Eight-Year Outcomes of Bilateral Lateral Rectus Recessions versus Unilateral Recession-Resection in Childhood Basic-Type Intermittent Exotropia

Purpose: To report 8-year outcomes from a randomized controlled trial (RCT) comparing bilateral lateral rectus muscle recession (BLRc) with unilateral recession-resection (R&R) for childhood intermittent exotropia (IXT).

Design: Eight-year follow-up of RCT cohort.

Participants: Of 197 randomized participants, 123 agreed to continue follow-up after the 3-year outcome visit (baseline age, 3-< 11 years; basic-type IXT, 15-40 prism diopters [Δ] by prism and alternate cover test [PACT]; baseline stereoacuity, ≤ 400 arcsec; no prior surgery).

Methods: After the RCT primary outcome at 3 years, annual follow-up from 4 through 8 years with treatment at investigator discretion.

Main outcome measures: Suboptimal surgical outcome by 8 years after randomization, defined as any of the following at any visit: exotropia of 10 Δ or more by simultaneous prism cover test (SPCT) at distance or near, constant esotropia (ET) of 6 Δ or more by SPCT at distance or near, loss of near stereoacuity by 0.6 log arcsec or more from baseline, or reoperation. Secondary outcomes included (1) reoperation by 8 years and (2) complete or near-complete resolution at 8 years, defined as exodeviation of less than 10 Δ by SPCT and PACT at distance and near and 10 Δ or more reduction from baseline by PACT at distance and near, ET of less than 6 Δ at distance and near, no decrease in stereoacuity by 0.6 log arcsec or more from baseline, and no reoperation or nonsurgical treatment for IXT.

Results: The Kaplan-Meier cumulative probability of suboptimal surgical outcome through 8 years was 68% (55 events among 101 at risk) for BLRc and 53% (42 events among 96 at risk) for R&R (difference, 15%; 95% confidence interval [CI], -2% to 32%; P = 0.08). Complete or near-complete resolution at 8 years occurred in 15% (7/46) for BLRc and 37% (16/43) for R&R (difference, -22%; 95% CI, -44% to -0.1%; P = 0.049). The cumulative probability of reoperation was 30% for BLRc and 11% for R&R (difference, 19%; 95% CI, 2%-36%; P = 0.049).

Conclusions: Despite no significant difference for the primary outcome, the 95% CI did not exclude a moderate benefit of R&R, which together with secondary outcomes suggests that unilateral R&R followed by usual care may yield better long-term outcomes than BLRc followed by usual care for basic-type childhood IXT using these surgical doses.

Financial disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.