Recovery of contractile function of myocardium stunned by a brief, transient period of regional ischemia is highly variable. In our experience, segment shortening (an index of regional systolic contractile function) assessed during the initial hours after a 15-minute period of coronary artery occlusion in anesthetized open-chest dogs ranged from -84 to +99% of normal preocclusion values. In this retrospective study, regression analysis was used to assess the effects of various parameters on segment shortening 2 hours after reperfusion. Parameters assessed included regional myocardial blood flow both during occlusion and after reperfusion, high-energy phosphate content of previously ischemic tissue, systemic hemodynamic parameters (heart rate, mean arterial pressure and double product), occluded bed size and segment shortening measured during coronary artery occlusion. Recovery of systolic contractile function was not influenced by the degree of ischemia during coronary artery occlusion, myocardial blood flow after reperfusion, high-energy phosphate content, hemodynamic parameters or occluded bed size (correlation coefficients, r, ranged from 0.001 to 0.37 [p = not significant]). Only the degree of dyskinesia/hypokinesia exhibited during coronary occlusion significantly and reliably predicted recovery of segment shortening measured 2 hours after reflow (r = 0.70, p < 0.001). Thus, recovery of systolic contractile function in the anesthetized canine model of the stunned myocardium is determined primarily by the degree of dysfunction exhibited during the preceding period of ischemia.