Prevalence and Correlates of Neurocognitive Disorders among HIV Patients on Antiretroviral Therapy at a Kenyan Hospital

Abstract

Background HIV-associated neurocognitive disorders (HAND) represent a spectrum of cognitive abnormalities affecting attention, concentration, learning, memory, executive function, psychomotor speed, and/or dexterity. Our objectives in this analysis are to determine the prevalence of HAND and the covariates in a Kenyan population. Methods We conducted a cross-sectional study in a convenient sample of people living with HIV on antiretroviral therapy (ART) attending routine care visits at the Kenyatta National Hospital HIV clinic between July and August 2015. Baseline demographics were obtained using interviewer-administered questionnaires; clinical data were abstracted from patient records. Trained research clinicians determined the neurocognitive status by administration of the International HIV Dementia Scale (IHDS), the Montreal Cognitive Assessment (MOCA) scale, and the Lawton Instrumental Activities of Daily Living (IADL) scale. Cognitive impairment was defined as a score of ≤26 on the MOCA and ≤10 on the IHDS. Descriptive analysis and logistic regression to determine predictors of screening positive for HAND were done with the significance value set at <0.05. Results We enrolled 345 participants (202 men; 143 women). The mean age of the study population was 42 years (±standard deviation (SD) 9.5). Mean duration since HIV diagnosis and mean duration on ART were 6.3 (±SD 3.7) and 5.6 years (±SD 3.4), respectively. Median CD4 count at interview was 446 cells/mm3 (interquartile range (IQR) 278–596). Eighty-eight percent of participants screened positive for HAND, of whom 87% had asymptomatic neurocognitive impairment (ANI) and minor neurocognitive disorders (MND) grouped together while 1% had HIV-associated dementia (HAD). Patients on AZT/3TC/EFV were 3.7 times more likely to have HAND (OR = 3.7, p=0.03) compared to other HAART regimens. In the adjusted analysis, women were more likely to suffer any form of HAND than men (aOR = 2.17, 95% CI: 1.02, 4.71; p=0.045), whereas more years in school and a higher CD4 count (aOR = 0.58, 95% CI: 0.38, 0.88; p=0.012), (aOR = 0.998, 95% CI 0.997, 0.999; p=0.013) conferred a lowered risk. Conclusion Asymptomatic and mild neurocognitive impairment is prevalent among people living with HIV on treatment. Clinical care for HIV-positive patients should involve regular screening for neurocognitive disorders while prioritizing women and those with low education and/or low CD4 counts.