A Randomized Trial Comparing Bilateral Lateral Rectus Recession versus Unilateral Recess-Resect for Basic Type Intermittent Exotropia
Sean P. Donahue MD, PhD1; Danielle L. Chandler, MSPH2; Jonathan M. Holmes, BM, BCh3; David K. Wallace, MD, MPH4; Raymond T. Kraker, MSPH2; Aaron M. Miller, MD5; Evelyn A. Paysse, MD6; David B. Petersen, MD7; Brian W. Arthur, MD8; on behalf of the Pediatric Eye Disease Investigator Group (PEDIG)
1Vanderbilt Eye Center, Nashville, TN,
2Jaeb Center for Health Research, Tampa, FL,
3Mayo Clinic, Rochester, MN,
4Duke Eye Center, Durham, NC,
5Houston Eye Associates, The Woodlands, TX,
6Texas Children’s Hospital, Houston, TX,
7Rocky Mountain Eye Care Associates, Salt Lake City, UT,
8Children’s Eye Research Center, Kingston, ON
Introduction: There is no consensus on the preferred surgical treatment for intermittent exotropia (IXT) in children.
Methods: 197 children, age 3-<11 years, with basic type IXT, largest deviation by Prism and Alternate Cover Test at any distance 15-40Δ, and near stereoacuity of at least 400 arc-seconds, were randomized to bilateral lateral rectus recession (BLRc) or unilateral recess-resect (R/R). The primary outcome measure was the proportion of subjects with “suboptimal surgical outcome,” defined as: exotropia ≥10Δ at distance or near using Simultaneous Prism and Cover Test (SPCT), constant esotropia ≥6Δ at distance or near using SPCT, or loss of ≥2 octaves (≥0.6 logarc-seconds) stereoacuity from baseline, at ANY of the masked examinations performed every 6 months between 6 months and 3 years. Reoperation was allowed, at investigator discretion, only after meeting suboptimal surgical outcome criteria; reoperation without meeting criteria was counted as a suboptimal surgical outcome for analysis.
Results: The cumulative probability of suboptimal surgical outcome occurring at ANY masked examination up to and including 3 years after surgery was 45.9% (43 of 101) in the BLRc group compared with 37.3% (33 of 96) in the R/R group (treatment group difference = 8.6%; 95% CI = -5.8% to 23.0%). Surgeons elected to reoperate by 3 years in 9 (9.8%) subjects in the BLRc group (8 met suboptimal surgical outcome criteria; 1 did not), and in 4 (4.6%) subjects in the R/R group (3 met suboptimal surgical outcome criteria; 1 did not) (treatment group difference = 5.2%; 95% CI = -2.3% to 12.7%). Among subjects who completed a full 3 years of follow up, 29.1% (25 of 86) in the BLRc group, and 16.9% (13 of 77) in the R/R group underwent reoperation or met suboptimal surgical outcome criteria at the 3 year visit (treatment group difference = 12.8%; 95% CI = -2.8% to 28.0%); this lower rate is primarily because several subjects in each group with suboptimal outcomes at earlier visits did not meet these criteria at 3 years.
Discussion: We did not find a statistically significant difference in suboptimal surgical outcomes by 3 years between children treated with BLRc compared with R/R; both treatment groups had a relatively high proportion with such outcomes although few patients underwent reoperation.
Conclusion: Given that there does not appear to be a clear advantage to either R/R or BLRc within the first 3 postoperative years, both techniques are reasonable surgical approaches.