The metabolic response to glycemic control by the artificial pancreas in diabetic man

Diurnal pattern of insulin action in type 1 diabetes: implications for a closed-loop system

We recently demonstrated a diurnal pattern to insulin action (i.e., insulin sensitivity [SI]) in healthy individuals with higher SI at breakfast than at dinner. To determine whether such a pattern exists in type 1 diabetes, we studied 19 subjects with C-peptide-negative diabetes (HbA1c 7.1 ± 0.6%) on insulin pump therapy with normal gastric emptying. Identical mixed meals were ingested during breakfast, lunch, and dinner at 0700, 1300, and 1900 h in randomized Latin square of order on 3 consecutive days when measured daily physical activity was equal. The triple tracer technique enabled measurement of glucose fluxes. Insulin was administered according to the customary insulin:carbohydrate ratio for each participant. Although postprandial glucose excursions did not differ among meals, insulin concentration was higher (P < 0.01) and endogenous glucose production less suppressed (P < 0.049) at breakfast than at lunch. There were no differences in meal glucose appearance or in glucose disappearance between meals. Although there was no statistical difference (P = 0.34) in SI between meals in type 1 diabetic subjects, the diurnal pattern of SI taken across the three meals in its entirety differed (P = 0.016) from that of healthy subjects. Although the pattern in healthy subjects showed decreasing SI between breakfast and lunch, the reverse SI pattern was observed in type 1 diabetic subjects. The results suggest that in contrast to healthy subjects, SI diurnal pattern in type 1 diabetes is specific to the individual and cannot be extrapolated to the type 1 diabetic population as a whole, implying that artificial pancreas algorithms may need to be personalized.

Forty-eight-hour first-trimester glucose profiles in women with type 1 diabetes mellitus: a report of three cases of congenital malformation

OBJECTIVE

Despite modern methods of treatment and near-normal HbA(1c) levels, women with type 1 diabetes mellitus are still at risk of having an infant with a congenital malformation (CM). We hypothesised that HbA(1c) levels are too gross a measure of glycaemic control and used a continuous glucose monitoring system (CGMS) to determine the diurnal glucose profiles during the first trimester of pregnancy. We present three cases of infants with a CM.

METHODS

Fifty-three women with type 1 diabetes used the CGMS for 48 h in the first trimester of pregnancy. Three of them gave birth to infants with a CM. HbA(1c) levels were determined at the time of the CGMS measurement and 6 to 8 weeks later.

RESULTS

The HbA(1c) levels at the time of the CGMS measurement were 6.0, 6.5 and 7.8% (normal range 4.0-6.0%) in the three women. The 48-h diurnal glucose profiles of these women showed a large variability with frequent hyperglycaemic episodes.

CONCLUSIONS

HbA(1c) levels are too gross a measure of glycaemic control to identify women at risk of giving birth to an infant with a CM. Even in women with normal or near-normal HbA(1c) levels, the diurnal glucose profiles reveal intermittent hyperglycaemic episodes that may cause the CM.

Pathogenesis of diabetes-induced congenital malformations

The increased rate of fetal malformation in diabetic pregnancy represents both a clinical problem and a research challenge. In recent years, experimental and clinical studies have given insight into the teratological mechanisms and generated suggestions for improved future treatment regimens. The teratological role of disturbances in the metabolism of inositol, prostaglandins, and reactive oxygen species has been particularly highlighted, and the beneficial effect of dietary addition of inositol, arachidonic acid and antioxidants has been elucidated in experimental work. Changes in gene expression and induction of apoptosis in embryos exposed to a diabetic environment have been investigated and assigned roles in the teratogenic processes. The diabetic environment appears to simultaneously induce alterations in several interrelated teratological pathways. The complex pathogenesis of diabetic embryopathy has started to unravel, and future research efforts will utilize both clinical intervention studies and experimental work that aim to characterize the human applicability and the cell biological components of the discovered teratological mechanisms.

Insulin resistance and action in hypertriglyceridemia

To define further the characteristics of insulin insensitivity associated with hypertriglyceridemia, the metabolic responses to the euglycemic insulin clamp were evaluated in 6 hypertriglyceridemic male patients and compared to 5 normal male controls. At baseline, the hypertriglyceridemic patients had elevated triglycerides (687 +/- 172 vs 78 +/- 7 mg/dl, P less than 0.005) and free fatty acids (702 +/- 36 vs 444 +/- 42 microM/l, P less than 0.005) concentrations. During the euglycemic insulin clamp, steady-state plasma glucose concentrations were similar in both groups (90.2 +/- 1.5 vs 88.8 +/- 2.3 mg/dl, ns) as were steady state insulin levels (142 +/- 10.1 vs 132.2 +/- 6.8 microU/ml**). The amount of glucose metabolized during the last hour of the clamp (M) was significantly reduced in the hypertriglyceridemic patient (2.9 +/- 0.4 vs 6.2 +/- 0.7 mg.min-1.kg-1, P less than 0.001). Changes in free fatty acid, glycerol, B-hydroxybutyrate, lactate and pyruvate during the euglycemic insulin clamp were similar indicating a preserved antilipolytic, antiketogenic and glycolytic intermediate (lactate + pyruvate) response to insulin and glucose infusion in the hypertriglyceridemic patients. In summary, hypertriglyceridemia is associated with insulin resistance as it relates to muscle glucose utilization. However, this is not universal, as a number of other insulin responsive pathways appear to be unaffected.

Studies of gut and hepatic metabolism in conscious rabbits

The present study was designed to develop the techniques for chronic catheterization of the hepatic and portal venous circulation in conscious rabbits and to apply these techniques to a study of hepatic metabolism in this species. Experiments were made after an 18-h fast and for 4 h after the initial feeding. Measurements of arteriovenous differences of substrates were combined with measurements of hepatic and gastrointestinal blood flow. Hepatic glucose production was suppressed by 60% at 1 h and had returned to control levels by 4 h. The hepatic uptake of lactate declined slightly at 1 h and had returned to control level 2 h after the meal. There was a marked and rapid fall in hepatic ketone body output after refeeding. Although amino acid concentrations displayed a transient increase 1 h after the meal, only the arterial concentration of branched-chain amino acids remained significantly elevated for 4 h. The total hepatic uptake of the gluconeogenic amino acids (alanine, serine, threonine) remained constant. Refeeding resulted in a doubling of arterial insulin concentrations at 1 h followed by a progressive decline over the next 3 h. It is concluded that in rabbits fed a mixed meal partial suppression of hepatic glucose output is mainly due to a decline in glycogenolysis rather than a decrease in gluconeogenesis, shortly after refeeding the liver is able to virtually shut off its ketone body production, the major gluconeogenic precursors (lactate, alanine, glycine, serine, and threonine) may contribute to approximately 40% of the glucose release.(ABSTRACT TRUNCATED AT 250 WORDS)

Alterations in alanine metabolism in diabetic dogs during short-term treatment with an artificial B cell

The flux rates of plasma glucose and alanine were studied isotopically (6-3H-glucose and U-14C-alanine simultaneously) in resting chronically diabetic dogs during short-term treatment with an artificial B cell where the insulin was infused into a peripheral vein. Despite perfect blood glucose control and normal glucose flux rates, the concentration and rates of appearance and disappearance of alanine were significantly elevated in the diabetic animals before, during and after an exogenous glucose load. The incorporation of the carbon moiety of alanine into circulating glucose was also increased, but diminished to a near-normal extent when exogenous glucose was given. The plasma clearance rates for alanine in the diabetic dogs were normal throughout the study. It is concluded that normal blood glucose control in diabetes does not necessarily mean normalization of the entire metabolic network. On the basis of peripheral hyperinsulinaemia alanine formation from glucose and branched chain amino acids is elevated in muscle. This may explain increased flux of alanine despite normal blood glucose control.

The metabolic response to the euglycemic insulin clamp in type I diabetes and normal humans

The euglycemic insulin clamp has been utilized extensively to measure in vivo tissue sensitivity to insulin under various circumstances. Insulin sensitivity is determined from the amount of glucose metabolized under steady state conditions. To assess the effect of abnormalities in other insulin responsive metabolic pathways on glucose metabolism and thus insulin sensitivity as measured by the glucose clamp, the concentration of lactate, pyruvate, 3-hydroxybutyrate, glycerol, alanine, and free fatty acids were measured at baseline and during a two-hour euglycemic clamp in 13 nonobese subjects with type I diabetes. The observed responses were compared to 11 normal controls. Insulin sensitivity as measured by M (glucose metabolized), MCRg (metabolic clearance of glucose), and M/I ratio (glucose metabolized per unit insulin) were all significantly decreased in the diabetic subjects (P less than 0.005). Free fatty acids (FFA) and 3-hydroxybutyrate were significantly elevated at baseline in the diabetic subjects (P less than 0.05) and decreased significantly at 60 and 120 minutes in both groups. Baseline blood pyruvate and lactate concentrations were similar in the control and diabetic subjects. Pyruvate increased significantly at 60 minutes in both groups (P less than 0.05) and returned to baseline in the control subjects but remained elevated at 120 minutes in the diabetic subjects (P less than 0.001). Lactate increased similarly in both groups and remained elevated at 60 and 120 minutes. In summary, insulin sensitivity as assessed by the euglycemic insulin clamp is decreased in type I diabetes. However, specific differences in the concentration of several other metabolites both at baseline and in response to hyperinsulinemia were also identified in the diabetic subjects.(ABSTRACT TRUNCATED AT 250 WORDS)

Review of glyburide after one year on the market

After one year on the American market, glyburide, a second-generation sulfonylurea, seems well accepted by physicians. If the hyperglycemia of patients with type II diabetes mellitus is not corrected by diet and exercise, glyburide can be used as adjunctive therapy. The drug is comparable in efficacy to the first-generation sulfonylurea chlorpropamide, but it has fewer reported side effects. Additionally, glyburide does not appear to interact with other medications and is well absorbed from the gastrointestinal tract. Although the drug had been reported to produce hypoglycemia, increased clinical experience--along with use of judicious dosages in appropriate patients--has decreased the incidence of hypoglycemia. Although still in the investigational stage, combination insulin/glyburide therapy may also benefit subgroups of type II diabetic patients. Glyburide is a useful and rational addition to therapy in properly selected patients with type II diabetes mellitus.

Characterization of insulin resistance in type I diabetes

Insulin sensitivity was assessed using the euglycaemic clamp technique in eight type I diabetic patients (after overnight blood glucose normalization with an artificial pancreas) and in six healthy subjects. Basal insulin concentrations were higher in diabetic patients (25 +/- 4 microU/ml) than in control subjects (17 +/- 1 microU/ml; P less than 0.05). Insulin infusion of 0.5, 1.0, 2.0 and 5.0 mU/kg per min during subsequent 2-h periods resulted in similar mean steady-state insulin concentrations in both groups. The mean dextrose requirements during the last 40 min of each period were nevertheless decreased in diabetic patients (1.6 +/- 0.5, 3.5 +/- 0.8, 6.5 +/- 0.7, 10.2 +/- 0.7 mg/kg per min) as compared with control subjects (4.7 +/- 0.3, 8.2 +/- 0.9, 10.2 +/- 0.9, 12.4 +/- 0.9 mg/kg per min). At low insulin concentrations dextrose requirements were diminished in all diabetic subjects. At the highest insulin levels, individual dose-response curves from only four patients were within the normal range. Under basal conditions, the monocyte receptor number was significantly reduced in diabetic patients (17,500 +/- 2,800 sites/cell) as compared with control subjects (26,700 +/- 2,500 sites/cell; P less than 0.05), whereas there were no differences regarding empty site affinities. Receptor data did not differ in patients with normal and decreased maximal dextrose requirements. Insulin resistance is apparently a common feature of type I diabetes at serum insulin concentrations of approximately 100 microU/ml. Normalization of the insulin effect by higher insulin concentrations is not possible in all patients. Insulin antibodies at concentrations observed in this study (less than 0.16 mU/ml) do not contribute significantly to insulin resistance; receptor and postreceptor defects are possibly more important.

Enhanced glycemic responsiveness to epinephrine in insulin-dependent diabetes mellitus is the result of the inability to secrete insulin. Augmented insulin secretion normally limits the glycemic, but not the lipolytic or ketogenic, response to epinephrine in humans

To determine if the enhanced glycemic response to epinephrine in patients with insulin-dependent diabetes mellitus (IDDM) is the result of increased adrenergic sensitivity per se, increased glucagon secretion, decreased insulin secretion, or a combination of these, plasma epinephrine concentration-response curves were determined in insulin-infused (initially euglycemic) patients with IDDM and nondiabetic subjects on two occasions: once when insulin and glucagon were free to change (control study), and again when insulin and glucagon were held constant (islet clamp study). During the control study, plasma C-peptide doubled, and glucagon did not change in the nondiabetic subjects, whereas plasma C-peptide did not change but glucagon increased in the patients. The patients with IDDM exhibited threefold greater increments in plasma glucose, largely the result of greater increments in glucose production. This enhanced glycemic response was apparent with 30-min increments in epinephrine to plasma concentrations as low as 100-200 pg/ml, levels that occur commonly under physiologic conditions. During the islet clamp study (somatostatin infusion with insulin and glucagon replacement at fixed rates), the heightened glycemic response was unaltered in the patients with IDDM, but the nondiabetic subjects exhibited an enhanced glycemic response to epinephrine indistinguishable from that of patients with IDDM. In contrast, the FFA, glycerol, and beta-hydroxybutyrate responses were unaltered. Thus, we conclude the following: Short, physiologic increments in plasma epinephrine cause greater increments in plasma glucose in patients with IDDM than in nondiabetic subjects, a finding likely to be relevant to glycemic control during the daily lives of such patients as well as during the stress of intercurrent illness. Enhanced glycemic responsiveness of patients with IDDM to epinephrine is not the result of increased sensitivity of adrenergic receptor-effector mechanisms per se nor of their increased glucagon secretory response; rather, it is the result of their inability to augment insulin secretion. Augmented insulin secretion, albeit restrained, normally limits the glycemic response, but not the lipolytic or ketogenic responses, to epinephrine in humans.

How laboratory dogs accommodate meals of different size but similar composition

Healthy laboratory dogs appear to absorb a mixed meal from the gut at a constant rate. This rate is apparently not affected by meal size. If this is the case, then duration of absorption should depend on total or integrated meal size, whereas metabolite and hormonal levels would be independent of the number of feedings. To explore these hypotheses further, we compared the metabolite and hormonal responses with a single mixed meal and one divided in two halves, provided in two feedings 4 h apart. We detected no effect of the second meal in the metabolic response levels of glucose, lactate, pyruvate, alanine, free fatty acids, or 3-hydroxybutyrate or the hormonal responses of insulin, pancretic glucagon, gastrin, or secretin. Only minor differences were detectable in the hormonal response levels of pancreatic polypeptide, gastric inhibitory peptide, and enteroglucagon, consistent with a response to a second meal. We conclude that the observed change in circulating metabolite or hormone concentration is independent of the size of meal eaten, but the duration of the excursion depends on meal size. Thus, during the bulk phase of nutrient uptake, the absorption mechanism of the laboratory dog appears to be saturated.

Minimizing the aggregation of neutral insulin solutions

Various solution additives affect the solubility and macroaggregation of insulin in buffered aqueous solutions at physiological pH. The solubility of insulin may be improved with the addition of small amounts of aspartic acid, glutamic acid, EDTA (ethylenediaminetetraacetic acid), lysine, Tris buffer, or bicarbonate buffer. In addition, the propensity of dissolved insulin to reaggregate and precipitate may be inhibited by such additives. Buffered physiological (pH 7.4) saline solutions containing 0.001-0.003 M lysine in the presence of 0.005 M EDTA or 0.01 M lysine in the absence of EDTA improve insulin solubility and are effective in minimizing aggregation. Solutions thus prepared may be suitable for application in intravenous insulin infusion devices and may be useful commercial insulin preparations.

Glucose metabolism studied isotopically in diabetic dogs: effect of restoration of peripheral normoinsulinaemia by the artificial B cell

Normoglycaemia, peripheral normoinsulinaemia, and normoglucagonaemia were restored acutely in chronically diabetic dogs, using an extracorporal artificial B cell with peripheral venous insulin administration. Glucose metabolism was analysed by a non-steady-state tracer technique with double-labelled glucose (6-3H- and U-14C-glucose), and the incorporation of the 14C label into plasma lactate was determined. In the basal state, glucose turnover rates were not different from those in non-diabetic controls; but recirculation of the glucose-C label through the Cori cycle, and lactate labelling from glucose utilization were decreased. The glycaemic response to an intravenous infusion of non-labelled glucose was distinctly enhanced. This was based on a reduction in the rates of glucose disappearance. Its rates of appearance (total endogenous glucose production) were, however, suppressed to a normal extent by the exogenous glucose. Accordingly carbon recycling was nearly totally suppressed during the glucose infusion as in the controls. It is concluded that metabolic recompensation in these fasting, resting diabetic dogs remained incomplete because the interval of normoinsulinaemia, which obviously applied only to the peripheral circulation, was not long enough.

Modulation of substrate transport to the brain

Variations of substrate transport across the cerebral capillary endothelium were examined in response to variations of the substrate demand of the brain tissue, and to variations of substrate concentration in the blood. The substrates examined included glucose and ketone bodies. The transport changes were measured in rats, using an indicator fractionation method modified by the reviewer. Four mechanisms appeared to contribute to the adjustment of substrate transport to variations in substrate demand. The first and least important mechanism was the change of concentration gradient across the endothelium that occurred when the substrate consumption rate changed. The second mechanism was the flow-dependency of the average capillary substrate concentration: the higher the perfusion rate, the higher the average capillary concentration. This mechanism failed to account for the changes of substrate transport observed during marked increases of the metabolic rate. The third and most important mechanism was a change of the capillary diffusion capacity, probably associated with a change of the number of perfused capillaries. The fourth mechanism, not previously described, was an adaptation of transport to permanent changes of substrate concentration in the blood. This mechanism appeared to reflect changes of the concentration (and affinity?) of transport proteins in the plasma membranes of endothelial cells, possibly in association with changes of cellular protein synthesis and gene expression.

The metabolic and hormonal response to acute normoglycaemia in type 1 (insulin-dependent) diabetes: studies with a glucose controlled insulin infusion system (artificial endocrine pancreas)

Twelve insulin deficient Type 1 (insulin-dependent) diabetic subjects were studied over an 11 1/2 h period during both subcutaneous insulin therapy and closed loop insulin delivery, using a glucose controlled insulin system (Biostator) programmed to maintain normoglycaemia. Results were compared with those from 21 age and weight-matched normal subjects. Using the Biostator, normoglycaemia was achieved in all diabetic subjects within 3.5 h and normal profiles maintained thereafter. Blood metabolite and hormone values were evaluated during the subsequent 8 h normoglycaemic period. Subcutaneous therapy resulted in abnormal glucose levels throughout the study period (mean 8 h value 8.3 +/- 0.7 compared with 5.6 +/- 0.3 mmol/l on feedback control and 5.5 +/- 0.1 mmol/l in normal subjects). The mean value of lactate and pyruvate over the final 8 h period was 25% higher in diabetic patients than in normal subjects with no difference between the two insulin treatments (blood lactate: 0.94 +/- 0.04 on subcutaneous insulin, 0.91 +/- 0.04 on feedback control and 0.74 +/- 0.03 mmol/l in control subjects). The pre-prandial peaks of blood glycerol and plasma non-esterified fatty acids were significantly decreased or absent during both feedback control and subcutaneous therapy in comparison with the normal subjects, whereas after the midday and evening meals, total ketone body levels were significantly higher in the diabetic patients. Peripheral serum free insulin levels were two-to fourfold greater in the diabetic than in the normal subjects. There were no significant differences between levels in diabetic patients receiving subcutaneous insulin or on the Biostator. Glucose turnover (1600-1800 h) was normal on feedback control (1.41 +/- 0.20 versus 1.55 +/- 0.18 mg X kg-1 X min-1 in the normal subjects) but was significantly decreased during subcutaneous insulin (1.04 +/- 0.09 mg X kg-1 X min-1). There was, in addition, a decrease in glucose recycling during both subcutaneous insulin therapy and feedback control in the diabetic subjects. These data suggest that although fine control of glucose metabolism both in terms of circulating concentrations and rates of production can be achieved by feedback-control, insulin infusion by the peripheral route is associated with significant metabolic abnormalities, at least in the short term. Longer term studies and examination of portal insulin delivery seem warranted.

Metabolic response of normal man and insulin-infused diabetics to postprandial exercise

Physical exercise is often performed during absorption of meals. We have characterized the metabolic response to 45 min of moderate exercise (approximately 55% of estimated maximal oxygen uptake) beginning 30 min after breakfast in seven healthy controls. Nine insulin-dependent diabetes were studied in an identical manner, with glycemia controlled by a closed-loop "artificial endocrine pancreas" controlled by a closed-loop "artificial endocrine pancreas" (AEP). Responses were compared to those during breakfast without exercise. In the controls, onset of exercise rapidly reversed the rise in both glycemia and insulin (IRI) that occurred with breakfast alone, both returning to fasting levels (glycemia, 80 +/- 3 mg/dl; IRI, 0.38 +/- 0.10 ng/ml). After exercise, small and transient increments occurred (glycemia, 33 +/- 6 mg/dl; IRI, 0.81 +/- 0.15 ng/ml). In the diabetics, prior overnight intravenous insulin normalized fasting glycemia (98 +/- 4 mg/dl), and its postbreakfast excursion was identical to that of controls, as were those of most measured substrates. Similarly, with exercise, glycemia returned rapidly to fasting levels, accompanied by an appropriate decrease in insulin infusion rates. "Free" IRI levels mirrored changes in infusion rates by the AEP, with a decrease in insulin requirement of 30% during exercise as compared to breakfast alone (P less than 0.05). Thus, in both diabetics treated with the AEP and in normals, the responses to postprandial exercise required rapid modulation of insulin delivery. To demonstrate the effect of postprandial exercise on preprogrammed open-loop insulin replacement, four diabetic subjects were studied during breakfast with and without exercise while receiving a fixed open-loop insulin infusion pattern (6.1 +/- 0.7 U over 140 +/- 8 min). The glycemic response to breakfast alone was entirely normalized. However, symptomatic hypoglycemia occurred in all subjects when exercise was initiated 30 min after breakfast. The diabetic responses to closed-loop insulin infusion provide important data in defining the appropriate preprogrammed open-loop insulin infusion pattern for postprandial exercise.

Blood-brain glucose transfer: repression in chronic hyperglycemia

Diabetic patients with increased plasma glucose concentrations may develop cerebral symptoms of hypoglycemia when their plasma glucose is rapidly lowered to normal concentrations. The symptoms may indicate insufficient transport of glucose from blood to brain. In rats with chronic hyperglycemia the maximum glucose transport capacity of the blood-brain barrier decreased from 400 to 290 micromoles per 100 grams per minute. When plasma glucose was lowered to normal values, the glucose transport rate into brain was 20 percent below normal. This suggests that repressive changes of the glucose transport mechanism occur in brain endothelial cells in response to increased plasma glucose.

The metabolic and hormonal responses to a mixed meal in unrestrained pancreatectomised dogs chronically treated by portal or peripheral insulin infusion

The metabolic and hormonal consequences of long term intravenous insulin replacement were studied in 11 pancreatectomised dogs. Insulin was delivered into the portal circulation of six animals for 164-224 days and into the peripheral circulation of the remainder for 123-365 days. Infusion rates were initially adjusted to achieve normoglycaemia in the fasting (0.37 +/- 0.01 mU Kg-1 min-1 portal; 0.45 +/- 0.03 mU kg-1 min-1 peripheral) and post-prandial states (2.57 +/- 0.07 mU kg-1 min-1 for 7 1/2 h portal; 3.16 +/- 0.18 mU kg-1 min-1 for 7 h peripheral). Animals were fed their usual mixed diet and blood samples were drawn from indwelling catheters at regular intervals for 24 h. A matched group of six normal dogs was similarly studied. Significantly less insulin was needed for glycaemic normalisation with portal (1.05 +/- 0.03 U kg-1 day-1) compared with peripheral (1.27 +/- 0.08 U kg-1 day-1) infusions, but post-prandial insulin levels were not normalised. Glucagon levels were normal and unaffected by the route of insulin infusion. Lactate and pyruvate responses were exaggerated post-prandially in the diabetic compared with the normal dogs. Fasting non-esterified fatty acid levels were suppressed with peripheral but normal with portal insulin infusion. There were only minor differences in the branched chain, essential and other non-essential amino acids except for alanine which was significantly above normal in the diabetic animals. Fasting levels of insulin, lactate, pyruvate and non-esterified fatty acids were normalised only with portal infusion while glucose, glucagon, 3-hydroxybutyrate and most amino acids were normalised regardless of the route of infusion. We conclude that the metabolic regulation achieved with portal insulin replacement is closer to normal than that achieved with peripheral infusion.

Treatment of diabetes mellitus by devices

A very important aspect of diabetes mellitus is whether or not normalization or near-normalization of blood glucose and/or other metabolites and hormones may reduce or eliminate the chronic complications of this disease. To answer this question and to provide a more "physiologic" approach to insulin administration, a constellation of devices have reached the stage of clinical investigation. These include small portable pump systems that can provide variable rates of insulin infusion via the subcutaneous intravenous or intraperitoneal routes. In addition, bedside artificial "beta cells" having the capability of providing insulin infusions, with the rate varying as a function of continuous glucose measurements, are available for short-term studies. Under development are implantable continuous infusion devices and implantable glucose sensors that could in the future lead to a miniaturized implantable glucose-controlled insulin administration system.

Insulin, glucagon, and amino acids during glycemic control by the artificial pancreas in diabetic man

The artificial endocrine pancreas (AEP) can normalize glycemia at rest and with meals. To determine whether insulin, glucagon, and amino acid profiles are also normalized, nine diabetics on subcutaneous insulin (S/C) and AEP control were compared to ten normal controls (NC). Glycemia was monitored continuously over 10 hr during which meals were consumed. Insulin infusion rate, and the levels of immunoreactive insulin (IRI) (in NC), free insulin (in S/C and AEP), C-peptide, glucagon, and amino acids are reported. Glycemia in AEP started at somewhat higher levels than in NC, but with breakfast and thereafter, it was identical. In S/C, hyperglycemia prevailed throughout, with no systematic change in free IRI. In AEP, both basal and peak free insulin levels, measured in four patients, were significantly higher than in NC. C-peptide values were significantly lower in diabetics and did not change with meals. Basal glucagon values were not different in the three groups and changes with meals were of small magnitude. Branched chain amino acids were higher in S/C and did not increase as in NC. In AEP, levels were lower than NC after the first two meals. Similarly, lysine and threonine were lower in AEP than in NC at the same times. Alanine, though similar at the onset, was lower 2 hr postbreakfast and higher 2 hrs postsupper in AEP and S/C compared to NC. These studies demonstrate that glycemic control with AEP is accompanied by hyperinsulinemia, which could account for the amino acid responses and the small alterations in immunoreactive glucagon (IRG) patterns. Further refinement is needed to obtain full normalization of metabolic profiles.

Blood glucose regulation using closed- and open-loop insulin delivery systems. II. Peripheral primed square wave infusions

Glucose was infused into anaesthetized dogs before and after pancreatectomy. In the diabetics blood glucose was regulated first by closed-loop and then by open-loop insulin delivery schemes. Insulin requirements for the latter were determined by resolving the former into a sequence of 3 different infusion rates: during the baseline and recovery periods, basal insulin was delivered at 0.37 +/- 0.02 mU/kg/min, while during the 60 min glucose infusion (10 mg/kg/min) there was an 8 min infusion at 4.96 +/- 0.37 mU/kg/min and a 52 min component at 1.85 +/- 0.08 mU/kg/min. With the open-loop method under these highly standardized conditions glycaemia was similar to normal controls but IRI levels were significantly higher, 13.5 vs 8.0 microU/ml (p less than 0.05) in the baseline and recovery periods and 74 vs 25 microU/ml (p less than 0.05) during the glucose infusion. It was concluded that: constant normoglycaemia can be maintained in the basal state by a constant rate of peripheral insulin delivery but at rates resulting in peripheral hyperinsulinaemia; the glycaemic response to glucose infusion can be normalized by a two component waveform of insulin delivery; and the closed-loop method can serve as a useful guide in determining insulin requirements.