Augmented lateral rectus muscle recession for treatment of infantile exotropia

The surgical outcome is usually not satisfactory for treatment of primary infantile exotropia (PIE) with large exodeviation angels of more than 50 prism diopters (PD). Here, we evaluate the effectiveness of augmented bilateral lateral rectus muscle recession (ABLRR) in treatment of PIE with large deviation angles. A retrospective analysis was performed for 25 patients with PIE who underwent ABLRR. ABLRR was performed using a hemi-hang-back technique. Ocular alignment, ocular motility and binocular vision were evaluated pre- and postoperatively. Fourteen males and 11 females with PIE were included in this study, with a mean age of (3.92 ± 2.02) years (range, 1–7 years). The average exodeviation angle was (−66.32 ± 7.84) PD (range, −60–−80 PD) preoperatively. All patients underwent ABLRR using the hemi-hang-back technique. Three of them underwent bilateral inferior oblique muscle anterior transposition in the same operation. The mean surgical dosage was (11.08 ± 1.19) mm for each lateral rectus muscle. The mean of deviation angel was −2.28 ± 5.08 PD postoperatively (P < 0.05). All patients did not have abduction deficiency after a large amount of lateral rectus recession. Twenty-one of 25 patients (84.0%) achieved orthophoria at the primary gaze position at the final visit, and 13 patients obtained binocular vision. ABLRR is an effective and safe surgical procedure for treatment of PIE with a large deviation angle.


Introduction
Infantile exotropia refers to the onset of exodeviation within 1 year after birth. It is broadly classified as intermittent and constant strabismus based on the clinical features [1]. Infantile exotropia is associated with neurological, ocular, or craniofacial anomalies, but may also present in the healthy infants, a condition referred as primary infantile exotropia (PIE) [2][3][4].
The etiology of PIE is not clear, with a relatively lower prevalence of 1 in 30,000 infants in the general population [5]. Clinically, PIE is characterized as a large exotropia at the primary eye position and usually combined with oblique muscle overaction, A-V pattern, and disassociated vertical deviation (DVD) [6]. Patients with PIE may fixate alternatively with each eye and have very poor fusions.
Strabismus surgery is only way to correct the large exodeviation. Through surgery, patients may obtain ocular alignment, providing the patients with the opportunity to establish binocular vision and improve their cosmetic appearance. However, the surgical outcome is usually not satisfied for patients with a large angle exodeviation of more than 50 prism diopters (PD) [7,8]. Here, we use the surgical method of augmented bilateral lateral rectus muscle recession (ABLRR) to correct the patients' larger exotropia. Large amount recession of lateral rectus is very common in correcting many incomitant strabismus with large exodeviation angles such as congenital fibrosis of extraocular muscles (CFEOM) and may release limitation from the tight LR muscle and improve eye position. Similar to the tight medial rectus in the patients with infantile esotropia, lateral muscle is also tight in PIE with large deviation angles due to adaptive changes in the length of lateral rectus. Previous studies showed that large recession of the tight muscles would be more effective than the recession-resection procedure in treatment of infantile esotropia [9]. Here, we evaluate the effectiveness of ABLRR in treatment of PIE C 60 PD.

Materials and methods
Medical records were retrospectively reviewed for the PIE patients who underwent strabismus surgery from July 2015 to September 2019. For eligibility, the patients had to meet the following criteria: (1) onset of exotropia within 1 year after birth; (2) constant exotropia at distance and near; (3) C 60 PD exodeviation angles at distance and/or near measured by the prism and alternate cover test (PACT) or by the Krimsky test; (4) normal anterior segment and fundus; (5) at least a follow-up period of 18 months after surgery. Patients were excluded if there had any congenital anomalies, neurologic disorders, restrictive or paralytic strabismus, intermittent exotropia, or prior strabismus surgery.
This study was approved by the Ethics Committee of Beijing Children's Hospital and adhered to the principles of the Declaration of Helsinki. Written informed consent was obtained from all the children's parents.

Ocular examination
All patients underwent anterior segment and fundus examinations. Visual acuity was evaluated using the Teller Card for younger children and the Snellen chart for older children. Refractive status was evaluated using a cycloplegic retinoscopy with 1% tropicamide. Spherical equivalents (SE) were determined through a handheld autorefractor (Welch Allyn VS100, China) and measured in diopters (D).
Ocular movement and ocular alignment were evaluated pre-and postoperatively. The deviation angle was measured using the prism and alternate cover tests at near (33 cm) and distance (6 m) for children who cooperated very well. The Krimsky method was used as an alternative for patients who were not able to cooperate. Binocular sensory status was evaluated with the Worth 4 dot test or the Bagolini striated glasses at near and distance, and by stereoacuity assessment at near using the Randot Preschool Stereoacuity test (Stereo Optical Co., Inc., Chicago, IL).

Operation and follow-up
An augmented bilateral lateral rectus recession (ABLRR) was performed for all patients using the hemi-hang-back technique [9]. Based on the preoperative calculation, bilateral LR recession of 9 mm may correct 50 PD exotropia, and each additional 1 mm of recession may increase 5 PD correction for exotropia. A modified hemi-hang-back technique was used due to the large amount recession. With the difference from the classic method, the double-armed needles were passed through the scleral at the location of 9 mm distance to the original insertion instead of at the halfway between the original insertion and the desired new position, which guaranteed correction of 50 PD exodeviation. While the residual exodeviation beyond 50 PD would be corrected by augmented recession of LR to the desired position using suture suspension from the beginning of 9 mm, which could avoid an inexact recession due to longer suture suspension. Anterior transposition of inferior oblique was performed to correct overaction of inferior oblique and disassociated vertical deviation (DVD) during the operation. All surgeries were performed under general anesthesia by the same surgeon.
The follow-up was scheduled regularly at postoperative day 1, 1 week, 1 month, and followed by twomonth intervals until 18 months. Ocular alignment and binocular sensory status were examined at each visit. Surgical success was defined as the distance deviation angle within ? 5--8 PD at the primary gaze position. Recurrence was defined as B -10 PD, and overcorrection was defined as C ? 6 PD. Stereoacuity of B 100 s of arc was considered as good. Orthoptic treatment was performed postoperatively for patients with exophoria or residual exotropia. Re-operation was considered for patients with consecutive esotropia C ? 20 PD or recurrent exotropia B -20 PD persisting for more than half a year postoperatively.
Statistical analysis SPSS version 22.0 for Windows (SPSS, Chicago, IL) was used for statistical analysis. The paired t-test was used to compare the mean deviation angles at distance pre-and postoperatively. A P value of \ 0.05 was considered statistically significant.

Results
Twenty-five patients met the criteria for inclusion, including 14 males (56%) and 11 females (44%) ( Table 1). The mean age at the time of surgery was 3.92 ± 2.02 years (range, 1-7 years). The mean exodeviation angle was -66.32 ± 7.84 PD (range, -60 to -80 PD) preoperatively. None of the patients had binocular vision and stereoacuity preoperatively. All patients underwent the surgical procedure of ABLRR using the hemi-hang-back technique, and the mean surgical dosage was 11.08 ± 1.19 mm for each LR muscle. In addition, three of 25 patients underwent anterior transpositions of bilateral inferior oblique for treating their DVD and inferior oblique overaction simultaneously with ABLRR.
Twenty patients had achieved ocular alignment at the primary gaze position after surgery. Three patients had an exophoria of less than 8 PD, and two patients had a residual exotropia of more than 15 PD on the postoperative day 1. These five patients accepted orthoptic treatment at the beginning of the postoperative 1 month. One patient underwent surgery to correct his residual exotropia after 8 months orthoptic treatment, while the other four patients still had residual exodeviation but had their appearance cosmetically improved. Twenty-one patients had achieved successful motor alignment at the final visit (18 months postoperatively). The mean of deviation angel had been reduced significantly from -66.32 ± 7.84 PD preoperatively to -2.28 ± 5.08 PD postoperatively (P \ 0.001) ( Table 2). The surgical success rate was 84.0% (21/25) according to the previously described motor criteria. In addition, thirteen of 25 patients had acquired binocular vision, and two patients had a stereoacuity of 200 s of arc (Table 1) postoperatively. Abduction deficiency was not found in all patients even though undergoing a large amount LR recession.

Discussion
Surgical treatment for congenital or primary infantile exotropia is a challenge for many strabismus surgeons, because preoperative evaluations, including ocular alignment, ocular movement, and deviation angles (especially at distance), are not easily performed for a lot of children who are not able to cooperate [10][11][12]. In addition, the surgical method would be hard to be determined for an uncooperative child if the surgeon chose the surgical protocol based on the traditional classification of exotropia, because the choice of surgical method is traditionally based on the types of exotropia which are categorized by the difference of the deviation angles at near and distance, or by the innervated and anatomic factors behind this difference [13]. On the other hand, the advantage of adjustable suture technique is also limited because the infant is not able to cooperate very well.
Von Noorden suggested a surgical dosage of bilateral 7 mm LR recession and resection of one medial rectus for correcting an exodeviation of more than 50PD and suggested that the technique of adjustable sutures would be helpful in patients with large-angle exotropia [14]. However, his surgical dosage could be possible not enough for correction of exodeviation of more than 65 PD. Because the dosage of bilateral 7 mm LR recession would be suitable for correction of 30 PD, the residual exodeviation of 35 PD would require a larger resection on one medial rectus, which would produce abduction deficiency and lateral incomitance due to large resection on one medial rectus; otherwise, two medial rectus resections are needed. On the other hand, cooperation is a prerequisite for performing adjustable sutures; however, medical compliance is the main problem for a majority of infants.
In this study, we use the surgical technique of ABLRR for 25 patients with PIE who had a large deviation angle of more than 60 PD. The mean    [15], a 13 mm recession would be predicted to produce limitation on abduction because LR is recessed behind the equator of eyeball. However, LR has a longer arc of contact and plenty of connective fascia, connecting tightly with inferior oblique muscle, superior and inferior recti through its pulley structure, even if severed completely, its contraction function would not be lost completely.
We think that one of advantages in ABLRR is to spare at least one horizontal muscle (medical rectus) in the treatment of large exodeviation, which is beneficial for patients with DVD who would undergo operations on the vertical recti, avoiding anterior ischemia. Second, long-term effects are relatively stable because of large recession. In addition, overcorrection is seldom present, but undercorrection could occur in ABLRR based on our observations from these limited cases in this study. For children under the age of six, overcorrection is more dangerous than undercorrection, because overcorrection could result in consecutive esotropia, leading to amblyopia. On the contrary, even occurrence of recurrent exodeviation caused by undercorrection, there are opportunities to operate using the two untouched medial rectus muscles.

Conclusions
ABLRR may be used to treat PIE with large deviation angles, due to its effectiveness and safety. Further study is necessary to compare the surgical outcome of ABLRR with other surgical methods in treatment of PIE.