- Email: [email protected]
Early screening for amblyogenic risk factors lowers the prevalence and severity of amblyopia
Major Articles Early Screening for Amblyogenic Risk Factors Lowers the Prevalence and Severity of Amblyopia Maya Eibschitz-Tsimhoni, MD,a Tatiana Friedman, MD,b Joel Naor, MD,c Naomi Eibschitz, MD,b and Zvi Friedman, MDb Purpose: To evaluate the efficacy of a mass screening program for amblyopia and amblyogenic risk factors in infants. Methods: Since 1968, children between the ages of 1 and 21⁄2 years in the city of Haifa, Israel, have been systematically screened for amblyopia and amblyogenic risk factors. The screening is performed by the Ophthalmology Department of Bnai-Zion Medical Center (formerly known as Rothchild Hospital). In 1995, we compared the prevalence and severity of amblyopia in two populations of 8-year-old children in elementary school: one group was a cohort of 808 children from the city of Haifa and its vicinity, who had been screened in infancy (between 1988 and 1990); and the second group, the control group, was a cohort of 782 children from Hadera and its vicinity, where this early screening program is not conducted. Amblyopia was defined as corrected visual acuity of ≤5/10 (20/40), or >1 line difference in corrected visual acuity between the two eyes. Referral rate, treatment rate, sensitivity, specificity, and positive predictive value and negative predictive value of the screening test in detecting factors that later resulted in the development of amblyopia were examined. Results: The prevalence of amblyopia in the 8-year-old population screened in infancy was found to be 1.0% compared with 2.6% in the 8-year-old population that had not been screened in infancy (P = .0098). The prevalence of amblyopia with visual acuity of ≤5/15 (20/60) in the amblyopic eye was 0.1% in the screened population compared with 1.7% in the non-screened population (P = .00026). In the screened infant population, 3.6% were referred from the screening examination to a confirmatory examination and 2.2% were treated. The screening examination had a sensitivity of 85.7% and a specificity of 98.6% for amblyopia. The positive predictive value of the screening examination was 62.1% and the negative predictive value was 99.6%. Conclusions: The screening program for amblyopia and amblyogenic risk factors in infants, followed by appropriate treatment, is effective in significantly reducing the prevalence and severity of amblyopia in children. (J AAPOS 2000;4:194-9)
S
creening infants to detect amblyopia and amblyogenic risk factors can detect strabismus, refractive errors, and media opacities earlier, thus facilitating prophylaxis and treatment of amblyopia. Since 1968, children between the ages of 1 and 21⁄2 years in the city of Haifa, Israel, have been systematically screened by the
Bnai-Zion Ophthalmology Department for amblyopia and amblyogenic risk factors. To determine the impact of early screening on the subsequent prevalence of amblyopia, we compared the prevalence of amblyopia in an 8-year-old population from Haifa with an 8-year-old population from Hadera, where systematic screening does not take place.
SUBJECTS AND METHODS From the Department of Ophthalmology, Kellogg Eye Center, University of Michigan, Ann Arbora; Department of Ophthalmology, Bnai-Zion Medical Center, Haifa, Israelb; and Department of Ophthalmology, Toronto Hospital, Toronto, Ontario, Canada.c Supported in part by grant 1815791 from the Technion–Israel Institute of Technology, Haifa, Israel. Presented as a poster at the 25th Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Toronto, Ontario, Canada, April 15-18, 1999. (Screening for amblyopia and amblyogenic conditions in infancy: a longitudinal study.) Submitted August 19, 1999. Revision accepted November 19, 1999. Reprint requests: Maya Eibschitz, MD, The University of Michigan, Kellogg Eye Center, 1000 Wall St, Ann Arbor, MI 48105 (e-mail: [email protected]). Copyright © 2000 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2000 $12.00 + 0 75/1/105274 doi:10.1067/mpa.2000.105274
194
August 2000
In Israel, the Ministry of Health, through its child welfare clinics, provides infants with free health care including examinations and immunizations. All children born in Israel are registered at the child welfare clinic nearest their home and attend the clinic periodically for examinations, immunizations, and general supervision. As of 1968, an ophthalmic screening examination for amblyopia and amblyogenic risk factors, in infants between the ages of 1 and 21⁄2 years, has been included as part of the routine examination at the child welfare clinics in the Haifa area. There are 55 child welfare clinics in Haifa and its vicinity. A total of approximately 9000 children are born in the city Journal of AAPOS
Journal of AAPOS Volume 4 Number 4 August 2000
Eibschitz-Tsimhoni et al
195
FIG 1. Longitudinal follow-up for all patients enrolled in study. Diagram shows distribution of screened and control populations according to screening test, confirmatory examination, and examination at 8 years of age. FN, False negative; FP, false positive; TN, true negative; TP, true positive.
of Haifa every year. A special form is sent, in advance, to the parents, indicating the importance of the screening examination, with general information on amblyopia. Ophthalmologic examinations are conducted by an ophthalmologist or an orthoptist trained in retinoscopy. The examination starts with a short history of the child and his or her family. The examination then continues with an external inspection followed by the Hirschberg corneal reflex test, monocular fixation-and-following test, ductions and versions examination, cover-uncover test, and alternate cover test. Since 1974, retinoscopy without cycloplegia (“rapid retinoscopy”), with use of only a +3.00 trial lens, was added to the screening test. By using rapid retinoscopy, examiners can detect children with significant ametropia, as well as children with opacities in the ocular media. The rapidity of the retinoscopy enables the ophthalmologist/orthoptist to perform the examination while avoiding fixation and accommodation by the children. The entire examination, including the retinoscopy, does not exceed 2 to 3 minutes per child. Each examination is recorded in the child’s medical file at the child welfare clinic, as well as in the child’s immunization book, which is kept by the parents. Children with a borderline amblyogenic condition are reexamined at a later date at the child welfare clinic. All children with strabismus, opacities in the ocular media, or suspected amblyogenic ametropia are referred to the eye clinic at the Bnai-Zion Medical Center, Haifa, for a complete examination including cycloplegic retinoscopy and fundoscopy. Treatment is instituted for
the confirmed amblyogenic conditions. Treatment and follow-up are continued until best visual acuity is achieved. In 1995, we compared two populations of regular elementary schoolchildren, aged 8 years, for the prevalence and severity of amblyopia. For the purpose of this study, amblyopia was defined as corrected visual acuity of ≤5/10 (20/40), or >1 line difference in corrected visual acuity between both eyes. Corrected Snellen visual acuity was measured with the child using his/her glasses, pinhole, or correction determined by retinoscopy. Excluded from the study were new immigrants and children who moved from another city. Twenty-nine schools (16 schools in Haifa and 13 schools in Hadera) were randomly selected by the Israeli Ministry of Education for participation in each community. All selected school administrators received information regarding the study and agreed to participate. Within selected schools, all children and their parents agreed to participate. The first group, the study group, was a cohort of 808 children from Haifa and its vicinity who had been screened (and treated, when necessary) in infancy between 1988 and 1990. Following the examination at the schools, each child’s immunization book was reviewed for the screening results. In patients who had abnormal findings at infancy and who were referred to our clinic, the medical records were also reviewed. Haifa is a city with a population of approximately 200,000, with about 9000 births per year.
196 Eibschitz-Tsimhoni et al
Journal of AAPOS Volume 4 Number 4 August 2000
FIG 2. Comparison of visual acuity in 8-year-old children with amblyopia (or treated for amblyopia) in the study group and visual acuity of children with amblyopia in the control group. Three categories are identified: non-screened, screened and treated, and screened and missed.
The second group, the control group, was a cohort of 782 children from Hadera and its vicinity, where an early screening program is not conducted. Hadera is a city with a population of approximately 100,000, with about 4000 births per year. Hadera is located 25 miles from Haifa. There is similar availability of pediatricians and ophthalmologists in both cities. Hadera’s community is similar to the community in Haifa in ethnic/racial mix, social classes, health care facilities, education, nutrition, and climate. This similarity enables group comparability. Examinations of schoolchildren from Haifa and Hadera were performed by the authors. The project was approved by the Israeli Ministry of Health. Statistical analysis was performed by the χ2 test and the Student t test. We used the test of equality of proportions by taking into account the finite population, adding the correction factor (1-n/N) for the variance. Sensitivity, specificity, and positive and negative predictive values of the screening examination to detect amblyopia were calculated.
RESULTS At the selected schools in Haifa, 988 children were examined. Of those examined, 808 had attended the screening examination as infants. At the time of the infant screening, amblyopia or risk factors for amblyopia were suspected in 29 of 808 (3.6%) infants. All 29 infants were then referred for a more thorough confirmatory evaluation at the outpatient Ophthalmology Clinic of the BnaiZion Medical Center. Of the 29 infants referred for further evaluation, 18 (2.2% of the screened population) were found to have an amblyogenic risk factor or amblyopia. These infants were treated and followed. In the remaining 11 infants, findings were not thought to be amblyogenic as determined by the full ophthalmic examination (Figure 1). This group of 808 children were reexamined by us when they were 8 years old. As summarized in Figures 1 and 2, 776 of 779 (99.6%) children, in whom amblyopia or
risk factors for amblyopia had not been found during infancy at the child welfare clinics, did not have amblyopia at the age of 8 years. Three were found to have amblyopia. Their corrected visual acuities were: 5/12 (20/48), 5/10 (20/40), and 5/9 (20/36) in the amblyopic eye. None of the 11 children, in whom findings from a full infant ophthalmic examination did not indicate amblyopia, were found to have amblyopia at 8 years of age. Of the 18 children who were treated for amblyogenic conditions as infants, 13 did not have amblyogenic loss of visual acuity and 5 had amblyogenic visual loss at 8 years of age. The corrected visual acuities of the 5 children with amblyopia were: 5/15 (20/60), 5/12 (20/48), 5/12 (20/48), 5/10 (20/40), and 5/9 (20/36) in the amblyopic eye. A total of 8 of 808 (1.0%) children with amblyopia were identified. The corrected visual acuities of 782 children, aged 8 years, from Hadera and its vicinity were measured in the same way as for the 808 children, aged 8 years, from Haifa. This group of 782 children were not screened in infancy and served as a control group. At 8 years of age, 20 of 782 (2.6%) children had amblyopia and 762 did not have amblyopia. Of the 20 children with amblyopia in this group, two had corrected visual acuities of 5/36 (20/144) in the amblyopic eye, four had 5/24 (20/96), four had 5/18 (20/72), three had 5/15 (20/60), two had 5/12 (20/48), four had 5/10 (20/40), and one had 5/9 (20/36). A comparison of visual acuity at the age of 8 years in children with amblyopia (or treated for amblyopia) in the study group and children with amblyopia in the control group is presented in Figure 2. When we compared the prevalence of amblyopia in the two populations of 8-year-old schoolchildren, we found it to be lower in the screened population than in the nonscreened population; 8 of 808 (1.0%) versus 20 of 782 (2.6%), P = .0098. The prevalence of more severe amblyopia with visual acuity of ≤5/15 (20/60) in the amblyopic eye was 1 of 808 (0.1%) in the screened population compared with 13 of 782 (1.7%) in the non-screened population, P = .00026.
Journal of AAPOS Volume 4 Number 4 August 2000
The screening examination for amblyopia had a sensitivity of 85.7%, as calculated from the prevalence of amblyopia in the 8-year-old study population. The specificity was 98.6%. The positive predictive value (PPV) was 62.1%, and the negative predictive value (NPV) was 99.6% (Figure 1). The number of children in each of the amblyopic subgroups (by cause for amblyopia) was too small to allow for subgroup analysis. In the control group, 15 children had refractive amblyopia, 4 children had strabismic amblyopia, and 1 child had occlusion amblyopia (ptosis). In the screened and treated group, 3 children had refractive amblyopia and 2 children had strabismic amblyopia. In the screened and missed group (children in whom amblyopia or risk factors for amblyopia had not been found during infancy but were found at the age of 8 years), 2 children had refractive amblyopia and 1 child had strabismic amblyopia.
DISCUSSION Many cases of amblyopia escape detection without effective screening. All lines of evidence support the concept that there is a limited period of susceptibility (“sensitive period”) for the developing visual system.1-8 The plasticity and responsiveness of the developing visual system in infants makes this age group ideal for ophthalmologic screening. Early identification and treatment of risk factors for amblyopia enables prevention of amblyopia. Diagnosis and treatment of amblyopia at an earlier age may lead to a better and more stable final visual result, with shorter treatment times, more rapid improvement in visual acuity, and better overall compliance with treatment regimens. Screening infants for risk factors for amblyopia to enable treatment was suggested by the eye department of the Bnai-Zion Medical Center in 1968, and has been performed ever since.9,10 In 1980, Friedman et al11 published the frequency of strabismic amblyopia and anisometropic amblyopia in 38,000 infants, screened between the ages of 1 and 21⁄2 years. Based on the frequency found, the authors advocated the importance of ophthalmic screening in infancy.11 In 1987, Neumann et al12 published a report assessing the results of preventive treatment for amblyopia in cases of constant unilateral esotropia. This study clearly showed that starting treatment at or before 24 months of age markedly increases the probability of mitigating the severity of or even eliminating the amblyopia. Starting treatment before the age of 18 months provides the best long-term stability of the achieved visual acuity. In 1985, Friedman et al13 published a report assessing the impact of early treatment in cases of marked ametropia discovered by rapid retinoscopy during screening between the ages of 1 and 21⁄2 years and confirmed by cycloplegic examination.13 Their study demonstrated that significant ametropia, without strabismus, carries a high risk for the development of amblyopia even when treated from a relatively early age.
Eibschitz-Tsimhoni et al
197
They suggested that without early screening and treatment, both the prevalence and severity of amblyopia in this group would have been higher. The psychologic and social benefits of early diagnosis and treatment are discussed elsewhere.14 Sondhi, Archer, and Helveston15 found that exodeviations are frequently seen up to the age of 6 months and that esodeviations are occasionally seen in infants who do not develop congenital esotropia, but not after 2 months of age. Any strabismus persisting after these ages should be considered abnormal and receive ophthalmologic evaluation.15 A study by Hopkisson et al16 demonstrated that high hypermetropia is common in newborns and astigmatism is common at 6 months of age. They found that the frequency of both conditions is reduced markedly at 1 year of age.16 We set the earliest age for screening at 1 year of age. Friedman et al11 have previously reported rapid retinoscopy as a practical method for diagnosing marked ametropia with or without strabismus during the screening of children between the ages of 1 and 21⁄2 years.11 Refractive amblyopia without strabismus is relatively frequent and accounts for 33% to 75% of all cases of amblyopia. An attempt was made to detect these cases during mass screening by rapid retinoscopy without cycloplegia. Our study was designed to determine the long-term effects and proficiency of the early screening program described. Since the ideal way to evaluate a screening test for amblyopia would be to verify the results with the actual presence of amblyopia, we used a confirmatory test of amblyopia at the age of 8 years on a sample of the same population that was screened in infancy. A statistically significant difference in the prevalence of amblyopia was found between the group screened in infancy and the control group (1.0% vs 2.6%, P = .0098), showing that prevention by early detection is effective. The prevalence of severe amblyopia was also found to be significantly reduced in the screened population. The number of infants referred for a confirmatory examination was small (29 infants, 3.6% of the screened population), demonstrating the feasibility of the screening program. The percentage of children requiring treatment was 2.2%. Sensitivity refers to the ability of a screening test to correctly identify the presence of disease (ie, the true positive results/ true positive + false negative [TP/TP+FN]). By knowing the false-negative rate for amblyopia, we could examine the sensitivity of the screening test to detect amblyopia. The sensitivity of this screening examination to detect amblyopia was 85.7%. The false-negative results were determined by the number of amblyopes in the screened 8-year-old population that had not been detected earlier. There were 3 patients with false-negative results in our study population (0.4%) and all 3 had only “mild” amblyopia. These patients with amblyopia may have been missed in spite of having one of the amblyogenic risk factors present during screening, or they may have developed the amblyogenic factors at a later date.
198 Eibschitz-Tsimhoni et al Specificity refers to the ability of a screening test to correctly identify the absence of disease (ie, true negative results/true negative + false positive [TN/TN+FP]). Again, by knowing the number of true-negative cases from studying the 8-year-old group for amblyopia, we could determine the specificity of the screening test for amblyopia. This screening program was found to have a specificity of 98.6%. In a condition like amblyopia, in which the prevalence of true-negative results is high, a very high specificity value is essential to make the screening examination efficient. Patients with false-positive results include those who accommodated on noncycloplegic retinoscopy and, thus did not have the suspected myopia in the confirmatory examination, and those who were suspected of having significant ametropia and found to have only mild ametropia by the confirmatory examination. Also, all strabismus cases, whether constant unilateral, alternating, intermittent, vertical, exotropic, or esotropic, are referred for further evaluation. Some of these patients have to be followed up or treated but are not amblyogenic. Positive predictive value (PPV) refers to the proportion of the population that screens positive and is found to have disease on diagnostic testing (ie, true positive results/true positive + false positive [TP/TP+FP]. The PPV has a high significance in a screening examination evaluation. The higher the PPV, the fewer children will be referred from the screening examination to a thorough time- and money-consuming examination in spite of being “healthy.” The PPV of our screening examination was 62.1%. The PPV must always be considered along with the actual percentage of referred children. Negative predictive value (NPV) refers to the proportion of a population that screens negative for a disease and has this confirmed by diagnostic testing (ie, true negative results/true negative + false negative [TN/TN+FN]). This is a very important value in the evaluation of a screening test because subjects who screen negative are discharged from further ophthalmologic surveillance. Our NPV was 99.6%, and only 3 of 779 children (0.4%) were found to have amblyopia in spite of passing the screening examination. All 3 children had only “mild” amblyopia. Screening programs for amblyopia have been previously reported.17-26 Many screening programs for decreased visual acuity in preschool children report referral rates in the range of 10% to 23% and treatment rates as high as 10% to 20% in spite of a prevalence of amblyopia of less than 5%.16,18-20 Others report low compliance with the screening and with further follow-up and treatment.26 The photoscreeners used to detect amblyopia and amblyogenic risk factors in infancy and childhood also show a very high referral rate (>20%), a high false-positive rate (up to 10%), and a high false-negative rate (>3%).27,28 These studies were designed to evaluate the ability to detect “amblyogenic factors” as determined by the authors and thus are not comparable to our study, which actually determines the prevalence of amblyopia in a previously screened population.
Journal of AAPOS Volume 4 Number 4 August 2000
Our examination technique has many advantages. The examination is objective, simple to perform, and welltolerated by the tested infants and their parents. It enables the examiners to acquire all the information needed in an average of 2 to 3 minutes and does not require further interpretation. The examination can be performed either by an ophthalmologist or an orthoptist. The only equipment needed for the examination is a retinoscope, a small target, a penlight, and a +3.00 lens. Access to care and follow-up and parental education are integral and crucial components of any vision screening program, as emphasized by Lichter.19 The problem of low compliance was reviewed in an inner city population.24 In conclusion, our rationale for screening infants for amblyopia is that early discovery and treatment provide better and more stable results of visual acuity and possibly of other visual functions known to be affected in amblyopia.1,29 Our method is very simple, quick, easy to perform, and well-tolerated by the tested children and their parents. Our results demonstrate that this program lowers the prevalence of amblyopia and, in particular, severe amblyopia. The prevalence of amblyopia in the population remains between 1% and 5% and is the leading cause of unilateral visual deficit among young adults.30,31 A child with amblyogenic risk factors that has been discovered and treated early can look forward to more than half a century of school and employment that demand performance of visually mediated tasks. Therefore, the most important benefit from screening is the reduction of preventable visual loss. We thank Monte A. Del Monte, MD, and Steven M. Archer, MD, for invaluable help in writing this article and for reviewing the statistical analysis of the patient data. References 1. Greenwald MJ, Parks MM. Amblyopia. In: Duane TD, Jaeger EA, editors. Clinical ophthalmology. Vol 1. Philadelphia (PA): Harper & Row; 1994. p. 1-21. 2. LeVay S, Wiesel TN, Hubel DH. The development of ocular dominance columns in normal and visually deprived monkeys. J Comp Neurol 1980;191:1-51. 3. Harwerth RS, Smith EL III, Duncan GC, Crawford ML, von Noorden GK. Multiple sensitive periods in the development of the primate visual system. Science 1987;232:235-8. 4. Worth CA. Squint: its causes, pathology and treatment. Philadelphia (PA): Blakiston; 1903. 5. Von Noorden GK. Mechanisms of amblyopia. Adv Ophthalmol 1977;34:92-115. 6. Hubel DH, Wiesel TN. The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol (Lond) 1970;206:419-36. 7. Wiesel TN. Postnatal development of the visual cortex and the influence of environment. Nature 1982;299:583-91. 8. Wiesel TN, Hubel DH. Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 1963;26:1003-17. 9. Neumann E, Eibschitz N, Hyams SW, Friedman Z. Ophthalmic screening in child welfare clinics in Israel with particular reference to strabismus and amblyopia. J Pediatr Ophthalmol Strabismus 1971;8: 257-60.
Journal of AAPOS Volume 4 Number 4 August 2000
Eibschitz-Tsimhoni et al
10. Friedman Z, Neumann E, Peleg B. Screening for strabismus and amblyopia in child welfare clinics. Metab Ophthalmol 1978;2:109. 11. Friedman Z, Neumann E, Hyams SW, Peleg B. Ophthalmic screening of 38,000 children, age 1 to 2.5 years, in child welfare clinics. J Pediatr Ophthalmol Strabismus 1980;17:261-7. 12. Neumann E, Friedman Z, Abel-Peleg B. Prevention of strabismic amblyopia of early onset with special reference to the optimal age for screening. J Pediatr Ophthalmol Strabismus 1987;24:106-10. 13. Friedman Z, Neumann E, Abel-Peleg B. Outcome of treatment of marked ametropia without strabismus following screening and diagnosis before the age of three. J Pediatr Ophthalmol Strabismus 1985; 22:54-7. 14. Friendly DS. Preschool visual acuity screening tests. Trans Am Ophthalmol Soc 1978;76:383-480. 15. Sondhi N, Archer SM, Helveston EM. Development of normal ocular alignment. J Pediatr Ophthalmol Strabismus 1988;25:210-1. 16. Hopkisson B, Arnold P, Billingham B, McGarrigle M, Shribman S. Can retinoscopy be used to screen infants for amblyopia? A longitudinal study of refraction in the first year of life. Eye 1992;6:607-9. 17. De Becker I, McPherson HJ, La Roche GR, Braunstein J, Cottle R, McIntyre LL, et al. Negative predictive value of populationbased preschool vision screening program. Ophthalmology 1992;99: 998-1003. 18. Preslan MW, Novak A. Baltimore vision screening project. Ophthalmology 1996;103:105-9. 19. Lichter PR. Vision screening and children’s access to eye care [editorial]. Ophthalmology 1992;99:843-4. 20. Kohler L, Stigmar G. Vision screening of four-year-old children. Acta Paediatr Scand 1973;62:17-27.
199
21. Kohler L, Stigmar G. Visual disorders in 7-year-old children with and without previous vision screening. Acta Paediatr Scand 1978;67:373-7. 22. Landenvall G. Visual assessment of school children seven years old. Findings at screening. Analysis of treated children. Acta Ophthalmol Scand 1992;70:412. 23. Yazawa K, Suga J, Wakita S, Sumitomo M, Vemura Y. The Tokyo metropolitan home vision screening program for amblyopia in 3year-old children. Am J Ophthalmol 1992;114:416-9. 24. Nordlow W, Joachimsson S. A screening test for visual acuity in four year old children. Acta Ophthalmol Scand 1962;40:453-62. 25. Lennerstrand G, Jakobsson P, Kvarnstrom G. Screening for ocular dysfunction in children: approaching a common program. Acta Ophthalmol Scand 1995;214:26-38. 26. Williamson TH, Andrews R, Dutton GN, Murray G, Graham N. Assessment of an inner city visual screening program for preschool children. Br J Ophthalmol 1995;79:1068-73. 27. Ottar WL, Scott WE, Hogado SI. Photoscreening for amblyogenic factors. J Pediatr Ophthalmol Strabismus 1995;32:289-95. 28. Schworm HD, Kau C, Reindl B, Horstmann S, Sollinger B, Boergen KP. Photoscreening for early detection of amblyogenic eye changes. Klin Monatsbl Augenheilkd 1997;210:158-64. 29. Levi DM. Visual acuity in strabismic and anisometropic amblyopia. Ophthalmol Clin North Am 1990;3:289-301. 30. Leske MC, Hawkins B. Screening relationship to diagnosis and therapy. In: Duane TD, Jaeger EA, editors. Clinical ophthalmology. Vol 5. Philadelphia (PA): Harper & Row; 1994. p. 1-19. 31. Krueger DE, Ederer R. Report on the National Eye Institute’s visual acuity impairment survey pilot study. Bethesda (MD): Office of Biometry and Epidemiology, NEI, NIH, PMS, DMMS, 1984.
An Eye on the Arts – The Arts on the Eye Bella had inherited her father’s lazy eye. His gaze veered shiftily to the right as if something very lewd was going on behind other people’s backs. As a child, Bella had thought of it as the evil eye, all-seeing, masterful. Until, that is, this congenital weakness of her father’s emerged in her…She had to wear a pair of glasses with the right lens patched over with sticking plaster. It gave her a lopsided, partial view of the world—a huge, pinkish blur before her and a sensation of an obstruction looming ahead which was never encountered yet never went away. She sometimes feared that this flesh-coloured wall that met her gaze was, in fact, a fresh layer of skin growing over the unused eye. Frequently she had to take the glasses off to reassure herself that her wall-up eye was still there, and still worked. “Is it sore?” her schoolfriends asked about the eclipsed eye, associating sticking plaster with pain. Some were convinced that Bella did not have a second eye; others that she had been badly scarred and it was too unsightly to expose. —Mary Morrissy (from A Lazy Eye p 54)
S
creening infants to detect amblyopia and amblyogenic risk factors can detect strabismus, refractive errors, and media opacities earlier, thus facilitating prophylaxis and treatment of amblyopia. Since 1968, children between the ages of 1 and 21⁄2 years in the city of Haifa, Israel, have been systematically screened by the
Bnai-Zion Ophthalmology Department for amblyopia and amblyogenic risk factors. To determine the impact of early screening on the subsequent prevalence of amblyopia, we compared the prevalence of amblyopia in an 8-year-old population from Haifa with an 8-year-old population from Hadera, where systematic screening does not take place.
SUBJECTS AND METHODS From the Department of Ophthalmology, Kellogg Eye Center, University of Michigan, Ann Arbora; Department of Ophthalmology, Bnai-Zion Medical Center, Haifa, Israelb; and Department of Ophthalmology, Toronto Hospital, Toronto, Ontario, Canada.c Supported in part by grant 1815791 from the Technion–Israel Institute of Technology, Haifa, Israel. Presented as a poster at the 25th Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Toronto, Ontario, Canada, April 15-18, 1999. (Screening for amblyopia and amblyogenic conditions in infancy: a longitudinal study.) Submitted August 19, 1999. Revision accepted November 19, 1999. Reprint requests: Maya Eibschitz, MD, The University of Michigan, Kellogg Eye Center, 1000 Wall St, Ann Arbor, MI 48105 (e-mail: [email protected]). Copyright © 2000 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2000 $12.00 + 0 75/1/105274 doi:10.1067/mpa.2000.105274
194
August 2000
In Israel, the Ministry of Health, through its child welfare clinics, provides infants with free health care including examinations and immunizations. All children born in Israel are registered at the child welfare clinic nearest their home and attend the clinic periodically for examinations, immunizations, and general supervision. As of 1968, an ophthalmic screening examination for amblyopia and amblyogenic risk factors, in infants between the ages of 1 and 21⁄2 years, has been included as part of the routine examination at the child welfare clinics in the Haifa area. There are 55 child welfare clinics in Haifa and its vicinity. A total of approximately 9000 children are born in the city Journal of AAPOS
Journal of AAPOS Volume 4 Number 4 August 2000
Eibschitz-Tsimhoni et al
195
FIG 1. Longitudinal follow-up for all patients enrolled in study. Diagram shows distribution of screened and control populations according to screening test, confirmatory examination, and examination at 8 years of age. FN, False negative; FP, false positive; TN, true negative; TP, true positive.
of Haifa every year. A special form is sent, in advance, to the parents, indicating the importance of the screening examination, with general information on amblyopia. Ophthalmologic examinations are conducted by an ophthalmologist or an orthoptist trained in retinoscopy. The examination starts with a short history of the child and his or her family. The examination then continues with an external inspection followed by the Hirschberg corneal reflex test, monocular fixation-and-following test, ductions and versions examination, cover-uncover test, and alternate cover test. Since 1974, retinoscopy without cycloplegia (“rapid retinoscopy”), with use of only a +3.00 trial lens, was added to the screening test. By using rapid retinoscopy, examiners can detect children with significant ametropia, as well as children with opacities in the ocular media. The rapidity of the retinoscopy enables the ophthalmologist/orthoptist to perform the examination while avoiding fixation and accommodation by the children. The entire examination, including the retinoscopy, does not exceed 2 to 3 minutes per child. Each examination is recorded in the child’s medical file at the child welfare clinic, as well as in the child’s immunization book, which is kept by the parents. Children with a borderline amblyogenic condition are reexamined at a later date at the child welfare clinic. All children with strabismus, opacities in the ocular media, or suspected amblyogenic ametropia are referred to the eye clinic at the Bnai-Zion Medical Center, Haifa, for a complete examination including cycloplegic retinoscopy and fundoscopy. Treatment is instituted for
the confirmed amblyogenic conditions. Treatment and follow-up are continued until best visual acuity is achieved. In 1995, we compared two populations of regular elementary schoolchildren, aged 8 years, for the prevalence and severity of amblyopia. For the purpose of this study, amblyopia was defined as corrected visual acuity of ≤5/10 (20/40), or >1 line difference in corrected visual acuity between both eyes. Corrected Snellen visual acuity was measured with the child using his/her glasses, pinhole, or correction determined by retinoscopy. Excluded from the study were new immigrants and children who moved from another city. Twenty-nine schools (16 schools in Haifa and 13 schools in Hadera) were randomly selected by the Israeli Ministry of Education for participation in each community. All selected school administrators received information regarding the study and agreed to participate. Within selected schools, all children and their parents agreed to participate. The first group, the study group, was a cohort of 808 children from Haifa and its vicinity who had been screened (and treated, when necessary) in infancy between 1988 and 1990. Following the examination at the schools, each child’s immunization book was reviewed for the screening results. In patients who had abnormal findings at infancy and who were referred to our clinic, the medical records were also reviewed. Haifa is a city with a population of approximately 200,000, with about 9000 births per year.
196 Eibschitz-Tsimhoni et al
Journal of AAPOS Volume 4 Number 4 August 2000
FIG 2. Comparison of visual acuity in 8-year-old children with amblyopia (or treated for amblyopia) in the study group and visual acuity of children with amblyopia in the control group. Three categories are identified: non-screened, screened and treated, and screened and missed.
The second group, the control group, was a cohort of 782 children from Hadera and its vicinity, where an early screening program is not conducted. Hadera is a city with a population of approximately 100,000, with about 4000 births per year. Hadera is located 25 miles from Haifa. There is similar availability of pediatricians and ophthalmologists in both cities. Hadera’s community is similar to the community in Haifa in ethnic/racial mix, social classes, health care facilities, education, nutrition, and climate. This similarity enables group comparability. Examinations of schoolchildren from Haifa and Hadera were performed by the authors. The project was approved by the Israeli Ministry of Health. Statistical analysis was performed by the χ2 test and the Student t test. We used the test of equality of proportions by taking into account the finite population, adding the correction factor (1-n/N) for the variance. Sensitivity, specificity, and positive and negative predictive values of the screening examination to detect amblyopia were calculated.
RESULTS At the selected schools in Haifa, 988 children were examined. Of those examined, 808 had attended the screening examination as infants. At the time of the infant screening, amblyopia or risk factors for amblyopia were suspected in 29 of 808 (3.6%) infants. All 29 infants were then referred for a more thorough confirmatory evaluation at the outpatient Ophthalmology Clinic of the BnaiZion Medical Center. Of the 29 infants referred for further evaluation, 18 (2.2% of the screened population) were found to have an amblyogenic risk factor or amblyopia. These infants were treated and followed. In the remaining 11 infants, findings were not thought to be amblyogenic as determined by the full ophthalmic examination (Figure 1). This group of 808 children were reexamined by us when they were 8 years old. As summarized in Figures 1 and 2, 776 of 779 (99.6%) children, in whom amblyopia or
risk factors for amblyopia had not been found during infancy at the child welfare clinics, did not have amblyopia at the age of 8 years. Three were found to have amblyopia. Their corrected visual acuities were: 5/12 (20/48), 5/10 (20/40), and 5/9 (20/36) in the amblyopic eye. None of the 11 children, in whom findings from a full infant ophthalmic examination did not indicate amblyopia, were found to have amblyopia at 8 years of age. Of the 18 children who were treated for amblyogenic conditions as infants, 13 did not have amblyogenic loss of visual acuity and 5 had amblyogenic visual loss at 8 years of age. The corrected visual acuities of the 5 children with amblyopia were: 5/15 (20/60), 5/12 (20/48), 5/12 (20/48), 5/10 (20/40), and 5/9 (20/36) in the amblyopic eye. A total of 8 of 808 (1.0%) children with amblyopia were identified. The corrected visual acuities of 782 children, aged 8 years, from Hadera and its vicinity were measured in the same way as for the 808 children, aged 8 years, from Haifa. This group of 782 children were not screened in infancy and served as a control group. At 8 years of age, 20 of 782 (2.6%) children had amblyopia and 762 did not have amblyopia. Of the 20 children with amblyopia in this group, two had corrected visual acuities of 5/36 (20/144) in the amblyopic eye, four had 5/24 (20/96), four had 5/18 (20/72), three had 5/15 (20/60), two had 5/12 (20/48), four had 5/10 (20/40), and one had 5/9 (20/36). A comparison of visual acuity at the age of 8 years in children with amblyopia (or treated for amblyopia) in the study group and children with amblyopia in the control group is presented in Figure 2. When we compared the prevalence of amblyopia in the two populations of 8-year-old schoolchildren, we found it to be lower in the screened population than in the nonscreened population; 8 of 808 (1.0%) versus 20 of 782 (2.6%), P = .0098. The prevalence of more severe amblyopia with visual acuity of ≤5/15 (20/60) in the amblyopic eye was 1 of 808 (0.1%) in the screened population compared with 13 of 782 (1.7%) in the non-screened population, P = .00026.
Journal of AAPOS Volume 4 Number 4 August 2000
The screening examination for amblyopia had a sensitivity of 85.7%, as calculated from the prevalence of amblyopia in the 8-year-old study population. The specificity was 98.6%. The positive predictive value (PPV) was 62.1%, and the negative predictive value (NPV) was 99.6% (Figure 1). The number of children in each of the amblyopic subgroups (by cause for amblyopia) was too small to allow for subgroup analysis. In the control group, 15 children had refractive amblyopia, 4 children had strabismic amblyopia, and 1 child had occlusion amblyopia (ptosis). In the screened and treated group, 3 children had refractive amblyopia and 2 children had strabismic amblyopia. In the screened and missed group (children in whom amblyopia or risk factors for amblyopia had not been found during infancy but were found at the age of 8 years), 2 children had refractive amblyopia and 1 child had strabismic amblyopia.
DISCUSSION Many cases of amblyopia escape detection without effective screening. All lines of evidence support the concept that there is a limited period of susceptibility (“sensitive period”) for the developing visual system.1-8 The plasticity and responsiveness of the developing visual system in infants makes this age group ideal for ophthalmologic screening. Early identification and treatment of risk factors for amblyopia enables prevention of amblyopia. Diagnosis and treatment of amblyopia at an earlier age may lead to a better and more stable final visual result, with shorter treatment times, more rapid improvement in visual acuity, and better overall compliance with treatment regimens. Screening infants for risk factors for amblyopia to enable treatment was suggested by the eye department of the Bnai-Zion Medical Center in 1968, and has been performed ever since.9,10 In 1980, Friedman et al11 published the frequency of strabismic amblyopia and anisometropic amblyopia in 38,000 infants, screened between the ages of 1 and 21⁄2 years. Based on the frequency found, the authors advocated the importance of ophthalmic screening in infancy.11 In 1987, Neumann et al12 published a report assessing the results of preventive treatment for amblyopia in cases of constant unilateral esotropia. This study clearly showed that starting treatment at or before 24 months of age markedly increases the probability of mitigating the severity of or even eliminating the amblyopia. Starting treatment before the age of 18 months provides the best long-term stability of the achieved visual acuity. In 1985, Friedman et al13 published a report assessing the impact of early treatment in cases of marked ametropia discovered by rapid retinoscopy during screening between the ages of 1 and 21⁄2 years and confirmed by cycloplegic examination.13 Their study demonstrated that significant ametropia, without strabismus, carries a high risk for the development of amblyopia even when treated from a relatively early age.
Eibschitz-Tsimhoni et al
197
They suggested that without early screening and treatment, both the prevalence and severity of amblyopia in this group would have been higher. The psychologic and social benefits of early diagnosis and treatment are discussed elsewhere.14 Sondhi, Archer, and Helveston15 found that exodeviations are frequently seen up to the age of 6 months and that esodeviations are occasionally seen in infants who do not develop congenital esotropia, but not after 2 months of age. Any strabismus persisting after these ages should be considered abnormal and receive ophthalmologic evaluation.15 A study by Hopkisson et al16 demonstrated that high hypermetropia is common in newborns and astigmatism is common at 6 months of age. They found that the frequency of both conditions is reduced markedly at 1 year of age.16 We set the earliest age for screening at 1 year of age. Friedman et al11 have previously reported rapid retinoscopy as a practical method for diagnosing marked ametropia with or without strabismus during the screening of children between the ages of 1 and 21⁄2 years.11 Refractive amblyopia without strabismus is relatively frequent and accounts for 33% to 75% of all cases of amblyopia. An attempt was made to detect these cases during mass screening by rapid retinoscopy without cycloplegia. Our study was designed to determine the long-term effects and proficiency of the early screening program described. Since the ideal way to evaluate a screening test for amblyopia would be to verify the results with the actual presence of amblyopia, we used a confirmatory test of amblyopia at the age of 8 years on a sample of the same population that was screened in infancy. A statistically significant difference in the prevalence of amblyopia was found between the group screened in infancy and the control group (1.0% vs 2.6%, P = .0098), showing that prevention by early detection is effective. The prevalence of severe amblyopia was also found to be significantly reduced in the screened population. The number of infants referred for a confirmatory examination was small (29 infants, 3.6% of the screened population), demonstrating the feasibility of the screening program. The percentage of children requiring treatment was 2.2%. Sensitivity refers to the ability of a screening test to correctly identify the presence of disease (ie, the true positive results/ true positive + false negative [TP/TP+FN]). By knowing the false-negative rate for amblyopia, we could examine the sensitivity of the screening test to detect amblyopia. The sensitivity of this screening examination to detect amblyopia was 85.7%. The false-negative results were determined by the number of amblyopes in the screened 8-year-old population that had not been detected earlier. There were 3 patients with false-negative results in our study population (0.4%) and all 3 had only “mild” amblyopia. These patients with amblyopia may have been missed in spite of having one of the amblyogenic risk factors present during screening, or they may have developed the amblyogenic factors at a later date.
198 Eibschitz-Tsimhoni et al Specificity refers to the ability of a screening test to correctly identify the absence of disease (ie, true negative results/true negative + false positive [TN/TN+FP]). Again, by knowing the number of true-negative cases from studying the 8-year-old group for amblyopia, we could determine the specificity of the screening test for amblyopia. This screening program was found to have a specificity of 98.6%. In a condition like amblyopia, in which the prevalence of true-negative results is high, a very high specificity value is essential to make the screening examination efficient. Patients with false-positive results include those who accommodated on noncycloplegic retinoscopy and, thus did not have the suspected myopia in the confirmatory examination, and those who were suspected of having significant ametropia and found to have only mild ametropia by the confirmatory examination. Also, all strabismus cases, whether constant unilateral, alternating, intermittent, vertical, exotropic, or esotropic, are referred for further evaluation. Some of these patients have to be followed up or treated but are not amblyogenic. Positive predictive value (PPV) refers to the proportion of the population that screens positive and is found to have disease on diagnostic testing (ie, true positive results/true positive + false positive [TP/TP+FP]. The PPV has a high significance in a screening examination evaluation. The higher the PPV, the fewer children will be referred from the screening examination to a thorough time- and money-consuming examination in spite of being “healthy.” The PPV of our screening examination was 62.1%. The PPV must always be considered along with the actual percentage of referred children. Negative predictive value (NPV) refers to the proportion of a population that screens negative for a disease and has this confirmed by diagnostic testing (ie, true negative results/true negative + false negative [TN/TN+FN]). This is a very important value in the evaluation of a screening test because subjects who screen negative are discharged from further ophthalmologic surveillance. Our NPV was 99.6%, and only 3 of 779 children (0.4%) were found to have amblyopia in spite of passing the screening examination. All 3 children had only “mild” amblyopia. Screening programs for amblyopia have been previously reported.17-26 Many screening programs for decreased visual acuity in preschool children report referral rates in the range of 10% to 23% and treatment rates as high as 10% to 20% in spite of a prevalence of amblyopia of less than 5%.16,18-20 Others report low compliance with the screening and with further follow-up and treatment.26 The photoscreeners used to detect amblyopia and amblyogenic risk factors in infancy and childhood also show a very high referral rate (>20%), a high false-positive rate (up to 10%), and a high false-negative rate (>3%).27,28 These studies were designed to evaluate the ability to detect “amblyogenic factors” as determined by the authors and thus are not comparable to our study, which actually determines the prevalence of amblyopia in a previously screened population.
Journal of AAPOS Volume 4 Number 4 August 2000
Our examination technique has many advantages. The examination is objective, simple to perform, and welltolerated by the tested infants and their parents. It enables the examiners to acquire all the information needed in an average of 2 to 3 minutes and does not require further interpretation. The examination can be performed either by an ophthalmologist or an orthoptist. The only equipment needed for the examination is a retinoscope, a small target, a penlight, and a +3.00 lens. Access to care and follow-up and parental education are integral and crucial components of any vision screening program, as emphasized by Lichter.19 The problem of low compliance was reviewed in an inner city population.24 In conclusion, our rationale for screening infants for amblyopia is that early discovery and treatment provide better and more stable results of visual acuity and possibly of other visual functions known to be affected in amblyopia.1,29 Our method is very simple, quick, easy to perform, and well-tolerated by the tested children and their parents. Our results demonstrate that this program lowers the prevalence of amblyopia and, in particular, severe amblyopia. The prevalence of amblyopia in the population remains between 1% and 5% and is the leading cause of unilateral visual deficit among young adults.30,31 A child with amblyogenic risk factors that has been discovered and treated early can look forward to more than half a century of school and employment that demand performance of visually mediated tasks. Therefore, the most important benefit from screening is the reduction of preventable visual loss. We thank Monte A. Del Monte, MD, and Steven M. Archer, MD, for invaluable help in writing this article and for reviewing the statistical analysis of the patient data. References 1. Greenwald MJ, Parks MM. Amblyopia. In: Duane TD, Jaeger EA, editors. Clinical ophthalmology. Vol 1. Philadelphia (PA): Harper & Row; 1994. p. 1-21. 2. LeVay S, Wiesel TN, Hubel DH. The development of ocular dominance columns in normal and visually deprived monkeys. J Comp Neurol 1980;191:1-51. 3. Harwerth RS, Smith EL III, Duncan GC, Crawford ML, von Noorden GK. Multiple sensitive periods in the development of the primate visual system. Science 1987;232:235-8. 4. Worth CA. Squint: its causes, pathology and treatment. Philadelphia (PA): Blakiston; 1903. 5. Von Noorden GK. Mechanisms of amblyopia. Adv Ophthalmol 1977;34:92-115. 6. Hubel DH, Wiesel TN. The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol (Lond) 1970;206:419-36. 7. Wiesel TN. Postnatal development of the visual cortex and the influence of environment. Nature 1982;299:583-91. 8. Wiesel TN, Hubel DH. Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 1963;26:1003-17. 9. Neumann E, Eibschitz N, Hyams SW, Friedman Z. Ophthalmic screening in child welfare clinics in Israel with particular reference to strabismus and amblyopia. J Pediatr Ophthalmol Strabismus 1971;8: 257-60.
Journal of AAPOS Volume 4 Number 4 August 2000
Eibschitz-Tsimhoni et al
10. Friedman Z, Neumann E, Peleg B. Screening for strabismus and amblyopia in child welfare clinics. Metab Ophthalmol 1978;2:109. 11. Friedman Z, Neumann E, Hyams SW, Peleg B. Ophthalmic screening of 38,000 children, age 1 to 2.5 years, in child welfare clinics. J Pediatr Ophthalmol Strabismus 1980;17:261-7. 12. Neumann E, Friedman Z, Abel-Peleg B. Prevention of strabismic amblyopia of early onset with special reference to the optimal age for screening. J Pediatr Ophthalmol Strabismus 1987;24:106-10. 13. Friedman Z, Neumann E, Abel-Peleg B. Outcome of treatment of marked ametropia without strabismus following screening and diagnosis before the age of three. J Pediatr Ophthalmol Strabismus 1985; 22:54-7. 14. Friendly DS. Preschool visual acuity screening tests. Trans Am Ophthalmol Soc 1978;76:383-480. 15. Sondhi N, Archer SM, Helveston EM. Development of normal ocular alignment. J Pediatr Ophthalmol Strabismus 1988;25:210-1. 16. Hopkisson B, Arnold P, Billingham B, McGarrigle M, Shribman S. Can retinoscopy be used to screen infants for amblyopia? A longitudinal study of refraction in the first year of life. Eye 1992;6:607-9. 17. De Becker I, McPherson HJ, La Roche GR, Braunstein J, Cottle R, McIntyre LL, et al. Negative predictive value of populationbased preschool vision screening program. Ophthalmology 1992;99: 998-1003. 18. Preslan MW, Novak A. Baltimore vision screening project. Ophthalmology 1996;103:105-9. 19. Lichter PR. Vision screening and children’s access to eye care [editorial]. Ophthalmology 1992;99:843-4. 20. Kohler L, Stigmar G. Vision screening of four-year-old children. Acta Paediatr Scand 1973;62:17-27.
199
21. Kohler L, Stigmar G. Visual disorders in 7-year-old children with and without previous vision screening. Acta Paediatr Scand 1978;67:373-7. 22. Landenvall G. Visual assessment of school children seven years old. Findings at screening. Analysis of treated children. Acta Ophthalmol Scand 1992;70:412. 23. Yazawa K, Suga J, Wakita S, Sumitomo M, Vemura Y. The Tokyo metropolitan home vision screening program for amblyopia in 3year-old children. Am J Ophthalmol 1992;114:416-9. 24. Nordlow W, Joachimsson S. A screening test for visual acuity in four year old children. Acta Ophthalmol Scand 1962;40:453-62. 25. Lennerstrand G, Jakobsson P, Kvarnstrom G. Screening for ocular dysfunction in children: approaching a common program. Acta Ophthalmol Scand 1995;214:26-38. 26. Williamson TH, Andrews R, Dutton GN, Murray G, Graham N. Assessment of an inner city visual screening program for preschool children. Br J Ophthalmol 1995;79:1068-73. 27. Ottar WL, Scott WE, Hogado SI. Photoscreening for amblyogenic factors. J Pediatr Ophthalmol Strabismus 1995;32:289-95. 28. Schworm HD, Kau C, Reindl B, Horstmann S, Sollinger B, Boergen KP. Photoscreening for early detection of amblyogenic eye changes. Klin Monatsbl Augenheilkd 1997;210:158-64. 29. Levi DM. Visual acuity in strabismic and anisometropic amblyopia. Ophthalmol Clin North Am 1990;3:289-301. 30. Leske MC, Hawkins B. Screening relationship to diagnosis and therapy. In: Duane TD, Jaeger EA, editors. Clinical ophthalmology. Vol 5. Philadelphia (PA): Harper & Row; 1994. p. 1-19. 31. Krueger DE, Ederer R. Report on the National Eye Institute’s visual acuity impairment survey pilot study. Bethesda (MD): Office of Biometry and Epidemiology, NEI, NIH, PMS, DMMS, 1984.
An Eye on the Arts – The Arts on the Eye Bella had inherited her father’s lazy eye. His gaze veered shiftily to the right as if something very lewd was going on behind other people’s backs. As a child, Bella had thought of it as the evil eye, all-seeing, masterful. Until, that is, this congenital weakness of her father’s emerged in her…She had to wear a pair of glasses with the right lens patched over with sticking plaster. It gave her a lopsided, partial view of the world—a huge, pinkish blur before her and a sensation of an obstruction looming ahead which was never encountered yet never went away. She sometimes feared that this flesh-coloured wall that met her gaze was, in fact, a fresh layer of skin growing over the unused eye. Frequently she had to take the glasses off to reassure herself that her wall-up eye was still there, and still worked. “Is it sore?” her schoolfriends asked about the eclipsed eye, associating sticking plaster with pain. Some were convinced that Bella did not have a second eye; others that she had been badly scarred and it was too unsightly to expose. —Mary Morrissy (from A Lazy Eye p 54)