Positive pressure ventilation at birth and potential pathways to newborn deaths in rural Tanzania

Abstract

Background: There are 2.6 million neonatal deaths that occur globally each year, with more than 80% of these deaths occurring in low-income countries. In Tanzania, available estimates report that approximately 40,000 newborn deaths occur each year, mainly due to intrapartumrelated causes, prematurity-related complications, and sepsis. The majority of intrapartum-related neonatal deaths can be avoided by improving care around births. Interventions that have the potential to reduce intrapartum-related neonatal deaths include foetal monitoring during labour, availability of emergency obstetric care, and newborn resuscitation at birth for non-breathing newborns. Low-income countries are faced with many challenges in providing this care, including unskilled providers and inadequate training strategies that do not support the acquirement and retention of skills in newborn resuscitation. Aims: The overall aim of this thesis was to investigate the causes of early newborn deaths and the contribution of intrapartum-related events and their association with ventilation immediately after birth. Furthermore, we wanted to describe the human factors and interactions that influence effective newborn resuscitation practices in this rural setting. Methods: We applied a mixed-methods design and conducted three studies from October 2014 to July 2017. An observational study of all admitted newborns, delivered at Haydom Lutheran Hospital (n=671) between October 2014 and July 2017, was conducted to determine the presumed causes of 7-day newborn deaths and potential pathways contributing to death in this setting (Study I). A study that included the admitted newborns who received positive pressure ventilation in the delivery room (n=232) between October 2014 and November 2016 was then performed to compare ventilation characteristics with the newborn outcome at 7 days (Study II). Infants who died within the first 30 minutes of birth were excluded from both Studies I and II because they died in the delivery room. Building on the findings of the quantitative studies, a third study was conducted, consisting of in-depth interviews with midwives who performed deliveries and newborn resuscitations at Haydom Lutheran Hospital to explore factors affecting the provision of effective ventilation during newborn resuscitation (Study III). Results: In Study I, intrapartum-related complications (birth asphyxia and meconium aspiration syndrome) contributed to almost two-thirds of all deaths within 7 days. Prematurity, presumed sepsis, and congenital abnormalities were other causes of death. Intrapartum hypoxia and prematurity were the major pathways leading to death. Severe hypoxia and hypothermia upon admission were important additional contributing factors. In Study II, we showed that depressed newborns at birth who eventually died within 7 days had an abnormal foetal heart rate during labour, presented signs of bradycardia immediately after birth, and had delayed heart rate responses to positive pressure ventilation. Abnormal foetal heart rate during labour, heart rate at the end of positive pressure ventilation, and duration of positive pressure ventilation were the perinatal predictors of death in this setting. These newborns developed seizures and moderate/severe encephalopathy, likely related to intrapartum hypoxia. Despite inconsistencies in adhering to the Helping Babies Breathe algorithm, the tidal volume and heart rate responses that were recorded did not significantly influence the outcome of death or survival. In Study III, midwives reported the importance of monitoring labour and being prepared for resuscitation before delivery. They also cited good teamwork and frequent ventilation training as factors to facilitate effective ventilation. Barriers to effective ventilation were identified as being anxious and/or feeling fear during ventilation, and difficulties in assessing clinical responses during ventilation. Conclusion: The findings in this PhD thesis demonstrate the contribution of intrapartum-related neonatal deaths to early newborn mortality in a rural sub-Saharan setting. Furthermore, the data demonstrate a link between intrapartum events, likely through interrupted placental blood flow, and a state of depression in the foetus at birth, as represented by low heart rate at birth, delayed heart rate responses to positive pressure ventilation, and, eventually, death. Hypothermia and hypoxia during admission likely played a role in increasing mortality. The included studies highlight the potential for improving intrapartum care through enhanced foetal monitoring during labour to identify those at risk, as well as the benefits of optimizing positive pressure ventilation during resuscitation in the delivery room. The latter should be the focus of frequent resuscitation training sessions to address the providers’ uncertainties and inconsistencies during resuscitation. Frequent resuscitation training should build the confidence of providers to quickly assess newborns immediately after birth, and to act without delay in order to optimize the provision of positive pressure ventilation.