Importance of blood samples for diagnosis and drug sensitivitytesting in HIV positive patients with suspected tuberculosis.

Background: Diagnosis of tuberculosis is difficult in HIV positive patients since they often present with atypical symptoms and are susceptible to pulmonary infections that mimic tuberculosis. Sputum collection may not be possible even in patients with pulmonary involvement since a productive cough is not always present. In such patients, blood smear and culture for AFB apart from serving as a diagnostic tool can be used for testing drug sensitivity. Objectives: This study was undertaken to explore the value of blood culture for diagnosis in patients with suspected TB. In addition, a comparison of drug sensitivity patterns of blood and sputum isolates in 10 of these patients was also carried out. Methods: Blood and sputum samples were processed, cultured and isolates tested for their drug susceptibility and for niacin production, nitrate reduction as well as catalase activity at 68o C. Results: All 24 blood samples were culture positive although only 6 were smear positive. On the basis of the biochemical investigations, 22 strains were identified as Mycobacterium tuberculosis. All the 10 sputum samples were culture positive despite 4 being smear negative. Comparison of drug sensitivity profiles from blood and sputum revealed concordance to five first or second line drugs in 5 of 10 patients. Additionally, 2 patients demonstrated discordance for only one first or second line drug. Conclusion: The study demonstrates the importance of blood culture in confirming diagnosis of tuberculosis and testing for drug sensitivity in HIV positive patients without a productive cough. The level of discordance in drug sensitivity profiles between blood and sputum the same individual is suggestive of infection with multiple strains. Testing for the occurrence of multistrain infections through individual colony examination of a single isolate is necessary since such infections would affect treatment of non-responder patients having HIV-TB dual infections.

The natural history of human immunodeficiency virus infection among adults in Mumbai.

BACKGROUND: Natural history studies of untreated HIV infection are useful for clinicians, public health experts and policymakers to improve and monitor care, plan services and control, and to model the epidemic. Several natural history studies on homosexual men and intravenous drug users have been published from developed countries. A few studies have emerged on heterosexual populations from Africa. With an emerging epidemic, a similar study was required in India. This study was designed to determine the progression of HIV disease in a prevalent cohort of adult HIV-seropositive patients. METHODS: A prevalent cohort of 1009 patients comprising 488 asymptomatic HIV-seropositive persons, 259 with AIDS-related complex (ARC), and 262 with acquired immunodeficiency syndrome (AIDS) were recruited for the study at Sir J.J. Hospital, Mumbai. A case-control study was conducted to determine the correlation of clinical features and other factors with disease progression. Disease progression was determined from the asymptomatic stage to that of ARC and AIDS using time series analysis. The incubation period from HIV to AIDS was also determined, using Weibull curves. RESULTS: The median incubation periods for progression were: HIV to AIDS-7.9 years and ARC to AIDS--1.9 years. The median survival after developing AIDS was 19.2 months. A comparison of progressors and non-progressors revealed that disease progression correlated with clinical features such as chronic fever (OR 5.6), persistent generalized lymphadenopathy (OR 4.7), persistent cough for >1 month (OR 3.5), chronic diarrhoea (OR 3.3), oral candidiasis (OR 3.2), >10% loss of body weight within 1 month (OR 2.9), incident tuberculosis (OR 2.8) and herpes zoster (OR 2.5). The annual incidence of active clinical tuberculosis was 86/1503 person-years (5.7/ 100 person-years), the median time to occurrence of active tuberculosis was 21.6 months and the annual incidence of mortality was 96/2009 person-years (4.8/100 person-years, 95% CI 3.4, 6.2). CONCLUSION: Progression to AIDS and death was faster among the heterosexual cohort in Mumbai than that reported for homosexual men and haemophiliacs in the USA and Europe. Strategies need to be developed to prevent the occurrence of tuberculosis among HIV-infected patients because that would help to reduce the morbidity and mortality. This is the first large study from the Indian subcontinent of a longitudinal follow up of HIV-infected persons. The findings will be useful for advocacy and assessing the impact of antiretroviral therapy (ART) in India.

Community prevalence of sexually transmitted diseases and human immunodeficiency virus infection in Tamil Nadu, India: a probability proportional to size cluster survey.

BACKGROUND: Human immunodeficiency virus (HIV) infection and AIDS is threatening the survival of many nations. To evaluate ongoing interventional strategies and burden of illness estimates, valid data on the prevalence of HIV are required. Often, in the absence of community prevalence data, estimates are based on surrogate markers such as prevalence of HIV in antenatal clinics. Even though the antenatal prevalence of HIV is easier to measure and can be repeated for evaluation, it is important to establish the association between antenatal and community prevalence of sexually transmitted diseases (STDs) and HIV, so that the validity of the estimates can be verified. METHODS: A 'probability proportional to size' cluster survey was conducted in three randomly selected districts of Tamil Nadu in India. The basic unit of the survey was households from rural and urban clusters. Adults 15-45 years of age from the selected households were eligible for recruitment. Demographic, behavioural and laboratory data were collected. Clinical examination was done to identify STD syndromes and blood, urine, vaginal/urethral and endocervical swabs were taken for laboratory diagnosis of STDs from the subjects. Direct smear examination for Trichomonas vaginalis; serological tests for syphilis, hepatitis B, HIV, herpes simplex virus 2, Chlamydia trachomatis; and culture of Neisseria gonorrhoeae and Haemophilus ducreyi were performed on the collected specimens. The data were analysed adjusting for cluster effect. RESULT: We selected and screened 1981 individuals (1157 women and 824 men) for STDs and HIV from 1114 households representing the 25 million projected adult population of Tamil Nadu. The overall community prevalence of STDs including HIV and hepatitis B in Tamil Nadu was 14.6% (CI: 14.1-15.1), and 8.3% (CI: 7.9-8.6) when HIV and hepatitis B were excluded. Community prevalence of HIV and hepatitis B infection was 1.8% (CI:1.7-1.9) and 5.3% (CI: 5.1-5.5), respectively. The distribution of HIV involved both rural and urban regions of Tamil Nadu. On clinical examination, at least one STD syndrome was noted in 486 (24.5%) of the women subjects; vaginal discharge was the most common and found in 421 women (38.4%). CONCLUSION: The prevalence of STD and HIV in Tamil Nadu is higher than expected and has extended into the non-high risk population (generalized epidemic).