A New Nomogram for Predicting 30-Day In-Hospital Mortality Rate of Acute Cholangitis Patients in the Intensive Care Unit

Abstract

Purpose. Acute cholangitis (AC) is a widespread acute inflammatory disease and the main cause of septic shock, which has a high death rate in hospitals. At present, the prediction models for short-term mortality of AC patients are still not ideal. We aimed at developing a new model that could forecast the short-term mortality rate of AC patients. Methods. Data were extracted from the Medical Information Mart for Intensive Care IV version 2.0 (MIMIC-IV v2.0). There were a total of 506 cases of AC patients that were included. Patients were given a 7 : 3 split between the training set and the validation set after being randomly assigned to one of the groups. Multivariate logistic regression was used to create an AC patient predictive nomogram for 30-day mortality. The overall efficacy of the model is evaluated using the area under the receiver operating characteristic curve (AUC), the calibration curve, the net reclassification improvement (NRI), the integrated discrimination improvement (IDI), and a decision curve analysis (DCA). Results. Out of 506 patients, 14.0% (71 patients) died. The training cohort had 354 patients, and the validation cohort had 152 patients. GCS, SPO2, albumin, AST/ALT, glucose, potassium, PTT, and peripheral vascular disease were the independent risk factors according to the multivariate analysis results. The newly established nomogram had better prediction performance than other common scoring systems (such as SOFA, OASIS, and SAPS II). For two cohorts, the calibration curve demonstrated coherence between the nomogram and the ideal observation $\left(P>0.05\right)$ . The clinical utility of the nomogram in both sets was revealed by decision curve analysis. Conclusion. The novel prognostic model was effective in forecasting the 30-day mortality rate for acute cholangitis patients.