Role of Public Health Laboratories in the Control of Tuberculosis
Robert C. Good, Ph.D.INTRODUCTION
Awareness of what will happen if tuberculosis program efforts are not sustained, both in this country and in the world as a whole, is necessary to maintain control of the disease. Experiences from the last few years are sufficient to warn of the need to be prepared for a resurgence of tuberculosis.
There are 800 to 900 laboratories in the United States that can be classed as public health laboratories supported by government (federal, state, or local) funding (Becker, 1999). These laboratories are part of health programs that are concerned with the control of diseases, especially contagious diseases. Tuberculosis is a contagious disease that has been targeted for elimination, defined as a case rate of less than 1/1,000,000 population by 2010, with an interim target of 3.5/100,000 population by the year 2000 (Centers for Disease Control, 1989). The case rate in 1998 was 6.5 per 100,000, down 35% from the rate in 1992 (Centers for Disease Control, 1999). The data indicate that the program for the elimination of tuberculosis is being successful, and a part of the success is associated with advances made in applied diagnostics, “including new methods that reduce the time needed to detect growth of Mycobacterium tuberculosis in diagnostic specimens” (Advisory Council for the Elimination of Tuberculosis, 1999). The ACET has also renewed the call for the development of a new vaccine against tuberculosis. In carrying out this developmental program, the TB/Mycobacteriology Branch at the Centers of Disease Control
Page 206(CDC) should be key, establishing and using animal models to test effectiveness and safety of potential vaccines.
To confirm the diagnosis of tuberculosis, the laboratory must isolate Mycobacterium tuberculosis from the patient's specimen. This has been true since the laboratory was first able to isolate tubercle bacilli, even if guinea pig inoculation were needed to improve the chances of isolation by passage in a highly susceptible host. When specific therapeutic regimens for the treatment of tuberculosis became available in the 1950s, the incidence of the disease began to decline to such an extent that 40 years later we were able to anticipate its eradication in the United States. As quickly as the number of cases declined, tuberculosis laboratory services in diagnostic mycobacteriology became less important in the overall activities of public health and clinical laboratories. Laboratories at that time were limited to the use of the slow, but reliable, methods that had been developed over decades. Therefore, detection of acid-fast bacilli by a specific stain could be accomplished in a local laboratory or on the ward of a hospital within hours of specimen collection, but isolation of tubercle bacilli and identification based on unique characteristics required inoculation of many tubes of solid media and subcultures to measure biochemical reactions. Drug susceptibility tests also required the inoculation of a solid medium prepared with assay levels of the drugs. Results of these tests could take from 6 weeks to 6 months to complete, and during this time patients with tuberculosis were continuing to be in contact with community members, continuing to infect others by exposing them to aerosols produced by coughing and sneezing.
The laboratories responsible for tuberculosis diagnosis were not ready for the expanded workload that resulted from the increased number of cases that began to occur in 1986, when, without precedent in this century, the tuberculosis case rate reversed a downward trend and began to increase. Nor were they prepared for testing the strains resistant to multiple drugs that were isolated from nosocomial outbreaks of tuberculosis in the last decade of the century.
New procedures for the isolation and identification of M. tuberculosis were developed, notably the radiometric BACTEC system (Roberts et al., 1983), high performance liquid chromatography (HPLC) analysis of mycolic acids for speciation (Butler et al., 1992), molecular procedures for identification (Shinnick and Jonas, 1994), and fingerprinting techniques which provide data useful in epidemiologic studies (van Embden et al., 1993). However, laboratories were slow to incorporate the new techniques for identification and drug susceptibility testing when they became avail
able (Huebner et al., 1993, Woods and Witebsky, 1993). Since the laboratory was limited by the slow growth of the tubercle bacillus in culture, the predominant attitudes were that there was no need to rush reporting; however, with the newer techniques, reports of identification and drug susceptibility could affect patient treatment and recovery rather than being an addendum that might be of epidemiologic significance only.
Understanding that a radical change needed to be made if laboratory results were to contribute to the early diagnosis and treatment of patients with tuberculosis, Tenover et al. (1993) acknowledged improved technologies available at that time and recommended that tuberculosis laboratories take positive steps to improve performance, particularly in the time taken for testing and reporting results. These recommendations are still applicable as we start the year 2000, with modification only in the addition of newer probes which can be used to probe a specimen directly. The recommendations are as follows:
(i) Promote the rapid delivery of specimens to the laboratory on a daily basis, even if this requires pickup of individual specimens to guarantee arrival within 24 h.
(ii) Use a fluorescent acid-fast staining procedure and use the time saved to immediately transmit the results by telephone or facsimile device for inclusion in patients' records.
(iii) Report patients who are suspected of having tuberculosis on clinical grounds or patients with acid-fast bacilli present on smears to the local health department promptly so that appropriate public health management (including contact investigation) can be initiated.
(iv) Inoculate a liquid medium as primary culture, and include inoculation of a slant of Lowenstein-Jensen medium.
(v) Identify growth in liquid medium as acid fast and use probes, NAP, or mycolic acid patterns to identify isolates as M. tuberculosis as soon as possible. Notify the local health department of the species identification as soon as it is known.
(vi) Determine the susceptibilities of M. tuberculosis isolates to primary drugs in a BACTEC or similar system.
(vii) Report the results of drug susceptibility testing to the clinician as soon as they are available by telephone or facsimile and follow up with hard copy. If drug resistance is present, the state tuberculosis control program should be notified promptly.
(viii) Maintain up-to-date records that include the results of quality control procedures.
(ix) Review all laboratory procedures and facilities to guarantee safety for workers.
Studies carried out on the basis of these recommendations should ensure reports of acid-fast examination of specimens within 24 h of specimen collection, identification of M. tuberculosis within 10 to 14 days, and reports of drug susceptibility tests within 15 to 30 days of specimen collection.” At several points in the presentation of this challenge, reporting is emphasized. Problems arise when reports are made only to a client, such as a physician, without including notification of the tuberculosis control program. Most states specify that in-state laboratories must notify the program, but out-of-state laboratories are not covered. The notification procedures are very detailed in Title 17, California Code of Regulations, Section 2505, Notification by Laboratories for in-state laboratories and out-of-state laboratories are covered. Tuberculosis controllers, public health laboratory directors, and health officers throughout the state have been made aware of this problem, but reports may not be received from the out-of-state laboratory (Royce, 1999). When these cases are not known to the tuberculosis control program, contact studies are not initiated, and additional cases can occur.
As we go into the new millennium tuberculosis laboratories are able to isolate and identify tubercle bacilli in one to 15 days, and drug susceptibility data are available 30 days after the collection of a specimen. This level of performance is not the case in all laboratories, but in those that have actively upgraded to meet the demands for the rapid reporting of results. This ability to perform depends on a commitment to respond rapidly. Such a laboratory will have well-trained personnel using stateof-the-art methodology and appropriate equipment. The laboratory will adhere to safety guidelines and develop a safe working environment. Using up to date communications equipment (computers, FAX, telephone), the laboratory will report results of tests to aid in patient care and community disease control. These are the laboratory goals that can be achieved if specimen loads are large enough to maintain efficiency and proficiency in the laboratory.
Every tuberculosis laboratory is a mycobacteriology laboratory which is responsible for the identification of any Mycobacterium species that may be associated with a pathologic condition. Therefore, methods should be selected that will distinguish any one of the more than 71 species that have been identified (Good and Shinnick, 1998). As the number of cases declines, the laboratory is faced with the necessity of identifying the etiologic agent of a condition to rule out involvement of the tubercle bacillus, the Mycobacterium species that causes a contagious disease.
CURRENT LABORATORY PROCEDURESSafety in the Tuberculosis Laboratory
Because of the highly infective hazard, the following steps are carried out in accordance with the recommended biosafety levels (CDC/NIH, 1999), that is, “Biosafty Level 2 practices and procedures, containment equipment, and facilities are required for non-aerosol-producing manipulation of clinical specimens such as preparation of acid-fast smears,” but “all aerosol-generating activities must be conducted in a Class I or II biological safety cabinet . . . Biosafety Level 3 practices, containment equipment, and facilities are required for laboratory activities in the propagation and manipulation of cultures of M. tuberculosis or M. bovis. . . .” This requirement, along with maintenance of facilities and equipment, increases the costs of constructing and operating tuberculosis laboratories; however, the safety of personnel must be ensured. Biological safety cabinets must be inspected and certified in situ at least annually by approved technicians, and personnel in the laboratory must be trained in their proper use, including location, control of room air flow, placement of materials and equipment within the cabinet, and understanding of the mechanics of air flow. Safety in the tuberculosis laboratory is essential in order to prevent infection of personnel and spread of tubercle bacilli through inappropriate disposal of contaminated wastes. Details of the necessary safety procedures have been published (CDC/NIH, 1999).Detection of M. tuberculosis
M. tuberculosis is one of the 22 slowly growing Mycobacterium species out of the 71 recognized or proposed species in the genus (Good and Shinnick, 1998). The number of species in the genus has increased to over 80 in the past few years as new species continue to be encountered (Metchock et al., 1999).Specimen Collection and Transport
Sputum is the typical specimen collected for detection of M. tuberculosis although others, such as blood, urine or tissue, may be collected depending on the site of disease. Once collected the specimen must be delivered to the laboratory as quickly as possible. It is inappropriate for clinics to hold collected specimens to send in a batch together. Commercial laboratories have regular pick-up of specimens, and the same should be true in the public health system. This is an easily controlled step, but many laboratories have continued to use the Postal Service so that specimen
delivery may take as long as a week. If necessary, public health laboratories must use a courier service to get specimens to the central laboratory within 24 hours of collection. This has not been a problem for water quality laboratories, and elaborate systems for collection and delivery have been devised by these programs that can be example for the delivery of mycobacterial specimens.
In the case of sputum, a specimen is collected on at least three, but up to five, days to optimize isolation of tubercle bacilli. If the specimen is to be shipped for any distance, the primary container is placed in a water tight secondary package and an outer package certified to meet performance tests specified by carriers (CDC/NIH, 1999). An “Infectious Substance” label must be attached to the exterior package. If the specimen is to be transported for a short distance within a building, the primary container should be placed in a water tight package and transported with care. Decontamination and Staining
Because the specimen usually contains bacteria other than mycobacteria, it must be decontaminated prior to culture on the rich media that were formulated to promote the growth of mycobacteria in general and of drug resistant tubercle bacilli in particular. The decontamination procedures are harsh, so the steps must be carried out carefully to keep from destroying mycobacteria. The usual decontamination agent is two percent sodium hydroxide in solution with the liquefying agent N-acetylL cysteine (NALC) (Kent and Kubica, 1985). Equal volumes of specimen and decontamination solution are mixed, incubated for 20 minutes, diluted to 50 ml. with neutralizing buffer and concentrated by centrifugation at 3000 to 3800 × g for 15 minutes (Metchock et al., 1999). The sediment is then used to prepare smears for staining and to inoculate various liquid and solid media. The decontamination conditions may be varied if bacteria other than mycobacteria grow on primary media, or if smear positive specimens do not then yield mycobacteria on culture.
All species in the genus Mycobacterium are acid-fast: that is, once the bacterial cell is stained with a basic fuchsin dye, it will resist decoloration by acidified alcohol. This very basic test does not permit determination of species, yet it is the only laboratory procedure used in many developing countries to support the clinical diagnosis of tuberculosis. The sensitivity of the procedure has been increased by using a fluorochrome stain and fluorescence microscopy. The advantage of the latter procedure, which requires more expensive instrumentation, is that the slides can be read with the low power or high dry objectives on the fluorescence microscope to observe 30 fields in about 90 seconds versus observation of 300 fields
on carbolfuchsin stained slides using an oil immersion lens on a standard light microscope in about 15 minutes. In either case, the appropriate number of fields must be observed before declaring a specimen negative for acid-fast bacilli (AFB). Many laboratories that do use fluorescence microscopy do a completely unnecessary and time-consuming confirmation step on AFB-positive smears, that is, restaining using the routine Ziehl-Neelsen procedure and observation with the light microscope. This is time that could be spent in better ways. Culture and Identification Current Procedures
Until 1993, the routine procedure in many tuberculosis laboratories was to inoculate the concentrated specimen onto one or more slants of solid medium, such as Lowenstein-Jensen medium, which is an egg-based formulation, or Middlebrook 7H10 or 7H11 medium, which are agarbased media, and then to incubate the medium for at least 21 days at 37° C in an atmosphere of 5 to 10 percent carbon dioxide in air. When growth occurred, supplemental media were inoculated for identification of the Mycobacterium species using biochemical tests to determine characteristics of the strain; but if growth did not occurr, the cultures were incubated for at least 6 weeks before judging them to be negative, a criterion that is still applicable. If M. tuberculosis was identified by the biochemical tests, the culture was tested for drug susceptibility. Completion of these tests for identification and drug susceptibility could take weeks or months because of common laboratory practices that included batching, that is, holding specimens at various points until a sufficient number was on hand to make processing efficient. Rapid Methods
High performance liquid chromatography (HPLC) is a rapid procedure for the identification of all mycobacteria. It was developed at the CDC to identify the spectrum of mycolic acids in the bacterial cell (Butler et al., 1992). When approximately 106 cells are available in culture (in 7 to 10 days), the culture can be analyzed, and within four hours the Mycobacterium species can be identified. The reference laboratory at CDC now uses this procedure for identification of cultures submitted, and many state and other government laboratories and universities have begun identifications by this technique. Instrumentation cost is about $40,000, but the cost for chemicals, etc., in minimal. Savings in time required to report the species can be weeks or more. The HPLC Users Group has established a
Standardized Method for Identification that is available at the CDC internet address.
Development of newer techniques to overcome the drawbacks of the BACTEC system promise to continue the advantages of rapid detection and identification. These include the Mycobacteria Growth Indicator Tube (MGIT), Becton Dickinson Diagnostic Instrument Systems, Sparks, MD, which detects oxygen utilization (Hanna et al., 1999a); the MB/BacT system, Organon Teknika Corporation, Durham, NC, which detects carbon dioxide evolution (Benjamin et al., 1998); and the LCx M. tuberculosis Assay, Abbott Laboratories, North Chicago, IL, which uses ligase chain reaction technology to detect M. tuberculosis (Ausina et al., 1997). The latter method is based on direct examination of the specimen while the first two methods are based on culture, and their performance is equivalent to the BACTEC system for detection and isolation of M. tuberculosis. These two culture methods are automated so that once inoculated with the processed specimen, the cultures do not have to be handled again until the time of subculture. Therefore, the advantages of the systems are decreased labor input, inclusion of a computerized data management system, use of a nonradioactive substrate, and a noninvasive technique to measure enzymatic activity of mycobacteria in culture.
The BACTEC MGIT 960 system is a fully automated, noninvasive setup for the growth and detection of mycobacteria by continuously monitoring 960 7-ml culture tubes. Hanna et al. (1999a) conducted a multicenter study to compare this system with the BACTEC 460 TB system and with Lowenstein-Jensen Medium and Middlebrook 7H11 selective plates for the detection of mycobacteria. From 3,330 specimens inoculated to all culture systems, 132 were positive for M. tuberculosis. The number of positive specimens detected by the BACTEC 460 plus solid media was 128 (97%), by MGIT 960 plus solid medium was 121 (92%), by the BACTEC 460 alone was 119 (90%), by MGIT 960 alone was 102 (77%), and by solid medium alone was 105 (79%). The mean times to detection were 14.4 days for MGIT and 15.2 days for BACTEC 460. These data suggest that the BACTEC 460 is the better of these two systems, but increasing concerns regarding the disposal of medium containing radioactive carbon limits its use.
Specimens can be examined directly for tubercle bacilli by PCR (Kox et al., 1997; Reischl et al., 1998; Wang and Tay, 1999; Wobeser et al., 1996), ligase chain reaction (Ausina et al., 1996; Moore and Curry, 1998; Rohner et al., 1998), and strand displacement amplification (Bergmann and Woods, 1998). In each of the studies, the particular method performed at an acceptable level, and all of them performed better with smear positive specimens than with smear negative specimens.
The Gen-Probe M. tuberculosis Amplified Direct Test (AMTDII), which
was developed for use in sputum specimens, was compared with culture and found to have a sensitivity of 93.6% and a specificity of 97.8% (Bradley et al., 1996). However, the test was more sensitive in patients with undiagnosed disease (74.7%) than in those who were receiving chemotherapy (29.2%). Also, sensitivity of the test was 95.5% in smear positive specimens and 70% in smear negative specimens. The authors concluded that AMTDII, when used in conjunction with routine smear and culture, is a useful rapid diagnostic test for suspected pulmonary tuberculosis.
Gladwin et al. (1998) concurred with that conclusion and proposed an algorithm using MTDT, acid-fast smear, and culture for the diagnosis and treatment of immunocompromised patients with suspected mycobacterial infection.
In a comparison of AMTDII, which detects rRNA, and the Abbott LCx Semiautomated Assay, based on ligase chain reaction, Piersimoni et al. (1998) found in direct tests that the sensitivity, specificity, and positive and negative predictive values for respiratory specimens were 92.8, 99.4, 98.5, and 97%, respectively, for AMTDII and 75.7, 98.8, 96.4, and 90.5%, respectively, for LCx. With extrapulmonary specimens, the values were 78.6, 99.3, 95.6 and 96.2%, respectively, for AMTDII and 55.6, 99.3, 93.7, and 92.1%, respectively, for LCx. The level of agreement between AMTDII and LCx assay results was 78.2%. The authors concluded “that although both nucleic acid amplification methods are rapid and specific for the detection of M. tuberculosis in clinical specimens, AMTDII is significantly more sensitive than LCx with both respiratory (p = .005) and extrapulmonary (p = .048) specimens.”
After recommendations of Food and Drug Administration Panels to approve the Gen-Probe MTD and the Amplicor M. tuberculosis test (Roche Diagnostic Systems), the American Thoracic Society convened a workshop to examine the data and technology, to develop a consensus addressing the appropriate use of rapid diagnostic tests for tuberculosis, and to identify future research needs and directions (American Thoracic Society Workshop, 1997). The consensus agreement from the workshop was that while these tests are a major improvement over standard techniques, there is insufficient information on their clinical and public health utility. Guidelines for interpreting the tests were advanced so that when the AFB smear and the direct amplification test (DAT) are both positive, the diagnosis of tuberculosis can be considered to be established. When the AFB smear is negative and the DAT is also negative, it is unlikely that M. tuberculosis will be grown from the specimen. When the results of the AFB smear and DAT do not agree, additional tests are needed. Results of the DAT tests must be interpreted within the overall clinical setting in which they are used.
The nucleic acid tests detect M. tuberculosis Complex, that is, M. tuber
culosis, M. bovis and M. africanum; however, M. tuberculosis is by far the most frequent isolate in the United States. The true species can be discerned through subsequent isolation and identification steps that will identify the species.
In addition to the identification tests based on biochemical reactions, nucleic acid probes, and mycolic acid patterns, some investigators have been developing serologic tests, but often with poor success. Zhou et al. (1996) reported a rapid membrane-based serologic assay using the 38-kDa antigen from M. tuberculosis. This assay had overall sensitivity, specificity, and positive and negative predictive values of 92, 92, 84, and 96%, respectively, for sputum-positive patients with tuberculosis, and 70, 92, 87, and 79% for sputum-negative patients. Only 2% of healthy control BCG-vaccinated subjects gave weak positive reactions in the assay. However, the primary problems with the serologic procedures have been cross-reactivity with sera from patients who were infected with other mycobacteria so that the reactions were not reliable for diagnosis.
A test to measure cellular immunity was developed by Lein et al. (1999) to discriminate patients infected with Mycobacterium avium from those infected with M. tuberculosis. The test is based on in vitro gamma interferon responses by peripheral blood mononuclear cells to a 6-kDa early secreted antigenic target which is found almost exclusively in M. tuberculosis Complex species. Significant responses were detected in 16 of 27 patients with pulmonary tuberculosis, but the antigen was not detected in specimens from any of the eight patients with M. avium disease or in any of the eight healthy controls. When purified protein derivative (Tuberculin PPD) was used as the antigen, the mononuclear cells from 23 of 27 patients with M. tuberculosis disease, two of eight patients infected with M. avium, and five of eight healthy controls gave significant gamma interferon responses. This reaction is one that may be developed into a sensitive test for the diagnosis of tuberculosis, but it is not sensitive enough in its present form to replace nucleic acid tests for diagnosis.
Although nucleic acid amplification techniques appear to be rapid and accurate procedures, Noordhoek et al. (1996) found that of 30 laboratories, only five correctly identified the presence or absence of mycobacterial DNA in 20 samples. Seven laboratories detected mycobacterial DNA in all positive samples, and 13 laboratories correctly reported the absence of DNA in negative samples. The authors concluded that many laboratories do not use the quality controls that are necessary to support the results of their tests. Ridderhof et al. (1998) sent 10 samples containing mycobacterial DNA to 86 participating laboratories (47 hospital, 23 health department, 13 independent, and 2 other laboratories) for detection of M. tuberculosis nucleic acids. Sensitivity for the detection of M. tuberculosis was 97.9% for all procedures used, but the rate of false positives was high
in the five specimens that did not contain M. tuberculosis nucleic acids. The authors believe that the failure to follow NCCLS recommendations that biological safety cabinets used for culture should not be used for nucleic acid amplification studies may be responsible for the high number of false positives. These two studies serve as warnings that should be remembered as there is greater emphasis placed on developing more rapid tests for detection of M. tuberculosis.
New products continue to be developed for the detection and identification of M. tuberculosis because of the large numbers of specimens that are processed yearly to either confirm the diagnosis of tuberculosis or to rule it out as a possible diagnosis. In addition to the direct commercial advantage enjoyed by a product that is accurate and rapid, the National Institutes of Health has provided money for research to universities and other research programs to encourage studies of the fundamental nature of M. tuberculosis. Technology developed through these two sources keep diagnostic mycobacteriology current. Tests for Drug Susceptibility
All isolates of M. tuberculosis are tested for their susceptibility to known antituberculosis drugs so that optimal therapy will be given as soon as possible. However, knowing patterns of resistance in a community is the best guide for initial therapy since results of drug susceptibility testing will take up to 30 days to obtain. When the results are available, a clinical decision must be made regarding continuation of primary treatment or adjustment based on the outcome of the tests. Perspective
Drug resistance is a decrease in susceptibility of a strain of M. tuberculosis to such an extent that it is different from wild strains that have never been in contact with the drug. However, some resistant bacilli can be present in numbers too small to affect treatment. Therefore, “Resistance on the part of the microorganisms is clinically significant when at least 1% of the total bacterial population develops at the so-called critical concentrations, that is, the weakest concentrations at which susceptible bacilli are unable to grow in the presence of the drug” Canetti et al. (1963). “Critical concentration” is further defined as follows: “For each drug it is necessary to determine the lowest concentration at which the bacilli may no longer be considered susceptible, but are to be regarded as resistant (“critical” concentration). This concentration should be determined in each laboratory according to the existing experimental conditions, on a series of so-called ‘wild' strains cultivated side by side with control strains.
Under the abovementioned test conditions, the ‘critical' concentrations are considered to be: 0.2 mg/ml for isoniazid; 5.0 mg/ml for streptomycin; 0.5 mg/ml for PAS; 1 mg/ml for thioacetazone; 20 mg/ml for ethionamide. If at these concentrations growth is observed to the extent of more than 20 colonies (on LJ medium with an inoculum of 2 to 10 × 103), the strain is to be considered resistant.”
Changes in test medium can cause the critical concentration to change as indicated in Table D-1 . In addition to the critical concentrations, other concentrations are often tested so that physicians may develop a successful therapeutic regimen. Therefore, on Middlebrook 7H10 medium isoniazid is routinely tested at 0.2, 1.0 and 5.0 mg per ml, and streptomycin is tested at 2.0 and 10.0 mg per ml. The critical concentration of a drug is always included so that there is a historical basis to determine if resistance increases in specific populations over time (Good and Shinnick, 1998).
The basic test to determine resistance of M. tuberculosis must provide information on the total numbers of viable and resistant tubercle bacilli present in the inoculum. Therefore, dilutions are inoculated to medium without the drug and to drug-containing medium. This has been carried over in newer tests such as the BACTEC system for determining drug susceptibility to provide information on the percent of resistant bacilli in the inoculum.
David (1970) determined that tubercle bacilli spontaneously mutate to resistance to four major antituberculosis drugs. The average mutation rates to isoniazid and streptomycin resistance were 3 × 10−8, to ethambutol 1 × 10−7, and to rifampin 2 × 10−10. The mutation rate for resistance to two drugs is less than 10−15. Inderlied and Salfinger (1999) discuss a special population hypothesis and the action of various antituberculosis drugs acting simultaneously to both prevent drug resistance and achieve a maximum therapeutic effect.
Please refer to the page image for an unflawed representation of this content.
Current Methods to Measure Susceptibility
Inderlied and Salfinger (1999) discuss the philosophy for susceptibility testing of primary isolates and provide detailed methodology for testing, particularly for the BACTEC procedure. A liquid medium is included for the primary isolation of AFB. M. tuberculosis strains grow more rapidly in a liquid medium, such as BACTEC, so that the bacilli can be identified in two weeks or less, and the liquid medium is ready to use as an inoculum for drug susceptibility tests in new vials of BACTEC medium. The radioactive substrate used in the BACTEC is a disadvantage, and the radioactive carbon dioxide in the vials must be read frequently. New tests have been developed that incorporate newer methodology to provide continuous monitoring for growth without the use of a radioactive substrate or individual handling of vials (Inderlied and Salfinger, 1999). Rapid Methods
The BACTEC MGIT 960 Antimycobacterial Susceptibility Test System (MIGIT 960 AST) was developed to overcome the drawbacks of the BACTEC 460 system. The preliminary report of a limited study by Hanna et al. (1999b) indicates through a multilaboratory evaluation that it is a rapid and reproducible system. The investigators found that the results were in “high” agreement with a proportion method.
Mycobacteria Growth Indicator Tubes (MIGIT) are available and can be used without the continuous reading system. In this case, the tubes are screened by testing for fluorescence under an ultraviolet source in a darkened room without the cost of the automated reader and recorder. Rüsch-Gerdes et al. (1999) reported the results of a multicenter study evaluating this procedure and comparing the results with those obtained in the BACTEC 460 system. A total of 441 isolates were tested for susceptibility to isoniazid (INH), streptomycin (SM), rifampin (RIF), and ethambutol (EMB). Discrepant results were obtained for three isolates (0.7%) with INH (susceptible by MIGIT, resistant by BACTEC 460TB), for four isolates (0.9%) with RIF (susceptible by MGIT, resistant by BACTEC 460TB), for six isolates (1.9%) with EMB (four susceptible by MGIT, resistant by BACTEC 460TB; two resistant by MIGIT, susceptible by BACTEC 460TB), and for four isolates (0.9%) with SM (two susceptible by MIGIT, resistant by BACTEC 460TB; two resistant by MIGIT, susceptible by BACTEC 460TB). When these cultures with discordant results were tested by the conventional proportion method, the results agreed about 50% of the time with the BACTEC 460TB procedure and about 50% of the time with the MIGIT manual system. Turnaround times were 3 to 14 days (median, 8.8
days) for MIGIT and 3 to 15 days (median, 7.8 days) for BACTEC 460TB. The difference in results from the two methods was not statistically significant. In this study the data demonstrate that the MGIT tube is an accurate nonradiometric alternative to the BACTEC 460TB system for rapid susceptibility testing of M. tuberculosis. This system, then, is one that could be used in a developing country where money for automation is not available.
Newer methods are being developed by using genetic probes (Martin-Casabona et al., 1997, for example) or a luciferase reporter phage (Riska et al., 1999). The capabilities that are available for study of the action of drugs on M. tuberculosis have expanded in the last decade, and future studies based on the initial findings should result in more rapid methods for susceptibility tests.
The relatedness of strains of M. tuberculosis isolated from case-patients who appear to form a cluster has depended on biochemical reactions, drug susceptibility patterns, and phage typing. For different reasons, all of these procedures were either inconvenient or inaccurate. A method of DNA fingerprinting was developed by Eisenach et al. (1988) based on detection of an insertion sequence, IS6110, in the mycobacterial genome. van Embden et al. (1993) proposed standardized procedures for this restriction fragment length polymorphism (RFLP) so that data from different laboratories could be compared. Typing has been used in a number of instances to provide additional data for epidemiologic evaluations.
As pointed out by Kamerbeek et al. (1997), widespread use of RFLP to differentiate strains of M. tuberculosis has been hampered by the need to culture the organism and the level of sophistication needed for typing. In consequence of this observation, these investigators developed a method which allows simultaneous detection and typing in clinical specimens and reduces the time between suspicion of the disease and typing from one or several months to one or three days. The method is referred to as spacer oligotyping or “spoligotyping.” In this study most of the isolates showed unique hybridization patterns, but outbreak strains shared the same spoligotype. The procedure was found also to differentiate M. bovis from M. tuberculosis.
In a study comparing molecular markers, Kremer et al. (1999) found that the RFLP typing methods were highly reproducible and concluded that, for epidemiological investigations, strain differentiation by IS6110 RFLP or mixed-linker PCR are the methods of choice. The IS6110 fingerprint patterns have high degrees of stability (Niemann et al., 1999); however, some multidrug resistant strains of M. tuberculosis may evolve too
quickly for reliable interpretation of strain typing results over a period of a few years (Alito et al., 1999).
The diversity of IS6110 fingerprints of M. tuberculosis was determined by examining isolates from 1,326 patients in three geographically separate states (Yang et al., 1998). A total of 795 different IS6110 fingerprint patterns were recognized, and the pattern diversity was similar in for all three states. Ninety-six percent of the fingerprint patterns were seen in only one state, demonstrating that most patterns are confined to a single location. The authors conclude that identical fingerprints of isolates from geographically separate locations reflect interstate transmission in the past, with subsequent intrastate spread of disease.
Many studies over the past few years have indicated the value of RFLP typing in the epidemiology of tuberculosis and for indicating laboratory cross contamination as a problem. Development of similar probes will further benefit studies of tuberculosis. As collections of data are analyzed for various areas, sites may be identified as centers for the spread of particular types. Collection of these data is only an exercise if the results are not shared with the proper tuberculosis control program for follow-up.
QUALITY ASSURANCE AND CONTROL
If there are problems that affect patient care, a quality assurance (QA) plan can be developed to define the problem, to propose changes to correct the problem, and to monitor the correction process (Sewell and MacLowry, 1999). Quality improvement (QI) as defined by these authors is a management tool used to define the customer's expectations, to describe and evaluate the processes used to provide services, and to continuously improve these processes and outcomes. QI focuses on the customers needs rather than process problems, and the customer in this case may be the patient, the health care provider, the third-party payers, or others who interact with the laboratory. Quality control (QC) includes evaluating all of the steps of a test procedure to determine if the result of the procedure is accurate. Since the methods used in the tuberculosis laboratory are high complexity tests, they must be verified regularly for accuracy and precision. Methods for verification and reliability of the tests should be included in the laboratory's policy and procedures manual. Proficiency Testing
The Clinical Laboratory Improvement Amendments (Title 42 CFR 493.825) require that all laboratories that perform any testing of mycobacteria enroll in a federally approved proficiency testing (PT) program. In
1994, there were 2862 mycobacteriology laboratories enrolled in one or more of six PT programs approved by the U.S. Department of Health and Human Services (CDC, 1995). Approved PT programs for mycobacteriology are the College of American Pathologists; the American Association of Bioanalysts; the states of New Jersey, New York, and Wisconsin; and the Commonwealth of Puerto Rico. Of the enrolled laboratories, 2,179 reported that they performed primary culture for M. tuberculosis. Of these, 1,166 (54%) referred AFB-positive isolates to another laboratory for identification and drug-susceptibility testing, while 699 (32%) performed primary culture with identification, and 314 (14%) performed all tests. The break in procedures indicated by this survey slows reports of results so that the entire process of detection, isolation, and drug susceptibility testing is out of phase.
The New York State Department of Health instituted a FAST Track for Tuberculosis Testing which has been highly successful (Salfinger et al., 1998). The unique FAST Track system, which is available to all physicians who provide health care in New York, includes the submission of specimens by overnight mail, 7-days-per-week service hours in the laboratory, nucleic acid amplification and susceptibility results reported by telephone, and same day reporting by FAX. The authors also list standards of laboratory practice, turnaround times, and quality assurance.
LEVELS OF SERVICE
In 1974 and again in 1983 there was concern that decentralization of tuberculosis management services might result in decreased proficiency as laboratories received fewer specimens for processing. Currently that fear is again surfacing as the number of tuberculosis cases continues to decline. Certainly, laboratory expertise will also decline if an adequate number of specimens is not processed to maintain laboratory efficiency. The laboratory that does not receive many specimens will lose financial support and will be in even greater decline.
The American Thoracic Society (1983) published Levels of Laboratory Service as a guide for referral of specimens. The original concept of having three levels of service has been debated, as well as prescribing the number of specimens that must be processed to maintain proficiency. The Level I Laboratory is to collect good specimens and ship to a higher level laboratory for culture. Level I laboratories may perform microscopic examination. If the Level I laboratory prepares and examines smears, then 10 to 15 specimens per week must be examined to maintain proficiency. The Level II Laboratory performs the same functions as Level I but also isolates organisms in culture, identifies Mycobacterium tuberculosis, and
performs susceptibility tests. Other mycobacterial isolates are referred to Level III for identification. Level III performs all procedures of Level I, identifies all mycobacteria, and performs susceptibility tests.
The Association of State and Territorial Public Health Laboratory Directors (ASTPHLD, now known as the Association of Laboratory Directors or ALD) in their second conference on the laboratory aspects of tuberculosis in 1995 proposed that in the new age of technology only two Levels of service should be recognized (Warren and Cordts, 1996). In this classification, Level I laboratories, housed in a Biosafety Level II facility, are those that arrange for transport of specimens for arrival in their laboratory within 24 hours of collection. The Level I laboratory may prepare acid-fast smears from concentrated specimens and examine them by fluorescent microscopy so that a report can be issued within 24 hours of receipt. If the laboratory chooses to stain specimens, it shall process at least 20 smears per week to maintain proficiency, with collection and transport of specimens for arrival in the laboratory within 24 hours of collection. Level I laboratories arrange for rapid transport of specimens to Level II laboratories for processing.
Level II laboratories processing at least 25 specimens per week will perform all the following tasks and must have a facility for Biosafety level III practices. Laboratories at this level must be able to perform the following:
A. Prepare acid-fast bacilli smears from concentrated specimens and examine them by fluorescent microscopy within 24 hours of receipt.
B. Use a combination of liquid and solid media for primary isolation of mycobacteria.
C. Identify all Mycobacterium species using rapid methods (such as DNA probes or high-performance liquid chromatography) to ensure identification within 10–14 days.
D. Perform susceptibility testing of M. tuberculosis to ensure reporting results within 15–30 days of specimen collection, by either implementation of the direct susceptibility test on agar plates (especially for new patients) or implementation of an indirect test in a liquid medium.
The levels of service concept is acceptable today. The ASTPHLD levels should be considered as guidelines until a more comprehensive program is in place. The processing of specimens in two different laboratories should be discouraged because of all of the time that is lost in shipment. A Level I laboratory may serve as an information source to direct specimens to the Level II laboratory.
Excellence in the laboratory comes from a permanent, well-trained staff who have pride in their performance. In my view, the Mycobacteriology Laboratory should not be included in the rotation exercises found in many diagnostic laboratories because of the expertise that is demanded in test performance, interpretation of results, and the requirement for performance under the safety codes for the protection of self and others.Clinical Laboratories Improvement Amendment of 1988
Tests for staining, isolation, identification, drug susceptibility testing, and special tests for epidemiologic studies of tuberculosis are all considered complex tests in the Clinical Laboratory Improvement Amendments (CLIA), Code of Federal Regulations, Chapter IV. Subpart M defines the qualifications of personnel for high complexity testing in paragraph 493.1489, that is, personnel should have a medical degree or “Have earned a doctoral, master's or bachelor's degree in a chemical, physical, biological or clinical laboratory science, or medical technology from an accredited institution;” or “Have earned an associate degree in a laboratory science, or medical laboratory technology from an accredited institution or” meet one or more other requirements that indicate acquisition of the individual skills required for performing specific laboratory tests, for proper instrument use, for performing preventive maintenance, and for calibration procedures.” Personnel must also “have the skills required to implement quality control policies and procedures of the laboratory;” have “an awareness of the factors that influence test results;” and have “the skills required to assess and verify the validity of patient test results through the evaluation of quality control values before reporting patient test results.” Qualifications of technologists as high complexity testing personnel under paragraph 1489(b)(3) are given in paragraph 493.1491. In summary, personnel performing the high complexity tests of the tuberculosis laboratory must have a level of training, preferably at least a bachelor's degree from an accredited institution, that indicates the comprehension and responsibility adequate for testing and reporting results of a critical nature.
In addition of the qualifications prescribed by CLIA, many states have criteria that personnel in the tuberculosis laboratory must meet. The local requirements mean that personnel in a commercial testing laboratory may not be able to transfer to a state laboratory even though they meet the CLIA requirements. State laboratories are obligated to work within the laws of their jurisdictions until those laws are changed.
Types of Training
Training should be available as an ongoing function of the laboratory through a collection of texts, reprints, and self-help documents, and through specific guidance by senior personnel. Outside training is available through the American Society for Microbiology (ASM), the Centers for Disease Control and Prevention (CDC), ALD, and the National Laboratory Training Network (NLTN) which is cosponsored by the ADL and CDC. The NLTN gives training courses applicable to the tuberculosis laboratory and in other clinical, environmental, and public health topics. Participation in these courses awards continuing educational credits to participants who successfully complete training.
All personnel should be encouraged to attend at least one training meeting per year, and be given as much support as possible to do so, such as compensated leave, support for travel and per diem, training fees, etc. An ongoing training program will pay dividends in maintaining an aware and capable staff. The excitement we experience in work activities is dependent on our ability to understand and interpret them. Need for Training
CDC recently surveyed 43 state public health laboratories, 8 nonstate public health laboratories, 87 hospital laboratories, and 7 commercial laboratories to get a first line response on what is perceived to be the training programs that would best serve their needs (Bird, 1999). Although the results are in the preliminary stages of analysis, some needs stand out. Most respondents believed that additional training was needed in safety, specifically in accident response, risk assessment, and effective use of the biological safety cabinet. Many respondents also believed management training was needed for quality assurance, CLIA standards, cost effectiveness and electronic reporting. Specific laboratory problems that respondents wanted addressed were the detection and prevention of false-positive reactions for M. tuberculosis, training in the use and interpretation of direct identification kits for specimens that are AFB positive, and training and possible research with the advanced technologies such as PCR and RFLP. Additional training is currently being planned by the Division of Laboratory Systems, Public Health Practice Program Office, CDC, to meet these needs.
Every laboratory can benefit by having a program of applied research to accompany the routine work load. Many research studies can concen
trate on method development in conjunction with trials of new products, resolving cases of cross-contamination, or investigation of local clusters of disease in collaboration with epidemiologists. Studies conducted by public health laboratories may concentrate on the background and follow-up of cases of disease due to Mycobacterium species other than M. tuberculosis, particularly using the Public Health Laboratory Information System (PHLIS) that is available to them from CDC. This system allows study of all mycobacterial isolates and drug susceptibility patterns by a number of parameters. Studies of the occurrence of these acid-fast bacteria and the diseases they cause are within the scope of the tuberculosis laboratory since many of the methods will be the same and, in general, information found with one species is helpful in understanding another. Studies to discover an environmental source for these infections would be interesting. A review of data reported by state laboratories to PHLIS can be found in an electronic publication (W. R. Butler and J. T. Crawford. Nontuberculous Mycobacteria Reported to the Public Health Laboratory Information System by State Public Health Laboratories, United States, 1993– 1996, which can be accessed on http://www.cdc.gov/ncidod/dastlr/mycobacteriology.htm). Additional data on the isolation of M. tuberculosis and results of drug susceptibility tests are not included in this report; however, each state has that information available to them, also.
Research projects should fit the interest of the investigator, but they should be of such a nature that an audience of peers will find the study informative. Development of an interest in one or more of the technologies available to the investigator may lead to discovery of a unique use that was not previously considered.
ROLE OF COMMERCIAL LABORATORIES AND THEIR RELATIONSHIP TO THE PUBLIC HEALTH LABORATORY
Public health laboratories are not in conflict with commercial laboratories; indeed, commercial laboratories do many things well that the public health laboratory needs to emulate; for example, the rapid pick-up of specimens (courier) and the rapid processing that is usually available from commercial laboratories. Many public health laboratories still depend on specimens that are held until a predetermined shipment size is reached, and are then placed in the mail for delivery by the U.S. Postal Service, resulting in a delay of several days. Public health laboratories submit reports to the client and to the tuberculosis program at stages of processing. In contrast, the commercial laboratory always sends a report to the client who submitted the specimen, but they may not always include notification to the health department (Royce, 1999).
McDade and Hughes (1998) point out that the failure of health care
providers to submit representative isolates of etiologic agents to public health authorities for confirmation and subtyping will preclude the prompt identification of disease outbreaks. Public health laboratory programs are even more compromised in their attempts to discover and control infectious diseases when out-of-state laboratories fail to comply with the reporting requirements. When the tuberculosis program is not notified of laboratory results, control measures are not activated, and contact cases may be lost to early treatment.
Skeels (1999) indicates two ways that private sector clinical laboratories contribute to public health efforts: (i) by diagnosing and reporting communicable diseases and (ii) by working with public health laboratories for the referral, confirmation, and typing of microbial isolates. He also reports that many clinical microbiology directors are no longer allowed by managed care organizations (MCOs) to perform “extra” testing on patient specimens for public health purposes that go beyond the needs of individual patient care. Decisions to do such testing are now made by administrators and insurers who require “critical path management” of patients, often with the view that public health testing is an unnecessary added cost. Skeels (1999) also discusses the findings of a study by the Lewin Group, which was commissioned by the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, as part of ongoing research regarding public health infrastructure. The Group reported that managed care and other health system changes are having negative effects on public health laboratories, making it more difficult for them to fulfill their public health mission.
Costs for diagnostic services vary widely between commercial and public health laboratories in that many of the latter do not charge for specimens submitted for study. However, some public health laboratories charge more than commercial laboratories, which must realize a profit. In answer to a survey question submitted to laboratory directors in preparation for this report, only eight of 45 state public health laboratory directors, but two of three territorial laboratory directors, indicated that they charged for tuberculosis laboratory diagnosis.
Dr. Leonid Heifets at the National Jewish Hospital in Denver is the head of an active charge-for-service, tuberculosis referral laboratory. The price for an AFB smear and inoculation of four units of medium (including BACTEC) for isolation of mycobacteria is $58.00. Prices for identification vary: $150.00 for an amplification test (MTD) with a raw specimen; $49.50 for the Gen-Probe test of a grown culture (including confirmation as M. tuberculosis and not another member of the M. tuberculosis Complex) with four biochemical tests; $30.75 for a conventional identification if it is ordered along with a drug-susceptibility test. Costs for susceptibility tests are not available at this time.
Because new arrangements are needed to identify specific functions of public- and private-sector laboratories, facilitate collaboration in areas of shared responsibility, and prevent unnecessary duplication of services, McDade and Hausler (1998) proposed the formation of local public health institutes to improve the strategic planning for public health. They further assigned certain functions to public and private sectors and identified a group of functions that are shared. There are advantages that will accrue from performance under this system that should be considered, chiefly that the role of each type of laboratory is clarified in an overall planned program that can be integrated on a local or regional level.
THE POTENTIAL ROLE OF REGIONAL LABORATORIES TO AUGMENT PROCEDURES OF STATE AND LOCAL LABORATORIES
As the number of cases of tuberculosis declines, work load in the laboratory will decrease even though there are an average of 20,621 isolates of nontuberculous mycobacteria identified per year by reporting state laboratories (W. R. Butler and J. T. Crawford, Nontuberculous Mycobacteria Reported to the Public Health Laboratory Information System by State Public Health Laboratories, United States, 1993–1996, which can be accessed on http://www/cdc.gov/ncidod/ dastlr/mycobacteriology.htm). If the trend continues, the number of isolations by some laboratories may be too rare to maintain their efficiency.
The concept of regionalized laboratories for fingerprinting isolates of M. tuberculosis is a potential model that can be used to increase the efficiency of public health tuberculosis laboratories. As noted above, the ATS and the ASTPHLD (ALD) endorse using laboratory activity extrapolated from the numbers of specimens processed as a guide for maintaining proficiency, over and above participation in a PT program. Use of genetic probes will not guarantee that other sophisticated procedures, such as drug susceptibility testing, are carried out for the characterization of bacilli identified as M. tuberculosis Complex. In areas where few specimens are processed for isolation and identification of M. tuberculosis, efficient use of funds and accuracy of reports mandate a regionalization of public health laboratories. Commercial laboratories have already established centralized laboratories in regions for various activities. The same approach will have to be used for public health laboratories as the number of specimens processed dwindles, because of the reduced number of cases.
Because of the sensitive nature of regionalization and the political ramifications in each locality, the problem should be considered in a special committee of the ALD. The people most involved in the problem can make the fairest appraisal of a solution.
FUNDING, GRANTS, CHARGES FOR SERVICE
State public health laboratories have received supplemental federal funding through CDC for several years to obtain equipment, to train laboratory personnel, to upgrade the level of effort needed to keep equipment and personnel current with new technology, and to develop the sophistication of the modern mycobacteriology laboratory in all of its aspects. The funding is on a five-year cycle and applications will be reviewed in 1999 for the next five-year cycle. Funding for the laboratories has been about eight million dollars per year. An additional eight hundred thousand dollars has supported the RFLP typing program.
Hospital and clinical laboratories develop relationships with commercial laboratories that may charge less for processing specimens than their state public health laboratory. Although the commercial laboratory is able to study specimens and report results in an efficient way in most instances, they should abide by the regulations of the state where specimens originate to ensure that reporting requirements for reportable diseases are all met (Skeels, 1999).
ROLE OF THE MYCOBACTERIOLOGY LABORATORY AT THE CENTERS FOR DISEASE CONTROL
The TB/Mycobacteriology Branch, Division of AIDS, STD, and TB Laboratory Research, NCID, CDC is central to the laboratory activities throughout the country. It is the ultimate reference laboratory to solve problems with difficult cultures and tests; to provide the background information through presentations, publications, and consultations on new and upcoming techniques applicable to the field; and to interact closely with the Division of Tuberculosis Elimination to support studies, give guidance regarding laboratory capabilities, and provide general back-up for their programs.
The activities of the CDC laboratory could be increased if funding were more available. Laboratory personnel need to attend meetings and conferences, many overseas. Practical research programs in the Branch need to be expanded to include studies in animals. This laboratory should be the central site to evaluate and measure the parameters of new vaccines that are being developed, but adequate animal quarters are not available for the work. The Branch should be in better contact with the state laboratories, and a good way to promote interaction would be to have the Branch, with a staff to monitor and supervise the expenditures, control funding for state laboratories.
The TB/Mycobacteriology Branch has active programs under way, and these are fully supported. The Branch has been providing training to
underdeveloped countries in collaboration with other segments of CDC; however, work in these areas has been at a rather low level since the programs are in other centers and divisions. The Branch should award training scholarships to offset per diem expenses of short term trainees from this country and abroad, and support travel and per diem for distinguished scientists. If the TB/Mycobacteriology Branch is to impact studies in developing countries, direct relationships must be established, and all support activities must be through the Branch Chief with his/her assignment of the most appropriate personnel for each project.
CURRENT STATUS OF PUBLIC HEALTH TUBERCULOSIS LABORATORIES
Tenover et al. (1993) issued a challenge to tuberculosis laboratories because available technologies were not being used to speed reports of results from tuberculosis laboratories. What is the status of laboratories in 1999?
Since 1993, tuberculosis laboratories, both in hospitals (Tokars et al., 1996) and state public health facilities (Bird et al., 1996; Denniston et al., 1997; Huebner et al., 1993), have worked to upgrade their facilities and procedures. The state laboratories have been very aggressive in a program of modernization.
In preparation of this report, questionnaires were sent to laboratory directors in all of the states, the District of Columbia, and the territories. Even though the turnaround time for submitting replies was short, 46 state laboratory (SL) directors and 4 directors of territorial laboratories (TL) responded (see Table D-2). All respondents accepted specimens for the laboratory diagnosis of tuberculosis, and all processed the specimens in-house except for two TLs that sent specimens to the SLs in Hawaii and California. Replies indicated that state and territorial laboratories have updated procedures to speed identification and drug susceptibility testing. As indicated by the responses, public health laboratories are using the modern procedures available to them, and reports are being made to state tuberculosis programs in a timely manner. Since this survey did not include any of hospital or commercial laboratories, a conclusion regarding their performance cannot be made.
In the past the ASTPHLD and various Centers of the CDC have cosponsored meetings to update the role of tuberculosis laboratories. These meetings should continue as the technology for methods of identification and drug susceptibility testing continues to develop. Tuberculosis laboratories that process initial specimens are now in a good position, but maintaining the laboratories as a first-line weapon for the elimination of tuberculosis in the United States is a major challenge that must be met.
Please refer to the page image for an unflawed representation of this content.
Advisory Council for the Elimination of Tuberculosis . 1999 . Tuberculosis elimination revisited: obstacles, opportunities, and a renewed commitment . MMWR 48(RR09) : 1–13 .
Alito, A. , N. Morcillo , S. Scipioni , A. Dolmann , M. I. Romano , A. Cataldi , and D. van Soolingen . 1999 . The IS6110 restriction fragment length polymorphism in particular multidrug-resistant Mycobacterium tuberculosis strains may evolve too fast for reliable use in outbreak investigation . J. Clin. Microbiol . 37 : 788–791 .
American Thoracic Society . 1983 . Levels of laboratory services for mycobacterial diseases: Official Statement of the American Thoracic Society . Am. Rev. Respir. Dis. 128 : 213 .
American Thoracic Society Workshop . 1997 . Rapid diagnostic tests for tuberculosis: what is the appropriate use? Am. J. Respir. Crit. Care Med. 155 : 1804–1814 .
Ausina, V. , F. Gamboa , E. Gazapo , J. M. Manterola , J. Lonca , L. Matas , J. R.Manzano , C. Rodrigo , P. J. Cardona , and E. Padilla . 1997 . Evaluation of the semiautomated Abbott LCx Mycobacterium tuberculosis assay for direct detection of Mycobacterium tuberculosis in respiratory specimens . J. Clin. Microbiol . 35 : 1996–2002 .
Becker, S. 1999 . Personal communication .
Benjamin, Jr., W. H. , K. B. Waites , A. Beverly , L. Gibbs , M. Waller , S. Nix , S. A. Moser , and M. Willert . 1998 . Comparison of the MB/BacT system with a revised antibiotic supplement kit to the BACTEC 460 system for detection of mycobacteria in clinical specimens . J. Clin. Microbiol . 36 : 3234–3238 .
Bergmann, J. S. , and G. L. Woods . 1998 . Clinical evaluation of the BD ProbeTec Strand Displacement Amplification Assay for rapid diagnosis of tuberculosis . J. Clin. Microbiol . 36 : 2766–2768 .
Bird, B. R. , M. M. Denniston , R. E. Huebner , and R. C. Good . 1996 . Changing practices in mycobacteriology: a follow-up survey of state and territorial public health laboratories . J. Clin. Microbiol . 34 : 554–559 .
Bird, B. R. 1999 . Personal communication .
Bradley, S. P. , S. L. Reed , and A. Catanzaro . 1996 . Clinical efficacy of the amplified Mycobacterium tuberculosis direct test for the diagnosis of pulmonary tuberculosis . Am. J. Respir. Crit. Care Med. 153 : 1606–1610 .
Butler, W. R. , I. Thiebert , and J. O. Kilburn . 1992 . Identification of Mycobacterium avium complex strains and some similar species by high-performance liquid chromatography . J. Clin. Microbiol . 30 : 2698–2704 .
Canetti, G. , S. Froman , J. Grosset , P. Hauduroy , M. Lagerova , H. T. Mahler , G. Meissner , D. A. Mitchison , and L. Sula . 1963 . Mycobacteria: laboratory methods for testing drug sensitivity and resistance . Bull. W. H. O. 29 : 565–578 .
Centers for Disease Control . 1989 . A strategic plan for the elimination of tuberculosis in the United States . MMWR 38 (Suppl. No. S-3) : 1–25 .
Centers for Disease Control . 1995 . Laboratory practices for diagnosis of tuberculosis— United States, 1994 . MMWR 44 : 587–590 .
Centers for Disease Control . 1999 . Progress Toward the elimination of tuberculosis—United States, 1998 . MMWR 48 : 732–736 .
Centers for Disease Control and Prevention and National Institutes of Health . 1999 . Biosafety in Microbiological and Biomedical Laboratories , J. Y. Richmond and R. W. McKinney , eds. 4th ed. U.S. Government Printing Office , Washington, D.C. , pp. 104–106 , 201–202 , 216–219 .
David, H. L. 1970 . Probability distribution of drug-resistant mutants in unselected populations of Mycobacterium tuberculosis . Appl. Microbiol . 20 : 810–814 .
Denniston, M. M. , B. R. Bird , and K. A. Kelley . 1997 . Contrast of survey results between state and a cohort of nonstate mycobacteriology laboratories: changes in laboratory practices . J. Clin. Microbiol . 35 : 422–426 .
Eisenach, K. D. , J. T. Crawford , and J. H. Bates . 1988 . Repetitive DNA sequences as probes for Mycobacterium tuberculosis . J. Clin. Microbiol . 26 : 2240–2245 .
Gladwin, M. T. , J. J. Plorde , and T. R. Martin . 1998 . Clinical application of the Mycobacterium tuberculosis direct test: case report, literature review, and proposed clinical algorithm . Chest 114 : 317–323 .
Good, R. C. , and T. M. Shinnick . 1998 . Mycobacterium , pp. 549–576 . In: A. Balows and B. I. Duerden (eds.), Topley & Wilson's Microbiology and Microbial Infections , 9th ed., vol. 2 , Systematic Bacteriology . Oxford University Press, Inc. , New York ,
Hanna, B. A. , A. Ebrahimzadeh , L. B. Elliott , M. A. Morgan , S. M. Novak , S. Rüsch-Gerdes , M. Acio , D. F. Dunbar , T. M. Holmes , C. H. Rexer , C. Savthyakumar , and A. M. Vannier . 1999a . Multicenter Evaluation of the BACTEC MGIT 960 System for recovery of mycobacteria . J. Clin. Microbiol . 37 : 748–752 .
Hanna, B. A. , A. Ebrahimzadeh , G. E. Pfyffer , F. Palicova , C. H. Rexer , T. Sanchez , C. Sathyakumar , and L. B. Heifets . 1999b . Evaluation of the BACTEC MGIT 960 antimycobacterial susceptibility test . Abstract, paper 864, 39th ICAAC, San Francisco, Ca., Sept 26–29, 1999 .
Huebner, R. E. , R. C. Good , and J. I. Tokars . 1993 . Current practices in mycobacteriology: results of a survey of state laboratories . J. Clin. Microbiol . 31 : 771–775 .
Inderlied, C. B. , and M. Salfinger . 1999 . Antimycobacterial agents and susceptibility tests , pp. 1601–1623 . In: P. R. Murray , E. J. Baron , M. A. Pfaller , F. C. Tenover , and R. H. Yolken (eds.), Manual of Clinical Microbiology , 7th ed. ASM Press , Washington, D.C.
Kamerbeek, J. , L. Shouls , A. Kolk , M. van Agterveld , D. van Soolingen , S. Kuijper , A. Bunschoten , H. Molhuizen , R. Shaw , M. Goyal and J. van Embden . 1997 . Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology . J. Clin. Microbiol . 35 : 907–914 .
Kent, P.T. , and G. P. Kubica . 1985 . Public Health Mycobacteriology: A Guide for the Level III Laboratory . Centers for Disease Control, U.S. Department of Health and Human Services , Atlanta, Georgia .
Kox, L. F. , H. M. Jansen , S. Kuijper , and A. H. Kolk . 1997 . Multiplex PCR assay for immediate identification of the infecting species in patients with mycobacterial disease . J. Clin. Microbiol . 35 : 1492–1498 .
Kremer, K. , D. van Soolingen , R. Frothingham , W. H. Haas , P. W. M. Hermans , C. Martin , P. Palittapongarnpim , B. B. Plikaytis , L. W. Riley , M. A.Yakrus , J. M. Musser , and J. D. A. van Embden . 1999 . Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis complex strains: interlaboratory study of discriminatory power and reproducibility . J. Clin. Microbiol . 37 : 2607–2618 .
Lein, A. D. , C. F. von Reyn , P. Ravn , C. R. Horsburgh, Jr. , L. N. Alexander , and P. Andersen . 1999 . Cellular immune responses to ESAT-6 discriminate between patients with pulmonary disease due to Mycobacterium avium complex and those with pulmonary disease due to Mycobacterium tuberculosis . Clin. Diagn. Lab. Immunol . 6 : 606–609 .
Martin-Casabona, N. , D. X. Mimó , T. González , J. Rossello , and L. Arcalis . 1997 . Rapid method for testing susceptibility of Mycobacterium tuberculosis by using DNA probes . J. Clin. Microbiol . 35 : 2521–2525 .
McDade, J. E. , and W. J. Hausler, Jr. 1998 . Modernization of public health laboratories in a privatization atmosphere . J. Clin. Microbiol . 36 : 609–613 .
McDade, J. E. , and J. M. Hughes . 1998 . The U.S. needs a national laboratory system . U.S. Medicine 34 : 9 .
Metchock, B. G. , F. S. Nolte , and R. J. Wallace, Jr. 1999 . Mycobacterium , pp. 399–437 . In: P. R. Murray , E. J. Baron , M. A. Pfaller , F. C. Tenover , and R. H. Yolken (eds.), Manual of Clinical Microbiology , 7th ed. ASM Press ,
Moore, D. F. , and J. I. Curry . 1998 . Detection and identification of Mycobacterium tuberculosis directly from sputum sediments by ligase chain reaction . J. Clin. Microbiol . 36 : 1028–1031 .
Niemann, S. , E. Richter , and S. Rüsch-Gerdes . 1999 . Stability of Mycobacterium tuberculosis IS6110 fragment length polymorphism patterns and spoligotypes determined by analyzing serial isolates from patients with drug-resistant tuberculosis . J. Clin. Microbiol . 37 : 409–412 .
Noordhoek, G. T. , J. D. van Embden , and A. H. Kolk . 1996 . Reliability of nucleic acid amplification for the detection of Mycobacterium tuberculosis: an international collaborative quality control study among 30 laboratories . J. Clin. Microbiol . 34 : 2522–2525 .
Piersimoni, C. , A. Callegaro , C. Scarparo , V. Penati , D. Nista , S. Bornigia , C. Lacchini , M. Scagnelli , G. Santini , and G. De Sio . 1998 . Comparative evaluation of the new GenProbe Mycobacterium tuberculosis Amplified Direct Test and the semiautomated Abbott LCx Mycobacterium tuberculosis Assay for direct detection of Mycobacterium tuberculosis Complex in respiratory and extrapulmonary specimens . J. Clin. Microbiol . 36 : 3601–3604 .
Reischl, U. , N. Lehn , H. Wolf , and L. Naumann . 1998 . Clinical evaluation of the automated COBAS AMPLICOR MTB Assay for testing respiratory and nonrespiratory specimens . J. Clin. Microbiol . 36 : 2853–2860 .
Ridderhof, J. , L. Williams , S. Legois , M. Bussen , B. Metchock , L. Kubista , J. Handsfield , R. Fehd . 1998 . Assessment of laboratory performance with nucleic acid amplification (NAA) tests for Mycobacterium tuberculosis (M. tb) . Abstract, Annual Meeting of the American Society for Microbiology .
Riska, P. F. , Y. Su , S. Bardarov , L. Freundlich , G. Sarkis , G. Hatfull , C. Carrière , V. Kumar , J. Chan , and W. R. Jacobs, Jr. 1999 . Rapid film-based determination of antibiotic susceptibilities of Mycobacterium tuberculosis strains by using a luciferase reporter phage and the Bronx box . J. Clin. Microbiol . 37 : 1144–1149 .
Roberts, G. D. , N. L. Goodman , L. Heifets , H. W. Larsh , T. H. Lindner , J. K. McClatchy , M. R. McGinnis , S. H. Siddiqi , and P. Wright . 1983 . Evaluation of the BACTEC radiometric method for recovery of mycobacteria and drug susceptibility testing of Mycobacterium tuberculosis from acid-fast smear-positive specimens . J. Clin. Microbiol . 18 : 689–696 .
Rohner, P. , E. I. M. Jalm , B. Ninet , C. Ionati , R. Weber , R. Auckenthaler , and G. E. Pfyffer . 1998 . Rapid diagnosis of pulmonary tuberculosis with the LCx Mycobacterium tuberculosis Assay and comparison with conventional diagnostic techniques . J. Clin. Microbiol . 36 : 3046–3047 .
Royce, S. E. 1999 . Personal communication .
Rüsch-Gerdes, S. , C. Domehl , G. Nardi , M. R. Gismondo , H.-M. Welscher , and G. E. Pfyffer . 1999 . Multicenter evaluation of the Mycobacteria Growth Indicator Tube for testing susceptibility of Mycobacterium tuberculosis to first-line drugs . J. Clin. Microbiol . 37 : 45–48 .
Salfinger, M. , Y. M. Hale , and J. R. Driscoll . 1998 . Diagnostic tools in tuberculosis: present and future . Respiration 65 : 163–170 .
Sewell, D. L. , and J. D. MacLowry . 1999 . Laboratory management , pp. 4–22 . In: P. R. Murray , E. J. Baron , M. A. Pfaller , F. C. Tenover , and R. H. Yolken (eds.), Manual of Clinical Microbiology , 7th ed. ASM Press , Washington, D.C.
Shinnick, T. M. , and V. Jonas . 1994 . Molecular approaches to the diagnosis of tuberculosis , pp. 517–530 . In: B. R. Bloom (ed.), Tuberculosis: Pathogenesis, Protection, and Control . ASM Press ,
Skeels, M. 1999 . Public health labs in a changing health care landscape . ASM News 65 : 479–483 .
Tenover, F. C. , J. C. Crawford , R. E. Huebner , L. J. Geiter , C. R. Horsburgh, Jr. , and R. C. Good . 1993 . The resurgence of tuberculosis: is your laboratory ready? J. Clin. Microbiol . 31 : 767–770 .
Tokars, J. I. , J. R. Rudnick , K. Kroc , L. Manangan , G. Pugliese , R. E. Huebner , J. Chan , and W. R. Jarvis . 1996 . U.S. hospital mycobacteriology laboratories: status and comparison with state public health department laboratories . J. Clin. Microbiol . 34 : 680–685 .
van Embden, J. D. A. , M. D. Cave , J. T. Crawford , J. W. Dale , K. D. Eisenach , B. Gicquel , P. Hermans , C. Martin , R. McAdam , T. M. Shinnick , and P. M. Small . 1993 . Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology . J. Clin. Microbiol . 31 : 406–409 .
Wang, S. X. , and L. Tay . 1999 . Evaluation of three nucleic acid amplification methods for direct detection of Mycobacterium tuberculosis complex in respiratory specimens . J. Clin. Microbiol . 37 : 1932–1934 .
Warren, N. G. , and J. R. Cordts . 1996 . Activities and recommendations by the Association of State and Territorial Public Health Laboratory Directors . In: L. B. Heifits (ed.), Clinics in Laboratory Medicine: Clinical Mycobacteriology 16 : 731–743 .
Wobeser, W. L. , M. Krajden , J. Conly , H. Simpson , B. Yim , M. D'Costa , M. Fuksa , C. Hiancheong , M. Patterson , A. Phillips , R. Bannatyne , A. Haddad , J. L. Brunton , and . 1996 . Evaluation of Roche Amplicor PCR assay for Mycobacterium tuberculosis . J. Clin. Microbiol . 34 : 134–139 .
Woods, G. I. , and F. G. Witebsky . 1993 . Current status of mycobacterial testing in clinical laboratories: results of a questionnaire completed by participants in the College of American Pathologists mycobacteriology E survey . Arch. Pathol. Lab. Med. 117 : 876–884 .
Yang et al . 1998 . Diversity of DNA fingerprints of Mycobacterium tuberculosis isolates in the United States . J. Clin. Microbiol . 36(4) : 1003–1007 .
Zhou, A. T. , W. L. Ma , P. Y. Zhang , and R. A. Cole . 1996 . Detection of pulmonary and extrapulmonary tuberculosis patients with the 38-kilodalton antigen from Mycobacterium tuberculosis in a rapid membrane-based assay . Clin. Diagnostic Lab. Immunol . 3 : 337–341 .