Relationship of LVEF and Myocardial Scar to Long-Term Mortality Risk and Mode of Death in Patients with Nonischemic Cardiomyopathy

Igor Klem, Michael Klein, Mohammad Khan, Eric Y. Yang, Faisal Nabi, Alexander Ivanov, Lubna Bhatti, Brenda Hayes, Edward A. Graviss, Duc T. Nguyen, Robert M. Judd, Raymond J. Kim, John F. Heitner, Dipan J. Shah

Research output: Contribution to journalArticlepeer-review

45 Scopus citations


BACKGROUND: Nonischemic cardiomyopathy is a leading cause of reduced left ventricular ejection fraction (LVEF) and is associated with high mortality risk from progressive heart failure and arrhythmias. Myocardial scar on cardiovascular magnetic resonance imaging is increasingly recognized as a risk marker for adverse outcomes; however, left ventricular dysfunction remains the basis for determining a patient’s eligibility for primary prophylaxis with implantable cardioverter-defibrillator. We investigated the relationship of LVEF and scar with long-term mortality and mode of death in a large cohort of patients with nonischemic cardiomyopathy. METHODS: This study is a prospective, longitudinal outcomes registry of 1020 consecutive patients with nonischemic cardiomyopathy who underwent clinical cardiovascular magnetic resonance imaging for the assessment of LVEF and scar at 3 centers. RESULTS: During a median follow-up of 5.2 (interquartile range, 3.8, 6.6) years, 277 (27%) patients died. On survival analysis, LVEF ≤35% and scar were strongly associated with all-cause (log-rank test P=0.002 and P<0.001, respectively) and cardiac death (P=0.001 and P<0.001, respectively). Whereas scar was strongly related to sudden cardiac death (SCD; P=0.001), there was no significant association between LVEF ≤35% and SCD risk (P=0.57). On multivariable analysis including established clinical factors, LVEF and scar are independent risk markers of all-cause and cardiac death. The addition of LVEF provided incremental prognostic value but insignificant discrimination improvement by C-statistic for all-cause and cardiac death, but no incremental prognostic value for SCD. Conversely, scar extent demonstrated significant incremental prognostic value and discrimination improvement for all 3 end points. On net reclassification analysis, the addition of LVEF resulted in no significant improvement for all-cause death (11.0%; 95% CI, −6.2% to 25.9%), cardiac death (9.8%; 95% CI, −5.7% to 29.3%), or SCD (7.5%; 95% CI, −41.2% to 42.9%). Conversely, the addition of scar extent resulted in significant reclassification improvement of 25.5% (95% CI, 11.7% to 41.0%) for all-cause death, 27.0% (95% CI, 11.6% to 45.2%) for cardiac death, and 40.6% (95% CI, 10.5% to 71.8%) for SCD. CONCLUSIONS: Myocardial scar and LVEF are both risk markers for all-cause and cardiac death in patients with nonischemic cardiomyopathy. However, whereas myocardial scar has strong and incremental prognostic value for SCD risk stratification, LVEF has no incremental prognostic value over clinical measures. Scar assessment should be incorporated into patient selection criteria for primary prevention implantable cardioverter-defibrillator placement.

Original languageEnglish (US)
Pages (from-to)1343-1358
Number of pages16
Issue number14
StatePublished - Apr 6 2021


  • cardiomyopathies
  • death
  • defibrillators,implantable
  • myocardial scar

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine
  • Physiology (medical)


Dive into the research topics of 'Relationship of LVEF and Myocardial Scar to Long-Term Mortality Risk and Mode of Death in Patients with Nonischemic Cardiomyopathy'. Together they form a unique fingerprint.

Cite this