- Email: [email protected]
Long-term results of the surgical management of intermittent exotropia
Long-term results of the surgical management of intermittent exotropia Stacy L. Pineles, MD, Noa Ela-Dalman, MD, Anna G. Zvansky, Fei Yu, PhD, and Arthur L. Rosenbaum, MDy PURPOSE
To examine long-term surgical success rates (.10 years) for patients with intermittent exotropia and the risk factors for failure of surgery in these patients.
METHODS
An attempt was made to contact all patients who underwent surgical treatment for intermittent exotropia between the years of 1970 to 1998 with a minimum postoperative follow-up of 10 years. Each patient underwent a detailed sensory and motor examination, including measurements of near and distance stereoacuity, cover testing, and ocular rotations. Patients were classified as achieving an excellent, fair, or poor outcome on the basis of motor and sensory outcomes. Risk factor analysis was performed to evaluate associations with a poor outcome and reoperations.
RESULTS
Of 197 patients identified, 50 were reevaluated. When combined motor/sensory criteria for surgical success were used, we found that 38% of patients achieved an excellent outcome, whereas 34% and 28% achieved a fair or poor outcome, respectively. When only the motor criteria were used, we found that 64% had an excellent outcome, whereas the remaining patients achieved either a fair (18%) or a poor (18%) outcome. During the follow-up period, 60% of patients required at least one reoperation. Multivariate risk factor analysis determined that anisometropia (p 5 0.03) was associated with a poor outcome, whereas postoperative undercorrection (p 5 0.04) and lateral incomitance (p 5 0.06) were associated with reoperations. Long-term surgical results in intermittent exotropia are less encouraging when sensory status is added to the evaluation. Patients with anisometropia, lateral incomitance, and immediate postoperative undercorrection are at increased risk for poor outcomes and to require reoperations. ( J AAPOS 2010;14:298-304)
CONCLUSIONS
I
ntermittent exotropia is a widely studied disease, yet its treatment remains controversial. Discrepancies among clinicians in the management of intermittent exotropia stem not only from a lack of consensus regarding optimal indications and timing for surgical interventions but also from a highly variable success rate in the published literature, ranging from 42% to 81%.1-7 The numerous studies published in the last 50 years regarding outcomes in the treatment of intermittent exotropia have been
Author affiliations: Jules Stein Eye Institute and Department of Ophthalmology, University of California, Los Angeles Presented at the 36th Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando, Florida, April 14-18, 2010. Dr. Pineles is the recipient of the Heed Fellowship and The Leonard Apt Fellowship; Dr. Rosenbaum was the recipient of Research to Prevent Blindness Physician-Scientist Merit Award. y Deceased Submitted February 18, 2010. Revision accepted June 2, 2010. Reprint requests: Stacy L. Pineles, MD, Jules Stein Eye Institute, 100 Stein Plaza, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095-7002 (email: [email protected]). Copyright Ó 2010 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2010/$36.00 1 0 doi:10.1016/j.jaapos.2010.06.007
298
limited in two important respects. First, there is a dearth of studies that include very long-term minimum followup of postoperative patients; second, most studies have failed to include stringent motor and sensory criteria in the assessment of surgical success. There have been only 2 studies of patients with intermittent exotropia with a minimum postoperative follow-up of at least 10 years, but both of these series included 30 or fewer patients.8,9 The fundamental lack of long-term data is especially important given that failure rates for surgical interventions in intermittent exotropia increase as postoperative follow-up time increases.2 Moreover, authors of multivariate analyses have indicated that the greatest risk factor for surgical failure in intermittent exotropia is increasing follow-up length.2 To truly define postoperative success rates, as well as preoperative predictors of surgical failure, one must study intermittent exotropia patients after an extended postoperative follow-up. In addition to the lack of lengthy follow-up, intermittent exotropia studies have been limited by variability in the criteria used to determine surgical success. The authors of most studies have used only postoperative motor alignment; they determined an alignment of orthotropia 8D-10D as a successful outcome.2,10-14 However, many
Journal of AAPOS
Volume 14 Number 4 / August 2010 authors believe that sensory data should be included in any outcome study of intermittent exotropia,3,5,15-17 especially in light of the excellent stereoacuity enjoyed by most patients with intermittent exotropia who do not undergo surgery. Finally, in addition to motor and sensory criteria, some surgeons believe that an objective measure of intermittent exotropia control, such as distance/near stereopsis disparities, should be included in the assessment of surgical success because it is an important patient-oriented outcome that defines the amount of time during which a deviation is manifest as a tropia. The purpose of this study was to describe the long-term (.10 years) surgical outcome of patients with intermittent exotropia by the use of stringent motor and sensory criteria as well as a measure of control. In addition, we sought to determine whether any preoperative or postoperative factors were associated with poor surgical outcomes in our population.
Methods This study was approved by the University of California Los Angeles Institutional Review Board and conformed to the requirements of the United States Health Insurance Portability and Accountability act; all subjects gave written informed consent before participation in conformity with the Declaration of Helsinki. The surgical records of all cases from 1970 to 1998 of one surgeon (ALR) were reviewed. An attempt was made to contact all patients who underwent surgery for intermittent exotropia between 1970 and 1998. Patients were excluded if there was any neurologic deficit, coexistent restrictive or paretic strabismus, constant exotropia on presentation, or history strabismus surgery. Intermittent exotropia was defined as a divergent strabismus that alternates between phoric and tropic phases. The following preoperative characteristics were recorded from the patients’ charts: age at onset, age at surgery, best-corrected visual acuity, cycloplegic refraction, preoperative motor alignment at distance and near, stereoacuity at distance and near, presence of an A or V pattern, and postoperative motor alignment on postoperative day 1. In addition, all subsequent surgical procedures and complications were noted. Surgical techniques were determined by the surgeon’s preference and are published elsewhere.18 In general, the surgeon operated for the largest angle of exotropia (ie, distance angle for pseudodivergence excess and near angle for convergence insufficiency). Lateral rectus recessions were usually chosen for divergence excess or pseudodivergence excess, and medial rectus resections were chosen for convergence insufficiency. For basic exotropias, surgeon preference changed from lateral rectus recessions to unilateral recession-resection procedures during the mid-1990s. Adjustable sutures initially were used by surgeon in 1975. The decision to reoperate was determined by surgeon and patient preferences, with indications including diplopia, progressive loss of stereopsis, consecutive esotropia .10D, or recurrence of exotropia. Patients were classified as having a basic intermittent exotropia if the distance deviation was within 10D of that at near. Divergence excess intermittent exotropia was defined as a distance deviation
Journal of AAPOS
Pineles et al
299
at least 10D greater than that at near. Pseudodivergence excess was defined as an exotropia that met criteria for divergence excess but had an increase in the near deviation to an amount within 10D of the distance deviation after prolonged monocular occlusion or with the use of 13.00 D lenses. Convergence insufficiency was defined as an intermittent exotropia at near measuring at least 10D greater than that at distance. Patients were considered to have lateral incomitance if the intermittent exotropia was at least 10D different between central and lateral gazes. All study examinations were performed by one of the coauthors (SLP) or (NED). Patients were questioned regarding subjective symptoms. Distance stereoacuity (Mentor B-Vat II BVS) and near stereoacuity (titmus test) were evaluated first. Visual acuity was assessed by use of projected Snellen optotypes after a manifest refraction. Ocular alignment was assessed by the use of cover/ uncover and alternate prism cover testing at distance (20 feet) in primary position and in the cardinal gaze positions. Motor alignment at near was assessed at 14 inches. All motor evaluations were done with the use of spectacle correction. On the basis of the study examination, patients were classified as having either an excellent, fair, or poor surgical outcome (Table 1). To be considered an excellent or fair outcome, patients met all 3 criteria listed in Table 1. All patients who did not meet the criteria for excellent or fair outcomes were considered to have a poor outcome. Surgical success rates were calculated by the use of the motor/ sensory criteria in Table 1. For comparison, success rates were also calculated by use of only the motor criteria in Table 1. In addition, patients were evaluated for the presence or absence of multiple surgeries. After patients were categorized into subgroups on the basis of surgical success and the need for multiple surgeries, an attempt was made to define pre- and postoperative risk factors for poor outcomes and multiple surgeries. Additional subgroup analysis was performed to evaluate overall outcomes in patients who presented before or after 8 years of age, and by exotropia type.
Statistical Analysis SAS version 9.1 (SAS Institute, Cary, NC) was used for statistical analysis. The associations between preoperative variables and postoperative day 1 alignments with the outcome status were assessed with the use of the Fisher exact tests. For postoperative day 1 alignments, univariate analysis comparisons were made between esotropic deviations and exotropic deviations. The associations between continuous variables (age, time periods before and after surgery, prism diopters of exotropia) and the outcome status were assessed by the use of Kruskal-Wallis tests. In addition, a multivariate logistic regression was performed to determine associations with surgical success (excellent or fair vs poor outcome) and reoperations; however, because of the large number of variables, only those variables with a univariate association near significance (p \ 0.2) were included. For the logistic regression models, postoperative day 1 alignment (prism diopters of deviation) was a continuous variable, with esotropic deviations denoted by negative numbers, and exotropic deviations having positive values. Direct comparison between means was assessed using a Student’s t-test. We considered p-values #0.05 as statistically significant. Finally,
300
Pineles et al
Volume 14 Number 4 / August 2010
Table 1. Criteria for surgical success Criteria Motor Near stereoacuitya Distance stereoacuity
Excellent D
0-8 X(T) \2 line worsening from preoperative measurement \2 line discrepancy from postoperative near stereoacuity
Fair D
D
#4 ET or 9-15 X(T) 2-4 line worsening from preoperative measurement 2- to 4-line discrepancy from post-operative near stereoacuity
Poor D
.4 ET, or .15D X(T) .4 line worsening from preoperative measurement .4 line discrepancy from postoperative near stereoacuity
X(T), intermittent exotropia; ET, esotropia. a Near stereoacuity not included in criteria if patient was too young for stereoacuity testing preoperatively. the rate of reoperations was illustrated by use of the Kaplan-Meier method.
Results Patient Characteristics A total of 197 patients met the inclusion criteria. Of these, 60 could be reached via telephone. Fifty patients were able to return for an examination. The remaining 10 patients did not reside in the area and therefore were unable to participate. In the 50 subjects, the mean age at the time of the study examination was 28.6 17.6 years (range, 11-75 years). The mean age at the time of the first surgery was 14.3 17.7 years (range, 2-65 years). The mean number of years since the first surgery was 14.3 4.8 (range, 10-29 years). Of the 50 subjects, 23 were men. Thirty patients (60%) required additional strabismus surgical procedures after the initial surgery; 28 of 30 reoperations were performed by the same surgeon. The pattern of exotropia was basic in 34 patients, convergence insufficiency in 9 patients, and pseudo-divergence excess in 7 patients. No patient had the true divergence excess type of exotropia. Surgical Success With Motor Criteria Only Patients were classified into groups on the basis solely of the motor criteria delineated in Table 1. Thirty-two (64%) of the patients had an excellent motor alignment postoperatively. Nine patients (18%) had a fair motor alignment, and 9 patients (18%) had a poor motor alignment. Surgical Success With Motor and Sensory Criteria Patients were reclassified into groups to determine surgical success on the basis of the motor and sensory criteria listed in Table 1. Nineteen (38%) of the patients achieved excellent results when we used the more stringent criteria. Seventeen patients (34%) had a fair outcome, and 14 (28%) had a poor outcome. Patient characteristics for the excellent and poor groups can be found in Table 2 (first 3 columns). Overall, 27% of patients in the fair group and 50% of patients in the poor group, respectively, lost 2 or more lines of near stereoacuity. The mean preoperative and postoperative near stereoacuity in those patients who had preoperative measurements taken (n 5 32) was 49 17 arcsec and 45 16 arcsec for the excellent group, 48 9 arcsec and 62 49 arcsec for the fair group, and 53
14 arcsec and 445 260 arcsec for the poor group. Although none of these changes in the mean stereoacuity for each group reached statistical significance, changes in individual patients’ stereoacuity scores in the fair and poor group were considered clinically significant by definition. There was no significant difference in the mean follow-up time from the first surgery amongst groups. Multiple Surgeries Thirty (60%) of the study patients required additional procedures after the initial surgery. A Kaplan-Meier survival curve detailing the cumulative rate of reoperation is depicted in Figure 1. Patient characteristics for this group can be found in Table 3. Surgery for a consecutive esotropia was performed for 6 of 30 patients, and the remaining 24 of 30 patients had surgery for residual or recurrent intermittent exotropia. The mean time to reoperation was 7.4 8 years (range, 6 months to 31 years). Seven patients underwent multiple reoperations, with 3 patients undergoing 2 reoperations, 3 patients undergoing 3 reoperations, and 1 patient undergoing 4 reoperations. Risk Factor Analysis for Poor Outcomes An analysis of pre- and postoperative risk factors was performed for both the motor only and motor/sensory success criteria. There were no statistically significant associations found when the combined motor and sensory criteria were used. The results of multiple comparisons for the motor criteria only are found in Table 2. Univariate analysis revealed that anisometropia (p 5 0.02) and a unilateral recession-resection procedure (p 5 0.04) were significantly associated with a poor motor outcome, whereas lateral incomitance (p 5 0.06) and a postoperative day 1 undercorrection (p 5 0.15) trended toward an association with a poor outcome. However, multivariate analysis comparing excellent and fair motor results to poor results revealed that anisometropia (p 5 0.03) was the only truly significant association, although lateral incomitance approached significance with a p-value of 0.07. Univariate analysis revealed that several variables had either significant or near-significant associations with a patient receiving additional surgery (Table 3), including postoperative day 1 undercorrection (p 5 0.006), a younger age of onset (p 5 0.06), lateral incomitance (p 5 0.06), and a younger age at surgery (p 5 0.08). In addition, a longer postoperative follow-up time from the last surgery was
Journal of AAPOS
Volume 14 Number 4 / August 2010
Pineles et al
301
Table 2. Comparisons of pre- and postoperative factors for 10-year postoperative status
Factors Age at first surgery (yrs) Age at examination (yrs) Age at onset (yrs) Time from onset to surgery, yr Time since first surgery, ys Preoperative deviation Pattern strabismus Type of exotropia Basic CI Pseudodiv excess Presence of anisometropia POD1 alignment:ET-ortho Lateral incomitance First surgery LR recession MR resection R&R Incom R&R Adjustable suture
Excellent motor/sensory (n 5 19)
Poor motor/sensory (n 5 14)
12.7 15.3 (2-47) 28.3 16.3 (11-60) 4.6 3.6 (0.5-13) 8.1 13.0 (0-37) 14.7 5.8 (10-29) 24.2 10.9 (12-54) 1 (5%)
19.1 22.8 (2-65) 34.1 24.0 (12-75) 9.3 13.5 (0.7-40) 9.8 14.4 (0-40) 12.8 3.3 (10-18) 26.9 9.1 (12-45) 3 (21%)
14 (74%) 1 (5%) 4 (21%) 2 (11%)
9 (64%) 3 (21%) 2 (14%) 3 (21%)
12 (63%) 1 (5%)
5 (36%) 3 (21%)
12 (63%) 2 (11%) 5 (26%) 0 7 (37%)
5 (36%) 2 (14%) 6 (43%) 1 (7%) 6 (43%)
p-value motor/sensory 0.56 0.66 0.66 0.93 0.32 0.28 0.29 0.14
0.63 0.17 0.29 0.38
1.0
Excellent motor only (n 5 32) 13.1 16.9 6 (2-65) 27.9 17.1 20 (11-75) 5.2 7.1 4 (0.5-40) 7.8 12.7 1.4 (0-44) 14.3 5.3 12.5 (10-29) 24.0 10.6 23.5 (12-54) 4 (13%)
Poor motor only (n 5 9) 19.7 22.3 3 (2-55) 36.3 24.2 20 (13-72) 8.1 12.8 2.5 (0.7-40) 11.5 16.5 2.0 (0.5-40) 13.3 3.2 13 (10-18) 25.3 5.3 25 (18-35) 2 (22%)
22 (69%) 2 (6%) 8 (25%) 2 (6%)
6 (67%) 2 (22%) 1 (11%) 4 (44%)
20 (63%) 2 (6%)
3 (33%) 3 (33%)
21 (66%) 4 (13%) 7 (22%) 0 12 (38%)
3 (33%) 0 5 (56%) 1 (11%) 4 (44%)
p-value motor only (multivariate analysis) 0.94 0.62 1.0 0.51 0.85 0.36 0.60 0.28
0.02 (0.03) 0.15 (0.5) 0.06 (0.07) 0.04 (0.14)a
0.72
The first 3 columns indicate surgical results using combined motor and sensory criteria; the last 3 columns indicate surgical results using motor criteria only. Results of multivariate logistic regression are shown in parentheses in the last column. CI, convergence insufficiency; Pseudodiv excess, pseudo-divergence excess; POD1, postoperative day 1; ET, esotropia; ortho, orthotropia; LR, lateral rectus; MR, medial rectus; R&R, combined recession-resection procedure; Incom R&R, incomitant recession-resection procedure (ie, recession on lateral rectus of one eye and resection on medial rectus of the fellow eye). a This p-value (0.04) indicates that a recession-resection procedure was significantly associated with a poor motor outcome when compared with the other possible surgical methods in a univariate analysis.
significantly associated with increased number of reoperations (p 5 0.0008). A multivariate analysis revealed that only postoperative day 1 undercorrection (p 5 0.04) was a truly significant association. Lateral incomitance could not be used as a possible associated factor in the multivariate analysis since all of the patients with lateral incomitance had reoperations. Subgroup Analysis Of those patients who presented before they were 8 years of age (n 5 42), 40% had an excellent result compared with 35% of those who presented after they were 8 years of age. Reoperations were required in 62% of patients who presented before they were 8 years of age compared with 50% of those who presented after they were 8 years of age. Of the 50 patients studied, 9 had convergence insufficiency pattern exotropia (mean age of onset, 10.4 12.1 years). Of these, 1 (11%) had excellent results. Of the 50 patients studied, 34 had basic type intermittent exotropia (mean age of onset, 4.1 3.1 years); 41% (14 of 34) of these patients had excellent outcomes. Finally, 7 patients had pseudo-divergence excess pattern exotropia (mean age of
Journal of AAPOS
onset 5 2.4 1.6 years); 57% (4 of 7) of these patients had excellent outcomes. Of convergence insufficiency, basic, and pseudo-divergence excess patients, 67%, 62%, and 43% underwent reoperations, respectively. Although none of the aforementioned differences amongst groups reached statistical significance on the basis of age or exotropia type, there was a significant difference in the mean age of the convergence insufficiency patients compared with the basic and pseudo-divergence excess groups (p 5 0.004 and 0.05, respectively).
Subjective Symptoms Upon questioning, 7 patients (14%) complained of subjective visual symptoms. Three patients complained of difficulty with depth perception. The remaining 4 patients complained of intermittent diplopia (2 were esotropic and 2 were exotropic). Of the 7 patients, 4 had been classified as a poor outcome, and 3 patients had a fair outcome. Three of the patients had convergence insufficiency intermittent exotropia, and the remaining 4 patients had basic intermittent exotropia.
302
Pineles et al
Volume 14 Number 4 / August 2010
Cumulative proportion without reoperation
1.0
Table 3. Comparisons of pre- and postoperative factors for risk of the need for multiple surgery
0.8 0.6
Factors
0.4 0.2 0.0 0
5
10
15
20
25
30
Length of follow-up (years)
FIG 1. Kaplan-Meier analysis depicting the proportion of intermittent exotropia patients not reoperation over time.
Discussion This study represents the largest single-surgeon series of patients with intermittent exotropia, with a minimum of 10 years’ follow-up from the first procedure. In our cohort, patients were almost evenly distributed amongst excellent, fair, and poor outcomes when stringent sensory and motor criteria were used. When we eliminated the sensory criteria from the analysis, the number of patients with an excellent result increased, and the percentage of patients with a poor result decreased. Within this classification scheme, we were able to detect several risk factors for a poor motor outcome that approached statistical significance. Outcomes of surgical interventions for intermittent exotropia have been studied since the 1960s.14,19 Despite the immense body of literature, there is still not a consensus on fundamental topics such as the optimal timing for interventions, choice of surgical technique, success rates, or the criteria for a successful outcome. For example, success rates in the literature vary widely from 42% to 81%,1-7 and success appears to be directly related to follow-up length and stringency of success criteria. There have been other authors who have evaluated surgical success in patients followed for a minimum of 1 year,3,5,6,8,11,20-24 but only 2 groups were examined after a minimum follow-up of at least 10 years.8,9 Zibrandtsen and colleagues9 reviewed the results of 25 intermittent exotropia patients who had at least 10 years of follow-up and found that approximately 50% of the patients had a good long-term result, with a small exophoria and acceptable binocularity. These authors did not attempt to systematically categorize success, nor did they evaluate risk factors for surgical failure. In 2008 Baker8 described a cohort of 30 patients prospectively re-evaluated with at least 20 years of follow-up. However, he did not make any attempt to classify success rates or risk factors for failure. Other authors claim to contain long-term follow-up data with the use of survival analyses, but they include subjects with follow-up time of less than 1 year in their analysis. Finally, few of the heretofore described studies actually included a measure of stereopsis in their analysis.2,3,5,15-17 Given that most patients with intermittent exotropia benefit from excellent near stereoacuity before surgical
Age at first surgery (yrs) Age at examination (yrs) Age at onset (yrs) Time from onset to surgery (yrs) Time since first surgery (yrs) Time since last surgery (yrs)a Preoperative deviation Presence of pattern strabismus Type of exotropia Basic Convergence insufficiency Pseudo-divergence excess Presence of anisometropia (n 5 number with amblyopia) Postoperative day 1 alignmentc Amblyopia Lateral incomitance First surgery Lateral rectus recession Medial rectus resection Combined recession-resection Incomitant recession-resection Adjustable suture
Multiple (n 5 30)
Single (n 5 20)
11.2 15.2 16.7 18.7 4.5 (2-55) 9 (2-65) 27.9 17.0 29.6 18.9 20 (12-72) 21.5 (11-75) 4.8 7.5 6.6 8.4 2.3 (0.5-40) 5 (0.5-40) 6.4 11.0 10.0 14.2 2 (0-44) 2.8 (0-40) 14.8 4.6 13.5 5.1 13.5 (10-29) 12 (10-29) 7.3 5.0 13.1 3.9 8 (1-16) 25 (12-54) 23.8 8.6 25.1 10.5 24 (12-45) 25 (12-54) 4 (13%) 2 (10%)
p-value (multivariate analysis) 0.08 (0.9) 0.85 0.06 (0.4) 0.64 0.13 (0.2) 0.0008 (0.008) 0.76 1.0 0.39
21 (70%) 6 (20%)
13 (65%) 3 (15%)
3 (10%)
4 (20%)
6 (20%) n51
2 (10%) n50
0.45
12 (40%)
15 (75%)
3 (10%) 6 (20%)
0 0
0.006 (0.04) 0.27 0.06b 0.85
17 (57%)
13 (65%)
2 (7%)
2 (10%)
10 (33%)
5 (25%)
1 (3%)
0
11 (37%)
7 (35%)
1.0
Results of multivariate logistic regression are shown in parentheses in the last column. a Time since the last surgery was not considered a true risk factor for multiple reoperations because it simply reflects that an intervening surgery occurred after the first operation. b Lateral incomitance could not be included in the multivariate logistic regression because all of the patients with lateral incomitance had multiple reoperations. c Alignment after final surgery, indicating number of patients with esotropia orthotropia on day 1 after surgery.
interventions, any degradation of stereoacuity that can be attributed to a surgical outcome (ie, postoperative esodeviations), must be evaluated in the functional outcome assessment.5 Pratt-Johnson and colleagues5 elucidated the importance of stereopsis in success criteria when they reported 81% success in motor alignment, but only 41% of patients achieving stereopsis of 40 arcsec in a group of 100 patients with divergence excess–type intermittent
Journal of AAPOS
Volume 14 Number 4 / August 2010 exotropia. In addition, we assert that control, as measured in this study by distance stereopsis (and its discrepancy from near stereopsis), is another important factor in the measurement of success.25 Control is an important patient-oriented outcome because it indicates the amount of time that exotropia is manifest in a patient’s everyday life; it has also been assessed by the use of control scales or parental report by other authors.26 Our data confirm several previously described findings. In our cohort, for an excellent motor result, the determined success rate of 64% falls well within the limits defined by previous authors; yet, when one considers sensory outcomes, our success rate of 38% falls below most previous studies. In addition, our finding of increased risk for the need for reoperation with early undercorrection confirms the previous findings of Raab and Parks,14 who first described early overcorrection as a predictor of the best long-term results. Our study also corroborates the previous findings of lateral incomitance and anisometropia as associations with a poor motor outcome.27,28 In addition to evaluating risk factors for surgical success and failure, we also evaluated risk factors for multiple surgeries. In an outcomes study, one must consider that subjects may have differing postoperative goals and risk tolerance. For example, if 2 subjects have similarly poor outcomes from their first surgery, one may choose to pursue a second surgery and obtain an excellent final result, whereas another subject may choose not to undergo a reoperation. Therefore, in an attempt to address this potentially confounding variable, we included a risk factor analysis for the need for multiple surgeries. The 2 factors identified that trended toward a significant association with patients undergoing multiple surgeries (postoperative day undercorrection and lateral incomitance) are well known risk factors for surgical failure as the result of postoperative exotropic drift and difficulty in surgical planning for the lateral incomitance (ie, asymmetric recessions and resections), respectively. Although not statistically significant, our cohort demonstrated a lower proportion of convergence insufficiency patients with excellent outcomes (11% vs 41% of basic and 57% of pseudo-divergence exotropia, respectively). Although it is unclear what caused poorer outcomes in our patients with convergence insufficiency, theoretical explanations include a tendency to undercorrect at near because of poorly tolerated esotropia at distance, diminished convergent ability predisposing to exotropic drift, or presentation at an older age. Our findings should be understood within the limitations of the study. Importantly, although patients were reevaluated prospectively, their surgeries and preoperative assessments were reviewed in a retrospective fashion. Inherent to study design, there were changes in surgical techniques and preferences during the 28-year period. The range of ages and follow-up time is wide, and therefore demonstrates variability within our patient population. Furthermore, there is some degree of selection bias in
Journal of AAPOS
Pineles et al
303
that although attempt was made to contact all surgical patients during the time period, only 60 patients could be contacted and only 50 patients could be evaluated. However, we believe that our cohort is relatively unbiased given that many of our cases had not been evaluated by us since the first postoperative year and therefore is less subject to the selection bias inherent to studies in which the authors only evaluate patients who have voluntarily maintained follow-up over time. Despite its limitations, our study is the largest single surgeon case series with at least 10 years of postoperative follow-up from the first procedure. Our results confirm that long-term surgical success in intermittent exotropia is less satisfying when sensory outcomes are considered. We assert that this difference is crucial for surgeons to acknowledge in attempting to classify functional surgical outcomes. Clinicians should consider preoperative counseling of patients to ensure their understanding of the increased risk of a poor surgical outcome or the need for multiple surgeries, especially in the presence of lateral incomitance, anisometropia, and immediate postoperative undercorrections. Future prospective studies will likely help in understanding how to further maximize outcomes.
Literature Search The authors performed a MEDLINE search without date restrictions for the following terms: intermittent exotropia, alternating exotropia, strabismus, and strabismus surgery. Additionally, articles not initially found on MEDLINE but referenced in review papers were also included. References 1. Clarke WN, Noel LP. Surgical results in intermittent exotropia. Can J Ophthalmol 1981;16:66. 2. Ekdawi NS, Nusz KJ, Diehl NN, Mohney BG. Postoperative outcomes in children with intermittent exotropia from a populationbased cohort. JAAPOS 2009;13:4-7. 3. Hardesty HH, Boynton JR, Keenan P. Treatment of intermittent exotropia. Arch Ophthalmol 1978;96:268-74. 4. Hardesty H. Management of intermittent exotropia. Binocul Vis Strabismus Q 1990;5:145. 5. Pratt-Johnson JA, Barlow JM, Tillson G. Early surgery in intermittent exotropia. Am J Ophthalmol 1977;84:689-94. 6. Richard JM, Parks MM. Intermittent exotropia: surgical results in different age groups. Ophthalmology 1983;90:1172-7. 7. Weinstein GS, Biglan AW, Hiles DA. Postoperative residual small angle exodeviations. Ophthalmic Surg 1982;13:478. 8. Baker JD. Twenty-year follow-up of surgery for intermittent exotropia. JAAPOS 2008;12:227-32. 9. Zibrandtsen P, Rindziunski E, Gregersen E. Ten years follow-up of surgery for intermittent exotropia. Acta Ophthalmol (Copenh) 1986;64:374-8. 10. Faridi UA, Saleh TA, Weings P, Twomey JM. Factors affecting the surgical outcome of primary exotropia. Strabismus 2007;15:127-31. 11. Gezer A, Sezen F, Nasri N, Gozum N. Factors influencing the outcome of strabismus surgery in patients with exotropia. JAAPOS 2004;8:56-60. 12. Jeoung JW, Lee MJ, Hwang JM. Bilateral lateral rectus recession versus unilateral recess-resect procedure for exotropia with a dominant eye. Am J Ophthalmol 2006;141:683-8.
304
Pineles et al
13. Olitsky SE. Early and late postoperative alignment following unilateral lateral rectus recession for intermittent exotropia. J Pediatr Ophthalmol Strabismus 1998;35:146-8. 14. Raab EL, Parks MM. Recession of the lateral recti: early and late postoperative alignments. Arch Ophthalmol 1969;82:203-8. 15. Beneish R, Flanders M. The role of stereopsis and early postoperative alignment in long-term surgical results of intermittent exotropia. Can J Ophthalmol 1994;29:119-24. 16. Son JH, Huh YS, Kim MM. Surgical outcomes of intermittent exotropia as a function of strabismic angle. Korean J Ophthalmol 2006;20:230-33. 17. Wu H, Sun J, Xia X, Xu L, Xu X. Binocular status after surgery for constant and intermittent exotropia. Am J Ophthalmol 2006;142: 822. e1-822.e7. 18. Santiago AP, Ing MR, Kushner BJ, Rosenbaum AL. Intermittent exotropia. In: Rosenbaum A, Santiago AP, editors. Clinical strabismus management: principles and surgical techniques. Philadelphia (PA): Saunders; 1999:163-75. 19. Burian HM, Spivey BE. The surgical management of exodeviations. Tr Am Ophth Soc 1964;62:276-307. 20. Kushner BJ. Selective surgery for intermittent exotropia based on distance/near differences. Arch Ophthalmol 1998;116:324-8.
Volume 14 Number 4 / August 2010 21. 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. 22. Chia A, Seenyen L, Long QB. Surgical experiences with two-muscle surgery for the treatment of intermittent exotropia. JAAPOS 2006;10: 206-11. 23. Koklanis K, Georgievski Z. Recurrence of intermittent exotropia: factors associated with surgical outcomes. Strabismus 2009;17:37-40. 24. Maruo T, Kubota N, Sakaue T, Usui C. Intermittent exotropia surgery in children: long term outcome regarding changes in binocular alignment. A study of 666 cases. Binocul Vis Strabismus Q 2001;16: 265-9. 25. Stathacopoulos RA, Rosenbaum AL, Zanoni D, et al. Distance stereoacuity: assessing control in intermittent exotropia. Ophthalmology 1993;100:495-500. 26. Hatt SR, Mohney BG, Leske DA, Holmes JH. Variability of control in Intermittent Exotropia. Ophthalmology 2008;115:371-6. 27. Moore S. The prognostic value of lateral gaze measurements in intermittent exotropia. Am Orthopt J 1969;19:69. 28. Gordon YJ, Bachar E. Multiple regression analysis predictor models in exotropia surgery. Am J Ophthalmol 1980;90:687-91.
Journal of AAPOS
To examine long-term surgical success rates (.10 years) for patients with intermittent exotropia and the risk factors for failure of surgery in these patients.
METHODS
An attempt was made to contact all patients who underwent surgical treatment for intermittent exotropia between the years of 1970 to 1998 with a minimum postoperative follow-up of 10 years. Each patient underwent a detailed sensory and motor examination, including measurements of near and distance stereoacuity, cover testing, and ocular rotations. Patients were classified as achieving an excellent, fair, or poor outcome on the basis of motor and sensory outcomes. Risk factor analysis was performed to evaluate associations with a poor outcome and reoperations.
RESULTS
Of 197 patients identified, 50 were reevaluated. When combined motor/sensory criteria for surgical success were used, we found that 38% of patients achieved an excellent outcome, whereas 34% and 28% achieved a fair or poor outcome, respectively. When only the motor criteria were used, we found that 64% had an excellent outcome, whereas the remaining patients achieved either a fair (18%) or a poor (18%) outcome. During the follow-up period, 60% of patients required at least one reoperation. Multivariate risk factor analysis determined that anisometropia (p 5 0.03) was associated with a poor outcome, whereas postoperative undercorrection (p 5 0.04) and lateral incomitance (p 5 0.06) were associated with reoperations. Long-term surgical results in intermittent exotropia are less encouraging when sensory status is added to the evaluation. Patients with anisometropia, lateral incomitance, and immediate postoperative undercorrection are at increased risk for poor outcomes and to require reoperations. ( J AAPOS 2010;14:298-304)
CONCLUSIONS
I
ntermittent exotropia is a widely studied disease, yet its treatment remains controversial. Discrepancies among clinicians in the management of intermittent exotropia stem not only from a lack of consensus regarding optimal indications and timing for surgical interventions but also from a highly variable success rate in the published literature, ranging from 42% to 81%.1-7 The numerous studies published in the last 50 years regarding outcomes in the treatment of intermittent exotropia have been
Author affiliations: Jules Stein Eye Institute and Department of Ophthalmology, University of California, Los Angeles Presented at the 36th Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando, Florida, April 14-18, 2010. Dr. Pineles is the recipient of the Heed Fellowship and The Leonard Apt Fellowship; Dr. Rosenbaum was the recipient of Research to Prevent Blindness Physician-Scientist Merit Award. y Deceased Submitted February 18, 2010. Revision accepted June 2, 2010. Reprint requests: Stacy L. Pineles, MD, Jules Stein Eye Institute, 100 Stein Plaza, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095-7002 (email: [email protected]). Copyright Ó 2010 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2010/$36.00 1 0 doi:10.1016/j.jaapos.2010.06.007
298
limited in two important respects. First, there is a dearth of studies that include very long-term minimum followup of postoperative patients; second, most studies have failed to include stringent motor and sensory criteria in the assessment of surgical success. There have been only 2 studies of patients with intermittent exotropia with a minimum postoperative follow-up of at least 10 years, but both of these series included 30 or fewer patients.8,9 The fundamental lack of long-term data is especially important given that failure rates for surgical interventions in intermittent exotropia increase as postoperative follow-up time increases.2 Moreover, authors of multivariate analyses have indicated that the greatest risk factor for surgical failure in intermittent exotropia is increasing follow-up length.2 To truly define postoperative success rates, as well as preoperative predictors of surgical failure, one must study intermittent exotropia patients after an extended postoperative follow-up. In addition to the lack of lengthy follow-up, intermittent exotropia studies have been limited by variability in the criteria used to determine surgical success. The authors of most studies have used only postoperative motor alignment; they determined an alignment of orthotropia 8D-10D as a successful outcome.2,10-14 However, many
Journal of AAPOS
Volume 14 Number 4 / August 2010 authors believe that sensory data should be included in any outcome study of intermittent exotropia,3,5,15-17 especially in light of the excellent stereoacuity enjoyed by most patients with intermittent exotropia who do not undergo surgery. Finally, in addition to motor and sensory criteria, some surgeons believe that an objective measure of intermittent exotropia control, such as distance/near stereopsis disparities, should be included in the assessment of surgical success because it is an important patient-oriented outcome that defines the amount of time during which a deviation is manifest as a tropia. The purpose of this study was to describe the long-term (.10 years) surgical outcome of patients with intermittent exotropia by the use of stringent motor and sensory criteria as well as a measure of control. In addition, we sought to determine whether any preoperative or postoperative factors were associated with poor surgical outcomes in our population.
Methods This study was approved by the University of California Los Angeles Institutional Review Board and conformed to the requirements of the United States Health Insurance Portability and Accountability act; all subjects gave written informed consent before participation in conformity with the Declaration of Helsinki. The surgical records of all cases from 1970 to 1998 of one surgeon (ALR) were reviewed. An attempt was made to contact all patients who underwent surgery for intermittent exotropia between 1970 and 1998. Patients were excluded if there was any neurologic deficit, coexistent restrictive or paretic strabismus, constant exotropia on presentation, or history strabismus surgery. Intermittent exotropia was defined as a divergent strabismus that alternates between phoric and tropic phases. The following preoperative characteristics were recorded from the patients’ charts: age at onset, age at surgery, best-corrected visual acuity, cycloplegic refraction, preoperative motor alignment at distance and near, stereoacuity at distance and near, presence of an A or V pattern, and postoperative motor alignment on postoperative day 1. In addition, all subsequent surgical procedures and complications were noted. Surgical techniques were determined by the surgeon’s preference and are published elsewhere.18 In general, the surgeon operated for the largest angle of exotropia (ie, distance angle for pseudodivergence excess and near angle for convergence insufficiency). Lateral rectus recessions were usually chosen for divergence excess or pseudodivergence excess, and medial rectus resections were chosen for convergence insufficiency. For basic exotropias, surgeon preference changed from lateral rectus recessions to unilateral recession-resection procedures during the mid-1990s. Adjustable sutures initially were used by surgeon in 1975. The decision to reoperate was determined by surgeon and patient preferences, with indications including diplopia, progressive loss of stereopsis, consecutive esotropia .10D, or recurrence of exotropia. Patients were classified as having a basic intermittent exotropia if the distance deviation was within 10D of that at near. Divergence excess intermittent exotropia was defined as a distance deviation
Journal of AAPOS
Pineles et al
299
at least 10D greater than that at near. Pseudodivergence excess was defined as an exotropia that met criteria for divergence excess but had an increase in the near deviation to an amount within 10D of the distance deviation after prolonged monocular occlusion or with the use of 13.00 D lenses. Convergence insufficiency was defined as an intermittent exotropia at near measuring at least 10D greater than that at distance. Patients were considered to have lateral incomitance if the intermittent exotropia was at least 10D different between central and lateral gazes. All study examinations were performed by one of the coauthors (SLP) or (NED). Patients were questioned regarding subjective symptoms. Distance stereoacuity (Mentor B-Vat II BVS) and near stereoacuity (titmus test) were evaluated first. Visual acuity was assessed by use of projected Snellen optotypes after a manifest refraction. Ocular alignment was assessed by the use of cover/ uncover and alternate prism cover testing at distance (20 feet) in primary position and in the cardinal gaze positions. Motor alignment at near was assessed at 14 inches. All motor evaluations were done with the use of spectacle correction. On the basis of the study examination, patients were classified as having either an excellent, fair, or poor surgical outcome (Table 1). To be considered an excellent or fair outcome, patients met all 3 criteria listed in Table 1. All patients who did not meet the criteria for excellent or fair outcomes were considered to have a poor outcome. Surgical success rates were calculated by the use of the motor/ sensory criteria in Table 1. For comparison, success rates were also calculated by use of only the motor criteria in Table 1. In addition, patients were evaluated for the presence or absence of multiple surgeries. After patients were categorized into subgroups on the basis of surgical success and the need for multiple surgeries, an attempt was made to define pre- and postoperative risk factors for poor outcomes and multiple surgeries. Additional subgroup analysis was performed to evaluate overall outcomes in patients who presented before or after 8 years of age, and by exotropia type.
Statistical Analysis SAS version 9.1 (SAS Institute, Cary, NC) was used for statistical analysis. The associations between preoperative variables and postoperative day 1 alignments with the outcome status were assessed with the use of the Fisher exact tests. For postoperative day 1 alignments, univariate analysis comparisons were made between esotropic deviations and exotropic deviations. The associations between continuous variables (age, time periods before and after surgery, prism diopters of exotropia) and the outcome status were assessed by the use of Kruskal-Wallis tests. In addition, a multivariate logistic regression was performed to determine associations with surgical success (excellent or fair vs poor outcome) and reoperations; however, because of the large number of variables, only those variables with a univariate association near significance (p \ 0.2) were included. For the logistic regression models, postoperative day 1 alignment (prism diopters of deviation) was a continuous variable, with esotropic deviations denoted by negative numbers, and exotropic deviations having positive values. Direct comparison between means was assessed using a Student’s t-test. We considered p-values #0.05 as statistically significant. Finally,
300
Pineles et al
Volume 14 Number 4 / August 2010
Table 1. Criteria for surgical success Criteria Motor Near stereoacuitya Distance stereoacuity
Excellent D
0-8 X(T) \2 line worsening from preoperative measurement \2 line discrepancy from postoperative near stereoacuity
Fair D
D
#4 ET or 9-15 X(T) 2-4 line worsening from preoperative measurement 2- to 4-line discrepancy from post-operative near stereoacuity
Poor D
.4 ET, or .15D X(T) .4 line worsening from preoperative measurement .4 line discrepancy from postoperative near stereoacuity
X(T), intermittent exotropia; ET, esotropia. a Near stereoacuity not included in criteria if patient was too young for stereoacuity testing preoperatively. the rate of reoperations was illustrated by use of the Kaplan-Meier method.
Results Patient Characteristics A total of 197 patients met the inclusion criteria. Of these, 60 could be reached via telephone. Fifty patients were able to return for an examination. The remaining 10 patients did not reside in the area and therefore were unable to participate. In the 50 subjects, the mean age at the time of the study examination was 28.6 17.6 years (range, 11-75 years). The mean age at the time of the first surgery was 14.3 17.7 years (range, 2-65 years). The mean number of years since the first surgery was 14.3 4.8 (range, 10-29 years). Of the 50 subjects, 23 were men. Thirty patients (60%) required additional strabismus surgical procedures after the initial surgery; 28 of 30 reoperations were performed by the same surgeon. The pattern of exotropia was basic in 34 patients, convergence insufficiency in 9 patients, and pseudo-divergence excess in 7 patients. No patient had the true divergence excess type of exotropia. Surgical Success With Motor Criteria Only Patients were classified into groups on the basis solely of the motor criteria delineated in Table 1. Thirty-two (64%) of the patients had an excellent motor alignment postoperatively. Nine patients (18%) had a fair motor alignment, and 9 patients (18%) had a poor motor alignment. Surgical Success With Motor and Sensory Criteria Patients were reclassified into groups to determine surgical success on the basis of the motor and sensory criteria listed in Table 1. Nineteen (38%) of the patients achieved excellent results when we used the more stringent criteria. Seventeen patients (34%) had a fair outcome, and 14 (28%) had a poor outcome. Patient characteristics for the excellent and poor groups can be found in Table 2 (first 3 columns). Overall, 27% of patients in the fair group and 50% of patients in the poor group, respectively, lost 2 or more lines of near stereoacuity. The mean preoperative and postoperative near stereoacuity in those patients who had preoperative measurements taken (n 5 32) was 49 17 arcsec and 45 16 arcsec for the excellent group, 48 9 arcsec and 62 49 arcsec for the fair group, and 53
14 arcsec and 445 260 arcsec for the poor group. Although none of these changes in the mean stereoacuity for each group reached statistical significance, changes in individual patients’ stereoacuity scores in the fair and poor group were considered clinically significant by definition. There was no significant difference in the mean follow-up time from the first surgery amongst groups. Multiple Surgeries Thirty (60%) of the study patients required additional procedures after the initial surgery. A Kaplan-Meier survival curve detailing the cumulative rate of reoperation is depicted in Figure 1. Patient characteristics for this group can be found in Table 3. Surgery for a consecutive esotropia was performed for 6 of 30 patients, and the remaining 24 of 30 patients had surgery for residual or recurrent intermittent exotropia. The mean time to reoperation was 7.4 8 years (range, 6 months to 31 years). Seven patients underwent multiple reoperations, with 3 patients undergoing 2 reoperations, 3 patients undergoing 3 reoperations, and 1 patient undergoing 4 reoperations. Risk Factor Analysis for Poor Outcomes An analysis of pre- and postoperative risk factors was performed for both the motor only and motor/sensory success criteria. There were no statistically significant associations found when the combined motor and sensory criteria were used. The results of multiple comparisons for the motor criteria only are found in Table 2. Univariate analysis revealed that anisometropia (p 5 0.02) and a unilateral recession-resection procedure (p 5 0.04) were significantly associated with a poor motor outcome, whereas lateral incomitance (p 5 0.06) and a postoperative day 1 undercorrection (p 5 0.15) trended toward an association with a poor outcome. However, multivariate analysis comparing excellent and fair motor results to poor results revealed that anisometropia (p 5 0.03) was the only truly significant association, although lateral incomitance approached significance with a p-value of 0.07. Univariate analysis revealed that several variables had either significant or near-significant associations with a patient receiving additional surgery (Table 3), including postoperative day 1 undercorrection (p 5 0.006), a younger age of onset (p 5 0.06), lateral incomitance (p 5 0.06), and a younger age at surgery (p 5 0.08). In addition, a longer postoperative follow-up time from the last surgery was
Journal of AAPOS
Volume 14 Number 4 / August 2010
Pineles et al
301
Table 2. Comparisons of pre- and postoperative factors for 10-year postoperative status
Factors Age at first surgery (yrs) Age at examination (yrs) Age at onset (yrs) Time from onset to surgery, yr Time since first surgery, ys Preoperative deviation Pattern strabismus Type of exotropia Basic CI Pseudodiv excess Presence of anisometropia POD1 alignment:ET-ortho Lateral incomitance First surgery LR recession MR resection R&R Incom R&R Adjustable suture
Excellent motor/sensory (n 5 19)
Poor motor/sensory (n 5 14)
12.7 15.3 (2-47) 28.3 16.3 (11-60) 4.6 3.6 (0.5-13) 8.1 13.0 (0-37) 14.7 5.8 (10-29) 24.2 10.9 (12-54) 1 (5%)
19.1 22.8 (2-65) 34.1 24.0 (12-75) 9.3 13.5 (0.7-40) 9.8 14.4 (0-40) 12.8 3.3 (10-18) 26.9 9.1 (12-45) 3 (21%)
14 (74%) 1 (5%) 4 (21%) 2 (11%)
9 (64%) 3 (21%) 2 (14%) 3 (21%)
12 (63%) 1 (5%)
5 (36%) 3 (21%)
12 (63%) 2 (11%) 5 (26%) 0 7 (37%)
5 (36%) 2 (14%) 6 (43%) 1 (7%) 6 (43%)
p-value motor/sensory 0.56 0.66 0.66 0.93 0.32 0.28 0.29 0.14
0.63 0.17 0.29 0.38
1.0
Excellent motor only (n 5 32) 13.1 16.9 6 (2-65) 27.9 17.1 20 (11-75) 5.2 7.1 4 (0.5-40) 7.8 12.7 1.4 (0-44) 14.3 5.3 12.5 (10-29) 24.0 10.6 23.5 (12-54) 4 (13%)
Poor motor only (n 5 9) 19.7 22.3 3 (2-55) 36.3 24.2 20 (13-72) 8.1 12.8 2.5 (0.7-40) 11.5 16.5 2.0 (0.5-40) 13.3 3.2 13 (10-18) 25.3 5.3 25 (18-35) 2 (22%)
22 (69%) 2 (6%) 8 (25%) 2 (6%)
6 (67%) 2 (22%) 1 (11%) 4 (44%)
20 (63%) 2 (6%)
3 (33%) 3 (33%)
21 (66%) 4 (13%) 7 (22%) 0 12 (38%)
3 (33%) 0 5 (56%) 1 (11%) 4 (44%)
p-value motor only (multivariate analysis) 0.94 0.62 1.0 0.51 0.85 0.36 0.60 0.28
0.02 (0.03) 0.15 (0.5) 0.06 (0.07) 0.04 (0.14)a
0.72
The first 3 columns indicate surgical results using combined motor and sensory criteria; the last 3 columns indicate surgical results using motor criteria only. Results of multivariate logistic regression are shown in parentheses in the last column. CI, convergence insufficiency; Pseudodiv excess, pseudo-divergence excess; POD1, postoperative day 1; ET, esotropia; ortho, orthotropia; LR, lateral rectus; MR, medial rectus; R&R, combined recession-resection procedure; Incom R&R, incomitant recession-resection procedure (ie, recession on lateral rectus of one eye and resection on medial rectus of the fellow eye). a This p-value (0.04) indicates that a recession-resection procedure was significantly associated with a poor motor outcome when compared with the other possible surgical methods in a univariate analysis.
significantly associated with increased number of reoperations (p 5 0.0008). A multivariate analysis revealed that only postoperative day 1 undercorrection (p 5 0.04) was a truly significant association. Lateral incomitance could not be used as a possible associated factor in the multivariate analysis since all of the patients with lateral incomitance had reoperations. Subgroup Analysis Of those patients who presented before they were 8 years of age (n 5 42), 40% had an excellent result compared with 35% of those who presented after they were 8 years of age. Reoperations were required in 62% of patients who presented before they were 8 years of age compared with 50% of those who presented after they were 8 years of age. Of the 50 patients studied, 9 had convergence insufficiency pattern exotropia (mean age of onset, 10.4 12.1 years). Of these, 1 (11%) had excellent results. Of the 50 patients studied, 34 had basic type intermittent exotropia (mean age of onset, 4.1 3.1 years); 41% (14 of 34) of these patients had excellent outcomes. Finally, 7 patients had pseudo-divergence excess pattern exotropia (mean age of
Journal of AAPOS
onset 5 2.4 1.6 years); 57% (4 of 7) of these patients had excellent outcomes. Of convergence insufficiency, basic, and pseudo-divergence excess patients, 67%, 62%, and 43% underwent reoperations, respectively. Although none of the aforementioned differences amongst groups reached statistical significance on the basis of age or exotropia type, there was a significant difference in the mean age of the convergence insufficiency patients compared with the basic and pseudo-divergence excess groups (p 5 0.004 and 0.05, respectively).
Subjective Symptoms Upon questioning, 7 patients (14%) complained of subjective visual symptoms. Three patients complained of difficulty with depth perception. The remaining 4 patients complained of intermittent diplopia (2 were esotropic and 2 were exotropic). Of the 7 patients, 4 had been classified as a poor outcome, and 3 patients had a fair outcome. Three of the patients had convergence insufficiency intermittent exotropia, and the remaining 4 patients had basic intermittent exotropia.
302
Pineles et al
Volume 14 Number 4 / August 2010
Cumulative proportion without reoperation
1.0
Table 3. Comparisons of pre- and postoperative factors for risk of the need for multiple surgery
0.8 0.6
Factors
0.4 0.2 0.0 0
5
10
15
20
25
30
Length of follow-up (years)
FIG 1. Kaplan-Meier analysis depicting the proportion of intermittent exotropia patients not reoperation over time.
Discussion This study represents the largest single-surgeon series of patients with intermittent exotropia, with a minimum of 10 years’ follow-up from the first procedure. In our cohort, patients were almost evenly distributed amongst excellent, fair, and poor outcomes when stringent sensory and motor criteria were used. When we eliminated the sensory criteria from the analysis, the number of patients with an excellent result increased, and the percentage of patients with a poor result decreased. Within this classification scheme, we were able to detect several risk factors for a poor motor outcome that approached statistical significance. Outcomes of surgical interventions for intermittent exotropia have been studied since the 1960s.14,19 Despite the immense body of literature, there is still not a consensus on fundamental topics such as the optimal timing for interventions, choice of surgical technique, success rates, or the criteria for a successful outcome. For example, success rates in the literature vary widely from 42% to 81%,1-7 and success appears to be directly related to follow-up length and stringency of success criteria. There have been other authors who have evaluated surgical success in patients followed for a minimum of 1 year,3,5,6,8,11,20-24 but only 2 groups were examined after a minimum follow-up of at least 10 years.8,9 Zibrandtsen and colleagues9 reviewed the results of 25 intermittent exotropia patients who had at least 10 years of follow-up and found that approximately 50% of the patients had a good long-term result, with a small exophoria and acceptable binocularity. These authors did not attempt to systematically categorize success, nor did they evaluate risk factors for surgical failure. In 2008 Baker8 described a cohort of 30 patients prospectively re-evaluated with at least 20 years of follow-up. However, he did not make any attempt to classify success rates or risk factors for failure. Other authors claim to contain long-term follow-up data with the use of survival analyses, but they include subjects with follow-up time of less than 1 year in their analysis. Finally, few of the heretofore described studies actually included a measure of stereopsis in their analysis.2,3,5,15-17 Given that most patients with intermittent exotropia benefit from excellent near stereoacuity before surgical
Age at first surgery (yrs) Age at examination (yrs) Age at onset (yrs) Time from onset to surgery (yrs) Time since first surgery (yrs) Time since last surgery (yrs)a Preoperative deviation Presence of pattern strabismus Type of exotropia Basic Convergence insufficiency Pseudo-divergence excess Presence of anisometropia (n 5 number with amblyopia) Postoperative day 1 alignmentc Amblyopia Lateral incomitance First surgery Lateral rectus recession Medial rectus resection Combined recession-resection Incomitant recession-resection Adjustable suture
Multiple (n 5 30)
Single (n 5 20)
11.2 15.2 16.7 18.7 4.5 (2-55) 9 (2-65) 27.9 17.0 29.6 18.9 20 (12-72) 21.5 (11-75) 4.8 7.5 6.6 8.4 2.3 (0.5-40) 5 (0.5-40) 6.4 11.0 10.0 14.2 2 (0-44) 2.8 (0-40) 14.8 4.6 13.5 5.1 13.5 (10-29) 12 (10-29) 7.3 5.0 13.1 3.9 8 (1-16) 25 (12-54) 23.8 8.6 25.1 10.5 24 (12-45) 25 (12-54) 4 (13%) 2 (10%)
p-value (multivariate analysis) 0.08 (0.9) 0.85 0.06 (0.4) 0.64 0.13 (0.2) 0.0008 (0.008) 0.76 1.0 0.39
21 (70%) 6 (20%)
13 (65%) 3 (15%)
3 (10%)
4 (20%)
6 (20%) n51
2 (10%) n50
0.45
12 (40%)
15 (75%)
3 (10%) 6 (20%)
0 0
0.006 (0.04) 0.27 0.06b 0.85
17 (57%)
13 (65%)
2 (7%)
2 (10%)
10 (33%)
5 (25%)
1 (3%)
0
11 (37%)
7 (35%)
1.0
Results of multivariate logistic regression are shown in parentheses in the last column. a Time since the last surgery was not considered a true risk factor for multiple reoperations because it simply reflects that an intervening surgery occurred after the first operation. b Lateral incomitance could not be included in the multivariate logistic regression because all of the patients with lateral incomitance had multiple reoperations. c Alignment after final surgery, indicating number of patients with esotropia orthotropia on day 1 after surgery.
interventions, any degradation of stereoacuity that can be attributed to a surgical outcome (ie, postoperative esodeviations), must be evaluated in the functional outcome assessment.5 Pratt-Johnson and colleagues5 elucidated the importance of stereopsis in success criteria when they reported 81% success in motor alignment, but only 41% of patients achieving stereopsis of 40 arcsec in a group of 100 patients with divergence excess–type intermittent
Journal of AAPOS
Volume 14 Number 4 / August 2010 exotropia. In addition, we assert that control, as measured in this study by distance stereopsis (and its discrepancy from near stereopsis), is another important factor in the measurement of success.25 Control is an important patient-oriented outcome because it indicates the amount of time that exotropia is manifest in a patient’s everyday life; it has also been assessed by the use of control scales or parental report by other authors.26 Our data confirm several previously described findings. In our cohort, for an excellent motor result, the determined success rate of 64% falls well within the limits defined by previous authors; yet, when one considers sensory outcomes, our success rate of 38% falls below most previous studies. In addition, our finding of increased risk for the need for reoperation with early undercorrection confirms the previous findings of Raab and Parks,14 who first described early overcorrection as a predictor of the best long-term results. Our study also corroborates the previous findings of lateral incomitance and anisometropia as associations with a poor motor outcome.27,28 In addition to evaluating risk factors for surgical success and failure, we also evaluated risk factors for multiple surgeries. In an outcomes study, one must consider that subjects may have differing postoperative goals and risk tolerance. For example, if 2 subjects have similarly poor outcomes from their first surgery, one may choose to pursue a second surgery and obtain an excellent final result, whereas another subject may choose not to undergo a reoperation. Therefore, in an attempt to address this potentially confounding variable, we included a risk factor analysis for the need for multiple surgeries. The 2 factors identified that trended toward a significant association with patients undergoing multiple surgeries (postoperative day undercorrection and lateral incomitance) are well known risk factors for surgical failure as the result of postoperative exotropic drift and difficulty in surgical planning for the lateral incomitance (ie, asymmetric recessions and resections), respectively. Although not statistically significant, our cohort demonstrated a lower proportion of convergence insufficiency patients with excellent outcomes (11% vs 41% of basic and 57% of pseudo-divergence exotropia, respectively). Although it is unclear what caused poorer outcomes in our patients with convergence insufficiency, theoretical explanations include a tendency to undercorrect at near because of poorly tolerated esotropia at distance, diminished convergent ability predisposing to exotropic drift, or presentation at an older age. Our findings should be understood within the limitations of the study. Importantly, although patients were reevaluated prospectively, their surgeries and preoperative assessments were reviewed in a retrospective fashion. Inherent to study design, there were changes in surgical techniques and preferences during the 28-year period. The range of ages and follow-up time is wide, and therefore demonstrates variability within our patient population. Furthermore, there is some degree of selection bias in
Journal of AAPOS
Pineles et al
303
that although attempt was made to contact all surgical patients during the time period, only 60 patients could be contacted and only 50 patients could be evaluated. However, we believe that our cohort is relatively unbiased given that many of our cases had not been evaluated by us since the first postoperative year and therefore is less subject to the selection bias inherent to studies in which the authors only evaluate patients who have voluntarily maintained follow-up over time. Despite its limitations, our study is the largest single surgeon case series with at least 10 years of postoperative follow-up from the first procedure. Our results confirm that long-term surgical success in intermittent exotropia is less satisfying when sensory outcomes are considered. We assert that this difference is crucial for surgeons to acknowledge in attempting to classify functional surgical outcomes. Clinicians should consider preoperative counseling of patients to ensure their understanding of the increased risk of a poor surgical outcome or the need for multiple surgeries, especially in the presence of lateral incomitance, anisometropia, and immediate postoperative undercorrections. Future prospective studies will likely help in understanding how to further maximize outcomes.
Literature Search The authors performed a MEDLINE search without date restrictions for the following terms: intermittent exotropia, alternating exotropia, strabismus, and strabismus surgery. Additionally, articles not initially found on MEDLINE but referenced in review papers were also included. References 1. Clarke WN, Noel LP. Surgical results in intermittent exotropia. Can J Ophthalmol 1981;16:66. 2. Ekdawi NS, Nusz KJ, Diehl NN, Mohney BG. Postoperative outcomes in children with intermittent exotropia from a populationbased cohort. JAAPOS 2009;13:4-7. 3. Hardesty HH, Boynton JR, Keenan P. Treatment of intermittent exotropia. Arch Ophthalmol 1978;96:268-74. 4. Hardesty H. Management of intermittent exotropia. Binocul Vis Strabismus Q 1990;5:145. 5. Pratt-Johnson JA, Barlow JM, Tillson G. Early surgery in intermittent exotropia. Am J Ophthalmol 1977;84:689-94. 6. Richard JM, Parks MM. Intermittent exotropia: surgical results in different age groups. Ophthalmology 1983;90:1172-7. 7. Weinstein GS, Biglan AW, Hiles DA. Postoperative residual small angle exodeviations. Ophthalmic Surg 1982;13:478. 8. Baker JD. Twenty-year follow-up of surgery for intermittent exotropia. JAAPOS 2008;12:227-32. 9. Zibrandtsen P, Rindziunski E, Gregersen E. Ten years follow-up of surgery for intermittent exotropia. Acta Ophthalmol (Copenh) 1986;64:374-8. 10. Faridi UA, Saleh TA, Weings P, Twomey JM. Factors affecting the surgical outcome of primary exotropia. Strabismus 2007;15:127-31. 11. Gezer A, Sezen F, Nasri N, Gozum N. Factors influencing the outcome of strabismus surgery in patients with exotropia. JAAPOS 2004;8:56-60. 12. Jeoung JW, Lee MJ, Hwang JM. Bilateral lateral rectus recession versus unilateral recess-resect procedure for exotropia with a dominant eye. Am J Ophthalmol 2006;141:683-8.
304
Pineles et al
13. Olitsky SE. Early and late postoperative alignment following unilateral lateral rectus recession for intermittent exotropia. J Pediatr Ophthalmol Strabismus 1998;35:146-8. 14. Raab EL, Parks MM. Recession of the lateral recti: early and late postoperative alignments. Arch Ophthalmol 1969;82:203-8. 15. Beneish R, Flanders M. The role of stereopsis and early postoperative alignment in long-term surgical results of intermittent exotropia. Can J Ophthalmol 1994;29:119-24. 16. Son JH, Huh YS, Kim MM. Surgical outcomes of intermittent exotropia as a function of strabismic angle. Korean J Ophthalmol 2006;20:230-33. 17. Wu H, Sun J, Xia X, Xu L, Xu X. Binocular status after surgery for constant and intermittent exotropia. Am J Ophthalmol 2006;142: 822. e1-822.e7. 18. Santiago AP, Ing MR, Kushner BJ, Rosenbaum AL. Intermittent exotropia. In: Rosenbaum A, Santiago AP, editors. Clinical strabismus management: principles and surgical techniques. Philadelphia (PA): Saunders; 1999:163-75. 19. Burian HM, Spivey BE. The surgical management of exodeviations. Tr Am Ophth Soc 1964;62:276-307. 20. Kushner BJ. Selective surgery for intermittent exotropia based on distance/near differences. Arch Ophthalmol 1998;116:324-8.
Volume 14 Number 4 / August 2010 21. 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. 22. Chia A, Seenyen L, Long QB. Surgical experiences with two-muscle surgery for the treatment of intermittent exotropia. JAAPOS 2006;10: 206-11. 23. Koklanis K, Georgievski Z. Recurrence of intermittent exotropia: factors associated with surgical outcomes. Strabismus 2009;17:37-40. 24. Maruo T, Kubota N, Sakaue T, Usui C. Intermittent exotropia surgery in children: long term outcome regarding changes in binocular alignment. A study of 666 cases. Binocul Vis Strabismus Q 2001;16: 265-9. 25. Stathacopoulos RA, Rosenbaum AL, Zanoni D, et al. Distance stereoacuity: assessing control in intermittent exotropia. Ophthalmology 1993;100:495-500. 26. Hatt SR, Mohney BG, Leske DA, Holmes JH. Variability of control in Intermittent Exotropia. Ophthalmology 2008;115:371-6. 27. Moore S. The prognostic value of lateral gaze measurements in intermittent exotropia. Am Orthopt J 1969;19:69. 28. Gordon YJ, Bachar E. Multiple regression analysis predictor models in exotropia surgery. Am J Ophthalmol 1980;90:687-91.
Journal of AAPOS