Influence of coronary collateral vessels on myocardial infarct size in humans. Results of phase I thrombolysis in myocardial infarction (TIMI) trial. The TIMI Investigators

The influence of coronary collateral vessels on infarct size in humans remains controversial, partly because no previous study has examined the impact of collaterals present at the onset of acute myocardial infarction on infarct size. The present study used the data base of the Thrombolysis in Myocardial Infarction (TIMI) Phase I trial to correlate the presence or absence of angiographically documented collaterals in the initial hours of myocardial infarct evolution with the size of the infarct as assessed by serial measurements of serum creatine kinase (CK). To avoid the confounding effects of reperfusion on enzymatic estimates of infarct size, this report is limited to those 125 patients who failed to recanalize at 90 minutes after administration of tissue plasminogen activator or streptokinase. Patients with angiographically documented collaterals (group A, n = 51) had significantly lower values of peak serum CK than patients without collaterals (group B, n = 74) (1,877 +/- 216 versus 2,661 +/- 212 IU/l, respectively [mean +/- SEM], p = 0.004). Similarly, CK-derived infarct size estimates were significantly lower in group A than in group B (20.6 +/- 2.5 versus 31.4 +/- 2.8 CK gram equivalents, p = 0.001). The infarct size observed in patients with collaterals was less for anterior infarctions as well as for infarctions of other locations; thus, the beneficial effects of collaterals were independent of the site of the infarct. In 65 of the 125 patients who failed to reperfuse, left ventricular ejection fraction (LVEF) was assessed by contrast ventriculography both at initial cardiac catheterization (before thrombolytic therapy) and at hospital discharge. Among the patients who had both studies, global LVEF tended to increase from pretreatment to hospital discharge in group A (from 50.6 +/- 1.8% to 53.4 +/- 1.8%, p = 0.10) but decreased in group B patients (from 50.3 +/- 1.8% to 47.8 +/- 1.7%, p = 0.02). At hospital discharge, global LVEF was greater in patients with coronary collaterals (53.5 +/- 1.7% versus 49.6 +/- 1.7%, p = 0.01). The results demonstrate that, in patients in whom thrombolytic therapy fails to induce reperfusion, the presence of coronary collateral vessels at the onset of myocardial infarction is associated with limitation of infarct size as assessed enzymatically and with improved ventricular function on discharge as assessed by LVEF.