Myocardial Scar by Pulse-Cancellation Echocardiography Is Independently Associated with Appropriate Defibrillator Intervention for Primary Prevention after Myocardial Infarction

J Am Soc Echocardiogr. 2020 Sep;33(9):1123-1131. doi: 10.1016/j.echo.2020.04.020. Epub 2020 Jul 1.

Abstract

Background: Myocardial scar burden impacts prognosis in patients with coronary artery disease who have experienced a myocardial infarction (MI). This has been demonstrated by late gadolinium enhancement cardiac magnetic resonance. Clinical experience with echocardiography suggests that the reflected ultrasound signal is enhanced in infarcted myocardial segments. Scar imaging with an ultrasound multipulse scheme (eScar) has been preliminarily validated in prior studies.

Objective: To assess whether scar burden, as detected by eScar, is associated with implantable cardioverter-defibrillator (ICD) shocks in post-MI patients.

Methods: We retrospectively selected 50 post-MI patients with an ejection fraction <35% who received an ICD for primary prevention and subsequently had at least one appropriate shock (cases). These were compared with 50 post-MI patients, matched for clinical variables and ejection fraction, who never experienced an appropriate defibrillator shock (controls). Subjects were assessed with the eScar technique at the time of implantation or during follow-up.

Results: An eScar was present in ≥1 segment in 40 of 50 (80%) cases vs 26 of 50 (52%) controls and was associated with appropriate ICD shocks (P = .004). Receiver operating characteristic curve analysis, using a threshold of ≥3 segments by eScar, showed an area under the curve (AUC) of 0.715. On models including clinical and echocardiographic variables, eScar remained significantly associated with ICD shocks (P = .050 or P = .033 depending on the model). Adding eScar to a multivariate logistic regression model including indexed left ventricular end-systolic volume led to an increase in AUC from 0.734 to 0.782 (P = .049), while substituting indexed left ventricular end-diastolic volume for indexed left ventricular end-systolic volume resulted in a nonsignificant increase in AUC from 0.747 to 0.785 (P = .098).

Conclusions: Presence and extent of eScar were independently associated with appropriate ICD shocks in this study of patients with prior MI and reduced ejection fraction. However, the addition of eScar assessment to the clinical multivariable model that included also indexed left ventricular end-diastolic volume did not provide significant incremental value.

Keywords: Appropriate shock; Echocardiography; Myocardial infarction; Myocardial scar; Pulse-cancellation ultrasound.

MeSH terms

  • Cicatrix / diagnostic imaging
  • Contrast Media
  • Defibrillators, Implantable*
  • Echocardiography
  • Gadolinium
  • Humans
  • Myocardial Infarction* / diagnostic imaging
  • Predictive Value of Tests
  • Primary Prevention
  • Retrospective Studies
  • Risk Factors

Substances

  • Contrast Media
  • Gadolinium