To the Editor:

We thank Dr. Docken for his letter and interest in our article. The definitions of relapse and recurrence that we used were similar to those cited from an earlier study (Salvarani C, Hunder GG. Musculoskeletal manifestations in a population-based cohort of patients with giant cell arteritis. Arthritis Rheum 1999;42:1250–66). We defined relapse as the return of signs and/or symptoms of GCA occurring after therapy (prednisone) was reduced, or less than 1 month after its discontinuation, along with regression of these signs or symptoms when the dosage of prednisone was increased, or therapy was resumed. Recurrence was defined as the return of signs and/or symptoms of GCA more than 1 month after discontinuation of therapy and their resolution when therapy was restarted. Thus, relapses and recurrences tend to prolong corticosteroid treatment, as Dr. Docken suggests, and increase the risk of steroid side effects. In this study, we did not record the specific findings that defined a relapse or recurrence in individual patients. Although a rise in the erythrocyte sedimentation rate (ESR) could fulfill the “signs or symptoms” part of these criteria, we looked for more evidence of active disease to make these diagnoses. We seldom made a diagnosis of relapse or recurrence on the basis of an asymptomatic rise in the ESR alone.

Because corticosteroids are currently the main form of therapy for GCA, a critical way to reduce the risk of adverse events is to carefully maintain the lowest dose of these drugs that suppresses the disease, as suggested by Dr. Docken. Nevertheless, it should also be recognized that there may be instances when clinically important flare-ups of GCA are manifest initially or only by an elevation of acute-phase reactants.

Anne Prøven MD*, Gene G. Hunder MD†, * Martina Hansens Hospital Baerum Postterminal, Norway, † Mayo Clinic Rochester, MN.