Mediating effect of lower extremity muscle on the relationship between obesity and osteoarthritis in middle-aged and elderly women in Korea: based on the 2009-2011 Korea National Health and Nutrition Examination Survey

Abstract

Objectives: This study investigated whether the lower extremity muscle mass index (LMI) mediates the relationship between general obesity, central obesity, and knee osteoarthritis in middle-aged and elderly women in Korea.

Methods: Data of 2,843 women aged ≥50 years were collected from the Korean National Health and Nutrition Examination Survey conducted between 2009 and 2011. General obesity and central obesity were evaluated based on body mass index (BMI) and waist circumference (WC), calculated through anthropometric measurements and body composition assessments. LMI was calculated by dividing the muscle mass in both legs-measured using the dual-energy X-ray absorptiometry-by body weight. Knee osteoarthritis was defined as a Kellgren-Lawrence scale (KL) grade of ≥2 as assessed through radiographic images.

Results: Knee osteoarthritis prevalence, indicated by KL grades, was significantly higher in the general obesity and central obesity groups compared to the normal group, and conversely, lower with varying LMI levels. Using mediation analysis with bootstrapping and adjusting for covariates, we found that LMI mediated the relationship between BMI and KL (β, 0.005; 95% confidence interval [CI], 0.000 to 0.010) and WC and KL grade (β, 0.002; 95% CI, 0.001 to 0.003), explaining 4.8% and 6.7% of the total effects of BMI and WC on KL grade, respectively.

Conclusions: The study suggested that LMI partially mediates the link between general obesity and/or central obesity and knee osteoarthritis, proposing that a higher proportion of lower limb muscle mass relative to body weight can alleviate the increased risk of knee osteoarthritis caused by obesity.

Keywords: Abdominal obesity; Body mass index; Knee osteoarthritis; Lower extremity; Muscles; Waist circumference.

Figure 1.

Flow chart of eligible participants in the study.

Figure 2.

Mediation analysis. Path coefficients of body mass index (BMI; A) or waist circumference (WC; B) on Kellgren-Lawrence grade (KL) through lower extremity muscle mass index (LMI). Path a: estimated coefficient for the regression with BMI (A) or WC (B) predicting LMI; Path b: estimated coefficient for the regression with LMI predicting KL; Path c=total effects for the regression with BMI (A) or WC (B) predicting KL; Path c`=direct effects for the regression with BMI (A) or WC (B) predicting KL independent of LMI.