Article Text


Clinical and laboratory observations of tuberculosis at a Mumbai (India) clinic
  1. D Gothi,
  2. J M Joshi
  1. Department of Respiratory Medicine, T N Medical College, B Y L Nair Hospital, Mumbai, India
  1. Correspondence to:
 Dr J M Joshi
 Professor and Head, Department of Respiratory Medicine, B Y L Nair Hospital, Mumbai 400008, India;


Objectives: To study the positivity of sputum acid fast bacilli (AFB) smears in patients with pulmonary tuberculosis using 24 hour sputum collection. To detect HIV seropositivity in patients suffering from tuberculosis, and to analyse the pattern of tuberculosis disease in this subgroup. To determine the outcome of patients treated with directly observed therapy.

Setting: The tuberculosis referral unit of a tertiary care hospital.

Design: A total of 893 consecutive patients with tuberculosis, diagnosed between 1 November 2000 and 30 September 2002, were included in the study. An HIV test was performed in all patients, with adequate counselling and informed consent. Treatment was prescribed as per World Health Organisation treatment categories.

Results: Out of 893 patients with tuberculosis, 695 had pulmonary tuberculosis and 198 had extrapulmonary tuberculosis. Out of the 695 pulmonary tuberculosis patients, 673 (96.8%) were sputum smear AFB positive. Overall, 71 patients (8.0%) were HIV positive. The pattern of tuberculosis was the same in HIV seropositive and seronegative patients. Treatment outcome could be analysed in 112 out of 150 patients: 78 patients (70%) were declared cured or completed treatment.

Conclusions: Sputum smear AFB could be a very sensitive test when a large quantity of sputum is used. The presence of HIV coinfection does not alter the clinical presentation. Only 70% of patients treated were cured/completed treatment, in spite of a strict directly observed therapy.

  • tuberculosis
  • HIV/tuberculosis coinfection
  • directly observed treatment
  • AFB, acid fast bacilli
  • DOT, directly observed therapy
  • DOTS, directly observed therapy, short course
  • ELISA, enzyme linked immunosorbent assay
  • RNTCP, Revised National Tuberculosis Control Programme
  • WHO, World Health Organisation

Statistics from

Tuberculosis is one of the biggest enigmas facing mankind. The National Tuberculosis Programme of India, established in 1962, created an infrastructure for tuberculosis control throughout the country. A comprehensive review in 1992 by a committee of national and international experts observed that the programme had not achieved the desired results.1,2 One reason for failure of the previous programme in India was under-utilisation of microscopy and excessive reliance on radiology, which resulted in overdiagnosis of tuberculosis and wastage of scarce resources. The World Health Organisation (WHO) declared tuberculosis a global emergency in 1993 and recommended the Revised National Tuberculosis Control Programme (RNTCP) with directly observed therapy, short course (DOTS) as a solution for its control.3,4

Two important components of the DOTS strategy are good quality smear microscopy and directly observed therapy (DOT).5 A 5–10 ml volume of sputum is recommended for the diagnosis of pulmonary tuberculosis.6–8 We routinely perform sputum smear acid fast bacilli (AFB) examination on a 24 hour sample in our reference laboratory, in order to obtain a large quantity of sputum. In developing countries, with the spread of HIV infection, the impact of tuberculosis has been intensified. The HIV and tuberculosis coepidemic has led to the development of strategies to decrease the burden of HIV related tuberculosis, comprising interventions against both tuberculosis and HIV. Hence, all patients with tuberculosis should be counselled for HIV testing. As per the RNTCP and DOTS strategy in India,5 treatment is to be fully supervised in the intensive phase and partially supervised in the continuation phase. However, the basic principle of DOT is that rifampicin should not be administered unsupervised. Hence, we give fully supervised treatment throughout therapy as recommended by the WHO.

The purposes of this study were thus to estimate, first, the sputum AFB smear positivity in patients with pulmonary tuberculosis using 24 hour sputum collection, second, the prevalence of HIV seropositivity and the pattern of disease in patients with pulmonary and extrapulmonary tuberculosis and, third, the outcome of treatment with a fully supervised DOT.


A total of 893 patients diagnosed with tuberculosis after the implementation of RNTCP and DOT, between November 2000 and September 2002, were included in the study. Personal and demographic data, along with the type of tuberculosis and associated conditions, were recorded. Diagnosis of pulmonary tuberculosis was made by positive sputum AFB smear. The methods of smear preparation and examination were as prescribed by the WHO but, instead of spot-overnight-spot samples, a 24 hour sputum collection was used to obtain a large quantity of sputum, as recommended by our reference microbiology laboratory. In patients with a negative sputum smear, diagnosis was established by sputum culture for Mycobacterium tuberculosis, bronchial aspirate smear or culture for M tuberculosis, or transbronchial lung biopsy revealing caseating granuloma with or without a positive smear, and/or culture for M tuberculosis. Extra-pulmonary tuberculosis was diagnosed by fine needle aspiration cytology or biopsy showing caseating granuloma with or without a positive smear and/or culture for M tuberculosis. One patient with suspected miliary tuberculosis required a trial of specific empiric antituberculosis treatment, and diagnosis was subsequently confirmed by liver biopsy. A complete blood count, tests for blood urea nitrogen and blood glucose, a liver function test, and a rapid immunoassay test for HIV were performed in all patients. Client centred two step HIV prevention counselling was followed, and HIV testing was informed, voluntary, and consented. The information given to the patient included a description of the ways in which HIV is transmitted, the importance of obtaining test results, and the explicit procedures for doing so and the benefits and consequences. In the second session, when the results of the HIV test were available, the counsellor discussed the test results, asked the client to describe the risk reduction steps attempted, helped the patient to identify and commit to additional behavioural steps, and provided appropriate referrals. HIV seropositivity was diagnosed by a three step rapid immunoassay. In the first step a micro-enzyme linked immunosorbent assay (ELISA) (J Mitra & Co Ltd) was performed. If micro-ELISA was positive, spot by HIV tri-dot (Biotech Inc) and spot by Comb-aids-recombinant protein and synthetic peptides (Span Diagnostic Ltd) were performed; the patient was considered HIV positive only if all three tests were positive.

All the patients were prescribed treatment as per WHO treatment categories, as shown in table 1. Of the 893 diagnosed cases, 723 patients were referred to their nearest DOT centre, and 20 patients were not willing to participate in DOT therapy and were treated with a fixed drug combination. This left 150 patients residing in the vicinity of our hospital who were treated at our DOT centre and whose treatment was fully supervised throughout. DOT included the involvement of a treatment organiser or a health visitor, who supervised each dose and visited patients at home if they failed to follow up. For the patients with pulmonary tuberculosis and positive initial sputum smears, sputum examination was repeated at 2–3, 4–5, and 6–8 months of treatment depending on WHO treatment categories. Treatment outcome was recorded as cured, completed treatment, defaulted, failed, or died based on the definitions given by the WHO (table 2). Of the 150 patients treated at our centre, outcome could be analysed in 112; the remaining 38 patients had not completed treatment at the time of data analysis.

Table 1

WHO treatment categories

Table 2



Our 893 patients with tuberculosis had an age range of 12 to 72 years (mean 42 years); 536 (60%) were males and 357 (40%) were females. In total, 695 patients were suffering from pulmonary tuberculosis, and 198 patients were diagnosed with extrapulmonary tuberculosis. Out of the 695 pulmonary tuberculosis patients, 673 (96.8%) were sputum smear positive for AFB, and only 22 (3.2%) were sputum smear negative. In 10 of the smear negative cases diagnosis was made by positive culture for M tuberculosis, and in six cases bronchial aspirate or transbronchial lung biopsy was positive for AFB. In the remaining six patients antituberculosis treatment had already been started at the time of referral, and, as there was clinical improvement, treatment was continued.

The commonest form of extrapulmonary tuberculosis was tuberculous lymphadenopathy, followed by pleural effusion (table 3). Six cases of miliary tuberculosis were diagnosed on transbronchial lung biopsy, while one patient with miliary tuberculosis required a trial of specific empiric antituberculosis therapy; diagnosis was subsequently confirmed by liver biopsy.

Table 3

Type of extrapulmonary tuberculosis (HIV and non-HIV combined; n = 198)

We found that 67 patients (7.5%) had diabetes mellitus, and 71 patients (8.0%) tested positive for HIV. Of the 71 patients with HIV seropositivity, 54 (76.1%) had pulmonary tuberculosis and only two of them (3.7%) were smear negative; nine patients (12.7%) were found to have pleural effusion, five (7.0%) had tuberculosis lymphadenopathy, one (1.4%) had both pleural and pericardial effusions, and two patients (3.7%) had miliary tuberculosis. Comparisons of pulmonary and extrapulmonary tuberculosis in HIV seropositive and seronegative patients are given in tables 4 and 5.

Table 4

Comparison of HIV status and pulmonary/extrapulmonary tuberculosis

Table 5

HIV status and types of extrapulmonary tuberculosis

Out of 150 patients who were treated at our centre, only 112 could be analysed, as the others had not completed the treatment at the time of data analysis. In all these patients the entire treatment, that is, intensive phase and continuation phase, was fully supervised. Overall, 78 patients (70%) were cured or completed treatment, 12 patients (11%) failed to respond, 16 patients (14%) defaulted, and six patients (5%) died during treatment.


The most significant finding of our study was that 96.8% of patients with pulmonary tuberculosis had positive sputum smears for AFB. Various studies have reported sputum smear positivities ranging from 39% to 74%.9–12 The reason for the high sputum smear positivity in our study could be the larger quantity of sputum (24 hour) submitted for AFB smear examination. A recent study by Warren et al documented 92% sputum smear AFB positivity when more than 5 ml of sputum was used for smear diagnosis of pulmonary tuberculosis.7 Use of 24 hour sputum appears to provide a sufficient volume for detection of AFB intermittently shed into respiratory secretions. A volume of 5–10 ml of sputum is recommended for the diagnosis of pulmonary tuberculosis,6 but most clinical laboratories do not comply with this guideline.8 Although AFB smear examination is an extremely simple test carried out in most laboratories, it requires dedicated effort to obtain accurate results. In developing countries, the AFB sputum smear may be the only test available for diagnosis.13 Hence, every effort should be made to establish reliable laboratories with adequate quality control.

Fine needle aspiration cytology or biopsy could establish the diagnosis in most cases of extrapulmonary tuberculosis. One patient with miliary tuberculosis required a trial of specific empiric antituberculosis therapy; diagnosis was subsequently confirmed by liver biopsy. Specific empiric antituberculosis therapy involves treatment with three specific antituberculosis drugs, isoniazid, pyrazinamide and ethambutol, in the absence of bacteriological or histological proof of tuberculosis.14 Rifampicin, being a broad spectrum antibiotic, is not included in empiric treatment as fever due to infections other than tuberculosis may respond to treatment, leading to misdiagnosis. Once the fever responds to specific empiric antituberculosis therapy, rifampicin is added to the treatment.

In this study, 8% of patients with tuberculosis were HIV seropositive. Various studies from India have reported rates of HIV seropositivity ranging from 0.4% to 4.9%, with a rising trend.15–18 Two recent studies, however, have reported HIV seropositivity in pulmonary tuberculosis patients to be 6.9% and 10%, respectively.19,20 The incidences of extrapulmonary tuberculosis in HIV seronegative and HIV seropositive patients in our study were almost the same, at 22% and 24%, respectively. Smear negative pulmonary tuberculosis was observed in 3.2% of HIV seronegative patients and in 3.7% of the HIV seropositive group. This indicates that the pattern of presentation of tuberculosis in HIV seropositive patients is the same as that in HIV seronegative patients. This finding has been highlighted in other recent reports.21,22 There is a close epidemiological link between HIV and tuberculosis,23,24 despite which the public health programmes have largely remained separate. The WHO has recommended that HIV and tuberculosis should be linked by voluntary counselling and testing for HIV in patients with tuberculosis.24 The benefits of voluntary counselling and testing for HIV in tuberculosis patients include early diagnosis, referral for appropriate clinical care, and support for those testing HIV positive.

DOT, which ensures that the patient takes medication under the direct supervision of a health assistant, has emerged as a solution to the problem of non-compliance and has been proved effective in various parts of the world.4 While all the patients at our centre received fully supervised treatment as recommended by the WHO, this is not routine practice under Indian RNTCP. Recent studies have shown that only one out of five DOTS workers ensures supervised swallowing of the doses, while others either ask the patient to go to the next room or hand over the entire weekly pack to be taken at home.25,26 So, what started off as DOTS ends up being POTS (partially observed treatment, short course) or even KNOTS (knowing but not observing treatment, short course).25

Despite claims of high cure rates when using DOT under RNTCP,5 only 70% of patients in our study were cured or completed treatment despite fully supervised therapy. The results are similar to those observed in some of the recent case-control studies.27,28 Reasons for the lower cure or completion rate in our study could be that the reliable microscopy laboratory used by our centre was better able to detect treatment failure, that there was a high default rate owing to the availability of loose drugs and loss of cases to private doctors, or that the cure rates claimed by the RNTCP are of questionable reliability.

Patients who were not willing to participate in DOTS were treated with a fixed drug combination. It is important to use methods of drug administration that avoid the danger of the use of rifampicin alone. Hence, if DOT is not possible, self administered treatment with a fixed drug combination could be used. Self supervised treatment with adequate counselling has shown results comparable to DOTS.29,30 The effectiveness of DOT therefore remains unclear, and further operational research is needed.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.