Enhanced External Counterpulsation Attenuates Atherosclerosis Progression Through Modulation of Proinflammatory Signal Pathway
Objective— Shear stress may be the most crucial local factor affecting atherogenesis. The present study investigated the effect of exposure to increased shear stress promoted by enhanced external counterpulsation (EECP) on the progression of atherosclerosis and the underlying inflammation-related molecular mechanisms in a porcine model of hypercholesterolemia.
Methods and Results— Hypercholesterolemic pigs were subjected to a 7-week EECP intervention while being fed a high-cholesterol diet. EECP resulted in a 34.38% increase of mean wall shear stress and a significantly lower pulsatility index in the brachial artery. The animals receiving EECP showed a marked reduction in atherosclerotic lesion size in the coronary artery and abdominal aorta compared with the hypercholesterolemic control group, associated with a decrease in macrophage accumulation. The expression of a set of genes involved in inflammation (including C-reactive protein [CRP], complement 3a, vascular cell adhesion molecule-1 [VCAM-1], and inducible nitric oxide synthase), mitogen-activated protein kinase (MAPK)-p38 phosphorylation, and nuclear factor-κB (NF-κB) activation, was attenuated.
Conclusion— These findings suggested that long-term EECP exerts a retarding effect on atherosclerosis by downregulating proinflammatory gene expression. The underlying mechanisms are related to chronic exposure to increased pulsatile shear stress promoted by EECP; this exposure suppresses the overactivation of the MAPK-P38/NF-κB/VCAM-1 signaling pathway induced by hypercholesterolemia.
Atherosclerosis remains the leading cause of morbidity and mortality in many countries. It is now recognized as a chronic inflammatory and immune disease. Oxidation of low-density lipoprotein can elicit a progressive inflammatory response in the arterial wall through activation of inflammation-relevant signaling pathways, resulting in endothelial cell adhesion molecule expression, leukocyte recruitment, macrophage lipid accumulation, and foam cell formation.1,2
Studies have identified that atherosclerotic lesions preferentially develop at distinct sites such as curves, branches, and bifurcations in arteries, areas that experience turbulent blood flow, low fluid shear stress, or flow reversal. Conversely, adjacent regions exposed to undisturbed flow and high mean shear stress are protected from plaque formation. High shear stress with laminal flow is generally beneficial, as it promotes vascular adaptation and dilatation through endothelium-mediated mechanisms and exerts antiinflammatory and prorepair functions on the vasculature.3
Enhanced external counterpulsation (EECP) is a noninvasive modality for the treatment of ischemic cardiovascular disease. EECP therapy is done by sequential inflation of 3 sets of cuffs wrapped around the lower extremities during diastole and deflation of the cuffs during systole. EECP enhances the aortic diastolic blood flow and coronary perfusion,4–6 leading to increased arterial wall shear stress in a pulsatile manner.7–8 Recently, the inhibitory effects of EECP on circulating proinflammatory biomarkers in patients with symptomatic coronary artery disease has been documented.9 However, the exact molecular mechanisms underlying the clinical benefits have not been fully clarified.
This study proposed to examine the effect of EECP on vascular atherosclerosis and inflammation in a porcine model of hypercholesterolemia. Because mitogen-activated protein kinases (MAPKs) and NF-κB play central roles in the inflammatory signal transduction, and vascular cell adhesion molecule-1 (VCAM-1) is of pivotal importance in monocyte-endothelial interaction and early atherosclerosis, we investigated these inflammation-relevant molecules and the proinflammatory MAPK-p38/NF-κB/VCAM-1 signaling pathway to provide insight into the molecular mechanisms of the action of EECP.
Materials and Methods
Thirty-five domestic male pigs were randomized to normal chow (control group; n=7) or a high-cholesterol diet (n=28) for 15 weeks. After 8 weeks of a high-cholesterol diet, when hypercholesterolemia was achieved, 17 pigs received EECP intervention while being maintained on the atherogenic diet (CHOL* EECP group). The remaining 11 pigs served as hypercholesterolemic controls (CHOL group). EECP was performed for 2 hours every other day, for a total of 34±2 hours, with the use of a modified clinical system. The lower extremities and the hips of the pigs were wrapped by 2 sets of modified cuffs. The pressure applied to the cuffs was set at 0.035 to 0.040 MPa/cm2. Effective hemodynamic changes of EECP were demonstrated by achieving a diastolic to systolic ratio greater than 1.2 using the plethysmographic technique. All pigs were anesthetized with 5 to 10 mg of midazolam IM and 10 mg/kg per hour 3% pentobarbital IV infusion. Experiments were performed in accordance with institutional guidelines.
Doppler ultrasound flow examinations and arterial wall shear stress calculations were performed in the right brachial artery as previously described.7,10 Pulsatility index (PI) as a measure of vascular impedance was calculated according to the formula of Gosling and King: PI=(peak systolic velocity−end diastolic velocity)/mean velocity.11 Blood pressure and flow volume measurements were performed in the common carotid artery. Maximum pulse flow difference (maximum flow volume−minimum flow volume) was calculated to evaluate the pulsatility of blood flow.
Peripheral blood samples were taken at baseline (0 weeks), before EECP (8 weeks), in the middle of the EECP treatment course (12 weeks), and at the end of the study (15 weeks) for measurement of blood lipid profiles (Hitachi 7170A, Hitachi, Tokyo, Japan).
Morphological Analysis of Atherosclerotic Lesions
The porcine abdominal aortas were en face stained with Sudan IV solution, cross-sectioned, and then stained with Oil Red O for evaluation of lesions. Left anterior descending coronary artery (LAD) cross-sections were stained with hematoxylin and eosin, elastica Gomori aldehyde fuchsin, and Oil Red O for microscopic quantification of the coronary atherosclerosis. To reveal the cellular and collagen components, smooth muscle α-actin immunohistochemistry and Masson trichrome staining were processed. Segments of the LADs were treated for scanning electron microscopy (Hitachi S520).
Quantitative Real-Time Polymerase Chain Reaction
Total RNA extracted from snap-frozen porcine aortic arches was used to measure VCAM-1 and inducible nitric oxide synthase (iNOS) mRNA expression.
Immunohistochemistry and Immunofluorescence
Serial paraffin-embedded cross-sections of LADs were incubated with primary antibodies specific for phospho-p38 (Cell Signaling Technology, Danvers, Mass.) or CRP (Santa Cruz Biotechnology, Inc, Santa Cruz, Calif.), complement 3a (Santa Cruz Biotechnology), macrophage (Santa Cruz Biotechnology), VCAM-1 (Santa Cruz Biotechnology), iNOS (Santa Cruz Biotechnology), and NF-κBp65 (Cell Signaling Technology). Immunofluorescence images were obtained with a confocal laser scanning fluorescence microscope (Fluoview FV500-IX81, Olympus, Tokyo, Japan).
Western Blot Analysis
The primary antibodies used in the Western blot were polyclonal anti-phospho-p38 and p38 (Cell Signaling Technology), anti-phospho-NF-κB and NF-κB (Cell Signaling Technology), anti-IκB (Cell Signaling Technology), and anti-CRP (Santa Cruz Biotechnology).
Data are reported as mean±SEM. When homogeneity of variance was assessed, statistical significance regarding multigroup comparison was determined by 1-way ANOVA followed by the Bonferroni post hoc test. When variables violated the assumptions for variance homogeneity, significance was alternatively evaluated by the Kruskal-Wallis ranking test and Mann–Whitney post hoc comparisons adjusted by the Holm method.
A detailed description of the methods is available in the supplemental materials, available online at http://atvb.ahajournals.org.
Measurement of Hemodynamic Changes
Doppler flow examination of the porcine right brachial artery showed that EECP dramatically enhanced diastolic blood flow velocity but produced no significant change in the systolic flow velocity. This resulted in a 36.30% increase in mean flow velocity (29.08±2.01 versus 21.34±1.69 cm/s, P=0.007<0.05), as well as a 34.38% increase in mean wall shear stress (27.46±8.02 versus 20.43±6.75 dyne/cm2, P=0.019<0.05). The PI during EECP was significantly lower compared with before EECP (1.52±0.06 versus 1.83±0.13, P=0.045<0.05), suggesting that EECP decreases the peripheral arterial resistance (details available in the supplemental online materials). We also observed a nonsignificant trend toward a modest decrease in peak systolic flow volume during EECP in the left common carotid artery compared with before EECP (362.31±38.19 versus 431.79±33.68 mL/min, P>0.05). In contrast, the diastolic peak flow volume was significantly elevated by EECP (540.13±57.31 versus 217.14±33.68 mL/min, P<0.001). During EECP therapy, there were 2 pulses per cardiac cycle, essentially doubling the number of arterial pulsations. EECP also significantly elevated blood pulse flow from 306.82±42.37 to 457.63±48.63 mL/min (P=0.041<0.05), demonstrating an increase in flow pulsatility by both pulsation frequency and magnitude.
Changes in Serum Cholesterol Level
At the end of study, compared with the normal control group, the 15-week high-cholesterol atherogenic diet resulted in a significant elevation of serum cholesterol level by 4.79-fold in the CHOL group (P<0.001) and 3.71-fold in the CHOL* EECP group (P<0.001). At the 4 sampling time points of 0, 8, 12, and 15 weeks, there were no significant differences in cholesterol levels between the CHOL group and the CHOL* EECP group (P>0.05), as shown in Figure 1A.
Analysis of the Extent and Composition of Arterial Atherosclerotic Lesions
Sudan IV en face staining and Oil Red O lipid staining revealed remarkable atherosclerotic lesions in the abdominal aortas from hypercholesterolemic control animals, especially at branch points, whereas the hypercholesterolemic animals receiving EECP exhibited a significant decrease in their percentage of sudanophilic plaque area (3.33±0.60% versus 12.03±2.15%, P<0.001 versus CHOL group), as shown in Figure 1.
The LAD atherosclerotic lesions were characterized by vascular endothelial cell (VEC) transfiguration and desquamation, smooth muscle cell (SMC) hyperplasia and migration, lipid deposition, elastic fibrin and collagen proliferation, and disarrangement in the intima and partial media, as shown in Figure 2. Typical necrotic core and fibrous cap were seldom observed, indicating that the early stage of atherosclerosis was induced by 7-week hypercholesterolemia. Quantitative microscopic inspection revealed that the LAD lesion to media area ratio was as high as 26.57±4.56% in the CHOL group but was only 7.32±1.59% in the animals receiving EECP (P<0.001 versus CHOL group), demonstrating the atheroprotective effect of EECP. EECP animals also presented a 53.78% reduction of SMC fraction (P<0.01<0.05 versus CHOL group) and a 30.44% reduction of collagen content (P=0.01<0.05 versus CHOL group), suggesting that lesion composition has been altered by EECP.
Coronary Artery Surface Morphology and Macrophage Accumulation
Scanning electron microscopic examination demonstrated many leukocytes (mostly monocytes) adherent to the endothelium of the coronary artery of hypercholesterolemic animals, whereas in the animals receiving EECP, there was less cellular adherence on the arterial endothelium, as shown in Figure 3A.
Macrophage accumulation in the coronary arteries was determined as an index of atherosclerotic burden and also an indicator of inflammatory response. The macrophage-positive staining was much more intense in the LADs of hypercholesterolemic control animals, whereas EECP animals demonstrated a 53.63% reduction in the positive area ratio (5.58±0.49 versus 9.30±0.71%, P<0.001; Figure 3B).
Proinflammatory Molecules in Coronary Artery and Aorta
As shown by immunohistochemical study in Figures 3C and 5⇓C, the brown positive staining of VCAM-1 and iNOS were much stronger in the LADs of hypercholesterolemic control animals and weaker in EECP animals.
Assessed by TaqMan Real Time-PCR, the VCAM-1 mRNA and iNOS mRNA expressions were significantly enhanced in the aortas of the hypercholesterolemic control animals (P=0.02<0.025 for VCAM-1 and P=0.004<0.017 for iNOS versus CHOL group, respectively). In animals receiving EECP, VCAM-1 mRNA levels were significantly decreased (P=0.002<0.017 versus the CHOL group). There was a trend toward lower iNOS mRNA levels in animals receiving EECP (P>0.05 versus CHOL group).
Arterial Complement Deposition and CRP Expression
We found marked deposition of C3a in the LAD atherosclerotic lesions of hypercholesterolemic control animals. The EECP animals demonstrated a 30.06% reduction of C3a positive area ratio compared with hypercholesterolemic control animals (12.98±0.66 versus 18.56±0.90%, P =0.001), as shown in Figure 4A, indicating a modifying effect of EECP on complement activation. C3a expression correlated well with the occurrence of atherosclerotic lesions.
As shown in Figure 4, the coronary arteries of hypercholesterolemic control animals presented extensive immunochemical CRP staining in the intima, media, and adventitia, with the greatest intensity on the edge of plaques. However, less staining was observed in animals receiving EECP. Western blot analysis demonstrated that aortic CRP protein expression was markedly elevated in the CHOL group (P=0.002<0.017 versus control group), whereas it was significantly decreased in the animals receiving EECP (P=0.009<0.025 versus CHOL group).
MAPK-p38 Activation in Coronary Artery and Aorta
Micrographs of porcine LADs with phospho-p38 antibody immunohistochemical staining for the 3 groups are shown in Figure 5. Phosphorylated (activated) p38 protein was localized mainly in the nuclei of endothelial cells and smooth muscle cells of hypercholesterolemic animals, with the greatest intensity on the edge of plaques. In LADs from the EECP animals, less staining was observed. Consistent with the immunohistochemical study, Western blot analysis demonstrated the p38 activity (expressed as the ratio of phospho-p38 to total p38) was significantly higher in the CHOL group (P<0.001 versus control group) and significantly lower in the CHOL*EECP group compared with the CHOL group (P=0.012<0.025). These results indicated that the abnormal phosphorylation of p38 induced by hypercholesterolemia could be suppressed by EECP treatment.
NF-κB Activation and IκB Degradation in Coronary Artery and Aorta
As shown in Figure 6, the confocal laser scanning fluorescence microscope images of NF-κB p65, visualized in green, were mainly confined to the cytoplasm in the animals fed normal chow, consistent with a quiescent state. In the LAD lesions of the CHOL group, NF-κB p65 protein was abundant not only in the cytoplasm but also in the nuclei (red plus green, visualized as yellow), showing increased nuclear translocation and activation. Decreased nuclear immunofluorescent intensity could be observed in animals receiving EECP, indicating pronounced inhibition of NF-κB p65 activation. Western blotting revealed that the amount of phospho-NF-κB p65 protein was significantly increased in the CHOL group (P<0.001 versus control group). The phospho-NF-κB p65 and total NF-κB p65 protein levels were significantly decreased in animals receiving EECP compared with the CHOL group (P=0.014<0.025, respectively). IκB protein degradation was significant in the CHOL group (P=0.002<0.017 versus control group), and an improved trend was demonstrated in animals receiving EECP (P>0.05). These results suggested that the NF-κB overexpression induced by hypercholesterolemia was suppressed by EECP treatment.
Atherosclerosis is a chronic lipid-driven inflammatory disease of arteries. Within the atherosclerotic lesions, the accumulation and proliferation of macrophages and T lymphocytes lead to enhanced cytokine secretion and activation of intracellular NF-κB and MAPK signaling pathways, which cooperate to induce more cytokines and other acute-phase reactants to further amplify inflammation.12 The complement also participates in the development of atherosclerosis. In vessels, the complement system may be activated by antigen-antibody immune complexes, CRP, modified lipoproteins, apoptotic cells, and cholesterol crystals. Because C3 is the central component in complement activation, inhibition of C3 activation leads to the blockade of biologically active end products, including C5b-9, C5a, and C3a,13,14 as well as the retardation of atherosclerosis.15
It has been widely appreciated that CRP is more than a biomarker of inflammation. Because of its capacity to activate complement, stimulate VCAM-1 to accelerate monocyte-endothelial cell adhesion, and promote uptake of oxidized LDL, CRP plays a direct proatherogenic role.16 CRP can active NF-κB and MAPK signaling pathways and evoke an inflammatory endothelial phenotype by decreasing endothelial nitric oxide synthase expression and activity.17 Studies find that the synthesis and secretion of CRP in the atherosclerotic lesion by paracrine/autocrine loops can result in local concentrations of CRP far in excess of plasma concentrations.
Hemodynamic shear stress may be the most pivotal factor affecting atherogenesis.18 Accumulating evidence shows that the magnitude of shear stress at the arterial wall is inversely correlated with susceptibility to atherosclerosis.19 Many endothelial genes contain shear stress–responsive elements in their promoters, which are inducible by shear stress. At sites susceptible to atherosclerosis, low time-averaged or oscillatory shear stress stimulates the endothelial cells to produce adhesion, vasoconstrictor, growth and coagulation molecules, which promote atherosclerosis in synergy with other systemic risk factors. At sites resistant to atherosclerosis, prolonged high shear stress induces the endothelial cells to express antiatherosclerotic genes, which suppress endothelial proinflammatory activation and inhibit cell cycles via multiple mechanisms, including modulating the activities of MAPKs and NF-κB signaling pathways.20,21 Experimentally, high shear stress with a clear direction causes only transient activation of proinflammatory and proliferative signaling pathways, which is downregulated, even to below the pre-shear levels, when such high shear stress is sustained.22,23 Our study supports these findings by demonstrating that EECP increases shear stress to the endothelium, attenuates vascular CRP expression, decreases macrophage accumulation and complement activation, and eventually ameliorates vascular inflammation.
Not only the magnitude of shear stress but also the flow patterns can influence the biology and pathology of endothelium. Studies have revealed that compared with steady laminar flow, pulsatile flow is more effective in suppressing monocyte adhesion to endothelium exposed to oxidized lipids and upregulating endothelial nitric oxide synthase to produce NO. Pulsatile shear stress exerts a greater atheroprotective effect.24 Furthermore, the duration of exposure to shear stress also influences endothelial gene expression. Acute shear stress induces genes related to endothelial activation, whereas under chronic shear stress, more endothelial genes are suppressed than induced, especially the proinflammatory or atherogenic genes.25 These findings suggest that any therapeutic approaches directed at optimizing arterial wall shear stress may be beneficial. But few methods have this kind of hemodynamic effect in vivo, and fewer could be used in human beings. Among them, regular physical exercise and control of hypertension can be considered as such modalities. The progression of atherosclerosis can be inhibited and even reversed in some patients who carry out regular physical exercise and modify their cardiovascular risk factors. Longitudinal regular exercise training demonstrates a long-term antiinflammatory effect by improving endothelial function and reducing inflammatory markers.26 In a manner similar to regular physical exercise, EECP is able to increase cardiac output and blood flow to change the hemodynamic patterns in the lesion-prone area. Our study provides some evidence that long-term EECP treatment increases arterial shear stress and pulsatility in vivo chronically, attenuating the progression of atherosclerosis.
Previous studies have shown that in patients with refractory angina, EECP treatment reduces arterial stiffness and improves arterial wall properties and wave reflection characteristics.27 In the present study, hypercholesterolemic animals subjected to ≈34 hours of EECP over a 7-week period displayed a significant shear stress augmentation in their brachial arteries, associated with a decreased PI. The PI reflects downstream vascular resistance of arteries and can be used as an estimate of vascular impedance and severity of vascular damage. PI values ≥1.2 are considered to indicate increased distal resistance to blood flow in human beings.28 Low mean velocity and high PI have been documented in multiple vessels by Doppler in patients with ischemic stroke diffuse intracranial disease, diabetes, transient monocular blindness, and cigarette smoking.29,30 In our study, although a normal range of PI has not been defined in the porcine model, significantly higher mean flow velocities and lower PI in the brachial arteries of hypercholesterolemic pigs were observed during EECP compared with baseline. This is indicative of vasodilation and reduced peripheral vascular resistance, contributing to the beneficial hemodynamic effects of EECP.
Many survival signaling pathways sensitive to atheroprone stimuli converge on MAPKs, including the p38 cascade, which regulates inflammation through activation of proinflammatory transcription factors such as NF-κB. Phosphorylation and activation of MAPKs was observed in endothelial cells exposed to low shear stress and mediated low shear stress–induced proinflammatory gene expression. Shear stress has also been found to modulate the expression of NF-κB. Analysis of vascular regions prone to atherogenesis revealed that low or disturbed shear stress primes the endothelial cells to respond to proatherogenic stimuli through upregulation of NF-κB.31
Our current data provide evidence that limiting MAPK-p38 and NF-κB activation in plaques could represent a molecular mechanism by which EECP modulates inflammatory responses to hypercholesterolemia, contributing to the observed clinical benefits of EECP.
EECP is a noninvasive device capable of improving blood flow, shear stress, and pulsatility systematically, thereby modulating the inflammatory response to hypercholesterolemia. Long-term EECP intervention may be an effective alternative therapeutic strategy to protect the cardiovascular system from inflammation and atherosclerosis.
We thank Ruide Hu, Huixi Wu, Lihong Che, Jinjun Zhang, Guifang Lin, Gang Dai, Luguang Liang, Minzhe Feng, and Yuetao Qian for excellent technical assistance.
Sources of Funding
This study was supported by National Natural Science Foundation (No. 50113-4103048), National Natural Science Foundation (No. 30871047), and the 10th National “Five-Year” Key Project (No. 2001BA706B07), China.
John C. K. Hui served as chief technology officer and senior vice president of Vasomedical and is the principal inventor of several patents related to enhance external counterpulsation (EECP).
Received on: September 24, 2009; final version accepted on: January 19, 2010.
de Winther MPJ, Kanters E, Kraal G, Hofker MH. Nuclear factor κB signaling in atherogenesis. Arteriolscler Thromb Vasc Biol. 2005; 25: 904–914.
Michaels AD, Accad M, Ports TA, Grossman W. Left ventricular systolic unloading and augmentation of intracoronary pressure and Doppler flow during enhanced external counterpulsation. Circulation. 2002; 106: 1237–1242.
Zhang Y, He XH, Chen XL, Ma H, Liu DH, Luo JY, Du ZM, Jin YF, Xiong Y, He JG, Fang DQ, Wang KJ, Lawson WE, Hui JCK, Zheng ZS, Wu GF. Enhanced external counterpulsation inhibits intimal hyperplasia by modifying shear stress: responsive gene expression in hypercholesterolemic pigs. Circulation. 2007; 116: 526–534.
Irace C, Cortese C, Fiaschi E, Carallo C, Farinaro E, Gnasso A. Wall shear stress is associated with intima: media thickness and carotid atherosclerosis in subjects at low coronary heart disease risk. Stroke. 2004; 35: 464–468.
Sharma VK, Tsivgoulis G, Lao AY, Malkoff MD, Alexandrov AV. Noninvasive detection of diffuse intracranial disease. Stroke. 2007; 38: 3175–3181.
Shagdarsuren E, Bidzhekov K, Djalali-Talab Y, Liehn EA, Hristov M, Matthijsen RA, Buurman WA, Zernecke A, Weber C. C1-esterase inhibitor protects against neointima formation after arterial injury in atherosclerosis-prone mice. Circulation. 2008; 117: 70–78.
Schepers A, de Vries MR, van Leuven CJ, Grimbergen JM, Holers VM, Daha MR, van Bockel JH, Quax PHA. Inhibition of complement component C3 reduces vein graft atherosclerosis in apolipoprotein E3-Leiden transgenic mice. Circulation. 2006; 114: 2831–2838.
Leung VWY, Yun S, Botto M, Mason JC, Malik TH, Song W, Paixao-Cavalcante D, Pickering MC, Boyle J, Haskard DO. Decay-accelerating factor suppresses complement C3 activation and retards atherosclerosis in low-density lipoprotein receptor-deficient mice. Am J Pathol. 2009; 175: 1757–1767.
Paul A, Ko KWS, Li L, Yechoor V, McCrory MA, Szalai AJ, Chan L. C-reactive protein accelerates the progression of atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 109: 647–655.
Verma S, Devaraj S, Jialal I. Is C-reactive protein an innocent bystander or proatherogenic culprit? C-reactive protein promotes atherothrombosis. Circ. 2006; 113: 2128–2151.
Caro CG. Discovery of the role of wall shear in atherosclerosis. Arteriolscler Thromb Vasc Biol. 2009; 29: 158–161.
Cheng C, Tempel D, van Haperen R, van der Baan A, Grosveld F, Daemen MJAP, Krams R, de Crom R. Atherosclerotic lesion size and vulnerability are determined by patterns of fluid shear stress. Circulation. 2006; 113: 2744–2753.
Hsiai TK, Cho SK, Reddy S, Hama S, Navab M, Demer LL, Honda HM, Ho CM. Pulsatile flow regulates monocyte adhesion to oxidized lipid-induced endothelial cells. Arteriolscler Thromb Vasc Biol. 2001; 21: 1770–1776.
Chen BPC, Li YS, Zhao Y, Chen KD, Li S, Lao J, Yuan S, Shyy JYJ, Chien S. DNA microarray analysis of gene expression in endothelial cells in response to 24-h shear stress. Physiol Genomics. 2001; 7: 55–63.
Pietryga M, Brazert J, Wendwe-Oegowaska E, Biczysko R, Dubiel M, Gudmundsson S. Abnormal uterine Doppler is related to vasculopathy in pregestational diabetes mellitus. Circulation. 2005; 112: 2496–2500.
Chao AC, Hsu HY, Chung CP, Chen YY, Yen MY, Wong WJ, Hu HH. Altered retrobulbar hemodynamics in patients who have transient monocular blindness without carotid. stenosis. Stroke. 2007; 38: 1377–1379.
Lim MH, Cho YI, Jeong SK. Homocysteine and pulsatility index of cerebral arteries. Stroke. 2009; 40: 3216–3220.
Chiu JJ, Chen LJ, Chang SF, Lee PL, Lee CI, Tsai MC, Lee DY, Hsieh HP, Usami S, Chien S. Shear stress inhibits smooth muscle cell-induced inflammatory gene expression in endothelial cells: role of NF-κB. Arteriolscler Thromb Vasc Biol. 2005; 25: 963–969.