- Email: [email protected]
obese subjects
Nutrition, Metabolism & Cardiovascular Diseases (2018) xx, 1e5
Available online at www.sciencedirect.com
Nutrition, Metabolism & Cardiovascular Diseases journal homepage: www.elsevier.com/locate/nmcd
Platelet number is negatively and independently associated with carotid intima-media thickness in apparently healthy overweight/ obese subjects G. De Pergola a,*, R. Zupo a, A. Cecere b, N. Bartolomeo c, V. Triggiani d, S. Paradiso a, L. Lampignano a, F. Silvestris a, M.M. Ciccone b a Clinical Nutrition Unit, Medical Oncology, Department of Biomedical Science and Human Oncology, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare 11, 70124, Bari, Italy b Section of Cardiovascular Disease, Department of Organ Transplantation, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare 11, 70124, Bari, Italy c Medical Statistics, Department of Biomedical Science and Human Oncology, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare, 70124, Bari, Italy d Section of Internal Medicine, Geriatrics, Endocrinology and Rare Diseases, Interdisciplinary Department of Medicine, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare, 70124, Bari, Italy
Received 18 May 2018; received in revised form 1 August 2018; accepted 2 August 2018 Handling Editor: F. Galletti Available online - - -
KEYWORDS Obesity; Overweight; Platelet; Carotid intima-media thickness; Atherosclerosis
Abstract Background and Aim: A significant change of platelet number may be a risk factor for atherosclerotic cardiovascular disease. The aim of this study was to investigate the association between platelet number and early signs of atherosclerosis, evaluated by carotid intima-media thickness (c-IMT), in a apparently healthy population mainly represented by obese subjects. Methods and Results: As many as 961 subjects, 686 women and 275 men, aged between 18 and 74 years, were enrolled in the study. Of these, 54 individuals (5.6% of all subjects) were normal weight, 259 individuals (27.0% of all subjects) were overweight, and 648 individuals (67.4% of all subjects) were obese. Waist circumference (WC) and blood glucose, insulin, total cholesterol, high and low density lipoprotein cholesterol, triglycerides and platelet count were also detected in all subjects, who underwent carotid echo color doppler ultrasound to measure c-IMT. c-IMT was significantly and positively associated to age (r Z 0.204, P < 0.0001), fasting glucose (r Z 0.073, P < 0.0240), total cholesterol (r Z 0.096, P Z 0.0031), and systolic and diastolic blood pressure (r Z 0.140, P < 0.0001 and r Z 0.119, P < 0.0003 respectively); c-IMT was significantly and negatively correlated with platelet count (r Z 0.165, P < 0.0001). Only age (P < 0.0001) and systolic blood pressure (P Z 0.0393), positively, and platelet number (P < 0.0001), negatively, were significantly and independently associated to c-IMT in a final multiple regression analysis. Conclusion: Lower platelet number represented an independent determinant of c-IMT in a population, mainly represented by obese patients. These results suggest that a decrease of platelet number may well be an early defensive mechanism in subjects developing the thickening of carotid artery. ª 2018 Published by Elsevier B.V. on behalf of The Italian Society of Diabetology, the Italian Society for the Study of Atherosclerosis, the Italian Society of Human Nutrition, and the Department of Clinical Medicine and Surgery, Federico II University.
* Corresponding author. Via Putignani 236, 70122, Bari, Italy. Fax: þ39 080 5478831. E-mail address: [email protected] (G. De Pergola). https://doi.org/10.1016/j.numecd.2018.08.001 0939-4753/ª 2018 Published by Elsevier B.V. on behalf of The Italian Society of Diabetology, the Italian Society for the Study of Atherosclerosis, the Italian Society of Human Nutrition, and the Department of Clinical Medicine and Surgery, Federico II University.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
2
Introduction Platelets are involved in the pathogenesis of atherothrombotic diseases, and platelet number is either a marker of inflammation [1] or an indicator of vascular repair activity and of cardiovascular disease (CVD) risk [2]. Obesity is a global epidemic, and central obesity in particular is an important risk factor for major cardiovascular events, due to the atherosclerotic alterations of coronary, cerebral and lower limb arterial vessels [3]. Abdominal obesity is significantly associated with increased platelet counts [4], and obese subjects show higher aggregation and reactivity, lower membrane fluidity and reduced sensitivity to antiplatelet drugs [5e7]. The mechanisms responsible for platelet dysfunction are: a) lower sensitivity to insulin and other molecules acting via intracellular cyclic nucleotides (nitrates and prostacyclin in particular); b) altered intracellular ionic milieu with elevated cytosolic Ca2þ; and c) higher oxidative stress, which stimulates isoprostane production from arachidonic acid [7]. Obesity is also characterized by higher circulating levels of leptin, that promotes platelet aggregation at plasma concentrations corresponding to those commonly found in obese patients (>50 ng/ml), whereas it has not this effect at usual concentrations in normal individuals (<10 ng/ml) [8]. Common carotid artery intima-media thickness (c-IMT) is used to estimate early atherosclerosis and the risk of coronary disease [9]. However, no study has previously examined whether platelet number is a determinant of cIMT in obese patients without a personal history of CVD. Thus, the primary goal of this study was to examine whether platelet number is related to c-IMT independently of factors well known to favor the thickening of the intimamedia common carotid artery, such as gender, age [10e13], BMI [10e13], waist circumference [10e13], systolic and diastolic blood pressure [10e13], insulinemia and insulin resistance [10e13], fasting glucose [10e13], and lipid blood levels (triglycerides [10], total [10,13], and LDL cholesterol [10,13]). To this purpose, a population (n Z 961) of apparently healthy subjects, mainly represented by obese individuals, was investigated. Methods Subjects Subjects were recruited consecutively from January 2008 to december 2017 at the Outpatient Clinic of Nutrition of the Medical Oncology Unit, Department of Biomedical Sciences and Human Oncology, University of Bari, School of Medicine, Policlinico, Bari, Italy. The subjects were referred to the Outpatient Clinic with the aim of improving the quality of their diet and/or to loose body weight. They were enrolled at the first visit whether they did not take any kind of drugs (including oral contraceptives or drugs for osteoporosis). Exclusion criteria concerned subjects suffering from endocrine diseases (diabetes mellitus, hypo or hyperthyroidism, hypopituitarism etc.), malignancies, unstable hypertension, chronic inflammatory diseases,
G. De Pergola et al.
renal and liver failure, angina pectoris, myocardial infarction, heart failure, genetic heart diseases, minor and major stroke, and inherited thrombocytopenias. The group of participants was made up of 961 subjects, 686 women and 275 men, aged between 18 and 74 years. Patients were defined normal weight if they had a BMI <25.0 (n Z 54, 5.6% of all subjects), overweight if they had a BMI between 25.0 and 29.9 (n Z 259, 27.0% of all subjects), and obese if they had a BMI 30 (n Z 648, 67.4% of all subjects). Subjects were examined by means of medical history, routine tests and electrocardiogram (ECG). None was following a rigid low-calorie diet or carrying out a high level of physical activity, while each of them kept on doing their common lifestyle. A written informed consent was obtained from each subject. All the assessment were performed in accordance with the guidelines of the Helsinki Declaration on Human Experimentation. Clinical and biochemical assessment Body weight was rounded to the nearest kg. Height was determined to the nearest cm. BMI was calculated as weight (kg) divided by the square of height (m). The waist circumference was measured at the narrowest part of the abdomen, or in the area between the 10th rib and the iliac crest (minimum circumference). The blood pressure of ambulatory patients was determined in a sitting position after at least 10-min rest and at least three different times, using a mercury manometer with appropriately sized cuff. Blood samples were drawn between 08:00 and 09:00 a.m. after an overnight fast. Serum insulin concentrations were measured by radioimmunoassay (Behring, Scoppito, Italy). Plasma glucose was determined using the glucose oxidase method (Sclavus, Siena, Italy), while the concentrations of plasma lipids (triglycerides, total cholesterol, HDL cholesterol) were quantified by automatic colorimetric method (Hitachi; Boehringer Mannheim, Mannheim, Germany). LDL cholesterol was calculated using the Friedewald equation [14]. Blood glucose, insulin, total cholesterol (TC), high (HDL), low (LDL) density lipoprotein cholesterol and triglycerides (TG) were detected after overnight fasting in all subjects. The Homeostasis Model Assessment of insulin resistance (HOMA-IR) [15] was obtained from fasting insulin and fasting glucose. Platelet count was determined by a Coulter Hematology analyzer (BeckmaneCoulter, Brea, CA). Blood samples were collected in the morning and analyzed within 3 h after venipuncture. EDTA served as an anticoagulant. Ultrasound measurement of the c-IMT All subjects underwent high-definition vascular echography according to the following protocols to measure c-IMT, an useful tool to evaluate early atherosclerosis. Ultrasonographic echo-color Doppler studies of left and right common carotid arteries were performed by the
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
Platelet number and early atherosclerosis in obesity
same physician with a Philips Sonos 5500 using a 7.5-MHz high-resolution probe. The patients were placed in a supine position, with the neck extended and rotated contralaterally by 45 and the common carotid arteries were examined on the sagittal axis with a lateral view. We used the method of Pignoli et al. to define c-IMT [16], as described in our previous studies [10e13]: by focusing and freezing images on the distal wall of the common carotid artery on the lengthwise axis during end-diastole, the cIMT was calculated as the distance between the leading borders of the first hyperechoic line and of the second hyperechoic line, separated by a hypoechoic space (“double-track pattern”). The measurements were performed bilaterally 1 cm proximally to the carotid bulb, for three times, and then c-IMT value was calculated as the arithmetic mean of each side. The c-IMT value considered for statistical analyses was the mean of right and left measurements. We excluded, for the IMT detection, arterial segments with atherosclerotic plaques. All the echo color Doppler investigations were performed at the Cardiovascular Department of Policlinico of Bari. Statistical analysis Leverage value, Cook’s Distance and Jackknife residual was used to identify the outliers. The sample characteristics were expressed as mean standard deviation or median (interquartile range [IQR]) values for continuous data, and number (%) for categorical data. Shapiro-Walk test was used to determine whether the variables showed a normal distribution and the variables not normally distributed was transformed. It was evaluated the association between c-IMT and each parameter; Mardia test was used to verify multivariate normality. When variables had a bivariate normal distribution Pearson’s correlation coefficient was used. When variables had a bivariate non-normal distribution nonparametric Spearman’s correlation coefficient was used. Independent t-test (t) or ManneWhitney U test (MW) are used to compare c-IMT between sex. Variables with p-value lower than 0.25 in the univariate analysis were included in a multivariate general linear model (GLM); the Akaike’s information criterion (AIC), adjusted R-square statistic and the significance level of the F-statistic were the criteria used to estimate the final model. Normality of residuals assumption for the GLM model was checked. All tests of statistical significance were two-tailed, and p-values less than 0.05 were considered statistically significant. Statistical analysis was performed by software SAS (version 9.4 for PC). Results Twelve outliers observations were excluded from the analysis; thus the database was made up of 949 patients. Table 1 sums up the general, anthropometric, hormonal, metabolic and cardiovascular parameters of the population enrolled in the study. Table 2 shows the univariate relationship between c-IMT (logarithmic transformation) and other investigated variables, and the final multiple regression model.
3
Table 1 General, anthropometric, metabolic, and cardiovascular parameters (n Z 949 subjects). Characteristics
Median (IQR)
Body mass index (Kg/m2) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Cholesterol (mg/dl) Age (years) Fasting blood glucose (mg/dl) HOMA-IR c-IMT (mm) Insulin (mUI/ml) Diastolic blood pressure (mmHg) Systolic blood pressure (mmHg) Platelet count (n 103/ml) Triglycerides (mg/dl) Waist circumference (cm) Sex (woman)
32.4 (28.8e36.8) 48 (40e55) 118 (98e144) 188 (162e217) 39 (29e49) 90 (84e97) 4.04 (2.55e6.27) 0.65 (0.55e0.75) 18 (11.9e27) 80 (75e85) 124 (115e134) 265 (227e304) 90 (60e129) 105 (97e115) 680 (71.6%)a
a
n (%).
Concerning the linear correlation analyses between c-IMT and all the other variables of the population, c-IMT was significantly and negatively correlated with platelet count, and significantly and positively associated to age, fasting glucose, total cholesterol, and systolic and diastolic blood pressure. Gender, BMI, waist circumference, insulin, HOMA-IR, triglycerides, LDL cholesterol, and HDL cholesterol did not show a significant correlation with c-IMT. The final multiple regression model showed that only age, platelet number and systolic blood pressure were associated to c-IMT, considered as the dependent variable. GLM residuals were normally distributed. Discussion The aim of the present study was to investigate the relationship between platelet number and early atherosclerotic process in a population mainly represented by obese subjects without manifested CV disease. c-IMT has been used to determine structural vascular changes, because it represents a well-known tool to quantify the presence of early atherosclerosis [17,18]. We found a strong negative and independent relationship between platelet number and c-IMT. It is to note that when the platelet number was introduced in the multiple regression model, the association between c-IMT and other CVD risk factors showing a significant linear correlation with c-IMT (fasting glucose and cholesterol) disappeared, reinforcing the importance of the inverse association between c-IMT and platelet count. The positive association between c-IMT and age or between c-IMT and systolic blood pressure were maintained independently of platelet count, thus confirming the results of previous studies [10e13]. Our findings strongly suggest that the inverse and independent association between platelet number and c-IMT may represent a compensatory mechanism protecting against atherothrombotic diseases, at least in people with excess of body fat. It may well be that some molecules (adhesion molecules, etc) coming from the arterial wall,
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
4
G. De Pergola et al.
Table 2 Univariate relationship between c-IMT (logarithmic transformation) and other investigated variables and the final multiple regression model. Characteristics
Univariate analysis Correlation coefficient 2
Body mass index (Kg/m ) HDL-cholesterol (mg/dl)a LDL-cholesterol (mg/dl)b Cholesterol (mg/dl) Age (years) Fasting blood glucose (mg/dl) HOMA-IRa Insulin (mUI/ml)a Diastolic blood pressure (mmHg) Systolic blood pressure (mmHg) Platelet count (n 103/ml)a Triglycerides (mg/dl)a Waist circumference (cm) Sex
d
0.019 0.054c 0.051c 0.096d 0.204c 0.073d 0.007c 0.024c 0.119d 0.140d 0.165c 0.046c 0.056d 0.184e
Multivariate analysis P 0.5538 0.0985 0.1194 0.0031 <.0001 0.0240 0.8390 0.4585 0.0003 <.0001 <.0001 0.1545 0.0835 0.2264
Estimate
St. Error
p
0.003
0.0006
<.0001
0.001 0.146
0.0005 0.0298
0.0393 <.0001
Variables included in the stepwise regression method: HDL-cholesterol, LDL-cholesterol, Cholesterol, Age, Fasting blood glucose, Diastolic blood pressure, Systolic blood pressure, Triglycerides, Waist circumference and Sex. Bold indicates statistically significant values. a Logarithmic transformation. b Square root transformation c Parametric Pearson Coefficient. d Non Parametric Spearman Coefficient. e Difference man-woman.
and progressively increasing with the thickening of c-IMT, inhibit megakaryocytes differentiation and platelet production in the bone marrow. Future studies could address this topic. To our knowledge, this is the first large study considering the arterial wall as a possible determinant of the number of PLT in an adult healthy population. The utility of this information is that the clinical meaning of the thickening of the arterial wall as parameter of cardiovascular risk is strongly influenced by the platelet number. Interestingly, Mediterranean diet has been shown to be responsible of a decrease of platelet number [1] and, since this study was performed in a typical Mediterranean area (Apulia), we cannot exclude that common diet of the subjects under investigation had any influence on the association between platelet count and c-IMT. Our results are only apparently in contrast with the finding that platelets are a possible marker of inflammation and of greater risk for cerebrovascular and coronary heart disease [1]. Actually, a recent prospective study (the Moli-sani study) showed that a higher platelet count was not associated with any death risk, whereas a lower platelet number was associated with increased risk of total, cancer, and non-cardiovascular/non-cancer mortality, whereas was unrelated to cardiovascular mortality [19]. It is clear that further studies are needed to clarify the relationship between platelet count and CVD risk. However, we know that platelets have an important role in vascular culture together with CD34-positive cells [20] and induce the differentiation of human CD34-positive cells into foam cells [21], which are a well-known contributing factor in the development of atherosclerotic lesions. In fact, CD34-positive cells have previously been observed in atherosclerotic lesions in humans [22]. In addition, platelet-rich plasma promotes angiogenesis [23].
The only study that has previously examined the relationship between platelet count and c-IMT was performed in a population of hypertensive subjects and showed a positive association between these two parameters [24]. However, in that study, platelet count was positively associated with c-IMT in hypertensive subjects, but not in non-hypertensive subjects [24]. Therefore, it may well be that the relationship between platelet number and c-IMT is strongly influenced by the simultaneous presence of hypertension. In our opinion, the strong aspect of this study is that we examined a high number of subjects (almost 1000), and the data were only taken in individuals without medications and, therefore, we did not have any interference of drugs on the statistical associations between c-IMT and CV disease risk factors. By contrast, a limitation is that we examined the number of platelets, but not the volume or the function of these elements. Another limitation of this study is its cross-sectional nature, which cannot allow to state the possible causality of associations. Moreover, the study population does not allow generalizability to other population settings. Finally, since we did not select the subjects on the basis of the platelets count, the number of platelets was not balanced according to sex and different age groups, and this limitation could not allow to achieve further analyses on the relationship between platelet number and c-IMT according to sex and age.
Conclusion Lower platelet number was a strong and independent determinant of c-IMT in a apparently healthy population, mainly represented by overweight and obese population.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
Platelet number and early atherosclerosis in obesity
These results suggest that a decrease of platelet number may be a defensive mechanism protecting against early atherosclerosis in people at higher cardiovascular risk, such as people exceeding body weight. Conflicts of interest None. References [1] Bonaccio M, Di Castelnuovo A, De Curtis A, Costanzo S, Persichillo M, Donati MB, et al., on behalf of the Moli-sani Project Investigators. Adherence to the Mediterranean diet is associated with lower platelet and leukocyte counts: results from the Molisani study. Blood 2014;123(19):3037e44. [2] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801e9. [3] Mathew B, Francis L, Kayalar A, Cone J. Obesity: effects on cardiovascular disease and its diagnosis. J Am Board Fam Med 2008 NovDec;21(6):562e8. https://doi.org/10.3122/jabfm.2008.06.080080. [4] Charles LE, Fekedulegn D, McCall T, Burchfiel CM, Andrew ME, Violanti JM. Obesity, white blood cell counts, and platelet counts among police officers. Obesity 2007;15:2846e54. [5] Ades PA, Savage PD, Toth MJ, Schneider DJ, Audelin MC, Bunn JY, et al. The influence of obesity and consequent insulin resistance on coronary risk factors in medically treated patients with coronary disease. Int J Obes 2008;32:967e74. [6] Raffaelli F, Nanetti L, D’Angelo M, Montecchiani G, Alidori A, Montesi L, et al. Interactions between lipoproteins and platelet membranes in obesity. Obesity 2009;17:1375e80. [7] Anfossi G, Russo I, Trovati M. Platelet dysfunction in central obesity. Nutr Metabol Cardiovasc Dis 2009;19:440e9. [8] Nakata M, Yada T, Soejima N, Maruyama I. Leptin promotes aggregation of human platelets via the long form of its receptor. Diabetes 1999;48:426e9. [9] Granér M, Varpula M, Kahri J, Salonen RM, Nyyssönen K, Nieminen MS, et al. Association of carotid intima-media thickness with angiographic severity and extent of coronary artery disease. Am J Cardiol 2006;97:624e9. [10] Ciccone MM, Cortese F, Gesualdo M, Donvito I, Carbonara S, De Pergola G. A glycemic threshold of 90 mg/dl promotes early signs of atherosclerosis in apparently healthy overweight/obese subjects. Endocr Metab Immune Disord Drug Targets 2016;16(4):288e95. [11] De Pergola G, Ciccone MM, Guida P, Morea G, Giannuzzo E, Cortese F, et al. Relationship between C3 levels and common carotid intima-media thickness in overweight and obese patients. Obes Facts 2011;4(2):159e63.
5 [12] De Pergola G, Pannacciulli N, Ciccone M, Tartagni M, Rizzon P, Giorgino R. Free testosterone plasma levels are negatively associated with the intima-media thickness of the common carotid artery in overweight and obese glucose-tolerant young adult men. Int J Obes Relat Metab Disord 2003;27(7):803e7. [13] Ciccone M, Vettor R, Pannacciulli N, Minenna A, Bellacicco M, Rizzon P, et al. Plasma leptin is independently associated with the intima-media thickness of the common carotid artery. Int J Obes Relat Metab Disord 2001;25(6):805e10. [14] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499e502. [15] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985 Jul;28(7): 412e9. [16] Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial thickness of the arterial wall: a direct measurement with ultrasound imaging. Circulation 1986;74:1399e406. [17] Lorenz MW, von Kegler S, Steinmetz H, Markus HS, Sitzer M. Carotid intimaemedia thickening indicates a higher vascular risk across a wide age range: prospective data from the Carotid Atherosclerosis Progression Study (CAPS). Stroke 2006;37: 87e92. [18] Greenland P, Abrams J, Aurigemma GP, Bond MG, Clark LT, Criqui MH, et al. Prevention Conference V: beyond secondary prevention: identifying the high-risk patient for primary prevention: noninvasive tests of atherosclerotic burden: writing Group III. Circulation 2000 Jan 4;101(1):E16e22. [19] Bonaccio M, Di Castelnuovo A, Costanzo S, De Curtis A, Donati MB, Cerletti C, et al. Moli-sani Investigators. Age- and sex-based ranges of platelet count and cause-specific mortality risk in an adult general population: prospective findings from the Moli-sani study. Platelets 2018;29(3):312e5. [20] Cangiano E, Cavazza C, Campo G, Valgimigli M, Francolini G, Malagutti P, et al. Different clinical models of CD34þ cells mobilization in patients with cardiovascular disease. J Thromb Thrombolysis 2011;32:1e8. [21] Daub K, Langer H, Seizer P, Stellos K, May AE, Goyal P, et al. Platelets induce differentiation of human CD34þ progenitor cells into foam cells and endothelial cells. FASEB J 2006;20:2261e559. [22] Torsney E, Mandal K, Halliday A, Jahangiri M, Xu Ǫ. Characterisation of progenitor cells in human atherosclerotic vessels. Atherosclerosis 2007;191:259e64. [23] Mammoto T, Jiang A, Jiang E, Mammoto A. Platelet rich plasma extract promotes angiogenesis through the angiopoietin1-Tie2 pathway. Microvasc Res 2013;89:15e24. [24] Shimizu Y, Sato S, Koyamatsu J, Yamanashi H, Nagayoshi M, Kadota K, et al. Platelets as an indicator of vascular repair in elderly Japanese men. Oncotarget 2016;7:44919e26.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
Available online at www.sciencedirect.com
Nutrition, Metabolism & Cardiovascular Diseases journal homepage: www.elsevier.com/locate/nmcd
Platelet number is negatively and independently associated with carotid intima-media thickness in apparently healthy overweight/ obese subjects G. De Pergola a,*, R. Zupo a, A. Cecere b, N. Bartolomeo c, V. Triggiani d, S. Paradiso a, L. Lampignano a, F. Silvestris a, M.M. Ciccone b a Clinical Nutrition Unit, Medical Oncology, Department of Biomedical Science and Human Oncology, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare 11, 70124, Bari, Italy b Section of Cardiovascular Disease, Department of Organ Transplantation, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare 11, 70124, Bari, Italy c Medical Statistics, Department of Biomedical Science and Human Oncology, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare, 70124, Bari, Italy d Section of Internal Medicine, Geriatrics, Endocrinology and Rare Diseases, Interdisciplinary Department of Medicine, University of Bari, School of Medicine, Policlinico, Piazza GiulioCesare, 70124, Bari, Italy
Received 18 May 2018; received in revised form 1 August 2018; accepted 2 August 2018 Handling Editor: F. Galletti Available online - - -
KEYWORDS Obesity; Overweight; Platelet; Carotid intima-media thickness; Atherosclerosis
Abstract Background and Aim: A significant change of platelet number may be a risk factor for atherosclerotic cardiovascular disease. The aim of this study was to investigate the association between platelet number and early signs of atherosclerosis, evaluated by carotid intima-media thickness (c-IMT), in a apparently healthy population mainly represented by obese subjects. Methods and Results: As many as 961 subjects, 686 women and 275 men, aged between 18 and 74 years, were enrolled in the study. Of these, 54 individuals (5.6% of all subjects) were normal weight, 259 individuals (27.0% of all subjects) were overweight, and 648 individuals (67.4% of all subjects) were obese. Waist circumference (WC) and blood glucose, insulin, total cholesterol, high and low density lipoprotein cholesterol, triglycerides and platelet count were also detected in all subjects, who underwent carotid echo color doppler ultrasound to measure c-IMT. c-IMT was significantly and positively associated to age (r Z 0.204, P < 0.0001), fasting glucose (r Z 0.073, P < 0.0240), total cholesterol (r Z 0.096, P Z 0.0031), and systolic and diastolic blood pressure (r Z 0.140, P < 0.0001 and r Z 0.119, P < 0.0003 respectively); c-IMT was significantly and negatively correlated with platelet count (r Z 0.165, P < 0.0001). Only age (P < 0.0001) and systolic blood pressure (P Z 0.0393), positively, and platelet number (P < 0.0001), negatively, were significantly and independently associated to c-IMT in a final multiple regression analysis. Conclusion: Lower platelet number represented an independent determinant of c-IMT in a population, mainly represented by obese patients. These results suggest that a decrease of platelet number may well be an early defensive mechanism in subjects developing the thickening of carotid artery. ª 2018 Published by Elsevier B.V. on behalf of The Italian Society of Diabetology, the Italian Society for the Study of Atherosclerosis, the Italian Society of Human Nutrition, and the Department of Clinical Medicine and Surgery, Federico II University.
* Corresponding author. Via Putignani 236, 70122, Bari, Italy. Fax: þ39 080 5478831. E-mail address: [email protected] (G. De Pergola). https://doi.org/10.1016/j.numecd.2018.08.001 0939-4753/ª 2018 Published by Elsevier B.V. on behalf of The Italian Society of Diabetology, the Italian Society for the Study of Atherosclerosis, the Italian Society of Human Nutrition, and the Department of Clinical Medicine and Surgery, Federico II University.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
2
Introduction Platelets are involved in the pathogenesis of atherothrombotic diseases, and platelet number is either a marker of inflammation [1] or an indicator of vascular repair activity and of cardiovascular disease (CVD) risk [2]. Obesity is a global epidemic, and central obesity in particular is an important risk factor for major cardiovascular events, due to the atherosclerotic alterations of coronary, cerebral and lower limb arterial vessels [3]. Abdominal obesity is significantly associated with increased platelet counts [4], and obese subjects show higher aggregation and reactivity, lower membrane fluidity and reduced sensitivity to antiplatelet drugs [5e7]. The mechanisms responsible for platelet dysfunction are: a) lower sensitivity to insulin and other molecules acting via intracellular cyclic nucleotides (nitrates and prostacyclin in particular); b) altered intracellular ionic milieu with elevated cytosolic Ca2þ; and c) higher oxidative stress, which stimulates isoprostane production from arachidonic acid [7]. Obesity is also characterized by higher circulating levels of leptin, that promotes platelet aggregation at plasma concentrations corresponding to those commonly found in obese patients (>50 ng/ml), whereas it has not this effect at usual concentrations in normal individuals (<10 ng/ml) [8]. Common carotid artery intima-media thickness (c-IMT) is used to estimate early atherosclerosis and the risk of coronary disease [9]. However, no study has previously examined whether platelet number is a determinant of cIMT in obese patients without a personal history of CVD. Thus, the primary goal of this study was to examine whether platelet number is related to c-IMT independently of factors well known to favor the thickening of the intimamedia common carotid artery, such as gender, age [10e13], BMI [10e13], waist circumference [10e13], systolic and diastolic blood pressure [10e13], insulinemia and insulin resistance [10e13], fasting glucose [10e13], and lipid blood levels (triglycerides [10], total [10,13], and LDL cholesterol [10,13]). To this purpose, a population (n Z 961) of apparently healthy subjects, mainly represented by obese individuals, was investigated. Methods Subjects Subjects were recruited consecutively from January 2008 to december 2017 at the Outpatient Clinic of Nutrition of the Medical Oncology Unit, Department of Biomedical Sciences and Human Oncology, University of Bari, School of Medicine, Policlinico, Bari, Italy. The subjects were referred to the Outpatient Clinic with the aim of improving the quality of their diet and/or to loose body weight. They were enrolled at the first visit whether they did not take any kind of drugs (including oral contraceptives or drugs for osteoporosis). Exclusion criteria concerned subjects suffering from endocrine diseases (diabetes mellitus, hypo or hyperthyroidism, hypopituitarism etc.), malignancies, unstable hypertension, chronic inflammatory diseases,
G. De Pergola et al.
renal and liver failure, angina pectoris, myocardial infarction, heart failure, genetic heart diseases, minor and major stroke, and inherited thrombocytopenias. The group of participants was made up of 961 subjects, 686 women and 275 men, aged between 18 and 74 years. Patients were defined normal weight if they had a BMI <25.0 (n Z 54, 5.6% of all subjects), overweight if they had a BMI between 25.0 and 29.9 (n Z 259, 27.0% of all subjects), and obese if they had a BMI 30 (n Z 648, 67.4% of all subjects). Subjects were examined by means of medical history, routine tests and electrocardiogram (ECG). None was following a rigid low-calorie diet or carrying out a high level of physical activity, while each of them kept on doing their common lifestyle. A written informed consent was obtained from each subject. All the assessment were performed in accordance with the guidelines of the Helsinki Declaration on Human Experimentation. Clinical and biochemical assessment Body weight was rounded to the nearest kg. Height was determined to the nearest cm. BMI was calculated as weight (kg) divided by the square of height (m). The waist circumference was measured at the narrowest part of the abdomen, or in the area between the 10th rib and the iliac crest (minimum circumference). The blood pressure of ambulatory patients was determined in a sitting position after at least 10-min rest and at least three different times, using a mercury manometer with appropriately sized cuff. Blood samples were drawn between 08:00 and 09:00 a.m. after an overnight fast. Serum insulin concentrations were measured by radioimmunoassay (Behring, Scoppito, Italy). Plasma glucose was determined using the glucose oxidase method (Sclavus, Siena, Italy), while the concentrations of plasma lipids (triglycerides, total cholesterol, HDL cholesterol) were quantified by automatic colorimetric method (Hitachi; Boehringer Mannheim, Mannheim, Germany). LDL cholesterol was calculated using the Friedewald equation [14]. Blood glucose, insulin, total cholesterol (TC), high (HDL), low (LDL) density lipoprotein cholesterol and triglycerides (TG) were detected after overnight fasting in all subjects. The Homeostasis Model Assessment of insulin resistance (HOMA-IR) [15] was obtained from fasting insulin and fasting glucose. Platelet count was determined by a Coulter Hematology analyzer (BeckmaneCoulter, Brea, CA). Blood samples were collected in the morning and analyzed within 3 h after venipuncture. EDTA served as an anticoagulant. Ultrasound measurement of the c-IMT All subjects underwent high-definition vascular echography according to the following protocols to measure c-IMT, an useful tool to evaluate early atherosclerosis. Ultrasonographic echo-color Doppler studies of left and right common carotid arteries were performed by the
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
Platelet number and early atherosclerosis in obesity
same physician with a Philips Sonos 5500 using a 7.5-MHz high-resolution probe. The patients were placed in a supine position, with the neck extended and rotated contralaterally by 45 and the common carotid arteries were examined on the sagittal axis with a lateral view. We used the method of Pignoli et al. to define c-IMT [16], as described in our previous studies [10e13]: by focusing and freezing images on the distal wall of the common carotid artery on the lengthwise axis during end-diastole, the cIMT was calculated as the distance between the leading borders of the first hyperechoic line and of the second hyperechoic line, separated by a hypoechoic space (“double-track pattern”). The measurements were performed bilaterally 1 cm proximally to the carotid bulb, for three times, and then c-IMT value was calculated as the arithmetic mean of each side. The c-IMT value considered for statistical analyses was the mean of right and left measurements. We excluded, for the IMT detection, arterial segments with atherosclerotic plaques. All the echo color Doppler investigations were performed at the Cardiovascular Department of Policlinico of Bari. Statistical analysis Leverage value, Cook’s Distance and Jackknife residual was used to identify the outliers. The sample characteristics were expressed as mean standard deviation or median (interquartile range [IQR]) values for continuous data, and number (%) for categorical data. Shapiro-Walk test was used to determine whether the variables showed a normal distribution and the variables not normally distributed was transformed. It was evaluated the association between c-IMT and each parameter; Mardia test was used to verify multivariate normality. When variables had a bivariate normal distribution Pearson’s correlation coefficient was used. When variables had a bivariate non-normal distribution nonparametric Spearman’s correlation coefficient was used. Independent t-test (t) or ManneWhitney U test (MW) are used to compare c-IMT between sex. Variables with p-value lower than 0.25 in the univariate analysis were included in a multivariate general linear model (GLM); the Akaike’s information criterion (AIC), adjusted R-square statistic and the significance level of the F-statistic were the criteria used to estimate the final model. Normality of residuals assumption for the GLM model was checked. All tests of statistical significance were two-tailed, and p-values less than 0.05 were considered statistically significant. Statistical analysis was performed by software SAS (version 9.4 for PC). Results Twelve outliers observations were excluded from the analysis; thus the database was made up of 949 patients. Table 1 sums up the general, anthropometric, hormonal, metabolic and cardiovascular parameters of the population enrolled in the study. Table 2 shows the univariate relationship between c-IMT (logarithmic transformation) and other investigated variables, and the final multiple regression model.
3
Table 1 General, anthropometric, metabolic, and cardiovascular parameters (n Z 949 subjects). Characteristics
Median (IQR)
Body mass index (Kg/m2) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Cholesterol (mg/dl) Age (years) Fasting blood glucose (mg/dl) HOMA-IR c-IMT (mm) Insulin (mUI/ml) Diastolic blood pressure (mmHg) Systolic blood pressure (mmHg) Platelet count (n 103/ml) Triglycerides (mg/dl) Waist circumference (cm) Sex (woman)
32.4 (28.8e36.8) 48 (40e55) 118 (98e144) 188 (162e217) 39 (29e49) 90 (84e97) 4.04 (2.55e6.27) 0.65 (0.55e0.75) 18 (11.9e27) 80 (75e85) 124 (115e134) 265 (227e304) 90 (60e129) 105 (97e115) 680 (71.6%)a
a
n (%).
Concerning the linear correlation analyses between c-IMT and all the other variables of the population, c-IMT was significantly and negatively correlated with platelet count, and significantly and positively associated to age, fasting glucose, total cholesterol, and systolic and diastolic blood pressure. Gender, BMI, waist circumference, insulin, HOMA-IR, triglycerides, LDL cholesterol, and HDL cholesterol did not show a significant correlation with c-IMT. The final multiple regression model showed that only age, platelet number and systolic blood pressure were associated to c-IMT, considered as the dependent variable. GLM residuals were normally distributed. Discussion The aim of the present study was to investigate the relationship between platelet number and early atherosclerotic process in a population mainly represented by obese subjects without manifested CV disease. c-IMT has been used to determine structural vascular changes, because it represents a well-known tool to quantify the presence of early atherosclerosis [17,18]. We found a strong negative and independent relationship between platelet number and c-IMT. It is to note that when the platelet number was introduced in the multiple regression model, the association between c-IMT and other CVD risk factors showing a significant linear correlation with c-IMT (fasting glucose and cholesterol) disappeared, reinforcing the importance of the inverse association between c-IMT and platelet count. The positive association between c-IMT and age or between c-IMT and systolic blood pressure were maintained independently of platelet count, thus confirming the results of previous studies [10e13]. Our findings strongly suggest that the inverse and independent association between platelet number and c-IMT may represent a compensatory mechanism protecting against atherothrombotic diseases, at least in people with excess of body fat. It may well be that some molecules (adhesion molecules, etc) coming from the arterial wall,
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
4
G. De Pergola et al.
Table 2 Univariate relationship between c-IMT (logarithmic transformation) and other investigated variables and the final multiple regression model. Characteristics
Univariate analysis Correlation coefficient 2
Body mass index (Kg/m ) HDL-cholesterol (mg/dl)a LDL-cholesterol (mg/dl)b Cholesterol (mg/dl) Age (years) Fasting blood glucose (mg/dl) HOMA-IRa Insulin (mUI/ml)a Diastolic blood pressure (mmHg) Systolic blood pressure (mmHg) Platelet count (n 103/ml)a Triglycerides (mg/dl)a Waist circumference (cm) Sex
d
0.019 0.054c 0.051c 0.096d 0.204c 0.073d 0.007c 0.024c 0.119d 0.140d 0.165c 0.046c 0.056d 0.184e
Multivariate analysis P 0.5538 0.0985 0.1194 0.0031 <.0001 0.0240 0.8390 0.4585 0.0003 <.0001 <.0001 0.1545 0.0835 0.2264
Estimate
St. Error
p
0.003
0.0006
<.0001
0.001 0.146
0.0005 0.0298
0.0393 <.0001
Variables included in the stepwise regression method: HDL-cholesterol, LDL-cholesterol, Cholesterol, Age, Fasting blood glucose, Diastolic blood pressure, Systolic blood pressure, Triglycerides, Waist circumference and Sex. Bold indicates statistically significant values. a Logarithmic transformation. b Square root transformation c Parametric Pearson Coefficient. d Non Parametric Spearman Coefficient. e Difference man-woman.
and progressively increasing with the thickening of c-IMT, inhibit megakaryocytes differentiation and platelet production in the bone marrow. Future studies could address this topic. To our knowledge, this is the first large study considering the arterial wall as a possible determinant of the number of PLT in an adult healthy population. The utility of this information is that the clinical meaning of the thickening of the arterial wall as parameter of cardiovascular risk is strongly influenced by the platelet number. Interestingly, Mediterranean diet has been shown to be responsible of a decrease of platelet number [1] and, since this study was performed in a typical Mediterranean area (Apulia), we cannot exclude that common diet of the subjects under investigation had any influence on the association between platelet count and c-IMT. Our results are only apparently in contrast with the finding that platelets are a possible marker of inflammation and of greater risk for cerebrovascular and coronary heart disease [1]. Actually, a recent prospective study (the Moli-sani study) showed that a higher platelet count was not associated with any death risk, whereas a lower platelet number was associated with increased risk of total, cancer, and non-cardiovascular/non-cancer mortality, whereas was unrelated to cardiovascular mortality [19]. It is clear that further studies are needed to clarify the relationship between platelet count and CVD risk. However, we know that platelets have an important role in vascular culture together with CD34-positive cells [20] and induce the differentiation of human CD34-positive cells into foam cells [21], which are a well-known contributing factor in the development of atherosclerotic lesions. In fact, CD34-positive cells have previously been observed in atherosclerotic lesions in humans [22]. In addition, platelet-rich plasma promotes angiogenesis [23].
The only study that has previously examined the relationship between platelet count and c-IMT was performed in a population of hypertensive subjects and showed a positive association between these two parameters [24]. However, in that study, platelet count was positively associated with c-IMT in hypertensive subjects, but not in non-hypertensive subjects [24]. Therefore, it may well be that the relationship between platelet number and c-IMT is strongly influenced by the simultaneous presence of hypertension. In our opinion, the strong aspect of this study is that we examined a high number of subjects (almost 1000), and the data were only taken in individuals without medications and, therefore, we did not have any interference of drugs on the statistical associations between c-IMT and CV disease risk factors. By contrast, a limitation is that we examined the number of platelets, but not the volume or the function of these elements. Another limitation of this study is its cross-sectional nature, which cannot allow to state the possible causality of associations. Moreover, the study population does not allow generalizability to other population settings. Finally, since we did not select the subjects on the basis of the platelets count, the number of platelets was not balanced according to sex and different age groups, and this limitation could not allow to achieve further analyses on the relationship between platelet number and c-IMT according to sex and age.
Conclusion Lower platelet number was a strong and independent determinant of c-IMT in a apparently healthy population, mainly represented by overweight and obese population.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001
Platelet number and early atherosclerosis in obesity
These results suggest that a decrease of platelet number may be a defensive mechanism protecting against early atherosclerosis in people at higher cardiovascular risk, such as people exceeding body weight. Conflicts of interest None. References [1] Bonaccio M, Di Castelnuovo A, De Curtis A, Costanzo S, Persichillo M, Donati MB, et al., on behalf of the Moli-sani Project Investigators. Adherence to the Mediterranean diet is associated with lower platelet and leukocyte counts: results from the Molisani study. Blood 2014;123(19):3037e44. [2] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801e9. [3] Mathew B, Francis L, Kayalar A, Cone J. Obesity: effects on cardiovascular disease and its diagnosis. J Am Board Fam Med 2008 NovDec;21(6):562e8. https://doi.org/10.3122/jabfm.2008.06.080080. [4] Charles LE, Fekedulegn D, McCall T, Burchfiel CM, Andrew ME, Violanti JM. Obesity, white blood cell counts, and platelet counts among police officers. Obesity 2007;15:2846e54. [5] Ades PA, Savage PD, Toth MJ, Schneider DJ, Audelin MC, Bunn JY, et al. The influence of obesity and consequent insulin resistance on coronary risk factors in medically treated patients with coronary disease. Int J Obes 2008;32:967e74. [6] Raffaelli F, Nanetti L, D’Angelo M, Montecchiani G, Alidori A, Montesi L, et al. Interactions between lipoproteins and platelet membranes in obesity. Obesity 2009;17:1375e80. [7] Anfossi G, Russo I, Trovati M. Platelet dysfunction in central obesity. Nutr Metabol Cardiovasc Dis 2009;19:440e9. [8] Nakata M, Yada T, Soejima N, Maruyama I. Leptin promotes aggregation of human platelets via the long form of its receptor. Diabetes 1999;48:426e9. [9] Granér M, Varpula M, Kahri J, Salonen RM, Nyyssönen K, Nieminen MS, et al. Association of carotid intima-media thickness with angiographic severity and extent of coronary artery disease. Am J Cardiol 2006;97:624e9. [10] Ciccone MM, Cortese F, Gesualdo M, Donvito I, Carbonara S, De Pergola G. A glycemic threshold of 90 mg/dl promotes early signs of atherosclerosis in apparently healthy overweight/obese subjects. Endocr Metab Immune Disord Drug Targets 2016;16(4):288e95. [11] De Pergola G, Ciccone MM, Guida P, Morea G, Giannuzzo E, Cortese F, et al. Relationship between C3 levels and common carotid intima-media thickness in overweight and obese patients. Obes Facts 2011;4(2):159e63.
5 [12] De Pergola G, Pannacciulli N, Ciccone M, Tartagni M, Rizzon P, Giorgino R. Free testosterone plasma levels are negatively associated with the intima-media thickness of the common carotid artery in overweight and obese glucose-tolerant young adult men. Int J Obes Relat Metab Disord 2003;27(7):803e7. [13] Ciccone M, Vettor R, Pannacciulli N, Minenna A, Bellacicco M, Rizzon P, et al. Plasma leptin is independently associated with the intima-media thickness of the common carotid artery. Int J Obes Relat Metab Disord 2001;25(6):805e10. [14] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499e502. [15] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985 Jul;28(7): 412e9. [16] Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial thickness of the arterial wall: a direct measurement with ultrasound imaging. Circulation 1986;74:1399e406. [17] Lorenz MW, von Kegler S, Steinmetz H, Markus HS, Sitzer M. Carotid intimaemedia thickening indicates a higher vascular risk across a wide age range: prospective data from the Carotid Atherosclerosis Progression Study (CAPS). Stroke 2006;37: 87e92. [18] Greenland P, Abrams J, Aurigemma GP, Bond MG, Clark LT, Criqui MH, et al. Prevention Conference V: beyond secondary prevention: identifying the high-risk patient for primary prevention: noninvasive tests of atherosclerotic burden: writing Group III. Circulation 2000 Jan 4;101(1):E16e22. [19] Bonaccio M, Di Castelnuovo A, Costanzo S, De Curtis A, Donati MB, Cerletti C, et al. Moli-sani Investigators. Age- and sex-based ranges of platelet count and cause-specific mortality risk in an adult general population: prospective findings from the Moli-sani study. Platelets 2018;29(3):312e5. [20] Cangiano E, Cavazza C, Campo G, Valgimigli M, Francolini G, Malagutti P, et al. Different clinical models of CD34þ cells mobilization in patients with cardiovascular disease. J Thromb Thrombolysis 2011;32:1e8. [21] Daub K, Langer H, Seizer P, Stellos K, May AE, Goyal P, et al. Platelets induce differentiation of human CD34þ progenitor cells into foam cells and endothelial cells. FASEB J 2006;20:2261e559. [22] Torsney E, Mandal K, Halliday A, Jahangiri M, Xu Ǫ. Characterisation of progenitor cells in human atherosclerotic vessels. Atherosclerosis 2007;191:259e64. [23] Mammoto T, Jiang A, Jiang E, Mammoto A. Platelet rich plasma extract promotes angiogenesis through the angiopoietin1-Tie2 pathway. Microvasc Res 2013;89:15e24. [24] Shimizu Y, Sato S, Koyamatsu J, Yamanashi H, Nagayoshi M, Kadota K, et al. Platelets as an indicator of vascular repair in elderly Japanese men. Oncotarget 2016;7:44919e26.
Please cite this article in press as: De Pergola G, et al., Platelet number is negatively and independently associated with carotid intimamedia thickness in apparently healthy overweight/obese subjects, Nutrition, Metabolism & Cardiovascular Diseases (2018), https:// doi.org/10.1016/j.numecd.2018.08.001