ABC | Volume 112, Nº1, January 2019

Original Article Dotta et al Regional QT dispersion as predictor of reperfusion Arq Bras Cardiol. 2019; 112(1):20-29 Table 1 – Baseline clinical and epidemiological characteristics of the patients (n = 104) Age (years) Men n (%) Type 2 DM n (%) SAH n (%) Dyslipidemia n (%) Smokers n (%) Symptom onset (min), m ± SD 55.6 ± 8.78 66 (62.9) 21 (20) 60 (57.1) 36 (34.3) 51 (48.6) 192.16 ± 94.35 Data expressed as mean and standard deviation (m±SD), or number and percentage, n (%), DM: diabetes mellitus; SAH: systemic arterial hypertension appropriate. Continuous variables were compared by Student’s t-test for independent samples or the Mann-Whitney test, as appropriate. Within-group comparisons were made by t-test for related samples or the Wilcoxon test. All tests were two-tailed, and a p-value < 0.05 was considered statistically significant. Area under the ROC (receiver operating characteristic) curves, based on C-statistics, were constructed to determine optimal cut-off values for some of the variables. All tests were performed using the Statistical Package for Social Sciences (SPSS) ® software version 17.0, da SPSS Inc, Chicago, IL, USA. Results Baseline characteristics of the population A total of 104 patients attending public primary care centers, with clinical and electrocardiographic diagnosis of STEMI, treated with a fibrinolytic agent (TNK) and submitted to coronary angiography within 2-24 hours thereafter were included in the study. Main demographic and clinical characteristics of these patients are described in Table 1. Age ranged from 35 to 74 years old, and most patients were men. Time (median and IQR) between symptom onset and initiation of thrombolytic therapy was 180 minutes (120-240 minutes). Localization of infarction AMI was classified according to the ventricular wall involved. For statistical analysis purpose, AMI was grouped into anterior (n = 42) and non-anterior wall infarction (n = 62). Distribution of QTc and QTcD by electrocardiographic criterion for reperfusion Sixty-seven (64%) patients met the electrocardiographic criterion for reperfusion. Electrocardiographic tracings were analyzed by two independent observers, with a Kappa coefficient of 0.84. Patients were categorized into two groups – patients with signs of reperfusion and patients without signs of reperfusion, considering only a ST-segment resolution of 50% or more. Values of QTc and QTcD before and after fibrinolysis are shown in Table 2. QTc and QTcD intervals in all 12 leads were not different between the groups. Regional QTcD increased in patients with criterion for reperfusion and, considering the involvement of ventricular wall, in patients with anterior wall myocardial infarction with criterion for reperfusion (p = 0.023) (Table 3). Distribution of QTcD by angiographic data Patients were categorized into two groups according to TIMI and Blush grades. Patients with optimal reperfusion, i.e ., TIMI 3 and Blush grade 3 – group T3B3 (+) – and those with TIMI < 3 and Blush < 3 – group T3B3 (-). Regional QTcD for anterior wall infarction significantly increased in the T3B3(+) group (p = 0.06), but not in non-anterior wall infarction (p = 0.77). To rule out the possibility of measurement bias in non-anterior wall infarction, regional QTcD in unipolar leads (V 1 -V 6 ) was compared with that in bipolar leads, with no statistically significant difference. Distribution of coronary flow by TIMI and Blush grades Distribution of the flow in the culprit artery according to TIMI grade flow 0, 1, 2 and 3 was 20.2%, 7.7%, 13.5% and 58.7%, respectively. Figure 1 depicts (a) percentage distribution of patients according to TIMI grade flow and the electrocardiographic criterion for reperfusion (ST-segment resolution); (b) distribution of patients (in relative frequency) according to TIMI and Blush scores (T3B3) and ST-segment resolution. Few patients with TIMI3 did not show adequate myocardial perfusion according to myocardial Blush grade. Distribution of myocardial blush grades 0, 1, 2 and 3 was 4.9%, 3.3%, 4.9% and 86.9%, respectively in patients with TIMI 3. With respect to the prediction of optimal coronary reperfusion [T3B3(+)], the criterion for reperfusion by ECG and analysis of QTcD showed a positive predictive value of 73%, negative predictive value of 89%, sensitivity of 93% and specificity of 73%. Baseline demographic characteristics according to TIMI/Blush were not different between T3B3(+) and T3B3(-), except for left ventricular ejection fraction, which was lower in T3B3(-) [(52.6 ± 9.8 vs 47.8 ± 8.5; p =0.009)] (Table 4). ECG parameters were not different before and 60 minutes after thrombolysis (Table 5). ROC curves were constructed to evaluate the classification performance of the regional QTcD and to establish the best cutoff point, as illustrated in Figure 2. If we consider only patients in which the classical electrocardiographic criterion for reperfusion failed to identify coronary reperfusion, there were 18 patients with ST-segment resolution in which optimal angiographic reperfusion was not achieved (failed reperfusion by ECG), and four patients without ST-segment resolution showed TIMI grade 3 and Blush grade 3 (failed rescue). In the groups with failed reperfusion by ECG, no difference was found in regional QTcD between pre-thrombolysis and post‑thrombolysis electrocardiographic analysis (p = 0.46) (Figure 3). Therefore, of the 104 patients who received TNK, we detected incorrect reperfusion in 22 cases (21%). Discussion QT interval between different ECG leads and this range of intervals is considered an index of spatial dispersion of ventricular recovery, serving as a signal of repolarization 22