Long-term Glucocorticoid Use in Rheumatoid Arthritis

To the Editor:

We read with interest the article by Hanly and Lethbridge concerning long-term patterns of glucocorticoid (GC) use in older patients with rheumatoid arthritis (RA)¹. Their report indicates that GC use has remained relatively stable over time, in contrast to greater use of disease-modifying antirheumatic drugs and biologic agents in the treat-to-target directive. They also report that rheumatologists prescribe lower doses than other physicians, and that the mean dose for rheumatologists has decreased over time.

The discussion decries continued use of GC, stating that in view of “the risks associated with chronic corticosteroid use, especially in older adults, renewed efforts are required to minimize their use in the long-term pharmacological management of RA.”¹ Of course, we fully agree that chronic dosing of prednisone (in North America, or prednisolone in Europe) in doses above 5 mg/d is generally undesirable. However, we find the blanket recommendation to discontinue GC entirely in RA unfortunate, in view of compelling evidence that doses of 5 mg/d prednisone or less provide effective and safe therapy, and has been of benefit to many patients with RA.

A recent placebo-controlled trial indicated that even patients in minimal disease activity or remission while taking tocilizumab on 5 mg/d prednisone suffered more flares when their prednisone was slowly tapered over a period of 16 weeks, and fully discontinued for only 8 weeks². This phenomenon is consistent with older data that documented the efficacy of 3 mg/day prednisone in a withdrawal clinical trial³, and similar effectiveness in most patients of an initial and ongoing dose of 3 mg prednisone to doses ≥ 5 mg/day⁴. Further, in 75 patients with RA treated with prednisone in doses < 5 mg for 4 to 8 years and in 73 patients for more than 8 years, the only clinically important adverse events were bruising and thinning of the skin, with no greater prevalence of hypertension, cataracts, or diabetes than expected⁴. Flares were rarely seen, possibly due to continuous control of inflammation by not discontinuing GC at all. Finally, 4 reports published before 1990 indicated that adrenal corticoid suppression was rare in patients taking 5 mg/d prednisolone and not seen in lower doses in those taking less than 5 mg/d^5,6,7,8.

A large placebo-controlled trial on 2 years of 5 mg/d prednisolone in seniors (65+ yrs) with RA will present results next year that can hopefully provide further support for continued small doses over long periods, if not indefinitely⁹. The diathesis leading to RA does not disappear, just as in diabetes and hypertension, and that is why therapy must be continued indefinitely in almost all patients.

In 2002, an editorial concerning a clinical trial documenting efficacy of 10 mg/day prednisolone versus placebo asked, “Are long-term very low doses of prednisone for patients with RA as helpful as high doses are harmful?”¹⁰ It was suggested that “a little corticosteroid, like a glass of wine may benefit many people, whereas a high dose…like a bottle of wine is harmful to all”; and also that “additional disease-modifying antirheumatic drugs [to prednisolone] appeared to be required.”¹⁰ Experience over 18 subsequent years has reinforced these concepts. Very low-dose GC has an excellent benefit-to-harm ratio and frequently provides optimal control for patients with RA.

• © 2021 The Journal of Rheumatology