O2044: Heart-rate Guided Resuscitation with Helping Babies Breathe: Neonatal Outcomes and Resuscitation Practices in the Democratic Republic of the Congo

Background: Ventilation of the neonatal lungs is critical to cardiorespiratory transition at birth. Heart rate (HR) is a sensitive indicator of the adequacy of spontaneous respiratory effort and thus need for ventilation, as well as the response to resuscitative interventions. Lack of guidance from HR monitoring during resuscitation may contribute to delayed and ineffective ventilation, and worse outcomes. In a clinical trial with midwives in the Democratic Republic of the Congo (DRC), we evaluated the impact of training in HR-guided resuscitation compared to Helping Babies Breathe (HBB) on neonatal outcomes and resuscitation practices. We hypothesized that HR-guided resuscitation would increase the proportion of newborns effectively breathing at three minutes by 50% compared to HBB.

Methods: We conducted a clinical trial of resuscitation training with continuous electronic HR monitoring in three urban health facilities in the DRC, enrolling all in-born neonates ≥28 weeks gestation (ClinicalTrials.gov: NCT03799861). We developed a HR-guided basic resuscitation training program by adapting the HBB materials to include assessment of HR to prompt both initiation and improvement of ventilation. During the trial, providers implemented HBB with continuous electronic HR monitoring followed by HR-guided resuscitation. We abstracted demographic/outcome data from delivery registers, and observed a convenience sample of resuscitations to document provider practices, newborn breathing status and continuous HR. We evaluated our primary outcome of effective breathing at three minutes after birth among newborns not breathing well at 30 seconds after birth using a method of least squares to fit general linear models.

Results: Among the 1,284 newborns with observational data, there was no difference in the primary outcome (adjusted Relative Risk [aRR] 1.08 [95% Confidence Interval {CI} 0.81, 1.45]), nor the secondary outcomes of time to effective breathing and HR≥100 at three minutes (Table 1). Among 145 receiving bag-mask ventilation (BMV), there was no change in time to effective breathing (p=0.122; Table 2) but time to BMV decreased 64.3 seconds (s) during HR-guided resuscitation (p < 0.001). Providers initiated both drying/stimulation and suctioning earlier during HR-guided resuscitation (-6.3s, p=0.009 and -27.2s, p=0.03, respectively). Among 10,906 enrolled in the trial, perinatal mortality was unchanged (aRR 1.19 [95%CI 0.96, 1.48]) and death before discharge increased (aRR 1.43 [95% CI 1.03, 1.99]). Expert review of stillborn cases demonstrated a misclassification rate of 33.3% during HBB with continuous HR monitoring (n=6 of 18 cases) compared to 5.9% in HR-guided resuscitation (n=1 of 17).

Conclusion: HR-guided resuscitation reduced time to BMV by greater than one minute compared to HBB. The increase in death before discharge and unchanged perinatal mortality may be due to resuscitation of newborns with a higher risk of mortality who were previously presumed stillborn. A cluster-randomized trial of HR-guided resuscitation is needed to evaluate its impact on neonatal mortality in low-resource settings.