CAUSE OR EFFECT? MICROVASCULAR DYSFUNCTION IN INSULIN RESISTANT STATES

Conventionally, it is understood that the vasculature is one of many tissues that are adversely affected in patients with type 2 diabetes mellitus (T2DM). Accelerated development of arterial atherosclerosis and microvascular dysfunction are two of the diabetic complications that cardiovascular clinicians commonly encounter. Microvascular dysfunction, defined as inadequate microvascular response to metabolic demand at rest or during a physiologic or metabolic challenge that normally requires a reduction in vascular resistance, occurs as a consequence of multiple pathways that become abnormal in response to obesity, insulin resistance (IR), hyperglycemia, and T2DM. These pathways are diverse and include reduced bioavailability of nitric oxide (NO) and other vasodilators (e.g. eicosanoids and other arachidonic acid metabolites), increased production or sensitivity to vasoconstrictors including endothelin or angiotensin-II, heightened response to alpha adrenergic signaling, functional abnormalities in cation channel (e.g. TRPV-1, Kv channels), diffuse glycosylation, and increased reactive oxygen species.^1,2 Tissues affected by secondary microvascular dysfunction include the heart, skeletal muscle, adipose tissue, and the kidney.

A somewhat under-appreciated concept is that microvascular dysfunction can actually contribute to the pathogenesis of IR and T2DM. Much of research supporting this notion has focused on the microcirculation of skeletal muscle which in humans is a major storage site for glucose. In limb skeletal muscle, insulin produces an augmentation in limb skeletal muscle blood flow in a dose-dependent fashion.^3,4 This vascular action of insulin is thought to influence the delivery of glucose, and possibly insulin itself, to muscle and other glucose storage sites.⁵ The measurement of blood flow alone may not tell the whole story of how vascular responses influence glucose homeostasis. At the capillary level, changes in muscle perfusion occur primarily as a result of changes in either the rate of blood flux through individual capillaries and/or changes in the number of actively perfused capillaries.⁶ Even under hyperinsulinemic conditions that increase myocellular glucose transport, the arterial-venous concentration difference for glucose is relatively low with A-V gradients of only 0.1–0.2 mM.⁵ Accordingly, changes in the blood flux rate through capillaries would result in only a small increase in glucose diffusion by increasing its plasma concentration at the terminal aspects of a capillary network. However, perfusion can also be manifest by capillary recruitment. Under resting conditions only 25–35% of capillaries in skeletal muscle are actively perfused so that the potential capillary blood volume (CBV) reserve is high.⁶ An increase in the total surface area for glucose diffusion and a reduction in intercapillary diffusion distance through capillary recruitment represents a much more effective response for increasing glucose uptake,⁷ and has been a keen subject of investigation.

Contrast-enhanced ultrasound, which was used by Keske, et al. to study vascular responses to glucose challenge in this journal issue has proven to be extremely valuable for assessing microvascular responses to insulin. It can not only rapidly assess tissue perfusion responses to metabolic stimuli, but it can also parametrically assess microvascular flux rate and CBV in the heart and in skeletal muscle.^6,8 Moreover, post-processing algorithms have been developed to eliminate the contribution of large vessels which reside inside muscle and adipose tissue in order to more carefully study microvascular responses.⁶ CEU has been used to demonstrate that capillary recruitment occurs within minutes of physiologic hyperinsulinemia or glucose challenge, preceding increases in limb glucose uptake by several minutes.^9,10 These data suggest that capillary recruitment is an early insulin-triggered event that is needed in order for insulin to maximally exert its metabolic actions in muscle. CEU studies performed in obese and IR animal models and in patients have indicated that a functional impairment skeletal muscle capillary recruitment in response to insulin, glucose challenge and exercise plays a role in the pathogenesis of IR and T2DM.^11–13 Temporal assessment of muscle perfusion with CEU has revealed that an increase in muscle CBV at rest or during glucose challenge acts as a potential compensatory response early after starting a high-fat high-calory diet, and that IR progresses more rapidly when compensatory CBV responses fail.¹⁴

Adipose tissue is another major storage site for both lipids and glucose that, like skeletal muscle, is under the regulatory influence of insulin.¹⁵ The idea that adipose tissue perfusion is abnormal in T2DM is a rather old one, long predating the study by Keske et al.,¹⁶ and has been studied using standard indicator-dilution techniques (i.e. ¹³³Xe) or by positron-emission tomography (PET) imaging.¹⁷ These studies have revealed that adipose microvascular responses may play a role in augmenting adipose lipid and glucose storage.¹⁸ It has also been established that adipose perfusion under basal and post-prandial conditions, like muscle, is lower in IR individuals, suggesting that adipose microvascular dysfunction may contribute to abnormal metabolic substrate storage.^15,19 Despite some differences in the pathways by which flow is augmented in muscle and adipose tissue, CEU performed in murine models of obesity and T2DM have demonstrated a close correlation between the degree of abnormal perfusion in muscle and adipose tissue.²⁰ Because adipose tissue is characterized rather large average intercapillary distances,²¹ particularly in obesity, and a small arterio-venous glucose gradient, it is quite likely that CBV (i.e. capillary surface area) is the major determinant of glucose uptake by adipose tissue at rest and in the post-prandial state. Morover, adipose tissue hypoxia in obesity has been proposed as an important contributor to inflammatory activation, abnormal adipokine signaling, and risk for vascular disease.²²

Although the scientific literature on adipose perfusion is rich, the study by Keske et al.¹⁶ is novel for several reasons. It is among the first that has used CEU perfusion imaging in humans to measure not only adipose perfusion at rest and during glucose challenge, but also CBV. In a previous study, CEU demonstrated that subcutaneous adipose tissue CBV response to glucose challenge is blunted in patients with T2DM.²³ The study by Keske and colleagues adds to this finding by providing correlations between abnormal adipose CBV and either clinical or biochemical variables. Although CEU perfusion imaging was performed on subcutaneous fat, the vascular responses correlated with the truncal fat mass, the expansion of which is thought to play a more important role in obesity-related IR. It is not yet clear whether this correlation implies a global adipose vascular abnormality associated with expansion of adipose mass. Consistent with earlier studies, some but not all of the perfusion responses correlated with indices of insulin sensitivity (QUICKI, glucose AUC, fasting insulin) and with triglycerides. What is not known is whether abnormalities in adipose perfusion contributed to abnormal glucose uptake and/or triglyceride uptake. Perhaps one of the most interesting findings was the lack of association between adipose perfusion at rest or after glucose challenge and any marker of systemic inflammation. This finding challenges the notion that modest reductions in adipose perfusion can lead to “dysfunctional fat” and inflammatory activation. It should be cautioned, however, that blood flow in more “metabolically active” visceral adipose tissue was not measured. The authors state that certain metabolic abnormalities tended to correlate more with abnormalities in adipose total perfusion than with CBV. However this statement needs to be tempered by the extremely low levels of contrast intensity used to measure CBV (resulting in wide coefficients of variation) and lack of normalization to blood pool for absolute quantification.

In summary, not only do IR and secondary hyperglycemia lead to microvascular dysfunction, but the the vice-versa is true as well. The bidirectional nature of the relationship between tissue perfusion and IR complicates the study of how adipose tissue perfusion influences metabolic syndrome. The results from the study by Keske et al. provide encouraging evidence that we now have tools that can be applied to better understand this process by examining temporal assessment of adipose perfusion during disease development, and by assessing metabolic consequences of interventions that alter microvascular function.