Case presentation

The day of presentation to our hospital was defined as hospital day 0. A 63-year-old man developed decreased visual acuity in the left eye approximately 17 days before presentation to our hospital. He had no history of immunodeficiency, diabetes mellitus, or malignancy. HIV serology was negative. HbA1c was 6.1%, physical examination and laboratory testing did not suggest collagen vascular disease, and systemic computed tomography showed no evidence of malignancy. No evidence of immunodeficiency was identified.

Thirteen days before presentation to our hospital, he visited a local ophthalmology clinic, where OCT obtained using a Topcon system showed a lesion centered on the fovea (Figure 1A). The exact Topcon model and scan protocol were not available in the record. Ten days later, three days before presentation to our hospital, he was referred to the ophthalmology department of another hospital because macular degeneration was suspected. Best-corrected visual acuity in the left eye was 0.2, and intraocular pressure (IOP) was 30 mmHg. A focal whitish lesion was present at the foveal center (Figure 1C). OCT obtained using ZEISS CIRRUS HD-OCT demonstrated marked hyperreflectivity involving nearly the entire inner retina at the fovea, with posterior shadowing and disruption of the ellipsoid zone (Figure 1B). This pattern was broader than the middle retinal band expected in PAMM. Fluorescein angiography showed mild macular hyperfluorescence without apparent retinal vascular occlusion or retinal vasculitis (Figure 1D). Macular OCT angiography was performed using CIRRUS HD-OCT 5000 with AngioPlex (Carl Zeiss Meditec, Dublin, CA, USA) with a fovea-centered 6 x 6-mm scan. The superficial and deep retinal slabs were generated automatically using the built-in AngioPlex software. The segmentation boundaries and corresponding B-scans were reviewed for obvious segmentation errors, and no manual correction was required. The segmentation boundaries and corresponding B-scans were reviewed for obvious segmentation errors, and no manual correction was required. No detectable perfusion abnormality was identified within the scanned macular area in either slab (Figures 1E, 1F). Dilated peripheral retinal examination and widefield fundus photography showed no visible peripheral necrotizing lesion (Figure 2); scleral depression was not performed. Although the presentation was atypical, ARN was included in the differential diagnosis. Oral valacyclovir 500 mg twice daily and topical antiglaucoma medications were initiated.

Figure 1: Sequential multimodal imaging of foveal-onset acute retinal necrosis (ARN) associated with Varicella-zoster virus (VZV)(A) Optical coherence tomography (OCT) at the first local ophthalmology clinic shows a lesion centered on the fovea. (B-F) Findings at the referring hospital three days before presentation to our hospital. (B) OCT demonstrates marked hyperreflectivity involving nearly the entire inner retina at the fovea, with posterior shadowing and ellipsoid zone disruption. (C) Color fundus photography shows a whitish lesion localized to the foveal center. (D) Fluorescein angiography shows mild macular hyperfluorescence without apparent retinal vascular occlusion or retinal vasculitis. (E) The superficial retinal slab of a 6 x 6-mm macular OCT angiography scan shows no detectable perfusion abnormality within the scanned area. (F) The deep retinal slab also shows no detectable perfusion abnormality within the scanned area. (G, H) Findings at the first visit to our hospital. (G) OCT shows persistent broad inner retinal hyperreflectivity and foveal ellipsoid zone disruption. (H) Color fundus photography shows the foveal lesion without additional lesions in the imaged posterior pole. (I) Three weeks after treatment initiation, OCT shows regression of the inner retinal hyperreflectivity and partial restoration of the ellipsoid zone outside the foveal center. (J) At the final follow-up, 42 weeks after treatment initiation, OCT shows stable retinal architecture and preservation of the ellipsoid zone outside the foveal center. Red arrows indicate the foveal lesion or the corresponding foveal OCT abnormality at each time point.

Figure 2: Widefield color fundus montage obtained at the referring hospital three days before presentation to our hospitalThe image shows the foveal lesion and no visible peripheral necrotizing retinitis within the photographed field. The red arrow indicates the foveal lesion. Scleral depression was not performed.

At the first visit to our hospital, best-corrected visual acuity was 1.5 in the right eye and 0.2 in the left eye. IOP was 12 mmHg in the right eye and 15 mmHg in the left eye while the patient was receiving topical antiglaucoma therapy. Laser flare photometry was 6.7 photon counts per millisecond in the right eye and 19.2 photon counts per millisecond in the left eye. Slit-lamp examination of the left eye showed fine keratic precipitates without anterior chamber cells. Anterior vitreous cells and vitreous opacity were not observed. Dilated fundus examination and widefield imaging showed no visible peripheral whitish lesion, retinal hemorrhage, or apparent retinal vasculitis. Color fundus photography documented the foveal lesion (Figure 1H). Scleral depression was not performed. OCT using ZEISS CIRRUS HD-OCT showed persistent broad inner retinal hyperreflectivity with foveal ellipsoid zone disruption (Figure 1G). Aqueous humor was collected at this visit for a 24-target multiplex ocular pathogen PCR assay while no peripheral retinal lesion was clinically evident.

Four days after presentation to our hospital, dilated fundus examination and ultra-widefield imaging documented a new inferotemporal peripheral retinal lesion between the 4- and 5-o'clock positions. The lesion measured approximately four disc diameters and was accompanied by vascular sheathing, punctate retinal hemorrhages, and vitreous opacity. These findings were clinically consistent with necrotizing herpetic retinitis. On the same day, the multiplex PCR result from the aqueous sample obtained at the first visit became available and was positive only for VZV DNA. HSV-1, HSV-2, Epstein-Barr virus, cytomegalovirus, human herpesviruses 6, 7, and 8, and all other tested infectious targets were negative. These findings established VZV-associated ARN.

Oral valacyclovir, which had been started at 500 mg twice daily three days before presentation, was increased to 1000 mg three times daily on hospital day four. Oral betamethasone 2 mg/day was initiated on the same day after seven days of oral antiviral therapy. Intravenous acyclovir 750 mg three times daily was started on hospital day 10 and continued until day 25. Betamethasone was increased to 6 mg/day intravenously on day 11, reduced to 4 mg/day on day 18, and to 2 mg/day orally on day 25. From day 42, corticosteroid therapy was switched to prednisolone 10 mg/day, reduced to 5 mg/day on day 56, 2.5 mg/day on day 70, and 2.5 mg every other day on day 84, and discontinued on day 98. After completion of intravenous acyclovir, oral valacyclovir 1000 mg three times daily was resumed on day 25, reduced to 1000 mg twice daily on day 42 and 500 mg twice daily on day 56, and discontinued on day 70.

The peripheral lesion rapidly regressed, and the broad inner retinal hyperreflectivity gradually decreased. At three weeks, the peripheral lesion had resolved, and OCT showed partial restoration of the ellipsoid zone outside the foveal center (Figure 1I). At six weeks, best-corrected visual acuity improved to 0.5. At the final follow-up, 42 weeks after treatment initiation, best-corrected visual acuity had improved to 0.6, with no retinal detachment, recurrent retinitis, or involvement of the fellow eye (Figure 1J).