The Enigma of Incisional Hernia Prediction Unravelled: External Validation of a Prognostic Model in Colorectal Cancer Patients

Aim Incisional hernia is the most common complication of abdominal surgery. Mesh-augmented closure appears to reduce incisional hernia rates in targeted patients, but identifying high-risk patients remains challenging. This study aims to validate the existing “Penn Hernia Calculator” on an external dataset of patients who underwent colorectal surgery, evaluating its performance and potential for generalisability. Methods External validation of the model was achieved using data from the Hughes Abdominal Repair Trial (HART), a prospective randomized control trial comparing closure methods in colorectal cancer patients undergoing abdominal surgery. The previously developed hernia prediction model was applied to the external dataset. Model performance was assessed using key metrics, including area under the curve (AUC-ROC and AUC-PR) and Brier score. Reporting was done in line with TRIPOD consensus. Results 802 patients who underwent colorectal surgery was identified, of which 674 patients with up to two years follow up were included. The average BMI was 28.1. Prevalence of diabetes, hypertension, and smoking were 15.7%, 16.3%, and 36.5%, respectively. Hernia occurrence rate up to a two-year follow-up was 24%. The external validation revealed an AUC-ROC of 0.66, AUC-PR of 0.72, and a Brier score of 0.2. Conclusion The hernia prediction model demonstrated moderate performance in the external validation, suggesting its potential for generalizability in diverse patient populations, specifically those undergoing colorectal surgery. The model may assist clinicians in identifying high-risk patients and implementing preventive measures to reduce hernia incidence.