IJCS | Volume 31, Nº2, March / April 2018

135 Barros et al. Acute pulmonary edema. Coronary artery disease. Int J Cardiovasc Sci. 2018;31(2)133-142 Original Article To characterize the study population and perform the bivariate analysis, the continuous variables were shown as means and standard deviation, when they had a normal distribution, or by medians and interquartile ranges. The normality hypothesis was verified by the Kolmogorov-Smirnov test. When searching for an association between the obstructive coronary artery disease predictive variables, Pearson's chi square test was used for the comparison of proportions, whereas the Mann-Whitney test was used for the comparison of medians and the Student's t test for the comparison of the means. The variables that showed statistical significance of up to 10% (p < 0.10) were eligible for the multivariate model. The logistic regression model was applied. The Hosmer-Lemeshow test was used for the model assessment, andmodel performance was assessed through the area under the curve (AUC) of the Receiver Operating Characteristic (ROC) curve. The statistical software Stata for Windows, version 12, was used for the statistical analyses. Results Of the 149 patients, 89 (59%) had severe obstructive CAD. Most patients were elderly and female (Table 1). A moderate lesion alone was observed in 9 patients (6%) of the 149 coronary angiograms evaluated. In most cases, the affected coronary artery was restricted to an epicardial vessel, with the right coronary artery being the one most frequently involved. An affected left main coronary artery and the involvement of three vessels were observed in 11 (12%) and 15 (17%) of the 89 patients, respectively, as described in Table 2. Table 3 shows the detailed involvement of the epicardial vessels. It was observed that severe coronary artery disease more often affected the population with diabetes, peripheral obstructive arterial disease, those with a history of coronary artery disease and stroke. As for the ejection fraction at the echocardiogram, the mean value was above 45% and in most cases, it was evaluated using the Teichholz method. The median troponin I on the first day was 0.086 ng/mL (normal value 0.014 ng/mL), and the interquartile range was 0.036 and 0.201 ng/mL in patients with APE. Troponin was considered positive when two independent cardiologists classified it as suggestive of an APE-related acute coronary event, a fact observed in 77 patients (51.3%) (Table 1). Mean systolic and diastolic blood pressures were high. The most often found alteration in the electrocardiogram wasSTdepression, verifiedinmost assessedpatients (53.5%). An intracoronary thrombus was visualized in 4 (3%) of the 122 patients in which CA was performed during hospitalization. One had elevation of CK-MB mass and troponin, with very high values – respectively, 86mg/mL (reference value of 4.54 mg/mL) and 1.29 ng/mL (reference value of 0.014 ng/mL) –, suggesting values compatible with coronary artery occlusion. The electrocardiogram showed ST depression of 1 mV in the inferior wall, V4 to V6 and ST elevation of 1 mV in aVR and V1. The CA showed a thrombus in the anterior descending artery and severe lesions in the diagonal and right coronary arteries. The univariate analysis of the possible predictors of obstructive CAD is described in Table 4. Ischemic chest pain, diabetes mellitus, systemic arterial hypertension, history of CAD, history of obstructive peripheral arterial disease, stroke, ejection fraction, myocardium segmental deficit at the echocardiography, presence of Q wave and evaluation of the absolutemeasurement of troponin on the first day were associated with obstructive CAD (p < 0.1). The variables that were included in the multivariate model and the independent variables of obstructive CAD were history of CAD (p < 0.000) and myocardium segmental deficit (p < 0.02), as described in Table 5. The model performance was evaluated by the area under the curve (AUC) of receiver operating characteristic (ROC) curve, as shown in Figure 1. The AUCwas shown to have a good discriminatory power with a C-statistics of 0.905 (95% confidence interval - 95%CI: 0.862-0.954). Discussion APE is a severe expression of acute heart failure (AHF) in the emergency room. It has been commonly interpreted in the presence of a positive necrosis marker as an expression of unstable coronary disease, as defined by Figueras et al. 2 and Pena-Gil et al. 6 However, it is known that AHF, alone, alters the marker – either CKMB or troponin – and therefore it is not a reliable marker for the diagnosis of acute coronary syndrome. 7 Thus, this diagnosis is difficult in the absence of clear and evident electrocardiographic alterations of an arterial occlusion, such as ST-elevationwith its respective classical evolution at the electrocardiogram.