INTRODUCTION

Coronavirus disease 19 (COVID-19) is caused by the Severe Acute Respiratory Syndrome coronavirus type 2 (SARS-CoV-2), previously named 2019 novel coronavirus (2019-nCoV)⁽¹⁾. The ongoing SARS-CoV-2 pandemic has been linked to ophthalmology since its beginnings. On December 30, 2019, Dr. Li Wenliang, a Chinese ophthalmologist, warned his colleagues through the social network WeChat about a SARS-like outbreak in Wuhan and its possible link to a local market⁽²⁾. On January 30, 2020, the World Health Organization (WHO) confirmed the outbreak as a public health emergency of international interest⁽³⁾, and on March 11, 2020 it was declared a pandemic⁽⁴⁾.

Coronaviruses (CoVs) are known to cause infections in humans and other mammals^(5,6), they gained public attention during the Severe Acute Respiratory Syndrome (SARS) outbreak in East Asia in 2003 and the spread of Middle Eastern Respiratory Syndrome (MERS) in 2012^(5,7). SARS-CoV-2 is a single-stranded, positive-sense, enveloped RNA virus⁽¹⁾. The main method of transmission is human-to-human through direct contact and droplets; transmission from asymptomatic carriers has also been reported⁽⁸⁾. Viral stability in aerosols on different surfaces has been demonstrated⁽⁹⁾, supporting evidence on indirect viral acquisition from fomites⁽¹⁰⁾, through the mucous membranes of the mouth, nose and eyes⁽¹¹⁾. At present, conjunctival transmission of CoVs has not been confirmed and remains controversial.

Out of the seven types of human coronaviruses (HCoV), HCoV-NL63 is the only one that has been confirmed to produce ocular disease, specifically conjunctivitis⁽¹²⁾, although the pathogenic mechanism of ocular infection has not been elucidated⁽²⁾. Ocular symptoms from SARS-CoV and MERS-CoV, which produce similar respiratory manifestations to SARS-CoV-2, have not been reported⁽²⁾. On the other hand, detection of SARS-CoV RNA in tears was confirmed in three out of 36 patients with SARS during the 2003 outbreak; further studies failed to confirm these results^(13,14).

Reports on SARS-CoV-2 conjunctival transmission, viral shedding through tears, and ocular manifestations in COVID-19 patients have emerged^(15-17). As the current pandemic progresses, so does the need for reliable information to help generate evidence-based recommendations on best practices. Incentives to rapidly produce scientific reports and positive results lead to the risk of disseminating poorly supported evidence and overloading the public with uncertain or even false information^(18,19). In this review we aim to identify the available information regarding COVID-19 and eye care and to classify it according to four main topics of interest: evidence of the virus in tears and the ocular surface, infection via the conjunctival route, ocular manifestations, and best practice recommendations.

METHODS

RESULTS

Search results

A total of 1018 records were retrieved from databases and ten others were found through referencing. After screening by title and/or abstract, 918 records were excluded. Duplicates and grey literature were removed, leaving 80 records which met the eligibility criteria. Correspondence and editorials accounted for 43 records (54%) which were reassessed and excluded.

A full text assessment for eligibility was carried out for the remaining documents. Further analysis led to the exclusion of three publications which were deemed opinion pieces, four due to incoherent translation; and three others, which focused on topics outside the scope of this review. One case report was excluded as it did not fulfill the CARE checklist guidelines. Finally, 26 records were included in the qualitative synthesis: 6 literature reviews, 10 case series or cross-sectional studies, 4 case reports, and 6 intervention descriptions (miscellaneous) (Figure 1).

Figure 1 Methodology flowchart. Adapted from the Preferred Information Elements for Systematic Testing and Meta-Analyses (PRISMA)23. 

DISCUSSION

We found that current recommendations regarding COVID-19 and ophthalmology are based on levels of evidence 4 and 5 according to the Oxford CBME methodology grading system. The amount of research conducted to date is limited; the nature of the disease and the scarcity of cases with ocular involvement pose challenging circumstances for research efforts. As the pandemic evolves so will the need for further data to bolster our understanding on COVID-19, its implications for eye care and the outcomes of the implemented strategies.

In a small percentage of patients, SARS-CoV-2 RNA has been isolated in the tear film. In the published literature, positive findings were reported in 0 to 7.14% of subjects across different studies; Sarma et al. found that the virus was present in 1.95% (95% C.I. 0.74 to 4.11) of samples⁽²⁸⁾. Possible explanations for a negative finding of viral particles on the ocular surface have been set forth: sensitivity of the tests; time of sampling collection; antimicrobial mechanisms of the conjunctiva; collection techniques; and washing of viral particles by tearing and passage to the nasopharynx through the lacrimal duct^(6,48).

The reviewed publications showed wide heterogeneity in their methodology. First, the time of sample collection, counted from the onset of any symptom, varied on average from 2 to 18 days, and in some studies was not even specified. Second, the number of samples taken was not consistent across the studies and even within them; some, such as case reports, performed only one test, and others up to four. Lastly, the methods for sample collection and processing included conjunctival swabbing and RT-PCR in the majority of the studies, while one collected tears using Schirmer strips and two also performed viral isolation through cell culture. Therefore, pooling of the data was not deemed appropriate. Until further evidence is available, we cannot rule out the possibility of viral particles being present in tears and conjunctiva; therefore, precautionary measures should be insisted upon.

Possible conjunctival transmission is mainly based on anecdotal reports of HCW who did not use eye protection⁽⁴⁰⁾, early conjunctival congestion symptoms reported in some patients, or the presence of viral RNA in the tear film^(15,16). Several hypothetical transmission mechanisms have been proposed. First, the virus adheres to ACE2 receptors found on the conjunctival and corneal epithelia^(49,50). Second, the nasolacrimal duct would serve as a pathway from the conjunctiva to the upper respiratory tract where the virus can infect the host⁽⁵¹⁾. However, some authors believe that this is not enough evidence to confirm the conjunctival route of transmission and suggest the following counter-arguments: a) the presence of the ACE2 protein in the conjunctiva is low compared to the lung; b) lactoferrin and secretory IgA present in the tears could eliminate the virus; and c) the presence of the virus in tears could be explained by fomites transmitted to the conjunctiva via splashed droplets or by direct contact with a contaminated hand⁽⁵¹⁾. Qiao et al. report that the overall incidence of COVID-19 among eye professionals across 10 hospitals was 2.52% (95% CI: 1.68-3.63%); the incidence of the disease was similar in ophthalmologists as that of general practitioners⁽⁵²⁾. Although there is currently no confirmed conjunctival transmission route, the authors agree with the WHO and other organizations´ recommendations that PPE must include eye protection such as goggles or face shields^(2,24).

A wide range of non-specific ocular manifestations has been reported for COVID-19; at this time, a characteristic presentation has not been clearly determined. The most common ocular manifestations were conjunctival hyperemia and watery discharge. This presentation varies, as can be observed with the report of Cheema et al. on a patient presenting with keratoconjunctivitis⁽³⁹⁾. The methodology of the reports also varies widely, as can be seen in the approaches taken by two of the studies. Wu et al. examined hospitalized COVID-19 patients who were more severely ill, including intubated patients, finding that patients with ocular manifestations presented more severe systemic disease or abnormal findings on blood tests⁽¹⁶⁾. Chen et al. obtained data via telephone, face-to-face surveys, or a smartphone application⁽³³⁾. If statistical measures to overcome the underlying studies limitations and the appropriateness of a pooled analysis are accepted, then Sarma et al. report that conjunctivitis/red eye is featured in 3.17% (95%CI 1.16 to 6.13) of patients⁽²⁸⁾. The normal prevalence of dry eye and allergic conjunctivitis could explain some of the reported ocular symptoms; other explanations may be related to poor hygiene or face mask misuse. Overall, severe eye manifestations have not been reported, and more specific observations will be needed to establish a particular set of findings.

Because of a lack of interventional studies, recommendations in current publications are based on lower-level evidence. Most derive from the experience gained and the general strategies implemented during the SARS outbreak in 2002-2004^(41,53). Administrative, environmental and PPE measures should be implemented, and attention should be given to protecting both HCW and patients. There is a need for studies that test or certify the effectiveness of these intervention measures during the current SARS-CoV-2 pandemic; however, erring on the side of safety is currently necessary. As better evidence continues to accumulate, it will be important to update these measures and adapt to a rapidly evolving scenario.

It was of particular interest to the authors that a considerable proportion of the analyzed literature referenced an anecdotal report about a Chinese respiratory expert who contracted COVID-19 despite having worn appropriate PPE except eye protection^(2,47,54). This report was cited in 40% of the screened publications and might have been the initial source for many of the recommendations on ocular protection⁽⁴⁷⁾. This phenomenon illustrates how quickly information with low levels of evidence can be widely disseminated.

This is the first review on the level of evidence for ophthalmology recommendations and COVID-19 conducted to date. It includes an independent full text review performed by five reviewers. This study was limited to SARS-CoV-2; therefore, data obtained from previous coronavirus epidemics might have been outside of its purview. Some of that information may be as important as what has been reviewed here; the article by Seah et al. on the evidence of ocular involvement in coronavirus cases might have served to counteract this shortcoming. An additional limitation is the dependence on third-party translation for Chinese publications; in order to address this limitation, translations were performed on multiple platforms, and an additional reviewer was added to these papers. Publication bias was approached by searching through grey literature, editorials, opinion pieces, pre-published works, and non-peer-reviewed articles alongside traditional publications, without language restrictions.

COVID-19 is a novel disease that has caused a pandemic unlike any we have experienced in modern medicine, not only because of the characteristics of the disease, but because of the speed of the spread of information. The pandemic has only begun to be studied properly; therefore, the scarcity of medical appraisals, randomized controlled trials and case control studies is not surprising. In this structured review we classified the available evidence and recommendations relating to ophthalmology during the COVID-19 pandemic. Overall, the level of available evidence for current recommendations is rising.

Currently, there is not sufficient evidence to rule out the possibility of viral particles being present in tears and conjunctiva. A few hypothetical mechanisms have been proposed suggesting that the conjunctiva may act as an entry route for SARS-CoV-2. With the varying degrees of evidence supporting or refuting it, conjunctival transmission remains controversial. Ocular manifestations are not common in COVID-19 patients; evidence suggests that they can resemble a viral infection of the ocular surface, with hyperemia and watery discharge as the cardinal signs. Most of the literature published to date consists of anecdotal reports, editorials, and opinion pieces with a high level of cross-referencing. The documents that currently contribute to expanding the knowledge of ocular involvement in COVID-19 are ranked as having low levels of evidence. Most recommendations are based on the strategies implemented in Asian countries during previous CoVs outbreaks, many of them are likely to prevail and set new standards of preventive measures in health systems. During the evolution of this worldwide phenomenon, reliable information will be crucial to elucidate recommendations to mitigate the propagation of SARS-CoV-2. As new studies and cases are reported, it will be fundamental to evaluate their level of evidence to correctly assess the recommendations and adapt them to the local circumstances.