A model for predicting return of spontaneous circulation and neurological outcomes in adults after in-hospital cardiac arrest: development and evaluation

Abstract

Introduction: In-hospital CA (IHCA) is associated with rates of high incidence, low return of spontaneous circulation (ROSC), low survival to discharge, and poor neurological outcomes. We aimed to construct and evaluate prediction models for non-return of spontaneous circulation (non-ROSC) and poor neurological outcomes 12 months after ROSC (PNO-12). Methods: We retrospectively analyzed baseline and clinical data from patients experiencing cardiac arrest (CA) in a big academic hospital of Jilin University in China. Patients experiencing CA between September 1, 2019 and December 31, 2020 were categorized into the ROSC and non-ROSC groups. Patients maintaining ROSC >20 min were divided into the good and PNO-12 subgroups. Results: Univariate and multivariate logistic regression identified independent factors associated with non-ROSC and PNO-12. Two nomogram prediction models were constructed and evaluated. Of 2,129 patients with IHCA, 851 were included in the study. Multivariate logistic regression analysis revealed that male sex, age >80 years, CPR duration >23 min, and total dose of adrenaline >3 mg were significant risk factors for non-ROSC. Before CA, combined arrhythmia, initial defibrillation rhythm, and advanced airway management (mainly as endotracheal intubation) also influenced outcomes. The area under the receiver operating characteristic curve in the prediction model was 0.904 (C-index: 0.901). Respiratory failure, shock, CA in the monitoring area, advanced airway management, and noradrenaline administration were independent risk factors for PNO-12. The AUC was 0.912 (C-index: 0.918). Conclusions: Prediction models based on IHCA data could be helpful to reduce mortality rates and improve prognosis.