Focal Temporal Scleral Bulge with Choroidal Thinning: An Under-Recognized Tomographic Feature

Purpose: To describe the swept source tomographic characteristics of focal temporal scleral bulge with choroidal thinning as a normal variation of the ocular anatomy in healthy eyes. Methods: Cross-sectional observational study. Swept-source optical coherence tomography (SS-OCT) examinations in healthy patients performed between October and December 2013 were analyzed. The presence of a focal scleral bulge with choroidal thickness (CT) variation was evaluated. In those cases with this finding we manually measured the extension of the scleral protuberance and the choroidal shape. Results: There were 166 eyes of 106 patients analyzed and 13% (22 eyes of 16 patients) demonstrated a focal scleral bulge. The scleral bulge was not visible ophthalmoscopically in any case. By SS-OCT, the scleral bulge had a mean basal diameter of 3225 microns (range 1954-4908 microns) and mean distance temporal to the foveola of 2261 microns (range 1148-4173 microns). Compared to normal submacular scleral thickness, the temporal scleral bulge was a mean of 107 microns thicker (range 31-171 microns). The mean overlying choroidal thickness was 177 microns (range 79-326 microns) compared with mean subfoveal thickness in the involved eye of 250 microns (range 99-431 microns), mean thickness at temporal edge of the bulge of 312 microns (range 195-529 microns). The overlying retinal pigment epithelium and inner retinal contour were normal in all cases. Conclusions: On SS-OCT, 13% of healthy eyes showed temporal scleral bulge with related choroidal thinning. The clinical significance of this novel sign is under evaluation.


Introduction
The choroid is a highly vascular layer of the eye, limited anteriorly by the retinal pigment epithelium (RPE) and posteriorly by the sclera. This tissue provides oxygen and nutrition to the outer retina, the retinal pigment epithelium (RPE) and the prelaminar portion of the optic nerve [1]. The choroid has been classically visualized by fluorescein and indocyanine green angiography. Choroidal thickness (CT) has been measured by ultrasound and magnetic resonance imaging, but with marked limitations due to poor resolution of imaging techniques [2][3]. Optical coherence tomography (OCT), and particularly, enhanced depth imaging (EDI-OCT) and swept source (SS-OCT) have improved resolution of the choroidal details with capability of precise measurement up to 6 microns [4][5][6].
There are several reported factors that can influence CT including diurnal fluctuations, [7] typically greater in men, increasing age, and increasing axial length [8][9][10]. Regarding diurnal variation, the choroid was found to be most congested with increased thickness at 8 am and thinnest at 5 pm [7]. Margolis et al established that each increasing decade of life was correlated with subfoveal CT decrease by approximately 15.6 microns [9]. Flores-Moreno et al suggested that each increasing mm in axial length was correlated with subfoveal CT decrease by 25.9 microns [10]. In addition, systemic and ocular characteristics have been reported to increase CT such as water drinking [11], sildenafil citrate [12], hypercholesterolemia [13], central serous chorioretinopathy [14], and inflammatory ophthalmic diseases including Vogt Koyanagi Harada syndrome [15], and posterior scleritis [16]. Decrease of CT was found with oral intake of nicotine [17] and coffee [18]. Pupillary dilation does not modify CT [19].
Several authors have studied the topographic distribution of CT within the macular area of normal eyes using various OCT devices and demonstrated marked asymmetry within the macular area [9,[20][21][22][23]. The thickest CT corresponded with the subfoveal area, decreasing rapidly in the nasal direction. CT was evidenced thicker temporally than nasally, and thicker superiorly than inferiorly.
Using SS-OCT we have recently observed a novel OCT finding in the temporal macular region, characterized by scleral inward bulge accompanied by focal choroidal thinning [24]. Herein, we report the quantitative and qualitative analysis of this finding in healthy patients analyzed using SS-OCT.

Methods
We performed a cross-sectional non-randomized observational study to evaluate the submacular choroid and sclera of clinically normal eyes that underwent high-penetrance OCT exam (SS-OCT, DRI Atlantis, Topcon Medical, Oakland, NJ, USA) between October 2013 and December 2013 at the Unit of Macula of the University and Polytechnic Hospital La Fe, Valencia, Spain.
The patient demographic and clinical data were collected including age, gender, and axial length (AL). Two different scan protocols of the SS-OCT were used including the posterior pole 3D cube (12 × 9 mm; 512 horizontal A-scans x 256 vertical A-scans) and the high-definition 12 mm radial scan (12 scans).
The SS-OCT scans with evidence of focal temporal scleral bulge with choroidal thinning were further analyzed by a single investigator (RDM). Eyes with clinically-evident macular pathology such as vascular diseases, pathologic myopia (more than 6 diopters), or posterior uveitis were excluded from analysis. In addition, eyes without poor or incomplete visualization of the sclerochoroidal boundary were excluded to avoid misinterpretation of data.
The scleral bulge was identified by the inward bowing on SS-OCT and different parameters were manually measured as depicted in Figure 1. The various measurements assessed included the maximum scleral elevation at the apex of the scleral bulge (distance between the innermost and the outermost point of the sclero-choroidal junction); the basal dimension of the scleral bulge in the horizontal axis (distance between both the nasal and the temporal limit of the scleral bulge); the distance from the apex of the scleral bulge to the foveola (measured from the point of minimum CT); the distance of the nasal margin of the scleral bulge to the foveola; and the distance from the apex of the scleral bulge to the temporal edge of the optic nerve (measured from the point of minimum CT) were also assessed. We also analyzed the minimum CT (measured at the apex of the scleral bulge); the subfoveal CT and the maximum temporal CT (evaluated temporal to the area of choroidal thinning). The location of the nasal edge of the scleral bulge (temporal to the foveola, subfoveal or nasal to the foveola) was also evaluated. The contour of the RPE and the inner retina were also analyzed.
In addition, analysis of the qualitative characteristics of the color fundus photographs, fluorescein angiography and/or indocyanine green angiography for correlation with the SS-OCT findings was performed.

General population and demographic data
Demographic data are summarized in Table 1. A total number of 173 healthy eyes of 111 patients were evaluated by SS-OCT in this study. Seven eyes of 5 patients were excluded due to an incomplete visualization of the sclerochoroidal boundary. Therefore, 166 healthy eyes of 106 patients (43 men and 58 women) were included for the final analysis of the SS-OCT scans.   ± 98 microns [range: 99-431]. The focal choroidal thinning was found in the temporal area in all cases, although the nasal edge of the scleral bulge was found temporal to the foveola in 17 eyes (77%); in the subfoveolar region in 3 eyes (14%); and nasal to the foveola in 2 eyes (9%). The RPE contour and the inner retinal contour were normal in all cases. No apparent sign in the biomicroscopic or angiographic evaluation of the fundus were associated with the SS-OCT finding.

Discussion
We report the presence of a focal temporal scleral bulge with choroidal thinning in 13% of the studied normal population (22 eyes of 159 eyes). The main SS-OCT characteristics of this novel finding include: focal scleral protuberance with thinning of the overlying choroidal tissue; lack of changes in the RPE or the inner profile of the retina; location temporal to the foveola (Figure 3).
The topography and variations of CT have been widely reported [9,[20][21][22][23]]. Margolis and Spaide described a progressive thinning of the CT in the horizontal section from the subfoveal to the temporal and nasal regions [9]. Later, Hirata et al. found the choroid to be thicker in the temporal and superior sectors compared with the nasal and inferior regions, [25] and Tan et al recently found similar results [23]. On the other hand, Esmaeelpour et al showed that CT variations were more evident in the superior macular region than inferiorly [21].
In the present study, we describe a frequent variation on the macular CT topography with a well-defined thinning constantly evidenced within the temporal macular region when present, with a variable distance to the foveola and a variable extent of the inward scleral bulge. This temporal thinning in choroidal thickness does not evidence any influence in the subfoveal choroidal thickness. Our results show a mean subfoveal choroidal thickness of 250 ± 98 microns, in agreement with previous data [25][26]. The origin of this novel finding is unknown. In our series, scleral tissue could not be entirely visualized, thus the involvement of the sclera in the development of this finding is not conclusive. We found a striking similarity of the OCT images of this temporal choroidal thinning with sclerochoroidal calcification as described by Shields et al within the superior macular region [27].
They hypothesized that the chronic tractional forces induced by the superior oblique muscle could play a significant role in the pathogenesis of sclerochoroidal calcification. We consider the possibility that variations in the inferior oblique muscle insertion could progressively cause scleral variation with subsequent choroidal thinning. These variations has been reported by Yalçin et al, showing the presence of double, triple, or multiple muscle bellies in approximately 90% of cases [28]. In addition, patients with this variation in the choroidal distribution seem to be older than the general population (p=0.008), suggesting a progressive change with age. Nevertheless, the unremarkable appearance of the fundus examination in our cases is against the hypothesis of calcification of the inferior oblique insertion, as that would likely cause a yellow discoloration of the fundus.  In addition, the mean distance of the insertion of the inferior oblique muscle to the foveola was reported by Siam et al. to be 4500 microns ( Figure 4) [29] and the mean distance of the scleral bulge was closer to the fovea in our cases (2261 ± 724 microns) (Figure 2). On the other hand, we have found some cases with a wider extension of the scleral bulge involving the nasal macular area. These features are not explained by the insertion variability of the inferior oblique muscle. This artery enters the sclera temporal to the fovea at a mean distance of 1500 microns, and the intrascleral course has a mean length of 4000 microns [29]. The location of the TLPCA correlates with the finding we report, but the absence of vascular tracts within the scleral bulge in our images would not support this hypothesis.
In summary, a focal temporal scleral bulge with choroidal thinning was detected in 13% of imaged ophthalmoscopically normal eyes. This novel finding is characterized by a scleral contour modification with no associated involvement of the RPE or the retinal contour, but with frequent anatomic thinning of the overlying CT topography. The limitations of our study include the absence of histopathologic correlation and the lack of an imaging technique that allows better visualization of depth in this anatomic change. Further studies are warranted in order to analyze the origin of this constant and frequent modification of the CT macular topography.