Clinical characteristics and long-term survival of patients with variant angina

Abstract

We studied 109 consecutive patients with variant angina who underwent cardiac catheterization over an 11 year period. All patients were followed for at least 6 months or until death, and 46 patients (22 treated medically and 24 treated surgically) were followed for 5 years or more. Of the 62 patients initially treated medically, 14 had nonfatal myocardial infarctions (12 within 1 month of catheterization) and 12 died (six within 6 months). Survival probabilities at 1, 3, and 5 years were 0.88, 0.84, and 0.77, respectively. Of the 48 surgically treated patients (including four patients initially treated medically and one initially treated with coronary angioplasty), four had nonfatal infarctions (three in the perioperative period) and three died (all in the perioperative period). The survival probability in these patients at 1 year was 0.94 and remained unchanged at 3 and 5 years. Only one nonfatal infarction and no deaths have occurred in the group of surgically treated patients subsequent to hospital discharge. Three additional patients were treated with coronary angioplasty. The single most important prognostic factor in medically treated patients was the presence or absence of fixed obstructive coronary artery disease. Infarction-free survival probabilities at 1 and 3 years in the 23 patients without significant coronary artery disease were 1.0 and 0.89, compared with 0.51 and 0.46 in the 39 patients with significantly coronary disease. Analysis by the Cox model showed that variant angina patients had a higher probability of death and nonfatal infarction than did those with nonvariant angina coronary disease if other important prognostic variables were held constant. The major independent prognostic variables for medically treated patients with variant angina were similar to those that were important in medically treated patients with coronary disease who did not have variant angina. Finally, we examined how well variant angina patients with and without significant (≥ 75%) coronary artery disease could be distinguished from each other with the use of noninvasive baseline characteristics. Although these two groups did have different distributions of some variables, neither univariable nor multivariable techniques were accurate enough to supplant cardiac catheterization for identifying variant angina patients without significant disease.