High cardiovascular risk in young Saudi males: Cardiovascular risk factors, diet and inflammatory markers

Clinica Chimica Acta 365 (2006) 288 – 296 www.elsevier.com/locate/clinchim

High cardiovascular risk in young Saudi males: Cardiovascular risk factors, diet and inflammatory markers Eman M. Alissa a,b,*, Suhad M. Bahjri a, Nabeel Al-ama a,d, Waqar H. Ahmed a,c, Gordon A.A. Ferns b b

a Faculty of Medicine, King Abdul Aziz University, PO Box 12713, Jeddah 21483, Kingdom of Saudi Arabia Centre for Clinical Science & Measurement, School of Biomedical & Molecular Science, University of Surrey, Guildford, Surrey GU2 7XH, UK c Department of Cardiology, King Fahd Armed Forces Hospital, Jeddah, Kingdom of Saudi Arabia d Department of Medicine, King Abdul Aziz University Hospital, Jeddah, Kingdom of Saudi Arabia

Received 22 May 2005; received in revised form 6 September 2005; accepted 8 September 2005 Available online 6 October 2005

Abstract Background: The relationship between coronary risk score (CRS), individual coronary risk factors and the serum inflammatory markers, high sensitivity C-reactive protein (hsCRP), ceruloplasmin (Cp), and soluble intercellular adhesion molecule-1 (sICAM-1) was studied in 140 Saudi males without clinically evident coronary heart disease (CHD). Methods: One hundred forty subjects without clinically evident CHD were categorized into age tertiles. Demographic data together with an estimate of CRS using Framingham and PROCAM algorithms were obtained, and serum lipid profile, glucose, hsCRP, sICAM-1, and Cp were measured. Macronutrient intake was assessed by a questionnaire. The relationship between CRS, biochemical markers and diet was assessed by univariate and multivariate analysis. Results: There was no significant difference in median hsCRP, sICAM-1 or Cp between the age groups. Serum Cp was positively associated with age (r = 0.224, p < 0.01) and FRS score (r = 0.174, p < 0.05). Serum sICAM-1 was negatively associated with PROCAM score (r = 0.183, p < 0.05). sICAM-1 was positively associated with HDL cholesterol (r = 0.36, p < 0.0001) among non-diabetics and negatively associated (r = 0.397, p < 0.05) among diabetic subjects. Age and dietary intake of saturated fatty acids together explained 7.9% of the variation in serum Cp level in a stepwise multiple regression model. Similarly 6.5% of the variation in serum sICAM-1 level was explained by the total cholesterol / HDL-C ratio. The youngest tertile of the group (<30 y) had the highest dietary intake of energy, fat and saturated fatty acids ( p < 0.05), and also had a high prevalence of obesity, smoking and sedentary lifestyle. Conclusion: We have demonstrated that there is a high prevalence of coronary risk factors and poor dietary intake within a Saudi male population, and that dietary factors are associated with serum sICAM-1 and ceruloplasmin but not hsCRP concentrations in this group. D 2005 Elsevier B.V. All rights reserved. Keywords: Inflammatory markers; C-reactive protein; Soluble adhesion molecules; Ceruloplasmin; Coronary risk factors; Saudi males; Dietary fatty acids

1. Introduction There is substantial evidence that coronary heart disease is a chronic inflammatory disease [1], and it has been proposed that inflammatory markers such as C-reactive * Corresponding author. Faculty of Medicine, King Abdul Aziz University, PO Box 12713, Jeddah 21483, Kingdom of Saudi Arabia. Tel.: +966 2 6644444x25242; fax: +966 2 66 434 99. E-mail address: [email protected] (E.M. Alissa). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2005.09.007

protein (CRP), interleukin-6 (IL-6) and soluble adhesion molecules such as intercellular adhesion molecule-1 (ICAM-1) may provide additional information in the assessment of coronary risk [2,3]. Prospective epidemiological data have generally supported this view, demonstrating an association between several inflammatory markers and cardiovascular events in apparently healthy middle-aged men [4], in women [5], in higher-risk middle-aged men [6], in the elderly with subclinical disease [7], and in patients with atherosclerosis

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

[8] although the data have not been totally consistent [9] and the use of CRP as a risk marker is not universally accepted [10]. Acute coronary events have been attributed to plaque rupture, a condition that may be associated with a local inflammatory response within the artery wall [1]. Hence it is also possible that serum inflammatory markers are predictors of clinical end-points in patients with established atherosclerosis [11,12]. In this context, CRP directly increases the endothelial expression of monocyte chemoattractant protein1 [13] and adhesion molecules [14]. The latter exist in a soluble form that can be measured in the circulation. Thus it has been suggested that CRP may have a direct proatherogenic role by disturbing endothelial function and promoting the formation of early atherosclerotic lesions. Several serum markers have been identified that may be used to assess different aspects of the inflammatory response: CRP is a sensitive acute-phase protein, produced by the liver in response to circulating inflammatory cytokines such as IL-6 and TNFa. Intercellular adhesion molecule-1 (ICAM-1) is expressed on leucocytes, smooth muscle cells and endothelial cells. It is shed from the cell surface and appears in serum as sICAM-1 and is probably an indicator of macrophage and endothelial cell activation. Ceruloplasmin is the major copper-containing plasma protein; it is an acute-phase reactant, elaborated by the liver. It has previously been reported to predict CHD in several prospective studies [15,16] and to be associated with coronary risk [17]. Ceruloplasmin is also an important antioxidant. However, previous epidemiological data are difficult to interpret because measurements were made in patients with a previous MI, which may have affected trace element levels. Changes in plasma copper and ceruloplasmin can be nonspecific secondary manifestations of inflammation accompanying atherosclerosis. A major difficulty in interpreting many of the studies relating it to atherosclerosis is the fact that plasma ceruloplasmin levels are often increased in the presence of acute or chronic inflammatory disease [18]. Hence, it is unclear whether increased ceruloplasmin is an independent coronary risk factor or merely an indicator of chronic inflammation. Algorithms currently used to estimate coronary risk, such as the Framingham risk score (FRS) and the Prospective Cardiovascular Munster in Europe (PROCAM) score do not take inflammatory markers into account. However, the FRS has been shown to be related to serum high sensitivity (hs)CRP levels [19]. CRP levels might also be associated with other classical determinants of cardiovascular risk. Similarly, intercellular adhesion molecules could be used as a marker of sub-clinical atherosclerosis since the focal expression of adhesion molecules seems to be mediated in part by modified lipoproteins or their constituents [20]. Most of the research on coronary risk assessment has been undertaken in western, Caucasian populations, and the relevance of this work to other populations has been questioned [21].

289

Dietary fatty acids have been shown to modulate lipoprotein concentrations [22,23], and immune and endothelial function [24,25]. There is considerable evidence that a diet enriched with u`-3 fatty acids protects against atherosclerosis [23]. It has also been reported that diets rich in monounsaturated fatty acids are as effective as diets rich in polyunsaturated fatty acids in lowering LDL-C [26]. Several mechanisms for their beneficial effects have been hypothesized including anti-adhesion, anti-arrhythmic, and lipoprotein altering effects [27]. Fatty acids may also contribute to atherosclerosis by altering cellular functions [28], which may be related to changes in cellular membrane fatty acid composition [29]. Because the lipid composition of plasma and tissues is closely related to dietary fat intake [30], exposure of endothelial cells to individual fatty acids can be directly influenced by the types of fatty acids in the diet [25]. However, concerns still remain with respect to the potential for increased lipid peroxidation following long-chain polyunsaturated N-3 fatty acids, particularly eicosapentaenoic acid and docosahexaenoic acid [31]. We investigated the relationship between some of the known serum inflammatory markers and coronary risk factors in Saudi males, who were without established coronary heart disease. We have also determined whether self-reported dietary fat intake was associated with these inflammatory markers.

2. Methods 2.1. Subjects One hundred and forty male subjects without clinically evident coronary disease were recruited from the outpatients’ departments of King Abdul Aziz University Hospital and The King Fahad Armed Forces Hospital, Jeddah, Kingdom of Saudi Arabia. They were categorized into age tertiles. On direct questioning, the subjects gave no history of symptoms of CHD, and had no personal history of acute coronary syndrome documented in their medical notes. They were aged between 16 and 87 y. Exclusion criteria included: established renal or hepatic disease, vascular disease (i.e., peripheral vascular disease, cerebrovascular disease), malignancy or those on treatment with statins, antioxidants or aspirin. Fasting blood samples were taken after an overnight fast. Venous blood samples were taken from an antecubital vein and placed into plain, or heparinized tubes. Tubes were centrifuged at 3000 g for 10 min. The serum obtained was separated and frozen at 80 -C until the time of analysis. Subjects were asked to complete an interviewer administered questionnaire, concerning their demographic characteristics, including age, health history, lifestyle habits and diet. The local ethics committee of the hospital approved the study.

290

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

2.2. Demographic and anthropometric information Age, physical activity, smoking habit and family history of heart disease and diabetes were obtained by questionnaire. Smoking habit was categorized as either non-smoker, or current and ex-smokers. Physical activity was graded by the participant according to the number of episodes of exercise undertaken per week and were categorized as active ( 3 times/week) or inactive (< 3 times/week) according to the recommendations of the American Heart Association consensus statement on primary prevention of coronary diseases and from the U.S. Surgeon General’s report [32]. A positive family history of heart disease was defined by a first degree relative with a myocardial infarction, or cardiac death before the age of 55 y. Height, weight, systolic and diastolic blood pressure were measured by routine methods. The presence of hypertension was screened for as part of routine clinical practice and defined as a systolic blood pressure > 140 mm Hg, and or a diastolic blood pressure >90 mm Hg, or current use of antihypertensive therapy [33]. Normal weight was defined as a body mass index (BMI) < 25, overweight as a BMI 25 –29.9, and obese as >30 kg/m2 [34]. Dyslipidemia was defined as total cholesterol level  5.2 mmol/l, LDL-C  3.36 mmol/l, HDL-C < 1.04 mmol/l [35]. Diabetes was defined as a history of diabetes, with a documented, or measured fasting glucose > 7 mmol/l, or treatment with either insulin or oral hypoglycemic agents [36]. 2.3. Assessment of coronary risk The FRS was derived from the NCEP ATP III algorithm, which is based on 6 coronary risk factors; age, gender, total cholesterol, HDL-C, systolic blood pressure (treated or untreated), and smoking habit [37]. An alternative risk prediction equation has been derived using data from the PROCAM study; this uses LDL-C rather than total cholesterol, and additionally includes triglycerides, the presence or absence of diabetes mellitus and family history of CHD [38]. 2.4. Routine laboratory measurements Glucose was measured enzymatically by a routine glucose oxidase method. Total cholesterol and triglycerides were measured enzymatically by a colorimetric end-point method. HDL cholesterol was measured using a phosphotungstate magnesium precipitation method. LDL was calculated using the Friedewald formula in samples where the triglycerides were <4 mmol/l. 2.5. Serum inflammatory markers CRP was measured using a high sensitivity sandwich immunoassay, using a commercial kit (DRG, Germany).

Serum sICAM-1 was measured using an immunoassay kit from R&D Systems (R&D Systems, Germany). Analytical sensitivity was 0.1 mg/l and < 0.35 ng/ml for hsCRP and sICAM-1, respectively. Intra- and inter-assay CVs were < 10% and <20%, respectively, for both analytes. Ceruloplasmin was estimated as its ferroxidase activity using odianisidine as a substrate [39]. 2.6. Assessment of dietary intake Dietary intake over the previous year was assessed using a previously validated semi-quantitative food frequency questionnaire (FFQ) [40,41]. The FFQ asked about the frequency of intake of 94 food items and beverages. Food items were classified into 11 categories: grains (5 items), bread (5 items), fruits (11 items), vegetables (19 items), legumes (4 items), nuts (4 items), meats, poultry and fish (12 items), dairy products (10 items), fat and oil (3 items), soft drinks (6 items), miscellaneous (12 items) and junk food (3 items). Subjects were asked to select either per day, per week, per month or almost never as a denominator and then to enter frequency of use. The nutrient database used was based on UK food composition tables [42] together with food composition tables for use in East Asia and the U.S. handbook of food composition [43]. The composition of traditional local foods, not included in the above tables, was derived from another local study [44]. The estimated dietary intake of all nutrients was calculated in terms of percentage recommended nutritional intake (%RNI for UK adults) for each individual, as there are no published data for a Saudi population. The most recent version of the United Kingdom Dietary Recommended Values (DRVs) [42] was used to standardize the pattern of nutrient intake. 2.7. Statistical analysis Data are presented as mean and standard deviation for normally distributed data, or as median and interquartile ranges for non-parametric data. Variables that showed skewed distribution were log transformed before analyses and then back transformed to their natural units for presentation. Statistical analyses were performed using ANOVA test for normally distributed parameters or Kruskall– Wallis test for non-normally distributed parameters to compare mean values of repeated measures of non-normally distributed parameters as well as to assess the relationship between 10-y risk of coronary disease and serum inflammatory marker concentrations. Associations between FRS and PROCAM scores as well as the individual risk factors and individual serum inflammatory markers were also assessed using Pearson’s and Spearman’s correlation coefficients. The association between fat intake with these inflammatory markers was assessed by Pearson’s correlation coefficient. Stepwise multiple regression analysis was used to model the association between each inflammatory marker with all independ-

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

ent variables with p value up to 0.1 to demonstrate their contribution to the inflammatory marker serum level. A p < 0.05 was considered statistically significant. All statistical analysis was carried out with SPSS, ver. 11.5 software.

3. Results 3.1. Demographic data The demographic and anthropometric data for the subjects are shown in Table 1. The frequency of coronary risk factors; hypertension, diabetes mellitus and obesity increased with age tertiles ( p < 0.05). There was no clear distinction between age groups for smoking habit and physical activity. However a significantly higher proportion of the young and middle aged group had a positive smoking habit compared to the older group of subjects ( p < 0.0001). A large proportion (approximately 30%) of the combined group of middle and older individuals had a history of diabetes mellitus. 3.2. Biochemical data The biochemical data for the groups segmented by age are shown in Table 2. A large proportion of all age groups had a

291

fasting serum total and LDL-cholesterol above NCEP ATP III guidelines. The middle age group appeared to have the worst overall lipid profile, with higher mean LDL cholesterol and triglycerides. Fasting glucose levels rose with age, consistent with the high frequency of diabetes in the oldest group. Although median serum concentrations of the inflammatory markers, hsCRP, ceruloplasmin and sICAM1 were all lowest for the youngest age group of subjects, the differences did not reach statistical significance ( p > 0.05). 3.3. Coronary risk score The proportion of subjects with a high or moderate risk score rose with age group, as might be expected, as both algorithms are dependent on age and other age-related conditions (see Table 3). The FRS and PROCAM coronary risk scores were highly correlated for the age groups combined ( p < 0.0001). 3.4. Dietary data Dietary total fat, saturated fat and cholesterol were all highest in the youngest age group ( p < 0.05) (Table 4). Total calorie intake decreased with age group; the proportion of subjects consuming excessive or adequate dietary calories was highest in the youngest group of subjects. This may be

Table 1 Demographic and anthropometric characteristics of the subjects investigated divided into tertiles by age Variable

n Age (y) Height (cm) Weight (kg) Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Hypertensive, n (%) Dyslipidemia, n (%) Diabetics, n(%) Body mass index, n (%) Normal Overweight Obese Family history, n (%) Heart disease Diabetes mellitus Smoking status, n (%) Never Former Current (<20 cigarette) Current (20 cigarette) Physical activity, n (%) <3 times/week 3 times/week

Age groups (y)

p

30

31 – 48

49

46 23.4 T 0.6 169.9 T 0.9 75.8 T 2.5 26.3 T 0.8 122.4 T 1.5 78.8 T 1.1 10 (22) 37 (80) 0 (0)

47 38.3 T 0.7# 167.9 T 1.2 79.6 T 2.3 28.1 T 0.7‘ 127.9 T 2.0 83.1 T1.3‘ 15 (32) 44 (94) 14 (30)

47 61.5 T 1.6l,. 168.2 T 1.2 81.3 T 2.2 28.8 T 0.8* 125.3 T 3.2 79.4 T 1.6 24 (51) 44 (94) 17 (36)

23 (50) 13 (28) 10 (22)

11 (23) 22 (47) 14 (30)

14 (30) 12 (25) 21 (45)

<0.05

11 (24) 25 (54)

12 (26) 25 (53)

10 (21) 17 (36)

NS NS

30 (65) 4 (9) 4 (9) 8 (17)

19 (40) 6 (13) 7 (15) 15 (32)

27 (57) 15 (32) 1 (2) 4 (9)

<0.001

13 (28) 33 (72)

14 (30) 33 (70)

12 (25) 35 (75)

NS

<0.0001 NS NS <0.05 NS <0.05 <0.05 NS <0.0001

Numeric data are presented as mean T SEM for normally distributed data and categorical data as number and percentage. Categorical data were compared by v 2 test, continuous variables were compared by Kruskal – Wallis test. NS: not significant, *p < 0.05 (30 vs. 49 y groups), lp < 0.001 (30 vs. 49 y groups), ‘p < 0.05 (30 vs. 31 – 48 y groups), #p < 0.001 ( 30 vs. 31 – 48 y groups), p < 0.05 (31 – 48 vs. 49 y groups), .p < 0.001 (31 – 48 vs. 49 y groups).

292

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

Table 2 Biochemical characteristics of the subjects investigated divided into tertiles by age Variable

Age groups (y)

p

30

31 – 48

49

n

46

47

47

Total cholesterol [mmol/l] Total cholesterol 5.2 mmol/l, n (%) Triglycerides [mmol/l] Triglycerides 1.7 mmol/l, n (%) HDL-C [mmol/l] HDL-C <1.04 mmol/l, n (%) LDL-C [mmol/l] LDL-C 3.36 mmol/l, n (%) Atherogenic index (TC / HDL) Glucose [mmol/l] (Whole group) (Diabetics) (Non-diabetics) Hs-C reactive protein (mg/l) sICAM-1 (ng/ml) Caeruloplasmin (U/l)

5.38 T 0.19 25 (54) 1.14 (0.86 – 1.5) 9 (20) 1.45 T 0.09 14 (30) 3.67 T 0.21 26 (57) 4.40 T 0.32 5.37 (4.9 – 5.8)

6.01 T 0.20 31 (66) 1.55 (0.95 – 2.5)‘ 21 (45) 1.49 T 0.09 12 (26) 4.2 T 0.19 33 (70) 4.6 T 0.29 5.80 (5.2 – 7.2)#

5.69 T 0.23 27 (57) 1.34 (0.77 – 2.1) 16 (34) 1.29 T 0.07 12 (26) 4.1 T 0.22 31 (66) 4.9 T 0.29 6.10 (5.4 – 9.3)l

– 5.37 (4.9 – 5.8) 1.83 (1.1 – 3.7) 211.6 (158.3 – 329.8) 33.1 (23.0 – 42.8)

9.2 (7.9 – 12.1) 5.46 (5.2 – 5.9) 2.21 (1.0 – 3.0) 219.1 (161.1 – 328.7) 35.6 (18.8 – 60.6)

10.2 (8.6 – 12.0) 5.5 (5.2 – 6.0) 2.12 (1.4 – 3.2) 219.5 (160.9 – 288.6) 36.9 (23.8 – 70.0)

NS NS <0.05 <0.05 NS NS NS NS NS <0.0001 1 NS NS NS NS NS

Numeric data are presented as mean T SEM for normally distributed data and as median (interquartile range) for non-normally distributed data. Continuous variables were compared by Kruskal – Wallis test. HDL-C: high density lipoprotein cholesterol, LDL-C: low density lipoprotein cholesterol, NS: not significant, sICAM-1: soluble adhesion molecule-1, lp < 0.001 (30 vs.  49 y groups), ‘p < 0.05 (30 vs. 31 – 48 y groups), #p < 0.001 (30 vs. 31 – 48 y groups).

explained in part by the lower proportion of these individuals on a diet (9%). The mean intake of total fat and cholesterol was above recommended values for all age groups. Subjects in the youngest tertile for age had the lowest PUFA : SFA ratio of all age groups ( p < 0.05). 3.5. Relationship between inflammatory markers and coronary risk There was no significant difference in median hsCRP, sICAM-1 or ceruloplasmin between the age groups (Table 2). The significant relationships between inflammatory markers and individual coronary risk factors are shown in Table 5. Other correlations did not attain significance. In the group as a whole, serum ceruloplasmin concentrations were positively correlated with FRS, age, hypertension, intake of SFA and MUFA. When smoking habits were dichotomized into yes/no, serum CRP levels were significantly higher Table 3 Risk factor scores of the subjects investigated divided into tertiles by age Variable

FRS Low, n (%) Moderate, n (%) High, n (%) Total PROCAM Low, n (%) Moderate, n (%) High, n (%) Total

Age groups (y) 30

31 – 48

49

46 0 0 46

(100) (0) (0) (100)

39 8 0 47

(83) (17) (0) (100)

19 23 5 47

(40) (49) (11) (100)

45 1 0 46

(98) (2) (0) (100)

45 2 0 47

(96) (4) (0) (100)

20 15 12 47

(43) (32) (26) (100)

FRS: Framingham risk score.

among smokers ( p < 0.05). Serum sICAM-1 was negatively associated with PROCAM score ( p < 0.05) as well as with total cholesterol / HDL cholesterol ratio ( p < 0.001) in the group as a whole. Among non-diabetic subjects, sICAM-1 was positively associated with HDL cholesterol (r = 0.36, p < 0.0001), whereas among diabetic subjects it was negatively associated with HDL cholesterol (r = 0.397, p < 0.05). 3.6. Multivariate analysis Stepwise multiple regression analysis was used for the data from all 140 individuals to identify the independent determinants of serum ceruloplasmin concentrations. All variables that showed a univariate relationship with serum ceruloplasmin ( p value < 0.1) were entered into the model. These included age, FRS and dietary saturated and monounsaturated fatty acids. Only age ( p < 0.01) and dietary SFA ( p < 0.05) remained as independent predictors of serum ceruloplasmin, explaining 4.3% and 3.6% of the variation in serum ceruloplasmin concentrations, respectively. Using a similar approach for serum sICAM-1 concentrations, the variables used for the multiple regression model included HDL-C, LDL-C, total cholesterol / HDL ratio, PROCAM score, and dietary saturated fat. In this model, 6.5% of the variation in serum sICAM-1 level was explained by the total cholesterol / HDL-C ratio ( p = 0.001).

4. Discussion The inflammatory nature of the atherosclerotic plaque has been well described and its potential importance in

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

293

Table 4 Dietary characteristics of the subjects investigated divided into tertiles by age Nutrient

RNI

Age groups (y)

n On a diet, n(%) Energy (kcal)

Total fat (g) % of energy Cholesterol (mg) Categories of cholesterol intake, n (%) SFA (gm) % of energy MUFA (g) % of energy PUFA (gm) % of energy P : S ratio

2755 2550 2380 2100 30%

(15 – 18) (19 – 59) (60 – 74) (75+)

<200 mg  200 mg 10% 10% 10% NA

p

30

31 – 48

49

46

47

47

4 (9) 2188.7 T 66.2l

11 (23) 2010.6 T 64.7

10 (21) 1840.0 T 62.8

NS <0.001

92.6 T 3.01* 38.3 T 0.72 313.2 T 15.5l 2 (4) 44 (96) 32.7 T 1.3 ‘,l 15.0 T 0.40 31.8 T 1.0 14.7 T 0.34 18.4 T 0.85 8.4 T 0.32 0.58 T 0.03

85.0 T 3.69 38.1 T1.02 272.1 T18.6 15 (32) 32 (68) 27.8 T 1.64 13.7 T 0.55 30.1 T1.37 15.1 T 0.48 18.2 T 0.83 9.1 T 0.32 0.71 T 0.04‘

79.5 T 3.59 38.5 T 0.88 238.1 T14.9 21 (45) 26 (55) 26.0 T 1.42 13.9 T 0.51 28.0 T 1.29 15.1 T 0.38 17.2 T 0.96 9.3 T 0.40 0.72 T 0.05*

<0.05 NS <0.05 <0.001 <0.05 NS NS NS NS NS < 0.05

Numeric data are presented as mean T SEM and categorical data as number and percentage. Categorical data were compared by v 2 test, continuous variables were compared by Kruskal – Wallis test. MUFA: monounsaturated fatty acids, NS: not significant, P : S ratio: polyunsaturated fatty acids: saturated fatty acids ratio, PUFA: polyunsaturated fatty acids, RNI: reference nutrient intake, SFA: saturated fatty acids, *p < 0.05 (30 vs. 49 y groups), lp < 0.001 (30 vs. 49 y groups), ‘p < 0.05 ( 30 vs. 31 – 48 y groups).

determining plaque rupture and the consequential vascular complications has received considerable attention over recent years [45]. It has been proposed that serum markers of inflammation, such as hsCRP may partially reflect inflammatory events within the plaque, and hence may be predictive of outcome [14]. This is supported by epidemiological studies that have consistently reported increased coronary risk among subjects with high serum levels of inflammatory markers [20,46,47]. Although this increased risk has been reported to be independent of other risk factors [46], other investigators have reported a positive relationship between coronary risk score and CRP concentrations [5]. Most previous studies have been undertaken in industrialized Caucasian populations, and it is difficult to be certain that these data are applicable to non-Caucasian subjects living in other geographical locations. The predictive power of serum inflammatory markers is likely to be Table 5 Correlation coefficients between classical CVD risk factors and serum inflammatory markers in the subjects investigated Serum inflammatory marker

Risk factor

sICAM – 1 (ng/ml)

PROCAM score TC/HDL-C SFA Age Hypertension FRS SFA MUFA

Caeruloplasmin (U/l)

r

p 0.183 0.268 0.168 0.224 0.190 0.174 0.175 0.167

0.030 0.001 0.047 0.008 0.024 0.039 0.039 0.049

FRS: Framingham risk score, HDL-C: high density lipoprotein cholesterol, MUFA: monounsaturated fatty acids, SFA: saturated fatty acids sICAM-1: soluble adhesion molecule-1.

altered by prevalent low-grade infection in these populations. It is also possible that diet and other lifestyle factors will have an impact on the strength of any association between serum inflammatory marker concentrations and coronary risk. Dietary fatty acids have been reported to affect both coronary risk [48] and immune responses [24,25]. In this present study we investigated the relationship between classical coronary risk factors, dietary fats and serum inflammatory markers in a male population sample from the Middle East. It is clear that in this population, as elsewhere in the Middle East, there has been a rapid increase in the prevalence of coronary risk factors, including obesity [49], diabetes mellitus [50], hypertension [51] and dyslipidemia [52]. A worrying feature within our population sample is the very high prevalence of obesity, positive smoking history and inactivity. Compounding this cluster of high-risk lifestyle characteristics is a poor diet, with a high intake of total and saturated fat, and cholesterol. These dietary features were significantly worse in the youngest tertile than for the older age groups, and may be related to the adoption of western-style dietary habits. 4.1. Serum inflammatory markers and coronary risk In our sample of subjects, the majority had low coronary risk scores (> 74% using either scoring system), and few had high scores (< 9% using either score). Although this may reflect the population from which the sample was drawn, it may make it difficult to detect weak correlations between risk scores and serum inflammatory markers. In general, the coronary risk scores calculated using the FRS or PROCAM

294

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

algorithms categorized subjects in a similar way as each other in a male urban Saudi population and both charts showed good agreement in classifying subjects correctly in the 3 categories (low, mid, high) as shown by j statistic of a value of 0.536, p < 0.001 (not shown in the manuscript). However, individual differences in risk calculated by these methods may arise due to the different components of the algorithm; hypertriglyceridemia was highly prevalent in our sample (33%), as was diabetes mellitus (22%). The differences in plasma triglycerides levels and the presence of diabetes mellitus may explain the individual differences in categorization between FRS and PROCAM. Of the serum inflammatory markers, only serum ceruloplasmin showed a gradual, non-significant rise in median concentrations with increasing category of risk. It has been reported that HDL inhibits TNFa and IL-1hinduced expression of cellular adhesion molecules in vitro [53]. It may therefore be expected that sICAM-1 and HDL concentrations in serum would be inversely related, while sICAM-1 and the total cholesterol / HDL ratio would be positively related. However, we found a negative association between serum sICAM-1 and total cholesterol / HDL-C ratio ( p < 0.05). This could be due to the fact that 75% of our study population was < 50 y whereas the other studies have been performed on elderly subjects [54]. In these studies serum sICAM-1 levels were shown to be inversely related to HDL-C in individuals with low HDL-C but not in those with normal or high HDL-C [54]. In our study, less than one third of the population had HDL-C levels < 1.04 mmol/l, many of whom were diabetics. Although our finding of a negative association between sICAM-1 and HDL cholesterol amongst the diabetics subjects ( p < 0.05) is consistent with other studies [55] we found a positive association between HDL and sICAM-1 among the nondiabetic subjects, and this remained significant even after adjustment for potential confounding factors ( p < 0.001). Ceruloplasmin concentrations were positively related to FRS (Table 5). Several prospective studies have indicated that serum ceruloplasmin concentrations may be an independent risk factor for CVD [16,17]. The reasons for this are likely to be complex, but it is possible that factors contributing to CVD risk such as diabetes mellitus, obesity, dyslipidemia and hypertension, are associated with a lowgrade inflammation that may be associated with high serum ceruloplasmin concentrations. Studies in western, Caucasian populations have consistently shown that serum hsCRP concentrations are associated with coronary risk score [56]. Increased levels of hsCRP predict future myocardial infarction independently of other cardiovascular risk factors in caucasian populations [4], and it has also been advocated that measurement of hsCRP in addition to traditional risk factors may improve the prediction of CVD [57,58]. However there is not complete consensus on this point [10]. We were unable to show a positive association between calculated coronary risk and hsCRP in our study population. There may be two

explanations for this; either the coronary risk calculations may not be valid in our population, or serum hsCRP concentrations may be being influenced by other factors, such as low-grade endemic infection. The high frequency of diabetes mellitus in our Saudi sample may also be associated with an underestimate of calculated risk as previously reported by Cushman et al. [59]. A cohort study will be necessary to assess the validity of the coronary risk scores in the Saudi population, and the relationship between serum inflammatory markers and clinical end-points. Although there was no significant relationship between serum sICAM-1 levels and FRS, serum sICAM-1 levels were found to be negatively correlated with PROCAM score. A previous study undertaken in an elderly population found a positive relationship between sICAM-1 levels and risk of future coronary events [54]. PROCAM risk charts take diabetes mellitus into account whereas the Framingham risk chart does not. We found a particularly strong negative correlation between sICAM-1 and HDL-C (r = 0.397, p < 0.05) among diabetic subjects in our population sample. Similar observations have been made in other studies on adhesion molecules in diabetic patients in whom low serum HDL concentrations are frequently seen [55,60]. Previous studies have reported a positive relationship between smoking habit and serum soluble adhesion molecules [47]. The reported effects of smoking on other inflammatory markers have been less consistent [60 – 62]. We were unable to identify any significant relationship between smoking habit and any of the inflammatory markers. Smokers had significantly higher CRP levels than non-smokers in the population as a whole. The positive relationship between inflammatory markers and obesity reported previously [63] has been explained by the fact that adipose tissue is a source of inflammatory cytokines, such as IL-6 and TNFa, which induce an acute phase response by the liver [53]. We were unable to demonstrate any significant independent relationship between BMI and any of the inflammatory markers in our population of Saudi males. 4.2. Serum inflammatory markers and dietary fat There has been great interest in the effects of dietary fatty acids on lipoprotein metabolism [24,25]. However, less attention has been paid to their effect on the immune system [12,33]. Dietary total fat, saturated fat and cholesterol were significantly higher in the youngest age group, and there was a gradual decrease in their intake with advancing age. Dietary PUFA intake was high in the middle tertile for age. Serum hsCRP concentrations were also highest in this group. This is consistent with the possibility that dietary PUFA may enhance the inflammatory response, having a potentially deleterious effect in inflammatory conditions including atherosclerosis. Thus, increased levels of acutephase proteins may reflect response to infection or inflammation.

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

Univariate analysis demonstrated that serum sICAM-1 levels were inversely related to dietary SFA. However multivariate analysis showed that this relationship could be explained by other factors entered into the model. Serum ceruloplasmin levels were positively related to dietary SFA and MUFA intake, and the former remained significant following multivariate analysis. This is consistent with previous reports on the pro-inflammatory effects of dietary MUFA [24] and SFA [22].

5. Limitations Ours was a cross-sectional, rather than a prospective study. Cross-sectional studies rely on the use of risk estimates rather than clinical events, and are therefore sensitive to deviations from the model parameters. Both Framingham and PROCAM studies are based on data from Caucasians with western lifestyles; it is therefore uncertain whether the models are suitable for non-Caucasian populations [37]. The Framingham algorithm does not consider serum triglycerides and relatively few diabetic patients were included in the Framingham cohort. The PROCAM algorithm was developed subsequently and attempts to take account of diabetes mellitus and HDL levels; however its use is limited to men aged 40 –65 y. As eluded to above, inflammatory markers may lose their predictive power in populations in which there is a state of persistent endemic low-grade infection. This may be the case in the western province of Saudi Arabia, where subclinical infections may be prevalent due to the large numbers of pilgrims visiting Jeddah and Mecca from different parts of the world.

6. Conclusions The relationship between two different coronary risk score and serum hsCRP and sICAM-1 was not significant in a male Saudi population, as has been shown in some caucasian populations, although a significant positive association was seen between FRS and ceruloplasmin concentrations. Within the sample of young Saudi males there was a high frequency of poor lifestyle habits, including positive smoking habit, obesity and high dietary fat. These are likely to result in an even greater prevalence of vascular disease in future decades.

Acknowledgements Eman Alissa was supported by a scholarship from the joint programme of the King Abdul Aziz University. The authors wish to thank Professor Ibrahim Ezzeldin and Dr. Bakor Bin Sadiq from King Faisal Specialist Hospital and Research Centre for their suggestions on statistical analysis.

295

References [1] Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 1999;340(2):115 – 26. [2] Ridker P, Rifai N, Pfeffer M, et al. Long-term effects of pravastatin on plasma concentrations of CRP. Circulation 1999;100:230 – 5. [3] Luc G, Arveiler D, Evans A, et al. Circulating intercellular adhesion molecule-1 and vascular adhesion molecule-1 and incident coronary heart disease: the PRIME study. Atherosclerosis 2003; 170(1):169 – 76. [4] Ridker P, Cushman M, Stampfer M, et al. Inflammation, aspirin, and the risk of CVD in apparently healthy men. N Engl J Med 1997; 336:973 – 9. [5] Ridker P, Hennekens C, Buring J, et al. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836 – 43. [6] Kuller L, Tracy R, Shaten J, et al. Relationship of CRP and CHD in the MRFIT nested case – control study. Am J Epidemiol 1996; 144:537 – 47. [7] Tracy R, Lemaitre R, Psaty B, et al. relationship of CRP to risk of CVD in the elderly: results from the cardiovascular health study and the rural health promotion project. Arterioscler Thromb Vasc Biol 1997;17:1121 – 7. [8] Liuzzo G, Biasucci L, Gallimore J, et al. The prognostic value of CRP and serum amyloid A protein in severe unstable angina. N Engl J Med 1994;331:417 – 24. [9] Danesh J, Whincup P, Walker M, et al. Low-grade inflammation and coronary heart disease: prospective study and updated meta-analysis. Br Med J 2000;321(7255):199 – 204. [10] Levinson S. Poor predictive value of high sensitivity (hs-CRP) indicates need for reassessment. Clin Chem 2004;50(10):1733 – 5. [11] Wallen NH, Held C, Rehnqvist N, et al. Increased serum intercellular adhesion molecule-1 and vascular adhesion molecule-1 among patients with stable angina pectoris who suffer cardiovascular death or non-fatal myocardial infarction. Eur Heart J 1999;20(14):1039 – 43. [12] Blankenberg S, Luc G, Ducimetiere P, et al. IL-18 and the risk of coronary heart disease in European men: Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation 2003; 108(20):2453 – 9. [13] Pasceri V, Chang J, Willerson J, et al. Modulation of CRP mediated monocyte chemoattractant protein-1 induction in human endothelial cells by anti-atherosclerosis drugs. Circulation 2001;103:2531 – 4. [14] Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of Creactive protein on human endothelial cells. Circulation 2000; 102(18):2165 – 8. [15] Mantarri M, Manninen V, Huttunen J, et al. Serum ferritin and ceruloplasmin as coronary risk factors. Eur Heart J 1994;15: 1599 – 603. [16] Reunanen A, Knekt P, Aaran R. Serum ceruloplasmin level and the risk of MI and stroke. Am J Epidemiol 1992;136:1082 – 90. [17] Ferns G, Lamb D, Taylor A. The possible role of copper ions in atherogenesis: the Blue Janus. Atherosclerosis 1997;133:139 – 52. [18] Atanasiu R, Stea D, Mateescu M, et al. Direct evidence of ceruloplasmin antioxidant properties. Moll Cell Biochem 1998; 189:127 – 35. [19] Albert M, Glynn R, Ridker P. Plasma concentration of C-reactive protein and the calculated Framingham coronary heart disease risk score. Circulation 2003;108(2):161 – 5. [20] Caterina R, Basta G, Lazzerini G, et al. Soluble vascular adhesion molecule-1 as a biohumoral correlate of atherosclerosis. Arterioscler Thromb Vasc Biol 1997;17:2646 – 54. [21] Leaverton P, Sorlie P, Kleinman J, et al. Representativeness of the Framingham risk model for coronary disease mortality: a comparison with a national cohort study. J Chronic Dis 1987;40:775 – 84. [22] Keys A, Anderson J, Grande F. Serum cholesterol response to changes in the diet: IV. Particular saturated fatty acids in the diet. Metabolism 1965;14:776 – 87.

296

E.M. Alissa et al. / Clinica Chimica Acta 365 (2006) 288 – 296

[23] Mori T, Beilin L. Omega-3 fatty acids, blood lipids and cardiovascular risk reduction. Curr Opin Lipidol 2001;12:11 – 7. [24] Yaqoob P, Knapper JA, Webb DH, et al. Effect of olive oil on immune function in middle-aged men. Am J Clin Nutr 1998;67(1):129 – 35. [25] Tsimikas S, Philis-Tismikas A, Alexopoulos S, et al. LDL isolated from Greek subjects on a typical diet or from American subjects on an oleate-supplemented diet induces less monocyte chemotaxis and adhesion when exposed to oxidative stress. Arterioscler Thromb Vasc Biol 1999;19:122 – 30. [26] Mensik R, Katan M. Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low-density and high-density lipoprotein cholesterol in healthy women and men. N Engl J Med 1989;321:436 – 41. [27] Connor W. Importance of n-3 fatty acids in health and disease. Am J Clin Nutr 2000;71:171S – 5S [Suppl]. [28] Hodgkin D, Boucek R, Purdy R. Dietary lipids modify receptor- and non-receptor-dependent components of alpha 1-adrenoceptor-mediated contraction. Am J Physiol 1991;261:R1465 – 9. [29] Berlin E, Shapiro S, Kliman P. Influence of saturated and unsaturated fats on platelet fatty acids in cholesterol-fed rabbits. Atherosclerosis 1987;63:85 – 92. [30] Reaven P, Parthasarathy S, Grasse B, et al. Effects of oleate-rich and linoleate-rich diets on the susceptibility of LDL to oxidative modification in mildly hypercholesterolemic subjects. J Clin Invest 1993;91:668 – 76. [31] Nenseter M, Drevon C. Dietary polyunsaturates and peroxidation of LDL. Curr Opin Lipidol 1996;7:8 – 13. [32] US department of Health and Human Services. Physical activity and health: a report of the surgeon general. Atlanta, GA: Centers for disease control and prevention (CDC), National centers for chronic disease prevention and health promotion; 1996. [33] Hennekens C, Albert C, Godfried S, et al. Adjunctive drug therapy of acute myocardial infarction—evidence from clinical trials. N Engl J Med 1996;335:1660 – 7. [34] World Health Organization. Obesity, preventing and managing the global epidemic: report of a WHO consultation on obesity. Geneva, Switzerland’ World Health Organization; 1997. [35] Cleeman JI, Grundy SM, Becker D, et al. Executive summery of the 3rd report of the NCEP expert panel on detection, evaluation and treatment of high blood cholesterol in adults (ATP III). JAMA 2001;285:2486 – 97. [36] The expert committee on the diagnosis and classification of diabetes mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1997;20:1183 – 202. [37] Wilson P, D’Agostino R, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97(18):1837 – 47. [38] Assman G. Lipid metabolism disorders and coronary heart diseases. Munchen’ MMV-Medizin-Verl; 1993. [39] Schosinsky K, Lehmann H, Beeler M. Measurement of ceruloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride. Clin Chem 1974;20:1556 – 63. [40] Willett W. Nutritional epidemiology. New York’ Oxford University press; 1998. [41] Alissa EM, Bahjri S, Al-ama N, Ahmed WA, Starkey B, Ferns GA. Dietary vitamin A may be a cardiovascular risk factor in a Saudi population. Asia Pac J Clin Nutr 2005;14(2):137 – 44. [42] Dietary reference values for food energy and nutrients for the UK. Report of the panel on dietary reference values of the committee on medical aspects of food policy. Department of Health; 2003. [43] Food composition tables for use in East Asia. US department of health, Education and welfare & UN FAO; 1972.

[44] Abalkhail B. Hypercholesterolemia and susceptibility to coronary vascular disease among white and blue collars in Jeddah. Saudi Med J 1998;19(2):123 – 9. [45] Fuster V, Corti R, Fayad ZA, et al. Integration of vascular biology and magnetic resonance imaging in the understanding of atherothrombosis and acute coronary syndromes. J Thromb Haemost 2003;1(7):1410 – 21. [46] Ridker P, Hennekens C, Roitman-Johnson B, et al. Plasma concentration of soluble intercellular adhesion molecule-1 and risks of future myocardial infarction in apparently healthy men. Lancet 1998; 351(9096):88 – 92. [47] Rohde L, Lee R, Rivero J, et al. Circulating cell adhesion molecules correlate with ultrasound-based assessment of carotid atherosclerosis. Arterioscler Thromb Vasc Biol 1998;18:1765 – 70. [48] Lemaitre R, King I, Mozaffarian D, et al. n-3 polyunsaturated fatty acids, fatal ischemic heart disease, and nonfatal myocardial infarction in older adults: the cardiovascular health study. Am J Clin Nutr 2003;77:319 – 25. [49] Alsaif M, Hakim I, Harris R, et al. Prevalence and risk factors of obesity and overweight in adult Saudi population. Nutr Res 2002;22:1243 – 52. [50] El-Hazmi M, Warsy A, Al-Swailem A, et al. Diabetes mellitus and impaired glucose tolerance in Saudi Arabia. Ann Saudi Med 1996;16(4):381 – 5. [51] Al-Nozha M, Ali A, Osman A. Arterial hypertension in Saudi Arabia. Ann Saudi Med 1997;17(2):170 – 4. [52] Al-Nuaim A. Serum total and fractionated cholesterol distribution and prevalence of hypercholesterolemia in urban and rural communities in Saudi Arabia. Int J Cardiol 1997;58:141 – 9. [53] Yudkin J, Kumari M, Humphries S, et al. Inflammation, obesity, stress and CHD: is IL-6 the link? Atherosclerosis 2000;148:209 – 14. [54] Calabresi L, Gomaraschi M, Villa B, et al. Increased soluble adhesion molecules in subjects with low HDL cholesterol. Arterioscler Thromb Vasc Biol 2002;22:656 – 61. [55] Abe Y, El-Masri B, Kimball K, et al. Soluble cell adhesion molecules in hypertriglyceridemia and potential significance on monocyte adhesion. Arterioscler Thromb Vasc Biol 1998;18:723 – 31. [56] Strandberg T, Tilvis R. C-reactive protein, cardiovascular risk factors and mortality in a prospective study in the elderly. Arterioscler Thromb Vasc Biol 2000;20:1057 – 60. [57] Ridker P, Glynn R, Hennekens C. CRP adds to the predictive value of total and HDL-C in determining risk of first myocardial infarction. Circulation 1998;97:2007 – 11. [58] Ridker P, Cook S. Clinical usefulness of very high and very low levels of CRP across the full range of FRS. Circulation 2004;109:1955 – 9. [59] Cushman M, Arnold A, Psaty B, et al. CRP and the 10-year incidence of CHD in older men and women—the cardiovascular health study. Circulation 2005;112(1):25 – 31. [60] Rohde L, Hennekens C, Ridker P. Cross-sectional study of soluble intercellular adhesion molecule-1 and cardiovascular risk factors in apparently healthy men. Arterioscler Thromb Vasc Biol 1999; 19:1595 – 9. [61] Onat A, Sansory V, Yildirim B, et al. C-reactive protein and coronary heart disease in western Turkey. Am J Cardiol 2001;88:601 – 7. [62] Mendall M, Patel P, Ballam L, et al. CRP and its relation to cardiovascular risk factors: a population based cross sectional study. Br Med J 1996;312:1061 – 5. [63] Cignarelli M, DePergola G, Picca G, et al. Relationship of obesity and body fat distribution with ceruloplasmin serum levels. Int J Obes Relat Metab Disord 1996;20:809 – 13.