Analysis of risk factors for poor wound healing after Nuss procedure for pectus excavatum and establishment of a predictive model.

The Nuss procedure is a widely used treatment for pectus excavatum (PE) that significantly enhances the effectiveness of correction and improves patient satisfaction. However, it is not without its limitations. One common complication that patients face is poor wound healing (PWH). Identifying risk factors for poor wound healing and accurately predicting its occurrence are critical for improving patient outcomes. Currently, there is limited research on poor wound healing in postoperative patients, and there is an urgent need for more robust clinical studies. Over the past decade, we collected data on about 40 indicators from 581 patients who underwent Nuss surgery in our hospital. Lasso regression and multivariate regression analysis (MRA) were used to identify factors significantly associated with poor wound healing. Based on these factors, a prediction model using categorical boosting (CatBoost) theory was developed to assess the risk of PWH. Approximately 9% of patients experienced poor wound healing after the Nuss procedure. Among the 35 potential risk factors, we identified several key clinical indicators closely related to poor incision healing, including muscle thickness (MT), Haller short diameter variation (Haller SDV), relative muscle thickness (RMT), orthopedic triangle (OT), absolute sternal rotation angle (ASRA), and body mass index (BMI). Furthermore, we discovered that the orthopedic triangle, a critical factor in orthopedic plate shaping, played a significant role in predicting PWH (χ² = 42.91, df = 3, p < 0.001). The optimal range for the orthopedic triangle was defined as 0.48 to 0.57. Using these identified factors, we developed the NOWI-model (Nuss surgical incision complication prediction model), which was based on CatBoost theory. The model performed well on the validation dataset, achieving an AUC of 0.96 (95% CI: 0.92,0.99) and an F1-score of 0.80 for predicting PWH. Additionally, an online calculator was created based on this model to help clinicians assess the risk of poor wound healing. We also found that the maximum chest wall elevation during correction should not exceed 4 cm. This study not only identified the high-risk factors for poor wound healing after Nuss surgery and defined the optimal range of some factors for the first time, but also provided a clinical prediction model and a tool for assessing poor wound healing risk. The NOWI model offers a reliable approach for predicting postoperative complications and improving patient care.