Anatomic and functional changes after either a permanent left anterior descending coronary artery occlusion (PO) or 2 h of occlusion followed by reperfusion (OR) in C57BL/6 mice were examined and compared with those in sham-operated mice. Both interventions generated infarcts comprising 30% of the left ventricle (LV) measured at 24 h and equivalent suppression of LV ejection velocity and filling velocity measured by Doppler ultrasound at 1 wk. Serial follow-up revealed that the ventricular ejection velocity and filling velocity returned to the levels of the sham-operated controls in the OR group at 2 wk and remained there; in contrast, PO animals continued to display suppression of both systolic and diastolic function. In contrast, ejection fractions of PO and OR animals were depressed equivalently (50% from sham-operated controls). Anatomic reconstruction of serial cross sections revealed that the percentage of the LV endocardial area overlying the ventricular scar (expansion ratio) was significantly larger in the PO group vs. the OR group (18 ± 1.7% vs. 12 ± 0.9%, P < 0.05). The septum that was never involved in the infarction had a significantly (P < 0.002) increased mass in PO animals (22.5 ± 1.08 mg) vs. OR (17.8 ± 1.10 mg) or sham control (14.8 ± 0.99 mg) animals. Regression analysis demonstrated that the extent of septal hypertrophy correlated with LV expansion ratio. Thus late reperfusion appears to reduce the degree of infarct expansion even under circumstances in which it no longer can alter infarct size. We suggest that reperfusion promoted more effective ventricular repair, less infarct expansion, and significant recovery or preservation of ventricular function.