Symmetrical Versus Asymmetrical Surgery for Asymmetrical Inferior Oblique Overaction

1. Background (Introduction): An excessive elevation of the eyeball on adduction, both on horizontal movement and in upgaze, is often due to inferior oblique over action (IOOA). It is a common disorder of ocular motility, and is usually bilateral and asymmetrical. The asymmetry may be due to difference in time of onset or difference in degree of severity of the IO overaction in the two eyes. IOOA may be primary and of unknown etiology, or secondary to a congenital superior oblique palsy. Primary IOOA is commonly associated with congenital esotropia, with the oblique overaction usually presenting after one year of age. In addition to congenital esotropia, primary IO overaction may be associated with exotropia or may occur as an isolated IO overaction without other strabismus. IOOA can be isolated or combined with other types of deviations. Primary IOOA has been reported to develop between one to six years of age in up to two-thirds of patients with infantile esotropia, and is usually bilateral. The cause is unknown. IOOA may be seen in 70% of patients with esotropia and in 30% of patients with exotropia. Secondary IOOA is often unilateral and is caused by paresis or paralysis of the superior oblique muscle. Depending on severity, IOOA is graded as: (+1) to (+4) overaction. A (+1) overaction indicates slight over elevation in adduction, and (+4) overaction indicates severe over elevation in adduction. It has been suggested that (+1), (+2), (+3) and (+4) overactions roughly translate to 5, 10, 15 and 20 prism diopter (PD) of hypertropia on side gaze. Clinically, in addition to the excessive elevation of the eye ball on adduction, IOOA may be associated with "V" pattern of strabismus and vertical deviation in primary position. A "V" pattern is relative divergence on the up gaze and convergence on down gaze. When the eyes converge more than 15 PD from upgaze to downgaze, the "V" pattern is said to be significant. With the eyes in the lateral gaze to the opposite side, alternate cover testing shows that the higher eye refixates with a downward movement and that the lower eye does so with an upward movement. Although patients with IOOA present with an excessively elevated eye in adduction, there is little vertical deviation in the primary position, when the condition is bilateral and symmetrical. In contrast, unilateral IOOA or bilateral and asymmetrical IOOA is associated with a significant vertical deviation in the primary position. All patients with strabismus undergo full ocular and orthoptic examinations. The angle of strabismus is measured by prisms and alternate cover test for distant and near in primary position of gaze. In distance fixation, the strabismus angle is also measured in secondary positions of gaze to detect presence of pattern deviation or incomitance. Surgery to weaken the IO muscle is indicated when the muscle is overacting and is associated with a significant "V" pattern or vertical deviation. There are various surgical techniques for weakening an overacting IO muscle. Most commonly used techniques these days are IO myectomy, graded recession and anteriorization. Normalization of the IOOA, vertical ocular alignment within + 5 PD of orthotropia and collapse of pattern are considered ideal outcomes for the surgery. There are many studies comparing IO myectomy and IO recession and all have concluded that they have a similar success rate. However, no study has been conducted to evaluate the effectivity of symmetrical weakening procedures (bilateral myectomy or bilateral equal graded recession) vs asymmetrical procedures (myectomy - recession / bilateral recession of different amounts) in achieving ocular alignment in patients with asymmetrical IOOA. The effect of presence of superior oblique underaction preoperative vertical deviation on the outcome of the surgery has also not been studied to date.
2. Aim of the study : To compare the effect of symmetrical vs asymmetrical Inferior oblique (IO) weakening procedures when employed in the treatment of asymmetrical Inferior oblique over action (IOOA) in : Primary objective: Normalizing IO action Secondary objectives: Correcting vertical deviation and "V" pattern
3. Material and Methods: A randomized, interventional clinical trial. All Patients with asymmetrical IOOA, requiring surgical correction (IO weakening procedure) from July 2017 to June 2020 will be prospectively recruited. Prior to surgery all patients will undergo full comprehensive ocular examination, standard orthoptic evaluation and cycloplegic refraction. Minimum of two investigators will assess the preoperative angles of strabismus in a standard manner for each patient. Standard 9-gaze photos will be taken for each patient. Standardized surgical procedure will be followed in all patients according to the grade of IOOA, extent of the "V" pattern and the degree of vertical deviation in primary position. Only patients with a minimal period of 3 months of follow up will be selected for analysis.
4. Randomization: Patients will be randomly allocated to symmetrical or asymmetrical surgery using permuted block randomization of sizes 2, 4 or 6 (SAS 9.1.3). If symmetrical surgery is planned, then surgical plan for more severe IOOA will be considered for both eyes.
5. Sample size: The study is designed to show that symmetrical surgery is non inferior to asymmetrical surgery. Sample size was estimated considering a study power of 0.8 with an alpha error of 0.05, aiming to detect a difference of 5 PD in the angle of strabismus between the two groups and a postoperative standard deviation (SD) of 5.3. Based on this estimation, a total of 14 patients was found to be adequate in each group.
6. Statistical analysis:

The baseline quantitative variables will be expressed using mean with SD across the two groups if the distribution of the variables are symmetric. If not, then median with interquartile range will be presented across the groups. The baseline categorical variables will be presented using frequencies and percentages across the groups.

Primary outcomes: The difference in pre- and post-operative measurements (mean and SD) will be calculated for both the groups. The 95% confidence interval (CI) for the difference in change will then be calculated. If there are any clinical differences in the baseline parameters, then analysis of co-variance will be used to compare the change across the groups adjusted for the baseline values. If the lower limit of the CI is within 5 units, then symmetrical surgery will be concluded non-inferior to asymmetrical surgery. Intention to Treat and per-protocol analysis will be done for the primary outcome.

Secondary outcomes: The secondary outcomes will be compared across the groups using independent t-test or Wilcoxon rank sum test depending on the distribution of the continuous variables. Fisher's exact test will be used to compare categorical variables across the groups.