Frequently Sampled Intravenous Glucose Tolerance Test Research Articles

Insulin secretion in insulin-resistant women with a history of gestational diabetes

Women with a history of gestational diabetes mellitus (GDM) tend to be insulin-resistant and hyperinsulinemic and are predisposed to the subsequent development of non—insulin-dependent diabetes mellitus (NIDDM). In the evolution of glucose intolerance, the first clinically detectable abnormality has not been defined and the relative importance of contributions of abnormal insulin secretion and insulin resistance is controversial. The present study was performed to evaluate the insulin secretory responses to oral and intravenous glucose and to mixed meals in women with a history of GDM, and to determine if the hyperinsulinemia present in these subjects is appropriate for the degree of insulin resistance. To address these questions, we studied the insulin secretory responses to oral glucose over a 3-hour period and to three mixed meals over a 24-hour period, and quantified the acute insulin response to glucose (AIR glucose) and insulin sensitivity (S 1) during frequently sampled intravenous glucose tolerance tests (FSIVGTTs). Studies were performed in seven subjects with a history of GDM and in seven matched controls. Insulin secretion rates (ISRs) were derived by deconvolution of peripheral C-peptide values using a two-compartment model and standard C-peptide kinetic parameters. Subjects with a history of GDM demonstrated (1) impairment in S 1 (2.15 ± 0.49 v 4.40 ± 0.56 × 10 −s · min −1 · (pmol/L) −1, P < .02); (2) an inappropriately low AIR glucose for the prevailing S 1 (percentile rank, 15.3% ± 7.4% v 62.6% ± 14.3%, P < .02); (3) basal glucose and insulin values that were not significantly elevated; (4) increased plasma glucose during a 24-hour meal study accompanied by increased serum insulin and ISR; and (5) normal temporal profiles of meal responses including normal ultradian insulin secretory oscillations, normal insulin clearance, and normal proinsulin to insulin molar ratio.

Influence of short-term submaximal exercise on parameters of glucose assimilation analyzed with the minimal model

After exercise, glucose uptake in tissues increases by insulin-dependent and -independent mechanisms. We evaluated whether these two effects of exercise on glucose disposal can be detected with the minimal model technique. Seven healthy volunteers were submitted at random order to two frequently sampled intravenous glucose tolerance tests (FSIVGTTs), one at rest and the other 25 minutes after a 15-minute exercise test. This exercise included 5 minutes of increasing workload on a cycloergometer followed by 10 minutes at 85% of the maximal theoretic heart rate. Bergman's minimal model of insulin action was used to analyze the two FSIVGTTs and produced the following parameters: coefficient of glucose tolerance ( K g ), ie, the slope of the exponential decrease in glycemia between 4 and 19 minutes after intravenous glucose; insulin sensitivity (S l); and glucose effectiveness at basal insulin (S g). S g was divided into its two components: basal insulin effectiveness ([BIE] S l × basal insulin) and glucose effectiveness at zero insulin ([GEZI] S g − BIE). After the exercise bout, subjects had an increased K g (3.44 ± 0.44 v 2.06 ± 0.28 × 10 −2 · min −1, P < .02), S l (11.43 ± 1.27 v 6.23 ± 0.97 × 10 −4 μU/mL · min −1, P < .01), and S g (4.40 ± 0.55 v 2.81 ± 0.36 × 10 −2 · min −1, P < .02). The increase in S g was mainly explained by a 60% increase in GEZI (3.6 ± 0.57 v 2.25 ± 0.36 × 10 −2 · min −1, P < .02), but also by an increase in BIE (0.80 ± 0.12 v 0.47 ± 0.08 × 10 −2 · min −1, P < .05). Thus, a FSIVGTT sensitively detects an acute increase in glucose assimilation after exercise, as demonstrated by an increase in K g and its two components S l and GEZI. GEZI seems to provide a measurement of the non—insulin-mediated recruitment of glucose transporters in exercised muscles. In addition, FSIVGTT protocols have to be carefully standardized for previous exercise, since minimal model measurements are sensitive to these acute effects of muscular activity.

Intravenous calcitriol normalizes insulin sensitivity in uremic patients

Recent studies suggest that secondary hyperparathyroidism and/or vitamin D deficiency are responsible for the insulin resistance in chronic renal failure. We investigated the effect of a 12-week intravenous treatment with 1,25 dihydroxycholecalciferol on glucose metabolism in 10 hemodialysis patients compared with 10 healthy control subjects by the frequently-sampled intravenous glucose tolerance test, analyzed with the minimal model technique. Compared to control subjects, the uremic patients featured elevated levels of parathyroid hormone (432 +/- 60 vs. 41 +/- 4 ng/liter, P < 0.001), insulin resistance (insulin sensitivity index, SI: 4.9 +/- 0.8 vs. 9.5 +/- 0.9 min-1/(microU/ml), P < 0.002), increased posthepatic insulin delivery (6.48 +/- 2.48 vs. 2.73 +/- 3.14 nmol/liter in 4 hr, P < 0.001) and a reduced C-peptide fractional clearance (0.033 +/- 0.004 vs. 0.085 +/- 0.009 min-1, P < 0.0002). Following treatment with 1,25 dihydroxycholecalciferol, the parathyroid hormone levels decreased significantly to 237 +/- 30 ng/liter (P < 0.05), the insulin sensitivity index (SI: 9.6 +/- 2.2, P < 0.05) reached a value similar to that of control subjects, and posthepatic insulin delivery decreased to 4.63 +/- 0.83 nmol/liter in 4 hr (P < 0.01), while all the other parameters remained unchanged. In summary, uremic patients with secondary hyperparathyroidism were found to be severely insulin resistant and hyperinsulinemic. Intravenous vitamin D treatment led to a significant reduction of parathyroid hormone levels and to a complete normalization of insulin sensitivity in the hemodialysis patients. Thus, intravenous 1,25 dihydroxycholecalciferol improves insulin resistance in uremic patients, acting per se or by reducing secondary hyperparathyroidism.

Increased levels of circulating islet amyloid polypeptide in patients with chronic renal failure have no effect on insulin secretion.

To elucidate the metabolism of islet amyloid polypeptide (IAPP) with respect to a possible renal elimination we investigated IAPP levels in 20 lean, nondiabetic patients with renal failure maintained on chronic hemodialysis (HD) and in 20 healthy controls. The basal levels of IAPP were significantly higher in uremic patients than in controls (15.1 +/- 3.2 vs. 3.2 +/- 0.2 pM, P < 0.001) suggesting renal excretion of IAPP. To investigate the impact of chronically elevated levels of endogenous IAPP on insulin secretion and insulin sensitivity, a frequently sampled intravenous glucose tolerance test (FSIGT) was performed in a subset of patients on hemodialysis and in age-matched healthy controls (C) and obese patients with normal (NGT) and with impaired glucose tolerance (IGT). Insulin sensitivity index (SI) was 8.7 +/- 1.5 in C (P < 0.05 vs. NGT, P < 0.01 vs. IGT), 5.4 +/- 0.9 in HD (P < 0.05 vs. IGT), 3.1 +/- 1.0 in NGT, and 2.0 +/- 0.5 in IGT. First phase insulin secretion was increased in patients on HD compared with those of several control groups. The results of this study therefore indicate a renal route of metabolism of IAPP. Increased endogenous circulating IAPP levels over a long period of time do not lead to a decrease in insulin release in patients on HD and do not cause the insulin resistance commonly seen in obesity and diabetes. Increased levels of circulating IAPP therefore are not likely to be a pathogenetic factor in the development of non-insulin-dependent diabetes mellitus (NIDDM).

A Comparison Between the Minimal Model and the Glucose Clamp in the Assessment of Insulin Sensitivity Across the Spectrum of Glucose Tolerance

An insulin-modified frequently sampled intravenous glucose tolerance test (FSIGTT) with minimal model analysis was compared with the glucose clamp in 11 subjects with normal glucose tolerance (NGT), 20 with impaired glucose tolerance (IGT), and 24 with non-insulin-dependent diabetes mellitus (NIDDM). The insulin sensitivity index (SI) was calculated from FSIGTT using 22- and 12-sample protocols (SI(22) and SI(12), respectively). Insulin sensitivity from the clamp was expressed as SI(clamp) and SIP(clamp). Minimal model parameters were similar when calculated with SI(22) and SI(12). SI could not be distinguished from 0 in approximately 50% of diabetic patients with either protocol. SI(22) correlated significantly with SI(clamp) in the whole group (r = 0.62), and in the NGT (r = 0.53), IGT (r = 0.48), and NIDDM (r = 0.41) groups (P < 0.05 for each). SI(12) correlated significantly with SI(clamp) in the whole group (r = 0.55, P < 0.001) and in the NGT (r = 0.53, P = 0.046) and IGT (r = 0.58, P = 0.008) but not NIDDM (r = 0.30, P = 0.085) groups. When SI(22), SI(clamp), and SIP(clamp) were expressed in the same units, SI(22) was 66 +/- 5% (mean +/- SE) and 50 +/- 8% lower than SI(clamp) and SIP(clamp), respectively. Thus, minimal model analysis of the insulin-modified FSIGTT provides estimates of insulin sensitivity that correlate significantly with those from the glucose clamp. The correlation was weaker, however, in NIDDM. The insulin-modified FSIGTT can be used as a simple test for assessment of insulin sensitivity in population studies involving nondiabetic subjects. Additional studies are needed before using this test routinely in patients with NIDDM.

Non-Insulin-Mediated Glucose Disappearance in Subjects With IDDM: Discordance Between Experimental Results and Minimal Model Analysis

Both insulin and glucose contribute to the regulation of glucose metabolism in vivo. We directly measured the ability of glucose per se to promote glucose disposal in subjects with insulin-dependent diabetes mellitus (IDDM). We compared our results with predictions of the minimal model of glucose metabolism. To identify minimal model parameters, a frequently sampled intravenous glucose tolerance test (FSIVGTT) was administered to each subject while they were connected to a Biostator (a device that monitors blood glucose and gives insulin to mimic normal insulin secretion). Data from this test reflected normal glucose tolerance and were in excellent agreement with minimal model predictions. The FSIVGTT was then repeated without the Biostator in the same diabetic subjects in order to directly measure the effect of glucose per se to promote glucose disposal in the absence of an incremental insulin effect (a basal insulin drip was maintained). To compare these results with minimal model predictions, the equations describing glucose disappearance in the absence of an incremental insulin effect were solved using parameters identified from the Biostator experiment. The glucose disappearance measured in the absence of an incremental insulin response was much slower than the minimal model predictions. Thus, the minimal model appears to overestimate the effect of glucose per se on glucose uptake and underestimate the contribution of incremental insulin.

Insulin sensitivity, insulin secretion, and glucose effectiveness in subjects with impaired glucose tolerance: A minimal model analysis

The aim of the present study was to estimate insulin secretion, insulin sensitivity (S I), and glucose effectiveness (S G) in non-obese Japanese subjects with impaired glucose tolerance (IGT). Ten IGT subjects (five men, five women) and 15 normal-tolerance subjects (seven men, eight women) without a family history of diabetes were studied. They underwent a modified frequently sampled intravenous glucose tolerance test (FSIGT); glucose (300 mg/kg body weight) was administered, and insulin (20 mU/kg over 5 minutes) was infused from 20 to 25 minutes after the administration of glucose. S I and S G were estimated by Bergman's minimal model method. No significant difference was observed in body mass index ([BMI] 22.1 ± 0.8 v 21.1 ± 0.5 kg/m 2), fasting plasma glucose (5.19 ± 0.18 v 5.07 ± 0.11 mmol/L), and insulin levels (50.7 ± 7.3 v 45.2 ± 4.5 pmol/L) of subjects with IGT and normal controls. The glucose disappearance rate (K G) was significantly lower in subjects with IGT than in normal-tolerance subjects (1.57 ± 0.20 v 2.09 ± 0.15%/min, P < .05). Pancreatic insulin secretion expressed as the integrated area of plasma insulin above the basal level during the first 20 minutes was lower in IGT subjects (2,556 ± 572 p pmol/L × min) than in normal-tolerance subjects (4,957 ± 800 pmol/L × min, P < .05). S I was not statistically different between the two groups (0.84 ± 0.13 × 10 −4 v 1.14 ± 0.15 × 10 −4 · min −1 · pmol/L −1). However, S G was significantly lower in subjects with IGT than in normal controls (0.013 ± 0.002 v 0.023 ± 0.002 min −1, P < .01). The basal insulin effect (BIE) of IGT subjects was not significantly different from that of normal subjects (0.004 ± 0.0004 v 0.005 ± 0.001 × min −1), whereas glucose effectiveness at zero insulin (GEZI) was significantly lower in IGT subjects than in normal subjects (0.010 ± 0.002 v 0.018 ± 0.002 min −1, P < .01). Thus, Japanese IGT subjects with normal S I are characterized by mild impairments in pancreatic insulin secretion and decreased S G and GEZI.

Glucose and insulin responses to intravenous glucose challenge in relatives of Nigerian patients with non-insulin-dependent diabetes mellitus

We analysed blood insulin and glucose concentrations before and during frequently sampled intravenous glucose tolerance tests (FSIGT) in 2 groups of Nigerian subjects: (A) Control group ( n = 18), without a positive family history of diabetes mellitus, and (B) Experimental group ( n = 16), comprising age-, sex- and body mass-matched first-degree relatives of patients with non-insulin-dependent diabetes mellitus (NIDDM). In comparison with Group A subjects, those in Group B had: (i) higher fasting plasma glucose level (mean ± S.E.M., 4.1 ± 0.1 vs. 3.8 ± 0.11 mmol/l, P < 0.05); (ii) similar fasting serum insulin levels (6.7 ± 5.0 vs. 5.8 ± 5.6 mU/l, P = NS); (iii) lower mean incremental area under the first-phase ( t = 0–10 min) post-glucose challenge insulin curve (376.9 ± 8.8 vs. 435.6 ± 5.6 mU/min l −1, P < 0.05); (iv) increased incremental area under the second-phase ( t = 10–182 min) post-glucose challenge insulin curve (432.9 ± 11.5 vs. 161.3 ± 8.7 mU/min l −1, P < 0.05); (v) reduced K G rate constant of glucose elimination (0.97 ± 0.12 vs. 1.41 ± 0.12%/min, P < 0.05). These results suggest that the subjects with a positive family history of NIDDM have a reduced beta-cell insulin secretory reserve (from reduced first-phase insulin response), tendency to rebound hyperinsulinemia during the latter phase of the insulin secretory response, a degree of tissue insulin insensitivity (as evident from high fasting plasma glucose despite similar insulin levels) and a diminished glucose disposal rate, in comparison with subjects without a family history of NIDDM. These features predict subsequent development of diabetes and suggest that as in Caucasians, first-degree relatives of Nigerian patients with NIDDM are at greater risk for future development of the disease.

Effect of heparin on insulin-glucose interactions measured by the minimal model technique: Implications for reproducibility using this method

The minimal model technique permits investigation of glucose and insulin interactions during a frequently sampled intravenous glucose tolerance test (FSIGT). The model allows measurement of several important variables including insulin sensitivity (S I), glucose effectiveness at basal insulin (S G), and glucose effectiveness at zero insulin (GEZI). In addition, from the FSIGT, first-phase insulin secretion (AIRglucose) and the glucose disappearance constant ( K g ) can be estimated. The estimates obtained using the minimal model technique could be of importance in understanding potential relationships between carbohydrate and lipid metabolism. In studying lipoprotein lipase (LPL) activity and mass as one of the elements of lipid metabolism, heparin administration is necessary. However, the effect of administration of this agent on factors important in carbohydrate metabolism has not been fully elucidated. Therefore, to assess the effect of heparin on insulin-glucose interactions, we performed paired studies in which subjects randomly underwent FSIGT studies 20 minutes following bolus injections of heparin or saline. None of the measured parameters was influenced by heparin administration just before the FSIGT. Since these estimates were not affected by heparin administration, these studies also permitted assessment of the coefficient of variation of replicate analysis for all values generated using the minimal model technique. The coefficients of variation of replicate analysis calculated from these two studies were 16.9% for S I, 16.6% for S G, 18.3% for GEZI, 20.6% for AIRglucose, and 14.5% for K g . These results suggest that heparin can be administered as part of a procedure that would allow concurrent assessment of various aspects of both carbohydrate and lipid metabolism. The measured variability of the different measures appears to result from biologic and laboratory factors. These coefficients of variation of replicate analysis should be considered when evaluating whether individual changes measured using the minimal model are real or might be accounted for by the variability of the experimental method.

Impaired glucose tolerance after endurance exercise is associated with reduced insulin secretion rather than altered insulin sensitivity

Paired frequently sampled intravenous glucose tolerance tests (FSIGT) were performed on five highly trained athletes within 2 hours of completing a 6-day ultramarathon run (E) and after 2 weeks of complete rest (R). Severe exercise increased free fatty acid (FFA) levels (E 1.2 ± 0.16 v 0.42 ± 0.07 mmol/L, P < .01) and norepinephrine levels (E 573 ± 141 v 224 ± 33 pg/mL, P < .01), with only moderate reductions in glucose tolerance (glucose disappearance [K g] E 1.06 ± 0.2 v R 1.7 ± 0.3 min −1 × 10 2, P < .05). The minimal model analysis of FSIGT data using the method of Bergman et al (Endocr Rev 6:45–86, 1985) showed a reduced second-phase insulin secretion ([Φ 2] E 5.2 ± 1.3 v 13 ± 2.2 μU/mL · min −2 per mg/dL, P < .05) and glucose disposition index ([SI × Φ 2] E 33.8 ± 10 v 73.9 ± 11 mg −1 · dL · min −3 × 10 4, P < .02). Insulin sensitivity (SI) and glucose-mediated glucose disposal (SG) were unchanged (SI E 6.9 ± 1.0 v 6.0 ± 0.6 min −1 per μU/mL × 10 4; SG E 1.8 ± 0.6 v 1.4 ± 0.3 min −1 × 10 2). Reduced glucose tolerance after prolonged extreme physical exercise was accompanied by reduced Φ 2 and not by alterations of SI or SG, despite the marked increase of FFA levels. Elevated norepinephrine levels, reflecting activation of the sympathetic noradrenergic system, was also associated with the reduction in K g. The reduction in Φ 2 would promote mobilization of FFA, the predominant metabolic substrate in these endurance events.

Effect of dexamethasone on insulin sensitivity, islet amyloid polypeptide and insulin secretion in humans.

The response of islet amyloid polypeptide and insulin and their molar ratios were investigated in eight healthy volunteers before and after treatment with dexamethasone by oral and frequently-sampled intravenous glucose tolerance tests. Following dexamethasone treatment the insulin sensitivity index decreased significantly from 6.5 +/- 1.3 to 4.1 +/- 1.0 (microU.ml-1).min-1, p < 0.05. The area under the curve representing above-basal levels of insulin during oral glucose tolerance test increased significantly following dexamethasone treatment from 48132 +/- 9736 to 82230 +/- 14846 pmol.l-1 x 3 h-1, p < 0.05, the area under the curve of islet amyloid polypeptide increased from 1308 +/- 183 to 2448 +/- 501 pmol.l-1 x 3 h-1, p < 0.05. The overall insulin/islet amyloid polypeptide molar ratios calculated from the area under the curve during the 3-h period of the oral glucose tolerance test was not significantly different before and after dexamethasone treatment (42 +/- 5 vs 40 +/- 4). During the oral glucose tolerance test the insulin/islet amyloid polypeptide ratio increased significantly from baseline to 30 min (p < 0.05), then declined towards initial values before and after dexamethasone treatment. In conclusion, dexamethasone induced a significant decrease in insulin sensitivity and a significant increase in insulin secretion during the oral glucose tolerance test. However, in contrast to previous animal experiments we did not find a change in the insulin/islet amyloid polypeptide ratio before and after dexamethasone treatment.

Insulin action and metabolism in patients with head and neck cancer

Cachexia is a phenomenon commonly observed in patients with cancer, but its etiology is poorly understood. Abnormalities of insulin action and metabolism that have been hypothesized to promote cancer cachexia were investigated in this study using a computerized euglycemic clamp and modified frequently sampled intravenous glucose tolerance test (FSIGT) in a group of malnourished patients with localized head and neck cancer, and in healthy, well-nourished, age- and, sex-matched controls. Glucose disposal rates, determined by the euglycemic clamp at three different rates of insulin infusion did not differ significantly between the two groups. However, mean plasma insulin concentrations at each level of insulin infused were significantly lower in the patients with cancer than in the control subjects resulting in unexpectedly higher calculated insulin metabolic clearance rate in the patients with cancer. Peripheral insulin sensitivity calculated from the slope of glucose disposal versus plasma insulin concentration did not differ between the two groups. Results from the FSIGT demonstrated no difference in insulin sensitivity or insulin-independent glucose disappearance between the two groups. However, whole body glucose appearance was significantly elevated in the patients with cancer. Thus, increased whole body glucose utilization in the absence of insulin resistance or increased insulin-dependent glucose disposal was observed. These data are consistent with the concept of a localized tumor acting as a glucose drain in which case increased glucose appearance and increased insulin clearance would defend against hypoglycemia.

A modified minimal model analysis of insulin sensitivity and glucose-mediated glucose disposal in insulin-dependent diabetes

Although glucose utilization is impaired in insulin-dependent diabetes mellitus (IDDM), it is unclear whether this is due to reductions in insulin sensitivity (S I) and/or glucose-mediated glucose disposal (S G). The minimal model of Bergman et al can be applied to a frequently sampled intravenous glucose tolerance test (FSIGT) to simultaneously estimate S I and S G, but cannot accommodate data from diabetics. Exogenous insulin approximating the normal pattern of insulin secretion was infused during FSIGTs in eight young non-obese C-peptide-negative IDDM subjects, but with the total dose modified to achieve sufficient glucose disappearance rates (K G) to allow analysis of data. The minimal model was modified to model the effects of the exogenous insulin on glucose kinetics to estimate S I and S G. Despite deliberately achieving supranormal plasma-free insulin levels during the FSIGT (“first-phase insulin” = 62 ± 9 SE mU/L; “second phase insulin” = 34 ± 9 mU/L), the diabetics showed low-normal K G values (1.3 ± 0.29 min −1 × 10 2). Using the model, good parameter resolution (fractional SD [FSD] < .5) was achieved (IDDM v controls: S I = 2.5 ± 0.6 v 8.3 ± 1.5 min −1 · mU −1 · L −1 × 10 4; SG = 1.6 ± 0.5 v 2.6 ± 0.2 min −1 × 10 2; P < .05). This reduction in S G was confirmed in the same IDDM subjects by FSIGT during basal insulin infusion only (S G = 1.0 ± 0.3 min −1 × 10 2). The reduction in S I was confirmed by our previous euglycemic clamp studies in a similar IDDM group, where S I = 4.2 ± 1.0 min −1 · mU −1 · L −1. It is feasible to use this modification of the Bergman minimal model to perform a simultaneous estimation of S I and S G in IDDM. Reductions in both S I and S G contribute to the impairment of glucose utilization seen in IDDM subjects treated by standard insulin regimens.

Reduced Beta‐Cell Secretion and Insulin Hepatic Extraction in Healthy Elderly Subjects

One factor responsible for the altered carbohydrate metabolism in elderly subjects is impaired insulin release; however, difficulties in directly measuring insulin secretion have limited studies on pancreatic activity and on the contribution of the liver to insulin delivery. This study investigated beta-cell performance and insulin hepatic extraction under dynamic conditions in normal elderly subjects. Two strictly comparable groups of 12 young controls (Y, 27 +/- 1 (SE) years, 73 +/- 3 kg) and 12 elderly men (E, 69 +/- 2 years, 73 +/- 3 kg) were chosen on the basis of normal OGTT and normal insulin sensitivity in order to investigate a "pure" age effect. The subjects underwent a 4-hour frequently sampled intravenous glucose tolerance test (FSIGT) (dose 0.3 g/kg). Although no significant differences were found between the fasting levels of glucose and insulin (respectively: E: 89 +/- 3 mg/dL versus Y: 87 +/- 2, P greater than .1; and E: 5.0 +/- 0.5 microU/mL versus Y: 6.8 +/- 1.0, P greater than .05), basal C-peptide was found to be lower in the old subjects: 0.43 +/- 0.06 ng/mL versus 0.70 +/- 0.11 (P less than .025). The patterns of glucose and insulin during the FSIGT were similar, whereas C-peptide concentration in E was systematically lower, suggesting a reduced insulin secretion. To verify this hypothesis, we analyzed FSIGT data with a mathematical model-based method that provides a noninvasive direct measurement of the time courses of insulin secretion and hepatic extraction.(ABSTRACT TRUNCATED AT 250 WORDS)

Alternative method for minimal model analysis of intravenous glucose tolerance data.

The minimal models of glucose-insulin kinetics were used to analyze sets of data obtained from human subjects and dogs during frequently sampled intravenous glucose tolerance tests (FSIGTs). Analysis of some data sets from both species resulted in poor identification of parameters. To improve the parameter resolution, the information base on which the parameters are estimated was enlarged. This was accomplished by incorporating into the analysis 1) glucose data obtained between 0 and 8 min of the FSIGT and some of the insulin data obtained prior to the insulin peak and 2) a second set of FSIGT data for each individual obtained during a physiological perturbation. As a result, data analysis was considerably enhanced, with parameter fractional standard deviation being routinely reduced to less than 0.5. Analysis of stimulated data with noise levels for glucose and insulin set between 0.05 and 0.15 confirmed the improvement in parameter estimates. This modified approach to analysis of FSIGTs therefore consistently leads to well-defined kinetic descriptions of experimental data in various situations and supports the usefulness of the minimal model in examining the complex interplay between the parameters that influence overall glucose tolerance.

Modified protocols improve insulin sensitivity estimation using the minimal model.

We attempted to improve the precision of the estimation of insulin sensitivity (S1) from the minimal model technique by modifying insulin dynamics during a frequently sampled intravenous glucose tolerance test (FSIGT). Tolbutamide and somatostatin (SRIF) were used to change the insulin dynamics without directly affecting insulin sensitivity. Injection of tolbutamide (100 mg) at t = 20 min provoked an immediate secondary peak in insulin response, resulting in a greater integrated incremental insulin than the standard FSIGT. SRIF, injected at t = -1 min, delayed insulin secretion in proportion to the dose without any change in magnitude. Computer simulation was used to assess the precision of S1 estimation. Insulin dynamics from both standard and modified protocols were adjusted in magnitude, with the shape unchanged and analyzed to determine the effect of the magnitude of insulin response. Fractional standard deviation was reduced from 73% with the standard insulin profile to 23% with tolbutamide and 18% with the highest dose of SRIF. In addition, the fractional standard deviation of S1 estimates decreased exponentially with increasing magnitude of insulin response. Modified FSIGTs require a smaller insulin response than the standard protocol to achieve the same precision.

MINMOD: a computer program to calculate insulin sensitivity and pancreatic responsivity from the frequently sampled intravenous glucose tolerance test

Insulin sensitivity and pancreatic responsivity are the two main factors controlling glucose tolerance. We have proposed a method for measuring these two factors, using computer analysis of a frequently-sampled intravenous glucose tolerance test (FSIGT). This ‘minimal modelling approach’ fits two mathematical models with FSIGT glucose and insulin data: one of glucose disappearance and one of insulin kinetics. MINMOD is the computer program which identifies the model parameters for each individual. A nonlinear least squares estimation technique is used, employing a gradient-type of estimation algorithm, and the first derivatives (not known analitically) are computed according to the ‘sensitivity approach’. The program yields the parameter estimates and the precision of their estimation. From the model parameters, it is possible to extract four indices: (1) S G , the ability of glucose per se to enhance its own disappearance at basal insulin, (2) S 1, the tissue insulin sensitivity index, (3) φ 1, first phase pancreatic responsitivity, and (4) φ 2, second phase pancreatic responsitivity. These four characteristic parameters have been shown to represent an integrated metabolic portrait of a single individual.