QRS duration (QRSd) and validated ECG parameters of ventricular repolarization associated with arrhythmic risk (T-peak-to-T-endTransmural , T-peak-to-T-endTotal , T-peak dispersion, QTc, QTc dispersion) were analyzed post-implant in 107 patients: 60 with CSP (HBP: n = 35, LBAP: n = 25) and 47 with BiVP. T-wave memory resolution and QTc shortening were analyzed on ECGs obtained ≥25 days post-implant. Twenty blinded measurements were obtained by both authors to assess Interobserver variability.
Although QRSd was shorter with HBP versus LBAP (119 ± 7 ms vs. 132 ± 9 ms, p = .02), there were no significant differences in any repolarization parameters between these methods of CSP. However, when comparing CSP (HBP + LBAP) to BiVP, both QRSd (125 ± 5 ms vs. 147 ± 7 ms, p < .0001) and repolarization parameters (T-peak-to-T-endTransmural : 83 ± 5 ms vs. 107 ± 8 ms; T-peak-to-T-endTotal : 110 ± 7 ms vs. 137 ± 10 ms; QTc: 470 ± 12 ms vs. 506 ± 12 ms; all p ≤ .0001) were significantly shorter with CSP. Improved T-peak-to-T-end values were unrelated to pre-implant QRSd or LV function. Interobserver variability was 4.6 ± 1.9 ms. Frontal QRS-T angle narrowing (132° to 104°, p = .001) and QTc shortening (483 ± 13 ms to 464 ± 12 ms, p = .008) were seen only with CSP.
In addition to improved depolarization, CSP reduced repolarization heterogeneity and provided greater T-wave memory resolution as compared to BiVP. Both modalities of CSP (HBP + LBAP) resulted in comparably reduced repolarization heterogeneity regardless of baseline QRSd and LV function. These observations may confer lower arrhythmogenic risk and warrant further study.