Whole body clearance of norepinephrine. The significance of arterial sampling and of surgical stress.

The whole body clearance of norepinephrine (NE) was measured in seven patients pre- and postoperatively. L[(3)H]NE was infused intravenously for 90 min and steady-state concentrations of L[(3)H]NE were measured at 75 and 90 min in both arterial and peripheral venous blood. Preoperatively, in the resting supine position, the clearance values based on arterial and venous sampling averaged 1.4 and 2.5 liter/min, respectively (P < 0.02). The difference in clearance values was due to a peripheral uptake of NE averaging 45%. The mean plasma NE increased from 1.70 nmol/liter preoperatively to 5.20 nmol/liter postoperatively (P < 0.02). The plasma appearance rate of NE averaged 2.4 nmol/min before surgery and it increased to 9.5 nmol/min postoperatively (P < 0.02). The plasma clearance of NE averaged 1.4 and 1.6 liter/min pre- and postoperatively, respectively (not significantly different). Our study demonstrates that the calculation of plasma NE clearance based on venous sampling results in values that are too high. Furthermore, such values may be influenced by individual variations in the peripheral uptake of NE, since we found no correlation between clearance values based on venous and arterial sampling. The increase in plasma NE postoperatively was due to an increase in the plasma appearance rate of NE because the clearance rate did not change.

Hemodynamics in diabetic orthostatic hypotension.

Hemodynamic variables (blood pressure, cardiac output, heart rate, plasma volume, splanchnic blood flow, and peripheral subcutaneous blood flow) and plasma concentrations of norepinephrine, epinephrine, and renin were measured in the supine position and after 30 min of quiet standing. This was done in normal subjects (n = 7) and in juvenile-onset diabetic patients without neuropathy (n = 8), with slight neuropathy (decreased beat-to-beat variation in heart rate during hyperventilation) (n = 8), and with severe neuropathy including orthostatic hypotension (n = 7). Blood pressure decreased precipitously in the standing position in the diabetics with orthostatic hypotension, whereas moderate decreases were found in the other three groups. Upon standing, heart rate rose and cardiac output and plasma volume decreased similarly in the four groups. The increases in total peripheral resistance, splanchnic vascular resistance and subcutaneous vascular resistance were all significantly lower (P less than 0.025) in the patients with orthostatic hypotension compared with the other three groups. The increase in plasma norepinephrine concentrations in the patients with orthostatic hypotension was significantly lower (P less than 0.025) than in the patients without neuropathy, whereas plasma renin responses to standing were similar in the four groups. We conclude that in diabetic hypoadrenergic orthostatic hypotension the basic pathophysiological defect is lack of ability to increase vascular resistance, probably due to impaired sympathetic activity in the autonomic nerves innervating resistance vessels; cardiac output and plasma volume responses to standing are similar to those found in normal subjects and in diabetics without neuropathy.

Decreased sympathetic vasomotor tone in diabetic orthostatic hypotension.

In normals, subcutaneous blood flow in the ankle region, measured by means of the 133Xe washout technique, decreases about 45% when the position of the ankle is changed from cardiac level to 50 cm below the heart. A sympathetic vascular axon reflex is responsible for this flow reduction. A normal response [mean blood-flow reduction, 49% (range, 40 to 64%)] was found in seven diabetics without neuropathy. In nine diabetics with autonomic neuropathy (beat-to-beat variation in heart rate less than or equal to 15 min-1 during hyperventilation), the reduction in blood flow on lowering of the ankle was also normal. A significant deterioration of the response was found in six diabetics with autonomic neuropathy, including orthostatic hypotension, in whom blood flow increased 23% (range, -35 to +77%). Autoregulation of blood flow, i.e., constancy of blood flow in the ankle region on lowering and elevating of the ankle 20 cm below and above cardiac level was intact in the patients with orthostatic hypotension, indicating a normal vascular smooth muscle function. In conclusion, in diabetic patients with orthostatic hypotension, the subcutaneous axon reflex is absent. This may be one of several factors responsible for their orthostatic hypotension.

Discrepancy between plasma C-peptide and insulin response to oral and intravenous glucose.

Plasma insulin, proinsulin, and C-peptide responses to 25 g glucose orally and intravenously administered were measured in 10 healthy males. Plasma insulin response was higher during the oral load in accordance with the "incretin" concept. However, the actual amount of insulin secreted, as measured by the plasma C-peptide response, was similar during the two glucose loads. The higher plasma insulin response after oral glucose was not due to crossreactivity with proinsulin in the insulin assay. These results suggest that the higher plasma insulin response during an oral glucose load is due at least partially to a lower hepatic extraction of insulin.

Maintenance of high-energy brain phosphorous compounds during insulin-induced hypoglycemia in men. 31P nuclear magnetic resonance spectroscopy study.

31P nuclear magnetic resonance (NMR) spectroscopy allows noninvasive studies of cerebral energy-rich phosphorous compounds in humans. In an attempt to characterize the relationship between peripheral blood glucose concentrations and whole-brain phosphate metabolism during insulin-induced hypoglycemia, 31P NMR spectra were obtained before and after intravenous injection of insulin (0.15 IU/kg body wt) in six men. Compared with prehypoglycemic measurements, no significant changes were found in brain content of Pi, sugar phosphates, phosphocreatine, phosphodiesters, and ATP, and brain pH remained constant during the experiment. These results show that the integrated brain profile of energy-rich phosphorous compounds is unaffected by experimental insulin-induced hypoglycemia in humans.

The effect of metabolic control on hemodynamics in short-term insulin-dependent diabetic patients.

Hemodynamics variables (heart rate, arterial blood pressure, cardiac output, hepato-splanchnic blood flow, forearm blood flow, and plasma catecholamines) were measured during good (median blood glucose 4.7 mmol/L) and poor (median blood glucose 16.3 mmol/L) metabolic control in eight young, short-term, insulin-dependent diabetic patients. The measurements were performed twice within 2 wk, in random order. Continuous subcutaneous insulin infusion (CSII) was applied for 1 wk in order to obtain good control. All eight patients had elevated cardiac output (median 9%) and forearm blood flow (median 34%) during poor compared with good metabolic control, P less than 0.01. In contrast, hepato-splanchnic blood flow was lower (median 12%) during poor compared with good metabolic control, P less than 0.05. Heart rate remained unchanged, while mean arterial blood pressure was slightly higher during poor control, P less than 0.05. Five of six patients had elevated plasma noradrenaline concentration during poor metabolic control. Due to the small number of patients investigated, no valid conclusion regarding the activity of the sympathoadrenal system can be drawn. Our study suggests that both increased cardiac output and reduced hepato-splanchnic blood flow (redistribution) contribute to the elevated blood flow previously demonstrated in various other organs and tissues in diabetic patients during poor metabolic control.

Hypoglycemia and hemostatic parameters in juvenile-onset diabetes.

Hypoglycemia was induced by intravenous infusion of insulin in six juvenile-onset diabetic subjects. Hemostatic parameters were assessed before insulin infusion and 0, 1, and 2 h after discontinuation of insulin infusion. The onset of hypoglycemia coincided with an enhancement of ADP-induced platelet aggregation in five of the patients; platelet aggregation returned to normal levels during the next 2 h. The recalcification time was significantly shortened at the onset of hypoglycemia and continued to decrease for the next 2 h despite the return of serum glucose to normal levels. Correspondingly, a significant increase in fibrinogen was found. The ethanol gelation test was positive in two patients 2 h after stopping insulin infusion. Platelet counts decreased significantly after stopping insulin infusion. Thus, insulin-induced hypoglycemia markedly affects hemostatic balance and may potentially lead to intravascular coagulation in juvenile-onset diabetic patients.

Serum growth hormone-binding protein in obesity: effect of a short-term, very low calorie diet and diet-induced weight loss.

GH-binding protein (GHBP) is increased in obesity. It is not known whether the increase in GHBP is reversible with weight loss or modulated by acute changes in nutritional intake. To address these questions, we measured GHBP in 18 obese subjects [body mass index (BMI), 40.9 +/- 1.1 kg/m2 (mean +/-SEM)] before and after an average weight loss of 30.3 +/- 4.6 kg and in 18 age- and sex matched normal subjects (BMI, 23.0 +/- 0.4 kg/m2) and studied the effects of a very low calorie diet over 4 days in 5 normal subjects and a subgroup of obese subjects before (n = 6) and after (n = 5) weight loss. GHBP was elevated in the obese subjects compared to levels in age- and sex-matched normal controls (1.48 +/- 0.1 vs. 0.53 +/- 0.1 nmol/L; P < 0.0001). GHBP was positively correlated to BMI and waist circumference (r = 0.71; P < 0.00001 and r = 0.73; P < 0.00001, respectively). In addition, GHBP was positively correlated to insulin as well as proinsulin levels (r = 0.60; P < 0.001 and r = 0.55; P < 0.001, respectively). After diet-induced massive weight loss, GHBP levels were restored to normal in obese subjects (BMI, 27.8 +/- 1.4 kg/m2). Multiple stepwise regression analysis revealed that changes in waist circumference and abdominal sagittal diameter during weight loss were the major determinants of and accounted for 54% of the fall in GHBP levels. Neither insulin nor proinsulin was an independent predictor. No changes were observed in GHBP in normal, obese, or reduced weight obese subjects after 4 days of a very low calorie diet, although mean insulin levels fell significantly in the normal subgroup as well as in the obese subgroup studied after weight loss. In summary, GHBP levels are elevated in obesity, are restored to normal by massive weight loss, and are unaffected by short term hypocaloric feeding. We conclude that GHBP may be regulated by the same or closely related factors that regulate fat mass and abdominal fat mass in particular, but not by insulin or acute changes in nutrition.

Lack of stimulation of 24-hour growth hormone release by hypocaloric diet in obesity.

Obesity is associated with a marked reduction in the spontaneous secretion of GH. To investigate the effect of acute alterations in calorie intake on GH release, 24-hr spontaneous GH release was measured during habitual calorie intake as well as during a short term, very low calorie diet (VLCD) in 6 obese subjects, 5 obese subjects after weight loss, and 5 normal, age- and sex-matched control subjects. Integrated 20-min samples were obtained over 24-h on two occasions in each subject using a constant blood withdrawal technique. In addition, basal levels of serum insulin-like growth factor-I (IGF-I), IGF-binding protein-1 (IGFBP-1), IGF-binding protein-3 (IGFBP-3), insulin, pro-insulin, and blood glucose were measured during habitual energy intake as well as during the hypocaloric diet. Twenty-four-hour GH release profiles and IGFBP-1 were decreased, and insulin as well as proinsulin levels were elevated in obese subjects compared to those in normal age- and sex-matched controls. No differences between obese subjects and normal controls were present regarding IGF-I, IGFBP-3, or IGF-I/IGFBP-3 molar ratio. In the last 24 h during the 96-h VLCD, an increase in 24-h GH release and basal IGFBP-1 levels and a decrease in basal insulin levels occurred in the normal controls, whereas no such changes were observed in the obese subjects. After caloric restriction 24-hr GH release, IGFBP-1 levels and insulin levels were similar in control subjects and obese subjects after weight loss. This suggests a reversible defect in GH release, rather than a persistent preexisting disorder. It is hypothesized that enhanced bioavailability of IGF-I, acting in concert with elevated proinsulin and insulin levels, may account for the lack of stimulation of 24-hr GH release by the hypocaloric diet in obese subjects. We conclude that the increase in 24-h spontaneous GH release and IGFBP-1 levels observed in normal subjects during the last 24 h of a 96-h VLCD is abolished in obese subjects. The lack of short term hypocaloric stimulation of spontaneous GH release may promote the retention of body fat and perpetuate the obese state.

No response of pancreatic hormones to hypoglycemia in diabetic autonomic neuropathy.

The responses of pancreatic hormones (i.e. glucagon, pancreatic polypeptide, and somatostatin) to insulin-induced hypoglycemia were investigated in 18 insulin-dependent diabetics without residual beta-cell function and in 6 normal subjects. Nine of the diabetics had autonomic neuropathy, and 9 had no neuropathy. After hypoglycemia, no significant increase in any of the 3 pancreatic hormones was found in the diabetics with autonomic neuropathy, whereas significant increments were found in the diabetics without neuropathy and in the normal subjects. These results suggest that autonomic nervous activity is of major importance for pancreatic hormone release during hypoglycemia in man.

Massive weight loss restores 24-hour growth hormone release profiles and serum insulin-like growth factor-I levels in obese subjects.

In the present study, we 1) determined whether the impaired spontaneous 24-h GH secretion as well as the blunted GH response to provocative testing in obese subjects are persistent disorders or transient defects reversed with weight loss and 2) investigated 24-h urinary GH excretion and basal levels of insulin-like growth factor-I (IGF-I), IGF-binding protein-3 (IGFBP-3), as well as insulin in obese subjects before and after a massive weight loss. We studied 18 obese subjects (age, 26 +/- 1 yr; body mass index, 40.9 +/- 1.1 kg/m2); 18 normal age-, and sex-matched control subjects; and 9 reduced weight obese subjects after a diet-induced average weight loss of 30.3 +/- 4.6 kg. Twenty-four-hour spontaneous GH secretion was estimated by obtaining 3240 integrated 20-min blood samples using a constant blood withdrawal technique and computerized algorithms. Body composition was determined using anthropometric measurements and dual energy x-ray absorptiometry scanning (DXA). In the obese subjects, 24-h spontaneous GH release profiles and the GH responses to insulin-induced hypoglycemia and L-arginine as well as basal IGF-I levels and the IGF-I/IGFBP-3 molar ratio were decreased, whereas insulin levels were elevated compared to those in normal subjects. In obese subjects, 24-h spontaneous GH secretion and serum IGF-I levels were inversely related to abdominal fat (r = -0.67; P < 0.01) and percent body fat (r = -0.69; P < 0.01), respectively. The decreased 24-h spontaneous GH release profiles, the decreased GH responses to insulin-induced hypoglycemia and L-arginine, the decreased basal IGF-I levels and IGF-I/IGFBP-3 molar ratio, as well as the elevated insulin levels were returned to normal after a massive weight loss in the obese subjects. In conclusion, the present study has shown reversible defects in 24-h spontaneous GH release profiles, basal IGF-I levels, and the IGF-I/IGFBP-3 molar ratio in obese subjects. The recovery of the 24-h GH release points to an acquired transient defect rather than a persistent preexisting disorder.

Adult growth hormone (GH)-deficient patients demonstrate heterogeneity between childhood onset and adult onset before and during human GH treatment. Adult Growth Hormone Deficiency Study Group.

The onset of adult GH deficiency may be during either adulthood (AO) or childhood (CO), but potential differences have not previously been examined. In this study the baseline and GH therapy (12.5 micrograms/kg per day) data from CO (n = 74; mean age 29 yr) and AO (n = 99; mean age 44 yr) GH-deficient adult patients have been compared. The first 6 months comprised randomized, double-blind treatment with GH or placebo, then all patients were GH-treated for a further 12 months. At baseline the height, body weight, body mass index, lean body mass, and waist/hip ratio of AO patients were significantly (P < 0.001) greater than in CO patients. Serum insulin-like growth factor-I (IGF-I) levels were below normal but were lower in CO than AO patients (P < 0.001), and the correlation with IGF binding protein-3 was stronger in CO than in AO patients. Osteocalcin concentration in CO patients was above the normal range and significantly greater than in AO patients. Both groups had significant psychosocial distress, but the deviation from normality was greater in AO patients. Throughout GH therapy there was a significant increase in lean body mass and significant decrease in percent body fat and sum of skinfolds in each group. Wais/hip ratio was decreased by long-term therapy in AO but not CO patients. Total and low density lipoprotein cholesterol levels were decreased from baseline at 6 months in AO but not CO patients and high density lipoprotein cholesterol was increased in both groups throughout therapy. IGF-I and IGF binding protein-3 were increased into the normal range by GH therapy in both groups. Mean osteocalcin level in AO patients was increased at 6 months with no further change with GH therapy, whereas in CO patients there was a steep increase up to 12 months but then a sharp decrease. Nottingham Health Profile scores showed significant improvements in physical mobility and energy at 18 months of therapy in AO patients but no consistent effects in CO patients. GH-induced side effects were mainly reported by AO patients; very few CO patients reported treatment-emergent adverse events. These results demonstrate significant differences in clinical and biochemical presentation and responses to therapy of the adult GH deficiency syndrome. This is consistent with the existence of two entities, developmental (CO) and metabolic (AO), and the different functions of GH at different periods of life.

Gene expression of a truncated and the full-length growth hormone (GH) receptor in subcutaneous fat and skeletal muscle in GH-deficient adults: impact of GH treatment.

In humans at least two GH receptors are significantly expressed. One is the full-length receptor (GHR); the other is a truncated form (GHRtr), that lacks most of the intracellular domain. This receptor may inhibit the action of the full-length receptor. Circulating GH-binding protein (GHBP) is a proteolytically cleaved product from both of these receptors. The clinical relevance of the different receptor types is unknown. We examined the gene expression of GHR and GHRtr in human adipose tissue and skeletal muscle and the influence of GH treatment on this expression. Furthermore, we studied the relationship of circulating GHBP and body composition to GHR and GHRtr gene expression. Eleven adult GH-deficient patients were studied before and after 4 months of GH substitution therapy. Abdominal fat obtained by liposuction and femoral muscle biopsies were taken at baseline and after 4 months. Gene expression of GHR and GHRtr in adipose tissue and skeletal muscle was determined and expressed relative to the expression of beta-actin. Gene expression of GHR in abdominal sc adipose tissue was not altered, whereas the expression of GHRtr increased significantly. In skeletal muscle inverse changes were seen in the expression of messenger ribonucleic acid (mRNA) levels for the two GH receptor forms: expression of GHR increased significantly, whereas mRNA levels for GHRtr decreased. As expected, body composition changed with reduction of body fat mass after 4 months of GH treatment. Levels of circulating GHBP decreased significantly. We conclude that GH treatment in GH-deficient adults changes the expression of mRNA for GHR and GHRtr in adipose tissue and skeletal muscle. Whether these changes are responsible for the observed changes in body composition in response to GH treatment and the observed changes in levels of circulating GHBP, however, needs further elucidation.

The effect of 30 months of low-dose replacement therapy with recombinant human growth hormone (rhGH) on insulin and C-peptide kinetics, insulin secretion, insulin sensitivity, glucose effectiveness, and body composition in GH-deficient adults.

The aim of the present study was to evaluate the long-term (30 months) metabolic effects of recombinant human GH (rhGH) given in a mean dose of 6.7 microg/kg x day (= 1.6 IU/day), in 11 patients with adult GH deficiency. Glucose metabolism was evaluated by an oral glucose tolerance test and an iv (frequently sampled iv glucose tolerance test) glucose tolerance test, and body composition was estimated by dual-energy x-ray absorptiometry. Treatment with rhGH induced persistent favorable changes in body composition, with a 10% increase in lean body mass (P < 0.001) and a 12% reduction of fat mass (P < 0.002); however, the glucose tolerance deteriorated significantly, and three patients developed impaired glucose tolerance. Fasting insulin level (P < 0.003) and the homeostasis model assessment insulin resistance score increased significantly, indicating a deterioration in insulin sensitivity; whereas the insulin sensitivity index, calculated from the frequently sampled iv glucose tolerance test, only decreased slightly. The clearance of C-peptide and insulin increased 100% and 60%, respectively, and the prehepatic insulin secretion was tripled during rhGH treatment; but related to the impairment in glucose tolerance, beta-cell response was still inappropriate. Our conclusion is that long-term rhGH-replacement therapy in GH deficiency adults induced a significant deterioration in glucose tolerance, profound changes in kinetics of C-peptide, and insulin and prehepatic insulin secretion, despite an increase in lean body mass and a reduction of fat mass. Therefore, rhGH treatment may precipitate diabetes in some patients already susceptible to the disorder.

Monitoring of celiac plexus block in chronic pancreatitis.

Pharmacological, percutaneous celiac plexus blockade is often inefficient in the treatment of pain in chronic pancreatitis. Lack of efficiency could be due to incomplete denervation of the plexus; however, a method for measuring the completeness of celiac plexus blockade is not yet available. We have, therefore, monitored the physiological completeness of pharmacological percutaneous celiac blockade with 40 ml 25% ethanol by measuring the effect of posture on heart rate, blood pressure, hepato-splanchnic vascular resistance, and pancreatic hormone concentrations before and after celiac plexus block in 6 patients with chronic pancreatitis. Blood pressure decreased and heart rate increased after the block (P less than 0.025), whereas no significant change was found in hepato-splanchnic vascular resistance nor in the change of these parameters during transition from the supine to standing position. Pancreatic hormones (C-peptide, free insulin, glucagon, pancreatic polypeptide and somatostatin) did not change in response to standing, either before or after the block. The cardiovascular variables were normalized the day after the block, and all the patients were in their habitual state regarding pain after 1 week. In conclusion, pancreatic hormone concentrations in response to standing are not useful for monitoring celiac plexus block, whereas heart rate, blood pressure and hepato-splanchnic blood flow may yield useful information. From such measurements it was concluded that permanent denervation of the celiac plexus was not achieved in our patients after injection of 40 ml 25% ethanol.

Changes in platelet function, blood coagulation and fibrinolysis during insulin-induced hypoglycaemia in juvenile diabetics and normal subjects.

Haemostatic parameters were assessed before insulin induced hypoglycaemia and 0, 1 and 2 hr after discontinuation of insulin infusion in 7 non-diabetics, aged 28 (22-31) years (mean and range), and 8 juvenile diabetics, aged 31 (27-35) years, with a mean duration of diabetes of 4 years. The patients were normoglycaemic for at least 10 hr before the study. Platelet aggregation in vitro was induced by lower adenosine diphosphate (ADP) concentrations in the diabetics than in the controls before hypoglycaemia and 0 and 60 min after insulin infusion. Platelet counts decreased significantly in the diabetics after hypoglycaemia, whereas no changes were seen in the control group. The activated partial thromboplastin time (APTT) was reduced in both groups and significantly lower in the diabetics than in the controls 120 min after insulin infusion. Fibrinogen and factor VIII R:Ag increased after insulin infusion; highest values were seen in the diabetics. The euglobulin clot lysis time (ELT) was reduced in both groups during insulin infusion; 120 min after end of insulin infusion ELT was significantly longer in the diabetics than in the control group.

Long-term effects of growth hormone (GH) on body fluid distribution in GH deficient adults: a four months double blind placebo controlled trial.

OBJECTIVE: Short-term growth hormone (GH) treatment normalises body fluid distribution in adult GH deficient patients, but the impact of long-term treatment on body fluid homeostasis has hitherto not been thoroughly examined in placebo controlled trials. To investigate if the water retaining effect of GH persists for a longer time we examined the impact of 4 months GH treatment on extracellular volume (ECV) and plasma volume (PV) in GH deficient adults. DESIGN: Twenty-four (18 male, 6 female) adult GH deficient patients aged 25-64 years were included and received either GH (n=11) or placebo (n=13) in a double blind parallel design. METHODS: Before and at the end of each 4 month period ECV and PV were assessed directly using 82Br- and 125I-albumin respectively, and blood samples were obtained. RESULTS: During GH treatment ECV increased significantly (before: 20.48+/-0.99 l, 4 months: 23.77+/-1.38 l (P<0.01)), but remained unchanged during placebo administration (before: 16.92+/-1.01 l, 4 months: 17.60+/-1.24 l (P=0.37)). The difference between the groups was significant (P<0.05). GH treatment also increased PV (before: 3.39+/-0.27 l. 4 months: 3.71+/-0.261 (P=0.01)), although an insignificant increase in the placebo treated patients (before: 2.81+/-0.18 l, 4 months: 2.89+/-0.20 l (P=0.37)) resulted in an insignificant treatment effect (P=0.07). Serum insulin-like growth factor-I increased significantly during GH treatment and was not affected by placebo treatment. Plasma renin (mIU/l) increased during GH administration (before: 14.73+/-2.16, 4 months: 26.00+/-6.22 (P=0.03)) and remained unchanged following placebo (before: 20.77+/-5.13, 4 months: 20.69+/-6.67 (P=0.99)) leaving no significant treatment effect (P=0.08). CONCLUSION: The long-term impact of GH treatment on body fluid distribution in adult GH deficient patients involves expansion of ECV and probably also PV. These data substantiate the role of GH as a regulator of fluid homeostasis in adult GH deficiency.

Metabolic control and B cell function in patients with insulin-dependent diabetes mellitus secondary to chronic pancreatitis.

Among 88 unselected patients with chronic pancreatitis 35% (95% confidence limits 25 to 46) had insulin-dependent diabetes, 31% (21% to 41%) had non-insulin-dependent diabetes or impaired glucose tolerance (by intravenous glucose tolerance test), and 34% (24% to 45%) had normal glucose tolerance. B cell function measured by C-peptide concentration after 1 mg glucagon IV correlated with the pancreatic enzyme secretion (meal stimulated duodenal lipase content). B cell function was preserved to a greater extent (P less than .01), and glycosylated hemoglobin and fasting level of glucose were lower (P less than .01 to .05) in the 31 patients with pancreatogenic diabetes than than in 35 otherwise comparable patients with type I (insulin-dependent) diabetes, yet daily insulin dose was similar in the two groups. Glucagon stimulated C-peptide was inversely correlated to glycosylated hemoglobin in insulin-dependent patients with pancreatogenic diabetes and in type I diabetes. Since body mass indices were identical in the two groups, better glucoregulation was not due to reduced food intake or malabsorption in pancreatogenic diabetes. Rather residual B cell function and/or different secretion of other pancreatic hormones in pancreatogenic diabetes may account for different metabolic control in type I IDDM compared with insulin-dependent pancreatogenic diabetes.

Norepinephrine kinetics during insulin-induced hypoglycemia.

Norepinephrine (NE) kinetics (plasma appearance rate, clearance, and forearm extraction) were measured during insulin-induced hypoglycemia in six healthy subjects. NE clearance did not change during hypoglycemia, indicating that the increase in plasma NE during hypoglycemia is due to an increased plasma appearance rate of NE. Forearm extraction of 3H-NE and of endogenous plasma epinephrine decreased significantly during hypoglycemia, probably due to an increase in regional blood flow.

Theophylline enhances glucose recovery after hypoglycemia in healthy man and in type I diabetic patients.

The principal mediators of glucose counterregulation (glucagon and epinephrine) use intracellular cyclic adenosine monophosphate (cAMP) to mediate glucose release. Since theophylline increases cAMP (by inhibiting its decomposition), we investigated the effect of theophylline on glucose recovery after insulin-induced hypoglycemia. Eleven healthy subjects and nine type I (insulin-dependent) diabetic patients each participated in two experiments in randomized order, receiving on both days an insulin bolus of 0.15 IU/kg soluble insulin. On one day, theophylline (intravenous [IV] bolus of 220 mg followed by IV infusion of 1 mg/kg/h) was administered from 1 hour before induction of hypoglycemia until the end of the study period. On the other day, NaCl was administered. Plasma glucose before induction of hypoglycemia was equal on the 2 study days. The plasma glucose area under the curve (AUC) was larger with theophylline than with NaCl (P = .04 for diabetic patients and P = .003 for healthy subjects). During the most active phase of glucose counterregulation, the rate of increase of plasma glucose was larger with theophylline (P = .003 for diabetic patients and P = .03 for healthy subjects). The incremental AUC for cAMP was larger with theophylline for diabetic patients (P = .01). For healthy subjects, cAMP was greater with theophylline 30 minutes after insulin (P = .03). In conclusion, glucose recovery after hypoglycemia is significantly increased when theophylline is administered in an asthma dosage before hypoglycemia is induced. This may be due to a significant enhancement of the cAMP response.