Menu
Menu
Menu
Menu
Background: High rates of adverse pregnancy outcomes globally raise the need to understand risk factors and develop preventative interventions. The Pregnancy Outcomes in the Era of Universal Antiretroviral Treatment in Sub-Saharan Africa (POISE Study) was a prospective, observational cohort study conducted from 2016 to 2017 in Blantyre, Malawi. We examine the associations between indicators of nutritional status, specifically mid-thigh circumference (MTC) and body-mass index (BMI), and adverse pregnancy outcomes, low birth weight (LBW), preterm birth (PTB), and small-for-gestational age (SGA), in a cohort of HIV-infected and HIV-uninfected women. Methods: Sociodemographic, clinical, laboratory, and maternal height, weight and MTC data were collected immediately before or after delivery at the Queen Elizabeth Central Hospital (QEHC) and 4 affiliated health centers in Blantyre, Malawi. LBW was defined as birth weight < 2.5 kg; PTB as gestational age < 37 weeks using Ballard score; and SGA as birth weight < 10th percentile for gestational age. Descriptive, stratified, and multivariable logistic regression were conducted using R. Results: Data from 1298 women were analyzed: 614 HIV-infected and 684 HIV-uninfected. MTC was inversely associated with LBW (adjusted odds ratio [aOR] = 0.95, p = 0.03) and PTB (aOR 0.92, p < 0.001), after controlling for HIV status, age, socioeconomic status and hemoglobin. The association between MTC and SGA was (aOR 0.99, p = 0.53). Similarly, higher BMI was significantly associated with lower odds of PTB (aOR 0.90, p < 0.001), LBW (aOR 0.93, p = 0.05), and SGA (aOR 0.95, p = 0.04). Conclusions: We observed an inverse relationship between MTC and adverse pregnancy outcomes in Malawi irrespective of HIV infection. MTC performs comparably to BMI; the ease of measuring MTC could make it a practical tool in resource-constrained settings for identification of women at risk of adverse pregnancy outcomes.
The parent POISE study was conducted at the Queen Elizabeth Central Hospital (QEHC) and 4 affiliated health centers in Blantyre, Malawi from January 2016 to September 2017. At these health facilities, eligible women were screened and enrolled in the study at delivery. Full details of the study have been published [12]. Briefly, POISE was a prospective observational study. Women were enrolled at the five health facilities at time of delivery (before or after delivery) and followed for one year post-delivery with their infants. Inclusion criteria were the following: confirmed HIV status, written informed consent and live singleton births. Participants were excluded if they were unable to provide informed consent or had multiple births. HIV-infected women were eligible for the study if they had a CD4 cell count ≥ 350 cells/mm3, were on ART for at least 1 week before delivery and did not have WHO stage 3 or 4 HIV disease. The parent study aimed to assess the impact of ART among clinically healthy HIV-infected women (i.e., they did have low CD4 cell count or stage 3 or 4 HIV disease stage). Participants were counseled and consented to enroll along with their infants. Eligible HIV-uninfected women were concurrently enrolled in the same health facilities at which the HIV-infected women were enrolled. Trained study nurses administered structured questionnaires and conducted physical examinations after obtaining informed consent at delivery. The questionnaires collected sociodemographic characteristics, medical history, and sexual and reproductive health information, which included information on risk factors and potential confounders. For HIV-infected women, additional information on ART use and adherence was included and blood samples were collected to measure the HIV viral load and CD4 cell count. The physical examination of all women following delivery included anthropometric measurements of height, weight, and MTC, which were the primary anthropometric indicators in this study. Weight was measured to the nearest 0.1 kg; height and MTC were measured to the nearest 0.5 cm. Body mass index (BMI) was calculated as kg/m2 using the height and weight measurements. Measurement of MTC was conducted by trained study workers on the right thigh while the woman was standing using a flexible measuring tape. The mid-point between the top of the femur and the knee was marked and MTC was determined in centimeters. Based on previous research conducted in Malawi, women who reported having electricity in the home were identified as having high socioeconomic status and women without electricity in the home were identified as having low socioeconomic status [12]. After birth, a physical examination of the infant was conducted, which included birth weight and other anthropometric measurements. Gestational age was calculated using the Ballard score within 36 h of birth by a trained study nurse. Gestational age was estimated using the date of last menstrual period when the Ballard score was missing. Estimates based on Ballard score and date of last menstrual period were compared to assess potential misclassifications [13]. Data were entered using a database deisgned in Microsoft Access and data cleaning was done in a regular interval in Microsoft Excel and Stata (version 14.2). Data double entry was done and anonymized prior to analysis. The primary outcomes of this study were PTB, LBW, and SGA. PTB was defined as a gestational age < 37 completed weeks. LBW was defined as birth weight < 2.5 kg. SGA was defined as birth weight less than the 10th percentile for gestational age, using the reference population as described by Oken et. al [14]. Descriptive analyses were conducted first on the sociodemographic and clinical data and primary outcomes of interest. Stratified analyses, graphical presentations, and data transformations were also used. We first assessed the associations between MTC and the outcomes of birth weight and gestational age as continuous variables using linear regression models. We modeled the associations between MTC and the three pregnancy outcomes (PTB, LBW, and SGA as binary outcomes) using univariable and multivariable logistic regression analyses, which controlled for potential cofounders. Univariable and multivariable models were also created to examine the associations between BMI and the three pregnancy outcomes. The following baseline risk factors were controlled for in the multivariable analyses: HIV status (infected/uninfected; all HIV-infected women were taking the standard national ART regimen in Malawi at the time of study conduct [tenofovir, lamivudine and efavirenz]), maternal age in years (continuous), hemoglobin level (< 10 g/dL/ ≥ 10 g/dL) and socioeconomic status measured as having electricity at home (yes/no). These were selected based on the risk factors assessed in the parent study, which were of biological and epidemiological importance [12]. The final multivariable model included these potential confounders in addition to the exposure of interest– MTC. The analyses were run at a p < 0.05 significance level, and crude and adjusted odds ratios with 95% confidence intervals are presented. All analyses were done in R statistical software package version 3.6.3 (Vienna, Austria).
N/A
