Send letter to journal:
Re: Re: The diagnosis of tuberculosis is difficult to verify

We thank Dr. T. Ho for his comment on our article reporting a young patient with tuberculous osteomyelitis [1]. We wrote the article from the perspective of emergency medicine. Although polymerase chain reaction (PCR) is a good adjunct to microbiological culture for diagnosing mycobacterium tuberculosis, it is not available to the majority of emergency physicians in Hong Kong. Nonetheless, we should discuss it briefly so that our article is more informative to readers.

Without argument, PCR provides an opportunity for early diagnosis and treatment. However, we should also note the limitation of the PCR especially when the PCR result is negative.

In 1998 Shah et al reported the accuracy of the AMPLICOR PCR test in diagnosing mycobacterium tuberculosis in tissue and body fluid specimens [2]. In this study, culture proof was adopted as the gold standard for diagnosing tuberculosis. Although 1032 patients were included in this study, only 34 specimens were positive for tuberculosis. Therefore, the sample size was too small and the 95% confidence interval of the sensitivity was too wide to suggest that PCR would not miss the diagnosis of mycobacterium tuberculosis. In this study, the PCR had a sensitivity of 76.4%, a specificity of 99.8% when results were compared with the gold standard. With the high specificity, PCR is a good "rule-in" test. However, PCR should not be used as a "rule-out" test because of the high false negative rate.

In 2000 Lim et al reported the accuracy of the AMPLICOR PCR test in diagnosing pulmonary tuberculosis in smear-negative respiratory tract specimens. Once again, the PCR test had a low sensitivity of 44% and a high specificity of 99% [3].

With evidence from both studies, a positive PCR test result facilitates early diagnosis, but a negative PCR test result cannot exclude mycobacterium tuberculosis. At the moment, microbiological culture remains the gold standard for diagnosing tuberculosis and a high index of suspicion for tuberculosis is the key to diagnosis.

References:

(1) Yuen MC, Tung WK. An uncommon cause of foot ulcer: tuberculous osteomyelitis. Emerg Med J 2001;18: 140-141
(2) Shah S, Miller A, Mastellone A, et al. Rapid diagnosis of tuberculosis in various biopsy and body fluid specimens by the AMPLICOR mycobacterium tuberculosis polymerase chain reaction test. Chest 1998;113: 1190-1194
(3) Lim TK, Gough A, Chin NK, et al. Relationship between estimated pretest probability and accuracy of automated mycobacterium tuberculosis assay in smear-negative pulmonary tuberculosis. Chest 2000;118: 641-647

Send letter to journal:
Re: The diagnosis of tuberculosis is difficult to verify

Yuen and Tung describe a case of tuberculous osteomyelitis of the foot [1] and the potential difficulties in making the diagnosis. The authors were fortunate enough to have typical histological biopsy findings which subsequently cultured Mycobacterium tuberculosis (TB), providing diagnostic confirmation and estimations of sensitivities. However, in many instances, the diagnosis of tuberculosis is difficult to verify. For instance, acid fast bacilli may not be identified on biopsy or may be non-tuberculous in origin. Additionally, subsequent culture confirmation can take several weeks or may fail completely, due to TB's fastidious nature.

Although the reliance on clinical suspicion is the basis for the diagnosis of many cases of TB, definitive confirmation is desirable in view of the long-term nature of therapy. It is also important to ensure that the organism is not resistant to the chemotherapeutic regime being employed, particularly with the increasing incidence of multi-drug resistant TB strains. A number of novel diagnostic techniques have been developed to facilitate this. The use of the polymerase chain reaction (PCR) to amplify specific TB DNA sequences allows a rapid confirmation of the diagnosis and an estimation of drug sensitivity [2]. These techniques have been successfully used on both clinical specimens and culture material [3]. Thus, acid fast bacilli can rapidly be identified as Mycobacterium tuberculosis and an estimation of rifampicin sensitivity can be obtained in a matter of days, free from the contraints of waiting up to several weeks for the standard culture to grow. These techniques should therefore be considered, particularly if the clinical findings are subtle or atypical.

References

(1) Yuen MC, Tung WK. An uncommon cause of foot ulcer: tuberculous osteomyelitis. Emerg Med J 2001; 18: 140-141.
(2) Telenti A, Imboden P, Marchesi F, et at. Detection of rifampicin-resistance mutations in Mycobacterium tuberculosis. Lancet 1993;341:647-650.
(3) Shah S, Miller A, Mastellone A, et al. Rapid diagnosis of tuberculosis in various biopsy and body fluid specimens by the AMPLICOR Mycobacterium tuberculosis polymerase chain reaction test. Chest 1998;113: 1190-1194.