Background Chronic inflammation and some anti-inflammatory medications in rheumatoid arthritis (RA) are not only responsible for accelerated atherosclerosis but also have effects on the development of traditional risk factors, among which type 2 diabetes mellitus (DM) is the most important. Although hydroxychloroquine (HCQ) and TNF inhibitors have been reported to decrease incident DM risk in RA, there are still a number of unanswered questions regarding the impact of newer biologics, statins, and the timing, dosing, and sustainability of this effect.

Objectives To investigate the impact of specific DMARDs and statins on the risk of incident DM in patients with RA.

Methods Study patients were participants without baseline DM and at least 1 year in the National Data Bank for Rheumatic Diseases (NDB) from 2000 through 2015. DM was determined by patient self-report or use of DM-specific medication. Multivariable Cox proportional hazard models were adjusted for age, sex, ethnicity, socioeconomic and smoking status, comorbidities, body mass index (BMI), and RA severity while examining the time-dependent association of HCQ, other DMARDs, glucocorticoids (GC), and statins.

Results During the 71,668 person-year followup of 13,669 RA patients, 1,139 developed DM. Adjusted HR (95% CI) for DM were 0.67 (0.57–0.80) for HCQ, 0.52 (0.31–0.89) for abatacept, 1.31 (1.15–1.49) for GC, and 1.56 (1.36–1.78) for statins (Table). There was no evidence of risk change through MTX, other synthetic and biological DMARDs. Although the risk reduction was significantly associated with higher HCQ doses and longer (>4 years) treatment durations, HCQ dose of <400mg/day, HR 0.71 (0.52–0.96), and ≥2-year duration, HR 0.72 (0.56–0.91), were also protective. DM-free survival of the patients who discontinued HCQ were quite similar to the patients who started HCQ and kept receiving the drug. Concomitant use of GC did not adversely affect the protective effect of HCQ, HR 0.69 (0.51–0.93). However, there was significant increased risk with statins and GC, HR 2.03 (1.65–2.50). In concomitant use of HCQ and statins, the protective effect of HCQ and increased risk with statins disappeared, HR 0.92 (0.68, 1.25).

Conclusions The risk of incident DM in RA was reduced with HCQ and abatacept and increased with GC and statins. HCQ not only confers a sustainable, dose and treatment duration-dependent protective effect but also abolishes the increased risk of DM as a consequence of GC or statins.

Disclosure of Interest None declared