Substrate Mapping and Voltage Mapping

The amplitude of a local bipolar electrogram reflects the density and electrical coupling of myocytes beneath the catheter tip. Healthy myocardium is tightly packed with well-coupled cells. When a depolarization wave sweeps through, the coordinated charge movement produces a tall, sharp signal, typically above 1.5 mV.

Scar tells a different story. Where infarction or fibrosis has replaced muscle, fewer myocytes survive, and those that remain are poorly coupled, separated by collagen. The signal shrinks. In regions of dense scar, the bipolar voltage drops below 0.5 mV or disappears entirely. The catheter records near-silence.

Between these extremes lies the border zone: a patchwork of surviving muscle fibers interspersed with fibrosis. Signals here are reduced (0.5 to 1.5 mV) and often fractionated, reflecting the tortuous, zigzag conduction through surviving bundles. This is precisely the tissue that supports slow conduction and, by extension, reentrant VT.

The standard voltage cutoffs for ventricular substrate mapping come from Marchlinski and colleagues:

These thresholds apply to endocardial ventricular mapping. Atrial tissue produces smaller signals overall, so some groups use lower cutoffs: < 0.5 mV for atrial scar and 0.5 to 1.0 mV for the atrial border zone. The principle is the same; the numbers shift because atrial walls are thinner and contain less muscle mass.

We construct a voltage map by moving the catheter systematically across the endocardial surface, collecting one data point at a time. At each position, the mapping system records two things: the catheter's three-dimensional coordinates and the local bipolar voltage amplitude. After hundreds or thousands of points, the system reconstructs a 3D shell of the chamber.

Each point is then color-coded by voltage. The conventional palette runs from red or purple (dense scar, < 0.5 mV) through yellow (border zone) to green or teal (normal, > 1.5 mV). The result is a heat map of tissue health. Areas of red immediately show where infarction or fibrosis has destroyed muscle. The yellow fringe around the scar marks the border zone, the hunting ground for VT channels.

Modern multi-electrode catheters (such as the Pentaray or Orion basket) can acquire thousands of points in minutes, creating high-density maps that reveal fine anatomical detail. A single pass across the scar border can resolve channels as narrow as a few millimeters.

Voltage alone tells us where scar is. The next step is finding the conduction channels hidden within it. Two signal features point directly to surviving fiber bundles conducting slowly through fibrosis.

Late potentials are low-amplitude, high-frequency deflections that persist after the end of the surface QRS complex. In normal tissue, local activation finishes well within the QRS. When signals continue after the QRS ends, they represent muscle bundles so slowly conducting that they are still depolarizing while the rest of the ventricle has already finished. These are the channels a reentrant wavefront can use as its slow limb.

Local abnormal ventricular activities (LAVAs) are fractionated, multi-component signals recorded in the border zone and scar. Where a normal electrogram has a single sharp deflection, a LAVA may show three, four, or more distinct components separated by isoelectric gaps. Each component represents a separate fiber bundle activating at a different time, confirming that conduction through that region is fragmented and slow.

Both late potentials and LAVAs mark tissue that is neither fully dead nor fully alive. These signals are the electrical fingerprint of the substrate that VT needs to survive.

Once the voltage map and late potential survey are complete, we have a detailed blueprint of the scar architecture. Several strategies exist to eliminate the channels that support VT, all without needing the arrhythmia to be present.

Each strategy has the same underlying logic: eliminate the slow conduction channels the circuit requires. The choice depends on scar complexity, the number of potential VT morphologies, and the operator's assessment of risk. Scar homogenization is the most aggressive and may reduce VT recurrence most effectively, but it requires more ablation and carries higher procedural risk. Dechanneling is more conservative but demands precise identification of channel entrances.

Some VT circuits live on the outer surface of the heart, beyond the reach of endocardial catheters. This is common in non-ischemic cardiomyopathy, where fibrosis may be mid-myocardial or epicardial rather than subendocardial. Chagas disease is a classic example: the parasite damages the epicardium preferentially, creating substrate that endocardial mapping cannot see.

To reach the epicardium, we perform subxiphoid pericardial access. A needle is advanced from just below the xiphoid process into the pericardial space, the thin sac surrounding the heart. Once a guidewire is in place, a sheath is introduced, and the mapping catheter can slide freely over the outer surface of the ventricles.

Voltage thresholds on the epicardium differ from the endocardium. Epicardial fat attenuates the electrogram signal, so normal epicardial voltages may be lower than 1.5 mV even in healthy tissue. Some operators use 1.0 mV as the lower bound of normal for epicardial maps. Interpreting the epicardial voltage map therefore requires awareness of local fat thickness, which varies by location (the lateral and inferior walls tend to carry more epicardial fat).

Epicardial ablation carries unique risks: the phrenic nerve runs along the lateral pericardium and can be paralyzed by ablation energy, and coronary arteries course over the epicardial surface. Before delivering energy, we must confirm the catheter is more than 5 mm from any major coronary branch using fluoroscopy or coronary angiography.