Surgical site infection (SSI) is a significant concern due to its potential to cause delayed wound healing and prolonged hospital stays. This study aims to develop a predictive model in patients with Crohn's disease.

We conducted single-factor and multi-factor logistic regression analyses to identify risk factors, resulting in the development of a logistic regression model and the creation of a nomogram. The model's effect was validated by employing enhanced bootstrap resampling techniques, calibration curves, and DCA curves. Finally, we investigated the risk factors for wall and intra-abdominal infections separately.

90 of 675 patients (13.3%) developed SSI. Several independent risk factors for SSI were identified, including higher postoperative day one neutrophil count (p=0.033), higher relative blood loss (p=0.018), female gender (p=0.021), preoperative corticosteroid use (p=0.007), Montreal classification A1 and L2 (p<0.05), previous intestinal resection (p=0.017), and remaining lesions (p=0.015). Additionally, undergoing strictureplasty (p=0.041) is a protective factor against SSI. These nine variables were used to develop an SSI prediction model presented as a nomogram. The model demonstrated strong discrimination (adjusted C-statistic=0.709, 95% CI: 0.659∼0.757) and precise calibration. The decision curve showed that the nomogram was clinically effective within a probability threshold range of 3% to 54%. Further subgroup analysis revealed distinct risk factors for wall infections and intra-abdominal infections.

We established a new predictive model, which can guide the prevention and postoperative care of SSI after Crohn's disease bowel resection surgery to minimize its occurrence rate.