Insulinoma in Childhood: Clinical, Radiological, Molecular and Histological Aspects of Nine Patients

R Padidela and others Insulinoma in childhood 170:5 741–747 Clinical Study

Insulinoma in childhood: clinical, radiological, molecular and histological aspects of nine patients

Raja Padidela, Miriam Fiest1,2, Ved Arya1,2, Virpi V Smith3, Michael Ashworth3, Dyanne Rampling3, Melanie Newbould4, Gauri Batra4, Jacqueline James5, Neville B Wright6, Mark J Dunne7, Peter E Clayton, Indraneel Banerjee and Khalid Hussain1

Pediatric Endocrinology, Royal Manchester Children’s Hospital, Manchester M13 9WL, UK, 1Developmental Endocrinology Research Group, Clinical and Molecular Genetics Unit, University College London, Institute of Child Correspondence Health, 30 Guilford Street, London WC1N 1EH, UK, Departments of 2Endocrinology and 3Histopathology, Great should be addressed Ormond Street Hospital for Children NHS Trust, London WC1N 3JH, UK and Departments of 4Pediatric to K Hussain Histopathology, 5Nuclear Medicine, 6Pediatric Radiology, Royal Manchester Children’s Hospital, Manchester M13 Email 9WL, UK and 7Faculty of Life Sciences, University of Manchester, Manchester, M13 9PT, UK [email protected]

Abstract

Background: Insulinomas are a rare cause of hyperinsulinaemic hypoglycaemia (HH) in children. The clinical features, investigations, management and histology of these rare pancreatic tumours in children have not been described in a large cohort of patients. Methods: We conducted a retrospective review of cases diagnosed between 2000 and 2012, presenting to two referral centres in the United Kingdom. Clinical, biochemical, imaging (magnetic resonance imaging (MRI) and 18 18 6-L- F-ﬂuorodihydroxyphenylalanine ( F-DOPA) PET/CT scanning) and histological data were collected. Results: Nine children (age range 2–14.5 years) were diagnosed during the study period at Great Ormond Street Hospital (nZ5) and Royal Manchester Children’s Hospital (nZ4). The combination of abdominal MRI scan (7/8) and 18F-DOPA PET/CT

European Journal of Endocrinology scan (2/4) correctly localised the anatomical location of all insulinomas. Before surgery, diazoxide therapy was used to treat hypoglycaemia, but only four patients responded. After surgical resection of the insulinoma, hypoglycaemia resolved in all patients. The anatomical localisation of the insulinoma in each patient was head (nZ4), uncinate process (nZ4) and tail (nZ2, one second lesion) of the pancreas. Histology conﬁrmed the diagnosis of insulinoma with the presence of sheets and trabeculae of epithelioid and spindle cells staining strongly for insulin and proinsulin, but not for glucagon or somatostatin. Two children were positive for MEN1, one of whom had two separate insulinoma lesions within the pancreas. Conclusions: We describe a cohort of paediatric insulinoma patients. Although rare, insulinomas should be included in the differential diagnosis of HH, even in very young children. In the absence of a single imaging modality in the preoperative period, localisation of the tumour is achieved by combining imaging techniques, both conventional and functional.

European Journal of Endocrinology (2014) 170, 741–747

Introduction

Hyperinsulinaemic hypoglycaemia (HH) is an important neurological deﬁcit. HH is caused by unregulated insulin cause of recurrent and severe hypoglycaemia during secretion, which results in severe and recurrent hypo- infancy and childhood. Though rare, it is one of the glycaemia. Inappropriate insulin secretion inhibits commonest causes of hypoglycaemia, causing signiﬁcant glycogenolysis, gluconeogenesis, lipolysis and ketogenesis

www.eje-online.org Ñ 2014 European Society of Endocrinology Published by Bioscientiﬁca Ltd. DOI: 10.1530/EJE-13-1008 Printed in Great Britain

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access Clinical Study R Padidela and others Insulinoma in childhood 170:5 742

that prevents generation of ketone bodies (alternative 2000 to 2012. The clinical data was collected retro- energy sources) leading to neuroglycopaenia and hypo- spectively from the medical notes. glycaemic brain injury. All patients were managed as per common protocols In children, congenital hyperinsulinism (CHI) is the for managing hypoglycaemia due to HH. The diagnosis of most common cause for HH and it typically presents in HH was based on clinical symptoms of hypoglycaemia and early infancy. The incidence of severe CHI is estimated finding detectable insulin at the time of hypoglycaemia. at 1:50 000 (1). Beyond infancy, HH may also be due to All children with HH were screened for mutations in the another diagnosis – insulinoma, a neuroendocrine tumour common genes causing CHI, ABCC8 and KCNJ11. of the pancreas, which has been described mostly in Abdominal MRI scan with contrast was performed in adults. It is estimated that insulinoma occurs once in eight children to identify signal changes correlating with 250 000 persons/year (2). The incidence of insulinoma has probable insulinoma in the region of the pancreas. Four not been estimated in children. Insulinoma can occur patients diagnosed after 2007 were also investigated with 18 18 either in isolation or in association with multiple 6-L- F-fluorodihydroxyphenylalanine ( -F-DOPA) endocrine neoplasia type 1 (MEN1), with a lifetime PET/CT scan as per routine investigation policy in children prevalence of 10% among adults carrying mutations in with HH. In one patient, a localised lesion had already MEN1 (3). Around 6% of insulinomas occur in patients been ascertained on 18-F-DOPA PET/CT scan and hence with MEN1 (3). Most insulinomas are benign: 5–10% are MRI scanning was not undertaken. malignant (2). The diagnosis of insulinoma was confirmed by the An insulinoma is clinically suspected in the presence histological examination of excised pancreatic tissue. On of Whipple’s triad (symptoms and/or signs compatible the basis of the findings of radiological and surgical with hypoglycaemia, a low measured plasma glucose exploration, insulinoma was further classified into single concentration, and resolution of symptoms and signs or multiple and benign or malignant. Once the diagnosis when glucose concentrations are raised). However, a of insulinoma was confirmed, genetic testing was carried particular problem for evaluating hypoglycaemia in out for MEN1. In those carrying MEN1 mutations, the infants and children is that it may not be possible to presence of pituitary, parathyroid and other pancreatic satisfy all components of Whipple’s triad, because young tumours were assessed (4). children cannot dependably recognise and/or commu- nicate their symptoms. Insulinomas are difficult to detect Results in adults, as symptoms can be non-specific and present for

European Journal of Endocrinology a long period before diagnosis. In children, the presenting Table 1 describes the clinical, radiological, molecular and clinical features are also non-specific; importantly, the surgical characteristics of patients with insulinoma in our biochemical features of HH do not differentiate between cohort. Five patients with insulinoma were managed at CHI and insulinoma. Childhood insulinomas are the Great Ormond Street Hospital for Children, and four considered to be exceedingly rare with only single case were managed at the Royal Manchester Children’s reports in the literature. Therefore, information about the Hospital. The cohort consisted of five females and four disease patterns of childhood insulinoma is limited. The males. rarity of insulinoma and the absence of large cohort The age of onset of clinical symptoms ranged from studies prompted us to examine two large cohorts of 2.0 to 14.5 years. Clinical symptoms developed in two children with HH. We describe the clinical, radiological, patients between 2 and 5 years of age, in four patients molecular and histological aspects of nine patients with between 5 and 10 years of age and in three patients childhood insulinoma. between 10 and 15 years of age. The duration of symptoms before the diagnosis of HH was established, ranging from 1 month to 3 years with an average of around 10 months. All patients developed Subjects and methods neurological symptoms of hypoglycaemia. Three patients To identify the study cohort, we reviewed the database of were treated for seizure disorders with antiepileptic children with HH managed at the two referral centres: medications (2M, 3M and 6F). The fasting tolerance was Great Ormond Street Hospital, London and Royal Man- limited and ranged from 2.5 to 6 h. Preoperatively, chester Children’s Hospital. Nine patients with a diagnosis euglycaemia was maintained in four patients with of insulinoma were identified from the database from diazoxide (1F, 3M, 5M and 7F) and in one patient with

www.eje-online.org

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access European Journal of Endocrinology

Table 1 Clinical, radiological, molecular and surgical characteristics of patients with insulinomas. lnclStudy Clinical

Patient ID and gender (M/F) 1F 2M 3M 4F 5M 6F 7F 8F 9M

Age at diagnosis 3 years 4 5 years 7 years 6 months 8 years 8 years 10 years 8 13 years 13 years 15 years months months Duration of 1 year 4 months 4 months 2 years 12 months 3 years 8 months 1 month 2 months 6 months symptoms before referral Fast tolerance (h) 3 4 4 4 3.5 6 4.5 2.5 5 Hypoglycaemia screen Blood glucose 1.9 2.2 1.5 1.1 0.8 1.6 2 2.1 1.9 (mmol/l)

Insulin (mIU/l) 5.7 4.5 5.8 66.3 85 9.0 7.0 44.6 14.4 childhood in Insulinoma others and Padidela R Medications for Responsive to Partial response Responsive to Not responsive Responsive to Nil, surgical Responsive to Not responsive to Partial response hyperinsulinism diazoxide to diazoxide diazoxide to diazoxide diazoxide resection diazoxide diazoxide and to octreotide and immediately octreotide octreotide after diagnosis MRI scan ND Lesion on the Lesion on the Lesion on the Normal Lesion on the Lesion on the Lesion on the Lesion on the head of uncinate head of appearance head of the uncinate uncinate head of pancreas process pancreas of pancreas process process pancreas pancreas anterolateral to the portal vein PET/CT scan Focal uptake in ND Poor uptake in ND Focal lesion at ND ND Diffuse uptake ND uncinate uncinate the junction process of process of of tail and pancreas pancreas body of pancreas Localisation of PET/CT and MRI scan and MRI, PET/CT and MRI scan and PET/CT and MRI scan and MRI scan and MRI scan and MRI scan and lesion perioperative postoperative perioperative postopera- perioperative histology postoperative histology postoperative histology histology histology tive histology histology histology histology Type of Focal Subtotal Partial Subtotal Partial Focal Focal lesion- Focal lesionect- Subtotal pancreatectomy lesionectomy lesionectomy ectomy of omy with head with pancreatico- pancreatico- jejunostomy jejunostomy Downloaded fromBioscientifica.com at09/30/202104:00:40AM Site and size of Uncinate Head, 1 cm Uncinate process Between head Tail, 1.2 cm Head, 1.2 cm Uncinate Uncinate Head, 1.2 cm lesion on process, 1 cm and tail, 1.5 cm and body, process, 0.8 cm process, 2 cm pancreas 1.5 cm Mutation for Negative Positive Positive Negative Negative Negative Negative Negative Negative MEN1 c.249_252del- c.628_631delA- GTCT (deletion CAG (deletion at codons at codons 170 www.eje-online.org

83–84) 210–211) :5 Additional endo- Nil Parathyroid Screening for Nil Nil Missed clinical Nil Nil Nil crinopathies adenoma after MEN associated follow-up 5 years of endocrino- presentation pathies ongoing 743 via freeaccess ND, not done. Clinical Study R Padidela and others Insulinoma in childhood 170:5 744

and the surrounding pancreatic tissue was compressed, but was otherwise normal. The neoplastic cells were spindle shaped and epithelioid with abundant eosinophilic cytoplasm and rounded or oval, regular nuclei (Fig. 3A, B and C of patient 3M). Two types of gross histology were seen, one which comprised sheets of uniform spindle-shaped cells with open nuclei with stippled chromatin (4F and 7F) and the others showing a trabecular pattern. Mitotic figures were very sparse. There was strong cytoplasmic staining of the tumour cells with antibodies to insulin and proinsulin. Apart from one Figure 1 patient (2M) in whom scattered glucagon and somato- Axial and coronal T2 Blade magnetic resonance images (patient statin-immunoreactive cells were seen, the lesions had 3M) through the upper abdomen show a well-defined focal no staining for glucagon, somatostatin or pancreatic area of high signal in the uncinate process of the pancreas polypeptide. There was no necrosis or haemorrhage and (arrow). no infiltration of the pseudocapsule, blood vessels or lymphatics or of the surrounding pancreas. diazoxide and additional cornstarch in feeds (2M). Three Some variations were noted. In patient 5M, the lesion patients were unresponsive to diazoxide and octreotide was non-encapsulated and poorly circumscribed with (4F, 8F and 9M). islands of normal pancreatic tissues within the lesion. All patients were negative for ABCC8/KCNJ11 Patient 3M had two disconnected lesions with a larger mutations. Preoperatively, abdominal MRI scan was lesion in the uncinate process and a smaller different performed in eight patients and was successful in lesion in the tail of the pancreas, both of which were identifying an abnormality of the pancreas in seven removed by partial pancreatectomy. patients while it was normal in one patient (5M). The Two patients were positive for known MEN1 pancreatic lesions on MRI scan typically had a low-signal mutations (2M and 3M). Both are under regular review density on T1-weighted images and a high-signal density for evolving endocrinopathies associated with the MEN1 on T2-weighted images (Fig. 1 of patient 3M). Four 18

European Journal of Endocrinology patients underwent -F-DOPA PET/CT scanning: in two patients (1F and 5M) increased uptake and localised retention of isotope 18-F-DOPA on the pancreas suggested a solitary lesion (Fig. 2 of patient 1F) and in one patient (3M) a well-defined reduced uptake was identified and correlated anatomically to an abnormality on MRI scan. RL We did not identify any atypical clinical presentation or histology feature in 3M to account for the reduced uptake of 18-F-DOPA on PET/CT scanning. One patient had diffuse uptake of 18-F-DOPA on PET/CT scan (8F). All children underwent pancreatic surgery, following which euglycaemia was achieved. In follow-up assessment, all children remained euglycaemic, suggesting that in 1–12 years of follow-up, recurrence of insulinoma was not observed. Perioperative frozen section histology was carried out in all cases to confirm the site of the lesion and the Figure 2 completeness of excision. The histological diagnosis of Fused 45–60 min 18-F-DOPA PET/CT scanning static image and insulinoma was confirmed in all cases. The lesions showed contrast enhanced arterial phase CT: transaxial slice (Siemens well-demarcated nodules of endocrine tissue without Biograph MCT). Focal intense 18F-DOPA-PET/CT accumulation is hyperplasia of the exocrine components (Fig. 3). There seen in the pancreatic head (arrow) close to the uncinate was a fibrous pseudocapsule at the perimeter of the lesion process (arrow; Patient 1F).

www.eje-online.org

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access Clinical Study R Padidela and others Insulinoma in childhood 170:5 745

syndrome. Patient 2M was diagnosed with parathyroid AB adenoma 5 years after presentation with insulinoma, which was surgically removed. In patient 3M, no endocrinopathy has been identified till date. 500 µm ×10 ×10 Discussion C Insulinomas have the highest incidence in the fifth and sixth decades (2). Presentation during childhood is extremely rare. We have described a relatively large cohort of patients who presented during childhood with insulinoma in the UK over a 12-year period. Our case series shows that insulinomas can occur at a very young age and is a rare but important diagnosis in the investigation of children with HH. While CHI is the most common cause ×143 for persistent and recurrent hypoglycaemia in infancy (5), D E the possibility of insulinoma should be considered, especially in those with HH presenting beyond infancy. The diagnosis of insulinoma in adults can be delayed, (i) with only 20% of cases being diagnosed within a year of ×10 ×10 onset of symptoms (6). In comparison, children in our F cohort with insulinoma also experienced symptoms of hypoglycaemia for a prolonged duration (an average of 10 months) before the diagnosis of HH was confirmed. Neuroglycopenic symptoms (with seizures and syncopal episodes) were frequent in this cohort, suggesting a greater severity at presentation, in contrast to adults with insulinoma. In our cohort, all children were initially treated with

European Journal of Endocrinology diazoxide, but only four children were responsive (high ×143 doses of 10–20 mg/kg per day) and one was partially responsive. This observation is in contrast to the better Figure 3 response to diazoxide in adults, before surgery (7, 8, 9). Histology of insulinoma (A, B and C) and focal congenital Patients with GCK/GLUD1 mutation can present hyperinsulinism (D, E and F). Conventional microscopy in beyond infancy with insidious onset of hypoglycaemic insulinoma (case 3M), endocrine cells in insulinoma are episodes, which are usually responsive to diazoxide organised in nests and trabecular pattern scattered throughout (10, 11). Clinical or biochemical evidence of protein a fibro-vascular stroma with well-localised fibrous septa sensitivity should prompt mutation analysis for GLUD1 separating it from normal tissue. There is an absence of exocrine (11). In our patients, the diagnosis of CHI was considered, tissue with the loss of normal pancreatic architecture. Islets but genetic investigations for GCK/GLUD1 were not outside the lesion have a normal appearance. Immuno- carried out as children did not present with clinical histochemistry staining with insulin (B) shows abundance of characteristics suggestive of GCK/GLUD1. Nonetheless, it insulin secreting cells within the lesion compared with outside would be prudent to consider mutations in GCK/GLUD1 the lesion. By contrast, the focal congenital hyperinsulinaemic causing CHI before pursuing imaging investigations for lesion has an abundance of insulin-secreting cells within the the diagnosis of an insulinoma. lesion along with the presence of exocrine pancreatic tissue It is understood that neuroendocrine cells take up and (panel E(i), insert). Endocrine cells have large pleomorphic retain 18F-DOPA in focal CHI (12, 13). Similar principles nuclei (F; bold arrows). Scale bars in panels A, B, D and E may apply to insulinoma, although the results of represent 500 mm; scale bars in panels C, E(i) and F represent 50 mm. 18-F-DOPA PET have been generally inconsistent in C and F are magnified sections of boxes in A and D respectively. published adult cohorts (14, 15). In our cohort, the success

www.eje-online.org

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access Clinical Study R Padidela and others Insulinoma in childhood 170:5 746

rate of 18-F-DOPA PET/CT scan was 50% (2/4), which is well localised with fibrous septa separating it from normal inconsistent with a recent report in which insulinoma was tissue with the absence of exocrine pancreatic components. localised in all three children who underwent 18-F-DOPA Insulinoma is also characterised by abundance of insulin PET/CT scan (16). secreting cells with large nucleoli (with very little In our cohort, MRI alone localised pancreatic lesions in pleomorphism) and abundant eosinophilic cytoplasm 88% (7/8) of patients. In adult studies of insulinoma, the arranged in a trabecular pattern or as sheets of uniform utility of MRI scanning has been variable, with success rates spindle-shaped cells (25). Mostly, there is an absence of varying between 30 and 85% (17, 18, 19), which could be glucagon-secreting cells, although immunohistochemical attributable to the small tumour size at the time of staining to glucagon, somatostatin and chromogranin has diagnosis (17). Patient 5M, in whom MRI scan was non- been previously reported in a proportion of patients with confirmatory, had a tumour size similar to that in other insulinoma (26). Similarly, one patient (2M) in our study children. Even though multiple modes of diagnostic had scattered glucagon and somatostatin immune-positive techniques have been applied for preoperative localisation, cells in the tumour. none can be considered as gold standard. The intra-arterial calcium stimulation has been used variably in adults and children (16, 18). Endoscopic ultrasonography and radio- nucleotide scanning using radioactive tracers of octreotide Conclusions have not been successful in children (16). Other modalities, We have described nine cases of insulinoma in children such as spiral CT scans, intraoperative ultrasound (20, 21) managed in two large cohorts over a 12-year period. The and GLP1 receptor scintigraphy (22), need further evalu- differential diagnosis of children presenting beyond ation in children. Based on our experience and evidence infancy with HH should include insulinoma, even in from the literature, we recommend MRI scan as a first line those in early childhood. Based on our experience and of investigation for insulinoma. Development of novel availability of MRI in most centres, we recommend MRI sequences including diffusion-weighted imaging and scan as the first line of investigation of childhood novel software programmes may further improve pre- insulinoma, with due consideration given to other operative localisation with MRI scan (17). imaging modalities as second-line investigations. All In our cohort, two patients were diagnosed with the children with insulinoma should be carefully followed MEN1 syndrome, one of whom developed a parathyroid up for tumour recurrence and evolving endocrinopathy adenoma later. In adults, the 20-year recurrence risk has should be screened in those carrying MEN1 mutations. European Journal of Endocrinology been observed to be higher (21%) in patients with MEN1 than in those without MEN1 (7%) (2). In view of the small but distinct possibility of recurrence and the prospect of Declaration of interest an evolving endocrinopathy, all patients in our cohort are The authors declare that there is no conflict of interest that could be being carefully followed up with suitable transition to perceived as prejudicing the impartiality of the research reported. adult endocrine services. In one of the earliest descriptions, Mann et al. (1969) (23) summarised 39 published cases of insulinoma in Funding children. From a historical perspective, although the entity No grant or fellowship was provided for writing this article. This work was partially supported by the Wellcome Trust (097820/Z/11/A). of CHI had not been defined at the time; most cases (15/24) were in children older than 4 years and were therefore probably insulinomas. In recent years, there has been a Author contribution statement clear histopathological differentiation of insulinoma from R Padidela was involved in acquisition of data, analysis and interpretation CHI. Focal lesions (Fig. 3D, E and F) are characterised by the of data and drafting of the manuscript. K Hussain was involved in study presence of a conglomeration of islet cells without a clear concept, design and revision of all the drafts. M Fiest and V Arya contributed towards data collection. V V Smith, M Ashworth, D Rampling, margin of demarcation with additional ductal components M Newbould, G Batra and M J Dunne contributed towards histopathology and the presence of insulin and glucagon staining. Focal analyses and description in the manuscript. J James contibuted towards the lesions have large endocrine cells with large cytoplasm and nuclear medicine and N B Wright towards radiology sections of this large nuclei which are pleomorphic, irregular and angular manuscript. I Banerjee and P E Clayton contributed towards critical revision of the manuscript. All authors reviewed the final draft and gave final O in shape and 3–5 times the size of the nearby acinar approval before submission. All individuals who contributed to this study nuclei (24). By contrast, insulinoma (Fig. 3A, B and C) is are included as authors.

www.eje-online.org

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access Clinical Study R Padidela and others Insulinoma in childhood 170:5 747

14 Kauhanen S, Seppanen M, Minn H & Nuutila P. Clinical PET imaging of References insulinoma and b-cell hyperplasia. Current Pharmaceutical Design 2010 1 Bruining GJ. Recent advances in hyperinsulinism and the pathogenesis 16 1550–1560. (doi:10.2174/138161210791164090) of diabetes mellitus. Current Opinion in Pediatrics 1990 2 758–765. 15 Tessonnier L, Sebag F, Ghander C, De Micco C, Reynaud R, Palazzo FF, 18 (doi:10.1097/00008480-199008000-00024) Conte-Devolx B, Henry JF, Mundler O & Taieb D. Limited value of F-F- 2 Service FJ, McMahon MM, O’Brien PC & Ballard DJ. Functioning DOPA PET to localize pancreatic insulin-secreting tumors in adults with insulinoma – incidence, recurrence, and long-term survival of patients: hyperinsulinemic hypoglycemia. Journal of Clinical Endocrinology and a 60-year study. Mayo Clinic Proceedings 1991 66 711–719. (doi:10.1016/ Metabolism 2010 95 303–307. (doi:10.1210/jc.2009-1357) S0025-6196(12)62083-7) 16 Peranteau WH, Palladino AA, Bhatti TR, Becker SA, States LJ, Stanley CA 3 Placzkowski KA, Vella A, Thompson GB, Grant CS, Reading CC, & Adzick NS. The surgical management of insulinomas in children. Charboneau JW, Andrews JC, Lloyd RV & Service FJ. Secular trends in Journal of Pediatric Surgery 2013 48 2517–2524. (doi:10.1016/j.jpedsurg. the presentation and management of functioning insulinoma at the 2013.04.022) mayo clinic, 1987–2007. Journal of Clinical Endocrinology and Metabolism 17 Druce MR, Muthuppalaniappan VM, O’Leary B, Chew SL, Drake WM, 2009 94 1069–1073. (doi:10.1210/jc.2008-2031) Monson JP, Akker SA, Besser M, Sahdev A, Rockall A et al. Diagnosis and 4 Thakker RV, Newey PJ, Walls GV, Bilezikian J, Dralle H, Ebeling PR, localisation of insulinoma: the value of modern magnetic resonance Melmed S, Sakurai A, Tonelli F & Brandi ML. Clinical practice guidelines imaging in conjunction with calcium stimulation catheterisation. for multiple endocrine neoplasia type 1 (MEN1). Journal of Clinical European Journal of Endocrinology 2010 162 971–978. (doi:10.1530/EJE- Endocrinology and Metabolism 2012 97 2990–3011. (doi:10.1210/jc. 10-0056) 2012-1230) 18 Guettier JM, Kam A, Chang R, Skarulis MC, Cochran C, Alexander HR, 5 Kapoor RR, Flanagan SE, Arya VB, Shield JP, Ellard S & Hussain K. Libutti SK, Pingpank JF & Gorden P. Localization of insulinomas to Clinical and molecular characterisation of 300 patients with congenital regions of the pancreas by intraarterial calcium stimulation: the NIH hyperinsulinism. European Journal of Endocrinology 2013 168 557–564. experience. Journal of Clinical Endocrinology and Metabolism 2009 94 (doi:10.1530/EJE-12-0673) 1074–1080. (doi:10.1210/jc.2008-1986) 6 Chen X, Cai WY, Yang WP & Li HW. Pancreatic insulinomas: diagnosis 19 Thoeni RF, Mueller-Lisse UG, Chan R, Do NK & Shyn PB. Detection of and surgical treatment of 74 patients. Hepatobiliary & Pancreatic Diseases small, functional islet cell tumors in the pancreas: selection of MR International 2002 1 458–461. imaging sequences for optimal sensitivity. Radiology 2000 214 483–490. 7 Gill GV, Rauf O & MacFarlane IA. Diazoxide treatment for insulinoma: (doi:10.1148/radiology.214.2.r00fe32483) a national UK survey. Postgraduate Medical Journal 1997 73 640–641. 20 Jaksic T, Yaman M, Thorner P, Wesson DK, Filler RM & Shandling B. (doi:10.1136/pgmj.73.864.640) A 20-year review of pediatric pancreatic tumors. Journal of Pediatric 8 Goode PN, Farndon JR, Anderson J, Johnston ID & Morte JA. Diazoxide Surgery 1992 27 1315–1317. (doi:10.1016/0022-3468(92)90284-E) in the management of patients with insulinoma. World Journal of 21 Kalafat H, Mihmanli I, Saribeyoglu K & Belli A. Intraoperative doppler Surgery 1986 10 586–592. (doi:10.1007/BF01655532) ultrasound: a reliable diagnostic method in insulinoma. Hepatogas- 9 Hirshberg B, Cochran C, Skarulis MC, Libutti SK, Alexander HR, troenterology 2007 54 1256–1258. Wood BJ, Chang R, Kleiner DE & Gorden P. Malignant insulinoma: 22 Christ E, Wild D, Forrer F, Brandle M, Sahli R, Clerici T, Gloor B, spectrum of unusual clinical features. Cancer 2005 104 264–272. Martius F, Maecke H & Reubi JC. Glucagon-like peptide-1 receptor (doi:10.1002/cncr.21179) imaging for localization of insulinomas. Journal of Clinical Endocrinology 10 Christesen HB, Tribble ND, Molven A, Siddiqui J, Sandal T, Brusgaard K, and Metabolism 2009 94 4398–4405. (doi:10.1210/jc.2009-1082) Ellard S, Njolstad PR, Alm J, Brock Jacobsen B et al. Activating 23 Mann JR, Rayner PH & Gourevitch A. Insulinoma in childhood. glucokinase (GCK) mutations as a cause of medically responsive Archives of Disease in Childhood 1969 44 435–442. (doi:10.1136/adc.44. European Journal of Endocrinology congenital hyperinsulinism: prevalence in children and character- 236.435) isation of a novel GCK mutation. European Journal of Endocrinology 2008 24 Arnoux JB, Verkarre V, Saint-Martin C, Montravers F, Brassier A, 159 27–34. (doi:10.1530/EJE-08-0203) Valayannopoulos V, Brunelle F, Fournet JC, Robert JJ, Aigrain Y et al. 11 Kapoor RR, Flanagan SE, Fulton P, Chakrapani A, Chadefaux B, Congenital hyperinsulinism: current trends in diagnosis and Ben-Omran T, Banerjee I, Shield JP, Ellard S & Hussain K. therapy. Orphanet Journal of Rare Diseases 2011 6 63. (doi:10.1186/1750- Hyperinsulinism-hyperammonaemia syndrome: novel mutations in 1172-6-63) the GLUD1 gene and genotype–phenotype correlations. European 25 Sempoux C, Guiot Y, Dahan K, Moulin P, Stevens M, Lambot V, Journal of Endocrinology 2009 161 731–735. (doi:10.1530/EJE-09-0615) de Lonlay P, Fournet JC, Junien C, Jaubert F et al. The focal form of 12 Hardy OT, Hernandez-Pampaloni M, Saffer JR, Suchi M, Ruchelli E, persistent hyperinsulinemic hypoglycemia of infancy: Zhuang H, Ganguly A, Freifelder R, Adzick NS, Alavi A et al. Diagnosis morphological and molecular studies show structural and and localization of focal congenital hyperinsulinism by 18F-ﬂuorodopa functional differences with insulinoma. Diabetes 2003 52 784–794. PET scan. Journal of Pediatrics 2007 150 140–145. (doi:10.1016/j.jpeds. (doi:10.2337/diabetes.52.3.784) 2006.08.028) 26 Lam KY & Lo CY. Pancreatic endocrine tumour: a 22-year clinico- 13 Lindstrom P. Aromatic-L-amino-acid decarboxylase activity in mouse pathological experience with morphological, immunohistochemical pancreatic islets. Biochimica et Biophysica Acta 1986 884 276–281. observation and a review of the literature. European Journal of Surgical (doi:10.1016/0304-4165(86)90174-1) Oncology 1997 23 36–42. (doi:10.1016/S0748-7983(97)80140-0)

Received 8 December 2013 Revised version received 28 February 2014 Accepted 5 March 2014

www.eje-online.org

Downloaded from Bioscientifica.com at 09/30/2021 04:00:40AM via free access