Treatment of amblyopia by occlusion has been described for more than 200 years and remains the accepted treatment; however, few population-based studies report on its efficacy. Such studies are difficult as no standard, accurate linear test of visual acuity exists that is applicable to all ages including infants. The Snellen acuity test used to measure the outcome of treatment cannot be compared with the single optotype tests used during the early stages of treatment.
It has long been clear that occlusion is less successful and eccentric fixation more prevalent in the older child.1 The risk of developing amblyopia is higher in the younger child,23 and clinicians have been aware that the density of amblyopia associated with nonaccommodative esotropia increases with the duration of the deviation. Many authors have noted that amblyopia presenting later in life requires more treatment.3-5 These clinical observations correspond to our understanding of visual development based on the initial work of Hubel and Weisel.6 Therefore, clinicians now aim to detect and treat amblyopia as early as possible during the critical period. As strabismus is detected earlier, fewer patients have accurate measures of visual acuity at the start of treatment.
The lack of published clinical trials has led to similar questions being asked today as in 19774:
* Should amblyopia be treated at all?
* Do the results justify the effort?
* What are the best methods of treatment?
The success of early treatment in the majority of cases prevents accurate assessment of visual acuity on presentation. This precludes the use of prospective studies, of lines of improvement in visual acuity, which would involve an unethical delay in starting occlusion. This retrospective study examines the outcome of treatment using the only quantitative measure available, ie, visual acuity on discharge. It examines the results for different diagnostic groups and analyzes a number of factors for their influence on the outcome of treatment.
SUBJECTS AND METHODS
Data relating to all children discharged from the orthoptic department after active treatment have been routinely collected in our small district general hospital since 1991. The method of collection described in 19937 has been modified to include all those children who failed to attend and those who did not complete treatment. Their details are added to the database 1 year after their last failed appointment. Consequently, this study includes all children discharged in a 5-year period from April 1991 to March 1996, and all those who failed to complete their treatment in the 4-year period from April 1991 to March 1995. Those who failed to attend between April 1995 and March 1996 were not added to the database at the time of study.
The medical records of all children were examined, and all those who had been prescribed occlusion therapy were assumed to have had amblyopia. Thirty six children who had previous treatment; 9 children with pathology, such as cataract and nystagmus that affected their visual acuity; and 11 children with handicap or global delay that affected the accuracy of visual acuity testing were excluded.
The medical records of 18 children were unavailable for review. However, as the linear visual acuity recorded on the department's database was not less than 6/12 for any of these children, their exclusion was not thought to have affected the final results significantly. The visual acuity of the amblyopic eye on discharge, ie, the eye with the lower visual acuity (where there was a difference between the two eyes), was recorded.
Linear visual acuity was tested using standard Snellen charts, which varied in the spacing of letters and therefore in the crowding effect. Matching of letters on the Snellen chart was used for illiterate children to obtain a linear optotype test. An attempt was made to record compliance with occlusion treatment by grading the orthoptist's comments into good, fair, and poor compliance, and by recording the number of missed appointments in the first 12 months of attendance.
We were unable to use the Thompson deprivation score8 to analyze whether socioeconomic deprivation influenced outcome as we were advised it could not be applied to an area where some postal codes represented large areas. However, local population census studies show mixed socioeconomic grouping.
Anisometropia was defined as a difference in refractive error between the two eyes of 1 diopter (D) or more in any meridian. Microtropia was defined as a condition where there was demonstrable stereoacuity and motor fusion with either no movement on cover test with central suppression on the 4-prism-diopter (?) test, or minimal deviation on cover test. Pure anisometropic amblyopia was defined as no deviation on cover test, and no demonstrable suppression scotoma.
Children were divided into two groups, which were analyzed separately. Patients in group 1 had a diagnosis of esotropia; other types of strabismus (16 patients), mainly exotropia, were excluded as it was felt occlusion had primarily been used for antisuppression treatment. Group 2 had a diagnosis of anisometropic amblyopia with or without microtropia. As 6/10 is the legal requirement for driving in the United Kingdom, treatment success was defined as an outcome of at least 6/12 linear visual acuity in the amblyopic eye. Those who did not achieve 6/12 were regarded as unsuccessful.
The following risk factors for treatment failure were analyzed: refractive error, age at presentation, age at start of treatment, visual acuity at start of treatment, fixation at start of treatment, type of occlusion, and compliance. Continuous measurements, such as age at presentation, were summarized as the median and interquartile range, categorical measurements, such as compliance were presented as frequencies and proportions. The significance of differences in risk factors between successful and unsuccessful groups was assessed using the MannWhitney U test for continuous measurements and Pearson's Chi-square or the Mantel-Haenszel test for ordered categories as appropriate.
During the study period, 246 children received treatment. Of these, 167 (68%) had strabismus, and 79 (32%) had anisometropia with or without microtropia. The type of occlusion used is shown in Table 1.
Group 1 consisted of 151 children with esotropia. The mean age at presentation was 3.1 years (SD 1.5). Of the 151 children, 43 (29%) had anisometropia of 1 D or more. At presentation, the visual acuity was tested on only 83 patients (55%). The age at onset was recorded in less than 50% of cases. The mean number of visits in the first year was 6.1 (SD 2.1).
Type of Occlusion Used During Treatment
Fig 1: Group 1 -Esotropia linear visual acuity in amblyopic eye on discharge.
Linear visual acuity in the amblyopic eye on discharge is shown in Fig 1. At discharge, 110 children (73%) achieved 6/9 or better; 24 (16%) achieved 6/12 or 6/18; and 17 (11%) achieved less than 6/18.
Risk Factor Analysis
There were 128 (85%) successful outcomes and 23 (15%) unsuccessful outcomes.
The presence of astigmatism was similar and not statistically significant (p=0.85) in the successful (n=32, 25%) and unsuccessful groups (n=6, 27%). The presence of 1 D or more of anisometropia was significantly associated with unsuccessful outcome (P<0.001). One D or more of anisometropia was present in 23% (n=29) of the successful group and 61% (rc=14) of the unsuccessful group.
Hypermétropie was less significant (P<0.01), 74% (?=?) of the unsuccessful group having 3 D or more hypermetropia compared with 50% (n=63) of the successful group.
Age at Presentation/start of treatment
Neither age at presentation (P=0.23), nor age at start of treatment (P=0.30) was statistically significant. The mean age at presentation was 3.0 years (SD 1.4) in the successful group and 3.5 years (SD 1.9) in the unsuccessful group.
Visual Acuity at Start of Treatment
Visual acuity was accurately measured in only a small group due to the young age of presentation. However, low visual acuity at the start of treatment did appear to be significantly correlated to poorer outcome (P<0.001).
Fig 2: Group 2- "Pure" anisometropic visual acuity on referral and on discharge.
Fixation at Start of Treatment
Fixation was recorded as a measure of low visual acuity, particularly in the younger age group where accurate acuity tests were not possible. A significant association between poor fixation or no movement to fixate and unsuccessful outcome was noted, although numbers were small (P<0.001). Poor fixation or no movement on cover test was recorded in 91% (n=10) of the unsuccessful group and only 33% (n=15) of the successful group.
Type of Occlusion
An unsuccessful outcome was associated with those children who had full-time occlusion at some point in their treatment (P=0.004). Full-time occlusion was prescribed in 28% (n=36) of the successful group and 57% (n=13) of the unsuccessful group. However, this can be attributed to additional hours of occlusion being prescribed where visual acuity has not responded to treatment.
Children recorded by the orthoptist as having fair or poor compliance had a significantly poorer outcome than those with good compliance (P<0.001). In the unsuccessful group, 70% (n=16) were recorded as having poor compliance; in the successful group only 17% (rc=20) were recorded. A significant association between number of appointments missed in the first year and poorer outcome (P=0.009) was noted.
Group 2 consisted of 79 children with no strabismus, ie, pure anisometropic amblyopia with or without microtropia.
The mean age of presentation of this group was slightly higher, 4.0 years (SD 1.1). The visual acuity of the amblyopic eye on referral using the Sheridan-Gardner single optotype test was recorded in 81% of children. Thirteen children were referred with 6/24 or worse, 43 with 6/18 or 6712, and 9 with 6/9. The visual acuity on discharge using the Snellen linear chart was recorded in 100% of children. Of the total, 75 children (95%) achieved 6/6 or 679; 4 children (5%) achieved 6/12; no child had less than 6/12 on discharge. Both results are shown in Fig 2.
The primary orthoptic screening service, which tests children between the ages of 3Mi and 3% years, referred 67 (85%) of the children; 12 (15%) were referred from other sources. The mean number of visits in the first year was slightly lower than in group 1: 5.6 visits (SD 1.5). Table 2 gives the levels of anisometropia for this group of children. Thirty nine children (50%) had 1 D or more of astigmatism in the amblyopic eye.
Group 1: Esotropia
In the largest full population study to date. Woodruff and colleagues9 reported on the outcome visual acuity for a combined group of 894 strabismic and anisometropic amblyopes: 48% achieved 6/9 or 6/6. Hiscox and colleagues10 reported an outcome of 6/9 or better in 40% of 368 patients studied. Our results are encouraging in that for esotropia alone- the diagnosis carrying the greatest risk of amblyopia- 73% achieved 679 or 676. However, the 11% of children who failed to achieve 6/18 remain a cause for concern.
The factors that appeared to influence outcome were the presence of anisometropia, visual acuity at the start of treatment, and compliance. Poor visual acuity or poor fixation at the start of treatment appeared to be a significant factor in outcome; however, it must be remembered that 45% of the children in this study were too young to test visual acuity accurately at this point.
Group 2 - Pure Anisometropia With or Without Microtropia
Neumann and colleagues11 described several early-onset esotropes with poor fixation who failed to respond to treatment. We feel there may be some validity in the suggestion that a few cases may have reduced visual acuity as a result of subclinical retinal abnormality affecting the macula rather than amblyopia.12"13 However, it was noticeable that the results of treatment in cases of poor initial visual acuity in this group were extremely variable. Neumann and colleagues,14 using full-time occlusion and studying only compliant esotropes younger than 3 years of age found that poor visual acuity was not a poor prognostic sign. Other studies where only compliant cases where included found equal visual acuity or acuity within one line of the fixing eye in 80% to 95%.514~1S Compliance is more difficult when the visual acuity is poor, which may explain the significance of lower visual acuity at the start of treatment.
The age at presentation in this and other studies910 was not found to be a statistically significant factor in outcome. However, the successful group in our study did have a lower mean age of presentation, 3.1 years (SD 1.4), than the unsuccessful group, 3.5 years (SD 1.9), and the overall age at presentation of children in our study was lower than Woodruff's cohort in which fewer children achieved 6/12 or better. Mean age at presentation in our esotropie group was 3.1 years, compared with 4.1 years for strabismus associated with anisometropia and 3.3 years for strabismus without anisometropia in Woodruff's study.16 This is confirmed by the fact that in our cohort, it was possible to test visual acuity on presentation in only 55% of children compared with 80% in Woodruffs cohort.
Two factors cloud any interpretation of age of presentation and its influence on outcome. First, the majority of children in our study presented early, with only 10% presenting after the age of 5. Second, the evaluation of age of presentation is undertaken without knowledge of the age of onset and should be treated with considerable caution.
Esotropia includes a wide spectrum of diagnoses with differing prognoses. With the exception of untreated infantile esotropia with a tendency to crossfixate,17 early onset esotropia is associated with higher degrees of amblyopia and less binocularity.
Small-angle esotropia with onset in the first 3 years of life may not be obvious to the untrained eye. Fifty percent of our parents seemed unaware of the onset of strabismus in their children in this study. Consequently, there may be a delay in presentation until the angle of deviation increases or the child is screened for visual defects. Therefore, although we find that age of presentation is not a statistically significant factor in visual outcome, it should not be assumed that early diagnosis is not important. Two children presenting at 3 years of age will have very different prognoses if one has a recent onset and the other an undetected strabismus in the first year of life.
The authors believe that age at presentation is not necessarily a useful measure without knowledge of duration of amblyopia. As it is not possible to determine the age of onset in all cases of amblyopia, we need further, preferably prospective, studies to compare outcome by different diagnostic groups of esotropia and the presence and degree of any binocular vision.
As the majority of esotropes in our group were diagnosed before age 5, we conclude that the most important factor for outcome is likely to be compliance. Relatively little emphasis has been placed on occlusion regimen in the past. In this series, as is common practice, full-time or part-time occlusion was used according to parental or orthoptic preference. However, part-time occlusion was often the first choice with an attempt to increase to full-time occlusion if improvement was not forthcoming. This can prove very demotivating as it can lengthen the most difficult period of initial occlusion when visual acuity is low.
Younger children require a shorter period of occlusion to obtain this initial improvement3·6·18 than older children. It should also be noted that when collecting the data, it was noticed that although age at start of treatment was recorded as the age at which occlusion was first prescribed, variable or noncompliance sometimes meant that effective treatment started considerably later. Therefore, in an attempt to improve our results further, we shall introduce an occlusion protocol aimed at giving maximum support for parents in the early stages.
An ideal regimen for any child with visual acuity of less than 6/18 or poor fixation may be to prescribe intensive occlusion with frequent review in the early stages, preferably by the same orthoptist, to provide continuity of care and support. If this shortens the period of active treatment, no increase in the overall level of resources should be required.
Group 2: Anisometropia
The incidence of anisometropia without strabismus was much higher in our study than in the recent seven-center study,10 but equates to the incidence of 30% for Oliver and Navratski18 in their study of 5000 children in 1970. In a study of 9000 men at entry into the armed forces, Helveston20 found an incidence of 52% of amblyopia without strabismus. This suggests that anisometropic amblyopia is at present underdetected in the United Kingdom. Shaw and Fielder21 stated that existing screening tests using single optotype tests (known to underestimate visual acuity in amblyopic eyes) are inadequate for screening. However, this study suggests that when used by orthoptists, it can successfully identify amblyopia prior to school entry.
Although some authors suggest the age at presentation does not influence outcome significantly,22 others have shown better results with early referral.23 Woodruff and Hiscox10 found significantly better results in their group in centers where children presented at an earlier age.
Regardless of the degree of anisometropia or the level of visual acuity on presentation, which in some cases may have been overestimated by the single optotype test, we have found treatment to be almost entirely successful in children who present at 4 years of age from orthoptic screening.
A weakness of this study is the failure to record 6/5 and 6/4, the actual visual acuity achieved in many of this group. Treatment involved fewer visits than the strabismic group in the first year. Compliance with occlusion was not a problem. Most parents were very motivated to undertake treatment after witnessing the reduced visual acuity evident at screening.
It is our clinical impression that glasses are well accepted prior to school age, and that anisometropic amblyopia does not recur provided glasses are worn. Therefore, very few further visits are required and discharge to the optician can often be as early as 6 years of age. In this department, these cases are managed entirely by orthoptists, including retinoscopy, requiring only one visit to an ophthalmologist on presentation, solely for examination of the fundus and media.
In conclusion, a limitation of any retrospective study of outcome visual acuity is that visual acuity at the start of treatment and lines of improvement are unknown. However, the results for our entire treated group are better than other full-population studies. There is no reason to suggest that the populations vary in any way other than compliance, age at presentation, and the percentage of pure anisometropes. The occlusion regimen was not unusual. Therefore, we conclude that early detection and a supportive and effectively monitored occlusion protocol are vital. Although we have shown that patients can be successfully treated within existing resources, there is still room for improvement.
1. Massie H. Fixing eye occlusion: survey of approximately 1000 cases of patients receiving occlusion of the fixing eye. Transactions of the Ophthalmological Society of Australia. 1965;24:39-46.
2. Keech RY, Kutschke CO. Upper age limit for the development of amblyopia J Pediatr Ophthalmol Strabismus. 1995;32:8993.
3. Ching FC, Parks ME, Friendly DS. Practical management of amblyopia. J Pediatr Ophthalmol Strabismus. 1986;23:12-16.
4. Flynn JT, Cassady JC. Current trends in amblyopia therapy. Ophthalmology. 1978;85:428-450.
5. Fulton AB, Mayer DL. Esotropie children with amblyopia: effects of patching on acuity. Graefes Arch Clin Exp Ophthalmol. 1998;226:309-312.
6. Hubel DH, Wiesel TN. Binocular interaction in striate cortex of kittens reared with artificial squint. J Neurophysiol. 1965;28:1041-1059.
7. Beardsell MR. Audit of outcome-meaningless data or a measure of the quality of care? British Orthoptic Journal. 1993;50:11-14.
8. Townsend P, Phillmore P, Beattie A. Health and Deprivation. Inequality and the North. London, England: Croom Helm; 1988.
9. Woodruff G, Hiscox F, Thompson JR, Smith LK. Factors affect ing the outcome of children treated with amblyopia. Eye. 1994;8:627-631.
10. Hiscox F, Strong N, Thompson JR, Minshull C, Woodruff G. Occlusion for amblyopia: a comprehensive surety of outcome Eye. 1992;6:300-304.
11. Neumann E, Friedmann Z, Abel-Peleg B. Presentation of stra· bismus amblyopia of early onset with special reference to the optimal age for screening. J Pediatr Ophthalmol Strabismus. 1987;24:106-110.
12. Williams C, Papakostopoulos D. Electro-oculographic abnormalities in amblyopia. Br J Ophthalmol. 1995;79:218-224.
13. Awaya S, Watanabe Y. Amblyopia. Current Opinion in Ophthalmology. 1995;6:9-14.
14. Newman R, Oliver M, Gottesman N, Shunshoni M. Prognosis for occlusive therapy for strabismic and anisometropic amblyopia and for different initial depths of amblyopia. Chibret International Journal of Ophthalmology. 1989;6:22-26.
15. Sjostrand J, Abrahamson M. Risk factors in amblyopia. Eye. 1990;4:787-793.
16. Woodruff G, Hiscox F, Thompson JR, Smith LK. The presenta tion of children with amblyopia. Eye. 1994;8:623-626.
1 7. Calcott C. The natural history of infantile esotropia; a study of the untreated condition in the visual adult. In: Tilleon G, ed. Advances in Amblyopia and Strabismus. Transactions of the VII International Orthoptic Congress, Nuremberg, Germany: Fahner Verlag.
18. Lithander J, Sjostrand J. Anisometropic and strabismic amblyopia in the age group 2 years and above: a prospective study of the results of treatment. Br J Ophthalmol. 1991;75:111-116.
19. Oliver M, Navratsky I. Screening of pre-school children for ocular anomalies. Br J Ophthalmol. 1971;55:467-471.
20. Helveston EM. The incidence of amblyopia ex anopsia in young adult malea in Minnesota in 1962-63. Am J Ophthalmol. 1965;60:75-77.
21. Shaw DE, Fielder AR, Minshull C, Rosenthal AR. Amblyopiafactors influencing age at presentation. Lancet. 1988;7:207209.
22. Hardman Lea SJ, Loades J, Rubenstein MP. The sensitive period for anisometropic amblyopia. Eye. 1989;3:783-790.
23. Collins A Hannah J. Anisometropia orthoptics and early diagnosis make the difference. Br J Ophthalmol. 1996;53:31-35.
Type of Occlusion Used During Treatment
Group 2 - Pure Anisometropia With or Without Microtropia