Methods
This retrospective cohort study was conducted from January 20 to March
10, with follow up till April 10, 2020. All cured adult patients with
COVID-19 who performed antibody test were enrolled in our study.
Patients were followed up in Guangzhou Eighth People’s hospital, a
government-designated hospital which admitted nearly 80% of the
COVID-19 cases in Guangzhou, the capital city of Guangdong Province in
southern China. This study was approved by the ethics committee of the
Guangzhou Eighth People’s Hospital. Because of the retrospective nature
of the study design and the grim scenario of COVID-19 pandemic, the
Ethics Committee assented to exempt of all informed consents.
Definition: COVID-19 was diagnosed as per the World Health
Organization’s interim guidelines [9]. High throughput sequencing or
RT-PCR were only performed in subjects with the following features: 1.
with a confirmed or suspected contact history of COVID-19; 2. presented
with symptoms; 3. with abnormal chest computed tomography (CT) imaging
related to COVID-19. A positive result on high throughput sequencing or
RT-PCR assay together with at least two of the above three clinical
features, confirmed the diagnosis of COVID-19. Criteria for
cured and discharged to home were as follows: vital signs were stable
for more than three days; the PCR test was negative two times
consecutively 24 hours apart; and the acute exudative lung lesions were
absorbed or cured on chest CT. Re-infection criteria were as
follows: typical clinical symptoms; chest CT indicative of new
infiltration; and two positive repeat PCR tests performed consecutively
at an interval of more than 24 hours. All confirmed re-infection cases
were reviewed by two senior COVID-19 experts. Person-to-person
transmission criteria were as follows: New confirmed COVID-19 cases
occurred after one with unprotected exposure to the cured within 2
weeks. Person-to-person transmission was assessed on the basis of the
reports of the cured patients. Since Guangzhou CDC released the new
diagnosed COVID-19 cases including the exposure to source of
transmission daily, the reports from CDC were also screened.
Follow up: All recovered or cured patients with COVID-19 were
quarantined at home for two weeks after being discharged. They were free
to go anywhere after two weeks. The cured patients were followed up
every two weeks. Follow-up consisted of assessing the general condition,
symptoms, living area, PCR assay, and antibody test. Additionally, these
recovered patients were required to report if people close to them had
been diagnosed with COVID-19. For patients with a positive PCR test, a
chest CT was performed immediately, and PCR test was re-performed
consecutively at an interval of more than 24 hours. The PCR assay and
antibody test were performed on the same day. If positive, IgM antibody
test would be repeated within two weeks. During the study, the
researchers screened the report from CDC in Guangzhou every day to
determine whether there were any new confirmed COVID-19 cases linked to
transmission by the cured patients.
IgM and IgG Testing: The serum SARS-CoV-2 antibodies (IgM and IgG) were
detected using colloidal Gold-based Immunoassays (ELISA kits, Livzon
Inc, Zhuhai, China). First, the kit was removed and kept for 30 minutes
at room temperature. Subsequently, 10 μl of plasma sample and 20 μl of
whole blood sample were added into the reaction pore until the liquid
was fully absorbed. Lastly, two drops of sample diluents were added into
the reaction hole until the liquid was fully absorbed. The result could
be read in 15 minutes.
Statistical Analyses: Shapiro-Wilk normality test was used to assess for
normal distribution of data. Continuous variables with normal
distribution were expressed as mean ± standard deviations (SD), while
those with non-normal distribution were expresses as median and inter
quartile range (IQR). Categorical variables were summarized as counts
and percentages. For continuous variables, Independent t-test or
Wilcoxon rank sum test were used. For comparison of categorical
variables, Chi-square test and Fisher’s exact test were used. Logistic
regression analyses were performed to examine the relationship between
independent variables and presence of IgG. Determinants with a Pvalue of 0.10 or less in univariate models were initially included in
the multivariate model and were then discarded using backward selection.
A P values < 0.05 means statistically significant. All
data were processed with SPSS version 22.0 for Windows (SPSS, Chicago,
IL, USA).