BackgroundPediatric out-of-hospital cardiac arrests (OHCAs) are frequently associated with a respiratory etiology. Despite the high proportion of preschool children with OHCAs, very few studies on this special population exist. This study characterizes the epidemiologic features of preschool pediatric OHCAs and analyzes the advantage of conventional (ventilations with chest compressions) bystander cardiopulmonary resuscitation (CPR) over compression-only bystander CPR (BCPR) on the one-month post-event neurological status of the patient.MethodsJapanese nationwide databases for all ambulance transport events and OHCAs occurring during a 4-year period between 2016 and 2019 were combined, totalling 3,608 patient events. Children ≤6-years-old were included; physician- and EMS-witnessed events, no prehospital resuscitation effort events, and neonatal patient events were excluded. Neurologically favorable 1-month survival rates were compared among groups using univariate and multivariate analyses before and after propensity score matching.ResultsFrom the combined database, 2,882 pediatric OHCAs meeting selection criteria were categorized as no BCPR (984), compression-only BCPR (1,428), and conventional BCPR (470). The proportion of bystander-witnessed cases was low (22.3%). Most OHCA witnesses were family members (88.5%), and most OHCAs occurred at home (88.0%). The neurologically favorable 1-month survival rates were: no BCPR 2.4%, compression only, 3.2%, and conventional 6.6% (P < 0.01). Multivariate logistic regression analysis before and after matching showed that conventional BCPR was associated with higher neurologically favorable 1-month survival than compression-only BCPR. Subgroup analyses after matching demonstrated that conventional BCPR was associated with better outcomes in nonmedical (adjusted odds ratio; 95% confidence interval, 2.83; 1.09–7.32) and unwitnessed OHCA cases (3.42; 1.09–10.8).ConclusionsConventional CPR is rarely performed by bystanders in preschool pediatric OHCA. However, conventional BCPR results in neurologically favorable outcomes in nonmedical and unwitnessed cases.