Dynamics of cognitive change in impaired HIV-positive patients initiating antiretroviral therapy

Abstract

Objective: To rigorously evaluate the time course of cognitive change in a cohort of individuals with HIV-associated neurocognitive disorders (HAND) initiating combination antiretroviral therapy (CART), and to investigate which demographic, laboratory, and treatment factors are associated with neuropsychological (NP) outcome (or "any NP improvement").

Methods: Study participants included 37 HIV+ individuals with mild to moderate NP impairment who initiated CART and underwent NP testing at 12, 24, 36, and 48 weeks thereafter. NP change was assessed using a regression-based change score that was normed on a separate NP-stable group thereby controlling for regression toward the mean and practice effect. Mixed-effect regression models adjusting for loss to follow-up were used to evaluate the time course of cognitive change and its association with baseline and time-varying predictors.

Results: In persons with HAND initiating CART, cognitive improvement happens soon after initiation (13% at week 12), but more often 24, 36, and up to 48 weeks after initiation (up to 41%), with fewer than 5% demonstrating significant worsening. In multivariate analyses, unique predictors of NP improvement included more severe baseline NP impairment and higher CART CNS penetration index. Greater viral load decrease was associated with NP improvement only in univariate analyses.

Conclusion: Clinically meaningful neuropsychological improvement seemed to peak around 24-36 weeks after combination antiretroviral therapy initiation and was prolonged over the 1-year study period. This study also provides new evidence that benefit may be maximized by choosing antiretroviral medications that reach therapeutic concentrations in the CNS.

Figures

Figure 1 Proportion of HIV+ individuals with neuropsychological improvement from baseline Proportion of participants with neuropsychological (NP) improvement at each session was derived from the categorical mean scaled score regression change score (MS-Reg-CS). The MS-Reg-CS provides a standard score that can be used as a continuous score or a categorical score (i.e., significant NP improvement as MS-Reg-CS ≥1.645 and NP decline as MS-Reg-CS ≤−1.645, based on a two-tailed 90% confidence interval; individuals within these ranges were classified as NP stable). Only one individual declined at week 36 (4.5%). In this figure, 95% confidence intervals are provided around the observed proportions of NP change at each visit. At each study time point, proportion of improving cases is different from zero.

Figure 2 Average change in neuropsychological performance across study time MS-Reg-CS = mean scaled score regression change score. A negative MS-Reg-CS indicates decline, while a positive MS-Reg-CS indicates improvement. Significant decline was defined as a MS-Reg-CS ≥−1.645 and improvement was defined as a MS-Reg-CS ≥+1.645, which is equivalent to a 90% confidence interval. Observed standard deviation for MS-Reg-CS: week 12 = 1.19; week 24 = 1.37; week 36 = 2.00; week 48 = 1.83. Mixed effect regression model (4*) showed that change in the MS-Reg-CS was significant at visit 12 (p = 0.002) as compared to baseline. Subsequent MS-Reg-CS at week 36 (p = 0.019) and week 48 (p = 0.003) visits were significantly different compared to week 12 visit, but not at week 24 visit (p = 0.11). *Wald test of random intercept model. A higher GDS represents lower performance. We illustrated three profiles of improvement depending on three different levels of baseline impairment (mild impairment: 0.5 ≤ GDS < 1; moderate impairment: 1 ≤ GDS <1.5; and severe impairment: GDS ≥1.5). Only the 37 patients’ average MS-Reg-CS was tested. In other words, the impairment subgroups were not statistically compared due to sample sizes and resulting power issues. To explore the advantage of using the MS-Reg-CS as the outcome variable, we conducted the same statistical analyses simply using the uncorrected mean scaled scores. We found NP improvement to be highly significant at all time points (p < 0.0001), which differs substantially from the above p values, reflecting an inflated improvement probably corresponding to uncorrected regression toward the mean and practice effect. Note that since the response measures change from baseline, the week 0 responses are not used in the model.