Author + information
- Received September 21, 2015
- Revision received February 25, 2016
- Accepted March 10, 2016
- Published online August 1, 2016.
- Avi Sabbag, MDa,
- Ilan Goldenberg, MDa,b,
- Arthur J. Moss, MDb,
- Scott McNitt, MSb,
- Michael Glikson, MDa,
- Yitschak Biton, MDb,
- Larry Jackson, MDc,
- Bronislava Polonsky, MSb,
- Wojciech Zareba, MD, PhDb and
- Valentina Kutyifa, MD, PhDb,∗ ()
- aSheba Medical Center, Ramat Gan, Israel
- bHeart Research Follow-up Program, Division of Cardiology, Department of Medicine at the University of Rochester Medical Center, Rochester, New York
- cDuke University School of Medicine, Durham, North Carolina
- ↵∗Reprint requests and correspondence:
Dr. Valentina Kutyifa, Heart Research Follow-up Program, Cardiology Division, University of Rochester Medical Center, 265 Crittenden Boulevard, PO Box 653, Rochester, New York 14642.
Objectives The study sought to analyze the risk of ventricular tachyarrhythmia (VTA) or death in black and white subjects implanted with implantable cardioverter-defibrillators (ICDs) or defibrillator and combined cardiac resynchronization therapy (CRT-D) in the MADIT-CRT (Multicenter Automatic Defibrillator Implantation with Cardiac Resynchronization Therapy) trial.
Background There are limited data on ethnic differences in the risk for VTA in mildly symptomatic heart failure patients with left ventricular dysfunction.
Methods The risk for first VTA (≥180 beats/min) or death was evaluated in black (n = 139) versus white (n = 1,638) patients enrolled in the MADIT-CRT trial using Kaplan-Meier survival analyses and Cox proportional hazards regression models after adjustment for relevant clinical covariates. Multivariate analysis was used to identify race-specific risk factors for VTA.
Results At 4 years of follow-up, the cumulative probability for a first VTA or death was significantly higher among black patients (42%) as compared with whites (34%; log-rank p value for the overall difference during follow-up = 0.01). Multivariate analysis confirmed significantly higher risk of VTA or death (hazard ratio: 1.60; 95% confidence interval: 1.18 to 2.17; p = 0.002), and higher risk of VTA alone (hazard ratio: 1.71; 95% confidence interval: 1.22 to 2.41; p = 0.002) in blacks compared to whites. The findings were similar in both ICD and CRT-D implanted patients, with no significant race-to-treatment-interaction (interaction p values >0.05). Independent risk factors for VTA among blacks included increased systolic blood pressure values and larger cardiac volumes.
Conclusions In the MADIT-CRT trial, black patients had a significantly higher rate of ventricular tachyarrhythmias or death compared to whites, with either an implanted ICD or CRT-D. (MADIT-CRT: Multicenter Automatic Defibrillator Implantation with Cardiac Resynchronization Therapy [MADIT-CRT], NCT00180271; Multicenter Automatic Defibrillator Implantation with Cardiac Resynchronization Therapy Post Approval Registry [MADIT-CRT PAR], NCT01294449; MADIT-CRT LONG-TERM INTERNATIONAL FOLLOW-UP REGISTRY–EUROPE, NCT02060110).
- cardiac resynchronization therapy with defibrillator
- implantable cardioverter-defibrillator
- mild heart failure
- ventricular tachyarrhythmias
The prevalence of heart failure (HF) continues to rise (1), becoming one of the leading causes of death in the United States with significant health burden. When compared to whites, blacks have a higher rate of HF (2), they present with HF symptoms at an earlier age and they have a 2.5 times higher mortality of HF (3). Blacks have also been shown to have a higher risk of sudden cardiac death.
Cardiac resynchronization therapy (CRT) reduces mortality (4), HF events (5,6) and improves functional capacity, and CRT has been shown to reduce the rates of ventricular tachyarrhythmia (VTA) in HF patients (7). However, previous studies have shown lower efficacy of several HF treatments in blacks, including an attenuated response to angiotensin-converting enzyme inhibitors, and less benefit from an implanted cardioverter-defibrillator (ICD) (8) or from CRT (9).
To our knowledge, the risk of life-threatening VTA or death, has not been evaluated in detail among blacks with HF and ICD or defibrillator and combined cardiac resynchronization therapy (CRT-D) thus far. Therefore the aim of our study was to evaluate the risk of VTA or death in blacks versus whites with an implanted ICD or CRT-D device.
The details of the MADIT-CRT (Multicenter Automatic Defibrillator Implantation with Cardiac Resynchronization Therapy) trial have been previously described (5). Briefly, the trial investigated the effect of CRT-D versus ICD alone on reducing mortality or HF events in patients with early HF, either in New York Heart Association (NYHA) functional class I or II due to ischemic etiologies or NYHA functional class II due to nonischemic cardiomyopathy, reduced ejection fraction (≤30%), and ventricular conduction delay as evidenced by QRS duration ≥130 ms. A total of 1,820 patients were enrolled between December 22, 2004, and April 23, 2008, at 110 hospital centers in North America and Europe and randomly assigned in a 3:2 ratio to receive either CRT-D or ICD. Race was self-reported and recorded at enrollment to the study. This current substudy included a total of 1,613 white (92%) and 139 black (8%) patients.
Device implantation and programming
Generally available ICD and CRT-D devices (Boston Scientific, St. Paul, Minnesota) were implanted in the MADIT-CRT trial using standard methods. Device programming and defibrillator threshold testing were performed as reported in the study protocol. ICD and CRT-D devices were programmed to a ventricular tachycardia (VT) (10) zone at 180 beats/min, and ventricular fibrillation (VF) zone at 210 beats/min. The VT zone first therapy was recommended to be programmed to burst-type antitachycardia pacing, and then shock therapies.
Study design and endpoints
The primary endpoints of this study were the occurrence of first VTA or death, and first VTA. Secondary endpoints included fast VTA (>200 beats/min) or death, and fast VTA during the follow-up period.
All VTA episodes were adjudicated by an independent committee that was blinded to treatment allocation by reviewing the stored electrograms of ICD and CRT-D devices. VTA episodes were defined as any type of therapy delivered for a VT or VF. The definition of VT was set to a rate from 180 to 250 beats/min. VF was defined as ventricular rate faster than 250 beats/min with disorganized ventricular electrograms.
We have previously shown that the clinical benefit of CRT-D in the MADIT-CRT trial was restricted to left bundle branch block (LBBB) patients (7). Furthermore, in a recent MADIT-CRT trial substudy we have shown that treatment with cardiac resynchronization therapy was associated with a significant reduction in VTA risk only in LBBB patients as compared with ICD-only patients, whereas among non-LBBB patients CRT-D was associated with corresponding risk increase (11). Accordingly, in the present study the analysis that evaluated the effect of CRT-D on VTA risk in black and white patients was restricted to LBBB patients.
Clinical characteristics of study patients were compared by race and arrhythmic status using the chi-square test for categorical variables, and the nonparametric Wilcoxon rank-sum test for continuous variables. The Kaplan-Meier method was used for graphical depiction of the time until first VTA and first VTA or death in black and white patients and statistically evaluated with the log-rank test.
The Cox proportional hazards regression model was used to evaluate the independent contribution of baseline clinical factors to the development of the endpoints. The best subsets method was used to determine the inclusion of adjustment variables with the added stipulation that variables to be significant at p < 0.05 for either VTA or death. The effect of treatment allocation on outcomes in black and white patients was assessed by including a treatment-by-race interaction term in the multivariate models.
We carried out separate Cox proportional hazards regression models for both races in order to identify different predictors for VTA and VTA or death in blacks or whites. In addition we developed a propensity score, which was used to stratify the multivariable Cox model using quintiles of the propensity score and present the results from both models. The variables utilized for the propensity score were on the basis of a logistic regression model with black versus white race as the response and used best subset analysis with Fisher scoring technique. Further, variables which met the significance level of <0.10 were used for the propensity score.
Analyses were performed with the SAS statistical software (version 9.4, SAS Institute, Cary, North Carolina). A 2-sided p value <0.05 was used for statistical significance.
A total of 139 (8%) black and 1,638 white patients were included in the current analysis, of whom 60% and 61% received a CRT-D device, respectively. The results of a comparison between black and white patients are summarized in Table 1. Blacks were younger, more often female, and had a higher proportion of nonischemic cardiomyopathy. Rates of diabetes and hypertension were higher among blacks whereas whites had a higher proportion of atrial fibrillation and worse renal function. There were no significant differences in QRS duration or QRS morphology or in the echocardiographic measurements of left ventricular function and left ventricular dimensions. Clinical characteristics by race and implanted device are shown in Online Table 1. Crude rates on mortality, any ventricular tachyarrhythmia event (VTE) and fast VTE in both groups as well as Kaplan-Meier analysis of cumulative probability of all-cause mortality are presented in Online Table 2 and Online Figure 1.
Probability of first VTA or death, by race
During follow-up, the rate of VTA or death was significantly higher in black as compared with white patients (Figures 2A and 2B). Thus, at 4 years of follow-up the cumulative probability for a first VTA or death was 42% in blacks as compared with 34% among white patients (p for overall difference during follow-up = 0.01) (Figure 1A). Similarly, the respective rates of VTA alone were also significantly higher among black versus white patients (35% vs. 26%; log-rank p for the overall difference during follow-up = 0.029) (Figure 1B). Consistent with the univariate findings, multivariate analysis showed that black patients experienced a significant 62% (p = 0.002) higher risk for VTA or death as compared with white patients, a 69% (p = 0.002) increased risk for VTA alone, and 93% increased risk for fast ventricular tachyarrhythmias (Table 2). Two additional multivariate models further correcting for a propensity score and treatment with digoxin yielded similar results.
Predictors of first ventricular arrhythmia or death
Independent predictors for first VTA or death in black and white patients are shown in Table 3. Among blacks, an increase in blood pressure was inversely related to the risk of VTA or death, with a significant 18% (p < 0.001) increase in the risk of VT or VF for every 5 mm Hg reduction in blood pressure values. An increase in left ventricular end-diastolic volume (LVEDV) was also directly related to VT or VF events, wherein each 10 ml/m2 increase in the indexed LVEDV was associated with a corresponding 13% (p = 0.03) increase in the risk of VT or VF. In addition, there was a trend towards a higher risk for ventricular tachyarrhythmia in patients with better baseline renal function wherein each 10 ml/min/1.73 m2 increase in estimated glomerular filtration rate was associated with a corresponding 14% (p = 0.08) increased risk for VTA or death. In contrast, among white patients there was no statistically significant association between systolic blood pressure values and the risk of VTA or death, whereas LVEDV was also directly correlated with VTA risk. Additional risk factors significantly associated with VTA among white patients included obesity, active smoking, female sex, QRS duration at baseline, history of myocardial infarction, and history of ventricular arrhythmias.
Effect of CRT-D on the risk of first VTA or death in LBBB patients
In patients with LBBB electrocardiographic morphology at baseline, treatment with CRT-D was associated with a lower rate of VTA or death as compared with ICD-only therapy among both black and white patients (Figure 2A): in blacks implanted with a CRT-D the rate of VTA or death at 4 years of follow-up was 36% as compared to 40% in blacks implanted with ICD only. In whites, the corresponding rates of VTA or death were 25% and 37% (p < 0.001) for the overall difference among the 4 groups during follow-up (Figure 2A). Similarly, the cumulative probability of VTA was also lower between both black and white patients treated with CRT-D (Figure 2B).
Multivariate analysis consistently showed that treatment with CRT-D was associated with a significant reduction in the risk of VTA or death and VTA alone as compared with ICD-only therapy in white patients (hazard ratio [HR]: 0.65; 95% confidence interval [CI]: 0.52 to 0.82; HR: 0.66; 95% CI: 0.51 to 0.85, respectively), whereas, possibly due to sample size limitations, these findings did not reach statistical significance among black patients (HR: 0.65; 95% CI: 0.32 to 1.30; HR: 0.59; 95% CI: 0.27 to 1.27, respectively). However, there was no significant treatment-by-race interaction (interaction p values for both endpoints >0.05), suggesting a similar effect of CRT-D versus ICD to reduce the risk of VTA or death, or VTA alone in black and white patients (Table 4).
In the present study we assessed differences in VTA risk between black and white patients with mildly symptomatic HF who were enrolled in the MADIT-CRT trial. Our main findings are that: 1) within the MADIT-CRT population, blacks have a 69% higher risk of ventricular tachyarrhythmia and a 62% higher risk of the combination of life-threatening tachyarrhythmia or death, driven mostly by a higher rate of first VTA events. The excess risk persists regardless of device therapy implemented and after adjustment for other relevant cofactors. These results were found to be robust and persisted after the addition of a propensity score to the model; 2) VTA risk in black patients was associated with specific risk factors, including lower systolic blood pressure values, and increased cardiac volumes; and 3) despite possible sample size limitations, our findings also suggest that treatment with CRT-D exerts similar antiarrhythmic effects in both black and white patients.
The disparities in baseline characteristics between blacks and whites are consistent with previous publications. Blacks are consistently found to be younger, have a higher proportion of women, and a higher proportion of nonischemic cardiomyopathy. These differences persisted when we compared patients with or without VTA or death in both races.
Our results, indicating a higher risk for ventricular arrhythmia in blacks, are consistent with previous reports on the basis of population studies (12). Blacks have been consistently shown to have a higher incidence of sudden cardiac death with events occurring at a younger age (12,13). This observation coupled with a significantly higher likelihood of HF (14), manifesting at a younger age (15) with a higher proportion of nonischemic etiology (16) may imply some disparities in mechanisms leading to HF, arrhythmia, and sudden cardiac death in this population.
One of the features characterizing the population of black patients with HF is a high proportion of hypertension, leading to hypertensive cardiomyopathy and left ventricular hypertrophy (LVH) (17). The high rate of hypertension has been postulated as a major reason for higher incidence of HF among blacks (14). Several studies have demonstrated an association between LVH and both atrial and ventricular arrhythmia as well as sudden cardiac death (18). A report on the basis of the MUSTT (Multicenter Unsustained Tachycardia Trial) trial (19) found LVH (defined by electrocardiographic criteria) to be a significant predictor of arrhythmic mortality. A recent meta-analysis (20) comprising of 10 studies with a total of 27,171 resulted in the same conclusions. Therefore we hypothesize that a higher proportion of LVH in black patients in our study may explain our results. Unfortunately the MADIT-CRT trial population includes patients with severe left ventricular dysfunction, which precludes the possibility of retrospectively establishing a diagnosis on the basis of left ventricular mass in our cohort, leaving this hypothesis a thought-provoking speculation.
Hypotension was found to predict arrhythmic events in blacks but not whites in our population. Both factors are likely a marker of decompensated HF and hemodynamic instability. The more prominent effect of hypotension in blacks may also be explained by higher rates of hypertension at baseline in this population.
Previous studies evaluating the arrhythmic risk in black patients with an ICD have resulted in conflicting conclusion. The SCD-HeFT (Sudden Cardiac Death in Heart Failure Trial) comprised 2,521 patients, of which 421 (17%) were self-identified African American (AA) and 1,932 (77%) as white randomly selected to either ICD or amiodarone treatment for prevention of SCD. In a subsequent substudy (16), the authors reported a similar unadjusted rate of appropriate ICD discharge in both AAs and whites. However, adjusted mortality risk was significantly higher in AAs compared to whites (HR: 1.27; p = 0.038). Of note, ICD therapy was associated with an equal reduction in mortality in both race groups compared to placebo (HRs of 0.65 in AAs and 0.73 in whites).
The PROSE-ICD (Prospective Observational Study of Implantable Cardioverter-Defibrillators) study recruited a total of 1,189 patients including 477 (40%) blacks who underwent ICD implantation for primary prevention. Over a median follow-up of 5.1 years, there was no difference in the likelihood of first appropriate shock between AA and non-AA (21). Notably, this may not clearly represent the difference in risk for VTA as an antitachycardia pacing zone was used more often in AA. However, AA had an increased risk of dying without receiving an appropriate ICD shock compared to non-AAs (HR: 1.33; 95% CI: 1.02 to 1.74). An analysis on the basis of the MUSTT trial concluded that blacks assigned to electrophysiology-guided therapy had higher mortality rates compared to whites. In fact, the survival of blacks randomly assigned to no EP-guided therapy was better than blacks receiving EP-guided therapy (22). The significance of the last report is limited by the fact that there were only 61 blacks in the electrophysiology-guided therapy group. In addition, both the SCD-HeFT and PROSE-ICD trial populations comprised patients with more severe HF (NYHA functional class III rates >30% vs. none in the MADIT CRT trial). It is plausible that racial differences may be more prominent at earlier stages of HF.
CRT-D has been previously shown to reduce mortality in non-white racial groups but no studies focused on CRT-D to reduce the risk of ventricular arrhythmias. A substudy (9) published by our group aimed at comparing rates of death or HF in blacks versus whites, found a similar magnitude of echocardiographic response to CRT-D and no significant race-to-treatment interaction. Similar to our current analysis, the study was underpowered to directly compare ICD to CRT-D among blacks but the results suggest that CRT-D is effective in reducing HF and mortality in blacks. The IMPROVE HF (Improve the Use of Evidence-Based Heart Failure Therapies in the Outpatient Setting) registry is a prospective observational cohort study. A subanalysis (23) on the basis of 6,994 patients with HF eligible for ICD/CRT-D treatment, including 818 blacks, of which 362 (44%) were treated with ICD/CRT-D, suggested that the efficacy of defibrillators was no different by racial groups. A similar decrease in mortality was observed among blacks and whites implanted with ICD/CRT-D with no significant treatment/race interaction.
Another MADIT-CRT trial substudy (24) has established that CRT-D may reduce the risk of first occurrence of ventricular tachyarrhythmias by 29% in patients with LBBB serving as the basis our current study. The combined evidence indicate that blacks are both at an increased risk for SCD and arrhythmia compared to whites however, they may derive a similar reduction in the risk of VTA or death from an implanted CRT-D. This further reinforces adherence to international practice guidelines indicating device implantation in eligible patients regardless of their race.
First, this was a post hoc analysis that was not pre-specified and therefore there are certain inherent bias related to post hoc analysis. Second, the groups were different in baseline clinical characteristics; attempt has been made to adjust for confounders, not the least of which are socioeconomical. The fact that even after multivariate adjustment relevant clinical covariates arrhythmic risk was still significantly higher among blacks compared with white patients suggests a possible existence of residual inherent differences. Notably, the observational nature of the study cannot preclude possible nonbiological mechanisms. Third, we had a small proportion of black patients; therefore we had a limited statistical power in some of the analyses presented. However, despite a modest sample size of black patients we had a substantial number of event (VTA = 43, VTA or death = 56) in this population. Therefore, our study is important because data on CRT-D effect to reduce VTA in blacks are scarce. Our intriguing results of blacks deriving the same benefit from CRT-D with reductions in VTA or death as whites need to be further assessed and validated in larger patient cohorts.
In the MADIT-CRT trial, black patients had a significantly higher rate of ventricular tachyarrhythmias or death compared to whites, with either an implanted ICD or a CRT-D. Treatment with CRT-D seems to be as effective in reducing ventricular arrhythmia or death in blacks as in whites.
COMPETENCY IN MEDICAL KNOWLEDGE 1: Blacks have a higher prevalence of HF, they present with HF symptoms at an earlier age, and they have a 2.5 times higher mortality of HF. On the basis of our result blacks HF patients have an increased risk of VTA or death.
COMPETENCY IN MEDICAL KNOWLEDGE 2: There is limited data evaluation the efficacy of CRT in blacks. Our data imply that the efficacy of CRT is similar in blacks and whites.
TRANSLATIONAL OUTLOOK: Additional research is needed to evaluate the risk of ventricular arrhythmia in blacks with HF and the efficacy of CRT in this population. Our results imply that black patients with HF are a high-risk group that benefits from CRT.
For supplemental tables and a figure, please see the online version of this article.
The MADIT-CRT study was supported by an unrestricted research grant from Boston Scientific, St. Paul, Minnesota, to the University of Rochester, Cardiology, Heart Research Follow-Up Program, Rochester, New York. Dr. Zareba has received research grant support from Boston Scientific. Dr. Kutyifa has received research grant support from Boston Scientific and Zoll. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- African American
- confidence interval
- cardiac resynchronization therapy
- defibrillator and combined cardiac resynchronization therapy
- heart failure
- hazard ratio
- implantable cardioverter-defibrillator
- left bundle branch block
- left ventricular end-diastolic volume
- left ventricular hypertrophy
- New York Heart Association
- ventricular tachycardia
- ventricular tachyarrhythmias
- ventricular fibrillation
- Received September 21, 2015.
- Revision received February 25, 2016.
- Accepted March 10, 2016.
- American College of Cardiology Foundation
- Go A.S.,
- Mozaffarian D.,
- Roger V.L.,
- et al.
- Berraho M.,
- Nejjari C.,
- Raherison C.,
- et al.
- Ouellet G.,
- Huang D.T.,
- Moss A.J.,
- et al.
- Zareba W.,
- Klein H.,
- Cygankiewicz I.,
- et al.
- Deo R.,
- Albert C.M.
- East M.A.,
- Jollis J.G.,
- Nelson C.L.,
- Marks D.,
- Peterson E.D.
- Haider A.W.,
- Larson M.G.,
- Benjamin E.J.,
- Levy D.
- Zimetbaum P.J.,
- Buxton A.E.,
- Batsford W.,
- et al.
- Zhang Y.,
- Kennedy R.,
- Blasco-Colmenares E.,
- et al.
- Ziaeian B.,
- Zhang Y.,
- Albert N.M.,
- et al.