Reply

REPLY:

We thank Min et al for their comments regarding our study. To reiterate our conclusion, our study suggested that the risk of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission through tears is low, not impossible. This was a conclusion made based on the analysis of 17 patients with coronavirus disease-2019 (COVID-19) and the available literature at the time of article submission on March 19, 2020. Majority of the limitations raised by Min et al have been acknowledged in our article.

First, Min et al probed the validity of the Schirmer strip collection method and the likely presence of viral material in tear samples. These limitations have already been acknowledged in the article. However, we point out that, apart from herpes simplex virus-1, the Schirmer strip has also been used to detect other herpes-family viruses including Epstein-Barr virus (types 1 and 2) and non-herpes viruses (e.g., adenovirus).^{1 , 2} They have also been used routinely for the study of proteins in tear films. To date, there is no known scientific rationale to indicate that these strips are not able to collect coronaviruses. Regarding the presence of viral material in tear samples, we have mentioned that, if the ocular surface tissue was infected, lysis of these cells (a part of the viral replication cycle), would have led to the release of viral particles or genetic material.

Second, in our article, we had already acknowledged the concerns of Min et al regarding the sensitivity of reverse transcriptase polymerase chain reaction to detect small quantities of SARS-CoV-2 RNA. To further improve the sensitivity, collected samples were also used to inoculate Vero-E6 cells to observe for cytopathic effect over a 4-day duration. The observation of cytopathic effect (which indicates infection) along with reverse transcriptase polymerase chain reaction would likely detect the presence of any SARS-CoV-2.

Finally, Figure 1 (in the original article) shows the full testing schedule for both nasopharyngeal and tear samples. As Min et al stated, it was difficult recruiting patients in early disease of <3 days. Only 2 tear samples were collected during this time period. We have previously acknowledged this in our article and explained that most patients presented to the hospital a couple of days after developing symptoms.

There have been multiple published case studies with conjunctival samples testing positive for SARS-CoV-2. Furthermore, ex vivo studies have also shown the ability of SARS-CoV-2 to infect conjunctival cells.³ However, to our knowledge, in the largest case series as of May 18, 2020, by Zhou et al, only 3 of 121 recruited COVID-19 patients (2.5%) had conjunctival samples that tested positive for SARS-CoV-2 RNA.⁴ Findings from other small case series were also similar.⁵ The nature of research in pandemic settings is that conclusions can change as more data are gathered. Early data can provide a suggestion of the possible implications of a novel pathogen like SARS-CoV-2. Unless there is a large study showing a large proportion of COVID-19 patients with positive ocular samples, respiratory droplet transmission should still be the primary concern of ophthalmologists owing to the close proximity to the patient during physical examination.