COVID-19 and sudden-onset ocular neurogenic palsy in prior healthy patients: a systematic review

ABSTRACT

Background

The aim of this systematic review is to identify cases of neurogenic ocular palsy in the presence of COVID-19 and to document patient characteristics, type of palsy and possible aetiologies.

Methods

A systematic search of PubMed, Medline and CINAHL databases was conducted on the 6th of January 2023 to identify cases of neurogenic ocular palsy in patients with current or previous COVID-19 infection. Data were pooled to summarise the neurogenic palsy, patient clinical characteristics and proposed palsy mechanisms.

Results

The combined database search yielded 1197 articles. Of these, 23 publications consisting of 25 patients met the inclusion criteria. Most patients were male (68%) and ranged in age from 2 to 71 years (median=32.7, SD=21.4). Seven patients (28%) were children aged 2 to 10 years old. Abducens palsies were most common (68%) and the most common ocular presentation was diplopia (76%) with an average time of onset 15 days from testing positive to COVID-19 or having symptoms of the virus. Proposed mechanism of development of a neurogenic palsy secondary to COVID-19 infection was classified into one of three categories: vascular/thrombotic, a viral neuro-invasive or inflammatory virus-mediated immune response.

Discussion

This study suggests that COVID-19 infection may be linked to oculomotor, trochlear and abducens nerve palsies and the underlying mechanisms may vary but are difficult to definitively establish. Further studies investigating the onset of neurogenic palsy secondary to COVID-19 infection is required.

KEYWORDS:

• Abducens nerve palsy
• COVID-19
• diplopia
• neuro-ophthalmological complications
• oculomotor nerve disease
• oculomotor nerve palsy
• ophthalmoplegia
• SARS-CoV-2 infection
• trochlear nerve palsy

Introduction

The coronavirus disease 2019 (COVID-19) resulting from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a respiratory virus that was declared a pandemic in March 2020.¹ It most commonly presents with symptoms of fever, dry cough and shortness of breath and can be detected by a Rapid Antigen Test (RAT) or Reverse Transcription Polymerase Chain Reaction (RT-PCR) test.² Whilst the respiratory manifestations of COVID-19 are commonly studied, less is known about the ocular presentation of the virus. Conjunctivitis, dry eyes, foreign body sensation, redness, itching and epiphora seem to be the most common associated symptoms.^3,4 The down-regulation of angiotensin-converting enzyme (ACE-2) receptors, which are proteins found on the surface of several cells in the body, is implicated in both ocular manifestations and multiple-organ injury secondary to the virus.⁵ There is a reported link between COVID-19 and peripheral nervous system complications such as neuroinflammation, vasodilation and thrombotic events; however, the underlying pathophysiology is not well established but may be caused by the neurotrophic and neuroinvasive nature of the coronavirus.^3,6

Given the nervous system complications that can occur due to COVID-19, there have also been reports of sudden-onset-acquired ocular neurogenic palsies as a result of the virus.^7–10 Acquired neurogenic palsies result in weakened ocular muscle strength and limitation of ocular movement and, of the three cranial nerve (CN) types involved with eye movement (oculomotor, trochlear and abducens), abducens nerve palsies have been reported as the most common.⁸ Patients generally present with an initial complaint of sudden-onset diplopia and this symptom typically resolves within 1 to 3 months, with full recovery.⁸

The underlying cause of an ocular neurogenic palsy secondary to COVID-19 is uncertain; however, based on the literature, there appear to be three broad causes: 1) a vascular/thrombotic, 2) neuro-invasive or 3) inflammatory immune response.^7,9,11,12 The vascular/thrombotic cause is linked to immune modulated thrombosis caused directly by the COVID-19 virus. The virus suppresses ACE-2 receptors that would normally act as an anti-coagulant. Dysfunctional microvasculature occurs due to the coagulated blood and this in turn causes the neurogenic palsy.^7,13,14 The ACE-2 receptors are also implicated in the neuro-invasive cause. It is thought that the COVID-19 virus can enter the brain via these receptors (or directly via the olfactory bulb) and bind to neurons, thereby inducing the CN palsy.^9,15 It is also possible that COVID-19 can cause dysfunctional nerve conduction via an excessive immune response post-infection. The dysregulated immune response leads to inflammation and an increase in cytokine levels. Cytokines, the small proteins involved in immune response signaling, create a “cytokine storm” and high levels of these proteins cause neurodegeneration and de-myelination leading to the nerve palsy.^3,9,16

The aim of this systematic review was to identify cases of neurogenic palsy in the presence of COVID-19 and to document the patient characteristics, type of palsy and possible etiologies. Recognizing any links between COVID-19 and neurogenic palsies will facilitate more efficient diagnosis and add to the body of literature in this emerging area.

Methods

Search strategy

A systematic search of three electronic databases (PubMed, Medline (1946-) on Ovid, and CINAHL) was conducted on 6 January 2023. Search results were limited to human subjects and those published in English. The search terms included SARS-CoV-2, COVID-19, coronavirus, palsy, diplopia, neurogenic, third nerve, oculomotor nerve, fourth nerve, trochlear nerve, sixth nerve and abducens nerve with Boolean operators used.

Eligibility criteria

Case reports and case series of patients, irrespective of age, positive for SARS-CoV-2 or COVID-19 with a diagnosis of ocular neurogenic palsy within 28 days of infection were included. The time to onset of the palsy of 28 days was set by the authors based on the findings of a systematic review of all ocular manifestations of COVID-19.⁴ This period of time also limits other systemic complications as the cause of the palsy.

Case reports or case series had to contain prior medical history and indicate whether there were preexisting comorbidities. Reports of patients with comorbidities including a history of stroke, diabetes mellitus, cancer and dementia were excluded.

Study selection

Endnote X9 (Clarivate, Philadelphia PA, USA) was used to import the search results and duplicates were automatically removed. The titles and abstracts of the articles were independently screened by two authors (RS and SB). Articles not meeting the inclusion criteria were removed, and the full text of the remaining articles were obtained and reviewed in full by authors RS and SB.

Data extraction, critical appraisal and data synthesis

All articles included in this study were independently appraised by authors RS and SB using the Joanna Briggs Institute (JBI) checklist for case reports¹⁷ or case series.¹⁷ The JBI appraisal tool was modified to provide a numerical score for quality where a “yes” received 2 points, “unclear” received 1 point and “no” received 0 points with a maximum of 16 points for case reports and 20 points for case series.¹⁸ Articles were graded by author MV using this method. Data were extracted using a self-designed data extraction table. Data collected included patient age and sex; whether they were symptomatic to COVID-19; neurogenic palsy classification; ocular symptoms; time to development of ocular symptoms and time to improvement or resolution of symptoms; treatment and possible palsy mechanism.

Due to inconsistent and heterogeneous patient results it was not possible to conduct a meta-analysis. Therefore, data were synthesized in a narrative summary.

Results

Search results

A combined search of the three electronic databases yielded 1197 articles and 17 were recognized as duplicates and removed through the Endnote X9 software. Using the title and abstract, 1180 articles were independently screened by two authors (RS and SB). The selection process is documented in a PRiSMA flow diagram (Figure 1). Both authors read the full text of the remaining 48 articles. Twenty-three published reports^{7,9–12,19–36} met the inclusion criteria, and all patients presented with COVID-19 infection and a neurogenic palsy.

Figure 1. Modified PRiSMA flow diagram of study selection process.

Study quality

The JBI tool for case reports was used for 20 publications, and the tool for case series was used for three publications. None were excluded through the critical appraisal process. Reduction in points for quality was due to lack of information required for data extraction (Table 1).

Table 1. Summary of patient characteristics and critical appraisal score (* denotes a case series and ^ denotes a case study).

Author/s & year	Country	Sex	Age	Covid-19 symptomatic	Time to ocular symptoms (days)	CN palsy	Ocular symptoms	Treatment	Time to improved or resolved symptoms (days)	Proposed mechanism	Case report or Case Series Quality score
Anilkumar et al.^Citation7	United Kingdom	F	44	No		Abducens	Diplopia	Paracetamol	14		14/16^
Baccarella et al.^Citation19	United States of America	M	9	Yes	7	Abducens	Diplopia	Acetazolamide, Steroids, Immunoglobulin	2	Inflammatory virus-mediated immune response	15/20*
Baccarella et al.^Citation19	United States of America	M	6	Yes		Abducens	Diplopia	Steroids, Immunoglobulin	150	Inflammatory virus-mediated immune response	15/20*
Belghmaidi et al.^Citation11	Morocco	F	24	Yes		Oculomotor	Diplopia	Chloroquine, Azithromycin, Vitamin C, Zinc	2	Viral neuro invasion	15/16^
Ben-David et al.^Citation20	Israel	M	44	No		Abducens	Diplopia	Chloroquine, Azithromycin		Inflammatory virus-mediated immune response	13/16^
Bingol Kiziltunc et al.^Citation21	Turkey	M	28	No	10	Abducens	Diplopia	Nil	60		13/16^
Börü et al.^Citation22	Turkey	M	42	Yes	13	Abducens		Favipravir, Steroids	9	Inflammatory virus-mediated immune response	14/16^
Capponi et al.^Citation23	Italy	F	9	No	17	Abducens		Orthoptics	60	Inflammatory virus-mediated immune response	14/16^
Corrêa et al.^Citation24	Brazil	F	25	Yes	14	Abducens and Facial		Steroids	7	Viral neuro invasion	17/20*
de Oliveira et al.^Citation25	Brazil	F	2	No		Oculomotor	Diplopia	Nil	30		12/16^
Dinkin et al.^Citation9	United States of America	M	36	Yes		Oculomotor	Diplopia	Immunoglobulin, Hydroxychloroquine	3	Viral neuro invasion	15/20*
Dinkin et al.,^Citation9	United States of America	F	71	Yes		Abducens	Diplopia	Hydroxychloroquine	14	Viral neuro invasion	15/20*
Elenga et al.^Citation26	French Guiana	M	10	No		Oculomotor	Diplopia	Steroids, Calcium, Vitamin D	7		16/16^
Falcone et al.^Citation10	United States of America	M	32	Yes		Abducens	Diplopia	Orthoptics, Choroquine, Oxygen			13/16^
Faucher et al.^Citation12	France	M	21	Yes	17	Oculomotor			7	Vascular and thrombotic	10/16^
Fitzpatrick et al.^Citation27	United States of America	M	67		21	Oculomotor	Diplopia		30	Vascular and thrombotic	12/16^
Hernandez-Garcia et al.^Citation28	Spain	M	3		20	Abducens	Diplopia	Nil	26	Inflammatory virus-mediated immune response	10/16^
Kubota et al.^Citation29	Japan	M	25	Yes		Abducens	Diplopia	Nil	14	Inflammatory virus-mediated immune response	14/16^
Manolopoulos et al.^Citation30	Greece	F	41	No		Abducens		Acetaminophen	120	Vascular and thrombotic	12/16^
Medeiros et al.^Citation31	Brazil	M	48	Yes	18	Abducens	Diplopia	Orthoptics, Azithromycin, Ivermectin	240	Viral neuro invasion	12/16^
Oliveira et al.^Citation32	Brazil	M	69	Yes	11	Abducens	Diplopia			Inflammatory virus-mediated immune response	12/16^
Ordás et al.^Citation33	Spain	M	62	Yes	14	Trochlear	Diplopia and blurred vision	Steroids	30	Inflammatory virus-mediated immune response	13/16^
Piazza et al.^Citation34	Italy	M	47	Yes		Abducens		Steroids	30	Viral neuro invasion	14/16^
Prajwal et al.^Citation35	United States of America	F	46	Yes		Oculomotor	Diplopia and lid pain	Nil	9		13/16^
Soto Insuga et al.^Citation36	Spain	M	7	Yes	17	Abducens	Diplopia and blurred vision	Azithromycin, Steroids, Acetazolamide	2	Inflammatory virus-mediated immune response	14/16^

Patient demographics

The data of 25 patients were available from the 23 reports included in this systematic review (Table 1). There were 17 males (68%) and 8 females (32%) with an age range of 2–71 years (mean = 32, SD = 21.4, median = 32.7). Seven patients (28%) in this study were children aged 2–10 years old. All patients had a history of COVID-19 infection of which 16 (64%) were symptomatic. Seven patients (28%) were from the United States of America, 4 (16%) were from Brazil, 3 (12%) were from France, 2 (8%) from Turkey and the remaining patients were each from French Guiana, Greece, Israel, Japan, Morocco and the United Kingdom (Table 1).

Clinical presentation

Seventeen patients (68%) presented with a right abducens nerve palsy (CN VI), seven (28%) with an oculomotor palsy (CN III) and one (4%) with a trochlear nerve palsy (CN IV). All patients, except one, had a palsy isolated to a single cranial nerve. The most common ocular presentation was diplopia (n = 16, 64%), diplopia with blurred vision (n = 2, 8%) or diplopia with eyelid pain (n = 1, 4%). The time from onset of COVID-19 symptoms or a positive test and ocular symptoms ranged from 2 days to 21 days (mean = 14.9, SD = 4.2) and the time to improvement or resolution of ocular symptoms ranged from 2 days to 240 days (mean = 41.1, SD = 59.7).

Proposed mechanism

The proposed causal mechanisms of the neurogenic palsies extracted from the reports were broadly classified as vascular/thrombotic, viral neuro-invasion or an inflammatory virus-mediated immune response. The most postulated mechanism was inflammatory virus-mediated immune response (n = 10, 40%) and of these, 3 (12%) were children diagnosed with pediatric multi-system inflammatory syndrome. Viral neuro-invasion was proposed as a mechanism for six patients (24%) and a vascular/thrombotic cause was suggested for three patients (12%). The most frequent causal mechanism for patients with an abducens palsy was an inflammatory virus-mediated response (n = 9) whilst those with an oculomotor palsy either had a vascular/thrombotic (n = 2) or viral neuro-invasive (n = 2) mechanism reported (Table 1).

Discussion

COVID-19 respiratory manifestations are now well known, as are some of the common ocular presentations, namely conjunctivitis, dry eyes, foreign body sensation, redness, itching and epiphora.^3,4 The findings of this systematic review suggest that, although uncommon, COVID-19 is associated with sudden-onset neurogenic palsy.

Based on the findings of this review, patients with ocular neurogenic palsy associated with COVID-19 are likely to be male and present with diplopia secondary to an abducens nerve palsy within 15 days of the onset of the virus. The overall age of the patients in this study varied greatly, as did the time to improvement or resolution of the diplopia.

Whilst the proposed mechanism of development of a neurogenic palsy secondary to COVID-19 infection has not been established, based on the literature, we classified the cases into one of three broad categories: vascular/thrombotic cause, viral neuro-invasive event or an inflammatory virus-mediated immune response.^7,9,11,12 The underlying cause for most of the patients with oculomotor nerve palsy was postulated to be viral neuro-invasive or vascular/thrombotic. In contrast, patients presenting with an abducens nerve palsy were more likely to be classified as having an inflammatory cause. Despite the prevalence of a proposed inflammatory virus-mediated immune response in those with abducens nerve palsy, inferences as to mechanisms cannot be made. In particular, the mechanisms were documented for each case as a proposed causal factor based on clinical presentation and investigation. Further research to establish more definite underlying causal mechanisms is required.

Given that this systematic review only included case reports or case series, there are limitations with respect to what conclusions can be drawn. There was much inconsistency between the cases in terms of reporting of symptoms, duration, time to resolution and investigations. Also, proposed mechanisms as to the cause of neurogenic palsies were postulated by the published authors, and for the purposes of this current review, subsequently grouped into three categories. The robustness of these hypotheses has not been unequivocally established, and other mechanisms may need to be considered. However, despite the paucity of literature and the only recent human exposure to SARS-CoV-2, the link between COVID-19 and neurogenic palsies appears strong. Further studies investigating the onset and pathogenesis of neurogenic palsy secondary to COVID-19 infection are required.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.