Long-term Outcomes of Photorefractive Keratectomy for Anisometropic Amblyopia in Children

Abstract

To evaluate the long-term visual acuity (VA) and refractive error responses to excimer laser photorefractive keratectomy (PRK) for treatment of anisometropic amblyopia in children. Prospective interventional case-control study. Eleven children, 2 to 11 years old, with anisometropic amblyopia who were noncompliant with conventional therapy with glasses or contact lenses and occlusion therapy were treated with PRK. A cohort derived retrospectively of 13 compliant and 10 noncompliant children with refractive errors similar to those of the PRK group who were treated with traditional anisometropic amblyopia therapy served as control groups. Photorefractive keratectomy for the eye with the higher refractive error. (1) Refractive error reduction and stability in the treated eye, (2) cycloplegic refraction, (3) VA, (4) stereoacuity, and (5) corneal haze up to 3 years after PRK. Compliant and noncompliant children with anisometropia amblyopia were analyzed as controls for refractive error and VA. Preoperative refractive errors were -13.70 diopters (D) (+/-3.77) for the myopic group and +4.75 D (+/-0.50) for the hyperopic group. Mean postoperative refractive errors at last follow-up (mean, 31 months) were -3.55 D (+/-2.2.5) and +1.41 D (+/-1.07) for the myopic and hyperopic groups, respectively. At last follow-up, cycloplegic refractions in 4 (50%) of 8 myopes and all hyperopes (100%) were within 3 D of that of the fellow eye. Five (63%) of 8 myopic children achieved a refraction within 2 D of the target refraction. Two (67%) of 3 hyperopic patients maintained their refractions within 2 D of the target. Refractive regressions (from 1 year after surgery to last follow-up) were 0.50+/-1.41 D (myopes) and 0.60+/-0.57 D (hyperopes). Seven children (77%) were able to perform psychophysical VA testing preoperatively and postoperatively. Five (71%) of the 7 children had uncorrected VA improvement of at least 2 lines, and 4 (57%) of 7 had best spectacle-corrected VA improvement of at least 2 lines, with 1 improving 7 lines. Five (55%) of 9 children had improvement of their stereoacuity at last follow-up. Subepithelial corneal haze remained negligible. The mean final VA of the PRK group was significantly better than that of the noncompliant control group (P = 0.003). The mean final refractive error for both myopic and hyperopic groups was also significantly better that that of the control groups (P = 0.007 and P<0.0001, respectively). Photorefractive keratectomy for severe anisometropic amblyopia in children resulted in long-term stable reduction in refractive error and improvement in VA and stereopsis, with negligible persistent corneal haze.