Search

Beta-cell function in growth hormone (GH)-deficient (GHD) adults is poorly documented. Beta-cell function was therefore studied in 10 GHD adults (age, 40+/-3 years; weight, 79.3+/-4.8 kg; body mass index [BMI], 27.5+/-1.3 kg x m(-2)) before and after 6- and 24-month recombinant human GH (rhGH) therapy (0.24 IU x kg(-1) x wk(-1)) compared with 10 age-, sex-, weight-, and BMI-matched control subjects. With rhGH therapy, fat-free mass (FFM) increased (48.2+/-4.9, 52.5+/-4.8, and 59+/-6.8 kg, respectively) and fat mass (FM) decreased (33.8%+/-2.8%, 28.0%+/-3.0%, and 29.4%+/-2.5%, respectively), as did serum cholesterol. Oral glucose tolerance initially deteriorated at 6 months, but improved toward the control value by 24 months. Fasting insulin (FI) increased significantly, as did the acute insulin response to oral glucose (deltaAIR(OGTT)/deltaG) at 30 minutes (FI: pretreatment 9.8+/-0.8, 6 months, 14.0+/-1.8, 24 months 12.5+/-1.6 v control 11.4+/-1.9 mU x L(-1); deltaAIR(OGTT)/deltaG: pretreatment 201+/-24, 6 months 356+/-41, 24 months 382+/-86 v control 280+/-47 mU x mmol(-1)). However, the acute insulin response to intravenous (IV) glucose (AIR(G)) and IV glucagon at euglycemia and hyperglycemia did not change with rhGH therapy and were similar to the control group values. Importantly, the expected reciprocal relationships (as observed for the control group) between the various insulin secretory parameters and insulin sensitivity (SI) either were not present or were statistically weak in GHD subjects, despite the 35% decrease in SI by 24 months of rhGH therapy. In particular, over time, there was an attenuation of insulin secretion with respect to the ongoing insulin resistance with rhGH therapy, particularly for AIR(G) at 24 months. After 5 days of rhGH withdrawal, insulin secretion decreased and SI improved in GHD subjects. It is concluded that the current long-term rhGH treatment regimens appear to impact on insulin secretion such that the normal relationships between insulin secretion and SI are altered despite the favorable impact on body composition and serum lipid profiles.

We tested the hypothesis that glucose intolerance develops in genetically prone subjects when exogenous insulin resistance is induced by dexamethasone (dex) and investigated whether the steroid-induced glucose intolerance is due to impairment of beta-cell function alone and/or insulin resistance. Oral glucose tolerance (OGTT) and intravenous glucose tolerance tests with minimal model analysis were performed before and following 5 days of dex treatment (4 mg/day) in 20 relatives of non-insulin-dependent diabetic (NIDDM) patients and in 20 matched control subjects (age: 29.6 +/- 1.7 vs 29.6 +/- 1.6 years, BMI: 25.1 +/- 1.0 vs 25.1 +/- 0.9 kg/m2). Before dex, glucose tolerance was similar in both groups (2-h plasma glucose concentration (PG): 5.5 +/- 0.2 [range: 3.2-7.0] vs 5.5 +/- 0.2 [3.7-7.4] mmol/l). Although insulin sensitivity (Si) was significantly lower in the relatives before dex, insulin sensitivity was reduced to a similar level during dex in both the relatives and control subjects (0.30 +/- 0.04 vs 0.34 +/- 0.04 10(-4) min(-1) per pmol/l, NS). During dex, the variation in the OGTT 2-h PG was greater in the relatives (8.5 +/- 0.7 [3.9-17.0] vs 7.5 +/- 0.3 [5.7-9.8] mmol/l, F-test p < 0.05) which, by inspection of the data, was caused by seven relatives with a higher PG than the maximal value seen in the control subjects (9.8 mmol/l). These "hyperglycaemic" relatives had diminished first phase insulin secretion (O1) both before and during dex compared with the "normal" relatives and the control subjects (pre-dex O1: 12.6 +/- 3.6 vs 26.4 +/- 4.2 and 24.6 +/- 3.6 (p < 0.05), post-dex O1: 22.2 +/- 6.6 vs 48.0 +/- 7.2 and 46.2 +/- 6.6 respectively (p < 0.05) pmol x l(-1) x min(-1) per mg/dl). However, Si was similar in "hyperglycaemic" and "normal" relatives before dex (0.65 +/- 0.10 vs 0.54 +/- 0.10 10(-4) x min(-1) per pmol/l) and suppressed similarly during dex (0.30 +/- 0.07 vs 0.30 +/- 0.06 10(-4) x min(-1) per pmol/l). Multiple regression analysis confirmed the unique importance of low pre-dex beta-cell function to subsequent development of high 2-h post-dex OGTT plasma glucose levels (R2 = 0.56). In conclusion, exogenous induced insulin resistance by dex will induce impaired or diabetic glucose tolerance in those genetic relatives of NIDDM patients who have impaired beta-cell function (retrospectively) prior to dex exposure. These subjects are therefore unable to enhance their beta-cell response in order to match the dex-induced insulin resistant state.

Fourteen GH-deficient (GHD) adults were compared with 12 age-, sex-, and body mass index-matched control subjects using a baseline tritiated glucose equilibration period and euglycemic-hyperinsulinemic (approximately 55 mU/L) clamp in conjunction with paired muscle biopsies for measurement of glycogen synthase fractional velocity (FV0.1). Despite similar basal rates of total glucose disposal (Rd), there was a 64% reduction in the insulin-stimulated rise (delta) in Rd in the GHD adults compared to that in controls [16.6 +/- 2.8 vs. 44.7 +/- 6.0 mumol/kg fat free mass (FFM)/min; P0.001], which was mainly due to a decreased glucose storage (GS) rate (delta GS, 12.6 +/- 2.9 vs. 39.5 +/- 7.5 mumol/kg FFM/min; P0.01). Furthermore, the insulin sensitivity indexes of Rd (0.39 +/- 0.07 vs. 0.85 +/- 0.11; P0.05) and GS (0.25 +/- 0.07 vs. 0.72 +/- 0.13 mumol/kg FFM/min per mU/L; P0.02) were reduced in GHD adults compared to the control values. The insulin sensitivity of the glycolytic pathway was also reduced by approximately 50% in GHD adults (P = 0.07 vs. controls). Insulin-stimulated FV0.1 was decreased in GHD adults (0.31 +/- 0.02 vs. 0.47 +/- 0.03; P0.005) despite similar basal FV0.1. Using multiple and stepwise regression analysis, duration of GH deficiency, fasting triglycerides and fasting insulin accounted for 67% of the variance in the insulin sensitivity index of Rd. In conclusion, the severe insulin resistance in GHD adults is mainly due to the inhibition of the GS pathway and glycogen synthase activity in peripheral tissues, which is related to the duration of GH deficiency, fasting triglycerides, and fasting insulin.

We determined the impact of variable insulinemia and glycemia on the in vivo partitioning of glucose effectiveness (GE) and insulin sensitivity (SI) and the in vitro intracellular processing of glucose metabolism. Six somatostatin- and [3-3H]glucose-infused dogs underwent euglycemic and hyperglycemic clamps at four physiological insulin (Ins) levels before a muscle biopsy. From the rates of glucose infusion (GINF), total glucose disposal (Rd), total glycolysis (GF), and glucose storage (GS), plots of delta GINF, delta Rd, delta GS vs. delta log Ins concentration were found to be linear for each dog, allowing calculation of the partitioning of GE and SI into their major in vivo sites (periphery vs. liver) and intracellular metabolic pathways (GS vs. GF). Insulinopenia induced a significant reduction in total GE. From insulinopenia to high insulinemia, the 2.3-fold increase in total GE was due to the increased peripheral glucose responsiveness of the GS pathway. Hyperglycemia induced a significant reduction in total SI, with approximately one-half of this reduction due to the decreased peripheral insulin responsiveness of the GF pathway. In skeletal muscle, both glycogen content and glycogen synthase fractional activity were positively correlated with log Ins concentration, Rd, and GS but negatively correlated with glucose 6-phosphate concentration. Moreover, both Rd and GS were negatively correlated with lactate concentration. We conclude that 1) the inhibition of GE and SI induced by insulinopenia and hyperglycemia, respectively, is due mainly to the reduced peripheral responsiveness of contrasting intracellular metabolic pathways; and 2) hyperinsulinemia and/or hyperglycemia stimulates glycogen synthesis and GF but not nonoxidative glycolysis.

The effect of 3 months of low dose (120 micrograms/kg.week or 0.24 IU/kg.week) recombinant human GH (rhGH) treatment on glucose tolerance, insulin secretion, and insulin- and glucose-mediated glucose disposal was examined in 10 GH-deficient adults. The frequently sampled iv glucose tolerance test was performed at baseline and after 1 week and 3 months of rhGH therapy and analyzed by the minimal model method of Bergman to provide estimates of the glucose decay rate, first and second phase insulin secretion (phi 1 and phi 2), fractional clearance of insulin, and glucose-mediated and insulin-mediated glucose disposal. Fasting glucose, insulin, C-peptide, nonesterified fatty acids (NEFA), and serum cholesterol and triglycerides were also measured. When the 1 week data were compared to baseline, there was a small but significant rise in mean (+/- SE) fasting glucose (4.62 +/- 0.17 vs. 5.1 +/- 0.15 mmol/L; P < 0.01), NEFA (0.70 +/- 0.09 vs. 1.1 +/- 0.12 mmol/L; P < 0.005), insulin (93.6 +/- 8.9 vs. 238.9 +/- 9.2 pmol/L; P < 0.0001), C-peptide (0.32 +/- 0.13 vs. 0.66 +/- 0.13 nmol/L; P < 0.005), and phi 1 (11.9 +/- 1.3 vs. 16.2 +/- 1.8 pmol/L.min/mmol.L x 10(2)) and phi 2 (1.43 +/- 0.17 vs. 3.15 +/- 0.25 pmol/L.min/mmol.L x 10(3); P < 0.05). Conversely, there were associated decreases in glucose decay rate (1.83 +/- 0.26 vs. 1.28 +/- 0.12 min-1; P < 0.05) and insulin-mediated glucose disposal (0.36 +/- 0.08 vs. 0.18 +/- 0.06 min/pmol.L x 10(-4); P < 0.005). There was no change in glucose-mediated glucose disposal or the fractional clearance of insulin. By 3 months, fasting insulin and C-peptide levels remained significantly elevated, whereas other parameters had returned to baseline. There was a minor reduction in serum cholesterol at 1 week (5.1 +/- 0.15 vs. 4.62 +/- 0.17 mmol/L; P < 0.01), which was not maintained at 3 months. Serum triglycerides remained unchanged throughout the study. We conclude that short term low dose rhGH treatment of GH-deficient adults induces a temporary state of mild glucose intolerance, hyperinsulinemia, insulin resistance, and raised NEFA levels at 1 week. By 3 months, these metabolic disturbances had returned to baseline for a persisting modest hyperinsulinemia. Whether this hyperinsulinemia will last over the longer term and/or has distant detrimental metabolic consequences in the individual must await further studies.

20 normoglycemic first degree relatives of non-insulin-dependent diabetes mellitus (NIDDM) patients were compared with 20 matched subjects without any family history of diabetes using the intravenous glucose tolerance test with minimal model analysis of glucose disappearance and insulin kinetics. Intravenous glucose tolerance index (Kg) was similar in both groups (1.60 +/- 0.14 vs 1.59 +/- 0.18, x 10(-2) min-1, NS). However, insulin sensitivity (Si) was reduced (3.49 +/- 0.43 vs 4.80 +/- 0.61, x 10(-4) min-1 per mU/liter, P = 0.05), whereas glucose effectiveness (Sg) was increased (1.93 +/- 0.14 vs 1.52 +/- 0.16, x 10(-2) min-1, P < 0.05) in the relatives. Despite insulin resistance neither fasting plasma insulin concentration (7.63 +/- 0.48 vs 6.88 +/- 0.45, mU/liter, NS) nor first phase insulin responsiveness (Phi1) (3.56 +/- 0.53 vs 4.13 +/- 0.62, mU/liter min-1 per mg/dl, NS) were increased in the relatives. Phi1 was reduced for the degree of insulin resistance in the relatives so that the Phi1 x Si index was lower in the relatives (11.5 +/- 2.2 vs 16.7 +/- 2.0, x 10(-4) min-2 per mg/dl, P < 0.05). Importantly, glucose effectiveness correlated with Kg and with basal glucose oxidation but not with total glucose transporter 4 (GLUT4) content in a basal muscle biopsy. In conclusion we confirm the presence of insulin resistance in first degree relatives of NIDDM patients. However, insulin secretion was altered and reduced for the degree of insulin resistance in the relatives, whereas glucose effectiveness was increased. We hypothesize that increased glucose effectiveness maintains glucose tolerance within normal limits in these "normoinsulinemic" relatives of NIDDM patients.

To determine the impact of variable plasma insulin concentrations and glycemia on the partitioning of whole body glucose metabolism between glycolysis and glucose storage, we estimated endogenous hepatic glucose production and rates of in vivo glycolytic flux (GF) and glucose storage (GS) in six normal dogs from the generation of plasma tritiated water (3H2O) and [3-3H]glucose specific activity during 150 min of somatostatin euglycemic (E) and hyperglycemic (H) clamps at hypoinsulinemic, basal, intermediate, and high insulin levels. During both E and H clamps, overall rates of GF and GS increased with the rising insulin levels, but the relative contributions to in vivo glucose disposal of GF decreased, whereas GS rose progressively with increasing insulin levels. The relative contribution of GS during H to overall glucose disposal was greater at the lower insulin level. In addition, in absolute terms, GF and GS were significantly higher (P < 0.05) during H than during E at all insulin levels. Moreover, the incremental rise in GF induced by H was equal for the low to intermediate insulin levels tested, independent of the prevailing free fatty acid (FFA) levels. However, when whole body glucose disposal rates were matched, GF and GS rates were independent of the coexisting glycemia, insulin, and/or FFA levels. We conclude that 1) insulin has a major impact on the intracellular fate of infused glucose, with a lesser but significant effect of hyperglycemia per se on these processes; 2) the magnitude of the hyperglycemia-induced increase in GF is independent of the prevailing insulin level from low to intermediate levels; and 3) in vivo GF and GS are dependent on the net rate of glucose uptake into cells but independent of absolute FFA levels or whether glucose uptake is stimulated by raised insulin or glucose levels.

The aim of this study was to determine the relative roles of changes in glucose-mediated glucose disposal (SG) and insulin sensitivity (SI) on the impairment of glucose disposal caused by epinephrine (EPI) infusion in type I (insulin-dependent) diabetes mellitus (IDDM). Seven non-obese young adult diabetics with minimal endogenous insulin secretion had EPI infusions at 25 ng/kg/min for 5.5 hours, after a basal overnight insulin infusion (12 mU/kg/h), and glucose infusion as required to maintain euglycemia. The EPI infusion produced approximately an eightfold increase in plasma EPI. At 2.5 hours, an intravenous glucose tolerance test (IVGTT) was performed with supplemental exogenous insulin infusion to achieve an approximation of normal endogenous insulin secretion. In random order, each subject also had a control (CTR) infusion of basal insulin before the IVGTT. The results were analyzed according to a modification of the minimal model of Bergman et al. EPI infusion was associated with (1) elevated basal plasma glucose (EPI v CTR, 9.8 +/- 0.3 SE v 7.7 +/- 0.7 mmol/L, P less than .05); (2) elevated plasma nonesterified fatty acids (NEFA, 0.9 +/- 0.1 v 0.3 +/- 0.1 mmol/L, P less than .05); and (3) profoundly reduced glucose disposal (KG 0.59 +/- 0.1 v 1.91 +/- 0.33 min-1 x 10(2), P less than .02). Further analysis showed that the reduced glucose disposal was attributable to a marked decrease in SI (EPI 0.9 +/- 0.5 v CTR 7.03 +/- 3.2 min-1.mU-1.L x 10(4), P less than .05) with no significant change in SG (EPI 2.5 +/- 0.2 v CTR 3.1 +/- 0.5 min-1 x 10(2), NS).(ABSTRACT TRUNCATED AT 250 WORDS)

The minimal model approach to analysis of intravenous glucose tolerance tests (IVGTT) yields estimates of parameters representing insulin sensitivity, glucose-mediated glucose disposal and pancreatic responsiveness. The precision of these estimates can deteriorate if the glucose and insulin data lack well-defined structure or freedom from data noise (random error). The precision of parameter estimates can be enhanced if data sets from two or more IVGTTs, obtained under different experimental conditions in the same subject, are analysed together in one data file. Following initial fitting using CONSAM, the conversational version of the modeling program SAAM, those parameters whose estimates remain at the same value under the different experimental conditions are constrained. This effectively reduces the number of adjustable parameters, and their estimates can then be fine-tuned with enhanced precision using the batch version of SAAM.

Six obese, non-insulin-dependent diabetic subjects were studied before and 3 mo after treatment with the sulfonylurea gliclazide, 40-80 mg b.i.d. Fasting plasma glucose fell significantly from 13.4 +/- 1.6 (SEM) to 8.6 +/- 1.2 mmol/L, accompanied by a significant reduction from 40.6 +/- 3.7 to 29.8 +/- 2.8 mM X h of the plasma glucose response to 75 g oral glucose. Fasting plasma insulin showed a nonsignificant increase from 24.8 +/- 2.0 to 31.3 +/- 2.3 mU/L. The percent specific binding of tracer 125I-insulin to erythrocytes and monocytes did not change significantly (from 9.8 +/- 1.7 to 8.5 +/- 0.7 for erythrocytes and 1.7 +/- 0.3 to 1.6 +/- 0.4 for monocytes). Glucose utilization was measured at three levels of insulin infusion (40, 100, and 300 mU/kg/h) by the euglycemic clamp technique. Overall there was a significant (P less than 0.05) increase in the disappearance rate (Rd) and metabolic clearance rate (MCRg) for glucose at the two higher insulin infusion rates (MCRg: 3.3 +/- 0.7 to 5.1 +/- 0.7 and 5.9 +/- 0.9 to 7.9 +/- 0.9 ml/kg/min), but not at the lowest infusion rate (MCRg: 3.6 +/- 0.8 to 3.3 +/- 0.6). Thus, the chronic hypoglycemic effect of gliclazide in obese diabetic subjects was associated with an improvement in insulin-mediated glucose utilization at high plasma insulin concentrations. This enhanced effect of insulin after gliclazide treatment was not accompanied by increased monocyte or erythrocyte insulin binding, which suggests that it was due to potentiation of postbinding insulin-sensitive pathways.

In six normal nonobese subjects, hyperinsulinemic euglycemic clamps were performed during paired sequential two-hour intravenous (IV) insulin infusions separated by an hour washout period. Each infusion was either 32 mU/kg/h of continuous insulin (CI) or 75% of this dose as 40-second pulses delivered every 13 minutes (PI). Six studies were performed with each of the following sequences in random order: PI-CI, CI-PI, and CI-CI. Based on the initial infusions, the insulin-dependent fractional glucose disappearance rate (X) during pulsatile insulin delivery (3.0 +/- 0.4 min-1 X 10(2), n = 6) was 73% of that of the continuous infusions (4.1 +/- 0.3 min-1 X 10(2), n = 12). This ratio was similar to that of the measured time-averaged plasma insulin areas (PI = 24.7 +/- 3.8 v CI = 31.4 +/- 3.5 mU/L). There was an average 23% enhancement of insulin's hypoglycemic effect during the second 12 CI infusions compared with the 12 initial CI infusions (X = 5.1 +/- 0.5 v 4.1 +/- 0.3 min-1 X 10(2), P less than .05). There was no significant difference between the enhancing effects of PI and CI infusions on insulin action in the subsequent CI's (X = 4.9 +/- 0.9 for PI-CI v X = 5.3 +/- 0.2 min-1 X 10(2) for CI-CI). First infusion PI significantly (P less than .05) decreased plasma C-peptide levels (0.34 +/- 0.05 to 0.20 +/- 0.06 mumol/L), whereas CI did not (0.33 +/- 0.02 to 0.32 +/- 0.07).(ABSTRACT TRUNCATED AT 250 WORDS)

Whereas the expected decrease in the binding of 125I-insulin to erythrocytes 4.5 h after a standard mixed breakfast was seen in six young nonobese normoglycemic controls (delta % bound/free = 2.0 +/- 0.5 SEM), 10 obese hyperglycemic subjects showed an increase (0.54 +/- 0.43) and eight obese normoglycemic subjects of similar age and weight, no change in binding (0 +/- 0.31), both being significantly different from the controls (p less than 0.05) but not from each other. Of the total 18 obese subjects, the 11 with fasting hyperinsulinemia (23 +/- 1.4 mU/l) showed an increase in binding of 0.82 +/- 0.40, significantly different (p less than 0.05) from the decrease of 0.30 +/- 0.25 shown by the seven normoinsulinemic (11 +/- 1.4 mU/l) subjects, despite their similar age, weight, and blood glucose concentrations. These changes in binding were explained by alterations in receptor affinity. Thus, an anomalous postprandial increase in erythrocyte insulin receptor affinity occurs in some obese subjects with or without diabetes, and is related to hyperinsulinemia rather than hyperglycemia.

These studies examined the effect of fenfluramine on insulin action and insulin secretion in healthy subjects and patients with non-insulin-dependent diabetes mellitus (NIDDM). In the first study, a double-blind crossover design was used in healthy subjects to compare the effect of short-term fenfluramine therapy (60 mg orally for 3 days) with placebo. Insulin secretion and whole-body insulin sensitivity (determined by frequently sampled intravenous glucose tolerance tests with analysis by the minimal-model method) were unchanged by fenfluramine. In the second study, involving patients with NIDDM inadequately controlled on submaximal to maximal doses of oral hypoglycemic agents, a double-blind crossover strategy was used to compare baseline studies (conducted after a run-in period) with fenfluramine (60 mg orally) or placebo for 4 wk. There was a signficant fall in fasting blood glucose after therapy with fenfluramine compared with the baseline study period (13.0 ± 1.2 vs. 8.4 ± 0.89 mM, mean ± SE, P < .01) with no significant fall in fasting serum insulin (20± 2 vs. 24 ± 3 (μU/ml) or C-peptide (1.3 ± 0.2 vs. 1.3 ± 0.1 nM). During euglycemichyperinsulinemic (1 mU · kg−1 · min−1) clamp studies there was a significant increase in insulin action from 12.7± 2.3 to 17.3 ± 1.8 min−1 103 μU · ml1 (P < .05), although clamp insulin levels were lower after fenfluramine treatment (136 ± 14 vs. 96 ± 9 μU/ml−1, P < .02), reflecting an enhanced metabolic clearance rate for insulin (12.7 ± 1.5 vs. 20.1 ± 2.1 ml · kg−1 · min−1 P < .025). When insulin action was normalized for the prevailing insulin level during the clamp, the increase of insulin action/in concentration was more marked (0.11 ± 0.22 to 0.22 ±0.04 min−1 103μU · ml−1

The structure of the insulin receptor in intact human erythrocytes was defined using the techniques of disuccinimidyl suberate (DSS) cross-linking of 125I-insulin and surface [125I]iodination followed by receptor immunoprecipitation. In contrast to a recent report, we found the erythrocyte insulin receptor to be similar in structure to that in classic target tissues for insulin, consisting of at least three species of molecular weight approximately 295,000, 265,000, and 245,000, containing disulfide-linked subunits of molecular weight approximately 130,000 and 95,000. The interconversion of the three oligomeric forms could mediate changes in receptor affinity as postulated in other tissues. The 95,000 subunit was detected by immunoprecipitation only if surface iodination was performed in a Tris/Hepes buffer using lodogen and not if phosphate-buffered saline or lactoperoxidase iodination was used. These findings indicate that the lack of a bioeffect of insulin in erythrocytes is not explained by a gross defect in the structure of their insulin receptors. The apparent identity of the insulin receptor structure in erythrocytes and insulin target tissues provides a firmer basis for the use of erythrocytes in some circumstances to reflect insulin receptor status.

The initial step in insulin action is binding to specific receptors. Two covalent receptor modifications possibly involved in producing pharmacodynamic effects as a result of insulin receptor binding are autophosphorylation and disulphide insulin binding. Insulin receptor numbers are 'down regulated' by insulin, but this effect may be minimised by pulsatile insulin secretion. Insulin receptor affinity is modulated rapidly by fasting, exercise and dietary composition. In non-insulin-dependent diabetes coupling of receptor binding to bioeffects is impaired. Binding is also reduced in those subjects with hyperinsulinaemia and non-insulin-dependent diabetes. Insulin-dependent diabetics have reduced insulin sensitivity, which is only partially reversed by conventional insulin therapy. 'Post-binding defects' in some diabetics could be related to defective covalent receptor modifications resulting from genetic receptor defects. High carbohydrate diets improve diabetes control through effects on the binding and coupling defects. In addition to stimulating insulin secretion, oral hypoglycaemics stimulate post-binding insulin action in vivo and in vitro. Insulin therapy in diabetes also tends to reverse post-binding defects. Pulsatile insulin delivery is more effective in lowering blood sugar than continuous administration, and produces less 'down regulation' of receptors. Combined insulin and sulphonylurea drugs reduce insulin requirements only in insulin-dependent diabetics with some endogenous insulin secretion, whereas metformin reduces insulin requirement in C-peptide negative insulin-dependent diabetes mellitus.

Insulin initiation and titration in primary care is necessary to respond to the growing epidemic of type 2 diabetes (T2D). The INITIATION study aims to evaluate the impact of implementing a new model of care with Primary Care Physician and Practice Nurse (PN) teams supported by a Credentialed Diabetes Educator-Registered Nurse (CDE-RN) and endocrinologist in initiating and titrating basal and prandial insulin for T2D patients in the Australian healthcare system over 24 weeks. This study also explores the feasibility and efficacy of retrospective continuous glucose monitoring (r-CGM) in comparison with self-monitoring of blood glucose (SMBG) among people with T2D in primary care. The study employs a before and after design with a nested exploratory trial of SMBG and r-CGM. A total of 102 insulin naive T2D patients with a glycated haemoglobin (HbA1c) level of >7.5% in the previous 6 months while treated with maximal oral therapy will be recruited and screened from 22 primary care practices in Melbourne, Australia. All patients will be commenced on a basal insulin regimen following randomization into one of the two blood glucose monitoring arms, with intensification to a “basal plus” regimen if required. The outcomes of the new model of care will be benchmarked with data collected over the same period from a specialist setting in Melbourne, Australia. This article describes the study protocol and insulin treatment algorithm employed in the first study to explore r-CGM use among T2D in primary care. Findings from the INITIATION study will inform development of a larger randomized controlled trial. ACTRN12610000797077 .