Low risk of SARS-CoV-2 transmission through the ocular surface.

Editor,

The coronavirus disease 2019 (COVID‐19) caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) has become a pandemic. It is worth noting that some infected medical staff presented first with ocular symptoms (e.g. conjunctival congestion; Lu et al. 2020), suggesting a potential ocular transmission route for SARS‐CoV‐2. To date, several ophthalmologists have died from COVID‐19 following an unclear route of infection at work, which further highlights the importance of precautionary practices in clinical work during the pandemic. To our knowledge, the presence of SARS‐CoV‐2 in the tears or conjunctival epithelium remains controversial (Seah et al. 2020, Wu et al. 2020). We believe there are several issues concerning the transmission capacity of SARS‐CoV‐2 via the ocular surface that are worth discussing.

First, SARS‐CoV‐2 may not be present in all tear samples from COVID‐19 patients. Several studies reported patients with ocular symptoms where SARS‐CoV‐2 was detected in the tears. However, the detection rate is very low, ranging from 3.33% to 5.26% (Wu et al. 2020, Xia et al. 2020). We have performed conjunctival and nasopharyngeal swab tests to detect the presence of SARS‐CoV‐2 in moderate, severe and critical hospitalized patients with COVID‐19. Of a total of 114 patients, 90 patients (79%) had positive nasopharyngeal swab tests. However, conjunctival swab samples from all patients were negative for SARS‐CoV‐2 nucleic acid (Table 1). We assumed that exposure of respiratory tract tissue to the virus remains the dominant mode of infection, although other transmission routes between humans may exist. However, as the virus continues to mutate, it is possible that in the subsequent epidemic phase, there may be more patients presenting mainly with ocular symptoms.

Table 1

Laboratory characteristics of 114 COVID‐19 patients.

	Patients (n = 114)	Moderate cases (n = 19)	Severe cases (n = 59)	Critical cases (n = 36)	P value
SARS‐CoV‐2 quantitative RT‐PCR test
Nasopharyngeal swab (positive cases), no. (%)	90 (79%)	16 (84%)	45 (76%)	29 (81%)
Conjunctival swab (positive cases), no.	0	0	0	0
Blood routine
White blood cell count (×109 cell per l), mean (SD)	7.4 (3.9)	5.5 (2.6)	6.6 (2.6)	9.8 (4.9)	0.0007
Neutrophils (×109 cell per l), mean (SD)	5.9 (3.9)	3.5 (2.1)	4.9 (2.6)	8.7 (4.9)	<0.0001
Lymphocytes (×109 cell per l), mean (SD)	1.1 (1.2)	1.5 (0.6)	1.3 (1.6)	0.6 (0.3)	0.0017
Haemoglobin (g/l), mean (SD)	127.8 (17.2)	132.6 (14.7)	124.8 (17.9)	130.0 (16.8)	0.16
Platelets (×109 cell per l), mean (SD)	225.3 (17.2)	225.5 (92.4)	251.6 (101.3)	182.1 (83.3)	0.0005
Coagulation function
Prothrombin time (s), mean (SD)	14.6 (1.5)	13.6 (0.6)	14.4 (1.1)	15.5 (1.9)	0.0027
D‐dimer (μg/l), mean (SD)	5.0 (7.5)	1.5 (4.7)	2.6 (4.6)	10.7 (9.2)	<0.0001
Blood biochemistry
ALT (U/l), mean (SD)	29.3 (18.9)	20.6 (10.8)	30.3 (21.1)	32.3 (17.4)	0.62
AST (U/l), mean (SD)	34.6 (20.3)	21.5 (4.6)	33.0 (19.8)	44.3 (21.9)	0.014
Albumin (g/l), mean (SD)	34.6 (7.8)	41.4 (8.5)	34.9 (6.6)	30.5 (6.5)	0.0021
Serum creatinine (μmol/l), mean (SD)	83.6 (101.3)	67.2 (29.7)	75.9 (29.5)	104.9 (174.6)	0.33
Glucose (mmol/l)	7.8 (3.4)	6.7 (2.7)	7.2 (2.3)	9.3 (4.7)	0.016
Infection‐related biomarkers
Erythrocyte sedimentation rate (mm/h), mean (SD)	41.7 (30.9)	16.4 (12.8)	47.7 (28.0)	45.4 (35.7)	0.74
C‐reactive protein (mg/l), mean (SD)	72.9 (77.9)	15.7 (22.7)	62.6 (71.3)	120.2 (81.3)	0.0008
Serum ferritin (ng/ml), mean (SD)	987.4 (1009.6)	339.3 (208.2)	885.3 (866.8)	1520.3 (1252.6)	0.0098
Interleukin‐6 (pg/ml), mean (SD)	39.8 (68.6)	6.4 (6.3)	31.5 (72.2)	76.2 (67.2)	0.0031
Tumor necrosis factor α, mean (SD)	9.7 (4.6)	7.0 (2.6)	9.5 (4.0)	11.8 (5.7)	0.038

ALT = alanine aminotransferase; AST = aspartate aminotransferase; SD = standard deviation.

Statistical analysis was performed between the severe and critical cases, using the unpaired t‐test with Welch’s correction.

Second, the virus may only be present in the conjunctival sac or tear fluid for a very short period of time, and at the time of sample collection, the virus has already entered the respiratory tract or been eliminated by treatment. Recently, Chen et al. reported a patient who presented with ocular symptoms 13 days after onset of the disease. The ocular symptoms lasted for approximately 5 days, and the conjunctival swab was positive during this period (Chen et al. 2020). However, we noticed that the viral load is significantly lower in conjunctival swabs than in nasopharyngeal swabs, suggesting the respiratory tract is the origin of the virus in this case.

Finally, our eyelids and eyelashes are natural barriers and prevent the eyeball being affected by dust and virus particles carried by droplets. The lacrimal drainage system and local immune system can also provide protection for the eye. Thus, for suspected COVID‐19 patients with ocular symptoms (e.g. conjunctival congestion) as the initial symptom, tests for viral nucleic acids in the conjunctival sac or tears during the early stages of disease are very likely to produce a positive result. For the majority of suspected COVID‐19 patients, the likelihood of ocular transmission is assumed to be low.

Considering current protective equipment (most people wear masks and do not wear goggles), among suspected individuals who are at high risk of infection through the ocular surface, performing conjunctival swabs simultaneously with nasopharyngeal swabs may be an option to detect potential infection in advance. Ophthalmologists need to remain vigilant during the COVID‐19 pandemic, and further studies are needed to determine whether SARS‐CoV‐2 can be transmitted from person to person via the conjunctival route. For the general populations, based on the current evidences, masks or goggles are not obligatory, which may exacerbate the strain on the limited medical resources.