- Email: [email protected]
Long-term ocular alignment after bilateral lateral rectus recession in children with infantile and intermittent exotropia
Long-term ocular alignment after bilateral lateral rectus recession in children with infantile and intermittent exotropia Jason C. S. Yam, FRCS, MPH, Patrick K. W. Wu, FRCS, FCOphthHK, Gabriela S. L. Chong, PhD, Ursula S. F. Wong, FRCS, FCOphthHK, Clement W. N. Chan, FRCS, FCOphthHK, and Simon T. C. Ko, FRCS, FCOphthHK PURPOSE
METHODS
RESULTS
CONCLUSIONS
To compare postoperative drift after bilateral lateral rectus recession for infantile exotropia (XT) and for intermittent XT and to compare initial postoperative alignment with longterm motor outcome. Medical records of all patients with infantile exotropia or intermittent exotropia who had undergone bilateral lateral rectus muscle recession surgery with a follow-up longer than 3 years were reviewed. The pre- and postoperative angles of deviation at distance and at near and postoperative drift at distance were compared. Surgical outcome was categorized as “success” (esotropia \6D or exotropia \11D), “recurrence” (.10D exotropia), or “overcorrection” (.5D of esotropia). The overall mean postoperative exotropic drift at 3 years was 10.4D in the infantile XT group and 7.2D in the intermittent XT group (P 5 0.05). Both groups had a low success rate at 3 years: 41% in the infantile XT group and 51% in the intermittent XT group (P 5 0.270). For patients with an initial esotropia of 0D to 10D, the success rate at 3 years was 86% in the infantile XT group (12 of 14) and 65% in the intermittent XT group (28 of 43). Postoperative exotropic drift is clinically similar in patients with intermittent versus infantile exotropia. Esotropia of 0D to 10D during the early postoperative period may be associated with the best long-term ocular alignment. ( J AAPOS 2012;16:274-279)
I
nfantile exotropia (XT) is a rare, divergent strabismus that is develops in infancy and persists beyond the age of 6 months.1 Intermittent exotropia is more common and tends to develop later in life. When considering the surgical approach to exotropia, an initial overcorrection may result in better long-term motor alignment.2,3 Specific recommendations for overcorrection of patients with intermittent exotropia vary: Raab and Parks4 advised overcorrection by 10D to 20D; Scott and colleagues,5 4D to 14D; McNeer,6 0D to 10D; and Souza-Dias and Uesugi,7 10D. It is unknown whether patients with infantile exotropia demonstrate the same postoperative exotropic drift and whether the same amount of initial overcorrection may be associated with the best surgical outcome. The purpose of this study was to compare the amount of postoperative drift in patients treated for infantile exotropia with that of patients treated for intermittent exotropia. Author affiliation: Department of Ophthalmology, Tung Wah Eastern Hospital, Hong Kong Submitted June 12, 2011. Revision accepted January 13, 2012. Correspondence: Jason C. S. Yam, FRCS, MPH, Department of Ophthalmology, Tung Wah Eastern Hospital, 19 Eastern Hospital Road, Causeway Bay, Hong Kong, People’s Republic of China (email: [email protected]). Copyright Ó 2012 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 doi:10.1016/j.jaapos.2012.01.005
274
It was also designed to determine the relationship between initial postoperative and long-term subsequent postoperative motor outcomes in infantile exotropia and to investigate what initial postoperative alignment might be associated with the best long-term motor outcomes.
Patients and Methods This retrospective study was exempt from ethics committee approval. The study protocol followed the principles in the Declaration of Helsinki. The medical records of patients diagnosed with infantile exotropia, defined as onset of exotropia before 12 months of age, and who underwent bilateral lateral rectus muscle recession surgery at Tung Wah Eastern Hospital and Pamela Youde Nethersole Eastern Hospital from 1993 to 2008 were reviewed. Exclusion criteria included history of prematurity (defined as a gestational age \37 weeks), trauma, or neurologic disease; anisometropia .1.50 D; myopia .3.00 D or hypermetropia .4.00 D; organic ocular or orbital pathology that would reduce vision; exotropia caused by previous eye muscle surgery; or follow-up \3 years.8,9 All patients were referred to a pediatrician to exclude neurologic disorders.10,11 Patients with ocular conditions most likely associated with the strabismus (amblyopia, latent nystagmus, oblique muscle overaction, dissociated vertical deviation, A pattern or V pattern) were not
Journal of AAPOS
Volume 16 Number 3 / June 2012
Yam et al
275
Table 1. Baseline characteristics of patients with infantile exotropia versus intermittent exotropia
Age of onset, months Age at diagnosis, months Age at surgery, months Associated features V pattern Inferior oblique overaction DVD Mean preoperative angle of deviation at, PD Near Distance Preoperative phase of deviation Constant Intermittent Preoperative refraction (SE)a Follow-up, months Surgical method for primary operation
Infantile exotropia (n 5 37)
Intermittent exotropia (n 5 144)
53 24 14 55 24
42 28 61 39 100 40
9 (24%) 6 (16%) 2 (5%)
32 (22%) 11 (8%) 0 (0%)
35 11 36 11
31 8 31 8
8 (22%) 29 (78%) 10.14 1.34 (12.50 to 56 24 BLR 37 (100%) BIOR 5 (14%)
2.12)
17 (12%) 127 (88%) 1.44 2.07 (13.12 to 49 15 BLR 144 (100%) BIOR 14 (10%)
7.06)
BIOR, bilateral inferior oblique recession surgery; BLR, bilateral lateral rectus muscle recession surgery; DVD, dissociated vertical deviation; PD, prism diopters; SE, spherical equivalent. a Data from 8 patients missing.
excluded. Patients with additional oblique muscle operations were not excluded. We also selected a comparison group from all patients with intermittent exotropia who underwent bilateral lateral rectus muscle recession surgery by the same surgeons (STCK or PKWW) during the period of 1993 to 2008. Intermittent exotropia was generally defined as a divergent deviation intermittently controlled by fusional mechanisms and intermittently breaking down into a manifest exotropia, with onset after 1 year of age.12 This group was designed to provide a comparison for the postoperative drift pattern and surgical outcomes between the infantile exotropia and intermittent exotropia. To avoid possible differences in case mix and outcomes between different surgeons, the first priority in selection of the control group was to choose patients operated by the same surgeon. Only those cases with bilateral lateral rectus muscle recession surgery were included because different types of surgery have been shown to result in different amounts of postoperative drift.13 Patients with additional oblique muscle surgeries were not excluded. For both infantile XT and intermittent XT groups, patients were further subdivided into pattern and comitant groups. Patients were defined as having an A pattern if the amount of exotropia in downgaze exceeded that in upgaze by at least 10D; V pattern was defined as an exotropia in up gaze exceeding that in down gaze by at least 15D. The following information was recorded when available: age at onset and at diagnosis, sex, intermittent versus constant phase before surgery, age at surgery, follow-up period, preoperative refraction, pre- and postoperative angles of strabismus, associated ocular motility disturbance, binocular sensory status, and associated amblyopia. Age of onset was defined as the age at which a parent or relative first observed ocular misalignment; when possible, these estimates were confirmed by the use of photographs of the patient. The preoperative refraction was defined as the mean spherical equivalent of both eyes. Amblyopia was de-
Journal of AAPOS
fined as best-corrected visual acuity of 6/12 or less in one or both eyes, or a bilateral difference of at least 2 best-corrected visual acuity lines. All patients underwent complete ophthalmologic and orthoptic examinations before surgery. The angle of deviation was measured at 6 m and 1/3 m by use of the prism and alternate cover test. The Krimsky test was used for uncooperative patients. All deviations recorded were determined by the use of appropriate spectacle correction. For binocular vision assessment, the Worth 4-dot and Titmus stereoacuity test (Stereo Optical, Chicago, IL) results were recorded. Stereopsis was graded into 4 categories: stereopsis better than 100 arcsec; stereopsis 100-400 arcsec; stereopsis 400-3000 arcsec; and nil stereopsis. For statistical analysis, patients with no measurable stereopsis were assigned to the worst measurable value of 3,000 arcsec. Surgery was recommended if the frequency or magnitude of exotropia deteriorated, or if tropia was present more than 50% of the time as determined either by examination or history. All patients underwent bilateral lateral rectus muscle recession; some underwent additional bilateral inferior oblique muscle recession for the correction of a V pattern. Surgery was performed after exodeviation had become stable (angle and degree of control) over 2 or 3 consecutive visits. All operations were performed by 1 of 2 surgeons (STCK or PKWW). Surgical dosages were applied using standard tables. We generally aim for slight overcorrection for all cases. Initial postoperative deviation at week 1 was recorded for each patient. Postoperative drift was then calculated for each patient at 1 month (3-6 weeks), 3 months (2-4 months), 6 months (5-7 months), 1 year (11-13 months), 2 years (23-25 months), and 3 years (35-37 months). Postoperative drift was defined as the difference between the angle of deviation at the particular follow-up period and the angle of deviation at week 1. Distance angle of deviation was used in the calculation of postoperative drift. Postoperative drift between the infantile exotropia and
276
Yam et al
Volume 16 Number 3 / June 2012
Table 2. Surgical outcomes of patients of infantile exotropia versus intermittent exotropia Infantile exotropia (n 5 37) Initial postoperative angle of deviation, PD Near Distance Final postoperative angle of deviation at near, PD Near Distance Primary surgical outcome within 3 years Successful Unsuccessful Recurrence Overcorrection Reoperation Final surgical outcomec Successful Unsuccessful Recurrence Overcorrection Final binocular vision status Mean stereopsis, arcsec Stereopsis \100 Stereopsis 100-400 Stereopsis 400-3000 Nil stereopsis Amblyopia Preoperative Final
Intermittent exotropia (n 5144)
Table 3. Mean postoperative drift in patients of infantile exotropia versus intermittent exotropia Mean postoperative drift,a prism diopters
P value Postoperative interval, month
4 12 3 11
39 2 10
0.464 0.497
13 12 13 12
9 11 9 10
0.042 0.027
14 (38%) 23 (62%) 22 (60%) 1 (2.7%) 5 (14%) BMRRes 4 BMRRec 1
65 (45%) 79 (55%) 68 (47%) 11 (7.6%) 12 (8%) BMRRes 5 BLRA 7
0.424a
15 (41%) 22 (59%) 20 (54%) 2 (5%) (N 5 34) 1169 1375 7 (21%) 13 (38%) 4 (12%) 10 (29%)
73 (51%) 71 (49%) 65 (45%) 6 (4%) (N 5 144) 327 697 53 (37%) 80 (55%) 11 (8%) 0 (0%)
0.270a
11 (8%) 2 (1%)
0.004a 0.001b
Infantile exotropia (n 5 37)
Intermittent exotropia (n 5 144)
P value
6.3 7.4 7.8 9.6 10.0 10.4
4.3 4.5 5.0 5.9 6.5 7.2
0.19 0.06 0.06 0.02 0.03 0.05
1 3 6 12 24 36 a
9 (24%) 6 (16%)
Positive value indicate exotropic drift.
Statistical Analysis 0.348
b
0.000 0.000b
BLRA, bilateral lateral rectus advancement surgery; BMRRec, bilateral medial rectus muscle recession; BMRRes, bilateral medial rectus resection surgery; PD, prism diopters. a 2 c test. b Fisher exact test. c Includes all reoperations. intermittent exotropia were compared at different points along the 3-year follow-up period. To evaluate the relation between initial and subsequent postoperative motor alignment, patients were divided into 6 groups according to ocular alignment 1 week postoperatively: overcorrected .20D, 11D to 20D, and 0D to 10D of esotropia; undercorrected 1D to 10D, 11D to 20D, and .20D of exotropia. Patients were assigned to one of the following long-term surgical outcome groups as determined by the measurement of distance deviation at 3 years’ follow-up: (1) success, defined as ocular alignment with esotropia \6D or XT \11D; (2) recurrence, defined as ocular alignment of .10D of exotropia; or (3) overcorrection, defined as .5D esotropia.3,14,15 Primary surgical outcome was defined as the surgical outcome after the primary operation. Patients requiring reoperations before the 3-year follow-up were counted as recurrence or overcorrection depending on the last distance angle of deviation before reoperation. Final surgical outcome was defined as surgical outcome after all operations (including reoperations) at 3 years’ follow-up.
All analyses were performed with statistical software (StatLab, SPSS for Windows, version 16.0; SPSS, Inc, Chicago, IL). The baseline characteristics and the amount of postoperative exotropic drift of the infantile XT group and the intermittent XT group were compared with the use of an independent t test and the Mann-Whitney U test as appropriate. Categorical outcomes of the 2 study groups were compared by the c2 test or the Fisher exact test as appropriate. Statistical significance was accepted for P values \0.05. Retrospective power calculation was performed for the postoperative drift analysis. Using the standard deviation for the drift for both group as approximately 8D, and power of the study as 0.8, we can detect a true difference of means of 4.2D between the drift of the 2 groups with our sample size.
Results We identified 48 infantile exotropia patients who underwent bilateral lateral rectus muscle recession surgery during the study period; 11 cases were excluded because of insufficient follow-up, leaving 37 cases for analysis. A total of 182 intermittent exotropia patients who underwent bilateral lateral rectus muscle recession surgery by the same surgeons during the same period were identified; 38 were excluded because of insufficient follow-up, leaving 144 cases available for analysis. Baseline clinical characteristics are summarized in Table 1. Compared with the intermittent XT group, the infantile XT group has a younger age of onset and age at surgery, larger preoperative misalignment, more hyperopic refraction, and longer follow-up. The surgical outcomes are summarized in Table 2. The infantile XT group has a larger final postoperative misalignment, poorer stereopsis, and a greater rate of amblyopia; however, 3-year primary surgical outcome, 3-year final surgical outcome, and the reoperation rate did not differ significantly between the 2 groups. The overall rate of primary successful outcome was 38% in the infantile XT versus 45% in the intermittent XT group (c2 test, P 5 0.424). The rate of final successful outcome after
Journal of AAPOS
Volume 16 Number 3 / June 2012
Yam et al
277
FIG 1. Postoperative drift in patients with infantile XT and intermittent XT. P 5 0.19 (1 month), 0.06 (3 months), 0.06 (6 months), 0.02 (12 months), 0.03 (24 months), and 0.05 (36 months).
primary surgery and reoperations was 41% in infantile exotropia and 51% in the infantile exotropia (c2 test, P 5 0.270). Postoperative Drift Mean overall drift for both groups was toward XT (Table 3 and Figure 1). The exotropic drift was similar between the 2 groups at 3 years postoperatively (independent t test, P 5 0.05). The drift pattern of the comitant and pattern groups within the infantile XT and intermittent XT groups was compared (Figure 2). For the infantile XT group, the comitant subgroup drifted greater than the pattern subgroup at 3 years’ follow-up (Mann-Whitney U test, P 5 0.04). For the intermittent XT group, the difference in drift between the comitant and pattern subgroups was not significant at 3 years (Mann-Whitney U test, P 5 0.19). There was no significant difference in drift when we compared those patients with inferior oblique muscle recession with those without inferior oblique muscle recession in both the infantile XT group (Mann-Whitney U test, P 5 0.230 at 3 years) and the intermittent XT group (Mann-Whitney U test, P 5 0.951 at 3 years). There was no statistical significant difference in drift between patients with constant phase versus intermittent phase of preoperative angle of deviation for both the infantile XT group (P 5 0.32) and intermittent XT group (P 5 0.52) at 3 years’ follow-up (Mann-Whitney test). Postoperative Motor Alignment The primary surgical outcome at 3 years as a function of initial alignment the infantile XT and intermittent XT groups is summarized in Table 4. In the infantile XT
Journal of AAPOS
FIG 2. Postoperative drift in patients with comitant and pattern infantile XT and comitant and pattern intermittent XT. The difference of drift between comitant and pattern infantile XT was statistically significant at 3 years postoperatively (P 5 0.04). The difference of drift between comitant and pattern intermittent XT was not significant at 3 years postoperatively (P 5 0.19).
group, all patients with successful primary outcome had an initial esotropia between 0D and 20D; initial esotropia in most patients (12 of 14) was between 0D and 10D. No patients who were initially undercorrected had a successful outcome. Patients with recurrence occupied a broad range of initial alignments, and most (19 of 22 patients) had initial undercorrection, ranging from 2D to 30D. One patient in the infantile XT group with large initial overcorrection, up to 22D, developed long-term overcorrection. There was an 86% chance of successful outcome for patients with initial alignment of 0D-10D. One patient with initial overcorrection .20D developed long-term overcorrection and required reoperation. In the intermittent XT group, of 65 patients with a successful outcome, 53 (82%) had initial alignment within 10D of orthotropia. Of 68 patients with longterm recurrence, 55 (81%) had initial undercorrection of 2D to 30D. All patients with long-term overcorrection had initial overcorrection of 2D to 30D. For patients with initial overcorrection, the chance for successful outcome was 60%; for those with initial undercorrection, less than 40%. By far the most favorable ratio of successful to unsuccessful outcomes occurred with initial alignments of 0D to 10D of esotropia in both the infantile XT and intermittent XT groups (86% vs 65%).
278
Yam et al
Volume 16 Number 3 / June 2012
Table 4. Three-year primary surgical outcomes as a function of initial postoperative alignment in infantile exotropia and intermittent exotropia, grouped by increments of 10 prism diopters Infantile exotropia (n 5 37) Initial postoperative alignment, PD
a
\ 20D 10 to 20D 0 to 10D 1 to 10D 10 to 20D .20D Total number (N)
Intermittent exotropia (n 5 144)
S
R
O
N
S
R
O
N
0 2 (67%)b 12 (86%) 0 0 0 14
0 1 (33%) 2 (14%) 12 (100%) 4 (100%) 3 (100%) 22
1 (100%) 0 0 0 0 0 1
1 3 14 12 4 3 37
0 8 (57%) 28 (65%) 25 (40%) 4 (21%) 0 65
0 2 (14%) 11 (26%) 38 (60%) 15 (8%) 2 (100%) 68
3 (100%) 4 (29%) 4 (9%) 0 0 0 11
3 14 43 63 19 2 144
N, total number of patients; O, overcorrection group; PD, prism diopters; R, recurrence group; S, success group. Minus angle denotes esotropia; plus angle, exotropia. b Percentage is of surgical outcome at that initial alignment group. a
Table 5. Binocular vision status and final 3-year surgical outcomes Infantile exotropia (n 5 34)
Intermittent exotropia (n 5 144) Stereopsis level
Final 3-year outcome Success Recurrence Overcorrection Total number a
\100
100-400
400-3000
nil
n
\100
100-400
400-3000
nil
n
5 (38%)a 2 (11%) 0 7
2 (15%) 11 (58%) 0 13
3 (23%) 1 (5%) 0 4
3 (23%) 5 (26%) 2 (100%) 10
13 19 2 34
45 (62%) 8 (12%) 0 53
26 (36%) 50 (77%) 4 (67%) 80
2 (3%) 7 (11%) 2 (33%) 11
0 0 0 0
73 65 6 144
Percentage is of stereopsis category of that surgical outcome group.
Binocular Function and Amblyopia Rate The level of stereopsis in both groups is given in Table 2 and the level of stereopsis according to the final motor outcome in Table 5. We were able to test stereoacuity in 34 of 37 patients in infantile XT group and all patients in intermittent XT group. The intermittent XT group had a better level of stereopsis than that of the infantile XT group (P \ 0.001). In both groups, patients with final success can achieve better stereopsis. The amblyopia rate in both groups is summarized in Table 2. The infantile XT group had a greater rate of amblyopia than the intermittent XT group preoperatively (P 5 0.004) and at 3 years’ followup (P 5 0.001).
Discussion Biglan and colleagues8 and Hunter and colleagues16 found that approximately equal numbers of patients with infantile exotropia have intermittent or constant exotropia at presentation and that clinical outcomes are similar regardless of presentation. Hunter and colleagues16 suggested that patients presenting with either constant or intermittent exotropia before 1 year of age may represent a spectrum of the same entity of infantile exotropia and should be included in analysis. In subsequent studies, researchers adopted this definition for case recruitment and analysis of infantile exotropia.9,17 It is not known whether the surgical overcorrection generally believed to be helpful for patients with intermittent exotropia is also beneficial for patients with infantile exotropia. Our results suggest that exotropic drift occurred in infantile exotropia during
a 3-year period and that the extent of drift was comparable at 3 years postoperatively to that seen in intermittent exotropia patients. Our subgroup analysis suggested that exotropic drift at 3 years may be smaller in patients with pattern deviation. This finding is consistent with the literature on intermittent exotropia patients.12 Yet, the drift between pattern and comitant subgroup in intermittent exotropia was not significantly different in our series. This may reflect either the absence of such mechanical differences or simply inadequate power to detect a difference due to the small sample size. Finally, the 2 subgroups were not well matched for comparison, as in the study by Pineles and colleagues.12 Our results suggest that initial overcorrection up to 10D of esotropia may be associated to the best-long term outcome. The importance of initial overcorrection for intermittent exotropia remains controversial.2-7,15,18 Oh and Hwan13 suggested that initial postoperative deviation was the only factor found to determine the outcome of exotropia. Choi and colleagues,18 however, found that initial overcorrection may be associated with lower probability of recurrence only within 2 years of surgery but not predictive beyond that. The disparities between the findings of different studies may be attributed to differences of the surgical procedure used, the definition of recurrence, and the follow-up period.18 Pratt-Johnson and colleagues19 cautioned that monofixation syndrome and amblyopia may develop in very young children when exotropia is overcorrected, especially if the deviating eye is left in an esotropic position for a prolonged period. In our series, the 18 cases of initial overcorrection
Journal of AAPOS
Volume 16 Number 3 / June 2012 (range, 2D-22D) all developed exotropic drift, with progression to \4D of esotropia in 15 patients (83%) 1 month postoperatively and 17 patients (94%) 3 months postoperatively. Only 1 patient with large initial overcorrection .20D remained overcorrected during the postoperative period and required reoperation. Our results have also shown that initial overcorrection does not lead to a higher rate of amblyopia development. It is difficult for patients with infantile exotropia to achieve binocular vision despite successful alignment. Biedner and colleagues20 reported that satisfactory alignment was achieved in 84% of cases but that no patient demonstrated fusion or stereoacuity. Our results demonstrated that patients with infantile exotropia had significantly poorer stereopsis than those with intermittent exotropia. In both groups, patients with successful final outcome can achieve better fine stereopsis than those with unsuccessful final outcomes. Thus, the level of attainable stereoacuity may be associated with the final motor alignment. Yet, it has also been suggested that stereoacuity may be associated with the initial presentation of constant or intermittent phase in infantile exotropia before surgery.16 In our study, of 7 patients in the infantile XT group with constant preoperative exotropia, 5 had nil stereopsis and 1 had stereopsis poorer than 400 arcsec. On the other hand, 19 of 27 patients in the infantile XT group with intermittent preoperative alignment achieved stereopsis better than 400 arsec. This finding is consistent with the findings of Hunter and colleagues,16 who reported that 7 of 10 testable patients achieved gross fusion after surgery and that 2 patients with intermittent exotropia at presentation acquired 200 arcsec of stereopsis but that no patients with constant exotropia achieved this level. Our study has several limitations. The number of recruited patients was relatively low; however, because infantile exotropia occurs only in 1 of 30,000 births in the general population,20 this limitation seems unavoidable. Other limitations included its retrospective nature, the contribution of two separate surgeons, imperfectly matched and different sample sizes between the 2 study groups, and nonuniform follow-up among individual cases. The onset of exotropia in this series was based on patient history confirmed with photographic documentation, which may have resulted in recall bias that would have been avoided by including only patients whose onset of symptoms were clinically confirmed. In conclusion, postoperative exotropic drift occurs in infantile exotropia, and the drift is clinically similar to that observed in intermittent exotropia up to 3 years after surgery. Initial overcorrection within 10D to orthophoria during the early postoperative period may be associated to the best long-term ocular alignment
Journal of AAPOS
Yam et al
279
over 3 years in both the infantile and intermittent exotropia.
Literature Search The authors conducted a MEDLINE (PubMed) search for English-language articles only for the period 1966 to 2010, using the following terms: infantile exotropia, congenital exotropia, intermittent exotropia, and exotropic drift. References 1. Kraft SP. Infantile exotropia. In: Rosenbaum AL, Santiago AP, editors. Clinical strabismus management: Principles and surgical techinques. Philadelphia, PA: W.B. Saunders; 1999:176-81. 2. Leow PL, Ko ST, Wu PK, Chan CW. Exotropic drift and ocular alignment after surgical correction for intermittent exotropia. J Pediatr Ophthalmol Strabismus 2010;47:12-16. 3. Ruttum MS. Initial versus subsequent postoperative motor alignment in intermittent exotropia. J AAPOS 1997;1:88-91. 4. Raab EI, Parks MM. Recession of the lateral recti. Early and late postoperative alignments. Arch Ophthalmol 1969;82:203-8. 5. Scott WE, Keech R, Mash AJ. The postoperative results and stability of exodeviations. Arch Ophthalmol 1981;99:1814-18. 6. McNeer KW. Observations on the surgical overcorrection of childhood intermittent exotropia. Am Orthoptic J 1987;37:135-50. 7. Souza-Dias C, Uesugi CF. Postoperative evolution of the planned initial overcorrection in intermittent exotropia: 61 cases. Binocular Vis Eye Muscle Surg Q 1993;8:141-8. 8. Biglan AW, Davis JS, Cheng KP, Pettapiece MC. Infantile exotropia. J Pediatr Ophthalmol Strabismus 1996;33:79-84. 9. Park JH, Kim SH. Clinical features and the risk factors of infantile exotropia recurrence. Am J Ophthamol 2010;150:464-7. 10. Baeteman C, Denis D, Loudot C, et al. Primary exotropia: Importance of cerebral MRI [in French]. J Fr Ophthalmol 2008;31:287-94. 11. Hunter DG, Ellis FJ. Prevalence of systemic and ocular disease in infantile exotropia: Comparison with infantile esotropia. Ophthalmology 1999;106:1951-6. 12. Pineles SL, Rosenbaum AL, Demer JL. Decreased postoperative drift in intermittent exotropia associated with A and V patterns. J AAPOS 2009;13:127-31. 13. Olitsky S. Early and late postoperative alignment following unilateral lateral rectus recession for intermittent exotropia. J Pediatr Ophthalmol Strabismus 1998;35:146-8. 14. Lee SY, Lee YC. Relationship between motor alignment at postoperative day 1 and at year 1 after symmetric and asymmetric surgery in intermittent exotropia. Jpn J Ophthalmol 2001;45:167-71. 15. Oh JY, Hwan JM. Survival analysis of 365 patients with exotropia after surgery. Eye (Lond) 2006;20:1268-72. 16. Hunter DG, Kelly JB, Buffenn AN, Ellis FJ. Long-term outcome of uncomplicated infantile exotropia. J AAPOS 2001;5:352-6. 17. Paik HJ, Yim HB. Clinical effect of early surgery in infantile exotropia. Korean J Ophthalmol 2002;16:97-102. 18. Choi J, Kim SJ, Yu YS. Initial postoperative deviation as a predictor of long-term outcome after surgery for intermittent exotropia. J AAPOS 2011;15:224-9. 19. Pratt-Johnson JA, Barlow JM, Tillson G. Early surgery in intermittent exotropia. Am J Ophthalmol 1977;84:689-94. 20. Biedner B, Marcus M, David R, Yassur Y. Congenital constant exotropia: Surgical results in six patients. Binocul Vis Eye Muscle Surg Q 1993;8:137-40.
METHODS
RESULTS
CONCLUSIONS
To compare postoperative drift after bilateral lateral rectus recession for infantile exotropia (XT) and for intermittent XT and to compare initial postoperative alignment with longterm motor outcome. Medical records of all patients with infantile exotropia or intermittent exotropia who had undergone bilateral lateral rectus muscle recession surgery with a follow-up longer than 3 years were reviewed. The pre- and postoperative angles of deviation at distance and at near and postoperative drift at distance were compared. Surgical outcome was categorized as “success” (esotropia \6D or exotropia \11D), “recurrence” (.10D exotropia), or “overcorrection” (.5D of esotropia). The overall mean postoperative exotropic drift at 3 years was 10.4D in the infantile XT group and 7.2D in the intermittent XT group (P 5 0.05). Both groups had a low success rate at 3 years: 41% in the infantile XT group and 51% in the intermittent XT group (P 5 0.270). For patients with an initial esotropia of 0D to 10D, the success rate at 3 years was 86% in the infantile XT group (12 of 14) and 65% in the intermittent XT group (28 of 43). Postoperative exotropic drift is clinically similar in patients with intermittent versus infantile exotropia. Esotropia of 0D to 10D during the early postoperative period may be associated with the best long-term ocular alignment. ( J AAPOS 2012;16:274-279)
I
nfantile exotropia (XT) is a rare, divergent strabismus that is develops in infancy and persists beyond the age of 6 months.1 Intermittent exotropia is more common and tends to develop later in life. When considering the surgical approach to exotropia, an initial overcorrection may result in better long-term motor alignment.2,3 Specific recommendations for overcorrection of patients with intermittent exotropia vary: Raab and Parks4 advised overcorrection by 10D to 20D; Scott and colleagues,5 4D to 14D; McNeer,6 0D to 10D; and Souza-Dias and Uesugi,7 10D. It is unknown whether patients with infantile exotropia demonstrate the same postoperative exotropic drift and whether the same amount of initial overcorrection may be associated with the best surgical outcome. The purpose of this study was to compare the amount of postoperative drift in patients treated for infantile exotropia with that of patients treated for intermittent exotropia. Author affiliation: Department of Ophthalmology, Tung Wah Eastern Hospital, Hong Kong Submitted June 12, 2011. Revision accepted January 13, 2012. Correspondence: Jason C. S. Yam, FRCS, MPH, Department of Ophthalmology, Tung Wah Eastern Hospital, 19 Eastern Hospital Road, Causeway Bay, Hong Kong, People’s Republic of China (email: [email protected]). Copyright Ó 2012 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 doi:10.1016/j.jaapos.2012.01.005
274
It was also designed to determine the relationship between initial postoperative and long-term subsequent postoperative motor outcomes in infantile exotropia and to investigate what initial postoperative alignment might be associated with the best long-term motor outcomes.
Patients and Methods This retrospective study was exempt from ethics committee approval. The study protocol followed the principles in the Declaration of Helsinki. The medical records of patients diagnosed with infantile exotropia, defined as onset of exotropia before 12 months of age, and who underwent bilateral lateral rectus muscle recession surgery at Tung Wah Eastern Hospital and Pamela Youde Nethersole Eastern Hospital from 1993 to 2008 were reviewed. Exclusion criteria included history of prematurity (defined as a gestational age \37 weeks), trauma, or neurologic disease; anisometropia .1.50 D; myopia .3.00 D or hypermetropia .4.00 D; organic ocular or orbital pathology that would reduce vision; exotropia caused by previous eye muscle surgery; or follow-up \3 years.8,9 All patients were referred to a pediatrician to exclude neurologic disorders.10,11 Patients with ocular conditions most likely associated with the strabismus (amblyopia, latent nystagmus, oblique muscle overaction, dissociated vertical deviation, A pattern or V pattern) were not
Journal of AAPOS
Volume 16 Number 3 / June 2012
Yam et al
275
Table 1. Baseline characteristics of patients with infantile exotropia versus intermittent exotropia
Age of onset, months Age at diagnosis, months Age at surgery, months Associated features V pattern Inferior oblique overaction DVD Mean preoperative angle of deviation at, PD Near Distance Preoperative phase of deviation Constant Intermittent Preoperative refraction (SE)a Follow-up, months Surgical method for primary operation
Infantile exotropia (n 5 37)
Intermittent exotropia (n 5 144)
53 24 14 55 24
42 28 61 39 100 40
9 (24%) 6 (16%) 2 (5%)
32 (22%) 11 (8%) 0 (0%)
35 11 36 11
31 8 31 8
8 (22%) 29 (78%) 10.14 1.34 (12.50 to 56 24 BLR 37 (100%) BIOR 5 (14%)
2.12)
17 (12%) 127 (88%) 1.44 2.07 (13.12 to 49 15 BLR 144 (100%) BIOR 14 (10%)
7.06)
BIOR, bilateral inferior oblique recession surgery; BLR, bilateral lateral rectus muscle recession surgery; DVD, dissociated vertical deviation; PD, prism diopters; SE, spherical equivalent. a Data from 8 patients missing.
excluded. Patients with additional oblique muscle operations were not excluded. We also selected a comparison group from all patients with intermittent exotropia who underwent bilateral lateral rectus muscle recession surgery by the same surgeons (STCK or PKWW) during the period of 1993 to 2008. Intermittent exotropia was generally defined as a divergent deviation intermittently controlled by fusional mechanisms and intermittently breaking down into a manifest exotropia, with onset after 1 year of age.12 This group was designed to provide a comparison for the postoperative drift pattern and surgical outcomes between the infantile exotropia and intermittent exotropia. To avoid possible differences in case mix and outcomes between different surgeons, the first priority in selection of the control group was to choose patients operated by the same surgeon. Only those cases with bilateral lateral rectus muscle recession surgery were included because different types of surgery have been shown to result in different amounts of postoperative drift.13 Patients with additional oblique muscle surgeries were not excluded. For both infantile XT and intermittent XT groups, patients were further subdivided into pattern and comitant groups. Patients were defined as having an A pattern if the amount of exotropia in downgaze exceeded that in upgaze by at least 10D; V pattern was defined as an exotropia in up gaze exceeding that in down gaze by at least 15D. The following information was recorded when available: age at onset and at diagnosis, sex, intermittent versus constant phase before surgery, age at surgery, follow-up period, preoperative refraction, pre- and postoperative angles of strabismus, associated ocular motility disturbance, binocular sensory status, and associated amblyopia. Age of onset was defined as the age at which a parent or relative first observed ocular misalignment; when possible, these estimates were confirmed by the use of photographs of the patient. The preoperative refraction was defined as the mean spherical equivalent of both eyes. Amblyopia was de-
Journal of AAPOS
fined as best-corrected visual acuity of 6/12 or less in one or both eyes, or a bilateral difference of at least 2 best-corrected visual acuity lines. All patients underwent complete ophthalmologic and orthoptic examinations before surgery. The angle of deviation was measured at 6 m and 1/3 m by use of the prism and alternate cover test. The Krimsky test was used for uncooperative patients. All deviations recorded were determined by the use of appropriate spectacle correction. For binocular vision assessment, the Worth 4-dot and Titmus stereoacuity test (Stereo Optical, Chicago, IL) results were recorded. Stereopsis was graded into 4 categories: stereopsis better than 100 arcsec; stereopsis 100-400 arcsec; stereopsis 400-3000 arcsec; and nil stereopsis. For statistical analysis, patients with no measurable stereopsis were assigned to the worst measurable value of 3,000 arcsec. Surgery was recommended if the frequency or magnitude of exotropia deteriorated, or if tropia was present more than 50% of the time as determined either by examination or history. All patients underwent bilateral lateral rectus muscle recession; some underwent additional bilateral inferior oblique muscle recession for the correction of a V pattern. Surgery was performed after exodeviation had become stable (angle and degree of control) over 2 or 3 consecutive visits. All operations were performed by 1 of 2 surgeons (STCK or PKWW). Surgical dosages were applied using standard tables. We generally aim for slight overcorrection for all cases. Initial postoperative deviation at week 1 was recorded for each patient. Postoperative drift was then calculated for each patient at 1 month (3-6 weeks), 3 months (2-4 months), 6 months (5-7 months), 1 year (11-13 months), 2 years (23-25 months), and 3 years (35-37 months). Postoperative drift was defined as the difference between the angle of deviation at the particular follow-up period and the angle of deviation at week 1. Distance angle of deviation was used in the calculation of postoperative drift. Postoperative drift between the infantile exotropia and
276
Yam et al
Volume 16 Number 3 / June 2012
Table 2. Surgical outcomes of patients of infantile exotropia versus intermittent exotropia Infantile exotropia (n 5 37) Initial postoperative angle of deviation, PD Near Distance Final postoperative angle of deviation at near, PD Near Distance Primary surgical outcome within 3 years Successful Unsuccessful Recurrence Overcorrection Reoperation Final surgical outcomec Successful Unsuccessful Recurrence Overcorrection Final binocular vision status Mean stereopsis, arcsec Stereopsis \100 Stereopsis 100-400 Stereopsis 400-3000 Nil stereopsis Amblyopia Preoperative Final
Intermittent exotropia (n 5144)
Table 3. Mean postoperative drift in patients of infantile exotropia versus intermittent exotropia Mean postoperative drift,a prism diopters
P value Postoperative interval, month
4 12 3 11
39 2 10
0.464 0.497
13 12 13 12
9 11 9 10
0.042 0.027
14 (38%) 23 (62%) 22 (60%) 1 (2.7%) 5 (14%) BMRRes 4 BMRRec 1
65 (45%) 79 (55%) 68 (47%) 11 (7.6%) 12 (8%) BMRRes 5 BLRA 7
0.424a
15 (41%) 22 (59%) 20 (54%) 2 (5%) (N 5 34) 1169 1375 7 (21%) 13 (38%) 4 (12%) 10 (29%)
73 (51%) 71 (49%) 65 (45%) 6 (4%) (N 5 144) 327 697 53 (37%) 80 (55%) 11 (8%) 0 (0%)
0.270a
11 (8%) 2 (1%)
0.004a 0.001b
Infantile exotropia (n 5 37)
Intermittent exotropia (n 5 144)
P value
6.3 7.4 7.8 9.6 10.0 10.4
4.3 4.5 5.0 5.9 6.5 7.2
0.19 0.06 0.06 0.02 0.03 0.05
1 3 6 12 24 36 a
9 (24%) 6 (16%)
Positive value indicate exotropic drift.
Statistical Analysis 0.348
b
0.000 0.000b
BLRA, bilateral lateral rectus advancement surgery; BMRRec, bilateral medial rectus muscle recession; BMRRes, bilateral medial rectus resection surgery; PD, prism diopters. a 2 c test. b Fisher exact test. c Includes all reoperations. intermittent exotropia were compared at different points along the 3-year follow-up period. To evaluate the relation between initial and subsequent postoperative motor alignment, patients were divided into 6 groups according to ocular alignment 1 week postoperatively: overcorrected .20D, 11D to 20D, and 0D to 10D of esotropia; undercorrected 1D to 10D, 11D to 20D, and .20D of exotropia. Patients were assigned to one of the following long-term surgical outcome groups as determined by the measurement of distance deviation at 3 years’ follow-up: (1) success, defined as ocular alignment with esotropia \6D or XT \11D; (2) recurrence, defined as ocular alignment of .10D of exotropia; or (3) overcorrection, defined as .5D esotropia.3,14,15 Primary surgical outcome was defined as the surgical outcome after the primary operation. Patients requiring reoperations before the 3-year follow-up were counted as recurrence or overcorrection depending on the last distance angle of deviation before reoperation. Final surgical outcome was defined as surgical outcome after all operations (including reoperations) at 3 years’ follow-up.
All analyses were performed with statistical software (StatLab, SPSS for Windows, version 16.0; SPSS, Inc, Chicago, IL). The baseline characteristics and the amount of postoperative exotropic drift of the infantile XT group and the intermittent XT group were compared with the use of an independent t test and the Mann-Whitney U test as appropriate. Categorical outcomes of the 2 study groups were compared by the c2 test or the Fisher exact test as appropriate. Statistical significance was accepted for P values \0.05. Retrospective power calculation was performed for the postoperative drift analysis. Using the standard deviation for the drift for both group as approximately 8D, and power of the study as 0.8, we can detect a true difference of means of 4.2D between the drift of the 2 groups with our sample size.
Results We identified 48 infantile exotropia patients who underwent bilateral lateral rectus muscle recession surgery during the study period; 11 cases were excluded because of insufficient follow-up, leaving 37 cases for analysis. A total of 182 intermittent exotropia patients who underwent bilateral lateral rectus muscle recession surgery by the same surgeons during the same period were identified; 38 were excluded because of insufficient follow-up, leaving 144 cases available for analysis. Baseline clinical characteristics are summarized in Table 1. Compared with the intermittent XT group, the infantile XT group has a younger age of onset and age at surgery, larger preoperative misalignment, more hyperopic refraction, and longer follow-up. The surgical outcomes are summarized in Table 2. The infantile XT group has a larger final postoperative misalignment, poorer stereopsis, and a greater rate of amblyopia; however, 3-year primary surgical outcome, 3-year final surgical outcome, and the reoperation rate did not differ significantly between the 2 groups. The overall rate of primary successful outcome was 38% in the infantile XT versus 45% in the intermittent XT group (c2 test, P 5 0.424). The rate of final successful outcome after
Journal of AAPOS
Volume 16 Number 3 / June 2012
Yam et al
277
FIG 1. Postoperative drift in patients with infantile XT and intermittent XT. P 5 0.19 (1 month), 0.06 (3 months), 0.06 (6 months), 0.02 (12 months), 0.03 (24 months), and 0.05 (36 months).
primary surgery and reoperations was 41% in infantile exotropia and 51% in the infantile exotropia (c2 test, P 5 0.270). Postoperative Drift Mean overall drift for both groups was toward XT (Table 3 and Figure 1). The exotropic drift was similar between the 2 groups at 3 years postoperatively (independent t test, P 5 0.05). The drift pattern of the comitant and pattern groups within the infantile XT and intermittent XT groups was compared (Figure 2). For the infantile XT group, the comitant subgroup drifted greater than the pattern subgroup at 3 years’ follow-up (Mann-Whitney U test, P 5 0.04). For the intermittent XT group, the difference in drift between the comitant and pattern subgroups was not significant at 3 years (Mann-Whitney U test, P 5 0.19). There was no significant difference in drift when we compared those patients with inferior oblique muscle recession with those without inferior oblique muscle recession in both the infantile XT group (Mann-Whitney U test, P 5 0.230 at 3 years) and the intermittent XT group (Mann-Whitney U test, P 5 0.951 at 3 years). There was no statistical significant difference in drift between patients with constant phase versus intermittent phase of preoperative angle of deviation for both the infantile XT group (P 5 0.32) and intermittent XT group (P 5 0.52) at 3 years’ follow-up (Mann-Whitney test). Postoperative Motor Alignment The primary surgical outcome at 3 years as a function of initial alignment the infantile XT and intermittent XT groups is summarized in Table 4. In the infantile XT
Journal of AAPOS
FIG 2. Postoperative drift in patients with comitant and pattern infantile XT and comitant and pattern intermittent XT. The difference of drift between comitant and pattern infantile XT was statistically significant at 3 years postoperatively (P 5 0.04). The difference of drift between comitant and pattern intermittent XT was not significant at 3 years postoperatively (P 5 0.19).
group, all patients with successful primary outcome had an initial esotropia between 0D and 20D; initial esotropia in most patients (12 of 14) was between 0D and 10D. No patients who were initially undercorrected had a successful outcome. Patients with recurrence occupied a broad range of initial alignments, and most (19 of 22 patients) had initial undercorrection, ranging from 2D to 30D. One patient in the infantile XT group with large initial overcorrection, up to 22D, developed long-term overcorrection. There was an 86% chance of successful outcome for patients with initial alignment of 0D-10D. One patient with initial overcorrection .20D developed long-term overcorrection and required reoperation. In the intermittent XT group, of 65 patients with a successful outcome, 53 (82%) had initial alignment within 10D of orthotropia. Of 68 patients with longterm recurrence, 55 (81%) had initial undercorrection of 2D to 30D. All patients with long-term overcorrection had initial overcorrection of 2D to 30D. For patients with initial overcorrection, the chance for successful outcome was 60%; for those with initial undercorrection, less than 40%. By far the most favorable ratio of successful to unsuccessful outcomes occurred with initial alignments of 0D to 10D of esotropia in both the infantile XT and intermittent XT groups (86% vs 65%).
278
Yam et al
Volume 16 Number 3 / June 2012
Table 4. Three-year primary surgical outcomes as a function of initial postoperative alignment in infantile exotropia and intermittent exotropia, grouped by increments of 10 prism diopters Infantile exotropia (n 5 37) Initial postoperative alignment, PD
a
\ 20D 10 to 20D 0 to 10D 1 to 10D 10 to 20D .20D Total number (N)
Intermittent exotropia (n 5 144)
S
R
O
N
S
R
O
N
0 2 (67%)b 12 (86%) 0 0 0 14
0 1 (33%) 2 (14%) 12 (100%) 4 (100%) 3 (100%) 22
1 (100%) 0 0 0 0 0 1
1 3 14 12 4 3 37
0 8 (57%) 28 (65%) 25 (40%) 4 (21%) 0 65
0 2 (14%) 11 (26%) 38 (60%) 15 (8%) 2 (100%) 68
3 (100%) 4 (29%) 4 (9%) 0 0 0 11
3 14 43 63 19 2 144
N, total number of patients; O, overcorrection group; PD, prism diopters; R, recurrence group; S, success group. Minus angle denotes esotropia; plus angle, exotropia. b Percentage is of surgical outcome at that initial alignment group. a
Table 5. Binocular vision status and final 3-year surgical outcomes Infantile exotropia (n 5 34)
Intermittent exotropia (n 5 144) Stereopsis level
Final 3-year outcome Success Recurrence Overcorrection Total number a
\100
100-400
400-3000
nil
n
\100
100-400
400-3000
nil
n
5 (38%)a 2 (11%) 0 7
2 (15%) 11 (58%) 0 13
3 (23%) 1 (5%) 0 4
3 (23%) 5 (26%) 2 (100%) 10
13 19 2 34
45 (62%) 8 (12%) 0 53
26 (36%) 50 (77%) 4 (67%) 80
2 (3%) 7 (11%) 2 (33%) 11
0 0 0 0
73 65 6 144
Percentage is of stereopsis category of that surgical outcome group.
Binocular Function and Amblyopia Rate The level of stereopsis in both groups is given in Table 2 and the level of stereopsis according to the final motor outcome in Table 5. We were able to test stereoacuity in 34 of 37 patients in infantile XT group and all patients in intermittent XT group. The intermittent XT group had a better level of stereopsis than that of the infantile XT group (P \ 0.001). In both groups, patients with final success can achieve better stereopsis. The amblyopia rate in both groups is summarized in Table 2. The infantile XT group had a greater rate of amblyopia than the intermittent XT group preoperatively (P 5 0.004) and at 3 years’ followup (P 5 0.001).
Discussion Biglan and colleagues8 and Hunter and colleagues16 found that approximately equal numbers of patients with infantile exotropia have intermittent or constant exotropia at presentation and that clinical outcomes are similar regardless of presentation. Hunter and colleagues16 suggested that patients presenting with either constant or intermittent exotropia before 1 year of age may represent a spectrum of the same entity of infantile exotropia and should be included in analysis. In subsequent studies, researchers adopted this definition for case recruitment and analysis of infantile exotropia.9,17 It is not known whether the surgical overcorrection generally believed to be helpful for patients with intermittent exotropia is also beneficial for patients with infantile exotropia. Our results suggest that exotropic drift occurred in infantile exotropia during
a 3-year period and that the extent of drift was comparable at 3 years postoperatively to that seen in intermittent exotropia patients. Our subgroup analysis suggested that exotropic drift at 3 years may be smaller in patients with pattern deviation. This finding is consistent with the literature on intermittent exotropia patients.12 Yet, the drift between pattern and comitant subgroup in intermittent exotropia was not significantly different in our series. This may reflect either the absence of such mechanical differences or simply inadequate power to detect a difference due to the small sample size. Finally, the 2 subgroups were not well matched for comparison, as in the study by Pineles and colleagues.12 Our results suggest that initial overcorrection up to 10D of esotropia may be associated to the best-long term outcome. The importance of initial overcorrection for intermittent exotropia remains controversial.2-7,15,18 Oh and Hwan13 suggested that initial postoperative deviation was the only factor found to determine the outcome of exotropia. Choi and colleagues,18 however, found that initial overcorrection may be associated with lower probability of recurrence only within 2 years of surgery but not predictive beyond that. The disparities between the findings of different studies may be attributed to differences of the surgical procedure used, the definition of recurrence, and the follow-up period.18 Pratt-Johnson and colleagues19 cautioned that monofixation syndrome and amblyopia may develop in very young children when exotropia is overcorrected, especially if the deviating eye is left in an esotropic position for a prolonged period. In our series, the 18 cases of initial overcorrection
Journal of AAPOS
Volume 16 Number 3 / June 2012 (range, 2D-22D) all developed exotropic drift, with progression to \4D of esotropia in 15 patients (83%) 1 month postoperatively and 17 patients (94%) 3 months postoperatively. Only 1 patient with large initial overcorrection .20D remained overcorrected during the postoperative period and required reoperation. Our results have also shown that initial overcorrection does not lead to a higher rate of amblyopia development. It is difficult for patients with infantile exotropia to achieve binocular vision despite successful alignment. Biedner and colleagues20 reported that satisfactory alignment was achieved in 84% of cases but that no patient demonstrated fusion or stereoacuity. Our results demonstrated that patients with infantile exotropia had significantly poorer stereopsis than those with intermittent exotropia. In both groups, patients with successful final outcome can achieve better fine stereopsis than those with unsuccessful final outcomes. Thus, the level of attainable stereoacuity may be associated with the final motor alignment. Yet, it has also been suggested that stereoacuity may be associated with the initial presentation of constant or intermittent phase in infantile exotropia before surgery.16 In our study, of 7 patients in the infantile XT group with constant preoperative exotropia, 5 had nil stereopsis and 1 had stereopsis poorer than 400 arcsec. On the other hand, 19 of 27 patients in the infantile XT group with intermittent preoperative alignment achieved stereopsis better than 400 arsec. This finding is consistent with the findings of Hunter and colleagues,16 who reported that 7 of 10 testable patients achieved gross fusion after surgery and that 2 patients with intermittent exotropia at presentation acquired 200 arcsec of stereopsis but that no patients with constant exotropia achieved this level. Our study has several limitations. The number of recruited patients was relatively low; however, because infantile exotropia occurs only in 1 of 30,000 births in the general population,20 this limitation seems unavoidable. Other limitations included its retrospective nature, the contribution of two separate surgeons, imperfectly matched and different sample sizes between the 2 study groups, and nonuniform follow-up among individual cases. The onset of exotropia in this series was based on patient history confirmed with photographic documentation, which may have resulted in recall bias that would have been avoided by including only patients whose onset of symptoms were clinically confirmed. In conclusion, postoperative exotropic drift occurs in infantile exotropia, and the drift is clinically similar to that observed in intermittent exotropia up to 3 years after surgery. Initial overcorrection within 10D to orthophoria during the early postoperative period may be associated to the best long-term ocular alignment
Journal of AAPOS
Yam et al
279
over 3 years in both the infantile and intermittent exotropia.
Literature Search The authors conducted a MEDLINE (PubMed) search for English-language articles only for the period 1966 to 2010, using the following terms: infantile exotropia, congenital exotropia, intermittent exotropia, and exotropic drift. References 1. Kraft SP. Infantile exotropia. In: Rosenbaum AL, Santiago AP, editors. Clinical strabismus management: Principles and surgical techinques. Philadelphia, PA: W.B. Saunders; 1999:176-81. 2. Leow PL, Ko ST, Wu PK, Chan CW. Exotropic drift and ocular alignment after surgical correction for intermittent exotropia. J Pediatr Ophthalmol Strabismus 2010;47:12-16. 3. Ruttum MS. Initial versus subsequent postoperative motor alignment in intermittent exotropia. J AAPOS 1997;1:88-91. 4. Raab EI, Parks MM. Recession of the lateral recti. Early and late postoperative alignments. Arch Ophthalmol 1969;82:203-8. 5. Scott WE, Keech R, Mash AJ. The postoperative results and stability of exodeviations. Arch Ophthalmol 1981;99:1814-18. 6. McNeer KW. Observations on the surgical overcorrection of childhood intermittent exotropia. Am Orthoptic J 1987;37:135-50. 7. Souza-Dias C, Uesugi CF. Postoperative evolution of the planned initial overcorrection in intermittent exotropia: 61 cases. Binocular Vis Eye Muscle Surg Q 1993;8:141-8. 8. Biglan AW, Davis JS, Cheng KP, Pettapiece MC. Infantile exotropia. J Pediatr Ophthalmol Strabismus 1996;33:79-84. 9. Park JH, Kim SH. Clinical features and the risk factors of infantile exotropia recurrence. Am J Ophthamol 2010;150:464-7. 10. Baeteman C, Denis D, Loudot C, et al. Primary exotropia: Importance of cerebral MRI [in French]. J Fr Ophthalmol 2008;31:287-94. 11. Hunter DG, Ellis FJ. Prevalence of systemic and ocular disease in infantile exotropia: Comparison with infantile esotropia. Ophthalmology 1999;106:1951-6. 12. Pineles SL, Rosenbaum AL, Demer JL. Decreased postoperative drift in intermittent exotropia associated with A and V patterns. J AAPOS 2009;13:127-31. 13. Olitsky S. Early and late postoperative alignment following unilateral lateral rectus recession for intermittent exotropia. J Pediatr Ophthalmol Strabismus 1998;35:146-8. 14. Lee SY, Lee YC. Relationship between motor alignment at postoperative day 1 and at year 1 after symmetric and asymmetric surgery in intermittent exotropia. Jpn J Ophthalmol 2001;45:167-71. 15. Oh JY, Hwan JM. Survival analysis of 365 patients with exotropia after surgery. Eye (Lond) 2006;20:1268-72. 16. Hunter DG, Kelly JB, Buffenn AN, Ellis FJ. Long-term outcome of uncomplicated infantile exotropia. J AAPOS 2001;5:352-6. 17. Paik HJ, Yim HB. Clinical effect of early surgery in infantile exotropia. Korean J Ophthalmol 2002;16:97-102. 18. Choi J, Kim SJ, Yu YS. Initial postoperative deviation as a predictor of long-term outcome after surgery for intermittent exotropia. J AAPOS 2011;15:224-9. 19. Pratt-Johnson JA, Barlow JM, Tillson G. Early surgery in intermittent exotropia. Am J Ophthalmol 1977;84:689-94. 20. Biedner B, Marcus M, David R, Yassur Y. Congenital constant exotropia: Surgical results in six patients. Binocul Vis Eye Muscle Surg Q 1993;8:137-40.