Tuberculin skin test conversion among HIV patients on antiretroviral therapy in Uganda

SETTING: A human immunodeficiency virus (HIV) clinic in a setting of high tuberculosis (TB) and HIV prevalence. OBJECTIVE: To study the incidence of and factors associated with tuberculin skin test (TST) conversion in HIV patients on antiretroviral therapy (ART). DESIGN: Prospective cohort study of TST-negative, ART-naive HIV patients (CD4 cell count < 250 cells/mu l) without active TB. TST was repeated at 2 months and, if negative, at 6 months. TST positivity was defined as an induration of >= 5 mm. Clinical examination, chest X-ray and CD4 cell counts were performed at baseline and follow-up. Proportions and incidence of TST conversion were calculated, and logistic regression analyses were performed. RESULTS: Of the 142 patients, 105 (75.5%) were females. The mean age was 35.9 years (standard deviation 8.1) and the median CD4 cell count was 119 cells/mu l (interquartile range 42-168). The incidence of TST conversion was 30.2/100 person years (95%CI 19.5-46.8). Conversion was not associated with clinical, CD4 cell count or chest radiography findings. CONCLUSIONS: A high incidence of TST conversion was observed, supporting the World Health Organization recommendation to provide isoniazid preventive therapy (IPT) to all HIV patients in high TB prevalence settings. If case-control programmes choose to provide IPT only to TST-positive patients, repeat TST should be considered following initiation of ART.