Share this article on:

Flow-mediated vasodilation: improving specificity for endothelial function evaluation

Cipollone, Francescoa; Muiesan, Maria Lorenzab

doi: 10.1097/HJH.0b013e32835d0dbd
Editorial Commentaries

aEuropean Center of Excellence on Atherosclerosis, Hypertension and Dyslipidemia, ‘G. d’Annunzio’ University, Chieti-Pescara

bDepartment of Clinical and Experimental Sciences, University of Brescia, Brescia, Italy

Correspondence to Maria L. Muiesan, Clinical and Experimental Sciences Department, University of Brescia, 2° Medicina Generale Spedali Civili 25121, Brescia, Italy. E-mail:

According to the most recent study of WHO, hypertension is estimated to cause over 7 million of deaths every year [1]. It is well known that hypertension represents a major risk factor for both coronary and cerebrovascular events, as it is a key factor for atherosclerosis development and progression. Indeed, its involvement has been proven since the first step of atherogenesis: the endothelial dysfunction.

Despite it is only a monolayer tissue, endothelium is a complex and crucial homeostatic organ that plays a key regulatory role in vascular tone and structure. Endothelial cell stimulation induces the production and release of several vasoactive molecules, including nitric oxide, prostacyclin and others, which are able to reduce vascular smooth muscle cell (VSMC) tone, leukocyte adhesion and homing, and platelet adhesion [2]. Under persistent pathological conditions such as hypertension, endothelial activity is greatly impaired, and endothelium undergoes rapid structural and functional changes that finally result in impaired vasorelaxation, oxidative stress and increased adhesiveness for circulating leukocytes (mainly monocytes and T-lymphocytes). The enhanced adhesion of blood leukocytes to the vessel wall represent a key event in atherogenesis and is a consequence of the upregulation of the expression of cellular adhesion molecules, including intercellular adhesion molecule 1, vascular cell adhesion protein 1 and selectins, in endothelial cells. After adhesion, leukocytes move to the tunica intima and promote self-enhancing inflammatory responses that leads to the development of foam cells, the histological hallmark of early atherosclerosis [3]. As this inflammatory process continues, these cells produce several mediators and growth factors that stimulate proliferating VSMC to synthesize several structural components (including collagen, elastic fibres and proteoglycans), which finally contribute to the progression towards advanced atherosclerotic lesions.

As endothelial dysfunction is an early step in atherogenesis, the possibility to assess endothelial function in vivo by using noninvasive tests represents a great opportunity for the identification of early (and still subclinical) damage in hypertensive patients. Due to its very short half-life and its volatile nature, the evaluation of the circulating level of nitric oxide as an in-vivo index of endothelial function represents a complex strategy in clinical practice. Moreover, endothelial function may greatly differ in different vascular districts and, in the same district, may change according to vessel diameter [2]. Another potential biomarker of endothelial cell dysfunction may be circulating endothelial precursor cells (EPCs), which are released by bone marrow and are apt to maintain and repair endothelial status [4]. Interestingly, their mobilization is regulated by nitric oxide level, and thus, their circulating level is impaired in patients with known cardiovascular risk factors [5]. However, despite their potential interesting clinical read-out, to date, the evaluation of circulating EPCs is still limited to research purposes.

In order to overcome these limitations, several methods based on the evaluation of endothelial-dependent vasodilatation have been proposed. These tests usually involve physiological or pharmacological studies. In particular, the first approach that has correlated endothelial dysfunction and cardiac events was based on the analysis of the change in coronary artery diameter during angiography before and after local infusion of acetylcholine [6]. Indeed, acetylcholine is able to induce nitric oxide release by intact and functional endothelium, which results in arterial vasodilatation; in contrast, a vessel with dysfunctional endothelium shows a vasoconstrictor response due to a direct muscarinic effect on VSMCs [5]. Other substances usually used are substance P, adenosine and bradikinin as well as L-N-methylarginine (L-NMMA), a specific nitric oxide antagonist. However, due to the invasive method of vascular response assessment (vascular angiography), it appears clear that also this method should be limited to research purposes. In this scenario, the possibility to test vessel reactivity by ultrasound evaluation of the brachial artery in order to evaluate the flow-mediated dilation (FMD) can allow a more easy-to-use approach to identify dysfunctional endothelium.

Different cardiovascular risk factors may impair FMD, and the decrease in FMD is parallel to the increase in cardiovascular risk burden. Most importantly, several studies have shown an association between the decline in FMD and the subsequent occurrence of cardiovascular events, in high or very high-risk patients [7,8]. Changes in FMD may be rapidly obtained by modifications of lifestyle factors, antihypertensive drugs, insulin-sensitizers or statin treatment, but the prognostic value of these changes have not yet been extensively explored. The FMD technique is based on the measurement by high-resolution ultrasound of the brachial (or femoral) arterial diameter. The diameter is measured in basal conditions and after reactive hyperemia obtained by 5 min of forearm ischaemia; the increase in blood flow induces an increase in shear stress and vasodilatation. The FMD method, firstly proposed by Celermajer et al. [9], is based on the demonstration that the change in arterial diameter reflects the biological activity of nitric oxide derived from the endothelium [10].

The improvement in ultrasound technology, which allows researchers to simultaneously measure arterial images and flow velocity, has played a substantial role in the widespread use of this method, in addition to the notion that, in respect to other stimuli of endothelial activity, the alteration in blood flow and shear stress seems to be the most ‘physiological’ one.

Despite the noninvasive nature of this technique, its low reproducibility has been advocated as a major limitation to its widespread use and could explain some of the discrepancies reported in the literature [11]. An Italian study [12] has recently demonstrated that adherence to a rigorous protocol, including operator training, standardized experimental settings and automated B-mode image edge detection system, provides accuracy and time-dependent reproducibility, in a multicentre setting. The results of the study have confirmed that changes in brachial artery diameter and flow might correspond to the intrinsic variability of endothelial response, as assessed by two different FMD studies performed in a single session.

In recent years, some aspects related to this technique have been investigated and are still disputed.

Reactive hyperemia is highly dependent on minimal forearm vascular resistance and is therefore related to structural changes in forearm small resistance arteries. It has been shown that the degree of reactive hyperemia and of brachial artery blood flow changes, as well as FMD, is strongly related to cardiovascular risk factors and events [13,14]. As the degree of dilation is related to shear stress amount, duration and pattern induced by reactive hyperemia, it was suggested to calculate shear rate from the brachial artery internal diameter and the time averaged mean blood flow velocity. Less often, shear stress has been calculated as the product of shear rate and blood viscosity, according to Poiseuille’ law. However, there is no consensus on the most appropriate blood flow velocity parameter to use for shear rate calculation, and in different studies, the peak postdeflation velocity, the difference between peak and baseline velocities or the time-integrated velocity over 10–40 s after tourniquet deflation have been proposed. Pyke and Tschakovsky [15] have found that the magnitude of FMD is related to total, rather than to peak, shear rate, whereas others suggest an independent effect of both on FMD [16].

It was also proposed to correct between-individual differences in eliciting shear rate stimulus, by normalizing FMD for shear rate, evaluating the ratio between FMD and shear rate or using shear rate as a covariate in a statistical analysis model. Several other factors, including baseline dimensions of the explored artery, may influence shear stress, and no consensus on the appropriate normalization of FMD for shear rate or stress has been reached.

Another potential limitation of FMD is that it is usually reported as the ratio (or percentage) of vessel diameter change from baseline diameter, and thus, it may be greatly influenced by baseline diameter. In a study published in the current issue of the Journal of Hypertension, Atkinson et al. [17] reanalysed the data published by Celermajer et al. [9] in 1992 and found that baseline arterial diameter may explain about 64% of FMD. Thus, the usual way to express FMD (as percentage) may weakly reflect vessel vasodilation being mainly dependent on baseline vessel diameter. In their in-depth analysis, the authors have investigated the statistical significance of FMD percentage changes and concluded that it may not correctly reflect the real endothelial (dys)function, by overestimating FMD in patients with smaller arteries or underestimating it in patients with larger arteries. Other authors have already focused on this aspect and have tried to propose a possible solution by normalizing data for anthropometric parameters (such as height or body size) [18,19]. Alternatively, it was proposed that resting brachial artery diameter may be adjusted for height, or absolute values of brachial artery diameter may express baseline endothelial activation [20]. Baseline brachial artery diameter measurements are less variable, easier to obtain than FMD and were shown to correlate significantly with cardiovascular disease (CVD) events after adjustment for Framingham risk score [18]. An additional solution is to use analysis of covariance analysis including the absolute change in vessel diameter together with the baseline diameter as a covariate: this solution appears to be statistically robust because it provides unbiased results even if the investigated values are not equal at baseline [21]. Some authors have calculated the integrated FMD response as the area under the dilation curve during the 120 s dilation period after cuff deflation in order to investigate the maintenance of the dilation reaction after its peak and found this parameter to be more strictly associated with cardiovascular risk than percentage FMD changes [22].

Finally, it has been proposed to evaluate the changes in brachial artery diameter during the low-flow state (and concomitant reduction in shear stress) induced by partial inflation of the pneumatic cuff used in FMD studies. Low-flow mediated constriction may represent an additional parameter of shear stress mediated vascular tone, possibly differentiating the increase in baseline endothelial activation from the abnormal response to endothelial stimulus [23]

In their current study, Atkinson et al. [17] correctly ask for a model that may allow the assessment of the FMD in every clinical setting, may not be influenced by the baseline vessel diameter and may be more reliable than ‘classical’ FMD percentage changes. Thus, in their article, they propose a logarithmic-derived general linear model that might identify an allometric scaling exponent of baseline diameter providing data carefully reflecting the extent of vasodilation and, therefore, the real endothelial functional status. This method appears to be an interesting alternative approach in order to express FMD, and clearly will impact on the evaluation of this parameter. The authors have derived this new measure analysing a small dataset (16 children and 48 adults), in which the model was able to provide a more accurate interpretation of FMD. In fact, the analysis of the data from this dataset according to the classical FMD method could generate the wrong conclusion that children (who obviously have smaller baseline artery diameters) have a significantly different FMD when compared with adults (with larger baseline arteries); in contrast, the author-proposed approach fully overcomes this apparent difference. This new method appears to have the potential of better defining FMD in different clinical settings. It may also lead to reconsider results of previous studies. Despite of its interest, it should be validated in larger population studies, and future prospective studies should definitely confirm its validity. It remains also to be established whether by this method an improvement in the normalization for shear rate can be obtained. Indeed, although this new way may have a great impact on the accurate analysis of endothelial function in clinical trials, many data provided by previous studies significantly link the FMD percentage values obtained by the ‘traditional’ formula with cardiovascular risk and hard cardiovascular endpoints. Thus, although the new approach appears in principle to be a more robust way to express FMD across a variety of patients at different ages and with different body size and may increase the specificity or reproducibility of FMD as a cardiovascular risk marker, the proven ability of FMD percentage to measure endothelial function and cardiovascular risk in patients should not be neglected.

Back to Top | Article Outline


Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline


1. World Health Organization. Global atlas on cardiovascular disease prevention and control. Mendis S, editor. Geneva: WHO; 2011. pp. 1–164.
2. Versari D, Daghini E, Virdis A, Ghiadoni L, Taddei S. Endothelial dysfunction as a target for prevention of cardiovascular disease. Diabetes Care 2009; 32 (Suppl 2):S314–S321.
3. Cipollone F, Cicolini G, Bucci M. Cyclooxygenase and prostaglandin synthases in atherosclerosis: recent insights and future perspectives. Pharmacol Ther 2008; 118:161–180.
4. Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, Li T, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997; 275:964–967.
5. Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance. Circulation 2007; 115:1285–1295.
6. Quyyumi AA. Prognostic value of endothelial function. Am J Cardiol 2003; 91:19H–24H.
7. Lerman A, Zeiher AM. Endothelial function: cardiac events. Circulation 2005; 111:363–368.
8. Muiesan ML, Salvetti M, Paini A, Monteduro C, Galbassini G, Poisa P, et al. Prognostic role of flow-mediated dilatation of the brachial artery in hypertensive patients. J Hypertens 2008; 26:1612–1618.
9. Celermajer DS, Sorensen KE, Gooch VM, Spiegelhalter DJ, Miller OI, Sullivan ID, et al. Noninvasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992; 340:1111–1115.
10. Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, et al. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation 1995; 91:1314–1319.
11. Flammer AJ, Anderson T, Celermajer DS, Creager MA, Deanfield J, Ganz P, et al. The assessment of endothelial function: from research into clinical practice. Circulation 2012; 126:753–767.
12. Ghiadoni L, Faita F, Salvetti M, Cordiano C, Biggi A, Puato M, et al. Assessment of flow-mediated dilation reproducibility: a nationwide multicenter study. J Hypertens 2012; 30:1399–1405.
13. Huang AL, Silver AE, Schvenke E, Schopfer DW, Keaney JF, Gokce N, Vita J. Predictive value of reactive hyperemia for cardiovascular events in patients with peripheral arterial disease undergoing vascular surgery. Arterioscler Vasc Biol 2007; 27:2113–2119.
14. Philpott AC, Lonn E, Title LM, Verma S, Buithieu J, Charbonneau F, Anderson TJ. Comparison of new measures of vascular function to flow mediated dilatation as a measure of cardiovascular risk factors. Am J Cardiol 2009; 103:1610–1615.
15. Pyke KE, Tschakovsky ME. Peak vs. total reactive hyperemia: which determines the magnitude of flow-mediated dilation? J Appl Physiol 2007; 102:1510–1519.
16. Stoner L, McCully KK. Peak and time-integrated shear rates independently predict flow-mediated dilation. J Clin Ultrasound 2012; 40:341–351.
17. Atkinson G., Batterham AM, Thijssen DHJ, Green DJ. A new approach to improve the specificity of flow-mediated dilation for indicating endothelial function in cardiovascular research. J Hypertens 2013; 31:287–291.
18. Packard GC, Boardman TJ. The use of percentages and size-specific indices to normalize physiological data for variation in body size: wasted time, wasted effort? Comp Biochem Physiol Part A 1999; 122:37–44.
19. Yeboah J, Folsom AR, Burke GL, Johnson C, Polak JF, Post W, et al. Predictive value of brachial flow-mediated dilation for incident cardiovascular events in a population-based study: the multiethnic study of atherosclerosis. Circulation 2009; 120:502–509.
20. Vanmolkot FHM, de Hoon JNJM. Reproducibility of forearm vasodilator response to intraarterial infusion of calcitonin gene-related peptide assessed by venous occlusion plethysmography. Brit J Clin Pharmacol 2005; 59:387–397.
21. Vickers AJ. The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: a simulation study. BMC Med Res Meth 2001; 1:6.
22. Kabutoya T, Hoshide S, Ogata Y, Iwata T, Eguchi K, Kario K. The time course of flow-mediated vasodilation and endothelial dysfunction in patients with a cardiovascular risk factor. J Am Soc Hypertens 2012; 6:109–116.
23. Gori T, Parker JD, Münzel T. Flow-mediated constriction: further insight into a new measure of vascular function. Eur Heart J 2011; 32:784–787.
© 2013 Lippincott Williams & Wilkins, Inc.