Search

This review addresses the current concepts in our understanding of the epidemiology, mechanisms, symptoms, clinical consequences, diagnosis and treatment of delayed gastric emptying in patients with diabetes. Upper gastrointestinal symptoms, particularly postprandial fullness, nausea, vomiting and abdominal bloating, occur in 30-50% of patients with diabetes. The use of scintigraphic techniques, and more recently breath test, has shown that as many as 50% of diabetic patients have gastroparesis. Diabetic gastroparesis comprises a decrease in fundic and antral motor activity, a reduction or a lack of the interdigestive migrating motor complex, gastric dysrhythmias, and pylorospasms. The mechanisms involved include: autonomic neuropathy, acute hyperglycaemia, and abnormalities in gastrointestinal hormones and neuropeptides. Other possible contributing factors such as hypothyroidism and H. pylori infection are discussed as well. Because treatment is possible by means of dietary advise, prokinetics or surgical procedures, it is important to identify risk factors for and to diagnose gastroparesis to prevent morbidity by controlling gastrointestinal symptoms, and to enhance glucoregulation. Understanding the current advances is key to the development of novel therapeutic strategies and for making rational choices in the management of diabetic gastroparesis.

The aim of this study was to compare oxidative stress status (OSS) with blood glucose and lipid changes during the fasting, postprandial and postabsorptive phases in type 1 diabetes mellitus.Twenty-three patients on intensive insulin treatment received a standard fat-rich breakfast and lunch. OSS was monitored at fasting (F), just after the post-breakfast glycemia peak (BP) (identified by continuous subcutaneous glucose monitoring), 3.5-h post-breakfast (B3.5), just after the post-lunch peak (LP), just after the post-lunch dale (LD) and 5 hours after lunch (L5).Whereas whole blood glutathione and plasma protein thiols increased in the postprandial period (from 6.52 +/- 1.20 (F) to 7.08 +/- 1.45 micromol/g Hb (BP), p = 0.005), ascorbate decreased gradually from 44 +/- 17 (F) to 39 +/- 19 micromol/L (LD), p = 0.015. Retinol and alpha-tocopherol also decreased from 27.1 +/- 7.0 (F) to 25.3 +/- 5.2 micromol/L (BP), p = 0.005. Uric acid decreased later, from 213 +/- 77 (BP) to 204 +/- 68 micromol/L (B3.5), p = 0.01, but then increased in LP (231 +/- 70 micromol/L) and LD to values higher than F (215 +/- 64, micromol/L, p = 0.01). Malondialdehyde increased gradually from 1.02 +/- 0.36 (F) to a maximum of 1.14 +/- 0.40 micromol/L (LP). In the postabsorptive phase (L5) all parameters except for thiols reverted to fasting concentrations.In type 1 diabetes lipid peroxidation increases during the postprandial phase in parallel to glucose and triglyceride changes. Blood antioxidants, however, followed diverse patterns of change.

Blood glucose, lipids, alpha-tocopherol, and malondialdehyde were monitored in type 1 diabetes mellitus patients for 8 hours after a standard fat-rich breakfast and lunch. Although glucose and triglycerides increased, alpha-tocopherol decreased and malondialdehyde increased in the postprandial phase. In the postabsorptive phase values returned to fasting levels. These results point to the possible relevance of postprandial alpha-tocopherol depletion and lipid peroxidation to an increased cardiovascular risk in these patients.

OBJECTIVE—Parietal cell antibodies (PCAs) are found in 20% of type 1 diabetic patients, denoting autoimmune gastritis and pernicious anemia, which may predispose to enterochromaffin-like (ECL) cell hyper/dysplasia and gastric carcinoid tumors. We evaluated whether chromogranin A (CgA), 5-hydroxyindole acetic acid (5-HIAA), and neuron-specific enolase (NSE) contribute to screening for ECL cell hyper/dysplasia. RESEARCH DESIGN AND METHODS—Sera from 93 type 1 diabetic patients (53 men and 40 women, 31 PCA+ and 62 PCA−, aged 45 ± 13 years) were analyzed for PCAs by indirect immunofluorescence and for CgA, NSE, and gastrin by radioimmunoassay. Urinary 5-HIAA was tested by high-performance liquid chromatography. Corpus atrophy and ECL cell proliferation were assessed in gastric biopsies. RESULTS—PCA+ patients had higher gastrin (P < 0.0001) and CgA levels (P = 0.003) and were more prone to autoimmune gastritis (odds ratio [OR] 17, P < 0.0001) and ECL cell hyper/dysplasia (OR = 23, P = 0.005) than PCA− subjects. ECL cell hyper/dysplasia was present in seven PCA+ patients who showed higher CgA levels (P < 0.0001) than subjects without ECL cell hyper/dysplasia, but NSE and 5-HIAA levels were similar. CgA levels correlated with gastrinemia (r = 0.50, P < 0.0001), PCA titer (r = 0.42, P = 0.001), and 5-HIAA levels (r = 0.38, P = 0.012). Logistic regression identified the CgA level (β = 0.01, P = 0.027) as an independent risk factor for ECL cell hyper/dysplasia when PCA, CgA, 5-HIAA, NSE, gastrin, sex, and age were tested. Multivariate linear regression demonstrated that CgA level was determined by ECL cell density (r = 0.59, P < 0.0001) and gastrin level (r = 0.67, P = 0.02). One PCA+ patient with elevated gastrin, CgA, and 5-HIAA levels had a gastric carcinoid tumor. CONCLUSIONS—PCA+ patients, particularly those with high gastrin and CgA levels, risk developing ECL cell hyper/dysplasia. The determination of CgA, but not NSE and 5-HIAA, may complement histology in evaluating ECL cell mass.

Pre-prandial glucose gives insufficient information on glycemic excursions throughout the day. We aimed to test a continuous subcutaneous glucose-monitoring device (GlucoDay) to describe postprandial glucose changes.In 23 T1DM patients, 24-h GlucoDay registrations were started about 14 h before receiving a standard breakfast B and 3 h later lunch L.The 3-min glucose values were computed into parameters describing the postprandial changes after B and L. Two-hour glucose was higher after B (243 +/- 69 vs 180 +/- 79 mg/dL after L, p0.0001). Maximum glycemia (313 +/- 105 mg/dL after B and 304 +/- 119 after L, p0.0001) was higher and was reached after 78 and 57 min respectively. Three-hour AUC was higher but 30-min AUC was lower after B (5725 +/- 2414 vs 7488 +/- 2208 min mg/dL after L, p = 0.004). Glucose spikes (maximum peak minus fasting plasma glucose) were similar after B and L but the difference between maximum and minimum values was smaller after B (165 +/- 110 vs 219 +/- 115 mg/dL after L, p = 0.020). Duration of hyperglycemic periods200, 140 or 126 mg/dL were not different after B or L, but time spent at glucose100 mg/dL was longer after L (p0.0001).These results illustrate the use of subcutaneous glucose registration to characterize postprandial glycemia patterns in T1DM. Application of such methods to evaluate this and other clinical situations in DM can lead to therapeutic and dietary adjustments and ultimately improve glycemic control.

OBJECTIVE—Approximately 15–20% of type 1 diabetic patients exhibit parietal cell antibodies (PCAs) targeting gastric H+/K+ATPase. We examined whether iron deficiency anemia, pernicious anemia, and autoimmune gastritis, which may predispose to gastric tumors, were more frequent in PCA+ than in PCA- patients. RESEARCH DESIGN AND METHODS—Gastric biopsies from 88 consecutively recruited type 1 diabetic patients (51 men and 37 women, 47 PCA+ and 41 PCA-, aged 42 ± 13 years) were evaluated using the updated Sydney system. Immunostaining was done for parietal cells, B- and T-cells, enterochromaffin-like (ECL) cells, and Helicobacter pylori (HP). PCAs were assayed by indirect immunofluorescence, H+/K+ATPase antibodies by enzyme immunoassay, and HP by serology, urea breath test, and histology. Pentagastrin tests were performed in 42 subjects. RESULTS—Autoimmune gastritis (AG) was present in 57% of PCA+ and 10% of PCA- cases (OR 12.5, P < 0.0001). PCA positivity (β = 1.44; P = 0.04) and hypergastrinemia (β = 0.01; P = 0.026), but not HP, age, diabetes duration, sex, and HLA-DQ type were risk factors for AG. Iron deficiency anemia (OR 3.9, P = 0.015), pernicious anemia (OR = 4.6, P = 0.022), and hypochlorhydria (OR = 20.0, P = 0.0002) were more frequent in AG+ individuals. HP infection was present in 47 patients but did not influence corpus histology or gastrinemia. (Pre)malignant lesions were found in 26% of PCA+ subjects: ECL cell hyperplasia in 7 AG+ patients, comprising 1 with a gastric carcinoid tumor, and corpus intestinal metaplasia in 11 AG+ patients, including 1 with linitis plastica. CONCLUSIONS—PCA+ type 1 diabetic patients should be screened for autoimmune gastritis, iron deficiency, and pernicious anemia. Particularly hypergastrinemic PCA+ patients with autoimmune gastritis are at increased risk for (pre)malignant gastric lesions.

OBJECTIVE—A worldwide increase in the incidence of childhood type 1 diabetes has been observed. Because in various countries the majority of new type 1 diabetic patients are diagnosed in adulthood, we investigated whether the rising incidence of this disorder in children reflects a global increase in the incidence of diabetes or a shift toward earlier clinical presentation. RESEARCH DESIGN AND METHODS—The incidence of type 1 diabetes presenting before age 40 years was prospectively measured in the Antwerp district over a 12-year period (1989–2000). The completeness of ascertainment was evaluated by the capture-recapture method. Trends in incidence during the study period were analyzed by Poisson regression. RESULTS—The incidence of type 1 diabetes diagnosed before age 40 years remained constant over the 12-year period, averaging 9.9 cases per 100,000 individuals per year. The incidence was similar in both sexes under age 15 years, but a marked male excess was noted for adult-onset disease, in particular after age 20 years, resulting in a male-to-female ratio of 0.9 under age 15 years vs. 1.6 thereafter (P = 0.001). During the 12-year observation period, there was a significant tendency toward increasing incidence under age 15 years at the expense of a decreasing incidence between ages 15 and 40 years (P = 0.025). The annual increase in incidence averaged 1.8% under age 15 years and 5.0% under age 5 years (P = 0.06). CONCLUSIONS—Our results indicate that in Belgium, the increasing incidence of childhood type 1 diabetes—especially for children under age 5 years—is not attributable to a global increase in disease incidence, but rather to earlier clinical manifestation. The results suggest that an environmental factor may preferentially accelerate the subclinical disease process in young diabetes-prone subjects.

The prandial glucose regulator repaglinide has a rapid onset of action, a short half-life and is metabolised mainly by the liver. Here we report the findings of a 10-week, double-blind, parallel, placebo controlled, randomised trial with repaglinide in 25 diet-treated, sulphonylurea-naive patients with Type 2 diabetes. Repaglinide was titrated, based on capillary blood glucose, from 0.5 mg to a maximum of 4 mg, preprandially with breakfast and dinner. After 10 weeks, repaglinide was associated with a decrease in HbA(1c) of 2.3%Hb relative to the placebo group (P=0.018). This reflected a 30% decrease within the repaglinide group from a mean HbA(1c) of 7.0 to 4.9%Hb (P

BACKGROUND Vitamin E supplementation has been proposed as adjunctive therapy to counteract the increased LDL oxidation in diabetes and thus prevent or delay cardiovascular complications. OBJECTIVE The objective of this study was to investigate the effect of a moderate pharmacologic dose of vitamin E for

The history of a 45-year-old male type 1 diabetic patient is presented. At the age of 29 years, he was diagnosed to have an autoimmune hepatitis with incipient liver cirrhosis. Five years later, a successful liver/pancreas transplantation was performed. Eighteen months later, however, pancreatic insufficiency occurred due to thrombosis of the pancreatic graft. Besides these conditions, iron deficiency, pernicious anemia, and autoimmune gastritis were also diagnosed. Serum parietal cell antibodies (PCA) and intrinsic factor antibodies (AIF) were positive. At 45, this patient was found to have a gastric carcinoid tumor. The clinical importance of PCA is discussed with regard to chronic atrophic gastritis and pernicious anemia, which both predispose toward gastric carcinoid tumors. Autoimmune type 1 diabetic patients who have a high prevalence of PCA should be screened for gastric autoimmune manifestations and tumors, as the history of this patient illustrates.

To investigate whether a serotoninergic drug such as dexfenfluramine (dF) may have some beneficial effects on energy expenditure (EE) during therapeutic weight reduction, a 3-month study was conducted in a double-blind, placebo-controlled trial. Thirty-two obese, premenopausal women received either dF or placebo (P) in addition to a very-low-calorie diet (VLCD) prescription. All patients started--when hospitalized at the metabolic ward--with a 500-kcal regimen and fulfilled the 3-month trial on a +/- 760-kcal protein-sparing modified fast. Although not statistically significant, women receiving dF lost more weight (16.0 +/- 1.4 v 12.8 +/- 1.3 kg, P = .111) over the 3-month study period. Resting metabolic rate (RMR) decreased significantly by 5% in the dF group (4.79 to 4.53 kJ/min) and by 9% in the P group (5.09 to 4.63 kJ/min). When expressed per kilogram body weight, RMR significantly increased from 0.050 to 0.057 kJ/min/kg in the dF group (P.001), versus 0.053 to 0.056 in the P group (NS). When expressed per kilogram fat-free mass (FFM), RMR remained stable in the dF group, whereas it significantly decreased in the P group (P = .024). No significant differences could be found between groups. Glucose-induced thermogenesis (GIT), expressed as percent increase above RMR, did not show significant differences between groups. When expressed per kilogram body weight, mean GIT increased in the dF group from 0.14% to 0.16% above RMR, with a significant decrease from 0.15% to 0.13% in the P group. Only during the first hour did GIT per kilogram body weight significantly (P = .038) increase in the dF group during the outpatient period (between day 16 and day 90). These results show that a serotoninergic drug seems capable of limiting the weight reduction-associated decrease in RMR and dietary-induced thermogenesis (DIT), certainly when expressed on a per-kilogram-weight basis.

Autoimmune gastritis and pernicious anemia are common autoimmune disorders, being present in up to 2% of the general population. In patients with type 1 diabetes or autoimmune thyroid disease, the prevalence is 3- to 5-fold increased. This review addresses the epidemiology, pathogenesis, diagnosis, clinical consequences, and management of autoimmune gastritis in type 1 diabetic patients.Autoimmune gastritis is characterized by: 1) atrophy of the corpus and fundus; 2) autoantibodies to the parietal cell and to intrinsic factor; 3) achlorhydria; 4) iron deficiency anemia; 5) hypergastrinemia; 6) pernicious anemia may result from vitamin B12 deficiency; and 7) in up to 10% of patients, autoimmune gastritis may predispose to gastric carcinoid tumors or adenocarcinomas. This provides a strong rationale for screening, early diagnosis, and treatment. The management of patients with autoimmune gastritis implies yearly determination of gastrin, iron, vitamin B12 levels, and a complete blood count. Iron or vitamin B12 should be supplemented in patients with iron deficiency or pernicious anemia. Whether regular gastroscopic surveillance, including biopsies, is needed in patients with autoimmune gastritis/pernicious anemia is controversial. The gastric carcinoids that occur in these patients generally do not pose a great threat to life, whereas the danger of developing carcinoma is controversial. Nevertheless, awaiting a consensus statement, we suggest performing gastroscopy and biopsy at least once in patients with autoantibodies to the parietal cell, iron-, or vitamin B12-deficiency anemia, or high gastrin levels.The high prevalence of autoimmune gastritis in type 1 diabetic patients and its possible adverse impact on the health of the patient provide a strong rationale for screening, early diagnosis, periodic surveillance by gastroscopy, and treatment.

OBJECTIVE—To investigate a possible association of BMI with retinopathy and neuropathy in type 1 diabetes. Retinopathy and neuropathy may not only be related to glycemic control and diabetes duration but also to blood pressure and BMI. RESEARCH DESIGN AND METHODS—A total of 592 type 1 diabetic patients without nephropathy were studied (M/F: 324/268; age: 41 ± 12 years; duration: 19 ± 11 years; HbA1c [A1C]: 7.9 ± 1.1%). Patients were subdivided according to BMI: 168 men and 146 women with BMI RESULTS—Hypertension (>130/85 mmHg) was present in 40%, retinopathy in 53%, and neuropathy in 43% of patients. Overweight subjects had more retinopathy (63 vs. 45%, P < 0.0001, odds ratio [OR] = 2.1) and neuropathy (49 vs. 38%, P = 0.008, OR = 1.6) than normal-weight patients. Patients with retinopathy were older (45 ± 12 vs. 37 ± 11 years, P < 0.0001) and had a longer diabetes duration (25 ± 10 vs. 12 ± 8 years, P < 0.0001), a higher A1C (8.0 ± 1.1 vs. 7.7 ± 1.1%, P = 0.001), and a higher BMI (25.8 ± 4.1 vs. 24.7 ± 4.2 kg/m2, P = 0.001) than individuals without retinopathy. The same results are found in neuropathy. Logistic regression analysis showed that diabetes duration (β = 0.15, P < 0.0001), blood pressure (β = 0.22, P = 0.0047), and A1C (β = 0.24, P = 0.01), but not BMI, lipid levels, sex, or age, were independent risk factors for retinopathy. Likewise, duration (β = 0.05, P < 0.0001), age (β = 0.04, P = 0.0001), A1C (β = 0.35, P < 0.0001), and sex (β = 0.74, P = 0.0001) but not BMI, lipid levels, or hypertension were independently associated with neuropathy. Men had more neuropathy than women (50 vs. 34%, P < 0.0001, OR = 1.9). Leptin and adiponectin levels did not differ between individuals with or without microvascular complications. CONCLUSIONS—Retinopathy and neuropathy are more prevalent in overweight (BMI ≥25 kg/m2) type 1 diabetic subjects. However, logistic regression analysis showed that diabetes duration and A1C remain the main determinants for retinopathy and neuropathy.

Objective: To investigate the impact of Vitamin E on lipids and peroxidation during statin treatment. Research Design and Methods:T1DM patients with high cholesterol received Atorvastatin 20 mg with either placebo (group AP, n = 11) or d--tocopherol 750 IU (group AE, n = 11) daily. They were monitored for blood biochemistry, low-density lipoprotein (LDL) subfractions and lipid peroxidation at inclusion and after 3 and 6 months. Results: Serum cholesterol and triglycerides decreased to the same extent (29 and 21% respectively) in both groups. Serum tocopherol decreased by 18% in AP and increased by 50% in AE (P < 0.0001, between-group comparison by repeated measures ANOVA) but relative to lipids it increased by 15% in AP and by 100% in AE. Copper-induced production of thiobarbituric reactive substances in the LDL + VLDL fraction increased by 18% in AP and did not change in AE (P = 0.02). The lagtime for the production of fluorescent products was prolonged by 13 min only in group AE (P = 0.028). Plasma malondialdehyde decreased by 35% in both groups (P = 0.002) but not when adjusted for lipids. Conclusions:In T1DM Vitamin E supplements do not affect the lowering of lipids and plasma malondialdehyde achieved by Atorvastatin. They reverse the increase of in vitro peroxidation caused by Atorvastatin but do not achieve the decreases observed in patients not receiving lipid-lowering drugs. These results indicate that the antioxidant effect of Vitamin E is attenuated when given in conjunction with this statin. © 2004 Elsevier Ireland Ltd. All rights reserved.

Approximately 15-20% of type 1 diabetic patients exhibit parietal cell antibodies (PCAs) targeting gastric H+/K+ATPase. We examined whether iron deficiency anemia, pernicious anemia, and autoimmune gastritis, which may predispose to gastric tumors, were more frequent in PCA+ than in PCA- patients.Gastric biopsies from 88 consecutively recruited type 1 diabetic patients (51 men and 37 women, 47 PCA+ and 41 PCA-, aged 42 +/- 13 years) were evaluated using the updated Sydney system. Immunostaining was done for parietal cells, B- and T-cells, enterochromaffin-like (ECL) cells, and Helicobacter pylori (HP). PCAs were assayed by indirect immunofluorescence, H+/K+ATPase antibodies by enzyme immunoassay, and HP by serology, urea breath test, and histology. Pentagastrin tests were performed in 42 subjects.Autoimmune gastritis (AG) was present in 57% of PCA+ and 10% of PCA- cases (OR 12.5, P0.0001). PCA positivity (beta = 1.44; P = 0.04) and hypergastrinemia (beta = 0.01; P = 0.026), but not HP, age, diabetes duration, sex, and HLA-DQ type were risk factors for AG. Iron deficiency anemia (OR 3.9, P = 0.015), pernicious anemia (OR = 4.6, P = 0.022), and hypochlorhydria (OR = 20.0, P = 0.0002) were more frequent in AG+ individuals. HP infection was present in 47 patients but did not influence corpus histology or gastrinemia. (Pre)malignant lesions were found in 26% of PCA+ subjects: ECL cell hyperplasia in 7 AG+ patients, comprising 1 with a gastric carcinoid tumor, and corpus intestinal metaplasia in 11 AG+ patients, including 1 with linitis plastica.PCA+ type 1 diabetic patients should be screened for autoimmune gastritis, iron deficiency, and pernicious anemia. Particularly hypergastrinemic PCA+ patients with autoimmune gastritis are at increased risk for (pre)malignant gastric lesions.

A worldwide increase in the incidence of childhood type 1 diabetes has been observed. Because in various countries the majority of new type 1 diabetic patients are diagnosed in adulthood, we investigated whether the rising incidence of this disorder in children reflects a global increase in the incidence of diabetes or a shift toward earlier clinical presentation.The incidence of type 1 diabetes presenting before age 40 years was prospectively measured in the Antwerp district over a 12-year period (1989-2000). The completeness of ascertainment was evaluated by the capture-recapture method. Trends in incidence during the study period were analyzed by Poisson regression.The incidence of type 1 diabetes diagnosed before age 40 years remained constant over the 12-year period, averaging 9.9 cases per 100,000 individuals per year. The incidence was similar in both sexes under age 15 years, but a marked male excess was noted for adult-onset disease, in particular after age 20 years, resulting in a male-to-female ratio of 0.9 under age 15 years vs. 1.6 thereafter (P = 0.001). During the 12-year observation period, there was a significant tendency toward increasing incidence under age 15 years at the expense of a decreasing incidence between ages 15 and 40 years (P = 0.025). The annual increase in incidence averaged 1.8% under age 15 years and 5.0% under age 5 years (P = 0.06).Our results indicate that in Belgium, the increasing incidence of childhood type 1 diabetes-especially for children under age 5 years-is not attributable to a global increase in disease incidence, but rather to earlier clinical manifestation. The results suggest that an environmental factor may preferentially accelerate the subclinical disease process in young diabetes-prone subjects.

OBJECTIVE—Delayed gastric emptying and/or gastrointestinal symptoms occur in 30–50% of diabetic patients. Known contributing factors are autonomic neuropathy and acute hyperglycemia, but the role of gastric autoimmunity has never been investigated, although 15–20% of type 1 diabetic patients exhibit parietal cell antibodies (PCAs). We studied gastric motility in diabetes in relation to PCA status, autonomic nerve function, HbA1c, thyroid-stimulating hormone (TSH), Helicobacter pylori (HP), acid production, and gastric histology. RESEARCH DESIGN AND METHODS—Gastric emptying of solids and liquids (measured by 13C-octanoic acid and 13C-glycine breath tests, respectively) was tested in euglycemic conditions in 42 type 1 diabetic patients (male/female: 29/13; 15 PCA+; mean age 40 ± 15 years; mean HbA1c 7.8 ± 0.9%). Gastrointestinal symptoms, autonomic nerve function (Ewing tests), PCA status (indirect immunofluorescence), gastric histology, and acid secretion (pentagastrin) were assessed. RESULTS—Solid gastric emptying was delayed in 40% and liquid emptying in 36% of patients. Gastric motility did not correlate with symptoms. PCA status, gastric morphology, and acid secretion were similar in those with and without gastroparesis. HbA1c level (β = 1.34, P = 0.011) was the only risk factor for delayed solid emptying in a logistic regression model testing HbA1c, autonomic nerve function, PCA, HP status, age, sex, diabetes duration, and TSH. Half-emptying time for liquids correlated with TSH level (r = 0.83, P < 0.0001) and autonomic neuropathy score (r = −0.79, P = 0.001). CONCLUSIONS—We found that ∼50% of type 1 diabetic patients studied had delayed gastric emptying that did not correlate with symptoms. Gastric autoimmunity did not contribute to diabetic gastroparesis. Metabolic control was worse in patients with delayed solid emptying.

A quarter of type 1 diabetic patients have thyrogastric autoantibodies (thyroid peroxidase and gastric parietal cell antibodies). Clinical, immune, and genetic risk factors help predict antibody status. First degree relatives of these patients may also frequently exhibit these antibodies. We assessed the prevalence of thyrogastric antibodies and dysfunction in first degree relatives in relation to age, gender, human leukocyte antigen-DQ type, beta-cell antibody (islet cell, glutamic acid decarboxylase-65, and tyrosine phosphatase antibodies), and proband thyrogastric antibody status. Sera from 272 type 1 diabetic patients (116 men and 156 women; mean age, 27 +/- 18 yr; duration, 10 +/- 9 y), 397 first degree relatives (192 men and 205 women; parents/siblings/offspring, 48/222/127; age, 22 +/- 10 yr), and 100 healthy controls were tested for islet cell antibodies and gastric parietal cell antibodies by indirect immunofluorescence and for tyrosine phosphatase, glutamic acid decarboxylase-65, and thyroid peroxidase antibodies by radiobinding assays. Glutamic acid decarboxylase-65 antibodies were present in 68% and 5%, islet cell antibodies were present in 36% and 2.5%, tyrosine phosphatase antibodies were present in 45% and 0.5%, thyroid peroxidase antibodies were present in 21% and 4.5%, and gastric parietal cell antibodies were present in 18% and 11% of diabetic patients and relatives, respectively. The presence of thyroid peroxidase antibodies in relatives was determined by age (beta = 0.22; P = 0.0001) and proband thyroid peroxidase antibodies status (beta = -2.6; P = 0.002; odds ratio = 11.1). Gastric parietal cell antibody positivity in relatives was associated with age (beta = 0.04; P = 0.026). Gastric parietal cell antibody-positive compared with gastric parietal cell antibody-negative relatives were more likely to have gastric parietal cell antibody-positive probands (P = 0.01; odds ratio = 3.0). beta-Cell antibody status and human leukocyte antigen-DQ type did not influence thyrogastric antibody status in relatives. (Sub)clinical dysthyroidism was found in 3%, iron deficiency anemia was present in 12% (26% gastric parietal cell antibody-positive and 9% gastric parietal cell antibody-negative subjects; P = 0.009), and pernicious anemia was found in 0.5% (5% gastric parietal cell antibody-positive and 0% gastric parietal cell antibody-negative subjects; P = 0.012) of relatives. They had less thyroid dysfunction (P < 0.0001) and pernicious anemia (P = 0.018) than diabetic probands. In conclusion, thyrogastric antibodies and dysfunction are more prevalent in type 1 diabetic patients than in first degree relatives. The presence of these antibodies in relatives is associated with age and proband antibody status, but not with beta-cell antibodies or human leukocyte antigen-DQ type.

In erythrocytes from diabetic patients, increased membrane lipid peroxidation might lead to abnormalities in composition and function. To study this relationship, we investigated the effects of a moderate pharmacologic dose of vitamin E for 1 y on erythrocyte membrane peroxidation in vitro and on its fatty acid composition, antioxidant capacity and rheological function. In a random and double-blind manner, type 1 diabetic patients (n = 44) were assigned to the following two groups: Group S received 250 IU (168 mg) d-alpha tocopherol 3 times daily for 1 y. Group P received placebo for 6 mo followed by d-alpha-tocopherol for an additional 6 mo. Variables were monitored every 3 mo. After 3 mo of supplementation, serum vitamin E doubled (P < 0.0005), thiobarbituric acid reactive substances in erythrocyte membranes incubated with tert-butyl hydroperoxide decreased by 25% (P = 0.006) and the lagtime of fluorescence increased from 28 +/- 16 to 41 +/- 28 min (P = 0.028). Patients who did not respond to supplementation (13 of 44) had lower serum lipids (P = 0.017) and body mass index (P = 0.024). We did not detect any significant effects of vitamin E supplementation on membrane lipid composition, antioxidant capacity or blood viscosity. Continuing supplementation for up to 1 y did not further affect serum vitamin E or membrane peroxidation. Stopping supplementation was followed by a return to inclusion values. These results show that the decrease in erythrocyte membrane peroxidation after vitamin E supplementation is moderate, saturable, reversible, restricted to some individuals and has no detectable effect on erythrocyte composition and function.

There exists much evidence suggesting a major role for the oxidized low density lipoprotein (LDL) and VLDL particles in the pathogenesis of atherosclerosis. Although obesity is characterized by dyslipidemia, less is known about the oxidation capacity of lipoproteins in obese subjects. We measured the oxidizability in vitro in 21 premenopausal women with a BMI of 36.7+/-5.4 and compared them to 18 age-matched controls (BMI 21.9+/-1.8). The oxidizability of the non-HDL fraction is evaluated by measuring the fluorescence and thiobarbituric acid reactive substances (TBARS; MDA nM/mg non-HDL) at different time intervals of incubation. TBARS formation increased linearly with the increase of lipids both in non-obese and obese subjects. TBARS, measured every 30 min, increased in non-obese controls up to a maximum of 59.6 at 180' in contrast to a maximum of 77.1 at 180' (P < 0.001) in obese, but healthy, normocholesterolemic subjects. At each measurement the TBARS were significantly higher (P < 0.01-0.001) in obese subjects. Also the lag-time (period from zero to the start of the particle oxidation process) was significantly lower (92.5 vs. 123.4; P < 0.001) in obese subjects, when compared to lean controls. BMI correlates significantly with TBARS formation and its log transformed values (maximum P < 0.001). The lag-time was negatively related (n=39 total group, r=- 0.57, P < 0.001) to body weight and BMI. A significant relationship exists between TBARS formation (up to r=0.59) and triglyceride levels and a negative relationship exists with HDL-cholesterol levels. In vitro oxidizability of non-HDL lipoproteins is significantly increased in obese, non-diabetic subjects and related to increased body weight and triglyceride levels. Further studies are necessary to explore the underlying mechanisms for this phenomenon.

The long-term results of dietary treatment for obesity are often very poor. To predict the effect of a hypocaloric diet, it may be important to consider factors that could influence energy expenditure, especially those altered in obesity. Elevated plasma levels of glucagon are associated with obesity. In this study, the relationship of glucagon to resting metabolic rate (RMR) and glucose-induced thermogenesis (GIT) has been investigated. RMR and GIT, after ingesting 100 g glucose, were measured by indirect calorimetry in 25 obese premenopausal women (body mass index [BMI], 37.2 +/- 4.7 kg.m-2). RMR was significantly related to fat-free mass (FFM) (r = .50, P less than .005). A significant relation could be found between RMR and fasting glucagon levels (r = .36, P less than .05). Plasma glucose and insulin levels were not predictive for RMR. Mean GIT increased with increasing waist-hip circumference ratio (WHR) (r = .71, P less than 0.0001), confirming previous findings of our group. No relation was found between GIT and glucagon levels, neither in the basal state nor after glucose. The only important metabolic determinant was area under the curve (AUC) for glucose (r = .45, P less than .01), suggesting a higher GIT in obese women with impaired glucose intolerance. This suggests that the control of energy metabolism by the concentration of glucagon may be more important in the fasting state than after a meal. Plasma glucagon concentration should be considered in the evaluation of RMR.

Evaluation of the obesity- and smoking-related atherogenic risk in an obese population, was studied by retrospectively reporting the 6-month dietary effect on metabolic lipoprotein parameters. In a non-smoker group, a weight reduction programme exhibits a marked improvement (P less than 0.05) of HDL and LDL cholesterol and atherogenic index. Smoking during successful, therapeutic weight reduction inhibits this advantageous improvement of lipid parameters, especially in women. In regard to the lipoprotein related vascular risk, obesity treatment without modifying smoking habits is less advantageous.

Chromium and selenium contents of thirty-five commercially available enteral feed were determined by atomic-absorption spectrometry after previous wet ashing. Daily chromium intake, based on a 2,000 kcal (8,368 kJ) diet, was below the proposed safe and adequate intake for 78% of the feedings. Hypocaloric diets especially provided extremely low doses of this element. For selenium nearly 80% of the diets provided intakes below the recommended 50 micrograms/day. Selenium concentration was positively correlated with the protein content (p = 0.007), whilst fat content and selenium concentrations showed a significant negative correlation (p = 0.002). No definite correlation could be found between the selenium and chromium content of the food (p greater than 0.1).

Low vitamin E levels have been found to be a frequent side effect in patients on total parenteral nutrition (TPN). In the present study, the vitamin E content of fat emulsions (Intralipid) was measured and the influence of the intravenously administered lipid emulsion on plasma vitamin E levels was investigated. The majority of vitamin E was provided in the beta + gamma-tocopherol fractions (68.7% of total tocopherol). Vitamin E levels in patients were significantly lower (p less than 0.05) as compared to age- and sex-matched normal controls. Although sufficient amounts of vitamin E (16.9 +/- 0.8 IU daily) were infused, according to RDA requirements, E-plasma levels decreased even further during the course of TPN. It was concluded that high amounts of biologically less active tocopherol isomers are not sufficient to maintain vitamin E plasma levels. Supplemental sources of alpha-tocopherol are needed and more attention should be paid to the different tocopherol isomers in the quantitation of the daily allowances of vitamin E.

In the diabetic surgical patient, the effects of absolute or relative insulin deficiency are superimposed on the metabolic alterations of stressed nondiabetic subjects. Stress hyperglycemia, ketosis, and insulin resistance in the perioperative period should be interpreted differently when adequate parenteral nutrition must be provided in a diabetic. Insulin must be added following rigid but sliding scales. Depending on stress levels the administration of calories must be chosen from different possible energy substrates: carbohydrates, lipids, and amino acids. Only during high stress levels with concomitant severe insulin resistance does enrichment of the feeding solution with branched-chain amino acids seem appropriate. In metabolically more stable situations, preference should be given to carbohydrate and fat calories. Although recent innovations such as nonglucose carbohydrates and medium-chain triglyceride infusions are very promising, their use in diabetics should be under close clinical and metabolic control to avoid major side effects such as intracellular substrate accumulation and ketosis.

The close relationship between vitamin E, lipid and prostaglandin metabolism stresses the need for an accurate definition of the status of vitamin E in a diabetic population. Plasma vitamin E and plasma lipids were determined in 34 type I and 21 type II normocholesterolemic and in 7 hypercholesterolemic diabetics. They were also measured in 62 age- and sex-matched controls, 34 normocholesterolemic controls for type I, 21 normocholesterolemic controls for type II and 7 hypercholesterolemic individuals. Plasma vitamin E levels were not significantly different in type I and type II diabetics as compared to their respective control groups. Vitamin E levels were significantly increased (p less than 0.001) in hypercholesterolemic individuals, both in diabetics and in non-diabetics. The vitamin E/cholesterol ratio in these subjects was, however, not different from that of normocholesterolemic. Plasma vitamin E was correlated with plasma lipids, especially with total and LDL-cholesterol (p less than 0.001). Since vitamin E is mainly transported by plasma lipoproteins, these strong correlations suggest that changes in plasma vitamin E should be considered as an epiphenomenon of altered plasma transport capacity.

The aetiology and pathogenesis of the Chronic Fatigue Syndrome (CFS) are still largely unresolved. Accompanying metabolic disorders such as selective n-6 fatty acid depletion suggest that oxidative stress and more specifically lipid peroxidation might play a role in its pathogenesis. In order to investigate this hypothesis, oxidant-antioxidant status and its impact on lipoprotein peroxidation in vitro was examined in 61 patients with unexplained fatigue lasting more than 1 month. They were subdivided into 2 groups: group CFS+ (33 subjects) fulfilled the 1988 Center of Disease Control criteria for CFS and group CFS- did not but was similar as regards age, sex distribution and clinical characteristics. Antioxidant status was similar in the 2 groups except for lower serum transferrin in the CFS + (mean (95 % CI) 2.41 (2.28-2.54) versus 2.73 (2.54-2.92) g/L in the CFS-, p = 0.009) and higher lipoprotein peroxidation in vitro: 6630 (5949-7312) versus 5581 (4852-6310) nmol MDA/mg LDL and VLDL cholesterol x minutes, p = 0.035). CFS intensified the influence of LDL cholesterol (p = 0.012) and of transferrin (p = 0.045) on peroxidation in vitro, suggesting additional pro-oxidant effects. These results indicate that patients with CFS have increased susceptibility of LDL and VLDL to copper-induced peroxidation and that this is related both to their lower levels of serum transferrin and to other unidentified pro-oxidising effects of CFS.