Combined Retinal Hamartomas Leading to the Diagnosis of Neurofibromatosis Type 2

Ophthalmic Genetics, 29:133–138, 2008Copyright c© Informa Healthcare USA, Inc.ISSN: 1381-6810 (print) / 1744-5094 (online)DOI: 10.1080/13816810802206507

CASE REPORT

Combined Retinal Hamartomas Leading to the Diagnosisof Neurofibromatosis Type 2

Elizabeth A. GrantOregon Health & Science University, School of Medicine, Portland, Oregon, USA

Karmen M. TrzupekOregon Health & Science University, Casey Eye Institute, Department of Ophthalmic Genetics, Portland, Oregon, USA

Jacob Reiss and Kate CrowKaiser Permanente Northwest, Department of Medical Genetics, Portland, Oregon, USA

Ludwine MessiaenUniversity of Alabama at Birmingham, Department of Genetics, Birmingham, Alabama, USA

Richard G. WeleberOregon Health & Science University, Casey Eye Institute, Department of Ophthalmic Genetics, Portland, Oregon, USA

Purpose: To report two cases of neurofibromatosis type 2 (NF2) initially presenting with isolatedbilateral combined hamartomas of the retina and retinal pigment epithelium (RPE). Methods:Retrospective observational case reports. Results: Two unrelated children presented to oph-thalmology with isolated combined hamartomas of the retina and RPE. Patient one presented toophthalmology at the age of 2 years; by 4 years, he developed what was thought to be a plexiformneurofibroma and, with more than 6 cafe au lait spots, was diagnosed with neurofibromatosistype 1 (NF1). By the age of 5, he had developed bilateral vestibular schwannomas, and wasdiagnosed with NF2. Subsequent molecular testing revealed a truncating mutation in exon 13(c.1396C>T; p.R466X) of the NF2 gene. Patient two presented to ophthalmology at the age of7 months; by age 6 she had developed two subcutaneous masses on her forehead, a mass in herleft lower abdomen, and in her gumline. Despite lack of pathological evidence of neurofibromaupon biopsy, molecular testing was initiated at age 6 and revealed a truncating mutation in exon8 (c.734delA) of the NF2 gene in the blood. Conclusions: Bilateral combined hamartomas ofthe retina and retinal pigment epithelium (RPE) in a young child should alert the clinician tothe possibility of neurofibromatosis type 2. The recognition of this rare finding as a presentingfeature of NF2 can lead to earlier diagnosis, which is vital to appropriate surveillance andpossible surgical intervention.

Keywords neurofibromatoses; hamartoma; schwannoma; retina; genotype; phenotype; neurofi-bromatosis 2; neurilemmoma

INTRODUCTIONThe neurofibromatoses represent two major disorders—

neurofibromatosis 1 and 2—with distinct genetic causationsbut somewhat overlapping symptoms.1 Neurofibromatosis 2

Received 15 April 2008; accepted 14 May 2008.This work was supported by a Center Grant from Foundation Fight-

ing Blindness, Owings Mills, MD and an unrestricted grant from Re-search to Prevent Blindness, New York, New York.

Address correspondence to Karmen M. Trzupek, Casey Eye Insti-tute, Oregon Health & Science University, 3375 SW Terwilliger Blvd.,Portland, Oregon 97239, USA. E-mail: [email protected]

(NF2) is a rare (1:50,000) autosomal dominant disease causedby mutations the NF2 tumor suppressor gene on chromo-some 22.2 The disease is 100% penetrant, with approxi-mately 50% of cases arising from de novo mutations withno family history of the disease.3 NF2 is typically diag-nosed in the second or third decade of life, following initialsymptoms of tinnitus, hearing loss, and/or balance dysfunc-tion secondary to vestibular schwannomas. Bilateral vestibularschwannomas occur in >90% of non-founder individuals withNF2.1 Schwannomas of other cranial and peripheral nerves,cranial meningiomas, ependymomas, and astrocytomas alsooccur.

133

134 E. A. GRANT ET AL.

The spine is the second most-common site of tumor for-mation: spinal tumors occur in at least two-thirds of affectedindividuals.4,5 The most common ocular finding in NF2 is a pos-terior subcapsular lenticular opacity which occurs in up to 50%of NF2 cases.3,6 Retinal hamartomas and epiretinal membranesrepresent the most common retinal manifestations of NF22,6–8

arising from loss of heterozygosity of the NF2 gene.2 Lessfrequently reported ocular findings include optic nerve menin-giomas and gliomas, neurotrophic keratopathy, and intraocularneurilemmomas.2,8

In contrast to NF2, Neurofibromatosis type 1 (NF1) is rela-tively common (1:3000) with clinical features including multiple“cafe au lait” pigmented lesions, skin-fold freckles, neurofibro-mas, optic pathway gliomas, and Lisch nodules on the irides.Like NF2, nearly 50% of affected NF1 patients have a positivefamily history for NF2 with the remaining 50% likely repre-senting de novo mutations. Penetrance is 100% with variableexpressivity.9

The diagnosis of NF2 is typically made on clinical findings,with vestibular schwannomas carrying the greatest diagnosticimportance. The occurrence of bilateral vestibular schwanno-mas alone is sufficient, but not required, for the diagnosis ofNF2.10,11 Individuals with NF2 do not usually present with mul-tiple cafe au lait spots or Lisch nodules, though they may havemore cafe au lait spots than the general population.12 Seventypercent of NF2 patients have cutaneous manifestations with 10%having more than 10 skin findings.1 The diagnosis of NF2 mayalso be made based on molecular genetic testing, and may es-pecially be helpful to establish an early diagnosis in founderpatients.12

In NF2, genotype-phenotype correlations have shown thatmissense mutations more frequently lead to a milder phenotype,whereas truncating mutations typically yield a more severe phe-notype with an earlier age of onset.13 Truncating mutations arealso more commonly associated with retinal findings, whichis consistent with the finding in several early studies that reti-nal manifestations in NF2 correlate with more severe disease.14

Splice-site mutations have somewhat more variability in pheno-type. Severe disease has also been more frequently associatedwith mutations located near the 5′ end of the gene: mutationsin exons 1–5 more frequently correspond to severe disease thanthose in exons 11–15.10,15 Somatic mosaicism is common inNF2, reportedly responsible for 25% of founder mutations, andcan lead to an attenuated or unilateral phenotype.16

Clinical data are useful in disease classification: age of on-set of symptoms has been validated in numerous studies as themost important predictor of subsequent disease progression dueto its correlation with tumor growth rates.17 The average lifeexpectancy in NF2 is 36 years, but this figure varies.12

There are genotype-phenotype correlations between the num-ber of NF2-associated intracranial meningiomas, spinal tumors,and peripheral nerve tumors and the type of NF2 mutation.18

No association has yet been found between the growth rate ofvestibular schwannomas or the occurrence of retinal abnormal-

ities with severity of NF2.10 Baser reviewed 18 patients withmutation-proven NF2 and retinal findings (retinal hamartomas,epiretinal membranes, or a combination of the two). The au-thors compared the type of mutation with retinal disease andfound that 60% of patients with retinal hamartomas had severedisease, which is similar to the incidence in the group withoutretinal hamartomas.19

We report two cases of NF2 in children who initially pre-sented with isolated bilateral combined retinal hamartomas.

CASE REPORT 1This patient was referred to the Ophthalmic Genetics Clinic

at 2 years of age with a history of right esotropia, pendularnystagmus, and a hypopigmented macula with surrounding hy-perpigmentation in the left eye. His past medical history wasotherwise benign. Upon fundus exam, the patient was found tohave bilateral combined hamartomas of the retina and RPE inboth eyes, left greater than right (Figure 1A and B). The rightmacula exhibited some tenting and wrinkling of the internal lim-iting membrane, while the left lesion was more extensive withpigmentary mottling and hyperplasia as well as abnormal vesselsdragged temporally from the left disc. A full-field electroretino-gram demonstrated mildly subnormal responses in the right eyeand moderately subnormal responses in the left eye with pro-longed cone and rod times. No Lisch nodules or lens opacitieswere noted.

Two years later, at age 41/2, the patient developed a mass onhis right eyelid near the accessory lacrimal gland, the biopsyof which was initially identified as a plexiform neurofibroma.Upon exam at age 5, several cafe au lait spots were noted inhis groin, flanks, knee, legs, chest, and back. The presence of>6 cafe-au-lait spots with a diameter of 5mm and a plexiformneurofibroma led to a clinical diagnosis of NF1. However, hesubsequently developed bilateral vestibular schwannomas, andwas diagnosed with NF2 based on clinical findings.

Despite aggressive resection, the vestibular schwannomas re-curred, leaving him profoundly deaf. He later developed tumorsof the trigeminal nerve bilaterally, the right giant cavernous si-nus, the right orbit, the larynx, the nasal septum, and myriadsof cutaneous tumors on his extremities—all of which were ini-tially diagnosed as neurofibromas on biopsy. He also developedhundreds of spinal cord tumors, identified as schwannomas. Inaddition to the hamartomas of the retina and RPE, his ocularmanifestations included a right sphenoid meningioma and anorbital meningioma, which were resected multiple times; theright eye became blind and painful and was eventually eviscer-ated. Most recently, at age 15, nine masses were removed fromhis right hand and wrist.

This patient has no known family history of NF2. At age16, molecular testing of the NF2 gene disclosed a previouslyreported nonsense mutation in exon 13 (c.1396C>T; p.R466X).Repeat histopathological studies were performed on availabletissue from two tumors previously thought to be plexiform

RETINAL HAMARTOMAS IN NF2 135

FIG. 1. Fundus photography of patient 1, showing bilateral combined hamartomas of the retina and RPE: (A) Right eye. (B) Lefteye.

neurofibromas: a cutaneous tumor of the finger, and theoriginal eyelid tumor. These studies—performed using newerimmunohistochemical stains including Collagen IV, S-100,and Neurofilament—enabled the pathologists to amend thediagnosis in both cases from a plexiform neurofibroma to aschwannoma.

CASE REPORT 2A seven-month-old female child was referred for ophthal-

mologic evaluation. She was born at term and required briefoxygen administration at delivery secondary to a nuchal cord.A heart murmur heard at two months of age was followed byan echocardiogram that showed a structurally normal heart withphysiologic tricuspid regurgitation. At three months of age, herparents noted poor visual tracking and horizontal nystagmus.Serology was negative for toxoplasmosis, syphilis, CMV, her-pes, and rubella.

At seven months of age, fundus examination revealed ab-normal temporal dragging of vessels bilaterally and elevatedmacular lesions left greater than right. Ultrasonography showedmacular lesions consistent with combined hamartomas of RPEand retina. Further physical exam revealed a small hemangiomabirthmark on her lower back but no other abnormal skin or softtissue findings. A full-field electroretinogram showed slightlyabnormal cone/rod responses, decreased more for the left eye,consistent with her macular lesions and not supportive of anunderlying primary retinal dystrophy. No Lisch nodules or lensopacities were noted.

An MRI at age 3 years revealed stable hamartomas; no masseswere reported. Upon examination at age 4, the patient presentedwith two masses: one on the right forehead and one on the leftlower abdomen, which were not suggestive of neurofibromatosis

upon pathologic examination. Later that year the patient had agrowth removed from her gumline diagnosed as an irritationfibroma and not a neurofibroma.

At age 6, the patient was noted to have two subcutaneousnodules on her forehead, frontal and bitemporal bossing, andfreckles on her face. The fundus examination disclosed bilateralcombined retinal hamartomas (Figure 2A and B). Macrocephaly(HC: >98th %) was noted, arousing slight suspicion of other eti-ologic possibilities including basal cell nevus syndrome (Gorlinsyndrome). By report, the patient’s father is also macrocephalic.

The patient did not have a known family history for NF2.Molecular testing was performed based on suspicion from theretinal findings: the results disclosed a single base pair deletionof the NF2 gene creating a frameshift mutation and resulting ina premature stop codon in exon 8 (c.734delA).

At age 8, the patient had an MRI that showed numerous smallT2 intense and enhancing lesions in the white matter of the brain,cranial nerves, posterior globes (left more than right, likely rep-resenting retinal hamartomas), and throughout the spinal cord.She also had multiple subcutaneous nodules seen and severalareas with low signals possibly representing vascular abnormal-ities, calcifications, or chronic hemorrhage.

DISCUSSIONThe differential diagnosis of a retinal lesion resembling a

combined hamartoma includes epiretinal membrane, choroidalmelanoma, choroidal nevus, congenital hypertrophy of the reti-nal pigment epithelium, adenoma and adenocarcinoma of theretinal pigment epithelium, and morning glory disc anomaly.21

Combined hamartomas of the retina and RPE are congenital,are relatively stable or evolve slowly with age, and are usu-ally unilateral. Once bilateral retinal hamartomas are confirmed,

136 E. A. GRANT ET AL.

FIG. 2. Fundus photography of patient 2, showing bilateral combined hamartomas of the retina and RPE: (A) Right eye; (B) Lefteye.

phakomatous etiologies (NF1, NF2, tuberous sclerosis, andGorlin syndrome) should be considered. Retinal hamartomasare more common in NF2 than NF1, but because NF1 is a moreprevalent disease, both should be considered in the differential.22

Combined retinal hamartomas have on rare occasion been re-ported in basal cell nevus syndrome (Gorlin syndrome)23 andtuberous sclerosis. According to a study in 2001, 44% of pa-tients with tuberous sclerosis complex (TSC) demonstrate reti-nal hamartomas, which were bilateral in 34% of patients. Retinalhamartomas in TSC can be flat, smooth, non-calcified, or resem-ble the classic “mulberry,” multinodular growth.24 Most, if notall, of the cases of retinal hamartomas associated with tuberoussclerosis have been found to harbor mutations in TSC2.25

Retinoblastoma must always be ruled out. Other rareinstances of retinal hamartomas have been observed inbranchio-oculo-facial syndrome26 and juvenile nasopharyngealangiofibroma27 and described in case reports. Retinal vascular(capillary) hamartomas have been associated with von Hippel-Lindau disease in at least five cases, but the lesions are distinctin appearance and behavior from combined hamartomas of theretina and RPE such as those seen in NF2.28

Case 1 illustrates the severity of disease that NF2 can repre-sent and several difficulties with the early diagnosis of NF2. Thepresence of a plexiform neurofibroma is highly supportive of thediagnosis of NF1. Indeed, this finding, in conjunction with 6 ormore cafe-au-lait spots, is sufficient to establish the diagnosis ofNF1.9 Incisional biopsy is often performed to verify the clinicalimpression of a plexiform neurofibroma. In this case, differenti-ating a plexiform neurofibroma from a plexiform schwanommaproved to be extremely difficult. Several reasons for possible am-biguity in tumor diagnosis exist, primarily relating to inadequatetissue samples and sections, and the overlapping pathological

features of neurofibromas and schwannomas. Only a very small,accessible portion of the tumor is typically biopsied; this lim-ited sampling restricts what is available to study. Even moreimportantly, plexiform neurofibromas and plexiform schwan-nomas share certain cell types and histopathological features.Plexiform schwannomas are characterized by fascicles and nestsof Schwann cells, surrounded by capsules or perineurial cells,which can resemble the histopathological features seen in plex-iform neurofibroma. Prominent nuclear palisading and, in par-ticular, Verocay body formation help to distinguish plexiformschwannomas.

However, foci resembling plexiform neurofibromas canoccur within a lesion otherwise designated as plexiformschwannoma,29 which raises the possibility that a “plexiformschwannoma” might be misconstrued as a plexiform neurofi-broma. Given a tumor sample with ambiguous pathology, thediagnosis is likely to be influenced by the clinical diagnosis ofthe patient: a sample taken from a patient with a clinical di-agnosis of NF1 may be more frequently labeled a plexiformneurofibroma, while that removed from a patient with a clinicaldiagnosis of NF2 may be identified as a schwannoma. The clin-ician must be aware of this bias and question biopsy results ifthey do not fit with the patient’s other clinical findings.

Several immunohistochemical stains are available to help dif-ferentiate neurofibromas from schwannomas. At the time of theoriginal eyelid and orbital biopsy for Case 1, however, thesestains were not available. Plexiform neurofibromas have beenreported to have a 10% incidence of malignant transformation,while plexiform schwannomas of the orbit reportedly rarelyundergo malignant transformation. Several cases of malignanttransformation of orbital schwannomas have been reported; caseone in this report may be another.30 If molecular testing had been

RETINAL HAMARTOMAS IN NF2 137

available at the time, NF2 could have been diagnosed follow-ing the patient’s initial presentation to the Ophthalmic GeneticsClinic, and more aggressive screening and intervention may havebeen possible.

In case 2, the suspicion of NF2 was made based on the oph-thalmologic diagnosis of bilateral combined hamartomas of theretina and RPE. Questionable growths on the patient’s foreheadmay have aided in the push toward NF2 genetic testing, butthe patient had several previous biopsies that did not indicateneurofibromatosis. All published case reports describing com-bined retinal hamartomas in NF2 were all already found to havetumor manifestations of the disease.14,31–34 In case 2, the com-bined retinal hamartomas aroused the suspicion of NF2 and thediagnosis was established through comprehensive NF2 genetictesting before any of the clinical diagnostic features becamemanifest. The retinal appearance alone provided sufficient rea-son to screen for NF2.

Current guidelines for screening affected and at-risk membersof NF2 families include screening for clinical findings of NF2from birth. Contrast-enhanced MRI of the entire spine and brainis recommended beginning at age 10–12. MRI is the modalityof choice due to its detailed pictures, minimal invasiveness, andability to track progression of even the smallest tumors.35 How-ever, limitations of MRI screening include reader bias, leading tounreliable tumor diagnosis and measurements, especially withsmall and multiple overlapping tumors and previous surgeries.A recent study looked at factors affecting the validity of MRIin NF2 and found that thin-slice, no-skip, small-pixel-size im-ages were statistically better, and that radiologist experience andaffiliation with a university or NF2 specialty center improved in-terobserver agreement.36

The patient will now be followed with serial MRIs for the de-tection of the likely progression of vestibular schwannomas andextensive tumor burden. Imaging will also include surveillanceof the retinal hamartomas for stability or progression. While thecourse of her disease remains to be seen, her combination ofmolecular and clinical findings, including early retinal findingsand a truncating frameshift mutation, may predict a severe formof NF2.15,18

Recognizing the prevalence of combined retinal hamartomasin NF2, as well as the association of retinal findings with moresevere disease, makes the diagnosis of NF2 based on ocular find-ings significant. In addition, emerging genotype-phenotype cor-relations may enable a molecular diagnosis to contribute valu-able prognostic information to the family regarding age of onsetand degree of disease severity. The lag time from first symptomsof NF2, predominantly hearing loss, to diagnosis was previouslyfound to be around 10 years.22 With new advances in imaging,increased knowledge of presenting signs, and the availability ofcomprehensive NF2 genetic testing, the diagnosis of NF2 couldbe much more timely.

A study in 2002 showed that NF2 patients who were re-ferred for care in specialty treatment centers had a lower risk ofmortality.37 With earlier diagnosis, presymptomatic screening

for intracranial masses can be initiated, and earlier surgical in-tervention made possible.

REFERENCES1. Evans DG, Sainio M, Baser ME. Neurofibromatosis type 2. J Med

Genet. 2000;37:897–904.2. Chan CC, Koch CA, Kaiser-Kupfer MI, et al. Loss of heterozygos-

ity for the NF2 gene in retinal and optic nerve lesions of patientswith neurofibromatosis 2. J Pathol. 2002;198:14–20.

3. Lalwani A. Neurofibromatosis type 2. In: Current Diagnosis& Treatment in Otolaryngology - Head & Neck Surgery. TheMcGraw-Hill Companies, Inc; 2004: Chapter 62.

4. Mautner VF, Tatagiba M, Lindenau M, et al. Spinal tumorsin patients with neurofibromatosis type 2: MR imaging studyof frequency, multiplicity, and variety. Am J Roentgenol.1995;165:951–955.

5. Parry DM, Eldridge R, Kaiser-Kupfer MI, Bouzas EA, Pikus A,Patronas N. Neurofibromatosis 2 (NF2): Clinical characteristicsof 63 affected individuals and clinical evidence for heterogeneity.Am J Med Genet. 1994;52:450–461.

6. Ragge NK, Baser ME, Klein J, et al. Ocular abnormalities inneurofibromatosis 2. Am J Ophthalmol. 1995;120:634–641.

7. Meyers SM, Gutman FA, Kaye LD, Rothner AD. Retinal changesassociated with neurofibromatosis 2. Trans Am Ophthalmol Soc.1995;93:245–252; discussion 252–247.

8. Kaye LD, Rothner AD, Beauchamp GR, Meyers SM, EstesML. Ocular findings associated with neurofibromatosis type II.Ophthalmology. 1992;99:1424–1429.

9. Gutmann DH, Aylsworth A, Carey JC, et al. The diagnostic eval-uation and multidisciplinary management of neurofibromatosis 1and neurofibromatosis 2. JAMA. 1997;278:51–57.

10. Ahronowitz I, Xin W, Kiely R, Sims K, MacCollin M, NunesFP. Mutational spectrum of the NF2 gene: A meta-analysis of 12years of research and diagnostic laboratory findings. Hum Mutat.2007;28:1–12.

11. Evans DG, Trueman L, Wallace A, Collins S, Strachan T. Geno-type/phenotype correlations in type 2 neurofibromatosis (NF2):Evidence for more severe disease associated with truncatingmutations. J Med Genet. 1998;35:450–455.

12. Evans DG. Neurofibromatosis 2. wwwgenetestsorg; 2006.13. Ruggieri M, Iannetti P, Polizzi A et al. Earliest clinical manifes-

tations and natural history of neurofibromatosis type 2 (NF2) inchildhood: A study of 24 patients. Neuropediatrics. 2005;36:21–34.

14. Ragge NK, Baser ME, Riccardi VM, Falk RE. The ocularpresentation of neurofibromatosis 2. Eye. 1997;11 (Pt 1):12–18.

15. Baser ME, Kuramoto L, Woods R, et al. The location of constitu-tional neurofibromatosis 2 (NF2) splice site mutations is associatedwith the severity of NF2. J Med Genet. 2005;42:540–546.

16. Evans DG, Wallace AJ, Wu CL, Trueman L, Ramsden RT, Stra-chan T. Somatic mosaicism: A common cause of classic disease intumor-prone syndromes? Lessons from type 2 neurofibromatosis.Am J Hum Genet. 1998;63:727–736.

17. Ruttledge MH, Andermann AA, Phelan CM, et al. Type ofmutation in the neurofibromatosis type 2 gene (NF2) frequentlydetermines severity of disease. Am J Hum Genet. 1996;59:331–342.

138 E. A. GRANT ET AL.

18. Baser ME, Kuramoto L, Joe H, et al. Genotype-phenotypecorrelations for nervous system tumors in neurofibromatosis 2: Apopulation-based study. Am J Hum Genet. 2004;75:231–239.

19. Baser ME, Kluwe L, Mautner VF. Germ-line NF2 mutations anddisease severity in neurofibromatosis type 2 patients with retinalabnormalities. Am J Hum Genet. 1999;64:1230–1233.

20. Bouzas EA, Parry DM, Eldridge R, Kaiser-Kupfer MI. Familialoccurrence of combined pigment epithelial and retinal hamartomasassociated with neurofibromatosis 2. Retina. 1992;12:103–107.

21. Elliot D, Schachat, AP. Combined Hamartoma of the Retinaand Retinal Pigment Epithelium. In: Retina. St. Louis: Mosby;2001:640–646.

22. Traboulsi E, Ragge, NK. The Phakomatoses. In: Traboulsi E (ed.),Genetic Diseases of the Eye. New York: Oxford University Press;1998:733–775.

23. De Potter P, Stanescu D, Caspers-Velu L, Hofmans A. Photo essay:Combined hamartoma of the retina and retinal pigment epitheliumin Gorlin syndrome. Arch Ophthalmol. 2000;118:1004–1005.

24. Rowley SA, O’Callaghan FJ, Osborne JP. Ophthalmic mani-festations of tuberous sclerosis: a population based study. Br JOphthalmol 2001;85:420–423.

25. Dabora SL, Jozwiak S, Franz DN, et al. Mutational analysis ina cohort of 224 tuberous sclerosis patients indicates increasedseverity of TSC2, compared with TSC1, disease in multipleorgans. Am J Hum Genet. 2001;68:64–80.

26. Demirci H, Shields CL, Shields JA. New ophthalmic manifes-tations of branchio-oculo-facial syndrome. Am J Ophthalmol.2005;139:362–364.

27. Fonseca RA, Dantas MA, Kaga T, Spaide RF. Combined hamar-toma of the retina and retinal pigment epithelium associatedwith juvenile nasopharyngeal angiofibroma. Am J Ophthalmol.2001;132:131–132.

28. Schmidt D, Neumann HP. Retinal vascular hamartoma in vonHippel-Lindau disease. Arch Ophthalmol. 1995;113:1163–1167.

29. Feany MB, Anthony DC, Fletcher CD. Nerve sheath tumours withhybrid features of neurofibroma and schwannoma: A conceptualchallenge. Histopathology. 1998;32:405–410.

30. Jakobiec FA, Font RL, Zimmerman LE. Malignant peripheralnerve sheath tumors of the orbit: A clinicopathologic study ofeight cases. Trans Am Ophthalmol Soc. 1985;83:332–366.

31. Sivalingam A, Augsburger J, Perilongo G, Zimmerman R,Barabas G. Combined hamartoma of the retina and retinal pigmentepithelium in a patient with neurofibromatosis type 2. J PediatrOphthalmol Strabismus. 1991;28:320–322.

32. Tonsgard JH, Oesterle CS. The ophthalmologic presentationof NF-2 in childhood. J Pediatr Ophthalmol Strabismus.1993;30:327–330.

33. Rettele GA, Brodsky MC, Merin LM, Teo C, Glasier CM. Blind-ness, deafness, quadriparesis, and a retinal malformation: the rav-ages of neurofibromatosis 2. Surv Ophthalmol. 1996;41:135–141.

34. Onder F, Cosar CB, Akbatur H, Kural G. Bilateral retinalinvolvement as a presenting sign of neurofibromatosis 2. J PediatrOphthalmol Strabismus 2002;39:49–51.

35. Evans DG, Baser ME, O’Reilly B, et al. Management of the pa-tient and family with neurofibromatosis 2: a consensus conferencestatement. Br J Neurosurg 2005;19:5–12.

36. Slattery WH, Lev MH, Fisher LM, Connell SS, Iqbal Z, Go JL.MRI evaluation of neurofibromatosis 2 patients: a standardized ap-proach for accuracy in interpretation. Otol Neurotol 2005;26:733–740.

37. Baser ME, Friedman JM, Aeschliman D, et al. Predictors ofthe risk of mortality in neurofibromatosis 2. Am J Hum Genet2002;71:715–723.