The Unipolar Electrogram Downslope as a Marker of Scar and Slow Conduction in Ventricular Tachycardia

Manisha Sahota1, Ursula Rohrer1, Fernando Campos2, Iulia Nazarov1, Sri Kousthubha Allampalli1, Janneke Burger3, Luca Azzolin4, Aurel Neic5, Ronak Rajani6, Pranav Bhagirath7, Gernot Plank8, Steven E Williams9, steven niederer1, Matthijs Cluitmans10, John Whitaker1, Martin Bishop1
1King's College London, 2UNESP, 3Amsterdam University Medical Center, 4Karlsruhe Institute of Technology, 5NumeriCor GmbH, 6Guy's and St Thomas' NHS Foundation, 7Kings College London, 8Medical University of Graz, 9University of Edinburgh, 10Maastricht University


Abstract

Background: Re-entrant ventricular tachycardia (VT) is often non-inducible or haemodynamically compromising, necessitating substrate mapping to evaluate electrophysiological properties of myocardium. Regions with abnormally slow conduction can form critical isthmuses that sustain VT and are key targets for catheter ablation. However, current mapping techniques remain limited in their ability to accurately characterise the diseased myocardial substrate.

Objective: We hypothesised that the minimum temporal derivative of the unipolar electrogram, dФe/dtₘᵢₙ, which reflects local tissue activation, could serve as a unique parameter for identifying arrhythmogenic myocardium, and evaluated its relationship with local myocardial conduction velocity (CV).

Methods: Electro-anatomic mapping (EAM) data from the left ventricular endocardium were analysed in 12 patients (8 with ischaemic and 4 with non-ischaemic cardiomyopathy) undergoing clinical VT catheter ablation. The dФe/dtₘᵢₙ was computed and its distribution compared in regions, defined by bipolar voltage, of scar (<0.5 mV), border zone (0.5-1.5 mV) and healthy tissue (>1.5 mV). Local CV was estimated from EAM local activation times using a triangulation approach and subsequently correlated with dФe/dtₘᵢₙ.

Results: A median of 2,037 EAM points were analysed (IQR:1,479-5,816) across the cohort. A consistent inverse linear association was observed between dФe/dtₘᵢₙ and local CV in all patients, with Spearman's correlation coefficients ranging from -0.49 to -0.06 (p<0.05), and a median of -0.22 (IQR:-0.30 to -0.18). Using bipolar voltage criteria to define tissue types, dФe/dtₘᵢₙ was significantly reduced in scar regions compared to border zone and healthy myocardium (-0.07±0.04 mV/ms vs. -0.48±0.21 mV/ms, p<0.001). Receiver operating characteristic (ROC) curve analysis identified an optimum dФe/dtₘᵢₙ threshold of -0.20 mV/ms for differentiating scar from healthy tissue, outperforming unipolar amplitude with an 88% sensitivity and 93% specificity (AUC=0.96, p<0.05).

Conclusion: Findings suggest dФe/dtₘᵢₙ could offer additional value for functional characterisation of the patient-specific pro-arrhythmic substrate and may complement established metrics in substrate mapping.