Author + information
- Received September 22, 2014
- Revision received January 1, 2015
- Accepted January 9, 2015
- Published online June 1, 2015.
- Paul A. Heidenreich, MD, MS∗∗ (, )
- Vivian Tsai, MD†,
- Haikun Bao, MS‡,
- Jeptha Curtis, MD‡,
- Mary Goldstein, MD, MS∗,
- Lesley Curtis, PhD§,
- Adrian Hernandez, MD, MS§,
- Pamela Peterson, MD, MS‖,
- Mintu P. Turakhia, MD, MAS∗ and
- Frederick A. Masoudi, MD, MS‖
- ∗Veterans Administration Palo Alto Healthcare System, Palo Alto, California
- †Palo Alto Medical Foundation, Palo Alto, California
- ‡Department of Medicine, Yale School of Medicine, New Haven, Connecticut
- §Duke Clinical Research Institute, Durham, North Carolina
- ‖University of Colorado Anschutz Medical Campus, Aurora, Colorado
- ↵∗Reprint requests and correspondence:
Dr. Paul Heidenreich, Veterans Administration Palo Alto Healthcare System, 111C Cardiology, 3801 Miranda Avenue, Palo Alto, California 94306.
Objectives This study sought to describe the use of CRT-D and its association with survival for older patients.
Background Many patients who receive cardiac resynchronization therapy with defibrillator (CRT-D) in practice are older than those included in clinical trials.
Methods We identified patients undergoing ICD implantation in the National Cardiovascular Disease Registry (NCDR) ICD registry from 2006 to 2009, who also met clinical trial criteria for CRT, including left ventricular ejection fraction (LVEF) ≤35%, QRS ≥120 ms, and New York Heart Association (NYHA) functional class III or IV. NCDR registry data were linked to the social security death index to determine the primary outcome of time to death from any cause. We identified 70,854 patients from 1,187 facilities who met prior trial criteria for CRT-D. The mean age of the 58,147 patients receiving CRT-D was 69.4 years with 6.4% of patients age 85 or older. CRT use was 80% or higher among candidates in all age groups. Follow-up was available for 42,285 patients age ≥65 years at 12 months.
Results Receipt of CRT-D was associated with better survival at 1 year (82.1% vs. 77.1%, respectively) and 4 years (54.0% vs. 46.2% , respectively) than in those receiving only an ICD (p < 0.001). The CRT association with improved survival was not different for different age groups (p = 0.86 for interaction).
Conclusions More than 80% of older patients undergoing ICD implantation who were candidates for a CRT-D received the combined device. Mortality in older patients undergoing ICD implantation was high but was lower for those receiving CRT-D.
Chronic resynchronization therapy (CRT) has been shown to improve survival and symptoms for patients with heart failure, reduced ejection fraction, and wide QRS intervals (1–5). However, the average age of patients included in the trials (mean 62 to 67 years of age) is lower than that of the general population of patients with heart failure in the United States. Several studies have found that many patients over age 75 years who receive CRT demonstrate improvement in symptoms and left ventricular ejection fraction (LVEF), however data for a survival benefit have been limited to subgroups of randomized trials, which were small in size (6).
We sought to describe the use of CRT with defibrillator (CRT-D) among older patients in the United States. We used data from the National Cardiovascular Disease Registry (NCDR) ICD registry to determine how often older candidates for a CRT and an implantable cardioverter-defibrillator (ICD) receive CRT-D or only an ICD. We also sought to determine the difference in outcome for US older patients receiving CRT-D compared to ICD alone among those who were candidates for CRT.
We used data from the NCDR-ICD registry, created in 2006 through a mandate from the Centers of Medicare and Medicaid Services (CMS), which requires all hospitals to report data for ICD implantations for primary prevention (7). Although hospitals are required to enter data only for Medicare beneficiaries, most institutions register all ICD implantations. Institutions use a standardized questionnaire to enter clinical data including patient clinical characteristics, device used, other treatments, and hospital outcome. Data are subjected to quality control checks for missing and improperly coded items, and a random audit is conducted annually through site visits (8).
We identified patients who were candidates for CRT (QRS ≥120 ms, LVEF ≤35%, New York Heart Association [NYHA] functional class III or IV, and had their first ICD implanted at a facility reporting for Medicare and non-Medicare patients. Of these patients, we included 72,563 after excluding patients with previously implanted pacemakers (16,640) and previous cardiac arrest (9,993) and cardiac transplants (277) and an unclear type of ICD (105). We also included 1,994 patients with epicardial leads.
The primary outcome was survival following ICD or CRT-D implantation determined by linking the patient’s social security number with the Social Security Death Index. Of the 72,563 patients, 1,632 (2.2%) could not be matched by social security number, and another 77 (0.1%) had a recorded death date prior to implantation date, leaving 70,854 patients for the mortality analysis.
Characteristics (Table 1) of the ICD recipients who received CRT were compared with those who did not, using the chi-square test for categorical variables and unbalanced Student t test for continuous variables. Two variables with missing values of more than 5% (B-type natriuretic peptide level and PR interval) were excluded from the primary analyses. Variables with low missing rates were imputed as the most common category for the categorical variables and the median value for continuous variables. For categorical variable with a missing rate >5%, a category was added to indicate “missingness.” The multiple logistic regression model was used to examine associations between patients, hospitals, and physicians’ characteristics and the use of CRT-D, and a forward stepwise selection method (entry p value of 0.05 and retention p value of 0.05) was used to identify variables most strongly associated with the use of CRT-D. Cox proportional hazard models were used to explore the impact of CRT-D use on survival among different age groups, with adjustment of patient, hospital, and physician’s characteristics. For multivariable analyses, we used the presence of ischemic heart disease (yes/no) to indicate cause of heart failure. All analyses were performed with SAS version 9.2 software (SAS Institute, Cary, North Carolina). All analyses were approved by the Yale University Human Investigation Committee.
Patient characteristics of the 70,854 patients included in the analysis are listed in Table 1 with groupings by age. Mean age was 69 years, and 69% were male. Those 75 years of age or older (n = 27,359) accounted for 39% of those undergoing ICD placement. Although most of the differences were statistically significant due in part to the large sample size, not all were clinically relevant. Older patients were more likely to be non-Hispanic white compared to younger age groups. Ischemia, as the cause of cardiomyopathy, increased with age as did LVEF (although the mean LVEF was <25% for all age groups). As expected, comorbidities including atrial fibrillation or flutter increased with age.
Hospital and physician characteristics
Characteristics of the 1,187 hospitals and training of the physicians performing the implantation are shown in Table 2. Older patients were less likely to be treated at hospitals that had public financing or were members of the Council on Teaching Hospitals than were younger patients. Older patients were also more likely than younger patients to be hospitalized in the Mid-Atlantic and Pacific U.S. Census regions. The reverse was true (oldest patients were less likely to receive CRT if a candidate) for those in southern regions (South Atlantic, South East and South West). Most physicians performing implantation were board certified in electrophysiology for both older and younger patients.
Use of CRT-D
Patients receiving CRT-D (n = 58,147) were slightly older (mean age of 69.4 years) than those receiving only an ICD (12,707 patients were a mean 68.9 years of age; p < 0.0001). CRT use among candidates was most common in those 65 to 84 years of age (83%) and was 80% or higher among candidates in all age groups (Figure 1). In a multivariate model, age showed a nonlinear relationship with CRT-D use. Compared to those younger than 55 year of age, the odds ratio for CRT-D was 1.16 (95% confidence interval [CI]: 1.08 to 1.26), for those 55 to 64 years of age, 1.20 (95% CI: 1.11 to 1.30) for those 65 to 74 years of age, 1.23 (95% CI: 1.14 to 1.34) for those 75 to 84 years of age, and 1.08 (95% CI: 0.97 to 1.20) for those 85 years of age and older. The c-statistic for the model of CRT use was 0.726.
Survival following implantation of an ICD or CRT-D is shown in Figure 2. Patients receiving CRT-D implants were more likely to survive than those receiving only an ICD (p < 0.0001). Follow-up was available for 42,285 patients age 65 years and older at 12 months. For those 65 years and older, receipt of CRT-D was associated with better survival at 1 year (82.1% vs. 77.1%, respectively) and 4 years (54.0% vs. 46.2%, respectively) compared with those receiving only an ICD (p < 0.001). CRT-D was associated with a similar increase in survival compared to ICD alone for both older and younger age groups (Figure 3). When a multivariate model of survival was stratified by age group there was a similar association with increased survival across age groups (p = 0.86 for interaction) (Figure 4).
Our study found a high rate of CRT use among candidates across all age groups. Use was nonlinear with age, with the highest rate of use among candidates 65 to 75 years of age (odds ratio of 1.20 compared to those under age 55 years) and slightly lower use in those over 75 years (odds ratio: 1.08). Receipt of CRT-D was associated with better survival at 1 and 4 years compared with those receiving only an ICD. The CRT benefit for survival was not significantly different for different age groups.
The number of patients 75 years of age and older who were enrolled in clinical trials of CRT-D use was relatively small, and there is uncertainty regarding the benefit of CRT in the oldest patients. Kron et al. (6) and Kron and Conti (9) examined the benefit of CRT in 174 patients above the age of 75 years from the MIRACLE (Multicenter InSync Randomized Clinical Evaluation) and MIRACLE-ICD (Multicenter InSync ICD Randomized Clinical Evaluation) trials. They were able to show a statistically significant improvement in NYHA functional class and LVEF for those age 75 years. However, they could not demonstrate an improvement in 9 other endpoints including quality of life, peak Vo2, exercise time QRS duration, and ventricular dimensions. In the CARE-HF (CArdiac REsynchronization in Heart Failure) trial, reduction in death or unplanned heart failure hospitalization was not clearly different in those older (n = 407) or younger than the mean age of 66.4 years (5). In the COMPANION (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure) trial, the benefit was also not significantly different between those above (n = 853) and below age 65 (4). In a substudy of then MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy), 331 patients 75 years of age or older had reduction in heart failure or death similar to that in younger patients (10). Although the lack of significant differences in CRT benefit between old and young patients are encouraging, these analyses have limited power (N) for older age groups, making it difficult to draw meaningful conclusions.
One purpose of the ICD registry was to determine whether the benefits of devices observed in clinical trials were consistent with the use of these devices in the community. Our study provides confirmation of the small subgroup analyses from clinical trials and indicates that CRT use in older adults in the United States is associated with improved survival compared to use of an ICD alone. This finding is also consistent with prior work suggesting a common mechanism for benefit with CRT across age groups. Nonrandomized studies have found similar reversals of left ventricular remodeling for old and young patients after CRT placement (11–14).
We found that a large fraction of CRT-D (39%) were implanted in patients 75 years old or older. This fraction of those receiving CRT in older adults was similar to those in the PREMIER registry (22% of CRT-D implants age ≥80 years) and single-center data from the Netherlands (44% age ≥70 years) (14) and Belgium (22% age ≥80 years) (11) but higher than that in a national registry from Israel (20% age ≥75 years) (15). Our study did not include patients receiving CRT without a defibrillator (CRT-P). Data from the PREMIER registry found that CRT-P made up 13% of all CRT implants and that these devices were used more frequently in the oldest patients (38% of CRT-P were placed in those ≥80 years of age).
Our study also demonstrates that among candidates for CRT and ICD, the oldest patients are appropriately receiving both devices. However, the use of CRT among all candidates is likely much lower. Data from the IMPROVE-HF study from 2005 to 2007 found that CRT was used only 39% of the time among candidates (16). Significant differences in use by age were noted for women. Females older than 76 years received CRT 30% of the time compared to 47% for females under 65 years of age. There was not an important interaction with age on CRT use for men. Overall there was a slightly lower use of CRT among older than among younger patients. Our findings suggest that once an older candidate is referred for an ICD, they will usually receive CRT if appropriate.
Although our study did not examine complication rates, other investigations have found complication rates for older patients are comparable to those for younger patients (10,17,18). Although in-hospital death is higher with older age, device-related complications requiring surgical intervention were <3% at 1 year regardless of age (14).
Other registries have noted less use of CRT in the elderly (19,20). In the Swedish Heart Failure Registry, 37% of patients over age 80 years were candidates for CRT but did not receive them compared to 32% for those 66 to 80 years of age and 23% for those 65 years and younger. Our study included only patients undergoing ICD therapy. Although we found there was not a marked age bias in use of CRT if undergoing an ICD, we do not know if there are differences in use of CRT-P or ICD use itself.
Our study has several potential limitations. As noted above, we could not determine the number of older patients who received CRT without an ICD (CRT-P). Although we know the number of ICD recipients who were candidates for ICD, we do not know how many patients were CRT-D candidates but did not undergo ICD implantation. We also do not know whether age was related to the degree of symptom improvement following CRT placement. Additional studies are needed to determine the age threshold at which the life-years gained from CRT become too small to justify the use of the device. Such a study will need to compare CRT patients with those receiving no device (ICD or CRT) and could not be done with this dataset. Although CRT use by race (21) and sex (22) has been examined, other studies are needed to explore the potentially complex interactions between race, sex, age, CRT use, and outcome. Our patient population was based on CRT guidelines at the time of data collection, and therefore, we did not include patients with NYHA functional class II symptoms. Finally, the outcomes other than primary (survival) should be interpreted in light of the multiple comparisons performed in the analysis.
Over 30% of those undergoing CRT-D in the United States are over the age of 75 years. Older patients undergoing ICD placement who were also candidates for CRT received the combined device (CRT-D) over 80% of the time. For those receiving CRT-D, mortality was lower than for those receiving an ICD alone, a finding that was consistent across all age groups.
COMPETENCY IN MEDICAL KNOWLEDGE: The use of CRT-D among candidates in the United States is high regardless of age. Mortality following implantation was improved for candidates if they received CRT-D.
TRANSLATIONAL OUTLOOK: Further studies of CRT in the elderly are needed to determine the age threshold at which the life-years gained from CRT become too small to justify the use of the device.
Dr. Heidenreich is supported by a grant from Veterans Administration Quality Enhancement and Research Initiative 04-326. Dr. Turakhia is supported by Veterans Health Services Research and Development Career Development Award CDA09027-1. Dr. Peterson is supported by Agency for Healthcare Research and Quality grant K08 HS019814-01. Dr. Curtis owns stock in Medtronics; and has received grant support from Boston Scientific through her institution. Dr. Hernandez has received grant support from Medtronics-Research. Dr. Turakhia has received grant support from Medtronics, iRhythm, Gilead Sciences; is an employee of U.S. Department of Veterans Affairs; has consulted for Medtronic, St. Jude Medical, and Precision Health Economics; and has received lecture honoraria from Biotronik. Dr. Masoudi is senior medical officer for American College of Cardiology. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- cardiac resynchronization therapy
- cardiac resynchronization therapy with defibrillator
- cardiac resynchronization therapy with pacing and no defibrillator
- implantable cardioverter-defibrillator
- left ventricular ejection fraction
- National Cardiovascular Disease Registry
- New York Heart Association
- Received September 22, 2014.
- Revision received January 1, 2015.
- Accepted January 9, 2015.
- 2015 American College of Cardiology Foundation
- Higgins S.L.,
- Hummel J.D.,
- Niazi I.K.,
- et al.
- Kron J.,
- Aranda J.M. Jr..,
- Miles W.M.,
- et al.
- Messenger J.C.,
- Ho K.K.,
- Young C.H.,
- et al.
- Suleiman M.,
- Goldenberg I.,
- Haim M.,
- et al.,
- On behalf of the Israeli Working Group on P, electrophysiology
- Zusterzeel R.,
- Curtis J.P.,
- Caños D.A.,
- et al.