Comparison of ganglion cell-inner plexiform layer thickness among patients with intermittent exotropia according to fixation preference: a retrospective observational study

Article information

J Yeungnam Med Sci. 2024;.jyms.2024.00864

Publication date (electronic) : 2024 October 25

doi : https://doi.org/10.12701/jyms.2024.00864

Yeon Ju Lim, Soo Jung Lee

Department of Ophthalmology, Inje University Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea

Corresponding author: Soo Jung Lee, MD, PhD Department of Ophthalmology, Inje University Haeundae Paik Hospital, Inje University College of Medicine, 875 Haeundae-ro, Haeundae-gu, Busan 48108, Korea Tel: +82-51-797-2310 • Fax: +82-51-797-2322 • E-mail: kris9352@paik.ac.kr

Received 2024 August 12; Revised 2024 September 18; Accepted 2024 September 24.

Abstract

Introduction

The relationship between structural differences and ocular dominance in the general population remains controversial. Some studies have found no significant differences in the thicknesses of the ganglion cell-inner plexiform layer (GCIPL) and retinal nerve fiber layer (RNFL) between the dominant and nondominant eyes [1-6]. In contrast, other studies have indicated significant differences that may be related to ocular dominance and structural asymmetry of the cortical visual areas, suggesting that anatomical differences in the retina are a factor in determining ocular dominance [7,8].

Intermittent exotropia (IXT) is characterized by occasional outward deviation of one or both eyes when fusional vergence breaks. It is classified as monocular or alternating exotropia depending on the presence or absence of fixation preference. The factors that determine the presence or absence of fixation preference in patients with IXT are not well understood, and limited research has been conducted on the role of the retinal layer structure in determining fixation preference. Na and Lee [9] reported a greater interocular difference in the subfoveal choroidal thickness in patients with monocular exotropia than in those with alternating exotropia, implying an association between fixation preference and choroidal thickness. However, no studies have investigated the differences in GCIPL thickness based on fixation preference. Therefore, the present study was designed to analyze the differences in GCIPL thickness between monocular and alternating IXT groups.

Methods

Patients aged between 4 and 9 years who consistently showed monocular or alternating IXT during three consecutive visits to Inje University Haeundae Paik Hospital from October 2018 to November 2022 were enrolled. Patients who developed exotropia before dissociation during a 30-second observation period were included. To exclude the effect of refractive error on GCIPL thickness, we only included patients with a spherical equivalent (SE) of ±1.25 diopters (D) or less and uncorrected visual acuity (UCVA) of 20/25 or better in both eyes. Patients with astigmatism of ≥1.00 D and anisometropia with a difference in SE of >1.00 D between both eyes were excluded. Patients with ophthalmic diseases (amblyopia, media opacity, and retinal diseases) other than strabismus, a history of ophthalmic surgery, neurological disorders, or low cooperation for spectral domain optical coherence tomography (SD-OCT) examination were also excluded. All patients underwent a comprehensive ophthalmic examination that included visual acuity (VA), refraction, and axial length (AL) measurements. VA was analyzed by converting the logarithm of the minimal angle of resolution (logMAR), and cycloplegic refraction was measured using autorefraction after administration of cycloplegic agents. AL was measured using an IOLMaster 500 (Carl Zeiss Meditec, Inc., Dublin, CA, USA). The presence or absence of fixation preference was determined using the cover-uncover test. In patients with monocular exotropia and fixation preference, the eye that returned to the primary position after occluder removal was considered the dominant eye. The results were verified for consistency using the sighting ocular dominance test, employing the hole-in-the-card method and a parental questionnaire. In the cover-uncover test, alternating exotropia was defined as both eyes remaining stationary when each eye was individually occluded and then uncovered.

Optical coherence tomography (OCT) was performed using SPECTRALIS OCT (Heidelberg Engineering, Heidelberg, Germany), and autosegmentation of the retinal layers was performed after a volume scan of the posterior pole. Macular GCIPL thickness was assessed by adding the thicknesses of the ganglion cell and inner plexiform layers for each of the nine sectors, based on the Early Treatment Diabetic Retinopathy Study (ETDRS) grid (Fig. 1). Images with GCIPL boundary-setting errors and Q scores <20 were excluded from the analysis.

Fig. 1.

Early Treatment Diabetic Retinopathy Study (ETDRS) grid in the right eye. Delineation of the nine macular sectors according to the ETDRS, within which we measure ganglion cell-inner plexiform layer thickness. C, central fovea sector. S1, T1, I1, and N1 are superior, temporal, inferior, and nasal sectors, respectively, of the inner circle subfield between 1 and 3 mm. S2, T2, I2, and N2 are superior, temporal, inferior, and nasal sectors, respectively, of the outer circle subfield between 3 and 6 mm.

Stereoacuity was measured using the Stereo Fly test with the subject wearing polarized glasses. The test booklet was held perpendicular to the subject’s visual axis at a distance of 40 cm. The Worth four dot test was performed to evaluate the fusional ability, and cases showing fusion at both near and far distances were categorized as having central fusion.

Statistical analyses were performed using MedCalc software (MedCalc Inc., Mariakerke, Belgium). Independent sample t-tests were used to compare the ages of patients with monocular and alternating exotropia. The chi-square tests were used to compare sex distribution and central fusion status. The Fisher exact test was used to compare the proportion of patients with stereopsis of 100 arcseconds or better. One-way analysis of variance (ANOVA) was used to compare the VA, refractive error, and AL of each eye between the monocular and alternating exotropia groups. An independent sample t-test, one-way ANOVA, and Tukey test were conducted to analyze the differences in GCIPL thickness in each sector between the two groups. Statistical significance was set at p<0.05.

Results

One hundred patients were included, 50 with monocular exotropia and 50 with alternating exotropia. Among the patients with monocular exotropia, 33 were right-eye dominant and 17 were left-eye dominant. The mean age was 6.5±1.5 years for the monocular exotropia group and 7.0±1.7 years for the alternating exotropia group. The mean best corrected VA and UCVA were 0.0 logMAR in both groups. The SE values for the dominant and nondominant eyes were 0.1±0.5 D and 0.1±0.7 D, respectively, in monocular exotropia, while the SE values for the right and left eyes were 0.1±0.7 D and 0.0±0.7 D, respectively, in alternating exotropia (p=0.935). The AL was 22.9±0.9 mm for both the dominant and nondominant eyes in monocular exotropia, and 22.9±0.8 mm for the right eye and 22.9±0.9 mm for the left eye in alternating exotropia (p=0.991). The proportion of patients with a stereoacuity of 100 arcseconds or better, which is the criterion for fine and useful central stereoacuity, did not differ between the monocular and alternating exotropia groups. In addition, the proportion of patients with central fusion did not differ significantly between the two groups (Table 1).

Table 1.

Patient demographics and refractive status

In the monocular exotropia group, GCIPL thickness in sector S1 was significantly thinner in the dominant eye than in the nondominant eye (91.2±7.4 μm vs. 93.3±5.2 μm, p=0.019), with no significant differences between the two eyes in other sectors. In the alternating exotropia group, there were no significant differences in GCIPL thickness between the eyes in any of the ETDRS sectors.

Furthermore, GCIPL thickness in the T1 sector was significantly thicker in the dominant eye of monocular exotropia than in the left eye of alternating exotropia (89.0±5.7 μm vs. 85.6±5.6 μm, p=0.021). In other sectors, there were no significant differences in thickness between the dominant, nondominant eyes in monocular exotropia, and the right or left eyes in alternating exotropia (Table 2).

Table 2.

Interocular difference in GCIPL thickness according to fixation preference

When comparing the interocular differences in GCIPL thickness between patients with monocular exotropia (absolute value of dominant eye minus nondominant eye) and those with alternating exotropia (absolute value of right eye minus left eye), the interocular difference in monocular exotropia was significantly greater than that in alternating exotropia in the center, S1, S2, T2, and I1 sectors (center, p=0.003; S1, p=0.007; S2, p<0.001; T2, p=0.015; and I1, p<0.001) (Table 3).

Table 3.

Comparison of absolute values of interocular differences in GCIPL thickness between monocular exotropia and alternating exotropia groups

Discussion

The visual cortex is involved in determining ocular dominance [2], and the neurons that form the visual cortex originate from ganglion cells in the retina. Therefore, we hypothesized that structural differences in the ganglion cell layer determine fixation preference in patients with strabismus, or vice versa. However, to the best of our knowledge, this relationship has not been investigated.

In the normal population, several studies have investigated how ocular dominance anatomically affects the retina and optic nerve using a different type of SD-OCT, the Cirrus OCT (Carl Zeiss Meditec, Inc.). These studies found no significant differences in any of the six zones, mean GCIPL thickness, or subfoveal thickness between dominant and nondominant eyes in healthy young adults. Therefore, these studies suggest that neither the inner nor the outer retinal layers play a significant role in determining ocular dominance, as the GCIPL represents the inner retinal layer and subfoveal thickness represents the outer retinal layer, which does not contain ganglion cells or nerve fibers [1-3]. In addition, when comparing right-eye dominance and left-eye dominance, there were no differences in mean GCIPL thickness between the right and left eyes [2,3]. In a study on healthy children using SPECTRALIS OCT, there were no significant differences in the mean and macular GCIPL thicknesses in each sector between the dominant and nondominant eyes [4]. These results [1-4] suggest that ocular dominance is more strongly associated with visual cortex development than with intraocular factors. However, considering that cortical development occurs through visual stimulation of the retina, intraocular factors cannot be completely excluded when determining ocular dominance [2,3].

In contrast, Choi et al. [7] reported that the inferonasal and inferior segments of the GCIPL are significantly thicker in the dominant eye than in the nondominant eye. These results support the hypothesis that ocular dominance is related to the macular GCIPL thickness profile.

In our study, 66% of the patients had right-eye dominance, which is consistent with the findings of previous reports [2-4,6,7,10]. Among the studies comparing interocular GCIPL thickness between the right and left eyes, one study using Cirrus OCT in healthy children and adolescents aged 5 to 17 years demonstrated that mean GCIPL thickness and GCIPL thicknesses in the superior and superonasal sectors were thinner in the right eye than in the left eye [11]. Similarly, another study using a different SD-OCT, the Topcon 3D OCT-2000 (Topcon Corp., Tokyo, Japan), in healthy participants aged 5 to 18 years also reported a thinner GCIPL in the superior hemisphere of the right eye than in that of the left eye [12]. These studies that reported significant differences in GCIPL thickness between the dominant and nondominant eyes, as well as between the right and left eyes, included patients with anisometropia or asymmetry in binocular astigmatism [7,11,12]. Conversely, a study of adults over the age of 18 years demonstrated that only the superonasal GCIPL was thinner in the right eye than in the left eye, while the mean and other sectoral GCIPL thicknesses did not differ significantly between the right and left eyes, indicating that laterality did not affect macular GCIPL thickness [13]. In another study, the mean GCIPL thickness and the thickness of all six sectors in adults exhibited high symmetry between the two eyes [14]. However, moderate-to-severe myopia or hyperopia may have been a potential bias in GCIPL thickness in these studies, as they included all cases of refractive error [1-4,7,11-14].

Reports using SPECTRALIS OCT in adults with refractive error within ±2 D have shown no difference in ganglion cell layer thickness between the right and left eyes [15,16].

Apart from refractive errors, variations in age, duration of exotropia, OCT modalities employed, and methods of sector division are likely to have contributed to the variation in results across studies. GCIPL is reported to be thinner in patients who are older, have greater myopia, and longer AL [13,17,18].

Our study only examined children without anisometropia who were close to emmetropia to eliminate any refractive errors that could affect GCIPL thickness. Since there were no significant differences in age, refractive error, or AL between the monocular and alternating exotropia groups, we were able to compare the GCIPL thickness without the influence of these factors. In the alternating group, there was no difference in the GCIPL thickness between the two eyes in any sector; however, in the monocular group, the dominant eye showed a thinner GCIPL than the nondominant eye only in sector S1. In most previous studies, there was no significant difference in the GCIPL thickness between the dominant and nondominant eyes in healthy participants, and in the absence of anisometropia, the GCIPL was also symmetrical between the right and left eyes [1-4,14]. However, regarding the peripapillary RNFL (pRNFL) in the general population, many reports have shown interocular asymmetry between the dominant and nondominant eyes, although some reports have indicated no difference [5,6]. Song and Kim [1] demonstrated that in adults aged between 23 and 38 years with myopia of ≤–6.00 D, the pRNFL in the superior sector was significantly thinner in the dominant eye than in the nondominant eye. In the dominant eye, the inferior pRNFL was thicker than the superior pRNFL, whereas in the nondominant eye, the inferior pRNFL was thinner. Choi et al. [8] also reported that the inferior RNFL was significantly thicker than the superior RNFL in dominant eyes. However, there was no difference in thickness between the superior and inferior RNFL in nondominant eyes, suggesting that anatomical differences in the retina play a role in determining ocular dominance. Considering the positive correlation between the GCIPL and RNFL thickness [7,13,17], it is presumed that the GCIPL thickness may follow a similar trend, which is consistent with the results of our study. This demonstrates that sector S1 is thinner in the dominant eye than in the nondominant eye in patients with monocular exotropia. Choi et al. [8] hypothesized that this is due to the dominant eye’s preference for the superior visual field, resulting in greater utilization of the ganglion cells in the inferior retina responsible for processing the superior visual field. Consequently, atrophy due to aging may occur at a slower rate in these ganglion cells.

In our study, we observed that the T1 sector of the GCIPL was thicker in the dominant eye with monocular exotropia than in the left eye with alternating exotropia. Oka et al. [19] also found that in patients with strabismus, changes in ganglion cell complex thickness occurred predominantly in the temporal retina rather than on the nasal side. This is likely because in unequal visual stimulation due to strabismus, the contralateral projection from the nasal retina remains unaffected, whereas the ipsilateral projection from the temporal retina is affected, resulting in suppression of the visual cortex. Adaptations in these retinocortical pathways result in morphological changes, such as the degeneration of ganglion cells in the temporal retina.

Differences between dominant and nondominant eyes were observed specifically in zone 1, which is the central macular zone. This may be due to the more accurate measurement of the central zone compared with the peripheral zone (zone 2), as the presence of retinal vessels passing through the peripheral macula can interfere with the assessment [10].

In certain ETDRS regions (center, S1, S2, T2, and I1), the absolute value of the interocular difference was significantly higher in patients with monocular exotropia than in those with alternating exotropia. This finding suggests that the presence or absence of fixation preference affects the retinal anatomy of patients with IXT.

Similar to the reduction in pRNFL thickness, a reduction in macular GCIPL thickness can help evaluate the extent and progression of optic nerve damage in glaucoma, optic neuropathy, and neurological conditions [20]. The GCIPL thickness is also thought to be related to the presence or absence of fixation preference. Ocular dominance is associated with the visual cortex and differences in GCIPL thickness between dominant and nondominant eyes. The visual cortex suppresses images in the nondominant eye to avoid diplopia [7]. However, in our study, there was no difference in the degree of fusion between monocular and alternating exotropia. Therefore, we could not assess how suppression affected the retinal layers in our study. Further research comparing healthy controls with patients with constant strabismus or strabismic amblyopia who exhibit suppression is required to understand these mechanisms.

The greater interocular difference in the GCIPL thickness observed in monocular exotropia than in alternating exotropia suggests that when fixation preference is present, structural variations in the retina may appear, even in children with good VA and emmetropia. It is likely that these differences may be more pronounced in adults due to the decrease in interocular symmetry of retinal layers with age [21]. Therefore, further studies involving a larger number of patients are needed to investigate long-term changes in GCIPL thickness with and without fixation preferences. Several animal studies have demonstrated that visual deprivation can lead to the degeneration of ganglion cells, changes in thickness and synapses in the inner plexiform layer, and a reduction in Müller fibers [22,23]. Therefore, studies on changes in these retinal microstructures, with and without fixation preference, would help us understand the underlying pathological mechanisms contributing to differences in GCIPL thickness only in certain sectors.

In conclusion, our findings suggest that in patients with IXT, the GCIPL was thinner in the dominant eye than in the nondominant eye, specifically in the S1 sector, in the monocular exotropia group. Additionally, the interocular asymmetry in thickness was greater in the monocular exotropia group than in the alternating exotropia group. These observations suggest that retinal anatomical differences play a role in determining fixation preference in patients with IXT or vice versa.

Article information Continued