Menu
Menu
Menu
Menu
Introduction: Over 15 million children who were exposed to HIV perinatally but uninfected (HEU) are alive globally, and they are faced with multiple risk factors for poor neurodevelopment. While children who are HIV-infected (HIV+) appear to have worse neurodevelopmental scores compared to children unexposed and uninfected with HIV (HUU), the evidence is mixed in children who are HEU. This small descriptive pilot study aimed to compare neurodevelopmental scores of children who are HIV+, HEU, and HUU in Kenya. Methods: This cross-sectional pilot study included children ages 18–36 months who were HIV+, HEU, or HUU. Neurodevelopment was assessed, along with sociodemographic, lab, and growth data. Statistical analysis included descriptive statistics, one-way ANOVA, chi-squared, and adjusted linear regression models. Results: One hundred seventy two were included (n = 24 HIV+; n = 74 HEU; n = 74 HUU). Mothers of children who were HEU experienced more depressive symptoms (p < 0.001). The only neurodevelopmental differences were found among groups was that children who were HIV+ had higher receptive language scores (p = 0.007). Lower height-for-age z-scores and being left home alone were associated with worse neurodevelopmental scores. Conclusions: Being stunted, left completely alone for at least an hour within the last week, and having higher sociodemographic status were associated with worse neurodevelopmental scores. The higher levels of depressive symptoms within mothers of children who are HEU warrants further investigation.
This was a descriptive cross-sectional pilot study assessing neurodevelopment among young children in Kenya as part of the Academic Model Providing Access to Healthcare (AMPATH) consortium. The AMPATH HIV care program represents a 20-year partnership among Indiana University School of Medicine (IUSM), Moi University School of Medicine (MUSM), and the Moi Teaching and Referral Hospital (MTRH) in Eldoret, Kenya. AMPATH has enrolled over 200,000 patients and currently provides care for approximately 15,000 children who are HEU and HIV+ in 65 clinics in western Kenya (9, 10). The children and their caregivers were recruited from the MTRH and AMPATH pediatric HIV clinics between 12/2017 and 9/2019, where care for prevention of mother-to-child transmission (PMTCT) of HIV has been freely provided since 2003 (11). Mixed convenience sampling of potential participants occurred within these clinics, with periods of consecutive sampling by study team members positioned within the clinic along with the clinic staff notifying the study team of eligible participants when the team was not physically present. Inclusion criteria included: (1) child between 18 and 36 months of age; (2) Kiswahili or English are the primary languages; (3) attending the MTRH maternal-child health (MCH) clinic or AMPATH HIV clinics; and (4) primary caregiver ≥18 years of age. Children aged 18–36 months are the population of interest in this study for the following reasons: (1) The young age ( ≤3 years) promotes early referral to intervention services during a critical period of child neurodevelopment, when intervention is most cost- and time-effective; (2) With 18 months as the lower age limit, cognitive and language domains may be tested with greater rigor and persisting delays will be more perceptible, while still allowing adequate time for intervention during this critical period; (3) From a feasibility standpoint, 18–36 months is the upper age limit of children born to mothers living with HIV routinely attending the MCH clinics in Kenya, due to the timing of their final HIV testing. We aimed to recruit 75 participants from each group (HEU, HIV+, and HUU), with HIV status obtained from maternal report. In total, 187 children were recruited. Twelve children could not complete ≥1 sub-sections of the Bayley-3, despite an additional time allowed to rest; two were excluded due to over-recruitment within the HEU cohort; and one did not return after consenting, resulting in 172 remaining study participants. Perinatal data were not collected from participants. However, standard PMTCT guidelines at this time included an efavirenz-based first-line ART for adults living with HIV ≥15 years of age (12, 13). The Bayley Scales of Infant and Toddler Development, 3rd edition (Bayley-3) is a neurodevelopment assessment that is internationally known and commonly used in research settings (1, 14). Our research team conducted a psychometric analysis of this culturally-adapted Bayley-3 within this setting in Kenya and found it to be valid and internally reliable (15). The cognitive, language (receptive/expressive), and motor (fine/gross) domains of the Bayley-3 were selected for use within this study, as they were the domains which had been culturally adapted. The Social-Emotional and Adaptive Behavior domains were not adapted due to the cross-cultural challenges in interpreting appropriate emotional status and behavior when compared to the normative population. One research assistant received training and approval to administer the Bayley-3 by a certified trainer and was blinded to study participants' HIV status prior to Bayley-3 administration. Caregivers completed a questionnaire containing demographic information and basic medical history, including birth history and HIV status. While clinical staff aided in identifying potentially eligible study participants, the HIV status was self-reported by caregivers. The caregiver also provided information on depressive symptoms using the Patient Health Questionnaire (PHQ-9) (16), UNICEF multiple cluster survey questions related to child stimulation (17), and socioeconomic status (SES), measured using the Wealth-Assets-Maternal Education-Income (WAMI index) (18). Anthropometric measurements were obtained at study staff. Each child's body mass and standing height was measured twice and if a discrepancy existed between the two measurements, the average was taken. Standing height was obtained in children <24 months to optimize measurement precision within the cohort. Weight was recorded to 0.1 kg and height was recorded to the 5 mm. Scale used: Salter 9,028 Razor Ultra Slim Technology Electronic Scale. Height measured with Seca 213 Portable Stadiometer Height-Rod. A blood sample was taken at enrollment to measure iron-deficiency anemia (IDA) using hemoglobin and serum ferritin. While the primary data regarding IDA is presented elsewhere (19), this variable was included within the regression analysis, as it is a known risk factor for worse neurodevelopmental outcomes (20). We classified a child as having IDA if (1) hemoglobin concentration was <118 g/L, based on World Health Organization-published algorithm for individuals living ≥2,000 meters in elevation, consist with local elevation (16); and (2) ferritin concentrations <12 μg/L. Study data were managed using REDCap, hosted at IUSM (21). At the time of this study's inception, data were limited regarding Bayley-3 scores within an international setting with children who were HEU and HUU. We used differences in rates of developmental delay among children who were HIV+ and HEU (22) to perform sample size calculations. With a margin of error of 5 points and 95% confidence interval, a sample size of 72 individuals per group was needed to determine a difference among groups. We aimed to recruit 75 per group to allow for a number who may not complete all study activities. Characteristics of children and caregivers were summarized for the total cohort and by HIV status. For continuous variables, the mean and standard deviation (SD) were reported, and differences among HIV groups were assessed using one-way ANOVA. For categorical variables, the frequency and percentage were presented, and proportions were compared using the chi-square test. Anthropometric measurements (height and weight) were converted to height-for-age z-scores, weight-for-age z-scores, and weight-for-height z-scores using the modeling defined by the WHO (23). Multivariable linear regression models were used to examine the associations between scores from neurodevelopmental subtests and predictors. Potential predictors were identified using two methods: (1) a priori using clinical judgement and (2) when corresponding p-values of 0.05 were found among the three groups (HIV+, HEU, and HUU) within univariable analyses. Main effect estimates, standard errors (SE), 95% confidence intervals (CI), and significance tests for the predictors were produced. All caregivers provided written informed consent for their child's participation in this study. The study was approved by Institutional Review Boards at IUSM in Indianapolis, USA and MUSM in Eldoret, Kenya.
N/A
