Author + information
- Received February 2, 2017
- Revision received May 17, 2017
- Accepted June 9, 2017
- Published online December 18, 2017.
- Freddy Del-Carpio Munoz, MD, MSca,∗ (, )
- S. Michael Gharacholou, MD, MSca,
- Christopher G. Scott, MSb,
- Vuyisile T. Nkomo, MD, MPHa,
- Francisco Lopez-Jimenez, MDa,
- Yong-Mei Cha, MDa,
- Thomas M. Munger, MDa,
- Paul A. Friedman, MDa,c and
- Samuel J. Asirvatham, MDa,d
- aDivision of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- bDivision of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, Minnesota
- cDepartment of Physiology and Biomedical Engineering, Mayo Clinic, Rochester, Minnesota
- dDivision of Pediatric Cardiology and Department of Physiology and Biomedical Engineering, Mayo Clinic, Rochester, Minnesota
- ↵∗Address for correspondence:
Dr. Freddy Del-Carpio Munoz, Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street Southwest, Rochester, Minnesota 55905.
Objectives The goal of this study was to evaluate whether prolonged ventricular conduction (paced QRS) and repolarization (paced QTc) times observed during ventricular stimulation predict ventricular arrhythmic events and death.
Background Abnormal ventricular conduction and repolarization can predispose patients to ventricular arrhythmias.
Methods Consecutive patients with left ventricular dysfunction (ejection fraction <50%) undergoing electrophysiology studies from January 2002 until May 2014 were identified at Mayo Clinic (Rochester, Minnesota). Patients were followed up until December 2014 for occurrence of ventricular arrhythmias and death.
Results Among the 501 patients included (mean age 65 years; mean left ventricular ejection fraction 33.1%), longer paced ventricular conduction was associated with longer baseline QRS duration, longer QT interval, and lower ejection fraction. On multivariable analysis, longer paced QRS duration was associated with higher risk of ventricular arrhythmia (hazard ratio [HR]: 1.11 per 10-ms increase; 95% confidence interval [CI]: 1.07 to 1.16; p < 0.001) and all-cause death or arrhythmia (HR: 1.09; 95% CI: 1.09 to 1.13; p < 0.001). A paced QRS duration >190 ms was associated with a 3.6 times higher risk of ventricular arrhythmia (HR: 3.6; 95% CI: 2.35 to 5.53; p < 0.001) and a 2.1 times higher risk of death or arrhythmia (HR: 2.12; 95% CI: 1.53 to 2.95; p < 0.001), independent of left ventricular function or baseline QRS duration. Longer QTc interval during ventricular pacing was associated with a higher risk of ventricular arrhythmia (HR: 1.03 per 10-ms increase; 95% CI: 1.02 to 1.12; p < 0.001) independent of paced QRS duration.
Conclusions Longer paced QRS duration and paced QTc interval predict ventricular arrhythmias in patients with cardiomyopathy. Ventricular conduction and repolarization prolongation during right ventricular pacing can determine the risk of ventricular arrhythmias.
No single clinical parameter can accurately identify patients at risk for ventricular arrhythmic events (1–7); the currently used parameters have, at best, moderate predictive value (6,8,9). The most widely used and accepted criterion for implantation of a defibrillator for primary prevention is severely decreased left ventricular (LV) function (10).
However, many patients who die of sudden cardiac death have moderate or only mildly decreased LV function, whereas in many others with severely decreased LV function, ventricular arrhythmias do not develop (11,12). Electrocardiographic (ECG) parameters associated with sudden arrhythmic death have poor predictive value and are generally not used clinically (9). Therefore, development of new and better risk markers of ventricular arrhythmic events is needed.
Ventricular conduction and repolarization abnormalities are associated with a higher risk of ventricular arrhythmias (13,14). Most analyses of ventricular conduction and repolarization are conducted in the baseline state, during normal depolarization. However, many abnormalities in these parameters can only be elicited under certain circumstances, such as exercise stress tests (15) or epinephrine administration (16) for long QT syndrome or antiarrhythmic agent infusion (17) for Brugada syndrome. Thus, provoking maneuvers during an apparently normal ventricular depolarization and repolarization assessment may reveal underlying abnormalities that can be associated with a higher risk of ventricular arrhythmias.
The goal of the present study was to evaluate the association of changes in ventricular conduction and repolarization times incited by ventricular stimulation with risk of ventricular arrhythmia and death in patients with LV dysfunction.
Consecutive patients with LV dysfunction undergoing electrophysiological testing from January 2002 through May 2014 at Mayo Clinic (Rochester, Minnesota) were enrolled. The duration changes in ventricular conduction and repolarization times during ventricular stimulation were measured. The study sample was followed up until December 2014 for development of ventricular arrhythmic events and all-cause and cardiovascular death. The study was approved by the Mayo Clinic Institutional Review Board.
ECG analysis and data acquisition
ECG data were acquired after retrieving and reviewing standard electrophysiology studies using the Prucka Mac-Lab/CardioLab system, version 5.0G (GE Healthcare, Little Chalfont, United Kingdom). All measurements were performed on the electronic system by using incorporated electronic calipers during simultaneous recording of the standard 12-lead electrocardiogram. QRS duration (from the earliest onset of the Q-, R-, or S- wave in any lead to the latest offset of the R- or S-wave in any lead) and QT interval (from the earliest onset of the QRS complex in any lead to the latest offset of the T-wave, the latest indication of ventricular repolarization) were measured. We evaluated baseline QT interval during atrial pacing at a cycle length of 600 ms and the paced QT interval during right ventricular (RV) stimulation at the same cycle length. The QT interval was then corrected for heart rate by using the Bazett formula. Measurements during pacing were considered only from the RV apex, which was confirmed by fluoroscopy and paced QRS morphology analysis.
In total, 501 patients underwent programmed ventricular stimulation for induction of ventricular arrhythmias: 256 (51.1%) as a diagnostic study only, 64 (12.8%) as part of ablation of supraventricular arrhythmias, 9 (1.8%) for ablation of atrial fibrillation, and 172 (34.3%) for ablation of ventricular extrasystoles or tachycardia.
Cardiovascular and all-cause deaths that occurred by December 2014 were reviewed. Immediate cause-of-death information was obtained from hospital documents, communication with family members, autopsy reports, and death certificates. Cardiovascular death was defined as non–sudden cardiac death. Sudden cardiac death was defined as death within 1 h after abrupt onset of symptoms or within 24 h after onset of symptoms if autopsy data did not reveal a noncardiac cause or after successful resuscitation from ventricular tachycardia and/or ventricular fibrillation.
Ventricular arrhythmic events were defined as symptomatic episodes of documented ventricular tachycardia or fibrillation requiring intervention (defibrillation shocks or antitachycardia pacing in patients with implantable defibrillators; documented ventricular arrhythmias requiring external shocks or overdrive pacing). In patients with implantable cardiac devices, arrhythmic events were classified after device-retrieved data related to the events were analyzed by 2 cardiac arrhythmia specialists; if findings were discrepant, a third specialist determined the event classification. Other sources of documentation (electrocardiograms, external ambulatory cardiac monitor, or telemetry tracings) were also reviewed. Ventricular arrhythmic events were evaluated blindly to ECG data, starting after the electrophysiology study date.
Continuous variables are expressed as mean ± SD and compared by using the Student t test or Mann-Whitney test, as appropriate; categorical variables are expressed as count (%), and comparisons were performed by using the Fisher exact test or chi-squared test. Area under the receiver-operating characteristic curve (AUC) analysis was used to examine the performance of paced and native QRS and QTc durations to detect ventricular arrhythmic events. Logistic regression analysis was used to evaluate arrhythmic event rate differences between the groups with longer and shorter paced QRS durations, stratified according to LV function and baseline QRS duration. Event-free survival was estimated by using the Kaplan-Meier method, and the log-rank test was used to compare the time to a ventricular arrhythmic event or all-cause death among groups with shorter versus longer paced QRS or QTc durations.
Cox proportional hazards models were used to evaluate the association of paced parameters with outcomes of interest. Covariates in univariable analysis included age, sex, baseline QRS duration, ventricular conduction defect, ejection fraction (EF), inducible ventricular arrhythmias, diabetes mellitus, hypertension, chronic kidney disease, ischemic cardiomyopathy, New York Heart Association (NYHA) functional class, and previous myocardial infarction. Significant covariates were included in multivariable models, including paced QRS, QTc, and JTc intervals. Subgroup analysis was performed including only patients with implantable cardioverter-defibrillators and patients with follow-up >1 year. Proportional hazards assumption was assessed on the plots of log(time) versus log(–log[survival]) stratified according to the variables. The assumptions were verified to be acceptable. For all tests, p values <0.05 were considered significant. Analyses were performed by using JMP Pro 10.0.0 (SAS Institute, Inc., Cary, North Carolina).
Patient demographic characteristics
Among the 501 patients included, the mean ± SD age was 65.2 ± 13.7 years, and 80% were men (Table 1). Approximately 74% of patients were in NYHA functional class II or III, and 301 (60%) had ischemic cardiomyopathy. The mean baseline QRS duration was 137 ± 44 ms. The mean left ventricular ejection fraction (LVEF) was 33.1 ± 9.6%; 41% of patients had an EF >35%. A total of 358 patients (71%) had, or subsequently underwent placement of, an implantable cardioverter-defibrillator. Median follow-up duration was 2.4 years (interquartile range: 0.7 to 4.8 years).
The cohort was divided into 2 groups according to the median value of paced QRS duration: >190 ms versus ≤190 ms. Baseline clinical differences between the groups are shown in Table 1. Patients with a paced QRS duration >190 ms more frequently had hyperlipidemia, worse NYHA functional class, and greater use of β-blockers and statins, and had longer baseline QRS duration and QT interval and lower LVEF (all, p < 0.01).
Paced QRS duration and ventricular arrhythmic events
Patients with paced QRS duration >190 ms had a higher risk of ventricular arrhythmic events than the group with shorter paced QRS duration (odds ratio [OR]: 5.7; 95% CI: 3.7 to 8.9; p < 0.001). When further grouping according to LVEF (≤35% vs. >35%), the group with shorter paced QRS and EF >35% had the lowest rate of arrhythmic events (17.3%), similar to those with shorter paced QRS duration and EF ≤35% (22.3%) (OR: 1.37; 95% CI: 0.7 to 2.8; p = 0.37) (Figure 1). Patients with a paced QRS duration >190 ms had increased risk of arrhythmic events compared with the first group for both lower EF (OR: 7.8; 95% CI: 4.4 to 14.5; p < 0.001) and higher EF (OR: 4.5; 95% CI: 2.2 to 9.3; p < 0.001).
Likewise, when grouping according to baseline QRS duration (≤120 ms vs. >120 ms) and paced QRS duration (≤190 ms vs. >190 ms), there was no difference in the rate of ventricular arrhythmic events between the groups with shorter paced QRS duration with shorter or longer baseline QRS duration (19.0% vs. 20.7%) (OR: 1.12; 95% CI: 0.55 to 2.26; p = 0.76). For the group with longer paced QRS duration, the odds of arrhythmic events were greater for both the longer (OR: 7.4; 95% CI: 4.3 to 13.3; p < 0.001) and shorter (OR: 3.2; 95% CI: 1.6 to 6.6; p < 0.001) baseline QRS duration compared with the group with shorter baseline and paced QRS durations (Figure 2).
Using AUC analysis to predict ventricular arrhythmic events based on ventricular conduction times, we found that a cutoff for a paced QRS duration of 190 ms (Figure 3A) yielded a sensitivity of 77% and a specificity of 63%. Compared with the predictive value of the baseline QRS duration (Figure 3B), a longer paced QRS duration was a better predictor of arrhythmic events (AUC: 0.74 vs. 0.67, respectively; p = 0.004).
In Kaplan-Meier analysis of time to ventricular arrhythmic event, patients with a paced QRS duration >190 ms had significantly lower event-free survival than patients with shorter paced QRS duration (log-rank test, p < 0.001) (Figure 4A). Furthermore, separate analyses dichotomizing for EF >35% or ≤35% found that, independent of LV function, the event-free survival for arrhythmic events was significantly decreased in the group with longer paced QRS duration (Figure 4B).
Cox univariable regression analysis showed that longer paced QRS duration was associated with a significantly higher risk of ventricular arrhythmic events (hazard ratio [HR]: 1.2; 95% CI: 1.1 to 1.2, per every 10-ms increase; p < 0.001). A paced QRS duration >190 ms was associated with a 4.5 times higher risk versus a paced QRS duration ≤190 ms (HR: 4.5; 95% CI: 3.2 to 6.6; p < 0.001). On multivariable analysis adjusting for baseline ventricular conduction defect, baseline QRS duration, LVEF, positive electrophysiological study, chronic kidney disease, and NYHA functional class, a paced QRS duration >190 ms remained a significant predictor of ventricular arrhythmic events (HR: 3.6; 95% CI: 2.4 to 5.5; p < 0.001) compared with paced QRS duration ≤190 ms (Table 2). Per every 10-ms increase in paced QRS duration, the risk of arrhythmic events increased by 11% (HR: 1.11; 95% CI: 1.07 to 1.16; p < 0.001). In addition, the risk of arrhythmic events was 1.1 times higher per every 10-ms increase in the difference between paced QRS and baseline QRS duration. Baseline QRS duration did not influence the risk of ventricular arrhythmic events.
Paced QRS duration and cardiovascular and all-cause death
For patients with a paced QRS duration >190 ms, the risk of cardiovascular death was significantly higher than for those with a shorter paced QRS duration (OR: 2.2; 95% CI: 1.3 to 3.8; p = 0.002). Further stratifying according to LVEF, a paced QRS duration ≤190 ms and EF >35% gave the lowest risk of cardiovascular death; compared with that group, patients with paced QRS duration >190 ms and EF ≤35% had a significantly higher risk of cardiovascular death (OR: 5.1; 95% CI: 2.4 to 12.0; p <0.001) (Figure 5). Compared with the reference group, patients with shorter paced QRS duration but worse LVEF had a higher risk of cardiovascular death (OR: 2.9; 95% CI: 1.2 to 7.4; p = 0.01), but those with longer paced QRS duration and similar higher LVEF did not (OR: 1.7; 95% CI: 0.5 to 5.0; p = 0.37). The Kaplan-Meier estimate for the time to cardiovascular death revealed a higher risk in patients with a paced QRS duration >190 ms (log-rank test, p < 0.001) (Figure 6).
The time to the composite endpoint of all-cause death or ventricular arrhythmic event was significantly shorter in patients with a paced QRS duration >190 ms versus ≤190 ms (log-rank test, p < 0.001) (Figure 7A). When further dichotomizing according to LVEF, event-free survival for all-cause death or ventricular arrhythmic events was significantly decreased in the longer versus shorter paced QRS duration group, independent of baseline LVEF (Figure 7B).
After adjusting for potential confounders, the risk of death or arrhythmias was 1.1 times higher per every 10-ms increase in paced QRS duration (HR: 1.1; 95% CI: 1.04 to 1.13; p < 0.001), and a paced QRS duration >190 ms remained associated with a higher risk of events (HR: 2.12; 95% CI: 1.5 to 3.0; p < 0.001) (Table 3).
On further subgroup analyses in patients who had implantable cardioverter-defibrillators and patients who had at least 1 year of follow-up, paced QRS duration remained associated with ventricular arrhythmic events on multivariable analyses, with similar HRs as above (Online Table 1).
Paced QT interval and outcomes
We evaluated the associations of QT, QTc, JT, and JTc intervals at baseline and during ventricular pacing. Among all these parameters, paced QTc intervals were associated with higher ventricular arrhythmic risk. Logistic regression and AUC analysis revealed that an RV-paced QTc interval cutoff of 620 ms had a sensitivity of 59% and a specificity of 77% to predict ventricular arrhythmic events (Online Figure 1).
We also introduced RV-paced QTc interval as a continuous variable or dichotomized at durations of ≤620 ms versus >620 ms as new covariates in multivariable regression models. After adjusting for baseline characteristics and including paced QRS duration, longer paced induced QTc interval was associated with a higher risk of arrhythmic events (HR: 1.05; 95% CI: 1.02 to 1.12, per 10-ms increase; p < 0.001). Similarly, longer paced JTc interval was associated with a higher risk of arrhythmic events (HR: 1.05; 95% CI: 1.02 to 1.07, per 10-ms increase; p < 0.001) (Online Figure 2, Online Table 2).
This study aimed to evaluate the association of ventricular conduction and repolarization duration observed during RV stimulation with ventricular arrhythmic events and death in patients with cardiomyopathy. We found that: 1) patients with longer paced QRS duration had longer baseline QRS duration, longer baseline QT interval, worse NYHA functional class, and lower LV function; 2) longer paced QRS duration was associated with a higher risk of ventricular arrhythmic events, and this association was independent of LV function or baseline QRS duration; 3) longer paced QRS duration was associated with a higher risk of cardiovascular death and the composite of ventricular arrhythmic events or all-cause death, independent of the degree of LV dysfunction; and 4) longer paced QTc or JTc intervals were associated with a higher risk of ventricular arrhythmic events.
QRS duration and its association with arrhythmic events and death
The association of QRS duration and increased ventricular arrhythmic risk is inconsistent (9). Although some population-based studies of persons with no known heart disease have shown that longer baseline QRS duration is associated with an increased risk, retrospective analyses of large clinical trials in patients with known heart disease have failed to confirm that association. A study of 2,049 men showed that a 10-ms increase in QRS duration was associated with a 27% increased risk of sudden cardiac death and that patients with a baseline QRS duration >110 ms had a 2.5-fold higher risk of sudden cardiac death than men with a QRS duration <96 ms (18). In contrast, the PainFREE RX II (Pacing Fast Ventricular Tachycardia Reduces Shock Therapies II) (19) and MADIT-II (Multicenter Automatic Defibrillator Implantation Trial II) (20) studies failed to demonstrate an association between QRS duration and risk of ventricular arrhythmias.
Ventricular activation during ventricular pacing
Ventricular electrical activation through the native electrical system differs substantially from the ventricular activation after RV pacing. Whereas during normal ventricular conduction the electrical impulse travels efficiently through the His-Purkinje system, during ventricular pacing, the impulse travels mostly through myocardial fibers. Therefore, the ventricular conduction time is different during normal ventricular conduction versus ventricular stimulation, and this difference seems to be more pronounced when paced ventricular activation occurs through diseased ventricular myocardium. Our study showed that patients with more severe cardiomyopathy and worse prognosis have slower conduction times, most likely because of more diseased myocardium.
Paced QRS duration as a marker of cardiomyopathy
RV-paced QRS duration has been associated with the degree of myocardial disease. Sumiyoshi et al. (21) evaluated paced QRS duration in patients with pacemakers and found that a paced QRS duration >180 ms was associated with decreased LV function compared with patients with a paced QRS duration ≤180 ms. Similarly, other investigators reported that in patients with implantable devices, longer paced QRS duration is associated with more severe cardiac dysfunction and is a better marker of subjacent cardiomyopathy than is baseline QRS duration (22). In addition, patients with cardiomyopathy, heart block, and a paced QRS duration >190 ms were more likely to have clinical heart failure events than those with a shorter paced QRS duration (23), and a prolonged paced QRS duration at pacemaker implantation was associated with a higher incidence of heart failure hospitalization (24).
Paced QRS duration as predictor of arrhythmic events
To our knowledge, no studies have evaluated the association of paced QRS duration and ventricular arrhythmic events among patients with LV dysfunction. A study evaluating 15 patients with aborted ventricular fibrillation sudden death and no underlying cardiomyopathy (25) showed that during programmed ventricular stimulation, the QRS duration of the third ventricular extrastimulus was significantly longer than that in a healthy control group. Another study evaluating intracardiac paced ventricular electrogram fractionation in response to ventricular extrastimuli showed that in patients with a history of ventricular fibrillation, the local intracardiac ventricular electrogram is prolonged at longer coupling intervals than in patients with no history of arrhythmic events (26).
Our study showed that patients with LV dysfunction at higher risk for ventricular arrhythmias had longer paced ventricular conduction and repolarization times; this finding can be easily elucidated by measuring the duration of paced QRS and QTc intervals. Ventricular pacing during electrophysiological testing or at implantation of pacemakers/defibrillators can unmask functional delayed ventricular activation that otherwise is not seen at baseline conditions and provides prognostic information on the risk of ventricular arrhythmic events and death.
Potential clinical impact
Pending further prospective validation studies, our findings can potentially be used as the basis to refine risk stratification of ventricular arrhythmias and sudden cardiac death. Potential subpopulations include patients with LV dysfunction and pacemaker implantation; patients undergoing diagnostic electrophysiology studies who have noninducible ventricular arrhythmias but perceived significant risk of ventricular arrhythmias; and patients with premature ventricular contractions with unknown arrhythmic risk, using QRS duration as a surrogate for paced QRS duration (depending on the origin of the premature ventricular contractions).
This research was a retrospective study with the inherent limitations of this design. All patients in our study required invasive electrophysiological testing. Therefore, these findings may not be applicable to the general population with cardiomyopathy. However, the sample size was large, and ECG measures were accurate, with reliable evaluation of outcomes.
The present study found that prolonged paced QRS and paced QTc interval duration are associated with a higher risk of ventricular arrhythmias and death. Further studies are required to validate these findings and explore the role of these parameters for risk stratification of ventricular arrhythmias and sudden cardiac death.
COMPETENCY IN MEDICAL KNOWLEDGE: The measurement of ventricular conduction and repolarization times during ventricular pacing (stimulation) can improve the identification of patients at increased risk for ventricular arrhythmias, beyond the information obtained from LV function, baseline conduction and repolarization times, or relevant clinical characteristics.
TRANSLATIONAL OUTLOOK: There is a need to further improve risk stratification for ventricular arrhythmic events and sudden cardiac death. More prospective studies should be performed to find better prediction tools.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors' institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- area under the receiver-operating characteristic curve
- confidence interval
- ejection fraction
- hazard ratio
- left ventricular
- left ventricular ejection fraction
- New York Heart Association
- odds ratio
- right ventricular
- Received February 2, 2017.
- Revision received May 17, 2017.
- Accepted June 9, 2017.
- 2017 American College of Cardiology Foundation
- Myerburg R.J.,
- Junttila M.J.
- Chugh S.S.,
- Jui J.,
- Gunson K.,
- et al.
- Goldberger J.J.,
- Basu A.,
- Boineau R.,
- et al.
- Goldberger J.J.,
- Buxton A.E.,
- Cain M.,
- et al.
- Goldberger J.J.,
- Cain M.E.,
- Hohnloser S.H.,
- et al.,
- American Heart Association Council on Clinical Cardiology,
- American Heart Association Council on Epidemiology and Prevention,
- American College of Cardiology Foundation,
- Heart Rhythm Society
- Epstein A.E.,
- DiMarco J.P.,
- Ellenbogen K.A.,
- et al.,
- American College of Cardiology Foundation,
- American Heart Association Task Force on Practice Guidelines,
- Heart Rhythm Society
- Narayanan K.,
- Reinier K.,
- Uy-Evanado A.,
- et al.
- Yap Y.G.,
- Duong T.,
- Bland J.M.,
- et al.
- Zimetbaum P.J.,
- Buxton A.E.,
- Batsford W.,
- et al.
- Brugada R.,
- Brugada J.,
- Antzelevitch C.,
- et al.
- Kurl S.,
- Makikallio T.H.,
- Rautaharju P.,
- Kiviniemi V.,
- Laukkanen J.A.
- Buxton A.E.,
- Sweeney M.O.,
- Wathen M.S.,
- et al.,
- PainFREE Rx II Investigators
- Singh J.P.,
- Hall W.J.,
- McNitt S.,
- et al.,
- MADIT-II Investigators
- Chen S.,
- Yin Y.,
- Lan X.,
- et al.,
- PREDICT-Heart Failure Study International Group
- Saumarez R.C.,
- Chojnowska L.,
- Derksen R.,
- et al.