Related Experiment Video
Updated: Jan 15, 2026
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Development and validation of a dynamic nomogram for predicting in-hospital mortality in acute massive cerebral
Xuhui Liu1, Xujie Wang2, Rongfei Xie2
1Department of Neurology, The Second Hospital of Lanzhou University, 82 Cuiying Men, Chengguan District, Lanzhou, 730030, Gansu, China.
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.
Insights
Smoking, high blood glucose, elevated homocysteine, low hemoglobin, and low Glasgow Coma Scale scores predict mortality in massive cerebral infarction (MCI) patients. Managing these factors can reduce death risk in stroke patients.
Area of Science:
- Neurology
- Cardiovascular Medicine
- Internal Medicine
Background:
- Massive cerebral infarction (MCI) is a severe ischemic stroke with high mortality.
- Identifying independent risk factors for MCI mortality is crucial for improving patient outcomes.
Purpose of the Study:
- To identify independent risk factors for in-hospital mortality in patients with massive cerebral infarction.
- To develop a predictive model for MCI mortality using logistic regression analysis.
Main Methods:
- Retrospective study of 159 hospitalized MCI patients.
- Data collected included patient history, coagulation, renal function, and biochemical markers (FBG, HCY, Hb).
- National Institutes of Health Stroke Scale (NIHSS) and Glasgow Coma Scale (GCS) scores were used for assessment.
Main Results:
- Smoking (OR=10.48), elevated fasting blood glucose (FBG) (OR=1.97), and elevated homocysteine (HCY) (OR=8.62) were associated with increased mortality.
- Low hemoglobin (Hb) (OR=0.96) and lower GCS scores (OR=0.67) were associated with decreased mortality.
- The multivariate logistic regression model demonstrated good predictive performance (AUC=0.943).
Conclusions:
- Smoking, elevated FBG and HCY, low Hb, and lower GCS scores are independent predictors of mortality in MCI.
- These findings highlight key modifiable and non-modifiable factors influencing MCI outcomes.
- Clinical management targeting these factors may reduce mortality rates in massive cerebral infarction patients.
