Blood glucose concentration dependent ACTH and cortisol responses to prolonged exercise

Summary. The purpose of this study is to investigate responses of serum ACTH and cortisol concentration to low intensity prolonged exercise. In experiment 1, 10 subjects fasted for 12 h and performed bicycle exercise at 49·3%V?O₂max (±4·3%) until exhaustion or up to 3 h. During the early part of the exercise, serum ACTH and cortisol concentrations did not increase from the pre-exercise values (ACTH: 44±5 μg/1, cortisol: 139±52 μg/1). Whilst the time to serum ACTH concentration increasing varied among the subjects (60·180 min), the increases of this hormone occurred for all subjects (175±85 ng/1, P<0·05) when blood glucose concentration decreased to a critical level of 3·3 mmol/1. At the end of the exercise, blood glucose concentration decreased to 2·60±0·21 mmol/1, and serum ACTH and cortisol concentrations increased to 313±159 μg/1 and 371±151 μg/1, respectively. In experiment 2, four subjects performed the same intensity exercise until exhaustion, and were then given 600 ml of 20 g glucose solution, and immediately afterwards, they were asked to repeat the same exercise. The subjects continued the exercise for between 30 to 90 min until again reaching exhaustion. During the second exercise, blood glucose concentration increased to the pre-exercise value (2·72±0·58 to 4·00±0·22 mmol/1, P<0·05) and simultaneously, serum ACTH concentration decreased considerably (354±22 to 119±54 ng/1, P<0·05). The results of the present study suggest that serum ACTH and cortisol concentration during low intensity prolonged exercise may be dependent on blood glucose concentration.     

Decreased leg glucose uptake during exercise contributes to the hyperglycaemic effect of octreotide

Aim: During prolonged infusion of somatostatin, there is an increase in arterial glucose concentration, and this increase persists even during prolonged exercise. The aim of the study was to measure glucose uptake in the leg muscles during infusion of the somatostatin analogue octreotide before and during leg exercise. Material and methods: Eight healthy male subjects were investigated twice in the fasting state: during 3 h infusion of octreotide [30 ng (kg min)^?1] or sodium chloride with exercise at 50% of maximal VO₂ in the last hour. Glucose uptake and oxygen uptake in the leg were measured using Fick’s principle by blood sampling from an artery and a femoral vein. Blood flow in the leg was measured using the indicator (indocyanine green) dilution technique. Results: After an initial decrease during rest, octreotide infusion resulted in a significant increase in arterial glucose concentrations compared to control conditions during exercise (mean ± SEM: 7·6 ± 0·6 versus 5·6 ± 0·1 mmol l^?1, P<0·01). During rest, octreotide did not change the leg glucose uptake (59 ± 10 versus 55 ± 11 μmol min^?1). In contrast, leg glucose uptake was significantly lower during exercise compared to control conditions (208 ± 79 versus 423 ± 87 μmol min^?1, P<0·05). During exercise, leg oxygen uptake was not different in the two experiments (20·4 ± 1·3 versus 19·5 ± 1·1 μmol min^?1). Conclusion: In conclusion, infusion of octreotide reduced leg glucose uptake during exercise, despite the same leg oxygen consumption and blood flow compared to control conditions. The hyperglycaemic effect of octreotide can partly be explained by the reduction in leg glucose uptake. Furthermore, the results suggest that a certain level of circulating insulin is necessary to obtain sufficient stimulation of glucose uptake in the exercising muscles.     

The forearm and leg perfusion techniques in man do not give the same metabolic information

Summary. The present study evaluates whether forearm and leg perfusion techniques give the same metabolic information. Seven patients hospitalized for operation of uncomplicated disease were investigated pre-operatively in the fasted state, while seven other patients who were on intravenous nutrition were studied in the fed state. Blood flow and the extremity exchange of glucose, lactate, glycerol, free fatty acids and amino acids were measured simultaneously across the forearm and the leg in all individuals. In the fasted state the arteriovenous difference (a-v) of glucose uptake was statistically significant across the forearm while it was statistically insignificant across the. leg (0·27 ± 0·06 vs. -0·04±0·13 mmol l¹). The a-v differences of glycerol (0·025 ± 0·028 vs. -0·043 ± 0·013 mmol l¹) and free fatty acids (0·10 ± 0·03 vs. -0·10 ± 0·04 mmol l¹) were positive across the forearm while they were negative across the leg (P < 0·01). In the fasted state the a-v difference of oxygen uptake (3·93 ± 0·67 vs. 3·21 ± 0·44 mmol l¹) and blood flow (4·1 ± 1·0 vs. 4·0 ± 0·7 ml min¹ 100 g¹) did not differ between the arm and the leg, but the a-v difference in carbon dioxide production was significantly higher (P<0.05) across the forearm (2·43 ± 0·37 vs. 1·29 ± 0·29 mmol l¹) compared to the leg. In the fed state all the above-mentioned differences between forearm and leg became statistically insignificant. In the fed state the a-v difference of the sum of all amino acids was not significantly different from zero balance across the forearm (-146 ± 103 mmol l¹) while there was a significant release from the leg (-175 ± 6 mmol l¹, P<0.05). In the fed state the flux of the sum of all amino acids became significantly positive across the arm while it was not significantly different from zero balance across the leg. In the fed state, forearm blood flow was significantly higher than leg blood flow (6·2 ± 0·5 vs. 4·0 ± 0·2 ml min¹ 100 g¹, P<0.001). The results in the present study demonstrate that the metabolic balance across regions of peripheral tissues may simultaneously differ considerably, i.e. being positive across the forearm and negative across the leg. This fact may imply that some previous claims may need reconsideration about ‘peripheral tissue metabolism’ associated with a certain clinical condition.     

Availability of glycogen and plasma FFA for substrate utilization in leg muscle of man during exercise

Summary. Six men performed two-legged cycle ergometer exercise at loads demanding 2–4 and 3–4 litres O₂ min^-1 (62 and 84% of Vo₂ max) for 20 min each or to exhaustion twice with 1 h rest between. An initial glycogen difference of 28 mmol glucose units kg^-1 of thigh muscle between the two legs was produced by one-legged exercise on a previous day followed by the consumption of a low carbohydrate diet. During the 1 h rest nicotinic acid (NA) was administered to inhibit lipolysis. Total body Fo₂ was unchanged by the NA administration. Work done by each leg, indicated by force on the pedals, was equal. RQ indicated a larger oxidation of fat in the leg with low glycogen. Muscle glycogen was 15 and 10 mmol kg^-1 in the normal and low glycogen leg at the end of the first exercise bouts and 3–8 mmol kg^-1 in both legs at exhaustion. The low glycogen leg extracted lactate from the blood whereas the normal leg released lactate and the uptake of glucose from the blood was greater by the low glycogen leg. These differences between the low glycogen and control legs did not persist during the NA condition when muscle glycogen content was equal in both legs. Further, the leg glucose uptake in the control and the NA conditions was positively related to the percentage of glycogen-empty muscle fibres and inversely to the glucose-6-P0₄ concentration. Thus the magnitude of the local glycogen stores of muscle influences the uptake and use of blood-borne substrates as well as determining endurance capacity during moderate to high intensity exercise.     

Effect of bicycle exercise on insulin absorption and subcutaneous blood flow in the normal subject

Summary. Elimination of 8 units ¹²⁵I-insulin and ^99mTc-pertechnetate from a subcutaneous depot on the thigh or the abdomen was studied at rest and during intense bicycle exercise in healthy postabsorptive volunteers. Disappearance rates of the tracers as well as plasma insulin and glucose concentrations were determined before, during and after the 20 min exercise period, and compared to corresponding values obtained during a non-exercise, control study on another day. Leg exercise caused a two-fold increase in the rate of ¹²⁵I-insulin disappearance from a leg depot (first-order rate constants rose from 0·68 ± 0·15 to 1·12 ± 0·12%·min^-1, P <0·05), but had no significant effect on the rate of disappearance from an abdominal depot (rate constants were 0·75 ±0·17 and 0·87±0·18%·min^-1 at rest and during exercise, respectively). ^99mTc-pertechnetate clearance from leg or abdomen showed no significant change during exercise, indicating that subcutaneous blood flow was unaltered. Leg, but not abdominal, injection of insulin was associated with a greater rise in plasma insulin during exercise than at rest. The average difference between exercise and control insulin area-under-curve in the leg group (1426 ± 594%·min) was significantly greater (P <0·05) than that from the abdominal group (298 ±251%· min). When the data from the two study groups were pooled, a direct relationship was found to exist between the change in ¹²⁵I-insulin disappearance rate and the change in plasma insulin concentration (r=0·61, P <0·02). Plasma glucose levels fell throughout the observation period both during the exercise and the control study, following leg as well as abdominal injection. The glucose decremental area was greater during exercise than at rest both following leg (P <0·05) and abdominal injection (P <0·01). The exercise-induced mean reduction in plasma glucose was 60% lower following abdominal injection, but this difference was not significant.     

Effects of physical exercise on insulin absorption in insulin-dependent diabetics. A comparison between human and porcine insulin

Summary. Nine insulin-dependent diabetics with undetectable plasma C-peptide (<0·05 nmol 1^-1) and without insulin antibodies (insulin binding to IgG<0·05 Ul^-1) received subcutaneous injections of 10 U ¹²⁵I-labelled soluble human or porcine insulin in the thigh on 2 consecutive days. Disappearance rates of ¹²⁵I were monitored continuously by external counting and plasma insulin levels were determined during rest for 30 min, bicycle exercise of moderate intensity for 40 min, and 60 min recovery. Subcutaneous blood flow was measured concomitantly in the contralateral thigh by the ¹³³Xenon clearance technique. During the initial period of rest human insulin was absorbed approximately 40% faster than its porcine analogue (first order rate constants 0·37±0·06 vs 0·27±0·06% min^-1, P<0·05) and the increment of the area under the plasma insulin curve was greater after hum-ii than after porcine insulin (184±46 vs 112±42 mUl^-1 min, P<0·05). Exercise enhanced the absorption rates for both ¹²⁵I-insulins to 0·50±0·06 and 0·48±0·10% min^-1 for human and porcine insulin, respectively (P<0·05). This increase was less pronounced for human compared to porcine insulin (49±19 vs 105±40%, P=0·06). During exercise plasma insulin rose to 37±5 mUl^-1 after human and 30±5 mUl^-1 after porcine insulin and the areas under the plasma insulin curves were similar. During the recovery phase the absorption rates decreased slightly compared to the exercise value for both insulins. The blood glucose lowering effect was similar for the two insulins. Subcutaneous blood flow was not significantly altered by exercise in either group. It is concluded that during rest human soluble insulin is more rapidly absorbed than porcine insulin. Physical exercise tends to increase porcine insulin absorption more and eliminates the basal difference in the absorption kinetics between human and porcine insulin. The increased insulin absorption during exercise is not coupled to corresponding changes in the subcutaneous blood flow.     

Glucose metabolism in chronic diabetic foot ulcers measured in vivo using microdialysis

Ten subjects with diabetes mellitus and unilateral chronic foot ulcer were investigated. Local tissue concentrations of glucose and lactate were measured using the microdialysis method at a distance of 0·5–1 cm from the edge of the ulcer and in normal skin in the contralateral foot. Subcutaneous blood flow in the area investigated was measured using the ¹³³Xe-washout technique. The interstitial glucose concentration in the ulcer was found to be lower than in intact skin (8·0 ± 1·0 mmol l^?1 vs. 8·5 ± 1·1 mmol l^?1) (P<0·02), and the interstitial lactate concentration was higher in the ulcer than in intact skin (3·2 ± 0·2 mmol l^?1 vs. 2·1 ± 0·3 mmol l^?1) (P<0·01). The subcutaneous blood flow was on average 40% higher in the ulcer than in the intact skin. The calculated local glucose uptake and lactate outputs were twofold higher in the ulcer than in the intact skin. However, the molar ratio between lactate output and glucose uptake was approximately two, both in the ulcer and in the intact skin, indicating that the glucose metabolism was qualitatively the same in the two regions.     

Influence of combined intravenous and oral glucose administration on splanchnic glucose uptake in man

Summary. The influence of intravenous plus oral glucose administration on splanchnic glucose handling was examined in healthy young individuals by combining the hepatic vein catheterization technique with the double glucose tracer method. After 1 h of steady state hyperglycaemia (11·7 Itim ) induced by intravenous glucose alone (hyperglycaemic clamp technique), subjects ingested 89 ± 1 g of glucose, and the hyper-glycaemic plateau was maintained for the subsequent 4 h by adjusting the exogenous glucose infusion rate. Over the 4-h absorptive period, only 51 ± 4 g of oral glucose (i.e. 58 ±4% of the ingested load) appeared in the systemic circulation, while 193 ± 15 g (1·072±0·083 mol) of glucose had to be infused exogenously to sustain the hyperglycaemia. Endogenous glucose production was suppressed by over 60%. Net splanchnic glucose balance switched from a positive value (i.e. net uptake) of 506 ± 2–56 uniol min^-1kg^-1with intravenous glucose alone (0·60 min) to a negative one (i.e. net output) of 12·50 ± 2·44 u. mol min^-1kg^-1during 4 h (60–300 min) of intravenous+oral glucose. The mean rate of splanchnic glucose uptake was estimated to be 6·39 ±4·67 ixmol min^-1kg^-1with intravenous glucose alone, and 8·83 ±4·28 u. mol min^-1kg^-1with intravenous+oral glucose. In either case, the large majority (80–90%) of the glucose appearing in the systemic circulation was disposed of by extrasplanchnic tissues. These results indicate that pre-existing hyperglycaemia and/or hyperinsulinaemia inhibit gastrointestinal glucose absorption, and that oral glucose administration does not result in a major redistribution of intravenous glucose between splanchnic and extrasplanchnic tissues.     

Interactions of stress hormones on lipid and carbohydrate metabolism in man with partial insulin deficiency

Abstract. The metabolic responses to 4-h infusions of adrenaline (3 μg kg^-1 h^-1) and cortisol (10 mg m^-2 h^-1 for 2 h followed by 5 mg m^-2 h^-1 for 2 h), separately and in combination, have been studied in six healthy subjects with concurrent somatostatin infusion (250 μg h^-1). A combined infusion of adrenaline, cortisol, glucagon (180 ng kg^-1 h^-1) and somatostatin has also been studied. Somatostatin plus adrenaline and somatostatin plus cortisol resulted in hyperglycaemia (at 240 min, somatostatin plus adrenaline 11·4 ± 0·4 mmol l^-1, P < 0·001; somatostatin plus cortisol 6·7 ± 0·3 mmol l^-1, P < 0·05; somatostatin alone 4·9 ± 0·4 mmol l^-1). No synergistic effect on blood glucose was seen with adrenaline and cortisol together. When glucagon was added, blood glucose rose more rapidly than without glucagon (9·3 ± 0·4 mmol l^-1v. 7·2 ± 0·5 mmol l^-1 at 45 min, P < 0·001), but plateau values were similar. Plasma NEFA levels were raised by somatostatin plus adrenaline (0·55 ± 0·04–1·82 ± 0·11 mmol l^-1 at 60 min). Somatostatin plus cortisol had no more effect on plasma NEFA than somatostatin alone. During the combined infusion of somatostatin plus adrenaline plus cortisol, a synergistic effect on plasma NEFA was observed (2·30 ± 0·11 mmol l^-1 at 60 min, P < 0·01 v. somatostatin plus adrenaline). This occurred despite a small escape of insulin secretion. The lipolytic actions of adrenaline are potentiated by elevated circulating cortisol levels in insulin-deficient man. Glucagon does not modify this response, but accelerates the development of hyperglycaemia.     

Interstitial lactate levels in human skin at rest and during an oral glucose load: a microdialysis study

In vitro data have suggested that the skin is a significant lactate source. The purpose of the present study was to measure lactate and glucose concentrations in intact human skin in vivo using the microdialysis technique. Microdialysis fibres of 216 μm were inserted intradermally and perfused at a rate of 3 μl min^–1. In the first experimental protocol, dialysis fibres were calibrated by the method of no net flux in eight subjects. Skin lactate concentrations of 2·48 ± 0·17 mmol l^–1 were significantly greater than lactate concentrations of 0·84 ± 0·15 mmol l^–1 in venous plasma (P<0·01). Glucose concentrations in skin and venous plasma were similar (5·49 ± 0·18 vs. 5·26 ± 0·24 mmol l^–1). In the second experimental protocol, changes in lactate and glucose levels were studied in 10 subjects after an oral glucose tolerance test (OGTT). After the OGTT, plasma glucose and lactate levels increased by 54% and 39% to peak levels at 30 and 60 min respectively. In comparison, skin glucose and lactate increased by 41% and 18% at 60 and 90 min. No changes in skin blood flow were observed during the OGTT. The data suggest that resting skin is a significant lactate source with no significant lactate production during OGTT. The cellular source of lactate in the skin remains undetermined to date.     

Muscle glycogen concentration during recovery after prolonged severe exercise in fasting subjects

The influence of 12 h of fasting after prolonged severe exercise on the muscle glycogen concentration was studied in 5 normal subjects. The subjects exercised in the post absorptive state at 70% of max. V_o2 till exhaustion, then rested for 12 h. No food was allowed during recovery. Blood samples and muscle biopsies were obtained before exercise, immediately after the cessation of exercise, and after 2, 4, 6, 9 and 12 h of recovery. Muscle glycogen content decreased from 70.4 ± 3.0 to 21.6 ±3.9 mmol glucosyl units/kg w.w. in response to exercise. After 4 h of recovery muscle glycogen had increased to 28.8 + 3.6 mmol glucosyl units/kg (P<0.025). During the next 8 h of recovery no further increase in glycogen concentration was observed. Mean plasma glucose concentration decreased from 5.25±0.16 to 4.37±0.18 mmol/1 during exercise (P<0.001). No change in the plasma glucose level was observed during recovery. Immunoreactive insulin (IRI) concentration decreased from 15.9±1.0 to 10.2±0.5 μU/ml (P<0.001) during exercise, and remained at this level during recovery. It is concluded that some muscle glycogen repletion may occur after prolonged, severe exercise even under fasting conditions. It is suggested that this may proceed through an increased hepatic gluconeogenesis.     

Effects of intense exercise training on endothelium-dependent exercise-induced vasodilatation

To determine whether intense exercise training affects exercise-induced vasodilatation, six subjects underwent 4 weeks of handgrip training at 70% of maximal voluntary contraction. Exercise forearm vascular conductance (FVC) responses to an endothelium-dependent vasodilator (acetylcholine, ACH; 15, 30, 60 μg min^?1) and an endothelium-independent vasodilator (sodium nitroprusside, SNP; 1·6, 3·2, 6·4 μg min^?1) and FVC after 10 min of forearm ischaemia were determined before and after training. Training elicited significant (P<0·001) increases in grip strength (43·4 ± 2·3 vs. 64·1 ± 3·5 kg, before vs. after, mean ± SEM), forearm circumference (26·7 ± 0·4 vs. 27·9 ± 0·4 cm) and maximal FVC (0·4630 ± 0·0387 vs. 0.6258 ± 0·0389 units, P<0·05). Resting FVC did not change significantly with training (0·0723 ± 0·0162 vs. 0.0985 ± 0·0171 units, P>0·4), but exercise FVC increased (0·1330 ± 0·0190 vs. 0.2534 ± 0·0387 units, P<0·05). Before and after the training, ACH increased exercise FVC above the control (no drug) exercise FVC, whereas SNP did not. Training increased (P<0·05) the exercise FVC responses to ACH (0·3344 ± 0·1208 vs. 0.4303 ± 0·0858 units, before vs. after training, 60 μg min^?1) and SNP (0·2066 ± 0·0849 vs. 0.3172 ± 0·0628 units, 6·4 μg min^?1). However, these increases were due to the increase in control (no drug) exercise FVC, as the drug-associated increase in exercise FVC above control did not differ between trials (P>0·6). These results suggest that exercise FVC is increased by both exercise training and stimulating the release of endothelium-dependent vasodilators. However, training does not affect the vascular response to these vasodilators.     

Impaired responses of plasma catecholamines to exercise in diabetic patients with abnormal heart rate reactions to tilt

Summary. The plasma catecholamine response to a standardized bicycle exercise test was evaluated in 24 insulin-dependent diabetic (IDDM) patients in whom the heart rate reactions to deep breathing (E/I ratio) and to tilt, the immediate acceleration and the transient deceleration (acceleration and brake indices), had been assessed as tests of autonomic neuropathy. Patients with an abnormal acceleration index (n= 8) showed, compared with non-diabetic (n= 18) controls who had participated in previous studies, an impaired increment in noradrenaline during exercise (80% of maximal working capacity) (MWC) (12·38 ± 1·46 nmol l^-1 vs. 18·74 ± 1·45 nmol I^-1; P<0·01) and adrenaline (50% of MWC: 0·25 ± 0·04 nmol I^-1 vs. 0·54 ± 0·08 nmol I^I–1; P<0·05). Similarly, patients with an isolated abnormal brake index (n= 6), i.e. with a normal acceleration index and a normal E/I ratio, showed compared with controls an impaired increment in noradrenaline (9·53 ± 1·66 nmol I^-1 vs. 18·74 ± 1·45 nmol I^-1; P<0·01) and adrenaline (1·41 ± 0·22 nmol I^-1 vs. 2·92 ± 0·51 nmol I^-1; P<0·05) during 80% of MWC. IDDM patients with abnormal heart rate reactions to tilt, an abnormal acceleration index or an abnormal brake index show impaired catecholamine responses to exercise, which can be demonstrated also in patients without signs of parasympathetic neuropathy.     

Effects of glucose on alanine-derived urea synthesis

Summary. The effect of glucose on alanine-stimulated urea synthesis was studied in six healthy volunteers during 6 h of constant alanine infusion, 2·8 mmol h^-1 kg^-1 b. wht., and during 12 h of constant glucose infusion, 4·0 mmol h^-1 kg^-1 b. wht., with superimposed alanine infusion. The urea nitrogen synthesis rate (UNSR) was determined at intervals of 2 h as urinary excretion rate corrected for accumulation and intestinal hydrolysis. UNSR depended on the blood alanine and glucagon concentration, but was not correlated with glucose, lactate. or insulin concentrations. The slope of the linear relation between UNSR and alanine concentration (the ‘Functional Hepatic Nitrogen Clearance’) was on the average 24·4 1 h^-1 and decreased to 12·8 1 h^-1 by glucose (mean difference k SE of the difference 10·6±7·3, P<0·01). The relation between glucagon and alanine concentration was linear, and the slope was decreased to 40 per cent by glucose (P<0·05). The slope of the linear relation between UNSR and glucagon was not changed by glucose. Thus the catabolism of alanine nitrogen is decreased by glucose because of a reduction of the urea synthesis. Data suggest that this may be due to a depression of the glucagon response to alanine.     

Carbohydrate utilization by exercising muscle following pre-exercise glucose ingestion

Summary. The effect on exercising muscle metabolism of prior ingestion of 200 g glucose was examined in six healthy subjects during 40 min leg exercise at 30% of maximal oxygen uptake. Leg glucose uptake during exercise was on average two- to three-fold higher after glucose (E+G) compared to exercise without glucose (E) and could account for 44–48% of the oxidative leg metabolism (control value: 19%, P<0·05-0·01). In contrast to E, which was associated with a significant release of leg lactate, pyruvate and alanine, E+G gave no leg production of lactate or alanine and an uptake of pyruvate. The respiratory exchange ratios (R) were higher during G + E and corresponded to a carbohydrate oxidation of 54–69% as against 46–49% (P0<·05-0·01) during E. Estimated from R-values and leg oxygen and glucose uptakes, carbohydrate oxidation during G<E was almost completely accounted for by blood glucose. During E, on the other hand, carbohydrate oxidation exceeded leg glucose uptake, indicating a small but significant muscle glycogen breakdown (P<0·01). The rate of glycogen utilization during E or G + E was too small to be detected by direct measurements of muscle glycogen content. The results demonstrate that glucose ingestion prior to light exercise is followed by increased uptake and more efficient oxidation of glucose, as well as by insignificant muscle glycogen degradation by exercising muscle. Although the present findings suggest a glycogen-conserving effect of glucose ingestion under these conditions, the main fuel shift is from fat to glucose oxidation.     

Insulin sensitivity in alcoholic cirrhosis

The amount-of-substance rate of glucose metabolism and its sensitivity to the concentration of insulin was quantified in 10 non-diabetic patients with alcoholic cirrhosis of varying severity, using the ‘glucose clamp technique’. Fasting glucose and insulin were 5.4±0.3 mmol/1 and 187±50 μmol/1 (mean ± SEM), respectively. During the hyperglycaemic clamp (blood glucose at 12.5 mmol/1) the glucose metabolic rate (divided by body mass) was 27± 4 μmol·min^?1·kg^?1 at an insulin concentration of 998± 158 pmol/1. Thus the insulin sensitivity of the tissue glucose metabolism was 22±7 m³·min^?1·kg^?1. During the euglycaemic clamp exogenous insulin was given to a concentration of 574± 72 pmol/1. The resulting glucose metabolic rate was 20± 4 μmol·min^?1·kg^?1 and the insulin sensitivity the same as during hyperglycaemia. The calculated systemic delivery rate of insulin (divided by body surface area) was 783± 172 pmol·min^?1·m^?2. Fasting glucagon was 32± 5 pmol/ and only partly depressed by glucose or insulin. In comparison with stated relevant control groups cirrhotics exhibit glucose intolerance characterized by decreased sensitivity to insulin, hyperinsulinaemia due to increased release, and hyperglucagonaemia with decreased suppressibility. There was no relation between clinical or biochemical data of the patients and the above results, suggesting that the abnormal glucose metabolism does not depend directly on the decreased liver function but on a disturbed pancreatic-hepatic-peripheral axis.     

Hypoglycaemia leads to an increased QT interval in normal men

Hypoglycaemia is presumed to be the cause of death in about 3% of insulin-treated diabetic patients. Some of these patients suffer from hypoglycaemic brain damage, but the majority have no evident brain damage and are supposed to have died from other causes such as a cardiac arrhythmia. The putative mechanism is a hypoglycaemia-induced prolongation of the QT interval which increases the risk of malignant ventricular tachycardia. The aim of the present study was to examine the electrocardiogram during and after hypoglycaemia in healthy men. To that end, hypoglycaemia was induced by an intravenous infusion of insulin (2·5 mU kg^?1 min^?1) in 10 healthy men to reach a venous blood glucose level of 2·1 ± 0·3 mmol l^?1 for 65 ± 9 min. Before hypoglycaemia, after 20 and 50 min of hypoglycaemia and 20 and 45 min after normalization of the blood glucose, the QT interval was measured by a ruler and corrected for the heart rate. Results are given as mean ± SD and comparisons were made with an ANOVA , except for symptom scores and plasma adrenaline where non-parametric tests were used. When this indicated significance, further analysis was performed with a two-tailed t-test. During hypoglycaemia the corrected QT interval increased from 380 ± 20 ms^½ to 440 ± 30 ms^½ (P<0·001), and the amplitude of the T wave decreased (P = 0·002). The serum potassium level decreased from 4·3 ± 0·3 mmol l^?1 to 3·5 ± 0·2 mmol l^?1 (P<0·001) and the plasma adrenaline concentration increased from 0·20 ± 0·04 nmol l^?1 to 2·46 ± 2·58 nmol l^?1 (P<0·01). The results of this study confirm that a prolongation of the QT interval occurs during hypoglycaemia, but the significance of this finding still has to be proven.     

Perfusion heterogeneity does not explain excess muscle oxygen uptake during variable intensity exercise

The association between muscle oxygen uptake (VO₂) and perfusion or perfusion heterogeneity (relative dispersion, RD) was studied in eight healthy male subjects during intermittent isometric (1 s on, 2 s off) one‐legged knee‐extension exercise at variable intensities using positron emission tomography and a‐v blood sampling. Resistance during the first 6 min of exercise was 50% of maximal isometric voluntary contraction force (MVC) (HI‐1), followed by 6 min at 10% MVC (LOW) and finishing with 6 min at 50% MVC (HI‐2). Muscle perfusion and O₂ delivery during HI‐1 (26 ± 5 and 5·4 ± 1·0 ml 100 g^?1 min^?1) and HI‐2 (28 ± 4 and 5·8 ± 0·7 ml 100 g^?1 min^?1) were similar, but both were higher (P<0·01) than during LOW (15 ± 3 and 3·0 ± 0·6 ml 100 g^?1 min^?1). Muscle VO₂ was also higher during both HI workloads (HI‐1 3·3 ± 0·4 and HI‐2 4·1 ± 0·6 ml 100 g^?1 min^?1) than LOW (1·4 ± 0·4 ml 100 g^?1 min^?1; P<0·01) and 25% higher during HI‐2 than HI‐1 (P<0·05). O₂ extraction was higher during HI workloads (HI‐1 62 ± 7 and HI‐2 70 ± 7%) than LOW (45 ± 8%; P<0·01). O₂ extraction tended to be higher (P = 0·08) during HI‐2 when compared to HI‐1. Perfusion was less heterogeneous (P<0·05) during HI workloads when compared to LOW with no difference between HI workloads. Thus, during one‐legged knee‐extension exercise at variable intensities, skeletal muscle perfusion and O₂ delivery are unchanged between high‐intensity workloads, whereas muscle VO₂ is increased during the second high‐intensity workload. Perfusion heterogeneity cannot explain this discrepancy between O₂ delivery and uptake. We propose that the excess muscle VO₂ during the second high‐intensity workload is derived from working muscle cells.     

Autonomic nervous function at rest in aerobically trained and untrained oldermen

Therelationship between aerobictraining, vagal influence on the heart and ageing was examined by assessing aerobic fitness andresting heart rate variability in trained and untrained older men. Subjects were 11 trained cyclistsand runners (mean age=6±61·6 years) and 11 untrained, age-matchedmen (mean age=66±1·2 years). Heart rate variability testing involvedsubjects lying supine for 25 min during which subjects’ breathing was paced andmonitored (7·5 breaths min^?1). Heart rate variability was assessedthrough time series analysis (HRV_ts) of the interbeat interval. Results indicated thattrained older men (3·55±0·21 l min^?1) hadsignificantly (P<0·05) greater VO _2maxthan that of control subjects (2·35±0·15 l min^?1).Also, trained older men (52±1·8 beats min^?1) hadsignificantly (P<0·05) lower supine resting heart rate than that of controlsubjects (65±4·2 beats min^?1). HRV_ts at highfrequencies was greater for trained men (5·98±0·22) than for untrainedmen (5·23±0·32). These data suggest that regular aerobic exercise inolder men is associated with greater levels of HRV_ts at rest.     

Effect of acute acidosis and alkalosis on leucine kinetics in man

Summary. The effects of acute pH changes on whole body leucine kinetics (1-¹³C-leucine infusion technique) were determined in normal subjects. Plasma insulin, glucagon, and growth hormone concentrations were kept constant by somatostatin and replacement infusions of the three hormones. When acidosis was produced by ingestion of NH4CI (4 mmol kg^-1 p. os; n = 8) arterialized pH decreased within 3 h from 7.39±0.01 to 7.31 ±0.01 (P<0.001) and leucine plasma appearance increased by 0.13 ±0.04 μmol kg^-1 min^-1 (P<0.02); in contrast, when alkalosis was produced by intravenous infusion of 4 mmol kg^-1 NaHCO₃ (n= 1, pH 7.47 ±0.01), leucine plasma appearance decreased by -0.09 ± 0.04 (xmol kg^-1 min^-1 (P<0.01 vs. acidosis). Whole body leucine flux also increased during acidosis compared to alkalosis (P<0.05), suggesting an increase in whole body protein breakdown during acidosis. Apparent leucine oxidation increased during acidosis compared to alkalosis (P=0.05). Net forearm leucine exchange remained unaffected by acute pH changes. Plasma FFA concentrations decreased during acidosis by -107 ±67 μmol l^-1 (P<0.05) and plasma glucose increased by 1.90±0.25 mmol l^-1 (P<0.02); in contrast, alkalosis resulted in an increase in plasma FFA by 83 ± 40 (μmol 1^-1 (P<0.02; P<0.01 vs. acidosis), suggesting an increase in lipolysis; plasma glucose decreased compared to acidosis (P<0.01). The data demonstrate that acute metabolic acidosis and alkalosis, as they occur in clinical conditions, influence protein breakdown, and in the opposite direction, lipolysis.