The artificial pancreas: evaluating risk of hypoglycaemia following errors that can be expected with prolonged at-home use

AIMS

Artificial pancreas systems show benefit in closely monitored at-home studies, but may not have sufficient power to assess safety during infrequent, but expected, system or user errors. The aim of this study was to assess the safety of an artificial pancreas system emulating the β-cell when the glucose value used for control is improperly calibrated and participants forget to administer pre-meal insulin boluses.

METHODS

Artificial pancreas control was performed in a clinic research centre on three separate occasions each lasting from 10 p.m. to 2 p.m. Sensor glucose values normally used for artificial pancreas control were replaced with scaled blood glucose values calculated to be 20% lower than, equal to or 33% higher than the true blood glucose. Safe control was defined as blood glucose between 3.9 and 8.3 mmol/l.

RESULTS

Artificial pancreas control resulted in fasting scaled blood glucose values not different from target (6.67 mmol/l) at any scaling factor. Meal control with scaled blood glucose 33% higher than blood glucose resulted in supplemental carbohydrate to prevent hypoglycaemia in four of six participants during breakfast, and one participant during the night. In all instances, scaled blood glucose reported blood glucose as safe.

CONCLUSIONS

Outpatient trials evaluating artificial pancreas performance based on sensor glucose may not detect hypoglycaemia when sensor glucose reads higher than blood glucose. Because these errors are expected to occur, in-hospital artificial pancreas studies using supplemental carbohydrate in anticipation of hypoglycaemia, which allow safety to be assessed in a controlled non-significant environment should be considered as an alternative. Inpatient studies provide a definitive alternative to model-based computer simulations and can be conducted in parallel with closely monitored outpatient artificial pancreas studies used to assess benefit.

Interstitial fluid glucose dynamics during insulin-induced hypoglycaemia

AIMS/HYPOTHESIS

Glucose sensors often measure s.c. interstitial fluid (ISF) glucose rather than blood or plasma glucose. Putative differences between plasma and ISF glucose include a protracted delay during the recovery from hypoglycaemia and an increased gradient during hyperinsulinaemia. These have often been investigated using sensor systems that have delays due to signal smoothing, or require long equilibration times. The aim of the present study was to define these relationships during hypoglycaemia in a well-equilibrated system with no smoothing.

METHODS

Hypoglycaemia was induced by i.v. insulin infusion (360 pmol.m(-2).min(-1)) in ten non-diabetic subjects. Glucose was sequentially clamped at approximately 5, 4.2 and 3.1 mmol/l and allowed to return to normoglycaemia. Subjects wore two s.c. glucose sensors (Medtronic MiniMed, Northridge, CA, USA) that had been inserted for more than 12 h. A two-compartment model was used to quantify the delay and gradient.

RESULTS

The delay during the fall in plasma glucose was not different from the delay during recovery (8.3+/-0.67 vs 6.3+/-1.1 min; p=0.27) and no differences were observed in the ratio of sensor current to plasma glucose at basal insulin (2.7+/-0.25 nA.mmol(-1).l) compared with any of the hyperinsulinaemic clamp phases (2.8+/-0.18, 2.7+/-0.021, 2.9+/-0.21; p=NS). The ratio was significantly elevated following recovery to normoglycaemia (3.1+/-0.2 nA.mmol(-1).l; p<0.001).

CONCLUSIONS/INTERPRETATION

The elevated ratio suggests that the plasma to ISF glucose gradient was decreased following hypoglycaemia, possibly due to increased skin blood flow. Recovery from hypoglycaemia is not accompanied by a protracted delay and insulin does not increase the plasma to s.c. ISF glucose gradient.

Modeling insulin action for development of a closed-loop artificial pancreas

Three models of glucose homeostasis are compared in terms of their steady-state dose-response characteristics, how they characterize glucose distribution kinetics, and how they characterize the dynamics of insulin action. The three models [minimal model, AIDA (Automated Insulin Dosage Advisor), and a model by Sorensen] are used to discuss a wider variety of questions related to metabolic modeling. Simulations are performed comparing each model's response to an intravenous glucose tolerance test, with and without incremental insulin responses, to existing data in individuals with type 1 diabetes mellitus. Predicted changes in blood glucose following a subcutaneous bolus of insulin or an incremental increase in basal insulin delivery are simulated. From these results, the models are evaluated as potential candidates for simulating changes in treatment and developing a closed-loop insulin delivery algorithm. While no consensus model is proposed, relevant issues needing to be addressed are highlighted.

Closed-loop insulin delivery-the path to physiological glucose control

The development of an artificial pancreas for the treatment of insulin-dependent diabetes is a highly desired endeavor for patients, physicians, scientists, and engineers. Historical algorithms and recent progress in research and technology are reviewed in the present article, together with aspects of beta-cell physiology that lead to normal glucose tolerance. Algorithms are evaluated for their ability to deliver insulin as to recreate, as closely as possible, glucose and insulin profiles observed in healthy individuals. Emphasis is placed upon how the algorithms compare to the beta-cell's secretory response, specifically first-phase and second-phase insulin secretion. Experimental closed-loop data employing intravenous and subcutaneous glucose sensors and implanted and external insulin pumps (Medtronic MiniMed, Northridge CA) are presented.

High glucose stimulates early response gene c-Myc expression in rat pancreatic beta cells

Glucose-induced insulin secretion from hyperglycemic 90% pancreatectomized rats is markedly impaired, possibly because of loss of beta cell differentiation. Association of these changes with beta cell hypertrophy, increased mRNA levels of the transcription factor c-Myc, and their complete normalization by phlorizin treatment suggested a link between chronic hyperglycemia, increased c-Myc expression, and altered beta cell function. In this study, we tested the effect of hyperglycemia on rat pancreatic islet c-Myc expression both in vivo and in vitro. Elevation of plasma glucose for 1-4 days (glucose infusion/clamp) was followed by parallel increases in islet mRNA levels (relative to TATA-binding protein) of c-Myc and two of its target genes, ornithine decarboxylase and lactate dehydrogenase A. Similar changes were observed in vitro upon stimulation of cultured islets or purified beta cells with 20 and 30 mmol.liter(-1) glucose for 18 h. These effects of high glucose were reproduced by high potassium-induced depolarization or dibutyryl-cAMP and were inhibited by agents decreasing cytosolic Ca(2+) or cAMP concentrations. In conclusion, the expression of the early response gene c-Myc in rat pancreatic beta cells is stimulated by high glucose in a Ca(2+)-dependent manner and by cAMP. c-Myc could therefore participate to the regulation of beta cell growth, apoptosis, and differentiation under physiological or pathophysiological conditions.

Gene expression of VEGF and its receptors Flk-1/KDR and Flt-1 in cultured and transplanted rat islets

BACKGROUND

Vascular endothelial growth factor (VEGF) and its two receptor tyrosine kinases, Flk-1/KDR and Flt-1, may play an important role in mediating the revascularization of transplanted pancreatic islets.

METHODS

Using semiquantitative multiplex reverse-transcribed polymerase chain reaction we determined the gene expression of VEGF and its receptors in cultured and transplanted rat islets.

RESULTS

After exposure of islet cells to hypoxia in vitro, increases were found in the gene expression of the VEGF120 and VEGF164 isoforms, with simultaneous increases in VE-cadherin, Flk-1/KDR, and Flt-1. In vivo studies consisted of analysis of islet grafts transplanted into both normal and diabetic recipients. Expression of both VEGF120 and VEGF164 in grafts was up-regulated for the first 2-3 days after transplantation, with the response being more prolonged in the diabetic rats. These increases were followed by reduced expression of VEGF on days 5, 7, and 9. Increases in the expression of VE-cadherin in islet grafts in normal and diabetic recipients tended to parallel VEGF expression, with the increases in both probably being caused by hypoxia. The early increases of VEGF expression were followed by a rise in the expression of VEGF receptors, which probably represents the early stages of angiogenesis. Graft expression of Flk-1/KDR and Flt-1 was enhanced at 3 and 5 days in the normoglycemic recipients, while in the diabetic recipients increases were found later on days 5, 7, and 14.

CONCLUSIONS

The delayed expression of VEGF receptors in the diabetic recipients could reflect impaired angiogenesis caused by the diabetic milieu; this delay could contribute to the less outcomes of grafts transplanted into a hyperglycemic environment.

Adaptation of beta-cell mass to substrate oversupply: enhanced function with normal gene expression

Although type 2 diabetes mellitus is associated with insulin resistance, many individuals compensate by increasing insulin secretion. Putative mechanisms underlying this compensation were assessed in the present study by use of 4-day glucose (GLC; 35% Glc, 2 ml/h) and lipid (LIH; 10% Intralipid + 20 U/ml heparin; 2 ml/h) infusions to rats. Within 2 days of beginning the infusion of either lipid or glucose, plasma glucose profiles were normalized (relative to saline-infused control rats; SAL; 0.45% 2 ml/h). During glucose infusion, plasma glucose was maintained in the normal range by an approximately twofold increase in plasma insulin and an approximately 80% increase in beta-cell mass. During LIH infusion, glucose profiles were also maintained in the normal range. Plasma insulin responses during feeding were doubled, and beta-cell mass increased 54%. For both groups, the increase in beta-cell mass was associated with increased beta-cell proliferation (98% increase during GLC and 125% increase during LIH). At the end of the 4-day infusions, no significant changes were observed in islet-specific gene transcription (i.e., the expression of islet hormone genes, glucose metabolism genes, and insulin transcription factors were unaffected). Two days after termination of the infusions, the glucose-stimulated plasma insulin response was increased approximately 67% in glucose-infused animals. No sustained effect on insulin secretory capacity was observed in the LIH animals. The increase in plasma insulin response after glucose infusion was achieved in the absence of any change in insulin clearance. We conclude that, in rats, an increase in insulin demand after an increase in glucose appearance or free fatty acid leads to an increase in beta-cell mass, mediated in part by an increase in beta-cell proliferation, and that these compensatory changes lead to increased insulin secretion, normal plasma glucose levels, and the maintenance of normal islet gene expression.

Islets in alginate macrobeads reverse diabetes despite minimal acute insulin secretory responses

BACKGROUND

Encapsulation of islets has been widely investigated as a treatment for diabetes. The characteristics and dynamics of insulin secretion by encapsulated islets in response to glucose and other secretagogues are not well understood.

METHODS

In our study, macroencapsulated syngeneic islets at 3-4 wk after transplantation were studied for insulin release in response to i.v. glucose (hyperglycemic clamps at 250 or 350 mg/dl plasma glucose), arginine (i.v. bolus, 100 mg/kg), glucagon-like peptide-1 (i.v. infusion for 20 min, 2.2 pmol/kg/min), and meal challenge. Syngeneic islets (6000 islets) were encapsulated in alginate macrobeads (2-3 mm diameter) with or without poly-L-lysine coating and transplanted into the peritoneal cavity of STZ-diabetic Lewis rats. Normal (nontransplanted) and diabetic Lewis rats transplanted with "naked" islets under the kidney capsule served as controls.

RESULTS

Animals transplanted with macrobeads displayed subnormal insulin responses to glucose, arginine, and glucagon-like peptide-1 despite achieving normoglycemia faster than animals with renal subcapsular islet transplants. Plasma insulin responses to meal challenges were blunted in animals with macrobeads resulting in increased plasma glucose excursions.

CONCLUSIONS

We conclude that, after transplantation into diabetic Lewis rats, macroencapsulated islets have significantly impaired insulin secretion despite achieving normal fed glycemic levels.

Beta-cell dysfunction in 48-hour glucose-infused rats is not a consequence of elevated plasma lipid or islet triglyceride levels

The abnormal insulin secretion found in human diabetics and animal models of diabetes has been attributed to the deleterious effects of chronic hyperglycemia and/or elevated circulating levels of nonesterified fatty acids (NEFAs). In this study, abnormal glucose-induced insulin secretion (GIIS) was generated by a 48-hour infusion of glucose and assessed by the isolated perfused pancreas technique. In these hyperglycemic animals, abnormal GIIS is accompanied by a decrease in plasma NEFAs, while plasma and, more importantly, islet triglycerides remain at levels comparable to those in the controls. It is concluded that the abnormal insulin secretion in this glucose infusion model was likely caused by 48 hours of hyperglycemia and not by changes in circulating or islet lipids.

Improved vascularization of planar membrane diffusion devices following continuous infusion of vascular endothelial growth factor

Improving blood vessel formation around an immunobarrier device should improve the survival of the encapsulated tissue. In the present study we investigated the formation of new blood vessels around a planar membrane diffusion device (the Baxter Theracyte System) undergoing a continuous infusion of vascular endothelial growth factor through the membranes and into the surrounding tissue. Each device (20 microl) had both an inner immunoisolation membrane and an outer vascularizing membrane. Human recombinant vascular endothelial growth factor-165 was infused at 100 ng/day (low dose: n = 6) and 500 ng/day (high dose: n = 7) for 10 days into devices implanted s.c. in Sprague-Dawley rats; noninfused devices transplanted for an identical period were used as controls (n = 5). Two days following the termination of VEGF infusion, devices were loaded with 20 microl of Lispro insulin (1 U/kg) and the kinetics of insulin release from the lumen of the device was assessed. Devices were then explanted and the number of blood vessels (capillary and noncapillary) was quantified using morphometry. High-dose vascular endothelial growth factor infusion resulted in two- to threefold more blood vessels around the device than that obtained with the noninfused devices and devices infused with low-dose vascular endothelial growth factor. This increase in the number of blood vessels was accompanied by a modest increase in insulin diffusion from the device in the high-dose vascular endothelial growth factor infusion group. We conclude that vascular endothelial growth factor can be used to improve blood vessel formation adjacent to planar membrane diffusion devices.

Can interstitial glucose assessment replace blood glucose measurements?

Current treatment regiments for individuals depending on exogenous insulin are based on measurements of blood glucose obtained through painful finger sticks. The shift to minimal or noninvasive continuous glucose monitoring primarily involves a shift from blood glucose measurements to devices measuring subcutaneous interstitial fluid (ISF) glucose. As the development of these devices progresses, details of the dynamic relationship between blood glucose and interstitial glucose dynamics need to be firmly established. This is a challenging task insofar as direct measures of ISF glucose are not readily available. The current article investigated the dynamic relationship between plasma and ISF glucose using a model-based approach. A two-compartment model system previously validated on data obtained with the MiniMed Continuous Glucose Monitoring System (CGMS) is reviewed and predictions from the original two-compartment model were confirmed using new data analysis of glucose dynamics in plasma and hindlimb lymph (lymph is derived from ISF) in the anesthetized dog. From these data sets, the time delay between plasma and ISF glucose in dogs was established (5-12 minutes) and a simulation study was performed to estimate the errors introduced if ISF is taken as a surrogate for blood. From the simulation study, the error component resulting from the differences in plasma and ISF glucose was estimated to be < 6% during normal day-to-day use in an individual with diabetes (error component calculated as the standard deviation of the ISF/plasma glucose differences under conditions where the maximal time delay was used). This difference is most likely within the variance between arterial and venous blood glucose. We conclude that the differences between plasma and ISF glucose will not be a significant obstacle in advancing the use of ISF as an alternative to blood glucose measurements.

Subcutaneous glucose predicts plasma glucose independent of insulin: implications for continuous monitoring

The present study investigated the relationship between blood and subcutaneous interstitial fluid (ISF) glucose by employing an amperometric glucose sensor specifically developed for 3-day continuous glucose monitoring. The apparent sensor sensitivity and ISF glucose equilibration delay were estimated on separate days during hyperglycemic clamps in four dogs in which insulin was either suppressed with somatostatin, allowed to change, or increased with an exogenous infusion. A 2-h sensor "settling-in" period was allowed before the clamps. During insulin deficiency, the sensor sensitivity and ISF glucose delay were 0.23 +/- 0.03 nA per mg/dl and 4.4 +/- 0. 8 min. Sensitivity was not affected by increases in endogenous (0.30 +/- 0.04 vs. 0.28 +/- 0.04 nA per mg/dl) or exogenous insulin (0.18 +/- 0.01 vs. 0.16 +/- 0.01 nA per mg/dl) nor was the delay (3.3 +/- 1.2 vs. 5.7 +/- 1.1 and 9.2 +/- 2.6 vs. 12.3 +/- 1.7 min; P > 0.05 for all). Sensor glucose accurately predicted plasma glucose without correcting for delays <10 min (r > 0.9 for all), whereas for longer delays a digital corrective filter was used (r = 0.91 with filter). We conclude that the relationship between blood and ISF glucose is not affected by insulin and that delays in ISF glucose equilibration can be corrected with digital filters.

Role of portal insulin delivery in the disappearance of intravenous glucose and assessment of insulin sensitivity

The contribution of portal insulin delivery to the disappearance of glucose administered intravenously was assessed in the present study. Paired insulin-modified intravenous glucose tolerance tests (IVGTTs) were performed in dogs in which insulin was administered into the portal vein or into a peripheral vein. Peripheral insulin levels were matched in the paired IVGTTs by adjusting the portal insulin dose in proportion to first-pass hepatic insulin extraction. Two sets of IVGTTs were performed. In the first set, hepatic insulin extraction was assumed to be 50% (insulin doses of 0.03 U/kg portal and 0.015 U/kg peripheral; n = 6); in the second set, the assumed extraction rate was reduced to 33% (0.0225 U/kg portal and 0.015 U/kg peripheral; n = 8). In the second set of experiments, a control "zero" dose (no insulin injection) was also performed. For these latter three IVGTTs, the exogenous glucose bolus was labeled with 3-[3H]glucose (25 microCi) to separately assess insulin's effects on the rate of glucose disappearance (Rd) and endogenous glucose production (EGP). For the paired IVGTT based on 33% extraction, the area under the insulin curves after the portal insulin injection was within 2% of that observed with peripheral insulin injection (1,820 +/- 711 vs. 1,791 +/- 661 microU/ml min; P = 0.79). For these conditions, neither the glucose profiles nor the minimal model estimate of insulin sensitivity (S(I)) was significantly influenced by the higher portal insulin delivery (S(I): 3.69 +/- 0.56 vs. 3.35 +/- 0.60 10(-4) min(-1) per microU/ml; portal vs. peripheral; P > 0.05). Analysis of the 3-[3H]glucose tracer dynamics failed to reveal any differences in the portal versus peripheral insulin effect on glucose disappearance or production. We conclude that portal insulin delivery per se does not significantly affect insulin's ability to normalize plasma glucose during acute glucose challenges.

Physiological insulinemia acutely modulates plasma leptin

Whether insulin acutely regulates plasma leptin in humans is controversial. We examined the dosage-response and time-course characteristics of the effect of insulin on leptin in 10 men (age 42+/-2 years [mean+/-SE]; BMI 29.3+/-2.0 kg/m2). Each individual underwent four 9-h euglycemic clamps (insulin at 20, 40, 80, and 400 mU x m[-2] x min[-1) and a control saline infusion. Although plasma glucose and insulin levels remained constant, leptin diminished from 9.1+/-3.0 to 5.9+/-2.1 ng/ml (P < 0.001) by the end of the control experiment. Conversely, plasma leptin showed a dosage-dependent increase during the insulin infusions that was evident within 30-60 min. The insulin-induced increase in leptin was proportionately lower in obese insulin-resistant men. Free fatty acids (FFAs) decreased during insulin and did not change during saline infusions. ED50 (the dose producing half-maximal effect) for insulin's effect on leptin and FFA was similar (138+/-36 vs. 102+/-24 pmol/l, respectively; P=0.11). To further define the role of physiological insulinemia, we compared the effect of a very low dosage insulin infusion (10 mU x m[-2] x min[-1]) with that of a control saline infusion in another group of 10 men (mean age 39+/-3 years; BMI 27.1+/-1.0 kg/m2). Plasma leptin remained stable during that insulin infusion, but fell by 37+/-2% in the control experiment. Thus physiological insulinemia can acutely regulate plasma leptin. Insulin could mediate the effect of caloric intake on leptin and could be a determinant of its plasma concentration. Inadequate insulin-induced leptin production in obese and insulin-resistant subjects may contribute to the development or worsening of obesity.

Diurnal and ultradian rhythmicity of plasma leptin: effects of gender and adiposity

Plasma leptin shows a nocturnal rise and a pulsatile pattern. This work was undertaken to determine the effects of gender and obesity on this pattern. Twenty-four-hour leptin profiles were evaluated in 31 subjects [17 male, 14 female; age: 36 +/- 2 yr (mean +/- SEM); body mass index: 27.5 +/- 1.0 kg/m2]. Plasma leptin profiles were higher in obese (body mass index > 27 kg/m2) than in lean subjects and higher in women than in men, regardless of fat mass. Leptin showed diurnal rhythmicity with peaks between 2200-0300 (median: 0120) and nadirs between 0800 and 1740 (median: 1033). Spectral analysis revealed 2 components (periodicities: 24 and 12 h) with higher relative amplitudes in lean than in obese subjects. The relative diurnal amplitude also was higher in men than in women, controlling for adiposity. Insulinemia, female sex, and age were negative determinants of diurnal rhythm relative amplitude. Pulse analysis revealed 3.6 +/- 0.3 pulses/24 h, occurring mostly 2-3 h after meals. Pulse frequency correlated negatively with fat mass and insulinemia (Spearman's r = -0.54 and -0.37, respectively; P < 0.05 for each). Thus, obesity is associated not only with higher leptin levels but also with blunted diurnal excursions and dampened pulsatility. This abnormal rhythmicity may contribute to leptin resistance in obesity. The significance of the sexual dimorphism in the diurnal amplitude is unclear, but it may be related to leptin's putative role as a metabolic signal to the reproductive axis.

Critical evaluation of the combined model approach for estimation of prehepatic insulin secretion

The combined model approach uses kinetic analysis of both plasma insulin and C-peptide dynamics to estimate prehepatic insulin secretion rates and parameters of insulin and C-peptide kinetics. The original model used single-compartment kinetics to describe both insulin and C-peptide despite knowledge that C-peptide follows two-compartment kinetics. The performance of the model under rapidly changing secretory conditions has come into question. Thus a more complex combined model is introduced, incorporating two-compartmental C-peptide disappearance. The addition of two-compartment C-peptide kinetics required a novel numerical approach to allow estimation of model parameters. This simulation study was undertaken to 1) compare the performance of the original combined model and 2) examine the numerical method used to identify parameters for the extended combined model with two-compartment C-peptide kinetics under simulated conditions of rapidly changing insulin and C-peptide. Monte Carlo simulation revealed that the original combined model does not provide accurate estimates of prehepatic insulin secretion under rapid kinetics. However, the extended combined model provides accurate reconstruction of prehepatic insulin secretory profile without separate quantification of C-peptide kinetics.

Method of insulin administration has no effect on insulin sensitivity estimates from the insulin-modified minimal model protocol

The effect of the method of insulin administration on insulin sensitivity estimates from the insulin modified minimal model (MINMOD) protocol was evaluated using the tolbutamide-boosted protocol as a reference. The study included 21 nondiabetic men ages 40 +/- 2 years (mean +/- SE) with a BMI of 26.6 +/- 1.1 kg/m2. Each subject underwent four frequently sampled intravenous glucose tolerance tests (FSIGTT), one with tolbutamide and three with the same insulin dosage (0.03 U/kg) given as a bolus or infusion over 5 or 10 min. The insulin sensitivity index (SI) of each subject was calculated from each FSIGTT with MINMOD. Insulin sensitivity indexes from the four FSIGTTs were highly correlated (r > 0.85, P < 0.001). SI(insulin) from the bolus and the 5- and 10-min infusion protocols were similar, but were 21 +/- 5, 29 +/- 5, and 23 +/- 4% lower than SI(tolbutamide), respectively. SG(tolbutamide) and SG(insulin) were not different among the four protocols and were significantly correlated (r > 0.55, P < 0.01). Thus the tolbutamide and insulin protocols must not be used interchangeably in any single cross-sectional or longitudinal study. When the same insulin dosage is used, the method of its administration has no bearing on insulin sensitivity estimates from the insulin-modified FSIGTT. The same method of insulin administration should be used, however, in any single study for purpose of standardization.

Differences between the tolbutamide-boosted and the insulin-modified minimal model protocols

The insulin-modified frequently sampled intravenous glucose tolerance test (FSIGTT) with minimal model analysis (MINMOD) was compared with the tolbutamide protocol and the glucose clamp in 35 nondiabetic subjects (age 38 +/- 2 years [mean +/- SE], BMI 27.2 +/- 0.9 kg/m2). Each subject underwent two FSIGTTs, one with tolbutamide (300 mg) and the other with insulin (0.03 U/kg) and a euglycemic hyperinsulinemic clamp (40 mU x m(-2) x min(-1)). Insulin sensitivity was determined from each FSIGTT with MINMOD and from the clamp. Insulin sensitivity indexes (S(I)) from the two FSIGTTs were significantly correlated (r = 0.77, P < 0.001), but S(I(insulin)) was 29 +/- 4% lower than S(I(tolbutamide)). Both S(I(insulin)) and S(I(tolbutamide)) correlated significantly with S(I(clamp)) (r = 0.70 and 0.71, P < 0.001 for each). Expressed in the same units (dl/min per pU/ml), S(I(tolbutamide)) was on average 13 +/- 6% lower than S(I(clamp)) (4.51 +/- 0.40 vs. 5.36 +/- 0.36 x 10(-2), P = 0.009), whereas S(I(insulin)) was 44 +/- 4% lower. S(G(tolbutamide)) and S(G(insulin)) were not different (1.88 +/- 0.10 vs. 2.01 +/- 0.09 x 10(-2) min(-1), P = 0.167) and were significantly correlated (r = 0.50, P = 0.002). Thus, insulin sensitivity estimates from both protocols correlate significantly with each other and with the clamp. They are quantitatively discrepant, however, possibly due to differences in the route of insulin delivery, saturation of insulin action, and/or tolbutamide-induced proinsulin release. Data obtained from these two MINMOD protocols are not directly comparable, and the same protocol must be used in any single cross-sectional or longitudinal study.

Extracellular glucose distribution is not altered by insulin: analysis of plasma and interstitial L-glucose kinetics

We examined the effects of insulin on leg blood flow, whole body extracellular glucose distribution, and glucose diffusion into the interstitial fluid (ISF) surrounding skeletal muscle cells in anesthetized dogs. Extracellular glucose distribution and glucose diffusion into the muscle ISF were assessed by studying the kinetics of L-[1-14C]glucose in plasma and hindlimb lymph. Femoral artery blood flow was not increased with insulin (7.9 +/- 0.7 vs. 7.1 +/- 1.4 ml.min-1.kg-1; P = 0.54). Plasma and lymph dynamics of L-glucose after intravenous administration were superimposable between saline and insulin infusion experiments, indicating that insulin did not affect L-glucose disappearance from plasma or appearance in muscle ISF. Plasma L-glucose kinetics were best described by a four-compartment model, and one of the remote pools (intermediate) predicted the lymph L-glucose dynamics well. Estimation of maximum glucose diffusion capacity indicated that this pool, rather than the slowest pool, represents insulin-sensitive tissues. In conclusion, our data indicate that insulin does not increase transcapillary glucose diffusion to insulin-sensitive cells. In addition, hindlimb lymph represents primarily skeletal muscle ISF, which is represented by an intermediate, rather than the slowest, remote pool from whole body compartmental analysis.

Causal linkage between insulin suppression of lipolysis and suppression of liver glucose output in dogs

Suppression of hepatic glucose output (HGO) has been shown to be primarily mediated by peripheral rather than portal insulin concentrations; however, the mechanism by which peripheral insulin suppresses HGO has not yet been determined. Previous findings by our group indicated a strong correlation between free fatty acids (FFA) and HGO, suggesting that insulin suppression of HGO is mediated via suppression of lipolysis. To directly test the hypothesis that insulin suppression of HGO is causally linked to the suppression of adipose tissue lipolysis, we performed euglycemic-hyperinsulinemic glucose clamps in conscious dogs (n = 8) in which FFA were either allowed to fall or were prevented from falling with Liposyn plus heparin infusion (LI; 0.5 ml/min 20% Liposyn plus 25 U/min heparin with a 250 U prime). Endogenous insulin and glucagon were suppressed with somatostatin (1 microgram/min/kg), and insulin was infused at a rate of either 0.125 or 0.5 mU/min/kg. Two additional experiments were performed at the 0.5 mU/min/kg insulin dose: a double Liposyn infusion (2 x LI; 1.0 ml/min 20% Liposyn, heparin as above), and a glycerol infusion (19 mg/min). With the 0.125 mU/min/kg insulin infusion, FFA fell 40% and HGO fell 33%; preventing the fall in FFA with LI entirely prevented this decline in HGO. With 0.5 mU/min/kg insulin infusion, FFA levels fell 64% while HGO declined 62%. Preventing the fall in FFA at this higher insulin dose largely prevented the fall in HGO; however, steady state HGO still declined by 18%. Doubling the LI infusion did not further affect HGO, suggesting that the effect of FFA on HGO is saturable. Elevating plasma glycerol levels did not alter insulin's ability to suppress HGO. These data directly support the concept that insulin suppression of HGO is not direct, but rather is mediated via insulin suppression of adipose tissue lipolysis. Thus, resistance to insulin control of hepatic glucose production in obesity and/or non-insulin-dependent diabetes mellitus may reflect resistance of the adipocyte to insulin suppression of lipolysis.

Transendothelial insulin transport is not saturable in vivo. No evidence for a receptor-mediated process

In vitro, insulin transport across endothelial cells has been reported to be saturable, suggesting that the transport process is receptor mediated. In the present study, the transport of insulin across capillary endothelial cells was investigated in vivo. Euglycemic glucose clamps were performed in anesthetized dogs (n = 16) in which insulin was infused to achieve concentrations in the physiological range (1.0 mU/kg per min + 5 mU/kg priming bolus; n = 8) or pharmacologic range (18 mU/kg per min + 325 mU/kg priming bolus; n = 8). Insulin concentrations were measured in plasma and hindlimb lymph derived from interstitial fluid (ISF) surrounding muscle. Basal plasma insulin concentrations were twice the basal ISF insulin concentrations and were not different between the physiologic and pharmacologic infusion groups (plasma/ISF ratio 2.05 +/- 0.22 vs 2.05 +/- 0.23; p = 0.0003). The plasma/ISF gradient was, however, significantly reduced at steady-state pharmacologic insulin concentrations (1.37 +/- 0.25 vs 1.98 +/- 0.21; P = 0.0003). The reduced gradient is opposite to that expected if transendothelial insulin transport were saturable. Insulin transport into muscle ISF tended to increase with pharmacologic compared with physiologic changes in insulin concentration (41% increase; 1.37 +/- 0.18 10(-2) to 1.93 +/- 0.24 10(-2) min-1; P = 0.088), while at the same time insulin clearance out of the muscle ISF compartment was unaltered (2.53 +/- 0.26 10(-2) vs 2.34 +/- 0.28 10(-2) min-1; P = 0.62). Thus, the reduced plasma/ISF gradient at pharmacologic insulin was due to enhanced transendothelial insulin transport rather than changes in ISF insulin clearance. We conclude that insulin transport is not saturable in vivo and thus not receptor mediated. The increase in transport efficiency with saturating insulin is likely due to an increase in diffusionary capacity resulting from capillary dilation or recruitment.

Free fatty acid as a link in the regulation of hepatic glucose output by peripheral insulin

Overproduction of glucose by the liver in the face of insulin resistance is a primary cause of hyperglycemia in non-insulin-dependent diabetes mellitus (NIDDM). However, mechanisms involved in control of hepatic glucose output (HGO) remain less than clear, even in normal individuals. Recent results have supported an indirect extrahepatic effect of insulin as the primary locus of insulin action to restrain HGO. One suggested extrahepatic site is the pancreatic alpha-cell. To examine whether insulin's extrahepatic site is independent of the alpha-cells, HGO suppression was examined independent of changes in glucagon secretion or insulin antagonism of glucagon action. Euglycemic glucose clamps (n = 40) with somatostatin infusion were performed in conscious dogs (n = 5). Paired experiments were conducted in which insulin was infused either portally (1.2, 3.0, 6.0 pmol.min-1.kg-1) or peripherally at half the portal infusion rate (0.6, 1.5, 3.0 pmol.min-1.kg-1). Additional zero and saturating portal-dose experiments (100 pmol.min-1.kg-1) were also performed. For the paired experiments, portal insulin infusion resulted in portal insulin concentrations approximately two to three times higher than in the corresponding peripheral insulin infusion experiments, while at the same time peripheral insulin concentrations were approximately matched. Equal peripheral insulin concentration resulted in equivalent HGO suppression irrespective of the portal concentrations. Thus, insulin affects a signal at a peripheral site, other than alpha-cell, that in turn suppresses hepatic glucose production. To investigate the nature of this signal, we measured alanine, lactate, and free fatty acids (FFAs).(ABSTRACT TRUNCATED AT 250 WORDS)

Insulin sensitivity accounts for glucose and lactate kinetics after intravenous glucose injection

Mathematical modeling was used to explore the interaction between glucose, insulin, and lactate during the frequently sampled intravenous glucose tolerance test (FSIGTT). Insulin-modified FSIGTs were performed in 25 lean volunteers, and an additional 5 volunteers underwent FSIGTs with glucose injection alone to illustrate the effect of insulin on both glucose and lactate kinetics. The model chosen as the best representation of the system extended the minimal model of glucose kinetics (MM) by including a two-compartment model of lactate kinetics. The model accounted for both glucose and lactate kinetics, provided traditional MM parameters of insulin sensitivity and glucose effectiveness, and descriptive parameters of lactate kinetics. Modeling suggested that lactate production was limited by the rate of glucose disappearance, with no indication of direct effects of insulin on lactate. Inclusion of lactate kinetics had no adverse effect on MM parameters (SG: 0.023 +/- 0.009 vs. 0.023 +/- 0.010 min-1, SI: 1.01 +/- 0.70 vs. 1.03 +/- 0.71 x 10(4).min-1.pmol-1.1; P > 0.50, lactate model vs. MM), and indicated that approximately 1.2% min-1 of total glucose disappearance during the FSIGT is converted to lactate. An additional benefit of including lactate kinetics was the significant improvement in precision in MM parameter estimates as reflected by the fractional standard deviations (FSDs). This effect was most prominent for SG, in which a threefold improvement in parameter precision was observed (FSD: 13.5 +/- 3.1 vs. 42.5 +/- 48.5; means +/- SD).(ABSTRACT TRUNCATED AT 250 WORDS)

Assessment of extracellular glucose distribution and glucose transport activity in conscious rats

The effects of insulin on extracellular glucose distribution and cellular glucose transport activity were studied by simultaneously analyzing the plasma kinetics of L-[1-14C]glucose and 3-O-[3H]methylglucose after an intravenous injection during saline or insulin infusion (euglycemic glucose clamp) in conscious rats (n = 7 for each). The time profiles of plasma L-glucose were almost superimposable in the two protocols, and compartmental analysis showed that neither distribution volumes nor distribution rate constants were affected with insulin (P > 0.05 for all), suggesting that glucose distribution within the extracellular space was not influenced with insulin. In contrast, the time profile of plasma 3-O-methylglucose (3-MG) was markedly altered with insulin; the initial decrease was much faster during insulin infusion than during saline infusion, indicating stimulation of 3-MG transport into intracellular spaces with insulin. The 3-MG data were analyzed using a comprehensive model separately describing extracellular distribution and cellular transport of 3-MG by incorporating information on extracellular distribution kinetics obtained from L-glucose data. The combined L-glucose and 3-MG kinetic analysis precisely estimated insulin's effect in vivo to stimulate glucose transport into and out of intracellular spaces. We conclude that 1) insulin does not affect extracellular glucose distribution kinetics or volumes in conscious rats and 2) insulin's effects on cellular glucose transport in vivo can be assessed by simultaneous analysis of plasma L-glucose and 3-MG kinetics.

OOPSEG: a data smoothing program for quantitation and isolation of random measurement error

We describe a data smoothing program, OOPSEG, which automatically quantitates and filters the random measurement error in a given data series. The measurement error is initially guessed, and the Optimal Segments technique is used to filter a corresponding amount of variation from the data, thus generating a relatively smooth curve. The residuals about this smooth curve are tested for serial correlation. If correlation is detected, a new measurement error is guessed, and the data are filtered again. This process continues until a smooth curve has been found for which the residuals do not exhibit serial correlation. Such residuals represent the random component of the data, i.e. (presumably) the measurement error. The corresponding smooth curve thus approximates the original, error-free curve. The smooth curve and the estimated coefficient of variation of the data are returned. OOPSEG handles end effects by performing an expansion of the data set, smoothing this expanded data set, then deleting the artificial points before assembling the final, smooth curve. Also, data may be smoothed as a sequence of independent regions to accommodate time courses with known changes in experimental conditions. We suggest that OOPSEG may have numerous applications in data analysis and experimental design.

Repeatability of insulin sensitivity and glucose effectiveness from the minimal model. Implications for study design

To determine the repeatability of insulin sensitivity measurements generated by the minimal model, we subjected 11 normal men to 3 frequently sampled intravenous glucose tolerance tests (FSIGTs) over the course of 12 days under conditions of fixed diet and limited physical activity. FSIGTs were analyzed by the minimal model using a full, 30-sample data set, as well as a reduced, 12-sample data set that has been proposed for population studies. Minimal model insulin sensitivity index (SI) calculated from the 30-sample data set averaged 0.8 +/- 0.083 x 10(-4) min-1.(pmol/l)-1 (range 0.10-1.64 x 10(-4) min-1.(pmol/l)-1 with an average interday coefficient of variation (CV) of 20.2 +/- 3.2% (range 6-44%). Glucose effectiveness (SG) was slightly less repeatable, with an average interday variability of 25.1 +/- 8.8%. The mean CVs for first-phase insulin secretion (20.1 +/- 3.5%) and total insulin secretion (21.4 +/- 3.2%) were similar to the CV for SI. Mean insulin sensitivity calculated from the 12-sample data set (0.82 +/- 0.08 x 10(-4) min-1.(pmol/l)-1 was not significantly different from the mean calculated from the full data set (P = 0.37, paired Student's t test). However, the mean CV of SI calculated from the reduced data set tended to be greater than that calculated from the full data set (27.7 +/- 5.4% vs. 20.2 +/- 3.2%).(ABSTRACT TRUNCATED AT 250 WORDS)

Dynamics of glucose production and uptake are more closely related to insulin in hindlimb lymph than in thoracic duct lymph

We previously reported a striking similarity between the dynamics of both glucose turnover and thoracic duct lymph insulin during euglycemic clamps (J Clin Invest 84:1620, 1989), which suggested that transendothelial insulin transport (TET) is rate-limiting for insulin action in vivo. Thoracic duct lymph, however, is primarily derived from insulin-insensitive tissues, which raises questions as to the physiological significance of this relationship. The relationship between glucose turnover and TET was thus examined in insulin-sensitive tissues by the simultaneous measurement of insulin in plasma, thoracic duct lymph, and hindlimb lymph during euglycemic clamps in normal anesthetized dogs (n = 8). Clamps consisted of two 3-h phases: a 0.6 mU.min-1.kg-1 insulin infusion (activation phase) followed by termination of the insulin infusion (deactivation phase). Lymph insulin was less than plasma insulin during both phases (P < 0.01) with steady-state hindlimb (120 +/- 12 pM) and thoracic duct lymph insulin (138 +/- 12 pM) 38 and 45%, respectively, lower than steady-state plasma insulin (222 +/- 24 pM) at the end of the activation phase (P < 0.05). Also, the rate of increase of lymph insulin was slower than plasma insulin during hormone infusion; half-time to steady-state was 8.8 +/- 2.0 min for plasma insulin, but longer for thoracic (25.8 +/- 3.5) and hindlimb lymph insulin (40.7 +/- 5.7 min). A very close relationship was observed during activation between the rate of increase of glucose uptake (Rd) and the increase in hindlimb lymph insulin (r2 = 0.92); this relationship was weaker for thoracic lymph (r2 = 0.74) and much weaker between glucose uptake and plasma insulin (r2 = 0.35). These data support the concept that interstitial insulin (represented by hindlimb lymph) is the signal that determines glucose uptake by insulin-sensitive tissues and that the rate of increase of glucose uptake is determined by transendothelial insulin transport into insulin-sensitive tissue. Also, during activation, hindlimb lymph insulin was a very strong predictor of the rate of suppression of hepatic glucose output (HGO) (r2 = 0.96), and the correlation with HGO was stronger than that for thoracic lymph (r2 = 0.85). The evidence that the rate of increase of Rd and the rate of suppression of HGO during insulin infusion are very strongly predicted by the time course of insulin in hindlimb lymph is consistent with the single-gateway hypothesis: the insulin transport rate across endothelium in insulin-sensitive tissue (skeletal muscle) determines the rate of glucose utilization and the suppression of hepatic glucose output.(ABSTRACT TRUNCATED AT 400 WORDS)

Quantitation of measurement error with Optimal Segments: basis for adaptive time course smoothing

We introduce a novel technique for estimating measurement error in time courses and other continuous curves. This error estimate is used to reconstruct the original (error-free) curve. The measurement error of the data is initially assumed, and the data are smoothed with "Optimal Segments" such that the smooth curve misses the data points by an average amount consistent with the assumed measurement error. Thus the differences between the smooth curve and the data points (the residuals) are tentatively assumed to represent the measurement error. This assumption is checked by testing the residuals for randomness. If the residuals are nonrandom, it is concluded that they do not resemble measurement error, and a new measurement error is assumed. This process continues reiteratively until a satisfactory (i.e., random) group of residuals is obtained. In this case the corresponding smooth curve is taken to represent the original curve. Monte Carlo simulations of selected typical situations demonstrated that this new method ("OOPSEG") estimates measurement error accurately and consistently in 30- and 15-point time courses (r = 0.91 and 0.78, respectively). Moreover, smooth curves calculated by OOPSEG were shown to accurately recreate (predict) original, error-free curves for a wide range of measurement errors (2-20%). We suggest that the ability to calculate measurement error and reconstruct the error-free shape of data curves has wide applicability in data analysis and experimental design.

Thoracic duct lymph. Relative contribution from splanchnic and muscle tissue

We have shown previously that thoracic duct lymph insulin dynamics are well correlated with tracer-determined whole-body glucose uptake and have suggested that thoracic duct lymph insulin is representative of insulin concentration in muscle interstitial fluid. However, thoracic duct lymph is comprised of interstitial fluid from all sub-thoracic tissue beds. To investigate the relative contribution of muscle interstitial fluid to total thoracic duct lymph flow, the distribution and elimination of [14C]inulin was investigated in eight experiments with conscious dogs. Both plasma and thoracic duct lymph were measured, and a three-compartment model that was hypothesized to consist of plasma, splanchnic interstitial fluid, and muscle interstitial fluid was identified. Identifications were performed with either a bolus protocol (n = 4) or an infusion protocol (n = 4), and the predicted [14C]inulin dynamics in the splanchnic and muscle interstitial fluid compartments were compared with measured values in thoracic duct lymph. Neither compartment predicted the thoracic duct [14C]inulin dynamics; however, a model based on a percentage contribution from each tissue bed fit the thoracic data well. The relative contribution of splanchnic interstitial fluid to the total thoracic duct lymph flow averaged 78 +/- 5% for the bolus protocol and 54 +/- 5% for the infusion protocol. Thus, we conclude that in the conscious animal, approximately 25-50% of thoracic duct lymph originates from muscle interstitial fluid.

Reduced sample number for calculation of insulin sensitivity and glucose effectiveness from the minimal model. Suitability for use in population studies

The FSIGT has been extensively applied to the minimal model of glucose kinetics to obtain noninvasive measures of Sl. The protocol has been modified by the addition of a bolus tolbutamide or insulin injection 20 min after glucose. Although the modified protocol has improved the Sl estimate, the method still requires a relatively large number of samples (n = 30). To reduce the total number of samples, we choose a sample schedule that minimizes the variance of the parameter estimates and the error in reconstructing the plasma insulin profile. With data from 10 subjects (BMI 30 +/- 7 kg/m2; Sl 0.9-10.2 x 10(-4) min-1.microU-1 x ml-1), a schedule consisting of 12 samples (0, 2, 4, 8, 19, 22, 30, 40, 50, 70, 90, and 180 min) was obtained. Estimates of Sl obtained from the reduced sampling schedule were then compared with those obtained with the full sampling schedule. In all 10 individuals, the Sl estimates were almost identical. A second, much larger data base consisting of 118 modified FSIGTs performed in 87 subjects (67 men, 20 women; BMI from 19.6 to 40 kg/m2 for men and 26.7 to 52.5 for women; Sl from 0.35 to 14.1 x 10(-4) min-1 x microU-1 x ml-1) was then used to independently assess the efficacy of the reduced sampling protocol. For this data base, the correlation between Sl, which was calculated from the full versus the reduced sampling schedule, was 0.95. The mean relative deviation was -1.5% (not significantly different from zero), and the SD of the relative deviation was 20.2%.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for entry of plasma insulin into cerebrospinal fluid through an intermediate compartment in dogs. Quantitative aspects and implications for transport

To study the route by which plasma insulin enters cerebrospinal fluid (CSF), the kinetics of uptake from plasma into cisternal CSF of both insulin and [14C]inulin were analyzed during intravenous infusion in anesthetized dogs. Four different mathematical models were used: three based on a two-compartment system (transport directly across the blood-CSF barrier by nonsaturable, saturable, or a combination of both mechanisms) and a fourth based on three compartments (uptake via an intermediate compartment). The kinetics of CSF uptake of [14C]inulin infused according to an "impulse" protocol were accurately accounted for only by the nonsaturable two-compartment model (determination coefficient [R2] = 0.879 +/- 0.044; mean +/- SEM; n = 5), consistent with uptake via diffusion across the blood-CSF barrier. When the same infusion protocol and model were used to analyze the kinetics of insulin uptake, the data fit (R2 = 0.671 +/- 0.037; n = 10) was significantly worse than that obtained with [14C]inulin (P = 0.02). Addition of a saturable component of uptake to the two-compartment model improved this fit, but was clearly inadequate for a subset of insulin infusion studies. In contrast, the three-compartment model accurately accounted for CSF insulin uptake in each study, regardless of infusion protocol (impulse infusion R2 = 0.947 +/- 0.026; n = 10; P less than 0.0001 vs. each two-compartment model; sustained infusion R2 = 0.981 +/- 0.003; n = 5). Thus, a model in which insulin passes through an intermediate compartment en route from plasma to CSF, as a part of a specialized transport system for the delivery of insulin to the brain, best accounts for the dynamics of this uptake process. This intermediate compartment could reside within the blood-CSF barrier or it may represent brain interstitial fluid, if CNS insulin uptake occurs preferentially across the blood-brain barrier.