Incremental value of cardiovascular magnetic resonance over echocardiography in the detection of acute and chronic myocardial infarction
1 Department of Cardiology, Maastricht University Medical Center, Maastricht, The Netherlands
2 Radiology, Maastricht University Medical Center, Maastricht, The Netherlands
3 Cardiovascular Research Institute Maastricht, Maastricht University Medical Center, Maastricht, The Netherlands
4 Epidemiology, Maastricht University Medical Center, Maastricht, The Netherlands
5 Department of Cardiology, Maastricht University Medical Center, P. Debyelaan 25, P.O. Box 5800, Maastricht, 6202 AZ, The Netherlands
Journal of Cardiovascular Magnetic Resonance 2013, 15:5 doi:10.1186/1532-429X-15-5Published: 16 January 2013
Although echocardiography is used as a first line imaging modality, its accuracy to detect acute and chronic myocardial infarction (MI) in relation to infarct characteristics as assessed with late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR) is not well described.
One-hundred-forty-one echocardiograms performed in 88 first acute ST-elevation MI (STEMI) patients, 2 (IQR1-4) days (n = 61) and 102 (IQR92-112) days post-MI (n = 80), were pooled with echocardiograms of 36 healthy controls. 61 acute and 80 chronic echocardiograms were available for analysis (53 patients had both acute and chronic echocardiograms). Two experienced echocardiographers, blinded to clinical and CMR data, randomly evaluated all 177 echocardiograms for segmental wall motion abnormalities (SWMA). This was compared with LGE-CMR determined infarct characteristics, performed 104 ± 11 days post-MI. Enhancement on LGE-CMR matched the infarct-related artery territory in all patients (LAD 31%, LCx 12% and RCA 57%).
The sensitivity of echocardiography to detect acute MI was 78.7% and 61.3% for chronic MI; specificity was 80.6%. Undetected MI were smaller, less transmural, and less extensive (6% [IQR3-12] vs. 15% [IQR9-24], 50 ± 14% vs. 61 ± 15%, 7 ± 3 vs. 9 ± 3 segments, p < 0.001 for all) and associated with higher left ventricular ejection fraction (LVEF) and non-anterior location as compared to detected MI (58 ± 5% vs. 46 ± 7%, p < 0.001 and 82% vs. 63%, p = 0.03). After multivariate analysis, LVEF and infarct size were the strongest independent predictors of detecting chronic MI (OR 0.78 [95%CI 0.68-0.88], p < 0.001 and OR 1.22 [95%CI0.99-1.51], p = 0.06, respectively). Increasing infarct transmurality was associated with increasing SWMA (p < 0.001).
In patients presenting with STEMI, and thus a high likelihood of SWMA, the sensitivity of echocardiography to detect SWMA was higher in the acute than the chronic phase. Undetected MI were smaller, less extensive and less transmural, and associated with non-anterior localization and higher LVEF. Further work is needed to assess the diagnostic accuracy in patients with non-STEMI.