Objective: To compare the effect of immediate versus deferred antiretroviral treatment (ART) on neuropsychological test performance in treatment-naive HIV-positive adults with more than 500 CD4(+) cells/mu l.
Design: Randomized trial.
Methods: The START parent study randomized participants to commence immediate versus deferred ART until CD4(+) less than 350 cells/ml. The START Neurology substudy used eight neuropsychological tests, at baseline, months 4, 8, 12 and annually, to compare groups for changes in test performance. Test results were internally standardized to z-scores. The primary outcome was the average of the eight test z-scores (QNPZ-8). Mean changes in QNPZ-8 from baseline were compared by intent-to-treat using longitudinal mixed models. Changes from baseline to specific time points were compared using ANCOVA models.
Results: The 592 participants had a median age of 34 years; median baseline CD4(+) count was 629 cells/mu l; the mean follow-up was 3.4 years. ART was used for 94 and 32% of accrued person-years in the immediate and deferred groups, respectively. There was no difference between the immediate and deferred ART groups in QNPZ-8 change through follow-up [-0.018 (95% CI -0.062 to 0.027, P = 0.44)], or at any visit. However, QNPZ-8 scores increased in both arms during the first year, by 0.22 and 0.24, respectively (P<0.001 for increase from baseline).
Conclusion: We observed substantial improvement in neurocognitive test performance during the first year in both study arms, underlining the importance of using a control group in studies assessing neurocognitive performance over time. Immediate ART neither benefitted nor harmed neurocognitive performance in individuals with CD4(+) cell counts above 500 cells/mu l. Copyright (C) 2018 Wolters Kluwer Health, Inc. All rights reserved.