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.
. Descriptive Epidemiology of Occupational Infections of Laboratory Workers A. INTRODUCTION The precise incidence of occupational infections among laboratory workers is not known. During the past century, however, an extensive literature has documented that such infections have occurred with regularity and have occasionally resulted in death. The overall mortality rate of reported cases is 4 per- cent [1001. Published reports have dealt largely with single cases or outbreaks, retrospective studies, and passively collected anecdotal information. For the most part, historical accounts of laboratory-associ- ated infections have listed only individual cases with no attempt at the difficult task of determining the size of the populations at risk. Furthermore, there is no central focus of responsibility or authority in the United States that maintains a comprehensive data base or conducts surveillance of occupational infec- tions in laboratory workers who regularly or occa- sionally handle microorganisms or viruses. B. THE EPIDEMIOLOGIC TRIAD The components of the epidemiologic triad asso- ciated with laboratory-acquired infections are the host, the infectious agent, and the environment. Although most of the data relevant to these three factors have been collected retrospectively and are incomplete, an evaluation of the information that is available pro- vides some insight into the complexity of the prob- lem. 1. The Host Comprehensive and current data are not available on the demography of laboratory workers (the host) who risk occupational infection with the agents they handle in their daily activities. Published surveys , however, indicate that the number of workers employed by public health and clinical laboratories is substantial, having been estimated at 250,000 in 1977. A more recent survey conducted in 1983 by the Occupational Safety and Health Administration (OSHA)  estimated that there are 370,000 em- ployees in clinical laboratories, 45,000 employees in federal government laboratories, and 127,000 em- ployees in academic laboratories. Additionally, the number of physicians' office laboratories in which one or more persons are employed may exceed 100,000 facilities. Of the combined total of more than 640,000 workers, as many as 500,000 may regu- larly or occasionally work with infectious agents or with blood, serum, urine, or other body fluids or tissues that may contain an infectious agent. The surveys cited above do not include all seg- ments of the general population of laboratory work- ers at potential risk of occupational exposure to in- fectious agents or their toxic or sensitizing metabolic products. Among the segments inadvertently ex- cluded are a substantial but unknown number of per- sons whose duties may involve either regular or oc- casional handling of infectious materials: e.g., those persons working in animal and avian disease diag- nostic or research laboratories, environmental labo ratories, industrial and biologics production labora- tories, forensic laboratories, and in laboratory animal production and care facilities. Consequently, the estimate of 500,000 workers who may be at risk of occupational infections probably represents a signifi- cant underestimation of the true number. 2. The Infectious Agent Data are also not uniformly available about the 8
DESCRIPTIVE EPIDEMIOLOGY OF OCCUPATIONAL INFECTIONS second epidemiologic component, the infectious agent. Details about the kinds of infectious agents encountered in the various categories of laboratories and the frequency with which such agents are handled are unknown. National morbidity data [311 indicate that hepatitis B virus (HBV) infections are wide- spread in the general population and that a pool of new cases and chronic carriers numbering several million may exist. For example, 1.0 to 1.5 percent of all admissions to large urban hospitals are positive for hepatitis B surface antigen (HBsAg) [451. A1- though blood samples from these patients are poten- tially infectious, their identity is usually unknown throughout the hospitalization. These data suggest that HBV represents the specific infectious agent most likely to be transmitted in the clinical labora- tory occupational setting. 3. The Environment The third component of the epidemiologic triad, the laboratory environment, is also poorly defined. The physical features, including safety equipment and adequacy of the facility, vary widely. Although a number of federal, state, and private sector organi- zations (e.g., Health Care Financing Administration, OSHA, College of American Pathologists, Joint Commission on the Accreditation of Healthcare Or- ganizations, American Association of Blood Banks, and various state agencies) do regulate, license, ac- credit, or inspect clinical laboratories, there is no consensus among these various organizations about the standards for laboratories operating under their respective jurisdictions. There is a current voluntary national code of laboratory practice  that de- scribes features of a laboratory facility recommended for work with infectious agents, but there is no na- tional authority that regulates laboratory operations conducted solely on an intrastate basis. C. LABORATORY-ASSOCIATED INFECTIONS 1. Infectious Agents Presenting the Highest Risk The series of survey summaries of laboratory- associated infections described by Pike in 1976  indicates that, as a group, bacterial infections were the most frequently reported occupationally associ- ated infections of laboratory workers during the first 9 seven and one-half decades of the twentieth century. Viral and rickettsial infections were more frequently reported during the latter half of this time period. Of the 3921 cases reported, the five most frequently recorded infections in rank order were: brucellosis, Q fever, typhoid fever, viral hepatitis (all ~es), and tuberculosis. Most of these diseases were prevalent and important public health problems in our recent past, and their etiologic agents were handled com- monly in clinical and diagnostic laboratories at that time. While the compilations of laboratory-associated infections cited above provide a historical perspec- tive of the hazards of occupational infection, these data are not necessarily indicative of present-day risk of infection. For example, brucellosis, the most fre- quently reported occupational infection, was formerly a major and widespread disease in human and animal populations. More than 6000 human cases were reported in the United States in 1947 [301. As the incidence of brucellosis declined in domestic animal reservoirs, there was a corresponding decline in cases in the human population. By 1963, the annual num- ber of cases reported had declined to 407 [431. Of this number, only one case, a Brucella suds infection, was specifically identified as being laboratory asso- ciated. Sharp decreases have been observed also in the annual number of reported cases of typhoid fever. In 1942, there were almost 6000 cases; in 1952, 2341 cases; and in 1984, 390 cases, of which approxi- mately 70 percent were acquired during foreign travel [301. Similarly, reported cases of tuberculosis de- creased from 121,000 in 1950 to 22,500 in 1984 [301. As the incidence of the "top five" diseases has de- creased in the general population, there has been generally a corresponding decrease in the number of laboratory specimens received and examined that contain these agents. This decrease in numbers of specimens containing the infectious agents has cer- tainly reduced the probability of occupational expo- sure. On the other hand, newly emerging diseases or newly recognized organisms may increase the risks of infection for laboratory personnel. Tuberculosis is an interesting example of the lat- ter situation. The recognition of the existence of species of mycobacteria other than Mycobacterium tuberculosis, M. bovis, and M. avium has required that specimens suspected of containing mycobacteria be subjected to increased laboratory manipulations, in order to recover as well as to identify the organ ~e . .
10 isms. Such increased manipulations are not without inherent risks to laboratory personnel. The recent report of Miller et al.  summarizes the collective experience of several investigators and shows that tuberculosis has ranked among the top six causes of laboratory-associated infections for more than 25 years. These published reports probably represent only the "tip-of-the-iceberg" of laboratory-associated cases of tuberculosis. For example, the laboratory tuberculosis consultant at the Centers for Disease Control (CDC) has, at the time of this writing, been asked to assist in 13 investigations of laboratory- associated tuberculous infection, none of which has been published [741. In these 13 investigations, 72 of 275 (26 percent) exposed individuals were found to have been infected with tubercle bacilli (i.e., re- cent tuberculin conversion); if untreated, 10 percent of these individuals would be expected to develop active tuberculosis. The fraction of exposed person- nel infected in the different outbreaks ranged from 19 to 55 percent. Airborne dissemination was sus- pected in all instances. It has recently been recognized that both acquired immunodef~ciency syndrome (AIDS) [32,33,119,122] and intravenous drug abuse  may be contributing factors to the recent increase in morbidity from tu- berculosis. Other factors that are suspected of play- ing a role are the influx of immigrants from Central America and the increasing numbers of homeless people. These factors can only cause an increase in the risk of mycobacterial infection for laboratory personnel. Despite excellent published surveys of the frequency of such infections, it appears obvious that many cases occur that are not reported in the medical literature. Q fever differs from the other "top five" occupa- tional infections in that this rickettsial disease has remained a relatively obscure public health problem of unknown incidence and sporadic distribution. However, Q fever is a proven and continuing hazard in those few facilities in which work with infected animals or human tissues is conducted, or in which the agent is propagated. This rickettsial agent is remarkably resistant to dessication and inactivation, and 10 or fewer organisms may produce infection via the respiratory route . Hepatitis of varied etiology continues to be a community as well as an occupational health hazard among certain high-risk groups, including laboratory workers. Over the past 15 years, the incidence of BIOSAFEI Y IN THE LABORATORY HBV infections has shown a progressive annual in- crease, while in the same period, the number of re- ported cases of hepatitis A virus (HAY) infection has decreased [311. In 1983, for the first time, the num- ber of HBV cases exceeded those caused by HAV. While laboratory hazards of HAV infection are re- stricted primarily to persons working with experi- mentally or naturally infected chimpanzees, HBV poses a persistent and continuing hazard to all cate- gories of laboratory workers handling clinical speci- mens of human origin. HBV has been demonstrated in a wide range of body fluids and tissues typical of those received and handled in clinical laboratories. The number of infectious virus particles may reach concentrations in excess of 100,000,000 per millili- ter of blood. Since the early 1970s, when procedures for the serologic differentiation of HAV and HBV came into general use, HBV has become the leading cause of occupationally acquired infection among laboratory and health care workers [811. Studies by Jacobsen and co-workers in Utah demonstrated a prevalence of clinical HBV infection in clinical labo- ratory personnel that was 14 times greater than that in the general population, i.e., 0.84 cases/100,000 versus 0.06 cases/100,000, respectively [701. Ele- vated HBV infection rates or incidences of serologic markers were demonstrated in public health labora- tory workers in the United Kingdom , in clinical chemistry workers in Denmark , and in small rural hospitals , as well as in large urban hospi- tals in the United States [661. Osterholm and An- drews  have demonstrated annual infection rates for HBV and non-A/non-B hepatitis that exceeded 10,500 cases/100,000 in the staffs of hospital dialy- sis units, while employees in nondialysis units of these hospitals exhibited a rate of approximately 500 cases/100,000. The ratio of HBV to non-A/non-B hepatitis infections was greater than 5 to 1. Lauer found HBV antigen on one-third of the surfaces of work areas, equipment, and laboratory implements sampled in a large, modern medical cen- ter laboratory . Collins found visible blood on the labels of 17 percent of tubes of blood specimens, and feces on the external surfaces of 6 percent of the containers of He stool samples received at a public health laboratory in England. Four to five percent of laboratory services request forms received by an- other public health laboratory in England were visi- bly blood stained [421. Bond et al. demonstrated that HBV remains viable and infectious after being dried . ~,
DESCRIPTIVE EPIDEMIOLOGY OF OCCUPATIONAL INFECTIONS in serum and held for seven days at ambient labora- tory environmental conditions of 25°C and 42 per- cent relative humidity t201. The primary routes of occupational infection with HBV, in rank order, are as follows: accidental paren- teral self-inoculation with infectious fluids (needle sticks) [45,811; exposure of the mucous membranes of the eyes, nose, or mouth to infectious materials; and, possibly, contamination of the skin with infec- tious materials. In a joint advisory notice, the U.S. Departments of Labor, and Health and Human Services, have in- formed employers about the serious occupational infection problems of HBV, human immunodefi- ciency virus (HIV), and other blood-borne diseases [1391. In this advisory, federal health officials esti- mated that as many as 18,000 health-care workers may be infected in a single year with HBV. Of these cases, as many as 12,000 may be occupationally associated. It was further estimated that nearly 10 percent of the cases will become long-term carriers of the virus, and that more than 200 health care workers may die as the result of the HBV infection or associated complications. The evidence is overwhelming that, of all indige- nous pathogens, HBV has the greatest potential for transmission within the occupational setting of the clinical laboratory. This conclusion is based upon the comparatively high frequency of asymptomatic carriers, the high titers of virus in blood and other body fluids, the stability of the virus on work sur- faces and other items in the laboratory, the low infec- tious dose, the multiple routes of infection, and the demonstrated occupational incidence of infection. An essential consideration in the occupational risk assessment of HBV and other infectious agents for which only Biosafety Level 2 (see Appendix A) practices are recommended is the lack of evidence suggesting that occupational transmission occurs by means of true infectious aerosols, i.e., inhalation of respirable particulates typically less than 5 microns . . In ~ .lameter. While not among the most prevalent occupational infections recorded by Pike (i.e., the "top five") , shigellosis is historically and currently a continuing occupational risk. The low oral infectious dose (on the order of 100 viable organisms) facilitates trans- mission in the occupational setting, as well as in the general population [1421. In a retrospective study of more than 20,000 British medical laboratory work 11 ers, shigellosis was the third most frequently recorded occupational infection, following viral hepatitis and tuberculosis [641. 2. Infectious Agents Presenting the Lowest Risk In contrast to the proven occupational infection hazard of HBV and the other "top five" agents, a number of infectious agents handled in laboratories have exhibited a consistent history of remarkably low incidence or absence of reported occupational infections. Examples of such agents include rabies virus, Creutzfeldt-Jakob agent (CJA), Vibrio chol- erae, Clostridium tetani, and HIV. While the conse- quences of infection with any of these five agents are serious, the cumulative history of laboratory experi- ence attests to the low risk of transmission in the laboratory setting. In the almost 100-year history of work with ra- bies virus in diagnostic and research laboratories, often in the most primitive of facilities and without preexposure immunization of personnel, only two documented cases of laboratory-associated infections have been recorded. Both cases occurred under con- ditions involving the manipulation of relatively large quantities of high-titer virus suspensions: one in a production facility  and the other in a research laboratory [291. Exposure of personnel to aerosols of high-titer virus suspensions was the most plau- sible explanation for each of these two unusual cases. Neither the quantity nor the concentration of the virus in the materials handled, nor the procedures performed, was typical of the conditions in a diag- nostic or clinical laboratory. Creutzfeldt-Jakob agent, a slow virus causing transmissible viral dementia, is an infectious agent that can be passed serially from human to human, and from human to susceptible nonhuman primates or rodents. Extensive experience with CJA and the clinical disease (CJD) up until the present time has indicated that "none of the people in closest contact with patients with CJD (wives, friends, employee contacts, members of the medical or nursing profes- sions, or paramedical personnel) appears to have a higher risk of contracting CJD than does the general population. Not a single case of CJD has yet been reported to have occurred in workers most exposed to infectious tissues from patients with CJD (neu
12 ropathologists, research scientists, and laboratory personnel). Thus, despite proven person-to-person transmissibility of the disease by invasive procedures, the risk of acquiring CJD by any means other than tissue penetration by contaminated materials must be very small indeed" [231. It should be noted that two accounts of the occurrence of CJD in laboratory work- ers were recently published, although the causal rela- tionship between the disease and occupational expo- sure was not established in either case [85,1131. While cholera is periodically epidemic in tropical and subtropical countries, only 12 laboratory-associ- ated infections have been reported during this cen- tury [1001. The very high oral dose required for infection, of the order of 100,000,000 viable organ- isms , is undoubtedly a major reason for the small number of laboratory-associated cases. Although Pike recorded five laboratory exposures to toxin of Clostridium tetani produced in vitro , there have been no recorded cases in laboratory work- ers of occupational infections or intoxications with C. tetani, C. botulinum, or their respective toxins. Few infectious agents have generated more con- cern and anxiety over potential occupational expo- sure and hazards of infection than has HIV. Active prospective surveillance, however, has shown that fewer than 1 percent of overt exposures (including needle sticks) of people attending patients with AIDS or with other manifestations of HIV infection, have resulted in seroconversion of the exposed individuals [351. The majority of those health care workers with reported occupationally acquired HIV infection have a history of needle stick exposure to blood of in- fected patients in the clinical setting. As of February 1988, there have been three reported seroconversions in laboratory workers. One of these three cases occurred in a medical technologist who spilled blood from an infected patient on her ungloved hands and forearms while manipulating an apheresis machine [351. The other two recorded cases occurred in em- ployees of large-scale virus production facilities propagating HIV for research or reagent use [36,1431. One of these two workers had an overt parenteral exposure to a concentrated virus preparation. The other worker had no recognized accidental occupa BIOSAFETY IN THE LABORATORY tional exposure or any risk behavior linked to HIV infection. Despite the low incidence of transmission in the laboratory, the potentially life-threatening conse- quences of HIV infection mandates that all labora- tory workers who handle blood, body fluids, tissues, or cultures utilize those laboratory practices and per- sonal protective measures identified as 'universal Precautions" by the CDC, which are recommended for the prevention of transmission of HIV and other blood-borne diseases [341. Common to each of these infectious agents with a demonstrated low risk of occupational infection is the fact that primary occupational infection is associ- ated with one of the following exposures: accidental parenteral inoculation (e.g., needle stick); contami- nation of the mucous membranes of the eyes, nose, or mouth with infectious droplets (par~culates typi- cally greater than 5 microns in diameter); ingestion; or penetration of the intact or broken skin by the agent. There is no documented risk of transmission by means of an infectious aerosol (particulates typi- cally less than 5 microns in diameter) generated dur- ing the manipulation of clinical materials or of diag- nostic quantities of the agent. 3. Other Infectious Agents Except for some exotic microbial agents, the oc- cupational risk of infection with virtually any bio- logical agent falls between the extremes observed with the"top five" and the"low-risk" groups of infectious agents described above. The recommended facilities, equipment, and microbiological practices necessary for the handling of infectious agents are detailed in Chapter 3 and in Appendix A, which is a reprinting of the Centers for Disease Control (CDC)/ National Institutes of Health (NIH) publication, Bio- safety in Microbiological and Biomedical Laborato- ries [1051. These guidelines should be followed when contemplating work with any potentially infec- tious agent. Recommendations for handling HIV, an infectious agent that was identified after the CDC/ NIH publication, are reproduced in Appendixes B and C [34,36,381.