Abstract Body

Background: Neurocognitive impairment is an important co-morbidity in HIV infection. Data describing neurocognitive function in patients from sub-Saharan Africa are generally sparse, with only limited data describing neurocognitive function at the time of failure of first-line therapy, and none describing the changes in neurocognitive function in response to second-line therapy.

Methods: We studied patients who were enrolled in EARNEST, a large multi-centre trial of second-line therapy conducted at 14 sites in sub-Saharan Africa. Eligible patients were >12 years old and failing first-line therapy according to WHO criteria after >12 months on an NNRTI-based regimen. Patients were randomised to take second-line therapy (open-label) with lopinavir/ritonavir (400mg/100mg twice daily) plus either 2-3 clinician-selected NRTIs, raltegravir, or as monotherapy after 12 weeks’ induction with raltegravir. Neurocognitive function was tested on those aged >18 years at baseline (switch to second-line), weeks 48 and 96 using colour trails tests 1 and 2, and the grooved pegboard test. Test results were converted to a composite z-score using US test norms.

Results: 1036 patients (>86% of those >18 y enrolled in the trial) had evaluable tests at baseline, and 915 patients at week 96. Patients were 58% female, mean age 38 years, median viral load 65,000 c/ml, median CD4 count 73 cells/mm3 at first-line failure. Mean (SD) composite z-score at baseline was -2.96 ± 1.74; z-scores were independently lower in older individuals, and those with lower body weight, higher viral load, lower haemoglobin, fewer years of education, fewer working hours per week, previous CNS disease, or taking fluconazole (all P<0.05 in a multivariable model). Neurocognitive function improved markedly after starting second-line therapy (mean (SE) increase in Z-score of +1.23 (0.04) at week 96), with no difference between the arms (P=0.35; Figure 1).

Conclusions: Patients in sub-Saharan Africa failing first-line therapy appear to have severely impaired neurocognitive function. In part this may reflect inadequate data for normalization, but the relationship with high viral load and indicators of general debility suggest much of the impairment is HIV-related. Neurocognitive function improves markedly on second-line therapy. The lack of difference between standardised treatment regimens with very different CNS penetration indicates that this need not be a primary consideration in the choice of an optimal drug regimen for second-line therapy.

viewer?TAG_ACTION=SECURE_FILE_DOWNLOAD&V
Figure 1: Changes in z-score over 96 weeks in the three randomised groups