Preoperative CT-derived thigh muscle quality as a predictor of postoperative outcomes after total knee arthroplasty in patients with hemophilia
TL;DR
Preoperative CT-derived thigh muscle quality was independently associated with 1-year functional recovery after TKA in haemophilic arthropathy and may support preoperative risk stratification.
Abstract
Background Postoperative recovery after total knee arthroplasty (TKA) for haemophilic arthropathy varies substantially, and objective imaging markers for preoperative risk stratification remain limited. This study investigated whether computed tomography (CT)-derived thigh muscle quality and tissue composition are associated with 1-year outcomes after TKA in patients with hemophilia. Methods This retrospective study included 67 patients with hemophilia who underwent primary unilateral TKA between January 2021 and March 2025. Preoperative CT was used to quantify thigh muscle volume, muscle quality, defined as mean muscle attenuation in Hounsfield units (HU), intermuscular adipose tissue, and subcutaneous adipose tissue. The excellent outcome was defined as both Knee Society Clinical Score (KSC) ≥70 and Knee Society Functional Score (KSF) ≥70 at 1 year. Multivariable linear regression, a prespecified five-variable logistic regression model, receiver operating characteristic (ROC) analysis, and inter-observer reproducibility assessment were performed. A secondary analysis compared CT-derived tissue parameters with age-, sex-, and body mass index-matched non-arthropathic controls. Results CT-derived parameters showed excellent inter-observer reproducibility. Muscle quality was higher in the excellent-outcome group than in the non-excellent-outcome group (36.07 ± 4.08 vs. 30.04 ± 6.12 HU; p < 0.001). At 1 year, the excellent-outcome group had higher KSC (88.21±5.80 vs 84.55±6.55; p=0.038) and KSF (83.81 ± 5.46 vs. 63.00 ± 11.05; p < 0.001). Muscle quality was independently associated with postoperative SF-12 PCS (p = 0.006), KSC (p = 0.008), and KSF (p = 0.002). In logistic regression, muscle quality was independently associated with excellent outcome (OR = 1.30 per 1-HU increase; 95% CI, 1.11–1.50; p = 0.004). ROC analysis identified an exploratory cutoff of 31.4 HU (AUC = 0.829; 95% CI, 0.72–0.94), sensitivity of 80.0%, and specificity of 88.9%. Compared with matched controls, patients with haemophilic arthropathy had lower muscle volume and muscle quality but higher adipose tissue volumes. Conclusion Preoperative CT-derived thigh muscle quality was independently associated with 1-year functional recovery after TKA in haemophilic arthropathy and may support preoperative risk stratification.