EXPONENTIAL FIT OF QT INTERVAL HEART-RATE RELATION DURING EXERCISE USED TO DIAGNOSE STRESS-INDUCED MYOCARDIAL-ISCHEMIA

The aim of this study was to analyze the dynamic changes of QT interval-heart rate relation during exercise, fitting their reciprocal variations to an exponential formula (QT = A - B . exp(-k . RR)), in order to see whether diagnostic contributions might so be derived. The authors studied 139 patients who underwent a simultaneous assessment of regional myocardial perfusion and ventricular function by means of two injections of 99mTc-methoxy-isobutyl-isonitrile at rest and at peak of a submaximal exercise test, using first pass radionuclide angiography with multielement gamma-camera and single photon emission computerized tomography, in order to detect and localize the presence of stress-induced myocardial ischemia. According to radionuclide results, patients were divided into three groups: group A, 7 individuals with no sign of stress-induced myocardial ischemia; group B, 79 patients with evidence of ischemia in 1 (16.5%), 2 (65.5%), or 3 (17.7%) main coronary territories; and group C, 53 patients with previous infarction and evidence of ischemia in other territories. Conventional analysis of the exercise test (greater-than-or-equal-to 0.1 mV ST depression) showed a pathological response in no individual of group A, in 34 patients of group B (43%), and in 27 patients of group C (50.9%); overall sensitivity was 46.2%, specificity 100%, and diagnostic accuracy 48.9%. Exponential coefficients A, B, and k showed wide overlap of values among the three groups, A and B (p < 0.001), and group C (p < 0.05). The algebraic sign of k was negative in 6 of 7 (85.7%) subjects of group A, in 11 of 79 patients of group B (13.9%), and in 10 of 53 patients of group C (18.9%). Using this criterion to define the response to exercise as abnormal overall sensitivity was 84.1%, specificity 85.7%, and diagnostic accuracy 84.2%. The sign of B and k coefficients represents the shift of the slope of the exponential curve relating QT interval (ie., the duration of the electrical systole) to heart rate (ie, the major determinant of myocardial oxygen consumption) during exercise. However, B values varied much interindividually, whereas k value variation was limited in the study group. Therefore, the change of k status might be indirectly related to the shift from aerobic to anaerobic metabolism, as supported by the higher sensitivity of the k coefficient to detect myocardial ischemia, in comparison to ST-segment depression. These results need now to be confirmed in larger series of patients, using multiparametric statistical analyses capable of accounting for the independent contribution of other variables that could theoretically affect QT interval changes during exercise.