Glucocorticoid dosing and relapses in giant cell arteritis-a single centre cohort study

Rheumatology (Oxford). 2022 May 5;61(5):1997-2005. doi: 10.1093/rheumatology/keab677.

Abstract

Objective: To investigate the relationship between real life glucocorticoid (GC) dosing and relapse rates in patients with new onset GCA in a single centre.

Methods: Complete clinical data taken from the inpatient and outpatient records of consecutive GCA patients followed beyond stopping GC were retrospectively analysed for GC doses, other immunomodulatory agents and relapses.

Results: We included 54 patients with GCA confirmed by biopsy or imaging and followed over their complete GC course. In the 25% dose percentile, patients who needed no pulse therapy at onset reached a dose of 15 mg prednisolone or lower at day 40, of 7.5 mg prednisolone or lower on day 169 (after 24 weeks), and were off prednisolone on day 496 (70 weeks). They were below British Society for Rheumatology recommended doses between week 4 and week 12 and above these after week 14. The cumulative prednisolone dose reached in this 25% quartile was 3.74 g. Of the 54 patients, 24 (44%) relapsed, only four of whom had stopped GC clearly (17-58 weeks) earlier than the 25% dose quartile and one was distinctly (>10%) below the 25% GC percentile. MTX treatment was not significantly associated with fewer relapses (P = 0.178).

Conclusion: Despite a long-term GC regimen with slow rates of reduction in the low dose range and high cumulative prednisolone doses, 44% of the patients relapsed. Only five (21%) of these relapses may have been prevented by adhering to the recommended GC regimen.

Keywords: giant cell arteritis; glucocorticoids; relapse; taper; treatment.

MeSH terms

  • Chronic Disease
  • Cohort Studies
  • Giant Cell Arteritis* / complications
  • Giant Cell Arteritis* / drug therapy
  • Glucocorticoids
  • Humans
  • Prednisolone / therapeutic use
  • Recurrence
  • Retrospective Studies
  • Treatment Outcome

Substances

  • Glucocorticoids
  • Prednisolone