Outcomes of Consecutive Exotropia Surgery at a Tertiary Eye Hospital in Saudi Arabia

Introduction

Consecutive exotropia is an outward ocular deviation that often develops immediately or several years after surgical or optical correction of esotropia.^1–5 The incidence of consecutive exotropia ranges from 3% to 29%.^1–3,6–10 Medial rectus (MR) muscle abnormalities, such as slipped muscle, lost muscle, and stretched scars, have been suggested as possible causes of this condition.^1,4,7 The risk factors that predispose patients to the development of consecutive exotropia include amblyopia, A and V patterns, developmental delay, correction of congenital esotropia before 6 months of age, poor binocularity, anisometropia, high hyperopia, nystagmus, dissociated vertical deviation, asymmetric surgery, multiple previous ocular surgeries, large MR recession, postoperative adduction limitation, and accompanying neurological diseases.^{1–5,7,8,11,12}

Nonsurgical treatment of consecutive exotropia includes botulinum toxin injection, prism glasses, or full myopic correction.^6,7 Unfortunately, the effects of these interventions are limited relative to those of surgical interventions.^6,7,10,11 Several surgical techniques with varying success rates have been described for the treatment of consecutive exotropia. These include MR advancement with or without resection, lateral rectus (LR) recession, and a combination of these techniques.^1–5 Therefore, this study aimed to examine surgical outcomes in patients with consecutive exotropia, compare the success rates of different surgical procedures, and identify potential risk factors for failure.

Materials and Methods

This retrospective cohort study was conducted at a tertiary eye hospital in Dhahran, Saudi Arabia. The medical records of all patients between 2007 and 2020 who were diagnosed with and operated on for consecutive exotropia after surgical correction of esotropia were reviewed. This study was approved by the institutional review board of the hospital, and the need for informed consent was waived due to the retrospective nature of the study.

Inclusion Criteria

We included patients with cosmetically unacceptable consecutive exotropia (≥10 prism diopters [PD]) or delayed functional disturbance (diplopia), ie, that occured more than 6 months after the surgical correction of esotropia.

Exclusion Criteria

Patients with a history of intraocular surgery, neurological abnormalities, or follow-up duration <3 months were excluded.

Evaluation

A complete history of exotropia, including the time of onset, duration, and intermittency, was recorded, and details regarding the age at which esotropia was diagnosed, type of esotropia, medical and surgical interventions, and type of strabismus surgery were obtained from the patient’s past medical records.

All patients underwent thorough preoperative evaluations, including visual acuity measurement, extraocular motility assessment, identification of amblyopia, anterior and posterior segment examination, and cycloplegic refraction testing. Additionally, all patients underwent a complete orthoptic examination. The near (33 cm) and distance (6 m) angles of deviation were measured using the alternate prism cover test, with the Krimsky test being used if the patient was unable to fixate or too young to cooperate.

The type of surgery performed was based on the surgeon preference and experience, the amount of angle deviation, presence of adduction limitation, and intraoperative findings (presence of a stretched scar and forced duction test results). The first surgeon performed unilateral MR advancement if the preoperative angle of deviation was ≤25 PD regardless of whether there was adduction limitation. If the deviation was >25 PD in the absence of adduction limitation, LR recession was the preferred surgery. Adjustable absorbable sutures were used by the first surgeon for the LR in cooperative adults in LR cases alone or if combined with MR advancement. In cases of under- or over-correction, adjustments were made within the first 3 days of surgery. The other surgeon considered the presence of adduction limitation, and performed MR advancement in patients with limitation, and bilateral LR recession in patients without limitation. Generally, If the deviation was >40 PD, the surgeons targeted three or more muscles. LR recession was performed using a limbal- or fornix-based approach using absorbable sutures. MR advancement involved moving the muscle back to its original insertion using absorbable sutures, and was combined with resection if a stretched scar was identified during surgery. If only one eye was operated on, we targeted the amblyopic eye, as in cases of amblyopia.

The amount of LR recession and MR advancement, and postoperative complications, were documented and collected. The angle measurements were repeated at each postoperative visit, and the minimum follow-up duration was 3 months.

The corrected distance visual acuity was converted to the logarithm of the minimum angle of resolution for analysis. Unilateral amblyopia was defined as a distance corrected visual acuity (DCVA) of <20/30 in either eye or as an interocular difference of two or more lines. Bilateral amblyopia was defined as a DCVA <20/30 in both eyes. Amblyopia severity was classified as mild (DCVA ≥20/40), moderate (DCVA <20/40 and ≥20/80), or severe (DCVA <20/80). Adduction deficit was graded between −1/2 and −4 (−1/2, 12.5% loss; −1, 25% loss; −2, 50% loss; −3, 75% loss; and −4, 100% loss or no movement past the midline). Surgical success was defined as near and distant angles of deviation ≤10 PD at the 12-month/last postoperative follow-up visit.

Statistical Analysis

Statistical analyses were performed using IBM SPSS for Windows (version 22; IBM Corp, Armonk, NY, USA), and the figure was constructed using Microsoft Excel (2019, Microsoft Corp., USA). The normality of the data was assessed using the Shapiro–Wilk test. Categorical data were compared using the chi-square test, and a difference of ≥10% was considered clinically significant. The dose–effect relationship represents the ratio of the change in deviation in PD obtained via surgery to the total dose of surgery in mm. It was established using linear regression with no intercept in the surgery group. Statistical significance was set at p < 0.05.

Results

Participant’s Characteristics

A total of 78 patients were diagnosed with consecutive exotropia during the study period. Nineteen patients were excluded for having: a short follow-up period (4/19), concurrent neurological disease (1/19), concurrent ocular disorder, such as cataract and keratoconus (3/12), or residual consecutive XT that was operated on elsewhere (3/19) or had not been operated on yet (8/19). Of the 59 enrolled participants, 33 (55.9%) were male and 26 (44.06%) were female. The mean age at the time of esotropia correction surgery (n = 52) was 5.5 ± 4.7 years (range, 0.8–27 years). The mean postoperative time from esotropia correction surgery to the development of consecutive exotropia (n = 42) was 8.9 ± 7.2 years (range, 1.0–33 years). The mean age at which patients (n = 52) developed consecutive exotropia was 15.8 ± 10 years (range, 3.0–53 years). Information regarding age at the time of previous surgery might be inaccurate, as this information was not documented in some of the referral reports, and we depended on the patient’s or their parent’s memory. The mean follow-up period after consecutive exotropia correction surgery was 7.8 ± 3.5 months. Follow-up was started at 2 weeks postoperatively and extended up to 12 months.

Amblyopia was observed in 30/59 patients (50.8%), of which 20.3%, 20.3%, and 10.2% had mild, moderate, and severe amblyopia, respectively. The mean spherical equivalent (n = 52) was 0.52 ± 1.70 D (range, −7.25 to +4.88 D) in healthy eyes and 0.84 ± 1.65 D (range, −4.25 to +5.75 D) in amblyopic eyes.

Among patients whose past surgical records were available, eight patients had infantile esotropia, seven had undergone bimedial rectus recession (BMRR); eight had partially accommodative esotropia and had undergone BMRR, and two had non-accommodative esotropia and had undergone BMRR. Among patients in whom the type of esotropia was undocumented, nine had undergone MR recession and LR resection, and 15 had undergone BMRR. The mean recession during BMRR in 22 patients was 5.5 ± 0.8 mm (range, 4.0–7.0 mm). Figure 1 shows the types of esotropia and surgical procedures performed.

Figure 1 Esotropia correction surgery per esotropia type.

Preoperatively, adduction limitation was noted in 39 (66.1%) patients, and the severity was as follows: −1/2 limitation, 28.8%; −1 limitation, 33.9%; and −2 limitation, 3.4%. The mean near deviation (n = 59) was 33 ± 14 PD (range, 10–65 PD). Only two patients had a small angle of esotropia at near (flick ET and 4 PD), and the mean distance deviation was 32 ± 14 PD (range, 10–65 PD).

Consecutive Exotropia Correction Surgery

Surgeries for consecutive exotropia (n = 59) included MR advancement with or without resection (27.1%, 16/59), LR recession (28.8%, 17/59), and combined surgery (44.1%, 26/59). Six (10.2%) patients underwent reoperation and nine (15.3%) required suture adjustment. The mean distance from the MR insertion to the limbus (n = 37), MR advancement (n = 39), MR resection (n = 8), and LR recession (n = 43) were 11.2 ± 1.8 mm (range, 6.5–15 mm), 5.5 ± 1.6 mm (range, 1–9.5 mm), 3.7 ± 1.2 mm (range, 1–5 mm), and 7.3 ± 1.9 mm (range, 3–10 mm), respectively. One-muscle, two-muscle, and three-or-more–muscle surgeries were performed in 17 (28.81%), 37 (62.71%), and 5 (8.47%) patients, respectively. The details are shown in Table 1.

Table 1 Number of Patients Who Underwent the Different Surgical Procedures

Success Rate

Table 2A shows the success rate of consecutive exotropia correction surgeries according to the near angle of deviation. At 12 months postoperatively or the last follow-up visit, the success rates in the near angle of deviation ≤10 PD group (n = 59) were 62.5%, 58.8%, and 76.9% for MR advancement, LR recession, and combined surgery, respectively (chi-square test, p = 0.402).

Table 2 Success Rates for Correction of Consecutive Exotropia by Surgery Type

Table 2B shows the success rates of consecutive exotropia correction surgeries according to the distance angle of deviation. At 12 months postoperatively or the last follow-up visit, the success rates in the distance angle of deviation group (n = 36) were 88.9%, 69.2%, and 85.7% for MR advancement, LR recession, and combined surgery, respectively (chi-square test, p = 0.427).

Factors Affecting Surgical Success

Table 3 shows the impact of factors including age at the time of esotropia surgery, amblyopia severity, number of muscles operated on, adduction limitation, and preoperative angle of deviation on the success rate. A lower preoperative near angle of deviation significantly impacted the success rate at distance (p = 0.045). Notably, there was a clinically significant linear trend for improved success rate at distance as the number of muscles operated on decreased, but the results were not statistically significant (linear-by-linear association, p = 0.836). Additionally, the success rate at distance was higher in children aged >5 years at the time of esotropia surgery, although this was not statistically significant (linear-by-linear association, p = 0.467). It is worth noting that the severity of amblyopia was associated with a higher success rate at near (p = 0.862) and distance (p = 0.892), although the difference was not statistically significant.

Table 3 Factors Affecting Success Rates at Near and Distance

Dose–Effect Relationship

The dose–effect relationship represents the ratio of the change in deviation in PD obtained via surgery to the total dose of surgery in mm. It was established using linear regression with no intercept in the surgery group. At the last follow-up visit or 12 months postoperatively, the mean dose–effect relationship according to the distance angle of deviation was 2.5 PD/mm (95% confidence interval [CI], 1.2–3.8, R2 = 0.525, n = 16, p = 0.001), 3.0 PD/mm (95% CI, 2.3–3.7, R2 = 0.825, n = 17, p < 0.001), and 3.0 PD/mm (95% CI, 2.6–3.3, R2 = 0.916, n = 26, p < 0.001) in the MR advancement, LR recession, and combined surgery groups, respectively.

Discussion

Consecutive exotropia is frequently observed after the correction of esotropia.^1–3 Previous studies suggest that correcting consecutive exotropia is challenging, and the target muscle and expected results are debated.^7,13 Some surgeons reversed what was previously done, that is, they advanced the recessed muscle, while others chose to work on muscles that had not been operated on previously.^7,8,13 Furthermore, some authors recommended a thorough preoperative evaluation, including the near and distance angles of deviation, limitation in adduction, type and dose of previous surgical correction, and intraoperative identification of a stretched scar, for selection of the appropriate surgical option.^3,9,13

Overall Success Rate

In this study, the overall success rates were 66.7% and 80.6% at near and distance, respectively, which were comparable to those in previous studies, despite the different surgical modalities. For example, in 2017, Chang and Lin achieved an overall success rate of 62.96%,² and in 2004, Donaldson et al reported a satisfactory alignment rate of 71%.¹ Table 4 shows the success rates documented in previous studies.

Table 4 Success Rates Reported in Previous Studies

Factors Affecting Surgical Success

In this study, we examined the correlation of several factors, such as the presence of amblyopia, number of muscles operated on, and preoperative angle of deviation, with the success rate of consecutive exotropia correction surgery. Consistent with the findings of Donaldson et al and Chang and Lin, our study showed no significant difference in the final success rate between amblyopic and non-amblyopic patients. However, this could be biased by the short follow-up period of some patients (3 months). In our study, 50.8% of patients had amblyopia, with no significant difference in the success rate at near and distance (p = 0.862 and p = 0.892, respectively). In the study by Donaldson et al, 20.3% of the 59 patients with consecutive exotropia had amblyopia. Although patients with amblyopia had a significantly higher preoperative angle of deviation, this did not influence the final surgical outcomes.¹ In the study by Chang and Lin, 33.3% of the patients had amblyopia, which did not affect surgical success (p = 0.683).² Lee et al evaluated the outcomes of consecutive exotropia surgery in 37 patients at 2 years postoperatively. They found that amblyopia was significantly associated with exodrift at 1 month postoperatively (p = 0.026); however, it did not have a significant impact on the final success rate, where success was defined as esodeviation ≤5 PD to exodeviation ≤10 PD.⁶

With regard to the number of muscles operated on, we found no statistically significant impact on the final success rate in patients who underwent one (18.6%), two (45.8%), and three-or-more (35.6%) muscle surgeries (p = 0.836). This was similar to the result of Lee et al, who reported that the mean number of muscles operated on in the success and failure groups was 1.4 ± 0.5 and 1.6 ± 0.7, respectively (p = 0.461).⁶

In this study, the preoperative level of near exotropia (≤27 PD, 28–36 PD, and ≥37 PD) had a significant influence on the final success rate of distance exotropia correction (p = 0.045); however, it did not significantly affect the success rate of near exotropia correction (p = 0.464). The preoperative level of distance exotropia (≤23 PD, 24–35 PD, and ≥36 PD) did not affect the success rate (p=0.086). Lee et al reported a preoperative mean distance exotropia of 26.7 ± 9.0 PD and near exotropia of 28.5 ± 11.3 PD, both of which did not significantly impact the success rate (p = 0.130, p= 0.242 respectively).⁶ Similarly, the presence of adduction limitation was not associated with worse outcomes (p = 0.770 at near, p = 0.747 at distance). This is consistent with the findings of Lee et al, who reported adduction limitation in 37.8% of their patients (p = 0.536).⁶ In contrast, Kasi et al, who investigated the long-term surgical outcomes in 46 patients with consecutive exotropia, reported that 39 patients had adduction limitation, which was significantly associated with suboptimal outcomes (p < 0.01).³

Dose–Effect Relationship

We analyzed the dose–effect relationship using the distance angle of deviation at 12 months or at the last follow-up in the MR advancement, LR recession, and combined surgery groups.

We found that 1 mm of MR advancement corrected 2.5 PD of distance-deviation, which was similar to that reported by Leon and Demer, who found it to range from 2.9 to 4.0 PD/mm in 20 patients.⁸ Marcon and Pittino reported a larger effect: 1 mm MR advancement corrected 4.0 PD of exotropia at 6 months postoperatively.⁹ MR advancement is an effective procedure; however, our study showed a lower dose–effect relationship than previous studies. In this study, correction of exotropia through LR recession was 3.0 PD/mm, which was comparable to the results of Lee et al, who reported a correction of 2.4 ± 0.5 D/mm.⁶ In the combined surgery group, we observed a change of 3.0 PD/mm, which is consistent with the results of previous studies. Chatzistefanou et al and Lee and Kim reported the mean effect in combined MR and LR muscle surgery to be 2.9 PD/mm and 3.1 ± 1.0 PD/mm, respectively.^6,13 Table 5 summarizes the dose–effect relationship reported in previous studies.

Table 5 Dose–Effect Relationships by Distance Angle

Limitations

This study had some limitations. First, it had a retrospective design. Second, the study included patients of two surgeons; thus, the surgeon’s experience influenced the selection of the surgical option and dosing. Further studies with a prospective design, larger sample sizes, and more homogenous groups are recommended to better predict the outcomes.

Conclusion

In this study, the overall success rate of surgical correction of consecutive exotropia was 80.6% at distance; MR advancement and combined MR advancement with LR recession had better surgical outcomes than LR recession alone, although the differences were not statistically significant. The only risk factor that negatively affected surgical success at distance was a higher preoperative near angle of deviation.