Ophthalmic Surgery, Lasers and Imaging Retina

Case Report 

Swept-Source OCT Angiography of Multiple Evanescent White Dot Syndrome With Inflammatory Retinal Pigment Epithelial Detachment

Swarup S. Swaminathan, MD; Fang Zheng, MD; Andrew R. Miller, BS; Giovanni Gregori, PhD; Janet L. Davis, MD; Philip J. Rosenfeld, MD, PhD

Abstract

A 30-year-old woman with photopsias and decreased vision was diagnosed with multiple evanescent white dot syndrome (MEWDS) with an atypical inflammatory subfoveal retinal pigment epithelial detachment (PED) and imaged using swept-source optical coherence tomography (SS-OCT) during several visits. SS-OCT imaging revealed a focal area of attenuated choriocapillaris underneath the PED. An attempt to treat the presumed macular inflammatory lesion with corticosteroids resulted in bilateral exudation consistent with central serous chorioretinopathy.

[Ophthalmic Surg Lasers Imaging Retina. 2018;49:145–151.]

Abstract

A 30-year-old woman with photopsias and decreased vision was diagnosed with multiple evanescent white dot syndrome (MEWDS) with an atypical inflammatory subfoveal retinal pigment epithelial detachment (PED) and imaged using swept-source optical coherence tomography (SS-OCT) during several visits. SS-OCT imaging revealed a focal area of attenuated choriocapillaris underneath the PED. An attempt to treat the presumed macular inflammatory lesion with corticosteroids resulted in bilateral exudation consistent with central serous chorioretinopathy.

[Ophthalmic Surg Lasers Imaging Retina. 2018;49:145–151.]

Introduction

Multiple evanescent white dot syndrome (MEWDS) is an acute inflammatory condition that is commonly associated with young women complaining of photopsias. Initially described by Jampol et al. in 19841, MEWDS presents with transient white dots on funduscopic exam, disruption of the photoreceptor-retinal pigment epithelium (RPE) layers on ophthalmic coherence tomography (OCT), and a wreath-like distribution of hyperfluorescent spots on fluorescein angiography (FA). Ordinarily, the RPE and choriocapillaris appear normal and normal choriocapillaris architecture has been confirmed on swept-source optical coherence tomography (SS-OCT) and SS-OCT angiography (SS-OCTA).2 However, there have been reports of subfoveal choroidal thickening, RPE detachment (PED), and focal choroidal excavation in association with MEWDS,3,4 which suggests that choroidal inflammation may occur in MEWDS.

In this report, we describe the SS-OCT characteristics of a patient with MEWDS who presented with an inflammatory subfoveal PED overlying an area of choriocapillaris flow impairment. Treatment with systemic corticosteroids resulted in bilateral central serous chorioretinopathy, and the choriocapillaris flow impairment persisted with poor recovery of vision.

Case Report

A 30-year-old woman presented to the Bascom Palmer Eye Institute Emergency Room complaining of decreased vision and photopsias in her right eye for 1 week. The patient was 5 months postpartum, but otherwise healthy. She denied a viral prodrome. When visual acuity (VA) was tested, she was only able to identify the letter “E” at a distance of 6 feet. Funduscopic examination of the right eye revealed parafoveal and peripapillary white dots (Figure 1A) along with a yellow foveal lesion (Figure 1B). The left eye appeared normal. Fundus autofluorescence showed hyperautofluorescent areas corresponding to loss of the ellipsoid zone seen on widefield SS-OCT imaging (Figures 1C and 1D). FA showed a typical wreath-like hyperfluorescent pattern (Figure 1E). Indocyanine green angiography (ICGA) did not show late hypofluorescent foci, but did demonstrate areas of choroidal hyperpermeability (Figure 1F). Spectral-domain OCT (SD-OCT) imaging showed discontinuous disruption of the photoreceptor ellipsoid zone along with a focal subfoveal PED containing hyperreflective material (Figure 2). Notably, the choroidal thickness, once controlled for age and axial length, was not significantly increased with a thickness of 334 μm.5 At presentation, SS-OCTA imaging showed a focal decrease in choriocapillaris flow in the area corresponding to the non-vascular inflammatory PED (Figure 3). No choroidal neovascularization was noted. Given prior SS-OCTA findings from our group,2 the patient was diagnosed with atypical MEWDS with the unusual finding of a PED associated with impairment of the subfoveal choriocapillaris flow. The patient was observed.

Imaging on initial presentation. (A) Color fundus images showing multiple white dots surrounding the macula and optic nerve. (B) Fundus image showing foveal hypopigmentation. (C) Fundus autofluorescence showing speckled hyperautofluorescent dots corresponding to the to the hyporeflective lesions noted on swept-source optical coherence tomography en face images of the ellipsoid zone (D). (D) En face outer retinal slab of the ellipsoid zone measure 20 μm to 40 μm above the retinal pigment epithelial layer. (E) Fluorescein angiography imaging demonstrating typical wreath-like hyperfluorescence around macula and optic nerve. (F) Indocyanine green angiography (ICGA) showing multiple areas of choroidal hyperpermeability but no hypofluorescent foci. Aside from ICGA, all of these images were consistent with a presentation of multiple evanescent white dot syndrome.

Figure 1.

Imaging on initial presentation. (A) Color fundus images showing multiple white dots surrounding the macula and optic nerve. (B) Fundus image showing foveal hypopigmentation. (C) Fundus autofluorescence showing speckled hyperautofluorescent dots corresponding to the to the hyporeflective lesions noted on swept-source optical coherence tomography en face images of the ellipsoid zone (D). (D) En face outer retinal slab of the ellipsoid zone measure 20 μm to 40 μm above the retinal pigment epithelial layer. (E) Fluorescein angiography imaging demonstrating typical wreath-like hyperfluorescence around macula and optic nerve. (F) Indocyanine green angiography (ICGA) showing multiple areas of choroidal hyperpermeability but no hypofluorescent foci. Aside from ICGA, all of these images were consistent with a presentation of multiple evanescent white dot syndrome.

Swept-source optical coherence tomography demonstrating outer retinal ellipsoid zone disruption (arrows) in addition to the subfoveal pigment epithelial detachment (arrowhead). Subfoveal choroidal thickness (334 μm) is not significantly thickened when compared to age-matched controls.

Figure 2.

Swept-source optical coherence tomography demonstrating outer retinal ellipsoid zone disruption (arrows) in addition to the subfoveal pigment epithelial detachment (arrowhead). Subfoveal choroidal thickness (334 μm) is not significantly thickened when compared to age-matched controls.

Swept-source optical coherence tomography angiography on initial presentation. Retinal flow images were normal (not shown.) (A) Cross-sectional B-scan with boundaries corresponding to the en face flow image (B) and structural image (C). (B) En face flow image corresponding to the choriocapillaris layer. (C) En face structural image corresponding to the choriocapillaris layer. (D) Cross-sectional B-scan with boundaries corresponding to the en face flow image (E) and structural image (F). (E) En face flow image corresponding to a deeper choroidal layer. (F) En face structural image corresponding to a deeper choroidal layer. A significant decrease in the inner choroidal flow in the area of the pigment epithelial detachment can be appreciated. The decrease flow signals with normal structural signals confirm a decrease in flow. No neovascularization was noted.

Figure 3.

Swept-source optical coherence tomography angiography on initial presentation. Retinal flow images were normal (not shown.) (A) Cross-sectional B-scan with boundaries corresponding to the en face flow image (B) and structural image (C). (B) En face flow image corresponding to the choriocapillaris layer. (C) En face structural image corresponding to the choriocapillaris layer. (D) Cross-sectional B-scan with boundaries corresponding to the en face flow image (E) and structural image (F). (E) En face flow image corresponding to a deeper choroidal layer. (F) En face structural image corresponding to a deeper choroidal layer. A significant decrease in the inner choroidal flow in the area of the pigment epithelial detachment can be appreciated. The decrease flow signals with normal structural signals confirm a decrease in flow. No neovascularization was noted.

The patient returned 2 weeks later with no change in VA, and the fundus appearance had improved with only the central area of hypopigmentation persisting (Figure 4A). The outer retinal disruption improved, but the configuration of the lesion had changed (Figure 4B). On subsequent imaging 1 week later, the lesion was slightly larger and less steep, with a mild improvement in the degree of disruption of the ellipsoid zone (Figures 4C and 4D). SS-OCTA imaging showed a persistent decrease in the inner choroidal flow beneath the hyperreflective PED (Figures 4E–4G). VA remained the same. Given the persistent PED, which was thought to be a remnant of the inflammatory process and the possibility that the MEWDS might have been induced by an underlying herpetic etiology, the patient was started on prednisone 1 mg/kg daily and acyclovir 800 mg five times daily.

Imaging from 2 and 3 weeks after presentation. (A) Color fundus images from 2 weeks after presentation showing a change in the configuration of the pigment epithelial detachment (PED). (B) Spectral-domain optical coherence tomography (SD-OCT) imaging showing some recovery of the outer retinal ellipsoid zone with a change in the configuration of the lesion. (C) Three weeks after initial presentation, SD-OCT imaging showing recovery of outer retina with a persistent deposit. (D) En face outer retinal slab of the ellipsoid zone measure 20 μm to 40 μm above the retinal pigment epithelial layer showing improvement of the ellipsoid zone when compared with Figure 1D. (E) Cross-sectional B-scan from the swept-source OCT angiography scans with the boundary segmentation for the slab encompassing the choriocapillaris shown in (F) and (G). (F) En face flow image corresponding to the choriocapillaris layer demonstrating persistent loss of choriocapillaris flow beneath the deposit. (G) En face structural image corresponding to the choriocapillaris layer showing an intact signal.

Figure 4.

Imaging from 2 and 3 weeks after presentation. (A) Color fundus images from 2 weeks after presentation showing a change in the configuration of the pigment epithelial detachment (PED). (B) Spectral-domain optical coherence tomography (SD-OCT) imaging showing some recovery of the outer retinal ellipsoid zone with a change in the configuration of the lesion. (C) Three weeks after initial presentation, SD-OCT imaging showing recovery of outer retina with a persistent deposit. (D) En face outer retinal slab of the ellipsoid zone measure 20 μm to 40 μm above the retinal pigment epithelial layer showing improvement of the ellipsoid zone when compared with Figure 1D. (E) Cross-sectional B-scan from the swept-source OCT angiography scans with the boundary segmentation for the slab encompassing the choriocapillaris shown in (F) and (G). (F) En face flow image corresponding to the choriocapillaris layer demonstrating persistent loss of choriocapillaris flow beneath the deposit. (G) En face structural image corresponding to the choriocapillaris layer showing an intact signal.

The patient was monitored on 60 mg prednisone with a slow taper during the next 6 weeks with no change in VA or exam. She subsequently returned 2 weeks later with a VA decrease from 20/200 to 20/400. Imaging showed collections of subretinal fluid in both eyes (Figures 5A–5D). When compared to the ICGA images from the initial presentation, the initial areas of focal hyperpermeability appeared to correspond to areas of subretinal fluid (Figures 5E and 5F). The patient was then diagnosed with central serous chorioretinopathy (CSCR) with an exacerbation from the steroid. Prednisone was tapered rapidly with subsequent resolution of the subretinal fluid.

Imaging while on steroid therapy. (A, B) After the use of oral prednisone, subretinal fluid collections can be appreciated on color fundus imaging of the right and left eyes, respectively. Lines indicate the location of the subsequent optical coherence tomography (OCT) B-scan images. (C, D) Spectral-domain OCT (SD-OCT) imaging showing a persistent subfoveal inflammatory pigment epithelial detachment with new subretinal fluid in the right and left eyes adjacent to the optic nerve. (E, F) Middle phase indocyanine green angiography (ICGA) images during the same visit confirming focal areas of hyperfluorescence suggesting choroidal hyperpermeability. These areas are similar to those seen in Figure 1F, which was the ICGA from the initial visit. Of note, there is a significant loss of choroidal vascular flow at the fovea in the right eye. Lines indicate the location of the subsequent OCT B-scan images. (G) SD-OCT image of the right eye showing a scan of the subfoveal lesion. (H) SD-OCT image of the left eye showing a normal fovea.

Figure 5.

Imaging while on steroid therapy. (A, B) After the use of oral prednisone, subretinal fluid collections can be appreciated on color fundus imaging of the right and left eyes, respectively. Lines indicate the location of the subsequent optical coherence tomography (OCT) B-scan images. (C, D) Spectral-domain OCT (SD-OCT) imaging showing a persistent subfoveal inflammatory pigment epithelial detachment with new subretinal fluid in the right and left eyes adjacent to the optic nerve. (E, F) Middle phase indocyanine green angiography (ICGA) images during the same visit confirming focal areas of hyperfluorescence suggesting choroidal hyperpermeability. These areas are similar to those seen in Figure 1F, which was the ICGA from the initial visit. Of note, there is a significant loss of choroidal vascular flow at the fovea in the right eye. Lines indicate the location of the subsequent OCT B-scan images. (G) SD-OCT image of the right eye showing a scan of the subfoveal lesion. (H) SD-OCT image of the left eye showing a normal fovea.

Four weeks later, the patient returned with no improvement in VA (Figures 6A–6D). An injection of intravitreal bevacizumab (Avastin; Genentech, South San Francisco, CA) was given to assess whether the PED was a vascular endothelial growth factor-mediated process. After a 4-week interval, the non-vascular inflammatory PED was still present, but her VA improved to 20/60-2 (Figure 6E); however, the patient reported no subjective improvement in vision. SS-OCT imaging demonstrated minimal change in the inflammatory lesion, which could represent a response to either the bevacizumab or a delayed response to the steroid therapy (Figures 6F–6H). SS-OCTA imaging remained unchanged, with a persistent focal decrease in the inner choroidal flow. The patient was observed without further intervention.

Imaging from weeks 13 and 17 after presentation, before and after bevacizumab therapy. (A) Color fundus image demonstrating persistence of the subfoveal retinal pigment epithelial detachment (PED) 13 weeks after presentation. The line indicates the location of the swept-source optical coherence tomography angiography (SS-OCTA) B-scan in (B). (B) SS-OCTA cross-sectional B-scan with the boundaries depicted for the choriocapillaris flow and structural en face images (C) En face flow image of the choriocapillaris 13 weeks after presentation showing a persistent decrease in flow underlying the inflammatory PED. (D) En face structural image of the choriocapillaris showing good signal intensity. (E) Color fundus image demonstrating persistence of the foveal lesion 4 weeks later, after an injection of intravitreal bevacizumab. The line indicates the location of the subsequent SS-OCTA B-scan in (F). (F) SS-OCTA cross-sectional B-scan with segmentation boundaries for the choriocapillaris. (G) En face flow image of the choriocapillaris with a persistent decrease in flow underlying the PED. (H) En face structural image of the choriocapillaris showing a good signal in the area of decreased choriocapillaris flow underlying the lesion.

Figure 6.

Imaging from weeks 13 and 17 after presentation, before and after bevacizumab therapy. (A) Color fundus image demonstrating persistence of the subfoveal retinal pigment epithelial detachment (PED) 13 weeks after presentation. The line indicates the location of the swept-source optical coherence tomography angiography (SS-OCTA) B-scan in (B). (B) SS-OCTA cross-sectional B-scan with the boundaries depicted for the choriocapillaris flow and structural en face images (C) En face flow image of the choriocapillaris 13 weeks after presentation showing a persistent decrease in flow underlying the inflammatory PED. (D) En face structural image of the choriocapillaris showing good signal intensity. (E) Color fundus image demonstrating persistence of the foveal lesion 4 weeks later, after an injection of intravitreal bevacizumab. The line indicates the location of the subsequent SS-OCTA B-scan in (F). (F) SS-OCTA cross-sectional B-scan with segmentation boundaries for the choriocapillaris. (G) En face flow image of the choriocapillaris with a persistent decrease in flow underlying the PED. (H) En face structural image of the choriocapillaris showing a good signal in the area of decreased choriocapillaris flow underlying the lesion.

Discussion

This case is an atypical presentation of MEWDS with a subfoveal inflammatory PED. Using SS-OCT angiography, we identified an inner choroidal flow void involving the choriocapillaris under the deposit, clearly demonstrating its non-vascular nature. Other atypical features of MEWDS included the lack of typical hypofluorescent lesions on ICG that are felt to result from irregular distribution of extravasated ICG dye into the subretinal tissues.6 However, there were subtle hyperfluorescent lesions on ICGA that may have foreshadowed her subsequent deterioration with fulminant CSCR when challenged with high doses of corticosteroids. Notably, no polyps or plaques were seen on ICGA to suggest polypoidal choroidal vasculopathy or other choroidal processes.

The lack of significant choroidal thickening is atypical for CSCR (Figure 2); given prior data,5 it appears as though the thickness of 334 μm would be considered within the expected range of choroidal thickness once corrected for age and axial length. However, given the choroidal hyperpermeability observed on ICG and the initial PED, an underlying diagnosis of CSCR may have been present at baseline. Although the PED may have been secondary to pre-existing CSCR, the internal hyperreflectivity of the PED had not been previously reported in CSCR except in association with choroidal neovascularization.

This case report demonstrates the utility of OCTA imaging over time to detect unsuspected findings even in well-defined disorders such as MEWDS. Although there may have been preexisting CSCR and a macular lesion in this young woman, there is a possibility that the acute inflammation of MEWDS affected tissues deep to the outer retina and produced a PED as described by Hashimoto et al.,4 which also resulted in scarring and poor vision. Similar loss of choriocapillaris perfusion may explain persistent poor vision in other patients with MEWDS.

MEWDS is usually a self-resolving disorder. However, the presence of choriocapillaris flow voids in the central macula may be a harbinger of poor vision. Whether the inflammatory PED resulted from MEWDS or CSCR is unknown, but the use of SS-OCT imaging will prove useful in following the disease course and explaining VA outcomes.

References

  1. Jampol LM, Sieving PA, Pugh D, Fishman GA, Gilbert H. Multiple evanescent white dot syndrome. I. Clinical findings. Arch Ophthalmol. 1984;102(5):671–674. doi:10.1001/archopht.1984.01040030527008 [CrossRef]
  2. Yannuzzi NA, Swaminathan SS, Zheng F, et al. Swept-source OCT angiography shows sparing of the choriocapillaris in multiple evanescent white dot syndrome. Ophthalmic Surg Lasers Imaging Retina. 2017;48(1):69–74. doi:10.3928/23258160-20161219-10 [CrossRef]
  3. Aoyagi R, Hayashi T, Masai A, et al. Subfoveal choroidal thickness in multiple evanescent white dot syndrome. Clin Exp Optom. 2012;95(2):212–217. doi:10.1111/j.1444-0938.2011.00668.x [CrossRef]
  4. Hashimoto Y, Saito W, Noda K, Ishida S. Acquired focal choroidal excavation associated with multiple evanescent white dot syndrome: Observations at onset and a pathogenic hypothesis. BMC Ophthalmol. 2014;14:135. doi:10.1186/1471-2415-14-135 [CrossRef]
  5. Abbey AM, Kuriyan AE, Modi YS, et al. Optical coherence tomography measurements of choroidal thickness in healthy eyes: Correlation with age and axial length. Ophthalmic Surg Lasers Imaging Retina. 2015;46(1):18–24. doi:10.3928/23258160-20150101-03 [CrossRef]
  6. Chang AA, Morse LS, Handa JT, et al. Histologic localization of indocyanine green dye in aging primate and human ocular tissues with clinical angiographic correlation. Ophthalmology. 1998;105(6):1060–1068. doi:10.1016/S0161-6420(98)96008-0 [CrossRef]
Authors

From Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Department of Ophthalmology, Miami.

Dr. Gregori and the University of Miami co-own a patent that is licensed to Carl Zeiss Meditec. Dr. Rosenfeld has received consulting honoraria from Carl Zeiss Meditec. The remaining authors report no relevant financial disclosures.

Drs. Gregori and Rosenfeld receive research support from Carl Zeiss Meditec. Additional research support was provided by a National Eye Institute Center Core Grant (P30EY014801) and an unrestricted grant from Research to Prevent Blindness, New York, NY to the Department of Ophthalmology, University of Miami Miller School of Medicine.

Address correspondence to Philip J. Rosenfeld, MD, PhD, 900 NW 17th Street, Miami, FL 33136; email: prosenfeld@med.miami.edu.

Received: April 10, 2017
Accepted: October 05, 2017

10.3928/23258160-20180129-12

Sign up to receive

Journal E-contents