Risk Factors for Accommodative Esotropia among Hypermetropic Children

Eileen E. Birch,1,2 Sherry L. Fawcett,1,2 Sarah E. Morale,1 David R. Weakley, Jr,2 and Dianna Hughbanks Wheaton1
PURPOSE. Identication of risk factors for accommodative esotropia may help to determine which children with hyperopia may benet from early spectacle correction or preventive therapy. METHODS. Participants in the family history study were 95 consecutive patients, aged 18 to 60 months, with accommodative esotropia. Participants in the binocular sensory function study were a subgroup of 41 children enrolled in the family history study within 1 month of onset, while the esodeviation was still intermittent. Participants in the hypermetropia study were 345 consecutive patients, ages 12 months to 8 years, with refractive error of 2.00 D or greater and no esodeviation before age 12 months. RESULTS. In the family history study, 23% of children with accommodative esotropia had an affected rst-degree relative, and 91% had at least one affected relative. In the binocular sensory function study, random-dot stereoacuity was abnormal in 41% of children, whereas an abnormal motion VEP, Worth 4-dot, or positive 4-PD base-out prism responses were present in 4% or less of the children. In the hypermetropia study, patients with a mean spherical equivalent of 3.00 D and signicant anisometropia had a 7.8-fold increased risk for accommodative esotropia over nonanisometropic patients. CONCLUSIONS. A positive family history, subnormal random-dot stereopsis, and hypermetropic anisometropia each pose a signicant risk for the development of accommodative esotropia. Assessment of these risk factors in conjunction with refractive screening should help to identify those children who are most likely to benet from early spectacle correction or preventive treatment. (Invest Ophthalmol Vis Sci. 2005;46:526 529) DOI: 10.1167/iovs.04-0618 reatment with spectacle correction of 4.00 D of hypermetropia has been shown to reduce the prevalence of accommodative esotropia in children aged 4 years by 50% and potentially reduces the risk of hypermetropic ametropic amblyopia.1 Despite the potential benet of early intervention, this treatment approach has not been widely adopted for at least two reasons. First, the prevalence of accommodative esotropia among children who had hypermetropia of 4.00 D during infancy is low (10%20%),1,2 and so the placement of spectacles on all children with 4.00 D hypermetropia would result in the treatment of many children who are not at risk for development of accommodative esotropia. Unnecessary treatment with spectacle correction during infancy and early childhood is expensive, places a compliance burden on families, and may interfere with emmetropization, especially if refractive error is fully corrected.3 Atkinson et al.4 found a signicantly slower course of emmetropization among hyperopic infants who were undercorrected by 1 D, although this was a transient effect. By 36 months of age, emmetropization was similar in the two groups. Full correction of refractive error may have a larger impact on emmetropization, since the greater reduction in accommodative demand may minimize the visual feedback from optical defocus that is thought to play a role in compensatory eye growth.57 Second, many children who have accommodative esotropia have hypermetropic refractive errors less than 4.00 D and would not be treated with this approach. For example, more than half of the patients in a group of 68 children with accommodative esotropia reported by Raab8 had less than 4.00 hypermetropia on the initial visit. Identication of additional risk factors for accommodative esotropia could play an important role in determining which children with hypermetropia 4.00 D may benet from early preventive treatment with spectacles. The purpose of this study was to examine three potential risk factors, commonly associated with accommodative esotropia9 11 that might be used along with refractive screening to identify children at highest risk for accommodative esotropia. The risk factors examined were family history, abnormal binocular sensory function, and signicant (1.00 D) anisometropia.

METHODS
Subjects
Participants in the family history study were 95 consecutive patients with accommodative esotropia, ages 18 to 60 months, enrolled in a prospective study of accommodative esotropia. Participants in the binocular sensory function study were a subgroup of those enrolled in the family history study who had a visit within 1 month of detection of the esotropia by the parent or pediatrician and within 1 month of diagnosis of accommodative esotropia by a pediatric ophthalmologist, while the esodeviation was still intermittent (n 41). The rationale for selecting this subgroup for study was that any binocular sensory abnormalities present at this early stage (when the child still enjoyed orthophoria throughout most of the waking hours) may have been present before the onset of the intermittent deviation and therefore represented a true risk factor. All participants in the studies of family history and binocular sensory function had onset of accommodative esotropia at 18 to 48 months of age associated with hypermetropia and/or high accommodative convergence/accommodation (AC/A) ratio. They were diagnosed by a pediatric ophthalmologist and referred to the Retina Foundation of the Southwest for participation in research studies of sensory outcomes after treatment during the years 1988 to 1998. None of the patients had known neurologic defects or other coexisting disease.
Investigative Ophthalmology & Visual Science, February 2005, Vol. 46, No. 2 Copyright Association for Research in Vision and Ophthalmology

From the 1Retina Foundation of the Southwest, Dallas, Texas; and the Department of Ophthalmology, University of Texas Southwestern Medical Center, Dallas, Texas. Supported by National Eye Institute Grant EY05236 (EEB). Submitted for publication May 28, 2004; revised September 13, 2004; accepted November 7, 2004. Disclosure: E.E. Birch, None; S.L. Fawcett, None; S.E. Morale, None; D.R. Weakley, Jr, None; D.H. Wheaton, None The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked advertisement in accordance with 18 U.S.C. 1734 solely to indicate this fact. Corresponding author: Eileen E. Birch, Retina Foundation of the Southwest, 9900 N. Central Expressway, Suite 400, Dallas, TX 75231; [email protected].
2

526

IOVS, February 2005, Vol. 46, No. 2

Participants in the hypermetropia study were drawn from a separate cohort composed of 345 consecutive new patients 1 year of age presenting to the Ophthalmology Service of Childrens Medical Center, Dallas Texas for the 42-month period from January 1, 1995, to June 30, 1998.1214 Patients were eligible if they had an overall refractive error of 2.00 D or greater (mean spherical error calculated as spherical error 0.5 astigmatic error; average of both eyes), and no history of or documented esodeviation before age 12 months; 345 patients were included in this study. Patients enrolled ranged in age from 12 months to 8 years. Informed consent was obtained from one or both parents before the childs participation. The research protocol conformed to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center.

Risk Factors for Accommodative ET

TABLE 1. Family History by Directed Interview

527

% Prevalence of accommodative ET among rst-degree relatives Prevalence of accommodative ET among second-degree relatives Proband has rst-degree and/or second-degree affected relative Proband has an affected relative ET, esotropia. 22 10 77 91

Family History
Parents of children with accommodative esotropia were asked to ll out a brief questionnaire about the family history of strabismus and other ocular disorders among close-degree relatives of the affected child (siblings, parents, grandparents, aunts, uncles, and cousins). These data were used to construct a preliminary pedigree. A staff member reviewed the pedigree with the parent(s), and corrections were made as needed. Accompanying siblings were examined. On follow-up visits to the laboratory and/or by telephone, directed interviews were conducted with the parents or other relatives to provide further additions and clarications to the pedigrees. Experienced staff conducted all interviews. Directed interviews included detailed questions on the nature of the strabismic deviation, the age of onset of strabismus, any surgical or nonsurgical treatment, refractive error, age at which glasses were initially prescribed, and whether there were any other familial medical conditions. Only family members who were reported to have eyes that were misaligned toward the nose, who had had onset during the preschool age range (18 months to 48 months), and who had been treated with glasses that magnied the eyes were classied as affected by accommodative esotropia. Family members with other forms of strabismus or amblyopia or for whom insufcient detail was available regarding the character of the strabismic deviation, the age at onset of strabismus, or the type of treatment they received were not classied as affected. Although ophthalmic examinations of all family members may have provided other useful information, in some cases neither hypermetropia 4.00 D nor esotropia persisted into adulthood after a successful early-childhood course of treatment. Pedigree analysis was performed to determine the prevalence of affected relatives as a function of degree of relationship with the proband.

of anisometropia. In addition, an ocular motility examination was performed to determine the presence or absence and constancy of esotropia, and acuity or xation preference testing was completed to determine presence or absence of amblyopia. Signicant anisometropia was dened as being 1.00 D or more. This amount of anisometropia was chosen as signicant, because it has been reported to increase the prevalence of amblyopia and reduces binocularity in patients without strabismus.12 The effects of moderate (1 to 2 D) and large (2 D) anisometropia were also compared with each other and with patients without anisometropia. The relative risk (derived from the odds ratio) of developing esotropia was determined.18 Logistic regression analysis was used to estimate the odds ratio for the adverse outcome of esotropia between subjects with and without anisometropia using a statistical analysis program (SAS Institute, Cary, NC). Age, mean spherical equivalent, and amblyopia were found to be associated with esotropia and considered to be possible confounders in the analysis. Adjusted odds ratios for esotropia were computed in a logistic regression model, with esotropia (yes/no) as the dependent variable; anisometropia (yes/no), age, and mean spherical equivalent as continuous variables; and presence of amblyopia (yes/no) as the independent variable. Because esotropia was not a rare outcome, the adjusted odds ratios were corrected to relative risks.

RESULTS
Family History
Of the 95 families invited to participate, 9 were excluded from the analyses. Two parents refused to provide sufcient information, two of the probands were adopted, three probands had one adopted parent, and two probands had a family history of other heritable conditions that might be associated with strabismus (one craniofacial disorder and one neurologic disorder). In the remaining 86 families studied, all interviews were completed to the level of the rst- and second-degree relatives (including parents, siblings, grandparents, aunts, and uncles), and 44 of the pedigrees were completed to the level of the third- to fth-degree relatives (including great grandparents, cousins, and more distant blood relatives). Data from a total of 2828 blood relatives were obtained, with a mean of 54 blood relatives per family. The total number of affected individuals was 214 (106 male, 108 female). Results are summarized in Table 1. Overall, 19 of 86 probands (22%) had affected rst-degree relatives and 66 of 86 probands (77%) had affected rst- and/or second-degree relatives. In the 44 pedigrees completed to the level of third- to fth-degree relatives, 40 (91%) of 44 probands had at least one affected relative.

Binocular Sensory Function

All children wore spectacle correction that fully corrected their intermittent esodeviations during binocular sensory testing. Random-dot stereoacuity was evaluated, using the Randot test (version 2) and the Preschool Randot Stereoacuity Test (for children aged 3 years) or the Randot Stereocards15 (for children 3 years of age) (all from Stereo Optical, Inc., Chicago, IL). Fusion was assessed with the Worth 4-dot test at near (33 cm) and, when possible, based on the childs age, at distance (3 m) and by the 4-PD base-out test. Monoxation (absence of foveal fusion) was determined by examining the asymmetry of the motion VEP in response to a 1-cyc/deg sine wave grating undergoing quadrature motion at 6 Hz. Asymmetry in the motion VEP response has been shown to be strongly associated with suppression of the W4D at distance but not at near xation, decient or absent stereopsis, and a positive 4-PD base-out test (i.e., with the clinical hallmarks of monoxation syndrome).16,17 Data were summarized by calculating the proportion of children with abnormal results on each test.

Binocular Sensory Function

The mean (SD) age of the 41 children tested at the onset of intermittent accommodative esotropia was 2.4 0.6 years, and the range was 1.5 to 3.6 years. The percentage of children with normal and abnormal outcomes on each of the four tests is summarized in Table 2. Not all children were cooperative for all four tests. The number of children who successfully completed each test is provided. At the onset of intermittent eso-

Hypermetropia Study
Each patient was evaluated, using cycloplegic refraction, to determine overall spherical equivalent refraction and overall spherical equivalent

528

Birch et al.

IOVS, February 2005, Vol. 46, No. 2 22.9% prevalence of rst-degree affected relatives reported herein is similar to the 18% reported by Aurell and Norrsell.19 Although the data support family history as a signicant risk for the development of accommodative esotropia, this does not imply that all children with a positive family history necessarily are candidates for intervention. On the one hand, the 22% prevalence of accommodative esotropia among parents and siblings clearly implies signicant risk; on the other hand, approximately 75% of rst-degree relatives are likely to be unaffected. Thus, although family history poses a signicant risk, it must be evaluated in the context of other ophthalmic and sensory examination data. It should be noted that strabismus is a common condition (affecting 2% 4% of the population) and, in the absence of a control group, it is difcult to determine whether the prevalence of affected second- to fthdegree relatives is signicantly higher among patients with accommodative esotropia versus the general population. Classication of family members as affected or unaffected in the pedigree analysis was accomplished with data gathered in directed interviews. Although ophthalmic examinations of all family members may appear at rst glance to be a superior methodology, the disease of accommodative esotropia may be uniquely difcult to assess in this manner. In particular, accommodative esotropia that has been successfully treated during childhood may leave no detectable symptoms in adulthood. Neither esodeviation nor hyperopia may be present and binocular sensory function may be normal. Indeed, the interview approach that we used may have underestimated the prevalence of accommodative esotropia among relatives, since those who appeared to have other forms of strabismus or amblyopia or for whom insufcient detail was available regarding the character of the strabismic deviation, the age at onset of strabismus, or the type of treatment they received were not classied as affected. The nding of subnormal random-dot stereopsis within a group of patients with recent diagnosis of intermittent accommodative esotropia suggests that, for at least some patients with accommodative esotropia, binocular sensory abnormalities may precede, and possibly contribute to, the onset of strabismus. This result is similar to results of two earlier studies of children with or at risk for accommodative esotropia which found that abnormal binocularity (fusion or stereopsis) precedes the onset of esotropia in some cases, but is not a necessary condition for the onset of esotropia.2,20 The increased risk for development of accommodative esotropia associated with anisometropia is particularly striking for children with lower amounts of hypermetropia. This suggests that screening for anisometropia among hypermetropic children may identify a subgroup which is at higher risk for development of accommodative esotropia than the overall group of hypermetropic children. Detection of anisometropia in children with hypermetropia 4.00 D may provide an avenue for identifying those children who are at signicant risk but would be missed if only spherical equivalent criteria were applied.
TABLE 3. Anisometropia and Relative Risk of Esotropia Anisometropia None (1 D) All patients (1 D) 1 D to 2 D 2 D Relative Risk* of Esotropia (95% CI) 1.0 1.68 (1.471.80) 1.68 (1.421.82) 1.68 (1.311.85)

TABLE 2. Binocular Sensory Function at the Onset of Intermittent Accommodative Esotropia Test Random-dot stereoacuity Worth 4-dot 4 PD base-out prism Motion VEP symmetry n 39 35 28 37 % Abnormal Results 41.0 0.0 3.6 0.0

deviation, random-dot stereoacuity was abnormal in 41% of children. No abnormal results were found for motion VEP response symmetry or fusion assessed by the Worth 4-dot test. Only 4% of children showed an abnormal 4-PD base-out prism test response.

Hypermetropia
Anisometropia (1.00 D) was present in 28% (97/345) of patients, whereas esotropia was present in 61% of patients (210/345). The relative risk (RR) for development of accommodative esotropia (either intermittent or constant) when anisometropia was present compared with when it was absent is summarized in Table 3. Anisometropia (1.00 D) increased the RR for development of accommodative esotropia to 1.68 when adjusted for age, overall spherical equivalent, and amblyopia. This increased RR was the same for patients with 1.00 to 2.00 D of anisometropia as for patients with 2.00 D of anisometropia. When the data were stratied by mean hypermetropic spherical equivalent (Table 4), anisometropia was found to signicantly increase the RR of development of accommodative esotropia at lower amounts of hypermetropia. Patients with a mean SE of 3.00 D and anisometropia had a 7.8 times increased RR for accommodative esotropia compared with nonanisometropic patients with the same overall SE. This was signicantly higher than the effect of anisometropia on the RR for development of accommodative esotropia in more hypermetropic individuals (RR 1.49 for 3.00 D spherical equivalent; P 0. 017). When the data were further stratied into 1-D groups, the largest effect on RR for development of accommodative esotropia was again most notable in patients with lower hypermetropic refractive error (Table 4). Anisometropic patients were more likely to be amblyopic than nonanisometropic patients (82%, 79/97 vs. 35%, 87/248). In addition esotropic patients were also much more likely to be amblyopic than nonesotropic patients (73%, 153/210 vs. 10%, 13/135). However, because of the signicant association between amblyopia and anisometropia (82% of anisometropic patients were amblyopic) and the strong correlation between both of these variables and esotropia (82% of anisometropic patients and 93% of amblyopic patients were esotropic), there was little opportunity to examine the effect of anisometropia free of amblyopia. Nevertheless, in the 179 nonamblyopic patients, anisometropia increased the RR for esotropia to 2.14, consistent with anisometropia as an independent risk factor.

DISCUSSION
Family history, subnormal random-dot stereopsis, and, in children with hypermetropia 4.00 D, anisometropia, each resulted in a signicant increase in the risk for the development of accommodative esotropia. More than 75% of children with accommodative esotropia had an affected parent, sibling, grandparent, aunt, or uncle, and 90% of children with accommodative esotropia had at least one affected relative. The

* Adjusted for signicant baseline factors (age, mean spherical equivalent, and amblyopia).

IOVS, February 2005, Vol. 46, No. 2

Risk Factors for Accommodative ET

529

TABLE 4. Anisometropia (1 D), Mean Spherical Equivalent, and Relative Risk of Esotropia Prevalence of Esotropia when Anisometropia Present (%) (n 97) 87 (84/97) 92 (11/12) 86 (73/85) 92 (11/12) 81 (13/16) 89 (17/19) 88 (23/26) 83 (20/24) Prevalence of Esotropia when Anisometropia Absent (%) (n 248) 51 (126/248) 12 (4/34) 57 (122/214) 12 (4/34) 38 (22/58) 73 (36/49) 57 (24/42) 63 (39/65) Relative Risk* of Esotropia with Anisometropia (95% CI) 1.68 (1.311.85) 7.79 (4.468.43) 1.49 (1.321.60) 7.79 (4.468.43) 2.14 (1.382.14) 1.22 (.861.33) 1.55 (1.161.69) 1.32 (.961.49)

Mean Spherical Equivalent (SE) in Diopters (D) All patients Two Subgroups 3 D 3 D 1-D Subgroups 2 to 3 D 3 to 4 D 4 to 5 D 5 to 6 D 6 D

* Adjusted for signicant baseline factors (age, mean spherical equivalent, and amblyopia).

One caveat in the interpretation of the anisometropia data must be noted. As reported in the results section, there was a signicant association between the presence of amblyopia and both anisometropia and esotropia. For the purpose of analysis of relative risk, amblyopia was not considered a baseline risk factor but rather was assumed to be an outcome of esotropia, anisometropia, or both. Nevertheless, even in nonamblyopic patients, anisometropia increased the RR for esotropia to 2.14, consistent with anisometropia as an independent risk factor. Moreover, from a practical standpoint, it may not matter whether the anisometropia or the amblyopia poses the greater risk for the development of accommodative esotropia. Given the high rate of co-occurrence, it may be simpler to screen for anisometropia than for amblyopia, particularly in preverbal children. In conclusion, the present study identied positive family history, subnormal random-dot stereopsis before development of a constant deviation, and, in children with hypermetropia 4.00 D, anisometropia as signicant risk factors for the development of accommodative esotropia. Assessment of the presence or absence of these additional risk factors in conjunction with the overall amount of hypermetropia should help to identify those children who are most likely to benet from preventive treatment before the onset of esotropia.

6.

7.

8. 9. 10. 11. 12.

13.

14.

15.

References
1. Atkinson J. Infant vision screening: prediction and prevention of strabismus and amblyopia from refractive screening in the Cambridge Photorefraction Program. In: Simons K, ed. Early Visual Development, Normal and Abnormal. New York: Oxford University Press; 1993:335348. 2. Dobson V, Sebris S. Longitudinal study of acuity and stereopsis in infants with or at-risk for esotropia. Invest Ophthalmol Vis Sci. 1989;30:1146 1158. 3. Bankes K. Do unnecessary spectacles make the eyes worse. Arch Dis Child. 1983;58:766. 4. Atkinson J, Anker S, Bobier W, et al. Normal emmetropization in infants with spectacle correction for hyperopia. Invest Ophthalmol Vis Sci. 2000;41:3726 3731. 5. Wildsoet C. Active emmetropization: evidence for its existence

16.

17.

18.

19.

20.

and ramications for clinical practice. Ophthalmic Physiol Opt. 1997;17:279 290. Graham B, Judge S. The effects of spectacle wear in infancy on eye growth and refractive error in the marmoset. Vision Res. 1999;39: 186 206. Smith E, Hung L. The role of optical defocus in regulating refractive development in infant monkeys. Vision Res. 1999;39:1415 1435. Raab E. Hypermetropia in accommodative esodeviation. J Pediatr Ophthalmol Strabismus. 1984;21:194 197. Ziakas N, Woodruff G, Smith L, Thompson J. A study of heredity as a risk factor in strabismus. Eye. 2002;16:519 521. Tomac S. Binocularity in refractive accommodative esotropia. J Pediatr Ophthalmol Strabismus. 2002;39:226 230. Ingram R. Refraction as a basis for screening children for squint and amblyopia. Br J Ophthalmol. 1977;61:8 15. Weakley DR Jr, Birch EE. The role of anisometropia in the development of accommodative esotropia. Trans Am Ophthalmol Soc. 2000;98:7179. Weakley D Jr, Birch E, Kip K. The role of anisometropia in the development of accommodative esotropia. J AAPOS. 2001;5:153 157. Weakley D Jr. The association between anisometropia, amblyopia, and binocularity in the absence of strabismus. Trans Am Ophthalmol Soc. 1999;97:9871021. Birch E, Morale S, Jeffrey B, OConnor A, Fawcett S. Measurement of stereoacuity outcomes at ages 1 to 24 months: Randot Stereocards. J AAPOS. In press. Birch E, Fawcett S, Stager D. Co-development of VEP motion response and binocular vision in normal infants and infantile esotropes. Invest Ophthalmol Vis Sci. 2000;41:1719 1723. Fawcett S, Birch E. Motion VEPs, stereopsis, and bifoveal fusion in children with strabismus. Invest Ophthalmol Vis Sci. 2000;41: 411 416. Zhang J, Yu K. Whats the relative risk?a method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280:1690 1691. Aurell E, Norrsell K. A longitudinal study of children with a family history of strabismus: factors determining the incidence of strabismus. Br J Ophthalmol. 1990;74:589 594. Wattam-Bell J, Braddick O, Atkinson J, Day J. Measures of infant binocularity in a group at risk for strabismus. Clin Vis Sci. 1987; 1:327336.