Soft Tissue-to-Bone Ratio Outperforms Body Mass Index in Predicting Periprosthetic Joint Infection in Total Knee Arthroplasty: A Retrospective Case-Control Study.

Abstract

Periprosthetic joint infection (PJI) remains one of the most devastating complications following total knee arthroplasty (TKA), often leading to prolonged treatment and poor patient outcomes. Traditionally, body mass index (BMI) has been used as a key risk factor in predicting PJI, but BMI fails to reflect local anatomical variations influencing surgical outcomes. This study explored whether the soft tissue-to-bone (STiB) ratio, a novel radiographic measurement, could serve as a more accurate predictor of PJI compared to BMI. We conducted a retrospective case-control study involving 25 culture-confirmed PJI cases and 75 matched TKA controls, with no evidence of infection at least 24 months after TKA. Participants were matched for BMI to reduce its confounding influence, allowing a focused comparison of STiB ratios between those who developed PJI and those who did not. Preoperative radiographs were analyzed to measure the STiB ratio at three different anatomical locations: antero-posterior (AP), Lateral 1, and Lateral 2. The ratios were calculated as the total soft-tissue structures divided by the osseous structures at each location. Multivariate logistic regressions and receiver operating characteristic (ROC) analyses were performed to assess predictive validity. The STiB ratio at Lateral 1 was significantly higher in the PJI group compared to the control group, with a mean of 1.92 versus 1.80 (P = 0.037). Logistic regressions showed that this site was an independent predictor of infection (odds ratio (OR) 2.2; 95% confidence interval (CI): 1.10 to 4.40; P = 0.025). The BMI was not statistically significant (P = 0.845). The ROC curves showed modest discriminative performance for Lateral 1 (Area under the curve (AUC) = 0.52). The STiB ratio, particularly at Lateral 1, may be a more accurate and clinically useful predictor of PJI risk than BMI. Its accessibility and objective nature make it a compelling addition to preoperative risk stratification protocols.