Etiology-specific predictors for short-term functional outcomes of OHCA: A Japanese Nationwide registry study.
The American journal of emergency medicine · 2026-04-23 · Registry study
Abstract
BACKGROUND: Out-of-hospital cardiac arrest (OHCA) functional outcomes vary widely by etiology, but cerebrovascular-origin OHCA remains poorly characterized. This study aimed to identify etiology-specific predictors of favorable functional outcomes after OHCA using a nationwide Japanese registry. METHODS: We analyzed 392,770 adult OHCA cases (311,632 cardiac-origin and 81,138 cerebrovascular-origin) recorded between 2005 and 2023 using data from the All-Japan Utstein Registry, a prospective nationwide population-based registry. Favorable outcome was defined as CPC 1-2 and OPC 1-2 at 30 days. Demographic, clinical, and time variables were evaluated using multivariable logistic regression. Missing time data were treated as an "unknown" category. A sensitivity analysis using Firth's penalized likelihood regression was performed on subgroups with recorded time data. RESULTS: Favorable outcomes were observed in 9.7% of cardiac-origin and 2.5% of cerebrovascular-origin OHCA. In cardiac-origin OHCA, early defibrillation (<5 min) was a strong predictor. Cerebrovascular OHCA showed a negligible association with age (OR 1.00, 95% CI 0.99-1.00), and defibrillation at 5-10 min was paradoxically associated with better outcomes. Time to ROSC >15 min was consistently associated with poor outcomes in both groups. Crucially, the sensitivity analysis demonstrated that the paradoxical associations were no longer statistically significant after applying Firth's correction. CONCLUSIONS: Determinants of favorable outcomes may differ fundamentally. The findings suggest that prognosis after cerebrovascular-origin OHCA may be largely influenced by the initial severity of cerebral injury, rather than by the timing of prehospital resuscitation efforts. However, given the limitation of registry-based research, prospective studies are warranted to confirm these observations.