Systematic review exploring the clinical features of optic neuritis after SARS-CoV infection and vaccination

Introduction

Optic neuritis (ON) is a clinical manifestation of central nervous system (CNS) inflammation that most commonly presents with pain, reduction in visual acuity (VA) and dyschromatopsia. ON is often the first indication of autoimmune demyelinating conditions; however, it can also be rarely precipitated following bacterial or viral infections.1

Cases of ON have also been reported following vaccination including rabies and influenza vaccines; however, no large cohort studies have been published on this topic.2 3

Since the SARS-CoV-2 pandemic, several cases of ON have been reported in the literature citing SARS-CoV-2 infection as the precipitating cause.4 There has also been speculation of a rare potential association between SARS-CoV-2 vaccination and ON; however, a causal association has not been established, and it is uncertain whether there is indeed any association between SARS-CoV-2 infection or vaccination and ON. Mild neurological symptoms such as headache and fatigue have been reported following SARS-CoV-2 vaccination and are also common with SARS-CoV-2 infection itself.5 Recently, a systematic review which explored the association between SARS-CoV-2 infection and vaccination and acute transverse myelitis (ATM) concluded that the incidence rate of ATM post-SARS-CoV-2 infection was as low as 0.5 per one million infections.6

The aim of this study was to characterise the symptoms and clinical features of previously reported episodes of ON following SARS-CoV-2 infection or vaccination. A narrative approach was used to compare descriptions of ON temporally associated with COVID-19 infection or vaccination with established characteristics of typical ON.

Method

A systematic literature search was conducted of four databases (PubMed, Medline, Embase and Google Scholar) on 20 November 2022. Keywords used in the search strategies included but were not limited to “optic neuritis”, “MOG antibody disease”, “NMOSD” and “COVID-19” (online supplemental table 1).

All studies identified were transferred to Rayyan Software,7 which was used to assist with the screening process. Duplicate studies were identified automatically by the software and removed manually. Two authors (IG and DC) assessed the studies in a three-stage process by study title screening first, followed by abstract reviewing and then by full text assessment. Publications were screened by the two authors independently at every stage and carried forward to the next stage of screening if they met the predefined inclusion criteria. In cases of disagreement a third author (CJS) assisted in resolving decision disparities. Data were extracted from all studies by two independent authors (IG and DC) and was compared and confirmed by a third author (CJS). Studies including patients with a clinical or radiological diagnosis of ON of likely demyelinating or inflammatory aetiology occurring within 3 months of SARS-CoV-2 infection or vaccination were included. The 3-month interval was chosen as it reflects the sequalae of acute and subacute infection before entering the ‘chronic and postacute’ phase after 12 weeks.8 Only studies including patients with SARS-CoV-2 confirmed on a PCR/NAAT test or those who had received a SARS-CoV-2 vaccination were eligible. There was no age restriction and case reports, and case series were included in the review due to the rarity of ON following SARS-CoV-2. Only studies that were published in English were entered.

Studies including patients diagnosed with ON more than 3 months after the last vaccination date or a positive SARS-CoV-2 PCR/NAAT test were excluded, as were those including patients with an existing or suspected diagnosis of any demyelinating condition, and those where patients had synchronous infections or other ophthalmological or systemic pathology potentially contributing to their symptoms. Neuropathies that were likely to be ischaemic or granulomatous in nature were also excluded.

Results

Study and patient characteristics

In total, the search strategy identified 1464 publications, of which 68 were eligible for inclusion. The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow chart is shown in figure 1. There were 34 studies of ON following COVID-19 infection,12–45 and an equal number of publications reported patients with ON post-COVID-19 vaccination.46–78 There were 19 case series and the remaining 49 were case reports. The minimum number of cases in a case series was 2 and the maximum 6. The 68 publications included a total of 93 patients and 125 eyes. The mean age of all the patients was 43.3 years with a range of 8–86 years and SD 17.7.

Request permission

Figure 1

PRISMA flow diagram demonstrating the identification and inclusion of studies.

Discussion

Several hypotheses suggesting possible mechanisms regarding the development of autoimmune phenomena following vaccinations have been reported in the literature. One of the reported theories is that of molecular mimicry, according to which, the structural homology between pathogens and self-proteins may be responsible for inducing an autoimmune response.79 More specifically, this mechanism was considered relevant in the cases of demyelination following hepatitis B virus vaccines. One study published in 2005, suggested that aminoacid similarities between the small hepatitis B virus surface antigen, myelin basic protein, and myelin oligodendrocyte glycoprotein (MOG), serve as targets of immunological cross-reactivity.80

Vaccinations have been previously associated with conditions that are presumed to interfere with myelin membranes. For example, Guillain-Barré syndrome is a well-recognised complication that may arise following a vaccination and is supported by evidence in the literature. The pathophysiological mechanism is thought to be secondary to the generation of antibodies/T-cells that could cross-react with epitopes on myelin due to generation of vaccine-associated products.81

Cases of CNS demyelination following vaccine administration have also been previously published in the literature. A PubMed search between 1979 and 2013 identified 71 cases of demyelination after various vaccines, with isolated ON being the most frequent manifestation.82 Despite these findings, a recent systematic review which examined the association between multiple sclerosis and several vaccines did not identify significant findings to establish causation.83

In this study, it was found that the majority of demyelinating episodes occurred after the administration of the first dose. One study following 1736 patients after SARS-CoV-2 vaccinations, reported that adverse events were more likely to be observed following the second dose.84 Furthermore, a large self-controlled case series study of people aged 13 years or older in England examined the association between myocarditis and COVID-19 vaccination. It is reported that occurrence of myocarditis was higher after the second dose.85 Similarly, a systematic review and meta-analysis with global data, reported that 82.1% of the included patients developed myocarditis after the second dose. It is suggested that the condition may be precipitated by cross-reactivity; however, no pathophysiological mechanism addressing increasing incidence with the second administration is reported.86

The CNS has been documented to be susceptible to several neuroinvasive and neurotropic viruses. A number of those viruses have also demonstrated neurovirulent properties as they have been associated with neurological disease.87 The viruses can penetrate the CNS via the blood–brain barrier, the blood–cerebrospinal fluid barrier as well as peripheral nerves and other routes.88

Respiratory viruses, such as RSV, have demonstrated capacity to infect the CNS and induce neurological sequelae after intranasal inoculation in murine studies.89 Similarly, human coronaviruses (HCoV), even though they are traditionally associated with respiratory pathology they can also invade the CNS causing a range of neurological symptoms including but not limited to anosmia, ageusia, headache, seizures and encephalopathy.90 Association of HCov with MS and demyelination has been previously reported; autopsy samples in patients with MS have shown statistically significant higher prevalence in OC43 coronavirus strain91 compared with neurologically intact controls. Therefore, there is a putative mechanism suggested in the literature and a sizeable number of case reports were identified in this study. However, in order to demonstrate causality, a putative correlation between COVID-19 infection and ON should also be established, with consideration to potential confounding factors. Due to the nature of the publications identified, the cohort studied may be affected by selection bias as identifying a patient for a case report is an author dependant decision.

The clinical features of ON in the cases identified in this study are interesting but show considerable overlap with the known features of ON due to other factors. One of the findings of this review was that 56.4% of the patients were female, in contrast with the Optic Neuritis Treatment Trial (ONTT), a multicentre collaborative study in the USA, in which the females comprised 77% of patients.92 Female predominance is observed for MS and other autoimmune diseases, while a further rise in the female to male ratio has been reported for the past decades.93 Potentially, the 1.3:1 female to male ratio in the current analysis suggests that the aetiology of the ON in this patient cohort is not reflective of typical autoimmune disease patterns.

A feature of typical ON is the presence of pain, particularly on eye movement. We found that 78% of the participants experienced pain which is lower than the previously quoted 92% in typical ON.92 Data from more recent studies show variable rates of ocular pain ranging between 46% and 77% depending on the presence of antibodies in serum and in a small Malaysian study the presence of pain was even found to be as low as 31.7%.94 95

A typical or MS associated ON is generally unilateral and bilateral cases are seen in paediatric age group. While aetiologies such as autoimmune antibody mediated, or infections or parainfectious are usually bilateral in presentation.96 In this study, we found the incidence of bilateral ON was as high as 33.3%, which is in line with a recent study published in 2016, in which 30% of patients with MOG-IgG antibodies experienced bilateral pathology.97 These findings may suggest that the higher incidence of bilateral ON in this study could be attributed to atypical aetiology.

The presence of IgG oligoclonal bands has been described as the immunological hallmark of multiple sclerosis, as they can be found in the CSF of 95% of patients with multiple sclerosis. Contrary to this finding, only 16.7% of patients with reported CSF findings in the infection cohort and 46.7% in the vaccination cohort tested positive for oligoclonal bands.98

One of the significant findings of the current study was that most of the patients with ON after a COVID-19 infection presented with severe visual impairment, with 64% presenting with a VA of 1.00 LogMAR (Snellen 6/60) or worse. In the vaccination cohort, 39% of patients had a severe visual impairment, similarly with 36% of the patients in the ONTT.92

In conclusion, patients with ON post-SARS-CoV-2 infection experienced more severe visual impairment compared with patients after a SARS-CoV vaccination. In addition, the proportion of patients suffering from bilateral ON was higher compared with typical ON.

Beyond the case reports summarised in this review, there is no current evidence to support causation between SARS-CoV-2 infection or vaccination and ON. To establish causation, further research is needed, in particular collection of longitudinal data to determine potential progression to MS, MOGAD or NMOSD in those affected.