Switching to second-line ART in relation to mortality in a large Tanzanian HIV cohort

Objectives: In a large cohort of HIV-infected Tanzanians, we assessed: (i) rates of first-line treatment failure andswitches to second-line ART; (ii) the effect of switching to second-line ART on death and loss to follow-up; and(iii) treatment outcomes on second-line ART by regimen.Methods: HIV-1-infected adults (≥15 years) initiated on first-line ART between November 2004 and September2012, and who remained on initial therapy for at least 24 weeks before switching, were studied. Survival analyseswere conducted to examine the effect of second-line ART on mortality and loss to follow-up in: (i) the whole cohort;(ii) all patients eligible for second-line ART by immunological failure (IF) and/or virological failure (VF) criteria;and (iii) patients eligible by VF criteria.Results: In total, 47296 HIV-infected patients [mean age 37.5 (SD 9.5) years, CD4 175 (SD 158) cells/mm3, 71%female] were included in the analyses. Of these, 1760 (3.7%) patients switched to second-line ART (incidencerate"1.7/100 person-years). Higher rates of mortality were observed in switchers versus non-switchers in all patientsand patients with ART failure using IF/VF criteria. Switching only protected against mortality in patientswith ART failure defined virologically and with the highest level of adherence [switching versus non-switching;>95% adherence; adjusted HR"0.50 (95%CI"0.26-0.93); P"0.03].Conclusions: Switching patients to second-line ART may only be beneficial in a select group of patients who arevirologically monitored and demonstrate good adherence. Our data emphasize the need for routine viral loadmonitoring and aggressive adherence interventions in HIV programmes in sub-Saharan Africa.