Development and validation of a dynamic nomogram for predicting in-hospital mortality in acute massive cerebral infarction: a retrospective study in a Chinese population.

Abstract

Background: Massive cerebral infarction (MCI) is a severe form of ischemic stroke that can result in adverse outcomes, including death. This study aimed to identify the independent risk factors associated with MCI mortality by developing a multivariate model using stepwise logistic regression analysis.

Methods: This retrospective study included 159 hospitalized patients between January 15, 2022, and October 20, 2023. The diagnosis of MCI was based on clinical symptoms, the National Institutes of Health Stroke Scale (NIHSS), the Glasgow Coma Scale (GCS), and brain MRI. Potential mortality-related predictors were identified by analyzing patient histories, coagulation profiles, renal function, and serum biochemical indicators such as fasting blood glucose (FBG), homocysteine (HCY), and hemoglobin (Hb).

Results: Among the 159 patients, optimized multivariate logistic regression analysis revealed that smoking (OR = 10.48, 95% CI 2.85-42.80), FBG (OR = 1.97, 95% CI 1.45-2.82), HCY (OR = 8.62, 95% CI 1.29-76.21), Hb (OR = 0.96, 95% CI 0.94-0.99), and GCS score (OR = 0.67, 95% CI 0.52-0.83) were significantly associated with in-hospital mortality (all P < 0.05). The model showed good discrimination (AUC = 0.943, 95% CI 0.903-0.982), with a marginal R-squared (R²M) of 0.660. Calibration and decision curve analyses suggested good predictive performance and potential clinical utility of the nomogram.

Conclusion: Smoking, elevated FBG and HCY, low Hb, and lower GCS scores were identified as independent predictors of mortality in MCI patients. Managing these factors may help reduce the risk of death.