Adjusted Endocardial Voltage Cut-Offs Proposed for Postinfarct LV Remodeling

Proposed reference cutoffs adjusted to define “normal” unipolar (UV) and bipolar (BV) voltage in patients with left ventricular (LV) remodeling better match the total infarct scar to provide an improved approach to electroanatomic voltage mapping (EAVM), according to study results published in the Journal of the American College of Cardiology: Clinical Electrophysiology.

The investigators of this study sought to determine the effect of LV structural remodeling on normal endocardial voltages of noninfarcted myocardium (NIM) in patients with postinfarct scar and to define new reference cutoffs of normal BV and UV adjusted for LV remodeling.

The study included 27 patients with postinfarct scar who underwent catheter ablation with real-time integration late gadolinium-enhanced (LGE) magnetic resonance imaging (MRI). Participants were categorized as having a nonremodeled LV (n=15) or a remodeled LV (n=12), which was defined by end-systolic volume index >50 m¹/m² with ejection fraction <47%. To discriminate between the scar area and relatively healthy NIM, all participants underwent EAVM, which was merged with scar meshes derived from LGE-MRI.

The investigators determined reference BVs and UVs by calculating the pooled NIM segments from all patients; they used fifth-percentile values to determine the cutoffs for “normal” BV/UV adjusted for LV remodeling.

Endocardial voltages of NIM in postinfarct patients were lower in patients with remodeled vs nonremodeled LV: the new cutoff references for normal BV and UV (fifth percentile values) were ≥2.1 mV and ≥6.4 mV in patients with remodeled LV and ≥3 mV and ≥6.7 mV in patients with nonremodeled LV, respectively.

Endocardial low-voltage areas defined by the adjusted cutoff values corresponded better to LGE-detected scar than low-voltage areas defined by uniform cutoffs, which underestimated total scar and scare core in both patient groups.

Participants who underwent successful ablation of ventricular tachycardia (n=15) contained >97% of all evoked delayed potentials applying the low-voltage area defined by remodeling-adjusted cutoffs; significantly smaller endocardial surface areas were covered when using uniform cutoffs.

Finally, whole-heart T1-mapping revealed more fibrotic NIM in patients with remodeled vs nonremodeled LV, which may correspond to the marked decrease of endocardial voltage in patients with remodeled vs nonremodeled LV.

A limitation to the study included underrepresentation of NIM segments from certain regions, as most patients presented with inferior/inferolateral scarring. The researchers determined derivation and validation of cutoff values from the same dataset, which may have introduced bias in the comparison of low-voltage area and scar detection by LGE-MRI.

The researchers of the study concluded that the proposed reference cutoffs to define “normal” LV endocardial voltage for LV remodeling postinfarct scar provide a more tailored approach to LV EAVM; the best match between endocardial low-voltage areas and total scar was achieved when remodeling-adjusted cutoffs were applied.

Reference

Sramko M, Abdel-Kafi S, Van der Geest RJ, et al. New adjusted cutoffs for “normal” endocardial voltages in patients with post-infarct LV remodeling [published online August 28th, 2019]. JACC Clin Electrophysiol. doi:10.1016/j.jacep.2019.07.007