IJCS | Volume 33, Nº3, May / June 2020

264 which is reflected by increased QTD. 9-11 Koide et al., 12 observed that even when a coronary patient, undergoing stress test, did not present ischemia criteria, the QTD was higher (62 ± 13 ms) compared to patients without coronary disease (40 ± 14 ms). Musha et al., 13 and col. showed an increase of QTD after exercise, which was not reduced by beta-blockers. 13 Naka,et al., 14 in a study with infarcted patients, found an increase in QTD due to residual ischemia;however, it did not increase in patients without residual ischemia. There are several limitations to QT interval measurement technique. Among them, T-P fusion during higher heart-rates and changes in Qt interval rates in relation to men and women 15,16 are worthy of note. The aims of this study are to evaluate whether the QTD index is sensitive to action potential changes in the presence of stress induced myocardial ischemia, as well as to define a cutoff point for QTD that could become a diagnostic criterion for myocardial ischemia. Methods An observational analytical study, where 80 patients underwent exercise testing and coronary angiography (CAT), with amaximum interval of 6months between the tests. The patients were aged 18 to 80 years. People who had any of the following conditions were excluded from the study: previous acute myocardial infarction (AMI), complete right [RBBB] or left [LBBB] bundle branch block, patients with long QT syndrome; patients with known ventricular dysfunction; unreadable ECG traces or ECG where less than eight electrocradiographic leads were available for QTI measurement. 17 A total sample of 80 patients was defined at the convenience of the researcher. Treadmill test was performedwith an analog-to digital converter of signals, Ergo PC 13 model in Micromed 2.3 version, with a simultaneous acquisition of twelve leads and record with speed of 25 mm/s and amplitude of 10 mm/mV. The protocols used for all patients were individualized aiming at reaching maximum heart rate. The test was considered suggestive of myocardial ischemia in case the patient presented at least onde one the ischemia criteria defined by the III Guidelines on ergometric tests of the Brazilian Society of Cardiology; 18 typical chest pain on exertion;ST-segment elevation or depression, equal to or greater than 1 mm, in relation to baseline ECG. The QT interval of each leadwas calculated by the mean of the three beats with less artifact. By using a cursor, one point was marked at the beginning of the QRS complex and another at the end of the T-wave (the point where the T wave returned to the isoelectric line) for each of the three beats. 11 After the measurement, we calculated the mean of the three values found, which would be the value to be considered as the QTI of the mentioned lead. The same procedure was performed for the 12 leads. Thus, for each patient in the study, at least, 24 QT intervals were measured at rest (standing) and 24 on ECG obtained within the first minute of the recovery stage. We decided to “measure the peak stress”within the first minute of the recovery phase in order to minimize the technique’s artifacts. At the end of the measurement of the QT intervals of all leads, we marked the highest and the lowest measure found, in the two phases studied: rest and effort. From these values, we calculated the QTI dispersion of these two phases, and also a delta QT dispersion value by determining the difference of QTD between effort and rest. In order to adjust the QT interval for the corresponding heart rate (HR), we used the Bazett’s formula. The adjustment enabled the calculation of the QTc (QT interval dispersion corrected for heart rate), and also the QTc “delta” – the difference between rest-stress QTc intervals. All the electrocardiographic measurements were done by a single observer. Figure 1 shows the sequence to measure a QT interval. Interobserver variability was determined by measurements performedby a second researcher, whowas blinded to themeasures obtained by the first observer. The secondECGexpertmeasured theQT interval in 12 patients randomly selected (patients were numbered from 1 to 63, and 12 numbers were raffled). The correlation between the measures was determined by Pearson’s correlation coeficiente. The Bland-Altman test was also used to assess interobserver variability (Figure 2). After catheterization, patients with a coronary stenosis of at least 70% of one or more arteries, or with ≥ 50% stenosis of the left coronary trunk (LCT), were classified as “people with obstructive coronaropathy”. In contrast, patients with with stenosis less than 70% in epicardial coronary arteries, or less than 50% in the left coronary trunk, were classified as “without obstructive coronary disease”. The Medcalc software was used for the statistical analysis. The data were expressed in absolute numbers, percentages and standard deviation. The classificatory variables were presented as tables, and the propotions were compared using the chi-square Barcelos et al. QT interval dispersion in exercise test Int J Cardiovasc Sci. 2020; 33(3):263-271 Original Article