Marys Medicine

Doi:10.1016/j.ijcard.2007.03.00

International Journal of Cardiology 125 (2008) 376 – 382 Association of endothelial function and vascular data with LDL-c and HDL-c in a homogeneous population of middle-aged, healthy military men: Evidence for a critical role of optimal lipid levels ☆ Martin Laclaustra a,⁎, Alejandro F. Frangi b, Andres G. Frangi a, Jose A. Casasnovas a, Pedro Cia a a Cardiovascular Research Group of Aragon, Aragon Institute of Health Sciences, Hospital Clinico Universitario "Lozano Blesa", Zaragoza, Spain b Computational Imaging Lab, Department of Technology, Pompeu Fabra University, Barcelona, Spain Received 17 May 2006; received in revised form 18 February 2007; accepted 25 March 2007 Available online 2 May 2007 Background: Modifying lipids levels underpins atherosclerosis prevention. Flow-mediated dilation (FMD) could advise which patients totreat and to what extent. Little is known about the influence of near-normal lipid levels on the endothelium and the mechanisms related todifferent lipid fractions. We studied associations between FMD and lipids, focusing on normal lipid levels.
Methods: An age-homogeneous sample of 171 healthy, untreated military men (mean age 35.5 ± 1.1 years) was studied: serum lipiddetermination and brachial artery ultrasound with a forearm ischemia cuff and automated measurement were performed. NCEP-ATP IIIgroups were used.
Results: Significantly smaller vessel diameters were found among individuals with high HDL-c (4.10 mm vs. 4.24 mm), optimal LDL-c(4.00 mm vs. 4.22 mm), and normal triglycerides (b150 mg/dl) (4.15 mm vs. 4.31 mm). Basal diameter correlated significantly with HDL-cand triglycerides. There were significant differences in FMD between low HDL-c compared to the rest (4.13% vs. 5.07%) and betweenoptimal and near-optimal LDL-c compared to the rest (5.28% vs. 4.56%). HDL-c and LDL-c correlated with FMD. The inverse relation ofhigh LDL-c and FMD is partially due to a decreased stimulus. Besides, stimulus heterogeneity may mask HDL-c link with FMD.
Conclusion: Those subjects naturally (not pharmacologically) in the healthy tail-end of the lipid distributions have the best endothelialfunction and smaller vessels. Functional vascular remodeling might precede anatomical remodeling and, in early stages, vessel size should beconsidered a risk indicator rather than an atherosclerotic sign. Furthermore, controlling the stimulus seems necessary for detecting therelationship between HDL-c and FMD, and should be performed regularly.
2007 Elsevier Ireland Ltd. All rights reserved.
Keywords: Endothelial function; Flow-mediated dilation; Lipids; Atherosclerosis ☆ Funding: Dr. Laclaustra is supported by the Instituto de Salud Carlos III, Spanish Ministry of Health, Spain. The AGEMZA Project was supported by Endothelial dysfunction has been described as the a grant (FIS 99/0600) from the Spanish Ministry of Health, Spain. The intermediate link between cardiovascular risk factors and the clinical research was also partially supported by a grant of the DiputacionGeneral de Aragon (P58/98), Spain. Dr. Frangi is supported by a Ramon y development of atherosclerosis When present, endothe- Cajal Research Fellowship and by Grants TEC2006-03617 from the Spanish lial responses are impaired all over the vascular tree. It is also MEC, CENIT-CDTEAM from CDTI, and FIS 04/040676 from ISCIII. AF is believed that microcirculation and conduit vessels are both also affiliated to the ISCIII CIBER-BBN (CB06/01/0061).
affected when endothelial dysfunction is present and that ⁎ Corresponding author. Aragon Institute of Health Sciences, Avda, both vascular beds could be equally used for exploring Gomez Laguna, 25, 3°, E-50009 Zaragoza, Spain.
E-mail address: (M. Laclaustra).
endothelial health. Nevertheless, atherosclerotic clinical 0167-5273/$ - see front matter 2007 Elsevier Ireland Ltd. All rights reserved.
M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 Fig. 1. Diameter and blood velocity curves as a function of time in a typical FMD test. The solid black line depicts diameter. The dotted gray line depicts bloodvelocity. The dotted horizontal line indicates the reference baseline diameter. Distal ischemia period is denoted as CUFF. NTG indicates the instant ofnitroglycerin administration.
diseases occur mainly in conduit vessels; therefore, different 2.1. Endothelial function testing responses to pro-atherosclerotic factors of the endothelium inconduit and in microcirculation vessels might be relevant.
Peripheral endothelial function was assessed using brachial Flow-mediated dilation (FMD) is a widely used method artery ultrasound with a flow-mediated dilation protocol that reflects endothelial dysfunction in conduit arteries.
similar to that originally described by Celermajer et al.
Modifying lipids levels underpins atherosclerosis preven- This followed the main guidelines of the International Brachial tion. FMD could advise which patients to treat and to what Artery Reactivity Task Force and the Working Group on extent. An association between lipid levels and endothelial Endothelin and Endothelial Factors of the European Society of function has been described, but there are few studies dealing Hypertension . Longitudinal images of the brachial artery with the influence of normal and near-normal lipid levels on above the antecubital fossa were continuously recorded at the endothelium in non-treated subjects. Very low LDL-c baseline and through vasodilator stimuli. Distal ischemia was levels in coronary patients under treatment associate with an induced by means of a pneumatic cuff inflated in the fore- increased FMD , but in general research has been focused arm (downstream of the studied portion of the vessel) to on the opposite extreme of lipid values .
suprasystolic pressure for 5 min. Eight minutes later, sub- We determined the association of low LDL-c levels and lingual nitroglycerin (25 μg administered in solution with a high HDL-c levels with endothelial function in healthy pipette) was used as an endothelium-independent vasodilator.
subjects. We explored it in terms of FMD and other relatedvascular parameters. We also explored lipids' influencewhen considering FMD in relation to flow stimulus.
Sample characteristics A cohort of military men underwent a thorough preventive Standard Deviation cardiovascular examination in the AGEMZA study . Inthis group, the population's age was around 35 years.
Participants with major clinical disease or undergoing pharma- cological treatments were excluded from further analysis (two Body mass index (kg/m2) subjects, one for each reason). All subjects gave written, in- Systolic blood pressure (mm Hg) formed consent to participate.
Diastolic blood pressure (mm Hg) This was a cross-sectional observational study. Health Cholesterol (mg/dl) Triglycerides (mg/dl) status and previously known diseases, habits, and family history were obtained through a questionnaire. Weight, height, and blood pressure were measured in accord with current quality standards. Fasting serum samples were Arterial diameter (μm) obtained from the subjects and analyzed in the University Nitroglycerin mediated dilation (%) Clinical Hospital "Lozano Blesa" (Zaragoza, Spain). Anal- Basal blood velocity (cm/s) yses were performed for total cholesterol, triglycerides, and Peak blood velocity in hyperemia HDL-c by standard enzymatic laboratory techniques. LDL-c was calculated using the Friedewald formula in subjects Average blood velocity during the first 54.33 ± 13.47 whose triglycerides levels were less than 400 mg/dl.
minute of hyperemia (cm/s) M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 Images of the brachial artery were obtained with a after cuff release, and another in the region of the curve prior to multifrequency (5.5–7.5 MHz) linear-array vascular trans- the administration of nitroglycerin. Flow-mediated and ducer on a Hewlett-Packard Sonos 4500 Imaging System nitroglycerin vasodilations were calculated as the peaks (Agilent Technologies Inc., Andover, Massachusetts, USA).
of diameter change measured as a percentage of baseline Vessel blood velocity was monitored with pulse-wave Doppler interrogation to detect reactive hyperemia. The From the Doppler velocity distribution chart provided by images of the vessel and the Doppler spectrum were digitized the echograph, we acquired time-averaged maximum velocity in end-diastole using a frame grabber.
measurements throughout the test using an automated A computerized image analysis method was used on the technique similar to that described by Tschirren et al.
image sequences . The method is based on pair-wise Basal blood velocity was defined as the average velocity image matching (registration). Registration provides a scaling during the initial minute at the beginning of the test. Peak value that models the transversal deformation of the artery hyperemia blood velocity was defined as the maximum occurring between two points in time. This method is more velocity after cuff release. Average hyperemic blood velocity robust to speckle noise than other edge-based techniques and was defined as the average velocity during the first minute of assesses the vasodilation curve more precisely than other hyperemia and used to evaluate the stimulus.
automated methods. To obtain a curve of absolute values(), the diameter of the initial artery image of each study 2.2. Statistical analysis was measured with a computerized tool by manuallyfitting a curve to each vessel wall and calculating the mean Data were analyzed with SPSS (Version 10.0, SPSS, Inc., distance between them. For each vasodilation curve, two Chicago, Illinois). Mean comparisons (t test) and Pearson's reference baseline diameters were established: one before correlation were computed to demonstrate associations be- vasodilation in a region free of motion artifacts selected just tween FMD and lipid fractions. Multivariate linear Fig. 2. Average basal artery diameter for each one of the National Cholesterol Education Program's lipid categories and its 95% confidence interval. Thosefigures at the bottom of the bars are the depicted average values and those above the graphic the number of subjects in each category.
M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 regression analysis was employed to model FMD, control- represented ("n" values of ). Among them, ling for the hyperemic stimulus. We divided the cohort into only 8% had high blood pressure. For those parameters with groups according to lipid levels as defined by the National non-normal distributions, medians and interquartile ranges Cholesterol Education Program Adult Treatment Panel III were: triglycerides 103 (69) mg/dl and FMD 4.43 (2.79)%.
in order to compare FMD and other vascular parameters(basal diameter and basal blood velocity) among them. In all 3.2. Relationship between flow-mediated dilation and lipids analyses, statistical significance was assumed when p b 0.05(two-tailed analysis). Statistical tests were conducted on Lipid levels were associated with basal vascular condi- transformed variables (square root of FMD, abbreviated tions. Arterial diameter tended to be smaller under better lipid sqrFMD; and logarithm of triglycerides) in those variables conditions, as can be seen in . We split the sample into that did not have normal distributions. In this article, two groups according to thresholds derived from the figure for continuous random variables are characterized as mean ± each lipid fraction. Individuals with HDL-c N60 mg/dl (high) standard deviation.
showed significantly smaller diameters (4095 μm vs.
4237 μm; p = 0.033) than the rest. Individuals with LDL-c b100 mg/dl (optimal) showed significantly smaller diameterstoo (4001 μm vs. 4223 μm; p = 0.007). The same happened 3.1. Sample characteristics for subjects with triglycerides b150 mg/dl (normal) (4149 μmvs. 4311 μm; p = 0.035). Accordingly, a significant linear A total of 171 subjects agreed to participate, satisfied the correlation was found for basal diameter with HDL-c (r = inclusion criteria, and fulfilled the imaging protocol for −0.163; p=0.033) and with the logarithm of triglycerides FMD. Their characteristics are reported in The (r = 0.234; p = 0.002). Basal flow, represented by basal blood sample corresponds to a population with favorable lipid velocity, also tended to rise as the lipid profile worsened levels, so pathological extreme values were scarcely (graphs are not shown) but no significant differences were Fig. 3. Average FMD for each one of the National Cholesterol Education Program's lipid categories and its 95% confidence interval. Those figures at the bottomof the bars are the depicted average values and those above the graphic the number of subjects in each category.
M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 found when comparing basal flow of the optimal groups with higher correlation coefficients than those studies that, like the rest of the sample.
ours, are based on a more homogeneous and healthier shows that FMD was also related to total population. The only study that is similar to ours is that of cholesterol and its fractions and presented a gradient.
Jensen-Urstad et al. and it reports moderate coefficients.
Again, we split the sample into two groups according to Thus, the modest value of the correlation coefficients in our thresholds derived from the figure for each lipid fraction.
study, when compared to those previously reported, can be There were significant differences between low HDL-c due to a shortened range of lipid levels and to a small number compared to the rest (means: 4.13% vs. 5.07%; medians: of extreme pathological values. In addition, our sample was 3.77% vs. 4.49%; sqrFMD t test: p = 0.042) and between restricted to men with a narrow age range, which may further optimal and near-optimal LDL-c compared to the rest reduce the variability of FMD in the sample.
(means: 5.28% vs. 4.56%; medians: 4.81% vs. 4.02%; A trial with statins, performed on a small sample with sqrFMD t test: p = 0.013). Furthermore, a significant linear LDL-c levels b160 mg/dl, showed endothelial function correlation was found for sqrFMD with HDL-c (r = 0.163; improvement in healthy subjects . Reaching very low p = 0.034) and with LDL-c (r = −0.201; p = 0.008).
levels in coronary patients undergoing treatment with statinsalso rendered better endothelial results In our study, 3.3. Multivariate analysis subjects with spontaneously low LDL-c levels also presentedbetter endothelial health. This reassures these levels as In the FMD test, endothelial function is assumed to be therapeutic goals, independently of the drug used to reach related to the vasodilator response of endothelium after a them. The same trends could be observed for HDL-c.
standard flow stimulus. However, intensity of the actual The presence of wider vessels has been described in high- stimulus is heterogeneous among subjects. This, represented risk and atherosclerotic individuals. In 55 year-old by average hyperemic blood velocity (dependent on small individuals, common carotid artery diameter correlated vessels' response to ischemia) correlated linearly with significantly with total cholesterol, HDL-c, and triglycerides sqrFMD (r = 0.369, p b 0.001).
In spite of this, until now there was no evidence that this We adjusted for average hyperemic blood velocity in a relationship between diameter and lipids also applies to normal linear regression model for sqrFMD explained by LDL-c. Its and optimal lipid levels in middle-aged people. We have found standardized beta coefficient decreased from −0.201 to just this evidence for HDL-c, LDL-c, and triglycerides.
−0.169. Conversely, the coefficient of HDL-c increased An expansive atherosclerotic artery remodeling has been from 0.163 to 0.205 after accounting for blood velocity in the described by other researchers However, before that bivariate model. In both cases, the lipid fraction under anatomic change occurs, some sort of toxic or compensatory consideration remained a significant term after adjustment.
reactive vasodilation must exist in patients with high Interestingly, a model trying to explain square-root FMD atherosclerotic lipid levels. Furthermore, differences appear with both cholesterol fractions, presented HDL-c as non- in vessel size soon after losing the theoretically optimal significant. After controlling for the blood velocity, the (near-fetal) levels. Hence, vessel size alone should be importance of the fractions reversed: HDL-c was significant considered a risk indicator rather than an atherosclerotic and LDL-c was not. Finally, we noticed that adjustment for signal as previously suggested Vessel size could be a BMI did not change any of the described models.
marker of endothelial dysfunction. In dyslipidemia, com-pensatory mechanisms could activate the nitric oxide pathway even in absence of shear stress; therefore, reserveregulatory capacity would be reduced when physiological We observed significantly different FMD for individuals stimuli appear. This hypothesis will have to be confirmed by with low HDL-c compared to the rest and also different future research. Probably similar changes happen in FMD for individuals with optimal and near-optimal LDL-c compared to the rest. A coherent progression was evidenced FMD changes with gender and age Usually, this along the range of lipid levels, and cholesterol fractions makes it difficult to discover subtle associations. Strategi- correlated significantly with FMD.
cally, our sample was homogeneous in terms of these A relationship between FMD and HDL-c and LDL-c has variables and we could focus multivariate analysis in con- already been shown, describing mainly differences between trolling the stimulus of the test, thus going one step further patients with extreme levels versus the normal population.
than previous work. Our cross-sectional study is not con- Most studies are based on pooled samples combining high- clusive for causality but we hypothesize here such a risk or diseased individuals (diabetes , coronary heart connection to discuss results from multivariate analysis.
disease , hypercholesterolemia , hypertryglicer- Controlling for the hyperemic stimulus increased HDL-c idemia , hypoalphalipoproteinemia and normal association strength with FMD but decreased that of LDL-c.
individuals, or even samples entirely made up of high-risk This finding certainly suggests that both lipids exert their individuals (obese , hyperlipidemic , suspected influence on the single parameter FMD through different coronary heart disease These scenarios render M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 We observed that LDL-c levels importantly influence from the Spanish Ministry of Health. The clinical research FMD, but part of this effect is through changes in was also partially supported by a grant from the Diputacion microcirculation's response to ischemia, that is to say, in General de Aragon (P58/98). Dr. Frangi is supported by a reactive hyperemia. Reactive hyperemia also represents Ramon y Cajal Research Fellowship and by Grants endothelial health to some extent but in a different TEC2006-03617 from the Spanish MEC, CENIT-CDTEAM vascular territory. After controlling this stimulus, HDL-c took from CDTI, and FIS 04/040676 from ISCIII. AF is also over influencing FMD. Therefore, HDL-c seems to actually affiliated to the ISCIII CIBER-BBN (CB06/01/0061). We have an impact on the intensity of the endothelial response to thank the Spanish Ministry of Defense for authorizing this increased shear stress in conduit vessels. To explore these study. We acknowledge the support of and collaboration HDL-c influences, therefore, hyperemia should necessarily be from the units that have made this research possible.
measured, as heterogeneity in the stimulus may mask theHDL-c link with FMD. Moreover, it is not possible to detect them using the endothelial explorations based exclusivelyon measuring post-ischemic reactive hyperemia, such as post- [1] Ross R. The pathogenesis of atherosclerosis: a perspective for the ischemic finger measurements . In the same way, also 1990s. Nature 1993;362:801–9.
some variants of FMD, like upstream cuff probably blunt [2] Reddy KG, Nair RN, Sheehan HM, Hodgson JM. Evidence that this association because dilation also depends on ischemic selective endothelial dysfunction may occur in the absence of angiographic or ultrasound atherosclerosis in patients with risk factorsfor atherosclerosis. J Am Coll Cardiol 1994;23:833–43.
Other mechanisms have been described for the HDL-c [3] Irace C, Ceravolo R, Notarangelo L, et al. Comparison of endothelial effect on FMD. Association of nitroglycerin-mediated function evaluated by strain gauge plethysmography and brachial dilation with HDL-c was described by previous studies artery ultrasound. Atherosclerosis 2001;158:53–9.
and a reduced smooth muscle sensitivity associated with low [4] Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the HDL-c was also observed in diabetic patients . Hints to ultrasound assessment of endothelial-dependent flow-mediated vaso-dilation of the brachial artery: a report of the International Brachial association between HDL-c and nitroglycerin-mediated Artery Reactivity Task Force. J Am Coll Cardiol 2002;39:257–65.
dilation were observed in our analysis (graphs are not [5] Kuvin JT, Patel AR, Sliney KA, Pandian NG, Karas RH. Comparison shown), but the differences were not statistically significant.
of flow-mediated dilatation of the brachial artery in coronary patients One more possible mechanism would be a reduced reserve of with low-density lipoprotein cholesterol levels b80 mg/dl versus the nitric oxide system due to an increased basal endothelium patients with levels 80 to 100 mg/dl. Am J Cardiol 2005;95:93–5.
[6] Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive activation , as we mentioned before.
detection of endothelial dysfunction in children and adults at risk ofatherosclerosis. Lancet 1992;340:1111–5.
[7] Lewis TV, Dart AM, Chin-Dusting JP. Endothelium-dependent relaxation by acetylcholine is impaired in hypertriglyceridemic To summarize our study: better vascular health, in terms humans with normal levels of plasma LDL cholesterol. J Am CollCardiol 1999;33:805–12.
of endothelial function, was found among those subjects [8] Kuvin JT, Patel AR, Sidhu M, et al. Relation between high-density naturally (not pharmacologically) in the healthy tail-end of lipoprotein cholesterol and peripheral vasomotor function. Am J the lipid distributions, suggesting that at least part of the endothelial health improvement described for aggressive [9] Laclaustra-Gimeno M, Gonzalez-Garcia MP, Casasnovas-Lenguas JA, lipid-lowering drugs can be explained by their effect on et al. Cardiovascular risk factor progression in young males at 15-yearfollow-up in the General Military Academy of Zaragoza (AGEMZA) lipids. In addition, it was remarkable that cholesterol Study. Rev Esp Cardiol 2006;59:671–8.
fractions and triglycerides had a strong influence on vessel [10] Casasnovas JA, Lapetra A, Puzo J, et al. Tobacco, physical exercise diameter even in a healthy population. Favorable levels were and lipid profile. Eur Heart J 1992;13:440–5.
associated with smaller arterial diameters. This discovery has [11] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the pathophysiological implications and deserves further re- concentration of low-density lipoprotein cholesterol in plasma, withoutuse of the preparative ultracentrifuge. Clin Chem 1972;18:499–502.
search. Moreover, we observed differential influences of [12] Deanfield J, Donald A, Ferri C, et al. Endothelial function and cholesterol fractions on vascular function and it will be dysfunction. Part I: Methodological issues for assessment in the necessary to further investigate pathophysiology of these different vascular beds: a statement by the Working Group on differences. Our work also established that controlling the Endothelin and Endothelial Factors of the European Society of flow stimulus seems necessary for detecting the relationship Hypertension. J Hypertens 2005;23:7–17.
[13] Frangi AF, Laclaustra M, Lamata P. A registration-based ap- between HDL-c and FMD, and should be thus performed proach to quantify flow-mediated dilation (FMD) of the brachial artery in ultrasound image sequences. IEEE Trans Med Imaging2003;22:1458–69.
[14] Laclaustra M, Frangi AF, Garcia D, Boisrobert L, Frangi AG, Pascual I.
Parameterization of flow-mediated dilation through principal compo-nent analysis. Physiol Meas 2007;28:301–20.
Dr. Laclaustra is supported by the Instituto de Salud [15] Tschirren J, Lauer RM, Sonka M. Automated analysis of Doppler Carlos III, Spanish Ministry of Health, Spain. The ultrasound velocity flow diagrams. IEEE Trans Med Imaging AGEMZA Project was supported by a grant (FIS 99/0600) M. Laclaustra et al. / International Journal of Cardiology 125 (2008) 376–382 [16] Executive summary of The Third Report of The National Cholesterol [28] Vogel RA, Corretti MC, Plotnick GD. Changes in flow-mediated Education Program (NCEP) Expert Panel on Detection, Evaluation, brachial artery vasoactivity with lowering of desirable cholesterol And Treatment of High Blood Cholesterol In Adults (Adult Treatment levels in healthy middle-aged men. Am J Cardiol 1996;77:37–40.
Panel III). JAMA 2001;285:2486–97.
[29] Holubkov R, Karas RH, Pepine CJ, et al. Large brachial artery diameter [17] Ravikumar R, Deepa R, Shanthirani C, Mohan V. Comparison of is associated with angiographic coronary artery disease in women. Am carotid intima-media thickness, arterial stiffness, and brachial artery Heart J 2002;143:802–7.
flow mediated dilatation in diabetic and nondiabetic subjects (The [30] Jensen-Urstad K, Jensen-Urstad M, Johansson J. Carotid artery Chennai Urban Population Study). Am J Cardiol 2002;90:702–7.
diameter correlates with risk factors for cardiovascular disease in a [18] Li XP, Zhao SP, Zhang XY, Liu L, Gao M, Zhou QC. Protective effect population of 55-year-old subjects. Stroke 1999;30:1572–6.
of high density lipoprotein on endothelium-dependent vasodilatation.
[31] Schoenhagen P, Ziada KM, Vince DG, Nissen SE, Tuzcu EM. Arterial Int J Cardiol 2000;73:231–6.
remodeling and coronary artery disease: the concept of "dilated" [19] Zhang X, Zhao SP, Li XP, Gao M, Zhou QC. Endothelium-dependent versus "obstructive" coronary atherosclerosis. J Am Coll Cardiol and -independent functions are impaired in patients with coronary heart disease. Atherosclerosis 2000;149:19–24.
[32] Celermajer DS, Sorensen KE, Spiegelhalter DJ, Georgakopoulos D, [20] Sorensen KE, Celermajer DS, Georgakopoulos D, Hatcher G, Robinson J, Deanfield JE. Aging is associated with endothelial Betteridge DJ, Deanfield JE. Impairment of endothelium-dependent dysfunction in healthy men years before the age-related decline in dilation is an early event in children with familial hypercholesterolemia women. J Am Coll Cardiol 1994;24:471–6.
and is related to the lipoprotein(a) level. J Clin Invest 1994;93:50–5.
[33] Meredith IT, Currie KE, Anderson TJ, Roddy MA, Ganz P, Creager [21] Lupattelli G, Lombardini R, Schillaci G, et al. Flow-mediated vasoactivity MA. Postischemic vasodilation in human forearm is dependent on and circulating adhesion molecules in hypertriglyceridemia: association endothelium-derived nitric oxide. Am J Physiol 1996;270:H1435–40.
with small, dense LDL cholesterol particles. Am Heart J 2000;140:521–6.
[34] Laclaustra M, Kaski JC, Frangi AF. Flow-mediated dilation: just a [22] Vaudo G, Marchesi S, Lupattelli G, et al. Early vascular damage in marker of local shear stress? Hypertension 2005;45:e11–2 [author primary hypoalphalipoproteinemia. Metabolism 2003;52:328–32.
reply e11–2].
[23] Brook RD, Bard RL, Rubenfire M, Ridker PM, Rajagopalan S.
[35] Bonetti PO, Pumper GM, Higano ST, Holmes Jr DR, Kuvin JT, Usefulness of visceral obesity (waist/hip ratio) in predicting vascular Lerman A. Noninvasive identification of patients with early coronary endothelial function in healthy overweight adults. Am J Cardiol atherosclerosis by assessment of digital reactive hyperemia. J Am Coll [24] Malik J, Melenovsky V, Wichterle D, et al. Both fenofibrate [36] Bots ML, Westerink J, Rabelink TJ, de Koning EJ. Assessment of and atorvastatin improve vascular reactivity in combined hyperlipidaemia flow-mediated vasodilatation (FMD) of the brachial artery: effects of (fenofibrate versus atorvastatin trial-FAT). Cardiovasc Res 2001;52:290–8.
technical aspects of the FMD measurement on the FMD response. Eur [25] Aggoun Y, Bonnet D, Sidi D, et al. Arterial mechanical changes in Heart J 2005;26:363–8.
children with familial hypercholesterolemia. Arterioscler Thromb Vasc [37] Watts GF, O'Brien SF, Silvester W, Millar JA. Impaired endothelium- dependent and independent dilatation of forearm resistance arteries in [26] Lupattelli G, Marchesi S, Roscini AR, et al. Direct association between men with diet-treated non-insulin-dependent diabetes: role of dyslipi- high-density lipoprotein cholesterol and endothelial function in daemia. Clin Sci (Lond) 1996;91:567–73.
hyperlipemia. Am J Cardiol 2002;90:648–50.
[38] Vallance P, Chan N. Endothelial function and nitric oxide: clinical [27] Jensen-Urstad K, Johansson J, Jensen-Urstad M. Vascular function relevance. Heart 2001;85:342–50.
correlates with risk factors for cardiovascular disease in a healthypopulation of 35-year-old subjects. J Intern Med 1997;241:507–13.

Source: http://www.dtic.upf.edu/~afrangi/articles/ijc2008.pdf