Full-time Bangerter Filters Versus Part-time Daily Patching for Moderate Amblyopia in Children (ATS10)

A Randomized Trial of Full-time Bangerter Filters Versus Part-time Daily Patching for the Treatment of Moderate Amblyopia in Children

This study is a randomized clinical trial designed to evaluate the non-inferiority of Bangerter filters compared to 2 hours of daily patching as a primary treatment for moderate amblyopia (20/40 to 20/80) in children ages 3 to < 10 years. Secondary objectives of this study are (1) to determine the time course of visual improvement with Bangerter filter treatment, (2) to compare patient quality of life, measured by a modified Amblyopia Treatment Index, between patients treated with patching vs. Bangerter filters, (3) to determine whether blurring the sound eye to a visual acuity worse than the amblyopic eye predicts improvement in acuity, and (4) to determine whether a change in fixation to the amblyopic eye is predictive of improvement in visual acuity. The primary outcome assessment is visual acuity at 24 weeks for both the amblyopic and sound eyes. The primary analytic approach for the amblyopic eye acuity will involve construction of a one-sided 95% confidence interval to assess non-inferiority based on a treatment group comparison of logMAR visual acuity scores adjusted for baseline visual acuity scores in an analysis of covariance (ANCOVA) model. Sound eye acuity data will be reported for each treatment regimen at the 24-week visit as mean change (logMAR lines) from baseline and as the distribution of the number of lines of change from baseline.

Amblyopia is the most common cause of monocular visual impairment in children, estimated to affect as many as 3.6% of the childhood population. The natural history of amblyopia is relatively unknown although it has been reported that visual acuity may deteriorate further without treatment. Although occlusion or patching of the sound eye has been the mainstay for amblyopia therapy, alternative treatment such as pharmacological or optical penalization may be as effective. In a randomized, controlled clinical trial of 419 children, 3 years to less than 7 years old with moderate amblyopia, patching was compared to atropine. Although improvement with atropine was initially slower, both treatments produced similar improvement after 6 months. Although both patching and atropine have been proven effective for treating amblyopia, neither treatment is without adverse side effects. Patching is associated with compliance difficulties, the need for continuous monitoring, and social stigma. Negative side effects observed in children treated with atropine include light sensitivity, facial flushing, and fever. In a randomized clinical trial comparing patching to atropine as a treatment for amblyopia, a questionnaire to assess the impact of patching and atropine treatment on the child and family indicated that both treatments were well tolerated overall, however, patching had lower compliance and higher social stigma than atropine. Bangerter filters, also known as Bangerter foils, have been used mainly as secondary amblyopia therapy following patching or atropine to either further improve or maintain the visual gain. One advantage of Bangerter filters compared to patching is that the lower density filters are not readily apparent and therefore would be expected to increase patient compliance due to reduced social stigma. Another advantage of Bangerter filters is that there is no opportunity for skin irritation from bandage adhesive, a commonly-reported side effect of patching. In addition, there is a theoretical advantage that Bangerter filters are less disruptive to binocular function during treatment compared to other modalities such as patching. Few data are available comparing Bangerter filters with patching for the treatment of amblyopia. Bonsall randomized 14 patients, 3 to 10 years old, with previously untreated strabismic/anisometropic amblyopia to either 6 hours of daily patching or full-time Bangerter filters. Baseline amblyopic eye acuity was 20/30 to 20/400 for the patching group and 20/30 to 20/200 for the Bangerter group. The Bangerter filter prescribed was the minimum density foil needed to elicit a switch in fixation from the sound eye to the amblyopic eye. Visual acuity was measured every 6-8 weeks until the amblyopic eye visual acuity was equal to that of the sound eye, an improvement that was achieved in 5 of the 14 at the time the study was stopped. The average time to achieve equal vision between the amblyopic and sound eyes was about 4.5 months (142 days) for the foil group versus about 9 months (272 days) for the patching group. Both forms of therapy were equally tolerated. Despite good preliminary data, a large randomized clinical trial comparing the effectiveness of Bangerter filters to patching for the treatment of amblyopia has yet to be conducted.

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) visual acuity testing protocol resulting in a Snellen acuity score that can range from 20/16 to 20/800 for ages 3 to <7; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds which resulted in a letter score that could range from 0 to 97 letters, with 0 being the worst and 97 being the best. Scores were converted to log of minimum angle of resolution (logMAR) equivalents for analyses (lower logMAR value is better than higher logMAR).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) visual acuity testing protocol resulting in a Snellen acuity score that can range from 20/16 to 20/800 for ages 3 to <7; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds which resulted in a letter score that could range from 0 to 97 letters, with 0 being the worst and 97 being the best. Scores were converted to log of minimum angle of resolution (logMAR) equivalents for analyses (lower logMAR value is better than higher logMAR).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). 'Worse' indicates acuity at 24 weeks is worse than acuity at baseline; 'Better' indicates acuity at 24 weeks is better than acuity at baseline.

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). Change from baseline to 24 weeks was calculated. A positive difference indicates improvement (one logMAR line = 5 letters or one Snellen line).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). A positive interocular difference indicates worse acuity in the amblyopic eye (one logMAR line = 5 letters or one Snellen line).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). A positive interocular difference indicates worse acuity in the amblyopic eye (one logMAR line = 5 letters or one Snellen line).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) visual acuity testing protocol resulting in a Snellen acuity score that can range from 20/16 to 20/800 for ages 3 to <7; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds which resulted in a letter score that could range from 0 to 97 letters, with 0 being the worst and 97 being the best. Scores were converted to log of minimum angle of resolution (logMAR) equivalents for analyses (lower logMAR value is better than higher logMAR).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). Change from baseline to 24 weeks was calculated. A positive difference indicates improvement (one logMAR line = 5 letters or one Snellen line).

The distribution of the number of participants in each patient characteristic category at the 24-week outcome examination was found (for example, the number of participants at 24 weeks who were 3 to <5 years old at the time of enrollment).

Mean and SD of Change in Visual Acuity in the Amblyopic Eye From Baseline to 24-Week Outcome Examination According to Patient Characteristics

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). Change from baseline to 24 weeks was calculated. A positive difference indicates improvement (one logMAR line = 5 letters or one Snellen line).

Distribution of Change in Randot Preschool Stereoacuity From Baseline to 24-Week Outcome by Treatment Group: All Subjects

The Randot Preschool Stereotest measures stereopsis from 800 to 40 seconds of arc. It is designed as a matching game in which the patient matches pictures in a test booklet wearing special glasses. A subject can fail the pretest (not see any pictures) or can score >800 (the worst), 800, 400, 200, 100, 60, or 40 (the best). If two shapes are identified correctly the patient moves to the next lower level. A failed test occurs when the patient cannot identify any shapes. A change of 1 level is a movement of 1 step in the scale (decrease shows improvement - ex. 100 to 60 is 1 level improved).

Distribution of Change in Randot Preschool Stereoacuity From Baseline to 24-week Outcome by Treatment Group: Subjects With Anisometropia and No Strabismus

The Randot Preschool Stereotest measures stereopsis from 800 to 40 seconds of arc. It is designed as a matching game in which the patient matches pictures in a test booklet wearing special glasses. A subject can fail the pretest (not see any pictures) or can score >800 (the worst), 800, 400, 200, 100, 60, or 40 (the best). If two shapes are identified correctly the patient moves to the next lower level. A failed test occurs when the patient cannot identify any shapes. A change of 1 level is a movement of 1 step in the scale (decrease shows improvement - ex. 100 to 60 is 1 level improved).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). Change from baseline to 24 weeks was calculated. A positive difference indicates improvement (one logMAR line = 5 letters or one Snellen line).

Visual acuity was measured in each eye using the Amblyopia Treatment Study (ATS) testing protocol resulting in a Snellen acuity score for 3 to <7 year olds; or with the electronic early treatment diabetic retinopathy study (E-ETDRS) method for 7 to <10 year olds, resulting in a letter score that could range from 0 to 97 letters (0 worst; 97 best). Scores were converted to log of minimum angle of resolution (logMAR)(lower logMAR indicates better score). Change from baseline to 24 weeks was calculated. A positive difference indicates improvement (one logMAR line = 5 letters or one Snellen line).

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family. Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores were summed and averaged for each individual (total sums could range from 0 to 90; means could range from 0 to 5). A mean across all individuals was computed from the individual means.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family. Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores were summed and averaged for each individual (total sums could range from 0 to 90; means could range from 0 to 5). A mean across all individuals was computed from the individual means.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (8 of these questions pertain to adverse effects of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 8 questions were summed and averaged for each individual (total sums could range from 0 to 40). A mean across all individuals was computed.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (8 of these questions pertain to adverse effects of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 8 questions were summed and averaged for each individual (total sums could range from 0 to 40). A mean across all individuals was computed.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (7 of these questions pertain to compliance of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 7 questions were summed and averaged for each individual (total sums could range from 0 to 35). A mean across all individuals was computed.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (7 of these questions pertain to compliance of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 7 questions were summed and averaged for each individual (total sums could range from 0 to 35). A mean across all individuals was computed.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (2 of these questions pertain to social stigma of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 2 questions were summed and averaged for each individual (total sums could range from 0 to 10). A mean across all individuals was computed.

The Parental Amblyopia Treatment Index consists of 18 likert-type questions evaluating the impact of treatment on the patient and family (2 of these questions pertain to social stigma of treatment). Questions are answered on a scale from 'Strongly Disagree' to 'Strongly Agree.' For analysis, values were coded numerically with integers from 5 (strongly agree) to 1 (strongly disagree) (0 was assigned to 'non-applicable' answers). The scores for these 2 questions were summed and averaged for each individual (total sums could range from 0 to 10). A mean across all individuals was computed.

Inclusion Criteria: Age 3 to < 10 years Amblyopia associated with strabismus, anisometropia, or both No ocular cause apparent for reduced visual acuity Visual acuity 20/40 to 20/80 (71 to 54 letters inclusive) in amblyopic eye Visual acuity 20/40 or better (>= 69 letters) in sound eye Interocular difference >= 3 logMAR lines (>= 15 letters) No amblyopia treatment other than spectacles in last 6 months *Any treatment more than 6 months prior to enrollment is acceptable Currently wearing spectacles Appropriate spectacles have been worn for 16 weeks prior to enrollment or visual acuity documented to be stable No myopia > -6.00 D spherical equivalent in either eye Cycloplegic refraction within 6 months prior to enrollment Ocular examination within 6 months prior to enrollment Exclusion Criteria: Current vision therapy or orthoptics Ocular cause for reduced visual acuity Prior intraocular or refractive surgery Known skin reactions to patch or bandage adhesives

Pediatric Eye Disease Investigator Group Writing Committee; Rutstein RP, Quinn GE, Lazar EL, Beck RW, Bonsall DJ, Cotter SA, Crouch ER, Holmes JM, Hoover DL, Leske DA, Lorenzana IJ, Repka MX, Suh DW. A randomized trial comparing Bangerter filters and patching for the treatment of moderate amblyopia in children. Ophthalmology. 2010 May;117(5):998-1004.e6. doi: 10.1016/j.ophtha.2009.10.014. Epub 2010 Feb 16.

Rutstein RP, Foster NC, Cotter SA, Kraker RT, Lee DH, Melia M, Quinn GE, Tamkins SM, Wallace DK; Pediatric Eye Disease Investigator Group. Visual acuity through Bangerter filters in nonamblyopic eyes. J AAPOS. 2011 Apr;15(2):131-4. doi: 10.1016/j.jaapos.2010.11.015. Epub 2011 Mar 21.

Wallace DK, Lazar EL, Melia M, Birch EE, Holmes JM, Hopkins KB, Kraker RT, Kulp MT, Pang Y, Repka MX, Tamkins SM, Weise KK; Pediatric Eye Disease Investigator Group. Stereoacuity in children with anisometropic amblyopia. J AAPOS. 2011 Oct;15(5):455-61. doi: 10.1016/j.jaapos.2011.06.007.