Background:  When viewed from the perspective of cross-sectional research studies, the vast majority of HIV-infected patients using ART in Africa achieve virologic suppression. However, the lack of frequent longitudinal virologic monitoring in large numbers of patients has limited our finer understanding of the occurrence and outcomes of detectable viremia. Indeed, early reports have identified an unexpected incidence of detectable viremia, which has been subsequently re-suppressed without ART change. We sought to describe the occurrence and outcomes of high-level viral rebound in a well characterized group of HIV-infected African adults.

Methods:  We evaluated HIV-infected adults initiating NNRTI-based ART in the Uganda AIDS Rural Treatment Outcomes (UARTO) cohort in Mbarara, Uganda. We collected quarterly plasma HIV RNA levels and medication event monitoring systems (MEMS) data, which counts how many times a day a pill bottle is opened. For this analysis, high-level rebound was defined as at least 1 HIV RNA measurement <400 copies/mL in the first 6 months of ART followed by a value > 10,000 copies/mL. 

Results:  Between June 2005 and August of 2008, 350 patients achieved at least 1 HIV RNA measurement <400 copies/mL in the first 6 months. Subsequently, the cumulative incidence of virologic rebound to >10,000 copies/mL at 0.5, 1, 1.5, and 2 years was 2.8, 8.8, 13.7, and 15.7%. Of the 37 individuals with virologic rebound, 34 had at least 1 subsequent HIV RNA measurement. Of these 34, 17 (50%) resuppressed to <400 copies/mL while remaining on an NNRTI-based regimen, 7 (21%) resuppressed after switch to a PI-based regimen, and 10 (29%) failed to resuppress (2 of whom had been switched to PI-based ART). MEMS data were available in 14 patients who experienced resuppression without ART change; treatment interruptions of at least 48 hours were seen in 5 patients (36%), sporadic non-adherence without any intervals >48 hours in 4 (29%), and adherence > 95% in the remaining 5 (36%).

Conclusions:  As has been seen in other African-based cohorts, many patients with high-level virologic rebound subsequently achieve virologic resuppression without switch to a new “salvage” regimen. While the mechanisms for this apparent transient high-level viremia require further investigation, this suggests that an episode of high level virologic rebound on an NNRTI does not always preclude subsequent effectiveness of the regimen.