- Email: [email protected]
The relationship between high-sensitive C-reactive protein and pulse wave velocity in healthy Japanese men
Atherosclerosis 174 (2004) 373–377
The relationship between high-sensitive C-reactive protein and pulse wave velocity in healthy Japanese men Hirofumi Tomiyama a , Tomio Arai a , Yutaka Koji a , Minoru Yambe a , Yoji Hirayama a , Yoshio Yamamoto b , Akira Yamashina a,∗ a
Second Department of Internal Medicine, Tokyo Medical University, 6-7-1 Nishi-Shinjuku, Shinjuku-ku, 160-0023 Tokyo, Japan b Health Care Center, Kajima Corporation, Tokyo, Japan Received 27 May 2003; accepted 7 January 2004 Available online 15 April 2004
Abstract Although pulse wave velocity (PWV) and high-sensitive C-reactive protein (hsCRP) are known as predictors of future cardiovascular events, their association has not been examined. The present study was conducted to evaluate their association in the general population. In 2668 Japanese men (43 ± 10 years old), PWV was obtained by volume rendering methods, and hsCRP was determined by the latex aggregation method. PWV showed a significant correlation with logarithm of hsCRP (r = 0.06, P < 0.01). The concentration of hsCRP in the highest quartile of PWV was higher than that in the other three groups (P < 0.01). However, multiple linear regression analyses demonstrated that logarithm of hsCRP was not significantly related to PWV, independent from conventional risk factors. Calculated Framingham risk score (FRS) was higher in the highest quartiles of both hsCRP and PWV than in the other groups (P < 0.05). Thus, while increased hsCRP related to increased PWV, they may be independent predictors of atherosclerotic cardiovascular risk. A prospective study to confirm the independency of their significance in predicting future cardiovascular events and to evaluate the usefulness of the combination of both parameters to screen subjects for cardiovascular risk is necessary. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: C-reactive protein; Pulse wave velocity; Inflammation; Atherosclerosis; Risk factors
1. Introduction Pulse wave velocity (PWV) is known as a predictor of cardiovascular events independent from conventional atherosclerotic risk factors [1]. However, the precise mechanisms of the contribution of an increased PWV to future cardiovascular events have not been fully evaluated [2]. Arterial inflammation significantly contributes to the development of atherosclerotic cardiovascular disease [3], and several recent studies have demonstrated that high-sensitive C-reactive protein (hsCRP), which is a marker of inflammation, is also a novel predictor of cardiovascular events [4,5]. Increased pulse pressure has also been demonstrated as a cardiovascular risk factor [6], and Abramson et al., recently, reported that increases in pulse pressure were associated with elevated CRP levels [7].
PWV is a more valid marker of arterial stiffness than pulse pressure, and several studies have demonstrated that PWV is a more powerful predictor of cardiovascular events than pulse pressure [1,2]. If inflammation relates to the predictive value of an increased PWV regarding future cardiovascular events, there would be, at least in part, a confounding relationship between PWV and hsCRP. However, their relationship has not been evaluated as yet. The present study was conducted to evaluate the association of PWV with hsCRP in 2668 apparently healthy Japanese men in a cross-sectional study to examine whether these two parameters are similar or independent markers for atherosclerotic cardiovascular risk.
2. Methods 2.1. Subjects
∗
Corresponding author. Tel.: +81-3-3342-6111; fax: +81-3-3342-4820. E-mail address: [email protected] (A. Yamashina).
We recruited 2678 Japanese men from among subjects who underwent annual physical check-ups, including the
0021-9150/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2004.01.032
374
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
measurement of PWV and of hsCRP, at a clinic affiliated to Tokyo Medical University. All of participants work at a construction company as an office worker. Their ages ranged from 30 to 74 years. Smoking and medical history of atherogenic diseases or other diseases requiring medical treatment was obtained by a questionnaire answered by each subject at the time of the health examination. Subjects with a plasma creatinine concentration of 176.8 mol/l (n = 2), atrial fibrillation (n = 4), abnormal ankle/brachial pressure index (ankle brachial index <0.9) (n = 4) were excluded from the present study. It is noted that brachial-ankle PWV is not accurate in the subjects with either atrial fibrillation or an abnormal ankle/brachial pressure index. None of the subjects had a history or symptoms related to bone disorders or peripheral artery diseases and none of the women was under estrogen replacement therapy. Informed consent was obtained from all the subjects. The study protocol was approved by the ethical committee of Tokyo Medical University. 2.2. Measurement of pulse wave velocity Brachial-ankle PWV was measured using a volumeplethysmographic apparatus (Form/ABI, Colin, Co. Ltd., Komaki, Japan). Details of the methodology were as described elsewhere [8,9]. The subject was examined while resting in the supine position. Cuffs were wrapped on both brachia and ankles. Pulse volume waveforms at the brachium and ankle were recorded using a semiconductor pressure sensor. Brachial-ankle PWV was measured after at least 5 min rest. The validation of this method has been reported previously; the interobserver coefficient of variation (CV) was 8.4% and the intra-observer CV was 10.0% [8].
that the impact of total cholesterol and low-density lipoprotein cholesterol on estimates of the risk of coronary heart disease was similar [10]. Smoking habit was defined as active smoking at the present state. High blood pressure—measured by oscillometric method at the time of the measurement of brachial-ankle PWV—was defined as a systolic blood pressure of 140 and/or a diastolic blood pressure of 90 mmHg or medication with anti-hypertensive drugs; diabetes mellitus was defined as a fasting blood glucose of 6.9 mmol/l or medication with hypoglycemic agents, and dyslipidemia was defined as total cholesterol of 6.2 mmol/l or medication with hypocholesterolemic drugs. 2.5. Statistics Data are expressed as the mean ± S.D. (error bars are represented in Fig. 1). Statistical analysis was performed using the SPSS software package (SPSS, Chicago, IL). The concentration of hsCRP was skewed rightward, and the baseline level was logarithmically normalized to allow for either linear or multiple regression analyses. Gender differences were assessed by unpaired t-test or chi-square test. The association between brachial-ankle PWV and the logarithm of hsCRP was subjected to linear regression analysis. One way analysis of variance with Bonferroni’s adjustment was used to assess the difference of hsCRP among quartiles of brachial-ankle PWV. This analysis was also applied to assess the difference in Framingham risk score (FRS) among the groups divided by quartiles of brachial-ankle PWV and quartiles of hsCRP. Multiple linear regression analysis was used to assess the independent relationship between brachial-ankle PWV and clinical variables.
2.3. Laboratory measurements Plasma total cholesterol, high-density lipoprotein cholesterol, electrolytes, and blood sugar levels were measured enzymatically. CV of laboratory measurements were as follows; total cholesterol (within: 0.38% and run: 0.32%), high-density lipoprotein cholesterol (within: 1.93% and run: 1.54%), and blood sugar (within: 0.80% and run: 0.83%). hsCRP was determined by the latex-aggregation method (Eiken Co. Tokyo, Japan). All blood samples were obtained in the fasting state in the morning. 2.4. Risk scoring and definition The experience of the Framingham study population was used to develop an algorithm that resulted in a calculated score to predict the 10-year risk of coronary heart disease in persons aged 30–74 years without atherosclerotic cardiovascular diseases (coronary heart disease and stroke) [10]. In this calculation score, we used total cholesterol instead of low-density lipoprotein cholesterol, Wilson et al. reported
Fig. 1. Plasma concentration of high-sensitive C-reactive protein in subjects distributed by their brachial-ankle pulse wave velocity. (1) Group of subjects in the lowest quartile of brachial-ankle pulse wave velocity; (2) group of subjects in the lower quartile of brachial-ankle pulse wave velocity; (3) group of subjects in the higher quartile of brachial-ankle pulse wave velocity; (4) group of subjects in the highest quartile of brachial-ankle pulse wave velocity. ∗ P < 0.05 vs. the group of subjects in the lowest quartile of brachial-ankle pulse wave velocity.
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
375
Table 1 Anthropometrics of the subjects Total number of subjects
2668
Age BMI (kg/m2 ) SBP (mmHg) DBP (mmHg) PP (mmHg) baPWV (cm/s) FBS (mol/l) TC (mol/l) HDL (mol/l) hsCRP (mg/l)
43 24 130 79 50 1320 5.2 5.2 1.5 1.2
± ± ± ± ± ± ± ± ± ±
10 3 15 11 8 220 0.8 0.9 0.5 2.8
Number of subjects with (percentage to total subjects) Smoking habit 1039 (39) Hypertension 616 (23) Diabetes mellitus 117 (4) Dyslipidemia 414 (16) BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; PP: pulse pressure; baPWV: brachial-ankle pulse wave velocity; FBS: fasting blood sugar; TC: total cholesterol; HDL: high-density lipoprotein cholesterol; hsCRP: high-sensitive C-reactive protein; smoking habit: active smoking at the present state; hypertension: measured by oscillometric method at the time of the measurement of brachial-ankle PWV—was defined as a systolic blood pressure of 140 and/or a diastolic blood pressure of 90 mmHg or medication with anti-hypertensive drugs; diabetes mellitus: fasting blood glucose of 6.9 mmol/l or medication with hypoglycemic agents; dyslipidemia: total cholesterol of 6.2 mmol/l or medication with hypocholesterolemic drugs.
3. Results The clinical characteristics, including medical history of the study population in both genders, are summarized in Table 1. Linear regression analysis showed a significant correlation between brachial-ankle PWV and logarithm of hsCRP (r = 0.06, P < 0.01). Fig. 1 illustrates the
Table 2 Results of multiple linear regression analysis to assess independent relationships between brachial-ankle pulse wave velocity and clinical variables
Significant independent variable SBP PP Age BMI FBS
β
P-value
0.48 0.09 0.30 −0.10 0.06
P P P P P
< < < < <
0.01 0.01 0.01 0.01 0.01
Not-independent variable DBP TC HDL Smoking habit log hsCRP R2 = 0.36; SBP: systolic blood pressure; DBP: diastolic blood pressure; PP: pulse pressure; BMI: body mass index; FBS: fasting blood sugar; TC: total cholesterol; HDL: high-density lipoprotein cholesterol; smoking habit: subjects with active smoking; log hsCRP: logarithm high-sensitive C-reactive protein.
Fig. 2. Framingham risk score in subjects divided by the quartiles of high-sensitive C-reactive protein and quartiles of brachial-ankle pulse wave velocity. (1) group of subjects in the lowest quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (2) group of subjects in the lower quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (3) group of subjects in the higher quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (4) group of subjects in the highest quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein. ∗ P < 0.05 vs. other groups.
mean concentration of hsCRP in subjects distributed by their brachial-ankle PWV. Quartile ranges of brachial-ankle PWV were 910–1176, 1177–1274, 1275–1421, and 1422–3263 cm/s. The concentration of hsCRP in the highest quartile of brachial-ankle PWV was higher than that in the other three groups (P < 0.01). Multiple linear regression analysis demonstrated that the logarithm of hsCRP did not show a significant correlation with brachial-ankle PWV, independent from conventional risk factors (Table 2). Fig. 2 illustrates the mean FRS in the groups divided by quartiles of hsCRP and quartiles of brachial-ankle PWV. Quartile ranges of plasma concentration of hsCRP were 0–0.2, 0.3–0.5, 0.6–1.0, and 1.1–43.5 mg/l. FRS was higher in the highest quartiles of both brachial-ankle PWV and hsCRP than in the other groups.
4. Discussion An increased PWV is an independent predictor of cardiovascular events [1]. The speculated underlying mechanisms are as follows: increased arterial stiffness, as reflected by an increased PWV, results in an increased cardiac ventricle load, reduced ejection fraction, increased myocardial oxygen demand, and aggravation of atherosclerosis via an increased stress on the arterial wall [2,11]. Recently, Abramson et al. found an association between pulse pressure and CRP, and they speculated that aggravated inflammation, in addition to the above mentioned mechanisms, also affected the prognosis of patients with increased arterial stiffness [7]. Increased arterial stiffness can aggravate inflammation in the systemic arterial tree via several mechanisms such as increased generation of oxygen species or increased expression of adhesion molecules on the endothelium [12].
376
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
However, it has not been concluded whether inflammation is a pathogenesis of atherosclerosis or it reflect the degree of atherosclerotic vascular damage [13]. Anyhow, CRP is a predictor of future cardiovascular events. In the present study, while the plasma concentration of hsCRP was elevated in individuals with an increased brachial-ankle PWV, their association was not significant independent from conventional atherosclerotic risk factors. Therefore, the confounding relationship between hsCRP and PWV as predictors of future cardiovascular events seems to be relatively weak. The measurement of hsCRP is simple and inexpensive, and it is highly applicable to the screening of the general population for risk factors of cardiovascular events as a first line screening tool. On the other hand, PWV has not been established as test for cardiovascular evaluation in clinical practice as yet. hsCRP in conjunction with standard cholesterol evaluation is reported as a quite practical approach to screen subjects for the presence of a cardiovascular risk [14]. The lack of a close relationship between brachial-ankle PWV and hsCRP may support the possibility that both parameters are independent predictors of cardiovascular events [11]. The fact that FRS, which is an established predictor of cardiovascular events, was highest in the highest quartile of both brachial-ankle PWV and hsCRP also supports their independence. Therefore, brachial-ankle PWV and hsCRP may be independent predictors of atherosclerotic cardiovascular risk. The usefulness of the combination of both parameters for the assessment of atherosclerotic cardiovascular risk should be evaluated. Another important hypothesis concerning the association between PWV and hsCRP is that hsCRP is a marker of the extent of atherosclerosis [15]. If this hypothesis is confirmed, an increased atherosclerotic burden may explain part of the increased risk of cardiovascular events in individuals with elevated hsCRP. Atheromas produce interleukin-6, and CRP is mainly produced in the liver by interleukin-6 [3]. In addition, recent studies have demonstrated that arterial tissue directly produces CRP [16]. Thus, CRP might be a marker of the extent of atherosclerosis. Although several studies have examined the association between hsCRP and intima-media thickness (IMT) of the carotid artery, which is not only a marker of subclinical atherosclerosis but also a predictor of the patients’ prognosis [17], their association was inconsistent [15,18]. However, IMT is a marker of regional subclinical atherosclerosis. In the systemic arterial tree, the extent of atherosclerosis is not homogenous [19], and markers of regional atherosclerosis might not fully reflect the extent of atherosclerosis in the systemic arterial tree. The degenerative and proliferate changes in either vascular smooth muscle cells or interstitial matrix cells associated with atherosclerosis lead to an increase of arterial stiffness [2]. The progression of atherosclerosis may be associated with a concomitant increase of PWV, and PWV is also thought as a marker of subclinical atherosclerosis in the systemic arterial tree. The present study suggests that hsCRP is a marker of the extent of atherosclerosis at least in part.
4.1. Study limitations The risk of increased PWV has been demonstrated using the carotid-femoral PWV, which is a conventional method [1,2]. Brachial-ankle PWV is a new available method, and no prospective study has demonstrated the usefulness of brachial-ankle PWV to predict future cardiovascular events. Such study is now underway in our institution. However, recent accumulated data tend to indicate that brachial-ankle PWV may be as useful as the carotid-femoral PWV [8,9]. Since the subjects in this study were men, the consistency of the present results should be confirmed in women. Most of subjects in the present study were middle-aged men, and they were a relative low risk population. The prognostic value of pulse pressure and hsCRP are significant especially in the elderly [6,5,14], and the evaluations of the association between PWV and hsCRP in the elderly (a high risk population) is the next step. In conclusion, while increased hsCRP related to increased brachial-ankle PWV, they may be independent predictors of atherosclerotic cardiovascular risk. A prospective study to confirm the independency of their significance in predicting future cardiovascular events and to evaluate the usefulness of the combination of both parameters to screen subjects for cardiovascular risk is necessary.
Acknowledgements This study was supported in part by a grant-in-aid from the Japanese Atherosclerosis Prevention Fund awarded to Prof. Akira Yamashina.
References [1] Laurent S, Boutouyrie P, Asmar R, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001;37:1236–41. [2] London GM, Cohn JN. Prognostic application of arterial stiffness: task forces. Am J Hypertens 2002;5:754–8. [3] Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–74. [4] Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke 2001;32:2575–9. [5] Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–65. [6] Franklin SS, Larson MG, Khan SA, et al. Does the relation of blood pressure to coronary heart disease risk change with aging? The Framingham Heart Study. Circulation 2001;103:1245–9. [7] Abramson JL, Weintraub WS, Vaccarino V. Association between pulse pressure and C-reactive protein among apparently healthy US adults. Hypertension 2002;39:197–202. [8] Yamashina A, Tomiyama H, Takeda K, et al. Validity, reproducibility, and clinical significance of noninvasive brachial-ankle pulse wave velocity measurement. Hypertens Res 2002;25:359–64. [9] Tomiyama H, Yamashina A, Arai T, et al. Influences of age and gender on results of noninvasive brachial-ankle pulse wave ve-
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
[10]
[11] [12]
[13]
[14]
locity measurement—a survey of 12517 subjects. Atherosclerosis 2003;16:303–9. Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837– 47. Cohn JN. Vascular wall function as a risk marker for cardiovascular disease. J Hypertens 1999:(Suppl 5):S41–4. Chappell DC, Varner SE, Nerem RM, Medford RM, Alexander RW. Oscillatory shear stress stimulates adhesion molecule expression in cultured human endothelium. Circ Res 1998;82:532–9. Mehta JL, Li DY. Inflammation in ischemic heart disease: response to tissue injury or a pathogenetic villain? Cardiovasc Res 1999;43:291– 9. Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 2003;107:363–9.
377
[15] Wang TJ, Nam BH, Wilson PW, et al. Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study. Arterioscler Thromb Vasc Biol 2002;22:1662–7. [16] Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 2000;102:2165–8. [17] Crouse JR, Goldbourt U, Evans G, et al. Risk factors and segment-specific carotid arterial enlargement in the atherosclerosis risk in communities (ARIC) cohort. Stroke 1996;27:69–75. [18] Blackburn R, Giral P, Bruckert E, et al. Elevated C-reactive protein constitutes an independent predictor of advanced carotid plaques in dyslipidemic subjects. Arterioscler Thromb Vasc Biol 2001;21:1962– 8. [19] Nissen SE. Who is at risk for atherosclerotic disease? Lessons from intravascular ultrasound. Am J Med 2002;112(Suppl 8A):27S–33S.
The relationship between high-sensitive C-reactive protein and pulse wave velocity in healthy Japanese men Hirofumi Tomiyama a , Tomio Arai a , Yutaka Koji a , Minoru Yambe a , Yoji Hirayama a , Yoshio Yamamoto b , Akira Yamashina a,∗ a
Second Department of Internal Medicine, Tokyo Medical University, 6-7-1 Nishi-Shinjuku, Shinjuku-ku, 160-0023 Tokyo, Japan b Health Care Center, Kajima Corporation, Tokyo, Japan Received 27 May 2003; accepted 7 January 2004 Available online 15 April 2004
Abstract Although pulse wave velocity (PWV) and high-sensitive C-reactive protein (hsCRP) are known as predictors of future cardiovascular events, their association has not been examined. The present study was conducted to evaluate their association in the general population. In 2668 Japanese men (43 ± 10 years old), PWV was obtained by volume rendering methods, and hsCRP was determined by the latex aggregation method. PWV showed a significant correlation with logarithm of hsCRP (r = 0.06, P < 0.01). The concentration of hsCRP in the highest quartile of PWV was higher than that in the other three groups (P < 0.01). However, multiple linear regression analyses demonstrated that logarithm of hsCRP was not significantly related to PWV, independent from conventional risk factors. Calculated Framingham risk score (FRS) was higher in the highest quartiles of both hsCRP and PWV than in the other groups (P < 0.05). Thus, while increased hsCRP related to increased PWV, they may be independent predictors of atherosclerotic cardiovascular risk. A prospective study to confirm the independency of their significance in predicting future cardiovascular events and to evaluate the usefulness of the combination of both parameters to screen subjects for cardiovascular risk is necessary. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: C-reactive protein; Pulse wave velocity; Inflammation; Atherosclerosis; Risk factors
1. Introduction Pulse wave velocity (PWV) is known as a predictor of cardiovascular events independent from conventional atherosclerotic risk factors [1]. However, the precise mechanisms of the contribution of an increased PWV to future cardiovascular events have not been fully evaluated [2]. Arterial inflammation significantly contributes to the development of atherosclerotic cardiovascular disease [3], and several recent studies have demonstrated that high-sensitive C-reactive protein (hsCRP), which is a marker of inflammation, is also a novel predictor of cardiovascular events [4,5]. Increased pulse pressure has also been demonstrated as a cardiovascular risk factor [6], and Abramson et al., recently, reported that increases in pulse pressure were associated with elevated CRP levels [7].
PWV is a more valid marker of arterial stiffness than pulse pressure, and several studies have demonstrated that PWV is a more powerful predictor of cardiovascular events than pulse pressure [1,2]. If inflammation relates to the predictive value of an increased PWV regarding future cardiovascular events, there would be, at least in part, a confounding relationship between PWV and hsCRP. However, their relationship has not been evaluated as yet. The present study was conducted to evaluate the association of PWV with hsCRP in 2668 apparently healthy Japanese men in a cross-sectional study to examine whether these two parameters are similar or independent markers for atherosclerotic cardiovascular risk.
2. Methods 2.1. Subjects
∗
Corresponding author. Tel.: +81-3-3342-6111; fax: +81-3-3342-4820. E-mail address: [email protected] (A. Yamashina).
We recruited 2678 Japanese men from among subjects who underwent annual physical check-ups, including the
0021-9150/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2004.01.032
374
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
measurement of PWV and of hsCRP, at a clinic affiliated to Tokyo Medical University. All of participants work at a construction company as an office worker. Their ages ranged from 30 to 74 years. Smoking and medical history of atherogenic diseases or other diseases requiring medical treatment was obtained by a questionnaire answered by each subject at the time of the health examination. Subjects with a plasma creatinine concentration of 176.8 mol/l (n = 2), atrial fibrillation (n = 4), abnormal ankle/brachial pressure index (ankle brachial index <0.9) (n = 4) were excluded from the present study. It is noted that brachial-ankle PWV is not accurate in the subjects with either atrial fibrillation or an abnormal ankle/brachial pressure index. None of the subjects had a history or symptoms related to bone disorders or peripheral artery diseases and none of the women was under estrogen replacement therapy. Informed consent was obtained from all the subjects. The study protocol was approved by the ethical committee of Tokyo Medical University. 2.2. Measurement of pulse wave velocity Brachial-ankle PWV was measured using a volumeplethysmographic apparatus (Form/ABI, Colin, Co. Ltd., Komaki, Japan). Details of the methodology were as described elsewhere [8,9]. The subject was examined while resting in the supine position. Cuffs were wrapped on both brachia and ankles. Pulse volume waveforms at the brachium and ankle were recorded using a semiconductor pressure sensor. Brachial-ankle PWV was measured after at least 5 min rest. The validation of this method has been reported previously; the interobserver coefficient of variation (CV) was 8.4% and the intra-observer CV was 10.0% [8].
that the impact of total cholesterol and low-density lipoprotein cholesterol on estimates of the risk of coronary heart disease was similar [10]. Smoking habit was defined as active smoking at the present state. High blood pressure—measured by oscillometric method at the time of the measurement of brachial-ankle PWV—was defined as a systolic blood pressure of 140 and/or a diastolic blood pressure of 90 mmHg or medication with anti-hypertensive drugs; diabetes mellitus was defined as a fasting blood glucose of 6.9 mmol/l or medication with hypoglycemic agents, and dyslipidemia was defined as total cholesterol of 6.2 mmol/l or medication with hypocholesterolemic drugs. 2.5. Statistics Data are expressed as the mean ± S.D. (error bars are represented in Fig. 1). Statistical analysis was performed using the SPSS software package (SPSS, Chicago, IL). The concentration of hsCRP was skewed rightward, and the baseline level was logarithmically normalized to allow for either linear or multiple regression analyses. Gender differences were assessed by unpaired t-test or chi-square test. The association between brachial-ankle PWV and the logarithm of hsCRP was subjected to linear regression analysis. One way analysis of variance with Bonferroni’s adjustment was used to assess the difference of hsCRP among quartiles of brachial-ankle PWV. This analysis was also applied to assess the difference in Framingham risk score (FRS) among the groups divided by quartiles of brachial-ankle PWV and quartiles of hsCRP. Multiple linear regression analysis was used to assess the independent relationship between brachial-ankle PWV and clinical variables.
2.3. Laboratory measurements Plasma total cholesterol, high-density lipoprotein cholesterol, electrolytes, and blood sugar levels were measured enzymatically. CV of laboratory measurements were as follows; total cholesterol (within: 0.38% and run: 0.32%), high-density lipoprotein cholesterol (within: 1.93% and run: 1.54%), and blood sugar (within: 0.80% and run: 0.83%). hsCRP was determined by the latex-aggregation method (Eiken Co. Tokyo, Japan). All blood samples were obtained in the fasting state in the morning. 2.4. Risk scoring and definition The experience of the Framingham study population was used to develop an algorithm that resulted in a calculated score to predict the 10-year risk of coronary heart disease in persons aged 30–74 years without atherosclerotic cardiovascular diseases (coronary heart disease and stroke) [10]. In this calculation score, we used total cholesterol instead of low-density lipoprotein cholesterol, Wilson et al. reported
Fig. 1. Plasma concentration of high-sensitive C-reactive protein in subjects distributed by their brachial-ankle pulse wave velocity. (1) Group of subjects in the lowest quartile of brachial-ankle pulse wave velocity; (2) group of subjects in the lower quartile of brachial-ankle pulse wave velocity; (3) group of subjects in the higher quartile of brachial-ankle pulse wave velocity; (4) group of subjects in the highest quartile of brachial-ankle pulse wave velocity. ∗ P < 0.05 vs. the group of subjects in the lowest quartile of brachial-ankle pulse wave velocity.
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
375
Table 1 Anthropometrics of the subjects Total number of subjects
2668
Age BMI (kg/m2 ) SBP (mmHg) DBP (mmHg) PP (mmHg) baPWV (cm/s) FBS (mol/l) TC (mol/l) HDL (mol/l) hsCRP (mg/l)
43 24 130 79 50 1320 5.2 5.2 1.5 1.2
± ± ± ± ± ± ± ± ± ±
10 3 15 11 8 220 0.8 0.9 0.5 2.8
Number of subjects with (percentage to total subjects) Smoking habit 1039 (39) Hypertension 616 (23) Diabetes mellitus 117 (4) Dyslipidemia 414 (16) BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; PP: pulse pressure; baPWV: brachial-ankle pulse wave velocity; FBS: fasting blood sugar; TC: total cholesterol; HDL: high-density lipoprotein cholesterol; hsCRP: high-sensitive C-reactive protein; smoking habit: active smoking at the present state; hypertension: measured by oscillometric method at the time of the measurement of brachial-ankle PWV—was defined as a systolic blood pressure of 140 and/or a diastolic blood pressure of 90 mmHg or medication with anti-hypertensive drugs; diabetes mellitus: fasting blood glucose of 6.9 mmol/l or medication with hypoglycemic agents; dyslipidemia: total cholesterol of 6.2 mmol/l or medication with hypocholesterolemic drugs.
3. Results The clinical characteristics, including medical history of the study population in both genders, are summarized in Table 1. Linear regression analysis showed a significant correlation between brachial-ankle PWV and logarithm of hsCRP (r = 0.06, P < 0.01). Fig. 1 illustrates the
Table 2 Results of multiple linear regression analysis to assess independent relationships between brachial-ankle pulse wave velocity and clinical variables
Significant independent variable SBP PP Age BMI FBS
β
P-value
0.48 0.09 0.30 −0.10 0.06
P P P P P
< < < < <
0.01 0.01 0.01 0.01 0.01
Not-independent variable DBP TC HDL Smoking habit log hsCRP R2 = 0.36; SBP: systolic blood pressure; DBP: diastolic blood pressure; PP: pulse pressure; BMI: body mass index; FBS: fasting blood sugar; TC: total cholesterol; HDL: high-density lipoprotein cholesterol; smoking habit: subjects with active smoking; log hsCRP: logarithm high-sensitive C-reactive protein.
Fig. 2. Framingham risk score in subjects divided by the quartiles of high-sensitive C-reactive protein and quartiles of brachial-ankle pulse wave velocity. (1) group of subjects in the lowest quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (2) group of subjects in the lower quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (3) group of subjects in the higher quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein; (4) group of subjects in the highest quartile of either brachial-ankle pulse wave velocity or high-sensitive C-reactive protein. ∗ P < 0.05 vs. other groups.
mean concentration of hsCRP in subjects distributed by their brachial-ankle PWV. Quartile ranges of brachial-ankle PWV were 910–1176, 1177–1274, 1275–1421, and 1422–3263 cm/s. The concentration of hsCRP in the highest quartile of brachial-ankle PWV was higher than that in the other three groups (P < 0.01). Multiple linear regression analysis demonstrated that the logarithm of hsCRP did not show a significant correlation with brachial-ankle PWV, independent from conventional risk factors (Table 2). Fig. 2 illustrates the mean FRS in the groups divided by quartiles of hsCRP and quartiles of brachial-ankle PWV. Quartile ranges of plasma concentration of hsCRP were 0–0.2, 0.3–0.5, 0.6–1.0, and 1.1–43.5 mg/l. FRS was higher in the highest quartiles of both brachial-ankle PWV and hsCRP than in the other groups.
4. Discussion An increased PWV is an independent predictor of cardiovascular events [1]. The speculated underlying mechanisms are as follows: increased arterial stiffness, as reflected by an increased PWV, results in an increased cardiac ventricle load, reduced ejection fraction, increased myocardial oxygen demand, and aggravation of atherosclerosis via an increased stress on the arterial wall [2,11]. Recently, Abramson et al. found an association between pulse pressure and CRP, and they speculated that aggravated inflammation, in addition to the above mentioned mechanisms, also affected the prognosis of patients with increased arterial stiffness [7]. Increased arterial stiffness can aggravate inflammation in the systemic arterial tree via several mechanisms such as increased generation of oxygen species or increased expression of adhesion molecules on the endothelium [12].
376
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
However, it has not been concluded whether inflammation is a pathogenesis of atherosclerosis or it reflect the degree of atherosclerotic vascular damage [13]. Anyhow, CRP is a predictor of future cardiovascular events. In the present study, while the plasma concentration of hsCRP was elevated in individuals with an increased brachial-ankle PWV, their association was not significant independent from conventional atherosclerotic risk factors. Therefore, the confounding relationship between hsCRP and PWV as predictors of future cardiovascular events seems to be relatively weak. The measurement of hsCRP is simple and inexpensive, and it is highly applicable to the screening of the general population for risk factors of cardiovascular events as a first line screening tool. On the other hand, PWV has not been established as test for cardiovascular evaluation in clinical practice as yet. hsCRP in conjunction with standard cholesterol evaluation is reported as a quite practical approach to screen subjects for the presence of a cardiovascular risk [14]. The lack of a close relationship between brachial-ankle PWV and hsCRP may support the possibility that both parameters are independent predictors of cardiovascular events [11]. The fact that FRS, which is an established predictor of cardiovascular events, was highest in the highest quartile of both brachial-ankle PWV and hsCRP also supports their independence. Therefore, brachial-ankle PWV and hsCRP may be independent predictors of atherosclerotic cardiovascular risk. The usefulness of the combination of both parameters for the assessment of atherosclerotic cardiovascular risk should be evaluated. Another important hypothesis concerning the association between PWV and hsCRP is that hsCRP is a marker of the extent of atherosclerosis [15]. If this hypothesis is confirmed, an increased atherosclerotic burden may explain part of the increased risk of cardiovascular events in individuals with elevated hsCRP. Atheromas produce interleukin-6, and CRP is mainly produced in the liver by interleukin-6 [3]. In addition, recent studies have demonstrated that arterial tissue directly produces CRP [16]. Thus, CRP might be a marker of the extent of atherosclerosis. Although several studies have examined the association between hsCRP and intima-media thickness (IMT) of the carotid artery, which is not only a marker of subclinical atherosclerosis but also a predictor of the patients’ prognosis [17], their association was inconsistent [15,18]. However, IMT is a marker of regional subclinical atherosclerosis. In the systemic arterial tree, the extent of atherosclerosis is not homogenous [19], and markers of regional atherosclerosis might not fully reflect the extent of atherosclerosis in the systemic arterial tree. The degenerative and proliferate changes in either vascular smooth muscle cells or interstitial matrix cells associated with atherosclerosis lead to an increase of arterial stiffness [2]. The progression of atherosclerosis may be associated with a concomitant increase of PWV, and PWV is also thought as a marker of subclinical atherosclerosis in the systemic arterial tree. The present study suggests that hsCRP is a marker of the extent of atherosclerosis at least in part.
4.1. Study limitations The risk of increased PWV has been demonstrated using the carotid-femoral PWV, which is a conventional method [1,2]. Brachial-ankle PWV is a new available method, and no prospective study has demonstrated the usefulness of brachial-ankle PWV to predict future cardiovascular events. Such study is now underway in our institution. However, recent accumulated data tend to indicate that brachial-ankle PWV may be as useful as the carotid-femoral PWV [8,9]. Since the subjects in this study were men, the consistency of the present results should be confirmed in women. Most of subjects in the present study were middle-aged men, and they were a relative low risk population. The prognostic value of pulse pressure and hsCRP are significant especially in the elderly [6,5,14], and the evaluations of the association between PWV and hsCRP in the elderly (a high risk population) is the next step. In conclusion, while increased hsCRP related to increased brachial-ankle PWV, they may be independent predictors of atherosclerotic cardiovascular risk. A prospective study to confirm the independency of their significance in predicting future cardiovascular events and to evaluate the usefulness of the combination of both parameters to screen subjects for cardiovascular risk is necessary.
Acknowledgements This study was supported in part by a grant-in-aid from the Japanese Atherosclerosis Prevention Fund awarded to Prof. Akira Yamashina.
References [1] Laurent S, Boutouyrie P, Asmar R, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001;37:1236–41. [2] London GM, Cohn JN. Prognostic application of arterial stiffness: task forces. Am J Hypertens 2002;5:754–8. [3] Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–74. [4] Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke 2001;32:2575–9. [5] Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–65. [6] Franklin SS, Larson MG, Khan SA, et al. Does the relation of blood pressure to coronary heart disease risk change with aging? The Framingham Heart Study. Circulation 2001;103:1245–9. [7] Abramson JL, Weintraub WS, Vaccarino V. Association between pulse pressure and C-reactive protein among apparently healthy US adults. Hypertension 2002;39:197–202. [8] Yamashina A, Tomiyama H, Takeda K, et al. Validity, reproducibility, and clinical significance of noninvasive brachial-ankle pulse wave velocity measurement. Hypertens Res 2002;25:359–64. [9] Tomiyama H, Yamashina A, Arai T, et al. Influences of age and gender on results of noninvasive brachial-ankle pulse wave ve-
H. Tomiyama et al. / Atherosclerosis 174 (2004) 373–377
[10]
[11] [12]
[13]
[14]
locity measurement—a survey of 12517 subjects. Atherosclerosis 2003;16:303–9. Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837– 47. Cohn JN. Vascular wall function as a risk marker for cardiovascular disease. J Hypertens 1999:(Suppl 5):S41–4. Chappell DC, Varner SE, Nerem RM, Medford RM, Alexander RW. Oscillatory shear stress stimulates adhesion molecule expression in cultured human endothelium. Circ Res 1998;82:532–9. Mehta JL, Li DY. Inflammation in ischemic heart disease: response to tissue injury or a pathogenetic villain? Cardiovasc Res 1999;43:291– 9. Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation 2003;107:363–9.
377
[15] Wang TJ, Nam BH, Wilson PW, et al. Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study. Arterioscler Thromb Vasc Biol 2002;22:1662–7. [16] Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 2000;102:2165–8. [17] Crouse JR, Goldbourt U, Evans G, et al. Risk factors and segment-specific carotid arterial enlargement in the atherosclerosis risk in communities (ARIC) cohort. Stroke 1996;27:69–75. [18] Blackburn R, Giral P, Bruckert E, et al. Elevated C-reactive protein constitutes an independent predictor of advanced carotid plaques in dyslipidemic subjects. Arterioscler Thromb Vasc Biol 2001;21:1962– 8. [19] Nissen SE. Who is at risk for atherosclerotic disease? Lessons from intravascular ultrasound. Am J Med 2002;112(Suppl 8A):27S–33S.