Ventral hernia repair in emergency settings. A machine learning model to predict post-operative complications.

Clinical Snapshot

Key Findings

• P Value: Not reported in abstract

• Effect Size: Random Forest AUC 0.95, accuracy 0.88, F1 score 0.86 vs. logistic regression AUC 0.82, accuracy 0.78

• Primary Outcome: Postoperative complications occurred in 181 of 557 patients (32.5%); major complications (Clavien-Dindo ≥ II) in 10%

• Nnt Or Sensitivity: Sensitivity and specificity not reported separately in the abstract; F1 score of 0.86 for Random Forest implies reasonable balance of precision and recall but individual values are not stated

• Confidence Interval: Not reported for any model performance metric

Clinical Application

The model requires prospective validation and software implementation before clinical deployment. Predictor variables (ASA score, operative duration, BMI, presence of sepsis, bowel obstruction) are routinely available in emergency surgical settings, suggesting feasibility of data capture. However, operative duration is only known post-hoc, limiting true preoperative risk stratification utility for that variable. Emergency ventral hernia repair is managed within Australian public hospital emergency surgical services and is relevant to general surgeons, acute care surgeons, and anaesthetists. The RACGP and RACS do not currently endorse ML-based perioperative risk tools for this indication. Existing risk stratification tools such as the American College of Surgeons NSQIP Surgical Risk Calculator are more established. Any future implementation in Australia would require TGA consideration if deployed as a clinical decision support software (SaMD) under the TGA's Software as a Medical Device framework. PBS implications are indirect but relevant if ML-guided triage influences resource allocation and length of stay. The Italian-specific dataset limits direct applicability without Australian validation data. Adult patients presenting with acute ventral hernia requiring emergency surgical repair, particularly those with elevated ASA classification, sepsis, anticipated prolonged operative duration, bowel obstruction, or elevated BMI

Abstract

BACKGROUND: Emergency ventral hernia repair remains a challenging procedure due to patient instability, contaminated surgical fields, and heterogeneity in hernia types and operative techniques. Predicting postoperative complications in this setting is difficult using traditional statistical methods. Machine learning (ML) may offer improved predictive accuracy by recognizing nonlinear patterns among multiple perioperative factors. METHODS: A retrospective multicenter analysis was performed using data from the ACTIVE (Acute Treatment for Incisional Ventral Hernias) study, including 557 adult patients undergoing emergent ventral hernia repair between 2018 and 2021 in 31 Italian surgical centers. Demographic, preoperative, intraoperative, and postoperative variables were analyzed. Three ML algorithms-Decision Tree, Random Forest, and Deep Learning Neural Network-were trained and validated using five-fold cross-validation after class balancing with SMOTE. Model performance was compared with traditional logistic regression using accuracy, area under the ROC curve (AUC), and F1 score. RESULTS: Postoperative complications occurred in 181 patients (32.5%), while major complications (Clavien-Dindo ≥ II) occurred in 10%. Random Forest achieved the best performance (AUC 0.95, accuracy 0.88, F1 score 0.86), outperforming logistic regression (AUC 0.82, accuracy 0.78). The most influential predictors were operative duration, ASA score, and sepsis for overall complications, while bowel obstruction and BMI were key factors for major complications. Surgical approach (open vs. laparoscopic) did not independently correlate with adverse outcomes, highlighting the complexity of patient- and case-specific interactions. CONCLUSIONS: Machine learning models can accurately predict postoperative complications following emergent ventral hernia repair, surpassing traditional regression methods. These findings suggest that ML-based decision tools could support risk stratification and optimize surgical planning in high-risk emergency settings. Prospective validation is warranted to integrate AI-assisted prediction into perioperative clinical workflows.

References

1. 1.Ortenzi, M., Crepaz, L., Anania, G., Balla, A., Morales-Conde, S., Mastronardi, M., Arezzo, A., Podda, M., & Azzolina, D. (2026). Ventral hernia repair in emergency settings. A machine learning model to predict post-operative complications. Minimally Invasive Therapy & Allied Technologies, advance online publication. https://doi.org/10.1080/13645706.2026.2665459