Abstract
This study aimed to examine whether dried fruit intake is causally associated with Osteoarthritis (OA). A two-sample Mendelian randomization (MR) analysis using the inverse-variance weighted (IVW), weighted median (WM), and MR-Egger regression methods was performed. We used the publicly available summary statistics data sets of genome-wide association studies (GWAS) meta-analyses for dried fruit intake in individuals included in the UK Biobank (n = 421,764; MRC-IEU consortium) as the exposure and a GWAS publicly available in PubMed for OA (total n = 484,598; case = 39,515, control = 445,083) as the outcome. We selected 41 single nucleotide polymorphisms at genome-wide significance from GWASs on dried fruit intake as the instrumental variables. The IVW method showed evidence to support a causal association between dried fruit intake and OA (beta = -0.020, SE = 0.009, P = .039). MR-Egger regression indicated no directional pleiotropy (intercept = 1E-05; P = .984), but it showed no causal association between dried fruit intake and OA (beta = -0.020, SE = 0.043, P = .610). However, the WM approach yielded evidence of a causal association between dried fruit intake and OA (beta = -0.026, SE = 0.012, P = .026). Cochran's Q test showed the existence of heterogeneity, but the statistics of I2 showed low heterogeneity. The results of MR analysis support that dried fruit intake may be causally associated with a decreased risk of OA.