It has become increasingly apparent that the atherosclerotic disease process begins early in life. Dynamic changes in vascular biology are involved in the initiation and progression of disease as well as in the destabilization of established plaques that gives rise to acute clinical events.¹ The vascular endothelium has been shown to be the central regulator of vascular health, accomplished through the production of a wide range of factors that affect vascular tone, cellular adhesion, thrombosis, smooth muscle cell proliferation, and vessel wall inflammation as described in Chapter 5. Because of its intimate interface between the circulating blood and the vessel wall, it is ideally placed to function as an active signal transducer for circulating modulators of vessel wall biology.² Alterations in endothelial function are the earliest pathological vascular changes that can be detected clinically. These typically precede the evolution of structural atherosclerotic disease, contributing mechanistically to lesion development and to later clinical complications.¹

Appreciation of the central role of the endothelium throughout the atherosclerotic disease process has led to the development of a wide variety of tests to evaluate its various functional properties. These techniques have provided valuable insights into the role of the endothelium in the maintenance of a healthy circulation and the pathogenesis of arterial disease. These different methods will be discussed in this chapter in relation to the opportunities provided for the detection of preclinical disease, understanding the impact of risk factors, and the vascular response to interventions.

The importance of the endothelium was first recognized by its role in the regulation of vascular tone. This is achieved by production and release of vasoactive molecules (including nitric oxide [NO], prostacyclin and other vasoactive prostanoids, endothelium-derived hyperpolarizing factor [EDHF], endothelin-1 [ET-1], and free radicals) as well as the response to and modification of circulating vasoactive mediators (including angiotensin, bradykinin, and thrombin). These agents predominantly act locally, but may also have wider systemic influences acutely on vascular tone and chronically on arterial structure and remodeling. In addition to its vasodilator functions, NO has an important function in the maintenance of vascular health through its inhibitory effects on inflammation, thrombosis, and cell proliferation. When exposed to factors that “activate” the endothelium, a switch in the biology occurs from a NO-dominant quiescent phenotype to an activated phenotype in which “uncoupled” eNOS (endothelial Nitric Oxide Synthase) at generates reactive oxygen species in the absence of its cofactor tetrahydrobiopterin (superoxide) and or the substrate l-arginine (hydrogen peroxide). Indeed, most conventional and many novel risk factors activate common pathways within the endothelium thus resulting in the dysfunction of this system. This dual role of eNOS in both the maintenance of a basal quiescent status and activation of the endothelium places the enzyme at the center of endothelial and therefore arterial homeostasis. A number of changes occur with activation of the endothelium including expression of proinflammatory chemokines, cytokines, and adhesion molecules; alteration of production of factors that modulate the local thrombogenic balance; release of endothelial microparticles; and ultimately senescence and detachment of endothelial cells from the arterial wall.

An improved understanding of the vascular biology of the endothelium has permitted the development of clinical tests that evaluate several of the functional properties of normal and activated endothelium.³ Ideally, such tests should be safe, noninvasive, reproducible, repeatable, cheap, and standardized between laboratories. The results should also reflect the dynamic biology of the endothelium throughout the natural history of atherosclerotic disease, define subclinical disease processes, as well as provide prognostic information for risk stratification in the later clinical phase. No single test currently fulfils all of these requirements, and a panel of several tests may therefore be needed to characterize these multiple facets of endothelial biology.

Clinical assessment of endothelial function can be split into two main areas: Firstly, evaluation of endothelium-dependent vasomotor function, which involves application of pharmacological and physiological techniques that can serve as functional “bioassays” of local NO bioavailability; secondly, development of assays for measurement of circulating biochemical and cellular markers of endothelial activation, damage, and repair capacity that can provide insight into systemic and regional changes in endothelial biology relevant to the pathogenesis of arterial disease.

Principles of Endothelial Vasomotor Function Testing

Endothelium-dependent vasomotion has been the most widely used clinical end point for the assessment of endothelial function because changes in arterial diameter and blood flow can be measured reliably in patients both invasively and noninvasively. Testing primarily involves pharmacological and/or physiological stimulation of net endothelial release of NO and other vasoactive compounds. NO activates guanylate cyclase in vascular smooth muscle leading to an increased production of cyclic guanosine monophosphate, a reduction in intracellular calcium leading to muscle relaxation, and vasodilatation. The magnitude of the induced vasodilatation typically reflects local bioavailability of NO and also the responsiveness of the vascular smooth muscle. Thus, in order to localize any observed defect to the endothelium, a comprehensive vascular function testing protocol will usually include assessment of the vasodilator response to an endothelium-independent dilator for comparison. These tests determine the effect of local endothelial NO bioavailability on vasomotor tone and also reflect its other important biological functions in health and disease.

Invasive Assessment of Vascular Function

Pharmacological assessment of vascular function is most commonly undertaken using intra-arterial infusions of acetylcholine (ACH) or bradykinin (BK) as endothelium-dependent agonists that mediate the release of NO, prostacyclin, and EDHF from endothelial cells. This is followed by infusion of an NO-donor such as sodium nitroprusside (NTP) or nitroglycerine (NTG) to assess endothelium-independent smooth muscle function. These studies are direct clinical analogs of Furchgott and Zawadski’s pioneering experimental work.⁴ These studies are invasive by nature. The original clinical investigations of endothelial function were undertaken in the coronary circulation, but brachial artery cannulation for pharmacological assessment of forearm vascular function is a useful alternative and commonly undertaken for clinical research purposes.

Assessment of Coronary Vasomotor Function

Direct assessment of coronary endothelial vasomotor function is undertaken at the time of cardiac catheterization and involves assessment of epicardial and microvascular responses to local infusion of endothelium-dependent pharmacologic probes, measured using quantitative coronary angiography and Doppler flow wire techniques. Inhibition of eNOS by intracoronary infusion of the l-arginine analog, l-N^G- monomethyl arginine (l-NMMA), increases coronary vascular resistance and constricts epicardial coronary arteries, confirming the importance of basal generation of NO in the maintenance of human coronary vasodilator tone.⁵ This effect is diminished in subjects with atherosclerosis or its risk factors, suggesting reduced bioavailability of NO in these individuals. Most human coronary vascular studies have employed muscarinic agonists to test endothelial function. The integrity of the endothelium is defined by the presence of a preserved vasodilator response to intra-arterial ACH, whereas abnormal function is characterized by a constrictor response of coronary epicardial vessels and/or a depressed microvascular vasodilator response. This occurs as a result of the smooth muscle response to direct muscarinic receptor stimulation overwhelming the depressed or absent dilator effect when availability of endothelium-derived NO is diminished.^5,6 Doses of ACH that result in final blood concentrations in the range of 10⁻⁸ to 10⁻⁵ mol are the most appropriate for assessment of the physiological range of responses.⁷ Co-infusion of l-NMMA almost completely abolishes the epicardial dilation and a significant proportion of the coronary microvascular dilation that is mediated by ACH, substance-P, and metabolic stress demonstrating the important contribution of NO release to the vasomotor responses provoked by these stimuli in vivo.^5,6,8,9

A comprehensive assessment of coronary anatomy and physiology has been termed a “functional angiogram.” The study should be performed after withholding cardioactive medications for at least 24 hours, and ideally for at least five half-lives. NTG should be withheld at least 4 hours prior to the study. After diagnostic cardiac catheterization, which will identify any important structural coronary lesions, the patient is anticoagulated with heparin and a guiding catheter is positioned in the left coronary ostium. A Doppler FloWire is then advanced into a straight, nonbranching segment of the proximal mid left anterior descending artery, and a narrow-guage infusion catheter is advanced over this with its tip just distal to the tip of the wire. Coronary blood flow reserve may then be assessed using the Doppler wire with intracoronary boluses of adenosine to achieve maximal hyperemia (18–42 μg). Endothelium-dependent coronary function is then determined by infusing ACH at increasing doses to achieve intracoronary concentrations (10⁻⁷–10⁻⁵ mol/L). The patient is closely monitored for symptoms, hemodynamic, and electrocardiographic changes. Intracoronary NTG (100–200 μg) is given to assess endothelium-independent vasomotor function, or a dose–response study to NTP maybe performed as an alternative. Quantitative coronary angiography for measuring epicardial diameter is performed at the end of each infusion and Doppler velocities are recorded. Coronary artery diameter (D) at the level of the Doppler wire and the average peak velocity (APV) from the Doppler signal are used to calculate coronary blood flow by means of the formula D² × APV/8.

Two to three straight, nonbranching segments of the study artery are also measured at baseline and following infusion of ACH and NTG and the epicardial endothelium-dependent and -independent vasomotor responses are determined by calculating the percentage change in arterial diameter from baseline³ (D_{ACH or NTG} − D_baseline/D_baseline × 100%) (Figure 6-1).

Similarly, endothelial function can be assessed by determining the responses to other agents such as BK or substance-P, although dysfunctional vasoconstrictor responses are not typically seen with these agents. Physiological responses to cold-pressor testing and flow-mediated dilatation of proximal epicardial coronary arteries in response to a more distal infusion of adenosine have also been used to assess coronary vascular function. These studies have provided important insights into the effects of atherosclerosis and its risk factors on coronary regulatory physiology and risk stratification (Figure 6-2) as well as the potential reversibility in response to agents such as statins and ACE-inhibitors.^{5,6,9,10,11,12,13,14}

These highly invasive studies are, of necessity, restricted to use in subjects with clinical indications for cardiac catheterization, limiting the research opportunities to the later more advanced stages of arterial disease using this methodology.

Invasive Assessment of Forearm Vascular Function

Assessment of endothelium-dependent vasodilatation in the forearm microcirculation uses similar methodology as used for coronary circulation. The brachial artery is cannulated and endothelium-dependent and -independent agonists and antagonists can be infused at doses that influence local physiology without systemic effects. Changes in forearm blood flow are measured using strain gauge venous occlusion plethysmography.¹⁵ Although invasive, the methodology is generally considered to be reasonably safe and the technique is used in many research laboratories worldwide. Ideally, FBF must be measured in both forearms to adjust the stimulated changes in the experimental arm for minor systemic fluctuations in basal blood flow and blood pressure that may occur during the study.^15,16 The majority of studies measure the percentage differences in forearm blood flow and resistance between the experimental and control arms in response to increasing doses of endothelium-dependent (e.g., ACH or BK) and -independent (e.g., NTP or NTG) agonists. The contribution of NO can also be evaluated by assessment of changes in blood flow and resistance following eNOS antagonists (e.g., l-NMMA and l-NAME), with endothelium-independent vasoconstrictor responses to, for example, phenylephrine used as a comparison. More detailed assessment of the contribution of NO to the endothelium-dependent responses to ACH or BK can be studied in the presence of a NO-clamp. This involves co-administration of NTP to counteract the constrictor effect of l-NMMA, thus allowing appropriate comparison of the responses under control conditions and NO-synthase blockade without the confounding effect of different baseline blood flow rates.^3,17 As usual, testing of endothelium-independent responses by construction of a dose–response curve to NTG or NTP is also necessary to determine the endothelial specificity of any differences in the responses to ACH or BK. Endothelium-dependent vasodilator responses in the forearm microcirculation involve different pathways, and although responses to the commonly used agents predominantly reflect NO release, the contribution of other mediators such as EDHF increases when NO bioavailability is diminished.^18,19

The main advantages of this technique are that, although invasive, it can be applied both to patients and healthy volunteers; it allows careful clinical pharmacological assessment of combinations of agonists and antagonists to test other pathways in addition to endothelial-derived and exogenous NO (e.g., indomethacin to block cyclooxygenase²⁰ and vitamin C to assess oxidative stress²¹). Furthermore, although forearm microcirculation is clearly not a target organ for atherosclerosis, the responses to ACH and modulation by vitamin C are predictive of cardiovascular outcome,²¹ suggesting that systemic, as well as coronary endothelial function, is an important marker of global cardiovascular risk. Despite the detailed data that can be obtained, the invasive nature of this technique limits its repeatability and prohibits its use in larger studies. Results are also difficult to standardize because baseline resistance vessel tone is variable within and between subjects, and testing protocols and setup differ between research laboratories. Therefore noninvasive techniques are required with broader applicability to larger study populations including younger adults and children.

Noninvasive Assessment of Vascular Function

Flow-Mediated Vasodilatation

Endothelial cells release NO and other endothelium-derived relaxing factors in response to mechanical stress. The precise mechanisms for the acute detection of shear forces and subsequent signal transduction to modulate vasomotor tone are not fully understood, but probably involve calcium-activated potassium channel opening, membrane hyperpolarization, and calcium-mediated activation of eNOS. Thus, in vivo, the endothelium can respond dynamically to shear stresses mediated by increased flow, by releasing NO that mediates vasodilatation. Using these principles, we developed and refined a technique for assessment of flow-mediated endothelium-dependent dilatation (FMD).²² This method involves measurement of the change in diameter of a conduit artery (most commonly the brachial artery) in response to increased flow, typically induced by a period of ischemia in the distal circulatory bed (Figures 6-3 and 6-4). FMD when performed under appropriate conditions is highly reproducible, is predominantly mediated by NO (Figures 6-5 and 6-6), and is depressed in subjects with atherosclerosis and its risk factors (Figure 6-7).^{23,24,25,26,27} These measurements also correlate well with coronary vascular endothelial vasodilator function²⁸ (Figure 6-8) and serological markers of endothelial perturbation as well as predicting long-term cardiovascular outcome (Figure 6-9).^29,30