Background: Cardiac resynchronization therapy (CRT) has emerged as an effective treatment strategy for patients with advanced heart failure (New York Heart Association classes II–IV heart failure, left ventricular ejection fraction (LVEF) ≤ 35%, and a QRS duration ≥ 120 ms). Approximately 30% of patients do not respond favourably to CRT. In recent years, the proportion of patients who respond favourably to CRT did not increase significantly. CRT is merely electrical therapy that improves left ventricular (LV) mechanical dyssynchrony by changing the sequence of electrical activation and restoring electrical synchrony. Therefore, ECG parameters of conduction anomalies and electrical dyssynchrony play an evolving and important role in understanding the mechanism of CRT response. Little is known about the effect of CRT on ventricular repolarization parameters. There are some data about the increase repolarization heterogeneity after CRT and potentially its proarrhytmic effect. On the other hand, some studies found amelioration of repolarization heterogeneity after CRT. However, interdependence between electrical and mechanical remodelling and time course of electrical remodelling have not been fully investigated. The present thesis evaluated the value of different ECG parameters in heart failure patients treated with CRT and clarified the time course and relationship between electrical and mechanical reverse remodelling after CRT.
Methods: The study population was selected from heart failure patients treated with CRT according to contemporary guidelines in University medical centre Ljubljana. All patients included in the study had transthoracic two-dimensional echocardiography performed at baseline and at follow-up visits (1, 3, 6, 9 and 12 months after implantation). Resting supine 12-lead ECGs and 20-minutes high-resolution Holter ECG were recorded before and after CRT implantation. At follow-ups, CRT was temporarily reprogrammed to VVI 40 bpm to allow native conduction. An absolute decrease in native pre-implant QRS duration ≥ 10 ms was defined as significant electrical remodelling (ER). Relative change in QRS duration was defined as the difference between post-implant QRS duration and baseline QRS duration divided by baseline QRS duration expressed in percent. All ventricular repolarization parameters (T-wave amplitude variability [TAV], QTa interval [QT apex], QTe interval [QT end], dispersion of QT, Tpeak to Tend interval [TpTe], and TpTe /QT ratio) were assessed from high-resolution 20-minutes ECG recordings. Ventricular tachyarrhythmia episodes (VTs) were classified as sustained ventricular tachycardia with appropriate ICD therapy or ventricular fibrillation. Heart failure hospitalization, cardiovascular death and cardiac transplantation were used as a combined end point to observe clinical outcome.
Results:
I. The association of electrocardiographic parameters from 12-leads ECG and response to CRT
The study population consisted of 101 patients with a mean age of 63.2 ± 10.9 years (66 (65.2%) males, 37 (36.6%) ischaemic aetiology, 77 patients (76.2%) in NYHA class III). After 12 months of biventricular pacing, 32 patients (31.7%) fulfilled the echocardiographic criteria of super-response (a relative reduction in LVESV ≥ 30%). There were no statistically significant differences in baseline ECG parameters when super-responder group was compared with non-super-responder group. Post-implant paced QRS duration was shorter in super-responder group (148 ± 22 ms vs. 162 ± 28 ms; P = 0.010). Additionally, only in super-responders significant QRS reduction was observed after implantation (super-responders: from 167 ± 26 ms to 148 ± 22 ms; P=0.010 vs. non-super-responders: from 165 ± 27 ms to 162 ± 28 ms; P = 0.536). Relative shortening of QRS complex was significantly higher in super-responders (12.1% (6.8-22.2) vs. 1.7% ([-11.9] to 11.8); P = 0.005). Only relative QRS shortening remained independently related to super-response to CRT in multivariable model adjusting for NYHA class, normal axis and LBBB with mid-QRS notching.
II. The effect of cardiac resynchronization therapy on ventricular repolarization parameters and their relation to ventricular tachyarrhythmias
Sixty-four patients (mean age of 63.9 ± 10.9 years, 47 (73%) males, 23 (36%) ischaemic cardiomyopathy) were enrolled in the analysis. Response to CRT, predefined as a relative reduction in LVESV ≥ 15%, was observed in 33 patients (51.6%). There were significant changes in repolarization parameters of native conduction over 12 months after CRT implantation. Significant changes over time was observed in QTec (P < 0.001), TpTe (P < 0.001), TpTe/QT ratio (P < 0.001) and dispersion of QTe (P < 0.014), whereas TAV values and QTa intervals were unchanged over first year after CRT. All repolarization parameters showed a similar pattern of changes with the increased repolarization heterogeneity in first months after CRT implantation, which then declined over time. Significant differences in repolarization parameters were observed between patients with and those without echocardiographic response to CRT. In responder group, despite significant prolongation in the first months, decline of repolarization parameters were noted during further follow-ups. On the other hand, in non-responder group progressive increase of repolarization parameters at follow-ups compared with pre-implant value were observed. Responder and non-responder group had similar pre-implant value of all repolarization parameters. Significant difference between groups in repolarization parameters was detected after 6 months of biventricular pacing. Appropriate device therapy was observed in 10 patients (15.6%) within 1 year after CRT implantation. More than half of the patients (60%) had VTs within the first month after CRT implantation. Distribution of VTs with the highest occurrence in the first month corresponded with the highest repolarization heterogeneity in the same period.
III. Time course and relationship between electrical and mechanical reverse remodelling after CRT
The study population included 62 patients (50 (81%) males, with mean age of 65.7 ± 10.3 years). There were significant changes in native QRS duration during 12 months after implantation (P = 0.003). Significant shortening of QRS duration was already observed at 1 month after CRT (pre-implant vs. 1 month: 185ms (175-194) vs. 180ms (170-186); P < 0.05), which persisted during further follow-ups. On the other hand, progressive LV structural remodelling was observed during 12 months with continuous reductions in LV volumes and improvement of LVEF (P < 0.001, for all). However, significant echocardiographic changes were observed only after 3 months of CRT (EDV: 231 ml (191-280) vs. 200 ml (167-267), P < 0.05; ESV: 167 ml (137-206) vs. 140 ml (112-196), P < 0.05; EF: 27% (24-31) vs. 31% (24-38), P < 0.05). In patients with ER, the reductions in QRS duration were already pronounced after one month of CRT (pre-implant vs. 1 month: 190 ms [179-196] vs. 180 ms [173-186]; P < 0.001) and persisted over follow-up. There were no changes in QRS duration during follow-ups in patients without ER. One month after implantation, relative changes in ESV were similar in both groups. At later follow-ups, patients with ER demonstrated progressive reverse LV remodelling compared with patients without ER. During a median follow-up of 27 months (18–37 months), the combined end point was observed in 15 (24.2%) patients. Compared with patients without ER, a superior survival free from combined end point was observed among patients with ER (chi-squared = 4.85, log-rank P = 0.028).
Conclusion: Some baseline ECG parameters from 12-leads ECG recording are associated with increased degree of echocardiographic response after CRT. Absolute post-implant QRS duration and acute relative shortening of QRS duration after initiation of CRT were correlated with super-response. However, there were dynamic changes of repolarization parameters of native conduction during 12 months after CRT. Increase repolarization heterogeneity in the first month after CRT corresponded with the high rate of VTs in the same periods. Additionally, electrical remodelling of native conduction is achieved before mechanical remodelling. Patients with electrical remodelling had better clinical outcome compared with patients without electrical remodelling.
|