Clinical outcomes of initial steroid pulse therapy in cardiac sarcoidosis: a nationwide multicenter study from the MYSTICS registry

Background/Introduction Corticosteroids are the main therapeutic option for cardiac sarcoidosis (CS). Although high-dose steroid pulse therapy is often administered in patients with severe or rapidly progressive disease, data regarding its prognostic impact remain limited. Purpose Using a large multicenter retrospective registry of patients with CS, we compared clinical outcomes among patients treated with standard-dose steroids, steroid pulse therapy, and those who did not receive steroid therapy as initial treatment. Methods Of 2,747 patients enrolled in the MYSTICS registry, 2,366 met the 2016 Japanese Circulation Society criteria and/or the 2014 American Heart Association statement for CS. After excluding 514 patients diagnosed before April 2012 and 4 with insufficient data, 1,848 patients diagnosed between April 2012 and March 2021 were analyzed. Results Patients were classified into no-steroid (n=207), steroid pulse (n=64), and standard steroid (n=1,577) groups. The primary endpoint was all-cause death; the secondary endpoint was cardiovascular death. Baseline characteristics differed among groups, including age, sex, diabetes, hypertension, hemoglobin, eGFR, prior CRT-D implantation, extracardiac sarcoidosis, cardiac uptake on Ga/PET, and late gadolinium enhancement on MRI. Baseline and 6-month prednisolone doses were similar between the standard steroid and steroid pulse groups (30 [30-30] vs 30 [30-30] mg/day, P=0.36; 10 [5-10] vs 10 [5-10] mg/day, P=1.00). During a mean follow-up of 1,914 days, 176 deaths occurred. 5-year cumulative mortality rates from CS diagnosis were 18.2%, 9.0%, and 7.6% in the no-steroid, standard steroid, and steroid pulse groups, respectively (log-rank P<0.01). After adjustment for age, sex, logBNP, and left ventricular ejection fraction, the risk of all-cause death relative to the standard steroid group was 1.09 (95% CI 0.59-2.00, P=0.79) and 1.13 (95% CI 0.74-1.73, P=0.58) for the steroid pulse group and the no-steroid group, respectively. Similarly, there was no significant difference in the risk of cardiovascular death across the treatment groups. Even when the treatment initiation was defined as time zero, the risk of all-cause death and cardiovascular death for steroid pulse therapy was similar to that with standard steroid therapy (HR 1.15, 95% CI 0.68-1.71, P=0.56 and HR 1.10, 95% CI 0.54-2.23, P=0.81, respectively). Conclusions In this multicenter retrospective CS registry, no significant difference in mortality was observed across initial steroid treatment strategies. The establishment of an optimal steroid strategy for CS requires further investigation.