Atypical bilateral acute retinal necrosis in a coronavirus disease 2019 positive immunosuppressed patient

Coronavirus disease 2019 (COVID-19) is caused by a novel coronavirus (severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2]).¹ In mid-March 2020 the World Health Organization proclaimed the outbreak of COVID-19 a global pandemic.² Acute retinal necrosis (ARN) is a very rare retinal disease with rates reported of approximately 0.5 new cases per million population.³ Here we present a case of an immunosuppressed patient, with viral retinitis, who was concurrently positive for COVID-19 but was systemically largely asymptomatic from it. She subsequently underwent vitreous sampling. PCR testing of this vitreous sample was negative for COVID-19 but positive for varicella zoster virus (VZV). To the best of our knowledge this is the first description of a case that has undergone vitreous PCR testing for COVID-19. We highlight some very unusual features of this case which may be relevant to the current COVID-19 pandemic.

A 75-year-old lady presented to eye casualty with floaters and reduced vision in her left eye over a 3-week period in April 2020. The patient was apyrexial with a chronic cough and no other systemic symptoms. She therefore underwent testing for COVID-19. The patient’s nasopharyngeal swabs tested positive for COVID-19 in a reverse transcription polymerase chain reaction (RT-PCR) assay. At presentation her visual acuity was 6/12 right eye and counting fingers left eye. On examination she displayed a left panuveitis with stellate diffuse keratic precipitates, anterior chamber cells, vitritis and extensive peripheral and mid-peripheral necrotising retinitis. In the right eye, she had a superior peripheral retinitis with minimal anterior or vitreous inflammation. No iris signs were found in either eye. Figures 1 and 2 show widefield retinal imaging at presentation. Two months prior to her diagnosis she completed a course of rituximab and chlorambucil chemotherapy for a relapse of diffuse large cell B-cell lymphoma (DLBCL). Post this cycle of chemotherapy she had complete metabolic recovery on a whole body fluorodeoxyglucose (FDG) positron emission tomography (PET) computed tomography (CT) scan – this scan was conducted 2 months prior to her presentation to our Ophthalmology service. The DLBCL was a Non-Hodgkins Lymphoma initially treated in 2014 with rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) chemotherapy.

She has a history of systemic lupus erythematosus in remission and a previous Coxsackie viral infection in 2012 with associated intermittent chest pain. The patient’s blood results on the day of diagnosis of the viral retinitis were: white blood cell count: 1.1 × 10⁹ cells/L, neutrophil count: 0.41 × 10⁹ cells/L, lymphocyte: count 0.31 × 10⁹ cells/L – all values are below normal limits. One month prior to presentation her CD4 count was 423 cell/mm³ which is also below the normal range. No CD8 count was available. At presentation, the patient had an intravitreal sample sent for viral PCR and an intravitreal injection of 2.4 mg/0.1 ml of foscarnet was administered to the left eye. She was initially started on 900 mg twice daily oral valganciclovir which was then changed to valaciclovir 1 g three times daily for 6 weeks, by the Haematologists due to renal impairment, after the PCR result. The vitreous sample PCR tested positive for Varicella Zoster Virus (VZV) and was negative for SARS-CoV-2. The clinical picture and the VZV positive vitreous sample point to a diagnosis of ARN.⁴

Two months after her initial diagnosis the viral retinitis in her right eye significantly improved. The left eye continued to have poor vision (hand movements) due to retinal thinning and the development of a cataract. The cataract precluded good quality retinal photos to be obtained at the follow up stage.

A picture similar to Cytomegalovirus virus retinitis or progressive outer retinal necrosis would be expected in an immunosuppressed state.⁴ It is however unusual to note the level of vitreous inflammation observed in this case. In such an immunosuppressed patient one would not expect such a gross vitritis. There is evidence to suggest that other coronaviruses can compromise the blood retinal barrier.⁵ In COVID-19 it may therefore be possible that there is a compromised blood retinal barrier which allows a greater immune response to be employed than one would otherwise expect. SARS-CoV-2 is also a zoonotic pathogen and a similar picture of gross intraocular inflammation with other coronaviruses is seen in feline and murine genres.⁶ It is unusual to note a presentation of ARN in an immunosuppressed patient.

Extra ocular VZV has been reported to occur in DLBCL on rituximab chemotherapy.⁷ This report should raise the awareness of VZV related viral retinitis in such patients which has been sparsely reported in the literature.^7,8

This case lends support to evidence that COVID-19 is either not detected, or detected in a very low percentage, of ocular samples in patients who have COVID-19.^1,9 This statement obviously has to take in to account the limitations of all the laboratory tests conducted. It therefore further supports the possibility that the virus seems to have less tropism for ocular tissue.⁹ As COVID-19 is such a new disease there is limited evidence on its testing in ocular tissue at this moment in time.

Healthcare policymakers and organisations are having to make decisions regarding the re-introduction of routine Ophthalmic surgery worldwide as the pandemic moves forwards. It is important that all evidence regarding the presence or absence of COVID-19 in ocular samples is reported in order to supply an evidence base to these decision makers. The clinical indications for performing a vitreous sample are rare, thus this case report starts a body of evidence that may be limited even as the pandemic progresses. Vitreous sampling for COVID-19 is of particular obvious relevance to vitreoretinal surgery but is also important in cataract surgery where vitreous may be encountered.

It will be interesting to see how further evidence elucidates how the SARS-CoV-2 virus interacts with ocular tissue when further cases of ocular infection in patients with concurrent COVID-19 are assessed.

We would advise a careful consideration of atypical viral retinitis in patients complaining of visual symptoms who have COVID-19.