Clinical epidemiology of twin deliveries in The Gambia and Burkina Faso: Secondary analyses of PregnAnZI2 clinical trial data

IntroductionTwins have higher risk of adverse clinical outcomes in all settings, but especially in low- and middle-income countries. This study aims to understand the clinical epidemiology of twins from varied West African settings, focusing on prevalence and determinants of twins, and adverse perinatal outcomes associated with twin deliveries. These context specific insights may help to inform public health and clinical approaches to improving health and survival of this vulnerable newborn group. MethodsWe conducted secondary analyses from the PregnAnZI-2 randomised clinical trial, which included data from pregnant women and their offspring who were delivered at eight peripheral health facilities in rural Burkina Faso and two urban health facilities in The Gambia. Logistic regression was used to identify associations between primary outcome (twin deliveries) and both risk factors and adverse outcomes and was informed by a novel conceptual framework. ResultsData from 12,192 newborns born from 11,983 women were analysed. The average twinning rate was 35 per 1000 total births (95% CI: 32-39): 27 per 1000 total births (95% CI: 23-31) in The Gambia and 45 per 1000 total births (95% CI: 40-51) in Burkina Faso. Risk factors for twinning included maternal ethnicity and parity. Twins had increased risk of low birth weight (OR =17.2, 95% CI:13.90-21.30, P= <0.001), low 5-min Apgar score (proxy for intrapartum related asphyxia) (OR =4.28, 95% CI: 2.79-6.33, P <0.001), intrapartum stillbirth (OR=4.69, 95% CI:2.47-8.22, P<0.001), and hospitalisation during the neonatal period (OR =1.69 95% CI:1.10-2.47, P= 0.011). Neonatal mortality was four-fold higher for twins than singletons (OR =4.23, 95% CI: 2.40-7.50, P <0.001), and among twin deaths, the second twin was more likely to die compared to the first twin (OR=2.46, 95% CI :1.05-7.59, P=0.048). ConclusionTwins are an important contributor to stillbirth and neonatal morbidity and mortality in West Africa. Efforts to identify twin pregnancies early, with optimisation of antenatal and intrapartum management could improve outcomes for this vulnerable newborn group. Trial RegistrationNCT03199547: Clinicaltrials.gov. Registered on 23rd June 2017