- Email: [email protected]
Risk Factors for Amblyopia in Children with Capillary Hemangiomas of the Eyelids and Orbit
Risk Factors for Amblyopia in Children with Capillary Hemangiomas of the Eyelids and Orbit Shirah R. Schwartz, MD,a Francine Blei, MD,b,c Emily Ceisler, MD,d Mark Steele, MD,d Louis Furlan, MD,d and Sylvia Kodsi, MDa Introduction: Capillary hemangiomas are the most common orbital tumors of childhood and can cause amblyopia secondary to occlusion of the pupil, anisometropia, or strabismus. We undertook this study to describe the clinical characteristics of children with capillary hemangiomas and to propose a classification system to guide clinical treatment decisions. Methods: A retrospective review of the records of 129 patients with 132 capillary hemangiomas in two pediatric ophthalmology practices was conducted. Hemangiomas were classified based on size. Presence of aniosometropic astigmatism, ptosis, pupillary occlusion, lid margin change, proptosis, globe displacement, and strabismus was recorded. Results: Thirty-one hemangiomas measured less than 1 cm in greatest dimension and were not associated with amblyogenic factors. Seventy-five patients had hemangiomas that measured greater than 1 cm, 40 of which were associated with amblyopia. Eighteen children had diffuse hemangiomas that could not be measured and 14 of these were associated with amblyopia. Five of seven hemangiomas in six patients with PHACES syndrome were associated with amblyopia. Conclusion: This study is the largest review of capillary hemangiomas of the orbit and eyelids. Our findings suggest that size greater than 1 cm in largest diameter is an important predictor of amblyogenic factors and approximately half of these patients will require treatment. Diffuse hemangiomas and hemangiomas in patients with PHACES syndrome will cause amblyopia in the majority of cases. (J AAPOS 2006;10:262-268) apillary hemangiomas are the most common vascular tumors of childhood and often develop in the eyelids and orbit. Histologically, these unencapsulated lesions are composed of lobules of endothelial cells, which line anastomosing blood filled vascular spaces and are interspersed with fibrous septa. Capillary hemangiomas usually present as small tumors at birth or during the first several months of life.1-3 The next 3 to 6 months following their appearance is known as the proliferative phase.1,4-7 During this phase, endothelial cells multiply and form clusters allowing for larger blood-filled spaces. Following the proliferative phase, the hemangioma enters the involutional phase, which is characterized by deposition of fibrofatty tissue around the lobules of endothelial
C
From the aDepartment of Ophthalmology, North Shore-Long Island Jewish Health Systems, Great Neck, New York; and bDepartment of Pediatrics, cDepartment of Plastic Surgery, dDepartment of Ophthalmology, New York University Medical Center, New York, New York Presented as a paper at the 2005 Annual meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando Florida, March 9-13 2005. Submitted February 22, 2005. Revision accepted November 29, 2005. Reprint requests: Sylvia Kodsi, MD, 600 Northern Blvd., Suite 220, Great Neck, NY 11021 Copyright © 2006 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2006/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.01.210
262
June 2006
cells, causing fibrosis and involution of the hemangioma.8 The involutional phase occurs over a period of 5 to 7 years. Complete resolution of the tumor occurs in approximately 75 to 90% of children by age 7.9,10 While most periocular capillary hemangiomas cause transient problems related to cosmesis, a significant percentage of periocular hemangiomas will cause ophthalmic complications which can lead to significant vision loss. The incidence of these complications as reported in the literature has ranged from 46% to as high as 80%.11-13 These ocular complications include refractive, occlusional, and strabismic amblyopia, proptosis, ocular displacement, and optic nerve compression.1 Some periocular hemangiomas require early treatment to avoid vision loss. Two previous researchers have attempted to categorize capillary hemangiomas by characteristics on clinical presentation. Haik and coworkers classified hemangiomas into three categories based on the color of the skin associated with the hemangioma: classic superficial strawberry nevus; subcutaneous hemangiomas that appear dark blue or purple through the overlying skin; and deep orbital tumors that present with proptosis without skin discoloration.1 Alternatively, Rootman developed a second classification system that categorizes hemangiomas according to level of skin involvement: strawberry nevus confined to the superficial dermis; superficial hemangiomas combined with subcutaneous involvement; and combined subcutaneJournal of AAPOS
Journal of AAPOS Volume 10 Number 3 June 2006
FIG 1. Group 1 hemangioma: measurable hemangioma less than 1 cm in largest diameter.
Schwartz et al
263
FIG 2. Group 2 hemangioma: measurable hemangioma greater than 1 cm in largest diameter.
ous and deep orbital involvement.14 Although there have been recent attempts to use these classification systems to guide treatment decisions, neither system is used commonly as a practical way to evaluate hemangiomas and predict the need for treatment secondary to the development of amblyopia.15 In this study of children with capillary hemangiomas, we sought to identify risk factors associated with the development of amblyopia and to devise an easy-to-use classification system to accurately predict the risk of ophthalmic complications and assess the need for early treatment.
METHODS In this retrospective chart review we examined the clinical records of 129 patients with 132 capillary hemangiomas involving the eyelids and/or orbit who had been examined by S.K. between July 1994 and July 2004 and by either M.S., E.C., or L.F. between September 1994 and June 2004. Diagnosis was based on the appearance of the lesion and the history of onset and growth. All hemangiomas were classified on the basis of their size. Group 1 included all hemangiomas less than 1 cm in largest diameter at the time of the examination (Figure 1). We chose 1 cm as an easy-to-use dimension that corresponds to the approximate diameter of the cornea making accurate measurement of the largest diameter of the hemangioma possible. Group 2 included all measurable hemangiomas 1 cm or larger in diameter (Figure 2). Group 3 included diffuse (unable to measure) hemangiomas involving the eyelids and/or orbits (Figure 3). Group 4 included diffuse hemangiomas (unable to measure) involving eyelids and/or orbits which were accompanied by stigmata of PHACES syndrome (Posterior fossa malformations, Hemangiomas, Arterial anomalies, Cardiac defects and coarctation of the aorta, Eye abnormalities, and Sternal abnormalities or midline developmental defects) (Figure 4).16 The patients
FIG 3. Group 3 hemangioma: diffuse hemangioma.
in group 4 have been previously described by Kronenberg et al.17 Size of the hemangioma was determined from either a measurement recorded in the patient’s chart or photographs of the children taken as part of the medical record. When photographs were used, the size of the hemangioma was estimated based on its size relative to corneal diameter (approximately 1 cm). Patient information recorded included the age of presentation to the ophthalmologist, sex, race, and pertinent medical history. Characteristics of the hemangioma recorded included eye involvement (left versus right); nasal versus temporal location within the orbit; and lid involvement. Additionally, it was noted if the hemangioma caused a lid margin change, ptosis, pupil occlusion, proptosis, globe displacement, strabismus, or anisometropic amblyopia. We defined lid margin changes as any change in the normal upper or lower lid contour either centrally or
Journal of AAPOS Volume 10 Number 3 June 2006
264 Schwartz et al
TABLE 1. Patient population demographics
No. of patients Male:female Prematurity* Twins
Group 1
Group 2
Group 3
Group 4
Total
29
76
18
6
129
12:17 7 4
28:48 15 5
4:14 3 1
1:5 2 0
45:84 27 10
*Prematurity ⱕ36 weeks gestational age.
FIG 4. Group 4 hemangioma: diffuse hemangioma in child with PHACES syndrome.
peripherally. We defined globe displacement as any displacement of the globe nasally, temporally, superiorly, or inferiorly as compared with the other eye. Proptosis was defined as any anterior displacement of the globe when compared with the other eye. Occlusion of the pupil to cause occlusional amblyopia was defined as any occlusion of the pupil in primary gaze. Cycloplegia for the purposes of cycloplegic refraction was obtained using cyclopentolate 1.0% in all cases and tropicamide 1.0% and neosynephrine 2.5% as appropriate. We considered a difference of 1.50 D or greater of astigmatism between the two eyes on cycloplegic refraction significant enough to produce anisometropic amblyopia based on the American Academy of Ophthalmology’s (AAO) preferred practice pattern of amblyopia in preverbal children.18
RESULTS A total of 129 patients were included in this study. Three of these patients had bilateral periocular capillary hemangiomas. Patients were grouped according to the size of the hemangioma as previously described. Thirty-one hemangiomas were included in group 1, 76 in group 2, 18 in group 3, and 7 in group 4. Of the 129 patients studied, 103 were Caucasian, 8 were African-American, 8 were Hispanic, 7 were Southeast Asian, and 3 were Asian. The average age of the child upon presentation to the ophthalmologist was 237 days. There were 45 (35%) males and 84 (65%) females. Twenty-seven of the 129 (21%) children were born prematurely, prior to 36 weeks gestation, and 10 children were twin births. The sex ratio, number of premature children, and number of twin births in each group are outlined in Table 1. Additional characteristics of the hemangiomas studied within each group are presented in Table 2. There were 70 (53%) left-sided hemangiomas and 62 (47%) right-sided
hemangiomas. There were 50 (38%) hemangiomas that caused ptosis and 56 (42%) that caused a lid margin change. Seven hemangiomas caused globe displacement and another five hemangiomas caused proptosis. Strabismus occurred in 18 patients with a total of 20 hemangiomas. A total of 45% (59/132) of hemangiomas included in this study caused amblyopia. Forty-three percent (57/132) of hemangiomas caused astigmatic anisometropic amblyopia and 11% (14/132) of hemangiomas caused deprivational amblyopia from occlusion of the pupil. Twelve hemangiomas caused both astigmatic anisometropic amblyopia and occlusion of the pupil. All 12 hemangiomas in this study that caused proptosis or globe displacement also caused astigmatic anisometropic amblyopia. Table 3 shows the correlation between nasal, temporal, and central location of the hemangioma and the presence or absence of amblyopia. Central hemangiomas were either centered on the lid or involved the whole lid. Table 4 shows the relationship between lid involvement of the hemangioma and the presence or absence of amblyopia. Those that did not involve the eyelids involved either the medial or the lateral orbit. None of the 31 hemangiomas in group 1 caused occlusion or anisometropic amblyopia. Using the 2 test, we found a significant difference between group 1 and groups 2, 3, and 4 in predicting the presence of amblyopia (P ⫽ 0.0002). Four hemangiomas on three patients in group 1 were associated with strabismus. One patient had a congenital esotropia; one had an accommodative esotropia, and one had an exotropia. Four hemangiomas caused ptosis and five caused a lid margin change. None of the hemangiomas in this group caused proptosis or globe displacement. Forty of the 76 (53%) hemangiomas included in group 2 caused amblyopia. All 40 hemangiomas caused anisometropic amblyopia and 7 of these 40 (18%) also caused occlusional amblyopia. Five of the seven hemangiomas associated with strabismus were associated with amblyopia. The two patients with strabismus who did not have amblyopia had an accommodative estropia. The presence of ptosis, lid margin change, proptosis, and globe displacement in group 2 is presented in Table 2. Twenty of 28 (71%) hemangiomas that caused ptosis and 23 of 33 (70%) hemangiomas that caused a lid margin also caused aniosometropic amblyopia. Within group 2 there was a statistically significant relationship between the presence of
Journal of AAPOS Volume 10 Number 3 June 2006
Schwartz et al
265
TABLE 2. Characteristics of hemangiomas and their effects Group 1
Group 2
Group 3
Group 4
Total
No. of hemangiomas Left eye:right eye Ptosis Eyelid margin change Globe displacement Proptosis Strabismus Amblyopia Astigmatism Occlusion of pupil
76 (58%) 42:34 28 (37%) 33 (43%) 3 (4%) 1 (1%) 7 (9%) 40 (53%) 40 (53%) 7 (9%)
18 (14%) 7:11 13 (72%) 11 (61%) 3 (17%) 2 (11%) 3 (17%) 14 (78%) 12 (67%) 4 (22%)
7 (5%) 4:3 5 (71%) 7 (100%) 1 (14%) 2 (29%) 5 (71%) 5 (71%) 4 (57%) 3 (43%)
132 70:62 50 (38%) 56 (42%) 7 (5%) 5 (4%) 20 (15%) 59 (45%) 56 (42%) 14 (11%)
31 (23%) 17:14 4 (13%) 5 (16%) 0 (0%) 0 (0%) 5 (16%) 0 (0%) 0 (0%) 0 (0%)
TABLE 3. Hemangioma location and amblyopia Hemangioma laterality Nasal Group Group Group Group Total
Temporal
Central
Amblyopia
No amblyopia
Total
Amblyopia
No amblyopia
Total
Amblyopia
No amblyopia
Total
0 27 5 0 32
10 15 1 0 26
10 42 6 0 58
0 4 3 0 7
10 11 2 0 23
10 15 5 0 30
0 9 6 5 20
11 10 1 2 24
11 19 7 7 44
1 2 3 4
TABLE 4. Hemangioma location and amblyopia Eyelid involvement Upper eyelid
Lower eyelid
Both eyelids
No eyelid involvement
No No No No Amblyopia amblyopia Total Amblyopia amblyopia Total Amblyopia amblyopia Total Amblyopia amblyopia Total Group Group Group Group Total
1 2 3 4
0 28 9 2 39
25 23 0 1 49
25 51 9 3 88
0 10 0 0 10
5 10 2 0 17
ptosis (P ⫽ 0.025) or lid margin change (P ⫽ 0.01) and the presence of amblyopia. Furthermore nasal location of the hemangioma when compared with temporal or central location was a statistically significant factor in predicting the presence of amblyopia (P ⫽ 0.03) In group 3, 14 of 18 (78%) hemangiomas induced amblyopia. Twelve of these patients had anisometropic amblyopia and 4 had occlusion amblyopia. All three hemangiomas that were associated with strabismus were also accompanied by amblyopia. A summary of the presence of ptosis, lid margin change, proptosis, and globe displacement in group 3 is presented in Table 2. Eleven of 13 (85%) hemangiomas that caused ptosis and 7 of 11 (64%) hemangiomas that caused lid margin changes also caused amblyopia. Five of the seven (72%) hemangiomas included in group 4 caused amblyopia. Four hemangiomas induced anisometropic amblyopia and three caused occlusion of the pupil. One hemangioma compressed the optic nerve causing vision loss. There were five hemangiomas on four
5 20 2 0 27
0 2 5 2 9
1 1 2 2 6
1 3 7 4 15
0 0 0 0 0
0 2 0 0 2
0 2 0 0 2
children that were accompanied by strabismus. Two patients had sensory exotropia and two had infantile esotropia. Both patients with infantile esotropia had developmental delays and neurological deficits, although only one patient was thought to have esotropia secondary to the hemangioma. In this group, there were five hemangiomas that induced ptosis and all seven caused a lid margin change. Two hemangiomas caused proptosis and one caused globe displacement.
DISCUSSION This study is the largest investigation of periocular capillary hemangiomas published to date. Our hypothesis was that grouping hemangiomas by size and character would allow definition of those with increased potential for contributing to amblyopia. Proper identification of problematic hemangiomas is crucial because the various treatments for periocular capillary hemangioma have well-known complications.19-24 Treatment options include but are not limited to oral steroids, intralesional injection and/or top-
266 Schwartz et al ical steroids, interferon-␣, argon laser, and surgical excision.25-32 In some children prompt treatment is necessary because the hemangioma can cause amblyogenic factors that can lead to permanent vision loss.33 Therefore, our goal was to help both pediatricians and ophthalmologists more easily differentiate between those benign hemangiomas that do not cause amblyopia and those that may result in permanent visual loss if left untreated. Similar to previous studies, our study group consisted mainly of female patients.34 Only 21% of patients in groups 3 and 4 were male, while 38% of children in groups 1 and 2 were male. A review of capillary hemangiomas by Stigmar et al found that not only were females more frequently affected by hemangiomas, but they were also more severely amblyopic.12 This is consistent with the results of the current study, which shows that larger hemangiomas were more likely to be present in females and be associated with amblyopia. It is possible that hemangiomas have a higher incidence in males than reported in the literature but because of the small size of the hemangiomas they may not come to the attention of a physician. Similar to other studies the majority of our study population was Caucasian.1 Prematurity is a known risk factor for the development of periocular hemangiomas and in our study 22% of the children in this study were born prematurely.32,35 Our results show no clear relationship between prematurity and the size of the hemangioma. Our study also has a greater percentage of twin births than previously reported.1 Twin births are also associated with an increased risk of prematurity and therefore may not be an independent risk factor for the development of hemangiomas. Multiple studies have examined the relationship between capillary hemangiomas and visual loss from amblyopia. Patients with capillary hemangiomas usually present within the first 3 months of age when the visual sensory system is developing, making amblyopia a great concern. Sensory deprivation caused by occlusion of the pupil by the eyelid or hemangioma and anisometropia related to pressure of the tumor on the anterior segment of the eye are possible causes of amblyopia. Rarely, posterior pressure by intraconal masses may also cause anisometropia. Although strabismus is often present in patients with capillary hemangiomas, both Stigmar and Robb suggested that strabismus is a secondary phenomenon and not the primary cause of amblyopia associated with capillary hemangiomas.11,12 Previous studies of hemangiomas have addressed the relationship between the size of the hemangioma and the need for treatment. In his study, Robb stated that “the prevalence of asymmetrical refractive errors in patients with hemangiomas of the eyelid and orbit might be expected to vary with the size of the lesions.”11 Stigmar et al stated that hemangiomas involving only one-third of the eyelid margin are less likely to affect vision and that the risk of anisometropic amblyopia is greatest in children
Journal of AAPOS Volume 10 Number 3 June 2006
with local but bulky hemangiomas.12 However, neither study quantitatively measured the size of the hemangiomas that were examined. In this study we found that the 31 hemangiomas in group 1 that measured less that 1 cm in largest diameter were not associated with anisometropia or occlusional amblyopia. It is likely that hemangiomas of this size are not large or bulky enough to cause the degree of lid ptosis needed to cause occlusion, nor can they apply enough pressure on the globe to cause astigmatism. Thus hemangiomas of this size are unlikely to need treatment to prevent the development of amblyopia but should be monitored closely for growth. Patients in group 1 with strabismus were primarily referred for the strabismus and the hemangiomas were incidentally noted on examination. It is unlikely that the strabismus in this group was related to the presence of the hemangioma because none of these patients had amblyopia. On the other hand, patients in group 3 with large, diffuse hemangiomas had amblyopia in 78% of cases. Sixty-seven percent of hemangiomas in this group were associated with anisometropic amblyopia and 22% were associated with pupillary occlusion. Of the four patients in group 3 that did not have amblyopia by the AAO’s guidelines, three patients were treated anyway. One patient had ⫹1.25 D of astigmatism and two patients had hemangiomas that demonstrated rapid progression and further growth would have caused occlusion of the pupil. As previously described, hemangiomas associated with PHACES syndrome are often large in size and involve the eyes as well as a large part of the face.16 Because of the large size of these hemangiomas, five of the seven hemangiomas in this group caused amblyopia. Three of seven hemangiomas caused pupillary occlusion and four of seven caused anisometropic amblyopia. Of note, one hemangioma in this group caused loss of vision secondary to ischemic and/or compressive optic neuropathy. This was thought to be secondary to the massive proptosis and retro-orbital component of the tumor. While none of the small hemangiomas in group 1 caused amblyopia and three-quarters (76%) of the very large hemangiomas in groups 3 and 4 caused amblyopia, approximately half (53%) of hemangiomas in group 2 caused amblyopia. Therefore, we looked at characteristics of these group 2 hemangiomas to determine which characteristics were associated with the presence or absence of amblyopia. Our study indicates that there is a statistically significant relationship between hemangioma location and the presence of amblyopia. While 64% of hemangiomas in group 2 that were located nasally caused amblyopia, only 27% of those that were located temporally required treatment. This is likely because temporal hemangiomas are not as confined as nasal hemangiomas by the orbital rim and the nasal bridge and therefore can transmit molding pressure over a larger area of the globe’s surface, making a
Journal of AAPOS Volume 10 Number 3 June 2006
more focal compression of the cornea (ie, astigmatism) less likely. The centrally located hemangiomas included in group 2 caused amblyopia in 47% of hemangiomas. This is more frequent than amblyopia caused by temporal hemangioma but less frequent than that caused by nasal hemangiomas. Additionally, there was a statistically significant relationship between the presence of ptosis or lid margin change in measurable hemangiomas greater than 1 cm in diameter and the presence of amblyopia. Over 71% of those hemangiomas in group 2 that caused ptosis and 70% of those hemangiomas that caused a change in lid margin contour caused amblyopia. Amblyopia was present in five of seven (71%) of the hemangiomas in group 2 accompanied by strabismus. Finally, as was true for the entire study population, all hemangiomas accompanied by proptosis or globe displacement also caused amblyopia. Therefore, hemangioma characteristics in patients with measurable hemangiomas greater than 1 cm that were more likely to have amblyopia include nasal location, ptosis, lid margin change, proptosis, globe displacement, and strabismus. In conclusion, this study shows that there is a clear relationship between the size of the hemangioma and the presence or absence of amblyopia. While hemangiomas less than 1 cm in greatest diameter are extremely unlikely to cause amblyopia, large diffuse hemangiomas and those hemangiomas associated with PHACES are extremely likely to cause amblyopia. Approximately half of hemangiomas that are greater than 1 cm but are not diffuse will cause amblyopia. Those hemangiomas that are nasally located cause ptosis, lid margin change, proptosis, globe displacement, or strabismus are more likely to cause amblyopia. Therefore, while all children with capillary hemangiomas require a full ophthalmologic examination, those patients with larger hemangiomas and those associated with characteristics that are predictive of amblyopia should be monitored closely for the development of amblyopia and the need for early treatment. References 1. Haik BG, Jacobiec FA, Ellsworth RM, Jones IS. Capillary hemangiomas of the lids and orbit: an analysis of the clinical features and therapeutic results in 101 cases. Ophthalmology 1979;86:760-89. 2. Henderson JW. Orbital tumors. 2nd ed. New York: ThiemeStratton; 1980. 3. Reese AB. Tumors of the eye. 3nd ed. Hagerstown (MD): Harper and Row; 1976. 4. Holmdahl K. Cutaneous hemangiomas in premature infants. Acta Paediatr 1955;44:370-9. 5. Jacobs A, Walton R. The incidence of birthmarks in neonates. Pediatrics 1976;58:218-22. 6. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412-20. 7. Mulliken JB, Young AE. Vascular birthmarks: hemangiomas and malformations. Philadelphia (PA): Saunders; 1988. p. 1-103. 8. Haik BG, Karcioglu ZA, Gordon RA, Pechous BP. Capillary hemangioma (infantile periocular hemangioma). Surv Ophthalmol 1994; 38:399-426.
Schwartz et al
267
9. Bowers RE, Graham EA, Tomlinson KM. The natural history of the strawberry nevus. Arch Dermatol 1960;82:59-72. 10. Margileth AM, Museles M. Cutaneous hemangiomas in children. JAMA 1965;194:135-8. 11. Robb RM. Refractive errors associated with the eyelids and orbit in infancy. Am J Ophthalmol 1977;83:52-8. 12. Stigmar G, Crawford JS, Ward CM, Thomson HG. Ophthalmic sequelae of infantile hemangiomas of the eyelids and orbit. Am J Ophthalmol 1978;85:806-13. 13. Bogan S, Simon J, Krohel GB, Nelson LB. Astigmatism associated with adnexal masses in infancy. Arch Ophthalmol 1987;105:1368-70. 14. Rootman J. Diseases of the orbit. Philadelphia (P)A: JB Lippincott; 1988. 15. Ceisler EJ, Santos L, Blei F. Periocular hemangioma: what every physician should know. Pediatr Dermatol 2004;21:1-9. 16. Frieden IJ, Reese V, Cohen D. PHACE syndrome: the association of posterior fossa brain malformations, hemangiomas, arterial anomalies, coarctation of the aorta and cardiac defects, and eye abnormalities. Arch Dermatol 1996;132:307-11. 17. Kronenberg A, Blei F, Ceisler E, Steele M, Furlan L, Kodsi S. Ocular and systemic manifestations of PHACES (posterior fossa malformations, hemangiomas, arterial anomalies, cardiac defects and coarctation of the aorta, eye abnormalities, and sternal abnormalities or ventral developmental defects) syndrome. J AAPOS 2005;9:16973. 18. American Academy of Ophthalmology: Amblyopia, Preferred Practice Pattern. San Francisco (CA): American Academy of Ophthalmology, 1994. 19. Egbert JE, Paul S, Engel WK, Summers G. High injection pressure during intralesional injection of corticosteroids into capillary hemangiomas. Arch Ophthalmol 2001;119:677-83. 20. Vazquez-Botet R, Reyes BA, Vazquez-Botet M. Sclerodermiform linear atrophy after the use of intralesional steroids for periorbital hemangiomas: a review of complications. J Pediatr Ophthalmol Strabismus 1989;26:124-7. 21. Egbert JE, Nelson SC. Neurologic toxicity associated with interferon alfa treatment of capillary hemangioma. J Pediatr Ophthalmol Strabismus 1997;1:190. 22. Motwani MV, Simon JW, Pickering JD, et al. Steroid injection versus conservative treatment of anisometropia amblyopia in juvenile adnexal hemangioma. J Pediatr Ophthalmol Strabismus 1995;32: 26-8. 23. Goyal R, Watts P, Lane CM, Berk L, Gregory JW. Adrenal suppression and failure to thrive after steroid injections for periocular hemangioma. Ophthalmology 2004;11:389-95. 24. Egbert JE, Schwartz GS, Walsh AW. Diagnosis and treatment of an ophthalmic artery occlusion during an intralesional injection of corticosteroid into an eyelid capillary hemangioma. Am J Ophthalmol 1996;121:638-42. 25. Walker RS, Custer PL, Nerad JA. Surgical excision of periorbital capillary hemangiomas. Ophthalmology 1994;101:1333-40. 26. Cruz OA, Zarnegar SR, Myers SE. Treatment of periocular capillary hemangioma with topical clobetasol propionate. Ophthalmology 1995;102:2012-5. 27. Elsas FJ, Lewis AR. Topical treatment of periocular capillary hemangioma. J Pediatr Ophthalmol Strabismus 1994;31:153-6. 28. Deans RM, Harris GJ, Kivlin JD. Surgical dissection of capillary hemangiomas. Arch Ophthalmol 1992;110:1743-47. 29. Loughnan MS, Elder J, Kemp A. Treatment of massive orbitalcapillary hemangioma with interferon alfa-2b: short-term results. Arch Ophthalmol 1992;110:1366-7. 30. Kushner BJ. The treatment of periorbital infantile hemangioma with intralesional corticosteroid. Plast Reconst Surg 1985;76:517-23. 31. Summers CG, Hordinsky MD. Argon laser treatment of periocular lesions: an experimental study. Ophthalm Surg 1987;18:100-2. 32. Hastings MM, Milot J, Barsoum-Homsy M, Hershon L, Dubois L, Leclerc JM. Recombinant interferon alfa-2b in the treatment of
Journal of AAPOS Volume 10 Number 3 June 2006
268 Schwartz et al vision-threatening capillary hemangiomas in childhood. J Pediatr Ophthalmol Strabismus 1997;1:226-30. 33. Ceisler E, Blei F. Ophthalmic issues in hemangiomas of infancy. Lymphatic Res Biol 2003;1:321-9.
34. Metry DW, Herbert AA. Benign cutaneous vascular tumors of infancy: when to worry, what to do. Arch Dermatol 2000;136:905-14. 35. Amir J, Metzker A, Krikler R, Reisner SH. Strawberry hemangiomas in preterm infants. Pediatr Dermatol 1986;3:331-2.
An Eye on the Arts – The Arts on the Eye
. . . As these new sights flooded my grandparents’ brains, they jostled with images from the day before. Ellis Island, rising like a Doge’s Palace on the water. The Baggage Room stacked to the ceiling with luggage. They’d been herded up a stairway to the Registry Room. Pinned with numbers from the Giulia’s manifest, they’d filed past a line of health inspectors who’d looked in their eyes and ears, rubbed their scalps, and flipped their eyelids inside out with buttonhooks. One doctor, noticing inflammation under Dr. Philobosian’s eyelids, had stopped the examination and chalked an X on his coat. He was led out of line. My grandparents hadn’t seen him again. “He must have caught something on the boat,” Desdemona said. “Or his eyes were red from all that crying.” Meanwhile, chalk continued to do its work all around them. It marked a Pg on the belly of a pregnant woman. It scrawled an H over an old man’s failing heart. It diagnosed the C of conjunctivitis, the F of favus, and the T of trachoma. But, no matter how well trained, medical eyes couldn’t spot a recessive mutation hiding out on a fifth chromosome. Fingers couldn’t feel it. Buttonhooks couldn’t bring it to light . . . — Jeffrey Eugenides (from Middlesex, Farrar, Straus and Giroux)
C
From the aDepartment of Ophthalmology, North Shore-Long Island Jewish Health Systems, Great Neck, New York; and bDepartment of Pediatrics, cDepartment of Plastic Surgery, dDepartment of Ophthalmology, New York University Medical Center, New York, New York Presented as a paper at the 2005 Annual meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando Florida, March 9-13 2005. Submitted February 22, 2005. Revision accepted November 29, 2005. Reprint requests: Sylvia Kodsi, MD, 600 Northern Blvd., Suite 220, Great Neck, NY 11021 Copyright © 2006 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2006/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.01.210
262
June 2006
cells, causing fibrosis and involution of the hemangioma.8 The involutional phase occurs over a period of 5 to 7 years. Complete resolution of the tumor occurs in approximately 75 to 90% of children by age 7.9,10 While most periocular capillary hemangiomas cause transient problems related to cosmesis, a significant percentage of periocular hemangiomas will cause ophthalmic complications which can lead to significant vision loss. The incidence of these complications as reported in the literature has ranged from 46% to as high as 80%.11-13 These ocular complications include refractive, occlusional, and strabismic amblyopia, proptosis, ocular displacement, and optic nerve compression.1 Some periocular hemangiomas require early treatment to avoid vision loss. Two previous researchers have attempted to categorize capillary hemangiomas by characteristics on clinical presentation. Haik and coworkers classified hemangiomas into three categories based on the color of the skin associated with the hemangioma: classic superficial strawberry nevus; subcutaneous hemangiomas that appear dark blue or purple through the overlying skin; and deep orbital tumors that present with proptosis without skin discoloration.1 Alternatively, Rootman developed a second classification system that categorizes hemangiomas according to level of skin involvement: strawberry nevus confined to the superficial dermis; superficial hemangiomas combined with subcutaneous involvement; and combined subcutaneJournal of AAPOS
Journal of AAPOS Volume 10 Number 3 June 2006
FIG 1. Group 1 hemangioma: measurable hemangioma less than 1 cm in largest diameter.
Schwartz et al
263
FIG 2. Group 2 hemangioma: measurable hemangioma greater than 1 cm in largest diameter.
ous and deep orbital involvement.14 Although there have been recent attempts to use these classification systems to guide treatment decisions, neither system is used commonly as a practical way to evaluate hemangiomas and predict the need for treatment secondary to the development of amblyopia.15 In this study of children with capillary hemangiomas, we sought to identify risk factors associated with the development of amblyopia and to devise an easy-to-use classification system to accurately predict the risk of ophthalmic complications and assess the need for early treatment.
METHODS In this retrospective chart review we examined the clinical records of 129 patients with 132 capillary hemangiomas involving the eyelids and/or orbit who had been examined by S.K. between July 1994 and July 2004 and by either M.S., E.C., or L.F. between September 1994 and June 2004. Diagnosis was based on the appearance of the lesion and the history of onset and growth. All hemangiomas were classified on the basis of their size. Group 1 included all hemangiomas less than 1 cm in largest diameter at the time of the examination (Figure 1). We chose 1 cm as an easy-to-use dimension that corresponds to the approximate diameter of the cornea making accurate measurement of the largest diameter of the hemangioma possible. Group 2 included all measurable hemangiomas 1 cm or larger in diameter (Figure 2). Group 3 included diffuse (unable to measure) hemangiomas involving the eyelids and/or orbits (Figure 3). Group 4 included diffuse hemangiomas (unable to measure) involving eyelids and/or orbits which were accompanied by stigmata of PHACES syndrome (Posterior fossa malformations, Hemangiomas, Arterial anomalies, Cardiac defects and coarctation of the aorta, Eye abnormalities, and Sternal abnormalities or midline developmental defects) (Figure 4).16 The patients
FIG 3. Group 3 hemangioma: diffuse hemangioma.
in group 4 have been previously described by Kronenberg et al.17 Size of the hemangioma was determined from either a measurement recorded in the patient’s chart or photographs of the children taken as part of the medical record. When photographs were used, the size of the hemangioma was estimated based on its size relative to corneal diameter (approximately 1 cm). Patient information recorded included the age of presentation to the ophthalmologist, sex, race, and pertinent medical history. Characteristics of the hemangioma recorded included eye involvement (left versus right); nasal versus temporal location within the orbit; and lid involvement. Additionally, it was noted if the hemangioma caused a lid margin change, ptosis, pupil occlusion, proptosis, globe displacement, strabismus, or anisometropic amblyopia. We defined lid margin changes as any change in the normal upper or lower lid contour either centrally or
Journal of AAPOS Volume 10 Number 3 June 2006
264 Schwartz et al
TABLE 1. Patient population demographics
No. of patients Male:female Prematurity* Twins
Group 1
Group 2
Group 3
Group 4
Total
29
76
18
6
129
12:17 7 4
28:48 15 5
4:14 3 1
1:5 2 0
45:84 27 10
*Prematurity ⱕ36 weeks gestational age.
FIG 4. Group 4 hemangioma: diffuse hemangioma in child with PHACES syndrome.
peripherally. We defined globe displacement as any displacement of the globe nasally, temporally, superiorly, or inferiorly as compared with the other eye. Proptosis was defined as any anterior displacement of the globe when compared with the other eye. Occlusion of the pupil to cause occlusional amblyopia was defined as any occlusion of the pupil in primary gaze. Cycloplegia for the purposes of cycloplegic refraction was obtained using cyclopentolate 1.0% in all cases and tropicamide 1.0% and neosynephrine 2.5% as appropriate. We considered a difference of 1.50 D or greater of astigmatism between the two eyes on cycloplegic refraction significant enough to produce anisometropic amblyopia based on the American Academy of Ophthalmology’s (AAO) preferred practice pattern of amblyopia in preverbal children.18
RESULTS A total of 129 patients were included in this study. Three of these patients had bilateral periocular capillary hemangiomas. Patients were grouped according to the size of the hemangioma as previously described. Thirty-one hemangiomas were included in group 1, 76 in group 2, 18 in group 3, and 7 in group 4. Of the 129 patients studied, 103 were Caucasian, 8 were African-American, 8 were Hispanic, 7 were Southeast Asian, and 3 were Asian. The average age of the child upon presentation to the ophthalmologist was 237 days. There were 45 (35%) males and 84 (65%) females. Twenty-seven of the 129 (21%) children were born prematurely, prior to 36 weeks gestation, and 10 children were twin births. The sex ratio, number of premature children, and number of twin births in each group are outlined in Table 1. Additional characteristics of the hemangiomas studied within each group are presented in Table 2. There were 70 (53%) left-sided hemangiomas and 62 (47%) right-sided
hemangiomas. There were 50 (38%) hemangiomas that caused ptosis and 56 (42%) that caused a lid margin change. Seven hemangiomas caused globe displacement and another five hemangiomas caused proptosis. Strabismus occurred in 18 patients with a total of 20 hemangiomas. A total of 45% (59/132) of hemangiomas included in this study caused amblyopia. Forty-three percent (57/132) of hemangiomas caused astigmatic anisometropic amblyopia and 11% (14/132) of hemangiomas caused deprivational amblyopia from occlusion of the pupil. Twelve hemangiomas caused both astigmatic anisometropic amblyopia and occlusion of the pupil. All 12 hemangiomas in this study that caused proptosis or globe displacement also caused astigmatic anisometropic amblyopia. Table 3 shows the correlation between nasal, temporal, and central location of the hemangioma and the presence or absence of amblyopia. Central hemangiomas were either centered on the lid or involved the whole lid. Table 4 shows the relationship between lid involvement of the hemangioma and the presence or absence of amblyopia. Those that did not involve the eyelids involved either the medial or the lateral orbit. None of the 31 hemangiomas in group 1 caused occlusion or anisometropic amblyopia. Using the 2 test, we found a significant difference between group 1 and groups 2, 3, and 4 in predicting the presence of amblyopia (P ⫽ 0.0002). Four hemangiomas on three patients in group 1 were associated with strabismus. One patient had a congenital esotropia; one had an accommodative esotropia, and one had an exotropia. Four hemangiomas caused ptosis and five caused a lid margin change. None of the hemangiomas in this group caused proptosis or globe displacement. Forty of the 76 (53%) hemangiomas included in group 2 caused amblyopia. All 40 hemangiomas caused anisometropic amblyopia and 7 of these 40 (18%) also caused occlusional amblyopia. Five of the seven hemangiomas associated with strabismus were associated with amblyopia. The two patients with strabismus who did not have amblyopia had an accommodative estropia. The presence of ptosis, lid margin change, proptosis, and globe displacement in group 2 is presented in Table 2. Twenty of 28 (71%) hemangiomas that caused ptosis and 23 of 33 (70%) hemangiomas that caused a lid margin also caused aniosometropic amblyopia. Within group 2 there was a statistically significant relationship between the presence of
Journal of AAPOS Volume 10 Number 3 June 2006
Schwartz et al
265
TABLE 2. Characteristics of hemangiomas and their effects Group 1
Group 2
Group 3
Group 4
Total
No. of hemangiomas Left eye:right eye Ptosis Eyelid margin change Globe displacement Proptosis Strabismus Amblyopia Astigmatism Occlusion of pupil
76 (58%) 42:34 28 (37%) 33 (43%) 3 (4%) 1 (1%) 7 (9%) 40 (53%) 40 (53%) 7 (9%)
18 (14%) 7:11 13 (72%) 11 (61%) 3 (17%) 2 (11%) 3 (17%) 14 (78%) 12 (67%) 4 (22%)
7 (5%) 4:3 5 (71%) 7 (100%) 1 (14%) 2 (29%) 5 (71%) 5 (71%) 4 (57%) 3 (43%)
132 70:62 50 (38%) 56 (42%) 7 (5%) 5 (4%) 20 (15%) 59 (45%) 56 (42%) 14 (11%)
31 (23%) 17:14 4 (13%) 5 (16%) 0 (0%) 0 (0%) 5 (16%) 0 (0%) 0 (0%) 0 (0%)
TABLE 3. Hemangioma location and amblyopia Hemangioma laterality Nasal Group Group Group Group Total
Temporal
Central
Amblyopia
No amblyopia
Total
Amblyopia
No amblyopia
Total
Amblyopia
No amblyopia
Total
0 27 5 0 32
10 15 1 0 26
10 42 6 0 58
0 4 3 0 7
10 11 2 0 23
10 15 5 0 30
0 9 6 5 20
11 10 1 2 24
11 19 7 7 44
1 2 3 4
TABLE 4. Hemangioma location and amblyopia Eyelid involvement Upper eyelid
Lower eyelid
Both eyelids
No eyelid involvement
No No No No Amblyopia amblyopia Total Amblyopia amblyopia Total Amblyopia amblyopia Total Amblyopia amblyopia Total Group Group Group Group Total
1 2 3 4
0 28 9 2 39
25 23 0 1 49
25 51 9 3 88
0 10 0 0 10
5 10 2 0 17
ptosis (P ⫽ 0.025) or lid margin change (P ⫽ 0.01) and the presence of amblyopia. Furthermore nasal location of the hemangioma when compared with temporal or central location was a statistically significant factor in predicting the presence of amblyopia (P ⫽ 0.03) In group 3, 14 of 18 (78%) hemangiomas induced amblyopia. Twelve of these patients had anisometropic amblyopia and 4 had occlusion amblyopia. All three hemangiomas that were associated with strabismus were also accompanied by amblyopia. A summary of the presence of ptosis, lid margin change, proptosis, and globe displacement in group 3 is presented in Table 2. Eleven of 13 (85%) hemangiomas that caused ptosis and 7 of 11 (64%) hemangiomas that caused lid margin changes also caused amblyopia. Five of the seven (72%) hemangiomas included in group 4 caused amblyopia. Four hemangiomas induced anisometropic amblyopia and three caused occlusion of the pupil. One hemangioma compressed the optic nerve causing vision loss. There were five hemangiomas on four
5 20 2 0 27
0 2 5 2 9
1 1 2 2 6
1 3 7 4 15
0 0 0 0 0
0 2 0 0 2
0 2 0 0 2
children that were accompanied by strabismus. Two patients had sensory exotropia and two had infantile esotropia. Both patients with infantile esotropia had developmental delays and neurological deficits, although only one patient was thought to have esotropia secondary to the hemangioma. In this group, there were five hemangiomas that induced ptosis and all seven caused a lid margin change. Two hemangiomas caused proptosis and one caused globe displacement.
DISCUSSION This study is the largest investigation of periocular capillary hemangiomas published to date. Our hypothesis was that grouping hemangiomas by size and character would allow definition of those with increased potential for contributing to amblyopia. Proper identification of problematic hemangiomas is crucial because the various treatments for periocular capillary hemangioma have well-known complications.19-24 Treatment options include but are not limited to oral steroids, intralesional injection and/or top-
266 Schwartz et al ical steroids, interferon-␣, argon laser, and surgical excision.25-32 In some children prompt treatment is necessary because the hemangioma can cause amblyogenic factors that can lead to permanent vision loss.33 Therefore, our goal was to help both pediatricians and ophthalmologists more easily differentiate between those benign hemangiomas that do not cause amblyopia and those that may result in permanent visual loss if left untreated. Similar to previous studies, our study group consisted mainly of female patients.34 Only 21% of patients in groups 3 and 4 were male, while 38% of children in groups 1 and 2 were male. A review of capillary hemangiomas by Stigmar et al found that not only were females more frequently affected by hemangiomas, but they were also more severely amblyopic.12 This is consistent with the results of the current study, which shows that larger hemangiomas were more likely to be present in females and be associated with amblyopia. It is possible that hemangiomas have a higher incidence in males than reported in the literature but because of the small size of the hemangiomas they may not come to the attention of a physician. Similar to other studies the majority of our study population was Caucasian.1 Prematurity is a known risk factor for the development of periocular hemangiomas and in our study 22% of the children in this study were born prematurely.32,35 Our results show no clear relationship between prematurity and the size of the hemangioma. Our study also has a greater percentage of twin births than previously reported.1 Twin births are also associated with an increased risk of prematurity and therefore may not be an independent risk factor for the development of hemangiomas. Multiple studies have examined the relationship between capillary hemangiomas and visual loss from amblyopia. Patients with capillary hemangiomas usually present within the first 3 months of age when the visual sensory system is developing, making amblyopia a great concern. Sensory deprivation caused by occlusion of the pupil by the eyelid or hemangioma and anisometropia related to pressure of the tumor on the anterior segment of the eye are possible causes of amblyopia. Rarely, posterior pressure by intraconal masses may also cause anisometropia. Although strabismus is often present in patients with capillary hemangiomas, both Stigmar and Robb suggested that strabismus is a secondary phenomenon and not the primary cause of amblyopia associated with capillary hemangiomas.11,12 Previous studies of hemangiomas have addressed the relationship between the size of the hemangioma and the need for treatment. In his study, Robb stated that “the prevalence of asymmetrical refractive errors in patients with hemangiomas of the eyelid and orbit might be expected to vary with the size of the lesions.”11 Stigmar et al stated that hemangiomas involving only one-third of the eyelid margin are less likely to affect vision and that the risk of anisometropic amblyopia is greatest in children
Journal of AAPOS Volume 10 Number 3 June 2006
with local but bulky hemangiomas.12 However, neither study quantitatively measured the size of the hemangiomas that were examined. In this study we found that the 31 hemangiomas in group 1 that measured less that 1 cm in largest diameter were not associated with anisometropia or occlusional amblyopia. It is likely that hemangiomas of this size are not large or bulky enough to cause the degree of lid ptosis needed to cause occlusion, nor can they apply enough pressure on the globe to cause astigmatism. Thus hemangiomas of this size are unlikely to need treatment to prevent the development of amblyopia but should be monitored closely for growth. Patients in group 1 with strabismus were primarily referred for the strabismus and the hemangiomas were incidentally noted on examination. It is unlikely that the strabismus in this group was related to the presence of the hemangioma because none of these patients had amblyopia. On the other hand, patients in group 3 with large, diffuse hemangiomas had amblyopia in 78% of cases. Sixty-seven percent of hemangiomas in this group were associated with anisometropic amblyopia and 22% were associated with pupillary occlusion. Of the four patients in group 3 that did not have amblyopia by the AAO’s guidelines, three patients were treated anyway. One patient had ⫹1.25 D of astigmatism and two patients had hemangiomas that demonstrated rapid progression and further growth would have caused occlusion of the pupil. As previously described, hemangiomas associated with PHACES syndrome are often large in size and involve the eyes as well as a large part of the face.16 Because of the large size of these hemangiomas, five of the seven hemangiomas in this group caused amblyopia. Three of seven hemangiomas caused pupillary occlusion and four of seven caused anisometropic amblyopia. Of note, one hemangioma in this group caused loss of vision secondary to ischemic and/or compressive optic neuropathy. This was thought to be secondary to the massive proptosis and retro-orbital component of the tumor. While none of the small hemangiomas in group 1 caused amblyopia and three-quarters (76%) of the very large hemangiomas in groups 3 and 4 caused amblyopia, approximately half (53%) of hemangiomas in group 2 caused amblyopia. Therefore, we looked at characteristics of these group 2 hemangiomas to determine which characteristics were associated with the presence or absence of amblyopia. Our study indicates that there is a statistically significant relationship between hemangioma location and the presence of amblyopia. While 64% of hemangiomas in group 2 that were located nasally caused amblyopia, only 27% of those that were located temporally required treatment. This is likely because temporal hemangiomas are not as confined as nasal hemangiomas by the orbital rim and the nasal bridge and therefore can transmit molding pressure over a larger area of the globe’s surface, making a
Journal of AAPOS Volume 10 Number 3 June 2006
more focal compression of the cornea (ie, astigmatism) less likely. The centrally located hemangiomas included in group 2 caused amblyopia in 47% of hemangiomas. This is more frequent than amblyopia caused by temporal hemangioma but less frequent than that caused by nasal hemangiomas. Additionally, there was a statistically significant relationship between the presence of ptosis or lid margin change in measurable hemangiomas greater than 1 cm in diameter and the presence of amblyopia. Over 71% of those hemangiomas in group 2 that caused ptosis and 70% of those hemangiomas that caused a change in lid margin contour caused amblyopia. Amblyopia was present in five of seven (71%) of the hemangiomas in group 2 accompanied by strabismus. Finally, as was true for the entire study population, all hemangiomas accompanied by proptosis or globe displacement also caused amblyopia. Therefore, hemangioma characteristics in patients with measurable hemangiomas greater than 1 cm that were more likely to have amblyopia include nasal location, ptosis, lid margin change, proptosis, globe displacement, and strabismus. In conclusion, this study shows that there is a clear relationship between the size of the hemangioma and the presence or absence of amblyopia. While hemangiomas less than 1 cm in greatest diameter are extremely unlikely to cause amblyopia, large diffuse hemangiomas and those hemangiomas associated with PHACES are extremely likely to cause amblyopia. Approximately half of hemangiomas that are greater than 1 cm but are not diffuse will cause amblyopia. Those hemangiomas that are nasally located cause ptosis, lid margin change, proptosis, globe displacement, or strabismus are more likely to cause amblyopia. Therefore, while all children with capillary hemangiomas require a full ophthalmologic examination, those patients with larger hemangiomas and those associated with characteristics that are predictive of amblyopia should be monitored closely for the development of amblyopia and the need for early treatment. References 1. Haik BG, Jacobiec FA, Ellsworth RM, Jones IS. Capillary hemangiomas of the lids and orbit: an analysis of the clinical features and therapeutic results in 101 cases. Ophthalmology 1979;86:760-89. 2. Henderson JW. Orbital tumors. 2nd ed. New York: ThiemeStratton; 1980. 3. Reese AB. Tumors of the eye. 3nd ed. Hagerstown (MD): Harper and Row; 1976. 4. Holmdahl K. Cutaneous hemangiomas in premature infants. Acta Paediatr 1955;44:370-9. 5. Jacobs A, Walton R. The incidence of birthmarks in neonates. Pediatrics 1976;58:218-22. 6. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412-20. 7. Mulliken JB, Young AE. Vascular birthmarks: hemangiomas and malformations. Philadelphia (PA): Saunders; 1988. p. 1-103. 8. Haik BG, Karcioglu ZA, Gordon RA, Pechous BP. Capillary hemangioma (infantile periocular hemangioma). Surv Ophthalmol 1994; 38:399-426.
Schwartz et al
267
9. Bowers RE, Graham EA, Tomlinson KM. The natural history of the strawberry nevus. Arch Dermatol 1960;82:59-72. 10. Margileth AM, Museles M. Cutaneous hemangiomas in children. JAMA 1965;194:135-8. 11. Robb RM. Refractive errors associated with the eyelids and orbit in infancy. Am J Ophthalmol 1977;83:52-8. 12. Stigmar G, Crawford JS, Ward CM, Thomson HG. Ophthalmic sequelae of infantile hemangiomas of the eyelids and orbit. Am J Ophthalmol 1978;85:806-13. 13. Bogan S, Simon J, Krohel GB, Nelson LB. Astigmatism associated with adnexal masses in infancy. Arch Ophthalmol 1987;105:1368-70. 14. Rootman J. Diseases of the orbit. Philadelphia (P)A: JB Lippincott; 1988. 15. Ceisler EJ, Santos L, Blei F. Periocular hemangioma: what every physician should know. Pediatr Dermatol 2004;21:1-9. 16. Frieden IJ, Reese V, Cohen D. PHACE syndrome: the association of posterior fossa brain malformations, hemangiomas, arterial anomalies, coarctation of the aorta and cardiac defects, and eye abnormalities. Arch Dermatol 1996;132:307-11. 17. Kronenberg A, Blei F, Ceisler E, Steele M, Furlan L, Kodsi S. Ocular and systemic manifestations of PHACES (posterior fossa malformations, hemangiomas, arterial anomalies, cardiac defects and coarctation of the aorta, eye abnormalities, and sternal abnormalities or ventral developmental defects) syndrome. J AAPOS 2005;9:16973. 18. American Academy of Ophthalmology: Amblyopia, Preferred Practice Pattern. San Francisco (CA): American Academy of Ophthalmology, 1994. 19. Egbert JE, Paul S, Engel WK, Summers G. High injection pressure during intralesional injection of corticosteroids into capillary hemangiomas. Arch Ophthalmol 2001;119:677-83. 20. Vazquez-Botet R, Reyes BA, Vazquez-Botet M. Sclerodermiform linear atrophy after the use of intralesional steroids for periorbital hemangiomas: a review of complications. J Pediatr Ophthalmol Strabismus 1989;26:124-7. 21. Egbert JE, Nelson SC. Neurologic toxicity associated with interferon alfa treatment of capillary hemangioma. J Pediatr Ophthalmol Strabismus 1997;1:190. 22. Motwani MV, Simon JW, Pickering JD, et al. Steroid injection versus conservative treatment of anisometropia amblyopia in juvenile adnexal hemangioma. J Pediatr Ophthalmol Strabismus 1995;32: 26-8. 23. Goyal R, Watts P, Lane CM, Berk L, Gregory JW. Adrenal suppression and failure to thrive after steroid injections for periocular hemangioma. Ophthalmology 2004;11:389-95. 24. Egbert JE, Schwartz GS, Walsh AW. Diagnosis and treatment of an ophthalmic artery occlusion during an intralesional injection of corticosteroid into an eyelid capillary hemangioma. Am J Ophthalmol 1996;121:638-42. 25. Walker RS, Custer PL, Nerad JA. Surgical excision of periorbital capillary hemangiomas. Ophthalmology 1994;101:1333-40. 26. Cruz OA, Zarnegar SR, Myers SE. Treatment of periocular capillary hemangioma with topical clobetasol propionate. Ophthalmology 1995;102:2012-5. 27. Elsas FJ, Lewis AR. Topical treatment of periocular capillary hemangioma. J Pediatr Ophthalmol Strabismus 1994;31:153-6. 28. Deans RM, Harris GJ, Kivlin JD. Surgical dissection of capillary hemangiomas. Arch Ophthalmol 1992;110:1743-47. 29. Loughnan MS, Elder J, Kemp A. Treatment of massive orbitalcapillary hemangioma with interferon alfa-2b: short-term results. Arch Ophthalmol 1992;110:1366-7. 30. Kushner BJ. The treatment of periorbital infantile hemangioma with intralesional corticosteroid. Plast Reconst Surg 1985;76:517-23. 31. Summers CG, Hordinsky MD. Argon laser treatment of periocular lesions: an experimental study. Ophthalm Surg 1987;18:100-2. 32. Hastings MM, Milot J, Barsoum-Homsy M, Hershon L, Dubois L, Leclerc JM. Recombinant interferon alfa-2b in the treatment of
Journal of AAPOS Volume 10 Number 3 June 2006
268 Schwartz et al vision-threatening capillary hemangiomas in childhood. J Pediatr Ophthalmol Strabismus 1997;1:226-30. 33. Ceisler E, Blei F. Ophthalmic issues in hemangiomas of infancy. Lymphatic Res Biol 2003;1:321-9.
34. Metry DW, Herbert AA. Benign cutaneous vascular tumors of infancy: when to worry, what to do. Arch Dermatol 2000;136:905-14. 35. Amir J, Metzker A, Krikler R, Reisner SH. Strawberry hemangiomas in preterm infants. Pediatr Dermatol 1986;3:331-2.
An Eye on the Arts – The Arts on the Eye
. . . As these new sights flooded my grandparents’ brains, they jostled with images from the day before. Ellis Island, rising like a Doge’s Palace on the water. The Baggage Room stacked to the ceiling with luggage. They’d been herded up a stairway to the Registry Room. Pinned with numbers from the Giulia’s manifest, they’d filed past a line of health inspectors who’d looked in their eyes and ears, rubbed their scalps, and flipped their eyelids inside out with buttonhooks. One doctor, noticing inflammation under Dr. Philobosian’s eyelids, had stopped the examination and chalked an X on his coat. He was led out of line. My grandparents hadn’t seen him again. “He must have caught something on the boat,” Desdemona said. “Or his eyes were red from all that crying.” Meanwhile, chalk continued to do its work all around them. It marked a Pg on the belly of a pregnant woman. It scrawled an H over an old man’s failing heart. It diagnosed the C of conjunctivitis, the F of favus, and the T of trachoma. But, no matter how well trained, medical eyes couldn’t spot a recessive mutation hiding out on a fifth chromosome. Fingers couldn’t feel it. Buttonhooks couldn’t bring it to light . . . — Jeffrey Eugenides (from Middlesex, Farrar, Straus and Giroux)