Seroprevalence and longevity of SARS-CoV-2 nucleocapsid antigen-IgG among health care workers in a large COVID-19 public hospital in Saudi Arabia: A prospective cohort study.

Abstract

Seroprevalence of SARS-CoV-2 IgG among health care workers (HCWs) is crucial to inform infection control programs. Conflicting reports have emerged on the longevity of SARS-CoV-2 IgG. Our objective is to describe the prevalence of SARS-CoV-2 IgG in HCWs and perform 8 months longitudinal follow-up (FU) to assess the duration of detectable IgG. In addition, we aim to explore the risk factors associated with positive SARS-CoV-2 IgG. The study was conducted at a large COVID-19 public hospital in Riyadh, Saudi Arabia. All HCWs were recruited by social media platform. The SARS-CoV-2 IgG assay against SARS-CoV-2 nucleocapsid antigen was used. Multivariable logistic regression was used to examine association between IgG seropositive status and clinical and epidemiological factors. A total of 2528 (33% of the 7737 eligible HCWs) participated in the survey and 2523 underwent baseline serological testing in June 2020. The largest occupation groups sampled were nurses [n = 1351(18%)], physicians [n = 456 (6%)], administrators [n = 277 (3.6%)], allied HCWs [n = 205(3%)], pharmacists [n = 95(1.2%)], respiratory therapists [n = 40(0.5%)], infection control staff [n = 21(0.27%], and others [n = 83 (1%)]. The total cohort median age was 36 (31-43) years and 66.3% were females. 273 were IgG seropositive at baseline with a seroprevalence of 10.8% 95% CI (9.6%-12.1%). 165/185 and 44/112 were persistently IgG positive, at 2-3 months and 6 months FU respectively. The median (25th- 75th percentile) IgG level at the 3 different time points was 5.86 (3.57-7.04), 3.91 (2.46-5.38), 2.52 (1.80-3.99) respectively. Respiratory therapists OR 2.38, (P = 0.035), and those with hypertension OR = 1.86, (P = 0.009) were more likely to be seropositive. A high proportion of seropositive staff had prior symptoms 214/273(78%), prior anosmia was associated with the presence of antibodies, with an odds ratio of 9.25 (P<0.001), as well as fever and cough. Being a non-smoker, non-Saudi, and previously diagnosed with COVID-19 infection by PCR were statistically significantly different by seroprevalence status. We found that the seroprevalence of IgG against SARS-CoV-2 nucleocapsid antigen was 10.8% in HCWs at the peak of the pandemic in Saudi Arabia. We also observed a decreasing temporal trend of IgG seropositivity over 8 months follow up period.