Risk Factors and Dynamic Nomogram Development for Surgical Site Infection Following Open Wedge High Tibial Osteotomy for Varus Knee Osteoarthritis: A Retrospective Cohort Study.

ABSTRACT

Background: As the worldwide population ages, the population receiving open wedge high tibial osteotomy (OWHTO) is growing, and surgical site infection (SSI) is a rare but fatal surgical complication. This study aimed to identify risk factors independently associated with SSI following OWHTO and develop a predictive nomogram. Methods: Clinical data of patients who received OWHTO and followed up for more than 12 months in our hospital were retrospectively reviewed. Multivariable logistic regression was performed to determine independent risk factors for SSI and to construct predictive nomograms. The study further illustrated the predictive performance of the model by using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). Results: A total of 1294 eligible patients were included in the study. Multivariate analysis revealed tobacco consumption (OR=3.44, p=0.010), osteotomy size ≥12 mm (OR=3.3, p=0.015), the use of allogeneic bone or artificial bone graft substitutes (allogeneic bone vs none, OR=4.08, p=0.037; artificial bone vs none, OR=5.16, p=0.047), Kellgren-Lawrence (K-L) grade IV (OR=2.5, p=0.046), systemic immune-inflammation index (SII) >423.62 (OR=6.2, p<0.001), high-sensitivity C-reactive protein (HCRP) >2.6 mg/L (OR=2.42, p=0.044), and a higher level of fasting blood glucose (FBG) (OR=1.32, p=0.022) were the independent predictors of SSI. The cutoff score of the model was 148, with a sensitivity of 76.0% and specificity of 81.0%. The concordance index (C-index) and Brier score of the nomogram were 0.856 and 0.017, and the corrected values after 1000 bootstrapping validations were 0.820 and 0.018, respectively. Furthermore, the ROC curve, calibration curve, and DCA exhibited excellent predictive accuracy and clinical applicability of the model. Conclusion: This study developed a dynamic nomogram based on seven predictors, which allowed surgeons to individualize risk stratification of patients and intervene promptly to reduce SSI rates.