To delineate and segment the lower limb muscle distribution in patients with varus knee osteoarthritis (KOA) using full-length anteroposterior digital radiographs (DR), quantitatively assess muscle abundance in different lower limb regions, and to explore its associations with KOA progression and the risk of sarcopenia (SP).

A total of 57 patients (80 lower limbs) diagnosed with KOA and hospitalized in the Joint Center of the Third Affiliated Hospital of Guangzhou University of Chinese Medicine between January 2023 and July 2023 were enrolled based on inclusion and exclusion criteria. Inclusion criteria were: knee pain within the past month, age ≥ 50 years, morning stiffness lasting <30 min, and radiographic evidence of joint space narrowing, subchondral sclerosis and/or cysts, and osteophyte formation. Exclusion criteria: valgus KOA or other types of arthritis, comorbidities affecting lower limb function, limb deformities of non-KOA origin, or inability to cooperate with data collection. Demographic data such as gender, age, height, weight, body mass index (BMI), and presence of diabetes or hypertension were collected. Muscle area indices were extracted from five anatomical regions-lateral femur, medial femur, medial tibia, crural interosseous region, and lateral fibula-based on full-length DR images. KOA-related radiographic parameters such as hip-knee-ankle angle (HKA), joint line convergence angle (JLCA), medial joint space width (MJSW), and lateral joint space width (LJSW) were also recorded. Appendicular skeletal muscle mass index (ASMI) was assessed using dual-energy X-ray absorptiometry (DXA). Patients were classified into SP-KOA and KOA groups according to the diagnostic criteria for sarcopenia proposed by the Asian Working Group for Sarcopenia (AWGS), and further categorized into mild KOA [Kellgren & Lawrence (K-L) grade one to two] and severe KOA (grade three to four) groups for comparative analysis. Group comparisons were conducted using Mann-Whitney U test, independent samples t test, and chi square test as appropriate. Binary logistic regression was used to assess the associations between regional muscle area indices and KOA severity or SP risk. Linear regression was employed to evaluate the relationship between muscle distribution ratios and the degree of varus deformity. Receiver operating characteristic (ROC) curves were constructed to determine predictive value.

Binary logistic regression analysis indicated that the tibial medial muscle area index was significantly associated with SP risk [odds ratio (OR)=0.068, 95% confidence interval (CI) (0.011, 0.407)], with area under the curve (AUC) of 0.732 [95% CI (0.625, 0.843), P<0.001], optimal cut-off value of 1.457, sensitivity of 85.3%, and specificity of 52.2%. Age [OR=1.11, 95% CI (1.024, 1.203)] and femoral lateral muscle area index [OR=0.321, 95% CI (0.127, 0.806)] were significantly associated with KOA progression [AUC=0.782, 95% CI (0.682, 0.881), P<0.001], with optimal thresholds of 68 years for age and 2.760 for femoral lateral muscle area index, respectively, yielding a sensitivity of 87.5% and specificity of 55.0%. Moreover, the medial-to-lateral femoral muscle area index ratio was significantly correlated with the degree of varus deformity: negatively correlated with HKA (t=-2.64, P<0.05), positively correlated with JLCA (t=2.38, P<0.05), and negatively correlated with MJSW (t=-3.07, P<0.05).

In patients with varus KOA, atrophy of the lateral femoral muscle and imbalance in medial-to-lateral femoral muscle distribution may contribute to disease progression, while medial tibial muscle atrophy may be associated with increased SP risk. Tailored rehabilitation strategies targeting specific muscle groups may help mitigate KOA progression and reduce the incidence of sarcopenia.