Authors' Reply

• Seroprevalence
• COVID-19

We agree with Rodríguez-Espinosa et al.¹ that earlier and stricter use of personal protective equipment by patients within the dialysis units may have reduced the high prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection within our center. Specific discussion around the risk of transmission within our center’s individual dialysis units has been addressed in an earlier article by Corbett et al.² We would, however, like to highlight that even with utilization of the same SARS-CoV-2 screening and management protocol, we have found a significant difference in prevalence rates between dialysis units within our center, which suggests a more complex relationship between infection rates in the community, patient characteristics, and clinical protocols.

We concur with the statement that regular screening of asymptomatic patients by RT-PCR testing would have detected more cases; however, it is also worth noting that false-negative results may occur even in symptomatic patients.³ The benefit of RT-PCR testing in asymptomatic patients is not known, but such practice does rely on a substantial resource investment, which may not be available at the time of infection surges. We agree that waiting for patients to become RT-PCR negative prior to deisolation may be a good practice. However, an ongoing challenge for the renal community is how to differentiate between viral RNA detection and infectivity.⁴ Patients may have prolonged carrier status, and there is a risk, especially in patients receiving in-center hemodialysis, that isolation facilities may become overburdened.

Related to our serologic study, as advances are made in the serologic testing for SARS-CoV-2, we believe that it is important to consider the type of assay, Ig isotype, and the target antigen being used to quantify seroprevalence.⁵ Rodríguez-Espinosa et al.¹ do not provide the details of the assay they used to define seroprevalence within their unit, and it maybe that the differences between the two cohorts are even greater. We do believe that serologic methods need to be uniform in order to compare more accurately seroprevalence in different populations.

• Copyright © 2020 by the American Society of Nephrology