Acute Zonal Cone Photoreceptor Outer Segment Loss

This case report describes an adolescent boy with acute visual field loss to the cone photoreceptor outer segment despite normal results of a fundus examination.

Key Points

Question

What is the pattern of dysfunction and structural change associated with acute visual field loss in the setting of a normal fundus examination?

Findings

In this case study of an adolescent boy, results of a fundus examination and short-wavelength and near-infrared fundus autofluorescence imaging were normal in both eyes, whereas near-infrared reflectance imaging showed a region of hyporeflectance temporally that corresponded with a dense cone scotoma. Photoreceptor outer segment loss detected by spectral-domain optical coherence tomography in this region co-localized with the area of cone dysfunction, but rod-mediated vision was normal.

Meaning

Localized, isolated cone dysfunction may represent the earliest photoreceptor abnormality or a distinct entity within the acute zonal occult outer retinopathy–complex group of diseases.

Abstract

Importance

The diagnostic path presented narrows down the cause of acute vision loss to the cone photoreceptor outer segment and will refocus the search for the cause of similar currently idiopathic conditions.

Objective

To describe the structural and functional associations found in a patient with acute zonal occult photoreceptor loss.

Design, Setting, and Participants

A case report of an adolescent boy with acute visual field loss despite a normal fundus examination performed at a university teaching hospital.

Main Outcomes and Measures

Results of a complete ophthalmic examination, full-field flash electroretinography (ERG) and multifocal ERG, light-adapted achromatic and 2-color dark-adapted perimetry, and microperimetry. Imaging was performed with spectral-domain optical coherence tomography (SD-OCT), near-infrared (NIR) and short-wavelength (SW) fundus autofluorescence (FAF), and NIR reflectance (REF).

Results

The patient was evaluated within a week of the onset of a scotoma in the nasal field of his left eye. Visual acuity was 20/20 OU, and color vision was normal in both eyes. Results of the fundus examination and of SW-FAF and NIR-FAF imaging were normal in both eyes, whereas NIR-REF imaging showed a region of hyporeflectance temporal to the fovea that corresponded with a dense relative scotoma noted on light-adapted static perimetry in the left eye. Loss in the photoreceptor outer segment detected by SD-OCT co-localized with an area of dense cone dysfunction detected on light-adapted perimetry and multifocal ERG but with near-normal rod-mediated vision according to results of 2-color dark-adapted perimetry. Full-field flash ERG findings were normal in both eyes. The outer nuclear layer and inner retinal thicknesses were normal.

Conclusions and Relevance

Localized, isolated cone dysfunction may represent the earliest photoreceptor abnormality or a distinct entity within the acute zonal occult outer retinopathy complex. Acute zonal occult outer retinopathy should be considered in patients with acute vision loss and abnormalities on NIR-REF imaging, especially if multimodal imaging supports an intact retinal pigment epithelium and inner retina but an abnormal photoreceptor outer segment.

Introduction

Acute zonal occult outer retinopathy (AZOOR) is an idiopathic syndrome characterized by acute or subacute changes in vision with minimal or no funduscopic or angiographic changes. Patients, typically young females, present with abrupt visual field loss, often with the perception of photopsias, variably impaired visual acuity, and abnormal full-field and/or localized electrophysiological responses. AZOOR can be asymmetric and shows stability or partial regression in most patients. Overlap between AZOOR and conditions within the white-dot syndromes led to the use of the term AZOOR complex. Strict diagnostic criteria of acute, regional, isolated outer retinal (photoreceptor) injury may help distinguish purer cases of acute photoreceptor degeneration from the numerous conditions within the AZOOR complex. We used multimodal imaging, psychophysics, and electroretinography to characterize the disease expression of a patient with acute visual field loss, outlining a diagnostic path that may contribute to a better understanding of conditions with acute outer retinal involvement.

Report of a Case

A healthy adolescent boy presented within 8 days of waking up complaining of a darkened nasal field in his left eye that was accompanied by photopsias. There were no prodromic symptoms or associated illnesses, history of retinotoxic exposures (medications, light), or family history of eye disorders. Written informed consent was obtained; the procedures adhered to the Declaration of Helsinki, and the study was approved by the institutional review board of the University of Pennsylvania.

Uncorrected visual acuity was 20/20 OU, and color vision (determined with the Farnsworth-Munsell D15 test) was normal in both eyes. His neuro-ophthalmic examination results were unremarkable. Goldmann perimetry (V-4e target) was full in peripheral extent in both eyes but showed a nasal and superior pericentral scotoma in the left eye. The anterior segment and vitreous were normal without cells; fundus examination results were also normal (Figure, A). Laboratory test results were negative for rapid plasma reagin (RPR) titer, fluorescent treponemal antibody absorption (FTA-ABS), herpes simplex virus (types 1 and 2), Bartonella, angiotensin-converting enzyme, rubeola, rubella, mumps, quantiferon-TB gold, cytomegalovirus antibody testing, Lyme disease, antinuclear antibody (ANA), antineutrophil cytoplasmic antibodies, HLA-B27, and basic chemistry and was positive for varicella zoster virus as a result of recent vaccination. A retinal autoimmune panel also showed negative results.

Figure. Acute Zonal Cone Photoreceptor Outer Segment Loss.

A, Color fundus photography and near-infrared reflectance (NIR-REF) images. Arrowheads indicate the boundary that corresponds to the location of the patient’s visual field defect. B, Short-wavelength (SW) and NIR en face fundus autofluorescence (FAF) imaging. C, Horizontal, 9-mm-long, spectral-domain optical coherence tomographic (SD-OCT) image through the fovea of the patient’s unaffected eye compared with the affected left eye. The nuclear layers are labeled (GCL indicates ganglion cell layer; INL, inner nuclear layer; and ONL, outer nuclear layer). D, Magnified SD-OCT cross sections from locations shown within the white rectangles in panel C. Distal retinal structures are numbered to the right: 1 indicates the external limiting membrane; 2, the ellipzoid zone (EZ); 3, the apical retinal pigment epithelium (RPE); and 4, the RPE. Overlaid bars correspond to the thickness of the ONL (blue) and the EZ-to-RPE distance (yellow). E, Multifocal electroretinography (mERG) amplitudes plotted against a pseudocolor scale. N indicates nasal retina; T, temporal retina. F, Light-adapted (achromatic stimulus) and dark-adapted (500-nm) sensitivity profiles measured with automatic static perimetry (200-millisecond duration; 1.7° diameter stimuli) in the patient compared with the normal reference range (indicated by the blue shaded band, which spans the mean value ±2 SDs). Dark-adapted photoreceptor mediation estimated with 2-color (500-nm and 650-nm) perimetry was rod mediated throughout the dark-adapted profile. The gap in the dark-adapted profile is the rod-free region near fixation (F). The light gray shaded portion of the graph represents the blind spot.

Spectral-domain optical coherence tomography (SD-OCT), en face near-infrared (NIR) reflectance (REF) and fundus autofluorescence (FAF) imaging with NIR and short-wavelength (SW) excitation lights was performed using an SD-OCT/scanning laser ophthalmoscope system (Heidelberg Engineering GmbH). NIR-REF imaging of the right eye showed the normal homogeneous gray background with superimposed retinal vasculature (Figure, A). The left eye showed an area of hyporeflectance in temporal-to-inferior retina that was separated from normal-appearing retina by a well-demarcated boundary (Figure, A [arrowheads]) that corresponded to the location of the patient’s visual field defect. Short-wavelength FAF imaging showed a normal homogeneous pattern of autofluorescence in both eyes with a dark center caused mostly by absorption of SW light by the macular pigment (Figure, B). Findings on NIR-FAF imaging were also normal in both eyes, with a brighter center reflecting greater melanin content and taller retinal pigment epithelial (RPE) cells in this region (Figure, B). Automated light-adapted static perimetry (HFA II-I; Carl Zeiss Meditec) using a 24-2 protocol confirmed an absolute scotoma corresponding to the region of NIR-REF abnormality in the left eye; the right eye was normal. Standard full-field flash electroretinography (ERG) was normal for rod- and cone-mediated responses, ruling out retinawide disease (not shown).

Spectral-domain OCT cross sections through the fovea demonstrated the normal appearance of the nuclear layers in both eyes (Figure, C). The left eye, however, had a downslope at approximately 2 mm of eccentricity in the temporal retina suggestive of retinal thinning (Figure, C). Magnified cross sections revealed approximation of the inner segment ellipsoid zone to the RPE, with loss of the signal from the interdigitation of the outer segments’ tips and the apical RPE (IZ) that coincided with the hyporreflective region on NIR-REF imaging (Figure, D). The changes likely represent shortening or loss of photoreceptor outer segments (POSs) and possible IZ abnormalities. Signals posterior to the RPE were normal. The parafovea and fovea, as well as locations nasal to the fovea, appeared normal, although the IZ was not as well defined compared with the right eye.

The SD-OCT thickness variables were quantified and related to co-localized measures of vision. At the fovea, the thickness of the outer nuclear layer (ONL) (115 μm; normal reference mean [2 SDs], 111 [30]) and the distance from the EZ to the IZ (25 μm; normal reference mean [2 SDs], 31 [8] μm) were within normal limits. At 2 mm of eccentricity in the temporal retina of the left eye, the ONL thickness was normal (65 μm; normal reference mean [2 SDs], 70 [10] μm), but the IZ was not discernible; the EZ to RPE distance was reduced compared with normal distances (18 μm; normal reference mean [2 SDs], 51 [11] μm). Multifocal ERG showed severely depressed responses throughout the temporal pericentral retina in the left eye; the right eye was normal (Figure, E). Achromatic and chromatic dark-adapted perimetry was performed following published methods. Thresholds were measured along the horizontal meridian corresponding to the retinal region scanned with SD-OCT. A light-adapted sensitivity profile showed a sharp transition from normal sensitivities near fixation and elsewhere in the nasal retina to a deep scotoma that co-localized with the region of structural and multifocal ERG abnormalities in the temporal retina (Figure, F). Surprisingly, rod function assessed with dark-adapted perimetry along the same region showed only 3-dB diffuse depression of sensitivity (Figure, F). The clinical picture remained grossly unchanged 9 months after presentation.

Discussion

Acute, zonal, severe and persistent POS loss and associated cone dysfunction in the absence of demonstrable RPE or choroidal changes points to the cone POS as the primary disease target in this patient, likely affected by a form of AZOOR. An autoimmune cause has been suspected in AZOOR. Antiretinal autoantibodies have been reported, which may interact with predisposing genetic factors and/or environmental disease triggers. Interocular asymmetry suggestive of a nongenetic cause for this retinopathy has been recognized in AZOOR associated with retinawide cone dysfunction. However, we could not identify an immune-mediated response or secondary causes of acute infectious or inflammatory retinopathies in our patient.

Although the topography of rod vs cone disease in AZOOR is rarely explored, the frequent localization of the abnormalities to the cone-rich central/pericentral retina, including an “annular” subtype reminiscent to that of our patient, suggests the pattern may be more frequent than previously reported. Grossly normal SW-FAF and NIR-FAF signals, which are dominated by RPE lipofuscin and melanin, respectively, excluded overt RPE abnormalities and long-term photoreceptor disease that would be expected to cause demelanization and/or changes in lipofuscin content within RPE cells (Figure, B). Although the fundus appearance was unremarkable, there was a dramatic change on NIR-REF imaging that co-localized with the loss of the POS and IZ signal and confirmed previous observations. Our findings may represent a stage of the AZOOR lesion that precedes the localized SW-hyperautofluorescence proposed as the earliest abnormality in this group of conditions. Long-term longitudinal observation is needed to determine whether this pattern will turn into a more recognizable AZOOR appearance and progress to include both photoreceptor subtypes as is known to occur in these conditions. Future studies using this method beyond the limitations of a case report are needed to determine the prevalence and significance of this pattern.

Conclusions

Acute changes limited to the POS with a preserved ONL, RPE, and anterior choroid witnessed in this patient suggest arrest of normal outer segment renewal and/or acute structural damage. The pattern raises hope that a focused search for the cause will lead to treatments and restoration of the POS physiologic features before photoreceptor death and irreversible vision loss takes place.