April 7, 2021
Reviewed by Razan El Ramahi, MBBS
Researchers at Oxford University Hospitals in the United Kingdom attempted to answer the above questions in a recent study published in the New England Journal of Medicine. A cohort of health care workers (HCWs) were categorized as either seropositive or seronegative by SARS-CoV-2 anti-spike (primary analysis) and/or anti-nucleocapsid IgG antibody test results. The study prospectively evaluated the incidence of positive SARS-CoV-2 polymerase chain reaction (PCR) in the cohort and ran from April 23, 2020, to November 30, 2020. In addition to PCR testing for symptomatic HCWs, PCR screening was offered every 2 weeks, and serologic testing was offered every 2 months. The HCWs attended voluntary PCR screening every 10 to 13 weeks on average, and seronegative HCWs presented more frequently, but this difference was not shown to significantly change the study results when sensitivity analysis was performed.
Among 12,541 HCWs who were tested, 11,364 (90.6%) were seronegative and 1,177 (9.4%) were seropositive. Among the 11,364 anti-spike seronegative HCWs, 223 tested positive for SARS-CoV-2 by PCR including 123 who were symptomatic. Only two HCWs tested positive for SARS-CoV-2 infection among the anti-spike seropositive HCWs, and both were asymptomatic. The incidence of PCR positivity was 1.09 per 10,000 days at risk in the anti-spike seronegative HCWs compared to 0.13 per 10,000 days at risk in the seropositive group. Results were similar when looking at anti-nucleocapsid IgG antibody results.
The authors concluded that among HCWs, the incidence of SARS-CoV-2 reinfection diagnosed by positive PCR was markedly decreased over 6 months following results of positive anti-spike or anti-nucleocapsid IgG antibodies. Whether the protection demonstrated in the study is attributed to measured antibody levels versus other unmeasured immune responses is not certain. The effect of positive antibody levels and other immune responses on protection beyond the study’s follow-up period requires more investigation. Furthermore, additional studies are needed in other populations since this cohort included presumably mostly healthy subjects who were younger than 65 years.