- Email: [email protected]
Asymptomatic insulinoma: A case report and autopsy series
diabetes research and clinical practice 98 (2012) 445–451
Contents available at Sciverse ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
Asymptomatic insulinoma: A case report and autopsy series Satoru Kishi a, Kentaro Sakamoto b, Masaya Mori c, Akihiro Isogawa a, Teruo Shiba a,b,* a
Division of Diabetic Metabolic Internal Medicine, Mitsui Memorial Hospital, Tokyo, Japan Division of Diabetes and Metabolism, Toho University Ohashi Medical Center, Tokyo, Japan c Department of Pathology, Mitsui Memorial Hospital, Tokyo, Japan b
article info
abstract
Article history:
Aims: We investigated the prevalence and characterization of asymptomatic pancreatic
Received 15 June 2012
tumors in response to our experience of asymptomatic insulinoma.
Received in revised form
Methods: A patient with a moderately low glucose level and pancreatic incidentaloma
15 August 2012
detected by CT was examined. Pancreas specimens from 423 autopsy cases were also
Accepted 28 August 2012
pathologically examined systematically by hematoxylin–eosin staining.
Published on line 17 October 2012
Results: The examined patient showed no profile characteristic of insulinoma by fasting or loading tests, however, ASVS led to diagnosis of insulin-producing tumor. The tumor was
Keywords:
resected with the pancreatic body and tail and revealed to be 10 mm in diameter, with 98.5%
Insulinoma
of the cells positive for insulin. Pathological evaluation confirmed a well-differentiated
Endocrine tumor
endocrine pancreatic tumor, which was suggestive of an incidentally detected asymptomatic insulinoma. Microscopic evaluations of pancreatic specimens from 423 autopsy cases
ASVS
revealed pancreatic monotonous lesions in 6 cases (1.42%). In 4 autopsy specimens large enough for immuno-histochemical evaluation, the lesions were positive for glucagon but negative for insulin. Conclusions: As concerns the present study, retrospective immunohistochemical investigation in autopsy cases revealed the presence of asymptomatic glucagonoma but no asymptomatic insulinoma. Advances in diagnostic imaging, however, might raise the probability of detecting early asymptomatic stages of insulinoma incidentally. ASVS appears to be sensitive even for asymptomatic incidental insulinomas. # 2012 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
Pancreatic endocrine tumors are a class of gastroenteropancreatic neuroendocrine tumors clinically classified as functioning or nonfunctioning according to the presence of symptoms caused by the inappropriate or excessive secretion of hormones. Insulinomas are rare and small neuroendocrine tumors with an incidence of approximately 1–4 cases per million per year [1–3]. Although rare, insulinomas are the most common functional pancreatic endocrine tumor,
accounting for 60–70% of the total [2]. Most insulinomas are smaller than 5 mm, but even at that size they develop specific symptoms of hypoglycemia due to excessive endogenous insulin secretion. The most recommended method for the evaluation of endogenous insulin hypersecretion is a supervised 72-h fasting test, a study with excellent sensitivity and specificity [4]. The patient we describe in this report was referred to our hospital because of a reduced blood glucose value detected in a general health check-up. None of various tests revealed hypoglycemic symptoms or evidence of continuous insulin
* Corresponding author at: Division of Diabetes and Metabolism, Toho University Ohashi Medical Center, 2-17-6, Ohashi, Meguro-ku, Tokyo 153-8515, Japan. Tel.: +81 3 3468 1251; fax: +81 3 3481 7305. E-mail address: [email protected] (T. Shiba). 0168-8227/$ – see front matter # 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.diabres.2012.08.007
446
diabetes research and clinical practice 98 (2012) 445–451
secretion. Computed tomography (CT), however, depicted what turned out to be a functioning insulinoma.
2.
Case report
A 61-year-old Japanese woman with an unremarkable medical history was referred to our outpatient clinic because of a low serum venous glucose value of 48 mg/dl in a fasting condition, detected by her physician in a general health check-up. She had no clinical symptoms of hypoglycemia and used no medications known to induce hypoglycemia, such as beta-blockers or anti-arrhythmic agents. She ate regular meals and drank a glass of whiskey with water a day. Her family history was also unremarkable. Her BMI was 22.3, and she reported no recent changes of body weight. The physical examination was unremarkable, revealing no goiter of the neck or palpable masses or organomegaly in the abdomen. All of the results of her general laboratory evaluation performed for screening were within the normal ranges and were inconsistent with hypoglycemic diseases such as hypothyroidism, hypopituitarism, catecholamine deficiency, insulin autoantibody syndrome, insulinoma, gastrinoma, glucagonoma, somatostatinoma, VIPoma, or IGF-I/II-producing tumor. However, enhanced CT of the abdomen performed before admission showed a coin lesion with high-density enhancement at the distal tail of the pancreas (Fig. 1). Magnetic resonance imaging (MRI) also showed high-intensity enhancing mass at the same location (data not shown). We started a 72-h fasting test after admission, but determined to terminate the test at 30 h because of continuous low level of venous blood glucose level less than 45 mg/dl though the patient did not complain any hypoglycemic symptoms. The result of fasting test did not fit any criteria for insulinoma such as Fajans Index, Turner Index, Grunt Index, or
Service’s criteria (Fig. 2A and B). More than 70% of patients with insulinoma are reported to manifest an exaggerated rise of plasma insulin under the glucagon load test [10]. When we injected 1 mg of glucagon intravenously at the end of the fasting test, the plasma glucose level remained above 42 mg/dl (Fig. 2A and B) [5–9]. As such, these results did not suggest continuous inappropriate insulin secretion from insulinoma. According to another report, the 75 g oral glucose tolerance test (75 g OGTT) leads to inappropriate insulin secretion and exaggerated hyperinsulinemic response in subjects with insulinoma [10]. Not having observed any signs of insulin hyper-secretion characteristic of insulinoma in the OGTT (data not shown), we decided to assess insulin secretion by performing arterial calcium stimulation with hepatic venous sampling (ASVS) [11]. First, we performed angiography, and then selectively injected a 5 ml bolus of calcium gluconate (0.025 mequiv. Ca2+/kg). We drew blood samples just before calcium injection, and at 30, 60, 120, and 180 s after calcium injection. Significant insulin gradients, calculated as the ratios to the hepatic venous insulin concentrations, were observed after calcium injection into the splenic artery (Fig. 2C). Though the angiographic examination showed no sign of a pancreatic lesion, these results, in combination, strongly suggested the presence of insulinoma at the pancreatic tail, possibly in association with the pancreatic tumor observed in the CT image [12]. As none of the examinations ruled out the possibility of a malignant insulinoma, the patient underwent a resection of the pancreatic body and tail with splenectomy. A 10 mm 10 mm 8 mm tumor was identified at the tail of the pancreas (Fig. 3D), and there was no evidence of invasion or metastases. Pathological evaluation revealed a well-differentiated benignly behaving endocrine tumor of the pancreas. Chromogranin-A, synaptophysin, and NSE were strongly positive for tumor cells in immunohistochemical studies by an immunoperoxidase method. In addition, 98.5% of the tumor cells were positive for insulin, 0% were positive for glucagon, and 1.5% were positive for somatostatin (Fig. 3A–C). Testing for NCAM was slightly positive, that for CK-19 was negative, and that for Ki-67 (MIB1 labeling index) was less than 2%, while the number of mitoses was less than 2 per 10 high-powered fields [13,14]. The endocrine tumor was classified as a benignly behaving insulinoma [15]. Her postoperative course was uneventful, and the fasting plasma glucose values were 108 and 93 mg/dl at a week and a month after the resection, respectively. No evidence of recurrence appeared on follow-up enhanced CT of the abdomen at 1 year after the operation.
3.
Fig. 1 – Computed tomography image of abdomen. The computed tomography image of the abdomen shows a 1.1 cm high-density enhancing mass in the tail of the pancreas.
Results and discussion
Insulinomas are a rare and small neuroendocrine tumor of a type usually found in patients between 30 and 60 years of age. In most cases, the tumors are discovered through the presentation of characteristic symptoms. Our case was a few years older than the patients with insulinoma typically reported, and presented without hypoglycemic symptoms. A common clinical feature of insulinoma is fasting hypoglycemia, but overall the clinical signs and symptoms of hypoglycemia are various and nonspecific. Hypoglycemic symptoms due to insulinoma can be characterized into two categories, namely, those associated with excess catecholamines, such as tremor, sweating,
diabetes research and clinical practice 98 (2012) 445–451
Fig. 2 – (A and B) 72 h fasting test. The result of the fasting test does not fit the criteria for Fajans Index (insulin/glucose > 0.3), Turner Index (glucose/insulin < 2.5), grunt index (Insulin T 100/glucose S 30 > 200), or Service’s criteria (insulin > 6 mU/ml, C Peptide > 0.6 ng/ml). There was no continuous secretion of insulin from the insulinoma. Insulin assay was performed by ARCHITECT (Abbott Laboratories, IL, USA). (C) Arterial calcium stimulation with hepatic venous sampling. The graphs show insulin levels from our patient with an insulinoma in the tail of the pancreas. After infusion of calcium gluconate into selected arteries (splenic artery, gastroduodenal artery, superior mesenteric artery, and hepatic artery), insulin levels were assayed from samples taken at various time intervals from the right hepatic vein. In the splenic artery, the insulin gradient peaked at 30 s.
447
448
diabetes research and clinical practice 98 (2012) 445–451
Fig. 3 – Surgical specimen and histological sections of the insulinoma with immunohistochemical demonstration. (A–D) Immunohistochemistry stains of the neuroendocrine tumor are positive for insulin (A), negative for glucagon (B), and negative for somatostatin (C). Insulinoma was excised from the tail of the pancreas (1.0 cm T 1.0 cm T 0.8 cm). The insulinoma was a well-circumscribed, white, solid tumor with a uniform internal consistency (D).
palpitations, hunger, anxiety, and nausea, and those associated with neuroglycopenia, such as headache, epilepsy, amnesia, disorientation, confusion, dizziness, seizure, diplopia, and hemiparesis. Though most insulinomas are smaller than 2 cm at the time of diagnosis, the diagnosis is usually based on symptoms of hypoglycemia secondary to excessive insulin secretion. The clinical history of the patient was unprecedented in two respects: she had been referred to our hospital because of a reduced blood glucose value detected in a general health check-up, and she had never recognized any of the hypoglycemic symptoms described above. Another characteristic feature of insulinoma is symptomatic hypoglycemia, a condition relieved by food intake and resulting weight gain. Even if our patient had failed to notice hypoglycemic symptoms, one would have expected her condition to induce snacking and weight gain. In fact, she had had no such complaints: normal weight had been maintained, she had no desire to eat snacks between meals or at night, and none of her family members noted any neuroglycopenic symptoms. It has been difficult to set a plasma glucose cut-off point for defining hypoglycemia, given that the physiological response to low blood glucose is hierarchical, as documented with the stepped hypoglycemic clamp technique in normal glucose subjects [16–18]. The early response elicited by a progressive decrease in plasma glucose is an inhibition of insulin secretion. Our patient, however, manifested neither an inappropriate insulin secretion nor an
exaggerated hyper-insulinemic response when her blood glucose was low. In a recent retrospective study of 237 patients with insulinoma, all 214 patients who provided reliable histories were symptomatic [19]. Analysis of these patients classified into three categories, namely, fasting, postprandial, or both, demonstrated that hypoglycemic symptoms in the postprandial state could not rule out insulinoma [19]. Our patient was unique for not manifesting symptoms in either the fasting or postprandial state. It may be that patients with insulinoma have reduced awareness of hypoglycemia and reduced counter-regulatory hormonal responses, just as diabetic patients may sometimes fail to notice hypoglycemia because of some hypoglycemia-associated autonomic failure [20]. The lack of autonomic symptoms was possibly caused by an adrenergic insensitivity induced by repeated hypoglycemia [21]. Those patients were often asymptomatic when their plasma glucose concentrations were below 35 mg/dl, but all of those with insulinoma were revealed to have symptoms at least once before undergoing their operations. Hypoglycemia itself can induce unawareness of the autonomic response and decrease the counterregulatory hormonal secretion to hypoglycemia, yet this unawareness of hypoglycemia has been reported to be reversible in patients with insulinomas, especially after surgery [21]. It thus may be possible to remain asymptomatic before surgery, just as we witnessed in our case with insulinoma.
449
diabetes research and clinical practice 98 (2012) 445–451
The enhanced CT performed before the fasting test provided another unique finding on our case, as it suggested, in the absence of symptoms or complaints caused by neoplasms, that the pancreatic tumor was classifiable as an incidentaloma. The lower blood glucose detected at the general health check-up would be expected of any of several tumor types, such as an insulinoma, gastrinoma, glucagonoma, somatostatinoma, VIPoma, or IGF-I/II producing tumor. Unable to rule out insulinoma in our case, we decided to perform the 72-h fasting test even in the absence of symptoms. No evidence of hypoglycemia with excessive insulin secretion appeared, however, during the fast. After we terminated the fast because of her persistently lower venous serum glucose, we performed a glucagon load test, having learned from the literature that glucagon administration following the fasting test leads to elevations in insulin and C-peptide in spite of further declines of glucose concentrations [9,10,22]. Our patient, however, did not present the paradoxical reaction reported, namely, the persistently inappropriate insulin secretion characteristic to insulinoma. Our case also showed no sign of inappropriate insulin secretion or exaggerated hyper-insulinemic response during the 75 g OGTT (data not shown), even though this test has been reported to induce hypoglycemia in some patients [10,23]. Arterial calcium stimulation with hepatic venous sampling (ASVS) has often been performed to identify and localize occult insulinoma, a tumor typically difficult to localize based solely on overt hypoglycemic symptoms. It was not the case in our incidental pancreatic tumor, but we still thought it prudent to perform ASVS in light of the lower venous blood glucose noted in the general health check-up. The ASVS results provided strengthened evidence of a tumor with insulin-secreting function at the distal tail of the pancreas, where high-density enhancement had been identified by the CT imaging performed before our patient’s admission. We therefore proceeded with resection, the recommendable course of action under the circumstances [11]. Immunological staining revealed insulin secretion by more than 98% of the cells composing the resected pancreatic tumor, and the histological characteristics were consistent with insulinoma with overt hypoglycemic symptoms. The resected insulinoma was large enough to elicit insulin-causing hypoglycemic symptoms. Also noteworthy was the relatively advanced age of the patient for the diagnosis of an insulinoma, 61 years old. The patient may have spent all of those years without any symptoms of hypoglycemia. The findings described above raise the possibility that an asymptomatic insulinoma had been present for much of that time. To investigate this issue,
we consecutively evaluated 423 autopsy cases with occult gastroentero-pancreatic neuroendocrine tumors available for systematic examination, from autopsies performed at Mitsui Memorial Hospital over the almost 8-year period between April 2002 and January 2010 (Table 1). The autopsy rate in our hospital was about 16.7% on average during this period, which was somewhat higher than the rate reported by hospitals registered as members of the Japanese Society of Pathology, namely, 6.7% (in 2008). The pancreas acquired by our autopsies had been sectioned from head to tail at 1–3 mm intervals. All specimens were by hematoxylin–eosin staining. Islets of Langerhans consisting of a single cell type were detected microscopically in 6 cases (1.42%). The lesions of the islets of Langerhans of the pancreas were composed of monotonous cells in all of these cases. Two of the lesions detected were too small for cell type evaluation by the following immunohistochemical studies. The lesions of the other four cases were positive for glucagon but negative for insulin. We could not find insulinomas in any of the 423 autopsied cases. The incidence of asymptomatic pancreatic endocrine tumors in our hospital (1.42%) was similar to that (1.6%) observed by Kimura et al., who reported that 70% of asymptomatic endocrine tumors produced multiple hormones and the other 30% produced single hormone [24]. None of the asymptomatic pancreatic tumor observed by Kimura’s group consisted of insulin-producing cells [24]. In the study by Polak et al., the most commonly found cell type in asymptomatic endocrine tumors of the pancreas cells produced multiple hormones, insulin, and pancreatic polypeptide [25,26]. In our study, glucagon producing cells were 100% in the detected occult gastroentero-pancreatic neuroendocrine tumors, and all of cases were asymptomatic. No cases of multiple endocrine neoplasia (MEN) type 1 were observed. Computed tomography (CT) scans are growing more prevalent, spurred by recent advances in CT technology. Though previously found only in autopsy cases, incidentalomas and asymptomatic pancreas tumors are now being detected by chance during radiological or magnetic resonance methods, mostly by CT. Yet incidentalomas of the pancreas are distinct and less frequently detected by CT. The likelihood of detection of a pancreatic endocrine tumor by CT depends upon the size and location of the tumor. In our case, enhanced CT detected a pancreatic tumor of 10 mm in diameter. In the process of differential diagnosis we treated the tumor as an incidentaloma before the tumor was characterized as a functional endocrine tumor by ASVS. Though tumors as small as 10 mm in diameter have been located, those less than 20 mm cannot be found consistently or reproducibly. The
Table 1 – Immuno-histochemical findings of endocrine tumor. Case 1 2 3 4 5 6 a
Chief pathologic cause of death Renal cancer Myeloma Lymphoma Lung cancer Lung cancer Stomach cancer
Size of lesion
Insulin
Glucagon
Somatostatin
0.7 mm 0.6 mm, 0.6 mm 0.6 mm 3.6 mm 3.4 mm 1.7 mm 1.3 mm 0.4 mm 0.4 mm 2.3 mm 1.5 mm 0.1 mm 0.1 mm, 0.1 mm 0.1 mm
<1% – <1%
40% 40% 100%
<1% – –
a
a
a
1%
90%
<1%
a
a
a
There is not enough sample volume to evaluate by immunohistochemical studies, because the tumor is very small.
450
diabetes research and clinical practice 98 (2012) 445–451
sensitivity of the dynamic CT scan in the detection of insulinomas or gastrinomas has been reported to range from 30% to 66% [11,27–30]. In Japan, the advancing effectiveness of imaging modalities (CT, MRI, etc.) is evidenced by the high rate (24%) of pancreatic endocrine tumors that have been found incidentally during routine health examinations of asymptomatic patients [3]. Endocrine societies revised the diagnostic consensus guidelines in 2009 [31], after the admission examination of our case. This posed a limitation on our clinical case study, as the revised threshold for serum insulin at the fasting test was lowered to below the insulin value our case presented. Our reporting here on this clinical case is based on the diagnostic criteria widely used at that time.
4.
Conclusions
We experienced a case who presented with a reduced blood glucose level in a general health check-up and was subsequently found to have an incidentaloma of the pancreas by a CT examination. This case appeared to be rare: in spite of its size (10 mm), the insulinoma caused no hypoglycemic symptoms and showed no typical reaction to the biochemical screening test for insulinoma. There have been no reports of a decryption of asymptomatic insulinoma by pathological study. As imaging modalities such as CT, MRI and ultrasound advance, they may have the potential to detect early-stage pancreatic neoplasms, especially insulinomas, before the tumors grow large enough to cause symptoms. ASVS has been suggested to be helpful not only for occult insulinoma with overt hypoglycemic symptoms, but also in making differential diagnosis and assessing the locations of asymptomatic insulinomas.
Conflict of interest There are no conflicts of interest.
references
[1] Service FJ, McMahon MM, O’Brien PC, Ballard DJ. Functioning insulinoma – incidence, recurrence, and long-term survival of patients: a 60-year study. Mayo Clin Proc 1991;66:711–9. [2] Halfdanarson TR, Rubin J, Farnell MB, Grant CS, Petersen GM. Pancreatic endocrine neoplasms: epidemiology and prognosis of pancreatic endocrine tumors. Endocr Relat Cancer 2008;15:409–27. [3] Ito T, Sasano H, Tanaka M, Osamura RY, Sasaki I, Kimura W, et al. Epidemiological study of gastroenteropancreatic neuroendocrine tumors in Japan. J Gastroenterol 2010;45:234–43. [4] Vezzosi D, Bennet A, Fauvel J, Caron P. Insulin C-peptide and proinsulin for the biochemical diagnosis of hypoglycaemia related to endogenous hyperinsulinism. Eur J Endocrinol 2007;157:75–83. [5] Service FJ. Hypoglycemic disorders. N Engl J Med 1995;332:1144–52. [6] Service FJ, O’Brien PC, McMahon MM, Kao PC. C-peptide during the prolonged fast in insulinoma. J Clin Endocrinol Metab 1993;76:655–9.
[7] O’Brien T, O’Brien PC, Service FJ. Insulin surrogates in insulinoma. J Clin Endocrinol Metab 1993;77:448–51. [8] Kao PC, Taylor RL, Service FJ. Proinsulin by immunochemiluminometric assay for the diagnosis of insulinoma. J Clin Endocrinol Metab 1994;78:1048–51. [9] Soh AW, Kek PC. Insulinoma in a patient with negative prolonged fast and glucagon-induced hypoglycemia. Endocr Pract 2010;16:838–41. [10] Marks V. Progress report. Diagnosis of insulinoma. Gut 1971;12:835–43. [11] Doppman JL, Miller DL, Chang R, Shawker TH, Gorden P, Norton JA. Insulinomas: localization with selective intraarterial injection of calcium. Radiology 1991;178:237–41. [12] Van Hoe L, Gryspeerdt S, Marchal G, Baert AL, Mertens L. Helical CT for the preoperative localization of islet cell tumors of the pancreas: value of arterial and parenchymal phase images. AJR – Am J Roentgenol 1995;165:1437–9. [13] Schmitt AM, Anlauf M, Rousson V, Schmid S, Kofler A, Riniker F, et al. WHO 2004 criteria and CK19 are reliable prognostic markers in pancreatic endocrine tumors. Am J Surg Pathol 2007;31:1677–82. [14] Bettini R, Boninsegna L, Mantovani W, Capelli P, Bassi C, Pederzoli P, et al. Prognostic factors at diagnosis and value of WHO classification in a mono-institutional series of 180 non-functioning pancreatic endocrine tumours. Ann Oncol 2008;19:903–8. [15] Kloppel G, Perren A, Heitz PU. The gastroenteropancreatic neuroendocrine cell system and its tumors: the WHO classification. Ann N Y Acad Sci 2004;1014:13–27. [16] Schwartz NS, Clutter WE, Shah SD, Cryer PE. Glycemic thresholds for activation of glucose counterregulatory systems are higher than the threshold for symptoms. J Clin Invest 1987;79:777–81. [17] Mitrakou A, Fanelli C, Veneman T, Perriello G, Calderone S, Platanisiotis D, et al. Reversibility of unawareness of hypoglycemia in patients with insulinomas. N Engl J Med 1993;329:834–9. [18] Fanelli C, Pampanelli S, Epifano L, Rambotti AM, Ciofetta M, Modarelli F, et al. Relative roles of insulin and hypoglycaemia on induction of neuroendocrine responses to, symptoms of, and deterioration of cognitive function in hypoglycaemia in male and female humans. Diabetologia 1994;37:797–807. [19] Placzkowski KA, Vella A, Thompson GB, Grant CS, Reading CC, Charboneau JW, et al. Secular trends in the presentation and management of functioning insulinoma at the Mayo Clinic, 1987–2007. J Clin Endocrinol Metab 2009;94:1069–73. [20] Fanelli CG, Epifano L, Rambotti AM, Pampanelli S, Di Vincenzo A, Modarelli F, et al. Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes 1993;42:1683–9. [21] Vea H, Jorde R, Sager G, Vaaler S, Sundsfjord J, Revhaug A. Pre- and postoperative glucose levels for eliciting hypoglycaemic responses in a patient with insulinoma. Diabet Med 1992;9:950–3. [22] Wiesli P, Schmid C, Perren A, Pfammatter T, Spinas GA, Keller U. Hypoglycemia in response to glucose and glucagon in insulinoma patients with a negative prolonged fast: functional and morphological properties. J Endocrinol Invest 2004;27:832–8. [23] Kar P, Price P, Sawers S, Bhattacharya S, Reznek RH, Grossman AB. Insulinomas may present with normoglycemia after prolonged fasting but glucosestimulated hypoglycemia. J Clin Endocrinol Metab 2006;91:4733–6.
diabetes research and clinical practice 98 (2012) 445–451
[24] Kimura W, Kuroda A, Morioka Y. Clinical pathology of endocrine tumors of the pancreas. Analysis of autopsy cases. Dig Dis Sci 1991;36:933–42. [25] Polak JM, Bloom SR, Adrian TE, Heitz P, Bryant MG, Pearse AG. Pancreatic polypeptide in insulinomas, gastrinomas, vipomas, and glucagonomas. Lancet 1976;1:328–30. [26] Heitz PU, Kasper M, Polak JM, Kloppel G. Pancreatic endocrine tumors. Hum Pathol 1982;13:263–71. [27] Kurosaki Y, Kuramoto K, Itai Y. Hyperattenuating insulinoma at unenhanced CT. Abdom Imaging 1996;21:334–6. [28] Norton JA, Shawker TH, Doppman JL, Miller DL, Fraker DL, Cromack DT, et al. Localization and surgical treatment of occult insulinomas. Ann Surg 1990;212:615–20.
451
[29] Galiber AK, Reading CC, Charboneau JW, Sheedy 2nd PF, James EM, Gorman B, et al. Localization of pancreatic insulinoma: comparison of pre- and intraoperative US with CT and angiography. Radiology 1988;166:405–8. [30] Wank SA, Doppman JL, Miller DL, Collen MJ, Maton PN, Vinayek R, et al. Prospective study of the ability of computed axial tomography to localize gastrinomas in patients with Zollinger–Ellison syndrome. Gastroenterology 1987;92:905–12. [31] Cryer PE, Axelrod L, Grossman AB, Heller SR, Montori VM, Seaquist ER, et al. Evaluation and management of adult hypoglycemic disorders: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2009;94:709–28.
Contents available at Sciverse ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
Asymptomatic insulinoma: A case report and autopsy series Satoru Kishi a, Kentaro Sakamoto b, Masaya Mori c, Akihiro Isogawa a, Teruo Shiba a,b,* a
Division of Diabetic Metabolic Internal Medicine, Mitsui Memorial Hospital, Tokyo, Japan Division of Diabetes and Metabolism, Toho University Ohashi Medical Center, Tokyo, Japan c Department of Pathology, Mitsui Memorial Hospital, Tokyo, Japan b
article info
abstract
Article history:
Aims: We investigated the prevalence and characterization of asymptomatic pancreatic
Received 15 June 2012
tumors in response to our experience of asymptomatic insulinoma.
Received in revised form
Methods: A patient with a moderately low glucose level and pancreatic incidentaloma
15 August 2012
detected by CT was examined. Pancreas specimens from 423 autopsy cases were also
Accepted 28 August 2012
pathologically examined systematically by hematoxylin–eosin staining.
Published on line 17 October 2012
Results: The examined patient showed no profile characteristic of insulinoma by fasting or loading tests, however, ASVS led to diagnosis of insulin-producing tumor. The tumor was
Keywords:
resected with the pancreatic body and tail and revealed to be 10 mm in diameter, with 98.5%
Insulinoma
of the cells positive for insulin. Pathological evaluation confirmed a well-differentiated
Endocrine tumor
endocrine pancreatic tumor, which was suggestive of an incidentally detected asymptomatic insulinoma. Microscopic evaluations of pancreatic specimens from 423 autopsy cases
ASVS
revealed pancreatic monotonous lesions in 6 cases (1.42%). In 4 autopsy specimens large enough for immuno-histochemical evaluation, the lesions were positive for glucagon but negative for insulin. Conclusions: As concerns the present study, retrospective immunohistochemical investigation in autopsy cases revealed the presence of asymptomatic glucagonoma but no asymptomatic insulinoma. Advances in diagnostic imaging, however, might raise the probability of detecting early asymptomatic stages of insulinoma incidentally. ASVS appears to be sensitive even for asymptomatic incidental insulinomas. # 2012 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
Pancreatic endocrine tumors are a class of gastroenteropancreatic neuroendocrine tumors clinically classified as functioning or nonfunctioning according to the presence of symptoms caused by the inappropriate or excessive secretion of hormones. Insulinomas are rare and small neuroendocrine tumors with an incidence of approximately 1–4 cases per million per year [1–3]. Although rare, insulinomas are the most common functional pancreatic endocrine tumor,
accounting for 60–70% of the total [2]. Most insulinomas are smaller than 5 mm, but even at that size they develop specific symptoms of hypoglycemia due to excessive endogenous insulin secretion. The most recommended method for the evaluation of endogenous insulin hypersecretion is a supervised 72-h fasting test, a study with excellent sensitivity and specificity [4]. The patient we describe in this report was referred to our hospital because of a reduced blood glucose value detected in a general health check-up. None of various tests revealed hypoglycemic symptoms or evidence of continuous insulin
* Corresponding author at: Division of Diabetes and Metabolism, Toho University Ohashi Medical Center, 2-17-6, Ohashi, Meguro-ku, Tokyo 153-8515, Japan. Tel.: +81 3 3468 1251; fax: +81 3 3481 7305. E-mail address: [email protected] (T. Shiba). 0168-8227/$ – see front matter # 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.diabres.2012.08.007
446
diabetes research and clinical practice 98 (2012) 445–451
secretion. Computed tomography (CT), however, depicted what turned out to be a functioning insulinoma.
2.
Case report
A 61-year-old Japanese woman with an unremarkable medical history was referred to our outpatient clinic because of a low serum venous glucose value of 48 mg/dl in a fasting condition, detected by her physician in a general health check-up. She had no clinical symptoms of hypoglycemia and used no medications known to induce hypoglycemia, such as beta-blockers or anti-arrhythmic agents. She ate regular meals and drank a glass of whiskey with water a day. Her family history was also unremarkable. Her BMI was 22.3, and she reported no recent changes of body weight. The physical examination was unremarkable, revealing no goiter of the neck or palpable masses or organomegaly in the abdomen. All of the results of her general laboratory evaluation performed for screening were within the normal ranges and were inconsistent with hypoglycemic diseases such as hypothyroidism, hypopituitarism, catecholamine deficiency, insulin autoantibody syndrome, insulinoma, gastrinoma, glucagonoma, somatostatinoma, VIPoma, or IGF-I/II-producing tumor. However, enhanced CT of the abdomen performed before admission showed a coin lesion with high-density enhancement at the distal tail of the pancreas (Fig. 1). Magnetic resonance imaging (MRI) also showed high-intensity enhancing mass at the same location (data not shown). We started a 72-h fasting test after admission, but determined to terminate the test at 30 h because of continuous low level of venous blood glucose level less than 45 mg/dl though the patient did not complain any hypoglycemic symptoms. The result of fasting test did not fit any criteria for insulinoma such as Fajans Index, Turner Index, Grunt Index, or
Service’s criteria (Fig. 2A and B). More than 70% of patients with insulinoma are reported to manifest an exaggerated rise of plasma insulin under the glucagon load test [10]. When we injected 1 mg of glucagon intravenously at the end of the fasting test, the plasma glucose level remained above 42 mg/dl (Fig. 2A and B) [5–9]. As such, these results did not suggest continuous inappropriate insulin secretion from insulinoma. According to another report, the 75 g oral glucose tolerance test (75 g OGTT) leads to inappropriate insulin secretion and exaggerated hyperinsulinemic response in subjects with insulinoma [10]. Not having observed any signs of insulin hyper-secretion characteristic of insulinoma in the OGTT (data not shown), we decided to assess insulin secretion by performing arterial calcium stimulation with hepatic venous sampling (ASVS) [11]. First, we performed angiography, and then selectively injected a 5 ml bolus of calcium gluconate (0.025 mequiv. Ca2+/kg). We drew blood samples just before calcium injection, and at 30, 60, 120, and 180 s after calcium injection. Significant insulin gradients, calculated as the ratios to the hepatic venous insulin concentrations, were observed after calcium injection into the splenic artery (Fig. 2C). Though the angiographic examination showed no sign of a pancreatic lesion, these results, in combination, strongly suggested the presence of insulinoma at the pancreatic tail, possibly in association with the pancreatic tumor observed in the CT image [12]. As none of the examinations ruled out the possibility of a malignant insulinoma, the patient underwent a resection of the pancreatic body and tail with splenectomy. A 10 mm 10 mm 8 mm tumor was identified at the tail of the pancreas (Fig. 3D), and there was no evidence of invasion or metastases. Pathological evaluation revealed a well-differentiated benignly behaving endocrine tumor of the pancreas. Chromogranin-A, synaptophysin, and NSE were strongly positive for tumor cells in immunohistochemical studies by an immunoperoxidase method. In addition, 98.5% of the tumor cells were positive for insulin, 0% were positive for glucagon, and 1.5% were positive for somatostatin (Fig. 3A–C). Testing for NCAM was slightly positive, that for CK-19 was negative, and that for Ki-67 (MIB1 labeling index) was less than 2%, while the number of mitoses was less than 2 per 10 high-powered fields [13,14]. The endocrine tumor was classified as a benignly behaving insulinoma [15]. Her postoperative course was uneventful, and the fasting plasma glucose values were 108 and 93 mg/dl at a week and a month after the resection, respectively. No evidence of recurrence appeared on follow-up enhanced CT of the abdomen at 1 year after the operation.
3.
Fig. 1 – Computed tomography image of abdomen. The computed tomography image of the abdomen shows a 1.1 cm high-density enhancing mass in the tail of the pancreas.
Results and discussion
Insulinomas are a rare and small neuroendocrine tumor of a type usually found in patients between 30 and 60 years of age. In most cases, the tumors are discovered through the presentation of characteristic symptoms. Our case was a few years older than the patients with insulinoma typically reported, and presented without hypoglycemic symptoms. A common clinical feature of insulinoma is fasting hypoglycemia, but overall the clinical signs and symptoms of hypoglycemia are various and nonspecific. Hypoglycemic symptoms due to insulinoma can be characterized into two categories, namely, those associated with excess catecholamines, such as tremor, sweating,
diabetes research and clinical practice 98 (2012) 445–451
Fig. 2 – (A and B) 72 h fasting test. The result of the fasting test does not fit the criteria for Fajans Index (insulin/glucose > 0.3), Turner Index (glucose/insulin < 2.5), grunt index (Insulin T 100/glucose S 30 > 200), or Service’s criteria (insulin > 6 mU/ml, C Peptide > 0.6 ng/ml). There was no continuous secretion of insulin from the insulinoma. Insulin assay was performed by ARCHITECT (Abbott Laboratories, IL, USA). (C) Arterial calcium stimulation with hepatic venous sampling. The graphs show insulin levels from our patient with an insulinoma in the tail of the pancreas. After infusion of calcium gluconate into selected arteries (splenic artery, gastroduodenal artery, superior mesenteric artery, and hepatic artery), insulin levels were assayed from samples taken at various time intervals from the right hepatic vein. In the splenic artery, the insulin gradient peaked at 30 s.
447
448
diabetes research and clinical practice 98 (2012) 445–451
Fig. 3 – Surgical specimen and histological sections of the insulinoma with immunohistochemical demonstration. (A–D) Immunohistochemistry stains of the neuroendocrine tumor are positive for insulin (A), negative for glucagon (B), and negative for somatostatin (C). Insulinoma was excised from the tail of the pancreas (1.0 cm T 1.0 cm T 0.8 cm). The insulinoma was a well-circumscribed, white, solid tumor with a uniform internal consistency (D).
palpitations, hunger, anxiety, and nausea, and those associated with neuroglycopenia, such as headache, epilepsy, amnesia, disorientation, confusion, dizziness, seizure, diplopia, and hemiparesis. Though most insulinomas are smaller than 2 cm at the time of diagnosis, the diagnosis is usually based on symptoms of hypoglycemia secondary to excessive insulin secretion. The clinical history of the patient was unprecedented in two respects: she had been referred to our hospital because of a reduced blood glucose value detected in a general health check-up, and she had never recognized any of the hypoglycemic symptoms described above. Another characteristic feature of insulinoma is symptomatic hypoglycemia, a condition relieved by food intake and resulting weight gain. Even if our patient had failed to notice hypoglycemic symptoms, one would have expected her condition to induce snacking and weight gain. In fact, she had had no such complaints: normal weight had been maintained, she had no desire to eat snacks between meals or at night, and none of her family members noted any neuroglycopenic symptoms. It has been difficult to set a plasma glucose cut-off point for defining hypoglycemia, given that the physiological response to low blood glucose is hierarchical, as documented with the stepped hypoglycemic clamp technique in normal glucose subjects [16–18]. The early response elicited by a progressive decrease in plasma glucose is an inhibition of insulin secretion. Our patient, however, manifested neither an inappropriate insulin secretion nor an
exaggerated hyper-insulinemic response when her blood glucose was low. In a recent retrospective study of 237 patients with insulinoma, all 214 patients who provided reliable histories were symptomatic [19]. Analysis of these patients classified into three categories, namely, fasting, postprandial, or both, demonstrated that hypoglycemic symptoms in the postprandial state could not rule out insulinoma [19]. Our patient was unique for not manifesting symptoms in either the fasting or postprandial state. It may be that patients with insulinoma have reduced awareness of hypoglycemia and reduced counter-regulatory hormonal responses, just as diabetic patients may sometimes fail to notice hypoglycemia because of some hypoglycemia-associated autonomic failure [20]. The lack of autonomic symptoms was possibly caused by an adrenergic insensitivity induced by repeated hypoglycemia [21]. Those patients were often asymptomatic when their plasma glucose concentrations were below 35 mg/dl, but all of those with insulinoma were revealed to have symptoms at least once before undergoing their operations. Hypoglycemia itself can induce unawareness of the autonomic response and decrease the counterregulatory hormonal secretion to hypoglycemia, yet this unawareness of hypoglycemia has been reported to be reversible in patients with insulinomas, especially after surgery [21]. It thus may be possible to remain asymptomatic before surgery, just as we witnessed in our case with insulinoma.
449
diabetes research and clinical practice 98 (2012) 445–451
The enhanced CT performed before the fasting test provided another unique finding on our case, as it suggested, in the absence of symptoms or complaints caused by neoplasms, that the pancreatic tumor was classifiable as an incidentaloma. The lower blood glucose detected at the general health check-up would be expected of any of several tumor types, such as an insulinoma, gastrinoma, glucagonoma, somatostatinoma, VIPoma, or IGF-I/II producing tumor. Unable to rule out insulinoma in our case, we decided to perform the 72-h fasting test even in the absence of symptoms. No evidence of hypoglycemia with excessive insulin secretion appeared, however, during the fast. After we terminated the fast because of her persistently lower venous serum glucose, we performed a glucagon load test, having learned from the literature that glucagon administration following the fasting test leads to elevations in insulin and C-peptide in spite of further declines of glucose concentrations [9,10,22]. Our patient, however, did not present the paradoxical reaction reported, namely, the persistently inappropriate insulin secretion characteristic to insulinoma. Our case also showed no sign of inappropriate insulin secretion or exaggerated hyper-insulinemic response during the 75 g OGTT (data not shown), even though this test has been reported to induce hypoglycemia in some patients [10,23]. Arterial calcium stimulation with hepatic venous sampling (ASVS) has often been performed to identify and localize occult insulinoma, a tumor typically difficult to localize based solely on overt hypoglycemic symptoms. It was not the case in our incidental pancreatic tumor, but we still thought it prudent to perform ASVS in light of the lower venous blood glucose noted in the general health check-up. The ASVS results provided strengthened evidence of a tumor with insulin-secreting function at the distal tail of the pancreas, where high-density enhancement had been identified by the CT imaging performed before our patient’s admission. We therefore proceeded with resection, the recommendable course of action under the circumstances [11]. Immunological staining revealed insulin secretion by more than 98% of the cells composing the resected pancreatic tumor, and the histological characteristics were consistent with insulinoma with overt hypoglycemic symptoms. The resected insulinoma was large enough to elicit insulin-causing hypoglycemic symptoms. Also noteworthy was the relatively advanced age of the patient for the diagnosis of an insulinoma, 61 years old. The patient may have spent all of those years without any symptoms of hypoglycemia. The findings described above raise the possibility that an asymptomatic insulinoma had been present for much of that time. To investigate this issue,
we consecutively evaluated 423 autopsy cases with occult gastroentero-pancreatic neuroendocrine tumors available for systematic examination, from autopsies performed at Mitsui Memorial Hospital over the almost 8-year period between April 2002 and January 2010 (Table 1). The autopsy rate in our hospital was about 16.7% on average during this period, which was somewhat higher than the rate reported by hospitals registered as members of the Japanese Society of Pathology, namely, 6.7% (in 2008). The pancreas acquired by our autopsies had been sectioned from head to tail at 1–3 mm intervals. All specimens were by hematoxylin–eosin staining. Islets of Langerhans consisting of a single cell type were detected microscopically in 6 cases (1.42%). The lesions of the islets of Langerhans of the pancreas were composed of monotonous cells in all of these cases. Two of the lesions detected were too small for cell type evaluation by the following immunohistochemical studies. The lesions of the other four cases were positive for glucagon but negative for insulin. We could not find insulinomas in any of the 423 autopsied cases. The incidence of asymptomatic pancreatic endocrine tumors in our hospital (1.42%) was similar to that (1.6%) observed by Kimura et al., who reported that 70% of asymptomatic endocrine tumors produced multiple hormones and the other 30% produced single hormone [24]. None of the asymptomatic pancreatic tumor observed by Kimura’s group consisted of insulin-producing cells [24]. In the study by Polak et al., the most commonly found cell type in asymptomatic endocrine tumors of the pancreas cells produced multiple hormones, insulin, and pancreatic polypeptide [25,26]. In our study, glucagon producing cells were 100% in the detected occult gastroentero-pancreatic neuroendocrine tumors, and all of cases were asymptomatic. No cases of multiple endocrine neoplasia (MEN) type 1 were observed. Computed tomography (CT) scans are growing more prevalent, spurred by recent advances in CT technology. Though previously found only in autopsy cases, incidentalomas and asymptomatic pancreas tumors are now being detected by chance during radiological or magnetic resonance methods, mostly by CT. Yet incidentalomas of the pancreas are distinct and less frequently detected by CT. The likelihood of detection of a pancreatic endocrine tumor by CT depends upon the size and location of the tumor. In our case, enhanced CT detected a pancreatic tumor of 10 mm in diameter. In the process of differential diagnosis we treated the tumor as an incidentaloma before the tumor was characterized as a functional endocrine tumor by ASVS. Though tumors as small as 10 mm in diameter have been located, those less than 20 mm cannot be found consistently or reproducibly. The
Table 1 – Immuno-histochemical findings of endocrine tumor. Case 1 2 3 4 5 6 a
Chief pathologic cause of death Renal cancer Myeloma Lymphoma Lung cancer Lung cancer Stomach cancer
Size of lesion
Insulin
Glucagon
Somatostatin
0.7 mm 0.6 mm, 0.6 mm 0.6 mm 3.6 mm 3.4 mm 1.7 mm 1.3 mm 0.4 mm 0.4 mm 2.3 mm 1.5 mm 0.1 mm 0.1 mm, 0.1 mm 0.1 mm
<1% – <1%
40% 40% 100%
<1% – –
a
a
a
1%
90%
<1%
a
a
a
There is not enough sample volume to evaluate by immunohistochemical studies, because the tumor is very small.
450
diabetes research and clinical practice 98 (2012) 445–451
sensitivity of the dynamic CT scan in the detection of insulinomas or gastrinomas has been reported to range from 30% to 66% [11,27–30]. In Japan, the advancing effectiveness of imaging modalities (CT, MRI, etc.) is evidenced by the high rate (24%) of pancreatic endocrine tumors that have been found incidentally during routine health examinations of asymptomatic patients [3]. Endocrine societies revised the diagnostic consensus guidelines in 2009 [31], after the admission examination of our case. This posed a limitation on our clinical case study, as the revised threshold for serum insulin at the fasting test was lowered to below the insulin value our case presented. Our reporting here on this clinical case is based on the diagnostic criteria widely used at that time.
4.
Conclusions
We experienced a case who presented with a reduced blood glucose level in a general health check-up and was subsequently found to have an incidentaloma of the pancreas by a CT examination. This case appeared to be rare: in spite of its size (10 mm), the insulinoma caused no hypoglycemic symptoms and showed no typical reaction to the biochemical screening test for insulinoma. There have been no reports of a decryption of asymptomatic insulinoma by pathological study. As imaging modalities such as CT, MRI and ultrasound advance, they may have the potential to detect early-stage pancreatic neoplasms, especially insulinomas, before the tumors grow large enough to cause symptoms. ASVS has been suggested to be helpful not only for occult insulinoma with overt hypoglycemic symptoms, but also in making differential diagnosis and assessing the locations of asymptomatic insulinomas.
Conflict of interest There are no conflicts of interest.
references
[1] Service FJ, McMahon MM, O’Brien PC, Ballard DJ. Functioning insulinoma – incidence, recurrence, and long-term survival of patients: a 60-year study. Mayo Clin Proc 1991;66:711–9. [2] Halfdanarson TR, Rubin J, Farnell MB, Grant CS, Petersen GM. Pancreatic endocrine neoplasms: epidemiology and prognosis of pancreatic endocrine tumors. Endocr Relat Cancer 2008;15:409–27. [3] Ito T, Sasano H, Tanaka M, Osamura RY, Sasaki I, Kimura W, et al. Epidemiological study of gastroenteropancreatic neuroendocrine tumors in Japan. J Gastroenterol 2010;45:234–43. [4] Vezzosi D, Bennet A, Fauvel J, Caron P. Insulin C-peptide and proinsulin for the biochemical diagnosis of hypoglycaemia related to endogenous hyperinsulinism. Eur J Endocrinol 2007;157:75–83. [5] Service FJ. Hypoglycemic disorders. N Engl J Med 1995;332:1144–52. [6] Service FJ, O’Brien PC, McMahon MM, Kao PC. C-peptide during the prolonged fast in insulinoma. J Clin Endocrinol Metab 1993;76:655–9.
[7] O’Brien T, O’Brien PC, Service FJ. Insulin surrogates in insulinoma. J Clin Endocrinol Metab 1993;77:448–51. [8] Kao PC, Taylor RL, Service FJ. Proinsulin by immunochemiluminometric assay for the diagnosis of insulinoma. J Clin Endocrinol Metab 1994;78:1048–51. [9] Soh AW, Kek PC. Insulinoma in a patient with negative prolonged fast and glucagon-induced hypoglycemia. Endocr Pract 2010;16:838–41. [10] Marks V. Progress report. Diagnosis of insulinoma. Gut 1971;12:835–43. [11] Doppman JL, Miller DL, Chang R, Shawker TH, Gorden P, Norton JA. Insulinomas: localization with selective intraarterial injection of calcium. Radiology 1991;178:237–41. [12] Van Hoe L, Gryspeerdt S, Marchal G, Baert AL, Mertens L. Helical CT for the preoperative localization of islet cell tumors of the pancreas: value of arterial and parenchymal phase images. AJR – Am J Roentgenol 1995;165:1437–9. [13] Schmitt AM, Anlauf M, Rousson V, Schmid S, Kofler A, Riniker F, et al. WHO 2004 criteria and CK19 are reliable prognostic markers in pancreatic endocrine tumors. Am J Surg Pathol 2007;31:1677–82. [14] Bettini R, Boninsegna L, Mantovani W, Capelli P, Bassi C, Pederzoli P, et al. Prognostic factors at diagnosis and value of WHO classification in a mono-institutional series of 180 non-functioning pancreatic endocrine tumours. Ann Oncol 2008;19:903–8. [15] Kloppel G, Perren A, Heitz PU. The gastroenteropancreatic neuroendocrine cell system and its tumors: the WHO classification. Ann N Y Acad Sci 2004;1014:13–27. [16] Schwartz NS, Clutter WE, Shah SD, Cryer PE. Glycemic thresholds for activation of glucose counterregulatory systems are higher than the threshold for symptoms. J Clin Invest 1987;79:777–81. [17] Mitrakou A, Fanelli C, Veneman T, Perriello G, Calderone S, Platanisiotis D, et al. Reversibility of unawareness of hypoglycemia in patients with insulinomas. N Engl J Med 1993;329:834–9. [18] Fanelli C, Pampanelli S, Epifano L, Rambotti AM, Ciofetta M, Modarelli F, et al. Relative roles of insulin and hypoglycaemia on induction of neuroendocrine responses to, symptoms of, and deterioration of cognitive function in hypoglycaemia in male and female humans. Diabetologia 1994;37:797–807. [19] Placzkowski KA, Vella A, Thompson GB, Grant CS, Reading CC, Charboneau JW, et al. Secular trends in the presentation and management of functioning insulinoma at the Mayo Clinic, 1987–2007. J Clin Endocrinol Metab 2009;94:1069–73. [20] Fanelli CG, Epifano L, Rambotti AM, Pampanelli S, Di Vincenzo A, Modarelli F, et al. Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes 1993;42:1683–9. [21] Vea H, Jorde R, Sager G, Vaaler S, Sundsfjord J, Revhaug A. Pre- and postoperative glucose levels for eliciting hypoglycaemic responses in a patient with insulinoma. Diabet Med 1992;9:950–3. [22] Wiesli P, Schmid C, Perren A, Pfammatter T, Spinas GA, Keller U. Hypoglycemia in response to glucose and glucagon in insulinoma patients with a negative prolonged fast: functional and morphological properties. J Endocrinol Invest 2004;27:832–8. [23] Kar P, Price P, Sawers S, Bhattacharya S, Reznek RH, Grossman AB. Insulinomas may present with normoglycemia after prolonged fasting but glucosestimulated hypoglycemia. J Clin Endocrinol Metab 2006;91:4733–6.
diabetes research and clinical practice 98 (2012) 445–451
[24] Kimura W, Kuroda A, Morioka Y. Clinical pathology of endocrine tumors of the pancreas. Analysis of autopsy cases. Dig Dis Sci 1991;36:933–42. [25] Polak JM, Bloom SR, Adrian TE, Heitz P, Bryant MG, Pearse AG. Pancreatic polypeptide in insulinomas, gastrinomas, vipomas, and glucagonomas. Lancet 1976;1:328–30. [26] Heitz PU, Kasper M, Polak JM, Kloppel G. Pancreatic endocrine tumors. Hum Pathol 1982;13:263–71. [27] Kurosaki Y, Kuramoto K, Itai Y. Hyperattenuating insulinoma at unenhanced CT. Abdom Imaging 1996;21:334–6. [28] Norton JA, Shawker TH, Doppman JL, Miller DL, Fraker DL, Cromack DT, et al. Localization and surgical treatment of occult insulinomas. Ann Surg 1990;212:615–20.
451
[29] Galiber AK, Reading CC, Charboneau JW, Sheedy 2nd PF, James EM, Gorman B, et al. Localization of pancreatic insulinoma: comparison of pre- and intraoperative US with CT and angiography. Radiology 1988;166:405–8. [30] Wank SA, Doppman JL, Miller DL, Collen MJ, Maton PN, Vinayek R, et al. Prospective study of the ability of computed axial tomography to localize gastrinomas in patients with Zollinger–Ellison syndrome. Gastroenterology 1987;92:905–12. [31] Cryer PE, Axelrod L, Grossman AB, Heller SR, Montori VM, Seaquist ER, et al. Evaluation and management of adult hypoglycemic disorders: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2009;94:709–28.