- Email: [email protected]
Glaucoma and Globe Enlargement Associated with Neurofibromatosis Type 1
Glaucoma and Globe Enlargement Associated with Neurofibromatosis Type 1 Jose Morales, MD,1 Imtiaz A. Chaudhry, MD, PhD,2 Thomas M. Bosley, MD3,4 Objective: To describe the features of glaucoma and globe enlargement sometimes associated with neurofibromatosis type 1 (NF1). Design: Single institution, retrospective, and cross-sectional study. Participants: Eighty medical records of patients treated at King Khaled Eye Specialist Hospital in Saudi Arabia with NF1 were reviewed, and 46 patients were examined. Methods: We reviewed the charts of patients with NF1 and examined available individuals, including gonioscopy, axial length, and ultrasound biomicroscopy in appropriate patients. Main Outcome Measures: Presence and type of glaucoma, anterior chamber angle abnormalities, globe axial length, ultrasound biomicroscopy, and visual outcome. Results: Two patient groups were identified: Group 1 had 56 patients with orbito-facial NF1, and group 2 had 24 patients with NF1 but without orbito-facial involvement. Group 1 included 13 patients with glaucoma (23%), which occurred only ipsilateral to the orbito-facial involvement and generally presented before age 3 years. Glaucoma surgery was required in all of these patients, and visual prognosis was poor. In group 1, mean axial length on the side affected by NF1 was 29.8⫾4.1 mm in patients with glaucoma and 25.6⫾2.0 mm in patients without glaucoma. Patients with glaucoma (P ⬍ 0.001) and without glaucoma (P ⬍ 0.0001) in group 1 had significantly larger globes on the affected side. Group 2 patients had a mean axial length of 23.6⫾1.6 mm for both eyes without significant globe asymmetry. Conclusions: In this Arab population, glaucoma associated with orbito-facial NF1 occurred less often than the 50% rate that is typically cited. Glaucoma presented early in life and only in patients with ipsilateral orbito-facial involvement. Glaucoma in this setting was always associated with globe enlargement. Glaucoma required surgery, and visual prognosis was poor because of glaucoma and concurrent pathology. Globe enlargement was most severe when associated with glaucoma but also present on the side with orbito-facial involvement in patients without glaucoma. The presence of neurofibroma may induce globe enlargement on the affected side in orbito-facial NF1. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Ophthalmology 2009;116:1725–1730 © 2009 by the American Academy of Ophthalmology.
Neurofibromatosis type 1 (NF1) is a dominantly inherited disorder of the neuroectoderm that results in the development of hamartomas, particularly of the skin, eyes, and nervous system. The syndrome primarily affects tissue derived from the neural crest, particularly sensory nerves, Schwann’s cells, and melanocytes.1 Glaucoma associated with NF1 is rare,2 and most ophthalmology literature about glaucoma and NF1 is derived from reports of 12,3 or 24 – 6 patients. The phenomenon of globe enlargement in NF1 without glaucoma has received limited attention,7 but a recent report from King Khaled Eye Specialist Hospital found globe enlargement on computed tomography and magnetic resonance imaging in 13 of 31 patients with orbito-facial NF1.8 This project describes the prevalence and clinical features of glaucoma and globe enlargement in patients with NF1 treated at King Khaled Eye Specialist Hospital in Riyadh, Saudi Arabia, over the last 25 years.
Materials and Methods We reviewed the medical records of all patients fulfilling National Institutes of Health criteria for NF19 who were examined between © 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
1982 and 2007 at King Khaled Eye Specialist Hospital, the major national ophthalmology referral site. We collected information regarding predominant signs and symptoms, age at onset of any symptoms, age at presentation to the hospital, age at the time of glaucoma surgery, presence of orbito-facial involvement and/or lid plexiform neurofibroma (LPNF), presence of ectropion uveae, laterality of NF1 and glaucoma, and initial and final visual acuity. Glaucoma was defined by the presence of consistently elevated intraocular pressure (IOP) (⬎23 mmHg) with evidence of glaucomatous optic nerve damage. The patient population was divided into 2 groups: group 1 with orbito-facial NF1 and group 2 with only cutaneous or neurocutaneous manifestations, without orbito-facial involvement. Twosample t tests were used to determine statistical significance for globe asymmetry. A glaucoma specialist (JM) performed a complete glaucoma examination on all patients with NF1 who were available for evaluation, including detailed gonioscopy, axial length measurements by A-scan, and photographic documentation of the anterior segment. Some group 1 patients with glaucoma also underwent ultrasound biomicroscopy evaluation of anterior chamber angle and ciliary body structures. Some patients had serial bilateral A-scan assessment of globe axial length over a period of years, which was used to investigate the possibility of progressive globe enlargement. An attempt was made to assign the most likely cause ISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2009.06.019
1725
Ophthalmology Volume 116, Number 9, September 2009 of glaucoma for each affected patient from the variety of hypothesized mechanisms,10 including infiltration of the angle by neurofibromatous tumor, ciliary body and choroidal infiltration by tumor causing mechanical displacement of iris diaphragm and secondary angle closure, and congenital angle abnormalities altering the angle and the outflow system. Institutional review board/ethics committee approval was obtained for both the retrospective and prospective components of this study, and examined patients signed informed consent. This study adhered to the Tenets of the Declaration of Helsinki.
Results Between 1982 and 2007, 95 Arab patients were diagnosed at King Khaled Eye Specialist Hospital with NF1, 80 of whom met strict National Institutes of Health NF1 criteria.9 Genetic testing was not performed. Group 1 consisted of 56 patients with orbito-facial NF1, and group 2 included 24 patients without orbito-facial involvement. Dr. Morales examined 46 patients, including 34 patients from group 1 (5 with glaucoma secondary to NF1) and 12 patients from group 2. Lid plexiform neurofibromas were present in 52 patients from group 1. Four patients from group 1 did not have LPNF but were allocated to this group because of orbital involvement by neurofibroma or glioma. Eight patients in group 1 had ectropion uvea with glaucoma (Fig 1), whereas 4 patients had variable degrees of ectropion uvea without glaucoma (Fig 2). Bilateral A-scan measurements were obtained in 31 patients in group 1 and 12 patients in group 2. Mean axial globe length was 29.8⫾4.1 mm (range, 26 –30 mm) on the affected side of group 1 patients with glaucoma and 25.6⫾2.0 mm in those without glaucoma. Relative globe enlargement was most prominent in group 1 patients with glaucoma, with axial length differences between affected and unaffected eyes ranging from 3 to 6 mm (P ⬍ 0.001). However, patients without glaucoma also had consistent globe enlargement on the affected side. The mean difference between the 2 globes of patients in group 1 without glaucoma was 2 mm (P ⬍ 0.0001), although 8 patients had differences ranging from 3 to 5 mm. Mean axial length of both globes in group 2 was 23.8⫾1.7 mm with no significant difference between eyes.
Figure 1. Slit-lamp photograph of patient 7 in Table 1 with unilateral orbito-facial NF1 and ipsilateral ectropion uvea and glaucoma showing 3 quadrant ectropion uvea and Lisch nodules. NF1 ⫽ Neurofibromatosis type 1.
1726
Figure 2. Slit-lamp photograph of patient from group 2 with unilateral orbito-facial NF1 and ipsilateral ectropion uvea without glaucoma showing temporal ectropion uvea and a distorted pupil dragged superonasally. NF1 ⫽ Neurofibromatosis type 1.
Clinical characteristics of the 13 patients in group 1 with glaucoma are detailed in Table 1. Age at the time of glaucoma diagnosis varied from birth (in 5 patients) to 13 years. Eleven patients were diagnosed by age 3 years, whereas only 2 patients were diagnosed somewhat later, at ages 6 and 13 years. Twelve patients with glaucoma in group 1 had unilateral glaucoma and ipsilateral LPNF. Patient 8 is the first reported patient with NF1 with both bilateral LPNF and bilateral glaucoma. He also had the largest globe dimensions, measuring 33.30 mm right eye (OD) and 36.50 mm left eye (OS), and developed bilateral retinal detachments by age 11 years. Patient 7 had an axial length of 28.15 mm in the affected eye and developed a retinal detachment at age 25 years. Table 2 details at least 2 sequential axial length measurements in 5 patients with glaucoma from group 1. Progressive globe enlargement was documented in patients 5 and 8. Globe enlargement occurred in both eyes of patient 5, but size increase was more prominent in the affected eye than in the unaffected eye (5.7 mm vs. 2.7 mm). Patient 8 had bilateral globe enlargement measuring 3.1 mm OD and 3.3 mm OS over 5 years. Gonioscopy on all patients with glaucoma from group 1 revealed 1 or more abnormalities in the affected eye(s) of each patient, including irregular patches or marked angle pigmentation, anterior insertion of the iris, and extensive anterior synechiae (Fig 3). Ultrasound biomicroscopy performed on 4 patients with glaucoma in group 1 showed thickening of the ciliary body in the involved eye (Fig 4). Eleven of 34 patients without glaucoma examined in group 1 had few angle abnormalities, usually ipsilateral to the orbito-facial involvement, consisting of pigmentary patches or marked angle pigmentation and localized areas of anterior synechiae (Fig 5). All 4 patients with ectropion uvea but without glaucoma in group 1 had variable degrees of peripheral anterior synechiae. At last follow-up, 3 of 13 patients with glaucoma in group 1 had undergone evisceration and 2 patients had enucleation of the affected eye. Three additional eyes were blind, both eyes of patient 8 with bilateral glaucoma and bilateral inoperable retinal detachments and 1 eye of patient 7 with unilateral glaucoma, who also developed an inoperable retinal detachment. Two patients had hand motion vision in the affected eye, and 1 patient had 4/400 vision. Two patients were too young to assess visual acuity reliably.
Morales et al 䡠 Glaucoma in Neurofibromatosis Type 1 Table 1. Group 1 Patients with Glaucoma Patient Sex
Age at Diagnosis
LPNF
Affected Glaucoma Side* Side
1 2
M M
Birth 5 mos
⫹ ⫹
Left Left
Left Left
3
F
1 mo
⫹
Left
Left
4
F
Birth
⫹
Left
Left
5
M
6 yrs
⫹
Left
Left
6
F
“Childhood”
⫹
Left
Left
7
F
13 yrs
⫹
Right
Right
8
M
Birth
⫹
Both
Both
9
F
2 yrs
⫹
Left
Left
10
F
3 yrs
⫹
Right
Right
11
F
Birth
⫹
Right
Right
12
M
3 yrs
⫹
Left
Left
13
M
Birth
⫹
Right
Right
Last Visual Acuity NA OD 20/20 OS Prosthesis OD 20/20 OS Prosthesis OD 20/40 OS Prosthesis OD 20/25 OS 20/400 OD HM OS HM OD NLP OS 20/20 OD NLP OS NLP OD 20/30 OS 4/400 OD Prosthesis OS 8/30 (Allen cards) CSM OD 20/20 OS HM OD: LP OS: 20/20
Ectropion Axial Length Uvea (mm)
UBM
Comment
No No
— —
No
—
Yes
—
Yes
OD: 24.45 OS: 30.50 —
Thick CB OS
Thick CB OD Severe RD OD
Yes
OD: 28.15 OS: 22.57 OD: 33.30 OS: 36.50 —
No
—
No
OD: 25.83 OS: 21.88 OD: 22.23 OS: 28.26
Yes Yes Yes
Yes
Not done
Bilateral severe RD
Thick CB OD Thick CB OS
Yes
CB ⫽ ciliary body; CSM ⫽ central, steady, maintains fixation; HM ⫽ hand motion; LP ⫽ light perception; LPNF ⫽ lid plexiform neurofibroma; NA ⫽ not able to assess because of early age; NLP ⫽ no light perception; OD ⫽ right eye; OS ⫽ left eye; OU ⫽ both eyes; RD ⫽ retinal detachment; UBM ⫽ ultrasound biomicroscopy results. *Glaucoma was always present only on the affected side(s).
Discussion Neurofibromatosis type 1 is one of the most common autosomal dominant genetic disorders with a prevalence estimated as 1 in 3000 people worldwide. Expression is highly variable even among members of the same family. Some individuals are not even diagnosed because they carry an asymptomatic, very mild form of the disease, whereas others may have a very severe form causing multiple compliTable 2. Sequential Axial Length Measurements in Selected Group 1 Patients with Glaucoma
Patient
Eye
Axial Length Scan 1
5 (yrs 1, 18)
OD OS OD OS OD OS OD OS OD OS
21.50 24.80 28.42 23.00 30.20 33.20 25.71 21.92 21.57 28.39
7 (yrs 19, 22, 25) 8 (yrs 3, 4, 8) 11 (yrs 3, 3.5, 4) 12 (yrs 8, 15, 16)
Axial Length Scan 2
Axial Length Scan 3
24.25 30.50 28.15 22.57 30.31 34.17 25.83 21.88 22.23 28.26
27.80 23.07 33.30 36.50 26.48 22.29 22.40 27.74
Progression (>3 mm) ⫹ ⫹ ⫹
OD ⫽ right eye; OS ⫽ left eye. Eye affected by orbito-facial involvement highlighted in bold letters.
cations.11 Glaucoma associated with NF1 has been considered rare, particularly among young children. For example, a well-established glaucoma group in Boston reported only 1 case among 300 patients with all types of childhood glaucoma examined over a period of 6 years.2 We describe 13 patients with glaucoma due to orbitofacial NF1 who were examined over a period of 25 years at a tertiary care eye hospital in the Middle East, constituting 16.2% of the 80 patients with a diagnosis of NF1 treated during that period. As far as we know, this is the largest reported series of patients with glaucoma associated with NF1. Lid plexiform neurofibromas are said to occur in approximately 10% of individuals with NF1,11 and yet 52 patients (65%) in our NF1 population had LPNF. This proportion may reflect the fact that ptosis and facial deformity caused by LPNF were among the main reasons for referral. Glaucoma was present in 23% patients in group 1 with orbito-facial NF1, all of whom had LPNF ipsilateral to the glaucomatous eye(s). This proportion was lower than the 50% rate that is typically quoted in glaucoma textbooks10,12 and articles,2,13 even though a substantial number of patients in this series had advanced disease. Glaucoma was diagnosed at a young age in most patients. Infiltration of the anterior chamber angle by neurofibroma is by far the most commonly cited mechanism for glaucoma associated with NF1,2,3,7,10 but other mechanisms have been considered, including malformation or immature development of the anterior chamber angle,2,18 and second-
1727
Ophthalmology Volume 116, Number 9, September 2009
Figure 3. Gonio-photographs of patients 5 (A, B, D) and 12 (C) with NF1 and glaucoma showing (A, B) excessive angle pigmentation obscuring normal structures and landmarks in the affected eye, (C) irregular patches of pigment scattered around the angle, and (D) combination of excessive angle pigmentation and extensive peripheral anterior synechiae. NF1 ⫽ Neurofibromatosis type 1.
ary angle closure due to ciliary body/choroidal infiltration.2 Assigning a single mechanism as the cause of glaucoma in group 1 was not possible. Most patients had a combination of potential mechanisms, including congenital angle abnor-
malities, pigmentary disturbances of the angle, and secondary angle closure by anterior synechiae. Glaucoma present at birth in several patients implied a possible congenital abnormality of the angle, whereas presentation after infancy
Figure 5. Gonio-photographs of patients from group 2 with NF1 without glaucoma showing (A) pigmentary patches and excessive pigmentation of angle plus peripheral anterior synechiae (PAS); (B) increased pigmentation obscuring angle landmarks and extensive PAS; (C) sectoral PAS and prominent long iris processes; and (D) extensive, broad-based PAS. NF1 ⫽ Neurofibromatosis type 1.
1728
Morales et al 䡠 Glaucoma in Neurofibromatosis Type 1
Figure 4. Ultrasound biomicroscopy of patients 5 (A) and 12 (B) with unilateral orbito-facial involvement and ipsilateral glaucoma showing (A) thickening of ciliary body and (B) thickening and posterior displacement of ciliary body from the posterior iris surface.
suggests a progressive element such as infiltration of the angle by neurofibromatous tissue or secondary angle closure due to progressive ciliary body and choroidal infiltration. Ectropion uvea occurred only in patients with orbitofacial NF1 and only on the same side. It was present in more than half of the patients with glaucoma in group 1. Although congenital ectropion uvea is considered a static condition, the association with glaucoma is considered the rule, and it has been advocated that any infant or child with this condition should be carefully followed for the development of glaucoma.6 The association of NF1 and ectropion uvea has been reported.6,14 Among patients without glaucoma in group 1, 4 patients had ectropion uvea and partial angle closure due to peripheral anterior synechiae, suggesting that ectropion uvea could be a factor contributing to development of glaucoma even if it may not be a sufficient cause if limited in extent. All examined patients with orbito-facial NF1 and glaucoma had a larger globe on the affected side. Globe enlargement was severe, with axial lengths ranging from 26 to 36 mm and with a difference between unaffected and affected eyes from 4 to 6 mm. This study is the first time that progressive globe enlargement in NF1 has been quantified, although it was suspected previously because of enlarging corneal diameters7 and progressive exophthalmos or buphthalmos. Two patients (patients 5 and 8) with progressive ocular growth had ultrasound measurements when young (1 and 3 years) and then at a significantly older age (18 years and 8 years, respectively). It is possible that ocular enlargement in orbito-facial NF1 occurs at a specific age and that patients who did not have significant globe growth between measurements (e.g., patients 7 and 12) were either too old or too young to exhibit changes. Severe, progressive globe enlargement was probably an important factor in retinal detachments that occurred in 2 patients. Hoyt and Billson7 reported a patient with progressive globe enlargement after a glaucoma procedure despite controlled IOP, drawing attention to the fact that globe enlargement can occur in NF1 independently of intraocular hyper-
tension. They suggested that globe enlargement could be caused by a local growth factor released by neurofibromatous tissue. Other authors have also concluded that ocular hypertension may not be the only factor contributing to megalophthalmos,15,16 and our observation of consistent globe asymmetry between affected and unaffected eyes in patients without glaucoma in group 1 supports this hypothesis. Asymmetric globe growth might also occur because intermittent or partial lid occlusion deprives the developing globe of a clear retinal image, disrupts the normal emmetropization process, and promotes the development of axial myopia.17 We believe that this is the first report of bilateral orbitofacial involvement and bilateral glaucoma in NF1. Patient 8 had severe bilateral LPNF involvement with extreme bilateral globe enlargement and bilateral glaucoma that required multiple surgeries for IOP control. By age 11 years, this patient had developed bilateral inoperable retinal detachments, presumably precipitated by extreme globe growth. Gonioscopic abnormalities were observed in all patients with glaucoma, consisting of irregular pigment patches and markedly increased pigmentation obscuring the structures of the angle, anterior iris insertion, and large bands of peripheral anterior synechiae. It was impossible to tell whether abnormally pigmented tissue was neoplastic in nature or simply a pigmentary disturbance in patients who develop pigmentary abnormalities elsewhere. All patients with glaucoma had undergone surgery before this examination, and synechial adhesions could have been a result of prior intervention or the neurofibromatous process. Gonioscopic abnormalities were also present in several patients with orbito-facial involvement without glaucoma, including pigmentary changes, which would support the hypothesis that systemic pigmentary abnormalities affected the anterior chamber angle. A recent study reported only subtle gonioscopic changes in 42 patients with NF1, consisting mostly of anterior insertion of the iris and frequent long iris processes.18 The population reported may have had more dramatic and diverse changes in the angle because a larger
1729
Ophthalmology Volume 116, Number 9, September 2009 proportion of patients had orbito-facial involvement. Ultrasound biomicroscopy performed in 4 patients with glaucoma attributed to NF1 had thickening of the ciliary body that could cause mechanical distortion and closure of the angle. Glaucoma at birth associated with NF1 can be initially misdiagnosed as unilateral congenital glaucoma, especially in patients in whom LPNF and café au lait spots are not evident at an early age.3,15 Patient 5, for example, did not have striking LPNF early in life, and the diagnosis of NF1 with glaucoma was delayed until Lisch nodules were noticed at age 15 years. In patients with unilateral infantile glaucoma, it is advisable to search for evidence of NF1 and other causes of unilateral glaucoma. Examination of family members for signs of NF1 and repeated observation for the development of LPNF, café au lait spots, and Lisch nodules can help with the diagnosis. Unilateral globe enlargement despite IOP control can also indicate NF1.7,15 The combination of NF1 and glaucoma seems to carry a grim visual prognosis. One third of our patients with NF1 glaucoma had either enucleation or evisceration of the affected globe. Two patients were too young to ascertain a reliable visual acuity level, but the rest were either blind or legally blind in the affected eye at the last examination. Severe visual loss or blindness was the result of a variety of factors, including retinal detachment, optic nerve glioma, severe amblyopia caused by anisometropia, and glaucomatous optic nerve damage. In conclusion, neurofibromatosis is a variable and complex disease that can affect several components of the globe. Globe enlargement on the affected side seems common in orbito-facial NF1 and is aggravated by glaucoma. Particular attention to globe asymmetry by either refraction or ultrasound measurement is advisable in patients with unilateral orbito-facial involvement in NF1. Further studies are warranted to investigate mechanisms of globe enlargement and possible modes of treatment.
References 1. Ebert EM, Albert DM. The phakomatoses. In: Albert DM, Jakobiec FA, Azar DT, Gragoudas ES, eds. Principles and Practice of Ophthalmology. 2nd ed. vol. 6. Philadelphia, PA: Saunders; 2000:5120 – 44.
2. Grant WM, Walton DS. Distinctive gonioscopic findings in glaucoma due to neurofibromatosis. Arch Ophthalmol 1968; 79:127–34. 3. Brownstein S, Little JM. Ocular neurofibromatosis. Ophthalmology 1983;90:1595–9. 4. Francois J, Katz C. Homolateral association of hydrophthalmia, plexiform neuroma of the upper lid and facial hemihypertrophy in Recklinghausen’s disease [in French]. Ophthalmologica 1961;142:549 –71. 5. Snell S, Treacher Collins E. Plexiform neuroma (elephantiasis neuromatosis) of temporal region, orbit, eyelid, and eyeball. Trans Ophthalmol Soc U K 1903;23:157–77. 6. Ritch R, Forbes M, Hetherington J Jr, et al. Congenital ectropion uveae with glaucoma. Ophthalmology 1984;91:326 –31. 7. Hoyt CS, Billson FA. Buphthalmos in neurofibromatosis: is it an expression of regional giantism? J Pediatr Ophthalmol 1977;14:228 –34. 8. Jacquemin C, Bosley TM, Svedberg H. Orbit deformities in craniofacial neurofibromatosis type 1. AJNR Am J Neuroradiol 2003;24:1678 – 82. 9. National Institutes of Health Consensus Development Conference. Neurofibromatosis: conference statement. Arch Neurology 1988;45:575– 8. 10. Weiss JS, Ritch R. Glaucoma in the phacomatoses. In: Ritch R, Shields MB, Krupin T, eds. The Glaucomas. vol. 2. St. Louis, MO: Mosby; 1996:899 –924. 11. Sippel KC. Ocular findings in neurofibromatosis type 1. Int Ophthalmol Clin 2001;41:25– 40. 12. Mattox C. Glaucoma in the phakomatoses. In: Epstein DL, Allingham RR, Schuman JS, eds. Chandler and Grant’s Glaucoma. 4th ed. Baltimore, MD: Williams & Wilkins; 1997:433–44. 13. Lieb WA, Wirth WA, Geeraets WJ. Hydrophthalmos and neurofibromatosis (von Recklinghausen). Confin Neurol 1959; 19:230 – 47. 14. Wilson ME. Congenital iris ectropion and a new classification for anterior segment dysgenesis. J Pediatr Ophthalmol Strabismus 1990;27:48 –55. 15. Burke JP, Leitch RJ, Talbot JF, Parsons MA. Choroidal neurofibromatosis with congenital iris ectropion and buphthalmos: relationship and significance. J Pediatr Ophthalmol Strabismus 1991;28:265–7. 16. Bardelli AM, Hadjistilianou T. Buphthalmos and progressive elephantiasis in neurofibromatosis: a report of three cases. Ophthalmic Paediatr Genet 1989;10:279 – 86. 17. Qiao-Grider Y, Hung LF, Kee CS, et al. Recovery from form-deprivation myopia in rhesus monkeys. Invest Ophthalmol Vis Sci 2004;45:3361–72. 18. Quaranta L, Semeraro F, Turano R, Gandolfo E. Gonioscopic findings in patients with type 1 neurofibromatosis (Von Recklinghausen disease). J Glaucoma 2004;13:90 –5.
Footnotes and Financial Disclosures Originally received: December 10, 2008. Final revision: May 20, 2009. Accepted: June 10, 2009.
4
Manuscript no. 2008-1470.
1
Glaucoma Division, King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia. 2
Oculoplastic and Orbit Division, King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia. 3
Department of Ophthalmology, King Saud University, Riyadh, Saudi Arabia.
1730
Neurology Division, Cooper University Hospital, Camden, New Jersey. Presented in part as a poster at: American Academy of Ophthalmology Annual Meeting, November 2007, New Orleans, Louisiana. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: Jose Morales, MD, Glaucoma Division, King Khaled Eye Specialist Hospital, PO Box 7191, Riyadh 11462, Saudi Arabia. E-mail: jmorales@kkesh. med.sa.
Neurofibromatosis type 1 (NF1) is a dominantly inherited disorder of the neuroectoderm that results in the development of hamartomas, particularly of the skin, eyes, and nervous system. The syndrome primarily affects tissue derived from the neural crest, particularly sensory nerves, Schwann’s cells, and melanocytes.1 Glaucoma associated with NF1 is rare,2 and most ophthalmology literature about glaucoma and NF1 is derived from reports of 12,3 or 24 – 6 patients. The phenomenon of globe enlargement in NF1 without glaucoma has received limited attention,7 but a recent report from King Khaled Eye Specialist Hospital found globe enlargement on computed tomography and magnetic resonance imaging in 13 of 31 patients with orbito-facial NF1.8 This project describes the prevalence and clinical features of glaucoma and globe enlargement in patients with NF1 treated at King Khaled Eye Specialist Hospital in Riyadh, Saudi Arabia, over the last 25 years.
Materials and Methods We reviewed the medical records of all patients fulfilling National Institutes of Health criteria for NF19 who were examined between © 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
1982 and 2007 at King Khaled Eye Specialist Hospital, the major national ophthalmology referral site. We collected information regarding predominant signs and symptoms, age at onset of any symptoms, age at presentation to the hospital, age at the time of glaucoma surgery, presence of orbito-facial involvement and/or lid plexiform neurofibroma (LPNF), presence of ectropion uveae, laterality of NF1 and glaucoma, and initial and final visual acuity. Glaucoma was defined by the presence of consistently elevated intraocular pressure (IOP) (⬎23 mmHg) with evidence of glaucomatous optic nerve damage. The patient population was divided into 2 groups: group 1 with orbito-facial NF1 and group 2 with only cutaneous or neurocutaneous manifestations, without orbito-facial involvement. Twosample t tests were used to determine statistical significance for globe asymmetry. A glaucoma specialist (JM) performed a complete glaucoma examination on all patients with NF1 who were available for evaluation, including detailed gonioscopy, axial length measurements by A-scan, and photographic documentation of the anterior segment. Some group 1 patients with glaucoma also underwent ultrasound biomicroscopy evaluation of anterior chamber angle and ciliary body structures. Some patients had serial bilateral A-scan assessment of globe axial length over a period of years, which was used to investigate the possibility of progressive globe enlargement. An attempt was made to assign the most likely cause ISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2009.06.019
1725
Ophthalmology Volume 116, Number 9, September 2009 of glaucoma for each affected patient from the variety of hypothesized mechanisms,10 including infiltration of the angle by neurofibromatous tumor, ciliary body and choroidal infiltration by tumor causing mechanical displacement of iris diaphragm and secondary angle closure, and congenital angle abnormalities altering the angle and the outflow system. Institutional review board/ethics committee approval was obtained for both the retrospective and prospective components of this study, and examined patients signed informed consent. This study adhered to the Tenets of the Declaration of Helsinki.
Results Between 1982 and 2007, 95 Arab patients were diagnosed at King Khaled Eye Specialist Hospital with NF1, 80 of whom met strict National Institutes of Health NF1 criteria.9 Genetic testing was not performed. Group 1 consisted of 56 patients with orbito-facial NF1, and group 2 included 24 patients without orbito-facial involvement. Dr. Morales examined 46 patients, including 34 patients from group 1 (5 with glaucoma secondary to NF1) and 12 patients from group 2. Lid plexiform neurofibromas were present in 52 patients from group 1. Four patients from group 1 did not have LPNF but were allocated to this group because of orbital involvement by neurofibroma or glioma. Eight patients in group 1 had ectropion uvea with glaucoma (Fig 1), whereas 4 patients had variable degrees of ectropion uvea without glaucoma (Fig 2). Bilateral A-scan measurements were obtained in 31 patients in group 1 and 12 patients in group 2. Mean axial globe length was 29.8⫾4.1 mm (range, 26 –30 mm) on the affected side of group 1 patients with glaucoma and 25.6⫾2.0 mm in those without glaucoma. Relative globe enlargement was most prominent in group 1 patients with glaucoma, with axial length differences between affected and unaffected eyes ranging from 3 to 6 mm (P ⬍ 0.001). However, patients without glaucoma also had consistent globe enlargement on the affected side. The mean difference between the 2 globes of patients in group 1 without glaucoma was 2 mm (P ⬍ 0.0001), although 8 patients had differences ranging from 3 to 5 mm. Mean axial length of both globes in group 2 was 23.8⫾1.7 mm with no significant difference between eyes.
Figure 1. Slit-lamp photograph of patient 7 in Table 1 with unilateral orbito-facial NF1 and ipsilateral ectropion uvea and glaucoma showing 3 quadrant ectropion uvea and Lisch nodules. NF1 ⫽ Neurofibromatosis type 1.
1726
Figure 2. Slit-lamp photograph of patient from group 2 with unilateral orbito-facial NF1 and ipsilateral ectropion uvea without glaucoma showing temporal ectropion uvea and a distorted pupil dragged superonasally. NF1 ⫽ Neurofibromatosis type 1.
Clinical characteristics of the 13 patients in group 1 with glaucoma are detailed in Table 1. Age at the time of glaucoma diagnosis varied from birth (in 5 patients) to 13 years. Eleven patients were diagnosed by age 3 years, whereas only 2 patients were diagnosed somewhat later, at ages 6 and 13 years. Twelve patients with glaucoma in group 1 had unilateral glaucoma and ipsilateral LPNF. Patient 8 is the first reported patient with NF1 with both bilateral LPNF and bilateral glaucoma. He also had the largest globe dimensions, measuring 33.30 mm right eye (OD) and 36.50 mm left eye (OS), and developed bilateral retinal detachments by age 11 years. Patient 7 had an axial length of 28.15 mm in the affected eye and developed a retinal detachment at age 25 years. Table 2 details at least 2 sequential axial length measurements in 5 patients with glaucoma from group 1. Progressive globe enlargement was documented in patients 5 and 8. Globe enlargement occurred in both eyes of patient 5, but size increase was more prominent in the affected eye than in the unaffected eye (5.7 mm vs. 2.7 mm). Patient 8 had bilateral globe enlargement measuring 3.1 mm OD and 3.3 mm OS over 5 years. Gonioscopy on all patients with glaucoma from group 1 revealed 1 or more abnormalities in the affected eye(s) of each patient, including irregular patches or marked angle pigmentation, anterior insertion of the iris, and extensive anterior synechiae (Fig 3). Ultrasound biomicroscopy performed on 4 patients with glaucoma in group 1 showed thickening of the ciliary body in the involved eye (Fig 4). Eleven of 34 patients without glaucoma examined in group 1 had few angle abnormalities, usually ipsilateral to the orbito-facial involvement, consisting of pigmentary patches or marked angle pigmentation and localized areas of anterior synechiae (Fig 5). All 4 patients with ectropion uvea but without glaucoma in group 1 had variable degrees of peripheral anterior synechiae. At last follow-up, 3 of 13 patients with glaucoma in group 1 had undergone evisceration and 2 patients had enucleation of the affected eye. Three additional eyes were blind, both eyes of patient 8 with bilateral glaucoma and bilateral inoperable retinal detachments and 1 eye of patient 7 with unilateral glaucoma, who also developed an inoperable retinal detachment. Two patients had hand motion vision in the affected eye, and 1 patient had 4/400 vision. Two patients were too young to assess visual acuity reliably.
Morales et al 䡠 Glaucoma in Neurofibromatosis Type 1 Table 1. Group 1 Patients with Glaucoma Patient Sex
Age at Diagnosis
LPNF
Affected Glaucoma Side* Side
1 2
M M
Birth 5 mos
⫹ ⫹
Left Left
Left Left
3
F
1 mo
⫹
Left
Left
4
F
Birth
⫹
Left
Left
5
M
6 yrs
⫹
Left
Left
6
F
“Childhood”
⫹
Left
Left
7
F
13 yrs
⫹
Right
Right
8
M
Birth
⫹
Both
Both
9
F
2 yrs
⫹
Left
Left
10
F
3 yrs
⫹
Right
Right
11
F
Birth
⫹
Right
Right
12
M
3 yrs
⫹
Left
Left
13
M
Birth
⫹
Right
Right
Last Visual Acuity NA OD 20/20 OS Prosthesis OD 20/20 OS Prosthesis OD 20/40 OS Prosthesis OD 20/25 OS 20/400 OD HM OS HM OD NLP OS 20/20 OD NLP OS NLP OD 20/30 OS 4/400 OD Prosthesis OS 8/30 (Allen cards) CSM OD 20/20 OS HM OD: LP OS: 20/20
Ectropion Axial Length Uvea (mm)
UBM
Comment
No No
— —
No
—
Yes
—
Yes
OD: 24.45 OS: 30.50 —
Thick CB OS
Thick CB OD Severe RD OD
Yes
OD: 28.15 OS: 22.57 OD: 33.30 OS: 36.50 —
No
—
No
OD: 25.83 OS: 21.88 OD: 22.23 OS: 28.26
Yes Yes Yes
Yes
Not done
Bilateral severe RD
Thick CB OD Thick CB OS
Yes
CB ⫽ ciliary body; CSM ⫽ central, steady, maintains fixation; HM ⫽ hand motion; LP ⫽ light perception; LPNF ⫽ lid plexiform neurofibroma; NA ⫽ not able to assess because of early age; NLP ⫽ no light perception; OD ⫽ right eye; OS ⫽ left eye; OU ⫽ both eyes; RD ⫽ retinal detachment; UBM ⫽ ultrasound biomicroscopy results. *Glaucoma was always present only on the affected side(s).
Discussion Neurofibromatosis type 1 is one of the most common autosomal dominant genetic disorders with a prevalence estimated as 1 in 3000 people worldwide. Expression is highly variable even among members of the same family. Some individuals are not even diagnosed because they carry an asymptomatic, very mild form of the disease, whereas others may have a very severe form causing multiple compliTable 2. Sequential Axial Length Measurements in Selected Group 1 Patients with Glaucoma
Patient
Eye
Axial Length Scan 1
5 (yrs 1, 18)
OD OS OD OS OD OS OD OS OD OS
21.50 24.80 28.42 23.00 30.20 33.20 25.71 21.92 21.57 28.39
7 (yrs 19, 22, 25) 8 (yrs 3, 4, 8) 11 (yrs 3, 3.5, 4) 12 (yrs 8, 15, 16)
Axial Length Scan 2
Axial Length Scan 3
24.25 30.50 28.15 22.57 30.31 34.17 25.83 21.88 22.23 28.26
27.80 23.07 33.30 36.50 26.48 22.29 22.40 27.74
Progression (>3 mm) ⫹ ⫹ ⫹
OD ⫽ right eye; OS ⫽ left eye. Eye affected by orbito-facial involvement highlighted in bold letters.
cations.11 Glaucoma associated with NF1 has been considered rare, particularly among young children. For example, a well-established glaucoma group in Boston reported only 1 case among 300 patients with all types of childhood glaucoma examined over a period of 6 years.2 We describe 13 patients with glaucoma due to orbitofacial NF1 who were examined over a period of 25 years at a tertiary care eye hospital in the Middle East, constituting 16.2% of the 80 patients with a diagnosis of NF1 treated during that period. As far as we know, this is the largest reported series of patients with glaucoma associated with NF1. Lid plexiform neurofibromas are said to occur in approximately 10% of individuals with NF1,11 and yet 52 patients (65%) in our NF1 population had LPNF. This proportion may reflect the fact that ptosis and facial deformity caused by LPNF were among the main reasons for referral. Glaucoma was present in 23% patients in group 1 with orbito-facial NF1, all of whom had LPNF ipsilateral to the glaucomatous eye(s). This proportion was lower than the 50% rate that is typically quoted in glaucoma textbooks10,12 and articles,2,13 even though a substantial number of patients in this series had advanced disease. Glaucoma was diagnosed at a young age in most patients. Infiltration of the anterior chamber angle by neurofibroma is by far the most commonly cited mechanism for glaucoma associated with NF1,2,3,7,10 but other mechanisms have been considered, including malformation or immature development of the anterior chamber angle,2,18 and second-
1727
Ophthalmology Volume 116, Number 9, September 2009
Figure 3. Gonio-photographs of patients 5 (A, B, D) and 12 (C) with NF1 and glaucoma showing (A, B) excessive angle pigmentation obscuring normal structures and landmarks in the affected eye, (C) irregular patches of pigment scattered around the angle, and (D) combination of excessive angle pigmentation and extensive peripheral anterior synechiae. NF1 ⫽ Neurofibromatosis type 1.
ary angle closure due to ciliary body/choroidal infiltration.2 Assigning a single mechanism as the cause of glaucoma in group 1 was not possible. Most patients had a combination of potential mechanisms, including congenital angle abnor-
malities, pigmentary disturbances of the angle, and secondary angle closure by anterior synechiae. Glaucoma present at birth in several patients implied a possible congenital abnormality of the angle, whereas presentation after infancy
Figure 5. Gonio-photographs of patients from group 2 with NF1 without glaucoma showing (A) pigmentary patches and excessive pigmentation of angle plus peripheral anterior synechiae (PAS); (B) increased pigmentation obscuring angle landmarks and extensive PAS; (C) sectoral PAS and prominent long iris processes; and (D) extensive, broad-based PAS. NF1 ⫽ Neurofibromatosis type 1.
1728
Morales et al 䡠 Glaucoma in Neurofibromatosis Type 1
Figure 4. Ultrasound biomicroscopy of patients 5 (A) and 12 (B) with unilateral orbito-facial involvement and ipsilateral glaucoma showing (A) thickening of ciliary body and (B) thickening and posterior displacement of ciliary body from the posterior iris surface.
suggests a progressive element such as infiltration of the angle by neurofibromatous tissue or secondary angle closure due to progressive ciliary body and choroidal infiltration. Ectropion uvea occurred only in patients with orbitofacial NF1 and only on the same side. It was present in more than half of the patients with glaucoma in group 1. Although congenital ectropion uvea is considered a static condition, the association with glaucoma is considered the rule, and it has been advocated that any infant or child with this condition should be carefully followed for the development of glaucoma.6 The association of NF1 and ectropion uvea has been reported.6,14 Among patients without glaucoma in group 1, 4 patients had ectropion uvea and partial angle closure due to peripheral anterior synechiae, suggesting that ectropion uvea could be a factor contributing to development of glaucoma even if it may not be a sufficient cause if limited in extent. All examined patients with orbito-facial NF1 and glaucoma had a larger globe on the affected side. Globe enlargement was severe, with axial lengths ranging from 26 to 36 mm and with a difference between unaffected and affected eyes from 4 to 6 mm. This study is the first time that progressive globe enlargement in NF1 has been quantified, although it was suspected previously because of enlarging corneal diameters7 and progressive exophthalmos or buphthalmos. Two patients (patients 5 and 8) with progressive ocular growth had ultrasound measurements when young (1 and 3 years) and then at a significantly older age (18 years and 8 years, respectively). It is possible that ocular enlargement in orbito-facial NF1 occurs at a specific age and that patients who did not have significant globe growth between measurements (e.g., patients 7 and 12) were either too old or too young to exhibit changes. Severe, progressive globe enlargement was probably an important factor in retinal detachments that occurred in 2 patients. Hoyt and Billson7 reported a patient with progressive globe enlargement after a glaucoma procedure despite controlled IOP, drawing attention to the fact that globe enlargement can occur in NF1 independently of intraocular hyper-
tension. They suggested that globe enlargement could be caused by a local growth factor released by neurofibromatous tissue. Other authors have also concluded that ocular hypertension may not be the only factor contributing to megalophthalmos,15,16 and our observation of consistent globe asymmetry between affected and unaffected eyes in patients without glaucoma in group 1 supports this hypothesis. Asymmetric globe growth might also occur because intermittent or partial lid occlusion deprives the developing globe of a clear retinal image, disrupts the normal emmetropization process, and promotes the development of axial myopia.17 We believe that this is the first report of bilateral orbitofacial involvement and bilateral glaucoma in NF1. Patient 8 had severe bilateral LPNF involvement with extreme bilateral globe enlargement and bilateral glaucoma that required multiple surgeries for IOP control. By age 11 years, this patient had developed bilateral inoperable retinal detachments, presumably precipitated by extreme globe growth. Gonioscopic abnormalities were observed in all patients with glaucoma, consisting of irregular pigment patches and markedly increased pigmentation obscuring the structures of the angle, anterior iris insertion, and large bands of peripheral anterior synechiae. It was impossible to tell whether abnormally pigmented tissue was neoplastic in nature or simply a pigmentary disturbance in patients who develop pigmentary abnormalities elsewhere. All patients with glaucoma had undergone surgery before this examination, and synechial adhesions could have been a result of prior intervention or the neurofibromatous process. Gonioscopic abnormalities were also present in several patients with orbito-facial involvement without glaucoma, including pigmentary changes, which would support the hypothesis that systemic pigmentary abnormalities affected the anterior chamber angle. A recent study reported only subtle gonioscopic changes in 42 patients with NF1, consisting mostly of anterior insertion of the iris and frequent long iris processes.18 The population reported may have had more dramatic and diverse changes in the angle because a larger
1729
Ophthalmology Volume 116, Number 9, September 2009 proportion of patients had orbito-facial involvement. Ultrasound biomicroscopy performed in 4 patients with glaucoma attributed to NF1 had thickening of the ciliary body that could cause mechanical distortion and closure of the angle. Glaucoma at birth associated with NF1 can be initially misdiagnosed as unilateral congenital glaucoma, especially in patients in whom LPNF and café au lait spots are not evident at an early age.3,15 Patient 5, for example, did not have striking LPNF early in life, and the diagnosis of NF1 with glaucoma was delayed until Lisch nodules were noticed at age 15 years. In patients with unilateral infantile glaucoma, it is advisable to search for evidence of NF1 and other causes of unilateral glaucoma. Examination of family members for signs of NF1 and repeated observation for the development of LPNF, café au lait spots, and Lisch nodules can help with the diagnosis. Unilateral globe enlargement despite IOP control can also indicate NF1.7,15 The combination of NF1 and glaucoma seems to carry a grim visual prognosis. One third of our patients with NF1 glaucoma had either enucleation or evisceration of the affected globe. Two patients were too young to ascertain a reliable visual acuity level, but the rest were either blind or legally blind in the affected eye at the last examination. Severe visual loss or blindness was the result of a variety of factors, including retinal detachment, optic nerve glioma, severe amblyopia caused by anisometropia, and glaucomatous optic nerve damage. In conclusion, neurofibromatosis is a variable and complex disease that can affect several components of the globe. Globe enlargement on the affected side seems common in orbito-facial NF1 and is aggravated by glaucoma. Particular attention to globe asymmetry by either refraction or ultrasound measurement is advisable in patients with unilateral orbito-facial involvement in NF1. Further studies are warranted to investigate mechanisms of globe enlargement and possible modes of treatment.
References 1. Ebert EM, Albert DM. The phakomatoses. In: Albert DM, Jakobiec FA, Azar DT, Gragoudas ES, eds. Principles and Practice of Ophthalmology. 2nd ed. vol. 6. Philadelphia, PA: Saunders; 2000:5120 – 44.
2. Grant WM, Walton DS. Distinctive gonioscopic findings in glaucoma due to neurofibromatosis. Arch Ophthalmol 1968; 79:127–34. 3. Brownstein S, Little JM. Ocular neurofibromatosis. Ophthalmology 1983;90:1595–9. 4. Francois J, Katz C. Homolateral association of hydrophthalmia, plexiform neuroma of the upper lid and facial hemihypertrophy in Recklinghausen’s disease [in French]. Ophthalmologica 1961;142:549 –71. 5. Snell S, Treacher Collins E. Plexiform neuroma (elephantiasis neuromatosis) of temporal region, orbit, eyelid, and eyeball. Trans Ophthalmol Soc U K 1903;23:157–77. 6. Ritch R, Forbes M, Hetherington J Jr, et al. Congenital ectropion uveae with glaucoma. Ophthalmology 1984;91:326 –31. 7. Hoyt CS, Billson FA. Buphthalmos in neurofibromatosis: is it an expression of regional giantism? J Pediatr Ophthalmol 1977;14:228 –34. 8. Jacquemin C, Bosley TM, Svedberg H. Orbit deformities in craniofacial neurofibromatosis type 1. AJNR Am J Neuroradiol 2003;24:1678 – 82. 9. National Institutes of Health Consensus Development Conference. Neurofibromatosis: conference statement. Arch Neurology 1988;45:575– 8. 10. Weiss JS, Ritch R. Glaucoma in the phacomatoses. In: Ritch R, Shields MB, Krupin T, eds. The Glaucomas. vol. 2. St. Louis, MO: Mosby; 1996:899 –924. 11. Sippel KC. Ocular findings in neurofibromatosis type 1. Int Ophthalmol Clin 2001;41:25– 40. 12. Mattox C. Glaucoma in the phakomatoses. In: Epstein DL, Allingham RR, Schuman JS, eds. Chandler and Grant’s Glaucoma. 4th ed. Baltimore, MD: Williams & Wilkins; 1997:433–44. 13. Lieb WA, Wirth WA, Geeraets WJ. Hydrophthalmos and neurofibromatosis (von Recklinghausen). Confin Neurol 1959; 19:230 – 47. 14. Wilson ME. Congenital iris ectropion and a new classification for anterior segment dysgenesis. J Pediatr Ophthalmol Strabismus 1990;27:48 –55. 15. Burke JP, Leitch RJ, Talbot JF, Parsons MA. Choroidal neurofibromatosis with congenital iris ectropion and buphthalmos: relationship and significance. J Pediatr Ophthalmol Strabismus 1991;28:265–7. 16. Bardelli AM, Hadjistilianou T. Buphthalmos and progressive elephantiasis in neurofibromatosis: a report of three cases. Ophthalmic Paediatr Genet 1989;10:279 – 86. 17. Qiao-Grider Y, Hung LF, Kee CS, et al. Recovery from form-deprivation myopia in rhesus monkeys. Invest Ophthalmol Vis Sci 2004;45:3361–72. 18. Quaranta L, Semeraro F, Turano R, Gandolfo E. Gonioscopic findings in patients with type 1 neurofibromatosis (Von Recklinghausen disease). J Glaucoma 2004;13:90 –5.
Footnotes and Financial Disclosures Originally received: December 10, 2008. Final revision: May 20, 2009. Accepted: June 10, 2009.
4
Manuscript no. 2008-1470.
1
Glaucoma Division, King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia. 2
Oculoplastic and Orbit Division, King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia. 3
Department of Ophthalmology, King Saud University, Riyadh, Saudi Arabia.
1730
Neurology Division, Cooper University Hospital, Camden, New Jersey. Presented in part as a poster at: American Academy of Ophthalmology Annual Meeting, November 2007, New Orleans, Louisiana. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: Jose Morales, MD, Glaucoma Division, King Khaled Eye Specialist Hospital, PO Box 7191, Riyadh 11462, Saudi Arabia. E-mail: jmorales@kkesh. med.sa.