Interpretation
Other studies have examined whether SARS-CoV-2 is detectable in vaginal swabs. In contrast to our study, the majority of studies report negative vaginal swabs for SARS-CoV-2 (1, 16-19). However, there are reports of SARS-CoV-2 detected in vaginal swabs of both pregnant (n=3), reproductive-aged (n=1), and post-menopausal (n=1) women, indicating that transmission during vaginal delivery might be possible, although the risk is likely to be low (3-6). In accordance with our findings, other studies have also found positive placental swabs, indicating a viral spread that is potentially either an ascending infection or infection through the bloodstream (n=5) (3, 4, 6, 20). Other studies have demonstrated SARS-CoV-2 in neonatal swabs of children born to women with symptoms of COVID-19 (16, 17, 21, 22), which we did not find.
There are several potential sources for viral presence in the vagina. The coronavirus binds to target cells through angiotensin-converting enzyme 2 (ACE2) receptors, which are upregulated in vaginal epithelium during pregnancy, making it possible for the virus to bind (23). Alternatively, it is possible that SARS-CoV-2 is detected because of exudation from the bloodstream rather than release from the epithelium itself. Detection of SARS-CoV-2 in the vagina could also be due to fecal or seminal contamination (15, 24).
Our study indicates that the neonate is not protected in the acute phase of maternal COVID-19 (between 0-16 days after a positive maternal pharyngeal swab) as there were no SARS-CoV-2 antibodies in cord blood even though the mother had produced antibodies. However, after 16 days 94.1% of offspring had antibodies in cord blood. Other studies have assessed the timespan between maternal SARS-CoV-2 diagnosis and the presence of antibodies in cord blood, and a study found IgG in 11 cord blood samples and both IgG and IgM in one case of 31 cases in total. In one case IgG was found as early as one day after maternal SARS-CoV-2 infection (3). However, IgG is usually not detected before 1-2 weeks after acute viral infection and the mother may therefore have been infected several days before diagnosis. Another study found that 11 infants of 83 seropositive mothers (13%) did not have antibodies in cord blood after a median time of six days (interquartile range 0-12 days) from diagnosis to delivery (25). This time span is shorter than we report with a range of 0-16 days. In accordance with our study, they found a positive correlation between the levels of maternal and cord-blood IgG (25).
The adverse fetal outcomes of participant 5 and 6 in our study are similar to a confirmed case of transplacental transmission of SARS-CoV-2 from Sweden, where a pregnant woman with a three-day history of COVID-19 presented with reduced fetal movements at GA 34w4d (7). After an emergency caesarean section due to a pathological CTG the mother and child were separated. Placental pathology was similar to the findings in our study as well as other studies (4, 7, 26-28). Authors concluded that the neonate had suffered from transient asphyxia attributed to intrauterine hypoxia secondary to placental dysfunction. The case from Sweden (7) and others (4), as well as participants 5 and 6 in our study, suggests that reduced fetal movements during COVID-19 should be handled with aggravated concern.
In our study, two cases did not have antibodies in neither maternal nor umbilical cord blood (participants 25 and 26). Participant 25 had a low CT value, but a negative pharyngeal swab four days later, while participant 26 had no symptoms, a high CT value, and a negative pharyngeal swab the day after testing positive. Possible explanations could be either a false positive pharyngeal swab, a false negative antibody tests or that the women did not have an antibody response. In three other cases, antibodies were not detected in cord blood of seropositive mothers, which in two of the cases was possibly due to a short timespan from SARS-CoV-2 infection to delivery (8 and 16 days, respectively).
In large cohort studies, the overall risk of severe neonatal outcomes as well as neonatal infection related to maternal COVID-19 seems low (29). However, our findings indicate that vaginal testing for SARS-CoV-2 in women with COVID-19 within eight days before going into labour could be considered. In case of a positive vaginal swab, mode of delivery should be discussed with the woman as passive immunity of the neonate is not guaranteed.