Background: Transient human immunodeficiency virus (HIV) viremia, a common phenomenon among patients taking antiretroviral therapy, is often attributed to lapses in adherence to the medication regimen. We investigated this relationship in a prospective observational cohort of 128 patients initiating a new regimen.
Methods: A case of transient viremia was defined as an HIV RNA level of 40-1000 copies/mL ("blip") sandwiched between 2 months of HIV RNA levels <40 copies/mL ("pre" and "post"). Adherence was most often measured with a composite adherence score (CAS), which is primarily based on electronically measured adherence. Case subjects' adherence and dose-timing was compared with (1) that of other patients (control subjects), who had undetectable virus loads for 3 consecutive months, and (2) that during periods of sustained undetectable virus loads among the case subjects themselves, if available.
Results: Among the 28 case subjects, mean CAS-measured adherence did not decrease before transient viremia; adherence during the pre, blip, and post periods were 86%, 84%, and 80%, respectively. Control subjects had lower adherence levels during the corresponding 3 months (77%, 79%, and 75%, respectively; P = .046). Among the 19 patients able to serve as their own controls, CAS-measured adherence was higher during the period of transient viremia than during control periods (P = .01). Similar relationships were found when comparing only electronically measured adherence on a week-wise basis. There were no significant differences in dose-timing error between case subjects and control subjects.
Conclusions: We found no evidence that transient HIV viremia is associated with decreases in adherence or differences in dose-timing. Other etiologies for transient viremia should be evaluated.