Myocardial infarction with normal coronary arteries: ten-year follow-up.

BACKGROUND: Patients with myocardial infarction without angiographically significant coronary artery lesions are considered, as a whole, to have a better prognosis. Different degrees of coronary involvement, within this wide group, may portend different degrees of risk. The aim of this study was to assess which clinical and angiographic covariates are more useful in defining the individual prognosis. METHLODS: We prospectively followed 53 consecutive patients admitted to our coronary care unit between 1985-1990 with myocardial infarction and a culprit lesion causing < or = 50% angiographic luminal narrowing.
RESULTS: Patients with normal angiograms (group A) were compared to those with minor parietal irregularities (group B) or discrete (< or = 50%) stenosis (group C). Group A patients were younger, had a lower peak creatine kinase release, and a higher ejection fraction. After a median follow up of 125 +/- 32 months, group A patients had a 100% 10-year survival compared to 77 and 58% for group B and group C patients respectively (p = 0.01). At univariate analysis, ischemic events correlated with the severity of coronary lesions (while group A patients had no ischemic events at follow-up, events occurred in 46 and 50% of group B and C patients respectively; p = 0.003) and with ejection fraction, that was lower in patients with events (56.0 +/- 17 vs 67 +/- 11%, p = 0.006). Cardiac death, too, correlated with the type of coronary lesions (p = 0.03) and with lower ejection fraction (p = 0.004). By means of multivariate analysis, on the contrary, only the vessel morphology was predictive of ischemic events at follow-up (p = 0.02), while the only significant predictor of death was ejection fraction (p = 0.0012).
CONCLUSIONS: Patients with myocardial infarction and strictly normal vessels have very few ischemic events at follow-up, and may be distinguished from both patients with non-significant lesions as well as those with minor angiographic irregularities. On the other hand, cardiac mortality correlates strongly and independently with a depressed ventricular function.