Pancreatectomized Patients Research Articles

Persistent Hyperinsulinaemic Hypoglycaemia of Infancy in 43 Children: Long-term Clinical and Surgical Follow-up

To describe the clinical, surgical, biochemical, radiological and electrophysiological features of 43 Saudi children with persistent hyperinsulinaemic hypoglycaemia of infancy (PHHI) who have been followed since 1983. Data from 43 patients were retrospectively analysed. PHHI was diagnosed on the basis of high intravenous glucose requirement, high insulin to glucose ratio, negative urinary ketones and normal tandem mass spectrometry. The patients were assessed radiologically by brain magnetic resonance imaging and/or computed tomography and electrophysiologically by brain stem auditory evoked potential, visual evoked response and electroencephalogram. Patients who failed medical therapy received near total pancreatectomy. The patients were severely hypoglycaemic and intolerant to fast. Hypoglycaemic convulsion was the most commonly presenting complaint. Eighteen patients were developmentally delayed and 14 of them had brain atrophy. All patients, except nine, did not respond to medical treatment and underwent surgery. Four pancreatectomized patients developed diabetes and two had malabsorption. One baby had 180 cm resection of gangrenous bowel most likely secondary to octreotide. No common bile duct injury was encountered. One patient was treated medically during childhood and developed diabetes and gained weight during adolescence. PHHI is a relatively common and serious disease among Saudi children. Early intervention is necessary to avoid neurological damage in patients who are severely hypoglycaemic and unresponsive to medical therapy. Surgically and probably medically treated patients are at a high risk of developing diabetes, which could be the natural outcome of this disease. Care and spending time during surgery to visualize the common bile duct help in avoiding its injury.

Insulin secretory defect plays a major role in the development of diabetes in patients with distal pancreatectomy

To investigate the pathogenesis of distal pancreatectomy (d-Px)–induced diabetes in Korean patients, we investigated insulin secretory and sensitivity indexes obtained by oral glucose tolerance testing in 20 patients that had received d-Px (10 with d-Px–induced diabetes and 10 with normal glucose tolerance with d-Px [NGT d-Px]) and in 164 control subjects (77 with type 2 diabetes mellitus and 87 with NGT) that did not receive d-Px. The pancreatectomized subjects had lower fasting serum insulin, homeostasis model assessment of pancreatic beta-cell function (HOMA- β) levels, and insulinogenic indices than the NGT controls. The HOMA- β values of nonobese NGT d-Px– and d-Px–induced diabetic subjects were 73.7% and 38.7% of those for nonobese NGT controls, respectively, and HOMA- β was significantly lower only for d-Px–induced diabetic subjects ( P < .01). In obese subjects, the HOMA- β values of obese d-Px–induced diabetic subjects were significantly lower than those of obese NGT controls ( P < .05). The insulin sensitivity was significantly lower in nonobese type 2 diabetes mellitus controls than in nonobese NGT d-Px or in nonobese d-Px–induced diabetic subjects ( P < .001 and .05, respectively). These results show that a reduced insulin secretory function is a typical feature of glucose homeostasis in distal pancreatectomized patients and that insulin secretory defect plays a major role in the development of diabetes in these patients. In addition, the study suggests that pancreatic resections of 60% or less and body mass index are not the main causes of diabetes onset after d-Px in this study.

Nasal glucagon delivery using microcrystalline cellulose in healthy volunteers

We developed an intranasal powder form of glucagon to improve metabolic status and fatty liver in patients with pancreatectomy. Microcrystalline cellulose, which is commonly used in commercial preparations for allergic rhinitis was used as an absorption enhancer. We compared the intranasal powder form with some spray solutions of glucagon with regard to glucagon absorption, concentration of blood glucose, stability and nasal irritation. The absorption of glucagon from the spray solution including 1.5% sodium glycocholate or 1% sodium caprate was 1.3- and 2.6-fold higher than that from the powder form mixed with microcrystalline cellulose at a ratio of 1:69, respectively. The C max values of plasma glucose were 2.18, 3.39 and 1.56 mmol l −1 in the spray solutions including sodium glycocholate and sodium caprate and in the powder form, respectively. However, glucagon in spray solutions was unstable, but that in the powder form was stable at 5 and 25 °C for at least 84 days. The spray solution caused strong irritation, but the powder form did not. These results suggested usefulness of the powder form of glucagon for treatment of pancreatectomized patients.

Characterization of Genes Encoding the Pancreatic .BETA.-cell ATP-sensitive K+ channel in Persistent Hyperinsulinemic Hypoglycemia of Infancy in Japanese Patients.

Persistent hyperinsulinemic hypoglycemia of infancy (PHHI) is a genetic disorder characterized by unregulated insulin secretion and profound hypoglycemia. Recently, mutations of SUR1 and Kir6.2, which constitute the pancreatic beta-cell ATP-sensitive potassium (K(ATP)) channel, have been shown to be associated with familial PHHI in certain ethnic groups. In the present study, we examined clinical symptoms, therapy, and variations in the SUR1 and Kir6.2 genes in eight Japanese patients with PHHI. Four patients were being treated with pharmacological agents and the other four had required pancreatectomy to normalize glucose levels. There was no difference in timing of the onset of hypoglycemia between the groups. There also was no difference in severity between the two groups, as assessed by blood glucose levels, plasma insulin levels, and birth weight. However, all of the pancreatectomized patients and none of the medically treated group had presented with seizures. By genetic screening, we found eleven nucleotide substitutions in the SUR1 gene, three of which result in amino acid changes, and three nucleotide substitutions in the Kir6.2 gene, two of which result in amino acid changes, but all of these genetic variants had been previously reported in normal subjects. These results indicate that the mechanism of hypoglycemia in these patients is different from those previously reported in PHHI patients, giving further support to the view that PHHI is a heterogeneous disorder.

Clinical Islet Transplantation After Allogeneic Orthotopic Liver Transplantation

ISLET transplantation as a treatment for type 1 diabetes mellitus has been applied mostly in patients with endstage diabetic nephropathy, either with previously established kidney allograft (islet after kidney transplant, IAK) or by a simultaneous islet–kidney transplantation (SIK) procedure, in the latter case utilizing pancreas and islets from the same donor. Immunosuppressive protocols for these patients are based on cyclosporine, usually in combination with corticosteroids and azathioprine or mycophenolate mofetil for induction and maintenance immune therapy, and in the majority of cases combined with anti– T-lymphocyte antibodies for a limited perioperative period. A small number of pancreatectomized patients (usually for treatment of upper abdominal malignancy) have received a simultaneous islet-liver transplant in previous studies. In most of these cases, induction/maintenance immunosuppression therapy has been based on tacrolimus (Prograf). In the present case study, for the first time an islet allograft (islet after liver transplantant, IAL) has been performed in a patient with type 1 diabetes who had previously received an orthotopic liver transplant for treatment of cryptogenic liver cirrhosis. Clinical and metabolic data are provided for the 6 month follow-up.

Treatment of isolated pancreatic islets to reverse pancreatectomy-induced and insulin-dependent type I diabetes in humans: A 6-year experience

THE SUCCESSFUL isolation and subsequent auto.1 transplantation of islets into pancreatectomized patients has promulgated the possible use of this procedure for the treatment of insulin-dependent type I diabetes (IDOM).1.2 However, our inability to diagnose and promptly treat islet allograft rejection combined with the deleterious effects of currently used immunosuppressive agents have severely limited the clinical utility of this approach. In an attempt to evolve a universally applicable strategy for the surgical treatment of 100M, we at the Thomas E. Starzl Transplantation Institute have performed, between January 1990 and May 1996. 42 pancreatic islet cell transplantations. We report herein, the results of our 6-year experience with this procedure in humans and discuss some of the approaches that we have evolved that may assist in the successful engraftment of allotransplanted islets.

Hyperinsulinemic hypoglycemia of infancy (nesidioblastosis) in clinical remission: high incidence of diabetes mellitus and persistent beta-cell dysfunction at long-term follow-up.

In persistent hyperinsulinemic hypoglycemia of infancy (PHHI), the long term outcome of the disease is not well documented. Previous reports suggested that partial pancreatectomy in infants does not endanger future islet function. We evaluated endocrine pancreatic function in 14 PHHI patients 6.5-21 yr after diagnosis. Eight underwent early subtotal pancreatectomy, and 6 were medically treated; all were in clinical remission. Intravenous glucose tolerance and C-peptide suppression tests were performed, with multiple determinations of hormone levels. The insulin response to glucose was blunted in all pancreatectomized and in 2 conservatively treated patients. Glucose disposal was reduced in 6 pancreatectomized patients and in 2 medically treated subjects. Six of the pancreatectomized patients, including two with normal glucose disposal at first evaluation, developed overt diabetes during puberty. None in the medically treated group became diabetic; however, only 2 patients have reached puberty. C-Peptide suppression in response to hypoglycemia was inadequate in 4 of 5 pancreatectomized and 3 of 5 nonpancreatectomized patients studied. These results show that children with PHHI have impaired insulin responses to glucose and lack of suppressibility of endogenous insulin secretion several years after clinical remission. Thus, the beta-cell defect responsible for the disease in infancy is not corrected with time despite the disappearance of spontaneous hypoglycemia. Insulin secretion seems more disturbed in pancreatectomized patients; the majority develop insulin-requiring diabetes during puberty. An effort should be made to treat PHHI patients medically to avoid pancreatectomy; this may reduce the incidence of diabetes at puberty.

すい全摘出患者に対するインスリン持続注入療法のありかた ＤＤＳの見地から

Total pancreatectomized patients has similar pathophysiological aspects with insulin-dependent diabetes mellitus (IDDM) in absolute insulin deficiency, both hepatic and peripheral insulin resistance, and, the development of diabetic micro—as well as macroangiopathy due to long-term hyperglycemia. Glucagon deficiency in pancreatectomized patients, however, causes more frequent hypoglycemia due to delayed hepatic glucose production and less insulin requirement than in IDDM patients. The optimal insulin therapy for these pancreatectomized patients is discussed from the viewpoint of drug delivery system (DDS). 1. Targeting : The study of hepatic and peripheral glucose handling in pancreatectomized diabetic dogs revealed that intraportal insulin administration results in a smaller gain in hepatic glucose handling than peripheral insulin administration. This means the superiority of intraportal insulin delivery which appears to be safer and more suitable for attaining stable glycemic control in these patients. 2. Controlled-release : Closed-loop insulin infusion system is optimal especially for postprandial glycemic regulation in these patients because of unstable enteric nutritional absorption, whereas it is still not available due to the lack of glucose sensor for long-term clinical usage. Therefore, frequent insulin dose adjustment based on self-monitoring of blood glucose (SMSG) is the second best. 3. The necessity of safer intraportal indweling cannula for long-term clinical usage : Intraportal cannulation may provoke portal thrombosis as well as infectious disease. These problems in biocompatibility promoted intraperitoneal delivery in stead. From these viewpoints, ultimate goal of future therapy for diabetes mellitus is discussed.

Hybrid artificial pancreas: islet transplantation inside membrane bioreactors.

The use of pancreatic islet transplantation in membrane bioreactors put in vascular circuits aims at resetting the glucose homeostasis in diabetic or pancreatectomized patients, avoiding immune host rejection. Our experience was carried out at following stages: porcine pancreas explantation and enzymatic separation of endocrine tissue from exocrine fraction by collagenase; evaluation of islet functionality (culture tests); in vitro tests of the islets-bioreactor system, to assess the metabolic response to the glucose; in vivo evaluation to assay the haemodynamic behaviour. The trials showed a good metabolic bioreactor functionality and a decreasing incidence of coagulative problems.

Metabolic Consequences of (Regional) Total Pancreatectomy

Little information has been reported on the metabolic characteristics of the totally pancreatectomized patient or the efficacy of medical management after radical pancreatic surgery. The prospective evaluation of 49 such patients, with 31% followed for 48 or more months, forms the basis of this report. The major immediate postoperative challenge is control of diarrhea and weight stabilization. Chronically patients have an increased daily caloric requirement (mean +/- SE, 56 +/- 1 kcal/kg), not wholly explained by moderate steatorrhea (fecal fat excretion, 16% +/- 2% of unrestricted fat intake). Despite persistent malabsorption, deficiencies in fat-soluble vitamin, magnesium, and trace element serum levels can be prevented in most patients. Pancreatogenic diabetes is characterized by (1) absence of the major glucoregulatory hormones insulin and glucagon, (2) instability, and (3) frequent hypoglycemia, with the latter parameters improving with rigorous home glucose monitoring. No patient has developed clinically overt diabetic micro- or macrovascular disease. Performance status in long-term survivors has been reasonable. However adverse chronic sequelae of the operation occur and include an unusual frequency of liver disease, characterized by accelerated fatty infiltration, and osteopenia, with an 18% reduction in radial bone mineral content noted in pancreatectomized patients studied more than 5 years after surgery.

Thyrotropin-releasing hormone: further extraction studies and analysis by fast protein liquid chromatography and radioimmunoassay

We describe the clinical application of a radioimmunoassay combined with fast protein liquid Chromatography (FPLC) for measuring TRH immunoreactivity (TRH-IR) in blood samples extracted previously with methanol or with Sep Pak C18 cartridges. Sensitivity of the RIA was 3 fmol/tube, displacement at 50% B/B0 was achieved by 55 fmol of unlabelled TRH. Our specific antibody K2B7 (km = 2.2 fM) showed no cross reaction with other peptides. No difference was observed between the mean values of TRH-IR in 19 euthyroid, 22 hyperthyroid, 18 hypothyroid and 10 hypophysectomised patients (45 +/- 17.8, 58 +/- 30, 40 +/- 22 and 36 +/- 12 fmol/ml, mean +/- SD, respectively), whereas TRH-IR was significantly lowered (p less than 0.05) in 6 euthyroid pancreatectomised patients (21 +/- 5 fmol/ml). The reversed phase FPLC analysis of the TRH-IR presented in the methanol extracts was shown to have the same retention time as synthetic TRH. TRH could not be measured in unextracted blood samples. TRH added in preincubated (60 min, at 37 C), before extraction, blood samples showed a loss of 83.4% of immunoreactivity. Our results demonstrate that this method is able to detect TRH-IR in human blood by whole methanol extraction and/or by Sep Pak C18 cartridges extraction. Furthermore the findings suggest that the main source of circulating TRH-IR may be of extrahypothalamic (pancreatic?) origin and that the basal peripheral TRH levels are not involved or they do not clearly represent a pathological condition.

Glucagon Secretion is Essential for Aminoacid-Induced Hyperfiltration in Man

The role of glucagon as a mediator of aminoacid-induced alteration of renal haemodynamics was evaluated in man in three different protocols. In the first it was shown that the increase in glomerular filtration rate (GFR) and renal plasma flow (RPF) observed during an aminoacid infusion was prevented by the additional infusion of somatostatin (SRIF), but reproduced by a glucagon infusion in the presence of SRIF. In the second protocol it was shown that, at variance with normal subjects, six totally pancreatectomised patients, thus deprived of pancreatic glucagon secretion, did not increase their GFR and RPF when infused with amino-acids, whereas they exhibited the expected hyperfiltration when infused with glucagon. In the third protocol it was shown that glucagon infusion in a renal artery did not alter the homolateral renal haemodynamics. It is concluded that glucagon secretion is a mandatory step in the cascade of events linking the infusion of aminoacids to the renal hyperfiltration. Other steps beyond glucagon secretion are necessarily involved because glucagon has no direct renal effect.

14 PERSISTENT HYPERINSULINENIC HYPOGLYCEMIA OF INFANCY – LONG-TERM EXPERIENCE WITH 28 PATIENTS

We studied 28 patients with Persistent Hyperinsulinemic Hypoglycemia of Infancy (PHHI). Twenty-two developed symptomatic hypoglycemia within 2 days of life (15 of these required surgery) while 5 developed first symptoms at age 2-7 months (1 required surgery). Twelve patients belonged to 5 families indicating a genetic basis for the disease (autosomal recessive). Non-pancreatectomized patients received frequent high carbohydrate feedings, raw cornstarch, diazoxide and/or somatostatin analogue (Sandostatin™, Sandoz, Switzerland, n=5). All 85-90% pancreatectomized patients (n=15) required some medical treatment for hypoglycemia post-operatively. Three patients died (2 of sepsis and 1 of acute gastric perforation). Diabetes developed only in the 1 patient who had a total pancreatecomy. Of 25 living patients, 16 entered remission between 0.3 & 8 years of age (follow-up 0.3-12 years, mean 5.5 years). Of the 22 living patients who were not retarded at diagnosis, 21 developed normally while 1 is retarded (IQ 65). Pathologic examination of 16 pancreata showed focal findings in 3, the rest being histologically normal for age. Beta-cells from 3 histologicaly normal pancreata were cultured on extra-cellular matrix and showed significant & unchanged insulin secretion at all levels of glucose tested (0-16.7 mmol), which suppressed with epinephrine or somatostatin and stimulated with IBMX or forskolin. In conclusion: PHHI appears to be caused by a functional β-cell defect, presumably genetic in origin, manifested by a lack of suppression of insulin secretion by low glucose, and which may resolve spontaneously with time. Intensive conservative treatment, with 85-90% pancreatectomy only when absolutely necessary, resulted in a good long-term prognosis in 95% of our patients in whom the diagnosis was made before the onset of neurologic damage.

Basal and meal-induced somatostatin-like immunoreactivity in healthy subjects and in IDDM and totally pancreatectomized patients. Effects of acute blood glucose normalization.

We investigated the impact of blood glucose normalization on plasma levels of somatostatin-like immunoreactivity (SLI) in subjects with C-peptide-negative insulin-dependent diabetes mellitus (IDDM) and in totally pancreatectomized patients. Patients were studied during hyperglycemia and during normoglycemia, which was attained by Biostator-directed feedback insulin infusion. The experiments were performed in the fasted state and after a standardized breakfast. In IDDM (n = 6), basal levels of SLI were significantly higher than in nondiabetic subjects (n = 18). In IDDM, normalization of hyperglycemia was followed by a 40% decline in basal SLI (P less than .05). After the meal, SLI increased to the same absolute levels with or without feedback insulin treatment; however, the incremental response was 60% higher during feedback insulin treatment (P less than .05). Treatment also suppressed fasting and postprandial levels of glucagon, whereas gastric inhibitory polypeptide (GIP) levels were unaltered. In four pancreatectomized patients, normoglycemia tended to lower plasma levels of SLI by 50% (P less than .1). After breakfast, an SLI response was noted during normoglycemia, whereas no significant effect of the meal was seen during hyperglycemia. We conclude that in IDDM and in totally pancreatectomized patients, administration of insulin with subsequent normalization of blood glucose is accompanied by a decline in plasma levels of SLI in the fasted state, whereas the apparent response to a meal is enhanced. These effects on plasma levels of SLI probably reflect to a major extent release of somatostatin from the gastrointestinal tract.

Basal and meal-induced somatostatin-like immunoreactivity in healthy subjects and in IDDM and totally pancreatectomized patients. Effects of acute blood glucose normalization

Basal and meal-induced somatostatin-like immunoreactivity in healthy subjects and in IDDM and totally pancreatectomized patients. Effects of acute blood glucose normalization

Glucagon immunoreactivity and chromatographic profiles in pancreatectomized humans. Paradoxical response to oral glucose.

The nature and origin of plasma immunoreactive glucagon (IRG) after pancreatectomy in humans remains controversial. Low plasma IRG levels and heterogeneity hamper accurate assessment. We studied plasma IRG levels and profiles in 12 patients 2-57 mo after a total pancreatectomy (with antrectomy and duodenectomy) for cancer (N = 9) or chronic pancreatitis (N = 3). After oral glucose, plasma IRG (with the COOH-terminal-specific 30K glucagon antibody) rose from 59 +/- 7 to a peak of 113 +/- 17 pg/ml at 60-120 min. Chromatographic profiles revealed four distinct IRG fractions. In every patient a plasma IRG fraction of 9000-15,000 Mr, detectable basally, increased markedly after oral glucose and accounted for the rise in total IRG observed in plasma. Nine of the 12 pancreatectomized subjects had no detectable 3500-Mr glucagon and the remaining 3 had very low levels. For the group as a whole, 3500-Mr IRG comprised 1-2% of the total recovered IRG. Two patients were also studied before pancreatectomy: suppressibility of glucagon (Mr 3500) was evident. After surgery this paradoxical response to oral glucose was demonstrated. Reproducibility of these responses was confirmed in two patients studied twice over 2 yr. Diabetic controls without pancreatectomy did not show this response. The absence or marked reduction of pancreatic glucagon was confirmed in five of the pancreatectomized patients after intravenous arginine or oral protein. Normal basal plasma IRG and profiles, oral glucose suppressibility, and arginine stimulation were present in five control patients with unresectable pancreatic malignancies.(ABSTRACT TRUNCATED AT 250 WORDS)

Insulin resistance is a prominent feature of patients with pancreatogenic diabetes

To compare in vivo insulin action in patients with diabetes secondary to pancreatic diseases (n = 9) to that in type I diabetic patients (n = 13) and in normal subjects (n = 8), we measured insulin-mediated glucose disposal by the euglycemic insulin clamp technique. Five of the nine patients with pancreatogenic diabetes had undergone total pancreatectomy. Similar plasma glucose (∼4.8 mmol/l) and insulin (∼70 mU/l) levels were maintained in all groups. The rate of glucose metabolism in the pancreatogenic diabetic patients (3.77 ± 0.55 mg/kg/min) was 47% lower ( P < 0.001) than in normal subjects (7.05 ± 0.57 mg/kg/min) and 21% lower ( P < 0.05) than in type I diabetic patients (5.54 ± 0.39 mg/kg/min). The rates of glucose uptake were similarly reduced in totally pancreatectomized patients and those with pancreatogenic diabetes due to other causes. During hyperinsulinemia induced by the clamp, glucose production (measured using 3- 3H-glucose infusion) was completely suppressed in both the pancreatogenic diabetic patients and the normal subjects indicating that the impairment of in vivo insulin action was localized to the peripheral tissues. However, basal glucose production was elevated in the pancreatogenic diabetic patients (2.75 mg/kg/min, P < 0.001) compared to the normal subjects (1.79 ± 0.07 mg/kg/min). Glucose production rates were comparable in the totally pancreatectomized patients and in the other patients with pancreatogenic diabetes. The fasting plasma insulin level was, however, lower in the totally pancreatectomized (3.2 ± 1.6 mU/L, P < 0.05) than the other pancreatogenic (11.5 ± 3.7 mU/L) diabetic patients. To examine the mechanisms of peripheral insulin resistance in the pancreatogenic diabetic patients, insulin binding and action were measured in isolated adipocytes. The pancreatogenic diabetic patients displayed normal insulin binding as well as normal rates of glucose transport and oxidation in adipocytes. In conclusion, patients with pancreatogenic diabetes demonstrated marked insulin resistance. Thus, impaired regulation of glucose production is a more likely explanation for the special clinical features of pancreatogenic diabetes than enhanced glucose utilization.

Pancreatic secretory trypsin inhibitor-like immunoreactivity in pancreatectomized patients

Pancreatic secretory trypsin inhibitor (PSTI) is a 6 000-dalton peptide, that occurs in high concentrations in the pancreas and in pancreatic juice. It is thought to be synthesized by pancreatic acinar cells. We have recently reported the finding of an identical trypsin inhibitor at high concentrations in the urine of patients with gynecological malignancy. Therefore, we have named the inhibitor tumor-associated trypsin inhibitor (TATI). We have now studied patients who have undergone total pancreatoduodenectomy for pancreatic cancer or chronic pancreatitis. By radioim-munoassay (RIA), we found normal levels of this inhibitor in the serum and urine of pancreatectomized patients. The absence of pancreas was confirmed by measuring serum trypsin. By gel filtration and HPLC it was found that PSTI/TATI occurring in pancreatectomized patients was indistinguishable from that found in connection with pancreatitis and ovarian cancer.

Pancreatic polypeptide: a marker for lean non-insulin-dependent diabetes mellitus?

Both basal and postprandial pancreatic polypeptide (PP) concentrations were exaggerated twofold in lean NIDDM patients, whereas they were normal in lean IDDM and obese NIDDM patients who were hyperglycemic as a result of partial insulin withdrawal. Insulin infusion from an artificial endocrine pancreas, which resulted in fasting euglycemia and near-normoglycemia postprandially, had no effect on PP responses in any of the diabetic patients. No postprandial PP responses were observed in totally pancreatectomized (TPX) patients. Excessive basal and postprandial concentrations of PP in diabetes appear to be related to both leanness and residual beta cell function and, therefore, potential markers for lean NIDDM.

Release of gastric inhibitory polypeptide in totally pancreatectomized patients.

The role of insulin in controlling gastric inhibitory polypeptide (GIP) release was assessed in seven totally pancreatectomized patients, in whom inhibitory effects of endogenous pancreatic hormones on GIP secretion were excluded. In these patients, glucose was ingested on two occasions, once alone and once during insulin infusion using an artificial endocrine pancreas to maintain blood glucose at levels similar to those in normal subjects. Plasma levels of GIP, the maximum response of plasma GIP, and the integrated incremental GIP for 180 min were similar. We conclude that insulin does not alter the secretion of GIP in response to oral glucose when the influence of endogenous pancreatic hormones is excluded.