Author + information
- Received July 4, 2005
- Revision received September 7, 2005
- Accepted December 1, 2005
- Published online June 20, 2006.
- Shajil Chalil, MRCP2,
- Zaheer R. Yousef, MD, MRCP,
- Sarkaw A. Muyhaldeen, MRCP1,
- Russell E. A. Smith, MD, FRCP3,
- Paul Jordan, FRCP3,
- Christopher R. Gibbs, MD, MRCP and
- Francisco Leyva, MD, FRCP3,4,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Francisco Leyva, Department of Cardiology, Good Hope Hospital, Rectory Road, Sutton Coldfield, West Midlands B75 7RR, United Kingdom.
Objectives This study was designed to determine whether cardiac resynchronization therapy (CRT) by means of biventricular pacing (BiVP) alters the QT interval (QTc) and QT dispersion (QTD), and whether such changes relate to the risk of developing major arrhythmic events (MAE).
Background Prolonged QTcis associated with MAE. Left ventricular pacing and BiVP alter QTc.
Methods A total of 75 patients with drug-resistant heart failure (New York Heart Association functional class III/IV) and QRS duration ≥120 ms underwent CRT. The QTcand QTD were measured before and 48 days after BiVP.
Results Over 807 days (range 93 to 1,543 days), 11 patients had a MAE. Compared to baseline, at 48 days after CRT, QTD increased in 47% of patients and QTcdecreased in 53%. The QTcat follow-up was higher in MAE patients compared with no-MAE patients (35.9 ± 14.2 ms vs. 0.52 ± 6.0 ms; p = 0.0323). Similar differential responses for QTD were observed (46.4 ± 13.5 ms in MAE vs. −5.1 ± 4.1 ms in no MAE, p < 0.0001). The MAE occurred in 29% of patients exhibiting an increase in QTD and in 3% of those exhibiting a decrease (p = 0.0017). In multiple regression analyses, change in QTD from baseline (ΔQTD) strongly predicted MAE, independent of ΔQTc, QRS duration, and left ventricular ejection fraction and end-diastolic volume (p < 0.001). Differences in survival curves were observed when patients were dichotomized according to whether QTD increased or decreased in relation to baseline values (p < 0.0001).
Conclusions The MAE in patients with BiVP are related to pacing-induced increases in QTD. Measures of ventricular repolarization at the time of pacemaker implantation may guide selection of patients for combined CRT and defibrillator therapy.
The Cardiac Resynchronization Heart Failure (CARE-HF) study has recently shown that in patients with heart failure (HF), cardiac resynchronization therapy (CRT) using biventricular pacing (BiVP) leads to a 36% reduction in all-cause mortality (1). This study also showed that, despite a reduction in all-cause mortality, CRT did not reduce the rate of sudden cardiac death (35% in the CRT group vs. 32% in the medical therapy group). This finding is in keeping with a recent meta-analysis of randomized trials of CRT showing that although BiVP reduces death from progressive HF, death from causes other than pump failure may be increased (2). In this respect, it has been suggested that BiVP may enhance arrhythmogenicity by reversing the normal depolarization pattern from endocardium to epicardium, which enhances transmural dispersion of repolarization and propagation of early after-repolarizations (3). These factors are known to facilitate the development of ventricular tachyarrhythmias (4,5). In this light, we hypothesized that pacing-induced changes in QT interval (QTc) and QT dispersion (QTD) relate to the risk of sudden cardiac death in patients undergoing CRT.
Data presented pertains to consecutive patients referred to our service and undergoing successful BiVP implantation between October 2000 and December 2003. The following inclusion criteria were adopted: 1) HF from any cause; 2) moderate-to-severe HF (New York Heart Association [NYHA] functional class III or IV) despite optimal medical treatment; 3) QRS duration ≥120 ms; and 4) left ventricular ejection fraction (LVEF) ≤35%. Exclusions included: 1) contraindications to cardiac pacing; 2) myocardial infarction or acute coronary syndrome within the previous 3 months; 3) presence of or indications for an implantable cardioverter-defibrillator (ICD); 4) severe structural valvular heart disease, 5) presence of comorbidities likely to threaten survival for 12 months; and 6) pulmonary edema requiring intravenous diuretics in the previous week. All patients gave written informed consent, and the study was approved by the North Birmingham Ethics Committee.
Patients referred for CRT underwent pacemaker implantation during an elective admission or after stabilization during the course of an unplanned admission for acute decompensated HF. Patients underwent a comprehensive clinical assessment that included assessment of NYHA functional class; a six-min hall walk test (6); a quality-of-life assessment using Minnesota Living with Heart Failure questionnaire (7); and transthoracic echocardiography on the day before implantation and at one month, three months, and every six months thereafter.
Major arrhythmic events (MAE)
Mortality data was collected through medical records and, where appropriate, from interviews with patients’ carers. Sudden cardiac death was defined as “a natural, unexpected death due to cardiac causes, heralded by an abrupt loss of consciousness within one hour of the onset of acute symptoms” (8). For the purposes of this study, MAE were defined as the combined rate of sudden cardiac death and/or resuscitation from a potentially fatal ventricular tachyarrhythmia.
Transvenous biventricular pacemaker implantation was undertaken using standard techniques under local anaesthesia. With reference to the 45° left anterior oblique radiographic view, coronary sinus leads were positioned in the following o’clock positions: 3 (61 %), 4 (20%), 5 (1.7%), 6 (11.9%), 9 (15.2%), and 11 (3.4%). There were five cases of left ventricular (LV) lead dislodgement and no device failures. Patients were entered in the study only after a successful implantation and were followed up in a dedicated CRT clinic. Patients in sinus rhythm (n = 56) underwent transmitral Doppler-directed optimization of atrioventricular delay (9) before discharge and at every scheduled visit thereafter. Backup atrial pacing was set at 60 beats/min, and the pacing mode was set to DDDR with an interventricular delay of 4 ms. For patients in chronic atrial fibrillation (n = 19), right ventricular (RV) and LV leads were implanted. In six of these cases, a standard dual-chamber generator was used, in which the LV lead was connected to the “atrial” port and the RV lead to the ventricular port, and the generator was programmed to VVT mode with the shortest sensed LV-to-RV delay of 30 ms. In the remainder of the patients with atrial fibrillation, a Medtronic InSync III generator (model 8042) was used, plugging the atrial port and programming the generator to a ventricular triggered mode. Generators used included the InSync models 8040 (n = 13) and 8042 (n = 32) (Medtronic Inc., Minneapolis, Minnesota), Frontier (n = 7) (St. Jude Medical, St. Paul, Minnesota), CRT 8000 (n = 1) (Vitatron, Arnhem, the Netherlands), Stratos (n = 4) (Biotronik, Berlin, Germany), Contak TR model 1241 (n = 2) and Renewal TR2 (n = 2) (Guidant Corp., Indianapolis, Indiana) and Medtronic Sigma DR (n = 1) (Medtronic Inc.).
Electrocardiograms (ECG) and echocardiography
Resting 12-lead ECGs were used to assess ventricular repolarization. The QT intervals at baseline and 1 month (mean 48 days) in each of the six precordial leads (V1to V6) were manually measured retrospectively for every patient by an experienced investigator blinded to the patients’ final outcome (10). The QTcfor each lead represented the mean of two consecutive QTcs, measuring from the beginning of the QRS complex to the visual return of the T-wave to the isoelectric line. Where the T-wave was complicated by a U-wave, the end of the T-wave was defined as the nadir between the T- and U-wave (11). The QT intervals were corrected for heart rate (QTc) (12). The QTD for each ECG recording was defined as the difference between the longest and shortest measured QT (13). Assessment of the ECG was done during the intrinsic rhythm at baseline and during biventricular pacing at follow-up.
Two-dimensional echocardiography was performed using the VIVID System 5 (General Electric, Milwaukee, Wisconsin). Standard left parasternal long-axis and short-axis, and apical, four-, five-, and two-chamber views were obtained. Digital images were stored for offline analysis (EchoPAC, Vingmed-General Electric). The LV volumes were estimated through Simpson’s equation by planimetry of apical four-chamber views. The frame at the beginning of the QRS complex was used to estimate left ventricular end-diastolic volume (LVEDV), whereas the frame with the smallest ventricular area just before mitral valve opening was used to estimate left ventricular end-systolic volume (LVESV). We derived LVEF from end-systolic volume and end-diastolic volume using standard formulae.
Continuous variables are expressed as mean ± SEM for normally distributed variables. Comparisons between normally distributed continuous variables were made using the Student ttest. The Mann-Whitney Utest was used for analysis of non-normally distributed data. The Cox proportional-hazards model was used to examine the risk of MAE in relation to baseline QTcand QTD as well as subsequent changes in these parameters (ΔQTcand ΔQTD) between baseline and the first clinic review at 48 days. Variables showing significant group differences were entered into multivariate regression models. The relationship between ΔQTD and survival was explored using Kaplan-Meier survival analysis. Differences in survival curves between the groups were assessed using the log-rank (Mantel-Cox) test. Differences between categorical variables were analysed using the chi-square test. Statistical analyses were performed using Statview (Cary, North Carolina). A two-tailed p value of <0.05 was considered statistically significant.
The characteristics of the study group are shown in Table 1.Compared with the no-MAE group, the MAE group had a higher NYHA functional class (p = 0.006), longer QRS complexes (p = 0.034) as well as higher LVESV (p = 0.010) and LVEDV (p = 0.022), and a lower LVEF (p = 0.025). There were no differences in heart rate before CRT. Heart rate increased by 6.58 beats/min at follow-up (p = 0.0023), but there was no difference between the MAE and no-MAE groups. No significant change in the ventricular response was observed in patients with atrial fibrillation (83.1 ± 5.3 beats/min vs. 87.8 ± 3.4 beats/min, p = NS).
After a follow-up period of 807 days (range 93 to 1,543 days), no differences from baseline in six-min walk test, quality of life, or NYHA functional class were observed in patients with MAE and no-MAE (Table 2).Responders, defined as patients who exhibited an increase in 6-min walk test distance above baseline at follow-up, were 7 (64%) in the MAE group and 41 (64%) in the non-MAE group (NS). By the end of the follow-up period, 11 patients had a MAE: 10 patients died suddenly and 1 was successfully resuscitated from ventricular fibrillation. The MAE occurred at a mean of 515 days (range 93 to 917 days) after delivery of CRT.
As shown in Figure 1,comparison of the groups with MAE and no MAE revealed no differences in baseline QTc. Although QTD was lower in the MAE group, this was not statistically significant. In the whole study sample, no changes in QTc(484.7 ± 5.3 ms vs. 491.4 ± 5.1 ms, p = NS) or QTD (47.3 ± 3.1 ms vs. 49.9 ± 3.7 ms, p = NS) were observed 48 days after implantation. When patients were grouped into MAE and no-MAE groups, however, significant differences emerged in the change at follow-up in QTc(ΔQTc, MAE: 35.9 ± 14.2 ms, no MAE: 0.52 ± 6.0 ms; p = 0.0323) and QTD (ΔQTD, MAE: 46.4 ± 13.5 ms, no MAE −5.1 ± 4.1 ms, p < 0.0001). In the whole-study sample, 47% exhibited an increase in QTD above baseline at follow-up, whereas 53% exhibited a decrease. The MAE occurred in 29% of patients exhibiting an increase in QTD above baseline and in 3% of those exhibiting a decrease (p = 0.0017). No differences in baseline QTcor QTD or changes in these measures at follow-up emerged when patients were grouped according to medical therapy with diuretics, angiotensin-converting enzyme inhibitors/angiotensin receptor antagonists, beta-blockers, or amiodarone. As shown in Figure 2,QRS duration at baseline correlated positively with ΔQTcfollowing implantation, but not with ΔQTD.
At baseline, there was no group difference in the Tpeak-endinterval. Following implantation, however, there was an overall reduction in the Tpeak-endinterval (−16.5 ± 5.0 ms for the whole cohort), which was more marked in the no-MAE group than in the MAE group (−20.0 ± 5.4 ms and −1.5 ± 12.8 ms, respectively, p = 0.047). Changes in the Tpeak-endinterval, however, failed to emerge as a significant predictor of MAE in Cox proportional hazards analyses.
In a Cox proportional hazards analyses, baseline QTcor QTD did not emerge as predictors of MAE. As shown in Table 3,ΔQTD emerged as a significant predictor of MAE independent of ΔQTcas well as baseline QRS duration, LVEF, and LVEDV. For Kaplan-Meier survival analyses, the pooled sample was stratified according to varying thresholds of QTD. The most significant difference in survival curves was observed when patients were dichotomized according to whether QTD increased or decreased from baseline values (Fig. 3).
We have shown that in patients with moderate-to-severe HF undergoing CRT, QTcand QTD at baseline are not predictors of MAE. The novel finding of this study is that pacing-induced increases in QTcand QTD are strong predictors of subsequent MAE in such patients. An increase in QTD above baseline values confers the greatest risk of future MAE. As expected from other studies (14), patients with MAE had a worse NYHA functional class, longer QRS complexes, a lower LVEF, and higher LVEDV and LVESV. The relationship between QTD and MAE, however, was independent of these variables.
Although QTD has been shown to predict mortality in population-based studies (15) and in patients with myocardial infarction (16) and HF (17), it has been generally disappointing as a predictor of arrhythmic events. Our findings of a pacing-induced increase in QTcand QTD is consistent with experimental studies showing that reversal of myocardial depolarization through epicardial stimulation, as occurs in LV pacing via the coronary sinus, promotes arrhythmogenesis. Using arterially perfused canine LV wedge preparations, Fish et al. (5) observed that reversing the direction of LV wall activation leads to an increase in QTcduration. This has been linked to increased transmural dispersion of repolarization resulting from earlier repolarization of the epicardium and delayed activation and repolarization of the midmyocardial M cells (5). Similar findings have been reported in clinical studies. In a series of patients with HF undergoing CRT, Medina-Ravell et al. (3) observed increases in the QT and JT interval durations following LV epicardial and BiVP, but not with isolated RV endocardial pacing.
On the other hand, several studies have suggested that BiVP exerts an antiarrhythmic effect. In a randomized crossover study using 24-h Holter monitoring, Paul et al. (18) showed that BiVP significantly reduced ventricular ectopic counts when compared to isolated RV pacing. In a cohort of participants of the Ventak trial (single-blinded, randomized, crossover study of BiVP plus ICD for three months, followed by three months of no pacing plus ICD), Higgins et al. (19) observed a lower frequency of antitachycardia therapies during BiVP compared to the periods of no pacing (16% vs. 34% respectively; p = 0.04). In contrast, the meta-analysis of CRT trials of Bradley et al. (2) showed that whereas CRT significantly reduced death from progressive HF, CRT tended to promote sudden, presumed arrhythmic, deaths. In our study, 47% of patients exhibited a pacing-induced increase in QTD above baseline, whereas 53% exhibited a decrease. This raises the possibility that BiVP has differential effects on the arrhythmogenic substrate, antiarrhythmic in some and arrhythmogenic in others. It is noteworthy that other treatments, such as class I antiarrhythmic agents, also have a differential effect on the arrhythmogenic substrate (20). In contrast, a study of 25 patients using a high-resolution 65-lead surface ECG (21) has shown that BiVP leads to a reduction in QTD from 114 ± 22 ms (sinus rhythm) to 90 ± 12 ms. The standard deviations quoted in this study, however, raise the possibility that some patients may have exhibited an increase in QTD.
With regard to the Tpeak-endinterval, Berger et al. (21) found that the biventricular pacing leads to an overall reduction of 81 ± 13.8 ms (mean ± SD) compared to sinus rhythm. It would appear from these standard deviations that a reduction in Tpeak-endinterval was not found in all patients. We found that the reduction was significantly less pronounced in the MAE group. The discrepancies between our findings and those of Berger et al. (21) may relate to the different methodologies used to collect ECG data or to the time difference between ECG recordings following pacing. The salient finding from our study is that CRT has a differential effect on the Tpeak-endinterval and that this is related to the development of MAE.
These findings may be relevant to the observation from the CARE-HF study that CRT reduces all-cause mortality but not sudden cardiac death (1). This raises the possibility that the favorable effects on mortality are negated by an increase in the risk of death from non-HF causes, among which are fatal arrhythmias. Whether pacing-induced arrhythmias in susceptible patients could be responsible for the apparent dilution of the survival benefit from CRT requires further investigation.
Examination of secondary end point data from the Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) trial (14) shows that all-cause mortality was lower in the pacemaker and pacemaker-defibrillator groups than in the pharmacological group after 12 months (15% and 12% compared to 19%, p = 0.059 and p = 0.003, respectively). Interestingly, the survival curves for all-cause mortality begin to separate between day 270 and 360 after randomization. This suggests that the benefit from ICD therapy is late and is in keeping with our data showing that the risk of MAE in CRT patients who developed pacing-induced increases in QTD also becomes apparent late after delivery of CRT, between day 350 and 450 after implantation.
This study has several limitations. First, it is small and observational. A relatively long follow-up period, however, has yielded sufficient MAE to achieve statistical significance in survival analyses. Second, the shortcomings of the standard definition of sudden cardiac death used in this study are well recognized. Thirdly, measurement of the QTccan be inaccurate. Manual measurement of QTc, however, is as reproducible as automatic measurement (10). We have adopted the horizontal plane leads V1to V6because these are more likely to reflect real local activity (10).
Our findings have emerged in the context that CRT delays death from progressive HF, but not from all causes. We have shown that CRT has differential effects on QTD, leading to a reduction in some patients and to an increase in others. Patients who exhibit an increase in QTD following CRT are at increased risk of MAE compared with those who exhibit a decrease. Further studies are needed to determine whether evaluation of QTD at the time of biventricular pacemaker implantation can be used to risk-stratify patients for the preferential use of CRT with ICD backup.
We are grateful to Mrs. Lisa Ball and Mrs. Janet Brashaw-Smith for their help with echocardiography and pacemaker follow-up. We are also grateful to Mr. Nick Irwin for his assistance.
↵1 Dr. Muyhaldeen held a research fellowship sponsored by Medtronic Inc.
↵2 Dr. Chalil holds a research fellowship sponsored by Medtronic Inc.
↵3 Drs. Smith, Jordan, and Leyva have received sponsorship from Medtronic Inc.
↵4 Dr. Leyva has received sponsorship from St. Jude Inc.
The first two authors contributed equally to this work.
- Abbreviations and Acronyms
- biventricular pacing
- Cardiac Resynchronization Heart Failure study
- cardiac resynchronization therapy
- heart failure
- implantable cardioverter-defibrillator
- left ventricle/ventricular
- left ventricular end-diastolic volume
- left ventricular ejection fraction
- left ventricular end-systolic volume
- major arrhythmic events
- New York Heart Association
- rate-corrected QT interval
- QT dispersion
- right ventricle/ventricular
- Received July 4, 2005.
- Revision received September 7, 2005.
- Accepted December 1, 2005.
- American College of Cardiology Foundation
- Medina-Ravell V.A.,
- Lankipalli R.S.,
- Yan G.-X.,
- et al.
- Yan G.X.,
- Antzelevitch C.
- Fish J.M.,
- Di Diego J.M.,
- Nesterenko V.,
- Antzelevitch C.
- Guyatt G.H.,
- Sullivan M.J.,
- Thompson P.J.
- Rector T.S.,
- Kubo S.H.,
- Cohn J.N.
- Myerburg R.J.,
- Castellanos A.
- Ritter P.,
- Padeletti L.,
- Gillio-Meina L.,
- Gaggini G.
- Coumel P.,
- Maison-Blanche P.,
- Badilini F.
- Brendrop B.,
- Elming H.,
- Jun L.,
- et al.
- Bazett H.C.
- Day C.P.,
- McComb J.M.,
- Campbell R.W.F.
- Bristow M.R.,
- Saxon L.A.,
- Borehmer J.,
- et al.,
- Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) Investigators
- de Bruyne M.C.,
- Hoes A.W.,
- Kors J.A.,
- Hofman A.,
- van Bremmel J.H.,
- Grobbee D.E.