Objective: To assess the clinical significance of HIV-1 RNA levels detectable using the Amplicor HIV-1 Monitor method but < 400 copies/ml versus levels undetectable by this method.
Design: Retrospective cohort study.
Methods: All plasma HIV-1 RNA results over 13 months in our institution were reviewed. The study population comprised all individuals that achieved an HIV-1 RNA level < 400 copies/ml and remained on stable antiretroviral therapy. Results of < 400 copies/ml were stratified as 'below quantifiable limits' (BQL) or 'below detectable limits' (BDL). We examined the incidence of virologic relapse, defined as an HIV-1 RNA level > 400 copies/ml, for individuals with viral loads of BQL or BDL. Cox proportional hazards regression analyses were performed to control for baseline CD4 cell count, the number of antiretroviral medications, and the use of protease inhibitors (PI) and/or non-nucleoside reverse transcriptase inhibitors (NNRTI).
Results: Virologic relapse occurred in 52 of 168 individuals over 29,576 person-days overall (incidence rate 1.8 cases/1,000 person-days). The relapse rate was three times greater following HIV-1 RNA levels of BQL rather than BDL [crude rate ratio 3.2; 95% confidence interval (CI) 1.8-5.8]. After adjusting for baseline CD4 cell count, number of antiretroviral medications, and use of PI and/or NNRTI, the rate of relapse was nearly four times greater for individuals with HIV-1 RNA levels of BQL (hazard ratio 3.7; 95% CI 2.0-6.7).
Conclusions: In a large clinic population, low-level HIV-1 RNA detected in plasma below the 400 copies/ml limit of quantifiability for Amplicor HIV-1 Monitor was associated with an increased rate of virologic relapse on therapy.