External validation of the IHXGboost-P model to predict incisional hernia after midline laparotomy.
Journal:
Hernia : the journal of hernias and abdominal wall surgery
Published Date:
Jan 16, 2026
Abstract
BACKGROUND: Incisional hernia (IH) is a significant complication that occurs after midline laparotomy and is associated with high morbidity and economic impacts. A fundamental goal of preventing IH is to determine which patients are considered low- or high-risk, as modifications in prevention techniques have been justified in high-risk patients. AIM: of this study was to externally validate the IHXGBoost-P model to assess its accuracy, generalizability, and clinical applicability in an independent cohort. METHODS: A prospective cohort study was conducted in a tertiary hospital in Mexico (March 2021-December 2022) to externally validate the IHXGBoost-P model. Patients older than 18 years who underwent midline laparotomy and have a minimum follow-up of 24 months were included. The performance of the model was evaluated via area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, precision, specificity and calibration metrics. RESULTS: Of the 438 patients analyzed, 62 (14.1%) developed IH. The model demonstrated good discriminative capacity (Accuracy: 0.94 ± 0.015) and calibration (Brier score: 0.051). Key predictors included the risk of surgical site infection (odds ratio (OR): 3.01, 95% CI: 2.32-3.91), previous surgery, and body mass index (BMI). The specificity (0.97 ± 0.013) was determined to be high and useful for identifying lowrisk patients. CONCLUSIONS: The IHXGBoost-P model is a reliable tool for predicting the risk of IH, with robust performance being observed in external validation. Its integration into clinical practice through a web application could optimize surgical decision-making to prevent IH.
