Due to the difficulty in making the decision to discontinue resuscitation efforts, clinicians often perform prolonged cardiopulmonary resuscitation (CPR) in children who have drowned. This study investigated the relationship between out-of-hospital CPR duration and neurological outcomes in children with drowning-induced cardiac arrest.

This retrospective cohort study used data from the All-Japan Utstein Registry from 2013 to 2021. We included patients aged ≤18 years with drowning-induced cardiac arrest who underwent CPR via emergency medical services. The study's primary outcome was a favorable neurological outcome (Cerebral Performance Category score of 1 or 2) at 1 month. We also calculated the dynamic proportion of 1-month outcomes as a function of out-of-hospital CPR duration. Moreover, we determined the sensitivity, specificity, and positive predictive value of 33 min of out-of-hospital CPR, which is the time point at which primary analysis showed a plateau in the neurological outcomes.

Of 14,849 children included in the registry, 992 cases of drowning-induced cardiac arrest were analyzed. The median out-of-hospital CPR duration was 20 (range, 2–164) min, with only 35 (3.5 %) patients showing favorable neurological outcomes at 1 month. The proportion of favorable neurological outcomes decreased rapidly for up to 33 min of out-of-hospital CPR and then plateaued to 0.3 %; only three patients achieved favorable neurological outcomes after 33 min of CPR. An out-of-hospital CPR duration of >33 min was associated with poor neurological outcomes (sensitivity, 0.17 [95 % confidence interval: 0.15–0.20]; specificity, 0.91 [0.77–0.98]; and positive predictive value, 0.98 [0.95–1.00]).

Prehospital EMS-initiated CPR duration for children with drowning-induced cardiac arrest was inversely associated with one-month favorable neurological outcomes. Favorable neurological outcomes after >33 min of out-of-hospital CPR were extremely rare, though accurately predicting the outcome remains challenging.