chapter 2 – vasomotion and capillary recruitment

Insulin-induced capillary recruitment is considered a significant regulator of overall insulin-stimulated glucose uptake. Insulin’s action to recruit capillaries has been hypothesized to involve insulin-induced changes in vasomotion. Data directly linking vasomotion to capillary perfusion, however, are presently lacking. We therefore investigated whether insulin’s actions on capillary recruitment and vasomotion were interrelated in a group of healthy individuals. We further assessed the role of capillary recruitment in the association between vasomotion and insulin-mediated glucose uptake. Changes in vasomotion and capillary density were determined by laser Doppler flowmetry (LDF) and capillary videomicroscopy in skin, respectively, before and during a hyperinsulinemic euglycemic clamp in 19 healthy volunteers. Insulin-induced increase in the neurogenic vasomotion domain was positively related to insulin-augmented capillary recruitment (r=0.51, P=0.04), and both parameters were related to insulin-mediated glucose uptake (r=0.47, P=0.06 and r=0.73, P=0.001, respectively). The change in insulin-augmented capillary recruitment could, at least statistically, largely explain the association between the neurogenic domain and insulin-mediated glucose uptake. Insulin-induced changes in vasomotion and capillary recruitment are associated in healthy volunteers. These data suggest that insulin’s action to recruit capillaries may in part involve action on the neurogenic vasomotion domain, thereby enhancing capillary perfusion and glucose uptake. Introduction Vasomotion, a rhythmic change in vascular diameter, is a typical feature of microvascular networks (1, 2). Whereas the physiological significance of vasomotion in microcirculatory flow remains unclear (2), hypotheses have been put forward suggesting its effects on capillary exchange of substances between blood and tissues (3, 4). In fact, theoretical modeling has demonstrated that vasomotion may strongly influence capillary exchange (5). Based on this assumption, insulin has been proposed to alter arteriolar vasomotion with a resultant increase in the capillary exchange surface termed ‘capillary recruitment’ (6, 7, 8). Insulin-induced capillary recruitment is considered a significant determinant of overall insulin-mediated glucose uptake, because it controls the access of insulin and glucose to muscle interstitium. Insulin delivery to skeletal muscle interstitium is a rate-limiting step in insulin-stimulated glucose uptake by skeletal muscle, and is slower in insulin-resistant subjects than in healthy subjects (9). Vasomotion causes a rhythmic variation in blood flow (flow motion), which can be detected by laser Doppler flowmetry (10). Vasomotion activity in the microvascular bed can, therefore, be explored by analysis of the component frequencies of the laser Doppler signal using spectral analyses such as Fast-Fourier analysis or a Wavelet transform (7, 8, 11, 12, 13). Distinct periodic oscillations in the laser Doppler signal have been attributed to, consecutively, the heart beat (spectral peaks at 0.4-1.6 Hz), respiration (0.15–0.4 Hz), myogenic activity in the vessel wall (0.06–0.15 Hz), neurogenic activity (0.02–0.06 Hz), and endothelial activity (0.01–0.02 Hz) (12). Using these analyses, human studies have suggested that systemic hyperinsulinemia affects microvascular vasomotion by increasing endothelial and neurogenic activity in skin and muscle (7, 8, 11), and that it is particularly the contribution of endothelial and neurogenic activity to microvascular vasomotion which is impaired in obese, insulin-resistant individuals (14). In rats, insulin acts to increase the myogenic component of vasomotor activity in muscle. Moreover, both muscle microvascular flow and myogenic activity were depressed in the α-methyl serotonin-induced acute insulin resistant state (13). Although these studies suggest that insulin’s recruitment of microvascular flow, and subsequent glucose uptake in skeletal muscle, may in part involve action on vasomotion, a direct link between changes in vasomotion and insulin-induced microvascular recruitment has yet to be demonstrated. The aim of the present study, therefore, was to investigate whether (changes in) microvascular vasomotion and (changes in) capillary recruitment are associated in a group of healthy individuals. We further assessed whether changes in vasomotion were associated with insulin-mediated glucose uptake and whether capillary recruitment could explain part of this association. Methods Subjects Nineteen healthy volunteers participated in this study (table 1). They were recruited through local advertisements. None had a history of cardiovascular disease, all were non-diabetic (15) and normotensive (<140/<90 mmHg) as determined by triplicate office blood pressure measurement. Participants were of Caucasian origin and nonsmokers. No medication was used during 4 weeks leading up to and on the day of study. The study protocol was approved by the local Ethics Committee and in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. Study design Measurements were conducted in a fasting state on an outpatient basis in a quiet, temperature-controlled room (23.0±1.0°C) after 30 minutes of acclimatization. Subjects had abstained from caffeine, alcohol and meals overnight. The microvascular measurements (laserDoppler/vasomotion and capillaroscopy) were performed simultaneously in the supine position for both the baseline microvascular measurement, as well as the microvascular measurement during hyperinsulinemia (figure 1). The hyperinsulinemic euglycemic clamp was chosen as the intervention. It utilizes high physiological insulin levels as seen in obese subject and microvascular measurements can be performed during a metabolic steady state. Similar (or in obese subject even higher) endogenous levels of insulin are reached for a comparable duration during a meal-test (16). However, this is accompanied by the secretion or increase of a myriad of vasoactive substances such as incretins and free fatty acids, hindering the investigation of insulin-specific effects on the microvasculature (16). Due to technical problems one videomicroscopy measurement and one laserDoppler registration failed on separate occasions. Skin microvascular measurements (laser Doppler) Skin temperature was registered continuously and was above 28°C at the start of all microvascular measurements. Skin blood flow was measured in conventional perfusion units (PU) by means of a laser Doppler system (Periflux 5010, Perimed, Stockholm, Sweden). Microvascular measurements were performed with one thermostatic laser Doppler probe (PF 457, Perimed, Stockholm, Sweden) positioned at the dorsal side of the wrist of the left hand (17). Vasomotion Wavelet analysis of LDF signals with a minimum of 30 minutes (with a sampling frequency of 32 Hz resulting in approximately 58.000 data points) in length was conducted to assess the frequency spectrum between 0.01 and 1.6 Hz as described above (18). Wavelet analysis was performed using the wavelet toolbox in Matlab (7.8.0.347; The Mathworks, Inc., Natick, MA, USA), as described earlier (13). Scales were chosen for a resulting frequency range from 0.01 to 1.6 Hz. To eliminate edge effects, the first and last 2000 samples were removed from the resulting wavelet transform (19). The relative amplitude was calculated for each of the five frequency bands by dividing the average amplitude within a band by the average amplitude of the entire spectrum. This normalisation takes into account the variation in the LDF signal strength between subjects and/or within subjects during an intervention (18, 19, 20). Skin microvascular measurements (capillary videomicroscopy) Nailfold capillary studies were performed as described in detail elsewhere (21). Briefly, nailfold capillaries in the dorsal skin of the third finger of the left hand were visualized by a capillary microscope (Zeiss), linked to a television camera (Philips LDH 070/20). A 3.2x objective (Zeiss 3.2/0.07) was used with a total system magnification of 99x. The number of perfused capillaries was counted off-line by an experienced investigator (M.P.d.B.) from a videotape. Capillary density at baseline was defined as the number of capillaries per square millimeter which were continuously perfused for 15 seconds during an observation-period of 30 seconds (visible flowmotion under baseline conditions). Capillary density after postocclusive reactive hyperemia (PRH) was determined by counting all visible capillaries after four minutes of arterial occlusion. Capillary recruitment was calculated as the relative increase in capillary density from the continuously filled fraction to capillary density after PRH. Insulin-augmented capillary recruitment (percentage-points) was defined as the increase in capillary recruitment during hyperinsulinemia from capillary recruitment during saline infusion. The day-to-day coefficients of variation (CV) of continuously perfused capillary density and peak capillary density were 3.4±2.0% and 3.9±1.6% respectively, as determined in 10 subjects on separate days (21). Insulin sensitivity Insulin sensitivity was assessed by the hyperinsulinemic, euglycemic clamp technique (22). Briefly, insulin (Actrapid, Novo Nordisk, Bagsvaerd, Denmark) was infused in a primed (0.4 U ml-1) continuous manner at a rate of 1 mU·kg·min-1. Euglycemia (5 mmol/L) was maintained by adjusting the rate of a 20% glucose infusion based on plasma glucose measurements performed at 5-10 minute intervals using an YSI 2300 STAT Plus analyser (YSI, Yellow Springs, USA). Whole body glucose uptake (M-value) was calculated from the glucose infusion rate during steady state of the clamp and expressed per kilogram of (lean) body weight. Anthropometrics Lean muscle mass was determined by bioelectrical impedance analysis (BF906, Maltron, Rayleigh, UK). Statistical Analysis All variables were first checked for normality of distribution. Data are presented as mean ± SD, or median and range when applicable. A paired samples t-test (Wilcoxon signed rank test for non-normally distributed data) was used to compare measurements during saline and insulin infusion. Correlation analyses (Spearman’s Rho for non-normally distributed data) were used to investigate correlations between (changes in) microvascular function and insulin sensitivity. In addition, multivariate analysis was performed to investigate whether the association between vasomotion and insulin sensitivity remained when allowing for capillary recruitment. A two-tailed P-value of <0.05 was considered significant. All analyses were performed using the statistical software package SPSS (version 18.0, SPSS, Inc., Chicago, USA).