Abstract: BACKGROUND: Frozen shoulder is a common disease in orthopedics, but there is no specific clinical indicator for diagnosis. There is a significant association between inflammatory cytokines and frozen shoulder, but the specific causal relationship is not yet clear. This study used summary statistical data from genome-wide association studies (GWAS) for Mendelian randomization analysis. GWAS data are based on large-sample genetic variation information, which can reduce environmental confounding factors and more reliably infer the causal relationship between inflammatory cytokines and frozen shoulder, filling the limitations of traditional observational studies that cannot determine causal relationships.
OBJECTIVE: To explore the causal relationship between inflammatory cytokines and frozen shoulder using bidirectional two-sample Mendelian randomization method.
METHODS: Using the summary statistical data of GWAS in the FinnGen database, the causal relationship between 41 inflammatory cytokines and frozen shoulder was analyzed. The FinnGen database is a collaborative initiative launched by the Finnish Institute for Health and Welfare (THL), the University of Helsinki, and other Finnish research institutions. It comprises 2 942 case samples and 167 641 controls of European ancestry, integrating genomic data, clinical phenotypes, and biochemical indicators from hundreds of thousands to millions of individuals. This resource supports genetic association studies for various diseases. This study was based on publicly accessible summary-level data and exempt from ethical review. Bidirectional Mendelian randomization analysis was performed using inverse variance weighting, weighted median, weighted model, simple model, MR-Egger regression, and sensitivity analysis (including MR-Egger, MR-PRESSO, Cochran’s Q test).
RESULTS AND CONCLUSION: Monocyte chemotactic protein 3 has significant causal effects in both positive and negative directions. In positive analysis, monocyte chemoattractant protein 3 was positively correlated with frozen shoulder risk [odds ratio (OR)=1.176, 95% confidence interval (CI): 1.034-1.338, P=0.014]; in reverse analysis, frozen shoulder was negatively correlated with monocyte chemoattractant protein 3 levels (OR=0.782, 95% CI: 0.625-0.979, P=0.032). In addition, a significant association was found between tumor necrosis factor β and the risk of frozen shoulder (OR=1.126, 95% CI: 1.002-1.264, P=0.046); in reverse analysis, stromal cell-derived factor 1α also showed a significant association with the risk of frozen shoulder (OR=1.1, 95% CI: 1.011-1.196, P=0.028), indicating a reliable correlation between tumor necrosis factor β, stromal cell-derived factor 1α, and frozen shoulder. This bidirectional Mendelian randomization study revealed the complex interaction between monocyte chemoattractant protein 3 and frozen shoulder, suggesting that monocyte chemoattractant protein 3 may be a potential therapeutic target. In addition, tumor necrosis factor β has been shown to be associated with the risk of frozen shoulder and may be a potential risk factor for frozen shoulder. And frozen shoulder is also associated with elevated levels of stromal cell-derived factor 1α, which has the potential to become a diagnostic biomarker for frozen shoulder. However, further research is needed to elucidate the biological mechanisms behind these causal relationships. Furthermore, the analysis of international databases provides candidate molecules and causal inference paradigms for Chinese research, but precise translation needs to be achieved in combination with local data.

Key words: frozen shoulder, inflammatory cytokines, monocyte chemotactic protein 3, stromal cell-derived factor 1α, Mendelian randomization analysis