Cognitive Impairment in Tuberculous Meningitis

Abstract

Background: Cognitive impairment is reported as a common complication in adult tuberculous meningitis (TBM), yet few studies have systematically assessed the frequency and nature of impairment. Moreover, the impact of impairment on functioning and medication adherence has not been described.

Methods: A cognitive test battery (10 measures assessing 7 cognitive domains) was administered to 34 participants with human immunodeficiency virus (HIV)-associated TBM 6 months after diagnosis. Cognitive performance was compared with that a comparator group of 66 people with HIV without a history of tuberculosis. A secondary comparison was made between participants with TBM and 26 participants with HIV 6 months after diagnosis of tuberculosis outside the central nervous system (CNS). Impact on functioning was evaluated, including through assessment of medication adherence.

Results: Of 34 participants with TBM, 16 (47%) had low performance on cognitive testing. Cognition was impaired across all domains. Global cognitive performance was significantly lower in participants with TBM than in people with HIV (mean T score, 41 vs 48, respectively; P < .001). These participants also had lower global cognition scores than those with non-CNS tuberculosis (mean global T score, 41 vs 46; P = .02). Functional outcomes were not significantly correlated with cognitive performance in the subgroup of participants in whom this was assessed (n = 19).

Conclusions: Low cognitive performance following HIV-associated TBM is common. This effect is independent of, and additional to, effects of HIV and non-CNS tuberculosis disease. Further studies are needed to understand longer-term outcomes, clarify the association with treatment adherence, a key predictor of outcome in TBM, and develop context-specific tools to identify individuals with cognitive difficulties in order to improve outcomes in TBM.

Keywords: HIV; cognitive impairment; functional impairment; treatment adherence; tuberculous meningitis.

Figure 1.

Processing of cognitive test battery data. First, standardization of scores used normative scores calculated from healthy control data collected by the CONNECT study. These data were collected between 2018 and 2020 from healthy human immunodeficiency virus (HIV)–negative community-dwelling individuals who presented to the same community health clinics in Gugulethu from which the people with HIV (PWH) comparator group was recruited, within the same area of Cape Town where participants with tuberculous meningitis (TBM) and tuberculosis were recruited. The groups therefore shared key demographics (age, ethnicity, language, and education), as well as psychosocial and socioeconomic characteristics. Demographically corrected z scores (mean [standard deviation (SD)], 0 [1]) were then calculated using standard regression-based norming processes. The z scores were then converted to demographically corrected T scores (mean [SD], 50 [10]). If participants had z scores >5 SDs below the mean, the conversion to a T score resulted in a negative T score. In these cases, we assigned a score of 0, the lowest possible T score, to maintain the clinical significance of the low performance. Next, cognitive performance data were summarized into domain-specific and global T scores by taking the average of T scores within each domain and then across domain T scores. T scores were then converted to deficit scores. The overall global deficit score (GDS) was calculated by averaging deficit scores. A cutoff GDS of ≥0.5 has been considered consistent with “cognitive impairment” on cognitive test performance [22]; for the purposes of this study, we termed this group as having “low performance on cognitive testing” while the clinical significance and functional impact in TBM is further explored, aligned with recent trends in the HIV literature [23].

Figure 2.

Consort diagram describing enrollments across 3 parent studies. Abbreviations: CNS, central nervous system; HIV, human immunodeficiency virus; HIV-TBM, HIV-associated tuberculous meningitis; LTFU, lost to follow-up; PWH, people with HIV.

Figure 3.

Global T scores. Scatterplot graph displays mean scores with standard deviations, as well as individual values plotted for participants with human immunodeficiency virus (HIV)–associated tuberculous meningitis (HIV-TBM), people with HIV (PWH) with non–central nervous system (CNS) tuberculosis (comparator group 2), and PWH with HIV only (no history of tuberculosis; comparator group 1).