Myocardial Viability and Long-Term Outcomes in Ischemic Cardiomyopathy

Abstract

Background: The role of assessment of myocardial viability in identifying patients with ischemic cardiomyopathy who might benefit from surgical revascularization remains controversial. Furthermore, although improvement in left ventricular function is one of the goals of revascularization, its relationship to subsequent outcomes is unclear.

Methods: Among 601 patients who had coronary artery disease that was amenable to coronary-artery bypass grafting (CABG) and who had a left ventricular ejection fraction of 35% or lower, we prospectively assessed myocardial viability using single-photon-emission computed tomography, dobutamine echocardiography, or both. Patients were randomly assigned to undergo CABG and receive medical therapy or to receive medical therapy alone. Left ventricular ejection fraction was measured at baseline and after 4 months of follow-up in 318 patients. The primary end point was death from any cause. The median duration of follow-up was 10.4 years.

Results: CABG plus medical therapy was associated with a lower incidence of death from any cause than medical therapy alone (182 deaths among 298 patients in the CABG group vs. 209 deaths among 303 patients in the medical-therapy group; adjusted hazard ratio, 0.73; 95% confidence interval, 0.60 to 0.90). However, no significant interaction was observed between the presence or absence of myocardial viability and the beneficial effect of CABG plus medical therapy over medical therapy alone (P = 0.34 for interaction). An increase in left ventricular ejection fraction was observed only among patients with myocardial viability, irrespective of treatment assignment. There was no association between changes in left ventricular ejection fraction and subsequent death.

Conclusions: The findings of this study do not support the concept that myocardial viability is associated with a long-term benefit of CABG in patients with ischemic cardiomyopathy. The presence of viable myocardium was associated with improvement in left ventricular systolic function, irrespective of treatment, but such improvement was not related to long-term survival. (Funded by the National Institutes of Health; STICH ClinicalTrials.gov number, NCT00023595.).

Copyright © 2019 Massachusetts Medical Society.

Figures

Figure 1 (facing page).. Kaplan–Meier Analysis of the Incidence of Death from Any Cause.

Panel A shows Kaplan-Meier curves for the incidence of death from any cause among patients who underwent a myocardial viability test, according to treatment group; results were compared with the use of a Cox proportional-hazards model with adjustment for baseline covariates. Panel B shows Kaplan–Meier curves for the incidence of death from any cause among patients without viable myocardium (left panel) and among those with viable myocardium (right panel), according to treatment group. Panel C shows the results of a Cox proportional-hazards model that tested for the interaction between myocardial viability and treatment, with adjustment for baseline covariates. CABG denotes coronary-artery bypass grafting.

Figure 2.. Incidence of Death from Any Cause, According to Changes in Left Ventricular Ejection Fraction.

Shown are the results of a landmark analysis that included data from the 318 patients who underwent myocardial viability testing and had paired imaging (i.e., assessed with the same imaging method at each time point) at baseline and at 4 months for measurement of left ventricular ejection fraction (LVEF). Kaplan-Meier estimates of death from any cause among patients with improvement in LVEF and among patients without such improvement were compared with the use of a Cox proportional-hazards model with adjustment for baseline covariates.

Figure 3.. Change in Left Ventricular Ejection Fraction, According to Myocardial Status and Treatment Group.

Shown is the least-squares mean change in LVEF from baseline to month 4 in the four subgroups of patients defined according to the presence or absence of viable myocardium and treatment assignment. I bars denote 95% confidence intervals.