ABC | Volume 113, Nº1, July 2019

Original Article Sincer et al Mean platelet volume, coronary artery fistula Arq Bras Cardiol. 2019; 113(1):71-76 of 120 patients were included: 70 with normal coronary arteries and 50 with CAF with no associated critical coronary artery stenosis. Coronary angiography was performed due to ischemic changes in ECG, positive exercise test or myocardial perfusion scintigraphy for ischemia. Clinical and laboratory findings were obtained by reviewing the patients’ files. Hypertension was defined as blood pressure > 140/90 mmHg or receiving antihypertensive medication. Diabetes mellitus was defined as having fasting glucose level > 126 mg/dL or receiving anti-diabetic medication. Presence of total cholesterol > 200 mg/dL or triglycerides > 150 mg/dL was accepted as hyperlipidemia. Patients with a history of acute coronary syndrome in the last 6 months, history of coronary artery stenting or bypass operation, idiopathic dilated or hypertrophic cardiomyopathy, congestive heart failure, moderate to severe renal failure, severe hepatic dysfunction, atrial fibrillation, severe valvular disease, systemic inflammatory diseases (e.g. rheumatoid arthritis, lupus erythematous), malignancy, history of blood transfusion in the last 3 months, recent infection (1 month), leukemia or thrombocytopenia, were excluded. The local institutional board approved the study. Peripheral venous blood samples were drawn from patients who were admitted for angiography or during regular follow-up checkups. Serum glucose, creatinine, total cholesterol, high-density lipoprotein cholesterol, and low‑density lipoprotein cholesterol levels were measured using an automatic biochemical analyzer (Architect C8000, USA). Complete blood count and platelet volume were determined using simultaneous optical and impedance measurements (Cell Dyn 3700; Abbott Diagnostics, Lake Forest, Illinois, USA). Platelet, lymphocyte, monocytes, white blood cell (WBC) and MPV values of each patient were recorded. Coronary angiographies were performed through the radial or femoral artery. The coronary angiographies were evaluated by three interventional cardiologists who were blinded to the clinical and laboratory data of the patients. The fistula location, drainage site , and the fistula shape were evaluated. Statistical analyses were carried out using the SPSS 18.0 Statistical Package Program for Windows (SPSS Inc., Chicago, Illinois, USA). The distribution of the variables in the study groups was analyzed by the Kolmogorov-Smirnov test. Normally-distributed variables were compared by t -test and expressed as mean ± standard deviation. Variables without normal distribution were compared with the Mann‑Whitney U-test and expressed as median (interquartile range). Qualitative variables were expressed as numbers and percentages. The differences between independent groups were assessed by Student’s t -test for normally-distributed quantitative variables and Mann-Whitney U-test for variables without normal distribution, whereas Chi-square test was used for qualitative variables. Pearson’s correlation test was used to assess the correlations of MVP with CAF presence. The univariate analysis was used to disclose the association of variables with CAF. Thereafter, to determine the independent prognostic factors of CAF, the multivariate logistic regression model with the forward-stepwise method was used with variables that were found to be significant in the univariate analysis. A receiver operating curve (ROC) analysis was performed to find MPV sensitivity and specificity, aiming to predict the presence of coronary fistulae. All results were considered statistically significant at the level of p < 0.05. Results Baseline patient demographics, including age and clinical risk factors, were similar between the groups, except that the number of females was significantly lower in the fistula group. Previous medications were also comparable between two groups (Table 1). Compared to the control group, HDL‑cholesterol levels were significantly higher in the CAF group. However, serum blood glucose, creatinine and other lipid levels were not significantly different (Table 2). Although the platelet, lymphocyte, monocytes and WBC counts were not significantly different between the two groups, mean MPV levels were significantly lower in the CAF group (8.84 ± 1.71fL vs. 10.43 ± 1.34, p < 0.001) (Table 2). In themultivariate logistic analysis, a forward-stepwisemodel including MPV, platelet, lymphocyte, monocytes, and WBC counts showed that only MPV was a significant predictor of CAF Table 1 – General characteristics of the study groups Baseline characteristics Fistula (n = 50) Control (n = 70) p Age (mean ±SD) (years)) 58 ± 12 55 ± 8 0.123 a Male/female 32/18 24/46 0.001 b Hypertension (%) 14 (28%) 13 (19%) 0.223 b Smoking 24 (48%) 35 (50%) 0.829 b Family history 10 (20%) 12 (17%) 0.115 b Diabetes mellitus 12 (24%) 9 (13%) 0.113 b Acetyl salicylic acid 17 (34%) 21(30%) 0.119 b Statin 15 (30%) 18 (26%) 0.143 b ACE inhibitor 2 (4%) 3 (4%) 0.938 b B-blocker 17 (34%) 20 (29%) 0.26 b ACE: angiotensin-converting enzyme; SD: Standard deviation, data were compared with Student’s t-test ( a ) and Chi-square test ( b ). 72