SURGICAL TREND ANALYSIS OF LONGITUDINAL LEVELS OF ENDURING DEEP SURGICAL SITE INFECTION RISK AFTER HIP AND KNEE ARTHROPLASTY

Introduction/Aim Surgical site infections (SSIs) remain a leading cause of morbidity following total joint arthroplasty and are widely used for quality benchmarking and public reporting. Despite advances in perioperative infection prevention, national data suggest that SSI rates have not demonstrated sustained declines. Increasing patient complexity may obscure improvements in infection prevention efforts. We sought to evaluate temporal trends in invasive SSI rates following hip and knee arthroplasty while assessing changes in patient risk factors and whether these influenced infection trends over time. Methods We performed a retrospective cohort study of 31,167 primary and revision total hip and knee arthroplasty procedures across hospitals within a single academic health system from 2018–2025 using a National Healthcare Safety Network (NHSN)–compliant database. Temporal trends in key prosthetic joint infection risk factors—including age, body mass index (BMI), diabetes, smoking status, ASA classification, and discharge disposition—were assessed. Multivariable logistic regression evaluated the association between calendar year and invasive SSI after adjustment for patient, procedural, and institutional factors. Interaction analyses between calendar year and major risk factors were performed overall and stratified by procedure type. Results Unadjusted invasive SSI rates decreased from 1.74% in 2018 to 0.81% in 2025 (OR per year 0.948, 95% CI 0.909–0.989, p = 0.013). Over the same period, patient complexity increased, including a rise in diabetes prevalence from 5.36% to 19.08% (OR per year 1.270, p < 0.001) and composite risk burden from 1.31 to 1.51 (+0.03 risk factors/year, p < 0.001), while smoking prevalence declined from 7.55% to 3.76% (OR per year 0.881, p < 0.001). In adjusted analyses, calendar year was not independently associated with SSI risk (OR 0.976, 95% CI 0.933–1.021, p = 0.297). Overall interaction analyses were not significant. However, procedure-specific analyses demonstrated that the effect of BMI on SSI risk in hip arthroplasty increased over time (p = 0.009), while the effect of current smoking on SSI risk in knee arthroplasty also changed over time (p = 0.020). Conclusions In this large contemporary cohort, invasive SSI rates remained stable after risk adjustment despite increasing patient complexity. Procedure-specific findings suggest evolving contributions of modifiable risk factors, particularly BMI in hip arthroplasty and smoking in knee arthroplasty. These findings support that improvements in perioperative optimization and infection prevention may be offsetting increasing baseline risk. Risk-adjusted interpretation of SSI metrics is essential for accurate benchmarking in modern arthroplasty practice.