Echocardiographic predictors of right ventricular involvement in cardiac sarcoidosis

Right ventricular (RV) involvement in cardiac sarcoidosis (CS) is associated with adverse cardiovascular outcomes. Echocardiography is recommended in the screening of sarcoidosis patients for cardiac involvement. Abnormal imaging should prompt onward performance of cardiac magnetic resonance (CMR) and 18fluorodeoxy glucose positron emission tomography (FDG-PET). Suspicion of RV involvement in the screening phase could prompt more expedient advanced imaging.

This study aimed to determine how well RV-focused echocardiographic parameters could predict RV involvement of CS. We defined RV involvement by the presence of RV impairment (RVEF <50%) and/or RV late gadolinium enhancement (LGE) on CMR and/or RV FDG-uptake on FDG-PET.

Consecutive patients referred for evaluation of suspected CS were prospectively recruited. All those with evidence of extracardiac sarcoidosis were included. Detailed 2D and 3D echocardiography with RV strain, CMR and FDG-PET was performed in all patients within 3 months. CS was diagnosed in line with the 2014 Heart Rhythm Society criteria. RV/pulmonary artery (PA) coupling was defined as the tricuspid annular plane systolic excursion (TAPSE):pulmonary artery systolic pressure (PASP) ratio. RV 4-chamber strain included the RV free wall and septal longitudinal strain components.

Among the 198 sarcoidosis patients enrolled, 97 (49%) were diagnosed with CS. The prevalence of pulmonary hypertension was 3%. RV involvement was present in 26 (13%) among which 14 had RV free wall or septal LGE, 17 had CMR RVEF <50% and 10 had RV FDG uptake. Six patients had both LGE and FDG uptake while 8 patients had both LGE and RVEF <50%.

Compared to those without, patients with RV impairment had lower RV fractional area change (38% [IQR 28-46] vs 43% [39-48], p=0.005), RV free wall strain (-17.7 ± 4.2% vs -22.5 ± 5.5%, p<0.001), RV 4-chamber strain (-13.9 ± 3.4% vs -18.5 ± 4.5%, p<0.001), 3DE RVEF (44 ± 7.4% vs 50 ± 7.5%, p=0.019) and greater RV basal diameter (3.99 ± 0.60cm vs 3.63 ± 0.58cm, p=0.004). There was no significant difference in TAPSE (p=0.091), PASP (p=0.811), RV/PA coupling (p=0.271) and tissue doppler RV S’ (p=0.203) values between the two groups.

RV 4-chamber strain (AUC: 0.797) had the best ability to predict RV involvement followed by RV free wall strain (AUC: 0.757) and 3DE RVEF (AUC: 0.730). The optimal cut-off for RV 4-chamber strain to predict RV involvement was -15.3% which had 74% sensitivity and 77% specificity.

In a prospective cohort, RV strain is a useful indicator of RV involvement among suspected CS patients and should be performed routinely in all screening studies of sarcoidosis patients.