Although metformin there is growing evidence of metformin’s pleiotropic effects, including possible effects on pain, there is a lack of study investigating the association of metformin with the prevalence of musculoskeletal pain among a large type 2 diabetes cohort.
Cross-sectional analyses were conducted with UK Biobank data from 21,889 participants with type 2 diabetes. Type 2 diabetes, metformin use, and musculoskeletal (back, knee, hip, and neck/shoulder) pain were self-reported. Participants reported musculoskeletal pain that had interfered with their usual activities in the last month (recent pain), and for more than 3 months (chronic pain). We performed logistic regression analyses for recent and chronic pain for each site and for multisite pain among participants with diabetes who did or did not take metformin.
Participants using metformin had lower odds of musculoskeletal pain for back [recent OR 0.91, 95%CI 0.85 to 0.97; chronic OR 0.87, 95%CI 0.81 to 0.93], knee [recent OR 0.91, 95%CI 0.85 to 0.97; chronic OR 0.87, 95%CI 0.81 to 0.94], and neck/shoulder regions [chronic OR 0.92, 95%CI 0.85 to 0.99] but not hip pain. Participants using metformin also had lower odds of reporting chronic multisite musculoskeletal pain. The associations were generally stronger among women.
People with diabetes taking metformin were less likely to report back, knee, neck/shoulder, and multisite musculoskeletal pain than those not taking metformin. Therefore, when treating these patients, clinicians should be aware that metformin may contribute to fewer reports of musculoskeletal pain. These effects should be investigated in future studies.