© 2002 by Oxford University Press
Utility of Clinically-directed Selective Screening to Diagnose HIV Infection in Hospitalized Children in Bombay, India
1 Departments of Paediatrics, LTM Medical College & LTMG Hospital, Bombay, India 2 Departments of Microbiology, LTM Medical College & LTMG Hospital, Bombay, India
The increasing prevalence of HIV infection in urban India together with limited financial resources necessitates judicious HIV testing. This prospective study was undertaken to determine the utility of selective screening for HIV infection based on five clinical risk factors reported in African children. The study was conducted at the Departments of Paediatrics and Microbiology, LTMG Hospital, Bombay, India between September 1998 and 2000. The children were enrolled after taking informed consent from their parents. The HIV seroprevalence rate was determined in children (aged 1 month to 12 years) consecutively admitted with severe malnutrition, serious pyogenic infections (pneumonia, pyogenic meningitis, septicaemia), disseminated tuberculosis, chronic diarrhoea and oral candidiasis, present either singly or in combination. Children above 18 months of age were diagnosed as being infected with HIV if they tested positive by two different HIV enzyme-linked immunosorbent assay (ELISA) tests. In children less than 18 months of age the diagnosis of HIV infection was made if they were ELISA positive and also fulfilled the WHO criteria for symptomatic HIV infection. Of a total 204 children (110 male, 94 female) screened, 24 (11.8 per cent) were diagnosed as HIV-infected. The HIV seropositive rate was highest in children having oral candidiasis (40.6 per cent), followed by chronic diarrhoea (18.2 per cent), disseminated tuberculosis (16.2 per cent), severe malnutrition (14.4 per cent), and serious pyogenic infections (11.2 per cent). Only the presence of oral candidiasis was a significant independent risk factor for predicting HIV infection (p < 0.0001). However, as the number of risk factors concomitantly present increased, the chances of the child being infected with HIV also increased significantly (p < 0.001). Our study shows that clinically-directed selective screening does have a practical role in diagnosing HIV infection in a resource-poor setting.