A Randomized Trial Comparing Part-time

Patching with Observation for Intermittent

Exotropia in Children 12 to 35 Months of Age
Pediatric Eye Disease Investigator Group*
Writing Committee: Brian G. Mohney, MD,1 Susan A. Cotter, OD, MS,2 Danielle L. Chandler, MSPH,3
Jonathan M. Holmes, BM, BCh,1 Angela M. Chen, OD, MS,2 Michele Melia, ScM,3 Sean P. Donahue, MD, PhD,4
David K. Wallace, MD, MPH,5 Raymond T. Kraker, MSPH,3 Melanie L. Christian, COA,6 Donny W. Suh, MD7

Purpose: To determine the effectiveness of part-time patching for treating intermittent exotropia (IXT) in
young children.
Design: Multicenter, randomized clinical trial.
Participants: Two hundred one children 12 to 35 months of age with untreated IXT meeting the following
criteria: (1) IXT at distance OR constant exotropia at distance and either IXT or exophoria at near, and (2) 15-prism
diopter (D) or more exodeviation at distance or near by prism and alternate cover test (PACT) but at least 10 D
exodeviation at distance by PACT.
Methods: Participants were assigned randomly to either observation (no treatment for 6 months) or patching
prescribed for 3 hours daily for 5 months, followed by 1 month of no patching.
Main Outcome Measures: The primary outcome was deterioration, defined as constant exotropia
measuring at least 10 D at distance and near or receipt of nonprotocol treatment for IXT.
Results: Of the 177 participants (88%) completing the 6-month primary outcome examination, deterioration
occurred in 4.6% (4 of 87) of the participants in the observation group and in 2.2% (2 of 90) of the participants in
the patching group (difference, 2.4%; P ¼ 0.27; 95% confidence interval, 3.8% to þ9.4%). Motor deterioration
occurred in 2.3% (2 of 87) of the observation group and in 2.2% (2 of 90) of the patching group (difference, 0.08%;
P ¼ 0.55; 95% confidence interval, 5.8% to þ6.1%). For the observation and patching groups, respectively,
6-month mean PACT measurements were 27.9 D versus 24.9 D at distance (P ¼ 0.02) and 19.3 D versus 17.0 D at
near (P ¼ 0.10); 6-month mean exotropia control scores were 2.8 versus 2.3 points at distance (P ¼ 0.02) and 1.4
versus 1.1 points at near (P ¼ 0.26).
Conclusions: Among children 12 to 35 months of age with previously untreated IXT, deterioration over
6 months was uncommon, with or without patching treatment. There was insufficient evidence to recommend
part-time patching for the treatment of IXT in children in this age group. Ophthalmology 2015;-:1e8 ª 2015 by
the American Academy of Ophthalmology.

*Supplemental material is available at www.aaojournal.org.

Intermittent exotropia (IXT), one of the most prevalent forms use of patching for very young children with IXT, we con-
of childhood strabismus,1e4 is characterized by an intermit- ducted a randomized trial of children 12 to 35 months of age
tent outward deviation of one or both eyes, often exacerbated to assess the effectiveness of part-time patching compared
by fatigue, inattention, or illness. Although a common con- with simple observation for reducing the risk of deterioration
dition, the best treatment and optimal timing for this disorder of IXT over a 6-month period.
remain unclear,5,6 particularly for young children who often
are unable to cooperate for a sensorimotor eye examination or
for some nonsurgical treatments such as orthoptic exercises. Methods
Given these challenges, part-time patching commonly is
The study was supported through a cooperative agreement with the
prescribed in young children as a temporizing measure7e10 to
National Eye Institute of the National Institutes of Health and was
delay potential surgery or until other forms of nonsurgical conducted by the Pediatric Eye Disease Investigator Group at 60
treatment become feasible. The reported potential benefits of clinical sites according to the tenets of the Declaration of Helsinki.
patching include the preservation of binocularity and a The protocol and Health Insurance Portability and Accountability
reduction in the frequency and/or magnitude of the Actecompliant informed consent forms were approved by insti-
exodeviation.11e19 Given the paucity of data regarding the tutional review boards, and a parent or guardian of each study

 2015 by the American Academy of Ophthalmology http://dx.doi.org/10.1016/j.ophtha.2015.04.025 1

Published by Elsevier Inc. ISSN 0161-6420/15
 Ophthalmology Volume -, Number -, Month 2015

Table 1. Eligibility Criteria

The following criteria must have been met for the patient to be enrolled in the study:
1. Age 12e35 months
2. IXT, a manifest deviation meeting all of the following criteria:
 IXT at distance OR constant exotropia at distance and either IXT or exophoria at near
 Exodeviation at least 15 D at distance or near measured by PACT, but at least 10 D at distance measured by PACT
3. No previous surgical or nonsurgical treatment for IXT other than refractive correction (e.g., vergence therapy, patching,
vision therapy or orthoptics, or deliberate over-minus with spectacles more than 0.50 D)
4. No vision therapy or orthoptics for any reason within the previous year
5. No previous amblyopia treatment other than refractive correction within 1 year
6. Investigator not planning to initiate amblyopia treatment
7. No hyperopia more than þ3.50 D SE in either eye
8. No myopia more than 6.00 D SE in either eye
9. Patients must be wearing refractive correction (spectacles or contact lenses) for at least 1 week if refractive error (based on cycloplegic refraction
performed within 6 months) meets any of the following:
 Myopia >0.50 D SE in either eye
 Anisometropia >1.00 D SE
 Astigmatism in either eye >2.00 D
Refractive correction must have met the following guidelines:
 Anisometropia SE within 0.25 D of the full anisometropic difference correction
 Astigmatism cylinder within 0.25 D of full correction and axis within 5
 For hyperopia and myopia, the spherical component can be reduced by investigator discretion provided reduction is symmetrical and results in
residual (i.e., uncorrected) SE refractive error that does not exceed þ3.50 D SE hyperopia or 0.50 D SE myopia
 Deliberate over-minus using refractive correction with more than 0.50 D of over-minus was not allowed; however, not prescribing the full
cycloplegic hyperopic correction (i.e., prescribing reduced plus) was not considered the same as over-minus for this protocol and was
therefore allowed
10. No atropine use within the previous week
11. Gestational age >34 weeks
12. Birth weight >1500 g
13. Investigator willing to observe the IXT untreated for 3 years unless specific deterioration criteria were met; investigator also willing to forgo
extraocular muscle surgery for the first 3 months in all cases, and from 3 months to 3 years unless specific deterioration criteria were met
14. Patient, parents, or both understood protocol, was willing to accept randomization to either observation or patching, and was willing to accept that
surgical or other nonsurgical treatment (other than patching in the patching group) of IXT would not be offered by the investigator unless
specific deterioration criteria were met
15. Parent had home phone (or access to phone) and was willing to be contacted by Jaeb Center staff
16. Relocation outside of area of an active PEDIG site within next 3 years not anticipated
17. No limitation of ocular rotations because of restrictive or paretic strabismus
18. No craniofacial malformations affecting the orbits
19. No prior strabismus surgery or botulinum injection, intraocular surgery, or refractive surgery
20. No ocular disorders that would reduce visual acuity (except refractive error)
21. No known skin reactions to patch or bandage adhesives
22. No strabismus surgery planned
23. No significant neurologic impairment such as cerebral palsy; patients with mild speech delays and/or common reading or learning disabilities
were not excluded
24. Investigator not planning to change refractive correction at this time

D ¼ diopter; IXT ¼ intermittent exotropia; PACT ¼ prism and alternate cover test; PEDIG ¼ Pediatric Eye Disease Investigator Group; SE ¼ spherical
equivalent; D ¼ prism diopter.

participant gave written informed consent. An independent data applicable). For study eligibility, the IXT had to meet the following
and safety monitoring committee provided oversight. The study is criteria: (1) IXT at distance OR constant exotropia at distance and
listed on www.clinicaltrials.gov (identifier, NCT01032330; either IXT or exophoria at near, and (2) 15 prism diopters (D) or
accessed February 12, 2015), and the full protocol is available at more of exodeviation (tropia or phoria) at distance or near by prism
www.pedig.net (accessed February 12, 2015). and alternate cover test (PACT), but at least 10 D exodeviation at
This 6-month randomized trial evaluated the short-term effect distance by PACT. Table 1 lists all eligibility criteria.
of part-time patching treatment compared with observation in
children 12 to 35 months of age. This report represents the first Treatment Regimens
phase of an ongoing 3-year study that also includes children from
3 to younger than 11 years20 and also aims to assess the long-term Data were entered on the Pediatric Eye Disease Investigator Group
natural history of IXT in the observation group. Relevant portions website and participants were assigned randomly (using a permu-
of the protocol are summarized below. tated block design stratified by site) with equal probability to either
observation or 3 hours of daily patching. Participants randomly
Eligibility Criteria assigned to observation received no treatment (other than refractive
correction, if needed) for 6 months. Participants randomly assigned
The study included children 12 to 35 months of age who had no to patching were prescribed patching (choice of eye or whether
prior treatment for IXT other than refractive correction (if alternating, at investigator discretion) for 3 hours daily for 5 months

2
 Mohney et al 
Patching vs. Observation for Exotropia

(in addition to refractive correction, if needed), followed by a 1- The primary analysis was a treatment group comparison of the
month washout period of no patching to help eliminate treatment proportion of participants with deterioration occurring within
effects maintained only during patching. Parents recorded the 6 months of randomization using a 1-sided Barnard’s test.22 The
number of patching hours completed per day on calendar logs that treatment group difference in the proportion with deterioration
were brought to the 6-month visit for review. Investigators were was estimated, along with a 2-sided exact 95% confidence inter-
discouraged from prescribing nonstudy treatment for IXT (treatment val to provide both an upper and lower limit on the potential
other than patching in the patching group or any treatment in the magnitude of difference. The primary analysis was limited to
observation group) before a participant met formal protocol- participants who completed cover-and-uncover and SPCT testing
specified motor deterioration criteria; however, exceptions were at the 6-month outcome examination. The primary analysis
permitted in cases of overwhelming parental concern. included 3 participants later found to be ineligible (all 3 in the
observation group). An alternative analysis used baseline data (age,
Testing Procedures and Follow-up Visits constant exotropia at distance, and distance SPCT magnitude) to
impute 6-month data by multiple imputation with the logistic
Follow-up consisted of a visit 3 months (2 weeks) after regression method23 for the 11 participants who missed the 6-
randomization and a primary outcome examination 6 months month outcome examination. Additional analyses adjusting the
(1 month) after randomization. At each follow-up visit, a study- primary outcome for potential confounding were performed using
certified examiner (pediatric ophthalmologist, pediatric optome- binomial regression or logistic regression when the binomial
trist, or certified orthoptist) masked to the participant’s treatment models did not converge.
group always assessed exotropia control first, followed later by An additional analysis was conducted using a post hoc alter-
ocular alignment at distance (6 m) and near (30 cm). Exotropia native definition of deterioration that counted the IXT as deterio-
control was measured using the Office Control Score,21 which rated only if the motor deterioration criteria were met (i.e.,
ranges from 0 (phoria, best control) to 5 (constant exotropia, regardless of whether nonstudy treatment was started without
worst control). Control levels 3 to 5 were assigned based on the meeting motor deterioration criteria). Secondary 6-month out-
proportion of time that an exotropia was present during a comes of exotropia control and deviation magnitude by PACT
30-second observation period before any dissociation. If no exo- were evaluated using analysis of covariance models adjusting for
tropia was observed, control levels 2 to 0 were assigned based on the baseline level of the outcome. Additional analyses adjusting
the longest time taken for re-establishment of fusion after 3 secondary outcomes for potential confounding were performed
consecutive 10-second periods of dissociation. After control using adjusted analysis of covariance models. All analyses
testing, ocular alignment was assessed at distance and near using followed the intention-to-treat principle. Analyses were conducted
the cover-and-uncover test, simultaneous prism and cover test using SAS software version 9.3 (SAS Inc, Cary, NC).
(SPCT), and PACT. If motor deterioration criteria appeared to be
met, the masked examiner retested ocular alignment after a
10-minute break to confirm motor deterioration.
At each follow-up visit, investigators judged patching compli- Results
ance based on parental discussion and review of study calendars.
Compliance was classified based on percentage of prescribed Baseline Characteristics
treatment completed (did not specify whether applied to percentage Between January 2010 and September 2012, 201 children were
of hours or percentage of days) as excellent (>75%), good (51%e enrolled at 60 sites, with 97 participants assigned to observation
75%), fair (26%e50%), or poor (25%). and 104 to patching (Fig 1). Average age was 24.66.6 months,
124 participants (62%) were female, and 125 participants (62%)
Primary Outcome were white. Intermittent exotropia was classified as basic type
(i.e., distance and near exodeviations within 10 D by PACT) in
The primary outcome measure was deterioration of the IXT within 132 participants (66%). At baseline, the observation group had
6 months after randomization. Motor deterioration was defined as a slightly worse control scores at distance (2.5 vs. 2.3 points) and
constant exotropia of 10 D or more at distance and near by SPCT, near (1.2 vs. 1.0 points) than the patching group and was more
confirmed by a retest, at either the 3- or 6-month visit. A constant likely to have constant exotropia at distance (9% vs. 3%); other
tropia was defined as a manifest tropia present 100% of the time baseline characteristics seemed similar (Tables 2 and 3).
during the examination, determined by cover-and-uncover tests
performed at least 3 different times during the examination,
including one before any dissociation. A participant’s IXT also was Visit Completion
classified as deteriorated for the primary analysis if nonstudy
treatment for IXT (i.e., any treatment in the observation group; any The 3-month visit was completed by 85% (82 of 97) and 87% (90 of
treatment other than patching in the patching group) was prescribed 104) of the observation and patching groups, respectively. The
without motor deterioration criteria being met. 6-month primary outcome visit was completed by 90% (87 of 97) and
by 87% (90 of 104) of participants, respectively. A masked examiner
Statistical Methods assessed the primary outcome in all but 3 cases. Comparing the
baseline data between the 177 participants who completed the
The original sample size of 336 subjects was chosen for the long- 6-month masked examination and the 24 participants who did not
term primary analysis 3 years after randomization. For the current complete the examination (10 in the observation group, 14 in the
6-month treatment group comparison, this sample size had 92% patching group) showed that age was 2.1 versus 1.8 years, white non-
power with a 1-sided type I error rate of 5% to detect a difference, Hispanic or Latino ethnicity was 64% versus 50%, mean distance
given expected rates of deterioration of 15% in the observation control score was 2.4 versus 2.1 points, mean near control score was
group versus 5% in the patching group. Recruitment ended when 1.1 versus 1.1 points, mean distance PACT results were 27.1 D
the study’s primary cohort (children 3 to younger than 11 years) versus 29.0 D, and mean near PACT results were 17.8 D versus 26.0
met its recruitment goal, at which time 201 children 12 to 35 D. The reasons for not completing the 6-month primary outcome
months of age had been recruited. were similar for both treatment groups (Fig 1).

3
 Ophthalmology Volume -, Number -, Month 2015

Table 2. Baseline Demographic and Clinical Characteristics by

Treatment Group

Observation Patching
(n [ 97) (n [ 104)
Characteristics No. % No. %
Female gender 58 60 66 63%
Race or ethnicity
White 65 67 60 58
Black 13 13 14 13
Hispanic or Latino 13 13 21 20
Other 6 6 9 9
Age at randomization (mos)
12e<18 22 23 23 22
18e<24 28 29 19 18
24e<30 24 25 34 33
30e<36 23 24 28 27
Mean (SD) 24.2 (6.5) 24.9 (6.8)
Range 12.4e35.7 12.2e35.7
Spectacle wear 5 5 9 9
Exotropia type*
Basic 61 63 71 68
Convergence insufficiency 2 2 1 1
High AC/A 2 2 0 0
Pseudodivergence excess 26 27 24 23
Figure 1. Diagram showing the flow of participants through study. True divergence excess 5 5 8 8
a
Due to poor cooperation, masked alignment testing could not be
completed for 5 participants at the 3-month visit (3 in observation and 2 in
AC/A ¼ accommodative convergence/accommodation; SD ¼ standard
patching) and 2 participants at the 6-month visit (1 in each group). deviation.
b
The observation group includes 1 participant who was lost to follow-up *One observation group participant is missing exotropia type.
and 2 participants who are continuing in the longer-term study follow-up
but did not complete the 6-month visit, and therefore are excluded from
the analysis. The patching group includes 4 participants who were lost to
follow-up and 3 participants who are continuing in the longer-term study interval, 3.8% to 9.4%; Table 4). Two of the 4 cases of
follow-up but did not complete the 6-month visit, and therefore are deterioration in the observation group did not meet the motor
excluded from the analysis. deterioration criteria (i.e., constant exotropia of at least 10 D by
c
In the observation group, 2 participants were withdrawn from the study for SPCT at distance and near), but were considered deteriorations
outcome-related reasons. One child’s mother and the pediatrician felt the for the primary analysis because patching was prescribed at the
IXT had resolved. The other child’s parents felt their child’s eyes were 3-month visit in response to parental concerns. Additional clin-
getting worse and wanted surgery. ical data for participants who experienced deterioration can be
d
In the patching group, parents withdrew 1 child because they were found in Table 5 (available at www.aaojournal.org). Adjusting the
transferring care to another ophthalmologist. primary outcome analysis for potential confounders including
presence of distance constant exotropia at baseline did not
produce meaningful differences (data not shown). An analysis
Treatment Compliance in Patching Group using multiple imputation with baseline data to impute the
6-month outcome for the 24 participants who did not complete
At the 3-month visit, patching compliance was judged as excellent the 6-month examination also yielded results similar to the primary
in 55 participants (60%), good in 11 participants (12%), fair in analysis (data not shown).
8 participants (9%), and poor in 16 participants (17%); 2 partici- A secondary outcome based solely on meeting motor deterio-
pants (2%) did not complete any patching. Patching compliance ration criteria regardless of whether nonstudy treatment had been
between 3 to 5 months after randomization was similar (data not started without meeting the criteria occurred in 2 (2.3%) of the 87
shown). Forty-eight participants (53%) stopped patching per pro- participants in the observation group and in 2 (2.2%) of the 90
tocol 3 to 5 weeks before the 6-month visit, 32 participants (35%) participants in the patching group (difference, 0.08%; P ¼ 0.55,
stopped earlier, and 11 participants (12%) stopped later. In- 1-sided hypothesis test; 2-sided exact 95% confidence
vestigators prescribed unilateral patching for 25 participants (24%), interval, 5.8% to 6.1%).
alternate patching for 57 participants (55%), and both unilateral For observation and patching groups, respectively, secondary
and alternate eye patching at different times for 10 participants 6-month outcomes showed that mean distance control scores were
(10%); the eye prescribed patching was not specified for 12 par- 2.8 versus 2.3 points (P ¼ 0.02), mean near control scores were 1.4
ticipants (12%). versus 1.1 points (P ¼ 0.26), mean distance PACT measurements
were 27.9 D versus 24.9 D (P ¼ 0.02), and mean near PACT
Outcome Analyses measurements were 19.3 D versus 17.0 D (P ¼ 0.10; Table 6).
Analyses that adjusted for potential confounders (in addition to
By the 6-month primary outcome examination, deterioration the baseline value of the outcome) did not produce meaningful
occurred in 4 (4.6%) of the 87 participants assigned to observation differences (data not shown). The distributions of change in
and in 2 (2.2%) of the 90 assigned to patching (difference, 2.4%; control scores and PACT magnitudes between baseline and
P ¼ 0.27, 1-sided hypothesis test; 2-sided exact 95% confidence 6 months are shown in Table 6.

4
 Mohney et al 
Patching vs. Observation for Exotropia

Table 3. Baseline Ocular Alignment According to Treatment Group

Distance Alignment Near Alignment

Observation (n ¼ 97) Patching (n ¼ 104) Observation (n ¼ 97) Patching (n ¼ 104)
Treatment Group No. % No. % No. % No. %
Deviation type
Constant exotropia 9 9 3 3 0 0 0 0
Intermittent exotropia 88 91 101 97 69 71 72 69
Exophoria d d d d 21 22 21 20
No exodeviation d d d d 7 7 11 11
Exotropia (D) by SPCT*
0 (no measurable tropiay) 28 29 41 39 63 65 75 72
1e9 8 8 7 7 11 11 8 8
10e14 4 4 5 5 4 4 3 3
16e18 8 8 13 13 3 3 4 4
20e25 30 31 21 20 12 12 9 9
30e35 16 16 12 12 3 3 3 3
40e50 3 3 3 3 1 1 1 1
>50 0 0 2 2 0 0 0 0
Median 20 14 0 0
Range 0e50 0e>50 0e45 0e40
Exodeviation (D) by PACTz
No exodeviation (orthophoria) d d d d 9 9 13 13
1e9 d d d d 7 7 8 8
10e14 0 0 0 0 24 25 15 14
16e18 9 9 14 13 9 9 13 13
20e25 44 45 44 42 30 31 30 29
30e35 39 40 36 35 14 14 18 17
40e45 3 3 8 8 3 3 5 5
50 2 2 2 2 1 1 2 2
Mean (SD) 27.4 (7.1) 27.2 (8.2) 18.1 (10.8) 19.4 (12.1)
Range 16e50 16e55 0e50 0e55
Exotropia control score21
No exodeviation 0 0 0 0 3 3 10 10
0: No XT unless dissociated, recovers <1 sec 2 2 2 2 31 32 28 27
1: No XT unless dissociated, recovers 1e5 sec 18 19 27 26 30 31 39 38
2: No XT unless dissociated, recovers >5 sec 38 39 33 32 17 18 15 14
3: XT <50% of 30 sec 19 20 27 26 12 12 11 11
4: XT >50% of 30 sec 11 11 11 11 4 4 1 1
5: Constant XT 9 9 4 4 0 0 0 0
Mean (SD) 2.5 (1.2) 2.3 (1.1) 1.2 (1.2) 1.0 (1.0)

PACT ¼ prism and alternate cover test; SD ¼ standard deviation; SPCT ¼ simultaneous prism and cover test; XT ¼ exotropia; D ¼ prism diopter; d ¼ not
applicable.
*One participant was missing SPCT at near.
y
Includes participants who did not have a tropia during the examination; had exotropia not detectable by the cover-and-uncover test; and had exotropia that
was not measurable because it was too brief, too small, or the participant was not cooperative enough to allow an SPCT.
z
PACT at distance was required to be at least 10 D for eligibility.

Table 4. Six-Month Primary Outcome According to Treatment Group

At 3-Month Visit By 6-Month Visit

(Interim Outcome) (Primary Outcome)*
Observation Patching Observation Patching
Outcome (n ¼ 82) (n ¼ 90) (n ¼ 87) (n ¼ 90)
Deterioration 2 (2.4) 0 (0) 4 (4.6) 2 (2.2)
Motor deterioration (constant exotropia 10 D at distance and near)y 2 (2.4) 0 (0) 2 (2.3) 2 (2.2)
Started nonprotocol treatment without meeting motor deterioration 0 (0) 0 (0) 2 (2.3) 0 (0)

D ¼ prism diopter.
Data are no. (%).
*The 6-month primary outcome is based on deterioration at either the 3-month or 6-month visit.
y
Measured by simultaneous prism and cover test.

5
 Ophthalmology Volume -, Number -, Month 2015

Table 6. Six-Month Exotropia Control and Prism and Alternate Cover Test Results According to Treatment Group

Observation (n [ 87) Patching (n [ 90) P Value*

Exotropia control at distancey
At 6 mos
Mean (SD) 2.8 (1.3) 2.3 (1.5) 0.02
Range 0e5 0e5
Change between baseline and 6 mosz
Mean (SD) 0.3 (1.3) 0.1 (1.5)
Range 3 to 4 4 to 3
3 points better 2 (2) 6 (7)
2 points better 3 (4) 10 (11)
1 point better 12 (14) 14 (16)
0 32 (37) 34 (38)
1 point worse 19 (22) 16 (18)
2 points worse 17 (20) 3 (3)
3 points worse 1 (1) 7 (8)
Exotropia control at neary
At 6 mos
Mean (SD) 1.4 (1.4) 1.1 (1.4) 0.26
Range 0e5 0e5
Change between baseline and 6 mosz
Mean (SD) 0.3 (1.1) 0.1 (1.3)
Range 3 to 3 5 to 3
3 points better 2 (2) 3 (3)
2 points better 2 (2) 7 (8)
1 point better 10 (12) 11 (12)
0 42 (49) 45 (50)
1 point worse 21 (24) 14 (16)
2 points worse 6 (7) 5 (6)
3 points worse 3 (4) 5 (6)
PACT at distance (D)y
At 6 mos
Mean (SD) 27.9 (8.2) 24.9 (9.4) 0.02
Range 0e45 0e45
Change between baseline and 6 mosx
Mean (SD) 0.6 (8.1) 2.0 (8.6)
Range 20 to 30 22 to 25
Decreased 8 D 8 (9) 22 (24)
Within <8 D 66 (77) 59 (66)
Increased 8 12 (14) 9 (10)
PACT at near (D)y
At 6 mos
Mean (SD) 19.3 (9.6) 17.0 (12.2) 0.10
Range 0e40 0e45
Change between baseline and 6 mosx
Mean (SD) 1.8 (11.4) 0.9 (12.8)
Range 30 to 25 30 to 25
Decreased 13 D 7 (8) 16 (18)
Within <13 D 66 (77) 60 (67)
Increased 13 D 13 (15) 14 (16)

PACT ¼ prism and alternate cover test; SD ¼ standard deviation; D ¼ prism diopter.
Data are no. (%) unless otherwise indicated.
*From analysis of covariance models adjusting for the baseline level of the outcome.
y
One observation group participant was missing control (distance and near) and PACT (distance and near) at the 6-month visit.
z
Change was calculated as control at baseline minus control at 6-month examination measures. Therefore, positive change was improvement (decreasing
control score) and negative change was worsening (increasing control score).
x
Change was calculated based on baseline PACT minus 6-month examination PACT. Therefore, positive change was improvement (decreasing angle) and
negative change was worsening (increasing angle). Change in PACT at distance and near was classified using thresholds of 8 D and 13 D, respectively,
because these amounts exceed the repeatability coefficients of 7.2 D and 12.8 D for PACT angles larger than 20 D at distance and near.24

Discussion previously untreated IXT. The rate of deterioration 6 months

after randomization was 2.2% for the patching group and
We found that deterioration of IXT over a 6-month period 4.6% for the observation group overall, or 2.3% for cases
was uncommon among children 12 to 35 months of age with that met motor deterioration criteria (i.e., excluding cases of

6
 Mohney et al 
Patching vs. Observation for Exotropia

treatment crossover), making it unlikely that part-time effects of observation or patching on sensory outcomes. In
patching provides a meaningful short-term benefit for contrast to the study’s limitations, the strengths are that it
young children with IXT. was a multicenter randomized trial, enrolled a fairly large
We are unaware of any randomized trial with which to cohort of children younger than 3 years of age with pre-
compare our results on the effectiveness of part-time viously untreated IXT, and had primary outcome assess-
patching in children younger than 3 years with IXT. We ments conducted by a masked examiner and confirmed by
have reported findings in children 3 to 10 years of age with a retest.
IXT as a parallel cohort in the same randomized trial.20 In this study of children 12 to 35 months of age with
Because those children were old enough to measure previously untreated IXT, motor deterioration over a
stereoacuity, deterioration was defined as either motor 6-month period was uncommon with or without part-time
deterioration, loss of stereoacuity (2 octaves), or receipt patching. This finding may assuage the concerns of some
of nonprotocol treatment.20 In the older cohort, motor and clinicians who believe that immediate surgical intervention
sensory deterioration occurred slightly less often in the is necessary to prevent IXT from becoming constant. We
patching group (0.6%; n ¼ 1 of 159) than in the found insufficient evidence to support the use of part-time
observation group (6.1%; n ¼ 10 of 165; P ¼ 0.004); patching for the treatment of IXT in children 12 to 35
however, there were no cases of motor deterioration months of age.
(0.0%; n ¼ 0 of 159) in the patching group and only a
single case in the observation group (0.6%; n ¼ 1 of
165).20 Both age cohorts and treatment groups taken References
together showed that children with previously untreated
IXT rarely demonstrate constant exotropia over the course
of 6 months, on the order of approximately 1%.20 1. Govindan M, Mohney BG, Diehl NN, Burke JP. Incidence and
Regarding secondary outcomes, there was a suggestion types of childhood exotropia: A population-based study.
Ophthalmology 2005;112:104–8.
of a slight positive effect from patching. Similar to the
2. Friedman DS, Repka MX, Katz J, et al. Prevalence of
current study, the parallel older cohort study20 found a amblyopia and strabismus in white and African American
slightly smaller mean distance exodeviation at the children aged 6 through 71 months: the Baltimore Pedi-
6-month outcome examination among those who were atric Eye Disease Study. Ophthalmology 2009;116:
patched compared with those who were observed (22.2 D 2128–34.
vs. 23.8 D; P ¼ 0.01) and no statistical difference in the 3. Multi-ethnic Pediatric Eye Disease Study Group. Preva-
mean near exodeviation (15.4 D vs. 17.6 D; P ¼ 0.11). lence of amblyopia and strabismus in African American and
Although the older cohort’s mean control score was better Hispanic children ages 6 to 72 months: the Multi-ethnic
in the patching group compared with the observation Pediatric Eye Disease Study. Ophthalmology 2008;115:
group at near (0.9 vs. 1.2 points; P ¼ 0.01), no statistical 1229–36.
4. McKean-Cowdin R, Cotter SA, Tarczy-Hornoch K, et al.
difference was found for the distance control score (2.0 vs.
Prevalence of amblyopia or strabismus in Asian and non-
2.3 points; P ¼ 0.09).20 The opposite was found for the Hispanic white preschool children: Multi-ethnic Pediatric
current younger cohort: the mean distance control score Eye Disease Study. Ophthalmology 2013;120:2117–24.
was better in the patching group (2.3 vs. 2.8 points; 5. Hatt SR, Gnanaraj L. Interventions for intermittent exotropia
P ¼ 0.02), but not the near control score (1.1 vs. 1.4 (Review). Cochrane Database of Systematic Reviews 2006:
points; P ¼ 0.26). The clinical significance of these 1–10.
small differences is unknown. 6. Hoyt CS, Pesic A. The many enigmas of intermittent exo-
There are several limitations to this study. The study tropia. Br J Ophthalmol 2012;96:1280–2.
recruited 60% of its planned sample size, resulting in less 7. American Academy of Ophthalmology Pediatric Ophthal-
statistical power than desired. In addition, the prestudy mology/Strabismus Panel. Preferred practice pattern guide-
lines. Esotropia and Exotropia. San Francisco, CA: American
sample size calculations were based on an assumption of a
Academy of Ophthalmology; 2012.
15% deterioration rate in the observation group over 8. Romano PE, Wilson MF. Survey of current management of
6 months, a frequency that seems to have been an over- intermittent exotropia in the USA and Canada. In: Campos EC,
estimate in light of our study results; this made it difficult ed. Strabismus and Ocular Motility Disorders. London: The
to determine if patching could lower an already very low Macmillan Press; 1990:391–6.
6-month deterioration rate. Our treatment regimen of 9. Romano PE, Robinson JA. Intermittent exotropia: survey of
3 hours of daily patching was based on consensus from current management. Transactions of the VIIth International
prestudy polling of Pediatric Eye Disease Investigator Orthoptic Congress. Nurnberg, Germany: 1991:227-31.
Group investigators; it is possible that patching more hours 10. Buck D, Powell C, Cumberland P, et al. Presenting features
may have different results. We acknowledge that 5 months and early management of childhood intermittent exotropia in
the UK: inception cohort study. Br J Ophthalmol 2009;93:
of patching followed by 1 month of observation is a rela-
1620–4.
tively short interval to study and that a longer period of 11. Flynn JT, McKenney S, Rosenhouse M. Management of
patching also may provide different results. The loss to intermittent exotropia. In: Moore S, Mein J, Stockbrige L, eds.
follow-up was higher than expected at 12%, although it Orthoptics: Past, Present, and Future: Transactions of the Third
was similar for both treatment groups. In addition, because International Orthoptic Congress, Boston, July 1-3, 1975. New
these children were too young for stereoacuity testing, we York: Stratton Intercontinental Medical Book Corp.; 1976:
cannot evaluate the possible beneficial or detrimental 551–7.

7
 Ophthalmology Volume -, Number -, Month 2015

12. Vishnoi SK, Singh V, Mehra MK. Role of occlusion in treatment 19. Suh YW, Kim SH, Lee JY, Cho YA. Conversion of
of intermittent exotropia. Ind J Ophthal 1987;35:207–10. intermittent exotropia types subsequent to part-time occlu-
13. Spoor DK, Hiles DA. Occlusion therapy for exodeviations sion therapy. Graefes Arch Clin Exp Ophthalmol 2006;244:
occurring in infants and young children. Ophthalmology 705–8.
1979;86:2152–7. 20. Pediatric Eye Disease Investigator Group. A randomized trial
14. Freeman RS, Isenberg SJ. The use of part-time occlusion for comparing part-time patching with observation for children 3-
early onset unilateral exotropia. J Pediatr Ophthalmol Stra- 10 years of age with intermittent exotropia. Ophthalmology
bismus 1989;26:94–6. 2014;121:2299–310.
15. Chutter CP. Occlusion treatment of intermittent divergent 21. Mohney BG, Holmes JM. An office-based scale for assessing
strabismus. Am Orthopt J 1977;27:80–4. control in intermittent exotropia. Strabismus 2006;14:147–50.
16. Iacobucci I, Henderson JW. Occlusion in the preoperative 22. Barnard GA. Significance tests for 2 x 2 tables. Biometrika
treatment of exodeviations. Am Orthopt J 1965;15:42–7. 1947;34:123–38.
17. Berg PH, Lozano MJ, Isenberg SJ. Long term results of part- 23. Rubin DB. Multiple Imputation for Nonresponse in Surveys.
time occlusion for intermittent exotropia. Am Orthopt J New York: Wiley & Sons; 2000.
1998;48:85–9. 24. Hatt SR, Leske DA, Liebermann L, et al. Variability of angle
18. Berg PH, Isenberg SJ. Treatment of unilateral exotropia by of deviation measurements in children with intermittent exo-
part-time occlusion. Am Orthopt J 1991;41:72–6. tropia. J AAPOS 2012;16:120–4.

Footnotes and Financial Disclosures

Originally received: February 19, 2015. Supported by the National Eye Institute, National Institutes of Health,
Final revision: April 21, 2015. Bethesda, Maryland (grant nos.: EY011751 and EY018810). The funding
Accepted: April 22, 2015. organization had no role in the design or conduct of this research.
Available online: ---. Manuscript no. 2015-290. Author Contributions:
1
Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota. Conception and design: Mohney, Cotter, Chandler, Holmes, Melia, Dona-
2 hue, Wallace, Kraker
Southern California College of Optometry at Marshall B. Ketchum Uni-
versity, Fullerton, California. Analysis and interpretation: Mohney, Cotter, Chandler, Holmes, Melia,
3 Kraker
Jaeb Center for Health Research, Tampa, Florida.
Data collection: Mohney, Cotter, Chandler, Holmes, Chen, Donahue,
4
Vanderbilt Eye Center, Nashville, Tennessee. Wallace, Kraker, Christian, Suh
5
Duke Eye Center, Durham, North Carolina. Obtained funding: Kraker, Holmes, Wallace
6
Concord Ophthalmologic Associates, Concord, Ohio. Overall responsibility: Mohney, Cotter, Chandler, Holmes, Chen, Melia,
7 Donahue, Wallace, Kraker, Christian, Suh
Wolfe Clinic, West Des Moines, Iowa.
*A complete list of participating members of the Pediatric Eye Disease Abbreviations and Acronyms:
Investigator Group (PEDIG) can be found in Appendix 1. IXT ¼ intermittent exotropia; PACT ¼ prism and alternate cover test;
SPCT ¼ simultaneous prism and cover test; D ¼ prism diopter.
Financial Disclosure(s):
The author(s) have no proprietary or commercial interest in any materials Correspondence:
discussed in this article. Brian G. Mohney, MD, c/o Jaeb Center for Health Research, 15310
Amberly Drive, Suite 350, Tampa, FL 33647. E-mail: pedig@jaeb.org.