Application of ripple mapping to visualize slow conduction channels within the infarct-related left ventricular scar

BACKGROUND: Ripple mapping (RM) displays each electrogram at its 3-dimensional coordinate as a bar changing in length according to its voltage-time relationship with a fiduciary reference. We applied RM to left ventricular ischemic scar for evidence of slow-conducting channels that may act as ventricular tachycardia (VT) substrate. METHODS AND RESULTS: CARTO-3(c) (Biosense Webster Inc, Diamond Bar, CA) maps in patient undergoing VT ablation were analyzed on an offline MatLab RM system. Scar was assessed for sequential movement of ripple bars, during sinus rhythm or pacing, which were distinct from surrounding tissue and termed RM conduction channels (RMCC). Conduction velocity was measured within RMCCs and compared with the healthy myocardium (>1.5 mV). In 21 maps, 77 RMCCs were identified. Conduction velocity in RMCCs was slower when compared with normal left ventricular myocardium (median, 54 [interquartile range, 40-86] versus 150 [interquartile range, 120-160] cm/s; P<0.001). All 7 sites meeting conventional criteria for diastolic pathways coincided with an RMCC. Seven patients had ablation colocating to all identified RMCCs with no VT recurrence during follow-up (median, 480 [interquartile range, 438-841] days). Fourteen patients had >/=1 RMCC with no ablation lesions. Five had recurrence during follow-up (median, 466 [interquartile range, 395-694] days). One of the 2 patients with no RMCC locations ablated had VT recurrence at 605 days post procedure. RMCCs were sensitive (100%; negative predictive value, 100%) for VT recurrence but the specificity (43%; positive predictive value, 35.7%) may be limited by blind alleys channels. CONCLUSIONS: RM identifies slow conduction channels within ischemic scar and needs further prospective investigation to understand the role of RMCCs in determining the VT substrate.