Reduced insulin sensitivity in young, normoglycaemic subjects alters microvascular tissue oxygenation during postocclusive reactive hyperaemia.

NEW FINDINGS: What is the central question of the study? Are measures of reduced insulin sensitivity in young, normoglycaemic subjects correlated with near-infrared spectroscopy-derived microvascular responsiveness [tissue oxygen saturation reperfusion rate (STO2 upslope)] during postocclusive reactive hyperaemia? What is the main finding and its importance? A sevenfold range of hepatic insulin sensitivity is significantly correlated (r = 0.44, P = 0.02) with STO2 upslope after transient tissue ischaemia. Near-infrared spectroscopy may be an important tool for determining altered microvascular function before onset of hyperglycaemia. Identification of pre-type 2 diabetes much earlier than with the present clinical criteria is important for pre-emptive measures against microvascular deterioration. ABSTRACT: Near-infrared spectroscopy (NIRS) measurement of postocclusive reactive hyperaemia (PORH) tissue oxygen saturation reperfusion rate [STO2 upslope (as a percentage per minute)] has recently been correlated with the percentage of flow-mediated dilatation (%FMD). Cardiovascular disease is associated with impairments in %FMD. Reduced insulin sensitivity may negatively affect the vascular system for many years before prediabetes/type 2 diabetes states. The aim of this study was to determine whether static and dynamic STO2 parameters during PORH are correlated with reduced insulin sensitivity in young, normoglycaemic subjects. Glucose and insulin were measured during an oral glucose tolerance test in 18- to 26-year-old, healthy subjects (11 men and 11 women), and STO2 was measured during PORH of antebrachial muscle. Hepatic (ISIHOMA ) and whole-body (ISICOMP ) insulin sensitivities were calculated. The STO2 upslope was negatively correlated with minimal STO2 (r = -0.5, P = 0.01). The change of STO2 from minimum to baseline (ΔSTO2 ) was significantly negatively correlated with fasting insulin (r = -0.5, P = 0.01) and a positively correlated with ISIHOMA (r = 0.65, P = 0.001). The minimum STO2 was significantly negatively correlated with ISIHOMA , and STO2 upslope was significantly positively correlated with ISIHOMA (r = 0.44, P = 0.02). The minimum STO2 (a measure of O2 extraction while the cuff was inflated), ΔSTO2 (a measure of the amount of reperfusion) and STO2 upslope (a measure of responsiveness of the microcirculation to ischaemia) were all positively correlated with ISIHOMA , one of the longest-used measures of insulin sensitivity. The NIRS-derived STO2 might be a useful tool for assessing how levels of reduced insulin sensitivity in young, normoglycaemic adults affect the microvasculature.

Effect of assuming constant tissue scattering on measured tissue oxygenation values during tissue ischemia and vascular reperfusion.

The purpose of this study was to determine the effects of assuming constant tissue scattering properties on tissue oxygenation measurements during a vascular occlusion test (VOT). Twenty-one subjects (21.8 ± 1.9 yr) completed a VOT [1 min baseline (BL), 5 min of tissue ischemia (TI), and 3 min of vascular reperfusion (VR)]. Absolute concentrations of oxygenated heme (oxy-[heme]), deoxygenated heme (deoxy-[heme]), total heme (total [heme), tissue oxygen saturation (StO2), and heme difference [heme]diff) were measured using frequency domain near-infrared spectroscopy (FD-NIRS) while 1) continuously measuring and incorporating tissue scattering ([Formula: see text]) and 2) assuming scattering remained constant. FD-NIRS measured [Formula: see text] increased during TI at 692 nm (P < 0.001) and decreased at 834 nm (P < 0.001). During VR, [Formula: see text] decreased at 692 nm (P < 0.001) and increased at 834 nm (P < 0.001). When assuming constant scattering, oxy-[heme] was significantly less at TIpeak (P < 0.05) while deoxy-[heme] and StO2 were significantly altered at BL, TIpeak, and VRpeak (all P < 0.001). Total [heme] did not change during the VOT. Absolute changes in deoxy-[heme], oxy-[heme], and StO2 in response to TI and VR were significantly exaggerated (all P < 0.001) and the rates of change during TI (slope 1) and VR (slope 2) in deoxy-[heme], oxy-[heme], StO2, and [heme]diff were significantly increased (all P < 0.05) when constant tissue scattering was assumed. These findings demonstrate the need for caution when interpreting NIRS data without continuously measuring tissue optical properties. Further, assuming tissue optical properties remain constant may have important consequences to experimental data and clinical conclusions made using NIRS.NEW & NOTEWORTHY NIRS measurements provide significant experimental and clinical insight. We demonstrate that absolute changes in tissue oxygenation measurements made with NIRS are overestimated and the kinetic responses of NIRS measurements are exaggerated by varying degrees among individuals if tissue scattering characteristics are assumed to remain constant during vascular occlusion tests.

Insulin resistance and metabolic syndrome criteria in lean, normoglycemic college-age subjects.

AIMS: The goal of this study was to determine insulin sensitivity in a fasted state and during an oral glucose tolerance test (OGTT), in normoglycemic (NGT), lean (L) (n = 35) and, for comparison, overweight/obese (OW/O) (n = 9) college-aged subjects. MATERIALS AND METHODS: Insulin sensitivity for 44 NGT, normotensive subjects, age 18-26 yrs., was determined by homeostasis model assessment (HOMA-IR) and from Matsuda index (ISI Matsuda). RESULTS: Subjects were normoglycemic fasted (4.59 + 0.35 mmol/L) and at two hours post OGTT (4.52  + 1.35 mmol/L). Besides anthropometric measures, there were significant differences between OW/O and L for fasting insulin (P < 0.001) and both measures of insulin sensitivity (P < 0.05). All subjects exhibited a 9-fold range in HOMA-IR (0.88 + 0.51, range 0.3-2.7) and an 8-fold range in ISI Matsuda (11.9 + 4.7, range 3.0-24.2). The latter was inversely correlated with systolic blood pressure (r = 0.35, P = 0.04) even though subjects were normotensive. In lean subjects, 2.3% were IR by HOMA-IR > 2.1, 5.7% by ISI Matsuda < 5.9, and 22.9% had >one criteria for metabolic syndrome (MetS); 28.6% had some negative metabolic biomarker. CONCLUSIONS: Insulin resistance is present in lean, NGT college-age subjects even without MetS criteria and is discernable with an easily applicable OGTT-derived index.

Relationship between brachial artery blood flow and total [hemoglobin+myoglobin] during post-occlusive reactive hyperemia.

The associations between macrovascular and microvascular responses reported previously during post-occlusive reactive hyperemia have been inconsistent. The purpose of this study was therefore to determine the temporal relationship between the reactive hyperemic responses within a conduit artery and the downstream microvessels. Conduit artery blood flow was measured in the brachial artery with pulsed Doppler ultrasound. A potential analog of microvascular flow, changes in skeletal muscle total[hemoglobin+myoglobin] (T[Hb+Mb]), was assessed with near-infrared spectroscopy (NIRS). We found a high degree of correlation between these two measures (r=0.91). Cross-correlation analysis revealed two distinct response patterns. In 10 of our 15 subjects there was time displacement between peak brachial artery blood flow (BABF) and T[Hb+Mb] responses; in the remaining 5 the peaks were coincident. Granger causality testing suggested that reactive hyperemia in the macrovessel determined hyperemia in the downstream microvessels in all 15 study subjects. Time constants for the on (τ1) and off (τ2) kinetics of each response were calculated; our initial hypothesis was that τ1 and τ2 for T[Hb+Mb] would correlate with τ1 and τ2 for BABF, respectively. However, only for τ2 was this observed (r=0.52; p<0.05). No similar relationship was observed for τ1. Adipose tissue thickness did not influence either time constant for T[Hb+Mb]. Taken together, our results show that the temporal characteristics of the hyperemic response in the conduit artery are qualitatively reflected in the downstream microvasculature, but mechanisms for quantitative differences remain to be identified.

Characterizing near-infrared spectroscopy responses to forearm post-occlusive reactive hyperemia in healthy subjects.

During post-occlusive reactive hyperemia (PORH) there is a temporary increase in the total hemoglobin + myoglobin (T[Hb+Mb]) signal as measured by near-infrared spectroscopy (NIRS). This transient increase predicts differences in the kinetic responses of deoxy[Hb+Mb] and oxy[Hb+Mb] during PORH. The purpose of this study was to determine whether sigmoidal (Gompertz or logistic) or exponential functions better describe these response curves during PORH. The fit of the three functions (exponential, Gompertz and logistic) to the NIRS responses, as determined from residual sum of squares, was compared using repeated measures ANOVA on Ranks. The Gompertz function provided a better fit to the oxy[Hb+Mb] response curve than did either the exponential or logistic function (χ (2) = 21.7, df = 2, p < 0.001). The logistic function provided a better fit for the deoxy[Hb+Mb] response (χ (2) = 22.9, df = 2, p < 0.001) than did either the Gompertz or exponential functions. For both NIRS signals, the better fitting sigmoidal functions fit the data well, with an average r value of 0.99 or greater. Adipose tissue thickness was correlated with parameters related to signal strength (amplitude, r = 0.86-0.89; baseline, r = 0.67-0.75; all p < 0.001) but was not related to kinetic parameters (time constant and inflection point; p > 0.05 for all comparisons). These results suggest that during PORH distinct sigmoidal mathematical functions best describe the responses of the oxy[Hb+Mb] (Gompertz) and deoxy[Hb+Mb] (logistic) as measured by NIRS. Further, differences in both the kinetic and amplitude aspects for the responses of oxy[Hb+Mb] and deoxy[Hb+Mb] predict the observed transient change in T[Hb+Mb]. Our methods provide a technique to evaluate and quantify NIRS responses during PORH, which may have clinical utility.

Effects of a high-fat meal on pulmonary function in healthy subjects.

Obesity has important health consequences, including elevating risk for heart disease, diabetes, and cancer. A high-fat diet is known to contribute to obesity. Little is known regarding the effect of a high-fat diet on pulmonary function, despite the dramatic increase in the prevalence of respiratory ailments (e.g., asthma). The purpose of our study was to determine whether a high-fat meal (HFM) would increase airway inflammation and decrease pulmonary function in healthy subjects. Pulmonary function tests (PFT) (forced expiratory volume in 1-s, forced vital capacity, forced expiratory flow at 25-75% of vital capacity) and exhaled nitric oxide (eNO; airway inflammation) were performed in 20 healthy (10 men, 10 women), inactive subjects (age 21.9 +/- 0.4 years) pre and 2 h post HFM (1 g fat/1 kg body weight; 74.2 +/- 4.1 g fat). Total cholesterol, triglycerides, and C-reactive protein (CRP; systemic inflammation) were determined via a venous blood sample pre and post HFM. Body composition was measured via dual energy X-ray absorptiometry. The HFM significantly increased total cholesterol by 4 +/- 1%, and triglycerides by 93 +/- 3%. ENO also increased (p < 0.05) due to the HFM by 19 +/- 1% (pre 17.2 +/- 1.6; post 20.6 +/- 1.7 ppb). ENO and triglycerides were significantly related at baseline and post-HFM (r = 0.82, 0.72 respectively). Despite the increased eNO, PFT or CRP did not change (p > 0.05) with the HFM. These results demonstrate that a HFM, which leads to significant increases in total cholesterol, and especially triglycerides, increases exhaled NO. This suggests that a high-fat diet may contribute to chronic inflammatory diseases of the airway and lung.

Matching of blood flow to metabolic rate during recovery from moderate exercise in humans.

It is unclear whether measurement of limb or conduit artery blood flow during recovery from exercise provides an accurate representation of flow to the muscle capillaries where gas exchange occurs. To investigate this, we: (a) examined the kinetic responses of femoral artery blood flow (QFA), estimated muscle capillary blood flow (Qcap) and estimated muscle oxygen uptake (VO2m) following cessation of exercise; and (b) compared these responses to verify the adequacy of O2 delivery during recovery. Pulmonary VO2 (VO2p) was measured breath by breath, QFA was measured using Doppler ultrasonography, and deoxy-haemoglobin/myoglobin (deoxy-[Hb/Mb]) was estimated by near-infrared spectroscopy over the rectus femoris in nine healthy subjects during a series of transitions from moderate knee-extension exercise to rest. The time course of Qcap was estimated by rearranging the Fick equation [i.e. Qcap(t) alpha VO2m(t)/deoxy-[Hb/Mb](t)], using the primary component of Vo2p to represent VO2m and deoxy-[Hb/Mb] as a surrogate for arteriovenous O2 difference. There were no significant differences among the overall kinetics of VO2m (tau, 31.4+/-8.2 s), QFA [mean response time (MRT), 34.5+/-20.4 s] and Qcap (MRT, 31.7+/-14.7 s). The VO2m kinetics were also significantly correlated (P<0.05) with those of both QFA and Qcap. Both QFA and Qcap appear to be coupled with VO2m during recovery from moderate knee-extension exercise, such that extraction falls (thus cellular energetic state is not further compromised) throughout recovery.

Effects of inspiratory muscle training on exercise responses in normoxia and hypoxia.

The purpose of this study was to determine the effects of inspiratory muscle training (IMT) on exercise in hypoxia (H) and normoxia (N). A 4-week IMT program was implemented with 12 healthy subjects using an inspiratory muscle trainer set at either 15% (C; n=5) or 50% (IMT; n=7) maximal inspiratory mouth pressure (PImax). Two treadmill tests (85% VO2max) to exhaustion and measures of diaphragm thickness (Tdi) and function were completed before and after training in H and N. Significant increases of 8-12% and 24.5+/-3.1% in Tdi and PImax, respectively, were seen in the IMT group. Time to exhaustion remained unchanged in all conditions. Inspiratory muscle fatigue (downward arrowPImax) following exercise was reduced approximately 10% (P<0.05) in IMT after both N and H. During H, IMT reduced (P<0.05) VO2 by 8-12%, cardiac output by 14+/-2%, ventilation by 25+/-3%; and increased arterial oxygen saturation by 4+/-1% and lung diffusing capacity by 22+/-3%. Ratings of perceived exertion and dyspnea were also significantly reduced. These data suggest that IMT significantly improves structural and functional physiologic measures in hypoxic exercise.

Human femoral artery and estimated muscle capillary blood flow kinetics following the onset of exercise.

The purpose of this study was to compare the kinetics of estimated capillary blood flow (Qcap) to those of femoral artery blood flow (QFA) and estimated muscle oxygen uptake (VO2m). Nine healthy subjects performed a series of transitions from rest to moderate (below estimated lactate threshold, 6 min bouts) knee extension exercise. Pulmonary oxygen uptake (VO2) was measured breath by breath, (QFA) was measured continuously using Doppler ultrasound, and deoxyhaemoglobin ([HHb]) was estimated by near-infrared spectroscopy over the rectus femoris throughout the tests. The time course of (Qcap) was estimated by rearranging the Fick equation (i.e. Qcap = VO2m/(a-v)O2), (arterio - venous O2 difference) using the primary component of VO2 to represent VO2m and [HHb] as a surrogate for (a - v)O2. The overall kinetics of QFA (mean response time, MRT, 13.7 +/- 7.0 s), VO2m (tau, 27.8 +/- 9.0 s) and Qcap (MRT, 41.4 +/- 19.0 s) were significantly (P < 0.05) different from each other. We conclude that for moderate intensity knee extension exercise, conduit artery blood flow (QFA) kinetics may not be a reasonable approximation of blood flow kinetics in the microcirculation (Qcap), the site of gas exchange. This temporal dissociation suggests that blood flow may be controlled differently at the conduit artery level than in the microcirculation.

Effects of pedal frequency on estimated muscle microvascular O2 extraction.

An increase in muscle contraction frequency could limit muscle blood flow QM compromising the matching of QM and muscle oxygen uptake VO2M. This study examined the effects of pedal cadence on skeletal muscle oxygenation at low, moderate and peak exercise. Nine healthy subjects [24.7+/-6.3 years (SD)] performed incremental cycling exercise at 60 and 100 rpm. Pulmonary VO2(VO2P) was measured breath-by-breath and vastus lateralis oxygenation was determined by near-infrared spectroscopy (NIRS). The deoxyhemoglobin signal ([HHb]) from NIRS was used to estimate microvascular O2 extraction (i.e., [HHb] proportional, variant VO2M/QM). The VO2P and [HHb] for low, moderate and at peak exercise were determined. The VO2P at 60 rpm (low=0.64+/-0.13, moderate=2.03+/-0.38 and peak=3.39+/-0.84 l/min) were lower (P<0.01) than at 100 rpm (1.29+/-0.23, 2.14+/-0.39 and 3.54+/-0.88 l/min, respectively). There was a progressive increase in [HHb] from low to peak exercise. However, there was no significant difference (ANOVA, P=0.94) for the 60 (in microM, low=24.0+/-9.5, moderate=30.5+/-13.8 and peak=36.7+/-16.5) and 100 contractions/min (in microM, low=25.7+/-11.6, moderate=32.1+/-14.0 and peak=35.4+/-16.5). We conclude that vastus lateralis O2 extraction was similar at 60 and 100 cpm, suggesting that the VO2M/QM in the microcirculation was not altered and, presumably, no impairment of QM occurred with the increase in pedal frequency.

Kinetics of estimated human muscle capillary blood flow during recovery from exercise.

The kinetic characteristics of muscle capillary blood flow (Qcap) during recovery from exercise are controversial (e.g. one versus two phases). Furthermore, it is not clear how the overall Qcap kinetics are temporally associated with muscle oxygen uptake (VO2m) kinetics. To address these issues, we examined the kinetics of Qcap estimated from the rearrangement of the Fick equation (Qcap=VO2m/C(a-v)O2) using the kinetics of pulmonary VO2 (VO2p, primary component) and deoxy-haemoglobin concentration ([HHb]) as indices of VO2m and C(a - v)O2 (arterio-venous oxygen difference) kinetics, respectively. VO2p (l min-1) was measured breath by breath and [HHb] (microm) was measured by near infrared spectroscopy during moderate (M; below lactate threshold, LT) and heavy exercise (H, above LT) in nine subjects. The kinetics of Qcap were biphasic, with an initial fast phase (tauI; M=9.3+/-4.9 s and H=6.0+/-3.8 s) followed by a slower phase 2 (tauP; M=29.9+/-8.6 s and H=47.7+/-26.0 s). For moderate exercise, the overall kinetics of Qcap (mean response time [MRT], 36.1+/-8.6 s) were significantly slower than the kinetics of VO2p (tauP; 27.8+/-5.3 s) and [HHb] (MRT for [HHb]; 16.2+/-6.3 s). However, for heavy exercise, there was no significant difference between MRT-[HHb] (34.7+/-10.4 s) and tauP for VO2p (32.3+/-6.7 s), while MRT for Qcap (48.7+/-21.8 s) was significantly slower than MRT for [HHb] and tauP for VO2p. In conclusion, during recovery from exercise the estimated Qcap kinetics were biphasic, showing an early rapid decrease in blood flow. In addition, the overall kinetics of Qcap were slower than the estimated VO2m kinetics.

Dynamics of skeletal muscle oxygenation during sequential bouts of moderate exercise.

In rat muscle, faster dynamics of microvascular P(O2) (approximately blood flow (Q(m) to O2 uptake (V(O2) ratio) after prior contractions that did not alter blood [lactate] have been considered to be a consequence of faster V(O2) kinetics. However, in humans, prior exercise below the lactate threshold does not affect the pulmonary V(O2) kinetics. To clarify this apparent discrepancy, we examined the effects of prior moderate exercise on the kinetics of muscle oxygenation (deoxyhaemoglobin, [HHb] alpha V(O2m)/Q(m)) and pulmonary V(O2) (V(O2p) in humans. Eight subjects performed two bouts (6 min each) of moderate-intensity cycling separated by 6 min of baseline pedalling. Muscle (vastus lateralis) oxygenation was evaluated by near-infrared spectroscopy and V(O2p) was measured breath-by-breath. The time constant (tau) of the primary component of V(O2p) was not significantly affected by prior exercise (21.5 +/- 9.2 versus 25.6 +/- 9.7 s; Bout 1 versus 2, P= 0.49). The time delay (TD) of [HHb] decreased (11.6 +/- 2.6 versus 7.7 +/- 1.5 s; Bout 1 versus 2, P < 0.05) and tau[HHb] increased (7.0 +/- 3.5 versus 10.2 +/- 4.6 s; Bout 1 versus 2, P < 0.05), while the mean response time (TD + tau) did not change (18.6 +/- 2.7 versus 17.9 +/- 3.9 s) after prior moderate exercise. Thus, prior moderate exercise resulted in shorter onset and slower rate of increase in [HHb] during subsequent exercise. These data suggest that prior exercise altered the dynamic interaction between V(O2m)and Q(m) following the onset of exercise.

Muscle capillary blood flow kinetics estimated from pulmonary O2 uptake and near-infrared spectroscopy.

The near-infrared spectroscopy (NIRS) signal (deoxyhemoglobin concentration; [HHb]) reflects the dynamic balance between muscle capillary blood flow (Q(cap)) and muscle O(2) uptake (Vo(2)(m)) in the microcirculation. The purposes of the present study were to estimate the time course of Q(cap) from the kinetics of the primary component of pulmonary O(2) uptake (Vo(2)(p)) and [HHb] throughout exercise, and compare the Q(cap) kinetics with the Vo(2)(p) kinetics. Nine subjects performed moderate- (M; below lactate threshold) and heavy-intensity (H, above lactate threshold) constant-work-rate tests. Vo(2)(p) (l/min) was measured breath by breath, and [HHb] (muM) was measured by NIRS during the tests. The time course of Q(cap) was estimated from the rearrangement of the Fick equation [Q(cap) = Vo(2)(m)/(a-v)O(2), where (a-v)O(2) is arteriovenous O(2) difference] using Vo(2)(p) (primary component) and [HHb] as proxies of Vo(2)(m) and (a-v)O(2), respectively. The kinetics of [HHb] [time constant (tau) + time delay [HHb]; M = 17.8 +/- 2.3 s and H = 13.7 +/- 1.4 s] were significantly (P < 0.001) faster than the kinetics of Vo(2) [tau of primary component (tau(P)); M = 25.5 +/- 8.8 s and H = 25.6 +/- 7.2 s] and Q(cap) [mean response time (MRT); M = 25.4 +/- 9.1 s and H = 25.7 +/- 7.7 s]. However, there was no significant difference between MRT of Q(cap) and tau(P)-Vo(2) for both intensities (P = 0.99), and these parameters were significantly correlated (M and H; r = 0.99; P < 0.001). In conclusion, we have proposed a new method to noninvasively approximate Q(cap) kinetics in humans during exercise. The resulting overall Q(cap) kinetics appeared to be tightly coupled to the temporal profile of Vo(2)(m).

Muscle contraction-blood flow interactions during upright knee extension exercise in humans.

To test for evidence of a muscle pump effect during steady-state upright submaximal knee extension exercise, seven male subjects performed seven discontinuous, incremental exercise stages (3 min/stage) at 40 contractions/min, at work rates ranging to 60-75% peak aerobic work rate. Cardiac cycle-averaged muscle blood flow (MBF) responses and contraction-averaged blood flow responses were calculated from continuous Doppler sonography of the femoral artery. Net contribution of the muscle pump was estimated by the difference between mean exercise blood flow (MBFM) and early recovery blood flow (MBFR). MBFM rose in proportion with increases in power output with no significant difference between the two methods of calculating MBF. For stages 1 and 5, MBFM was greater than MBFR; for all others, MBFM was similar to MBFR. For the lighter work rates (stages 1-4), there was no significant difference between exercise and early recovery mean arterial pressure (MAP). During stages 5-7, MAP was significantly higher during exercise and fell significantly early in recovery. From these results we conclude that 1) at the lightest work rate, the muscle pump had a net positive effect on MBFM, 2) during steady-state moderate exercise (stages 2-4) the net effect of rhythmic muscle contraction was neutral (i.e., the impedance due to muscle contraction was exactly offset by the potential enhancement during relaxation), and 3) at the three higher work rates tested (stages 5-7), any enhancement to flow during relaxation was insufficient to fully compensate for the contraction-induced impedance to muscle perfusion. This necessitated a higher MAP to achieve the MBFM.