Use of Cardiac Resynchronization Therapy Defibrillator in US Hospitals

Key Points

Question  Has use of cardiac resynchronization therapy defibrillator (CRT-D) changed with the American College of Cardiology/American Heart Association/Heart Rhythm Society guideline update in 2012?

Findings  In this study of more than 85 000 new CRT-D implants from 2012 to 2015 through the National Cardiovascular Data Registry for implantable cardioverter-defibrillators, there was an increase in guideline-concordant implantation during the period. The major fraction of guideline-discordant implants were clustered at a minority of hospitals.

Meaning  Nationally, an increase in guideline-concordant CRT-D implantation was noted, suggesting improved patient selection in those most likely to benefit from this therapy; continued, focused opportunities for improvement in the use of CRT-D may be useful in addressing guideline-discordant implants at a minority of hospitals.

Importance  Cardiac resynchronization therapy (CRT) provides significant reduction in morbidity and mortality in select patients with left ventricular systolic dysfunction and specific parameters of electrocardiographic evidence of dyssynchrony. Relative to the 2012 American College of Cardiology/American Heart Association/Heart Rhythm Society guideline update for patient selection, little is known about the contemporary use of CRT in the United States.

Objective  To describe the use of CRT defibrillator (CRT-D) in the period around guideline revision.

Design, Setting, and Participants  All patients undergoing new CRT-D implantations in the National Cardiovascular Data Registry for implantable cardioverter-defibrillators from January 1, 2012, to December 31, 2015, at 1710 participating hospitals were identified for this population-based study. Rates of CRT-D implantation that were concordant and discordant with the 2012 American College of Cardiology/American Heart Association/Heart Rhythm Society update of the 2008 guidelines for device-based therapy were determined. Analysis began in January 2012.

Main Outcomes and Measures  Increase in guideline-concordant CRT-D implantation.

Results  Among 135 253 patients undergoing initial CRT-D implantation, 88 923 were included in the study cohort, of which 73 859 implants (83.1%) were guideline concordant. The proportion of guideline-concordant devices increased from 81.2% (16 710 of 20 481) in 2012 to 84.2% (20 515 of 24 356) in 2015 (P for trend < .001). Significant clustering was noted with 33% (565 of 1710) of hospitals accounting for greater than 70% (10 545 of 15 065) of guideline-discordant CRT-D implants. Conduction abnormalities, in particular, underlying right bundle branch block (3597 [23.9%] vs 7425 [10.1%]; P < .001) and nonspecific intraventricular conduction delay (3341 [22.2%] vs 4769 [6.5%]; P < .001) were more common in those who received guideline-discordant devices.

Conclusions and Relevance  Rates of guideline-concordant CRT-D implantation increased during the study. The major fraction of guideline-discordant implants were clustered at a minority of hospitals. Conduction abnormalities, particularly non–left bundle branch block and nonspecific intraventricular conduction delay, correlated with guideline-discordant implants indicating continued opportunity for dissemination and understanding of guideline updates.

In select patients, cardiac resynchronization therapy (CRT) is an important therapeutic option that provides significant morbidity and mortality benefits. Meta-analyses and randomized clinical trials evaluating CRT have shown improvement in heart failure–related hospitalizations, mortality, and multiple quality of life measures.^1-4 As such, rates of CRT implantation in the United States have increased substantially.^5,6 However, benefits of CRT are greatest among patients with left bundle branch block (LBBB) and QRS width greater than 150 milliseconds in normal sinus rhythm, as outlined in the 2012 American College of Cardiology/American Heart Association/Heart Rhythm Society (ACC/AHA/HRS) update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities.⁷

Previous work, conducted before guideline restructuring, found that a substantial fraction of patients received CRT defibrillator (CRT-D) outside of guideline indications.⁸ Furthermore, data have shown no benefit and potential harmful outcomes with CRT in patients with right bundle block morphology.^9,10 Additionally, a prior study that found worse outcomes in patients receiving non–evidence-based devices.¹¹ In this study, we sought to describe the changing state of CRT-D use in the United States relative to the 2012 guideline revision for device-based therapies. Focusing on trends, factors associated with guideline-concordant and guideline-discordant implantation, and hospital-level variation in guideline-discordant CRT implantation, we hope to understand use patterns and ultimately inform potential strategies to optimize evidence-based device use.

The National Cardiovascular Data Registry (NCDR) for implantable cardioverter-defibrillators (ICD) is the largest national registry of device implantations across the United States. The ICD registry has been described previously.^12,13 More than 75% of hospitals enter data on all ICD or CRT implantation regardless of insurance or indication. Quality assurance is achieved through automatic system validation, education and training of staff, reporting of completeness, and random on-site auditing. Institutional review board approval and patient consent were waived because data were from a national quality registry.

Guideline-concordant CRT-D implantation was defined as a procedure meeting criteria for class I, class IIa, and class IIb indications under the 2012 ACC/AHA/HRS guidelines published on October 12, 2012.⁷ Guideline-discordant CRT-D implantation was defined as a procedure that meets a class III (New York Heart Association class I/II functional status and non-LBBB QRS morphology with QRS duration <150 milliseconds) indication under the 2012 ACC/AHA/HRS guidelines⁷ or in any patient with an ejection fraction greater than 35%.

We evaluated all patients undergoing new CRT-D implantation in the NCDR ICD registry from January 1, 2012, to December 31, 2015, at 1710 participating hospitals. Analysis began in January 2012. Given gradual dissemination and implementation of guideline updates, the intent of this analysis was to evaluate patterns of CRT-D use around the period of guideline revision. Patients were excluded if necessary data were not available for the determination of CRT guideline concordance. The necessary variables included ejection fraction, New York Heart Association class status, QRS morphology, and QRS duration. Next, we excluded 4038 patients with an independent pacing indication (second- or third-degree heart block or bradycardic cardiac arrest) and 6676 patients with a preexisting pacemaker (eFigure 1 in the Supplement).

Comparisons of patient and procedural characteristics were completed using t test for continuous variables and χ² test for categorical variables. A hierarchical logistic regression model was constructed to determine predictors of receiving a guideline-discordant CRT-D system. We then plotted the distribution of guideline-discordant CRT-D implantation rates by hospital to determine hospital-level frequency of guideline-discordant CRT-D use. To quantify the magnitude of clustering, a median odds ratio (OR) for guideline-discordant CRT-D implantation was calculated across all hospitals. A 2-tailed P value was considered significant at .05.

Of 135 253 patients analyzed, 88 923 received initial CRT-D implantation and fit inclusion criteria. Table 1 displays demographic, comorbidities, electrocardiographic data, and hospital-level data split by guideline-concordant or guideline-discordant CRT use. Among the study cohort, 73 859 (83.1%) were deemed guideline concordant when analyzed against the 2012 ACC/AHA/HRS update of the 2008 guidelines at the time of device implantation. Of 73 859 individuals with guideline-concordant implants, 37 165 (50.3%) were performed under class I, 29 224 (39.6%) under class IIa, and 7470 (10.1%) under class IIb indications. Left bundle branch block was much more prevalent in those receiving guideline-concordant devices (61 028 of 73 859 [82.6%] vs 7425 of 15 064 [49.3%]; P < .001). Conversely, right bundle branch block (3597 of 15 064 [23.9%] vs 7425 of 73 859 [10.1%]; P < .001) and nonspecific intraventricular conduction delay (3341 of 15 064 [22.2%] vs 4769 of 73 859 [6.5%]; P < .001) were more common in those who received guideline-discordant implants.

The Figure displays a histogram of the percentage of guideline-discordant initial CRT-D implantations by percentage of hospitals. Guideline-concordant CRT-D use was greater than 70% (1197 of 1710) for more than 70% (51 700 of 73 895) of hospitals. Guideline-discordant CRT-D implants were clustered with 18% (308 of 1710) of hospitals accounting for 50% (7532 of 15 064) and 33% (565 of 1710) of hospitals accounting for more than 70% (10 545 of 15 065) of guideline-discordant implants. No significant differences in hospital-level characteristics were noted between facilities accounting for most guideline-discordant CRT-D implants compared with the remainder of the hospitals. Across all hospitals, the median (interquartile range) guideline-discordant CRT-D implantation rate was 15.4% (9.1%-21.9%) with a median OR of 1.42 (95% CI, 1.36-1.48). The median OR can be interpreted as the odds that patients with similar covariates undergo guideline-discordant CRT-D implantation at 1 randomly selected hospital among those studied compared with another. In a sensitivity analysis, no significant change in guideline-discordant clustering of CRT-D implants was noted after elimination of low-implant centers (ie, analysis of hospitals with >10 CRT-D implants yearly).

During the study, the rate of guideline-discordant CRT-D use declined from 18.4% (3771 of 20 481) in 2012 to 15.8% (3841 of 24 356) in 2015 (P value for trend <.001) (eFigure 2 in the Supplement). Table 2 displays predictors of guideline-discordant CRT-D use by demographic, clinical, and hospital-level characteristics. Non-LBBB conduction defects showed significantly higher odds of guideline-discordant device use (OR for right bundle branch block: 14.7 [95% CI, 13.4-16.1], P < .001 and OR for nonspecific intraventricular conduction delay: 7.9 [95% CI, 7.2-8.7], P < .001).

We sought to describe CRT-D use compared with the period of publication and dissemination of the 2012 ACC/AHA/HRS updated guideline recommendations among patients receiving new CRT-D implants. Analysis of more than 85 000 patients undergoing CRT-D implantation indicated that (1) 83.1% of all CRT-D device implants were guideline concordant, (2) most guideline-discordant CRT-D implants were clustered at a minority of hospitals, and (3) the proportion of guideline concordance increased over time. Factors comprising guideline definitions including QRS morphology, QRS width, and nonsinus rhythm were associated with guideline-discordant device use.

In select patients, randomized clinical trials have shown significant improvement in morbidity and mortality following CRT implantation,^1,2 although newer data have shown limited benefit and potential harm in subsets of patients receiving CRT.^9,14 Prior analyses, limited to a subset of patients undergoing CRT-D implantation and conducted prior to the 2012 guideline update, determined 23.7% of implants were outside of guideline recommendations.^7,8 The intent of the present study was to address the gap in knowledge pertaining to the use of CRT-D under the 2012 guideline revision over a population that included new CRT-D implants.

Our findings demonstrate an increase in guideline-concordant patient selection for CRT-D implantation during the study. Evidence has increasingly shed light on the nonuniform benefit of CRT, with particular discrimination based on QRS duration and morphology.^10,15 Subset analyses from landmark trials^1,2 have not shown support for patients with heart failure and concomitant right bundle branch block. Analyses evaluating CRT benefit for patients with nonspecific intraventricular conduction delay have shown similar findings.¹⁶ Our data show non-LBBB as a predictor of guideline-discordant CRT-D implantation, although guideline discordance requires analysis of a combination of patient factors. In addition to patient selection with particular attention to QRS morphology, this suggests continued room for improvement in dissemination and understanding of guideline changes.

Importantly, significant clustering of guideline-discordant use of CRT-D was seen at a small proportion of implanting hospitals. Hospital-level analyses between this cohort of hospitals compared with the remainder did not reveal differences in hospital-level characteristics attributable to disproportionate guideline-discordant CRT-D implantation rates. Among other factors, this may indicate patient or physician preference, patient-level characteristics, clinician knowledge or understanding of guideline metrics, or practice environment as contributors to hospital-level clustering. These findings suggest continued opportunity for further study seeking to understand and address drivers of guideline-discordant CRT-D use.

The results of our analysis should be considered in light of the following limitations. Assessment of functional status and life expectancy greater than 1 year are important considerations prior to device implantation and may be difficult to measure through observational data. Our study consists of hospitals participating in the NCDR and may not be generalizable to all facilities performing device implantation. However, the NCDR ICD registry captures a large representation of US hospitals performing device implantation and for the duration of the study period, reporting of device implantation was mandated by Centers for Medicare & Medicaid Services.¹² Our analysis may not have captured all approaches to CRT implantation (ie, surgical placement of an epicardial coronary sinus pacing lead) or newer options such as His-bundle pacing. Given limited available provider characteristics through the data set, we did not focus on operator characteristics associated with CRT-D implantation.

In this large national registry, rates of guideline-concordant CRT-D implantation consistent with the 2012 ACC/AHA/HRS guidelines have increased from 2012 to 2015. The major fraction of guideline-discordant CRT-D implants were clustered at a minority of hospitals. This suggests continued opportunities for improvement in the use of CRT-D to target those most likely to be benefit from this therapy.

Corresponding Author: Amneet Sandhu, MD, MSc, Division of Cardiology, University of Colorado School of Medicine, 12631 E 17th Ave, B130, Aurora, CO 80045 (amneet.sandhu@ucdenver.edu).

Accepted for Publication: April 18, 2019.

Published Online: June 19, 2019. doi:10.1001/jamacardio.2019.1755

Author Contributions: Dr Sandhu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Sandhu, Minges, Varosy, Borne, Zipse, Marzec, Peterson, Masoudi, Bradley.

Acquisition, analysis, or interpretation of data: Sandhu, Bao, Minges, Varosy, Borne, Marzec, Peterson, Masoudi, Bradley.

Drafting of the manuscript: Sandhu, Borne.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Sandhu, Bao, Marzec.

Obtained funding: Sandhu.

Administrative, technical, or material support: Sandhu, Minges, Varosy, Zipse, Marzec, Masoudi.

Supervision: Sandhu, Varosy, Borne, Zipse, Bradley.

Conflict of Interest Disclosures: Dr Minges reports receiving salary support to provide analytic services to the American College of Cardiology National Cardiovascular Data Registry during the conduct of the study. Dr Peterson reports receiving grants from the National Heart, Lung, and Blood Institute and personal fees from the American Heart Association outside the submitted work. Dr Masoudi reports serving as chief scientific advisor for the American College of Cardiology National Cardiovascular Data Registry outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the National Cardiovascular Data Registry for implantable cardioverter-defibrillators, which is an initiative of the American College of Cardiology Foundation.

Role of the Funder/Sponsor: The American College of Cardiology oversees the National Cardiovascular Data Registry, funds the collection of data in the implantable cardioverter-defibrillators registry, and representatives of the implantable cardioverter-defibrillators Research and Publications committee approved the final manuscript. The funder had no role in the design and conduct of the study; management, analysis, and interpretation of the data; preparation or review of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this article represent those of the authors and do not necessarily represent the official views of the National Cardiovascular Data Registry or its associated professional societies identified at http://www.ncdr.com. Additionally, the views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the US Department of Veterans Affairs or the United States government.