Response to: 'Statin therapy now more common than steroids in clinical practice' by Pareek and Chankurdar.

We appreciate the interest shown by Pareek and Chandurkar concerning our recent study on diabetes mellitus (DM) risk associated with disease-modifying antirheumatic drugs and statins in patients with rheumatoid arthritis (RA). They pointed out that statin-associated DM risk in patients with RA might be a more important concern than the risk attributed to glucocorticoids (GCs) due to longer treatment durations. We agree that statin treatment is potentially continued for a longer duration than GC when initiated in patients with RA, as in the case of our cohort (median (IQR) duration for statins: 13 (6–26) months vs GC: 7 (0–17) months). However, GCs are more frequently used in patients with RA than statins (GC: 65% vs statins: 25% in our cohort), and the DM risk increase starts at the first month of treatment and continues to increase as the duration increases. Although there are no data regarding timing of statin-associated risk increase, studies from the general population suggest that DM risk further increases with longer duration of statin use and intensive regimens. These factors were not accounted for in our study and may partially explain the higher HR observed with statins compared with GC. However, we also like to emphasise that statins’ effects on DM risk might be potentiated by the chronic inflammation of RA along with the use of other medications influencing DM risk. The JUPITER ( Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin) trial showed in patients with slightly elevated C reactive protein (≥2 mg/L) a relatively greater DM risk increase (HR (95% CI)=1.26 (1.04 to 1.51)) than in other statin trials. 7 Although this risk might be partly attributed to the potency of rosuvastatin, it is likely that even low-grade inflammation might influence DM risk given the non-significant DM risk increase (HR (95% CI) 1.14 (0.84 to 1.55)) observed in another rosuvastatin trial in patients with heart failure (a well-known diabetogenic state). Lastly, despite the anti-inflammatory effects of both GCs and statins, the potency of this effect might also modify the insulin resistance and DM risk emerged by these medications. Pareek and Chandurkar also highlight that concomitant hydroxychloroquine (HCQ) use with statins or GCs might attenuate the DM risk increase associated with both medications. This therapeutic advantage of HCQ was also demonstrated in our study (see table 1) as shown by Pareek et al for statins, despite the differences in patient characteristics, statin and HCQ doses and durations. This is an important point since both statins and GCs play essential roles in RA and cardiovascular risk management and cannot be avoided because of DM risk. Concomitant HCQ use might be considered in patients with RA who are on these medications and at high risk for DM. However, for DM risk reduction with HCQ in RA, longer treatment durations may be required (∼2 years+) than in those without inflammatory diseases. 10–12 To conclude, despite the DM risk increase with statins and treatment duration concern pointed out by Pareek and Chandurkar, we believe that statins exert beneficial cardiovascular effects with more complex interactions in patients with RA than in the general population. We hope to elucidate the net risk/benefit ratio of statins in our ongoing research on statins in additional RA cohorts.