Global gains after Helping Babies Breathe.

Implementation of basic neonatal resuscitation in lowand middle-income settings consistently results in lives saved on the day of birth. Wrammert et al. explore the causes and timing of newborn deaths in a large maternity facility in Nepal before and after the introduction of basic newborn resuscitation through Helping Babies Breathe (HBB), implemented along with quality improvement cycles aimed at changing providers’ practice (1). The intervention resulted in significant decrease in fresh stillbirth, all-cause mortality within 24 hours of birth and cause-specific mortality due to intrapartum-related events through discharge or 28 days. However, overall mortality to discharge did not significantly change, remaining around 12 per 1000 live births. Deaths from prematurity were postponed, but rates of survival for prematurity, infection and congenital anomalies were unchanged. Reductions in fresh stillbirth and first-day neonatal deaths with resuscitation open the potential for improved long-term survival. However, many implementation trials have reported only immediate outcomes or shown results similar to those in this study, with no change in overall 28-day neonatal mortality or perinatal mortality. Introduction of HBB in eight high-volume Tanzanian hospitals showed a significant reduction in fresh stillbirth (19 to 14.4/1000 births, RR 0.76; 95% CI 0.64–0.98) and deaths in the first 24 hours (13.4 to 7.1/1000 live births, RR 0.53; 95%CI 0.43– 0.65) with corresponding decrease in asphyxia-related mortality (2). Follow-up to 7 and 28 days was not available. Subsequent observations in one of the Tanzanian hospitals substantiate that babies not breathing and not moving at birth are often classified as fresh stillbirths, when in fact a heart rate is present and basic neonatal resuscitation – drying, clearing the airway as needed, specific stimulation to breathe and bag-and-mask ventilationwith air as necessary – effectively reverses apnoea. Babies with deeper cardiorespiratory depression either do not respond or require additional support beyond resuscitation to survive (Ersdal HL personal communication). Implementation of HBB in 12 primary health centres, rural district hospitals and urban hospitals in Karnataka, India showed similar reduction in fresh stillbirth from 17 to 9/1000 births (OR 0.54, 95% CI 0.37–0.78) and unchanged predischarge mortality of 0.1%. A populationbased Maternal and Newborn Health Registry system permitted follow-up through 28 days and showed that neonatal mortality was unchanged (18/1000 live births pre and 19/1000 post, OR 1.09, 95%CI 0.80–1.47). Preterm/low birth weight accounted for the largest proportion of neonatal deaths (47.9 and 42.7% pre and post), followed by birth asphyxia (30.1 and 28.2% pre and post); neonatal deaths attributed to infection rose from 6.6 to 13.6% (p ≤ 0.001). Again, resuscitation training improved recognition that not all infants who fail to breathe are stillborn and reduced predischarge deaths from asphyxia (3). The stable neonatal mortality rate suggested that resuscitation did not simply defer death. A recent pre–post study of scale-up of HBB training in 71 rural and semi-urban facilities in India and Kenya showed a differential effect on stillbirth and perinatal death (fresh stillbirths and neonatal deaths within 7 days) by initial level of mortality. Significant reductions occurred in one geographic cluster of facilities with high mortality both for stillbirths (25.7 to 16.4/1000 births; estimated mean difference 11.27, 95% CI 0.95, 21.59, p = 0.03) and perinatal deaths (38.5 to 28.2/1000 births; estimated mean difference 11.71, 95% CI 0.39, 23.03, p = 0.04). In facilities with lower baseline rates of stillbirth (7.7–11.6/1000 births) and perinatal death (10.0–25.5/1000 births), there was no significant reduction post scale-up of resuscitation (4). In a post hoc analysis stratified by birthweight, the rates for LBW (<2500 g) fresh stillbirths, early neonatal deaths and perinatal deaths did decrease at one of these sites. Thus, basic neonatal resuscitation can reduce the proportion of newborns classified as fresh stillbirths without increasing neonatal mortality. Basic resuscitation consistently decreases deaths in the first 24 hours, but additional measures are needed to sustain those gains. Much as basic neonatal resuscitation can save lives on the day of birth, Essential Newborn Care can help preserve the gains in survival through the first days and weeks of life. All babies need early assessment (weight, temperature and physical exam) todeterminewhether theycan receive routine care or need special support. All babies need protection from hypothermia, adequate feeding and prevention or recognition/treatment of infection. Hypothermia is under-recognised, prevalent in both hot and cold climates, present in facilities and the community and strongly linked to neonatal mortality in a dose–response relationship (5). Coverage of essential newborn care practices, with the exception of breastfeeding, remains very low (under 25%) in the 75 countries monitored in the Countdown to the 2015 for the Millennium Development Goals (6). Implementation of training packages in Essential Newborn Care has produced