Hemorrhagic myocardial infarction after coronary reperfusion detected in vivo by magnetic resonance imaging in humans: Prevalence and clinical implications

With the advent of thrombolytic therapy, hemorrhagic myocardial infarction (HMI) has been observed in experimental and human autopsy studies. However; its clinical implications remain undetermined, because of the absence of a reliable method to detect its presence in vivo. This study was designed to evaluate the clinical implications of HMI detected by magnetic resonance ( MR) imaging in vivo after coronary reperfusion. Thirty-nine patients with a cute myocardial infarction (AMI) were studied. Percutaneous transluminal co ronary angioplasty (PTCA) was used to reopen the occluded coronary artery. Electrocardiogram (ECG)-gated T2*-weighted gradient-echo MR imaging was per formed to detect intramyocardial hemorrhage, using a 1.5-T magnet within 2 weeks after coronary reperfusion (average, 5.7 days). Thirteen patients (33 %) showed intramyocardial hemorrhage as a distinct hypointense zone by grad ient-echo MR imaging and 26 patients showed homogeneous intensity consisten t with absence of intramyocardial hemorrhage. Coronary, angiograms showed l esser development of collateral flow in the patients with HMI than in those without (81% vs. 37%). Infarct size, estimated 1 month after coronary repe rfusion by thallium-201 scintigraphy, was lai-ger among patients with HMI t han in those without (37 +/- 14% vs. 21 +/- 14%, respectively p < 0.05). Le ft ventricular ejection fraction at 1 month follow-up showed less recovery in patients with HMI than in those without (47 +/- 9 to 51 +/- 10%; p = 0.4 7, vs. 53 +/- 10 to 60 +/- 9%, respectively, p < 0.05). ECG-gated T2*-weigh ted gradient-echo MR imaging offers a noninvasive means of detection of int ramyocardial hemorrhage in patients with reperfused AMI. HMI occurred even after primary PTCA and may be a common finding associated with severely inj ured myocardium.