Read on Twitter
I agree that #SARSCoV2 #COVID19 #coronavirus ocular exposure needs more research, but I'd hold off on assuming that this route is responsible for a substantial portion of new infections for a couple reasons.
(Disclaimer: this is not an argument against eye protection) https://twitter.com/eliowa/status/1299813623838146560
(Disclaimer: this is not an argument against eye protection) https://twitter.com/eliowa/status/1299813623838146560
1. There is conflicting information about ACE2 expression on the conjunctiva. (The retinal cells above aren't on the surface of the eye). This study shows there isn't, but only looked at 38 samples. ACE2 expression probably does vary from person to person. https://onlinelibrary.wiley.com/doi/full/10.1002/jmv.25981
1 (cont). This study shows there is ACE2 on the conjunctival surface, but only looked at 8 patients. The authors attributed the difference to a modified staining protocol. So it's still unclear if ACE2 expression on the ocular surface is common to all. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7300893/
2. ACE2 is not sufficient for SARS-CoV-2 cell entry. After ACE2 binding, the virus is taken up into a membrane--bound compartment called the endosome. Coronaviruses replicate in the cytoplasm outside the endosome, so they have to escape that compartment to replicate.
2 (cont). Spike protein has to be primed (clipped by proteases) after entry in order to fuse with the endosomal membrane and disgorge the viral genome into the cytoplasm where replication occurs. It's unclear whether the cofactors needed for priming are present in ocular cells.
2 (cont). At least one study shows that ACE2 transcripts (mRNA, which encode proteins) and one of these co-factors, TMPRSS2, are in the conjunctiva. However, the presence of a transcript doesn't guarantee the presence of the protein it encodes. https://www.nature.com/articles/s41591-020-0868-6
2 (cont). So it remains unknown whether the surface cells of the eye are indeed susceptible to infection, or how much exposure is required to cause an infection.
3. Dose might also be an issue. Ocular exposure is likely to be lower than exposure to aerosols or droplets. A person breathing will inhale and exhale liters of air, and can have a much higher exposure to respiratory particles containing virus.
3 (cont). Ocular exposure, on the other hand, would be by passive exposure to aerosols, a direct hit from droplets with an unfortunate trajectory, or from rubbing one's eye with virus on the hand. These are likely to be lower dose exposures than from breathing.
4. We do know that animals can be experimentally infected with SARS-CoV-2 via the ocular route. However, this study used 10e6 TCID50s (1 million infectious units). It's hard to say if that dose is remotely close to a typical exposure in the real world.
https://www.biorxiv.org/content/10.1101/2020.03.13.990036v1.full.pdf
https://www.biorxiv.org/content/10.1101/2020.03.13.990036v1.full.pdf
4 (cont). We also don't know how the minimum infectious dose via ocular infection differs from the minimum infectious dose via other routes, in humans or animals. I'd expect it to be much higher via ocular vs intranasal, intratracheal, intrabronchial, or aerosol routes in animals
5. While we don't know minimum infectious dose by any route, we do know that different routes have minimum infectious doses. This is determined by the prevalence and distribution of susceptible cells as well as the unique immune environments of those tissues.
5 (cont). Eyes are very special in terms of their immune environment. Because there are some cells in the eye that are highly specialized and can't regenerate the way many other cells can, it is referred to as an "immunologically privileged" site.
5 (cont). Immune privilege means that killing infected cells is not an option, so there are other specialized immune responses to deter infection that don't require cytolytic (cell killing) means of control. There are a whole host of antiviral proteins present in tears.
5 (cont). This includes antiviral molecules that block replication, surfactants that disrupt viral envelopes, and signaling molecules that prevent virus from spreading from one cell to another. As a result, eyes are more resistant to infection.
6. Some (not Dr. Perencevich AFAIK) have cited conjunctivitis as evidence that ocular infection occurs. It's not. Conjunctivitis is common in respiratory infections and can occur from the virus traveling to the eyes via the naso-lacrimal system as well as the other way around.
While barriers to ocular infection can be overcome by a high enough dose, we just don't have enough information right now to assume that is common in real-world situations, or that it is as much of a risk as exposure by inhalation.
Face shields and eye protection are sensible precautions, but they don't replace other measures such as masks. They offer no protection against inhalation exposure, especially by smaller particles that are not propelled through the air.
Again, this is not an argument against face shields. It is, however, a caution that face shields are not alternatives to masks or other risk reduction measures. We should not ignore the potential for ocular transmission but we should evaluate it more rigorously.
