Researchers examine effect of horizontal strabismus surgery on refraction

Key Points

San Diego-Horizontal strabismus surgery may cause statistically significant changes in cylinder after strabismus surgery, and clinically significant changes may affect a large number of eyes undergoing the procedure, suggest results of a retrospective review. In addition, the surgery may affect the eye that did not receive the operation, according to Cintia F. Gomi, MD, assistant project scientist at the University of California, San Diego, and the Ratner Children's Eye Center, San Diego.

"Patients should be aware of this possibility of a change in their refraction, and the final refraction should be made after surgery, especially in young children," Dr. Gomi said.

The refractive status of the eyes changes throughout life and depends on various factors, including axial length and the refractive index of several ocular structures, such as the corneal lens, she said, adding that any change in the shape, elasticity, or viscosity of these structures could affect the refractive index and, therefore, the refraction.

Retrospective chart review

Contributing to the debate on the importance and mechanisms of the refractive change, she and her colleagues performed a study to investigate the effects of the horizontal muscle surgery on refraction in patients who underwent unilateral strabismus surgery in comparison with the fellow eye, which served as a control.

The Ophthalmology Departments of the University of California, San Diego, and the University of São Paulo, Brazil, collaboratively performed the retrospective chart review. The study enrolled patients who underwent unilateral surgery for horizontal strabismus from January 2000 to December 2005. Patients who had a refraction before and after surgery and a minimum follow-up of 2 months were eligible. Those who had any vertical deviation, surgery on any vertical or oblique muscle, or no postoperative follow-up were excluded.

All patients aged fewer than 25 years underwent refraction with cycloplegia. Data collected from the charts included date of birth, type of deviation, pre-and postoperative refraction, the type of surgery performed, and which muscles received the operation.

After a review of 204 charts, 59 patients met the inclusion criteria. The study population consisted of 27 males and 32 females with a mean age of 16.6 years and a mean follow-up of 7.14 months. The surgery was performed in 28 right eyes and 31 left eyes; 34 patients presented with esotropia and 25 with exotropia.

For esotropia, four eyes underwent a recession with single medial rectus and 30 underwent a combined surgery with medial rectus and lateral rectus recession. All 25 eyes with exotropia underwent a lateral rectus recession with a medial rectus resection. The mean recession of the medial rectus was 3.88 mm; when the combined surgery was performed, the medial rectus was recessed an average of 5 mm and the lateral rectus was resected at an average of 7.63 mm. For exotropia, the lateral rectus was recessed 6.84 and the medial rectus was resected at 6.5 mm.

Dr. Gomi and colleagues calculated the absolute change in sphere, cylinder, and spherical equivalent (SE) in the surgical and control eyes within both groups then compared the results.

"We saw change in cylinder that was significant between pre-and post-surgery on the surgical eyes only and a mean change that was significant between the surgical and control eyes," Dr. Gomi said.

The mean preoperative values in surgical eyes were +0.20 ± 3.91 D for sphere, +0.96 ± 1.21 D for cylinder, and +0.68 ± 3.77 D for SE. The mean preoperative values for the control eyes were +0.61 ± 2.22 D for sphere (p = 0.284), +0.60 ± 0.75 D for cylinder (p = 0.002), and +0.92 ± 2.12 D for SE (p = 0.790).

Postoperatively, the values for surgical and control eyes, respectively, were +0.85 ± 1.13 D versus +0.52 ± 0.50 D for sphere (p = 0.056) and +0.63 ± 0.92 D versus 0.44 ± 0.45 D for SE (p = 0.210). For cylinder measurements, however, the values for surgical versus control eyes were +0.57 ± 0.72 D and +0.28 ± 0.29 D, respectively (p = 0.001).