Purpose:
The relationship between keratoconus (KC) and osteoarthritis (OA) has been hypothesized, yet the causal link remained obscure. We aimed to evaluate the genetic correlation and causality between KC and OA through a genetic perspective.
Methods:
Linkage Disequilibrium Score Regression and bidirectional 2-sample Mendelian randomization (MR) analysis were conducted. Data were used from genome-wide association study on KC (4669 cases and 116,547 controls), OA (encompassing 24,955 patients with knee OA, 15,704 patients with hip OA, 39,427 patients with knee or hip OA, and 378,169 control participants), and KC data in the FinnGen consortium for replication and meta-analysis. Simple model, weighted model, inverse-variance weighted, weighted median, and MR-Egger regression were employed to assess the causal effects. MR pleiotropy residual sum and outlier method, MR-Egger method, and leave-one-out analysis were also applied to detect pleiotropy.
Results:
No statistically significant genetic correlation between KC and OA (all P > 0.05) were observed. MR estimates obtained from all 5 methods after meta-analysis revealed nonsignificant odds ratio values of KC-related traits to knee OA, hip OA, and OA (all adjusted P > 0.05). No evidence of causal relationships of knee and hip OA with KC-related traits was detected after meta-analysis of all 5 MR methods (all adjusted P > 0.05).
Conclusions:
The large MR analysis indicated that KC may not be causally associated with a risk of OA.