Development and validation of a predictive model for surgical site infection following joint surgery.

Abstract

Background: Surgical site infections (SSIs) are common complications after joint arthroplasty, leading to increased morbidity and healthcare costs. Traditional models, like the National Nosocomial Infections Surveillance (NNIS) system, have limitations in predicting SSI risk due to a lack of patient-specific factors. This study aimed to create and validate a predictive model focusing on hypoproteinemia to enhance SSI risk assessment in joint surgery patients.

Methods: A retrospective cohort study of 726 patients undergoing joint arthroplasty between 2020 and 2022 was conducted. Data included demographics, laboratory values, and surgical details. Univariate and multivariate analyses identified key predictors, including hypoproteinemia, to develop a predictive nomogram. Model validation was performed using receiver operating characteristic curves, calibration, and decision curve analysis (DCA), comparing it to the NNIS model.

Results: Hypoproteinemia was a significant independent predictor of SSI, with the new model outperforming the NNIS system (area under the curve: 0.829 vs. 0.534). Calibration analysis showed excellent agreement between predicted and observed probabilities, with a mean absolute error of 0.009. DCA further confirmed the model's clinical utility, showing a higher net benefit across various thresholds compared to traditional approaches.

Conclusions: Hypoproteinemia is a critical risk factor for SSI in joint arthroplasty. The new predictive model offers improved risk stratification, supporting a more personalized approach to perioperative management in orthopedic surgery.