Search

In diabetes prevention and care, invasiveness of glucose measurement impedes efficient therapy and hampers the identification of people at risk. Lack of calibration stability in non-invasive technology has confined the field to short-term proof of principle. Addressing this challenge, we demonstrate the first practical use of a Raman-based and portable non-invasive glucose monitoring device used for at least 15 days following calibration. In a home-based clinical study involving 160 subjects with diabetes, the largest of its kind to our knowledge, we find that the measurement accuracy is insensitive to age, sex, and skin color. A subset of subjects with type 2 diabetes highlights promising real-life results with 99.8% of measurements within A + B zones in the consensus error grid and a mean absolute relative difference of 14.3%. By overcoming the problem of calibration stability, we remove the lingering uncertainty about the practical use of non-invasive glucose monitoring, boding a new, non-invasive era in diabetes monitoring.

Background: Noninvasive glucose monitoring (NIGM) in diabetes is a long-sought-for technology. Among the many attempts Raman spectroscopy was considered as the most promising because of its glucose specificity. In this study, a recently developed prototype (GlucoBeam, RSP Systems A/S, Denmark) was tested in patients with type 1 diabetes to establish calibration models and to demonstrate proof of concept for this device in real use., Methods: The NIGM table-top prototype was used by 15 adult subjects with type 1 diabetes for up to 25 days at home and in an in-clinic setting. On each day, the subjects performed at least six measurement units throughout the day. Each measurement unit comprised two capillary blood glucose measurements, two scans with an intermittent scanning continuous glucose monitoring (CGM) system, and two NIGM measurements using the thenar of the subject's right hand., Results: Calibration models were established using data from 19 to 24 days. The remaining 3-8 days were used for independent validation. The mean absolute relative difference of the NIGM prototype was 23.6% ± 13.1% for the outpatient days, 28.2% ± 9.9% for the in-clinic day, and 26.3% ± 10.8% for the complete study. Consensus error grid analysis of the NIGM prototype for the complete study showed 93.6% of values in clinically acceptable zones A and B., Conclusions: This proof of concept study demonstrated a practical realization of a Raman-based NIGM device, with performance on par with early-generation CGM systems. The findings will assist in further performance improvements of the device.

Education programs for intensive insulin therapy were found to be valuable in improving glycemic control, but, due to low prevalence of type 1 diabetes in children and adolescents, access to those programs varies considerably in rural areas. We report on a telemedical care program that overcomes geographical isolation of patients on intensive insulin therapy. Sixty-one children and adolescents under the age of 26 participated in a telemedical care program. They stored daily information on blood glucose, injected insulin, meals and exercise in a glucosemeter with electronic memory and transferred the data via modem go a remote diabetes center outside of the region. By individual telephone consultations from home, they reviewed the data with a diabetologist at the diabetes center and adjusted their intensive insulin therapy in order to achieve predefined treatment goals. Patients were trained for 19 (6-48) weeks in the program and measured blood glucose 4.9 (1.7-4.9) times daily. Compared to the run-in-period, mean blood glucose had decreased (167 to 158 mg/dL, p < 0.01), standard deviation of blood glucose had decreased (81 to 70 mg/dL, p < 0.001), and frequency of hypoglycemia had decreased (5.2 to 3.3 in 4 weeks, p = 0.01) at the end of the program. The proportion of blood glucose values within the target range (80-180 mg/dL) had increased (47-55%, p < 0.001). HbAlc was reduced by 0.4% (-3.8 to +2.2%, p < 0.05). Telemedical care for intensive insulin therapy is safe, can improve glycemic control, and reduce the number of hypoglycemias in children and adolescents with type 1 diabetes.

The term insulin resistance includes both physiological and pathological metabolic changes. Physiological insulin resistance, for example, is to be found in puberty or in advanced old age; additionally, it may be caused by endocrinological diseases. A pathologically elevated insulin requirement is to be found in the insulin resistance syndrome, type A (genetic) or type B (insulin receptor antibodies). In recent times, however, a massively elevated insulin requirement has been observed in increasing numbers of type 2 diabetics, in whom a subcutaneous insulin resistance is suspected--a condition that can be verified by various tests. The treatment of choice for breaking down this resistance is the intravenous administration of sufficiently high-dosed insulin that is as closely matched to the sugar profile as possible. In many cases the return to subcutaneous insulin--which again must be adequately high dosed--becomes possible during the further course of the disease. In addition, a case of insulin resistance caused by patient manipulation is described.

Objective: The importance of screening for diabetic retinopathy has been established, but the best method for screening has not yet been determined. We report on a trial of assessment of digital photographs by telemedicine compared with standard retinal photographs of the same fields and clinical examination by ophthalmologists., Research Design and Methods: A total of 129 diabetic inpatients were screened for diabetic retinopathy by slit-lamp biomicroscopy performed by an ophthalmologist and by two-field 50 degrees non-stereo digital fundus photographs assessed by six screening centers that received the images by electronic mail. Conventional 35-mm transparencies of the same fields as the digital photographs were assessed by a retinal specialist and served as the reference method for detection of diabetic retinopathy. Slit-lamp biomicroscopy was the reference method for the detection of macular edema., Results: The prevalence of any form of diabetic retinopathy was 30% (n = 35); of sight-threatening retinopathy including macular edema, the prevalence was 6% (n = 7). The assessment of digital images by the six screening centers resulted in a median sensitivity of 85% and a median specificity of 90% for the detection of moderate nonproliferative or sight-threatening diabetic retinopathy. Clinically significant macular edema (n = 4) was correctly identified in 15 of the 24 grading reports. An additional seven reports referred the patients for further investigation because of concurrent diabetic retinopathy., Conclusions: Telescreening for diabetic retinopathy by an assessment of two-field 50 degrees non-stereo digital images is a valid screening method. Although detection of clinically significant macular edema using biomicroscopy is superior to digital or standard non-stereo photographs, only few patients with sight-threatening diabetic retinopathy are missed.

Simultaneous pancreas and kidney transplantation (PKT) is associated with a deterioration of hemorheology. We investigated the determinants of plasma and blood viscosity (hct. 35%) after PKT (n = 49), in type 1 diabetes (n = 26) and in healthy controls (n = 24). Patients after PKT were subdivided due to their graft function (intact pancreas and kidney graft, n = 26; pancreas rejected, intact kidney graft, n = 23). We examined the correlations of total serum protein, albumin, fibrinogen, alpha 2-macroglobulin, total cholesterol, LDL cholesterol, HDL cholesterol and triglycerides with plasma and blood viscosity (hct. 35%) measured at a continuous shear range of 600-0.2 s-1 with a rotational viscometer (Haake, Germany). Total protein was strongly associated with plasma viscosity in all examined groups (r > 0.5, p < 0.03), it determined blood viscosity over the whole shear range in type 1 diabetic patients, but only at high shear rates after PKT (> or = 100 s-1). The strong association of albumin and blood viscosity in type 1 diabetes and in healthy controls (shear rates > or = 10 s-1) was not found after PKT. Fibrinogen correlated with plasma and blood viscosity (> or = 25 s-1) after PKT (p < 0.03) but no in type 1 diabetic patients or healthy controls. Alpha 2-macroglobulin correlated with plasma and high shear blood viscosity after PKT only after pancreas rejection, no correlation was found after successful PKT. It also correlated with plasma and blood viscosity at low and high shear rates in type 1 diabetes. Total cholesterol and low shear blood viscosity correlated positively in successfully transplanted patients (r > 0.44), but negatively after pancreas rejection (r > -0.44). No correlation was found in type 1 diabetic patients, a positive association was found in healthy controls for plasma and low shear blood viscosity. LDL cholesterol correlated negatively (after pancreas rejection) or positively (healthy controls) with low shear blood viscosity (p < 0.03) and positively with plasma viscosity. HDL cholesterol was negatively associated with high shear blood viscosity in all groups (p < 0.05), except after successful PKT, where no association was found. It did not correlate with plasma viscosity in any group. Triglycerides did not contribute significantly to blood viscosity in the examined groups. The metabolic alterations after PKT influence plasma proteins, lipids and corpuscular elements of blood with regard to their effect on rheology.

The positive influence of simultaneous pancreas and kidney transplantation (PKT) on the development of diabetic microvascular lesions is well established. On the other hand, little is known on its impact on diabetic macrovascular disease, which is still the major cause of death in diabetes, including patients after PKT. In order to evaluate the influence of PKT on the cardiovascular risk profile, we performed a cross-sectional study on 55 patients. Special attention was given to the hemorheological parameters fibrinogen and plasma viscosity, two important cardiovascular risk factors, which so far have found no attention in the field of PKT research. The patients were subdivided into three groups according to their graft function: group 1-26 patients after successful PKT (no insulin dependency, serum creatinine <2 mg%), group 2-23 patients after PKT and rejection of the pancreas graft (insulin dependency, serum creatinine <2 mg%), group 3-6 patients after PKT with pancreas rejection and renal insufficiency (insulin dependency, serum creatinine >2 mg%, no dialysis). There was a high prevalence of arterial hypertension after PKT (group 1: 65%, group 2: 70%, group 3: 100%). Serum lipids were in the normal range as long as renal function was intact. In renal insufficiency, however, LDL-cholesterol and triglycerides were significantly elevated (p < 0.05). Fibrinogen was significantly raised after PKT (p < 0.001), as was plasma viscosity when the pancreas graft was rejected (p < 0.02). There was a tendency towards elevated fibrinogen levels with decreasing graft function. In conclusion, a number of cardiovascular risk factors were identified in patients after PKT, predominantly arterial hypertension and impaired hemorheology, with elevated fibrinogen levels and plasma viscosity. There is a further enhancement with decreasing graft function.

Plasma viscosity is mainly determined by large non-spherical proteins. In Type 1 diabetes mellitus, plasma viscosity increases with deterioration of diabetic control. Since protein glycation and formation of advanced glycosylation end products (AGEs) alter the structural and functional properties of proteins, AGEs might influence the rheological properties of plasma proteins. Therefore, we investigated the influence of plasma-AGEs on plasma viscosity in 34 normoalbuminuric diabetic patients (17 Type 1, 17 Type 2) with normal renal and liver function. In an additional experiment, 6 ml plasma of 9 healthy volunteers were incubated under sterile conditions for 14 days at 37.5 degrees C in the presence of 5.2 and 32.9 mmol l(-1) glucose. In diabetic patients, plasma-AGE levels were not correlated with plasma viscosity. Plasma-AGE levels in healthy controls (246 +/- 37 U ml[-1], mean +/- SD) were raised significantly (p<0.001) after the incubation at 37.5 degrees C (392 +/- 57 U ml[-1] and 552 +/- 58 U ml[-1], respectively). However, no difference was found in plasma viscosity pre- and post-incubation (pre-incubation: 1.25 +/- 0.04 mPas, post-incubation: 1.23 +/- 0.03 and 1.24 +/- 0.03, respectively). We conclude that there is no influence of plasma-AGEs on plasma viscosity.

In a retrospective analysis of 752 consecutive medical reports of patients with insulin- or non-insulin-dependent diabetes mellitus, we investigated the completeness of documentation of indicators of quality of care. The medical reports are the currently used form of documentation which is sent to the General Practitioner after the patient's discharge from hospital. The indicators of care were data on clinical history, physical examination, laboratory results and secondary complications. The documentation was incomplete; e.g. in 8.0% of insulin-dependent (IDDM) and in 26.4% of non-insulin-dependent diabetics (NIDDM), HbA1c was missing. In 7.6%, the type of diabetes was not stated. The frequency of recorded secondary complications was lower than it has to be expected considering metabolic control and duration of diabetes of the studied group. Documentation was more complete for IDDM patients. The reports of NIDDM patients with incipient or overt diabetic nephropathy revealed less frequent recordings of data on lipid metabolism and blood pressure compared to the group without nephropathy. The documentation of indicators of quality of care in medical reports for general practitioners is incomplete for many diabetic inpatients. Standardized methods of documentation are required urgently.

The hemorheological properties of blood play an important role in determining blood flow. Blood inertia, as characterized by blood density, controls blood flow in the large arteries, whereas blood viscosity becomes increasingly important with decreasing vessel diameter. In order to evaluate the impact of glycemic control on the rheological properties of blood, we examined viscosity (shear range: 600-0.2 s-1) and density of plasma and whole blood in 26 Type-1 diabetic patients and in 24 healthy controls, matched for age and sex. The diabetic subjects were subdivided into two groups according to their degree of glycemic control: 14 patients with good (HbA1c = 7.1 +/- 0.6%), and 12 patients with poor control (HbA1c = 8.7 +/- 0.7%). Diabetic patients as a whole did not differ from healthy controls in any of the rheological parameters. Subdivision of the patients due to their degree of glycemic control led to a marked rheological separation of Type-1 diabetic subjects with significantly lower plasma (P < 0.008) and whole blood viscosity (P < 0.03 at 10 and 25 s-1), and plasma density (P < 0.05) in well controlled patients. Compared with healthy controls, well controlled diabetic patients had significantly lower values of viscosity (P < 0.005) and density (P < 0.05) of plasma. Poorly controlled patients, on the other hand, did not differ from healthy controls in the examined rheological parameters. There seems to be a positive influence of good glycemic control on hemorheology in Type-1 diabetic patients.

Findings of the Nationale Verzehrstudie and Verbundstudie, Ernährungserhebung und Risikofaktorenanalytik studies are discussed in light of their clinical significance. Obesity increases the statistical risk of developing a number of diseases, of which the most important are cardiovascular morbidity and mortality. Obesity causes insulin resistance and plays a part in the development of the metabolic syndrome. The Verbundstudie, Ernährungserhebung und Risikofaktorenanalytik survey shows that in the mid-1980s roughly one third of the German population was overweight. However, criteria for defining the overweight condition are variable. Overall data in Germany are comparable to those of several other population surveys, but regional increases are observed. Analysis of dietary intakes shows a marked deviation from recommended nutritional standards, especially in the distribution of total caloric intake as carbohydrate (42.8%) and fat (38.4%). Almost 9% of the population derived more than 20% of their energy intake from snacks. The data cannot provide a full explanation as to why people become fat. This may be due to the well-known problem of underrating and underreporting intakes, particularly by obese subjects.

The development of implantable, remote-controlled insulin pumps dates back to the early 1970's when it was recognized that conventional insulin therapy may be inadequate to control microvascular complications. For the first prototypes the intraperitoneal access route was favoured because of a physiological portal/peripheral insulin gradient. With intraperitoneal insulin delivery excellent metabolic control can be obtained with glycohaemoglobin values close to the upper normal range. Although long-term studies in insulin-dependent diabetic patients show comparable results with respect to glycaemic control and intermediary substrate levels with intensive conventional therapy, the advantage of intraperitoneal insulin delivery may lie in the low rate of hypoglycaemic episodes.

Delayed gastric emptying is known as an important organic cause for brittle diabetes. We proposed the interval from the start of a meal to the rise in blood glucose, defined as blood glucose latency (T BG) as an index for gastric emptying and a non-invasive test for diabetic gastropathy. In order to validate this test we compared it in 22 type 1 diabetic patients with an established scintigraphic method for the measurement of gastric half-emptying time (T1/2) and found the following correlation: T BG = 4.4 + 0.162 x T1/2; r 0.79, P < 0.001. We therefore suggest measuring the blood glucose latency as a simple non-invasive screening method.

Both short-term care (blood glucose monitoring) and long-term care (clinical examinations) of diabetes generate an ample amount of data for each patient. Health care in hospitals has to provide services with respect to both demands. The quality of control depends on obtaining the right finding at the right time and taking the individually adequate measures. These repetitive activities follow to a certain extent standardized algorithms and computer-programs are able to support demands like this; however, up to now no attempts have been made to provide useful tools for this environment. DIALIN is a data-bank especially designed for the use in hospitals or outclinics and Camit is a diabetes management system for advanced evaluation of long-term blood glucose monitoring data. The expert-system DIACONS up to now determines diabetes type and adequate initial therapy from data of patients' history alone; it operates on the DIALIN-databank via SQL. DIALIN has proven to be a useful tool for data-processing in hospitals. Camit was well accepted by patients in a feasibility study. DIACONS has been tested with 83 diabetic patients to provide the correct diabetes-type and the proper initial regimen with a precision of 96% compared to the correspondence between two independent experts. The combination of all three systems is a step towards the Munich Medical Information system MAMIS.

After 2 wk of good control 9 type I diabetic patients were hospitalized and treated in a randomized fashion with different human insulin mixtures (regular: NPH), containing 10, 15, or 20% regular insulin. In the case of the 15:85% mixture, mean blood glucose was found to be in the range of 70-150 mg/dl. Significant differences occurred at certain time points between the curves obtained with the 10:90% and the 15:85% mixtures, and the 10:90% and the 20:80% preparations, respectively. Prolongation of the interval between hormone injection and start of breakfast from 30 to 45 min abolished the postprandial glucose excursion in the 15:85% and the 20:80% mixture, but not the 10:90% mixture.

Six to 12 hr after IP injection of 400 mg/kg of D-galactosamine in rats a 5-fold increase in plasma insulin was observed. In addition, impaired glucose assimilation was present after an IV Load in spite of unchanged fasting glucose levels. In streptozotocin-diabetic rats (100 mg/kg IV) plasma insulin remained diminished 12 h after induction of D-galactosamine hepatitis. Under identical conditions of preparation and incubation, the liver plasma membranes of D-galactosamine-treated rats, in both normal and diabetic states, bound only 40--60% as much insulin per mg of membrane protein as those of the control rats. Scatchard analysis suggested that this was due to a decrease in the number of receptor sites in the membranes of the D-galactosamine-injected rats. No difference in the insulin degrading capacity and in insulin-receptor dissociation of the plasma membranes between control and D-galactosamine-treated groups was found. These data suggest that a reduction in the number of hepatic insulin receptors in galactosamine hepatitis can lead to insulin resistance and hyperinsulinaemia.

A portable insulin dosing device (Siemens) was used together with a programmable pocket calculator and a glucose analyzer for short-term adaptation of continuous intravenous insulin infusion to blood glucose alterations. A special algorithm was developed which utilizes a given blood glucose value and the glucose rate of change obtained from two to four consecutive samples as input variables. In contrast to current techniques of feedback-regulation, which require continuous glucose monitoring, intermittent blood sampling allows greater mobility of patients. With the semi-automatic feedback system, euglycaemic control was obtained for 12-h periods in ten Type 1 (insulin-dependent) diabetic patients (maximum value 9.50 mmol/l, minimum value 2.83 mmol/l). Severe hypoglycaemia occurred in no case and additional control by glucose infusion appeared to be unnecessary. Light exercise after termination of insulin dose for standard meals led to glycaemic excursions with a rapid decrease (mean 1.08 +/- 0.09 mmol/l), followed by a rebound (0.59 +/- 0.07 mmol/l) in each patient. The amplitude of these excursions decreased with increasing distance from the peak of the meal dose. Comparison of feedback-control alone with feedback by glucose plus preprogrammed dose (4 U/h) at the onset of the test meal revealed lower post-prandial glucose levels (post-prandial maximum +/- SEM: 6.49 +/- 0.18 versus 7.71 +/- 0.79 mmol/l) and a lower infusion rate of insulin for the combined regimen (mean post-prandial maximum +/- SEM: 8.4 +/- 1.2 versus 12.0 +/- 0 IU/h). The system is useful for programming of portable infusion devices and studies based on euglycaemic control in unrestrained patients.

Partially purified, non-suppressible, insulin-like material (NSILA-S) was studied with respect to its effect on the levels of 3',5'-cyclic adenosine monophosphate (cAMP) and its mechanism of action in the control of this nucleotide in rat fat cells. NSILA-S prevents the rise of cAMP in fat cells under the influence of isoproterenol with similar kinetics to insulin. A maximal effect is observed at about 70 ng/ml with a biological activity equivalent to 200 muU/ml of insulin. NSILA-S inhibits norepinephrine-stimulated adenylate cyclase activity in fat cell ghosts and partially purified plasma membrane preparations. At 10 mM Mg2+, the inhibition is characterized by an effect of Vmax without change in affinity towards ATP (apparent KM 30 muM). Similarly there is no observed change in affinity towards Mg2+. With respect to inhibition of norepinephrine-stimulated adenylate cyclase, the dose-response curve of NSILA-S is similar to that already found with intact cells. The effect of norepinephrine is inhibited throughout the dose-response range between 5 X 10(-7) and 5 X 10(-4) M. In contrast to previous observations with insulin in ghosts, NSILA-S inhibits the basal adenylate cyclase activity. Cyclic nucleotide phosphodiesterase activity in homogenates as measured at 1.0 muM substrate is increased by 90% after previous incubation of fat cells with NSILA-S. The study suggests that the anti-lipolytic effect of NSILA-S is mediated by a lowering of cAMP through inhibition of the adenylate cyclase and/or stimulation of the phosphodiesterase system.

For the continuous intravenous application of insulin, a portable open-loop system was developed consisting of a delivery unit with a miniaturized pump and an insulin reservoir which is connected with an electronic control unit. The infusion rates were either preprogrammed or patient-controlled. Blood glucose control with both systems was tested in eight juvenile-type diabetics, among them two of the brittle type. Diabetic control during a 1-2-day pre-infusion period was compared with 2-3 days of continuous insulin infusion; as judged by the mean blood glucose value (MBG), the mean amplitude of glycemic excursions (MAGE), and glucosuria, all patients were significantly better controlled by the open-loop systems than by conventional therapy with subcutaneous insulin. The use of portable open-loop systems offers a promising approach to an improvement of metabolic control in insulin-requiring diabetics.

Control of the levels of cAMP in the early phase after addition of catecholamines and the effect of insulin is discussed under consideration of own findings from experiments with isolated fat cells of the rat. Data on the kinetics of cAMP are interpreted in the light of results from several groups of a rapid activation of phosphodiesterase activity along with the adenylate cyclase system. Comparison of energy metabolism of fat cells with the formation of cAMP under conditions of near-maximal activation of the adenylate cyclase system by isoproterenol shows that about half of the cellular ATP turnover is used for information transfer. Insulin reduces cAMP concentrations in the presence of isoproterenol within one min of incubation when added either together with or after the catecholamine. Experiments with propranolol and the phosphodiesterase inhibitor, methyl isobutylxanthine suggest an effect of insulin on formation and breakdown of cAMP.

The effects of the thyroid gland upon the nervous system with neurologic-psychiatric symptoms are discussed. With hyperthyreosis in younger patients hyperactive and psychotic signs occur, in older patients so-called apathic thyreotoxicosis. Neurologically there appear acute and chronic thyreotoxic myopathies, paroxysmal paralyses, the so-called hyperthyreotic chorea and a series of disturbances of eye muscles. In a thyreotoxic crisis psychotic preliminaries may proceed to a coma. With hyperthyreoses there occur intellectual impairment, showing but also psychoses as well as multiple neurological defect syndromes and epileptic seizures. Neurological defects as a consequence of thyreostatic therapy are rare, mostly these are disturbances of periphal nerves.

In 8 subjects in whom portal vein catheters had been inserted 5-6 days previously during cholecystectomy, arterial and portal concentrations of glucose, lactate, pyruvate, glycerol, alanine, free fatty acids, beta-hydroxybutyrate and acetoacetate revealed no significant differences. This provided the basis for the calculation of hepatic balances from arterio-hepatic venous substrate-differences in 17 healthy volunteers. In eight of them metabolic balances were determined during elevated hepatic lactate supply. Kinetics of the substrates throughout the whole test period in 9 controls showed no gross interference from the catheterization or infusion procedure. The elevated hepatic lactate concentration caused a doubling of hepatic glucose output, which could almost entirely be accounted for by a fivefold increase of hepatic lactate uptake. This acceleration of hepatic gluconeogenesis was accompanied by a significant increment of hepatic free fatty acid uptake, whereas hepatic ketone body production did not change. These data seem to support the view that hepatic energy requirements caused by an accelerated gluconeogenesis might be covered from enhanced free fatty acid oxidation.

Insulin binding experiments were performed with liver plasma membranes from guinea pig, calf and chicken. Bound insulin was separated from free insulin by a simple and rapid centrifugation of membranes through a layer of silicon oil. 125 I-labeled beef insulin was displaced from receptor sites by unlabelled guinea pig, beef and chicken insulin. The receptors of animals with insulins of different biological activity show similar basic characteristics and affinities to the different insulin molecules and thus are not specialised for the interactions with the homologous insulin molecule. The binding capacity of the membranes for beef insulin seems to be inversely related to the affinity of the homologous insulin to the receptor, guinea pig membranes showing the highest and chicken membranes the lowest receptor concentration.

In eight type I diabetics-three of them under intensified insulin therapy and five on continuous subcutaneous insulin infusion (CSII)-blood-glucose concentrations fell significantly during 20-minute isometric and dynamic muscle work at about 50% of individual maximal work capacity and subsequent intensive exercise of five minutes until exhaustion. However, 6 1/2 hours later blood-glucose levels (221 +/- 39 mg/dl after isometric exercise and 219 +/- 24 mg/dl after dynamic exercise) were clearly above the mean level measured at the same time of day on the three preceding days of 138 +/- 12 mg/dl (P less than 0.05). No hypoglycaemia occurred. This blood-sugar profile can be explained by a significant rise in cortisol and growth hormone and their long-lasting blood-sugar raising effect.