Background: Evidence supporting methodologies to measure local repolarisation time (RT) in the intact human heart using intracardiac electrograms is based on a limited number of studies reporting conflicting results. The Wyatt method (RTWyatt = dV/dtmax for both positive and negative T-waves, see Figures A–B) has a solid theoretical background, has been proven to correlate with ERP in patients and has been validated in animal studies. However, the alternative method (RTAlt = dV/dtmin for positive T-waves, RTAlt = dV/dtmax for negative T-waves, see Figures A–B) has been suggested to provide better correlation to RT derived from monophasic action potentials (MAP) in human hearts.
Aims: To compare the Wyatt versus alternative methods in the intact human heart using: (A) Effective refractory period (ERP) obtained during programmed ventricular stimulation and (B) Monophasic action potential like signals (MAP-like) serendipitously recorded during catheter ablation for ventricular tachycardia.
Methods: Local ERP was estimated during S1S2 pacing (S1S1 = 600 ms, 8 consecutive stimulations) established from the distal pole of a decapolar catheter placed in the RV (n = 10), LV (n = 10) and CS (n = 4) in 11 patients with structurally normal ventricles undergoing EPS for supraventricular tachycardia ablation. Unipolar electrograms were simultaneously recorded from the adjacent electrode and RTWyatt and RTAlt were measured.
Furthermore, unipolar electrograms recorded from the distal pole of ablation and diagnostic catheters during electroanatomical mapping in 50 patients undergoing catheter ablation for ventricular tachycardia were used to create unfiltered bipolar electrograms, which were analysed with a customised algorithm to identify MAP-like signals. MAP-like signals and repolarisation markers (RT50, RT70 and RT90, i.e. instant at which the recovery phase of a MAP-like signal reached 50%, 30% and 10% of its amplitude) were further revised using customised Graphical User Interfaces. Analyses of RTWyatt and RTAlt were performed blindly to ERP and markers from MAP-like signals.
Results: In total, ERP was estimated in 24 cardiac sites. RTWyatt but not RTAlt provided a good approximation of local ERP (ERP-RTWyatt = –2.7 ± 13.9 ms, ERP-RTAlt = –68.9 ± 14.3 ms, see Figure A). In total, 310 MAP-like signals were identified in 27 patients. Markers of RT from MAP-like signals (RT50, RT70 and RT90) showed higher Spearman’s correlation coefficient for RTWyatt than RTAlt (0.85 versus 0.70, 0.82 versus 0.65, 0.86 versus 0.70, respectively). As shown in Figure B, RTWyatt showed significantly lower mean absolute error than RTAlt for RT50 (31.9 ± 29.5 versus 74.3 ± 45.1 ms, respectively) and RT70 (24.6 ± 23.3 versus 46.6 ± 3.9.1 ms, respectively), whereas RTAlt showed lower mean absolute error than RTWyatt for RT90
(43.2 ± 25.2 versus 31.1 ± 28.2 ms, respectively).
Conclusions: This comprehensive assessment of RT markers in the intact human heart shows that (1) The Wyatt method provides RT estimates ranging between RT50 and RT70, while the Alternative method provides RT estimates exceeding APD90. (2) Correlation to RT markers derived from MAP-like signals is higher for the Wyatt than the alternative method. (3) The Wyatt, but not the alternative, method provides an accurate marker of local refractoriness.