Skip to content


Open Access

Late gadolinium enhancement score (LGE-Score) for prediction of extensive late gadolinium enhancement in hypertrophic cardiomyopathy

  • Raymond H Chan1,
  • Barry J Maron2,
  • Iacopo Olivotto3,
  • Gabriele Egidy Assenza3,
  • Tammy S Haas2,
  • John R Lesser2,
  • Christiane Gruner1,
  • Andrew Crean1,
  • Harry Rakowski1,
  • James E Udelson4,
  • Ethan J Rowin4,
  • Massimo Lombardi3,
  • Franco Cecchi3,
  • Benedetta Tomberli3,
  • Paolo Spirito3,
  • Francesco Formisano3,
  • Elena Biagini3,
  • Claudio Rapezzi3,
  • Carlo Nicola De Cecco3,
  • Camillo Autore3,
  • Susie N Hong5,
  • Michael C Gibson5,
  • Warren J Manning5,
  • Evan Appelbaum5 and
  • Martin Maron4
Journal of Cardiovascular Magnetic Resonance201517(Suppl 1):Q59

Published: 3 February 2015


Atrial FibrillationCardiac Magnetic ResonanceLate Gadolinium EnhancementHypertrophic CardiomyopathyCardiac Magnetic Resonance Imaging


Extensive fibrosis detected by late gadolinium enhancement (LGE) in contrast enhanced cardiac magnetic resonance (CMR) has recently been identified as a prognostic marker for adverse events in hypertrophic cardiomyopathy (HCM) patients. However, use of CMR in all patients with HCM may be limited by cost and availability. We therefore sought to develop a score based on clinical and imaging variables to predict the probability of extensive LGE, defined by ≥15% of left ventricular (LV) myocardium with LGE.


We assessed the relation between clinical and imaging variables with extensive LGE in an international multicenter cohort of 1293 HCM patients. We used logistic regression to construct integer risk weights for independent variables which predict the presence of extensive LGE. These weights were summed for each patient to create a score (LGE-Score) to predict the probability of extensive LGE.


Extensive LGE was found in 109 of 1293 patients (8.4%). There were 5 independent predictors for extensive LGE: LV ejection fraction (LVEF) (p<0.0001), history of non-sustained ventricular tachycardia (NSVT) (p<0.0001), history of atrial fibrillation (p=0.02), maximal wall thickness (p=0.0001), and significant resting LV outflow tract gradient (LVOT) >30mmHg (p=0.02). Their associated additive risk-weights in the LGE-Score (in parentheses) were as follows: LVEF ≥70%(0), 50-70% (+4 points), <50% (+8 points); history of NSVT (+3 points); history of atrial fibrillation (+2 points); maximal wall thickness<16.25mm (0), 16.25-19.2mm (+1 points), 19.2-22.9mm (+2 points), ≥22.9mm (+3 points); significant resting LVOT gradient (-2 points). The model has an area under receiver operator curve of 0.794. Using a cutoff LGE-Score of ≤+4 points, the negative predictive value for extensive LGE was 98%.


Extensive LGE (myocardial fibrosis/scarring) identifies HCM patients at significantly increased risk for sudden death events, however CMR is not available to all patients. For those who are unable to undergo CMR imaging, this novel predictive score can be used to identify those HCM patients who are highly unlikely to have extensive myocardial scarring.


Figure 1
Figure 1

ROC curve for LGE-Score.

Authors’ Affiliations

Toronto General Hospital, Toronto, Canada
Minneapolis Heart Institute Foundation, Minneapolis, USA
Azienda Ospedaliera Universitaria Careggi, Florence, Italy
Tufts Medical Center, Boston, USA
Beth Israel Deaconess Medical Center, Boston, USA


© Chan et al; licensee BioMed Central Ltd. 2015

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.