Incremental value of the exercise treadmill test in determining the extent of coronary artery disease: a discriminant...

This study was undertaken to determine whether it was possible to use a multivariate discriminant function analysis to assess the severity of coronary artery disease in patients with angina pectoris. One hundred and forty-three patients diagnosed as having angina pectoris who formed a 'learning group' were subjected to an exercise test on a treadmill using a Bruce protocol. The items of data from the test least likely to be influenced by beta blockade were considered for the analysis along with two discrete variables (the gender and the use of beta blockers). All the patients subsequently underwent coronary angiography and the severity of the coronary artery disease was graded according to a scoring system. In addition, patients in 'learning group' were classified into two 'Groups', (I) those with normal vessels or with disease in one or two of the major arteries, (II) those with disease in the left main stem or in all three major arteries. The multivariate analysis yielded a score based upon 4 factors: (i) gender, (ii) percentage increase in double product at the highest level of exercise achieved, (iii) duration of ST depression in the recovery period and (iv) the summation of the ST depression in the standard electrocardiogram excluding lead aVR. There was a significant overall correlation between the discriminant function score and the coronary artery score for severity of the disease. In addition, it was found that, of patients having discriminant function score below 0, 96% (69/72) were in Group I. The post-test risk of a positive test was 0.005.(ABSTRACT TRUNCATED AT 250 WORDS)