In this retrospective cohort study, we investigated the clinical features of the cured or recovered COVID-19 patients for the first time. Although they were PCR or IgM positive, these patients displayed no clinical manifestations of infection, and no signs of new acute infection were found on chest CT, indicating that these patients did not meet the re-infection criteria. Based on these findings, a positive result on PCR or IgM assay should not be considered indicative of COVID-19 re-infection. There might be several reasons for absence of re-infections. Firstly, the patients with COVID-19 were discharged from hospitals after following strict criteria, and the duration of hospital stay was more than 14 days, far exceeding that in community acquired pneumonia [10], which means that the SARS-CoV-2 was more likely to be have been eradicated. Secondly, 82.4% of the patients were positive for antibody, which might prevent a repeat infection by the virus. Thirdly, an effective prevention and control strategy ensured that the cured patients were kept away from other confirmed COVID-19 patients. Finally, the medical staffs working in the front line have not been infected till date, which effectively prevented secondary infections and spread of the disease in the hospital [11]. Given that no re-infections were found in patients with negative results for IgG, the role of adaptive immunity in cured patients need to be further studied and verified.
The diagnosis of early COVID-19 is based on the PCR and antibody tests. Positive results mean that the patient is infectious in the early stage. For those who have recovered, the clinical significance of the PCR and antibody tests has not been clarified. Our study found that was resurgence of positive results of PCR or IgM tests in some patients after being discharged home. Among people who were in close contact with the cured patients, no one was diagnosed with COVID-19 as reported by both the cured patients and the Guangzhou CDC. The incubation period of COVID-19 is 3-14 days, and our follow-up period for cured patients was more than 14 days. This might have helped in excluding the cases in the incubation period of the infection. Based on these findings, we believe that the cured patients cannot cause human-to-human transmission. They also indicate that a positive result of the PCR or IgM assay does not mean that the cured patient is infectious.
IgG antibodies usually appears 3-40 days after the onset of symptoms [8]. In our study, 82.4% patients produced IgG antibodies. However, IgG antibodies were not detected in 17.6% patients when tested after 53 days on an average from the onset of the disease, which means that these patients might not produce IgG antibodies. IgM antibodies appeared in 35 patients when tested after 53 days on an average following the onset of symptoms, and disappeared in 12 patients during the follow up period. Therefore, IgM antibodies might be present in some COVID-19 patients for a long time.
All COVID-19 patients were discharged home after they had negative PCR test results on two consecutive occasions, 24 hours apart. However, positive results of PCR or IgM were again observed in some patients during the follow up period. The positive PCR turned to negative in the subsequent retest. Current research has not been able to explain the cause of the positive PCR retests, or confirm whether it is caused by a virus residue. Interestingly, the percentage of positive anal swabs in the cured patients was much higher than the positive pharyngeal swabs. PCR positivity of anal swabs was reported in several studies, which has led to a discussion on the possibility of fecal-oral transmission [12, 13]. The reason for PCR positive anal swabs may be that the virus enters the digestive tract from the patient's mouth. However, whether the virus remains active is unknown. During the follow-up, we did not find any new confirmed COVID-19 cases that came into contact with the cured patients who demonstrated positive PCR test results from anal swabs. Although PCR positive, fecal-oral transmission could not be confirmed in our study, and further research is needed.
Compared with the IgG negative group, the IgG positive group patients were older, with longer hospital stay, higher proportion of antibiotic use, higher proportion of severe cases, and higher proportion of CT abnormalities. Further logistic regression analysis showed that the treatment of antibiotic and length of stay were risk factors for IgG antibody production. The mechanism of antibody production associated with antibiotic treatment and long-term hospitalization is not clear. Although diabetes, cancer, and other diseases may cause a decline in immunity, they do not affect the production of antibodies. Similarly, although the use of glucocorticoids may inhibit the immune system, it also has no effect on the production of antibodies.
Studies found that IgG and IgM have a good diagnostic value in the early stage of the disease [6-8]. However, the value of IgG and IgM in the diagnosis of cured COVID-19 patients is not clear. In our study, we found that both IgM and IgG have no relationship with PCR. Therefore, for the cured patients, IgG and IgM neither have a diagnostic value, nor can they be used to replace the PCR test. Since neither re-infection nor human-to-human transmission was found in the cured patients, IgG and IgM cannot be used to guide the prevention and control of COVID-19.
This study has the following limitations. Firstly, since this was an observational study, no interventions such as re-exposure of the cured patients to SARS-CoV-2 were performed. Therefore, it is hard to judge whether the cured patients were immune to the virus. Secondly, this was a single center study carried out in Guangzhou, a mild epidemic area. Accordingly, the conclusions of this study might not be suitable for extrapolation to other areas. Thirdly, our conclusions were based on a small sample size, which need to be further verified in a study with a large sample size. Nevertheless, our study results clarified some clinical features of the cured patients and maybe be of considerable importance for the prevention and control of COVID-19.