ABC | Volume 113, Nº6, December 2019

Original Article Celebi et al. NT-pro BNP and left ventricular aneurysm Arq Bras Cardiol. 2019; 113(6):1129-1137 appropriate. Continuous variables were compared by Student’s t-test for independent samples or the Mann-Whitney test, as appropriate. All tests were two-tailed, and a p-value < 0.05 was considered statistically significant. We analyzed the effects of different variables on the occurrence of ventricular aneurysm in univariate analysis and determined the variables whose unadjusted p-value was < 0.10 as potential risk markers and these were included in the full model. We composed the model by using forward elimination at multivariate regression analysis, and we eliminated potential risk markers by using likelihood ratio tests. Area under the ROC (receiver operating characteristic) curves, based on C-statistics, were performed to determine the optimal cut-off value for NT-pro BNP to predict the LVA. Results A total of 1,519 patients were enrolled (mean age 56.7 ± 11.7 years). Figure 1 describes the enrollment of patients for this study. The baseline characteristics of the patients are summarized in Table 1. The time from onset of chest pain to arrival at the hospital was 6 ± 12 hours. Of the 1,519 patients, primary PCI was performed in 67%, and 26% received fibrinolytic therapy. Seven percent of patients did not receive any reperfusion therapy. Among the patients who developed LVA, the myocardial infarction localization was anterior wall in 90.4%, isolated inferior wall in 1.3%, inferior-posterior in 2%, and inferior-right ventricle in 6.3%. In addition, anterior-located STEMI patients developed LVA more frequently than did patients with infarction on the other side (7.8% vs 1.2% p < 0,01). Patients who developed LVA had lower reperfusion rate than did patients without LVA (42.1% vs. 15.2%, p = 0.021) (Table 2). The LVA rate was lower in patients who received PPCI than in patients who received fibrinolytic therapy or no reperfusion therapy (5.3 vs. 9.2 p < 0.01, 5.3 vs. 14.7; p = 0.03). The type of fibrinolytic agent used had no effect on LVA development. Patients with LVA had a lower rate of P2Y12 inhibitor use (Table 3). When the culprit artery was the left anterior descending artery, the LVA risk was more than in the other coronary arteries (Figure 2). The Gensini scores were similar between groups (38.7 ± 30.8 vs. 37.9 ± 29.9, p = 0.924). However, Rentrop scores were significantly higher in patients without LVA (1.96 ±1.32 vs. 1.51 ± 0.76, p = 0.001). Time to reperfusion therapy was shorter in patients without LVA (4.1 ± 6.3 vs 6.2 ± 5.9, p < 0.05). The basal NT-pro BNP level was significantly higher in patients with LVA (523.5 ± 231.1 pg/mL vs. 192.3 ± 176.6 pg/mL, p < 0.001) (Figure 3). Multivariate logistic regression analysis determined the predictors of LVA after MI (Table 4). Previous CABG, post-MI heart failure, younger age, smoking, no-reflow phenomenon and high NT-pro BNP at admission predicted LVA formation after acute STEMI. Roc analysis showed that the cut-off value of NT-pro BNP at admission for LVA development was 400 pg/mL. The sensitivity and specificity were 78.3% and 94.7%, respectively. (Area under curve: 0.860 0.751-0.968 95% CI) (Figure 4). Roc analysis showed that the cut-off value for peak cTnI for LVA development was 78 pg/mL (Area under the curve: 0.720 0.541-1.320 95% CI) and for peak CK-MB was 312.86 IU/mL (Area under the curve: 0.640 0.314-0.986 95% CI). Discussion Our study showed two main issues: 1) the factors affecting the development of LVA in the new treatment era of acute STEMI, and 2) that NT-pro BNP measured during the acute phase of STEMI is useful for predicting LVA development. Recent studies have determined that LVA incidence after acute STEMI was reduced from 10-30% to 8-15% after the developments in the treatment of acute STEMI. 12,13 In accordance with these data, we found that the incidence of LVA after acute STEMI was 10.3%. Although there are some controversial issues, previous studies have determined that single-vessel disease, total LAD occlusion, poor collateral supply to the infarct-related artery, hypertension and female gender were major determinants for the development of LVA after acute MI. 14-16 However, most of these data were reported before the modern treatment era. In our study, we determined that patients who received fibrinolytic therapy or no reperfusion therapy developed LVA more frequently than did patients who received PPCI. The type of fibrinolytic agent used had no effect on LVA formation. Patients who received reperfusion therapy earlier had less frequent LVA formation. These data showed that earlier reperfusion prevents the development of LVA. Moreover, in our study, P2Y12 inhibitors were found to be a determinant of LVA formations. Hirai et al. 17 showed that good collateral coronary circulation has a beneficial effect on the prevention of LVA formation. 17 Similarly, we found that the Rentrop score was significantly higher in patients without LVA. We also determined that the severity of coronary disease had no effect on the development of LVA. Additionally, gender and risk factors such as diabetes or hypertension had no effect on the development of LVA. The biomarkers of cardiac function may provide useful information in evaluating cardiac outcomes after acute STEMI. Mayr et al. 18 found that NT-pro BNP on day 3 after admission correlated with acute and chronic infarct size and left ventricular ejection fraction after acute myocardial infarction. 18 Kleczy ń ski et al. 19 showed that the assessment of NT-pro BNP level 6 months after STEMI is a useful marker of infarct size and left ventricle function at long-term follow up. 19 Fazlınezhad et al. 20 showed that BNP level is a predictor of acute MI complications such as left ventricular pseudoaneurysm. 20 We showed that NT-pro BNP assessment within the first 12 hours of chest pain is a good indicator to predict LVA. This relationship may be associated with ischemic remodeling of the left ventricle, which triggers NT-pro BNP secretion. The answer as to why we observed high NT-pro BNP levels before the development of LVA may be associated with the properties of NT-pro BNP. The level of NT-pro BNP may reflect the severity of the ischemic insult, even when myocardial necrosis has not occurred. It has been shown that in experimental acute myocardial infarction, NT-pro BNP synthesis is augmented not only in infarcted tissue but also in non-infarcted tissue. 21 In addition, NT-pro BNP levels have been shown to increase transiently after uncomplicated percutaneous transluminal coronary angioplasty, even when intracardiac filling pressures remain unchanged. 22 Therefore, we may suggest that transient ischemia increases wall stress and induces BNP synthesis and the release in proportion to the degree of ischemic insult. Afterwards, this 1131