Potential drawbacks of SARS-CoV-2 seroprevalence surveys.

Sir,

I read with interest the recent meta-analysis of Galanis et al. [1] who provided an estimation of the seroprevalence of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) antibodies in healthcare workers. Although seroprevalence surveys can be regarded as important tools for exploring the burden of disease and progression of herd immunity, especially in front-line healthcare staff, there are important drawbacks and limitations that need to be considered when analysing their pooled outcomes.

Reliable evidence has been provided that seroconversion varies widely in patients with SARS-CoV-2 infection, with up to one-third of all asymptomatic subjects failing to mount a detectable humoral response with anti-SARS-CoV-2 immunoglobulin G (IgG) and thus remaining completely underdiagnosed [2]. The progressive, time-dependent decline of anti-SARS-CoV-2 antibodies is another important issue, whereby it could be proven that >50% of anti-SARS-CoV-2 IgG-seropositive subjects may become seronegative as early as 2 months after initial molecular diagnosis [3]. Since the first peak of the SARS-CoV-2 outbreak was reached in most countries during the first months of 2020 [4], it is conceivable that many infected individuals may have lost any detectable humoral response later in the same year. Last but not least, the diagnostic sensitivity of anti-SARS-CoV-2 antibody testing is extremely variable, with consensus-positive interpretation that is often unsatisfactory, and which may hence be associated with a non-negligible risk of obtaining false-negative test results [5].

Together, these three aspects suggest that serologic evidence of previous SARS-CoV-2 infection may often be unreliable, so actual calculated values may provide a consistently underestimated figure of the real spread of SARS-CoV-2 infection within healthcare facilities.

Conflict of interest statement

None declared.

Funding sources

None.