A Randomized Trial of Bilateral Lateral Rectus Recession Versus Unilateral Lateral Rectus Recession With Medial Rectus Resection for Intermittent Exotropia

Intermittent exotropia (IXT) is the most common form of childhood onset exotropia with an incidence of 32.1 per 100,000 in children under 19 years of age. Intermittent exotropia is characterized by an exotropia that is not constant and is mainly present when viewing at distance, but may also be present at near. Normal binocular single vision (BSV) is typically present at near when the exotropia is controlled, with evidence of normal (occasionally sub-normal) stereoacuity. Although the natural history of the condition is largely unknown, many children with IXT are treated using either surgical or non-surgical interventions. The rationale for intervention in childhood IXT is that extended periods of misalignment may lead to entrenched suppression, resulting in loss of BSV. Intervention may also aim to address the social effects caused by the appearance of misaligned eyes. Many children treated for IXT are currently treated surgically.

There is poor agreement as to which type of surgery is most effective for the correction of IXT and the debate has long been related to differentiation between IXT sub-types. Based on distance-near angle disparity, IXT sub-types are classified as: 1) basic (similar magnitude of misalignment at distance and near); 2) true divergence excess (larger at distance); 3) pseudo divergence excess (initially larger at distance, but near angle increases following occlusion or with addition of plus lenses at near); 4) convergence insufficiency (larger at near). Basic and pseudo divergence excess appear to be the most common of the sub-types, and are also the types for which there is most disagreement regarding the optimum surgical approach. The two most common procedures are bilateral lateral rectus recession (BLRrec) and unilateral lateral rectus recession combined with a medial rectus resection in the same eye (R&R). Traditionally, BLRrec has been advocated where there is a larger distance angle, and R&R where there is a similar angle at distance and near. A survey of American strabismus surgeons published in 1990 found that the majority performed BLRrec for both basic and divergence excess types. Similarly, we found by polling our investigator group that the majority still perform a BLRrec for basic type IXT. Nevertheless, controversy still exists as to which of these surgical approaches is superior. Advocates of the BLRrec procedure tend to hold that surgery should be based purely on the distance angle of deviation. Proponents of R&R surgery suggest resection of the medial rectus best addresses the exodeviation at near.

The proposed advantage of the R&R procedure is that resecting the medial rectus, with a possible longer term initial overcorrection, is necessary for a stable and superior long-term outcome. Nevertheless, those who favor the BLRrec procedure suggest that the more profound and prolonged initial overcorrection occurring with R&R is not only unnecessary, but may in fact be harmful. A persistent overcorrection may be associated with the development of diplopia, amblyopia, and loss of stereoacuity. On the other hand, critics of the BLRrec procedure suggest that long-term recurrence rates are higher. Poor motor outcomes are likely to require reoperation and therefore the long-term success rates of these surgeries have public health importance in terms of cost to society.

Evaluating initial and long-term surgical outcomes in the proposed randomized clinical trial (RCT) will answer questions regarding the failure rates of these surgeries and also provide needed data on the potential harm of each procedure.

Only one prospective randomized clinical trial addresses success rates of BLRrec versus R&R for IXT. After between 12-15 months of follow up, 82% of 17 patients undergoing an R&R had a satisfactory outcome compared to 52% of 19 patients undergoing a BLRrec. Nevertheless, there are some important limitations of this previous study. The sample size was very small. The study population was a sub-group of patients with basic type IXT, excluding patients with basic IXT whose angle of deviation increased at far distance or following occlusion, thus limiting the generalizability of the results. In addition, outcomes were assessed unmasked, potentially biasing the results. One observational study of 103 patients (90% of whom had basic type IXT) found 1-year success rates of 56% for BLRrec and 60% for R&R. A retrospective study of 115 patients with basic type IXT reported success rates of 69% for BLRrec and 77% for R&R after an average of 15 months of follow up. Other studies comparing surgery types are limited not only by retrospective study design but also by inclusion of other types of exotropia, making it difficult to interpret results. In addition, many different criteria for success are used, precluding meaningful comparison of success rates between studies. This lack of evidence makes it very difficult to counsel parents of children with IXT regarding the likely success and complication rate of either procedure, limiting our ability to make informed management decisions. Establishing the respective failure rates through the proposed study will allow physicians to offer patients the type of surgery with the highest chance of long-term success, minimizing suboptimal results and repeat surgeries.