Conventional Versus Compression-Only Versus No-Bystander Cardiopulmonary Resuscitation for Pediatric Out-of-Hospital Cardiac Arrest

BACKGROUND: Conventional cardiopulmonary resuscitation (CPR) (chest compression and rescue breathing) has been recommended for pediatric out-of-hospital cardiac arrest (OHCA) because of the asphyxial nature of the majority of pediatric cardiac arrest events. However, the clinical effectiveness of additional rescue breathing (conventional CPR) compared with compression-only CPR in children is uncertain. METHODS: This nationwide population-based study of pediatric OHCA patients was based on data from the All-Japan Utstein Registry. We included all pediatric patients who experienced OHCA in Japan from January 1, 2011, to December 31, 2012. The primary outcome was a favorable neurological state 1 month after OHCA defined as a Glasgow-Pittsburgh Cerebral Performance Category score of 1 to 2 (corresponding to a Pediatric Cerebral Performance Category score of 1-3). Outcomes were compared with logistic regression with uni- and multivariable modeling in the overall cohort and for a propensity-matched subset of patients. RESULTS: A total of 2157 patients were included; 417 received conventional CPR, 733 received compression-only CPR, and 1007 did not receive any bystander CPR. Among these patients, 213 (9.9%) survived with a favorable neurological status 1 month after OHCA, including 108/417 (25.9%) for conventional, 68/733 (9.3%) for compression-only, and 37/1007 (3.7%) for no-bystander CPR. In unadjusted analyses, conventional CPR was superior to compression-only CPR in neurologically favorable survival (odds ratio [OR] 3.42, 95% confidence interval [CI] 2.45-4.76; P<0.0001), with a trend favoring conventional CPR that was no longer statistically significant after multivariable adjustment (ORadjusted 1.52, 95% CI 0.93-2.49), and with further attenuation of the difference in a propensity-matched subset (OR 1.20, 95% CI 0.81-1.77). Both conventional and compression-only CPR were associated with higher odds for neurologically favorable survival compared with no-bystander CPR (ORadjusted 5.01, 95% CI 2.98-8.57, and ORadjusted 3.29, 95% CI 1.93-5.71), respectively. CONCLUSIONS: In this population-based study of pediatric OHCA in Japan, both conventional and compression-only CPR were associated with superior outcomes compared with no-bystander CPR. Unadjusted outcomes with conventional CPR were superior to compression-only CPR, with the magnitude of difference attenuated and no longer statistically significant after statistical adjustments. These findings support randomized clinical trials comparing conventional versus compression-only CPR in children, with conventional CPR preferred until such controlled comparative data are available, and either method preferred over no-bystander CPR.