Outcomes of Strabismus Surgery for Esotropia in Children With Down Syndrome Compared With Matched Controls

William W. Motley, III, MS, MD; Andrew T. Melson, BA; Michael E. Gray, MD; Shelia R. Salisbury, PhD

ABSTRACT
Purpose: Strabismus surgery dosages used in children with various neurodevelopmental disorders have been the subject of controversy. Few data have been reported regarding surgical results in individuals with Down syndrome (DS). Methods: A retrospective, casecontrol study was performed in which children with DS and previous bilateral medial rectus recession surgery were matched with similar control patients without DS. Surgical results were compared using a random coefficients model for repeated measurements for each group. Results: Sixteen patients with DS were matched with 16 control patients. Mean preoperative esotropia was 28.4 prism diopters (PD) in the DS group, and 27.9 PD in the control group. No significant difference was found in surgical dosages between the two groups (P = .2402). Median surgical dosage was 4.4 mm in the DS group and 4.5 mm in the control group. Preoperative and 4-month and 24-month postoperative mean angles of esotropia were not different between groups (P = .8050). The 4-month postoperative mean angles of esotropia for the DS and control groups were 3.15 and 2.66 PD, respectively. The 24-month mean angles of esotropia for the DS and control groups were 7.09 and 6.60 PD, respectively. Conclusion: Standard bilateral medial rectus recession surgical dosages need not be modified for individuals with DS. [J Pediatr Ophthalmol Strabismus 20XX;XX:XX-XX.]

INTRODUCTION Down syndrome (DS) occurs in 1 of 733 live births, making it the most common genetic etiology of intellectual disability.1 Strabismus is a common ophthalmic manifestation of DS.2-6 Esotropia occurs in 19% to 70% of individuals with DS, whereas exotropia and hypertropia occur in 0% to 6% and 0% to 3%, respectively.2,6

Dosing of medial rectus recession in the treatment of esotropia in children with neurodevelopmental disorders may be challenging.7-11 Some authors have advocated using a reduced surgical dosage for bilateral medial rectus recession (BMR) in children with neurodevelopmental disorders to avoid overcorrection.8,9 However, others have reported frequent undercorrection using this ap-

From Abrahamson Pediatric Eye Institute (WWM, ATM, MEG), Cincinnati Childrens Hospital Medical CenterUniversity of Cincinnati; and Biostatistics and Epidemiology (SRS), Cincinnati Childrens Hospital Medical Center, Cincinnati, Ohio. Originally submitted August 20, 2011. Accepted for publication January 4, 2012. Posted online February 14, 2012. Presented at the annual meeting of the American Association for Pediatric Ophthalmology and Strabismus, March 30-April 3, 2011, San Diego, California. Supported in part by a Challenge Grant from Research to Prevent Blindness, Inc., New York, New York, to the University of Cincinnati Department of Ophthalmology (James J. Augsburger, MD, Chairman). Dr. Motley is on the speakers bureau of Alcon Laboratories. The authors have no financial or proprietary interest in the materials presented herein. Address correspondence to William W. Motley, III, MS, MD, Cincinnati Childrens Hospital Medical Center, Division of Pediatric Ophthalmology, 3333 Burnet Avenue, MLC 4008, Cincinnati, OH 45229. E-mail: william.motley@cchmc.org doi: 10.3928/01913913-20120207-04

Journal of Pediatric Ophthalmology & Strabismus Vol. xx, No. x, 20XX

TABLE 1

Standard Surgical Dosage for Symmetrical Bilateral Medial Rectus Recession

Esotropia (PD) 15 20 25 30 35 40 50
PD = prism diopters.

Dosage (mm) 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Figure. Mean postoperative esotropia angle for the Down syndrome (DS) group (n = 16) and matched controls (n = 16) predicted by random coefficients model for repeated measures.

proach.10 Individuals with DS are sometimes included but make up a small minority of subjects in these studies of medial rectus recession surgical outcomes in children with neurodevelopmental disorders.8-11 Few authors have reported strabismus surgery results specifically among individuals with DS and previously published studies have been limited to noncontrolled interventional case series.12-14 The current casecontrol study compares the effect of BMR using standard surgical dosages in children with DS with the effect of BMR in children without neurodevelopmental disorders. PATIENTS AND METHODS Study approval was obtained from the Cincinnati Childrens Hospital Medical Center Institutional Review Board and conformed to the requirements of the United States Health Insurance Portability and Accountability Act. The study was performed with adherence to the tenets of the Declaration of Helsinki. The medical records of all patients with DS who had undergone BMR at Cincinnati Childrens Hospital Medical Center during the years 2003 to 2009 were reviewed. Demographic data, maximum angle of preoperative angle of esotropia in full cycloplegic correction, surgeon, surgical dosage for BMR, and maximum postoperative angle of esotropia in corrective lenses were extracted from the medical records. Matched control group subjects were consecutive, developmentally normal pediatric patients who had undergone BMR during the years 2003 to 2009 at Cincinnati Childrens Hospital Medical Center. Exclusion criteria for study group and control group subjects were extraocular
2

muscle or intraocular surgery prior to BMR, low vision, incomplete operative report, and no documented postoperative alignment measurement. Consecutive control group subjects were paired with study group subjects and matched by surgeon, age at surgery ( 3 years), and preoperative angle of esotropia in full cycloplegic refractive corrective lenses (10 prism diopters [PD]). The nonparametric Wilcoxon signed rank test was used to compare the actual BMR surgical dosages used in each group with standard surgical dosages listed in Table 1. All available postoperative measurements were used to construct a random coefficients model for repeated measures to assess difference between the two groups in mean esotropia angle over time. Analysis was performed using the Proc Mixed procedure with KenwardRoger correction. All tests were two-sided and P values of less than .05 were considered statistically significant. SAS software (version 9.2; SAS Institute, Inc., Cary, NC) was used. RESULTS Twenty-two pediatric patients with DS who had undergone BMR were identified. Six patients were excluded due to incomplete operative reports or no documented postoperative alignment measurement. Sixteen studycontrol matched pairs were analyzed. Surgeries were performed by six faculty pediatric ophthalmologists. Age at time of surgery and preoperative angle of esotropia were compared between the groups and revealed no bias in the matching variables. The median age was 4.1 years (interquartile range (IQR) = 3.2 to 6.8 years) in the DS group and 4.6 years (IQR = 3.5 to 5.9 years) in the control group (P = .37). The mean preoperative angles of esotropia in the DS and control groups were 28.4 and 27.9 PD, respectively. Median surgical dosage was 4.4 mm in the DS group and 4.5 mm in the control group (P = .2402).
Copyright SLACK Incorporated

TABLE 2

Preoperative and Postoperative Alignment (Esotropia in PD)

DS Group Visit Preoperative 4-month postoperative 24-month postoperative Mean 28.4 3.15 7.09 95% CI 25.131.7 -0.246.54 2.3011.88 Mean 27.9 2.66 6.60 Control Group 95% CI 24.731.2 -0.736.05 1.8411.36

PD = prism diopters; DS = Down syndrome; 95% CI = 95% confidence interval.

There was no significant difference between the standard surgical dosages listed in Table 1 and the actual surgical dosages used for either the DS group (P = .5078) or control group (P = 1.0). The median difference between the standard surgical dose and actual dose was 0.0 (IQR = -0.375 to 0) for the DS group and 0.0 (IQR = -0.125 to 0.375) for the control group. The results of the random coefficients model for repeated measures are shown in the figure. The predicted mean angle of postoperative esotropia for each group was plotted against time. The difference between the mean angle of esotropia for the DS and control groups was less than 1 PD for all time points. The DS and control group plots were parallel, demonstrating no significant difference in BMR effect by group over time (P = .8050). The mean postoperative alignment of each group did not change significantly over time. The mean preoperative and postoperative alignment for the DS and control groups are listed in Table 2. DISCUSSION Our data reveal a similar surgical effect in a group of individuals with DS compared with developmentally normal controls. Over 24 months, the difference remained clinically and statistically insignificant. The random coefficients model for repeated measures is not simply a regression of all available data points for each group, but rather a statistical test that computes slope and intercept for each subject when longitudinal data points are available at varying time intervals. The slopes and intercepts of all subjects in one group are combined to model the mean group effect over time. Surgical results of both groups can be compared at multiple postoperative time intervals. Variability of postoperative alignment for each group is evaluated by the 95% confidence intervals.

Previously published studies on BMR results in children with various neurodevelopmental disorders have suggested that standard surgical dosages for BMR frequently result in significant overcorrection.7-9,11 Pickering et al. reported that the mean surgical effect in children with neurodevelopmental disorders was 5.28 PD greater than expected by standard surgical dosage table.9 Habot-Wilner et al. later published that reducing the surgical dosage for BMR in children with neurodevelopmental disorders by an average of 0.84 mm per muscle resulted in undercorrection in 7 of 16 patients (44%).10 In contrast, small noncontrolled interventional case series limited to patients with DS undergoing BMR have suggested that favorable surgical results are common when standard surgical dosages are used.12-14 Hiles et al. reported surgical success in 9 of 10 patients (90%) with DS and overcorrection in 1 patient.14 Ruttum et al. reported success in 14 of 21 (66%) patients with DS, undercorrection in 6 patients, and overcorrection in 1 patient.13 Yahalom et al. reported success in 12 of 14 patients (86%) with DS and undercorrection in 2 patients.12 In the current study, we found that the presence of DS had no effect on BMR surgical results. Surgeons at our institution used similar surgical dosages for children with and without DS and there was no significant difference in postoperative alignment results over time. Therefore, we conclude that standard BMR surgical dosages need not be modified for individuals with DS.
1. Patterson D. Molecular genetic analysis of Down syndrome. Hum Genet. 2009;126:195-214. 2. Yurdakul NS, Ugurlu S, Maden A. Strabismus in Down syndrome. J Pediatr Ophthalmol Strabismus. 2006;43:27-30. 3. Da Cunha RP, Moreira JB. Ocular findings in Down syndrome. Am J Ophthalmol. 1996;122:236-244. 4. Jaeger EA. Ocular findings in Downs syndrome. Trans Am Ophthalmol Soc. 1980;158:808-845. 5. Shapiro MB, France TD. The ocular features of Downs syn-

REFERENCES

Journal of Pediatric Ophthalmology & Strabismus Vol. xx, No. x, 20XX

drome. Am J Ophthalmol. 1985;99:659-663. 6. Haugen OH, Hovding G. Strabismus and binocular function in children with Down syndrome: a population-based, longitudinal study. Acta Ophthalmol Scand. 2001;79:133-139. 7. Holman RE, Merritt JC. Infantile esotropia: results in the neurologic impaired and normal child at NCMH (six years). J Pediatr Ophthalmol Strabismus. 1986;23:41-44. 8. Pickering JD, Simon JW, Ratliff CD, Melsopp KB, Lininger LL. Alignment success following medial rectus recessions in normal and delayed children. J Pediatr Ophthalmol Strabismus. 1995;32:225-227. 9. Pickering JD, Simon JW, Ratliff CD, Lininger LL, Melsopp KB, Pinto GL. Exaggerated effect of bilateral medial rectus recession in developmentally delayed children. J Pediatr Ophthalmol Strabismus. 1994;31:374-377.

10. Habot-Wilner Z, Spierer A, Glovinsky J, Wygnanski-Jaffe T. Bilateral medial rectus muscle recession: results in children with developmental delay compared with normally developed children. J AAPOS. 2006;10:150-154. 11. Van Rijn LJ, Langenhorst AE, Krijnen JS, Bakels AJ, Jansen SM. Predictability of strabismus surgery in children with developmental disorders and/or psychomotor retardation. Strabismus. 2009;17:117-127. 12. Yahalom C, Mechoulam H, Cohen E, Anteby I. Strabismus surgery outcome among children and young adults with Down syndrome. J AAPOS. 2010;14:117-119. 13. Ruttum MS, Kivlin JD, Hong P. Outcome of surgery for esotropia in children with Down syndrome. Am Orthoptic J. 2004;54:98-101. 14. Hiles DA, Hoyme SH, McFarlane F. Downs syndrome and strabismus. Am Orthoptic J. 1974;24:63-68.

Copyright SLACK Incorporated