OUP user menu

Leukocyte count and fibrinogen are associated with carotid and femoral intima-media thickness in a risk population for diabetes

Theodora Temelkova-Kurktschiev , Carsta Koehler , Elena Henkel , Markolf Hanefeld
DOI: http://dx.doi.org/10.1016/S0008-6363(02)00547-3 277-283 First published online: 1 November 2002

Abstract

Objective: To investigate the relationship of the inflammatory parameters—leukocyte count and fibrinogen level—to the intima-media thickness (IMT) of the common carotid artery and the common femoral artery, as well as to a variety of risk factors within the metabolic syndrome in a risk population for diabetes. Methods: A total of 597 subjects were analyzed from the Risk factors in Impaired glucose tolerance for Atherosclerosis and Diabetes (RIAD) study, who were at risk for the development of type 2 diabetes. IMT of the common carotid and common femoral artery was determined by B-mode ultrasound. Leukocyte count and fibrinogen level, as well as various risk factors for atherosclerosis, were measured by established methods. Results: In univariate analysis, leukocyte count and fibrinogen level correlated significantly to carotid and femoral IMT. Leukocyte count was significantly correlated to body mass index, waist to hip ratio, blood pressure, plasma triglycerides, high-density lipoprotein cholesterol (inversely), fasting and postprandial plasma glucose, insulin and proinsulin, PAI(active), tPA and microalbuminuria, as well as to smoking and physical activity (inversely). Fibrinogen level was significantly correlated with body mass index, systolic blood pressure, plasma triglycerides, fasting plasma glucose, HbA1c, PAI(active), tPA and von Willebrandt factor, as well as with smoking and low physical activity. In multivariate analysis, leukocyte count was an independent determinant of the maximal carotid IMT and fibrinogen level of femoral IMT. Conclusions: Our study indicates that low-grade inflammation is correlated to IMT, as an indicator of early atherosclerosis, and is strongly associated to a variety of risk factors within the metabolic syndrome in a population at risk for type 2 diabetes.

Keywords
  • Atherosclerosis
  • Diabetes
  • Infection/inflammation
  • Leukocytes

Time for primary review 25 days.

1 Introduction

It is now established that atherosclerosis is a disease of a complex origin, in the initiation and progress of which multiple risk factors are involved, such as hypercholesterolemia, smoking, diabetes, hypertension etc. [1–3]. However, known risk factors do not account for all the cardiovascular events and therefore additional risk factors were suggested to be involved [4,5]. Thus, there is a considerable amount of evidence indicating an important role of inflammation in the pathogenesis of atherothrombosis [6–14]. Macrophage and T-cell infiltration is a major feature of the atherosclerotic plaque, especially at the sites of plaque rupture [8]. The level of systemic markers of inflammation, such as circulating C-reactive protein, amyloid A protein and other acute-phase reactants are elevated in patients with unstable angina pectoris and are an indicator of poor prognosis [15]. Moreover, many studies have convincingly demonstrated that chronic subclinical inflammation is a strong predictor of atherothrombotic events [9–14]. The Caerphilly and Speedwell collaborative heart disease studies [9] showed that fibrinogen and white blood cell (WBC) count are major risk factors for future ischemic heart disease. Also, in the Framingham study [11] the degree of elevation of WBC count within the normal range was found to be a marker for increased risk of cardiovascular disease (CVD). Among non-smoking men for each 1.0×109/L-cell difference in increment in WBC count, the CVD risk increased by 32% [11]. In the Physicians Health Study [14] the baseline plasma concentration of C-reactive protein was found to predict the risk of future myocardial infarction and stroke. Another inflammatory parameter, lipoprotein-associated phospholipase A2 was shown to be an independent predictor of coronary heart disease events in the WOSCOPS study [16].

Although inflammatory markers have been found to predict future atherothrombotic events, little is known so far about the role of chronic subclinical inflammation in the initial stages of atherogenesis. The Bruneck study [17] reported an increased occurrence/progression of atherosclerotic plaques, detected by ultrasound, among individuals with chronic infections who had a prominent inflammatory response. Recently, inflammation was shown to be associated with intima-media-thickening in dialysis patients [18]. The arterial intima-media thickness (IMT) is an accepted marker of early atherosclerosis, which is associated with a number of risk factors for atherosclerosis [19,20]. Carotid IMT was found to be increased in patients in the early diabetic stages and already in prediabetic subjects with impaired glucose tolerance [21,22]. Elevated levels of inflammatory markers were also found in type 2 diabetic patients with features of the metabolic syndrome [23]. Recently the National Cholesterol Education Program (NCEP) declared patients with diabetes being at the same risk as individuals with a documented coronary heart disease and regarded diabetes as a coronary heart disease risk equivalent [24]. So far there are no data available on the association of IMT and low-grade inflammation in a risk population for diabetes. Therefore, the aim of the present study was to investigate the relationship of the inflammatory parameters—leukocyte count and fibrinogen level—to IMT of the common carotid artery and the common femoral artery, as well as to a variety of risk factors within the metabolic syndrome in a risk population for diabetes.

2 Materials and methods

A total of 597 subjects (288 men and 309 women) were analyzed, who were consecutive participants of the RIAD study, a prospective survey on the Risk factors in IGT for Atherosclerosis and Diabetes. Details on study design have been previously published [21,22,25]. In brief, subjects (40–70 years of age) were examined, who were at risk for the development of diabetes, such as family history of type 2 diabetes, obesity and/or hyper/dyslipoproteinemia. Known diabetes, medication affecting glucose tolerance, liver and kidney diseases, thyroid gland functional disorders and acute infections were exclusion criteria. The investigation conforms with the principles outlined in the Declaration of Helsinki [26]. Basic characteristics of the examined subjects are shown in Table 1.

View this table:
Table 1

Basic characteristics

ParametersMean (S.D.)
N (men/women)597 (288/309)
Age (years)55.2 (8.0)
Body mass index (kg/m2)27.0 (4.3)
Waist to hip ratio0.89 (0.09)
HbA1c (%)5.6 (0.6)
Plasma glucose fasting (mmol/l)6.0 (0.9)
Plasma glucose 2 h in OGTT (mmol/l)7.6 (2.8)
Blood pressure systolic (mmHg)134.9 (18.3)
Blood pressure diastolic (mmHg)80.0 (9.9)
Total cholesterol (mmol/l)5.8 (1.1)
High-density-lipoprotein cholesterol (mmol/l)1.4 (0.4)
Triglycerides (mmol/l)a1.3 (1.0 to 2.1)
Albuminuria (mg/l)a5.5 (1.3 to 14.0)
  • a Median values and interquartile range.

Venous blood was drawn after an overnight fast of at least 10 h for the measurement of inflammation parameters and a variety of risk factors for atherosclerosis [21,22,25]. A standard oral glucose tolerance test (OGTT) was performed with 75 g glucose (Glucodex, Rougier, Chambly, Quebec, Canada). The ultrasound examination was conducted on the day of blood collection for laboratory analysis, so that both study participants and physician were unaware of the corresponding laboratory values.

2.1 Ultrasound measurement

B-mode ultrasound of the common carotid artery (CCA) and the common femoral artery (CFA) was performed with an Acuson 128XP computed sonography system using a 10 MHz linear array transducer, as published elsewhere [21,22,25]. In brief, the thickness of the intima-media complex was assessed as described by Pignoli [19]. For both vessels the intima-media thickness was determined only of the far wall of the artery, since it is known to be of a higher precision than that of the near arterial wall [27]. The reproducibility of the IMT measurement was found to be good, as previously published [25]. The patients were examined in a supine position. Measurements were conducted in plaque-free portions of the 10-mm linear segment proximal to the carotid bulb. For each patient two measurements were performed bilaterally and the values averaged, which presented the IMTmean of the CCA. In addition, the maximal IMT of the vessel was determined independent on the localization, visually judged by the observer. The IMT of the common femoral artery was assessed in its distal part, bilaterally in double measurements and the values averaged.

2.2 Laboratory examination

Venous blood was collected in EDTA monovettes and plasma immediately separated by centrifugation (4000 rev./min for 8 min at 4 °C). Fibrinogen was measured by the method of Clauss (Fibrinogen; Boehringer Mannheim, FRG; coefficient of variation −2.9 to 5.5%). Total leukocyte count was performed with standard techniques.

Plasma glucose was measured by the hexokinase method. HbA1c was examined by high-performance liquid chromatography on a Diamat analyzer (Bio-Rad Laboratories, Munich, Germany). Triglycerides and total cholesterol were measured enzymatically on a Ciba Corning Express Plus analyzer (Boehringer Mannheim). HDL cholesterol was determined after precipitation with dextran sulfate on a Ciba Corning Express Plus analyzer (Boehringer Mannheim).

Proinsulin was analyzed by highly specific enzyme immunoassay (DGR Instruments, Marburg, FRG). Specific insulin was also measured by enzyme immunoassay (Medgenix Diagnostics Fleurus, Belgium). Specific insulin showed no cross-reactivity to human proinsulin.

The concentration of active plasminogen activator inhibitor-1 (PAI-1) antigen was determined using commercially available enzyme immunoassay (Immuno AG, Heidelberg, FRG). Tissue plasminogen activator (tPA) was measured by enzyme immunoassay (TintElize; Biopool, Umea, Sweden) and von Willebrandt factor antigen by electroimmunoassay (Immuno AG). Urine was collected as fresh morning urine samples. Albuminuria was measured by nephelometry (Nephelometer BNII; Behring, Marburg, Germany).

2.3 Statistics

Data evaluation was conducted using the SPSS/PC+ programme. The distribution of values was assessed by the Kolmogorov–Smirnov test for homogeneity of variances. Data are presented as mean and standard deviation or median values and interquartile range for nonparametric variables. The correlation of inflammatory parameters to IMT and to atherosclerosis risk factors was assessed using Pearson or Spearman correlation coefficients, as appropriate. In addition, partial correlation after adjustment for age, sex and hematocrit was evaluated. IMT was assessed in trend by ANOVA in tertiles of leukocyte count and glucose tolerance stages. Multivariate analysis was conducted by stepwise multiple linear regression including risk factors for atherosclerosis, such as age, sex, body mass index, waist to hip ratio, blood pressure (systolic and diastolic), total cholesterol, triglycerides, high-density-lipoprotein cholesterol, Lp(a), plasma glucose (fasting and 2 h postchallenge), smoking, albuminuria, fibrinogen, leukocyte count, insulin (fasting and 2 h postchallenge), plasminogen activator inhibitor and von Willebrandt factor.

3 Results

In univariate analysis a highly significant correlation was found between total leukocyte count and IMTmean and IMTmax of the CCA (Fig. 1). Fibrinogen level correlated significantly to both IMT of the common carotid and the common femoral artery. This relationship remained significant after adjustment for age and sex, as well as after excluding subjects under lipid lowering treatment. To exclude the effect of nonspecific concentration phenomena through changes in haematocrit, the relationship to IMT was additionally adjusted for haematocrit, and remained significant.

Fig. 1

Correlation between leukocyte count (leuko) and fibrinogen level (fib) and intima-media thickness. IMTMAX, maximal IMT of the common carotid artery; IMTMEAN, average IMT of the common carotid artery; IMT_FC, IMT of the common femoral artery.

The two inflammatory parameters were highly significantly correlated to each other (Table 2). Leukocyte count was also significantly correlated to body mass index, waist to hip ratio, blood pressure, plasma triglycerides, high-density lipoprotein cholesterol (inversely), fasting and postprandial plasma glucose, insulin and proinsulin, PAI(active), tPA and microalbuminuria, as well as to smoking and physical activity (inversely). Fibrinogen level was significantly correlated with body mass index, systolic blood pressure, plasma triglycerides, fasting plasma glucose, HbA1c, PAI(active), tPA and von Willebrandt factor, as well as with smoking and low physical activity (Table 2). The significant correlations were confirmed after adjustment for age, sex and hematocrit.

View this table:
Table 2

Correlation of white blood cell count (WBC) and fibrinogen level to atherosclerosis risk factors

WBC (P)Fibrinogen (P)
Age−0.05 (NS)0.17 (<0.001)
Sex−0.05 (NS)0.08 (NS)
Smoking0.12 (0.006)0.14 (0.001)
Physical activity−0.11 (0.01)−0.15 (<0.001)
Body mass index0.28 (<0.001)0.22 (<0.001)
Waist to hip ratio0.16 (<0.001)0.05 (NS)
Blood pressure systolic0.11 (0.013)0.10 (0.02)
Blood pressure diastolic0.11 (0.013)0.07 (NS)
Total cholesterol−0.04 (NS)0.10 (0.06)
Triglycerides0.23 (<0.001)0.13 (0.005)
HDL cholesterol−0.17 (<0.001)0.01 (NS)
Fasting plasma glucose0.17 (<0.001)0.10 (0.03)
2 h pp plasma glucose0.26 (<0.001)0.08 (0.07)
Glycosylated haemoglobin A1c0.07 (NS)0.11 (0.01)
Specific insulin0.19 (<0.001)0.08 (0.06)
Proinsulin0.12 (0.007)−0.04 (NS)
Plasminogen activator inhibitor (active)0.13 (0.005)0.10 (0.04)
Tissue plasminogen activator0.14 (0.003)0.15 (0.001)
Fibrinogen0.26 (<0.001)
Von Willebrandt factor−0.05 (NS)0.11 (0.02)
Albuminuria0.11 (0.02)0.09 (0.06)

In Fig. 2 we show carotid IMTmax in tertiles of leukocyte count and stages of glucose tolerance. The three-dimensional presentation of IMTmax reveals that in all tertiles for leukocyte count a significant rise in trend is to be found parallel to the glucose tolerance stages. On the other hand diabetic patients from the highest tertile of leukocyte count showed a significantly increased IMTmax of the CCA in comparison to the diabetics of the lowest leukocyte count tertile. Similar results were observed for IMT mean. Fibrinogen level in the different glucose tolerance stages was not associated with significant IMT changes.

Fig. 2

Maximal intima-media thickness (IMT) of the common carotid artery (CCA) in mm in tertiles of leukocytes count and glucose tolerance stages. NGT, normal glucose tolerance; IGT, impaired glucose tolerance; DM, type 2 diabetes mellitus; *P<0.05 ANOVA in trend; **P<0.05 vs. DM of the lowest tertile for leukocytes count.

In multivariate analysis leukocyte count was an independent determinant of carotid IMTmax (Table 3a), but not of IMTmean, and fibrinogen level of femoral IMT (Table 3b).

View this table:
Table 3

(a) Determinants of IMTmax of the common carotid artery; (b) determinants of IMT of the common femoral artery

Variableβ-coefficientP
(a) IMTmax of the CCA  
Age0.322<0.001
Male sex0.230<0.001
Plasma glucose 2 h0.178<0.001
HDL-cholesterol−0.1240.013
Albuminuria0.1120.011
Leukocyte count0.1040.016
Total cholesterol0.1010.019
(b) IMT of the common femoral artery 
Age0.370<0.001
Male sex0.289<0.001
Lp(a)0.1940.002
Fibrinogen0.1520.013

4 Discussion

Type 2 diabetic patients were recently shown to be at the same risk for future myocardial infarction as non-diabetic subjects who already had one [28]. Even after taking into consideration major cardiovascular risk factors, such as cholesterol, smoking and blood pressure, diabetic subjects still exhibit 2–4 times higher cardiovascular disease mortality than non-diabetic individuals [3]. Furthermore, atherosclerosis was shown to start already in the early stages of diabetes and significant thickening of the intima-media complex of the common carotid artery was described in newly detected type 2 diabetes and in impaired glucose tolerance [21,22]. In this analysis we demonstrate for the first time that inflammatory parameters are independent determinants of early atherosclerosis, evaluated as increased IMT, in a population at a high risk for the development of type 2 diabetes.

In recent years it was found that markers of inflammation, such as C-reactive protein, leukocyte count, fibrinogen, and lipoprotein associated phospholipase A2 are strong predictors of cardiovascular events [6–14]. The increased IMT in a specific population of dialysis patients suggested that low-grade inflammation could be also involved in the initiation of atherosclerosis [18]. In the present work both carotid and femoral IMT are strongly correlated to inflammatory parameters (Fig. 1). Interestingly, in all levels of leukocyte count a significant rise of IMT was observed parallel to glucose intolerance and this was mostly expressed in the highest leukocyte count tertile. On the other hand, the increase of leukocyte count in diabetic subjects was associated with an additional significant rise in IMT.

Previously, inflammatory parameters were reported to be strongly correlated to various features of the metabolic syndrome [29–32]. Thus, white blood cell count was strongly related with plasma glucose, insulin, body mass index, blood pressure, plasma triglycerides, fibrinogen and high-density lipoprotein cholesterol (inversely) in healthy male individuals [31]. C-reactive protein was also shown to be associated with obesity, insulin resistance and endothelial dysfunction in healthy subjects [30,32]. In a non-diabetic population from the Insulin Resistance Atherosclerosis Study without clinical coronary artery disease, C-reactive protein, WBC count and fibrinogen were correlated with several components of the insulin resistance syndrome, as well as to insulin sensitivity, measured by a frequently sampled intravenous glucose tolerance test [29]. Our study confirms the previous findings that subclinical inflammation is an important part of the metabolic syndrome, since both leukocyte count and fibrinogen level were significantly correlated to the different parameters of this syndrome (Table 2).

Interestingly, white blood cell count and fibrinogen level were positively correlated with plasminogen activator inhibitor and tissue plasminogen activator, which are markers of endothelial dysfunction. It could, however, be speculated why the leukocyte count is a significant predictor of the maximal but not of the average IMT of the common carotid artery. Our results suggest that the inflammation process could better reflect a focal than a generalised endothelial damage.

It has been suggested that overnutrition and obesity through the synthesis of interleukin-6 and TNF-alpha could mediate the systemic metabolic impairments of the metabolic syndrome [33,34]. In our study body mass index and low physical activity were significantly correlated to leukocyte count and fibrinogen level. Besides, obesity was not an independent determinant of early atherosclerosis, so that the effect could be really mediated through the clustering to the other risk factors. It is also possible that the multiple risk factors within the metabolic syndrome may yield macrovascular disease and this be reflected by increased levels of inflammatory markers [35]. However, leukocyte count, CRP and other inflammatory parameters were found predictive of future cardio- and cerebrovascular events [6–14]. Further studies would be necessary to clarify the nature of the link between inflammation, dysglycemia and atherosclerosis, which could also represent different impairments of a common soil, strongly interrelated to each other in a vicious cycle.

In conclusion, our study indicates that low-grade inflammation is correlated to intima-media thickness, as an indicator of early atherosclerosis, and is strongly associated to a variety of risk factors within the metabolic syndrome in a population at risk for type 2 diabetes.

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
View Abstract