Incident diabetes associated with hydroxychloroquine, methotrexate, biologics and glucocorticoids in rheumatoid arthritis: A systematic review and meta-analysis

To evaluate the impact of disease-modifying antirheumatic drugs on the risk of developing diabetes in rheumatoid arthritis (RA) patients without diabetes. Electronic database searches of PubMed, EMBASE and Cochrane Library plus a hand search of conference proceedings were performed from inception to October 2019. The studies assessing the association between diabetes and antirheumatic agents in RA patients in cohort or case-control design were included. Data were pooled using fixed-effects or random-effects meta-analysis according to I2 and pooled hazard ratios (HRs), and 95% confidence intervals (CIs) were used as summary statistic. A total of 15 studies involving 552,019 patients with RA (11 for hydroxychloroquine, 7 for methotrexate, 6 for tumor necrosis factor inhibitors [TNFi], and 8 for glucocorticoids) were included. In pooled analysis, a reduced risk of diabetes was reported with hydroxychloroquine (meta-HR 0.61, 95% CI 0.56–0.66), methotrexate (meta-HR 0.81, 95% CI 0.75–0.87), TNFi (meta-HR 0.63, 95% CI 0.55–0.71), while glucocorticoids was associated with an increased risk of developing diabetes in a dose-dependent manner (Any dose: meta-HR 1.46, 95% CI 1.39–1.53; <10 mg/day prednisolone or equivalent: meta-HR 1.30, 95% CI 1.13–1.51; ≥10 mg/day prednisolone or equivalent: meta-HR 2.25, 95% CI 1.88–2.70). Hydroxychloroquine, methotrexate and TNFi were associated with decreased risk of diabetes, and glucocorticoids with increased risk in RA patients. These important findings may aid clinical decision-making in the management of RA. Large, prospective, well-designed studies are needed in the RA patients with high-risk diabetes.