Reactive hypoglycemia

Spontaneous hypoglycemia: should we mind the gap? Long-term follow-up of healthy people who met Whipple's triad criteria

CONTEXT

Patients discharged as "healthy" with the symptoms of spontaneous hypoglycemia, commonly known as Whipple's triad, need more attention.

OBJECTIVE

Characterization and long-term follow-up of symptom development in patients with spontaneous hypoglycemia discharged as "healthy". The objective was to ascertain whether any conditions related to the symptoms were diagnosed during the follow-up period.

METHODS

Retrospective analysis of patient data and evaluation of a specific questionnaire on the development of symptoms of spontaneous hypoglycemia. In addition, patient questionnaires were evaluated and primary care physicians were asked about possible diseases not recorded at baseline that occurred during the follow-up period.

SETTING

Center for Endocrinology, Diabetology, and Osteology at the University Hospital Marburg, Inpatient Department, Germany.

PATIENTS

All patients who presented to our center for the 72-hour fast between 2005 and 2018 and were discharged without an internal medicine diagnosis were included.

INTERVENTIONS

Survey by questionnaire, via telephone interview.

MAIN OUTCOME MEASURES

Patient-reported information on current symptoms compared to original symptoms, diagnosis of insulinoma or diabetes mellitus during follow-up, matched with primary care physician data, and metabolic and biometric data such as body mass index (BMI), homeostasis model assessment for insulin resistance (HOMA IR), insulin sensitivity Matsuda Index (ISI-M), and area under the curve.

RESULTS

A total of 41 datasets were evaluated at baseline and 38 patients were followed for an average of approximately 10 years. In total, 61% of respondents still reported the same symptoms as at baseline. No insulinoma was missed in these patients. Only two of the 38 patients developed diabetes mellitus.

CONCLUSION

The high percentage of patients who are discharged as "healthy" and still have symptoms after many years is disturbing. It is possible that the symptoms are not due to low blood glucose. We urge caution with use of the term "healthy". We advocate a multidisciplinary therapeutic approach after an organic cause of hypoglycemia has been ruled out. Psychosomatic treatment seems to be useful. In addition, more research should be conducted on this topic.

Reactive hypoglycemia owing to an intrahepatic congenital portosystemic shunt in an older patient

An 85-year-old woman was admitted to our hospital because of hypoglycemia and impairment of consciousness several hours after breakfast. Because the hypoglycemia predominantly occurred 2-4 h after meals, we diagnosed reactive hypoglycemia. An oral glucose tolerance test showed prolonged hyperinsulinemia following the postprandial hyperglycemia, with a subsequent rapid decrease in blood glucose concentration. The post-stimulus plasma C-peptide concentration was relatively low compared to the plasma insulin concentration. Abdominal computed tomography revealed an intrahepatic congenital portosystemic shunt (CPSS). On the basis of these findings, we concluded that the reactive hypoglycemia was induced by the CPSS, via a reduction in hepatic insulin extraction. Treatment with an alpha-glucosidase inhibitor resolved the reactive hypoglycemia. CPSS comprises anomalous vascular connections between the portal vein and the systemic venous circulation, and reactive hypoglycemia is a rare complication of this malformation, which has most frequently been reported in children, with only a few cases reported in adults. However, this case indicates that even in adult patients, imaging studies should be conducted to rule out CPSS as the cause of the reactive hyperglycemia.

Nutritional guidelines for the management of insulin resistance

Obesity and its related co-morbidities, namely type 2 diabetes (T2D), pose a significant global public health problem. Insulin resistance (IR) in muscle and liver is the core pathophysiologic defect that underlies obesity preceding and predicting the onset of T2D in susceptible humans. There is a broad population with IR that has no indication for prescription of medications, who still need medical consultation and specific advice in this respect. This prevalent need can be achieved by appropriate diet, exercise, and other behavioral therapies for lifestyle interventions. Despite a well-recognized role of IR in the progression to metabolic diseases, no specific nutritional recommendations exist to manage this condition, to the best of our knowledge. An international panel of experts reviewed and critically appraised the updated literature published about this topic. This review primarily examines the evidence for areas of consensus and ongoing uncertainty or controversy about diet and exercise approaches for IR. The aim of this article is to present the most common IR states, namely obesity and Polycystic Ovary Syndrome (PCOS), and provide nutritional advice to manage IR, hyperinsulinemia, and reactive hypoglycemia. These nutritional guidelines could prevent progression or worsening of IR with resultant beta-cell failure and, as a result, T2D.

Diagnosis of insulinoma using the ratios of serum concentrations of insulin and C-peptide to glucose during a 5-hour oral glucose tolerance test

The 72-hour fast test is the current standard for the diagnosis of insulinoma. However, to conduct this test patients require hospitalization due to the chance of severe hypoglycemic episodes. Thus, it is costly and stressful for the patient. An out-patient test would serve the patient better and be more economical. Our aim was to evaluate the value of insulin to glucose and C-peptide to glucose ratios during a prolonged 5-hour oral glucose tolerance test (5-hour OGTT) in qualitative diagnosis of insulinoma, and to identify the optimal threshold for clinical screening. Initially, 15 subjects with pathological insulinoma and 12 control subjects with reactive hypoglycemia were enrolled in the study. A further 75 subjects with symptoms of hypoglycemia as a chief complaint at their initial clinic visit were subsequently screened. Serum insulin, C- peptide levels and blood glucose were quantified after a 5-hour OGTT in all participants and the ratios of serum concentrations of insulin and C-peptide to glucose were calculated. Subjects with insulinoma had significantly different insulin-to-glucose and C-peptide-to-glucose ratios from reactive hypoglycemia at the times of fasting, 4-hour post glucose load and 5-hour post glucose load. Higher specificity (73.08%) and sensitivity (82.67%) were achieved with the combined insulin-to-glucose ratio at the 5-hour post load and the C-peptide-to-glucose ratio at fasting. In combination, ratios of insulin and C-peptide release relative to blood glucose levels, measured during a 5-hour OGTT, may have important clinical value in the diagnosis of insulinoma.

Post-absorptive glucose lowering in normal healthy individuals: an epidemiological observation

Post-absorptive glucose lowering (PALG) is observed in individuals with glucose intolerance and in healthy individuals. We report a prevalence of about 23% among healthy Asian Indians. Individuals with PALG are characterized by leaner phenotype, low body fat percentage, increased insulin sensitivity and higher fasting glucose levels.

Hypoglycemia

Under physiologic conditions, glucose plays a critical role in providing energy to the central nervous system. A precipitous drop in the availability of this substrate results in dramatic symptoms that signal a medical emergency and warrant immediate therapy aimed at restoring plasma glucose to normal levels. A systemic approach to the differential diagnosis is useful in identifying the cause of hypoglycemia. Once established, a specific and/or definitive intervention that addresses that underlying problem can be implemented. In most cases, this systemic approach to diagnosis and therapy is rewarded with a good outcome for the patient.

Postprandial reactive hypoglycemia in an oldest-old patient effectively treated with low-dose acarbose

We recently encountered a 96-year-old Japanese woman who suffered from frequent hypoglycemia. Endocrinological and imaging data eliminated the possibility of insulinoma, whereas oral glucose tolerance testing revealed impaired glucose tolerance and subsequent reactive hypoglycemia. The patterns between insulin or C-peptide secretions and glucose excursions demonstrated that the discrepancy occurred in the late postprandial stage. Administration of small doses of alpha-glucosidase inhibitor (alpha-GI) dramatically inhibited the rapid rise and subsequent precipitous fall of plasma glucose. Reactive hypoglycemia may be one of the important cause of hypoglycemia in the elderly, and alpha-GI could effectively and safely prevent such hypoglycemic attacks in those patients.

Biochemical hypoglycemia in female nurses during clinical shift work

Female nurses in Japan commonly experience symptoms similar to those of hypoglycemia while working. Biochemical hypoglycemia can lead to impaired cognition; thus, it is important to know the prevalence of biochemical hypoglycemia among nurses. Five hundred and sixty-eight female nurses (53% of the target population) in four hospitals completed questionnaires. They determined their blood glucose levels 12 times, at four points during three shifts. Fifty-seven nurses (10%) recorded biochemical hypoglycemia (<or=3.0 mM) at least once. Multivariate analysis revealed three independent risk factors for hypoglycemia: higher coefficient of variation of blood glucose level, lower body mass index, and not smoking. Subjective symptoms were not associated with hypoglycemia. More research on hypoglycemia is needed in order to improve working environment for nurses.

Classification of hypoglycemic disorders

The classification of symptoms of hypoglycemia, namely, reactive versus fasting, based on segregation by timing in relation to meals is no longer useful from a clinical point of view. Every patient with neuroglycopenic symptoms, regardless of relation to food ingestion, requires comprehensive evaluation. Identification of the possible cause of hypoglycemia and selection of diagnostic modalities are facilitated by considering whether the patient appears to be healthy (with or without compensated coexistent disease) or ill (with a disease known to have a proclivity to develop hypoglycemia, or is hospitalized). Medications may mediate hypoglycemia in anybody, whether they appear healthy or ill.

Pancreatic alpha-cell function in idiopathic reactive hypoglycemia

Idiopathic reactive hypoglycemia (IRH) is a well-documented but overdiagnosed syndrome. The presence of transient hypoglycemia and enhanced insulin secretion and/or increased insulin sensitivity before the onset of IRH is well documented. However, the data regarding glucagon secretion are sparse. Therefore, this study assessed glucagon and insulin responses to (1) oral ingestion of 100 g glucose oral glucose tolerance test (OGTT) and (2) a 100-g protein meal after an overnight fast in a randomized sequence at intervals of 7 to 10 days in five subjects with previously well-documented IRH and six normal subjects. Basal plasma glucose and insulin levels were not significantly different in both groups. However, basal glucagon was significantly higher (P < .025) in IRH subjects (347 +/- 83 ng/L) compared with normals (135 +/- 20 ng/L). In IRH subjects during the OGTT, hypoglycemia (2.7 +/- 0.11 mmol/L) occurred at 150 +/- 16 minutes and was preceded by a markedly higher (P < .01) peak glucose concentration (11.7 +/- 0.6 mmol/L) at 36 +/- 6 minutes in comparison to normals (8.8 +/- 0.4 mmol/L), indicating the presence of impaired glucose tolerance in these subjects. Similarly, the plasma insulin increase was significantly higher (P < .01) but delayed in IRH subjects compared with normals. In contrast, glucagon suppression was not significantly different in both groups, although glucagon failed to increase following hypoglycemia in IRH. During a protein meal, plasma glucose declined in both groups, with a significantly (P < .05) greater decrease in IRH subjects (-0.8 +/- 0.2 mmol/L) compared with normals (0.5 +/- 0.1 mmol/L). However, the glucagon increase was significantly (P < .01) blunted in IRH subjects (61% +/- 15%) in comparison to normals (152% +/- 39%). Thus, basal hyperglucagonemia with normal glucose concentration may suggest the presence of a hyposensitivity of the glucagon receptor in IRH. Moreover, the lack of appropriate suppression during the OGTT despite marked hyperglycemia, the lack of an increase at the onset of hypoglycemia, and the inhibited response to a protein meal in IRH subjects compared with normals denote altered glucagon secretion in IRH. Therefore, it is likely that glucagon receptor downregulation and impaired glucagon sensitivity and secretion may contribute to postprandial hypoglycemia in IRH.

Increased insulin sensitivity and basal insulin effectiveness in postprandial reactive hypoglycaemia

Glucose clamp experiments have shown that patients with reactive postprandial hypoglycaemia (PRH) frequently have an increased glucose disposal, but the relative involvement of insulin sensitivity (SI) and glucose effectiveness (Sg) in this process remains unknown. The minimal model approach was used to compare 13 patients in whom moderate reactive hypoglycaemia ( < 3.3 mmol) had been previously diagnosed and 13 matched controls. The intravenous glucose tolerance test (IVGTT, 0.5 g/kg glucose IV) with 0.02 U/kg insulin given at the 19th min and frequent sampling over 180 min shows that PRH patients exhibit a higher glucose tolerance coefficient Kg (2.99 +/- 0.26 vs 2.19 +/- 0.12; P < 0.02), higher SI [22.9 +/- 6.4 vs 7.18 +/- 0.14 min-1/(microU/ml). 10(-4); P < 0.01] and higher Sg (3.84 +/- 0.35 vs 2.92 +/- 0.79 min-1. 10(-2); P < 0.05). The increase in Sg is explained by an increase in its component basal insulin effectiveness (BIE: 1.2 +/- 0.27 min-1.10(-2) in PRH subjects vs 0.58 +/- 0.07; P < 0.05) rather than an increase in Sg at zero insulin. The increase in BIE results from the high values of SI. In 4 PRH subjects SI and Sg were within the normal range, and the increase in Kg evidenced in the 9 others was explained by an increase in SI alone in 3 cases, in Sg alone in 1 case, and both SI and Sg in 5 cases. Thus, in sedentary subjects, the previously reported rise in tissue glucose assimilation is mainly explained by an increased insulin-mediated glucose disposal rather than non-insulin-mediated glucose disposal.

Evaluation of a standardized hyperglucidic breakfast test in postprandial reactive hypoglycaemia

The oral glucose tolerance test is not specific for diagnosing postprandial reactive hypoglycaemia, since it too frequently induces low blood glucose values in subjects who have never complained of symptoms of this. By contrast, the mixed meal tests are deceptive for this purpose because they do not induce hypoglycaemia in subjects who have complained of of hypoglycaemic symptoms. We investigated the frequency of hypoglycaemia after a standardized hyperglucidic breakfast test in three groups of subjects:group A, 43 control subjects; group B, 38 postprandial reactive hypoglycaemic patients; group C, 1193 asymptomatic subjects undergoing assessment of glycoregulation. In the 38 subjects with suspected reactive hypoglycaemia the mean blood glucose nadir was 3.48 +/- 0.08 mmol/l, i.e. lower than in control subjects (4.83 +/- 0.13 p < 0.0001). Blood glucose levels less than 3.3 mmol/l were found in 47.3% of subjects with suspected postprandial reactive hypoglycaemia (group B), i.e more frequently than in control subjects (group A: 2.2% p = 1.6 x 10(-6)) and asymptomatic subjects (group C: 1% p = 8 x 10(-22)). This markedly higher frequency of low blood glucose values in subjects with postprandial symptoms compared with control and asymptomatic subjects suggests that this test detects a tendency to hypoglycaemia after a standardized hyperglucidic breakfast. Since this test mimics average French eating habits, the results suggest that the patients undergo such symptoms in their everyday life, and that the hyperglucidic breakfast test is a simple alternative to ambulatory glucose sampling for diagnosis of postprandial reactive hypoglycaemia.

Reactive hypoglycaemia in association with disordered islet function and abnormal hepatic glucose-6-phosphatase activity: response to diazoxide

Severe reactive hypoglycaemia was confirmed in a non-diabetic male patient by a counter-regulatory hormone (GH, cortisol and catecholamine) response to profound hypoglycaemia induced by an intravenous glucose load. There was also evidence of disordered pancreatic islet cell paracrine regulation with hyperinsulinaemia and absent glucagon response to hypoglycaemia. A defect in the patient's hepatic glucose-6-phosphatase enzyme system was documented. Because of severe symptoms, dietary control was insufficient, but the patient responded clinically and biochemically to 18 months of oral diazoxide therapy. He also showed good biochemical response to a single dose (100 micrograms IM) of the somatostatin analogue octreotide.

Glucagon: prehospital therapy for hypoglycemia

STUDY OBJECTIVE

This study evaluated the efficacy of glucagon for prehospital therapy of hypoglycemia in patients without IV access.

DESIGN

Prospective clinical trial.

SETTING

Prehospital in a busy, urban emergency medical services system.

TYPE OF PARTICIPANTS

Fifty consecutive patients presenting with documented hypoglycemia (ChemStrip BG less than or equal to 80 mg/dL) and symptoms of decreased level of consciousness, syncope, or seizure were enrolled.

MEASURES AND MAIN RESULTS

Data collected included pretreatment (ChemStrip BG) and post-treatment serum glucose (hospital assay) as well as assessment of level of consciousness by a quantitative measure, the Glasgow Coma Score, and by a qualitative scale (0 to 3). The mean pretreatment blood glucose of 33.2 +/- 23.3 mg/dL increased after treatment to 133.3 +/- 57.3 mg/dL. Qualitative level of consciousness increased from a mean of 1.26 +/- .96 to 2.42 +/- .94 and Glasgow Coma Score increased from a mean of 9.0 +/- 4.19 to 13.04 +/- 3.68. The mean time until response was 8.8 minutes in those who responded to both level of consciousness criteria 82% (41 of 50). Glucagon administered for hypoglycemia resulted in a glucose increase in 98% (49 of 50) with headache as the only side effect noted in 4% (two of 50) of patients (P less than .0001).

CONCLUSION

Glucagon is safe and effective therapy for hypoglycemia in the prehospital setting.

Somatostatin effect in postprandial hypoglycemia

The effect of the administration of oral glucose with or without a simultaneous intravenous somatostatin infusion on blood glucose, immunoreactive insulin, C-peptide, and glucagon levels in seven patients with idiopathic postprandial hypoglycemia was studied. Oral glucose alone induced an excessive insulin response and hypoglycemia, whereas a slight suppression in glucagon levels without any increase at the hypoglycemic nadir was observed. The simultaneous administration of somatostatin significantly reduced the insulin response and induced a slower rise of blood glucose; no hypoglycemia developed. Only minor variations in glucagon were observed with respect to the basal test. A rebound in insulin, C-peptide, and glucagon levels was observed at the end of the somatostatin infusion. These data show that somatostatin can suppress glucose-induced hypoglycemia in these subjects, thus suggesting that its long-acting analogues might be worth a therapeutic trial in severe idiopathic postprandial hypoglycemia.

The effects of blood sugar level changes on cognitive function, affective state, and somatic symptoms

In an attempt to find out whether decreased blood sugar level is associated with impaired cognitive function, adverse emotional changes, or somatic symptoms, 35 subjects who suspected that they had hypoglycemia were given 5-hr glucose tolerance tests (GTTs). Nine blood samples were taken during the GTT, and the subject's mood, Serial Sevens Test (SST) performance, and somatic symptom reports were recorded on each occasion of sampling. The subjects reported significantly more negative affect and somatic symptoms after glucose nadir than before nadir. SST performance deteriorated at glucose nadir. These effects were more pronounced for subjects with high hypoglycemic index scores than for subjects with low index scores. The impairment in SST performance was greater for subjects who showed rapid decreases in blood sugar than for subjects who showed slow decreases. Dividing subjects by high and low nadirs did not reveal any differences in symptomatology.

Oral glucose tolerance test: indications and limitations

The oral glucose tolerance test has been used in many clinical situations. From a practical standpoint, however, it has limited applicability. The test should never be done to evaluate reactive hypoglycemia. It can be used in the patient whose fasting plasma glucose level is less than 140 mg/dl but who manifests symptoms compatible with complications of diabetes. Finally, all pregnant patients should undergo testing for gestational diabetes with a 50-g glucose screen at 24 to 28 weeks of pregnancy, and a subsequent glucose tolerance test should be performed if the result of that test is abnormal.

Symptomatic reactive hypoglycemia during glucose tolerance test in lithium-treated patients

Glucose, insulin, glucagon, and cortisol responses during a five-hour oral glucose tolerance test (GTT) were evaluated in nine patients with bipolar affective disorders who were receiving lithium treatment and in seven control patients with bipolar affective disorders who were not receiving any treatment. Both the lithium-treated and the control patients were in stable mood at the time of GTT. During GTT mean nadir serum glucose of 48 +/- 2 mg/dL in the lithium-treated patients was significantly lower (P less than 0.001) than mean nadir serum glucose of 62 +/- 2 mg/dL observed in the control subjects. Seven of these nine lithium-treated patients, but none of the control patients, experienced hypoglycemic symptoms coinciding with low serum glucose concentration. In response to hypoglycemia, mean serum cortisol significantly rose (P less than 0.01) to 22 +/- 3 micrograms/dL in the lithium-treated patients, whereas mean serum cortisol levels gradually declined to 10 +/- 2 micrograms/dL in the control patients at 300 minutes. Despite symptomatic postglucose hypoglycemia, plasma glucagon levels in the lithium-treated patients were similar to those observed in the control patients. These findings suggest that chronic lithium treatment is associated with a symptomatic and biochemical hypoglycemia during GTT, which is characterized by a rise in serum cortisol but by lack of appropriate rise in plasma glucagon concentrations.

Hypoglycemia: causes, neurological manifestations, and outcome

During a 12-month prospective study there were 125 visits to the Harlem Hospital Emergency Room for symptomatic hypoglycemia. Sixty-five patients had obtundation, stupor, or coma; 38 had confusion or bizarre behavior; 10 were dizzy or tremulous; 9 had had seizures; and 3 had suffered sudden hemiparesis. Diabetes mellitus, alcoholism, and sepsis, alone or in combination, accounted for 90% of predisposing conditions; others included fasting, terminal cancer, gastroenteritis, insulin abuse, and myxedema. Average blood glucose levels were lower among comatose than among obtunded patients, but overlap was considerable, and overall there was little correlation among cause, blood glucose levels, and symptoms. Although mortality was 11%, only one death was attributable to hypoglycemia per se, and only four survivors had focal neurological residua.

Relativism in the diagnosis of hypoglycemia

It is a common assumption in our medicalized society that diagnostic labels simply describe facts, rather than express the diagnostician's own concepts of anomaly and dysfunction. Variability among physicians in diagnosing reactive hypoglycemia illustrates some fundamental problems of reifying the concept of disease. Although most physicians consider the same three variables in this diagnosis, their interpretations of them may vary widely. The disagreement centers around the credibility given standardized, technological measures versus the individual patient's experiences. Most physicians consider lab tests definitive in diagnosing hypoglycemia, while bypassing patients' subjective reports. Some, however, rely more heavily on patient-generated criteria to diagnose the condition. This lack of agreement about the criteria for diagnosing hypoglycemia, coupled with the psychological nature of many of the symptoms, make hypoglycemia and psychological disorder reasonable alternative diagnoses. These diagnoses, in turn, have very different effects on patients. These variations in physicians' diagnostic criteria illustrate that diagnostic labels are cognitive constructs, subject to the influence of the physician's culture, expectations, values and interests.

The therapeutic potential of glucose tolerance factor

Glucose Tolerance Factor (GTF) is synthesized in vivo from absorbed dietary chromium, and acts as a physiological enhancer of insulin activity, binding to insulin and potentiating its action about three-fold. Since GTF is well absorbed orally, the development of sufficiently concentrated and stable supplementary sources of this agent may enable convenient and physiologically appropriate pharmacological modulation of insulin activity. A review of the numerous physiological actions of insulin suggests a number of therapeutic applications for GTF, in such diverse ailments as diabetes mellitus, hyperlipidemia, reactive hypoglycemia, obesity, cancer, protein malnutrition or malabsorption, endogenous depression, Parkinsonism, hypertension and cardiac arrhythmias. GTF supplementation may also have value in preventive medicine.

Diagnosis of pancreatic islet hyperplasia causing hypoglycemia in a patient with portacaval anastomosis

A patient with biopsy-proved biliary cirrhosis and previous gastrojejunostomy and portacaval anastomosis experienced episodes of severe hypoglycemia. She was found to have hyperinsulinemia and hyperglucagonemia. An oral glucose tolerance test showed postgastrectomy hypoglycemia. Results of the intravenous tolbutamide test were diagnostic for insulinoma, but results of the intravenous glucagon test and prolonged fast (96 hours) were not. Failure, on two occasions, to suppress C-peptide normally during insulin-induced hypoglycemia led to a diagnosis of pancreatogenous hyperinsulinemia. The pancreas showed a 10-fold increase in islet volume, with intensely positive staining with anti-insulin and anti-glucagon antiserums in addition to anti-somatostatin and anti-pancreatic polypeptide antiserums. Incidental findings at pancreatic exploration were a mesothelioma, which did not stain with anti-insulin antiserum, and, at autopsy one year later, a hepatoma.

Glucose-insulin interaction in obese individuals with asymptomatic reactive hypoglycemia

The interrelationship of glucose and insulin was investigated in obese nondiabetic subjects with asymptomatic reactive hypoglycemia. Results were compared to those obtained from obese control subjects and normal individuals. The diagnostic criteria for asymptomatic reactive hypoglycemia were the appearance of blood glucose values of 40 mg/dl and below during the postabsorptive phase of a 6-h OGTT and the absence of related symptoms. The blood glucose nadir occurred earlier in obese hypoglycemics than in obese controls. Maximum insulin response was similar in both obese groups, but occurred significantly later in obese hypoglycemics than in obese subjects without hypoglycemia and normal subjects. In obese hypoglycemics the blood glucose nadir was inversely proportional to the time of the insulin peak (i.e. the later the insulin peak the lower the blood glucose nadir) but correlated poorly to maximum insulin values. Delayed insulin response was found to be the major abnormality in asymptomatic reactive hypoglycemia and a probable cause of the decreased ability to maintain post-hyperglycemic glucose homeostasis. Decreased glucose tolerance in some obese hypoglycemics pre-treated with prednisolone suggests that asymptomatic reactive hypoglycemia could be the manifestation of an early diabetic stage.