That lede encapsulates the underlying subtext of the most recent New England Journal of Medicine publication regarding the spread of coronavirus. This article is very simple, and – importantly – very limited.
These authors from Columbia University in New York, New York, simply supply a short research letter describing their experience screening pregnant patients for SARS-CoV-2 infection at their obstetrical intake prior to delivery. Between March 22nd and April 4th, there were 215 women who delivered at New York-Presbyterian and Columbia University Irving. Four were febrile and symptomatic at admission and all tested positive. However, an additional 29 patients who were not experiencing symptoms of Covid-19 also tested positive by nasophargyneal swab and RT-PCR. Thus, the big freak out: 15% of this convenience sample were asymptomatic at time of diagnosis.
Many important limitations may preclude its generalizability to clinical practice and public health policy:
- New York was in the midst of a substantial outbreak during sampling, with certainly higher background prevalence than most communities.
- Full follow-up was not completed to determine whether the “asymptomatic” patients were truly “asymptomatic” or actually “presymptomatic”. Truly asymptomatic infection would imply vastly larger pools of infected and/or recovered citizens. This would drive down case-fatality rates and contribute to “herd immunity”. Presymptomatic infection is distinctly different, however, as these patients would be reflected in our current methods of (under)counting cases.
- Pregnant women are not representative of the general populace, neither in their overall immune system physiology, nor with respect to the frequency with which they leave the home and access healthcare systems during a high-risk period.
- It’s simply a convenience sample of 215 patients, diminishing the precision of their measurements.
The big takeaway here mostly has implications for testing. In areas with high prevalence of circulating infection – particularly circulating, undetected infection – routine screening at hospital admission should probably be performed. Right now, that is basically everywhere – so, in the interests in protecting healthcare workers and staff, all patients being admitted to the hospital ought to be screened, if possible. The other implication is reinforcement of our larger testing needs, and the already-known risks of presymptomatic viral shedding. Contact tracing and testing will be critical to stamping out local flare ups of disease – and we will need millions more tests than we currently perform.
“Universal Screening for SARS-CoV-2 in Women Admitted for Delivery”
https://www.nejm.org/doi/full/10.1056/NEJMc2009316