Development and validation of a nomogram for predicting 28-day mortality in patients with ischemic stroke

• DOI: 10.1371/journal.pone.0302227
• Publication Date: 2024-04-24T17:25:38Z
• Authors: Lingyan Fang; Menglu Zhou; Fengkai Mao; Mengyuan Diao; Wei Hu; Guangyong Jin
• Keywords: Nomogram; Concordance; Stroke

Background/aim We aimed to construct a validated nomogram model for predicting short-term (28-day) ischemic stroke mortality among critically ill populations. Materials and methods collected raw data from the Medical Information Mart Intensive Care IV database, comprehensive repository renowned its depth breadth in critical care information. Subsequently, rigorous analytical framework was employed, incorporating 10-fold cross-validation procedure ensure robustness reliability. Leveraging advanced statistical methodologies, specifically least absolute shrinkage selection operator regression, variables pertinent 28-day were meticulously screened. Next, binary logistic regression utilized establish nomogram, then applied concordance index evaluate discrimination of prediction models. Predictive performance assessed by integrated improvement (IDI) net reclassification (NRI). Additionally, we generated calibration curves assess calibrating ability. Finally, evaluated nomogram’s clinical benefit using decision curve analysis (DCA), comparison with scoring systems clinically under common conditions. Results A total 2089 individuals identified assigned into training (n = 1443) or validation 646) cohorts. Various risk factors, including age, ethnicity, marital status, underlying metastatic solid tumor, Charlson comorbidity index, heart rate, Glasgow coma scale, glucose concentrations, white blood cells, sodium potassium mechanical ventilation, use heparin mannitol, associated individuals. 0.834 obtained dataset, indicating that our had good discriminating IDI NRI both cohorts proved positive predictive performance, compared other systems. The actual predicted incidence showed favorable on (P > 0.05). DCA revealed that, used conditions, constructed yielded greater benefit. Conclusions Utilizing array fourteen readily accessible variables, prognostic formulated rigorously provide precise prognostication within cohort.