Effects of emollient therapy with sunflower seed oil on neonatal growth and morbidity in Uttar Pradesh, India: a cluster-randomized, open-label, controlled trial

Abstract

ABSTRACTBackgroundNewborn oil massage is a widespread practice. Vigorous massage with potentially harmful products and forced removal of vernix may disrupt skin barrier integrity. Hospitalized, very-preterm infants treated with sunflower seed oil (SSO) have demonstrated improved growth but community-based data on growth and health outcomes are lacking.ObjectivesWe aimed to test whether SSO therapy enhances neonatal growth and reduces morbidity at the population level.MethodsWe conducted an open-label, controlled trial in rural Uttar Pradesh, India, randomly allocating 276 village clusters equally to comparison (usual care) and intervention comprised of promotion of improved massage practices exclusively with SSO, using intention-to-treat and per-protocol mixed-effects regression analysis.ResultsWe enrolled 13,478 and 13,109 newborn infants in demographically similar intervention and comparison arms, respectively. Adherence to exclusive SSO increased from 22.6% of intervention infants enrolled in the first study quartile to 37.2% in the last quartile. Intervention infants gained significantly more weight, by 0.94 g · kg−1 · d−1 (95% CI: 0.07, 1.82 g · kg−1 · d−1, P = 0.03), than comparison infants by intention-to-treat analysis. Restricted cubic spline regression revealed the largest benefits in weight gain (2–4 g · kg−1 · d−1) occurred in infants weighing <2000 g at birth. Weight gain in intervention infants was higher by 1.31 g · kg−1 · d−1 (95% CI: 0.17, 2.46 g · kg−1 · d−1; P = 0.02) by per-protocol analysis. Morbidities were similar by intention-to-treat analysis but in per-protocol analysis rates of hospitalization and of any illness were reduced by 36% (OR: 0.64; 95% CI: 0.44, 0.94; P = 0.02) and 44% (OR: 0.56; 95% CI: 0.40, 0.77; P < 0.001), respectively, in treated infants.ConclusionsSSO therapy improved neonatal growth, and reduced morbidities when applied exclusively, across the facility–community continuum of care at the population level. Further research is needed to improve demand for recommended therapy inside hospital as well as in community settings, and to confirm these results in other settings.This trial was registered at www.isrctn.com as ISRCTN38965585 and http://ctri.nic.in as CTRI/2014/12/005282.