Clinical Features of Crowded Orbital Syndrome on Magnetic Resonance Imaging

ABSTRACT

We have previously reported strabismus due to mismatch of orbital volume and globe as ‘crowded orbital syndrome’ (COS). In this study we have used magnetic resonance imaging (MRI) to investigate its clinical features. This has revealed that a globe with a similar axis occupies a larger volume in the orbit in patients with COS than in controls without strabismus. This suggests that strabismus with high myopia may easily occur in those with relatively small orbits and axial elongation. In acquired esotropia and/or vertical strabismus, a mismatch of orbital volume and globe axis should be investigated with MRI.

Introduction

Patients who visit an ophthalmologist with the main complaint of acquired double vision at a distance often have no intracranial or orbital disease. Almost all of these cases are related to high myopia, especially among patients with divergence insufficiency esotropia without any obvious eye movement disorders. We presented such cases as divergence insufficiency associated with high myopia, which occurs due to a relatively large eyeball with high myopia that causes mechanical movement restriction within the orbit.¹ On the basis of its pathology, this condition is also called crowded orbital syndrome (COS). In divergence insufficiency esotropia with high myopia, the space between the superior rectus and lateral rectus muscles widens, and the eyeball is dislocated. In extreme cases, it can also cause strabismus fixus.²

Some of these patients have vertical strabismus in addition to esotropia. Therefore, in this study, we used magnetic resonance imaging (MRI) to identify the clinical characteristics of patients with esotropia or vertical strabismus who have high myopia and aimed to clarify the relationship between the movement of the ocular muscles and eyeball in these cases.

Methods

This retrospective study was approved by the institutional review boards of the Inouye Eye Hospital (Approval No.: 201911–1) and Kitasato University Hospital (Approval No.: B17-098). Among the patients who visited the Inouye Eye Hospital or the Department of Ophthalmology, Kitasato University Hospital with double vision as the chief complaint, those with intracranial, intraorbital, and systemic neurological disorders or strabismus since childhood were excluded. Patients with adult-onset esotropia or vertical strabismus with concomitant high myopia, who underwent ocular axis measurements or refraction tests and MRI, and were confirmed to have no eye movement disorders were included in this study (n = 60). Patients who underwent the alternate prism cover test (APCT) at distance vision and near vision, which helps to categorise patients with esotropia and vertical strabismus, and had an axis of ≥ 25 mm or refractive power of ≤ −6 D, were retrospectively included and allocated to the COS group. Informed consent was obtained from all patients for the use of their medical data.

Magnetic resonance (MR) images of patients with symptoms and conditions, such as headache and optic neuropathy, were also retrospectively acquired. Patients without double vision or eye movement disorder with an axis of ≥ 25 mm or myopia of ≤ −6 D and without corneal diseases that could cause refractive errors or ophthalmoscopic nuclear cataracts were assigned to the high myopia (HM) control group (n = 26). Patients with refractive power within ± 3.5 D were assigned to the non-high myopia control (NC) group (n = 20).

The MRI performed included coronal T1-weighted spin-echo imaging with 3 mm slice thickness and axial T2-weighted spin-echo imaging with 3 mm slice thickness. ImageJ (National Institutes of Health and the Laboratory for Optical and Computational Instrumentation, University of Wisconsin, USA) was used to determine the angle formed between the centre of the superior rectus muscle and that of the lateral rectus muscle with the centre of the eyeball as the apex, 6 mm anterior to the optic nerve-globe junction, on coronal MR images. This was set as the dislocation angle (Figure 1).^1,Figure 2Figure3

Figure 1.

Coronal magnetic resonance image showing the dislocation angle. The angle formed between the centre of the superior rectus muscle (A) and centre of the lateral rectus muscle (B) with the centre of the eyeball (O) as the apex, 6 mm anterior to the optic nerve-globe junction, was set as the dislocation angle

Figure 2.

Axial magnetic resonance image of the eyeball showing globe-axis distance, and x, y, and z segments. The distance from the common tendinous ring to the eyeball was set as the globe-axis distance. The length of the line segment connecting the leading edge of the zygomatic bone on the outer orbital surface and the leading edge of the maxilla on the inner orbital surface was set as the transverse diameter of the anterior part of the orbit (x). A straight line parallel to x, touching the posterior part of the eyeball, was drawn, and the length of the straight line-segment bisected by the outer and inner surfaces of the orbit was set as y. A perpendicular line was drawn to connect x and y, denoted as z, which represented the distance between the two segments x and y

Figure 3.

T1-weighted magnetic resonance images: coronal section (a) and axial section (b). The arrow in the coronal section shows no temporal tilting of the lateral rectus muscle, in contrast to that seen in sagging eye syndrome, whereas the lateral rectus muscle is in contact with the eyeball, as seen in heavy eye syndrome. The axial elongation is mild, as shown in the axial section

The distance from the common tendinous ring to the eyeball on axial MR images was set as the globe-axis (GA) distance. The length of the line segment connecting the leading edge of the zygomatic bone on the outer orbital surface and that of the maxilla on the inner orbital surface was measured and set as the transverse diameter of the anterior part of the orbit (x). A straight line parallel to x that touched the posterior part of the eyeball was drawn, and the length of this straight line segment bisected by the outer and inner surfaces of the orbit was set as y. A perpendicular line was drawn between x and y, denoted as z, which represented the distance between the two segments x and y (Figure 2).

The COS, HM, and NC groups were included in the comparative study. Age and parameters such as dislocation angle, axial length, GA, and x, y, and z, were compared. Axial length was not measured in the NC group. Only the right eye was used for measurements. The Kruskal-Wallis test was used for the three-group (COS group, HM group, NC group) comparisons of each parameter, and those with significant differences were compared using the Mann-Whitney U test between two groups. SPSS (IBM SPSS Statistics Ver. 22.0, Chicago, USA) statistical software was used for statistical analyses. The results are presented as mean ± standard deviation. The significance level was set at p < .05.

Results

The characteristics measured are presented in Table 1. There were 23 men and 37 women in the COS group, six men and 20 women in the HM group, and 10 men and 10 women in the NC group. Age showed no significant difference among the three groups (p = .0756). The dislocation angle was smaller in the HM group than that in the COS group, and smaller in the NC group than that in the HM group (p < .001).

Table 1.

Features of orbit in the crowded orbital syndrome, high myopia without strabismus, and non-high myopia groups

	Age	Dislocation angle (°)	GA (mm)	x (mm)	y (mm)	z (mm)	Axis (mm)
COS group n = 60	56.3 ± 11.7	120.9 ± 10.1	25.0 ± 2.6*	38.2 ± 1.7	28.5 ± 24.4	15.8 ± 2.3	27.57 ± 1.74
HM group n = 26	46.2 ± 18.8	114.9 ± 6.9*	26.2 ± 2.2#	38.2 ± 1.6	29.5 ± 2.3	14.6 ± 2.1*	27.0 ± 1.74 (n = 19)
NC group n = 20	52.1 ± 16.2	105.8 ± 5.0ᶲᶴ	28.96 ± 3.84ᶲᶴ	38.1 ± 1.6	31.3 ± 2.8ᶴ	12.5 ± 1.9ᶲᶴ	No measurements
p value	p = .0756	p < .001	p < .001	p = .8252	p < .001	p < .001	p = .28

Notes. COS group: group with severe myopia and strabismus; HM group: group with severe myopia and no strabismus; NC group: group with non-severe myopia and no strabismus; GA: globe-axis (refer to the main text for the definitions of x, y, and z). Mann-Whitney U test:

* p <.017 for the COS Group compared with the HM Group;

# p = .03 for the COS Group compared with the HM Group;

ᶲ p <.017 for the HM Group compared with the NC Group;

ᶴ p <.017 for the COS Group compared with the NC Group.

Axial length was 25.06–33.47 mm (27.57 ± 1.74 mm) in the COS group. In the HM group, the axial length could be measured only in 19 cases and was 25.37–32.65 mm (27.08 ± 1.74 mm). Axial lengths showed no significant difference between the COS and HM groups (p = .28; Mann-Whitney U test).

The GA distance tended to be shorter in the COS group than in the HM group (p = .03), and shorter in the HM group than that in the NC group (p < .001). This indicates that the ratio of eyeball size in the muscle cone is larger in the COS group than in the HM group; that is, size mismatch between the orbit and eyeball may cause abnormal eye position. There was no significant difference in x among the three groups (p = .825). However, y was significantly different among the three groups (p < .001), but not between the COS and HM groups or between the HM and NC groups. There was a significant difference in y between the COS and NC groups (p < .001). The values were in the following order: COS group < HM group < NC group, and the lateral width of the orbit posterior to the eyeball tended to be significantly smaller in the COS group than in the NC group. There was a difference in z among the three groups (p < .001), with a significant difference between any two groups (p < .017). That is, the length of the eyeball within the orbit was in the order of COS group > HM group > NC group.

In 76.7% (n = 46) of patients with COS, divergence insufficiency was confirmed by the presence of the following: 1) esotropia at distance vision, indicating phoria or phoria-tropia at near vision, or 2) a difference exceeding +10 prism dioptres in eye position between distance vision and near vision. The eye position in the COS group ranged from 6 ⊿ esotropia to 65 ⊿ esotropia on APCT for distance vision and 1 ⊿ esotropia to 65 ⊿ esotropia on APCT for near vision. Results on APCT ranged from 1⊿ to 11⊿ for both distance and near vision in patients with vertical strabismus. Among the 60 patients in the CO group, 37 had esotropia alone without vertical strabismus at either distance or near, whereas 23 patients had esotropia and vertical strabismus at either distance or near.

Discussion

In patients with divergence insufficiency esotropia and high myopia, the space between the superior rectus and lateral rectus muscles widens, eventually leading to eyeball dislocation. In strabismus fixus cases, the dislocation angle is extremely large, occasionally exceeding 180°.³ Because of high myopia, esotropia may have developed before patients experienced eyeball dislocation. Our results show that even a relatively smaller dislocation angle compared to strabismus fixus can cause abnormal eye position, subsequently resulting in divergence insufficiency esotropia and vertical strabismus without obvious limitation of eye movements. The orbital size was possibly small in the COS group; the GA distance was shorter in the COS group than that in the HM group.

The proportion of the eyeball occupying the orbital volume was large in eyes with greater axial lengths, therefore such eyeballs are prone to dislocate between the superior and lateral rectus muscles, which tend to have low resistance. This dislocation is presumed to be caused by divergence insufficiency esotropia and vertical strabismus. The z value represented the intraorbital eyeball length. Analysis of z values indicated whether the volume of the eyeball inside the orbit increased with the elongation of the eye axis. We found that the z value was higher in the COS group than in the HM group, indicating that patients with divergence insufficiency or vertical strabismus had larger eyeballs in the orbits than those with high myopia without strabismus. Since there was no significant difference in GA distance and axial length between COS and HM groups, there might be some morphological factors such as the condition of the soft tissue of eyelids and the orbit, the shape and size of the globe, and the shape of the orbital aperture,⁴ which could lead to a larger eyeball inside the orbit in patients with COS without causing ocular protrusion.

The COS group in this study usually showed no temporal tilting of the lateral rectus muscle, unlike in sagging eye syndrome.^5,6 However, the lateral rectus muscle was in contact with the eyeball, as observed in patients with heavy eye syndrome (Figure 3).^7,8 It is also observed in relatively young people, who are not expected to have sagging eye syndrome. These young patients do not have deep superior sulci in the upper eyelids. In terms of MRI characteristics, the COS group might overlap with heavy eye syndrome. Many patients in the COS group who were mildly affected had shorter ocular axes and smaller dislocation angles than those with heavy eye syndrome reported previously.^2,3,7 According to Tan et al.,⁸ heavy eye syndrome presents with esotropia and limited abduction due to the superotemporal globe shift relative to the extraocular muscles. In general, heavy eye syndrome refers to strabismus fixus or a state similar to it. COS differs from heavy eye syndrome in that it does not have associated eye movement disorders.

Regarding management for COS, patients are informed that progressive myopia can cause diplopia, for which prism glasses or strabismus surgery is advised. We often perform bilateral medial rectus recession in cases of COS. For the heavy eye syndrome, the Yokoyama procedure is typically performed. However, for COS it is not yet possible to make a clear choice between the Yokoyama procedure and the medial rectus recession. Based on our previously successful experience with the Yokoyama procedure³ in cases with a mild form of heavy eye syndrome and severe COS, we performed the Yokoyama procedure on a patient with only COS as well, and the diplopia disappeared after surgery.⁹

Ji et al. reported that orbital volume was smaller in women than in men and that it differed between ethnicities.¹⁰ According to the study by Furuta, the mean orbital volume in the Japanese population is 23.6 ± 2.0 cm³ in men and 20.9 ± 1.3 cm³ in women,¹¹ which seems to be smaller than that reported for Caucasians (25.17 ± 0.06 mL).¹² In Japan, the prevalence of high myopic esotropia caused by a narrow orbit, such as that observed in heavy eye syndrome, or mild cases of strabismus fixus – mentioned here as COS – may not be as low as previously believed. In acquired esotropia and/or vertical strabismus, we should consider that there could be a mismatch of orbital volume and globe axis, which should be confirmed using MRI.

Funding Statement

This work was not funded.

Declaration of interest statement

The authors report no conflicts of interest.