Background and Purpose: In acute ST segment elevation myocardial infarction (STEMI), rapid restoration of epicardial coronary blood flow and myocardial perfusion limits infarct size and improves survival. Primary percutaneous coronary intervention (PCI) is superior to systemic fibrinolysis when instantly performed by experienced operators. The "Herzinfarktverbund Essen" (HIVE) is an urban STEMI network supporting direct patient transfer for primary PCI to four PCI centers covering a city area of 600,000 inhabitants. Integrated health care is an optional part of the HIVE allowing for reimbursement of medical innovations such as the evaluation of infarct size and the presence and extent of microvascular obstruction by contrast-enhanced cardiac magnetic resonance (CMR). The aim of this study was to assess the prognostic impact of contrast-enhanced CMR in the patient cohort of a regional STEMI network. Patients and Methods: Within the 1st year (09/2004 to 08/2005) of the HIVE registry, 489 patients with acute myocardial infarction were treated in the four primary PCI centers. In one of the centers, including 143 patients, early CMR imaging using a standardized MR protocol for infarct quantification was performed whenever possible. Patients with hemodynamic instability, emergency coronary artery bypass grafting, resuscitation or death prior to CMR, claustrophobia, and other general contraindications to MRI had to be excluded, leaving 67 patients (54 male; mean age 61 ± 12 years) for final evaluation. CMR was performed 4.5 ± 2.5 days after admission on a 1.5-T MR scanner (Sonata, Siemens Medical Solutions, Erlangen, Germany) including steady-state free precession (SSFP) cine imaging for left ventricular function and single-shot inversion-recovery SSFP imaging for delayed enhancement (DE) and no-reflow (NR) evaluation following injection of 0.2 mmol/kg body weight gadodiamide (Omniscan®, GE Healthcare Buchler, Munich, Germany). NR and DE volumes were calculated from single-shot short-axis stacks taken within the 1st minute following gadodiamide infusion by manual planimetry and summation of disks. 1-year follow-up data (telephone interview) for major adverse cardiac events (MACE: cardiac death, myocardial infarction, and rehospitalization for congestive heart failure, angina pectoris, or revascularization) were available for all patients. Results: DE as a measure of infarct size was 9% ± 7% (range 0-33%) of left ventricular mass (LVM), and mean volume of microvascular obstruction was 2% ± 3% (range 0-17%). Microvascular obstruction was present in 61% of patients. 16 MACE (one cardiac death, one myocardial infarction, and 14 rehospitalizations for congestive heart failure or unstable angina pectoris with PCI in six cases) occurred within the follow-up period of 430 ± 63 days. Patients with MACE had larger infarcts (14% ± 10% vs. 8% ± 6% DE), lower left ventricular ejection fraction (LVEF 44% ± 17% vs. 48% ± 14%) and larger NR (3% ± 5% vs. 2% ± 3%). Using a stepwise logistic regression model, only NR > 0.5% of LVM was independently related to outcome (odds ratio = 3.9, confidence interval 1.1-13.9). Conclusion: NR as a correlate of microvascular obstruction remains independently related to prognosis in patients with acute myocardial infarction treated by PCI.