Discrepancy between MRI and echocardiography in assessing functional left ventricular parameters and scar characteristics in patients with chronic ischemic cardiomyopathy

Studies have demonstrated that infarct size estimated by CMR-LGE was an independent determinant of adverse LV remodeling and dysfunction. We sought to assess relationship between different scar characteristics and left ventricular remodeling and dysfunction using late gadolinium enhancement CMR (LGE-CMR) and echocardiography in patients with ischemic cardiomyopathy. Forty-eight patients with post-infarction left ventricular (LV) dysfunction underwent CMR and 2D echocardiographic studies. Various scar characteristics were assessed by a freely available software and were correlated with functional parameters. All patients had LGE in CMR indicating prior myocardial infarction (MI). A statistically significant but modest negative association was found between left ventricular ejection fraction (LVEF) and number of segments with LGE (r = −0.4, p = 0.005). Additionally, there was a statistically significant modest to moderate positive relationship between LV end diastolic volume (LV EDV) and absolute total scar mass (r = 0.38, p = 0.007), absolute scar core mass (r = 0.32, p = 0.026), peri-infarct zone as absolute (r = 0.45, p = 0.001) and as percent of LV (r = 0.29, p = 0.045) and number of segments with LGE (r = 0.32, p = 0.029). Similarly, statistically significant modest positive correlations were observed between LV end systolic volume (LV ESV) and absolute total scar mass (r = 0.37, p = 0.009), absolute scar core mass (r = 0.32, p = 0.02), peri-infarct zone as absolute (r = 0.4, p = 0.004) and number of segments with LGE (r = 0.38, p = 0.007). There was a mild to moderate correlation between LVEF as assessed by TTE and LVEF measured by CMR (r = 0.49, p < 0.001). The mean difference in LVEF between the two methods was 7.5 ± 9.2% with a p value <0.001. Bland–Altman limits were wide ranging from −10.5 to 25.5%. Different scar characteristics as assessed by CMR were associated with the extent of LV remodeling and dysfunction. This highlights the potential importance of myocardial scarring assessment in risk stratification of patients with ischemic cardiomyopathy. Wide agreement limits for ejection fraction assessment by TTE and CMR suggest that both methods are not interchangeable. Given its 3D approach and superior image quality, CMR may be the preferred technique for volume and ejection fraction estimation.