Background:  A5199 is a trial that compared the neurological and neuropsychological test performance effects of 3 ARV regimens in participants with HIV-1 in resource-limited settings. The association of these neurological outcomes with background and other disease events was also monitored. Here we report an association of diagnosed TB and these outcomes.

Methods:  Participants from Brazil, India, Malawi, Peru, South Africa, Thailand, and Zimbabwe were randomized to 3 ARV treatment arms as part of the ACTG PEARLS study. Standardized neurological and brief neuropsychological exams (grooved pegboard, timed gait, semantic verbal fluency, and finger tapping) were administered every 24 weeks from February 2006 to May 2010. Associations with neurological and neuropsychological function were estimated from linear and logistic regression models using generalized estimating equations (GEE). TB diagnoses were made according to NIAID’s Division of AIDS (DAIDS) Appendix 60 criteria.

Results:  Baseline characteristics of the 860 participants were:  53% female, 49% African; median age 34 years; median (Q1, Q3) screening CD4 of 173 (98, 232) cells/mm³; and median entry plasma log₁₀ HIV-1 RNA of 5.0 copies/mL (4.5 to 5.5). At baseline, there were 35 cases of pulmonary, 9 cases of extrapulmonary (1 with both), and one case of central nervous system (CNS) TB. There were 91 observations in 53 persons with pulmonary, 35 observations in 19 persons with extrapulmonary, and 4 observations in 2 persons with CNS TB over the 192 weeks of follow-up. TB observations decreased over time with ART and continued follow-up. GEE models were constructed to assess the concurrent association between the TB diagnoses with neurological and neuropsychological performance, with the co-variates of country, randomized treatment from A5175, screening HIV-1 RNA strata, screening CD4 count, baseline neuropsychological/neurologic functioning, age, sex, and years of education. Those participants with TB were significantly slower in both grooved pegboard dominant (p <0.001) and non-dominant hands (p <0.001). Participants who had TB performed worse in finger-tapping non-dominant hand (p = 0.01). TB was positively associated with diffuse CNS disease on exam (p = 0.03).

Conclusions:  TB co-infection was associated with poorer neurological and neuropsychological functioning. Whether this association was related to indirect effects on CNS function such as medication, immune activation and enhanced CNS HIV infection or more direct CNS effects of TB remains to be determined.