The development and validation of a nomogram to determine neurological outcomes in cardiac arrest patients

Abstract

Abstract Objectives This study aimed to investigate the variables that influence neurological functional restoration in cardiac arrest patients and construct a nomogram to predict neurofunctional prognosis. Patients and methods We extracted the data from the Dryad database. Associations between patient variables and neurological outcomes were examined by logistic regression models. On the basis of these predictors, a prognostic nomogram was constructed. The identification and calibration of the prognostic nomogram were evaluated through the receiver operating characteristic (ROC) curve, the calibration curve, and the concordance index (C-index). Results A total of 374 cardiac arrest individuals were recruited in the research. Sixty percent of the participants had an adverse neurological result. The multivariable logistic regression analysis for poor neurological recovery, which showed patient age ≥ 65 years, previous neurological disease, witnessed arrest, bystander cardio-pulmonary resuscitation(CPR), cardiac arrest presenting with a non-shockable rhythm, total epinephrine dose ≥ 2.5 mg at the time of resuscitation and acute kidney injury(AKI) remained independent predictors for neurological outcomes. Conclusions The novel nomogram based on clinical characteristics is an efficient tool to predict neurological outcomes in cardiac arrest patients, which may help clinicians identifying high-risk patients and tailoring personalized treatment regimens.