Concordance and reliability between electroanatomical mapping using high density mapping catheter and ADAS-3D software for myocardial tissue characterization in non-ischemic cardiomyopathy patients

Risk assessment for ventricular arrhythmias (VAs) and sudden cardiac death (SCD) in non-ischemic cardiomyopathy (NICM) is a clinical challenge. Late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) is the gold standard for identifying scar core zone (CZ) and border zones (BZ), the primary substrates for VA. The ADAS 3D software (GALGO Medical) processes LGE-CMR data to map CZ and BZ distribution across myocardial segments, from endocardium (ENDO) to the epicardium (EPI). For patients unable to undergo LGE-CMR, myocardial tissue characterization may alternatively be achieved via electroanatomical mapping (EAM) using a high-density mapping catheter equipped with omnipolar technology (OT) to delineate scar regions and their electrophysiological profiles.

To assess the concordance and reliability of myocardial tissue characterization between ADAS 3D and EAM performed with a high-density mapping catheter using OT.

In this multicentric retrospective observational study, we included 36 NICM patients who underwent left ventricular (LV) EAM for arrhythmia and SCD risk assessment, all of whom had prior LGE-CMR imaging. ADAS 3D processed the LGE-CMR data to generate color-coded pixel intensity maps of CZ and BZ, illustrating their extension from the ENDO to EPI layers. EAM was performed with the Ensite X navigation system (ABBOTT) and the Advisor high density Grid mapping catheter (AHD-G, ABBOTT) using OT, bipolar (BIP), and unipolar (UNI) configurations. Concordance between EAM and ADAS 3D in CZ and BZ mapping was evaluated using Cohen's Kappa coefficient for agreement and Bland-Altman analysis for reliability across all EAM modes

The agreement analysis confirmed the UNI configuration’s capability to explore epicardial regions (Cohen’s Kappa 0.93, p < 0.001; positive predictive value [PPV] 97.2%, negative predictive value [NPV] 86.5%) and demonstrated superior reproducibility of OT over BIP when compared to ADAS 3D in ENDO layers (OT vs BIP: K Cohen 0.97 vs. 0.74; PPV 96% vs. 76%; NPV 100% vs. 96%). Reliability analysis revealed measurement discrepancies in CZ and BZ areas across ENDO and EPI layers, with OT displaying the least systematic bias and the narrowest limits of agreement.

The high concordance observed between the two methods supports the diagnostic accuracy of both ADAS 3D and AHD-G EAM, particularly favoring the OT configuration. In our study, OT achieved unprecedented agreement rates compared to published data and demonstrated the lowest discrepancy in reliability analyses relative to UNI and BIP configurations. This success is attributed to OT technology's ability to overcome the limitations associated with bipolar blindness. While CMR remains the standard, OT EAM offers detailed insights into myocardial electrical vectors and may serve as an alternative for myocardial characterization in patients unable to undergo LGE-CMR.

Bland-Altman plot border zone