Risk factors and predictive model construction for severe complications following repair of perforated peptic ulcer

TL;DR

This study identifies risk factors and builds a highly accurate model to predict severe complications after surgery for perforated peptic ulcers.

Contribution

A novel prediction model with high accuracy (AUC 0.961) for severe postoperative complications in perforated peptic ulcer repair is developed.

Findings

Alcohol use history, ASA score, admission nutritional score, CRP level, and preoperative albumin level are independent risk factors for severe complications.

The developed nomogram showed excellent discriminative ability (AUC = 0.961) and good calibration.

The model was confirmed to have clinical utility through decision curve analysis.

Abstract

To identify risk factors and construct a prediction model for severe complications(Clavien-Dindo Classification III-V)following repair of perforated peptic ulcers.

Clinical data from 230 patients who underwent perforated peptic ulcer repair at Haiyan County People's Hospital and Jiaxing Second Hospital between January 2018 and June 2024 were retrospectively analyzed. Univariate and multivariate logistic regression analyses were performed to screen relevant variables, followed by the development of a risk prediction model for severe complications, with predictive performance validated using receiver operating characteristic (ROC) curve analysis.

The cohort comprised 230 patients (185 males, 45 females) with a mean age of 62.2 ± 17.8 years, predominantly presenting with gastric perforations. Severe postoperative complications occurred in 42 cases (18.3%). In ERAS patients, advantages were observed in terms of the incidence of severe complications and length of hospital stay; however, these differences did not reach statistical significance. Analytical results indicated that alcohol use history, ASA score, admission nutritional score, CRP level, and preoperative albumin level were independent risk factors for severe complications (P < 0.05). The nomogram constructed based on multivariate analysis showed excellent discriminative ability (AUC = 0.961), with calibration curves indicating good agreement between predicted and observed outcomes. Decision curve analysis confirmed the clinical utility of this model.

This prediction model demonstrates high accuracy for severe complications after peptic ulcer perforation repair, providing valuable guidance for clinical monitoring and early preventive interventions.

Figures