SARS-CoV-2 took the world by storm at the dawn of 2020. The novel single-stranded RNA beta coronavirus affected the health and livelihoods of millions across the globe on a scale only encountered once approximately every 100 years (Wesselingh and Wesselingh, 2023). There is currently a plethora of case reports and case series of neuro-ophthalmic manifestations of COVID-19 and its vaccine counterparts (Luís et al., 2020; Betsch and Freund, 2021; Tisdale et al., 2021). As the end of this pandemic is reached, this review seeks to summarise the current knowledge of neuro-ophthalmic manifestations of both COVID-19 infection and vaccination.

Electronic searches for published literature were conducted using EMBASE and MEDLINE on the 30th of July 2023, see Figure 1 . The search strategy comprised of controlled vocabulary and free-text synonyms for the following terms in various combinations: “coronavirus, COVID-19, SARS- CoV-2, 2019-nCoV, vaccination, vaccine, immunisation and neuro-ophthalmology”. No time range limits were set for the literature search. Published English abstracts for articles written in a different language were screened if available.

The incidence of neuro-ophthalmic sequelae following COVID-19 disease is far greater than that following vaccination against the virus. As of December 2023, there have been more than 770 million confirmed cases of COVID-19 (including nearly 7 million deaths) reported to WHO (2023b). At the same timepoint, 13 billion vaccine doses had been administered, with nearly 5.6 billion people vaccinated with a complete primary series. The 48 cases of adverse neuro-ophthalmic sequelae following infection and 31 cases following vaccination included in this review must be viewed in the context of the significant difference in denominator between these two groups.

It is difficult to conclude direct causation as opposed to correlation between COVID-19 and the neuro-ophthalmic sequelae, as the literature currently available primarily consists of individual case reports and small case series. Optic neuritis was the most frequently reported neuro-ophthalmic sequelae following COVID-19 infection and vaccination in this review, and prior to the onset of the pandemic is known to occur at a stable rate in the general population (Braithwaite et al., 2020). Optic neuritis can be associated with specific identifiable causes, such as multiple sclerosis, infection, or may be idiopathic, and the onset may seem to develop randomly without a trigger or cause. With neuro-ophthalmic conditions developing at a spontaneous low rate in the population at any time, many of the reported neurological sequelae following COVID-19 infection and vaccination in this review may have simply been unrelated coincidences.

In their multi-centre observational study, Zhao et al. analysed the rates of diagnosis of neuro-ophthalmic conditions as well as retinal detachment (RD) and acute angle closure glaucoma (AACG) (Zhao et al., 2024). They found that all conditions were diagnosed at higher rates following the introduction of COVID-19 vaccines compared to pre-COVID and pre-vaccine periods. The authors discuss that a likely explanation for the increased diagnostic rates following introduction of vaccinations is that patients tended to postpone seeking treatment even for acute ocular conditions such as RD and AACG during the pandemic, with clinic visit rates only noted to increase when vaccines became available to protect against the virus. Unlike neuro-ophthalmic conditions such as optic neuritis and cranial neuropathies, RD and AACG do not have plausible causal mechanisms in association with COVID-19 infection and vaccination, therefore the parallel fluctuation in diagnostic rates may point more towards correlation as opposed to causation. Nevertheless, the temporal association of the neuro-ophthalmic events occurring concurrent with or after SARS-CoV-2 infection or vaccination has led to the proposition of various causal pathomechanisms.

One theory of the pathogenesis of optic neuritis following COVID-19 infection is through direct viral inflammation of the optic nerve, subsequently causing myelin damage (Deane et al., 2021). This was demonstrated in mice inoculated with the mouse hepatitis virus MHV-A59 whereby upon isolation, the optic nerve sheaths and parenchyma were infiltrated with inflammatory cells (Shindler et al., 2008). It has been suggested that a similar mechanism may underlie optic neuritis as a result of COVID-19 infection as the mouse hepatitis virus shares a common genus with SARS-CoV-2 (Körner et al., 2020). An alternate theory is optic nerve ischemia secondary to a systemic state of hypercoagulability associated with SARS-CoV-2 infection. The increased thrombin generation and endothelial inflammation results in a hypoxic state which may induce an ischemic optic neuropathy (Deane et al., 2021). However, ischemia is less likely to be reversible, therefore the observation that most cases of optic neuritis concurrent with or following COVID-19 infection demonstrated improvement after treatment detracts from the likelihood of ischemia being the underlying causal mechanism.

Eleven of the 25 cases of new-onset optic neuritis in this review demonstrated MOG antibody positivity. It has been hypothesised that much like other viral infections, the aberrant post-infectious immune response following COVID-19 triggers the onset of a para-infectious demyelinating process (Tenembaum et al., 2007; Bosello et al., 2023). One suggested mechanism is through molecular mimicry, where viral antigens induce a pathological immune response against CNS antigens including antibodies against MOG (Tisdale et al., 2021; Bosello et al., 2023). However, it has also been pointed out that despite auto-antibodies targeting MOG, molecular mimicry is unlikely the cause due to lack of homology between the COVID-19 viral protein and the MOG protein (Johnsson et al., 2022). Instead, loss of self-tolerance in the context of an activated adaptive immune system is suggested to be culprit (Wesselingh and Wesselingh, 2023).

There are many cases in the literature of new-onset IIH following COVID-19 infection, with a limited consideration of how CSF hydrodynamics is affected by COVID-19 infection. Under physiological conditions, the choroid plexus secretes CSF (Wang et al., 2022) and is a known site of blood brain barrier breakdown in COVID-19 (Fullard et al., 2021; Yang et al., 2021). SARS-CoV-2 has tropism for the choroid plexus epithelium, meninges, and brain vasculature as the SARS-CoV-2 entry proteins ACE2 and TMPRSS2 are both expressed at these sites, which has been suggested as the underlying mechanism of dysregulated CSF hydrodynamics seen in COVID-19 (Yang et al., 2021). Endothelial dysfunction in COVID-19 also impairs CSF absorption via the arachnoid villi and astrocytic foot processes surrounding cerebral vasculature (Wang et al., 2022). Additionally, COVID-19 leads to a systemic state of hyperviscosity and hypercoagulability which increases venous pressure and induces formation of cerebral venous sinus thromboses, also leading to increased intracranial pressure (Al-Mufti et al., 2021; Mukharesh et al., 2022).

The pathogenesis of myasthenia gravis following COVID-19 is under investigation (Sriwastava et al., 2021). It is known that a systemic state of inflammation such as that following SARS-CoV-2 infection reduces the availability of AChRs at the postsynaptic neuromuscular junction (Koneczny and Herbst, 2019). Additionally, an external virus induces antibody production, which evokes an immune response that subsequently cross-reacts with AChRs as a result of molecular resemblance (Sriwastava et al., 2021). The affinity of COVID-19 for the ACE2 receptor found on multiple organs including the brain, lungs, and heart leads to the formation of autoantibodies and subsequently a downstream cascade involving proinflammatory cytokines, chemokines, depletion of B and T cells, and increased interleukins and TNF-a (Baig et al., 2020). This increases the risk of a cross-reactive autoimmune attack against the body’s own receptors (Baig et al., 2020). The treatment target is to improve symptoms, control antibody production and reduce disease severity, and typically encompasses pyridostigmine and immunosuppressive agents (Sriwastava et al., 2021).

PRES occurs due to cerebral oedema secondary to endothelial dysfunction that presents as visual disturbance, headache, seizures and altered consciousness (Triplett et al., 2022). SARS-CoV-2 has affinity for the viral receptor ACE2, which is also expressed at the capillary endothelium. The inflammatory nature of COVID-19 infection leading to vascular endothelial injury is suggested to be the causal mechanism of PRES development. The literature of PRES following COVID-19 infection present with similar symptoms to the three cases described in this review including altered mental status, seizure activity, visual impairment and visual agnosia (Franceschi et al., 2020; Kaya et al., 2020; Kishfy et al., 2020; Hixon et al., 2021). Neuroimaging in these three cases is also comparative with that described in the literature of T2/FLAIR hyperintensities diffusely distributed in the cerebral white matter without corresponding diffusion restriction on DWI (Hixon et al., 2021). The limiting factor is that none of the three cases underwent cerebrospinal fluid testing, hence direct infection resulting in the development of PRES cannot be excluded despite its uncommon occurrence.

Concerns for neurological complications of the vaccines began arising in 2020 when two patients developed transverse myelitis following the AstraZeneca vaccine (Goss et al., 2021). Serious systemic complications have been reported after vaccination, such as cerebral venous thrombosis as well as specific immune-mediated conditions such as myocarditis, pericarditis and thrombocytopenia (Chen et al., 2022). Most adverse neurological sequelae occurred after AstraZeneca vaccinations, followed by Pfizer-BioNTech and Moderna vaccines (Sriwastava et al., 2022). However the rates of each vaccination vary widely, for example as of November 2023, the total vaccine doses administered in the European Union were Pfizer-BioNTech 654 million, Moderna 155 million, AstraZeneca 67 million, Johnson & Johnson 19 million, Sinopharm/Beijing 2 million, Sputnik V 2 million (Mathieu et al., 2021). In this review, Pfizer-BioNTech was associated with nearly half of the neuro-ophthalmic complications reported, followed by AstraZeneca and Moderna, but it is clearly difficult to compare the rates of complications between vaccines when the absolute risks are so low and the numbers of vaccine doses administered varied so widely.

Optic neuritis was the most common complication following COVID-19 vaccination in this review. Optic neuritis following vaccine administration has previously been reported for other anti-viral vaccines, including but not limited to measles/mumps/rubella, influenza, rabies, tetanus/diphtheria/pertussis, hepatitis B and Herpes Zoster (Stübgen, 2013; Cheng and Margo, 2022a; Bhatti et al., 2023). There is no definitive causal relationship between vaccination and optic neuritis onset (Bhatti et al., 2023), although propositions of the etiopathogenesis include reaction to a vaccine component (e.g. protein or preparation adjuvant), molecular mimicry, production of proinflammatory cytokines, and/or activation of the innate and adaptive immune system (Bhatti et al., 2023). Nevertheless, reports of serious neurological sequelae following COVID-19 vaccination are still considered rare, especially in comparison with that caused by COVID-19 infection which has been found to be 617-fold higher in an analysis from the vaccine adverse event reporting system (VAERS) (Frontera et al., 2022).

As we approach the end of the SARS-CoV-2 pandemic, there is an increased awareness of the neuro-ophthalmic complications brought on by viral insult. The various COVID-19 vaccines that are currently available are part of the global effort to protect the most vulnerable of the human population. The incidence of neuro-ophthalmic consequences following infection with COVID-19 is hundred-folds higher and associated with more harrowing systemic effects than that following vaccination against the virus. Several proposed causal mechanisms are discussed in this review, but larger studies are required to make any definitive conclusions. As we approach the post-COVID era, neuro-ophthalmic presentations should nevertheless always necessitate careful exploration of COVID-19 infection history and vaccination status.

JS: Writing – original draft. HD: Writing – review & editing.