1455-1.full

7/27/2019 1455-1.full

1/7

Neonatal Hyperinsulinemic Hypoglycemia:Heterogeneity of the Syndrome and Keys for Differential

Diagnosis*C. SEMPOUX, Y. GUIOT, A. LEFEVRE, C. NIHOUL-FEKETE, F. JAUBERT,J-M. SAUDUBRAY, AND J. RAHIER

Department of Pathology, University Hospital St. Luc (C.S., Y.G., A.L., J.R.), 1200 Brussels, Belgium;Departments of Infantile Surgery (C.N-F.), Pathology (F.J.), and Medical Genetics (J-M.S.), HopitalNecker Enfants Malades, 75743 Paris, France

ABSTRACTThe two major forms of infantile persistent hyperinsulinemic hy-

poglycemia require different treatments, but are difficult to differ-entiate during surgery. Indeed, one is characterized by focal adeno-matous hyperplasia often macroscopically invisible, whereas theother consists of a diffuse, but discreet, -cell abnormality. We eval-

uated, in a large series of persistent hyperinsulinemic hypoglycemia,the reliability of two criteria in differentiating these two forms: themean -cellnuclearradius (MNR) andthe -cellnuclearcrowding, i.e.the number of nuclei per 1000 m2 -cell (BCNC). The values of thelargest MNRandof BCNC incasesbearinga focallesion(respectively,3.27 m 0.25 and 14.62 1.78) were significantly different from

those in the diffuse pathology (4.25 m 0.43 and 10.00 1.55)Setting the threshold value of MNR at 3.70 m and that of BCNC a12.00 enabled correct classification of 90.9% of the diffuse and 100%of the focal forms.-Cell nuclear analysis can thus contribute to a subclassification

of the syndrome, not allowed by clinical or biological data. If per-

formed during surgery it could help in determining the extent ofpancreatectomy necessary to cure the patient, as the diffuse formwith abnormal nuclei in the whole pancreas, requires subtotal tonear-total pancreatectomy, whereas the focal form, devoid of abnormal insular -cell nuclei, can be cured by partial pancreatectomy(J Clin Endocrinol Metab 83: 14551461, 1998)

ALTHOUGH numerous studies have been devoted tothe syndrome of persistent hyperinsulinemic hypo-glycemia of neonates and infants (PHHI), its pathogenesis isnot completely elucidated (116). Linkage analysis hasmapped hyperinsulinism in familial cases to chromosome 11(17), and other genetic molecular analysis has revealed sul-fonylurea receptor gene abnormalities that might explaininsulin hypersecretion in such familial cases (18). The ab-sence of functional KATP channels has recently been shownto induce insulin hypersecretion in sporadic PHHI (19).However, morphological diagnosis of the disease remainsdifficult, especially when the clinical history of recurrenthypoglycemic events is not known by the pathologist, as incertain cases of sudden infant death. Nesidioblastosis, de-fined as a persistent endocrine cell proliferation buddingfrom pancreatic ducts (20), can be observed in the pancreasof normoglycemic neonates (3, 4, 6, 10, 13) and, therefore, isnot pathognomonic of this syndrome. Furthermore, reliablemorphological criteria for the evaluation of nesidioblastosishave yet to be defined.

There is no consensus about the optimal clinical manage-

ment of the disease (2133). Diazoxide is usually the firstdrug to be tried, but it is often not effective, especially in

severe hyperinsulinemic hypoglycemia of neonatal onsetSomatostatin has also been recommended and, with the ar-rival of long acting analogs, pancreatectomy or further re-sections have been avoided in certain patients (24, 26, 28, 33)When medical treatment fails, surgery is unavoidable toprevent permanent brain damage. The extent of the pancre-atectomy is, however, a matter of debate, as some researchersrecommend an initial 95% pancreatectomy in all patients(22), which may increase the long term risk of diabetes.

Several studies have clearly established the existence oftwo forms of PHHI, one characterized by focal pancreaticadenomatous hyperplasia (focal PHHI) and the other char-acterized by a diffuse -cell abnormality (diffuse PHHI) (510, 13, 16, 34).

The distinction between these two forms should be ofparamount importance, because infants suffering from thefocal form may be cured by a partial pancreatectomy (35)However, for several reasons, few clinicians take this dis-tinction into account when determining the extent of pan-createctomy. Firstly, the experience of most clinicians is gen-erally limited because of the rarity of the disease. Secondly

contrary to adult insular adenomas, focal lesions in infantsare often difficult and sometimes impossible to identify mac-roscopically, as they are lobulated like the normal pancreaticparenchyma. Thirdly, the selective catheterization recommended by our group guides the surgeon in locating a focalesion, but is, as yet, not readily available in all hospitalsFourthly, this technique is not infallible in differentiating thefocal from the diffuse form of PHHI (36, 37).

In a previous study of 16 cases (13), we have drawn at-tention to thefact that morphology could help to differentiate

Received June 16, 1997. Revision received November 14, 1997. Ac-cepted December 15, 1997.

Address all correspondence and requests for reprints to: JacquesRahier, M.D., Ph.D, Department of Pathology ANPS 1712, University ofLouvain Medical School, Cliniques St. Luc, 10 avenue Hippocrate, 1200Brussels, Belgium.

* This work was supported by Grant 3461593 from the Fonds de laRecherche Scientifique et Medicale, Brussels.

0021-972X/98/$03.00/0 Vol. 83, No.Journal of Clinical Endocrinology and Metabolism Printed in U.S.ACopyright 1998 by The Endocrine Society

1455

7/27/2019 1455-1.full

2/7

TABLE 1. Descriptions of the patients

Case no. Sex Birth wt (g)

Age at

Extent of pancreatectomy Follow-upOnset ofsymptoms (days)

Pancreasresection (months)

1 M 4700 1 6 Subtotal No symptomatic hypoglycemia2 F 2680 1 1 Subtotal Recurrent hypoglycemia

31 wk gestation Requires steroids3 M 4520 1 4 90% Diabetes4 F 3800 1 2 85% Requires increased nutritional supply

5 F 2700 1 4 Partial Recurrent hypoglycemia33 wk gestation 10 Near-total Diabetes

6 M 4350 1 2 85% Recurrent hypoglycemia7 F 3000 1 14 days 90% No symptomatic hypoglycemia8 F 3900 3 4 Partial (isthmic) Recurrent hypoglycemia

5 Near-total Diabetes9 M 3800 1 13 days 85% No symptomatic hypoglycemia

10 F 3220 1 6.5 90% Recurrent hypoglycemiaRequires steroids

11 F 3500 1 4 80% Recurrent hypoglycemia34 wk gestation 13 Near-total Diabetes

12 M 4280 1 1.5 Subtotal No symptomatic hypoglycemia13 M 3800 1 9 Subtotal Diabetes14 M 4110 1 12 Subtotal Mild glucose intolerance15 M 2960 1 3.5 Partial Recurrent hypoglycemia

36 wk gestation 6 Near-total No symptomatic hypoglycemia16 M 3390 2 5 Subtotal Requires increased nutritional supply

17 F 3880 1 3.5 Near-total Diabetes38 wk gestation

18 M 2850 5 3 Subtotal Diabetes19 F 3350 1 13 Near-total Diabetes20 M 3880 1 4 Near-total No symptomatic hypoglycemia21 F 5000 1 mo 6.5 Partial Recurrent hypoglycemia

8 Near-total Transient diabetes22 F 2460 1 3.5 Near-total Recurrent hypoglycemia

31 wk gestation Requiring diazoxide23 F 4845 1 3 Subtotal Recurrent hypoglycemia

Requires increased nutritional supply24 M 5620 12 3 Subtotal No symptomatic hypoglycemia25 F 3260 2 5 Subtotal Transient diabetes26 M 3330 1 1.5 Partial Recurrent hypoglycemia

Near-total Requires increased nutritional supply27 F 3840 1 3 Subtotal Recurrent hypoglycemia requiring dia28 M 3700 1 16 days 80% No symptomatic hypoglycemia

29 F 3900 3 1.5 90% Recurrence requiring diazoxide30 M 3 1 Subtotal No symptomatic hypoglycemia31 F 2700 2 3 Subtotal Recurrent hypoglycemia

13 Near-total Diabetes32 M 4800 2 2 Subtotal Recurrent hypoglycemia

10 Near-total Diabetes33 F 4500 1 8 Subtotal Recurrent hypoglycemia

10 Near-total Diabetes34 F 3800 1 6.5 Subtotal Recurrent hypoglycemia

38 wk gestation 10 Near-total Diabetes

7/27/2019 1455-1.full

3/7

the two forms of the syndrome. Indeed, we showed that thediffuse form of PHHI could be characterized by the presenceof numerous abnormal large -cell nuclei, whereas such nu-clei were not observed in the insular -cells of cases bearinga focal lesion.

In the present study, we have evaluated -cell nuclearanalysis and other morphological criteria from a large ret

rospective series to differentiate the two forms of the syn-drome, with the hope that these criteria might be used todetermine the extent of pancreatectomy during surgery.

Materials and Methods

Partial or near-total pancreatectomy specimens were obtained from50 infants suffering from diazoxide-resistant hyperinsulinemic hypoglycemia of neonatal onset. Infants with late-onset hypoglycemia (3months of age) and hypoglycemic children were excluded from thisstudy.

Each resected pancreas was cut into thin (1- to 2-mm) slices. Aftefixation and embedding, 5-m sections were cut from each slice, stainedwith hematoxylin-eosin, and processed by immunoperoxidase technique to reveal insulin and ensure recognition of even very small focalesions of adenomatous hyperplasia. For insulin immunodetection, the

sections were incubated overnight with a monoclonal antiinsulin anti-body (HUI-018, Novo-Biolabs, Bagvaerd, Denmark) at a dilution o1:2000. After rinsing, the sections were successively treated by antimouse IgG-biotin conjugate (1089-285, Boehringer Mannheim, Mannheim, Germany) at a dilution of 1:500 and a streptavidin peroxidaseconjugate (1089-153, Boehringer Mannheim) at a dilution of 1:1000Before counterstaining in Mayers hematoxylin, peroxidase activity wasrevealed by immersion in a solution of 3,3-diaminobenzidine hydrochloride (50 mg/mL Tris-HCl, pH 7.4; Amersham, Aylesbury, UK) containing 0.02% H

2O2.

After verification that the morphological features were identical ineach section, two sections were selected at random for -cell nucleameasurements. Themeasurements were made by oneobserverwho wanotawareof thefinaldiagnosis (diffuse or focal). Thetwo whole sectionwere screened to analyze the -cell nuclei. In a first series of 20 cases (6focal and 14 diffuse lesions), we initially measured the mean nuclearradius (MNR) of 500 randomly selected insular -cell nuclei at a 1200magnification with a camera lucida connected to an Orthoplan Leitzmicroscope (Leitz, Wetzler, Germany) and a semiautomatic image analyzer (Videoplan Kontron, Munich, Germany). Thereafter, we measured only the largest insular -cell nuclei, calculating the mean nuclearradius of the 50 largest nuclei (50-MNR). On the same slides and withthe same device at a 300 magnification, we measured -cell nucleacrowding (BCNC), which is the number of-cell nuclei per 1000 m2 o-cell cytoplasm. This parameter was chosen rather than the nuclear/cytoplasmic ratio because the cytoplasmic limits of individual -cellwere difficult to define accurately on these specimens. This has theadvantage of being independent of the nuclear size variation of the-cells. The BCNC was systematically measured in all islets present inthe sections from the two randomly selected paraffin blocks.

Results are expressed as the mean sd. The statistical significance othe results was assessed by the Wilcoxon rank-sum test.

Results

Seventeen of the 50 specimens studied showed focal ad-enomatous hyperplasia at microscopic examination (Table 1patients 3450). This lesion is very different from adenomasoccurring in children and adults, as in neonates it consists ofan agglomerate of apparently normal islets with a peripheraring of non--cells (Fig. 1, A and B). In every case but onethe lesion was unifocal. The size of the lesions varied from2.513.0 mm in diameter. Their limits were often irregularWithin the lesions, the nuclear polymorphism of the -celnuclei was evident (Fig. 2A), but outside the lesions, -celT

ABLE

1.

Continued

Caseno.

Sex

Birthwt(

g)

Ageat

Extentofpancreatectomy

Follow-up

Histologicalfindings

Onsetof

symptoms(days)

Pancreas

resection(months)

Focal

lesion

Abnormal

insular

-cellnuclei

35

F

4720

1

1

Partial

Normoglycemiawithoutrecurrence

36

M

4070

1

1

Subtotal


37

F

2900

2

5.5

Subtotal


38

F

3370

1

2.5

Partial(65%)


39

F

4155

1

1.5

Partial(65%)


40

M

2250

1

2

Partial(cephalic)


41

F

3935

2mo

4.5

Partial(cephalic)


42

F

3380

2

2

Partial(caudal)


43

M

3450

1

3.5

Partial(cephalic)


44

F

3900

1

3

Partial

Recurrenthypoglycemia

3.5

Subtotal


45

F

4220

1

3

Partial(cephalo-isthmic)


46

M

4250

1

4

Partial(isthmic)


47

F

3140

2

5

Partial(corporeo-caudal)


48

F

4135

2

5

Partial(corporeo-caudal)


49

M

3130

3

2.5

Partial(isthmicandhemicorporeal)


50

F

3000

3

2

Partial(caudal)


DIFFERENTIAL DIAGNOSIS OF NESIDIOBLASTOSIS 1457

7/27/2019 1455-1.full

4/7

nuclei were regular and small, the -cell cytoplasm wasscanty, and these cells often appeared squeezed together(Fig. 2B). The MNR measured in 500 randomly selected in-sular -cells was 2.95 0.09 m (Fig. 3A), and the MNR ofthe 50 selected largest radii of insular -cell nuclei (50-MNR)was 3.27 0.25 m (Fig. 3B). The BCNC was 14.62 1.78-cell nuclei/1000 m2 (Fig. 3C). Fifteen of these 17 patients

were cured as a result of the first operation, including the 13patients treated by partial pancreatectomy, whereas hypo-glycemia recurred in only two infants: in one case a secondfocal lesion was found during the second operation (no, 34),and in another case the lesion had not been completely re-sected (no. 44).

In the other 33 infants (Table 1, patients 133), no focallesion could be found after systematic and detailed macro-scopic and microscopic analyses of the whole resected spec-imens. Nine of these infants, despite their 8090% pancre-atectomy, suffered from severe recurrent hypoglycemicattacks requiring near-total pancreatectomy. All of thesepan-creatic specimens (including second operation specimens)displayed similar morphological features; the islets were not

evidently more abundant than in the pancreas of infants witha focal lesion. -Cells with very large nuclei were observedin all but 3 cases (no. 6, 7, and 9). These -cells with anextremely large nucleus also showed a large and clear cy-toplasm and were easily recognizable even at low magnifi-cation (Fig. 2C). Their frequency varied from case to case. Incertain infants, they were present in almost every islet,whereas in others, only a few islets had such abnormal -cell

nuclei. These abnormal nuclei were evenly distributedthroughout the pancreas and were found in each section ofthese 33 specimens of neonatal-onset PHHI. The -cell MNRwas 3.15 0.02 m. This value was statistically higher thanthat measured in the insular -cells of cases with a focalesion (P 0.05), but the overlap would not allow a reliablediscrimination between the two groups (Fig. 3A). The mean

nuclear radius of the 50 largest -cell nuclei was 4.25 0.43(Fig. 3B), a value significantly higher than that of insular-cells of infants with a focal lesion (P 0.001), with only 3patients nuclei having radii smaller than 3.70 m (no. 6, 7and 9). The nuclear crowding of the -cells was 10.00 1.55-cell nuclei/1000 m2 (P 0.001 vs. cases with a focalesion), with only 3 patients having densities above 12.00-cell nuclei/1000 m2 (no. 7, 9, and 25; Fig. 3C), two ofwhom corresponded to the smaller 50-MNR (no. 7 and 9; Fig3B).

Setting the threshold values for 50-MNR at 3.70 m and12.00 for the BCNC analysis enabled correct classification of90.9% of the diffuse forms at PHHI and 100% of the focalcases.

Discussion

The recognition that a focal lesion(s) could be responsiblefor PHHI is not new (20, 38), and some researchers havesuggested that all cases of PHHI result from the presence ofa focal lesion, often so small that it escapes diagnosis (6). Theexistence of 2 forms of PHHI suggested by our previous

FIG. 1. A, Focal adenomatous hyperplasia corresponds to an agglomerate of apparentlynormal islets (arrowheads; hematoxylin-eosin stainingmagnification,2.5). B, Indeed, as in normal islets, non--cells are located at the periphery (immunostaining of somatostatin; magnification10).

1458 SEMPOUX ET AL. JCE & M 1998Vol 83 No 5

7/27/2019 1455-1.full

5/7

7/27/2019 1455-1.full

6/7

analysis of a series of 16 cases (13) has been confirmed (5).Although now well documented, this concept of 2 distinctforms requiring different types of treatment is not taken intoaccount by most researchers who recommend an initial near-total pancreatectomy in all cases of PHHI resistantto medicaltherapy regardless of whether itspathology is focal or diffuse(22). This probably results from the difficulties of the differ-ential diagnosis.

In the present study, we have tried to define precisely thepathognomic morphological criteria enabling this diagnosis.We have thus combined conventional microscopy, to searchfor the presence of abnormal -cell nuclei and evaluate thenucleo-cytoplasmic ratio, with morphometry, to accuratelyanalyze the characteristic differences between islets in bothfocal and diffuse forms of PHHI. The morphometric analysisdid not change the classification into focal or diffuse, but itclearly established the relevance of the criteria selected forthe differential diagnosis. In the first series of 20 cases, where

we measured 500 insular -cell nuclei, although the MNR ofthe -cell nuclei was significantly higher with a diffuse lesionthan with a focal lesion, this parameter did not allow a cleardistinction between the 2 groups, as 21% of the patients withdiffuse pathology were within the range of those with a focalesion. The frequency of the large -cell nuclei characteristicof the diffuse form of PHHI was probably too low in certain

cases to influence the MNR. For this reason, we decided tofocus our measurement specifically on the 50 largest -celnuclei (50-MNR). This method of evaluation proved to be afast measurement and increased the sensitivity of the anal-ysis, because theresults were notinfluenced by thefrequencyof these abnormal nuclei. It enabled differential diagnosis

between the 2 forms of the disease. Instead of measuring thenucleo-cytoplasmic ratio, we also decided to measure BCNCThis method, giving the mean amount of cytoplasm per-cell, proved to be fast and easy, and again had the advan-tage of being independent of the nuclear size variation of the-cells.

Both 50-MNR and BCNC are linked to cellular functionIndeed, an increase in nuclear size has repeatedly been dem-

onstrated in hyperfunctional endocrine cells (3942). On theother hand, although an increase in cytoplasmic size is notpathognomonic of hyperfunction, they are generally linkedThus, it is likely that hyperinsulinism in the diffuse form ofthe disease is related to an increased function of the -cellsas their number has been demonstrated to be normal (10, 1316). It has recently been suggested that in this form, hyper-insulinism results from the absence of functional KATPchannels in -cells (19), but we still do not know whetherall cases of diffuse PHHI are the consequence of thisabnormality.

In the focal form of the disease, the 50-MNR value ofinsular -cells located away from the focal lesion was neverhigher than 3.70 m. This was quite the opposite in cases

without focal lesion, where numerous abnormal-cell nuclewere observed, and where all but three cases had 50-MNRvalues higher than 3.70 m. As two of these three cases alsohad quite distinct values for the BCNC (cases 7 and 9), it isnot excluded that they actually belong to the group of focallesions and that the lesion has not been recognized becauseof its small size. On the other hand, we have no explanationfor case 6, whose 50-MNR value is lower than 3.70 m andwhose BCNC remains inferior to 12.00, or for case 25, whohas a high 50-MNR value and BCNC. This might suggest thaother forms of PHHI exist, different from those we havealready reported.

Despite these exceptions, it can be claimed that the pres-ence of obvious abnormal -cell nuclei in the different partsof the pancreas has never been observed in the presence ofa focal lesion. It is interesting to note that in all cases devoidof abnormal nuclei, hyperinsulinemic hypoglycemia has al-ways been radically cured by partial pancreatectomy, some-times even limited to the focal lesion itself, as long as the focalesion was unique and totally resected. On the other hand,when abnormal -cell nuclei were observed in the differentparts of the pancreas, none of the infants with hypoglycemiaof neonatal onset was cured by partial pancreatectomy. Pre-liminary results suggest that the recognition of these abnormal nuclei may be performed on frozen sections during

FIG. 3. A, Mean radius of 500 nonselected -cell nuclei (microns)measured outside the focal lesions () and in cases without focallesion ( ). B, Mean radius of 50 selected -cell nuclei (microns) mea-sured outside the focal lesions () and in cases without focal lesion( ). , Threshold value. C, BCNC (number per 1000m2) measuredoutside the focal lesions () and in cases without focal lesion ( ). , Threshold value. Numbers refer to Table 1.

1460 SEMPOUX ET AL. JCE & M 1998Vol 83 No 5

7/27/2019 1455-1.full

7/7

surgery (43). This could obviously help the surgeon, as it isnow clear that the two forms of PHHI require different treat-ments: the diffuse form, associated with abnormal -cellnuclei in the whole pancreas and a low BCNC requires sub-total pancreatectomy, whereas this cannot be justified in thecase of a focal lesion, which can be recognized by the absenceof abnormal -cell nuclei, and of which the patient can be

cured by partial pancreatectomy restricted to the lesion.

Acknowledgments

We gratefully acknowledge the kind help from our colleagues (Prof.Brauner and Dr. Poggi, Hopital Necker (Paris, France); Profs. Czerni-chow and Aigrain and Dr. Touati, Hopital Debre (Paris, France); Profs.Creusy and Vittu, Hopital St. Antoine (Lille, France); Profs. Craen andDe Rom, Universitair Ziekenhuis (Ghent, Belgium); Prof. Geis, HopitalHautepierre(Strasbourg, France); and Profs. de Ville, Maes,Moulin, andOtte, Cliniques Universitaires St. Luc, Universite Catholique de Lourain(Brussels, Belgium),who allowed us tostudythe pancreasof their youngpatients andprovidedus with clinical data. We thank Prof. J-C.Henquinand Dr. C. de Burbure for critical reading of the manuscript, and D.Dubois, S. Lagasse and M. April for technical, photographic, and edi-torial assistance.

References1. Ariel I,Kerem E,Schwartz-Arad D, et al. 1988 Nesidiodysplasia. A histologic

entity? Hum Pathol. 19:12151218.2. Bishop AE, Polak JM, Chesa PG, Timson CM, Bryant MG, Bloom SR. 1981

Decrease of pancreatic somatostatin in neonatal nesidioblastosis. Diabetes.30:122126.

3. Dahms BB, Landing BH, Blaskovics M, Roe TF. 1980 Nesidioblastosis andother islet cell abnormalities in hyperinsulinemic hypoglycemia of childhood.Hum Pathol. 11:641649.

4. Falkmer S, Sovik O, Vidnes J. 1981 Immunohistochemical, morphometric,and clinical studies of the pancreatic islets in infants with persistent neonatalhypoglycemiaof familialtype withhyperinsulinismand nesidioblastosis.ActaBiol Med Ger. 40:3954.

5. Goossens A, Gepts W, Saudubray JM, et al. 1989 Diffuse and focal nesidio-blastosis. A clinicopathological study of 24 patients with persistent neonatalhyperinsulinemic hypoglycemia. Am J Surg Pathol. 13:766775.

6. Goudswaard WB, Houthoff HJ, Koudstaal J, Zwierstra RP. 1986 Nesidio-blastosis and endocrine hyperplasiaof the pancreas:a secondaryphenomenon.Hum Pathol. 17:4654.

7. Gould VE, Memoli VA, Dardi LE, Gould NS. 1983 Nesidiodysplasia andnesidioblastosis of infancy: structural and functional correlations with thesyndrome of hyperinsulinemic hypoglycemia. Pediatr Pathol. 1:731.

8. Heitz PU , Kloppel G, Hacki WH, Polak JM, Pearse AGE. 1977 Nesidioblas-tosis: the pathologic basis of persistent hyperinsulinemic hypoglycemia ininfants. Morphologic andquantitativeanalysis of seven cases based on specificimmunostaining and electron microscopy. Diabetes. 26:632642.

9. Heitz PU, Kloppe l G, Polak JM. 1979 Morphology of the endocrine pancreasin persistent hypoglycaemia in infants. In: Andreani D, Lefevre PJ, Marks V,eds.Currentviewson hypoglycaemia andglucagon. Proceedingsof theSeronoSymposia. London: Academic Press; 355365.

10. Jaffe R, Hashida Y, Yunis EJ. 1980 Pancreatic pathology in hyperinsulinemichypoglycemia of infancy. Lab Invest. 42:356365.

11. Kohnert KD, Falt K, Odselius R, et al. 1988 Production of pro-insulin, C-peptide, and insulin in nesidioblastosis, focal islet-cell adenomatosis, andgenuine insulomas. A correlated radioimmunochemical, immunohistochem-ical, and ultrastructural investigation with particular regard to the occurrenceof argyrophil and pro-insulin immunoreactive cells. Diabetes Res. 8:151163.

12. Kramer JL, Bell MJ, Deschryver K, Bower RJ, Ternberg JL, White NH. 1981

Clinical and histologic indications for extensive pancreatic resection in nesid-ioblastosis. Am J Surg. 143:116119.13. Rahier J, Falt K, Muntefering H, Becker K, Gepts W, Falkmer S. 1984 The

basic structural lesion of persistent neonatal hypoglycaemia with hyperinsu-linism: deficiency of pancreatic D cells or hyperactivityof B cells? Diabetologia.26:282289.

14. Rahier J. 1989 Relevance of endocrine pancreas nesidioblastosis to hyperin-sulinemic hypoglycemia. Diabetes Care. 12:164166.

15. Sovik O, Vidnes J, Falkmer S. 1975 Persistent neonatal hypoglycaemia. Aclinical and histopathological study of three cases treated with diazoxide andsubtotal pancreatectomy. Acta Pathol Microbiol Scand [A]. 83:155166.

16. Witte DP, Greider MH, Deschryver-Kecskemeti K, Kissane JM, White NH1984 The juvenile human endocrine pancreas: normal vs idiopathic hyperin-sulinemic hypoglycemia. Semin Diagn Pathol. 1:3042.

17. Glaser B, Chiu KC, Anker R, et al. 1994 Familial hyperinsulinism maps tochromosome 11p1415.1, 30 cM centromeric to the insulin gene. Nat Genet7:185188.

18. Thomas PM, Cote GJ, Wohllk N, et al. 1995 Mutations in the sulfonylureareceptorgene in familialpersistenthyperinsulinemichypoglycemia of infancyScience. 268:426429.

19. Kane C, Shepherd RM, Squires PE, et al. 1996 Loss of functional K-ATPchannels in pancreatic B-cells causes persistent hyperinsulinemic hypoglycemia of infancy. Nat Med. 2:13441347.

20. Laidlaw GF. 1938 Nesidioblastoma, the islet tumor of the pancreas. Am Pathol. 14:125134.

21. Daneman D, Ehrlich RM. 1993 The enigma of persistent hyperinsulinemihypoglycemia of infancy. J Pediatr. 123:573575.

22. FillerRM,WeinbergMJ,CutzE, WessonDE,EhrlichRM. 1991Current statuof pancreatectomy for persistent idiopathic neonatalhypoglycemia due to islecell dysplasia. J Pediatr Surg. 26:6075.

23. Gauderer M, Stanley CA, Baker L, Bishop HC. 1978 Pancreatic adenomas ininfants and children: current surgical management. J Pediatr Surg. 13:591596

24. Glaser B, Hirsch HJ, Landau H. 1993 Persistent hyperinsulinemic hypoglycemia of infancy: long-term octreotide treatment without pancreatectomyJ Pediatr. 123:644 650.

25. HamptonRichR, DehnerLP,Okinaga K,Deeb LC,UlstromRA,Leonard AS1978 Surgical management of islet-cell adenoma in infancy. Surgery84:519526.

26. Hirsch HJ, Loo S, Evans N, Crigler JF, Filler RM, Gabbay KH. 1977 Hypo

glycemia of infancy and nesidioblastosis. Studies with somatostatin. N EngJ Med. 296:13231326.27. Horev Z, Ipp M, Levey P, Daneman D. 1991 Familial hyperinsulinism: suc

cessful conservative management. J Pediatr. 119:717720.28. Kitson HF, McCrossin RB, Jimenez M, Middleton A, Silink M. 1980 Soma

tostatin treatment of insulin excess due to B-cell adenoma in a neonate. J Pediatr. 96:145148.

29. Labrune Ph, Bonnefont JP, Nihoul-Fekete C, et al. 1989 Evaluation des methodes diagnostiques et thera peutiques de l hyperinsulinisme du nouveau-ne edu nourrisson. Arch Fr Pediatr. 46:167173.

30. Landau H,PerlmanM, MeyerS, et al. 1982 Persistent neonatal hypoglycemidue to hyperinsulinism: medical aspects. Pediatrics. 70:440446.

31. Poggi-Travert F, Rahier J, Brunelle F, Fekete C, Sa udubray JM. 1994 Hyperinsulinismes de lenfant: a propos dune serie de 56 cas (1984 1994)Journees parisiennes de pediatrie. Paris: Medecine-Sciences Flammarion.

32. Soltesz G, Aynsley-GreenA. 1984Hyperinsulinism in infancy and childhoodIn: Frick HVP, von Harnack GA, Kochsiek K, Martini GA, Prader A, edsAdvances in internal medicine and pediatrics. Berlin, Heidelberg: SpringerVerlag; 151202.

33. Thornton PS, Alter CA, Katz LEL, Baker L, Stanley CA. 1993 Short- andlong-term use of octreotide in the treatment of congenital hyperinsulinismJ Pediatr. 123:637643.

34. de Lonlay P, Fournet JC, Rahier J, et al. 1997 A somatic deletion of theimprinted 11p15 in sporadic persistent hyperinsulinemic hypoglycemia oinfancy is specific of focal adenomatous hyperplasia and endorses partiapancreatectomy. J Clin Invest. 4:802807.

35. Lyonne t S, Bonn efont J P, Saud ubray J M, Nihou l-Fek ete C, Brunel le F. 1989Localization of focal lesion permitting partial pancreatectomy in infants [Letter]. Lancet. 2:671.

36. Brunelle F, Negre V, Barth MO. 1989 Pancreatic venous sampling in infantsand children with primary hyperinsulinism. Pediatr Radiol. 19:100103.

37. Dubois J, Brunelle F, Touati G, et al. 1995 Hyperinsulinism in childrendiagnostic valueof pancreatic venous samplingcorrelated withclinical, pathological and surgical outcome in 25 cases. Pediatr Radiol. 25:512516.

38. Schwartz JF, Zwiren GT. 1971 Islet cell adenomatosis and adenoma in aninfant. J Pediatr. 79:232238.

39. Bowen RE,SwartzFJ. 1976 The ultrastructure of polyploid B-cells in the isletof normal mice. Diabetologia. 12:171180.

40. HellmanB, Hellerstrom C. 1959 Size differences of the B-cell nuclei in the isletissue of normal and alloxan-treated rats. Acta Pathol Microbiol Scand45:113122.

41. Kracht J. 1958 Morphologische Kriterien zur Beurteilung der InselaktivitatEndokrinologie. 36:146158.

42. Wagner D, Richart R. 1968 Polyploidy in the human endometrium with theArias-Stella reaction. Arch Pathol. 85:475480.

43. Rahier J, Sempoux C, Fournet JC, et al. 1998 Partial or near-total pancreatectomy for persistent neonatal hyperinsulinaemic hypoglycaemia: apathologists role. Histopathology. 32:1519.

DIFFERENTIAL DIAGNOSIS OF NESIDIOBLASTOSIS 1461

1455-1.full

Documents

Transcript of 1455-1.full