Prognostic implications of extensive right ventricular late gadolinium enhancement in cardiac sarcoidosis: insights from the ILLUMINATE-CS-CVi registry

Background In cardiac sarcoidosis (CS), the presence of late gadolinium enhancement (LGE) in the left ventricle (LV) on cardiac magnetic resonance (CMR) is known to be associated with worse clinical outcomes, particularly fatal ventricular arrhythmic events (FVAEs). However, the prognostic value of LGE presence and extent in the right ventricle (RV) is not fully understood. Purpose To evaluate the prognostic value of RV-LGE for FVAEs in patients with CS. Methods We analyzed 198 patients (median 61 years, 38.9% male) who underwent CMR within 6 months of diagnosis using a retrospective multicenter registry in Japan (ILLUMINATE-CS-CVi registry). CS was diagnosed between 2001 and 2017 based on either the 2016 Japanese Circulation Society criteria or the 2014 Heart Rhythm Society criteria. The presence and extent of RV-LGE were visually assessed in a blinded manner for the outcomes. We divided the RV into 9-segments based on short-axis slice levels (basal, mid, and apical) and location in the RV free wall (anterior, lateral, and inferior). Patients were categorized into three groups according to the sum of RV-LGE segments: 0-, 1-, and ≥2-segment groups. The primary endpoint was documented FVAEs. Results RV-LGE was present in 73 (36.9%) out of 198 patients; 35 (17.7%) were in the 1-segment group and 38 (19.2%) were in the ≥2-segment group. The basal inferior wall was the most common site of RV-LGE involvement, present in over 50% of affected patients. A greater number of RV-LGE segments was associated with lower LV and RV ejection fraction and greater extent of LV-LGE. During the median follow-up of 4.0 years (interquartile range, 2.1–6.8 years), 43 FVAEs were observed. After adjusting for covariates including LV ejection fraction and the number of LV-LGE segments, the ≥2-segment group was associated with a significantly higher risk of FVAEs compared with the 0-segment group (hazard ratio [HR], 2.46; 95% confidence interval [CI], 1.17–5.19; P = 0.018), whereas the 1-segment group showed no significant difference compared with the 0-segment group (HR, 1.27; 95% CI, 0.54–2.99; P = 0.58). The number of RV-LGE segments, as a continuous variable, was also independently associated with FVAEs (HR, 1.21; 95% CI, 1.05–1.40; P = 0.008). Conclusion The presence of RV-LGE was observed in more than one-third of patients with CS, and a greater number of RV-LGE segments was associated with FVAEs independent of the extent of LV-LGE, which may imply that its quantification provides better risk stratification.