The dilemma of COVID-19 recurrence after clinical recovery.

Dear Editor,

We read with interested the recent article published by Batisse et al. (1) regarding the possible recurrences of COVID-19 symptoms after recovery and their discussion on the possible hypothesis of reactivation or reinfection.

In this specific context, the duration of immunization after clinical recovery is still unknown and this could be of particular concern regarding the management and spread of infection. According to the WHO's guidelines on clinical management, a patient can be discharged from hospital after two consecutive negative real-time reverse-transcriptase polymerase-chain-reaction (RT-PCR) tests at nasopharyngeal swabs at least 24 hours apart in a clinically recovered patient (https://ecdc.europa.eu/en/publications-data/covid-19-guidance-discharge-and-ending-isolation). Nevertheless, some recent reports described patients with recurrent RT-PCR tested positive again after clinical recovery (2, 3, 4, 5), but these reports usually included a small number of patients followed-up for a limited period of time (6, 7).

We collected clinical data of COVID-19 positive patients who had cured and discharged from two hospitals of ASST Rhodense, in Milan Province, Northern Italy, from March 9th to June 30th 2020. We considered patients with a positive RT-PCR test for COVID-19 on nasopharyngeal swab who were subsequently discharged when symptoms disappeared and two negative nasopharyngeal swabs repeated after 24-48 hours from each other were obtained. After discharging, patients were followed-up in designated medical wards or in a designated nursing home, where nasopharyngeal swabs were periodically collected (usually every week or anytime the patients developed clinical symptoms). We included in the analysis all the patients with a recurrence of COVID-19 infection, defined as a new positive nasopharyngeal swab after two negative tests.

A total of 1146 patients were hospitalized and then discharged for COVID-19 in our hospitals during the time-frame considered. Among these, 125 (10.9%) had a recurrence of COVID-19 infection. Table 1 summarized the clinical and demographic characteristics of this population; mean age was 65,7 years (95% CI 26-95) and most of patients were primarily hospitalized for interstitial pneumonia (n=103, 82.4%). The mean time to clinical recovery and two negative nasopharyngeal swabs was 27.7 days (95% CI 11-51); after that, the mean time to recurrence was 19.9 days (95% CI 3-43). Recurrence of COVID-19 infection was mainly diagnosed by chance during follow-up surveillance (n=96, 76.8%), whereas 29 patients (23.2%) developed clinical symptoms (fever in 16, malaise/fatigue in 9 and respiratory failure in 4 patients). After a mean time of 14.8 days (95% CI 6-36), 102 subjects (81.6%) had two additional negative nasopharyngeal swabs and were considered clinically recovered for the second time. During follow-up, 11 patients (8.8%) died and 12 (9.6%) were still positive when database was closed. Patients who died were older than others (mean age 86.4 years, 95% CI 77-92) and 8 of them (72.7%) had clinical symptoms at the time of recurrence (4 fever and 4 respiratory failure). The mean time from recurrence of COVID-19 infection to death was 8 days (95% CI 5-11).

Table 1

Main demographic and clinical characteristics of a cohort of 125 subjects with recurrent COVID-19 infection.

Female, n (%)	64 (51.2)
Age, years old (mean, 95% CI)	65.7 (26-95)
Hospitalized for interstitial pneumonia, n (%)	103 (82.4)
Time to first clinical recovery, days (mean, 95% CI)	27.7 (11-51)
Time to recurrence, days (mean, 95% CI)	19.9 (3-43)
Time to second clinical recovery, (n=102), days (mean, 95% CI)	14.8 (6-36)

Currently, there is a certain possibility of RT-PCR rendering false negative results due to sampling procedures, sources of samples and the sensitivity/specificity of the nucleic acid test kit (8). At the moment, it is impossible to discriminate if the positive nasopharyngeal swab results are due to real recurrence of COVID-19 infection or intermittent shedding of RNA fragments, especially in asymptomatic subjects. It is therefore possible that recurrences should be persistent infections in which nasopharyngeal swab resulted falsely negative at discharge. Alternatively, it cannot be excluded that truly negative discharged patients suffered reactivation or were re-infected with another COVID-19 strain, especially in elderly or in subjects with comorbidities (5). In our cohort, a certain amount of patients (23.2%) with RT-PCR recurrences developed new clinical symptoms, considering this interpretation plausible. To our knowledge, no studies have been conducted to investigate the contagiousness of patients with recurrence of viral RNA shedding. If these patients were contagious, they could represent a potential source of infections for the community.

At our knowledge, this is the largest cohort of subjects with recurrent COVID-19 infection. Our data confirmed that more than 10% of patients clinically recovered from COVID-19 infection had re-positive RT-PCR at nasopharyngeal swab during post-discharge follow-up (6, 7); most of these subjects were asymptomatic at the time of recurrence.

In conclusion, our data confirm that recurrence of COVID-19 infection is a fairly frequent phenomenon. Little is known on how to manage these patients and how this will impact the evolution of the pandemic in the future.