Causal analysis of body composition measurements in osteoarthritis knee: a two-sample mendelian randomization study

Abstract

Background: To analyse the causal associations of different physical measures with osteoarthritis knee (KOA).

Methods: Exposure factors (weight, body mass index (BMI), body fat percentage, waist circumference, hip circumference, waist-hip ratio (WHR), and basal metabolic rate (BMR)), and outcome factor KOA were analyzed by inverse-variance weighted (IVW) method, along with heterogeneity test, sensitivity and pleiotropy analyses. Meta-analysis was used to combine the effect values of IVW methods in different data sources.

Results: Weight, BMI, body fat percentage, waist circumference, hip circumference and BMR analyses showed causal association with increased KOA risk, while WHR analysis indicated a reduction of the incidence of KOA. P-value for all the results was less than 0.05 and F-value large than 20. All results were negative for heterogeneity tests and sensitivity analyses, and there was pleiotropy in weight and BMR. Meta-analysis results showed that the results of Odds Ratios (95% Confidence Intervals) for Weight (1.43(1.35-1.51)), BMI (1.40(1.10-1.78)), body fat percentage (1.56(1.44-1.68)), waist circumference (1.40(1.10-1.78)), hip circumference (1.37(1.30-1.44)), WHR (0.86(0.71-1.04)) and BMR (1.36(1.27-1.46) were consistent with the ones by Mendelian randomization analyses.

Conclusions: Body fat percentage may be a better indicator of KOA than BMI. In addition, weight and BMR may have a causal effect in KOA, but WHR does not have a causal relationship. BMI, body fat percentage, waist circumference, and hip circumference has a causal effect on KOA.

Keywords: Body composition measurements; Mendelian randomization; Meta-analysis; Osteoarthritis knee.