Objectives
This study aims to estimate the prevalence and longevity of detectable SARS-CoV-2 antibodies as well as T and B memory cells during infection with SARS-CoV-2 and after recovery. In addition, prevalence of COVID-19 reinfection, and the preventive efficacy of previous infection with SARS-CoV-2 were investigated.
Methods and analyses
A synthesis of existing research was conducted. The Cochrane Library for COVID-19 resources, the China Academic Journals Full Text Database, PubMed, and Scopus as well as preprint servers were searched for studies conducted between 1 January 2020 to 1 April 2021. We included studies with the relevant outcomes of interest. All included studies were assessed for methodological quality and pooled estimates of relevant outcomes were obtained in a meta-analysis using a bias adjusted synthesis method. Proportions were synthesized with the Freeman-Tukey double arcsine transformation and binary outcomes using the odds ratio (OR). Heterogeneity between included studies was assessed using the I2 and Cochrans Q statistics and publication bias was assessed using Doi plots.
Results
Fifty-four studies, from 18 countries, with a total of 12 011 447 individuals, followed up to 8 months after recovery were included. At 6-8 months after recovery, the prevalence of SARS-CoV-2 specific immunological memory remained high; IgG 90.4% (95%CI 72.2 to 99.9, I2=89.0%, 5 studies), CD4+ 91.7% (95%CI 78.2 to 97.1, one study), and memory B cells 80.6% (95%CI 65.0 to 90.2, one study) and the pooled prevalence of reinfection was 0.2% (95%CI 0.0 to 0.7, I2 = 98.8, 9 studies). Individuals previously infected with SARS-CoV-2 had an 81% reduction in odds of a reinfection (OR 0.19, 95% CI 0.1 to 0.3, I2 = 90.5%, 5 studies).
Conclusion
Around 90% of people previously infected with SARS-CoV-2 had evidence of immunological memory to SARS-CoV-2, which was sustained for at least 6-8 months after recovery, and had a low risk of reinfection.