OPERATING THEATERS George F. Mallison* Airborne microbial contamination in operating theaters may contribute, in part, to the risk of surgical wound infection; but endogenous infection, infection acquired postoperatively, and infection transmitted by contact are of greater importance. Other factors such as operative technique and host susceptibility also have major influences on the risk of acquiring wound infection. Unidirectional air- flow (UAF) systems in operating theaters have a potential for reducing the risk of wound in- fection in selected circumstances, but there are no controlled studies to support their efficacy thus far. Hospitals are not encour- aged to install these systems until controlled clinical trials document a significant reduc- tion in patient disease, and until population groups that might benefit from these systems are more precisely defined. Air filtration and delivery systems that are apparently far more efficient than traditional air conditioning systems in reducing airborne par- ticulate and microbial contamination have recently become available for use in hospitals (11). These systems incorporate ultra-high efficiency particulate filters and high rates of air change in an attempt to achieve clean, relatively unidirectional airflow. They have been variously described as laminar, linear, or unidirectional airflow (UAF) systems, but the latter term appears more de- scriptive of their effects in hospital operating theaters (5,21). Several in- vestigators have enthusiastically recommended conversion to UAF in operating theaters (11,21), but others have questioned the value of such systems for routine use (7,9). A hospital's decision regarding installation of UAF systems should be based upon an assessment of the epidemiology of surgical wound in- fections and data concerning the efficacy of these systems. THE EPIDEMIOLOGY OF SURGICAL WOUND INFECTIONS Airborne contamination in the operating theater is but one of the many factors influencing the risk of surgical wound infection. Bacterial Diseases Division, Center for Disease Control, Bureau of Epidemiology, Atlanta, Georgia.
210 Surgical wound infections may be acquired from numerous potential human and extra-human reservoirs both at time of operation and in the post- operative period (16,18). A patient's own flora is one potential reservoir: normal fecal flora is responsible for a substantial proportion of infections following intra-abdominal procedures (24); nasal carriers of staphylococci appear to have a higher risk of acquiring staphylococcal wound infection than non-carriers (25) ; and transient bacteremia at the time of operation may seed a surgical wound and result in subsequent wound infection (13). Exogenous reservoirs such as contaminated air, surgical instruments, medications, or colonized personnel are frequently implicated in outbreaks of wound infection (20,22). Pathogenic microorganisms may be transmitted to a surgical wound by direct contact, by contamination of a common vehicle, or indirectly by airborne spread. Despite numerous attempts to document the role of airborne transmis- sion in the acquisition of surgical wound infection, its relative importance is still uncertain. Burke demonstrated that air adjacent to open wounds at the time of operation is often contaminated with potential pathogens, and he con- sidered the air to be the final common pathway responsible for much of the universal contamination of clean wounds that he demonstrated (4). Howe and Marston, on the other hand, demonstrated no relationships between variations in airborne contamination and the incidence of infections in clean surgical wounds in a large uncontrolled study (14). Ultraviolet (UV) light is effective in re- ducing airborne microbial contamination (17), but a large and well-controlled study of the effects of UV light in the operating room showed a statistically significant reduction in infections only in very clean surgical wounds when UV light was used at the time of operation; and UV light had no influence on the incidence of wound infection in other types of surgery (17). Contamination of a surgical wound with pathogenic organisms is nec- essary but not sufficient for development of a clinical wound infection (10); from 30 percent to 100 percent of wounds of clean surgical procedures have been shown to be contaminated at the time of closure (4,17), and most healed without evidence of wound infection. Factors influencing the risk of infection include alterations ...n host susceptibility, surgical technique, and patterns of anti- microbial usage (8,17).
211 EFFECTIVENESS OF UNIDIRECTIONAL AIRFLOW Valid environmental testing of UAF systems requires sophisticated techniques and careful study design (15); many environmental studies purporting to demonstrate reductions in airborne contamination with UAF provide little meaningful data because of invalid sampling techniques or test conditions that are not applicable to the operating theater. Nonetheless, airborne microbial counts are probably reduced by these systems. However, current practices in operating theaters in the United States generally result in very low levels of airborne contamination. The benefits resulting from a further lowering of this contamination level by use of UAF must be measured by effects on patient disease; expensive or complicated alterations do not seem justified unless the incidence of infection is significantly decreased. Charnley and associates pioneered the clinical use of UAF in total- hip replacements. As they introduced progressively sophisticated measures to reduce airborne contamination, they noted a definite trend towards a reduction in surgical wound infections (6). However, changes in operative technique, alterations in surgical skill, variations in patient selection, and changes in ward care activities occurring concurrently were not evaluated and played an unmeasured role in these uncontrolled studies. Furthermore, they also intro- duced measures to minimize direct contact transmission. While some investigators have noted high infection rates following total-hip replacement (19,26), others have apparently achieved infection rates comparable to Charnley's excellent results without use of UAF systems (23,26). In a controlled clinical trial, Alpert reported a threefold reduction in surgical wound infections with use of a patient surgical isolator, but the study design was insufficiently described to allow interpretation of these data (1). DISCUSSION The exact proportion of surgical wound infections resulting from air- borne contamination in the operating theater is unknown (25), but we believe that other factors are generally of greater importance. A higher relative risk of acquiring airborne infection may occur in certain populations (the very old or young; patients with impaired immune systems, or those undergoing prolonged procedures that result in extensive tissue damage, or that implant a foreign
212 body). In these select populations, UAF has theoretical advantages. However, trials such as the UV light study (17) suggest that UAF alone may have only minimal effect on the incidence of wound infection, even in select populations. Thus, until the effectiveness of UAF in reducing patient disease is documented in controlled clinical trials, we do not recommend that hospitals routinely install UAF devices. REFERENCES 1. Alpert, S., T. Salzman, C. Sullivan, et al. * Wound bacteriology and infec- tion in patients operated upon by the surgical isolator technique or in a conventional environment. Surg. Forum 22:65-66, 1971. 2. Bassett, D.C.J., K. J. Stokes, and W.R.G. Thomas, Wound infection with Pseudomonas multivorans: A water-borne contaminant of disinfectant solu- tions. Lancet 1:1188-1191, 1970. 3. Bernard H. R., W. R. Cole, Bacterial air contamination and its relation to postoperative sepsis. Ann. Surg. 156:12-18, 1962. 4. Burke, J. F., Identification of the sources of staphylococci contaminating the surgical wound during operation. Ann. Surg. 158:898-904, 1963. 5. Charnley, J., Operating-theatre ventilation (Letter to the editor). Lancet 1:1053-1054, 1970. 6. Charnley, J., and N. Eftekhar, Post-operative infection in total prosthetic replacement arthroplasty of the hip-joint with special reference to the bacterial content of the air of the operating room. Brit. J. Surg. 56:641-649, 1969. 7. Committee on Operating Room Environment-American College of Surgeons: Special air systems for operating rooms. Bull. Amer. College Surgeons, May 18, 1972. 8. Condie, J. D., and D. J. Ferguson, Experimental wound infections: Contami- nation versus surgical technique. Surg. 50:367-371, 1961. 9. Eickhoff, T. C., Hospital Infections. Disease-A-Month, Sept. 1972. 10. Elek, S. D., and P. E. Conen, The virulence of Staphylococcus pyogenes for man. A study of the problems of wound infection. Brit. J. Exp. Path. 38:573-586, 1958. 11. Fox, D. G., A study of the application of laminar flow ventilation to opera- ting rooms. Public Health Monograph No. 78, 1-50, 1969. 12. Health Facilities Planning and Construction Service: General Standards of Construction and Equipment for Hospital and Medical Facilities. Washington GPO, 1969.
213 13. Howe, C. W., Experimental wound sepsis from transient Esaherichia coli bacteremia. Surg. 66:570-574, 1969. 14. Howe, C. W., and A. T. Marston, A study on sources of post-operative Staphylococcal infection. Surg. Gynecol. Obstet. 115:226-275, 1972. 15. Kethley, T. W., Air quality specifications (microbial): The sampling pro- blem. Svensk Farmacentisk Tidskrift. 73:854-866, 1969. 16. Lindbom, G., Studies of the epidemiology of Staphylococcal infections. II. Staphylococcal infections in a thoracic surgery unit. Acta. Chir. Scand. 128:421-434, 1964. 17. National Academy of Sciences-National Research Council, Post-operative wound infections: The influence of ultraviolet irradiation of the operating room and of various other factors. Ann. Surg. 160 (Suppl): 1-192, 1964. 18. O'Riordan, C., J. L. Adler, H. H. Banks, et al. , Wound infections on an orthopaedic service. Amer. J. Epidemiol. 95:442-450, 1972. 19. Patterson, F. P., and C. S. Brown, The McKee-Farrar total-hip replacement: Preliminary results and complications of 368 operations performed in five general hospitals. J. Bone Joint Surg. 54-A:257-275, 1972. 20. Penikett, E.J.K., R. Knox and J. Liddell, An outbreak of post-operative sepsis. Brit. Med. J. 1:812-814, 1958. 21. Scott, C. C., Laminar/linear flow system of ventilation: Its application to medicine and surgery. Lancet 1:989-993, 1970. 22. Shooter, R. A., G. W. Taylor, G. Ellis, et al ⢠, Postoperative wound infection. Surg. Gynecol. Obstet. 103:257-262, 1956. 23. Stinchfield, F. E., and E. S. White, Total hip replacement. Ann. Surg. 74: 655-662, 1971. 24. Thomsen, V. F., S. Larsen, and 0. B. Jepsen, Post-operative wound sepsis in general surgery. IV. Sources and routes of infection. Acta. Chir. Scand. 136:251-260, 1970. 25. Williams, R.E.O., R. Blowers, L. P. Garrod, et al., Hospital Infection: Causes and Prevention. Lloyd-Luke Ltd., London, 1969. 26. Wilson, P. D., H. C. Amstutz, A. Czerniecki et al., Total hip replacement with fixation by acrylic cement. J. Bone Joint Surg. 54-A:207-236, 1972.
214 PERSONAL ENVELOPE SYSTEM IN THE CONTROL OF OPERATING ROOM AIR CONTAMINATION Jo Miller* Geoffrey Richards* G. Ross Murphy* John Prentis* It is now a well-established fact that, in the modern operating room, the large majority of airborne microorganisms emanate from the skin and res- piratory tracts of the surgical team, including the circulating nurses, the anesthetists and other personnel. This air contamination is of particular con- cern to the orthopaedic surgeon, as it represents one of the sources of wound contamination which may lead to infection, particularly in the presence of the large implants used in major joint reconstruction. There has been considerable interest in controlling the level of this air contamination, and for the past decade most of the attention has been directed towards the use of so-called "laminar flow" clean rooms. These installations do not completely eliminate the problem, but they appear to reduce the level of air and wound contamination (8,10,19,23), and have, for the moment, gained considerable acceptance, in spite of the fact that they tend to be cramped, noisy, hot and expensive. More recently, attention has been devoted to other somewhat more simplistic ap- proaches in an attempt to reduce air and subsequent wound contamination. We have postulated that if the "spilling" into the air of organisms from the skin and respiratory tract of the surgical team could be prevented or blocked at its source by the use of new highly specialized operating room cos- tumes, then the level of air contamination and the degree of risk to the sur- gical wound would be markedly reduced. It is the purpose of this paper to pre- sent the preliminary results of research in this regard. REVIEW OF THE LITERATURE The surgical operating room of the modern hospital is synonymous with cleanliness and sterility, but complicating infection persists as a problem, involving 3-10 percent of all elective operations (1,22,25), and the availabil- ity of antibiotics has not changed the overall incidence of this problem in Department of Orthopaedic Surgery and Department of Microbiology, The Montreal General Hospital and McGill University, Quebec, Canada.
215 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis recent decades. Altemeier (1) states that the average cost for sepsis complicating elective surgery is $7,000 per patient, and the cost to the North American community probably exceeds ten billion dollars a year. It is probable that sepsis originates, in the large majority of cases, with contamination occurring during the course of surgery (4). It is now recognized that airborne microorganisms in the operating room are an im- portant source of contamination and a threat to the clean surgical wound (9,14, 15, 23). Many studies have shown that the air in the empty operating room is almost free of organisms (10,21), but becomes increasingly contaminated as the room is occupied by the surgical team. It is obvious that most of the contami- nating organisms originate from the skin and respiratory tract of the people who staff the room (3,9,11). There is great variation from individual to in- dividual in the number and character of organisms released into the air. Some individuals described as "shedders" or "dispersers" may release unusually large numbers of Staphylococcus aIbus, the organism most commonly found on the skin, while others may release significant numbers of very virulent pathogens, either because this is "normal" for that person or because of the presence of a local purulent lesion. The normal individual shedsnany thousands of viable particles per minute (26), and this desquamation is in no way controlled by the caps, masks and gowns in current use (7,12,13,16,20) which are essentially identical to those used 60 years ago (17,24). Some studies have suggested that closely woven or impervious fabrics might be of some value (7,13,20). Some effort has been made to identify the exact portion of the body responsible for the major part of air contamination (20), and some new types of attire have appeared designed to control the problem. These include hoods of various designs, new masks with or without exhaust systems, and gowns of synthetic fabrics, which are said to be more impervious than those of standard cotton fabric.
216 Personal Envelope System THE PREVENTION OF AIR CONTAMINATION It is postulated that if the spill of microorganisms from the surgical team into the air is to be prevented, two requirements must be satis- fied: the first is that costumes must be designed to cover all parts of the body which are sources of bacterial contamination; and secondly, that a mate- rial must be used which is completely impervious to airborne organisms. This leads to a third requirement: that moisture and heat accumulating beneath the costume must be carried away, and to satisfy this, a total-body exhaust system would be essential. An additional advantage of an exhaust system is that it produces a negative pressure within the costume, so that at the points where essential parts, such as arms, protrude from the costume, the flow of air will be from outside in. METHODOLOGY The experiments to be described were all carried out in an empty operating room of standard dimensions, supplied by filtered outside air, with approximately twenty changes per hour. The room was subjected to routine cleaning by the operating room housekeeping staff each day preceding the experi- ments. The air in the operating room was sampled using a Casella sampler which passes 25 cu/f t of air per minute over a bacteriological culture plate. Each of the sequential plates represented a five-minute period of sampling. A "mock operating room team" was assembled, made up of a small number of individuals who took part in most of the experiments. A standard activity pattern was developed which yielded reproducible results in terms of air con- tamination, providing a consistent baseline upon which the investigations were superimposed. In all experiments, the operating room was allowed to stand empty for 15 minutes, followed by the introduction of the "mock operating room team" which carried out its assignments for a period of 5 minutes. Following this, the room was once again allowed to "rest" for 15 minutes with air sampling be- fore the next experiment was carried out.
217 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis . ! 9i 11t OPERATING ROOM CHARACTERISTICS The air in the empty operating room was found to be virtually free of organisms, but the entrance of personnel was always attended by a sharp increase in contamination; within 15 minutes after the room was vacated the counts returned again to the baseline. These findings were consistent and re- producible from day to day. Reservoir Effect It was postulated that organisms introduced into the operating room air during one experiment might settle onto various surfaces, and might be remobilized by activity and turbulence during a subsequent experiment, resulting in artifactual data. To explore this point, the operating room was occupied by the "mock operating room team" dressed only in standard scrub suits, caps and masks; they pursued the standard activity pattern for five minutes with the ex- pected increase in air contamination. The room was then vacated and allowed to remain empty for 15 minutes, at which time large fans which had previously been placed in the room were started in order to produce air turbulence. This failed to produce an increase in the level of airborne microorganisms. It ap- pears, therefore, that there is no reservoir effect, and that organisms intro- duced into an operating room are either carried away by the airconditioning, or settle out on surfaces, and are not effectively remobilized. Opening and Closing of Doors It has been said that the opening and closing of doors in operating rooms produces turbulence which increases the level of air contamination. With samplers functioning in an empty operating room, the doors were opened and closed repeatedly from the outside without any increase in the level of air con- tamination. When the experiment was repeated by individuals inside the operat- ing room opening and closing doors, there was a significant rise in the air count. Obviously, the opening and closing of doors and the attendant turbulence in itself is not of importance.
218 Personal Envelope System Number of People and Level of Activity Studies were carried out to document the relationship between the number of people in the operating room, their level of activity, and the associated level of airborne organisms. In sequential experimental periods, increasing numbers of people followed the standard activity pattern in the operating room with air sampling. Figure 1 indicates the direct relationship between the number of people and the level of air contamination. * 200 100 . C.P.P. a. 5 5 £ 5 5 Ca Fig. 1. This histogram summarizes the relationship between the number of people in the operating room and the level of air con- tamination. As in all of our studies, the "mock operating room team" occupied the room for five-minute experimental periods, divided by 15-minute intervals, during which time the operating room was empty. Air sampling continued throughout. In sequential experimental periods two members of the "mock operating room team" pursued increasing levels of activity, with proportional increases in the level of air contamination, as summarized in Figure 2.
219 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis 200-tC.P.P. 100 _ 5 5 5 o. o_ 5 5 MINS. Fig. 2. The relationship between the level of activity of the "mock operating room team" and the concentration of airborne organisms is summarized in this histogram. THE COMPLETELY OCCLUSIVE "ENVELOPE" SYSTEM A relatively crude experiment was devised to establish whether or not the operating room environment could be protected from the organisms normally released from the skin or respiratory tract. A large occlusive polyethylene "bag" (Fig. 3) was devised with legs, but no openings for arms and head. Mem- bers of the "mock operating room team," dressed in this costume, entered the operating room and pursued a standard activity pattern for five minutes. Fifteen minutes later, as a control, the same members entered dressed only in scrub suits, caps and masks, and again pursued the same activity pattern for five minutes.
220 Personal Envelope System Fig. 3. Occlusive polyethylene "bag" used experi- mentally. Mock operations carried out with the team dressed in these garments resulted in no increase in airborne organisms.
221 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis Fifteen minutes later, the same group re-entered the operating room, once again dressed in the polyethylene bags. The findings are summarized in Figure 4, which, shows the members of the operating room team did not contaminate the air when dressed in plastic bags, but, in contrast, produced heavy contamination when dressed in scrub suits. It is interesting to note that 15 minutes later, when they re-entered the operating room in plastic bags, they were unable to re- mobilize the organisms released during the scrub suit phase moments earlier, once again verifying that there is no reservoir effect in the well-maintained operating room. 250- 200- 150. 103. 50. b C.P.P. i i; 5 u3 0- >->->- tâ lâ ^â D-_ tX Q_ BOB 9= &= m rn <r O_ Q_ O_ if E £ 5 90 KINS. Fig. 4. "Mock operating room team" dressed in plastic bags produced no increase in the number of airborne organisms, in contrast to heavy increase when they wore standard scrub suits, caps, masks and gowns.
222 Personal Envelope System FABRIC STUDIES The occlusive bag was utilized as a vehicle to study the effective- ness of various fabrics which are commonly used in the manufacture of operating room costumes. Bags were made of operating room cotton (Fig. 5), a patented closed woven fabric, and disposable paper fabric. The same members of the "mock operating room team" wore these costumes in sequential experiments, and the findings are summarized in Figure 6. The degree of air contamination is directly proportional to the inadequacies of the material in containing desquamated or- ganisms. The fabric is, of course, subjected to the various deformations, stresses and local pressure differentials which would be generated in any cos- tume in common operating room usage. Fig. 5. Experimental costumes were also manufactured from cotton and other fabrics.
223 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis 2SC 150 100 t- V) I 3 B u x D p * a '! 0 u II i 8 a Ul 3 Fig. 6. This histogram summarizes the effectiveness of different materials in containing desquamated and expired micro- organisms.
224 Personal Envelope System To compare the effectiveness of these bags with the jump suits worn over street clothes by visitors and with scrub suits, studies were carried out which are summarized in Figure 7. 150- 130. 50 - C.P.P. t 55 i >->â¢>- E E E 1 I 60 90 WINS Fig. 7. The effectiveness of various materials in containing microorganisms is contrasted in this histogram with the effective- ness of scrub suits, and with jump suits over street clothes.
225 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis SUMMARY AND CONCLUSIONS The results of the experiments reported here indicate that the empty operating room is a suitable "laboratory" for the study of air contami- nation emanating from members of the operating room team. This contamination can, in the experimental setting, be controlled by the use of a completely occlusive and impervious costume. Various materials have been evaluated and have been shown to be more or less effective with respect to their ability to contain organisms from the skin and respiratory tract. Studies are continuing and involve the development of a practical one-piece scrub suit, with arms pro- truding, made of impervious material, and used with a total-body exhaust system. All individuals in the surgical operating room will be required to wear this garment, which may diminish markedly the level of air contamination during the course of surgery. The surgeon, his assistant and the scrub nurse will, as at present, use the sterile gown to prevent contamination by contact. The importance of an occlusive head gear is still being studied. REFERENCES 1. Altemeier, W. A., Trauma Workshop Report. J. Trauma 10:1084-1086, 1970. 2. Anspach, W., Personal Communication. 3. Bernard, H. R., R. Spiers, Jr., F. W. O'Grady, and R. A, Shooter, Reduction of dissemination of skin by modification of operating room clothing and by ultraviolet light irradiation. Lancet 2:461, 1965. 4. Burke, J. E., Identification of the sources of staphylococci contaminating the surgical wound during operation. Ann. Surg. 158:898-904. November 1963. 5. Charnley, J., Instructions for Using the Charnley Ventilated Operating Gown and Mask. Internal Publication No. 22, Wrightington Hospital, Wigan, England. 6. Charnley, J., Post-operative infection after total hip replacement with special reference to air contamination in the operating room. Clin. Orthop. 87:167, 1972.
I 226 Personal Envelope System 7. Charnley, J. and M. Eftekhar, Penetration of gown material by organisms from the surgeon's body. Lancet 1:172, 1969. 8. Charnley, J. and M. Eftekhar, Post-operative infection in total prosthetic replacement arthroplasty of the hip joint, with special reference to the bacterial content of the air of the operating room. Brit. J. of Surg. 56:641-9, September 1969. 9. Coriell, L. L., W. S. Blackemore, and G. J. McGarrity, Medical applications of dust free rooms. II. Elimination of airborne bacteria from an opera- ting theater. JAMA 203:134-142, May 18, 1968. 10. Coventry, Personal Communication. 11. Davies, R. R., and W. D. Noble, Dispersal of bacteria and desquamated skin. Lancet 2:1295, 1962. 12. Devenish, E. A., and A. A. Miles, Control of Staphylococcus aureus in an operating theater. Lancet 1:1088, 1939. 13. Dineen, P. , Penetration of surgical draping material by bacteria. JAMA 43:82, Oct. 1, 1969. 14. Favero, M. S. , J. R. Puleo, J. N. Marshall, and G. S. Oxborrow, Comparison of contamination levels among hospital operating rooms and industrial clean rooms. Appl. Microbiol. 16-3:480-486, March 1968. 15. Ford, C. R., D. E. Peterson, and C. R. Mitchell, Microbial studies of air in the operating room. J. Surg. Res. 7:376-382, August 1967. 16. Ford, C. R., D. E. Peterson, and C. R. Mitchell, An appraisal of the role of surgical face masks. Ann. J. Surg. 113:787-890, June 1967. 17. Fowler, S., The Operating Room and the Patient, W. B. Saunders, Phila- delphia, 1951. 18. Jones et al. , Unidirectional air flow and surgical face mask exhaust system in the prevention of airborne surgical infection. Am. J. Surg. 124:49-51, 1972. 19. McDade, J. J. , J. G. Whitcomb, E. W. Rypka, and C. M. Franklin, Micro- biological studies conducted in a vertical laminar airflow surgery. JAMA 203:147-152, Jan. 8, 1968. 20. Medical Research Council Report of Aseptic Methods in the Operating Suite. Lancet pp. 705-709, 763-768, and 831-839, 1968. 21. Miller, Jo, and G. Richards, Unpublished data.
227 Jo Miller Geoffrey Richards G. Ross Murphy John Prentis 22. National Research Council, Post-operative wound infections; The influ- ence of ultraviolet radiation of the operating room and various other factcr.rs. Ann. Surg. , Suppl. 160:2, 1964. 23. Nelson, J. P., Horizontal clean flow rooms. Clin. Orthop. 96, 1973. 24. Neuber, G., 1886. 25. Public Health Laboratory Service, Incidence of surgical wound infections in England and Wales. Lancet 2:659, 1966. 26. Ulrich, J. A., Microbiology of Surgery Suites. Symposium on Clean Room Technology in Surgical Suites, NASA-Midwest Research Institute, Cape Kennedy, pp. 11-32, May 1971. 27. Williams, R.E.O. and R. A. Shooter, Infection in Hospitals, F. A. Davis Co., Philadelphia, p. 200, 1963.
228 CLEAN AIR SYSTEMS C. L. Nelson* T. L. Gavan* J. Schwartz* One of the most controversial issues in surgery today is the use of laminar airflow systems in the operating room. The controversy and mis- understanding are compounded with the term "laminar flow" since there is no really true laminar flow as it is used in the operating room. Many statements have been made concerning laminar airflow systems, or, more precisely, clean air systems. Questions have been raised regarding the design of units, the direction of airflow, premarketing testing, the fed- eral role in legislation, and the medico-legal implications of infection as related to clean air systems. Unfortunately, because of these implications and a lack of information regarding clean air systems, the issue has become almost too controversial to discuss. Therefore, only meager information on clean air systems has been available to the orthopaedist. Rumor, impression, and dogma have been substituted for the scientific study and statistical evalu- ation that we usually rely upon. The concept of clean air is of common interest to the ecologist and the surgeon, although the adjective "clean" has somewhat different connotations to each of these scientists. Attached to the concept of clean air and laminar airflow is an unfortunate emotional reaction. One must separate fact from hy- pothesis and attribute to these clean air systems only those facts we know. If unproved features are dogmatically attributed to the clean air systems, they will detract from what appears to be a logical, useful, and efficient adjunct in reducing bacterial contamination of surgical wounds. The orthopaedists' interest in clean air was further stimulated when total-hip replacement arthroplasty was developed for the treatment of diseases of the hip. An infection in the total-hip replacement arthroplasty or total-knee arthroplasty sites may result in failure of the operation. This is in contrast to the usually excellent results obtained with these operations. Also, total-hip and total-knee replacement arthroplasty may be a more sensitive indicator of infection than many commonly used laboratory bacterial testing systems. The large dead space surrounding the operative area and the small *Department of Orthopaedic Surgery, University of Arkansas Medical Center
229 C. L. Nelson T. L. Gavan J. Schwartz amount of living tissue in contact with the large endoprosthesis are factors that encourage bacterial growth. Furthermore, materials such as acrylic bone cement and high density polyethylene, which are not entirely inert, may also increase the likelihood of infection. HISTORICAL CONSIDERATIONS In considering the merits of the clean-air system in total-hip replacement arthroplasty, the historical background of infections and past attempts to eradicate infection should be reviewed. Lister (16) proposed the concepts of antiseptic surgery, and with the application of these principles the rate of infection decreased. His concepts led to the present era of aseptic surgery, which includes preoperative skin cleansing combined with a sterile scrub of the operating room personne1, sterile rubber gloves and masks, sterile draping equipment, sterile clothing, and sterile instruments. One must concede that there is no substitute for strict adherence to- standard aseptic surgical techniques. In 1888, Moynihan (20) stated that two-thirds of his pa- tients died of infection after he had opened the belly. In 1915, Brewer (4) showed that the infection rate after clean operations was 39 percent. In 1933, Meleney (18) reported that adherence to aseptic technique, rigid restriction of movement within the operating room, careful preparation of the patient, and gentle handling of tissue reduced the incidence of serious wound infection from 4 to 1.7 percent and of minor infection from 10 to 5.4 percent. In a similar study, McKissock et al. (17) reduced the percentage of surgical infec- tions from 15 to 1.1 percent. More recently, Henderson and Kornblum (12) were able to hold the percentage of serious infections in 3,290 operations to 1.7 percent; Steel (29) reduced his infection rate from 15 to 0.58 percent. From these studies, similar criteria for reducing the incidence of infection have been established: rigid restriction of movement in the operating room, especially restriction of the inadvertent visitor; careful, thorough prep- aration of the patient and the operating room personnel; gentle handling of tissue; and an intangible -- the alerting of every one in the hospital to the problem of infection.
230 4 Clean Air Systems SOURCES OF INFECTION Almost everyone agrees that surgical wounds, at the time of operation, can become infected. There is less agreement whether most infec- tions are endogenous or exogenous, animate or inanimate. However, certain facts are known. Wound infections are directly related to the type of organism that may find its way into the wound, the host's ability to combat infection, and the number of bacteria deposited in the wound; and of course, there must be a chance of deposition of bacteria into the wound for the development of infection. A single individual sheds 5,000 to 55,000 particles per minute, depending on how recently he showered and on the kind of clothing worn. The physical activity of the surgeon and others in the operating room directly affects the particle and bacterial counts and also affects the circulation of bacteria in the air. Conventional operating room air may contain as many as 10 to 15 bacteria per cubic foot and as many as 250,000 particles per cubic foot. The fine cotton surgical gowns and drapes have apertures large enough to allow particulate matter and accompanying bacteria to pass through. Charnley and Eftekhar (6) have shown that 50 percent of the surface of such gowns is contaminated at the conclusion of a total-hip replacement arthroplasty. Since the surgeon and operating room personnel are sources of bacteria, and since the activity pumps particulate matter over the wound, there is little question that if one releases particles with attached bacteria above the wound, they will be deposited into the wound. Even the most staunch nonbeliever in airborne contamination will agree that bacteria should not be deposited into an open wound. In light of the knowledge of airborne contamination of the wound site, what attempts have been made to eradicate contamination and clean the air of the operating room? Ultraviolet light has been used to reduce the incidence of wound infection by reducing the number of airborne bacteria. The results are inter- esting. Ultraviolet light does decrease the amount of bacterial contamination in the operating room, but a double blind study, using dummy lamps in one room,
231 C. L. Nelson T. L. Gavan J. Schwartz showed that the infection rate was almost identical with and without the ultraviolet light. In reviewing the results of this study, it is clear that all the cases in this portion of the study were the so-called unclean cases (11 2l)- When only ultraclean cases were considered such as those in recon- structive surgery, the difference in the two groups was statistically signifi- cant. Cleaning the airborne contamination by ultraviolet light reduced the infection rate. In genera1, in this study there would have been 30 fewer deep wound infections if ultraviolet light had been used in ultraclean cases. Therefore, there are data and logic in support of attempts to reduce airborne contamination. Altemeier and Levenson (1) have pointed out that in- fections developed postoperatively in an estimated 1,391,000 patients at a cost of $7,000 per patient, with an overall cost of 9.8 billion dollars for the control of wound infections. This is a significant sum, but only the individual affected and the surgeon can truly understand the suffering and disability that occur with serious deep wound infection. Monetary and humani- tarian needs justify all rational attempts to reduce wound infection. LAMINAR AIRFLOW SYSTEMS The first use of laminar airborne contamination control was by the aerospace industry in the early 1960s. It was noted that any small particulate matter could contaminate gyroscopic equipment in missile guidance systems and cause malfunction. Particulate matter as small as that in cigarette smoke could be responsible for contamination. Whitfield (31) pioneered the use of laminar airflow systems to reduce contamination. - Laminar airflow can be described as the flow in which the entire body of air within a confined area moves at a uniform velocity along parallel flow lines. This principle was mechanized and then applied to the aerospace industry. Recently this type of clean air technology has been applied to the operating room. In this type of environment, the system filters out particu- late matter with attached microbial organisms before it enters the surgical area. This is done by a high efficiency filter system. The filter system now used in surgical suites provides what is known as Class 100 air. Class 100 air is one in which each cubic foot of air contains not more than 100 particles 0.5 micron or larger in diameter.
232 Clean Air Systems Laminar flow clean air really does not exist in the operating room. Persons or objects in the room, such as the surgeon or surgical tools, disturb the airflow pattern, and the air becomes turbulent. Thus, in all laminar clean rooms the air is turbulent. "Clean air room" is a more accurate and acceptable description than the term "laminar." It is also clear that in designing and using clean air systems, it is necessary to properly plan and use the surgical area, so that any turbulence or eddying of current is reduced, and does not occur over the surgical site. One type of clean air system that has been used is a wall-less horizontal system consisting of two parts - a self-contained blower filter system, which propels a horizontal flow of Class 100 air, and a vacuum aspira- tor system for the surgeon (22). A blower module placed at one end of the operating room propels a flow of clean air across the operating room. The horizontal system sends airflow in one direction, and in this manner particles are removed from the operating area. This system produces approximately 200 air changes per hour in the room, compared with 12 air changes in the standard operating room. The 200 changes developed by the horizontal system are for the entire room; over the work site there are about 500 changes of air per hour, and particles shed by the surgeon and staff are swept away from them and across the room. The air then rebounds from the opposite wall, turns laterally along the wall and enters the filtration system, the module, at the side. A high efficiency particulate air filter lies in the front of the module behind the perforated metal cover. The filter is 99.9 percent effective in removing particulate matter with attached bacteria and viruses. A bacteriological study was designed to determine the effectiveness of a horizontal-wall-less clean air system in reducing bacterial contamination in the operating theater. One operating theater was chosen for study and all investigations were done in this room. The same surgeon, number of operating room personnel, and nursing staff were used for each clinical evaluation. Pa- tients undergoing total-hip replacement arthroplasty were chosen for clinical study, none showing evidence of previous or secondary infection. The patient and operating room personnel were identically prepared prior to surgery, and all personnel in the operating theater wore hoods, gowns and foot coverings. The doors of the operating theater were taped closed and any entry was noted.
233 C. L. Nelson T. L. Gavan J. Schwartz Five TDL slit samplers were positioned in specified locations and labeled A,B,C,D, and E. The letters designated the following sites: A - in- strument table; B - side opposite surgeon; C - anesthesia area; D - wound site; E - clean air filter exit (see schematic representation). The sampler at the wound site was somewhat differently positioned from the rest. A metal tube k\ inches by 15/16 inches was connected to the slit sampler which was attached to a tigon tubing 21 inches by 1 6/16 inches and'a metal tube 9 6/16 inches by 15/16 inches. The metal tube and flexible plastic tube were passed beneath the surgical coverings, incorporated into the drapings and taped into position. The tube did not directly contact the patient's skin and was positioned only after adhesive skin covering had been applied to the thigh. Three tissue samples were taken at surgery, incubated in blood agar plates at 35° C and colony forming unit (CFU) counts made at 48 hours. The TDL slit sampler used in this study was made by Engineering Development and Products, Incorporated, Decatur, Georgia. The sampler is a spring drive model-A which operates by a simple gear mechanism. A plastic Petri dish (100x15 millimeters containing 10 cc of blood agar medium) was placed on an adjustable stage which rotated at a speed of one revolution per hour. The medium did not touch the sampler except at the level indicator, which is a contact point in the center of the Petri dish. The top edge of the indicator is the point of reference used in determining the starting point from the dish. A witness mark on the dish is matched to the reference indica- tor. All samplers were calibrated with a Wright's respirometer, and were found to draw one cubic foot, plus or minus 100 cubic inches of air per minute (including the sampler with tubing attached). The plates were collected after surgery and placed in a 35° C incubator for approximately 48 hours. The plates were examined with a Quebec colony counter, and the colonies counted counter-clockwise in twelve equal divisions on the plates. Each division represented a five-minute period. The colony counts were recorded in colony forming units per minute per cubic foot of air. The presence of lint on the plates is also recorded, and indicated as
234 Clean Air Systems degree by either scattered lint (which means slight) or distinct lint (which means a definite line of lint.) RESULTS The operating theater studied showed the following levels of bacterial contamination when the operating room was completely empty (Fig. 1) Figure I Clean Air Unit Off Room Empty A = 2.4 CFU/CU FT /MIN B = 1.8 CFU/CU FT /MIN C = 1.9 CFU/CU FT /MIN D = 2.2 CFU/CU FT /MIN E = 2.0 CFU/CU FT /MIN The same empty operating room showed a significant decrease in the levels of contamination of air sampling when a clean air system was turned on (Fig. II). Figure II Clean Air Unit On Room Empty A = 0.2 CFU/CU FT /MIN B = 0.1 CFU/CU FT /MIN C = 0.2 CFU/CU FT /MIN D = 0.4 CFU/CU FT /MIN The bacteriological contamination of the operating theater was sampled in five different sites during total-hip replacement arthroplasties in a standard operating theater without the clean air system functioning. The levels were significantly high (Fig. III).
235 C. L. Nelson T. L. Gavan J. Schwartz Figure III AVERAGE CFU/CU FT/MIN.AT 5 SITES DURING TOTAL HIP REPLACEMENT STANDARD O.R. (15) A 7.2 B 9.9 C 8.5 D 6.3 E 9.8 Under identical conditions, the sampling was performed during total- hip arthroplasty when a clean-air system and an aspirator system were turned on. The levels of bacterial contamination were significantly lower (Fig. IV). Figure IV AVERAGE CFU/CU FT/MIN AT 5 SITES DURING TOTAL-HIP REPLACEMENT CLEAN AIR SYSTEM OJ (15) A B C D E A third portion of the study was performed when the clean «ir system was functioning during a total-hip replacement arthroplasty, but the aspirator system was shut off. The bacteriological levels were statistically unchanged. PART II The data in Part I showed that the use of an aspirator system did not further reduce the level of bacterial contamination. The aspirator system was then replaced with a prototype surgical isolator-aspirator system, con- sisting of a self-contained aspirator vacuum, and a one-piece Barbac surgical gown with face shield that completely covered the surgeon to the knees. There was a further significant reduction in the bacterial contamination (Fig. V). 0.1 p <0.02 1.24 p <0.02 2.32 p <0.05 1.4 p <0.001 0 p <0.001
236 Clean Air Systems Figure V AVERAGE CFU/CU FT/MIN AT 5 SITES DURING TOTAL-HIP REPLACEMENT CLEAN AIR SYSTEM QN. & SURGICAL ISOLATOR (6) A 0.6 B 0.88 C 0.71 p< 0.02 D 0.43 p< 0.05 E 0. The surgical isolator-aspirator system alone was then used during total-hip replacement arthroplasty in the standard operating room without a clean air system. The data in a small number of cases showed a difference at the wound site (Fig. VI). Figure VI CLEAN AIR UNIT BE! SURGICAL ISOLATOR 3 PATIENTS AVERAGE CFU/CU FT/MIN AT 5 SITES A = 4.2 CFU/CU FT /MIN B = 3.7 CFU/CU FT /MIN C = 3.7 CFU/CU FT /MIN D = 1.9 CFU/CU.FT /MIN p< 0.05 E = 3.6 CFU/CU FT /MIN Eighty-two different colony forming units were surveyed and morphologically identified (Fig. VII). Figure VII BACTERIA IDENTIFIED Staphylococcus Coagulase Negative 52% Diphtheroids 447, Bacillus Species 27. Yeast Forms 17. Staphylococcus Coagulase positive 1% C
237 ⢠J3 Figure VIII FOURTEEN PATIENTS WITH INFECTED TOTAL- HIP REPLACEMENT ARTHROPLASTY Staphylococcus Coagulase Negative 6 Staphylococcus Coagulase Positive 1 Beta Hemolytic Streptococcus 1 Gamma Streptococcus 1 Enterococcus 1 Klebsiella 1 Pseudomonas Aeruginosa 3 This study confirms the findings of Charnley, Ritter, and Herndon (7,13,14,27,28) that clean air systems do significantly reduce bacterial contamination, and points out the importance of containment of the bacterial contamination produced by operating room personnel at the time of surgery. EFFECTIVENESS OF CLEAN AIR SYSTEMS Is there any proof that these systems have value? Charnley and Eftekhar (7) have shown that with an increased airflow the infection rate was reduced from 8.9 to 1.3 percent. During this period, no doubt, they changed and improved their technique, altered the criteria for surgery, and more rigidly enforced aseptic technique. Nevertheless, this significant decrease in the incidence of wound infection combined with the other data, suggest that the clean air system is a valuable adjunct in reducing operating room infection. The results of the other studies point out the capabilities of these systems in reducing the amount of bacterial contamination and, in turn, infection. Re- sults of a study in Albuquerque, New Mexico, showed a 0.79 percent wound in- fection rate in clean air systems compared with a 1.4 percent wound infection rate in a control group. One of the few controlled studies that has been done involved 300 total-hip replacement arthroplasties , divided into two groups (25). One group of patients were operated on in a nonclean air environment under similar con- ditions, and with similar insertions. In 134 patients operated in nonclean air room, the infection rate was 5.2 percent. In 270 patients operated on in a clean air system, infection rate was 1.1 percent. This, too, is not statis- tically significant at present, but again, there appears to be a definite trend.
238 Clean Air Systems DISCUSSION There was a statistically significant difference between a standard operating room, and the same operating room with the clean air system function- ing. When both rooms were empty, there was approximately a ten-fold decrease in bacterial contamination in all areas sampled. There was also a significant difference between the standard operat- ing room and the clean room when tested during total-hip arthroplasty. A sig- nificant difference was found in all areas measured. It was interesting that the aspirator system furnished with the unit did not cause a significant dif- ference in the bacterial contamination level whether the aspirator system was on or off. This was, in part, due to the fact that the system did not cover the neck and, consequently, did not efficiently aspirate the particulate ma- terial from the body. The addition of a surgical isolator-aspirator system in place of the standard aspirator system showed a statistically significant difference in areas C and D (p<0.02 p<0.05 respectively). When the operating room per- sonnel used the surgical isolator-aspirator in a standard operating theater, there was a statistically significant decrease in bacterial contamination at the wound site. The data point out the importance of isolating the surgeon, since the addition of a more effective "packaging" of the surgeon further re- duced the bacterial contamination, even though the ventilating environment had not been changed. It should be pointed out that the limited number of samplings in the study may underlie the lack ot a statistically significant effect for the surgical isolator at some sites. None of the patients studied became infected. The bacteriological identification settle plates show that the majority of organisms identified were those usually found on the skin. The organisms found in the wound were not in the same proportion as those found in the infected cases (Fig. VIII). This finding has been pre- viously noted.
239 C. L. Nelson T. L. Gavan J. Schwartz Charnley (5), who has popularized the concept of reduced airborne contamination, has recently reported his results as well. In 5,800 total- hip replacement arthroplasties done between 1960 and 1970, the incidence rate of infection fell from 7 to 0.5 percent. It is his feeling that re- duced exogenous infection in the operating room resulted from measures taken to eliminate airborne contamination. Charnley has stated that of all the precautions taken against infection in the operating room, the most impor- tant was clean air, but this measure alone did not reduce the infection rate below 1.5 percent. He has stated that a reduction below 1.5 percent in 1969 through 1970 was due to measures taken to avoid penetration of bac- teria through the surgeon's gown. His studies are some of the longest with the largest numbers and longest follow-up of those yet published. At present, there are no hard statistical data universally agreed upon that either prove or disprove the decrease in the infection rate when using a clean-air system. Perspective can be gained when one considers op- erating rooms of fifty years ago. In looking at the operating theaters of fifty years ago, one wonders how one could possibly operate under these sur- gical conditions and not expect a relatively high infection rate. In the future, one will look upon our present system and wonder how one could ex- pect to implant large prostheses and not have high infection rates. In summary, there is little accepted scientific data that prove the effectiveness of clean air in reducing infection although Charnley's ex- perience appears to offer support for this relationship. The expectation of decreased bacterial contamination by improved mechanical barriers and clean air systems seems reasonable. Certainly it is not harmful to use a clean air system when it is employed properly. It is also acceptable to reduce bacterial contamination at the wound site; in fact, this has been a Listerian principle known to surgery since the 1800s. Lastly, the systems that are available are capable of reducing bacterial contamination of the wound site not only through the use of clean-air systems, but through im- proved methods of surgical isolation (2,3,8,9,10,13,14,23,24,26,27,28,30).
* I i 240 Clean Air Systems The absolute provable data regarding infection may not be available for the next ten years. Even at the end of ten years, the answer may not be irrefutably settled statistically. It becomes apparent that each surgeon must identify the problems in his operating room technique, in the type of surgery he does, and then choose whether an improved barrier system or clean air sys- tem is necessary in his environment. REFERENCES 1. Altemeier, W. A., and S. Levenson, Trauma workshop report: Infections, immunology, and gnotobiosis. J. Trauma, 10:1084-1086, 1970. 2. Amstutz, H. C., Prevention of operative infections. Clean Air Symposium - Part I. Cleve. Clin. Q., 40:125-131, 1973. 3. Anspach, W. E. Jr., and M. Bakels, Local air blanket protection of surgical wounds to prevent airborne contamination. Clean Air Symposium - Part II. Cleve. Clin. Q., 40:229-239, 1973. 4. Brewer, G.E., Studies in aseptic technique. JAMA 64:1369-1372, 1915. 5. Charnley, J., Postoperative infection after total-hip replacement with special reference to air contamination in the operating room. Clin. Orthop., 87:167-187, 1972. 6. Charnley, J., and N. Eftekhar, Penetration of gown material by organisms from the surgeon's body. Lancet, 1:172-173, 1969. 7. Charnley, J., and N. Eftekhar, Postoperative infection in total prosthetic replacement arthroplasty of the hip joint. Brit. J. Surg., 56:641-649, 1969. 8. Charnley, J., Clean air in the operating room. Clean Air Symposium - Part I. Cleve. Clin. Q., 40:99-114, 1973. 9. Dixon, R. E., The role of airborne bacteria in theatre-acquired surgical wound infection. Clean Air Symposium - Part I. Cleve. Clin. Q., 40: 115-123, 1973. 10. French, M.L.V., M. A. Ritter, J. B. Hart, and H. E. Eitzen, A systems analysis approach to postoperative wound infections. Clean Air Symposium - Part II. Cleve. Clin. Q., 40:221-227, 1973.
241 11. Hart, D. , R. W. Postlewait, I. W. Brown, Jr., et^ ail. , Postoperative wound infections: A further report on ultraviolet irradiation with comments on the recent (1964) National Research Cooperative Report. Ann. Surg. , 167: 728-741, 1968. 12. Henderson, E.D. , and S.S. Kornblum, Studies on the epidemiology of staphylo- coccal wound infection in previously clean surgical cases on an orthopaedic service. Instruct. Lect. Am. Acad. Orthop. Surg., 18:282-287, 1961. 13. Herndon, C. H. , Personal communication, February 24, 1972. 14. Herndon, C. H. , The clean air operating room at University Hospitals of Cleveland. Clean Air Symposium - Part II. Cleve. Clin. Q., 40:183- 190, 1973. 15. Laufman, H. , Confusion in application of clean air systems to operating rooms. Clean Air Sympoisum - Part II. Cleve. Clin Q. 40:203-209, 1973. 16. Lister, J. , On the antiseptic principles in the practice of surgery. Lancet, 2:353-356, 1867. 17. McKissock, W. , J. Wright, and A. A. Miles, The reduction of hospital infections of wounds. A controlled experiment. Brit. Med. J. , 2:375- 377, 1941. 18. Meleney, F. L. , The control of wound infections. Ann. Surg., 98:151- 153, 1933. 19. Morris C. , Medical-legal implications of clean air systems. Clean Air Sympoisum - Part II. Cleve. Clin. Q. , 40:161-181, 1973. 20. Moynihan, B. G. , Infection acquired in hospitals. Lancet, 2:885-891, 1955. 21. National Research Council, Postoperative wound infections: The influence of ultraviolet radiation of the operating room and of various other factors. Ann. Surg., 160 (Suppl. ) :1-192, 1964. 22. Nelson, C. L. , Experience with a wall-less horizontal clean air system during total hip replacement, Tn_ Symposium on Clean Room Technology in Surgical Suites, Cocoa Beach, Fla. NASA, Midwest Research Institute, pp. 107-114, 1971. 23. Nelson, C. L. , Clean Air Symposium - Part I. Cleve. Clin. Q., 40:97-98, 1973. 24. Nelson, C. L. , Introposition Arthroplasty Orthopaedic Clinics. 25. Nelson, J. P., Personal communication, April 27, 1972. 26. Nelson, J. P., A. R. Glassburn, Jr., R. D. Talbott, and J. P. McElhinney, Horizontal flow operating room clean rooms. Clean Air Symposium - Part II. Cleve. Clin. Q. , 40:191-202, 1973.
242 27. Ritter, M. A., and J. B. Hart, Evaluation of wall-less horizontal laminar flow clean air system for hospital operating rooms. Presented at Symposl on Bio-Clean Room and Its Application to Surgery Suites, San Francisco, November 1971. 28. Ritter, M. A., J. B. Hart, M.L.V. French, and H. E. Eitzen, A systems analysis approach to postoperative wound infections. Phase I. Clean Air Symposium - Part II. Cleve. Clin. Q., 40:211-219, 1973. 29. Steel, H. H., Surgical infections - orthopaedic considerations. Instruct Lect. Am. Acad. Orthop. Surg., 18:288-293, 1961. 30. Whitcomb, J. G., W. Whitfield, J. G. King, e_t al., Ultraclean operating rooms. Lovelace Clin. Rev., 2:65-69, 1965. 31. Whitfield, W. J., A new approach to clean room design (SC-4673 RR.) Sandia Corporation, Albuquerque, New Mexico, March 1962. «
243 HORIZONTAL FLOW OPERATING ROOM CLEAN ROOMS J. Phillip Nelson* INTRODUCTION Our experience with operating room clean rooms at St. Luke's Hospital in Denver, Colorado began in March 1971, with the installation of what is be- lieved to be the first functional horizontal flow class 100 operating room clean room (5). A second horizontal flow class 100 clean room was installed nine months later for purposes of evaluation of the effectiveness of the clean room and a helmet-aspirator-paper gown system. This second clean room was construc- ted by the Martin-Marietta Corporation under contract to NASA (6) and included portability and communication features. Since their installation, these rooms have been used for essentially all of our orthopaedic procedures, and the helmet- aspirator-paper gown system has been used for the majority of major joint im- plants. Two other helmet-aspirator-paper-gown systems have also been evaluated. Figures 1 through 6 illustrate the clean rooms and personnel isolator systems. Our studies (2,3,4,7) have included measurement of airborne bacterial and particle concentrations, wound contamination rates, sterile surface contami- nation rates, and continuing analysis of deep wound infections following joint implants. In addition, we have evaluated the effect of substituting 99.0 per- cent High Efficiency Particulate Air (HEPA) filters for 99.99 percent HEPA fil- ters on the bacteriological and particle characteristics in a functional oper- ating room (OR) clean room (1). Credit for a substantial amount of this work is given to Mr. Michael Wardle and his associates from the Jet Propulsion Lab- oratory. AIRBORNE BACTERIAL SAMPLING Table 1 summarizes our experience with airborne bacterial sampling in a horizontal flow class 100 operating room clean room using a Sartorious Mem- brane Sampler at the wound. In general, there is a significant reduction in airborne bacterial concentrations when comparing the clean room to the regular operating room, which is enhanced by improving the scrubbed personnel barrier sys- tem. These trends were also corroborated using a Reyniers slit sampler. The bac- teria cultured were predominantly Staph epidermidis, diphtheroids and micrococcus. *Instructor, Department of Orthopaedics, University of Colorado, Denver, Colorado.
244 O <U N CO â¢H Q, i-j CO O iH O U o o e o o â¢H C U cd 9 n) ,H 4J U cu B â¢H O »-i O I GO â¢Ul 4-1 VJ 0) E a) cx o 00 M-l â¢H at 4J g N â¢H O O iH VJ co C en n) n) CD rH H u o O 00 o c > C 0) Pi O OOM-I â¢H t
245 a fl) u 01 n s*. cu in 3 O EC w I ca O iH U O C c ⢠o vO 03 M ⢠QJ 60 O. 0) O C0 M t-> e â¢H 0) n) u O 4J ⢠td o ⢠(0 00-H M rt o â¢U 4J 4J CO 0) )-i â¢H -r-l M CX C â¢H .u i XI (0 ⢠3 >, 60 J3 03
246 Horizontal Flow Operating Room Clean Rooms OPERATING ROOM TYPE GARMENT Table 1. Airborne Bacteria Sampling at Wound BACT./CU.FT. Standard Cotton Hip Nail - Prosthesis (6) 3.87 1 4 C.R. Cotton THA (30) .84 C.R. H-A-P2 THA (14) .11 C.R. Cotton TKA5 ( 2) .56 C.R. H-A-P TKA (11) .48 C.R. Cotton THA + TKA (32) .84 C.R. H-A-P THA + TKA (25) .27 1. Clean Room 2. Helmet-Aspirator-Paper Gown 3. Approximately 50 cubic feet of air sampled per case 4. Total Hip Arthroplasty 5. Total Knee Arthroplasty PARTICLE COUNTS Table 2 demonstrates that the Clean Room with standard garments markedly reduces the level of 0.5y and 5y size particles at the wound. The level is further reduced by improving the effectiveness of the personnel barrier system. WOUND SAMPLING Tables 3 and 4 reflect our experience with swab wound sampling in the regular operating room, clean room, and clean room with Helmet-Aspirator- Paper Gown (H-A-P) system. The data indicate that the clean room reduces wound contamination and this effect again is enhanced by the H-A-P system. Table 5 indicates the types of bacteria cultured from wounds, and these types and their distribution correspond to those found in the airborne sam- pling data.
247 J. Phillip Nelson Table 2. Volumetric Air Sampling of Surgical Field Airborne Particulates in Horizontal Flow Class 100 Clean Room Operating Room .'i GARMENT PROCEDURE MEAN N0. PARTICLES/CU.FT. (No.) 0.5 0.5u 5u Cloth THA and TKA (65) H-A-P THA and TKA (55) 3,108 1,830 266 197 Table 3. Wound Bacterial Contamination Rates - Total Experience CULTURE NO. CASES N0. CULTURES POS. CULTURES RATE (%) REGULAR 0.K. SUPERFICIAL 56 PLUS COTTON DEEP 107 OTHER 14 OVER -ALL 108 CLEAN ROOM O.R. SUPERFICIAL 344 PLUS COTTON DEEP 551 OTHER 27 OVER-ALL 590 CLEAN ROOM 0.R. SUPERFICIAL 107 PLUS H-A-P DEEP 120 OTHER 18 OVER -ALL 122 56 107 14 177 344 556 29 929 107 120 21 248 10 25 4 39 16 32 1 49 1 4 2 7 17.9 23.4 28.6 22.0 4.7 5.8 3.5 5.3 .9 3.3 9.5 2.8
248 Horizontal Flow Operating Room Clean Rooms Table 4. Wound Bacterial Contamination Rates Total Hip and Total Knee Surgery REGULAR O.R. PLUS COTTON CLEAN ROOM O.R. PLUS COTTON CLEAN ROOM O.R. PLUS H-A-P CULTURE NO. CASES NO. CULTURES POS. CULTURES RATE (7.) SUPERFICIAL 38 38 6 15.8 DEEP 43 43 12 27.9 OTHER 10 10 3 30.0 OVER -ALL 43 91 21 23.1 SUPERFICIAL 121 121 4 3.3 DEEP 123 123 6 4.9 OTHER 11 11 0 0 OVER-ALL 123 255 10 3.9 SUPERFICIAL 98 98 2 2.0 DEEP 99 99 2 2.0 OTHER 9 9 1 11.1 OVER-ALL 99 206 5 2.4 Table 5. Types of Bacteria Cultured from Clean Surgical Wounds STAPH. EPID. 72.5% 57.1% DIPHTHEROIDS . 7.5 19.0 BACILLIS SUB. 7.5 0 NON-HEMO. STREP. 0 16.7 ANEROB. STREP. 2.5 0 MICROCOCCUS 2.5 0 ENTEROCOCCI 2.5 0 ENTEROBACT. HAFINA 0 2.4 HERELLEA VAG. 2.5 0 CLOSTRIDIUM PERFRINIG. 2.5 0 E. COLI 0 2.4 MORAXELLA 0 2.4
249 J. Phillip Nelson STERILE SURFACE SAMPLE Table 6 reflects our experience with glove and drape bacterial sampling at the conclusion of the procedure. The improved personnel barrier system significantly reduces the rate of this contamination. Table 6. Microbial Sampling of Surgical Gloves and Drapes in a Clean Room SURFACE SYSTEM NO. OPERATIONS NO. SAMPLES % POSITIVE Gloves Cotton 22 98 20 H-A-P 31 180 3.2 Dxapes Cotton 57 57 23 H-A-P 36 36 8.3 99.99 PERCENT HEPA FILTERS VS 99.0 PERCENT HEPA FILTERS Tables 7 and 8 summarize our studies with regard to use of the less efficient filters under operating conditions, and the data indicate that the less efficient filter is as effective in providing particulate and micro- biological control at the wound as the 99.99 percent filter. Table 7. Comparison of Particle Counts in Horizontal Laminar Flow Surgery Room with 99.99% and 99.0% Filters MEAN NO. PARTICLES/FT 3 99.99% FILTERS 99.0% FILTERS 1 1 2 2 0.5y 5.0y 0.5y 5.0y Wound Site - THA 2159 238 2410 40 Back Table - - 2436 9 Upstream of Activity 1 1 1328 1 1. Particle counts taken with Model 540 Coulter Counter. 2. Particle counts taken with Model 550 Coulter Counter.
- 250 Horizontal Flow Operating Room Clean Rooms Table 8. Comparison of Airborne Bacterial Counts in Horizontal Laminar Flow Surgery Room with 99 99% and 99.0% Filters .3 99.99% FILTERS MEAN NO BACTERIA/FT" 99.0% FILTERS Garment HAP-1* (No.)1 HAP-1 (No.) HAP -2** (No.) At Wound - THA .11 (14) .12 (5) .08 (18) .35 (5) * Bubble helmet-aspirator-paper gown system. ** Hood helmet-aspirator-paper gown system. 1. Number of cases. Approximately 50 cubic feet using Sartorius membrane sampler. of air sampled per case INFECTION STUDIES We have kept careful records of deep wound infections following total-hip and total-knee replacements. Table 9 summarizes our results with a follow-up of at least six months on all cases. There have been changes in our technique, such as consistent use of perioperative antibiotics, and use of two different operative exposures. However, draping, operative time, use of irrigation solutions, closure and the use of tube drains and dressings have all remained constant. A thorough analysis of these infections, includ- ing their economic significance is now in progress. Table 9. Total-Hip and Total-Knee Deep Wound Infections Regular O.R . Inf. % Clean Room O.R . Inf. % THA 131 8 6.10 466 4 .86 TKA 1 0 0 90 1 1.11 THA + TKA 132 0 6.06 556 5 .90
251 CONCLUSIONS J. Phillip Nelson The Class 100 Horizontal Flow Clean Room is effective in substantially reducing ariborne bacterial and particle concentrations at the site of the wound, and wound contamination. Sterile surface contamination, wound contamination, and airborne bacterial and particle concentration rates are further reduced by use of an improved scrubbed personnel barrier system. The 99,0 percent filter is as effective as the 99.99 percent filter in maintaining.these reductions. The deep wound infection rate in our series of total-joint replacements is lower in the Clean Room Environment as compared to the Regular Operating Room. REFERENCES 1. Marsh, C. R. and J. P. Nelson, Evaluation of 99.0% Efficient Filters in a Horizontal Laminar Flow Surgical Clean Room. Unpublished material. 2. Nelson, J. P., Bacterial Studies in a Horizontal Flow Operating Clean Room. Unpublished material. 3. Nelson, J. P., A. R. Glassburn, Jr., R. D. Talbott, and J. P. McElhinney, Clean room operating rooms. Clinical brthop. and Rel. Res., 96:179-187, 1973. 4. Nelson, J. P., A. R. Glassburn, Jr., R. D. Talbott, and J. P. McElhinney, Horizontal flow operating room clean rooms. Cleve. Cl. Quart., 40:19l-201, 1973. 5. Nelson, J. P., Use of Total Horizontal Laminar Flow Systems in Surgery. Symposium on Clean Room Technology in Surgery Suites. Medical Research In- stitute and NASA. Reprint MRI-1064, Cape Kennedy, 1971. 6. Tevebaugh, M. D., and J. P. Nelson, Experimental System and Its Evaluation for the Control of Surgically Induced Infection. Final Report from Martin- Marietta Corp. to NASA (Applications Technology Office). Contract NASW- 2210, MCR-72-80, Washington, 1972. 7. Wardle, M. D., J. P. Nelson, p. LaLime and C. S. Davidson, A surgeon body- exhaust, clean-air operating room system. Orthop. Rev., 3:43-51, 1974.
252 , #; HELMET ASPIRATOR SYSTEMS J. Phillip Nelson* INTRODUCTION Since the great majority of airborne bacteria originate from operating room personnel, improved personnel barrier garmenting should result in reduced numbers of bacteria in the ambient air. Helmet aspirator systems represent a significant advance in our ability to prevent shed bacteria from reaching the operating room air. The principles of these systems are: 1. Full-length gown made of wetting-resistent , small porosity materials . 2. Helmet or mask assembly with clear plastic face plate. 3. Yoke for shoulders of head harness to support helmet or mask. 4. Hood for head which fits beneath neck portion of gown or one piece hood-gown. 5. Tubing in helmet or hood which connects to a long, pliable, lightweight tubing. 6. Suction pump for aspiration of air. Ideally, the gown should fit loosely so that a negative pressure may be maintained in the gown-body space. Air is sucked from the floor upwards around the body and is exited with air from the head region through the helmet tubes. Not only does this air movement evacuate shed bacteria, but it also provides significant cooling through increased evaporation. Air movement at a rate of approximately 6 cfm has been found to be most comfortable in terms of cooling effect and noise level. The umbilical tubing traverses down the wearer's back and then proceeds along the floor to the suction manifold. It is desirable to have the hood or helmet portion sterilizable so that procedures wherein there is considerable activity or wherein bone chips might fly about may be done more safely. Evacuated air is usually recycled through the clean room High Effi- ciency Particulate Air (HEPA) filter system. Exhaust tubing is made of pliable plastic with metal wire helical reinforcement and it is usually one inch in diameter. instructor, Department of Orthopaedics, University of Colorado, Denver, Colorado
253 J. Phillip Nelson The sequence for putting on the assemblies is usually: 1. Head harness prior to scrub. 2. Hand scrub. 3. Hood or hood faceplate in operating room. 4. Gowning. PROBLEMS WITH DESIGN INCLUDE: 1. Faceplate scratching. This occurs if faceplates or helmets are used repeatedly and because materials are soft plastics. Harder plastics such as Lexan might resolve this problem. 2. Visual distortion. In some models there is a significant cur- vature in the faceplate and difference of thickness of the plastic which causes mild distortions about the periphery. 3. Discomfort from head harness. This usually occurs if the face- plate harness is too heavy or if the headband is too tight. If a shoulder yoke-bubble helmet is used, the assembly is too heavy and often causes fatigue rapidly. 4. Faceplate fogging. Insufficient air exchange is the cause of this problem. 5. Noise from air movement. This is unavoidable but since the frequencies are of low intensity and if tubing is of sufficient diameter, noise is usually neither irritating or distracting. 6. Communication. Because the entire head is enclosed either with plastic or fabric and because of the air movement noise, hearing is impaired and one must use either an electronic communication system when the head is enclosed in a bubble or speak in a louder voice with the cloth system. 7. Mobility and agility. The ability of team members to move about is reduced to some degree with all systems. This effect is greatest with the Charnley and bubble helmets and least with the standard gown helmet systems. 8. Sterility. Care in putting on the units is necessary, particu- larly in the Charnley over-the-head type and in applying sterile faceplate hoods. â¢
254 Helmet Aspirator Systems TYPES OF SYSTEMS 1. Charnley â Loose fitting full-length, one-piece gown and hood. Head harness with cheek exhaust tubes and clear plastic face- plate. Necessitates overhead donning with potential for con- tamination (Figs. 1 and 2). 2. Martin-Marietta Bubble Helmet â Plexiglass space-type helmet resting on shoulder yoke. Special paper gowns. Heavy, easily scratched (Figs. 3 and 4). 3. Goodrich â Open faceplate, plastic helmet with aspirator. Neck relatively unprotected. Standard-type gown (Fig. 5). 4. Bechtol, Leinbach â Head harness, cheek tubes and clear plastic faceplate. Requires use of hood and paper or finely woven gown (Figs. 6 and 7). 5. Laminair â Welder's mask with metal cheek tubes. Replaceable plastic shield. Hood and paper or finely woven gown. Relatively heavy (Figs. 8 and 9). 6. Envirco-Hutter â Lightweight plastic head harness with face shield incorporated in hood. Face shield snaps onto hood harness. Face shield head-hood assembly may be sterilized (Figs. 10 and 11). OUR STUDIES Our studies, comparing the Helmet-Aspirator-Plexiglass system (H-A-P) to cloth gown systems, have indicated an approximate ten-fold reduction in par- ticle generation and an approximate three-fold reduction in airborne bacteria and sterile surface contamination. In addition, wound contamination rates in the clean room have fallen from 5.3 to 2.8 percent using the H-A-P system as opposed to the standard cloth garments. CONCLUSIONS Comfortable, functional helmet-aspirator low-porosity gown systems which appear to significantly reduce release of shed particles and bacteria from scrubbed operating room personnel are now available.
255 Fig. 1. Charnley Head Harness and Assembly. , Fig. 2. Charnley Head Harness and Full Body Gown.
256 Fig. 3. Martin-Marietta Bubble Helmet and Shoulder Yoke. Fig. 4. Martin-Marietta Bubble Helmet and Gown.
257 I Fig. 6. Bechtol Aspirator System Fig. 5. Goodrich Helmet Fig. 7. Leinbach Aspirator System
258 Fig. 8. Laminair Head Harness Assembly Fig. 9. Laiminair Head Harness, Hood and Gown. Fig. 10. Envirco-Hutter Head Harness Assembly. Fig. 11. Envirco-Hutter Head Harness, Hood and Gown.
259 FACTORS TO BE CONSIDERED IN ANALYZING POSTOPERATIVE CLEAN, REFINED WOUND INFECTIONS J. Phillip Nelson* The following outline was constructed in an attempt to organize acquisition of information which might contribute to defining the etiology of postoperative wound sepsis, particularly in the clean, refined wound. The outline is meant to give general guidance to those who wish to develop a pro- tocol for the investigation of wound sepsis. Modification of the outline by addition or deletion of certain factors thought to relate to wound sepsis may be necessary. Since large numbers of patients and a vast amount of information would be required to achieve statistical significance for a single variant with regard to the incidence of wound sepsis, it is suggested that an adequate information handling system, plus the assistance of an infection control nurse and microbiologist be enlisted. I. The Patient. A. Statistical Information. 1. Name or initials and date of admission. 2. Hospital and Hospital Number. 3. Age, Weight, Sex. B. Diagnosis. 1. Primary - reason for surgery. 2. Secondary - all other recorded diagnoses. C. History. 1. Previous surgery - should include all surgery and complications thereof. 2. Medicines - should include all medicines taken on regular basis during preceding year. 3. Allergies. D. Laboratory. 1. Blood - CBC, Sed Rate and other indicated tests. 2. Urine - analysis, culture and sensitivity for pus cells. *Instructor, Department of Orthopaedics, University of Colorado, Denver, Colorado.
260 3. X-rays - appropriate results. 4. Cultures - appropriate results such as urine, joint aspiration, etc. a. The results of preoperative nasal, perineum and skin at the operative area may give useful information relative to the carrier state and for correlation with postoperative sepsis. E. Infection. 1. Specific inquiry by history and examination should be made for evidence of infection in teeth, respiratory system, urinary system, and skin. 2. Specific inquiry should be made for history of any type of previous bacterial infection. II. Preoperative Preparation. A. Preoperative hospital days. B. Intestinal preparation. C. Urinary or intravenous catheters. D. Skin preparation. 1. Shave or depilatory - method, time before surgery. 2. Washing. a. Area. b. Number of times. c. Length of wash. d. Materials used for wash - i.e., soap, hexachlorophene, iodoform, etc. III. Antibiotics A. Local or Systemic. B. Specific type. C. Administration. 1. Route. 2. Schedule relative to surgery. 3. Dose. IV. Operating Room. A. Personnel. ,
261 1. Names for each case including: a. Surgeon(s). b. Anesthesiologist. c. Circulating nurse. d. Scrub nurse. e. Visitors. 2. Bacteriology of each member of the O.K. team, specifically with regard to carriage or shedding of certain types of bacteria. B. Air Handling System (Conventional and Clean Room). 1. Air mix - i.e., recirculated vs. fresh air. 2. Air exchanges per hour. 3. Filter condition. 4. Filter efficiencies. 5. Temperature. 6. Humidity. C. Other Antibacterial Systems. 1. Local air exchange systems - describe. 2. Ultraviolet light - describe. D. Barrier Technique. 1. Describe type of: a. Hood or cap. ) b. Helmet or mask. ) Specific brand name with type c. Gown. ) of generic named material for d. Drapes. ) each. e. Gloves. ) 2. If aspirator system used, describe. E. Physical Layout of Operating Room. 1. Doors and windows. 2. Relationship to corridor and other rooms. 3. Size. 4. Lighting. 5. Cabinetry.
262 F. Surgical Procedure. 1. Type. 2. Implant. 3. Length in time. 4. Irrigation - amount and type. 5. Closure. 6. Dressing. 7. Drains. 8. Unusual occurrences - i.e., break' in technique, violation of contaminated viscus, etc. G. Bacterial Sampling 1. Surface. a. Wound and sterile surfaces. (1) Method - i.e., swab, irrigation, tissue. (2) Timing. (3) How handled in laboratory. (4) Time of reading. (5) Results - qualitative and quantitative. 2. Airborne, a. Method. (1) Type of sampler. (2) Placement of sampler. (3) Rate of sampling. (4) Results - qualitative and quantitative in viable organisms per cubic foot. 3. Personnel. a. Area - i.e., nasal, perineum, skin or clothes. b. Method. c. Results. H. Sterilization. 1. General methods of sterilization (i.e., steam with pressure, gas, irradiation, soaking) should be recorded for differing types of materials. 1
263 V. Postoperative Care. A. Area - i.e., recovery room, ICU, ward, or private room. B. Dressing changes - when, where, and by whom. C. Activity. D. Wound complications. 1. Dehiscence. 2. Hematoma. E. Drains, catheters and TV's removed. 1. Cultured? F. Anticoagulation. G. Antibiotics. 1. See III. 2. Other antibiotics - define some parameters and reason for use. H. Other infections - i.e., respiratory, urinary, etc. 1. Record time of onset, appropriate diagnostic procedures, results of cultures and sensitivities. VI. Wound Infection. A. Define with regard to: 1. Location - i.e., superficial or deep to deep fascia. 2. Time of appearance. 3. Local signs of inflammation. 4. Systemic signs of inflammation. 5. Laboratory signs of inflammation. 6. Drainage and sinus tracts. 7. Special diagnostic procedures - i.e., arthrogram or sinogram. B. Cultures. 1. How obtained? 2. How handled in laboratory? - i.e., methods of culturing including aerobic and anaerobic. 3. Result of culture. 4. Sensitivities of organisms. 5. Bacterial typing.
264 C. Treatment. 1. Antibiotics - dose, route, interval and duration. 2. Surgical - describe. D. Outcome of infection. '
265 HUMAN BACTERIAL SHEDDING Dick K. Riemensnider* Studies, ranging from very simple to complex, of bacterial skin population and their dissemination have been conducted by numerous investigators. Two efficient methods for recovering airborne bacterial particles shed from in- dividuals are volumetric air sampling and a total rinse procedure in a sterile environment. Work by Duguid and Wallace, Ulrich, and others showed that large numbers of microorganisms can be shed into the environment by individuals as they carry out daily work routines. While at the Center for Disease Control in Atlanta, the author conducted many experiments on bacterial shedding. The Microbiotank, a large 7 ft X 3 ft cylindrical stainless steel chamber capable of being sterilized, was used in studying individuals for different time periods in various modes of dress to determine their bacterial shed rate. Bacteria were collected by a rinse procedure of sterile buffered water and a refrigerated continuous flow centrifuge. Enumeration was done with the pour-plate technique. This work showed not only that large numbers of microorganisms were shed, but that proper dress reduced shedding. Most individuals were quite consistent in their shedding rate. Daily, weekly, monthly, and yearly, the number of organisms shed was within the same log. One individual studied over a two-year period had the same shed rate each time. Shed rates were in the range of 5 -5 viable particles per minute. The author believes that, as a result of the collection method, skin scale clumps were broken up into smaller aggregates than those recovered by volumetric air sampling. One series of studies undertaken to reduce shedding demonstrated the effectiveness of changing the type of clothing and the clothing materials. In *Environmental Microbiology Consultant, Division of Facilities Utilization, Health Resources Administration, Public Health Service, U.S Depart- ment of Health, Education, and Welfare, Rockville, Maryland.
Jill 266 Dick K. Riemensnider all instances, sterile clothing was used to assure that the bacteria were coming from the subject. In comparing a Dacron coverall with a hood and sterile scrub suit, the T test at five percent level showed a statistically significant re- duction with the Dacron suit. A nonwoven material produced an even greater re- duction. In a study with conventional operating room garb, increased reductions occurred as mask, surgical gown, and gloves were added, i.e., the more body covered with sterile clothing, the more reduction. Studies were done with various bathing procedures and it was found that hexachlorophene soaps, upon prolonged usage, definitely decreased the shed rate. The shed rate was decreased to almost zero in a study using both the Gemini and the Apollo space suite, where air was exhausted outside the sterile environment. This investigator believes that proper operating room clothing together with adequate air handling systems (filtered air and good peripheral exhaust) can control bacterial shedding problems in the operating room. BIBLIOGRAPHY 1. Duguid, J. P., and A. T. Wallace, Air infection with dust liberated from clothing. Lancet 2:845-849, 1948 2. Riemensnider, D. K., Quantitative Aspects of Shedding of Microorganisms by Humans. Proceedings Spacecraft Sterilization Technology, NASA SP-108, 97-103, 1966. 3. Riemensnider, D. K,. Reduction of microbial shedding from humans. Contain. Control 6:19-20, 1967. 4. Sciple, G. W., D. K. Riemensnider, and C.A.J. Schleyer, Recovery of micro- organisms shed by humans into a sterilized environment. Appl. Microbiol. 15:1388-1392, 1967. 5. Ulrich, J. P., Isolation and Identification. Proceedings of the National Conference on Institutionally Acquired Infections, P.H.S. Publication No. 1188, 80-83, 1964.
267 HORIZONTAL LAMINAR AIRFLOW - ITS EFFECTS ON REDUCING POSTOPERATIVE WOUND INFECTIONS Merrill A. Ritter* Harold Eitzen** Morris L. V. French** Jack B. Hart*** Since the development of the unidirectional filtered air (laminar airflow) for use in operating rooms, four prime questions have arisen as to its beneficial effect when compared to a conventional air handling system: 1) Does laminar airflow reduce the amount of bacteria about the wound site? 2) Does laminar airflow keep the surgical instruments cleaner? 3) Does laminar airflow reduce wound contamination? 4) Does laminar airflow reduce infection rates? METHODS AND MATERIALS I Seven hundred and twenty-one consecutive clean orthopaedic cases have been evaluated microbiologically (1), 167 without laminar airflow, 377 with a horizontal wall-less unit (A)1 and 177 with a horizontal walled (walls retractable) unit (B)2. During the evaluation of unit B we randomly deter- mined whether or not the side walls were pulled out, for only total-hip and total-knee arthroplasty cases. This unit, which was capable of higher speeds at either end, was not installed as such, so that with the walls out there was an even flow of air through the tunnel as noted in other walled units. During the study we also evaluated the effect of a body exhaust system3. During and before this time, total-hip arthroplasty cases have been cultured upon opening the wound and just before closure. Infected cases have *Associate Professor of Orthopaedic Surgery, Indiana University School of Medicine. **Assistant Professors of Clinical Pathology, Indiana University School of Medicine ***Assistant Professor of Mechanical Engineering, Indiana University- I Purdue University at Indianapolis ^gnew-Higgins, Beech Grove, California 2 Weber, Grand Rapids, Michigan Surgivent, Grand Rapids, Michigan
268 Merrill A. Ritter Harold Eitzen Morris L. V. French Jack B. Hart all been identified. Ninety-two total-hip arthroplasty cases were performed in a conventional operating room without laminar airflow, 278 total-hip arthro- plasties in the wall-less laminar airflow unit A, and 87 in the unit B. Seventy-three total-hip arthroplasties were in unit A before our microbiologi- cal studies began. Three hundred and seventy have been followed for more than one year. To evaluate the problem of instrument contamination, two studies were undertaken. The first study evaluated 20 open heart surgeries in a conventional operating room, whereby hemostats were removed from a rack on the sterile instru- ment table and placed into jars of thioglycolate broth every 15 minutes, and tes- ted for both aerobic and anaerobic bacteria. In ten cases the hemostats were placed in the jars by the gloved scrub nurse's hand and in ten cases with sterile pick-ups. Each case was continued for only three hours. There were 20 bilateral total-hip operations evaluated in the horizontal laminar airflow operating room over a three-hour period. In ten cases the hemostats were removed from the instru- ment table located next to the laminar airflow unit with a gloved hand and ten with sterile pick-ups. The second study evaluated both skin and deep knife blades. There were 102 skin and 209 deep knives from cases in a laminar airflow room versus 106 skin and 165 deep knives from cases in a conventional operating room evaluated for contamination by dropping them into the test tubes of thioglycolate broth after their normal usage. Both the jars and test tubes were evaluated for contamination over a 14-day period. RESULTS All three types of laminar airflow reduced the airborne contamination in the operating room statistically (p.<0.005). Of interest was unit B without walls, which gave just as good control as it did with walls, or as did unit A. This was surprising because the velocity at the wound site (midway between the two ends of the unit and six feet from the filter bank) without walls was about An 85 feet per minute, but zero from there out laterally at the six-foot mark. The plates around the operating team and room illustrate that the room itself was cleaner.
269 Horizontal Laminar Airflow 2 Table 1 presents the number of colony forming units (CFU)/ft /hr for cases in a conventional operating room, a Horizontal Wall-less Laminar Airflow (LAF) unit A and a Horizontal Wall unit B with and without walls. The sites of air settle plate (ASP) sampling were adjacent to the wound, on the back instru- ment table; plates one through four were four feet above the floor and directly behind the surgical team; plates five and six were at the periphery of the room, four feet high. Table 1. Airborne Contamination Levels in Operating Rooms Equipped with Conventional and Laminar Airflow Handling Systems Air Settle Plates Conventional Oper . Room Wall-less LAF LAF with Walls LAF without Walls Wound 293 22 16 26 Back Table 312 8 2 1 1 261 90 2 2 2 250 87 4 7 3 315 133 35 58 4 298 117 24 86 5 190 54 116 135 6 259 70 84 99 To evaluate the problem more specifically, Table 2 summarizes only total-hip arthroplasty cases. Even though fewer cases had been done in unit B, there seemed to be little difference beween the two units or three types of laminar airflow rooms. Contamination levels as seen in-Table 3, even though not statistically significant, and definitely not yet affecting the infection rates, showed that a body exhaust system further reduced the environmental contamination around the operating site.
270 Merrill A. Ritter Harold Eitzen Morris L. V. French Jack B. Hart Table 2. Airborne Contamination Levels During Total-Hip Replacement Surgery Utilizing Laminar Airflow Wall-less LAF LAF with LAF without Walls Walls No. Cases 205 50 37 Air Settle Plates Wound 16 18 17 Back Table 6 2 1 1 100 2 2 2 90 4 7 3 147 35 58 4 124 24 86 5 60 116 135 6 81 84 99 The. number of CFU/ft /hr for just total-hip arthroplasty cases. The areas of ASP sampling were as described in Table 1. Table 3. Effect of a Body Exhaust System Used in Conjunction with Laminar Airflow on Airborne Contamination Levels Body Exhaust 2 Colony Forming Units/ ft /hr Off Body Exhaust On Set Up Evaluated Instrument Table Wound Site Instrument Table Wound Site Without Walls 2 (38)* 25 (37) 0.5 (38) 17 (38) With Walls 4 (40) 20 (40) 1 (71) 15 (77) * ( ) Number Tested
271 Horizontal Laminar Airflow The effectiveness of laminar flow on the surgical instruments (Table 4 and Table 5) was exemplified by the reduction in the number of con- taminated hemostats per case and per 15 minutes in the room with laminar air- flow versus the conventional ventilation (p < 0.05). In the conventional operating room, the number of contaminated hemostats was statistically (p < 0.05) less when removed with sterile pick-ups. There was no statistical difference between the number of contaminated hemostats in the laminar airflow room whether they were picked up with the gloved hand or sterile pick-ups. Spearman's cor- relation coefficient revealed a correlation between time and the number of con- taminated hemostats picked up by the gloved hand (.7517). Table 4. The Effect of Different Air Handling Systems on Hemostat Contamination Conventional Operating Room Laminar Airflow Operating Room Scrub Nurse's Gloved Hand 6.4 1.1 Sterile Pick-Ups 4.3 0.5 The average number of contaminated hemostats per case (12 sampled/ case). Table 5. The Effect of Different Air Handling Systems On Hemostat Contamination Conventional Operating Laminar Airflow Room Operating Room Scrub Nurse's Gloved Hand 5.33 0.92 Sterile Pick-Ups 3.58 0.49 The average number of contaminated instruments per fifteen minutes for ten cases .
272 Merrill A. Ritter Harold Eitzen Morris L. V. French Jack B. Hart Table 6 revealed a statistically significant reduction in the number of contaminated skin and deep knife blades in a laminar airflow operating room versus those from the conventional operating room. There was no statistical difference between the skin and deep knife blades contaminated from either group. Table 7 revealed a statistical reduction in not only the number of positive opening (p < 0.05) and closing cultures (p < 0.001), but also the num- ber of infections (p < 0.05) in total-hip arthroplasty cases done in a laminar airflow operating room. Table 6. The Effect of Different Air Handling Systems on Knife Blade Contamination Microbial Contamination of Knife Blades Laminar Airflow On Laminar Airflow Off Probability No. Tested No . Contaminated No. Tested No. Contaminated Skin 103 3 (3Z) 106 16 (15%) P < 0.005 Deep 209 9 («) 165 33 (207.) P < 0.005 Total 311 12 271 49 P < 0.005 The number of contaminated skin and deep knife blades tested from clean orthopaedic cases done in a conventional operating room with and without the laminar airflow system in use. Table 7. The Effect of Different Air Handling Systems on Opening and Closing Cultures and Postoperative Infections During Total- Hip Surgery Conventional Operating Room Mo. Tested No. Positive Percent Probability Opening Cultures 92 Closing Cultures 92 Infections 92 11 11.4 P < 0.05 22 23 P < 0.001 6 6.5 P < 0.05 Laminar Airflow Operating Room Opening Cultures 225 Closing Cultures 225 Infections 278 8 1.3 P < 0.05 15 6.6 P < 0.001 3 1.07 P < 0.05 The number of positive openiog and closing cultures and infections in 370 total-hip arthroplasties done with and without laminar airflow for more than one year.
273 Hotizontal Laminar Airflow DISCUSSION The data most interesting to laminar airflow skeptics are now becoming available, as noted in the answers to the previously posed questions: 1. Several horizontal laminar airflow units have reduced environmen- tal airborne contamination more than 93 percent (p < 0.005) and is further aided by the use of a body exhaust system. 2. Surgical instruments (hemostats, 82 percent, and knife blades, 80 percent) are kept statistically cleaner (p < 0.05) by the addition of laminar airflow in the operating rooms. 3. Opening and closing cultures both show a statistically significant (p < 0.05 and p < 0.001) reduction of 69 percent and 72 percent respectively in the number of contaminated wounds treated under laminar airflow conditions. 4. Our infection rates have been statistically (p < 0.05) reduced from 6.5 percent to 1.07 percent for total-hip arthroplasties done under horizontal laminar airflow conditions. CONCLUSION Not only does horizontal laminar airflow control and keep the environ- ment cleaner, but it also reduces the contamination of surgical instruments and wounds, thus lowering infection rates. REFERENCE 1. Ritter, M. A., M.L.V. French, J. B. Hart, and H. A. Eitzen, A systems analysis approach to postoperative wound infections. Clev. Clinic Quart. 40:211-227, Winter 1973.
274 LOCALIZED SURGICAL ISOLATORS (Sterile Environment Capsule) Dana M. Street* The most sterile operating theater might be found on the moon but is still too inaccessible. One can clean the air of the entire room as in our laminar flow rooms, or the immediate area of the patient including the surgeon and assistants but excluding the anesthetist as in the "greenhouse. " or just the region of the operative field excluding the surgeon with a lo- calized surgical isolator. The smaller the space, the easier it is to ob- tain sterility and the less the cost. My first experience with the use of isolators in surgery was in 1962 when associated with Jerome Landy who was using isolators for breeding germ-free research animals at the University of Arkansas. An isolator was exhibited at a meeting of the American Academy of Orthopaedic Surgeons in 1963, and a movie depicting intramedullary nailing of the humerus in an isolator was shown at the meeting in 1964. The isolator, measuring five feet long by three feet in diameter, was constructed of flexible but heavy transparent vinyl, and incorporated two pairs of long-sleeved, heavy, black rubber gloves (Fig. 1). The isolator or capsule was inflated by air passed through a High Efficiency Particulate Air (HEPA) filter. The undersurface of the capsule was cemented to the prepared operative field with Vi-spray, and the incision made through the floor of the capsule. The surgeon and assistants scrubbed their hands, but no other sterile precautions, such as gowns and masks, were considered necessary. A pass-through port was located at one end, but most of the instruments, towels, and sponges were placed in- side and gas sterilized with the capsule prior to surgery. While the above arrangement functioned satisfactorily, there were a number of disadvantages. The large capsule of heavy vinyl was cumbersome and expensive as were the long, thick, rubber gloves. The air blower and filter also were costly items. Therefore, in 1963, the feasibility of a light-weight, inexpensive, disposable capsule was investigated. The patient was prepared and draped in the usual fashion, and the team gowned and gloved, so that the capsule could *Professor and Chief, Division of Orthopedic Surgery, County of Los Angeles/Harbor General Hospital, Torrance, California.
275 Dana M. Street be removed at any time during the procedure if necessary. Wearing the usual operating gloves, the arms were introduced through sleeves in the capsule eliminating the heavy rubber gloves. The capsule was at first inflated with nitrous oxide from the anesthesia machine eliminating the air blower and fil- ter. Later, less expensive nitrogen was used. Fig. 1. Isolator for Germ-free Animals. 1. Heavy vinyl capsule; 2. Heavy, black, rubber gloves; 3. Pass-through port; 4. HEPA filter; 5. Blower.
276 Localized Surgical Isolators The first capsules were kitchen table fabrications of Saran Wrap and Scotch Tape, which built up an electrostatic charge attracting dust and lint. Later experimental models, by Aeroplast in 1965, were of a con- ductive film (Fig. 2). While designed to be disposable, the supply was limited, and they were used repeatedly by placing a large Vi-drape patch in the floor. New capsules were cemented to the field using Vi-spray, the patched ones by the usual application of the Vi-drape which held somewhat better. \J\J U U n n A a Fig. 2. Modified Isolator - Disposable, Germ-free Environment Capsule. 1. Lightweight plastic capsule; 2. Elastic in sleeves; 3. Snorkle intake. Gas from the N20 or N« tanks was thought to be sterile, but was passed through cotton plug filters in the intake tube which could catch any possible oil droplets, and also could be cultured for organisms. The cotton plugs tended to slow the rate of inflation. At first, the instruments were placed in a tray and covered with a separate plastic bag which was cemented underneath the nurse's end of the capsule and the floor was incised to get to them (Fig. 3). Later, it was found satisfactory for the nurse to pass them in through one of her sleeves, wrapping the sharper ones in a towel. The Bovie cord, suction, and fiber- optic cord were also carried in through sleeves. The seal between the end of the capsule sleeves and the surgeon's gloves was held by elastic around the wrists. When the nurse withdrew one arm in order to introduce an
277 Dana M. Street instrument she would close the empty sleeve with her other hand to prevent escape of gas and collapse of the capsule. Still another method for larger cases was to open the end of the capsule, introduce a tray of instruments, then seal it again. Fig. 3. Capsule in use. Standard draping and gowning. Instruments introduced through floor from tray in separate bag.
278 Localized Surgical Isolators Settle-plate cultures were obtained from various locations in the operating room outside of the capsule, and from two plates inside - one at the head end and one at the foot end near the nurse's sleeves. The wound was cultured just prior to closure in some cases, as was the surface of the capsule, inside and out. With a clean surgery infection rate of two percent and before the advent of the total hip, the need for such a sterile environment was not so pressing, and the project was discontinued in 1965 after only a handful of cases. In these cases the settle plates remained negative inside, positive cultures of Staph. epidermidis were obtained from the top of the capsule and plates outside ranged up to 40 colonies. Colony counts vary with loca- tion in the room and over a wide range, but average ten colonies per hour in most of our rooms. The project was reactivated in 1970 with the total-hip procedures, using the original supply of capsules from Aeroplast Corporation. In the next ten cases, using gas only filtered by cotton, three showed positive plates of three, ten, and nine colonies after three, one, and six hours of surgery. Unfortunately, the cotton plugs in these cases were not cultured. Also, plates freshly poured in surgery were used which-did not have time to demonstrate their prior sterility. No infection developed. However, there was sufficient cause to doubt the sterility of gas taken directly from the tank. In subsequent cases the gas has been bubbled through a 1/4 percent Neomycin in distilled water solution, and, with few exceptions, cultures inside have been negative (Table 1). Besides sterilizing the nitrogen, passing it through an antiseptic solution introduces additional benefits. First, the humidity is raised 75 to 85 percent; and the wound does not dry out as under a draft of low- humidity air. Second, the solution-laden atmosphere (particle count 300,000) may have an antiseptic action on the capsule and its contents (a settle plate exposed outside the capsule for one hour then taken inside for one hour was reduced to three colonies, while a plate beside it outside grew 46 colonies using 1/4 percent Neomycin. Other more effective solutions are shown in Table 2).
279 Table 1. Case # Col. Dana M. Street Colony Counts Inside Capsule Case # Col. Case # Col. 7 0-0 24 0- 36 0-0 8 0- 25 0-0 37 1-0 9 10-10 26 0- 38 0-0 10 0-0 27 0- 39 0- 11 0-0 29 0- 40 0- 12 0-0 30 0- 41 0- 13 0-0 31 0- 42 0- 14 9- 32 1-0 45 0- 15 1-0 33 0- Total colony counts on settle plates in 26 cases in which counts were recorded out of 45 total surgeries ranging from 1-6 hours. Known tech- nical break in cases 9, 14, & 32. No infections. Table 2. Effect of Solution on Colonies Solution Gas Gas Caps. Hum. Vapor Control P1. 15 30 45 60 N2 Humid. min min min min 1/4% Y! 21% 62% 41% 6 8 6 2 6 Neo. " G 53% 81.5 28.5 8 13 21 10 7 " YI 27% 85 64 13 10 14 6 35% G 53 86 33 8 10 4 3 6 Ale. 4% Y2 15.5 83 67.5 15 8 6 5 4 Neo. 70% ii n 73.5 58 39 10 9 0 0 Alc. 10 min 20 min 30 min 40 n + I2 n n 75.5 60 5 5 2.7 0 0 10 min 20 min 30 min 40 min 3.3 2.3 1 0 0 Settle plates exposed in operating room then brought inside capsule and exposed for stated periods.
B K .. 280 Localized Surgical Isolators Disadvantages of the surgical isolator are: 1. A sense of claustrophobia on the part of the surgeon. 2. Visual impairment a. from reflective glare (eliminated by fiber- optic lamps near or inside the capsule. b. from blood splatter on the inside surface (cleared with a damp sponge). Fogging has not been encountered The disadvantages are minor compared with the advantages of true sterility and lack of dehydration. While admittedly this was a preliminary study with only 51 ex- periences involved, it demonstrates great potential for sterility at the wound site at the very low cost of an estimated $20-$30 per patient.
281 VERTICAL LAMINAR AIRFLOW SYSTEM Robert S. Turner* INITIAL SURGICAL APPLICATION William Randolph Lovelace II, co-founder of the Lovelace-Bataan Medical Center, and Willis Whitfield, discussed applications of laminar air- flow rooms in December, 1961. Bacteriologic studies were started in April, 1962, in a laminar air- flow room at Sandia Laboratories. On the basis of the studies, it was concluded that a laminar airflow operating room would reduce the possibility of airborne contamination from pathogenic bacteria. The vertical laminar airflow system was then installed in an existing operating room in the Bataan Memorial Hospital in Albuquerque, and has been used since January 3, 1966 (Fig. 1). Because of operating room regulations and the permanent flooring in the existing operating room, it was necessary to adopt an outflow system of peripheral vinyl drapes (Fig. 2). After the vertical laminar airflow system was installed, air sampler bacterial counts were compared in a conventional operating room identical in dimensions to the laminar airflow operating room (Talbe 1)". Settling plates were also used to compare bacterial colony counts in the conventional operat- ing room with the vertical laminar airflow operating room (Table 2). Air sampler studies demonstrated that almost no bacteria existed on the floor at the foot of the operating table, even though no hoods or special aspirating devices were used, when the vinyl drapes were closed (Fig. 3). The incisional area was noted to be nearly free of viable bacteria particles dur- ing these air sampler studies (Fig. 4). *Chairman, Department of Orthopaedic Surgery, Lovelace-Bataan Medical Center, Albuquerque, New Mexico.
282 Fig. 1. Vertical Laminar Flow Operating Room, Bataan Memorial Hospital, Albuquerque, used since January 3, 1966. High Efficiency Particulate Air (HEPA) filters remove par- ticulate matter 0.3 microns or greater in diameter. VERTICAL LAMINAR FLOW OPERATING ROOM (CURTAIN TYPE) AIR SUPPLY PLENUM HEPA FILTERS -BLOWER -PRE-FILTER COOLED OR WARMED MAKE-UP AIR ^ â¢x,_r r r C C r/-/ ioiiiiii1i iiiiiiiiiiiiiiiiiiiiiiiiiiiiiii Fig. 2. Functional diagram of the Vertical Laminar Air Flow Operating Room adapted to an existing operating room. Note the pre-fliters, and the introduction of fifty new air changes to the recirculating system, allowing a total of 600 air changes per hour. c
283 Robert S. Turner Table 1. COMPARATIVE BACTERIOLOGICAL STUDIES Andersen Sampler Technique Bacterial Count per Cubic Foot of Air per Win. Time Conventional O.R. Ultra-Clean O.R. Pre-op. determ. 3.0 bact. per cu. ft./1 0.3 bact . per cu. ft./1 1st 10 minutes 15 4 n n n n 1.3 " n M n 2nd " 15.1 " " " " 0.4 " n M n 3rd " " 10.0 " ' 0.7 " n n H 4th " 15.7 " " " " 0.4 " n n H 5th " " 11.3 " " " " 0.4 " n n n 6th " 23.4 0.6 " n n n 7th " " 16.0 " " " " 0.3 " 8th " " 10.5 " " " " 0.4 " n n n 9th " 9.3 0.2 " n M n AVERAGE 14.4 bact. per cu. ft./1 0.5 bact. percu. ft./1 Table 2. COMPARATIVE BACTERIOLOGICAL STUDIES Settling Plate Technique Anesthesia table Top of cautery unit Operative field Sterile Mayo stand Sterile back table Open room Conventional O.R. 4 colonies/plate 9-14 " " 12 11-19 7-21 " " 7-10 " " Ultra-Clean O.R. 0-1 colony 0 colony 0 colony 0-1 colony 0 colony 0 colony AVERAGE 9-12 colonies/plate 0 colony
Vertical Laminar Airflow Systems 4-r <r Ki .. i- 3 u. x V) 2 y Pre-operotive preparation 0700 Sampler on floor-foot of table "Patient brought into room Surgical team enters lastic curtains closed* room air laminar down-flow Clock motor failed- sampling stopped 0800 0900 TIME (HOURS) 1000 1100 Fig. 3. Aortic Vascular Surgery Case. Note the marked decrease of the viable particle count when laminar airflow was obtained with closure of the vinyl drapes. Sampler on surgeon's side of NUMBER OF table- sampling probe at incision V1ABLE PARTICLES/ CU FT OF A1R Sampling probe Sampling probe 1 - placed at incision removed o ooco-o-o-oo-oo o-o-oo-o-o Q- i i 1 i i i 1 i i i ⢠⢠â 1 | TââI 1 | 0700 080O 0900 1000 1100 TIME (HOURS) Fig. 4. Aortic Vascular Surgery Case. Note the near absence of viable particulate matter at the wound site even though no operating hoods or aspirating devices were used.
285 Robert S. Turner REVIEW OF RESULTS (POSTOPERATIVE INFECTION RATES) The postoperative surgical results were those obtained in a community service-type hospital, with approximately 25 to 30 percent of the patients re- ferred from outside the Albuquerque community. The laminar airflow room has been used predominantly by several surgical specialists in a multispecialty group practice in a medical-center-type setting. There has been a definite tendency to use the laminar airflow operat- ing room for major cases requiring considerable operating time. Over a 54-month period, from October 1966 to March 1971, there was a postoperative infection rate of .79 percent after 3,408 operations performed in the vertical laminar airflow operating room. The infection rate was also low for two control operating rooms, as there was a .93 pecent postoperative infection rate after 4,162 operations in one control room and a 1.14 percent postoperative infection rate after 4,091 operations in another control room. Thus, there was a 1.04 percent postoperative infection rate following the 8,253 operations in the two control rooms (Table 3). Table 3. VERTICAL LAMINAR FLOW OPERATING ROOM October 1966 - March 1971 54 Months Room #1 (Lam. Flow) 3408 operations 27 infections Room #2 (Control) 4162 operations 39 infections Room #3 (Control) 4091 operations 46 infections % Rate 0.79 0.93 1.14
286 Robert S. Turner These figures were collected on a monthly basis on a confidential report form that listed the operations done during a specific month by a specific surgeon. The surgeon graded the operation as to whether it was a clean or contaminated procedure, and stated the outcome of the operative wound healing. The figures were tabulated monthly. No reports were entered in the patient's clinical chart. This form of reporting was started by Whitcomb, and is being continued with the full cooperation of the surgical staff. Bataan Memorial Hospital expanded its surgical suite in 1971, and a horizontal laminar airflow operating room was constructed with dimensions of 7.92 meters by 6.70 meters (26 by 22 feet). The horizontal laminar airflow operating room has been used since May 24, 1971, and its recent use has been much more intensive than the vertical laminar airflow room, because its larger dimensions allow better surgical working conditions. From January 1, 1966 to January 1, 1973, there were 284 spine fusions performed in both laminar airflow operating rooms, with one postopera- tive infection. During the seven-year time interval from January 1, 1966 through January 1, 1973, 207 hip arthroplasties and total-joint replacements, as well as 45 knee arthroplasties and total-joint replacements, were performed in the laminar airflow rooms with five postoperative infections. The post- operative infection rate for these selected 536 major orthopaedic procedures performed in the laminar airflow operating rooms was 1.12 percent. From January 1, 1966 to January 1, 1973, there were 28,485 surgical operations at the Bataan Memorial Hospital with 334 postoperative infections for an overall postoperative infection rate of 1.17 percent. The postoperative infection rate of .72 percent following 4,400 operations performed in both laminar airflow operating rooms from January 3, 1966 to January 1, 1973 compared with the postoperative infection rate of 1.205 percent after the 24,085 operations performed in all of the conventional operating rooms in the Bataan Memorial Hospital during the same time period, is significantly different at the p^O.05 competence level. The validity of this comparison is questionable, however, because of the numerous variables involved.
287 Vertical Laminar Airflow Systems PROBLEMS WITH USAGE OF THE VERTICAL LAMINAR AIRFLOW OPERATING ROOM Portable lighting was necessary for use as the high efficiency particulate air filters were located in the ceiling. Usually two portable operating room lights were required. This increased the amount of congestion in the 10 by 12 foot work area. Circulating nurses had to open and close the vinyl drapes for ingress and egress. There was a definite tendency to leave one or more of the drapes partly open to facilitate this activity. Often, an exterior door would be left open for the same reason, and this would obviously alter the effective- ness of the airflow system. Working space was found to be rather cramped as anesthesia equipment and the instrument tables were all placed within the enclosure. Surgical nurses felt ill at ease, for fear of contaminating their work areas through con- tact with the adjacent vinyl drapes or other equipment in the working area. The vinyl drapes are placed on a track so that they can be opened and closed. When closed, some personnel had a "closed in" claustrophobic-type feeling. The vinyl drapes only partially obscured visibility. PROBLEMS OF ACCEPTANCE OF THE VERTICAL LAMINAR AIRFLOW ROOM BY PERSONNEL Almost all personnel were immediately aware of the sound level of approximately 65 decibels (A scale). Many nurses, anesthetists, and anesthesio- logists expressed reluctance to work constantly in the room because of the sound level. As personnel turnover occurred through the years, it was noted that constant re-education was needed for the operating room staff to aid in under- standing the principle of how the vertical laminar airflow system worked. This re-education was also necessary to aid the personnel in understanding the potential significance of eliminating airborne pathogenic bacteria. SPECIAL CONSIDERATIONS A few patients with scientific background who live in the Albuquerque area are aware of the laminar airflow operating room system and have inquired as to whether their surgery would be performed in such a room.
288 Robert S. Turner Lawyers in the Albuquerque area are also aware of the potential significance of laminar airflow operating rooms. To the best of my knowledge, no decision has been made by any court in New Mexio relating to the use of, or failure to use, a laminar airflow operating room. .
289 MICROBIOLOGY OF HUMAN SKIN AND ITS RELATION TO POSTSURGICAL INFECTIONS John A. Ulrich* The work of numerous investigators has supported the observation that the surgical team is the primary source of microbial aerial contamination in the operating room. During the pandemic of Staphylococcus aureus, 80/81, infections in the late fifties, it was believed the nose was the prime site in which this organism was carried. It is true that Staphylococcus aureus did reside and was found in the nares of many operating personnel. However, after carrying on a project for more than two and one-half years, sampling the personnel in surgeries at least once a week, and after more than 14,000 cultures, we had a correlation of only .13 between nasal carriers and infections originating in the surgeries. It was obvious that this was not the major source of the organism causing our problem, so we sought elsewhere for the real source. The most obvious source was one of the largest organs in our body which is in constant contact with the environment: the skin. Thus, we began to study the skin to see if it was one of the problem sources of potential pathogens in surgery. The human skin is an involved structure with a complex bacteriology. It is more than just a contaminated surface. First, there are microbial popula- tions classified according to source which can be divided into two different types, transient and indigenous. A transient population is one you pick up from contaminated external sources and then transfer elsewhere; it can be washed off rather easily. Secondly, there is also an indigenous population which is always with us. It is one of our important main lines of defense against other invading organisms. Therefore, we do not want to destroy these bacteria; we only want to control them in operating rooms and other sensitive areas. Most of this population does not grow on the surface of the skin. It comes from a deeper source, usually the sebaceous glands and the ducts of the sebaceous glands. It is expressed to the surface via the hair follicles. We use a number of techniques to track down the dynamics of the bacterial population of the skin. The contact plate method is very simple and gives good reproducible * Chief, Microbiology Service, University of New Mexico, Albuquerque, New Mexico.
290 John A. Ulrich data. It is nothing more than an aluminum milk-bottle cap filled with agar. The agar is placed in contact with the skin, immediately removed and incubated, and then quantified after incubation. Using this technique we can develop full-body patterns for the human being (Table 1). Notice that the bacterial colony counts on the forehead, temple, eye, nose, upper lip, cheek, jaw, ear, chin, back of the neck and the neck are relatively high counts. Since these are total colony counts on a plate the size of a milk-bottle cap, it indicates that the skin has a high density of the bac- terial population in these areas. The axillary population is much higher in most people than indicated here. The use of underarm deodorants controls and reduces the population in this area. Normally, without the use of deodorants, the bac- terial population of the axilla is a little higher than around the head and the neck. The trunk and the outer arms have relatively low populations. The hand is high. This is due not only to the indigenous population, but also to the transient population. Generally, most of the lower parts of the trunk and the legs are relatively low in microbial population. Areas which are high include the groin, perineum, navel, and buttocks. The inner thigh, the ankles and feet, although not listed, are found to be high population areas. In summary, each human being has essentially the same pattern of distribution with very high bacterial populations on the head and the neck, the axilla, the hands, the peri- neum, the groin, and the feet. The other areas are less heavily contaminated or populated. Note that one of the most highly populated sites is the angle of the jaw. Whereas everyone has essentially the same pattern of bacterial popula- tion on the skin, each of us is individual in the number of bacteria per square centimeter that we carry. Table 2 is a compilation of data gathered from a num- ber of individuals, both males and females. Some have relatively high counts; some have relatively low counts. Whatever the level of the count, it remains at the same level for long periods of time. It can be changed from time to time by special techniques which I will not discuss. The important aspect is that each person has a unique and stable microbial level which can be shed into the air in the operating room. Usually an individual who has a very high skin count is also a good shedder. Each of us constantly sheds into the air. *
291 Microbiology of Human Skin Table 1. Bacterial Skin Populations (Contact Plate Method: Right Side) Plate Count Body Region Site Head and Neck Forehead 348 Temple 560 Under eye TNTC* Nose TNTC Upper lip TNTC Cheek 584 Angle of jaw TNTC Behind ear TNTC Under chin TNTC Back of neck 211 Neck 316 Upper trunk and arms Axilla 106 Upper arm â lateral 42 â volar 8 Elbow â lateral 8 â volar 24 Forearm â lateral 41 â volar 11 Wrist â lateral TNTC â volar 53 Hand â lateral 224 Palm Pseudomonas Lower trunk and arms Shoulder 43 Subclavicular 83 Breast 50 Lateral chest 15 Lateral waist 15 Groin TNTC Perineum TNTC Navel TNTC Subscapular 128 Buttock TNTC Hip 104 Lateral thigh 232 Inner thigh TNTC Popliteal space 241 *Too numerous to count
292 Table 2. John A. Ulrlch Bacterial Skin Populations in Different Individuals (Contact Plate Method) 1M 2M 7M 4F 5F Jaw TNTC* TNTC 145 94 TNTC Arm 2 26 109 13 ** Breast 6 86 11 8** 278 Abdomen 47 219 29 8** 121 Groin 128 132 Cont. 74 113 Buttock 248 207 TNTC 106 52 * too numerous to count ** spore formers We cannot scrape or wash these organisms away because of the subsurface replicating population. The technique by which deep or subsurface populations are studied is called tape stripping. A sticky sterile tape, made by the Minne- sota Mining Company, called Tape No. 850, is used. Regular Scotch tape cannot be used as it contains antibacterial agents. The sterile tape is removed from the roll by the operator and pressed onto the skin, smoothed over with a sterile applicator, and removed, put into a Petri dish, agar poured over it, incubated and quantified. Each time one of these tapes is applied to the skin and pulled off, it removes a layer of cells and any bacteria that happen to be in that layer. Normally 20 layers of cells are removed serially in these studies. You cannot remove further layers, because minimal bleeding occurs. The type of data ob- tained using this technique is shown in Table 3. These are bacterial counts from the tape strips. Note again, the varying bacterial levels among subjects. Some people are relatively low-level carriers; some are relatively high; and some are intermediate in levels of populations they carry. After removing ten layers of cells, relatively large numbers of bacteria are still present in the skin. Thus, the bacteriology of the skin is not only a surface phenomenon, but much more complex. In between tape 10 and tape 11, a series of six two- minute surgical scrubs were performed, and yet we continued to find demonstrable
293 Microbiology of Human Skin numbers of bacteria 20 cell layers down. These subsurface organisms cannot be controlled at the present time. These are the organisms that repopulate us and that we want to maintain. Table 3. Deep Bacterial Skin Populations (Bacterial Colony Counts per Tape: Tape-Stripping Method) Subject and Sex Tape XII M XIV M XV F XVI F 1 72 157 167 35 2 36 145 182 36 3 31 167 260 88 4 51 115 201 46 5 23 160 215 41 6 20 42 132 27 7 29 39 86 10 8 29 56 101 42 9 23 58 61 43 10 43 55 57 37 11 4 24 5 10 12 1 20 3 7 13 5 12 5 11 14 14 8 6 13 15 3 8 8 6 16 8 6 7 12 17 8 10 5 8 18 5 3 7 6 19 6 10 5 7 20 5 6 5 15 The next technique is known as the Price technique and it is a modified surgical scrub. When Price first published this technique, he per- formed ten two-minute surgical scrubs. The amount of time to put on the detergent, to scrub, and to flush are checked by a stopwatch. After the two minutes, the detergent is flushed from the skin into the measured amount of
294 John A. Ulrich water and the bacteria are quantified from the basin contents. Table 4 is a compilation of the type of data gathered by this technique. Table 4. Removal of Bacteria from Hands and Arms (Mean Bacterial Count per Basin: Price Technique) Subject, Sex, and Number of Tests Basin XIII M (43) XIV M (18) XV F (25) XVI F (29) 1 40* 465 194 196 2 64 295 182 219 3 71 214 156 210 4 65 183 133 175 5 62 154 129 170 6 44 147 124 151 * Multiply by 105 for total count in basin ' Each one of these basins represents an individual surgical scrub. After the first two-minute scrub, subsequent scrubs are done consecutively, one right after the other. These particular data were gathered while using castile soap without a germicide. The observation that is important and interesting is that skin bacteria cannot be completely removed even through six serial scrubs. One of the reasons is the deep population. The ducts and the glands act like little tubes of toothpaste; and as you scrub, massage, and squeeze, bac- teria are expressed onto the skin surface. Eventually, the surface population will be reduced to a minimal level, but it will continue at that level through subsequent scrubs. Apparently a triggering mechanism takes place as bacteria are expressed from the deeper sources and rapid reproduction commences. This indicates, even with the use of germicidal soaps, that this population cannot be destroyed. It remains in the deeper sites, and the surface population is replaced in a short period of time.
295 Microbiology of Human Skin f Microbial shedding from humans is more germane to the intent of this meeting. The discussion on skin microbiology is to indicate what happens to these bacterial populations that are normally on skin. Some of them die off, but most of them are shed (6). The average individual with normal skin sheds on the average 10,000 viable particles every minute (2-8). With abnormal skin such as an eczematoid process or folliculitis, the shedding rate increases tremendously. Fortunately, in surgery, barrier techniques are employed. Gowns, caps, gloves, and shoe covers are worn and these barriers reduce the amount of material shed into the air. Cotton gowns are imperfect barriers, and examina- tion of these cloths with a magnifying glass shows large openings through which bacteria may penetrate. A number of layers form a somewhat better barrier. Even though gowned and draped, personnel still shed bacteria into the air, and these must be controlled in another way. Surface sampling of floors in surgical suites indicates that the majority of bacteria recovered are human skin types. Gram-positive cocci and diphtheroids are most common, but gram-negative bacteria are noted by their absence. Most of the bacteria on the floor arrive by the airborne route. We should know more about airborne bacteria in surgery. Figure 1 shows data gathered using slit-type quantitative samplers. Two samplers were used simultaneously. This operating room did not have laminar airflow, but had an ordinary turbulent type of air delivery. Counts from the samplers followed one another closely, even though they were placed in different sites in the room. The level of organisms in the air before the surgical team came in was one organism per cubic foot. This was a low level with only the Case1la operator in the room, and a few other personnel in and out. The symbol "S" signifies a coagulase (+) Staphylococcus on that particular plate. After the team arrives and begins work on the patient, the airborne bacterial count increases. The data in Figure 2 were gathered at a time when one surgical case was being finished and another started. There is an average of ten surgical personnel moving about the surgery suite. The patient has under- gone surgery and he is undraped at this point, and of course the activity in a surgery increases greatly at this time. Nurses move around, picking up the
296 John A. Ulrich nrt rort -1 In rHtdrum Prior to Ht up of optratlna. room ⢠SompKrI (on goll.ry Mdgt) o SomplorEditor center of room) ©CooguloM poiltivt ttophylococcl in In f 10 19 20 25 JO 55 40 45 »0 5S 800 ! Tkiw Fig. 1. Preoperatlve microorganism levels. Efftcl ol activity In op«at.ng room Sompl«I(<o<ltndofl«loil Sompltr n (southwMt comer of room) ,00 9 to I9 20 29 V> J9 40 49 90 99 IQOO Tlmt Fig. 2. Effect of activity in operating room on microorganism levels.
297 Microbiology of Human Skin dirty instruments; the patient is undraped, picked off the table, and put back onto the cart. The floor is mopped. There is an excellent correla- tion between the degree of activity and the airborne bacterial population. This is probably due to two causes: we are scuffing up settled organisms from the floor, and probably more important, as our activity increases, our clothes rub and contact the skin and we break off more skin particles. The legend indicates one patient is taken out and another brought in. When a patient is brought in the activity is very great (moving him to the table, scrubbing and getting him draped). New instruments are brought in and tables set up. There is a great deal of movement and so the airborne bac- terial population stays high. As soon as personnel activity slows down, the number of organisms in the air drops rather dramatically. The samplers, in spite of being in different parts of the room, collect similar numbers of or- ganisms in the same time periods. One of the important aspects in maintain- ing low aerial counts is to reduce personnel activity as much as possible. Once surgery is in progress activity goes down and again we reach a level of one organism per cubic foot. Another important factor is the number of people present during the surgery. The data in Figure 3 show that more airborne microorganisms are re- leased if more people are present. These data were gathered during an inter- esting case which had an increased number of spectators. Normally, the sur- gical teams comprise about ten people, but in this case there were about double that number. These observers were standing quietly. The airborne bacterial levels, instead of being around one per cubic foot, averaged at least ten times the expected number. In an ordinary turbulence air system operating room, it does not matter whether these people are sitting in a gallery or standing around the table. Bacteria, being the size that they are, follow the gas laws and diffuse much as gas does. Although there may be a point source, these organisms will fill the room in a relatively short period of time. The above is true in any conventional turbulent flow surgery, but not in the laminar flow type that we will be discussing during the meeting. The levels of airborne bacteria in operating rooms are usually very low compared to other hospital areas. Figure 4 shows the microbial aerial com- bination in a surgical hall. It is of interest that the number of people
298 John A. Ulrich traversing a certain point is much greater than expected. The levels in the hall average 15 bacteria per cubic foot. Many coagulase (+) Staphylococci are present in the hall air. Most of the organisms found in hall samples were human skin bacteria. â¢etivlly if* pxiannil ln optrolmg room © CoofluKm poiitin @ Numtor ol pi-ioii In i â¢* © ® 41 ss © © ⢠Sompltr I (en MII ind of oollwy i«dg>] o Sonpltr E (MMlhMBl coratr of room] ttl0 » JO M 40 4» go n ,00 t 10 II «0 II M Si 40 Fig. 3. Increased activity and personnel in'operating room. Surgical Mil 9*1 floor i $ (S) Oo«ylOH | Aytoconi (is) Numbw ol P..IWI ln Ml (JtHkto, Mmp«n« >vto4 . Sompl.r : iooii.m O.R â¢> © o Sompltr C locroil from oH.:« l J 1 1 1 40 45 SO it ,,00 5 10 1S M » JO J9 Fig. 4. Levels of microorganisn in surgical hall.
299 Microbiology of Human Skin The data in Figure 5 indicate what happens when an infected patient Is brought into surgery. This was a patient with a severe staphylococcal in- fection. Within a minute and a half after arrival (the bandages were not yet removed), the samplers began to pick up the phage-type of staphylococci found in his wound. He continued to shed these organisms all during the procedure in relatively large numbers. Interestingly enough, two minutes after the pa- tient was removed from surgery, we could not find the organisms in the operat ing room (walls, floor, or air). They had most likely been swept out in the exhaust air. <w â¢-,> .â¢!- dop otxaii 1 Sompl.r I (toil .na of :idg<) iw< i- o Sompltr O (vttt ,,00 S 10 Fig. 5. Patient with deep abscess.
300 John A. Ulrich The rapidity with which microorganisms can be disseminated is indicated in Figure 6. This was a rather fortuitous-type of situation in which the surgeon happened to cut into a deep abscess of which he was not aware. One minute after the incision the phage-type of organism that was isolated from the abscess was demonstrated on the sampler plates. Despite complete removal of the abscess, the organism was present in the operating room air as long as the patient was present. As soon as he was removed from the room the staphylococcus disappeared. The important aspects are the speed with which the organism disseminates, and that it is present in sufficient numbers for easy sampling. Polltnt ..Ih dmlnlng wxind Somptolltoil md of golltry ltdgt) o Sompltr n(vnt sidt of ltdgt) <D COOO.UIOM po.iir.e ilophytococci 90 99 2°° 9 '0 <9 20 29 Fig. 6. Patient with draining wound.
301 Microbiology of Human Skin Table 5 shows data that relate surgical team personnel to infections originating in surgery. Twelve patients were infected by one surgeon in a peridH of two weeks with the same phage-type staphylococcus. The surgeon had subsurface boils that had not yet erupted, but he was shedding large numbers of organisms. His skin was checked using contact plates and in Table 6 it was noted he was colonized with the proper phage-type of staphylococcus. Every site sampled during the third week was positive. At five weeks the jaw was negative, but the other sites were positive. At six weeks some sites be- came negative, and at nine weeks the 80/81 staphylococci remained only in the area where the boils originated. During this period of time he received auto- genous vaccine because the staphylococci could not be controlled or removed with antibiotics. Table 5. Surgical Infections Related to a Shedder Thursday Surgical Tuesday Thursday Schedule Friday Saturday Thursday Pt. 1, Sa Pt. 2, Sa Pt. 3, Sa Pt. 9, Sa Pt. 10, Sa Pt. 12, Sa Pt. 4, Sa Pt. 11, Sa ^ Pt. 5, Sa Pt. 6, Sa Pt. 7, Sa Pt. 8, Sa Table 6. Colonization of a Staphylococcus Aureus Shedder Site Duration Weeks 5 6 Nose 80/81 52/52A/80/81 52/52A/ 80/81 47/53/54 Jaw 80/81 Negative Negative Negative Arm 80/81 80/81 Negative Negative Shoulder (front) 80/81 Negative Lesion (rt. chest) 80/81 80/81 Negative Back (midline) 80/81 80/81 Negative Waist 80/81
i 302 John A. Ulrich The fact that he carried these organisms on the skin is not of prime importance. The important aspect is the shedding of the pathogen into the air. We differentiate between carriers and shedders, and find that carriers, by and large, in surgery are not dangerous. It is only when they become shedders that they become dangerous. The shedding studies shown for this individual in Table 7 were performed three weeks following the outbreak. In doing a shedding study the individual is placed in a small room with an air sampler, and the air is sampled for three different ten-minute periods. First of all, he sits quietly, then he moves around the room rather moderately, and finally he moves about rapidly. Some non- typable staphylococci and some not related were isolated from the room before the subject was allowed to enter. When sitting very quietly he was not shed- ding. When he started to move about, he began to shed and we began to pick his organisms out of the air. The same tests were repeated at six weeks when his skin became negative (except around the waist) and he was not shedding. He was allowed to go back into surgery with no further problems, even though he did continue to carry the 80/81 staphylococci on the skin in the area of the healed boils. Activity Table 7. Shedding Studies Total Count/ Cubic Feet Phage Type Room only Sitting Moderate movement Rapid movement Room only Sitting Moderate movement Rapid movement 6 13 41 51 Six Weeks 3 6 12 10 47/53/83, no type No type 80/81 80/81 (2) No type No type
303 » Microbiology of Human Skin In summary, the human being is the important source of airborne organisms in surgery. It is very rare that the mechanical equipment, the walls, or the instruments are sources if they are cared for properly. Most of the airborne organisms in surgery are gram-positive. What has been discussed here refers only to an operating room and not to other parts of the hospital, such as nursing stations where direct contact of individual to individual, nurse to patient, and then on to another patient, is probably the prime method by which organisms are spread. In an operating room, the use of barriers indicate the airborne route to be the most important. REFERENCES 1. Bernard, H. R. , R. Speers, Jr., and F. W. O'Grady, et^ ai.. , Modification of operating room clothing and by ultraviolet irradiation. Lancet 2:458 September 4, 1965. 2. Bethune, D. W., R. Bowers, and M. Parker, Dispersal of Staphylococcus aureus by patients and surgical staff. Lancet 1:7383:480, February 27, 1965. 3. Duguid, J. P., and A, T. Wallace, Air infection with dust liberated from clothing. Lancet 2:6535:845, November 27, 1948. 4. Hare, R., and G. C. A. Thomas, The transmission of Staphylococcus aureus. Brit.Med. J. 2:4997:840, October 13, 1956. 5. Noble, W. C., and R. G. Davies, Studies on the disposal of staphylococci. J. Clin Pathol. 18:16, January 19, 1965. 6. Riemensnider, D. K., Space craft sterilization technology. NASA SP-108, 97-103, 1966. 7. Selwyn, S., and D. Chalmers, Dispersal of bacteria from skin lesions: A hospital hazard. Brit. J. Dermatol. 77:349, July 1965. 8. Speers, R., H. Bernard, F. O'Grady, and R. Shooter, Increased dispersal of skin bacteria into the air after shower-bath. Lancet 1:7383:478, February 27, 1965.
304 LOCAL CONTROL OF AIRBORNE BACTERIA J.C.N. Westwood* E. Griddle E. J. Synek P. Neals AIRBORNE CONTAMINATION IN POSTOPERATIVE WOUND INFECTION The Significance of Airborne Contamination The relative importance of airborne bacteria contamination in causing postoperative wound infection cannot be settled with any finality since it depends on local operating room conditions, and the techniques and discipline of a particular operating team. The better these are, the more relatively im- portant will be the danger from airborne contamination. Even under standard conditions for general surgery, however, it is indisputable that airborne con- tamination can result in infection, and the author has yet to hear any other acceptable explanation for the development of streptococcal infection derived from the anus of the anaesthetist (8,14,15,19). Whatever may be its relative importance under specified conditions, an analysis of the possible causes of exogenous infection makes it clear that airborne bacteria cannot be safely ignored. The basic facts are simple. For postoperative wound infection to occur, bacteria must reach the wound. They may do so either endogenously or exogenously. The former cannot be prevented by aseptic technique, the latter can. Bacteria from the external environment can reach the wound only if carried on some object which makes physical contact with it. The number of objects making such contact is limited and the possible mechanisms for their contamination are also limited. They may be listed as follows: 1. The skin of the patient Ineffective sterilization or isolation of the patient's skin may lead to direct or indirect contamination of the wound. An integral part of aseptic technique entails measures to prevent this occurrence. 2. The hands of the operating team These are isolated by impervious sterile surgical gloves subject to only four routes of contamination: Department of Microbiology and Immunology, Faculty of Medicine, University of Ottawa, Canada.
305 J.C.N. Westwood E. Griddle E. J. Synek P. Neals a. Sterilization failure. b. Puncture due to faulty manufacture or damage during the operation. c. Contact with a contaminated object. If the object is external this is a breach of aseptic technique. Con- tamination by contact with the surgeon's gown (3,4) through which skin bacteria have penetrated is a case in point. If the contaminated object is within the wound, then endogenously derived bacteria are already present. d. Airborne. The remaining objects which can act as vehicles for wound contamina- tion are the instruments, towels and drapes, sponges, suture material, irri- gating fluids, dressings, topical applications, and air. These are subject only to three possible routes of bacterial contamination - sterilization failure, contact with hands or body of scrub-nurse or surgeon, and via the air. This list may not be exhaustive, but the message is plain. Apart from the special risks of puncture of the operating team's gloves and failure to protect from the patient's own skin, only three sources of contamination can be identified and the first two of these represent breaches of standard aseptic technique. The third is the air. In fact, if standard technique is effectively carried out, then the airborne route of contamination, either directly to the wound itself or in- directly through the various intermediaries listed constitutes the only means by which bacteria from the external environment can reach the wound. Under these circumstances, it is clearly necessary to examine the sources from which the air itself derives its contamination, and there is no question that in the operating room, the source of primary importance is the operating team itself. Riemensnider and Sciple (17,20) have shown the startling degree to which the human body sheds bacteria-laden particles, and more recent studies (1,2,12) have demonstrated that these are mainly derived from the
306 Local Control of Airborne Bacteria perineal area, which is not only the external "cess-pit" of the human body with regard to enteric bacteria, but is also one of the primary sites of Staphylococcus aureus carriage. It was Charnley's recognition of the over- whelming importance of people as the source of dangerous contamination in the operating room that led to his success in controlling wound infection by physical isolation of the operating team (3,5). If it be accepted that the argument so far presented is soundly based and logically developed, then two conclusions are inescapable: 1. When standard aseptic technique is effectively applied, the airborne route is the only remaining one by which bacteria from the external environment may reach the wound. 2. The air is the only unsterile object permitted to make un- restricted contact with the wound and with every other object contacting the wound. To these conclusions may be added the observation that they are not wholly theoretical, since episodes exist even in general surgery which can be explained in no other way than by airborne contamination. If it be granted that the air is in fact a vehicle for bacterial contamination, one must decide whether it should be controlled and, if so, how? Techniques of Air Handling Modern technology has made the production of adequate quantities of sterile air a simple matter so that the achieving of a bacteria-free environ- ment at the wound site has become simply an exercise in ventilation technique. In the classical approach, filtered air is introduced peripherally to the op- erating room, and is also scavenged peripherally. With turbulent ventilation, about 60 percent scavenging efficiency is the effective limit, but the intro- duction of the laminar-flow ventilation principle has raised this limit too close to 100 percent provided that a rectilinear space is maintained without physical interruption of the airflow. The room must in fact be rectangular or cylindrical and empty. Any departure from these conditions will disrupt the laminar-flow pattern, and tend toward turbulent ventilation. Clearly, the ideal cannot be achieved in an operating room. Furthermore, any ventilation
307 t J.C.N. Westwood E. Criddle E. J. Synek P. Neals system in the operating room relying on the peripheral introduction of sterile air leaves; the operating team the main source of contamination, because of its location between the sterile air supply and the wound site which is to be pro- tected. Not only does the latter represent a ventilation deadspace, but it is also the seat of rising convection currents, termed "wound thermals," and demonstrable by Schlieren photography, whose velocity exceeds that of achiev- able laminar-flow systems (9). It was the recognition of the surgical team as the major source of bacterial contamination, and the impossibility of achieving laminar flow, that led Charnley to advocate the complete isolation of each member of the team with an individual airflow vented to the exterior. Despite these limitations, a tremendous amount of work has been put into the application of laminar-flow ventilation to surgery, with marked and clearly demonstrable reduction of bacterial air loading even at the wound site (6,7,13,16,18,21). In 1968, Marsh and Beck (11), recognizing the primary source of con- tamination and the consequent basic weakness of any ventilation system relying on a peripheral source of air, introduced the logical concept of generating the supply of sterile air at the wound site itself. Under these conditions, an expanding bubble of sterile air would be generated between the critical area to be protected and the surgical team, so that contamination from the latter, and from any other source in the operating room would continuously be driven away from the wound, which would itself be constantly bathed in the flow of sterile air. Furthermore, the rising column of sterile air is aided rather than opposed by the "wound thermals." The work of my colleagues and myself represents an independent develop- ment of this concept to the point of practical applicability under operating room conditions. In the system which we have developed, sterile air is intro- duced from a simple fan-filter unit into a cloth channel sewn into the final surgical drape so that the wound site is surrounded by a diffusing collar. Sterile air emerges from this diffusing collar at a velocity of 100-200 ft/min, and effectively isolates the wound from external airborne contamination at
308 Local Control of Airborne Bacteria challenge velocities up to 300 ft/min. At this breakthrough point, quantitative assessments have shown a penetration of one particle in 20,000. Under normal conditions of operating room contamination, this would permit 70 hours of continuous operating time in a three-mile-an-hour wind for each viable particle penetrating. Under operating room conditions, this approaches complete protection. As an adjunct to the wound-site diffuser or isolator, it is also nec- essary to protect the instruments and other paraphernalia on the instrument trolley and Mayo stand, and the hands of the operating team. This is achieved by means of a modified "clean station" providing sterile air, which sweeps over the instrument trolley and unites with that from the wound-site diffuser, to give a sterile air corridor within which the surgical team operates. It should be emphasized that in the absence of the wound-site diffuser the clean station cannot give effective protection at the wound site. In our present design the two fan-filter units are housed in a single cabinet but operate independently, and either can, in fact, be used without the other. In an operating room already supplied with a laminar-flow sterile-air ventilation system, the clean station would be superfluous, and the wound-site isolator alone would be required. This could be housed in a unit of about 2 ft x 2 ft x 2 ft 6 in. The system is flexible, relatively inexpensive, and mobile. While its main value would appear to lie in critical orthopaedic or cardiovascular pros- thetic surgery, it could also provide an inexpensive means for upgrading sub- standard surgical facilities, and would be capable, in addition, of providing class 100 operating room facilities under field conditions in military or di- saster emergency hospitals. In conclusion, we would like to express a personal opinion. Whatever may be the virtues of one ventilation system over another, there can be no justification in critical surgical procedures for tolerating evident hazards which can be eliminated by simple existing technology. The American College of Surgeons has recently taken the position (10) that "there is no convincing evidence at this time that laminar, clean airflow, in itself, has a favorable influence on the incidence of surgical wound infections." We would like to
309 J.C.N. Westwood E. Criddle E. J. Synek P. Neals suggest that the issues with which we are now concerned no longer lie, for logistical reasons, within the realm of effective statistical evaluation, and statistical impotence is not a valid excuse for the rejection of simple con- clusions based on known facts. More legitimate is their rejection on grounds of cost benefitâthat the benefits to be gained do not justify expenditure on special air-handling systems. Compromise between the desirable and the prac- tical is a familiar situation, but should be accepted for what it is, and should not be shrouded in political smoke screens aimed at facilitating evasion of the issue. REFERENCES 1. Ayliffe, G.A.J. , J. R. Babb, and B. J. Collins, Dispersal and skin carriage of Staphylococci in healthy male and female subjects and patients with skin disease. Proc. IVth International Symposium on Aerobiology. "Airborne transmission and airborne infection." J. F. Hers and K. C. Winkler, Eds., pp. 435-437, 1973. 2. Blowers, R. , J. Hill, and A. Howell, Shedding of Staphylococcus aureus by human carriers. Proc. IVth International Symposium on Aerobiology. "Airborne transmission and airborne infection." J. F. Hers and K. C. Winkler, Eds., pp. 432-434, 1973. 3. Charnley, J. and N. Eftekhar, Post-operative infection in total prosthetic replacement arthroplasty of the hip-joint. Brit. J. Surg. 56:641-649, 1969a. 4. Charnley, J. and N. Eftekhar, Penetration of gown material by organisms from the surgeon's body. Lancet 1:172-174, 1969b. 5. Charnley, J., Experience with germ-free environments in surgery in relation to design. Proc. 3rd International Symposium on Aerobiology, pp. 191-198, 1970. 6. Coriell, L. L., W. S. Blakemore, and G. J. McGarrity, Medical applications of dust-free rooms. II. Elimination of airborne bacteria from an operating theater. JAMA 203:1038-1046, 1968. 7. Fox, D. G. and M. Baldwin, Contamination levels in a laminar flow operating room. Hospitals 42:108-112, 1968. 8. Gryska, P. F. and A. G. O'Dea, Postoperative streptococcal wound infection - The anatomy of an epidemic. JAMA 213:1189-1191, 1970.
I 310 Local Control of Airborne Bacteria 9. Ingrain, F. A., Air handling systems for hospital operating rooms. Mechanical Contracting and Engineering 61:46, 1967. 10. Laufman, H., Confusion in application of clean air systems to operating rooms. Proc. IVth International Symposium on Aerobiology. "Airborne transmission and airborne infection." J. F. Hers and K. C. Winkler, Eds., pp. 575-580, 1973. 11. Marsh, C. R. and W. C. Beck, Target zone protection from air contamina- tion by a new airflow modality - Vortex airflow. Guthrie Clin. Bull. 38:52, 1968. 12. May, K. R. and N. P. Pomeroy, Bacterial dispersion from the body surface. Proc. IVth International Symposium on Aerobiology. "Airborne transmis- sion and airborne infection." J. F. Hers and K. C. Winkler, Eds., pp. 426-432, 1973. 13. McDade, J. J., J. G. Whitcomb, E. W. Rypka, et. al., Microbiological studies conducted in a vertical laminar airflow surgery. JAMA 203:125- 130, 1968. 14. Mclntyre, D. M., An epidemic of Streptococcus pyogenes puerperal and postoperative sepsis with an unusual carrier site - The anus. Amer. J. Ob. Gyn. 101:308-314, 1968. 15. McKee, W. M., J. M. di Caprio, C. E. Roberts, Jr., and J. C. Sherris, Anal carriage as the probable source of a streptococcal epidemic. Lancet 11:1007-1009, 1966. 16. Nelson, J. P., A. R. Glassburn, Jr., R. D. Talbott, and J. P. McElhinney, Horizontal flow operating room clean rooms. "Clean Air Symposium - Part II." Cleveland Clinic Quarterly 40:191-202, 1973. 17. Riemensnider, D. K., Quantitative aspects of shedding of microorganisms of humans. Spacecraft Sterilization Technology NASA-SP 108:97, 1966. 18. Ritter, M. A., J. B. Hart, M.L.V. French, and H. E. Eitzen, A systems analysis approach to postoperative wound infections. "Clean Air Symposium - Part II." Cleveland Clinic Quarterly 40:211-219, 1973. 19. Schaffner, W., L. B. Lefkowitz, J. S. Goodman, and M. G. Koenig, Hospital outbreak of infections with group A streptococci traced to an asymptomatic anal carrier. New Eng. J. Med. 280-1224-1225, 1969. 20. Sciple, G. W., D. K. Riemensnider, and C. A. Schleyer, Recovery of microorganisms shed by humans into a sterilized environment. Applied Microbiol. 15:1388-1392, 1967. 21. Ulrich, J. A. , Microbiology of surgery suites. Symposium on Clean Room Technology in Surgery Suites. NASA Midwest Research Institute, Cape Kennedy, pp. 11-32, May 1971.
311 PRINCIPLES OF HIGH VELOCITY DIRECTIONAL AIRFLOW SYSTEMS Willis J. Whitfield* I was asked to speak on the "Principles of High Velocity Directional Airflow Systems." First, I shall define "high velocity" and "directional" airflow as they are related to these types of clean rooms. "High Velocity Airflow" is high velocity only when compared to the "low velocity" (20-30 ft/min) systems that you heard reports on earlier today. Airflow velocities of the system that I will be describing operate over a range of 80 to 100 ft/min, or about one mile per hour. These must also be considered low velocities when compared to other velocities with which we are familiar. "Directional" airflow means that the entire body of air in the room is moving in a single direction, either ceiling-to-floor (vertical flow) or one wall to the opposite wall (horizontal flow). A better descrip- tion might be "unidirectional turbulent flow." Also, an adequate description is contained in Federal Standard 209B for this type of airflow, designated as "Laminar Airflow." I shall be very brief in describing the principles of the unidirec- tional airflow system as many of you have been working with these systems for a number of years. Very likely the rest of you have been exposed to the idea many times. More information is contained in Sandia Laboratories Report SC-M-69-1291. The basic idea is a straightforward one, and may be described as "the entire body of air in a room that flows uniformly in the same direction throughout the room." Several elements are necessary to achieve this action including a filter bank, plenums, blowers, a containment system and tempera- ture control. Figure 1 depicts one arrangement of those elements which pro- vide very rigid control over airborne particle contaminants of which microbes are a part. This system is commonly referred to as the vertical downflow type. Air enters the room through the entire ceiling and exits the room through the entire floor. This arrangement provides a very short path for *Division Supervisor, Biosystems Research Department, Sandia Laboratories, Albuquerque, New Mexico
/ ;_ 312 Willis J. Whitfield removal of contamination from the room, and offers a high degree of isolation for any given location within the room. Usually, this type of room is the most expensive to install. ILL | AIR FLOW i i i I I I I I I i <> HEPA FILTER Ml I M HIM II IN INI I I I I I V O 111 .WALL /FLOOR GRATED ' BAS1C LAMINAR AIR FLOW CONCEPT Fig. 1. The same principle is applied to the curtain unit except that air exits beneath the curtain instead of through the floor. Another variation of the basic principle is the crossflow, or hori- zontal flow room. In this case, air enters the room through one entire wall and exits through the opposite wall. A variation of the horizontal flow idea is the "tunnel" room which is the same as the horizontal unit, except the air exits through an open end of the room.
313 Principles of High Velocity Directional Airflow Systems The unidirectional systems described here have the capability of maintaining close control of airborne microbes, thereby, drastically limiting microbial fallout onto surfaces within the room. This has been shown by Turner (2), McDade (3), and others with a large number of tests in operational surgical suites. RELATIONSHIP OF AIRBORNE CONTAMINATION TO PATIENT CONTAMINATION In order to better understand the need for a clean room in surgery, a review is presented here of common modes of microbial contamination transfer to a patient. Such a transfer becomes of concern when the contamination affects the patient in a damaging manner, such as by infection or allergy. 1. Self contamination occurs when the patient's microbial flora cause problems as a result of surgery or medical treatment. 2. Direct transfer is a high probability mode of transferring contamination directly to the patient by contact with surgical instruments, the surgical team or monitors. The clean room has the potential of prevent- ing airborne microbial contamination from collecting or "settling out" on the surgical instruments, personnel, trays, or clothing which can be transferred directly or indirectly to the patient. 3. Direct fallout or impingement is the mode by which airborne microbes are transferred directly from the air to a patient. The clean room eliminates this mode of contamination transfer to the extent of air cleanliness immediately surrounding the patient. Air cleanliness near the patient depends upon: (a) the clean room's filtration capability of cleaning the air entering the clean room; (b) clean room air velocity, direction and patterns in the immediate vicinity of the patient to provide fast removal of microbial particles shed by personnel and instruments. Also, correct airflow characteristics will reduce cross- transfer of contamination within the clean room*
I 314 Willis J. Whitfield (c) the amount of contamination generated or carried to the immediate vicinity of the patient; (d) clean room work procedures, which.can reduce microbial contamination brought near the patient. It should be emphasized that even though direct fallout or impingement rates are small, the indirect effects of airborne contamination are serious. It is for this reason that airborne contamination must be controlled before very clean surfaces can be obtained and maintained. Very often the effects of contamination can be greatly reduced with extreme care without the use of clean rooms. However, techniques involving extreme care in cleaning, sequencing of procedures, and trial and error methods used to lessen the danger of contamination without a clean room, many times cost more in time, effort, and risk than a clean room would cost. DETERMINING THE NEED FOR CLEAN ROOMS IN SURGICAL SUITES For low-risk surgery where infection rates are low to begin with, the benefits of the clean rooms will also be low. For high-risk surgery the clean room will likely be of much greater benefit. Unless clean rooms are used for all types of surgery, decisions must be made as to which surgical cases require the microbial control provided by a clean room. The surgeon is the most logical person to make these decisions since he can best assess the risk of a given surgical procedure, the condition of the patient, and determine the selection of instruments and other equipment for his personal use. One thing is missing that is necessary for the surgeon to make such decisions - discrete airborne microbial levels or standards. Standardization is needed to provide the surgeon with a method of designating airborne micro- bial levels for operating rooms to meet specific surgical procedure needs. I am convinced that the technology exists today to create such standards.
315 Willis J. Whitfield REFERENCES 1. Lindell, K. F., W. J. Whitfield, and D. M. Garst, Design Requirements for Laminar Airflow Clean Rooms and Devices, Sandia Laboratories Report SC-M-69-129, May 1969. 2. Turner, R. S., "Laminar Air Flow," reprint from the Journal of Bone and Joint Surgery, 56-A:430-435, March 1974. 3. McDade, J. J., J. G. Whitcomb, E. W. Rypk, W. J. Whitfield, and C. M. Franklin, "Microbiological Studies Conducted in a Vertical Laminar Airflow Surgery, JAMA 203:125-130, January 8, 1968.
316 ULTRACLEAN ENVIRONMENTS - A BRITISH POSITION William Whyte* INTRODUCTION It is my intention in this paper to report on the basic design data that exist at the present time in the United Kingdom on ultraclean systems. I shall not consider the way in which conventional turbulently ventilated operat- ing rooms are designed to prevent airborne infection: firstly, because of the time available and, secondly, I would be repeating what has recently been re- ported in the UK by a Joint Working Party (7). All current thinking on the design of ventilation systems and air movement control in the operating suite has been incorporated into this report. It is natural that this paper will draw heavily on my own research work, but I should also like to report on some of the work carried out in the small number of UK centers where there are ultraclean systems. This does not, however, include any more than a fleeting reference to the extensive work of Professor John Charnley whose research it is hoped, will be well known by, and extensively referred to, by others at this workshop. It is worth noting that although the number of ultraclean systems in the USA must be counted in their hundreds, there must certainly be no more than twenty in the UK at the present time. To prevent airborne contamination of the wound, three methods of ap- proach are possible: 1. to protect the wound so that airborne bacterial particles cannot get to it; 2. to wear impervious clothing to prevent bacteria being liberated at the source; and 3. to dilute and, if possible, remove all bacterial particles by ventilation. The contention that the solution to wound contamination lies in pro- tecting the wound is held by the exponents of the plastic film isolator system, a modification of systems built to produce germ-free animals. However, the other approach to the ultraclean environment, which has at the present time in Building Services Research Unit, Department of Mechanical Engineering, University of Glasgow, Glasgow, Scotland.
317 Ultraclean Environments in the UK the greatest number of adherents, is the use of unidirectional or laminar flow systems (to dilute or remove bacterial particles) with or without the use of special clothing (to prevent the dispersal of bacteria). I propose to discuss what has been done and what is known about these two approaches, firstly, by discussing the plastic film isolator type of system, and then, in more depth, the unidirectional systems and the role of clothing in reducing the chance of wound contamination. PLASTIC FILM ISOLATORS For many years large quantities of animals have been born and raised germ free by the use of the plastic film isolator technique. Instinctively one would expect such a successful system would have an application to the oper- ating room, and this is so, having been applied recently in the UK to surgery and now on sale commercially. This system has recently been described by the pioneer of notobiotics, Trexler (8), and its application to orthopaedic surgery is described by McLauchlan ejt al. (6). Shown in Figure 1 is the diagram of the apparatus used by McLauchlan and his colleagues, its function and use being more fully described in their paper. Simply, it consists of a thin-film plastic bag which is inflated by a continuous supply of sterile air supplied by a blower unit through a filter unit. The bag is stuck to the skin at the wound site and an incision made from the Inside of the bag through the isolator wall and patient's skin. The'opera- tion is carried out by the operating team using the access sleeves for their arms. Instruments and materials are passed through the entry port and along with the operating wound are isolated from outside contamination. Various in- genious methods are used to ensure the integrity of the system against bac- terial contamination and to keep its use as convenient as possible. Airborne bacterial sampling has demonstrated the sterility of the system. These units are commercially available in the UK and are presently being used in several centers in the UK. Hundreds of hip arthroplasties have been carried out under their protection. UNIDIRECTIONAL FLOW SYSTEMS AND OPERATING ROOM CLOTHING It has been explained above that the plastic isolator system will protect the wound so that no airborne bacteria can be deposited in it. There
1 318 William Whyte are, however, two other distinct ways of preventing airborne contamination. On the correct assumption that in modern operating theaters almost all the airborne bacteria in the area of the wound are derived from operating room personnel, it is possible either to prevent the bacteria being dispensed by them using occlusive clothing, or to dilute and hopefully remove them from the air by ventilation. Unfortunately for reasons which are explained below, it is not possible at this time to produce a completely sterile environment by either method alone. It is necessary to use both methods. SUPPORTING YOKE FLUTTER VALVE AIR-SUPPORTED ENTRY PORT CURTAIN SWAB SUMP FILTER UNIT ACCESS SLEEVE BLOWER UNIT COOLING JACKET AIR SUPPLY Fig. 1. Diagram of plastic film isolator. DILUTION BY UNIDIRECTIONAL FLOW SYSTEMS - AVAILABLE DESIGN DATA Over the last few years at Glasgow we have been assembling research data on the performance of laminar flow systems in an attempt to optimize their performance. Shown in Figure 2 is a sectional drawing of our laminar flow system as installed in Gartnavel District General Hospital, Glasgow. This is a slightly modified version of the system which was first installed four years ago in another hospital (10). It should be noted that this system has the fa- cility for changing both air direction and velocity during an operation. There were a number of problems which existed, indeed still exist, on the design of these systems and I should like to consider in turn the prob- lems as I see them.
319 Ultraclean Environments Fig. 2. Sectional drawing of optional-direction, variable speed, laminar-flow system: a. primary filters; b. silencers; c. variable speed fan; d. variable direction flap normally either up or down; e. High Efficiency Particulate Air (HEPA) filters; f. front plastic curtain; g. laminated glass; h. body-exhaust duct; i. personal ex- haust tubes; j. return air grill; k. side flap; 1. skeletal lamp; m. audio controls. 1. Downflow or crossflow? One of the basic questions we set out to solve was whether or not crossflow was better or worse than downflow in terms of airborne bacteria at the wound site. The crossflow system is in general cheaper to build and install and because of the open end there is better communi- cation and accessibility. Unfortunately, our studies showed that using conven- tional operating clothing the downflow system was around five times better in terms of airborne bacteria at the wound side during major orthopaedic surgery. By measuring the airborne bacterial count within 10 cm of the wound site by means of a-specially adapted Casella high-volume sampler drawing 700 1/min. (25 cfm) of air, the counts as shown in Figure 3 were ascertained. Sampling was carried out for up to 10 minutes and then either the air velocity or direction was changed. This was carried out through the entire operation and it may be clearly seen
I 320 William Whyte from this strictly controlled experiment that downflow, in terms of airborne bacterial particles sampled at the wound, was around five times better than crossflow at 0.3 to 0.5 m/s (60-100 ft/min). The full results of this experi- ment are contained in our paper (12). 200 a l00 O DOWN 50 4-0 30 20 â¢1.0 01 0-2 O-3 04 VELOCITY m/« OS 06 Fig. 3. Airborne bacterial counts at different velocities under cross- and downflow conditions. 2. Velocity requirements. Also shown in Figure 3 is the bacterial concentration to be found at the wound site at velocities varying from 0.1 m/s to 0.6 m/s (20-120 ft/min). As seen from this graph, the higher the velocity of the unidirectional airflow the lower the bacterial count. However, as in so many situations, the effort (and hence cost) expended in achieving very low airborne bacterial counts in an operating room is not proportional to results. This means that a point must be drawn where moderate effort and good design will achieve a highly efficient system. A study of Figure 3 will show this point to be around 0.3 m/s (60 ft/min). Allowing a small margin of safety, we now run our system at 0.35 (70 ft/min) using a downflow of air.
321 Ultraclean Environments 3. Walls or no walls? Unfortunately, although a downflow system of air movement is the best bacteriologically, it is not the best in terms of access or freedom of communication. Because of the apparent necessity of having walls Co prevent entrainment of air from outside the clean area, the operation team is shut in and communication and access for X-ray equipment, etc. is very difficult. The access problem may be overcome by removable walls, but this rannot be considered a fully satisfactory solution. We have studied this prob- lem and suggest as a possible solution the use of a downflow partial-walled system. These partial walls replace the full walls being located around the periphery of the clean area, and hang down to head height. A photograph of a partial-walled system at the London Hospital is shown in Figure 4. I It Fig. 4. Partial-walled system at the London Hospital.
322 William Whyte By constructing a l/10th scale water model and ensuring the Reynolds number to be the same in the water model as the airflow in the laminar flow oper- ating room, it was possible to predict the type of flow and the effectiveness of the system in preventing penetration by contamination from the outer dirty area (15). This effectiveness is assessed by injecting dye as a challenge to the system, an example of this being shown in Figure 5. .. EXHAUST Fig. 5. Dye challenge of partial-walled, water-model system. We also have had the opportunity of studying at the London Hospital (14) a partial-walled operating room. An assessment of this system highlighted some design faults, but by means of bubble generators, smoke challenge tests and velocity readings it was possible to show that: a. the type of airflow was similar to that of the water model; b. the partial wall was preferable to no wall; c. entrainment was possible but the amount of contamination which reached the operating table was very smallâabout 1/10,000th of that on the outside area, but it was l/100th at the periphery of the clean area where the instrument trolleys could be sited; d. in order to achieve the same airborne bacterial concentrations
323 Ultraclean Environments as the full walled system, the airflow would require to be increased by 25 percent because of the air moving sideways (as shown in Fig. 3) instead of being forced down by the constraining walls. it.. Obstructions and thermals. Low-speed unidirectional flow systems, as suggested by Blowers (2), have been shown to be disrupted by ther- mals rising from operating room personnel and operating lamps. On studying this problem in the higher speed laminar flow systems by means of Schlieren photography (an optical method by which hot air may be seen), it was possible to demonstrate (13) that the most intense heat source in our system, which was a quartz-halogen operating light giving off 1 kw of heat, would be overcome by the downflow of air at speeds over 0.3 m/s (60 ft/min). It is therefore ap- parent that the airflow in laminar flow systems will not be disrupted by ther- mals . Investigations were also carried out on obstructions of the laminar flow system by use of bubble generators which produce small neutral-buoyancy detergent bubbles. Shown in Figure 6 is the effect on unidirectional airflow of the large lamp of the type used in the great majority of British operating theaters. The air in the turbulent area behind this lamp will have high con- centrations of contamination if it is in contact with bacteria being dispersed by the operating team. In a similar fashion if it is in contact with the wound site, it will subject the wound to high concentrations of bacteria. Such ob- structions should be kept well clear of the operating site, reduced in size or completely removed. 5. Filter efficiency and noise levels. It is common practice to fit filters that are of an efficiency greater than 99.9 percent against 0.3y par- ticles. This practice is obviously a direct copy from the industrial situation where these HEPA filters are required to contend with dust particles of 0.5y and greater in size. As the size distribution of bacterial particles in hospital areas is much larger than this (the average size being around 12 to 13y), it would appear that there is reason for considering filters that are less effi- cient. As these industrial type HEPA filters have a high pressure loss, they require large fans which, in turn, generate a great deal of noise if situated close or near to the operating theater. Use of less efficient filters would be of much benefit.
324 William Whyte Fig. 6. Airflow around operating room lamp. I have studied (9) the efficiency of airconditioning filters in a hospital ward when air was being recirculated. As it was a hospital ward, very high efficiency filtration was not considered a priority and HEPA filters were not used, but when a final filter around 80 percent against 0.5y particles was installed, the bacterial count in the recirculated air was reduced from 1.93 3 3 bacterial particles/ft before the filters, to 0.013 ft after the filters--a 99.3 percent reduction. Considering the always present possibility of contami- nation which could account for at least part of the final bacterial concentra- tion and also that any particles passing through this final filter would not be of the size that would deposit in the wound, it would appear that there is an excellent indication that HEPA filters are not essential. A standard has now been accepted by the DHSS (7) for filtration of recirculated air in the operating theater of 90 percent against 0.3y to 0.5y particles.
325 Ultraclean Environments Use of these filters would approximately halve the pressure drop at the filter and hence reduce the noise level but, notwithstanding this, it would appear that a great variety of systems are on sale which suffer from excess noise, and it would be an advantage that rioise standards be worked toward the achievement of acceptable noise levels. This may be impracticable at the pres- ent time, but we should insist that noise levels, as measured in the operating area, be quoted for a given air velocity. PREVENTION OF DISSEMINATION OF BACTERIA BY CLOTHING It has been well established that people disseminate large numbers of bacteria into the air. Duguid and Wallace (4) estimated indirectly that a per- son with ordinary clothing would, during vigorous activity, disperse 10,000 bac- terial particles per minute and that ordinary operating room clothing did very little to prevent this. The importance of this lies in the fact that, in an operating room which is plenum ventilated by filtered air, almost all of the bacteria to be found in the air are derived from the operating team and that Staphylococcus aureus is given off into the air in large numbers by carriers of this patho- genic organism (1). It seems obvious that operating theater clothing should be designed to prevent dissemination. Unfortunately, materials which are impervious to bacteria also prevent exchange of heat and water vapor from the body and, hence, are intolerable to wear. The present approach is either to produce clothes which are more impermeable to bacteria yet not uncomfortable to wear, or to use impermeable clothing and ventilate the person. There are also some intermediate stages between these two possibilities. In a series of as yet unreported ex- periments, we have compared, in a dispersal chamber, several types of gowns and trousers. We attempted to simulate both conventional turbulent and laminar- flow ventilation in the chamber but, for the purposes of this paper, I should like to mention only the results obtained in the laminar flow situation. Three subjects were used, and it was demonstrated that, by using disposable gowns of a conventional pattern made from nonwoven fabric instead of conventional linen gowns, the airborne bacterial counts could be reduced by 60 percent; but by using total-body exhause systems of the Charnley style, the count could be
326 William Whyte reduced by over 98 percent. In order to test the validity of these experiments, as carried out in the test chamber, a comparison was made in the operating room between 1) ordinary linen gowns, 2) nonwoven fabric disposable gowns, and 3) Charnley-style total-body exhaust gowns. These showed in the laminar flow unit a very similar drop in airborne concentration when nonwoven fabric gowns were used, but the total-body exhaust gowns did not give quite as good a performance as in the dispersal chamber. There was some evidence that this slightly smaller drop was due to bacteria being dispersed from the patient on the operating table. The bacterial count in the air of our laminar flow system when the total-body exhaust gowns are being used is around 0.63/m which is a reduction by a factor greater than 650 in airborne count at the wound site compared to our plenum ventilated conventional operating room when linen clothes are being used. AIRBORNE COUNTS, WOUND CONTAMINATION, AND INFECTION It is obvious from the results cited above that both the plastic film isolator type of ultraclean system and the laminar flow type, substan- tially reduce the number of bacteria in the air and, therefore, ipso facto must also reduce the number of bacteria to be found in the wound after surgery. The evidence for this statement may be ascertained by washing the wound after surgery and counting the number of organisms isolated. A standard technique has been suggested by the Medical Research Council's Working Party on Ultra- Clean Air in Surgical Operating Rooms for use in clinical trials of ultraclean systems, and is being used by several centers in the UK. The results of one such comparison will be presented by Mr. George Bentley during this workshop. He compared a conventionally ventilated operating room with an ultraclean one with respect to the number of bacteria in wounds after surgery. The question still remains, however: Will the number of bacteria in the wound be reduced to a level sufficient to affect the incidence of infection? Or to pose the question another way: Is airborne contamination important? It should not be necessary at this workshop to remind the participants of the ex- cellent work carried out by Professor Charnley in the UK. He maintains that a substantial reduction in deep infection after hip arthroplasties will be gained
327 Ultraclean Environments by use of ultraclean air in operating rooms (3). Equally well, it should not be necessary to mention the criticism of the sequential method of his trial where variables other than a reduction of airborne bacterial particles, were introduced and the low infection rates obtained by many other orthopaedic sur- geons in conventionally ventilated operating rooms (see Journal of Clinical Orthopaedics and Related Research. September 1973). Unfortunately in the UK there are relatively few centers with ultraclean systems, and even fewer with the years of experience, or number of operations, where Charnley's work could be confirmed. We ourselves, to date, have been more concerned with an assess- ment of the system from a bacteriological standpoint, rather than from a clinical one, and have only recently started a fully controlled clinical trial. This trial is being run according to the protocol set up by the Medical Research Council Working Party on Ultra-Clean Air in Surgical Operating Rooms for a multicentered clinical trial of ultraclean systems in the UK. How this trial will be run, and an outline of the protocol will be reported by Dr. Lidwell at this workshop. It wil1, therefore, not be necessary to duplicate such infor- mation. However, it is worth reporting the clinical results obtained in Edinburgh by Dr. Gould and his colleagues. They have now for several years been comparing a laminar flow system with an unventilated operating theater. Shown in Table 1 are the results reported for the years 1971-72 and 1972-73 (5). It should be reported that a perfect balance between surgeons operating on one system or the other was not obtained, but it would appear there is an indica- tion from these results that deep infection will be reduced by ultraclean sys- tems. A complete answer must await the results of the multicentered clinical trial.
I 328 William Whyte Table 1. Infection of Hip Arthroplasties No. of operations No. of superficial infections No. of deep infections 1970 - 1971 Unidirectional flow 80 5 (6 .2) 0 Without unidirectional flow 110 5 (3 .7) 3 (2.8) 1972 - 1973 Unidirectional flow 104 2 (1 .0) 0 Without unidirectional flow 76 5 (6 .3) 3 (4.0) SUMMARY This paper reports the known basic design data that are available at present in the UK for producing ultraclean environments for the operating room. Two systems are considered. These are the plastic film isolator which is used to protect surgical wounds from environmental contamination and unidirectional flow systems, with or without special clothing. The plastic isolator system is briefly described. Most of the remainder of the paper discusses the effect of different unidirectional flow systems and operating room clothing on reducing airborne contamination. The main effects are summarized in Figure 7. These data were gathered by us at two different times. The data on the effect of clothing in the downflow system were collected some time after the comparison of downflow with crossflow so that the data do not quite intermesh, and averages have been taken. Other subjects considered are: 1. velocity requirements; 0.35 m/s (70 ft/min) being suggested; 2. the use of partial walls for theaters with communication or access problems; 3. less efficient filters than at present (90 percent efficiency against 0.3y to 0.5y particles); 4. noise standards;
329 B Ultraclean Environments 5. the non-influence of thermals; 6. obstructions; and 7. comparative sepsis figures, the lack of which should be overcome by a multicentered trial to be mounted in the UK. CONVENTIONAL OPERATING | UNIDIRECTIONAL OPERATING ROOMS | ROOMS (Velocity O-35m/s) .. . _ Q.&-O-7 (ti3/» [CROSSFLOW | 8-IO persons ⢠, , 1 -,._». conventional clothing conventional clothing LjS-i*^"^ 3 4OO-5OO/m3 ~' | DOWNFLOW ] TOTAL » BODY ^^"^ conventional clothing *'^.. . _» 8/m3 EXHAUST 0.63/m3 «i 1 disposable clothing 2.5/m3 I ... , 1 »_ greater than 650 times reduction I I I Fig. 7. Vector diagram of the influence of unidirectional system and clothing on airborne contamination. i
330 William Whyte ACKNOWLEDGMENTS I should like to thank the editor of the Journal of Hygiene for permission to use Figures 3, 4, and 6, and Mr. J. McLauchlan and Dr. J. C. Gould for supplying current information which I have incorporated into this text. My research studies have been supported by the Medical Research Council, United Kingdom. REFERENCES 1. Bethune, D. W., R. Blowers, M. Parker, and E. A. Pask, Dispersal of Staphylococcus aureus by patients and surgical staff. Lancet i:480, 1965. 2. Blowers, R. , and Beryl Crew, Ventilation of operating theatres. J. Hyg. (Camb.) 58:427, 1960. 3. Charnley, J., Post-Operative Infection after Total Hip Replacement with Special Reference to Air Contamination in the Operating Room. Internal Publication No. 45, Wrightington Hospital, 1972. 4. Duguid, J. P., and A. T. Wallace, Air infection with dust liberated from clothing. Lancet ii:845, 1948. 5. Gould, J. C., F. J. Bone, and J.H.S. Scott, Bacteriology of surgical theatres with and without unidirectional airflow. Bulletin de la Societe Internationale de Chirurgie 33:53, 1974. 6. McLauchlan, J., M. F. Pilcher, P. C. Trexler, and R. C. Whalley, The sur- gical isolator. Br. Med. J. i:322, 1974. 7. Medical Research Council, Ventilation in Operation Suites. The Report of a Joint Working Party, London, 1972. 8. Trexler, P. C., An isolator system for the maintenance of aseptic environ- ment. Lancet 1:91, 1973. 9. Whyte, W., Bacteriological aspects of air-conditioning plants. J. Hyg. (Camb.) 66:567, 1969. 10. Whyte, W., B. H. Shaw, and R. Barnes, An experimental laminar-flow oper- ating room. Lancet ii:905, 1971. 11. Whyte, W., and B. H. Shaw, The design and comparative advantages of laminar- flow systems. In Airborne Transmission and Airborne Infection. Oosthoek, Utrecht, Netherland, p. 140, 1973.
331 Ultraclean Environments 12. Whyte, W., B. H. Shaw, and R. Barnes, A bacteriological evaluation of laminar-flow systems for orthopaedic surgery. J. Hyg. (Camb.) 71:559, 1973, 13. Whyte, W., and B. H. Shaw, The effect of obstructions and thermals in laminar-flow systems. J. Hyg. (Camb.) 72:415, 1974. 14. Whyte, W. , B. H. Shaw, and M.A.R. Freeman, An evaluation of a partial- walled laminar-flow operating room. J. Hyg. (Camb.) 73:61, 1974. 15. Whyte, W., B. H. Shaw, and P. V. Bailey, An assessment of partial walls for a down-flow laminar-flow system. In Proceedings of the International Symposium on Contamination Control, 1974. In Press.
333 AGENDA WORKSHOP ON CONTROL OF OPERATING ROOM AIRBORNE BACTERIA COMMITTEE ON PROSTHETICS RESEARCH AND DEVELOPMENT November 8-10, 1974 National Academy of Sciences Washington, D.C. George T. Aitken, Chairman APPENDIX A Friday, November 8 Lecture Room 8:30 am 8:40 am 8:50 am 9:05 am 9:20 am 9:30 am 10:00 am 10:20 am 10:30 am 10:45 am 10:55 am 11:15 am Moderator - George T. Aitken Welcome American Academy of Orthopaedic Surgeons Introduction and Objectives Review of Postoperative Wound Infections The Orthopedist's Special Interest in Surgical Infection Operating Room Air as a Source of Wound Contamination and Infection John C. Wilson, Jr. George T. Aitken William A. Altemeier Mark B. Coventry Ruth B. Kundsin COFFEE Moderator - John C. Wilson, Jr. Operating Room Bacteriology Carriers versus Shedders Skin Carriage Sampling and Significance Physics of Operating Room Air Bacteria Types of Operating Room Air Bacteria Human Bacterial Shedding Current Operating Room Air Handling Standards Standard Techniques to Reduce Operating Room Bacteria Moderator - Jo Miller Ultraclean Environments - The British Position The Case for Clean AirâResults of Three Years of Airborne Monitoring John A. Ulrich Dick Riemensnider Kenneth L. Credle Howard P. Hogshead William Whyte John A. Feagin, Jr.
i 334 A. M. Wiley Robert H. Fitzgerald, Jr. Harlan C. Amstutz 11:30 am Culture Plate Studies in the Orthopaedic Operating Room 11:40 am Airborne Bacterial Contamination in the Operating Room 11:50 am Experiences with Operating Room Environment at UCLA 12:00 noon DISCUSSION 12:30 pm LUNCH 1:30 pm Participants in five multidisciplinary groups will discuss the morning's work Moderator - John A. Ulrich 4:00 pm Plenary session for group leaders to present conclusions of group. Discussion 5:30 pm ADJOURN Saturday, November 9 Lecture Room Moderator - Philip D. Wilson, Jr. Improved Operating Room Barrier Materials Peter Dineen Personal Envelope Systems Jo Miller Helmet Aspirator Systems J. Phillip Nelson Ultraviolet Light J. Leonard Goldner Low Velocity Unidirectional Ultrafiltered Air Systems 8:30 am 8:45 am 8:55 am 9:05 am 9:20 am 9:35 am 10:05 am 10:15 am 10:25 am 10:35 am 10:45 am 10:55 am Lewis L. Coriell COFFEE Moderator - Harlan C. Amstutz Willis J. Whitfield Robert S. Turner Principles of High Velocity Directional Airflow Systems Vertical Laminar Airflow System Horizontal Flow Operating Room Clean Rooms J. Phillip Nelson Horizontal Laminar Airflow - Its Effects on Reducing Postoperative Wound Infections Merrill A. Ritter Bacteriological Evaluation of Clean Air Systems Carl L. Nelson Local Airflow as a Means of Protecting Surgical Wounds from Airborne Contamination William E. Anspach, Jr.
335 11:05 am Local Control of Airborne Bacteria J.C.N. Westwood 11:15 am Localized Surgical Isolators (Sterile Environment Capsule) Dana M. Street Moderator - William A. Altemeier 11:25 am A Controlled Comparison of Air and Wound Contamination in Conventional and Vertical Laminar Flow Operating Theatres George Bentley 11:45 am Evaluation of Methods for the Control of Surgical Sepsis Owen M. Lidwell 11:55 am What is the Hard Evidence on the Role of Airborne Bacteria in Wound Infection? Harold Laufman 12:05 pm Defining Biologic Clean Air and Its Clinical Significance William C. Beck 12:15 pm Summary of Proceedings of the International Symposium on Contamination Control: Micro- biological Aspects of Clean Room Technology as Applied to Surgeryâwith Special Reference to Unidirectional Airflow Systems Michael D. Wardle 12:25 pm Two Considerations in the Application of Laminar Clean Air Edward 0. Goodrich, Jr. 12:35 pm Experiences with Ultraviolet Light in the Operating Room J. Drennan Lowell 12:45 pm LUNCH 1:30 pm Group discussions of the morning's session Moderator - Kenneth L. Credle 4:00 pm Plenary Session 5:30 pm ADJOURN Sunday, November 10 Lecture Room 9:00 am Plenary Session Report of the Moderator of Plenary Session on Basic Data John A. Ulrich 9:30 am Discussion 10:00 am COFFEE 10:15 am Report of the Moderator of Plenary Session on Practical Application Kenneth L. Credle 10:45 am Discussion 11:15 am Summary, Conclusions, and Recommendations George T. Aitken
1â. KlL
337 APPENDIX B PARTICIPANTS WORKSHOP ON CONTROL OF OPERATING ROOM AIRBORNE BACTERIA COMMITTEE ON PROSTHETICS RESEARCH AND DEVELOPMENT COMMITTEE ON PROSTHETIC-ORTHOTIC EDUCATION ASSEMBLY OF LIFE SCIENCESâDIVISION OF MEDICAL SCIENCES NATIONAL ACADEMY OF SCIENCESâNATIONAL RESEARCH COUNCIL ORTHOPAEDIC SURGEONS George T. Aitken, M.D., Chairman 260 Jefferson Avenue, S.E. Grand Rapids, Michigan 49502 Past Chairman - Committee on Prosthe tics Research and Development (CPRD) Past President - American Academy of Orthopaedic Surgeons (AAOS) Harlan C. Amstutz, M.D. University of California Medical Center Los Angeles, California 90024 Professor and Chief, Division of Orthopaedic Surgery - UCLA Member of AAOS Committee on Operating Room Environment & Surgical Asepsis William E. Anspach, Jr., M.D. 648 North U.S. One North Palm Beach, Florida 33408 Member of AAOS Committee on Operating Room Environment & Surgical Asepsis George Bentley, F.R.C.S. University of Oxford Nuffield Orthopaedic Centre Headington, Oxford 0X3 7LD, England Associate Professor Mark B. Coventry, M.D. Mayo Clinic 200 First Street, S.W. Rochester, Minnesota 55901 Professor of Orthopaedic Surgery - Mayo Medical School Member of AAOS Committee on Operating Room Environment & Surgical Asepsis Lt. Col. John A. Feagin, Jr. , M.D. Letterman Army Medical Center Presidio of San Francisco, California 94129 Assistant Chief, Orthopaedic Service Director, Joint Replacement Service Robert H. Fitzgerald, Jr., M.D. Mayo Clinic 200 First Street, S.W. Rochester, Minnesota 55901 Instructor in Orthopaedics - Mayo Medical School
338 J. Leonard Goldner, M.D. Duke University Medical Center Durham, North Carolina 27710 Professor and Chairman, Division of Orthopaedic Surgery - Duke University Medical Center Howard P. Hogshead, M.D. 333 East Ashley Street Jacksonville, Florida 32202 Director of Rehabilitation - Cathedral Health and Rehabilitation Center Member of AAOS Committee on Operating Room Environment & Surgical Asepsis J. Drennan Lowell, M.D. Peter Bent Brigham Hospital 721 Huntington Avenue Boston, Massachusetts 02115 Director, Orthopaedic Clinical Services - Peter Bent Brigham Hospital Assistant Professor of Orthopaedic Surgery - Harvard Medical School Jo Miller, M.D., F.R.C.S., F.A.C.S. Montreal General Hospital 1650 Cedar Avenue, Room 646 Montreal 109, Quebec, Canada Orthopaedic Surgeon-in-Chief, Montreal General Hospital Member of AAOS Committee on Operating Room Environment & Surgical Asepsis Associate Professor of Surgery, McGlll University Carl L. Nelson, M.D. University of Arkansas Medical Center Little Rock, Arkansas 72201 Professor and Chairman, Department of Orthopaedic Surgery - University of Arkansas Medical Center Member of AAOS Committee on Operating Room Environment & Surgical Asepsis J. Phillip Nelson, M.D. Orthopaedic Associates, P.C. 1707 East 18th Avenue Denver, Colorado 80218 Instructor, Department of Orthopaedics - University of Colorado Active Staff - St. Luke's Hospital Chairman of AAOS Committee on Operating Room Environment & Surgical Asepsis Merrill A. Rltter, M.D. 1815 North Capitol Avenue Indianapolis, Indiana 46202 Associate Professor of Orthopaedic Surgery - Indiana University School of Medicine Dana M. Street, M.D. Harbor General Hospital 1000 West Carson Street Torrance, California 90509 Professor and Chief, Division of Orthopaedic Surgery - Harbor General Hospital
339 Robert S. Turner, M.D. Lovelace Clinic 5200 Gibson Boulevard, S.E. Albuquerque, New Mexico 87108 Chairman, Department of Orthopaedic Surgery - Lovelace Bataan Medical Center A. M. Wiley, M.Ch., F.R.C.S. 25 Leonard Avenue Suite 101 Toronto, Ontario M5T 2R2, Canada Assistant Professor of Surgery - University of Toronto John C. Wilson, Jr., M.D. 2010 Wilshire Boulevard #803 Los Angeles, California 90057 President - AAOS Philip D. Wilson, Jr., M.D. Hospital for Special Surgery 535 East 70th Street New York, New York 10021 Surgeon-in-Chief - Hospital for Special Surgery Professor of Orthopaedic Surgery - Cornell University Medical College Past President - AAOS GENERAL SURGEONS William A. Altemeier, M.D. Cincinnati General Hospital Cincinnati, Ohio 45229 Chairman, Department of Surgery - Cincinnati General Hospital Professor of Surgery - University of Cincinnati Medical College William C. Beck, M.D., F.A.C.S. Donald Guthrie Foundation for Medical Research Guthrie Square Sayre, Pennsylvania 18840 Chairman, Committee on Operating Room Environment - American College of Surgeons John F. Burke, M.D. Massachusetts General Hospital Boston, Massachusetts 02114 Chief of Staff - Shriners Burns Institute Visiting Surgeon - Massachusetts General Hospital Associate Professor of Surgery - Harvard Medical School Richard E. Clark, M.D. 4960 Audubon Avenue St. Louis, Missouri 63110 Cardiothoracic Surgeon - Barnes General Hospital Associate Professor of Surgery - Washington University School of Medicine
340 Peter Dineen, M. D. New York Hospital Cornell Medical Center 525 East 68th Street New York, New York 10021 Professor of Surgery - Cornell Medical Center Edward 0. Goodrich, Jr., M.D. , F.A.C.S. Coronado Building Santa Fe, New Mexico 87501 Chairman, Subcommittee on Air Environment of the Committee on Operating Room Environment - American College of Surgeons Col. Gerald Klebanoff , M.D. Headquarters USAF/SGPC Washington, D.C. 20314 Consultant in Surgery Harold Laufman, M.D. , Ph.D. Montefiore Hospital and Medical Center 111 East 210th Street Bronx, New York 10467 Director, Institute of Surgical Studies Past Chairman, Committee on Operating Room Environment - American College of Surgeons MICROS IOLOGISTS Lewis L. Coriell, M.D. , Ph.D. Institute for Medical Research Copewood Street Camden, New Jersey 08103 Director - Institute for Medical Research Bacteriologist Ruth B. Kundsin, Sc.D. Peter Bent Brigham Hospital 721 Huntington Avenue Boston, Massachusetts 02115 Hospital Epidemiologist - Peter Bent Brigham Hospital Director, Surgical Bacteriology Laboratory - Environmental Sepsis Labora tory - Peter Bent Brigham Hospital Principal Research Associate - Harvard Medical School Dick K. Riemensnider Bureau of Health Resources Administration Division of Facilities Utilization, Room 1233 5600 Fishers Lane Rockville, Maryland 20852 Environmental Microbiologist, Bureau of Health Resources Administration Department of Health, Education, and Welfare
341 John A. Ulrich, M.D. University of New Mexico Medical School 915 Stanford Albuquerque, New Mexico 87131 Chief, Microbiology Service Michael D. Wardie, M.S. Jet Propulsion Laboratory California Institute of Technology 4800 Oak Grove Drive Pasadena, California 91103 Senior Microbiologist J.C.N. Westwood, M.B., B.Chir. University of Ottawa 275 Nicholas Street Ottawa, Ontario KIN 6N5, Canada Professor and Head, Department of Microbiology and Immunology - Faculty of Medicine - University of Ottawa William Whyte, M.D. Building Services Research Unit Department of Mechanical Engineering University of Glasgow 3 Lilybank Gardens Glasgow G12 8RZ, Scotland Bacteriologist ENGINEERS Kenneth L. Credle, P.E. 14317 Woodcrest Drive Rockville, Maryland 20853 General Engineer, Public Assistance Division - Federal Disaster Assistance Administration, Department of Housing and Urban Development Consultant to American Hospital Association Owen M. Lidwell, Ph.D. Central Public Health Laboratory Colindale Avenue London NW9 5HT, England External Scientific Staff - Medical Research Council Deputy Director - Cross Infection Reference Laboratory George F. Mallison Bureau of Epidemiology Center for Disease Control Atlanta, Georgia 30333 Assistant Director, Bacterial Diseases Division - Center for Disease Control
342 R. Claude Marsh Envirco, Inc. Subsidiary of Bio-dynamics 6701 Jefferson, N.E. P.O. Box 6468 Albuquerque, New Mexico 87107 Health Physicist Director of Research and Development - Envirco, Inc. Norman Petersen, S.M. Phoenix Laboratories Center for Disease Control 4402 North 7th Street Phoenix, Arizona 85014 Sanitary Engineer Director - Phoenix Laboratories J. Clyde Sell, P.E. Office of Architecture and Engineering Bureau of Health Resources Development Division of Facilities Utilization Department of Health, Education, and Welfare Rockville, Maryland 20852 Mechanical Engineer Willis J. Whitfield, Ph.D. Sandia Laboratories Albuquerque, New Mexico 87115 Division Supervisor, Biosystems Research Department - Sandia Laboratories Walter E. Williams Baltimore City Hospital 4940 Eastern Avenue Baltimore, Maryland 21224 Hospital Engineer Director of Hospital Services - Baltimore City Hospital OTHERS E. E. Harris, M.R.C.S. Committee on Prosthetics Research and Development National Academy of Sciences-National Research Council 2101 Constitution Avenue, N.W. Washington, D.C. 20418 Staff Surgeon - CPRD Charles V. Heck, M.D. Executive Director American Academy of Orthopaedic Surgeons 430 N. Michigan Avenue Chicago, Illinois 60611
343 Eugene F. Murphy, Ph.D. Director, Research Center for Prosthetics Veterans Administration 252 Seventh Avenue New York, New York 10001 A. Bennett Wilson, Jr. Executive Director Committee on Prosthetics Research and Development National Academy of Sciences-National Research Council 2101 Constitution Avenue, N.W. Washington, D.C. 20418
1
COMMITTEE ON PROSTHETICS RESEARCH AND DEVELOPMENT ASSEMBLY OF LIFE SCIENCES-NATIONAL RESrAKCM COUNCIL FRANK W. CLIPPINGER, JR., M.D. Acting Chairman Professor of Orthopaedic Surgery Duke University Medical Center DUDLEY S. CHILDRESS, Ph.D. Director Prosthetics Research Laboratory Northwestern University Medical School MARY DORSCH, C.P.O. President Dorsch-United Proothetics-Orthotics, Inc. ROBERT E. FANNIN, C.O. Hanger-Goldsby, Mobile, Ala. SIDNEY FISHMAK, Ph.D. Coordinator, Prosthetics and Orthotics New York University Post-Graduate Medical School VICTOR li. FRANKEL, M.D. Professor, Orthopaedic Surgery and Biomedical Engineering Case Western Reserve University RICHARD M. HERMAN, M.D. Director of Research Krusen Center for Research and Engineering Moss Rehabilitation Hospital RICHARD E. HOOVER, M.D. Department of Ophthalmology The Johns Hopkins University LEON K. KRUGER, M.D. Depcrtraent of Orthopedics Wesson Memorial Hospital, Springfield, Mass. JAMES M. MORRIS, M.D. Associate Professor of Orthopaedic Surgery University of California Medical Center at San Francisco COMMITTEE ON PROSfHETIC-ORTHOTIC EDUCATION ASSEMBLY OF LIFE SCIENCES-NATIONAL RESEARCH COUNCIL HER3ERT E. PEDERSEN, M.D., Chairman Professor and Chairman Peparti"snt of Orthopedic Surgery Wayne State University School of Medicine HARLAN C. AM5TUTZ, M.D. Professor and Chairman Division cf Orthopaedic Surgery University of: California School of Medicine at Los Angeles FRANK W. CLIPPINGER, JR., M.D. Professor of Orthopaedic Surgery Duke University Medical School CLINTON L. COMPERE, M.D. Director, Rehabilitation Engineerins Program Northwestern University Medical School DOROTHY J. HICKS Director Rehabilitation Nursing Program University of Washington School of Nursing DOUGLAS A. I10BSON, P. Eng. Technical Director Rehabilitation Engineering Program Crippled Children's Hospital School Memphis, Tenn. BELLA J. >!AY, Ed.D. Associate Professor and Chairman Department of Physical Therrpy School of Allied Health Professions Medical College of Georgia J. WARREN PERRY, Ph.D. Dean, School of Health Related Professions State University of New York at Buffalo JACQUELI>: PERRY, M.D. Chief, Research and Development Group Kinesiology Service Ranoho Lcs Amigos Hospital, Inc. ALLEN S. RUSSEK, M.D. Director, Amputee-Prosthetic Service New York University Medical Center AUGUSTO SARMIENTO, M.D. Professor and Chairman Department of Orthopaedics and Rehabilitation University of Miami School of Medicine ROY SNELSON, C.P.O., Project Director Amputee and Problem Fracture Service Rancho Los Araigos Hospital, Inc. WALTER A. L. THOMPSON, M.D. Professor and Chairman Department of Orthopedic Surgery New York University School of Medicine STAFF: A. Bennett Wilson, Jr., Executive Director Hector W. Kay, Aosistcrr.- Zxesutiva Director E. E. Harris, Sdiff S'trcson Gustav F. Haas, Staff Vxginesr Peter J. Nelson, Staff Engineer Michael J. Quiglcy, Staff Prosthetist/Orthotist G. E. Sharpies, Staff Officer
25071 25072 25073 25074 25075 25076 25077 25078 25079 26070 GENUINE BLACK LIGHT BLUE DARK BLUE L1GHT GRAY LIGHT GREEN DARK GREEN TANGERINE RED EXECUTIVE RED YELLOW PRE83BOARD ACCO INTERNATIONAL INC. CHICAGO. ILLINOIS 60619
