Head-to-head comparison of 1 week versus 6 months CMR-derived infarct size for prediction of late events after STEMI

Int J Cardiovasc Imaging. 2013 Oct;29(7):1499-509. doi: 10.1007/s10554-013-0239-1. Epub 2013 Jun 4.

Abstract

Infarct size (IS) at 1 week after ST-elevation myocardial infarction (MI) diminishes during the first months. The incremental prognostic value of IS regression and of scar size (SS) at 6 months is unknown. We compared cardiovascular magnetic resonance (CMR)-derived IS at 1 week and SS at 6 months after MI for predicting late major adverse cardiac events (MACE). 250 patients underwent CMR at 1 week and 6 months after MI. IS and SS were determined as the extent of transmural late enhancement (in >50 % of wall thickness, ETLE). During 163 weeks, 23 late MACE (cardiac death, MI or readmission for heart failure after the 6 months CMR) occurred. Patients with MACE had a larger IS at 1 week (6 [4-9] vs. 3 [1-5], p < .0001) and a larger SS at 6 months (5 [2-6] vs. 3 [1-5], p = .005) than those without MACE. Late MACE rates in IS >median were higher at 1 week (14 vs. 4 %, p = .007) and in SS >median at 6 months (12 vs. 5 %, p = .053). The C-statistic for predicting late MACE of CMR at 1 week and 6 months was comparable (.720 vs. .746, p = .1). Only ETLE at 1 week (HR 1.31 95 % CI [1.14-1.52], p < .0001, per segment) independently predicted late MACE. CMR-derived SS at 6 months does not offer prognostic value beyond IS at 1 week after MI. The strongest predictor of late MACE is ETLE at 1 week.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Female
  • Heart Failure / etiology
  • Humans
  • Kaplan-Meier Estimate
  • Magnetic Resonance Imaging, Cine*
  • Male
  • Middle Aged
  • Myocardial Infarction / complications
  • Myocardial Infarction / diagnosis*
  • Myocardial Infarction / mortality
  • Myocardial Infarction / pathology
  • Myocardium / pathology*
  • Patient Readmission
  • Predictive Value of Tests
  • Prognosis
  • Proportional Hazards Models
  • Prospective Studies
  • Registries
  • Risk Factors
  • Time Factors