Kerstin Rebrin

Feasibility of Automating Insulin Delivery for the Treatment of Type 1 Diabetes

An automated closed-loop insulin delivery system based on subcutaneous glucose sensing and subcutaneous insulin delivery was evaluated in 10 subjects with type 1 diabetes (2 men, 8 women, mean [±SD] age 43.4 ± 11.4 years, duration of diabetes 18.2 ± 13.5 years). Closed-loop control was assessed over ∼30 h and compared with open-loop control assessed over 3 days. Closed-loop insulin delivery was calculated using a model of the β-cell’s multiphasic insulin response to glucose. Plasma glucose was 160 ± 66 mg/dl at the start of closed loop and was thereafter reduced to 71 ± 19 by 1:00 p.m. (preprandial lunch). Fasting glucose the subsequent morning on closed loop was not different from target (124 ± 25 vs. 120 mg/dl, respectively; P > 0.05). Mean glucose levels were not different between the open and closed loop (133 ± 63 vs. 133 ± 52 mg/dl, respectively; P > 0.65). However, glucose was within the range 70–180 mg/dl 75% of the time under closed loop versus 63% for open loop. Incidence of biochemical hypoglycemia (blood glucose <60 mg/dl) was similar under the two treatments. There were no episodes of severe hypoglycemia. The data provide proof of concept that glycemic control can be achieved by a completely automated external closed-loop insulin delivery system.

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 ɑ-cell. To examine whether insulins extrahepatic site is independent of the ɑ-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 ɑ-cell, that in turn suppresses hepatic glucose production. To investigate the nature of this signal, we measured alanine, lactate, and free fatty acids (FFAs). There was no clear relationship between alanine or lactate and HGO suppression; however, there was an extremely strong relationship between plasma FFAs and HGO both at steady state and during dynamic changes in insulin. These data suggest, but do not prove, that insulin acts to suppress HGO as follows: Insulin slowly traverses the capillary endothelium in adipose tissue; elevated insulin in adipose tissue interstdtium inhibits lipolysis, thus decreasing FFA levels; and decreased FFAs act as a signal to the liver to suppress endogenous glucose production.

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 insulins 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.

Determination of Plasma Glucose During Rapid Glucose Excursions with a Subcutaneous Glucose Sensor

Continuous glucose monitoring has the potential to improve glucose management and reduce the risk of hypoglycemia in individuals with diabetes. Accurate sensors may also allow the development of a closed-loop insulin delivery system. The purpose of this work was to determine the delay time associated with a subcutaneous glucose sensor during rapidly changing glucose excursions. Subcutaneous glucose sensors (Medtronic MiniMed, Inc., Northridge, CA) were inserted in five healthy men. After a 2-h stabilization period, a 3-h hyperglycemic (approximately 11 mM) clamp was performed followed by a 90-min period in which plasma glucose was allowed to decline to as low as 2.8 mM. Sensors were calibrated using two points (basal and hyperglycemia), and the calibrated sensor glucose measurements were compared with those from a reference analyzer (Beckman Instruments, Fullerton, CA). Response time was estimated from a first-order kinetic model. Plasma glucose levels, determined with the subcutaneous sensor, were highly correlated with those obtained with the reference glucose analyzer (r(2) = 0.91, p < 0.001; mean absolute difference of approximately 8%). The half-time for the sensor response was estimated to be 4.0 +/- 1.0 min. The subcutaneous glucose sensor has the potential to facilitate the detection of hypoglycemia and improve overall glycemic control when used in a real-time monitor. The rapid response should be sufficient to allow a fully automated closed-loop insulin delivery system to be developed based on the subcutaneous sensing site.

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.

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.

Modeling β-Cell Insulin Secretion - Implications for Closed-Loop Glucose Homeostasis

Glucose sensing and insulin delivery technology can potentially be linked to form a closed-loop insulin delivery system. Ideally, such a system would establish normal physiologic glucose profiles. To this end, a model of beta-cell secretion can potentially provide insight into the preferred structure of the insulin delivery algorithm. Two secretion models were evaluated for their ability to describe plasma insulin dynamics during hyperglycemic clamps (humans; n=7), and for their ability to establish and maintain fasting euglycemia under conditions simulated by the minimal model. The first beta-cell model (SD) characterized insulin secretion as a static component that had a delayed response to glucose, and a dynamic component that responded to the rate of increase of glucose. The second model (PID) described the response in terms of a proportional component without delay, an integral component that adjusted basal delivery in proportion to hyper/hypoglycemia, and a derivative component that responded to the rate of glucose change. Both models fit the beta-cell response during the clamp, and established fasting euglycemia under simulated closed-loop conditions; however, the SD model did not maintain euglycemia following simulated changes in insulin sensitivity or glucose appearance, whereas the PID model did. The PID model was more stable under the simulated closed-loop conditions. Both the SD and PID models described beta-cell secretion in response to a rapid increase in glucose. However, the PID model could maintain fasting euglycemia and was more stable under closed-loop conditions, and thus is more suited for such conditions.

The role of the independent variable to glucose sensor calibration.

In vivo subcutaneous glucose sensor accuracy depends on the calibration method. Sensor accuracy was assessed during standard oral glucose tolerance tests in six non-diabetic subjects each wearing six subcutaneous glucose sensors (Medtronic MiniMed). Paired blood glucose (B(G)) and sensor current readings were used for retrospective sensor calibration using either B(G) or sensor current as the independent variable. Sensor accuracy after calibration was assessed using three criteria: linear regression between B(G) and sensor glucose (S(G)); correlation; and mean absolute difference (MAD), defined as 100 x |B(G) - S(G)|/B(G). Calibration with B(G) as the independent variable resulted in unbiased estimates of regression slope (1.02, not different than 1, p< 0.01) and y-intercept (-1.06 mg/dL, not different than 0, p< 0.01). In contrast, calibration with sensor current as the independent variable resulted in biased estimates of slope (0.76, different than 1, p< 0.01) and y-intercept (31.25 mg/dL, different than 0, p< 0.01). However, with sensor current as the independent variable, the MAD was lower than the corresponding value for calibration with B(G) at the x-axis (15.00 +/- 0.47% vs. 18.35 +/- 0.63%, p< 0.01). The Pearson correlation coefficient between B(G) and S(G) was higher when using sensor current as the independent variable (R = 0.82 vs. R = 0.79 when using glucose on the x-axis). We suggest that despite the fact that calibration with sensor current as the independent variable leads to a bias in the estimate of B(G), it is a more appropriate calibration method when the primary concern is minimization of the MAD between S(G) and B(G).

Use of Subcutaneous Interstitial Fluid Glucose to Estimate Blood Glucose: Revisiting Delay and Sensor Offset:

Background: Estimates for delays in the interstitial fluid (ISF) glucose response to changes in blood glucose (BG) differ substantially among research groups. We review these findings along with arguments that continuous glucose monitoring (CGM) devices used to measure ISF delay contribute to the variability. We consider the impact of the ISF delay and review approaches to correct for it, including strategies pursued by the manufacturers of these devices. The focus on how the manufacturers have approached the problem is motivated by the observation that clinicians and researchers are often unaware of how the existing CGM devices process the ISF glucose signal. Methods: Numerous models and simulations were used to illustrate problems related to measurement and correction of ISF glucose delay. Results: We find that (1) there is no evidence that the true physiologic ISF glucose delay is longer than 5–10 min and that the values longer than this can be explained by delays in CGM filtering routines; (2) the primary impact of the true ISF delay is on sensor calibration algorithms, making it difficult to estimate calibration factors and offset (OS) currents; (3) inaccurate estimates of the sensor OS current result in overestimation of sensor glucose at low values, making it difficult to detect hypoglycemia; (4) many device companies introduce nonlinear components into their filters, which can be expected to confound attempts by investigators to reconstruct BG using linear deconvolution; and (5) algorithms advocated by academic groups are seldom compared to algorithms pursued by industry, making it difficult to ascertain their value. Conclusions: The absence of any direct comparisons between existing and new algorithms for correcting ISF delay and sensor OS current is, in part, due to the difficulty in extracting relevant details from industry patents and/or extracting unfiltered sensor signals from industry products. The model simulation environment, where all aspects of the signal can be derived, may be more appropriate for developing new filtering and calibration strategies. Nevertheless, clinicians, academic researchers, and the industry would benefit from collaborating when evaluating those strategies.

Evaluation of the Effect of Gain on the Meal Response of an Automated Closed-Loop Insulin Delivery System

A continuous closed-loop insulin delivery system using subcutaneous insulin delivery was evaluated in eight diabetic canines. Continuous glucose profiles were obtained by extrapolation of blood glucose measurements. Insulin delivery rate was calculated, using a model of β-cell insulin secretion, and delivered with a Medtronic MiniMed subcutaneous infusion pump. The model acts like a classic proportional-integral-derivative controller, delivering insulin in proportion to glucose above target, history of past glucose values, and glucose rate of change. For each dog, a proportional gain was set relative to the open-loop total daily dose (TDD) of insulin. Additional gains based on 0.5 × TDD and 1.5 × TDD were also evaluated (gain dose response). Control was initiated 4 h before the meal with a target of 6.7 mmol/l. At the time of the meal, glucose was similar for all three gains (6.0 ± 0.3, 5.2 ± 0.3, and 4.9 ± 0.5 mmol/l for 0.5 × TDD, TDD, and 1.5 × TDD, respectively; P > 0.05) with near-target values restored at the end of experiments (8.2 ± 0.9, 6.0 ± 0.6, and 6.0 ± 0.5, respectively). The peak postprandial glucose level decreased significantly with increasing gain (12.1 ± 0.6, 9.6 ± 1.0, and 8.5 ± 0.6 mmol/l, respectively; P < 0.05). The data demonstrate that closed-loop insulin delivery using the subcutaneous site can provide stable glycemic control within a range of gain.