Predicting postoperative mortality after colorectal surgery: a novel clinical model

Area of Science:

• Colorectal Surgery
• Surgical Outcomes
• Clinical Prediction Models

Background:

• In-hospital mortality following colorectal surgery poses a significant clinical challenge.
• Accurate prediction of mortality risk is crucial for patient management and resource allocation.
• Existing prediction models may lack clinical practicality or discriminative power.

Purpose of the Study:

• To develop a clinically practical and discriminative prediction model for in-hospital mortality in colorectal surgery patients.
• To externally validate the developed model in an independent patient cohort.
• To construct a simplified scoring system for ease of clinical application.

Main Methods:

• Multivariate logistic regression analysis was employed to identify predictors of in-hospital mortality.
• A simplified Identification of Risk in Colorectal Surgery (IRCS) score was constructed using significant predictors.
• The model's predictive performance was assessed using the area under the receiver operating characteristic curve (AUC) in a validation cohort.

Main Results:

• Key predictors of in-hospital mortality included emergency surgery, tumor stage, age, pulmonary failure, and cardiac failure.
• The IRCS score demonstrated strong predictive performance with an AUC of 0.83 in the validation population.
• The IRCS score outperformed the CR-POSSUM score (AUC 0.76) in predicting in-hospital mortality.

Conclusions:

• The developed IRCS score is a robust and clinically useful predictor of in-hospital mortality after colorectal surgery.
• The IRCS score offers a practical and discriminative tool for risk stratification in this patient population.
• The model's effectiveness is maintained despite its relatively low number of parameters.