Growth hormone (GH) has insulin-antagonistic effects, and GH secretion is augmented during fasting and hypoglycemia. In the present study, 10 patients aged 21 to 28 years with childhood-onset GH deficiency (GHD) were studied during a 24-hour fast and a hypoglycemic glucose clamp before and after 9 months of GH replacement. During the 24-hour fast, blood glucose, serum insulin, and serum free fatty acid (FFA) levels were measured. In the hypoglycemic clamp, the counterregulatory hormones (plasma catecholamines, serum glucagon, and serum cortisol), serum insulin-like growth factor (IGF) binding protein-1 (IGFBP-1), serum FFA, and glucose uptake were measured. The GH dose was adjusted to the response of serum IGF-I, and the median GH dose was 0.14 IU/kg/wk (range, 0.08 to 0.19). At the end of the study, serum IGF-I levels were normalized in all but one patient, in whom serum IGF-I was above the normal range. Nine months of GH treatment did not cause any significant changes in the blood glucose level, insulin to glucose ratio, or serum FFA level during the 24-hour fast, and none of the patients experienced hypoglycemia either before or after GH treatment. However, GH therapy resulted in increased insulin resistance during hypoglycemia, without changes in the counterregulatory hormonal responses, serum IGFBP-1, or serum FFA.

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 ms1/2 to 440 +/- 30 ms1/2 (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.

During hypoglycaemia, counter-regulatory hormones are released, cognitive function is impaired and cerebral blood flow is increased. In the immediate period after normalisation of blood glucose only counter-regulatory hormones seem to be normalised. The aim of this study was to follow the changes in cerebral blood flow during a prolonged recovery period following moderate hypoglycaemia in normal man. In 15 healthy men, hypoglycaemia was induced by an intravenous infusion of insulin (2.5 mU/kg per min) to a blood glucose of 2.2 +/- 0.3 mmol/l (mean +/- S.D.) and was kept at this level for 66 +/- 11 min. The cerebral blood flow was measured by a single photon emission computed tomography camera (SPECT) recording the clearance of intravenously administered xenon-133. Measurements were performed before, at the beginning and at the end of the hypoglycaemic period, as well as 23 +/- 5, 51 +/- 7 and 97 +/- 7 min after normalisation of the blood glucose. The basal cerebral blood flow was 50.2 +/- 5.2 ml/100 g per min, increased to 55.6 +/- 4.9 ml/100 g per min (P < 0.001) during hypoglycaemia, and remained at this level at all measurements after normalisation of blood glucose. There was no relation between the rate of fall in blood glucose or level of hypoglycaemia and increment in cerebral blood flow or the actual blood flow during hypoglycaemia. The values of plasma adrenaline, serum ACTH, serum cortisol and symptom scores increased significantly during hypoglycaemia. The adrenaline level was back to the basal level at the first measurement after normalisation of blood glucose, while the ACTH level was normalised at the subsequent measurement and the cortisol level at the last measurement. In conclusion, the results show that despite normalisation of counter-regulatory hormones and hypoglycaemic symptoms, the cerebral blood flow remains elevated for at least 97 +/- 7 min following 66 +/- 11 min of moderate hypoglycaemia, indicating that additional factors which are not coupled to the cerebral metabolism influence this vasculatory response.

The restoration of cognitive function was studied in 10 healthy men aged 26 years (25.5 +/- 1.2 years; mean +/- SD) after insulin-induced hypoglycaemia (arterialized blood glucose 2.5 +/- 0.4 mmol l(-1)) for 62 +/- 8 min. Another group of six men participated in a single blind sham study for comparison. The hypoglycaemic event caused a significant increase (p = 0.006) in serum adrenaline levels. Ratings of adrenergically mediated symptoms increased during hypoglycaemia (p = 0.006), as did neuroglycopenic symptoms (p = 0.002), although neuroglycopenia ratings increased in both studies. During hypoglycaemia, P300 amplitudes in a relatively demanding visual search task decreased (p = 0.02), whereas easier tasks were unaffected. The amplitudes were restored after 40 min of normoglycaemia. Reaction time deteriorated after restoration of normoglycaemia, suggesting an effect of hypoglycaemia on learning. Thus, hypoglycaemia at a blood glucose level that is common among patients treated with insulin causes clear cognitive dysfunction, although restoration of the cognitive dysfunction to normal was fast.

The effect of rapid lowering of blood glucose on cerebral blood flow (CBF) was studied in 10 Type 1 (insulin-dependent) diabetic patients (age 23.5 +/- 3.8 years; mean +/- S.D.) with longstanding, poor metabolic control (HbAlc 11.2 +/- 1.0%; normal value 4.0-5.3%) using an intravenous xenon 133 single photon emission computed tomography technique. After a fall in blood glucose, during 81 +/- 11 min (mean +/- S.E.M.), from 18.2 +/- 1.4 mmol/l to 9.2 +/- 0.9 mmol/l CBF was unchanged, but increased from its initial value of 48.8 +/- 2.9 ml/100 g per min to 57.1 +/- 2.4 ml/100 g per min (P < 0.001) when the blood glucose level was restored. The CBF was higher in the right compared to the left hemisphere at all measurements (1.8 +/- 0.5 ml/100 g per min, P < 0.01; 1.9 +/- 0.5 ml/100 g per min, P < 0.05; 2.1 +/- 0.7 ml/100 g per min, P < 0.05, respectively). The change in CBF was inversely correlated with time for fall of blood glucose, but there was no correlation with absolute levels of blood glucose. The respiratory end-tidal PCO2 decreased during the low blood glucose level, but there was no correlation between the PCO2 and CBF. The cerebral volume was unchanged during the study. The results indicate that in patients with chronic hyperglycemia a rapid fall in blood glucose may cause a rise in CBF of the same magnitude as previously shown during absolute hypoglycemia in patients with well controlled diabetes mellitus and in normal subjects.(ABSTRACT TRUNCATED AT 250 WORDS)