Central nervous system antiretroviral penetration and cognitive functioning in largely pretreated HIV-infected patients.

Prolonged time on effective antiretroviral therapy (ART) should be associated with a low incidence of neurocognitive impairment (NCI). We investigated the rate of NCI in 162 largely pretreated patients with HIV RNA suppression according to CNS antiretroviral drug penetration in comparison with 67 patients on their first ART line. Cognitive performance (Trailmaking A, B, Digit Symbol, Grooved Pegboard; demographically adjusted and converted to Z scores, NPZ4) was evaluated, and CNS penetration effectiveness (CPE) ranks of 1 to 4 were assigned and summed per regimen. After a median of 173.2 months on therapy (1,909.3 patients-year), the rate of NCI was similar in both groups (mean NPZ4, -0.24 vs -0.2). Pretreated patients received regimens with a CPE <7 in a large proportion (30 vs 9 %; p < 0.01). Patients in monotherapy had worse NPZ4 score than patients receiving triple therapy (-0.78 vs -0.18; p = 0.02; effect sizes 1.38), and a lower CPE score was observed in patients with a CD4+ count nadir <200 cells/ml with NCI (6.5 vs 7.3, p = 0.04). In the multivariate model, only the lowest CD4+ count and hepatitis C virus coinfection were associated with NPZ4, whereas CPE <7 showed a trend to association (p = 0.06) probably due to patients on monotherapy (estimate -0.33, p < 0.01). In conclusion, the rate of NCI in largely pretreated patients was similar to that observed in patients on their first regimen, and nadir CD4+ count continue to be critical. CNS drug penetration should be considered in cases of high risk for NCI.