This study evaluated the prognosis and prognostic factors of patients with cardiac sarcoidosis (CS), an underdiagnosed disease. Patients from a retrospective multicentre registry, diagnosed with CS between 2001 and 2017 based on the 2016 Japanese Circulation Society or 2014 Heart Rhythm Society criteria, were included. The primary endpoint was a composite of all-cause death, hospitalization for heart failure, and documented fatal ventricular arrhythmia events (FVAE), each constituting exploratory endpoints. Among 512 registered patients, 148 combined events (56 heart failure hospitalizations, 99 documented FVAE, and 49 all-cause deaths) were observed during a median follow-up of 1042 (interquartile range: 518–1917) days. The 10-year estimated event rates for the primary endpoint, all-cause death, heart failure hospitalizations, and FVAE were 48.1, 18.0, 21.1, and 31.9%, respectively. On multivariable Cox regression, a history of ventricular tachycardia (VT) or fibrillation [hazard ratio (HR) 2.53, 95% confidence interval (CI) 1.59–4.00, P < 0.001], log-transformed brain natriuretic peptide (BNP) levels (HR 1.28, 95% CI 1.07–1.53, P = 0.008), left ventricular ejection fraction (LVEF) (HR 0.94 per 5% increase, 95% CI 0.88–1.00, P = 0.046), and post-diagnosis radiofrequency ablation for VT (HR 2.65, 95% CI 1.02–6.86, P = 0.045) independently predicted the primary endpoint. Although mortality is relatively low in CS, adverse events are common, mainly due to FVAE. Patients with low LVEF, with high BNP levels, with VT/fibrillation history, and requiring ablation to treat VT are at high risk. Clinical characteristics, findings on cardiovascular imaging, prognosis, and prognostic factors in patients with cardiac sarcoidosis: summary results of ILLUMINATE-CS.

Endomyocardial biopsy (EMB) and late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR) imaging performed at baseline are both used to evaluate the extent of myocardial fibrosis. However, no study has directly compared the effectiveness of these diagnostic tools in the prediction of left ventricular reverse remodeling (LVRR) and prognosis in response to therapy in patients with idiopathic dilated cardiomyopathy (IDCM). Seventy-five patients with newly diagnosed IDCM who were undergoing optimal therapy were assessed at baseline using LGE-CMR imaging and EMB; the former measured LGE area and the latter measured collagen volume fraction (CVF) as possible predictive indices of LVRR and cardiac event-free survival. Among all the baseline primary candidate factors with P < 0.2 as per univariate analysis, multivariate analysis indicated that only LGE area was an independent predictor of subsequent LVRR (β = 0.44; 95 % confidence interval (CI) 0.87–2.53; P < 0.001), as indicated by decreasing left ventricular end-systolic volume index over the 1-year follow-up. Kaplan–Meier curves indicated significantly lower cardiac event-free survival rates in patients with LGE at baseline than in patients without (P < 0.01). By contrast, there was no significant difference in prognosis between patients with CVF values above (severe fibrosis) and below (mild fibrosis) the median of 4.9 %. Cox proportional hazard analysis showed that LGE area was an independent predictor of subsequent cardiac events (hazard ratio 1.06; 95 % CI 1.02–1.10; P ≤ 0.01). The degree of myocardial fibrosis estimated by baseline LGE-CMR imaging, but not that estimated by baseline EMB, can predict LVRR and cardiac event-free survival in response to therapy in patients with newly diagnosed IDCM.