CMR DENSE and the Seattle Heart Failure Model Inform Survival and Arrhythmia Risk After CRT

Kenneth C. Bilchick, Daniel A. Auger, Mohammad Abdishektaei, Roshin Mathew, Min Woong Sohn, Xiaoying Cai, Changyu Sun, Aditya Narayan, Rohit Malhotra, Andrew Darby, J. Michael Mangrum, Nishaki Mehta, John Ferguson, Sula Mazimba, Pamela K. Mason, Christopher M. Kramer, Wayne C. Levy, Frederick H. Epstein

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


Objectives: This study sought to determine if combining the Seattle Heart Failure Model (SHFM-D) and cardiac magnetic resonance (CMR) provides complementary prognostic data for patients with cardiac resynchronization therapy (CRT) defibrillators. Background: The SHFM-D is among the most widely used risk stratification models for overall survival in patients with heart failure and implantable cardioverter-defibrillators (ICDs), and CMR provides highly detailed information regarding cardiac structure and function. Methods: CMR Displacement Encoding with Stimulated Echoes (DENSE) strain imaging was used to generate the circumferential uniformity ratio estimate with singular value decomposition (CURE-SVD) circumferential strain dyssynchrony parameter, and the SHFM-D was determined from clinical parameters. Multivariable Cox proportional hazards regression was used to determine adjusted hazard ratios and time-dependent areas under the curve for the primary endpoint of death, heart transplantation, left ventricular assist device, or appropriate ICD therapies. Results: The cohort consisted of 100 patients (65.5 [interquartile range 57.7 to 72.7] years; 29% female), of whom 47% had the primary clinical endpoint and 18% had appropriate ICD therapies during a median follow-up of 5.3 years. CURE-SVD and the SHFM-D were independently associated with the primary endpoint (SHFM-D: hazard ratio: 1.47/SD; 95% confidence interval: 1.06 to 2.03; p = 0.02) (CURE-SVD: hazard ratio: 1.54/SD; 95% confidence interval: 1.12 to 2.11; p = 0.009). Furthermore, a favorable prognostic group (Group A, with CURE-SVD <0.60 and SHFM-D <0.70) comprising approximately one-third of the patients had a very low rate of appropriate ICD therapies (1.5% per year) and a greater (90%) 4-year survival compared with Group B (CURE-SVD ≥0.60 or SHFM-D ≥0.70) patients (p = 0.02). CURE-SVD with DENSE had a stronger correlation with CRT response (r = −0.57; p < 0.0001) than CURE-SVD with feature tracking (r = −0.28; p = 0.004). Conclusions: A combined approach to risk stratification using CMR DENSE strain imaging and a widely used clinical risk model, the SHFM-D, proved to be effective in this cohort of patients referred for CRT defibrillators. The combined use of CMR and clinical risk models represents a promising and novel paradigm to inform prognosis and device selection in the future.

Original languageEnglish
Pages (from-to)924-936
Number of pages13
JournalJACC: Cardiovascular Imaging
Issue number4
StatePublished - Apr 2020

Bibliographical note

Publisher Copyright:
© 2020 American College of Cardiology Foundation


  • cardiac magnetic resonance
  • cardiac resynchronization therapy
  • heart failure
  • implantable cardioverter-defibrillator
  • risk models

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'CMR DENSE and the Seattle Heart Failure Model Inform Survival and Arrhythmia Risk After CRT'. Together they form a unique fingerprint.

Cite this