| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Department of Medicine (P.K., P.P.), Great Western Hospital, Swindon SN1 4JU, United Kingdom; North Cumbria Acute Hospitals National Health Service (NHS) Trust (S.S.), Whitehaven CA28 8JG, United Kingdom; and Departments of Endocrinology (A.B.G.), Hepatobiliary Surgery (S.B.), and Academic Radiology (R.H.R.), Barts and the London NHS Trust, London EC1A 7BE, United Kingdom
Address all correspondence and requests for reprints to: Prof. A. B. Grossman, Department of Endocrinology, St. Bartholomew’s Hospital, London EC1A 7BE, United Kingdom. E-mail: A.B.Grossman{at}qmul.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionCase ReportsDiscussionReferences |
|---|
Methods: We employed prolonged fasts and standard glucose tolerance tests, plus imaging studies and surgical pathology, in two patients with suspected insulinomas.
Results: The prolonged 72-h fast was normal in both the patients, whereas in both cases a prolonged oral glucose tolerance test clearly demonstrated the induction of severe hyperinsulinemia followed by significant hypoglycemia. Surgical removal confirmed the presence of insulinomas in each case.
Conclusions: Although the sensitivity of the 72-h fast is high and still plays an important role in the diagnosis of an insulinoma, we suggest that a "normal" test result should be interpreted in the light of clinical symptoms.
|
Introduction |
|---|
|
TopAbstractIntroductionCase ReportsDiscussionReferences |
|---|
|
Case Reports |
|---|
|
TopAbstractIntroductionCase ReportsDiscussionReferences |
|---|
A 34-yr-old male presented with a 10-yr history of symptoms consistent with paroxysmal hypoglycemia (sweating, tremor, dizziness, blurred vision, paresthesia), principally occurring after meals. He had previously sought medical advice but was advised that he had no organic problem. As the symptoms continued unabated, he was reviewed by a nutritionist who advised him to avoid "high-glycemic-index food." However, his symptoms continued and he started self-testing of his blood glucose by home capillary sampling, which showed levels of glucose in the range of 1.5–3.0 mmol/liter (27 - 54 mg/dl; normal, >4.0) when he was symptomatic. He had no significant past medical or family history, and clinical examination did not show any features suggestive of Addison’s disease or diabetes mellitus.
All standard baseline investigations were normal. Three overnight 18-h fasts revealed normal glucose levels (lowest reading, 4.5 mmol/liter; 81 mg/dl). He had a prolonged oGTT (Table 1
, upper portion), which showed induced symptomatic hypoglycemia with neuroglycopenia with concomitant hyperinsulinemia and raised C-peptide levels.
|
, lower portion); again, he developed neuroglycopenia associated with hypoglycemia and hyperinsulinemia. Subsequent to this he underwent a supervised 72-h fast (Table 2), followed by a brief period of exercise (5 min running up and down one flight of 20 stairs). Blood was sampled for glucose twice daily and at the end of the fast and exercise test, but at no point did he develop hypoglycemia or neuroglycopenia. Imaging with computerized tomography (CT), magnetic resonance (MR) (Fig. 1A), octreotide scintigraphy, and endoscopic ultrasound uniformly revealed the presence of a tumor in the head of the pancreas. A calcium stimulation catheter study localized the tumor to the vascular supply of the gastro-duodenal artery (Table 3). He underwent surgical resection of the tumor, which was 1.5 cm in diameter; histology confirmed the morphological and immunohistochemical features to be consistent with an insulinoma. Postoperatively, he became symptom free, and an oGTT now failed to induce hypoglycemia.
|
|
|
A 76-yr-old male presented to a neurologist with a 1-yr history of episodes of slurred speech, dizziness, and sweating. He felt these were more likely to occur if he missed breakfast, and eating could avert minor episodes. There was no significant past medical or family history. Clinical examination did not show any features suggestive of any specific pathology. Fasting blood glucose was normal, as were his routine basal biochemistry tests. Serum prolactin and glucagon levels were within normal limits. A 24-h urinary catecholamine collection was normal. An oGTT was arranged and showed significant hypoglycemia at 120 min (Table 4). However, subsequent to this, a supervised 72-h fast was organized, but at no stage did he become symptomatically or biochemically hypoglycemic. A diagnosis of "reactive hypoglycemia" was made and the patient was reassured. His symptoms appeared to subside and he declined any further investigations at that point. Two years later, he re-presented with a recurrence of his symptoms but felt that "glucose tablets" alleviated these symptoms. He mentioned that he could avoid symptoms by omitting breakfast but was then likely to have a severe attack after lunch. He had three additional overnight fasting glucose samples, and all were normal. However, he was observed in one of his attacks, during which his blood glucose fell to 2.0 mmol/liter (36 mg/dl), with a concomitant and inappropriate plasma insulin of 311 pmol/liter (44.4 mU/liter; normal, <18 pmol/liter in the presence of hypoglycemia) and undetectable sulfonylurea levels. A second 72-h fast was organized and again showed glucose levels persistently above 4 mmol/liter (72 mg/dl) throughout. On one occasion he showed a blood glucose level of 3.7 mmol/liter (67.3 mg/dl) with a concomitant plasma insulin of 27 pmol/liter (3.9 mU/liter) and proinsulin of 70 pmol/liter (normal, <5). A profile of his fasting gut peptides was normal (gastrin, 10 pmol/liter; normal, <40; vasoactive intestinal peptide, 11 pmol/liter; normal, <30; glucagon, 12 pmol/liter; normal, <50; neurotensin, 11 pmol/liter; normal, <100).
|
) confirmed the presence of a tumor, which was subsequently excised laparoscopically; it was 1.3 cm in diameter, and histology and immunocytochemistry confirmed the tumor to be an insulinoma with no abnormal features. There was postsurgical resolution of all symptoms, although no formal postoperative glucose tolerance test was performed. |
Discussion |
|---|
|
TopAbstractIntroductionCase ReportsDiscussionReferences |
|---|
The presence of Whipple’s triad (symptoms of hypoglycemia associated with low blood glucose and rapid alleviation of symptoms after glucose administration) with a concomitant detection of plasma immunoreactive insulin in the presence of low blood glucose levels (<2.2 mmol/liter; 40 mg/dl) constitutes the diagnostic basis of insulinoma as the cause of organic hypoglycemia (9, 10, 11, 12). An insulinoma can be associated with a variety of symptoms: weight gain is common. Neuroglycopenic symptoms may be particularly prominent and include confusion, bizarre behavior, personality change, seizure, and coma. Delay in diagnosis of insulinoma is common (13) because the symptoms largely precede occurrence of a readily apparent tumor and there may be misattribution of symptoms to psychiatric, cardiac, or neurological disorders (14, 15).
The diagnosis of an insulinoma is based on the suspicion of hypoglycemia as the cause of the symptoms, followed by confirmation of hypoglycemia when the patient is symptomatic in the presence of readily detectable insulin levels. The latter has become more difficult to demonstrate because newer insulin assays are more specific for monomeric insulin, and such patients may secrete a variety of insulin precursors and fragments. This should then be followed by attempts to localize the tumor using conventional radiology such as CT scanning and/or MR or high-sensitivity tests such as the preoperative calcium stimulation test and endoscopic or intraoperative ultrasound (16, 17, 18, 19). It is important to remember that invasive studies should only be carried out in patients with biochemical confirmation of the diagnosis.
However, a major diagnostic problem remains: the need to demonstrate hypoglycemia. Although most authorities advise a prolonged 72-h fast, it has recently been suggested that the great majority of cases will be successfully diagnosed even by 48 h (20). The purpose of the fast is 2-fold: the first is to diagnose hypoglycemia as the cause of the patient’s symptoms, and the second is to determine the etiology of the hypoglycemia. Serial glucose and insulin levels are obtained during the fast until the patient becomes symptomatic. A "normal" insulin level does not exclude the disease because the absolute insulin level is not elevated in all patients with insulinoma. Concomitant insulin levels at or above 6 mU/ml (41.8 pmol/liter) with elevated C-peptide levels of at least 0.2 nmol/liter (0.6 ng/ml) are necessary, with an absence of sulfonylurea or similar drugs in blood or urine (although some of the newer-generation oral hypoglycemics may not be readily assayable) (21, 22). Current assays use an insulin cutoff of greater than 3.0 mU/liter (18 pmol/liter). Measurements of proinsulin and C-peptide are also valuable in patients suspected of having organic hypoglycemia: proinsulin levels of greater than 40% of the total are suggestive of a malignant islet cell tumor (7). In both of our patients there was no hypoglycemia to prolonged fasting, although it could be argued that there may not have been rigorous adherence to the regimen imposed. This is a salient point, and it has been suggested that ketonuria should be sought to confirm suppression of insulin, together where appropriate with measurement of C-peptide levels and postglucagon glucose stimulation (9). Nevertheless, both patients were rigorously observed throughout the procedure, and we found no evidence that prohibited food or fluid intake occurred. It therefore seems highly probable that there was adherence to the fast.
Approximately 75% of patients with insulinomas are diagnosed after a 24-h fast and 90–94% at 48 h. Although some experts advocate conducting the prolonged fast for only 48 h (20, 23), others disagree, arguing that prolonging the fast up to 72 h minimizes misdiagnosis and maximizes the probability of diagnosing an insulinoma (24). Fajans and Vinik (25) suggested the "minifast," whereby patients are observed for symptoms, and plasma glucose, insulin, C-peptide, and sulfonylurea levels are measured after an overnight fast from 1800 h the previous evening. The high success rate of this test is based on the fact that some 75% of patients had developed symptoms within 18 h of beginning the fast (25). However, in our two cases we illustrate the presence of an insulinoma with a normal 72-h fast, whereas the oGTT demonstrated severe hyperinsulinemia followed by significant hypoglycemia. A literature search reveals only very few other such instances of "normal" 72-h fasts in patients with confirmed insulinoma (3, 4). There is also evidence in the literature suggesting a role for the 5-h oGTT in the diagnosis of insulinoma (26). Recent research has suggested using serial measurements of ß-hydroxybutyrate during the 72-h fast as an indicator of possible negative results: hypoglycemia secondary to hyperinsulinemia leads to suppression of ß-hydroxybutyrate, and in addition its measurement helps verify that the patient was indeed fasting throughout the study. A generous glucose response to intravenous glucagon at the end of the study also helps confirm mediation by insulin or a related insulin-like factor (27).
These cases illustrate that it is important to consider the diagnosis of an insulinoma in patients with reactive rather than fasting hypoglycemia. In both cases the presence of an insulinoma was confirmed immunohistochemically, and in both there was complete resolution of all symptoms on removal of the tumor. The first patient also had a biochemically normal response to hypoglycemia postoperatively, although, unfortunately, this was not performed in the second patient. However, we believe that the evidence is sufficiently strong to suggest that the prolonged 72-h fast may occasionally be misleading in patients with confirmed insulinomas. Although the sensitivity of the 72-h fast is indeed high and still plays an important role in the diagnosis of an insulinoma, a normal test should be interpreted in the light of clinical symptoms.
|
Footnotes |
|---|
First Published Online September 26, 2006
Abbreviations: CT, Computerized tomography; MR, magnetic resonance; oGTT, oral glucose tolerance test.
Received July 5, 2006.
Accepted September 18, 2006.
|
References |
|---|
|
TopAbstractIntroductionCase ReportsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. A. Placzkowski, A. Vella, G. B. Thompson, C. S. Grant, C. C. Reading, J. W. Charboneau, J. C. Andrews, R. V. Lloyd, and F. J. Service Secular Trends in the Presentation and Management of Functioning Insulinoma at the Mayo Clinic, 1987-2007 J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1069 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. E. Cryer, L. Axelrod, A. B. Grossman, S. R. Heller, V. M. Montori, E. R. Seaquist, and F. J. Service Evaluation and Management of Adult Hypoglycemic Disorders: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., March 1, 2009; 94(3): 709 - 728. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
