Imaging of scar in patients with ventricular arrhythmias of right ventricular origin: cardiac magnetic resonance versus electroanatomic mapping.

INTRODUCTION

Assessment of late gadolinium enhancement (LGE) at cardiac magnetic resonance is often used to detect scar in patients with arrhythmias of right ventricular (RV) origin. Recently, electroanatomic mapping (EAM) has been shown to reliably detect scars corresponding to different cardiomyopathic substrates. We compared LGE with EAM for the detection of scar in patients with arrhythmias of RV origin.

METHODS AND RESULTS

Thirty-one patients with RV arrhythmias and biopsy-proven structural heart disease (18 ARVC and 13 myocarditis), and 5 with idiopathic RV outflow tract arrhythmias underwent LGE analysis and EAM with scar validation through EAM-guided endomyocardial biopsy. EAM scars were present in 23 (64%) patients (all with structural heart disease), whereas LGE was present only in 12 (33%). In 2 cases, EAM provided a false-positive diagnosis of a small scar in the basal perivalvular area. LGE correctly diagnosed EAM scar in 48% of patients, resulting in high positive (92%) but low negative (50%) predictive values. The distribution of LGE was significantly associated with the distribution of EAM scars (P < 0.001 in the free wall, P = 0.003 in the outflow tract, and P = 0.023 in the posterior/inferior wall). Presence of LGE reflected a higher extension of EAM scars (34.4 ± 16.5% vs 7.9 ± 10.1% of the RV area, P < 0.001). At receiver operating characteristic (ROC) analysis, an extension of scar ≥20% of the RV area was the best cut-off value to detect LGE (sensitivity 83%, specificity 92%). Of note, LGE missed 10 of 11 (91%) patients with EAM scars <20% of RV area.

CONCLUSIONS

LGE is significantly less sensitive than EAM in identifying RV cardiomyopathic substrates. Absence of LGE does not rule out the presence of small scars, and EAM with biopsy should be considered to increase the diagnostic yield.