Mortality in HIV infected individuals in Pune, India.

Background & objectives: With the presence of HIV epidemic for more than two decades in India, rise in the number of HIV related deaths is expected. Data on mortality in HIV infected individuals from prospective studies are scanty in India. We report here data on mortality in a systematically followed cohort of HIV infected individuals at Pune, Maharashtra, India Methods: A total of 457 HIV infected individuals were enrolled in a prospective study in Pune between September 2002 and November 2004. They were evaluated clinically and monitored for CD4 counts at every quarterly visit. Mortality data were collected from the records of hospital facilities provided by the study. If the death occurred outside such hospitals; relatives of the participants were requested to inform about the death. Results: Median CD4 count in study participants was 218 cells/µl (95% CI: 107-373) at baseline. The median duration of follow up was 15 months (IQR: 12, 22). Mortality was higher in antiretroviral therapy (ART) naive patients compared to those who received treatment (16.59 vs. 7.25 per 100 person years). Participants above 35 yr of age, CD4 count less than or equal to 100 cells/µl at baseline, tuberculosis at any study time point and ART status were independently associated with high mortality [(RR=1.97; 95% CI: (1.23, 3.14), P=0.005, (RR=33.20, 95%CI (7.59, 145.29), P<0.001, (RR=2.38, 95% CI (1.38, 4.09), P= 0.002 and RR=5.60, 95% CI (3.18, 9.86), P<0.001, respectively]. Interpretation & conclusions: High mortality at advanced immunosuppression highlights the importance of early detection of HIV infection. Emphasis needs to be given at timely diagnosis and management of tuberculosis and ART initiation. It is important to create awareness about availability of free antiretroviral drugs in the government ART roll out programme.

Mortality & clinical characteristics of hospitalized adult patients with HIV in Pune, India.

BACKGROUND & OBJECTIVE: In India, data regarding mortality and clinical characteristics of hospitalized HIV-infected patients are sparse, which may limit the effectiveness of new hospital-based HIV programmes providing antiretroviral therapy (ART). The objective of our study was to determine mortality and clinical characteristics of hospitalized HIV-infected individuals in a high HIV prevalence region of India. METHODS: A retrospective chart review was done of known HIV-infected adults admitted to the Medical Service of a large, public hospital in Pune, India, from January 2002 to November 2003. RESULTS: A total of 655 HIV-infected patients were identified; 489 (74.7%) were male and 4 (0.6%) were on ART. The most common illnesses reported were tuberculosis (55.8%), diarrhoea (4.2%), and alcoholic liver disease (3.7%) . The inpatient mortality was 172 (26.3%). The most common causes of death of the 172 people were tuberculosis (52.9%) and cryptococcal meningitis (7.6%). In multivariate analysis, factors associated with increased mortality were male sex (adjusted odds ratio (AOR) 1.92, 95% CI: 1.08-3.41), haemoglobin level < 7 g/dl (AOR 2.75, 95% CI:1.23-6.14), length of stay < 2 days (OR 5.78, 95%, CI: 1.82-18.4), and cryptococcal meningitis (OR 4.44, 95% CI:1.19-16.6). INTERPRETATION & CONCLUSION: In the era prior to widespread ART, a high inpatient mortality of 26 per cent was found among hospitalized HIV-infected individuals. Thus, while hospitalization is an important access and referral point for HIV care and treatment, earlier identification of HIV-infected persons must occur to ensure they will optimally benefit from the government's ART programme.

Mother-to-child transmission of HIV among women who chose not to exclusively breastfeed their infants in Pune, India.

BACKGROUND & OBJECTIVE: The percentage of HIV cases attributed to mother-to-child transmission (MTCT) has increased several fold in recent years. No reports are available on HIV MTCT rates among HIV-infected choosing not be exclusively breastfeed their infants in India. We examined HIV MTCT rates among 41 Indian women in a prospective cohort who chose predominantly not to exclusively breastfeed. METHODS: Of the 41 women, 27 (66%) received MTCT prophylaxis: 3 received short course zidovudine (AZT), 19 single-dose nevirapine (NVP), and 5 both AZT and NVP. Maternal HIV-I RNA levels (viral load) were measured at the time of delivery. Infants were tested for HIV-1 infection by PCR up to 11 times is first year of life and viral load was measured in PCR positive infants. RESULTS: All infants received single dose NVP. Thirty two (76%) infants were exclusively formula-fed, 10 (24%) were mixed fed. Four infants were diagnosed with HIV infection for an overall 12- month transmission probability of 8 per cent [95% confidence interval (CI) of 3.2 to 22.1%]. Restricting analysis to 31 women who exclusively formula-fed, only one (3.1%) transmission event occurred. The 41 HIV-infected women gave birth to 42 live-born infants. INTERPRETATION & CONCLUSION: Our data from a small cohort of HIV-infected women suggest that short-course AZT or single dose NVP are effective in reducing MTCT in an Indian setting. Larger studies are needed to assess HIV MTCT rates in India, but in this small study rates were comparable to that observed among women who chose not to exclusively breastfeed in other resource-limited settings.

Incident HIV infection among men attending STD clinics in Pune, India: pathways to disparity and interventions to enhance equity.

Systematic disparities in rates of HIV incidence by socioeconomic status were assessed among men attending three sexually transmitted disease (STD) clinics in Pune, India, to identify key policy-intervention points to increase health equity. Measures of socioeconomic status included level of education, family income, and occupation. From 1993 to 2000, 2,260 HIV-uninfected men who consented to participate in the study were followed on a quarterly basis. Proportional hazards regression analysis of incident HIV infection identified a statistically significant interaction between level of education and genital ulcer disease. Compared to the lowest-risk men without genital ulcer disease who completed high school, the relative risk (RR) for acquisition of HIV was 7.02 (p < 0.001) for illiterate men with genital ulcer disease, 3.62 (p < 0.001) for men with some education and genital ulcer disease, and 3.02 (p < 0.001) for men who completed high school and had genital ulcer disease. For men with no genital ulcer disease and those with no education RR was 1.09 (p = 0.84), and for men with primary/middle school it was 1.70 (p = 0.03). The study provides evidence that by enhancing access to treatment and interventions that include counselling, education, and provision of condoms for prevention of STDs, especially genital ulcer disease, among disadvantaged men, the disparity in rates of HIV incidence could be lessened considerably. Nevertheless, given the same level of knowledge on AIDS, the same level of risk behaviour, and the same level of biological co-factors, the most disadvantaged men still have higher rates of HIV incidence.