Discontinuation of potent antiretroviral therapy: Predictive value of and impact on CD4 cell counts and HIV RNA levels

Objective To characterize predictors and consequences of discontinuing antiretroviral therapy (ART) in terms of CD4 cell count, HIV RNA, and reported side-effects in a large cohort of HIV-infected women. Design Cohort study. Methods A total of 1058 HIV-infected women initiated potent ART before September 1999. For each 6 month period after October 1996 we determined the proportion of potent ART users who downshifted to non-potent ART and who discontinued all ART. We examined the role of CD4 cell count and HIV RNA with regard to ART discontinuation. Results Between October 1996 and September 1999, 1058 individuals contributed 3362 visits at which potent ART was reported in the previous 6 months. Overall rates of 6 month downshifting and discontinuation were 10.0% and 6.7%. The proportion of individuals discontinuing all ART increased from 2.9% in late 1996 to 9.1% in mid 1999 (P < 0.001). Individuals with high HIV RNA levels were more likely to discontinue (P < 0.05). Compared to those who continued on potent ART, individuals who discontinued experienced large declines (P < 0.001) in CD4 cell counts and were more than three times more likely (P < 0.001) to experience HIV RNA increases. However, over one-third of those discontinuing ART reported side-effects and this subset had smaller CD4 cell count declines as compared to discontinuers not reporting side-effects (P = 0.147). Conclusions In a large cohort of HIV-infected women, an increasing proportion of potent ART users discontinued ART over 3 years. Higher HIV RNA levels predicted discontinuation. Immediate immunological/virological deleterious consequences were observed. Side-effects were the most common reason for discontinuation and CD4 cell count declines were larger among those who did not cite side-effects as the reason for discontinuation.