Abstract
Purpose
To determine the amount and time course of binocular visual acuity improvement during treatment of bilateral refractive amblyopia in children age 3 to <10 years old
Design
Prospective, multicenter noncomparative intervention
Methods
113 children (mean age = 5.1 years) with previously untreated bilateral refractive amblyopia were enrolled at 27 community- and university-based sites and were provided optimal spectacle correction. Bilateral refractive amblyopia was defined as 20/40 to 20/400 best-corrected binocular acuity in the presence of ≥ 4.00 D hypermetropia by spherical equivalent and/or ≥ 2.00 D astigmatism in each eye. Best-corrected binocular and monocular visual acuities were measured at baseline and at 5, 13, 26 and 52 weeks. The primary study outcome was binocular acuity at one year.
Results
Mean binocular visual acuity improved from 0.50 logMAR (20/63) at baseline to 0.11 logMAR (20/25) at one year (mean improvement 3.9 lines, 95% confidence interval [CI] = 3.5 to 4.2). Mean improvement at one year for the 84 children with baseline binocular acuity of 20/40 to 20/80 was 3.4 lines (95% CI = 3.2 to 3.7) and for the 16 children with baseline binocular acuity of 20/100 to 20/320 was 6.3 lines (95% CI = 5.1 to 7.5). The cumulative probability of binocular acuity of 20/25 or better was 21% at 5 weeks, 46% at 13 weeks, 59% at 26 weeks, and 74% at 52 weeks.
Conclusions
Treatment of bilateral refractive amblyopia with spectacle correction improves binocular visual acuity in children age 3 to <10 years old, with most improving to 20/25 or better within one year.
Introduction
Bilateral refractive amblyopia can develop in children with large amounts of uncorrected hypermetropia and/or astigmatism in both eyes. Treatment consists of prescribing the appropriate refractive correction with the possible addition of occlusion or pharmacologic penalization if asymmetric visual acuity is present after correction is provided. The prevalence of bilateral amblyopia at the time of entry into school was estimated in one study to be 0.5% (4 of 830).1 The presumed mechanism of bilateral refractive amblyopia is pattern vision deprivation. Abnormal binocular interaction with suppression may also contribute in those cases with concomitant strabismus.2 There are few published studies of treatment for bilateral amblyopia.3-9 Most have been limited by small numbers of subjects and short follow-up times. To address these limitations, we designed a prospective cohort study to determine the amount and time course of binocular visual acuity improvement during usual treatment of previously untreated bilateral refractive amblyopia.
Methods
The Pediatric Eye Disease Investigator Group (PEDIG) conducted this study at 27 community- and university-based clinical sites, and it was supported through a cooperative agreement with the National Eye Institute of the National Institutes of Health. The respective institutional review boards (IRBs) approved the protocol and HIPAA-compliant informed consent forms. The parent or guardian of each study participant gave written informed consent.
Some IRBs required that children above a certain age give their assent for participation; assent was given by each child for whom the IRB required it. The major aspects of the protocol are summarized herein. The complete protocol is available at http://public.pedig.jaeb.org.
The major eligibility criteria included age 3 to <11 years old, binocular visual acuity 20/40 to 20/400 in optimal refractive correction, cycloplegic refractive error in each eye of ≥ 4.00 D hypermetropia (spherical equivalent) and/or ≥ 2.00 D astigmatism (including some eyes with myopic astigmatism), no myopia greater than -6.00 D of spherical power in plus cylinder form, no previous treatment for amblyopia except one month or less of spectacle wear terminating three or more months prior to enrollment, no amblyopia treatment planned other than spectacles, and no cause for reduced visual acuity suspected other than bilateral refractive amblyopia.
At a screening visit, visual acuity was measured using trial frames or a phoropter with correction from a cycloplegic refraction (using cyclopentolate 1%). The method for determining adequacy of cycloplegia was at investigator discretion. Children who were potentially eligible were prescribed spectacles in which anisometropia, astigmatism, and myopia were fully corrected and in which hypermetropia was either fully corrected or under-corrected symmetrically by no more than 1.50 D in both eyes. There was no untreated control group.
Spectacles were worn, for the first time, for 10 to 30 minutes before visual acuity was measured at the baseline visit. A study-certified vision tester measured visual acuity first binocularly, then for each eye separately. Children age 3 to 6 years were tested using the ATS single-surround HOTV visual acuity testing protocol which yields a line score (Snellen score),10 whereas children age 7 to 10 years were tested with the electronic ETDRS (E-ETDRS) testing protocol which yields a letter score.11 If either eye's monocular acuity tested worse at baseline than at the screening visit, acuity was to be retested. In children whose prescribed spectacles contained hypermetropic correction, because of the possibility that the reduced acuity was due to incomplete relaxation of accommodation, the retesting was to be completed using a −1.00 D lens over the spectacles. Stereopsis was measured using the Randot Preschool Stereoacuity Test (Stereo Optical Company, Chicago, Illinois) and ocular alignment was assessed using the simultaneous prism and cover test.
Protocol-specified follow-up visits were conducted at 5, 13, 26, and 52 weeks after the baseline examination. If monocular acuity was 20/25 or better in both eyes at the 5-week or 13-week visit, then subsequent visits prior to the one-year examination were skipped.
At each follow-up visit, visual acuity was measured with spectacle correction first binocularly and then monocularly for the right eye and then the left eye. At the one-year examination, additional testing included the Randot Preschool Stereoacuity Test and a refraction (manifest or cycloplegic). A refraction was also performed any time the investigator suspected that refractive error was not optimally corrected. Whenever a significant change in refractive error was detected (as defined in the protocol), monocular and binocular acuities were retested using the new refractive correction in a trial frame.
Spectacle correction changes during follow up were at investigator discretion. Additional amblyopia treatment with patching and/or atropine was also initiated at investigator discretion; however, it was suggested that treatment be started only after monocular visual acuity had stopped improving in each eye.
Statistical Methods
Visual acuity data for patients less than 7 years old were combined with visual acuity data from patients 7 years or older by converting to a common logMAR scale both the HOTV line scores from the younger patients and the E-ETDRS letter scores from the older patients. A change of 0.1 logMAR was considered to be a one line change in acuity (equivalent to a 5-letter change using the E-ETDRS testing method).
The primary study outcome was binocular visual acuity at one year. Mean lines of binocular acuity improvement from baseline to one year were computed along with a 95% confidence interval (CI). The one-year cumulative probability of reaching a binocular acuity of 20/25 or better and a 95% CI were computed using the Kaplan-Meier product-limit method.
The associations of baseline characteristics with lines of visual acuity improvement at one year and with the proportion of children achieving binocular acuity 20/25 or better during follow up were evaluated using linear regression and proportional hazards regression, respectively. All regression models included baseline binocular visual acuity as a covariate. The average of the two eyes was used to assess spherical equivalent and astigmatism.
The associations of refractive error type (hypermetropia vs. astigmatism) with lines of visual acuity improvement at one year and with the proportion achieving binocular acuity 20/25 or better during follow up were evaluated using linear regression and proportional hazards regression, respectively, in a subset of children who had either significant hypermetropia only or significant astigmatism only. Baseline binocular visual acuity was included in the models as a covariate.
Among children who at baseline had significant hypermetropia but no significant astigmatism, the association between baseline hypermetropia and baseline binocular acuity was evaluated using linear regression.
The association of binocular acuity improvement at one year with stereoacuity improvement at one year was assessed using Pearson's correlation.
Mean lines of monocular acuity improvement from baseline to one year and a 95% CI were computed using data from both eyes of each child and using generalized estimating equations to account for the within-subject correlation.
All reported P values are two-tailed. Analyses were conducted using SAS version 9.1 (SAS Institute, Cary, NC).
Results
Baseline Characteristics
Between August 2004 and June 2005, 113 children with a mean age of 5.1±1.3 years were enrolled into this study at 27 clinical sites. Mean baseline binocular visual acuity was 0.50 logMAR (approximately 20/63). Twenty-three children (20%) had ≥ 2 lines of interocular difference (IOD) in visual acuity. Additional baseline characteristics are included in Table 1.
Table 1.
Baseline Characteristics of Study Cohort of Patients Aged 3-<10 Years with Previously Untreated Bilateral Refractive Amblyopia (N=113)
| N (%) | |
|---|---|
| Gender: Female | 50 (44) |
|
| |
| Race / Ethnicity | |
| White | 70 (62) |
| African-American | 16 (14) |
| Hispanic or Latino | 21 (19) |
| Other | 6 (5) |
|
| |
| Age | |
| Mean (SD) years | 5.1 (1.3) |
| Range, years | 3.0 to 9.2 |
| 3 to <4 | 24 (21) |
| 4 to <5 | 30 (27) |
| 5 to <6 | 38 (34) |
| 6 to <7 | 11 (10) |
| 7 to <8 | 5 (4) |
| 8 to <9 | 4 (4) |
| 9 to <10 | 1 (1) |
| 10 to <11 | 0 (0) |
|
| |
| Binocular Visual Acuity | |
| Mean LogMAR [Snellen equivalent], (SD) | 0.50 [20/63] (0.18) |
| Range [Snellen equivalent] | 0.30 to 1.20 [20/40 to 20/320] |
| 20/40 to 20/50 | 56 (50) |
| 20/60 to 20/80 | 38 (34) |
| 20/100 to 20/160 | 16 (14) |
| 20/200 to 20/320 | 3 (3) |
| Worse than 20/320 | 0 (0) |
|
| |
| Interocular Acuity Difference | |
| Mean (SD) lines | 1.1 (1.6) |
| Range lines | 0.0 to 8.0 |
| 0 to <1 line | 50 (44) |
| 1 to <2 lines | 40 (35) |
| 2 to <3 lines | 11 (10) |
| >=3 lines | 12 (11) |
|
| |
| Type of Refractive Amblyopia in Both Eyes* | |
| Significant hypermetropia only | 40 (35) |
| Significant astigmatism only | 46 (41) |
| Both significant hypermetropia and astigmatism | 18 (16) |
| Mixed† | 9 (8) |
|
| |
| Spherical Equivalent‡ | |
| Mean (SD), spherical equivalent in diopters | 4.7 (3.1) |
| Range, spherical equivalent in diopters | -2.75 to 11.00 |
| <0.00D | 6 (5) |
| 0.00 to <4.00 D | 41 (36) |
| 4.00 to <7.00 D | 36 (32) |
| >=7.00 D | 30 (27) |
|
| |
| Astigmatism‡ | |
| Mean (SD), cylinder in diopters | 2.4 (1.5) |
| Range, cylinder in diopters | 0.0 to 6.50 |
| <2.00 D | 43 (38) |
| 2.00D to <4.00 D | 54 (48) |
| 4.00D to <6.00 D | 14 (12) |
| >=6.00 D | 2 (2) |
|
| |
| Anisometropia | |
| Mean (SD), spherical equivalent in diopters | 0.4 (0.4) |
| Range, spherical equivalent in diopters | 0.00 to 2.50 |
| 0.00 D | 25 (22) |
| >0.00 D to <0.50 D | 40 (35) |
| 0.50 to <1.00 D | 37 (33) |
| 1.00 to <1.50 D | 8 (7) |
| >=1.50 | 3 (3) |
|
| |
| Hypermetropic spectacle correction prescribed§ | |
| Fully corrected | 20 (19) |
| Undercorrected <=0.50 D | 6 (6) |
| Undercorrected >0.50 to 1.00 D | 49 (46) |
| Undercorrected >1.00 to 1.50 D | 31 (29) |
|
| |
| Stereoacuity, seconds of arc|| | |
| 40 | 4 (4) |
| 60 | 5 (5) |
| 100 | 5 (5) |
| 200 | 5 (5) |
| 400 | 10 (9) |
| 800 | 15 (14) |
| None detected (>800) | 66 (60) |
|
| |
| Strabismus present¶ | 15 (13) |
Significant hypermetropia refers to spherical equivalent ≥+4.00 D; significant astigmatism refers to cylinder ≥ 2.00 D.
Seven children had significant hypermetropia and significant astigmatism in one eye and significant hypermetropia and no significant astigmatism in the other eye, and two children had significant hypermetropia and significant astigmatism in one eye and significant astigmatism and no significant hypermetropia in the other eye.
Refers to the average amount between the two eyes.
Excluded are seven children who had no hypermetropia: 6 had myopia and 1 had neither hypermetropia or myopia. Per protocol, hypermetropic spectacle correction could be either fully corrected or undercorrected symmetrically in both eyes up to 1.50 D.
Three children were unable to perform stereoacuity testing.
Strabismus refers to presence of a horizontal tropia at distance or at near, or history of strabismus or strabismus surgery.
Among 40 children with significant bilateral hypermetropia only (hypermetropia ≥ 4.00 D by spherical equivalent and astigmatism <2.00 D), higher levels of hypermetropia were associated with worse baseline binocular visual acuity (P=<0.001). Mean baseline binocular acuity was 0.46 logMAR (approximately 20/63) in the 17 children with 4.00 to <7.00 D of hypermetropia and 0.67 logMAR (approximately 20/100) in the 23 children with >=7.00 D of hypermetropia. Mean baseline binocular acuity was worse among the 40 children with significant bilateral hypermetropia only compared with the 46 children with significant bilateral astigmatism only (mean acuity 0.58 vs. 0.44 logMAR, respectively; P=<0.001).
At the baseline visit, there were 8 children (7%) whose monocular acuity in one or both eyes tested 2 or more lines worse from the enrollment acuity and whose spectacles contained hyperopic correction. By protocol, these children should have had their acuity retested using a -1.00 lens, but none completed this retesting. Four of these children (50%) had a baseline binocular acuity which was 2 or more lines worse than the better of their enrollment monocular acuities.
Study Completion
The study was completed by 101 (89%) of the 113 children (Figure 1). Children not completing the study were similar to children completing the study in terms of baseline characteristics including age (4.5 versus 5.1 years, P=0.09), baseline binocular visual acuity (mean 0.53 versus 0.49 logMAR, P=0.70), interocular acuity difference (1.3 vs. 1.1 lines, P = 0.68), spherical equivalent (5.3 vs. 4.7 D, P =0.52), and astigmatism (2.0 vs 2.4. D, P = 0.35).
Figure 1. Flow Diagram of One-year Follow Up For Patients with Bilateral Refractive Amblyopia (N=113).
aSkipped indicates visit was skipped because patient had monocular visual acuity 20/25 or better in both eyes at a previous visit. Note: some patients skipped both the 13-wk and the 26-week visit.
bOne patient who completed the 52-week visit did not have binocular acuity measured.
Treatment During Follow Up
Of the 109 children entering the study and completing at least one follow-up visit, 96 (88%) were treated with spectacles alone during follow up and 13 (12%) received additional amblyopia treatment (patching for 12 children and both patching and atropine for 1 child). Compliance with spectacle wear was reported as excellent (spectacles worn 75 to 100% of the waking hours) at every completed visit for 74 children (68%).
Binocular Visual Acuity Improvement
Binocular visual acuity (mean at baseline 0.50 logMAR, 20/63) improved by an average of 2.3 lines (95% CI = 2.0 to 2.6) to a mean of 0.26 logMAR (20/40) after 5 weeks and by an average of 3.9 lines (95% CI = 3.5 to 4.2) to a mean of 0.11 logMAR (20/25) at one year (Table 2). Mean improvement at one year for the 84 children with baseline binocular VA of 20/40 to 20/80 was 3.4 lines (95% CI = 3.2 to 3.7) and for the 16 with baseline binocular VA of 20/100 to 20/400 was 6.3 lines (95% CI = 5.1 to 7.5). Overall, the cumulative probability of reaching binocular acuity 20/25 or better was 21% at 5 weeks, 46% at 13 weeks, 59% at 26 weeks, and 74% (95% CI = 66% to 82%) at one year (Figure 1). All 15 children who first achieved 20/25 or better binocular acuity at the one-year visit had completed the three prior follow-up visits.
Table 2.
Binocular Visual Acuity Outcomes During One Year Of Spectacle Wear in Patients Aged 3-<10 Years with Previously Untreated Bilateral Refractive Amblyopia (N=113)*
| Baseline Binocular Acuity | |||
|---|---|---|---|
| All Patients (N=113) | 20/40 to 20/80 (N=94) | 20/100 to 20/320 (N=19) | |
| Change in Binocular Acuity From Baseline to One Year, lines | N (%) | N (%) | N (%) |
| Mean (SD) | 3.9 (1.8) | 3.4 (1.3) | 6.3 (2.3) |
| Range | 0.0 to 12.8 | 0.0 to 7.0 | 3.0 to 12.8 |
| 0 | 1 (1) | 1 (1) | 0 (0) |
| 1 | 5 (5) | 5 (6) | 0 (0) |
| 2 | 11 (11) | 11 (13) | 0 (0) |
| 3 | 31 (31) | 30 (36) | 1 (6) |
| 4 | 25 (25) | 22 (26) | 3 (19) |
| 5 | 13 (13) | 12 (14) | 1 (6) |
| 6 | 6 (6) | 2 (2) | 4 (25) |
| 7 | 5 (5) | 1 (1) | 4 (25) |
| 8 | 2 (2) | 0 (0) | 2 (13) |
| >8 | 1 (1) | 0 (0) | 1 (6) |
|
| |||
| Binocular Acuity at One Year | |||
| Mean LogMAR [Snellen equivalent] | 0.11 [20/25] | 0.09 [20/25] | 0.18 [20/32] |
| (SD) LogMAR | (0.13) | (0.12) | (0.16) |
| Range LogMAR [Snellen equivalent] | -0.10 to 0.50 [20/16 to 20/63] | -0.10 to 0.40 [20/16 to 20/50] | -0.08 to 0.50 [20/16 to 20/63] |
| 20/16 | 8 (8) | 7 (8) | 1 (1) |
| 20/20 | 28 (28) | 26 (31) | 2 (13) |
| 20/25 | 33 (33) | 28 (33) | 5 (31) |
| 20/32 | 18 (18) | 15 (18) | 3 (19) |
| 20/40 | 8 (8) | 6 (7) | 2 (13) |
| 20/50 | 4 (4) | 2 (2) | 2 (13) |
| 20/63 | 1 (1) | 0 (0) | 1 (6) |
| Worse than 20/63 | 0 (0) | 0 (0) | 0 (0) |
|
| |||
| Binocular Acuity 20/25 or Better At Any Follow-up Visit | 79 (74)† | 70 (78)† | 9 (55)† |
SD = standard deviation
13 children are missing binocular acuity at one year: 10 children with baseline binocular acuity 20/40 to 20/80 and 3 children with baseline binocular acuity 20/100 to 20/320. Therefore, the effective Ns for change in binocular acuity and for binocular acuity at one year are both 100 children overall.
Percentages are cumulative probabilities derived from Kaplan-Meier product-limit method.
We evaluated the effect of a single outlier on analyses related to mean improvement in binocular acuity—a 9-year-old child with astigmatism whose binocular acuity improved 12.8 lines at one year. In all analyses, excluding data from this outlier did not substantively change results except where noted.
Factors Predictive of Binocular Acuity Improvement
Table 3 shows the binocular acuity outcome data stratified by baseline demographic and clinical characteristics. The number of lines of binocular acuity improvement was greater with worse baseline binocular acuity (P<0.001). The cumulative probability of achieving 20/25 or better binocular acuity during follow up was greater in children with better baseline binocular acuity (P<0.001).
Table 3.
Binocular Acuity Outcomes During One Year Of Spectacle Wear in Patients Aged 3-<10 Years with Previously Untreated Bilateral Refractive Amblyopia Stratified by Baseline Characteristics (N=113)
| Binocular Acuity | ||||||
|---|---|---|---|---|---|---|
| Lines Improvement At One Year* | 20/25 or Better At Any Follow-up Visit† | |||||
| N | Mean | P Value | N | N (%) | P Value | |
| Gender | 0.28 | 0.09 | ||||
| Female | 42 | 3.9 | 50 | 39 (84) | ||
| Male | 58 | 3.9 | 63 | 40 (67) | ||
|
| ||||||
| Race/ethnicity | 0.11 | 0.93 | ||||
| White | 62 | 3.9 | 70 | 47 (71) | ||
| Non-white | 38 | 3.9 | 43 | 31 (79) | ||
|
| ||||||
| Age, in years | 0.17 | 0.71 | ||||
| 3 to <4 | 19 | 3.7 | 24 | 16 (75) | ||
| 4 to <5 | 26 | 4.0 | 30 | 19 (66) | ||
| 5 to <6 | 35 | 4.0 | 38 | 28 (79) | ||
| 6 to <7 | 10 | 3.1 | 11 | 9 (82) | ||
| 7 to <10 | 10 | 4.3 | 10 | 7 (70) | ||
|
| ||||||
| Strabismus | 0.84 | 0.46 | ||||
| Present | 12 | 3.7 | 15 | 12 (91) | ||
| Absent | 88 | 3.9 | 98 | 67 (72) | ||
|
| ||||||
| Hypermetropia‡, spherical equivalent in diopters | 0.94 | 0.75 | ||||
| 4.00 to <7.00 D | 29 | 3.3 | 36 | 23 (71) | ||
| >=7.00 D | 27 | 4.7 | 30 | 17 (60) | ||
|
| ||||||
| Astigmatism§ | 0.46 | 0.78 | ||||
| 2.00 to <4.00 D | 47 | 3.7 | 54 | 38 (76) | ||
| 4.00 to <6.00 D | 14 | 3.5 | 14 | 10 (71) | ||
| >=6.00 D | 2 | 5.3 | 2 | 2 (100) | ||
|
| ||||||
| Anisometropia | 0.39 | 0.56 | ||||
| 0.00 D | 21 | 4.6 | 25 | 17 (76) | ||
| >0.00 D to <0.50 D | 35 | 3.7 | 40 | 24 (62) | ||
| 0.50 to <1.00D | 33 | 3.8 | 37 | 28 (80) | ||
| 1.00 to <1.50D | 8 | 3.9 | 8 | 7 (88) | ||
| >=1.50 D | 3 | 2.2 | 3 | 3 (100) | ||
|
| ||||||
| Binocular Visual Acuity | <0.001 | <0.001 | ||||
| 20/40 to <20/50 | 49 | 3.1 | 56 | 47 (89) | ||
| 20/60 to 20/80 | 35 | 3.9 | 38 | 23 (63) | ||
| 20/100 to 20/320 | 16 | 6.3 | 19 | 9 (55) | ||
|
| ||||||
| Interocular Acuity Difference | ||||||
| 0 to <1 line | 44 | 4.1 | 0.05 | 50 | 35 (74) | 0.84 |
| 1 to <2 lines | 36 | 3.8 | 40 | 27 (72) | ||
| 2 to 3 lines | 9 | 3.9 | 11 | 10 (91) | ||
| >= 3 lines | 11 | 3.3 | 12 | 7 (64) | ||
|
| ||||||
| Stereoacuity, seconds of arc|| | 0.52 | 0.47 | ||||
| 40 | 4 | 3.7 | 4 | 3 (75) | ||
| 60 | 5 | 3.6 | 5 | 4 (80) | ||
| 100 | 4 | 3.5 | 5 | 5 (100) | ||
| 200 | 5 | 2.9 | 5 | 5 (100) | ||
| 400 | 9 | 3.2 | 10 | 6 (60) | ||
| 800 | 13 | 4.0 | 15 | 11 (79) | ||
| None detected (>800) | 57 | 4.1 | 66 | 43 (71) | ||
N is the number of children completing the study in the specified strata. P values from individual linear regression models with lines of binocular acuity improvement as the dependent variable and baseline binocular visual acuity and the factor of interest as covariates. Age, hypermetropia, astigmatism and anisometropia were evaluated as continuous data. Stereoacuity was evaluated dichotomously as >800 seconds of arc vs. 800 seconds of arc or better.
N is the number of children in the study in the specified strata (i.e., includes all children regardless of whether they completed the study). Percentages cited are cumulative probabilities derived from Kaplan-Meier product-limit method. P values from individual proportional hazards regression models the proportion of children achieving binocular acuity 20/25 or better as the dependent variable and baseline binocular visual acuity and the factor of interest as covariates. Age, hypermetropia, astigmatism, and anisometropia were evaluated as continuous data. Stereoacuity was evaluated dichotomously as >800 seconds of arc vs. 800 seconds of arc or better.
Analysis of hypermetropia is limited to children who were eligible for the study based on significant hypermetropia (spherical equivalent >= +4.00 D) in both eyes. Hypermetropia refers to the average amount of spherical equivalent between the two eyes.
Analysis of astigmatism is limited to children who were eligible for the study based on significant astigmatism (>=2.00 D) in both eyes. Astigmatism refers to the average amount of astigmatism between the two eyes.
Three children were unable to perform stereoacuity testing.
Among children who completed the study, the 34 children with significant bilateral hypermetropia only appeared to have had greater binocular acuity improvement than the 44 children with significant bilateral astigmatism only (estimated difference between groups adjusted for baseline acuity = 0.06 logMAR, P = 0.04), however this association no longer existed when data were analyzed excluding the single outlier who had 12.8 lines of improvement (estimated difference between groups adjusted for baseline acuity = 0.04 logMAR, P = 0.18). Children with significant bilateral hypermetropia only were similar to children with significant bilateral astigmatism only in the cumulative probability of reaching 20/25 or better binocular acuity over one year (69% vs. 86%, P = 0.13).
Binocular Visual Acuity Improvement in Children Also Treated with Patching and/or Atropine
Among the 13 children who were treated with spectacles and patching and/or atropine, binocular visual acuity (mean at baseline 0.53 logMAR, 20/63) improved by an average of 3.3 lines (95% CI = 2.1 to 4.4) to a mean of 0.22 (20/32) at one year. The cumulative probability of reaching binocular acuity 20/25 or better was 49% (95% CI = 25% to 78%) at one year.
Monocular Acuity Improvement
Monocular visual acuity (mean at baseline 0.57 logMAR, 20/80) improved by an average of 2.3 lines (95% CI = 2.1 to 2.6) to a mean of 0.33 logMAR (20/40) after 5 weeks and by an average of 3.9 lines (95% CI = 3.7 to 4.2) to a mean of 0.17 logMAR (20/32) at one year. At one year, 13 children (13%) had monocular acuity 20/40 or worse in both eyes and 34 children (34%) had monocular acuity 20/40 or worse in one eye.
Change in IOD
Of the 23 children who had 2 or more lines of IOD at baseline, 17 (74%) were treated with spectacles alone and 6 (26%) were treated with patching and/or atropine in addition to spectacles. Twenty of these children had IOD measured at one year, and 2 or more lines of IOD persisted after one year in 5 of the 14 (36%) treated with spectacles alone and in 2 of the 6 (33%) who also had patching and/or atropine.
Of the 90 children who had less than 2 lines of IOD at baseline, 83 (92%) were treated with spectacles alone and 7 (8%) were treated with patching and spectacles. Eighty-one of these children had IOD measured at one year, and 2 or more lines of IOD were present after one year in 10 (13%) of the 75 treated with spectacles alone and in 1 (17%) of the 6 who also had patching.
Stereoacuity
Of the 94 children who had stereoacuity tested at both baseline and the one-year examination, stereoacuity had improved a mean of 1.9 levels (95% CI = 1.4 to 2.3), with 56 children (60%) improving two levels or more (Table 4). Improvement in stereoacuity was associated with improvement in binocular visual acuity (P=0.02).
Table 4.
Stereoacuity Outcomes During One Year Of Spectacle Wear in Patients Aged 3-<10 Years with Previously Untreated Bilateral Refractive Amblyopia (N=113)
| Change in Stereoacuity From Baseline to One Year, levels* | |
| Mean (SD) | 1.9 (2.1) |
| Range | -4.0 to 6.0 |
|
| |
| Stereoacuity at One Year, in seconds of arc† | N (%) |
| 40 | 13 (13) |
| 60 | 12 (12) |
| 100 | 18 (19) |
| 200 | 11 (11) |
| 400 | 17 (18) |
| 800 | 3 (3) |
| >800 | 23 (24) |
Negative values represent decrease in stereoacuity. Change in steroacuity could not be calculated for 19 patients: 5 were unable to perform testing (3 at baseline and 2 at one year) and 14 did not have testing completed at one year.
16 children were missing stereoacuity outcomes at one year: 2 children were unable to perform testing and 14 did not have testing completed.
Discussion
In this prospective multicenter study of 113 children with previously untreated bilateral refractive amblyopia, binocular visual acuity improved an average of 3.9 lines after one year of treatment, with spectacles as the sole treatment in all but 13 children. Binocular visual acuity of 20/25 or better was achieved by 73% of children within one year of starting treatment. Although there was no untreated control group, the observed improvement substantially exceeded any expected learning or age effect.10-12 Visual acuity improvement was accompanied by a corresponding improvement in stereopsis, with 60% of children improving by at least 2 levels on the Randot Preschool Stereoacuity Test. Smaller published series of patients with similar inclusion criteria have also demonstrated improvement in bilateral refractive amblyopia. Klimek et al found that 21 of 36 children (58%) with bilateral refractive amblyopia achieved a visual acuity of 20/25 or better in at least one eye with a mean follow-up of 3.3 years.8 Schoenleber et al. reported that 10 of 12 children (83%) improved to 20/40 or better in both eyes with a mean follow up of 22 months.5 Our study had 12 months of follow up and was not designed to assess maximal improvement on treatment. Therefore, additional improvement in visual acuity and/or stereopsis could potentially occur beyond one year.
At baseline, the children in our cohort had modestly reduced visual acuity (mean of 20/63), and the acuity deficit was usually symmetrical (79% had less than 2 lines of IOD). There was an approximately equal number of children with high bilateral hypermetropia only and those with high bilateral astigmatism only. The refractive error characteristics of our cohort differs from that of other samples that include a larger proportion of children with high astigmatism such as Native Americans.13
Binocular visual acuity was our primary outcome because we felt it best represents visual function in the “real world” setting. Most children had symmetrical amblyopia at baseline, so it is not surprising that there was little difference between monocular and binocular visual acuity outcomes. Improvement of bilateral amblyopia with spectacles alone can result in resolution of amblyopia in one eye and persistent amblyopia in the other eye, requiring additional amblyopia treatment with occlusion therapy or atropine. However, only 13 of 113 children (12%) in our cohort received patching and/or atropine treatment, in contrast to 13 of 36 children (36%) in the study by Klimek et al.8 This difference may be because our protocol specifically discouraged investigators from treating with patching or atropine until the visual acuity in each eye stopped improving.
The mechanism of bilateral refractive amblyopia is presumed to be pattern vision deprivation; that is, failure of both eyes to achieve a clear foveal image results in abnormal development of the visual cortex.2 Children with uncorrected hypermetropia without significant astigmatism can generally accommodate sufficiently to achieve clear retinal images and thus avoid the development of amblyopia. It is not known why some children with uncorrected hypermetropia develop amblyopia and some do not. Perhaps reduced accommodative amplitudes, which have been found in some children with bilateral refractive amblyopia, play a role.5, 6 The relatively large amount of accommodation required for clear vision can result in the development of refractive accommodative esotropia in some children with high hypermetropia. It has also been suggested that children with bilateral hypermetropic amblyopia subconsciously choose orthophoria and bilaterally reduced vision over esotropia with diplopia.5 The presence of strabismus will generally prompt a referral to an eye doctor, whereas those children with high hypermetropia without strabismus can be detected by screening programs.14 When strabismus is absent, noncompliance with glasses can be a problem and parents may deny the need for treatment because there is no obvious disability.
In cases of bilateral amblyopia with concomitant strabismus, abnormal binocular interaction with suppression may also contribute to the development of amblyopia in the non-preferred eye. We observed strabismus in 15 of 113 children (13%) in our study, 12 of whom had esotropia. In contrast, Klimek et al. detected strabismus in 23 of 36 children (64%), 22 of whom had esotropia.8 However, their study included only children with hypermetropia >= 4.50 D, whereas we included children with astigmatism without significant hypermetropia.
In conclusion, we observed substantial improvement of binocular best-corrected visual acuity during treatment of bilateral refractive amblyopia with spectacle correction; 73% of children 3 to <10 years of age achieved visual acuity of 20/25 or better after one year. Improvements in visual acuity and stereopsis were generally achieved with spectacles alone, as only 12% of the children in our cohort received additional amblyopia treatment with patching or atropine.
Figure 2. Cumulative Probability of Binocular Visual Acuity 20/25 or Better During Follow Up of Patients with Bilateral Refractive Amblyopia (N=113).
Acknowledgments
Funding/Support: Supported through a cooperative agreement from the National Eye Institute EY11751
Financial Disclosure: none
Institutional Review Board/Informed Consent: The respective institutional review boards (IRBs) approved the protocol and HIPAA-compliant informed consent forms.
Contribution of Authors - Design of the study (DW, DC, RB, RA, EB, JF, JH, DK, GQ, MR, ST); all authors were involved in the collection, management, analysis, and interpretation of the data and the preparation, review, and final approval of the manuscript.
Writing Committee: Lead authors: David K. Wallace, M.D., M.P.H.; Danielle L. Chandler, M.S.P.H.; Roy W. Beck, M.D., Ph.D.; Additional writing committee members (alphabetical): Robert W. Arnold, M.D.; Darron A. Bacal, M.D.; Eileen E. Birch, Ph.D.; Joost Felius, Ph.D.; Marcela Frazier, O.D.; Jonathan M. Holmes, B.M., B.Ch.; Darren Hoover, M.D.; Deborah A. Klimek, M.D.; Ingryd Lorenzana. O.D.; Graham E. Quinn, M.D., M.S.C.E.; Michael X. Repka, M.D.; Donny W. Suh, M.D.; Susanna Tamkins, O.D.
The Pediatric Eye Disease Investigator Group
Clinical Sites that Participated in this Protocol
Sites are listed in order by number of patients enrolled into the study. The number of patients enrolled is noted in parenthesis preceded by the site location and the site name. Personnel are listed as (I) for Investigator, (C) for Coordinator, and (V) for Visual Acuity Tester.
Cranberry TWP, PA - Everett and Hurite Ophthalmic Association (27)
Darren L. Hoover (I); Pamela A. Huston (C); Joan M. Addison (V); Barbara R. Fuchs (V); Jody L. Parker (V); Pamela M. Racan (V)
Birmingham, AL - University of Alabama at Birmingham School of Optometry (16)
Robert P. Rutstein (I); Marcela Frazier (I); Kristine T. Hopkins (I); Wendy L. Marsh-Tootle (I); Katherine K. Weise (I); Cathy H. Baldwin (C); Sophocles Sophocleous (V)
Erie, PA - Pediatric Ophthalmology of Erie (16)
Nicholas A. Sala (I); Rhonda M. Hodde (C); Veda L. Zeto (C)
Miami, FL - Bascom Palmer Eye Institute (11)
Susanna M. Tamkins (I); Eva M. Olivares (C); Bruce D. Bailey (V); Alexis H Strauss (V); Ana C. Rosa (C); Elias J. Silverman (C); Hannah Park (V)
Streamwood, IL - Advanced Vision Center (10)
Ingryd J. Lorenzana (I); Elva R. Banuelos (V); Manuela Villagrana (V)
Anchorage, AK - Ophthalmic Associates (6)
Robert W. Arnold (I); Mary Diane Armitage (C)
Dallas, TX - Pediatric Ophthalmology, P.A. (5)
David R. Stager, Sr. (I); David R. Stager, Jr. (I); Joost Felius (C); Mary K. Alters (C); Alexia B. Perez (V)
Milford, CT - Eye Physicians & Surgeons, PC (5)
Darron A. Bacal (I); Marla Doheny (C); Donna Martin (C)
South Charleston, WV - Children's Eye Care & Adult Strabismus Surgery (5)
Deborah L. Klimek (I); Lisa L. Winter (C); Bounthavy Lisa Greenlee (V)
West Des Moines, IA - Wolfe Clinic (5)
Donny W. Suh (I); Marilee McNeece (C); Jan M. Jansen (C); Shannon L. Craig (V); Rhonda J. Swisher (V); Laura J. Watkin (V); Kimberly Walters (C)
Memphis, TN - Southern College of Optometry (4)
Erin R. Nosel (I); Kristin K. Anderson (I); Sean M. Skierczynski (C); Christopher W. Lievens (V)
Milwaukee, WI - Medical College of Wisconsin* (4)
Jane D. Kivlin (I); Mark S. Ruttum (I); Veronica R. Picard (C)
Rochester, NY - University of Rochester Eye Institute* (4)
Matthew D. Gearinger (I); Nancy M. Fedick (C); Lynne M. Addams (V)
Waterbury, CT - Eye Care Group, PC (4)
Andrew J. Levada (I); Tabitha L. Matchett (C); Cheryl Schleif (C); Shelley Klein-Weiss (V); Gina Silva (V)
Fullerton, CA - Southern California College of Optometry (3)
Susan A. Cotter (I); Carmen N. Barnhardt (I); Raymond H. Chu (I); Kristine Huang (I); Monique M. Nguyen (I); Susan M. Shin (I); Erin Song (I); Sue M. Parker (C); Jamie H. Morris (C)
Lancaster, PA - Family Eye Group (2)
David I. Silbert (I); Don D. Blackburn (I); Eric L. Singman (I); Noelle S. Matta (C)
Portland, OR - Pacific University College of Optometry (2)
Richard London (I); James J. Kundart (I); Jayne L. Silver (C); Cheryl L. Bruce (C)
Rochester, MN - Mayo Clinic* (2)
Jonathan M. Holmes (I); Brian G. Mohney (I); Melissa L. Rice (I); Rebecca A. Nielsen (C); David A. Leske (C); Tracee L. Shevlin (C); Debbie M, Priebe (V); Virginia Karlsson (V)
Brooklyn, NY - SUNY Downstate Medical Center (1)
Janine N. Smith (I)
Chapel Hill, NC - UNC Dept. of Ophthalmology (1)
David K. Wallace (I); John David Wright, Jr. (I); Melissa W. Compton (V); Madonna R. Petty (V)
Dallas, TX - Pediatric Ophthalmology, P.A. (1)
Priscilla M. Berry (I); Mary K. Alters (C); Joost Felius (C); June M. Gartlir (V)
Fall River, MA - Center for Eye Health Truesdale Clinic (1)
John P. Donahue (I); Mary E. Silvia (C); Deborah P. Branco (V); Marisa F. Sousa (V); Christine J. Bazinet (V)
Ft. Lauderdale, FL - Nova Southeastern University College of Optometry, The Eye Institute (1)
Yin C. Tea (I); Marc B. Taub (I); Deborah M. Amster (I); Annette Bade (C); Jacqueline Rodena (V)
Minneapolis, MN - University of Minnesota* (1)
C. Gail Summers (I); Erick D. Bothun (I); Stephen P. Christiansen (I); Ann M. Holleschau (C); Sara J. Downes (V); Kathy M. Hogue (V); Kim S. Merrill (V)
Providence, RI - Rhode Island Eye Institute (1)
John P. Donahue (I); Nicole L. Waterman (V)
Saint Paul, MN - Associated Eye Care (1)
Susan Schloff (I); Ann Hickson (I); Valori E. Olson (C)
Sharon, MA - Daniel M. Laby, M.D. (1)
Daniel M. Laby (I); Ricky Laby (C)
Amblyopia Treatment Study Steering Committee: Roy W. Beck, Eileen E. Birch, Susan A. Cotter, Sean P. Donahue (2005), Donald F. Everett, Allison R. Edwards (2005-2006), Stephen R. Glaser (2006), Richard W. Hertle, Rhonda Hodde (2005), Jonathan M. Holmes, Pamela A. Huston (2006), Deborah L. Klimek (2006), Don W. Lyon, Noelle S. Matta (2004), Pamela S. Moke (2004), Graham E. Quinn (2004), Michael X. Repka, Nicholas A. Sala (2005), Mitchell M. Scheiman, David K. Wallace, David R. Weakley
PEDIG Coordinating Center - Tampa, FL: Roy W. Beck, Brian D. Becker, Gladys N. Bernett, Nicole M. Boyle, Christina M. Cagnina-Morales, Debora A. Cagnina, Esmeralda L. Cardosa, Danielle L. Chandler, Laura E. Clark, Sharon R. Constantine, Quayleen Donahue, Mitchell Dupre, Allison R. Edwards, Kyle L. Baccus Horsley, Heidi A. Gillespie, Raymond T. Kraker, Stephanie V. Lee, Lee Anne Lester, Shelly T. Mares, Holly McCombs, B. Michele Melia, Pamela S. Moke, Stephanie Morgan-Bagley, Jim L. Pyner, Heidi J. Strayer, Dana Williams
National Eye Institute - Bethesda, MD: Donald F. Everett
PEDIG Executive Committee: Roy W. Beck, Eileen E. Birch, Susan A. Cotter, Sean P. Donahue (2005-2006), Donald F. Everett, Jonathan M. Holmes, Raymond T. Kraker (2005-Current), Pamela S. Moke (2004), Michael X. Repka, Nicholas A. Sala (2006), David K. Wallace (2006)
Footnotes
The writing committee and a list of the members of the Pediatric Eye Disease Investigator Group (PEDIG) participating in the study appear in the acknowledgements.
Center received support utilized for this project from an unrestricted grant from Research to Prevent Blindness Inc., New York, New York.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Haase W, Muhlig HP. The incidence of squinting in school beginners in Hamburg (author's translation) Klin Monatsbl Augenheikd. 1979;174:232–235. [PubMed] [Google Scholar]
- 2.von Noorden GK. Binocular vision and ocular motility: Theory and management of strabismus. 5th. St Louis; Mo: Mosby: 1996. [Google Scholar]
- 3.Fern KD. Visual acuity outcome in isometropic hyperopia. Optom & Vis Sci. 1989;66:649–658. doi: 10.1097/00006324-198910000-00001. [DOI] [PubMed] [Google Scholar]
- 4.Cavazos H, Haase W, Meyer E. Ametropic amblyopia. Strabismus. 1993;1:63–67. doi: 10.3109/09273979309087719. [DOI] [PubMed] [Google Scholar]
- 5.Schoenleber DB, Crouch ER. Bilateral hypermetropic amblyopia. J Pediatr Ophthalmol Strabismus. 1987;24:75–77. doi: 10.3928/0191-3913-19870301-06. [DOI] [PubMed] [Google Scholar]
- 6.Werner DB, Scott WE. Amblyopia case reports-bilateral hypermetropic ametropic amblyopia. J Pediatr Ophthalmol Strabismus. 1985;22:203–205. doi: 10.3928/0191-3913-19850901-09. [DOI] [PubMed] [Google Scholar]
- 7.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 Ped Ophthalmol Strabismus. 1985;22:54–57. doi: 10.3928/0191-3913-19850301-05. [DOI] [PubMed] [Google Scholar]
- 8.Klimek DL, Cruz OA, Scott WE, et al. Isoametropic amblyopia due to high hyperopia in children. J AAPOS. 2004;8:310–313. doi: 10.1016/j.jaapos.2004.05.007. [DOI] [PubMed] [Google Scholar]
- 9.Haase W. Visual acuity in cases of monocular and bilateral amblyopia: treatment during school age. Metabolic Ophthalmology. 1978;2:147–148. [Google Scholar]
- 10.Holmes JM, Beck RW, Repka MX, et al. The amblyopia treatment study visual acuity testing protocol. Arch Ophthalmol. 2001;2001;119:1345–1353. doi: 10.1001/archopht.119.9.1345. [DOI] [PubMed] [Google Scholar]
- 11.Cotter SA, Chu RH, Chandler DL, et al. Reliability of the Electronic Early Treatment Diabetic Retinopathy Study testing protocol in children 7 to <13 years old. Am J Ophthalmol. 2003;136(4):655–661. doi: 10.1016/s0002-9394(03)00388-x. [DOI] [PubMed] [Google Scholar]
- 12.Moke PS, Turpin AH, Beck RW, et al. Computerized method of visual acuity testing: adaptation of the amblyopia treatment study visual acuity testing protocol. Am J Ophthamol. 2001;132:903–909. doi: 10.1016/s0002-9394(01)01256-9. [DOI] [PubMed] [Google Scholar]
- 13.Harvey EM, Dobson V, Miller JM. Prevalence of High Astigmatism, Eyeglass Wear, and Poor Visual Acuity Among Native American Grade School Children. Optometry & Vision Science. 2006;83:206–212. doi: 10.1097/01.opx.0000214333.84822.71. [DOI] [PubMed] [Google Scholar]
- 14.Arnold RW, Donahue SP. The yield and challenges of charitable state-wide photoscreening. Binocul Vis Strabismus Q. 2006;21(2):93–100. [PubMed] [Google Scholar]