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 Table 2 – Laboratory data of the study cohort Fistula (n = 50) Control (n = 70) p Creatinine(mg/dL) 0.80 (0.30) 0.80 (0.22) 0.50 a Fasting plasma glucose (mg/dL) 98 (57) 101(17) 0.90 a LDL-cholesterol (mg/dL) 111 (48) 101(56) 0.47 a HDL-cholesterol (mg/dL) 42 (16) 44(24) 0.04 a Triglycerides (mg/dL) 150 (115) 157(107) 0.49 a Total cholesterol (mg/dL) 190 (54 187(54) 0.56 a Hematocrit (%) 42 (7) 41(6) 0.45 a Monocytes (x10 3 µL) 0.61(0.36) 0.62(0.24) 0.42 a Hemoglobin (gr/dL) 13.9 ± 1.8 13.3 ± 1.9 0.12 b MPV(fL) 8.84 ± 1.71 10.43 ± 1.34 < 0.001 b Platelet count (k/mm 3 ) 271 ± 82.72 268 ± 78 0.85 b Lymphocytes (x10 3 µL) 2.37 ± 0.751 2.46 ± 1.01 0.60 b WBC (x10 3 µL) 8.2 ± 3.1 6.35 ± 2.01 0.18 b a Data without normal distribution shown as median (interquartile range), acomparison with Mann Whitney U test, b comparison with t-test. MPV: mean platelet volume; WBC: White blood cells; LDL: low-density lipoprotein cholesterol; HDL: high-density lipoprotein cholesterol. (p < 0.001, 95% CI for OR: 0.438 (0.306–0.629). A significant negative correlation was found between MPV and fistulae in Pearson’s correlation test (r: –0.454, p < 0.001). An MPV level < 9.6 fL had a sensitivity, specificity, positive predictive value and negative predictive value of 80%, 68%, 71% and 78% respectively (AUC = 0.766, 95% CI, 0.678–0.854) for CAF prediction. The overall accuracy of MPV when determining the presence of CAF was 75% (Figure 1). The right ventricular diameter was slightly increased in the CAF group. Other echocardiographic measurements were similar (Table 3). The angiographic features of the fistula group are summarized in Table 4. Discussion In this study, we showed that MPV levels were significantly reduced in the CAF group when compared to that with normal coronary arteries. Coronary arterial fistulas are usually asymptomatic. Usual symptoms include dyspnea, angina or fatigue on exertion and, occasionally, arrhythmias. Congestive heart failure may occasionally occur in childhood. 8-11 In adults, myocardial ischemia may occur due to coronary steal. 12 In our study, the right ventricular end-diastolic diameters of the fistula group were statistically larger than those in the control group regarding echocardiographic findings. The pathophysiological changes in CAF depend on the pressure difference between the fistula origin and its location and size. 1 This pressure difference determines the length, the width and tortuosity of the fistula. 1 A left-to-right shunt is found in over 90% of the cases. 1 When the fistula drains to the right side of the heart, there is a continuous flow from the fistula due to presence of low pressures in the right chambers, the volume load is increased to the right heart and, consequently, to the pulmonary vascular bed, the left heart chambers. However, when the fistula drains into the left heart chambers, pulmonary blood flow does not increase. Therefore, similar to our findings, right heart chamber dilation is more frequent than left-heart dilation. Especially in the case of a large-caliber CAF, the shifting of blood away from the normal coronary circulation may result in the coronary steal phenomenon and, consequently, in ischemia. 1 Automatic blood count analyzers calculate MPV in routine assays. It is affected by inflammation and it is accepted as a marker of thrombocyte activation. 13 A number of diseases, such as metabolic syndrome, myocardial infarction, acute ischemic stroke and diabetes mellitus have been associated with increased MPV. 13,14 Conversely, lower MPV has been reported in subjects with rheumatoid arthritis, Nasal Polyps and ankylosing spondylitis. 15,16 Similarly, several studies in the literature reported decreased levels of MPV in irritable bowel disease. 17,18 It has been proposed that lower and higher MPV values in different conditions are related with high and low-grade inflammation states. Gasparyan et al. 19 reported that diseases characterized by marked inflammation (e.g. rheumatoid arthritis) are associated with lower MPV values, while low-grade inflammation (e.g. nasal polyps, Behcet’s disease) are associated with increased MPV levels. 19 MPV was found to be higher in most cardiovascular diseases. 13,14 However, interestingly, MPV in CAF patients was found to be lower than in the control groups. Therefore, we speculate that, unlike acute coronary syndromes, CAF may be associated with a low but continuous inflammatory burden. Unfortunately, this is the first study in the literature that evaluated MPV in CAF. Another author suggested that activated platelets tend to enlarge and cause an elevation in MPV value, and might be utilized in such active inflammatory process, leaving smaller platelets, thus causing a reduction in MPV. 20 On the other hand, other authors speculate that overproduction of pro-inflammatory cytokines and acute 73