Adverse Events in HIV-Infected Persons Receiving Antiretroviral Drug Regimens in a Large Urban Slum in Nairobi, Kenya, 2003-2005.

OBJECTIVE: This article describes toxicities to antiretroviral therapy (ART) among HIV-infected patients receiving care at a clinic in a large urban slum in Nairobi, Kenya. METHODS: Patients were treated with nonnucleoside reverse transcriptase inhibitor-based ART and followed at scheduled intervals. Frequencies and cumulative probabilities of toxicities were calculated. RESULTS: Among 283 patients starting ART, any and severe clinical toxicity were recorded as 65% and 6%, respectively. Cumulative probabilities for remaining free of any and severe clinical toxicities at 6, 12, and 18 months, were 0.47, 0.26, and 0.17, respectively and 0.98, 0.95, and 0.89, respectively. The probability of remaining free from elevated and grade 3 or 4 serum aminotransferase (AST) at 6, 12, and 18 months were 0.62, 0.42, and 0.21, respectively, and 0.99 at 6, 12, and 18 months. CONCLUSIONS: ART toxicities were frequent, but severe toxicities were less common. In resource-limited settings, ART toxicity should not represent a barrier to care.

A program to provide antiretroviral therapy to residents of an urban slum in Nairobi, Kenya.

OBJECTIVE: To evaluate retention in care and response to therapy for patients enrolled in an antiretroviral treatment program in a severely resource-constrained setting. METHODS: We evaluated patients enrolled between February 26, 2003, and February 28, 2005, in a community clinic in Kibera, an informal settlement, in Nairobi, Kenya. Midlevel providers offered simplified, standardized antiretroviral therapy (ART) regimens and monitored patients clinically and with basic laboratory tests. Clinical, immunologic, and virologic indicators were used to assess response to ART; adherence was determined by 3-day recall. A total of 283 patients (70% women; median baseline CD4 count, 157 cells/ mm(3); viral load, 5.16 log copies/mL) initiated ART and were followed for a median of 7.1 months (n = 2384 patient-months). RESULTS: At 1 year, the median CD4 count change was +124.5 cells/mm(3) (n = 74; interquartile range, 42 to 180), and 71 (74%) of 96 patients had viral load <400 copies/mL. The proportion of patients reporting 100% adherence over the 3 days before monthly clinic visits was 94% to 100%. As of February 28, 2005, a total of 239 patients (84%) remained in care, 22 (8%) were lost to follow-up, 12 (4%) were known to have died, 5 (2%) had stopped ART, 3 (1%) moved from the area, and 2 (< 1% ) transferred care. CONCLUSIONS: Response to ART in this slum population was comparable to that seen in industrialized settings. With government commitment, donor support, and community involvement, it is feasible to implement successful ART programs in extremely challenging social and environmental conditions.