Tuberculosis Reference Laboratory

    As a part of the Public Health Laboratories in the Department of Health, HKSAR, the Tuberculosis (TB) Reference Laboratory (the Laboratory) supports the TB & Chest Service in the Department in providing a range of laboratory services for the clinical management of tuberculosis (TB) and other mycobacterial diseases, as well as organizing and participating in TB surveillance and epidemiological studies. The Laboratory also takes part in the TB Notification System by submitting demographic information of patients with positive Mycobacterium tuberculosis (MTB) culture isolation to the TB & Chest Service. The Laboratory is now receiving patients' specimens, mainly sputum, from all Government Chest Clinics, some hospitals under the Hospital Authority, private laboratories and practitioners for routine microscopy examination and mycobacterial culture. Moreover, positive mycobacterial cultures from most hospitals under the Hospital Authority and also from private laboratories are also sent to the Laboratory for bacterial identification and drug susceptibility tests.

    In order to maintain high accuracy of the laboratory testing, the Laboratory organizes as well as participates in various quality assurance programmes. In collaboration with Hong Kong Medical Technology Association, it has organized an "Acid Fast Bacilli (AFB) smear Quality Assurance Programme" for most clinical bacteriology laboratories in Hong Kong. Moreover, the Laboratory also joins the External Quality Assurance Programme organized by the United Kingdom NEQAS (Microbiology Quality Assessment) and World Health Organization Supranational Reference Laboratory for mycobacteria smear and culture, and drug susceptibilities testing respectively.

    Following is a list of services presently provided by the Laboratory. In general, these can be divided into eight major categories:

1. Direct microscopy for the detection of AFB
2. Direct detection of MTB from sputum by nucleic acid amplification method
3. Primary isolation of mycobacteria from clinical specimen
4. Positive culture identification
5. Drug susceptibility tests
6. MTB strain typing for outbreak investigation
7. Urine isoniazid metabolite tests
8. Collaborative studies and training

1. Direct microscopy for the detection of acid-fast bacilli

    Direct microscopy is used for the rapid diagnosis of TB and other mycobacterial diseases, as a relatively long period of time is required for mycobacteria to be detected by culture methods. Patients with positive smears are considered to have higher bacterial load, and thus more likely to spread TB. Results from direct microscopy can be available within 24 hours. Clinically, it is important to detect the presence of mycobacteria as rapidly as possible for implementation of appropriate patient care and public health measure. Moreover, direct microscopy on successive specimen can also be used to monitor the success of chemotherapy for smear positive patients.

     In the Laboratory, fluorochrome-staining method, which is found to be more sensitive, is used for the detection of acid-fast bacilli. Using this staining method, a minimum concentration of 5,000 to 10,000 bacilli per mL of specimen is required for the detection of positive smear. When compared with culture, the overall sensitivity of the direct smear has been reported to range from 22 to 43%. Moreover, despite the high specificity of direct smear for detection of mycobacteria, this test cannot be used to differentiate between various mycobacterial species.

2. Direct detection of Mycobacterium tuberculosis from sputum by nucleic acid amplification method.

     Due to the slow growth of Mycobacterium tuberculosis (MTB), a usual period of 3 to 8 weeks is required to obtain a definite TB diagnosis by culture examination. In this connection, the Laboratory provides another option for TB diagnosis by using a Food and Drug Agency, U. S. A. (FDA) approved nucleic acid amplification (NAA) assay to detect and identify MTB directly from smear-positive clinical respiratory specimen. This test was found to have a sensitivity of >95% and a specificity of >98%, and the result can be obtained within 2 days. This NAA assay is particularly useful when rapid differentiation between MTB infection and non-tuberculosis mycobacteria infection is required, such as for AIDS patients.

     Despite the rapidity of this test, sensitivity for this NAA assay against smear negative specimen is about 50% and only limited data are available for guiding the interpretation of the test for non-respiratory specimen. In this connection, discussion with the Laboratory's medical microbiologists can be arranged for individual cases on the usefulness and limitation of using this relatively expensive NAA assay for primary TB diagnosis.

3. Primary isolation of mycobacteria from clinical specimen.

    Isolation of definite pathogen e.g. MTB or repeated isolation of opportunistic pathogens from clinical specimen may confirm a definitive diagnosis of TB or other mycobacterial infections. Due to the slow growth of most mycobacteria including MTB, 3 to 8 weeks are required for conventional cultures. When specially requested for individual cases, rapid culture using broth medium can be arranged, with which about 90% of positive culture can be obtained within 3 weeks.

    Since a long incubation period is required for performing mycobacteria culture, it is inevitable to have contamination due to growth of other bacteria and fungi. In general, a 5% level of contamination should be expected.

4. Positive culture identification.

    All positive cultures have to be identified for the confirmation of pathogen. In the Laboratory, mycobacteria will always be identified to the species level if possible. Batches of mycobacteria identification methods are available in the Laboratory. These identification methods include conventional biochemical tests, analysis of mycobacterial fatty acid or mycolic acid by chromatography, and genetic investigation through the use of nucleic acid probes, restriction fragment analysis of DNA amplification products, and DNA sequencing.

    Since the amount of mycobacteria isolates is always limited for performing biochemical identification tests, additional time is required for performing sub- culture in order to obtain sufficient growth in order to do the identification tests. However, for the identification of positive cultures as MTB and Mycobacterium avium-intracellulare complex, rapid identification by using the DNA probes can be arranged when specially requested, and in most cases, results can be available within 2 working days.

5. Drug susceptibility tests (DST)

    With the emergence of drug resistant MTB isolates, DST for MTB become a very important tool for revealing the cause of treatment failure and providing a guide on the choice of anti-TB drugs. The Laboratory routinely performs DST against all clinical MTB isolates for the 4 first line anti-TB drugs (streptomycin, isoniazid, rifampicin and ethambutol). For MTB isolated from re-treatment, relapse or failure cases, drug susceptibilities on ethionamide, amikacin and ofloxacin will also be included. Moreover, tests on kanamycin, capreomycin, cycloserine and pyrazinamide will also be added if required.

    The Laboratory is using a well calibrated, standardized absolute concentration method to perform the DST with results available usually in 4 weeks. Moreover, DST will also be conducted in broth medium for those MTB isolates obtained by broth culture. In these cases, results can be ready in 2 to 3 weeks.

     Apart from MTB, the Laboratory also conducts DST for M. avium-intracellulare complex (MAC) by Bactec macrobroth dilution method. For performing DST on other anti-mycobacterial agents or against other mycobacteria, our medical officers or microbiologist can be consulted for further discussion.

6. MTB strain typing for outbreak investigation.

    Epidemiological studies of tuberculosis is essential in the prevention and control of TB. DNA fingerprinting (restriction fragment length polymorphism (RFLP) using the insertion sequence IS6110 as standard probe) has proven to be a powerful epidemiological tool for this purpose. In the Laboratory, DNA fingerprinting can be arranged for suspected TB outbreaks.

7. Urine isoniazid metabolite test

    The method facilitates the monitoring of isoniazid drug uptake by TB patients and also the use of isoniazid as a marker for drug compliance by assessing the regularity with which other drugs prescribed for self- administration are actually ingested. Test results usually can be available within a week.

8. Collaborative studies and training

    In collaboration with the TB & Chest Service, the Laboratory is actively involved in a number of anti-TB drug trials, as well as epidemiological studies. With approval from the Department of Health Headquarters, we also provide short term training for visiting laboratory officers and technical staff from this region.

From the Tuberculosis Reference Laboratory,
Department of Health, Hong Kong SAR.

Dated 17th December, 2001.

Biological safety cabinet

Inoculation

Specimen decontamination for TB culture