Introduction

,1–5,6–13].

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) humoral immunity suggests that more than 90% of seroconversion rates occur after acute primary infection with variable degrees of decline in antibody levels over time [14,15,1,2,3,4,5,6,7,16].

The aim of this prospective longitudinal study was to comprehensively characterize the relative incidence of COVID-19 reinfections in relation to serological response among individuals who had recovered from COVID-19 after the first wave. The study included a wide spectrum of unselected patients ranging from asymptomatic to severely infected, assessed over a 10-month follow-up period.

Methods

Study setting and patient population

,17].

The target population was a cohort of all consecutive adult in- and outpatients (≥ 18 years) attending the Infectious Disease Department with a diagnosis of COVID-19 from March 1 (the day of the first COVID-19 diagnosis at our hospital) until May 30, 2020. Further definitions of acute COVID-19 and baseline conditions are summarized in Supplementary Table 1.

Serological test collection, reinfection follow-up, and definitions

SARS-CoV-2 antibody concentrations were measured at the serological follow-up visits each month (± 15 days) after symptom onset for the first 4 months, and every other month up to 10 months (± 15 days), from March 2020 to February 2021 (CORMOR 3–4® protocol). Patients attending at least two serological follow-ups were included in the study (Fig. 1).

Fig. 1
figure 1

Reinfection and serological follow-up (up to February 2021): flow diagram of in- and outpatients with COVID-19 included

Full size image

,18].

Specifically, patients with symptoms or signs of recurrent illness (fever, rhinorrhea, sore throat, cough, dyspnea, sputum, myalgia, fatigue, thoracic pain, vomiting, diarrhea, dysgeusia or anosmia, conjunctivitis, rash) and/or a positive PCR test for SARS-CoV-2 were instructed to contact the research team by phone, in order to schedule a prompt visit (within 24 h) at the infectious disease outpatient clinic or go to the emergency department for medical examination and PCR test for SARS-CoV-2. In addition, during this contact, patients were asked about previous not-reported episodes of symptoms/signs. Systematic SARS-CoV-2 PCR test was performed at regular intervals (every 2/4 weeks) only for healthcare workers (HCWs) in accordance to Hospital and Nursing homes/long-term facility protocols.

At the time of reinfection, for definitive analysis, we classified patients according to their most recent (within 2 months) antibody status into three groups: (1) seronegative in the absence of any IgM-/IgG-positive serological samples; (2) seroreverted in the presence of a decline in IgM/IgG antibody levels below the positivity threshold after initial seroconversion; and (3) seropositive in the presence of persistence of IgM-/IgG-positive serological sample.

Antibody measurements

Serum concentrations of the anti-SARS-CoV-2 specific antibodies IgG and IgM were assessed using iFlash-SARS-CoV-2 (Shenzhen YHLO Biotech Co., Ltd., China, distributed in Italy by Pantec SRL), a paramagnetic particle chemiluminescence immunoassay (CLIA) for the determination of IgM and IgG antibodies against SARS-CoV-2 N and S protein. In accordance with the manufacturer’s instructions, the IgM and IgG thresholds for positivity were considered to be 10.0 kAU/L.19.

Statistical analysis

Patients’ demographic and clinical characteristics were presented with absolute values and percentages for categorical variables and means or medians (standard deviation (SD) or interquartile ranges (IQRs)) for continuous variables. The Shapiro–Wilk test was used to assess whether data were normally or non-normally distributed. Categorical variables were compared using the chi square (χ2) test or Fisher’s exact test, while quantitative variables were compared using the t-test or Mann–Whitney U test, as appropriate. Statistical analysis was performed using STATA 16.1.

Results

Study population at onset of acute COVID-19

Overall, during the study period, a total of 1067 patients received the COVID-19 diagnosis. After excluding 211 patients for refusing to participate in the research, 138 nursing home/long-term facility residents who were not capable of giving their consent due to cognitive decline, 38 who were lost to follow-up, 51 for incomplete serological follow-up and 81 deaths, a total of 546 patients were included (Fig. 1). Demographic and clinical characteristics of patients at baseline are summarized in Table 1. The mean age of our study population was 53 years (SD 15.4; range 18‒94), 292 (53.5%) were female and the majority (480/521, 92.1%) were native Italians. One hundred and fifteen were HCWs. During the acute phase, most patients (502, 91.9%) were symptomatic and presented mild illness (374, 68.5%). One hundred and forty-seven (27.2%) had been hospitalized (22 in the intensive care unit) (Table 1).

Table 1 Patients’ baseline characteristics and clinical presentation at acute COVID-19 onset
Full size table

Serological dynamics of SARS-CoV-2 IgM and IgG after primary infection

A complete description of the serological evolution of the study population is presented in another work that is currently in progress. In brief, the overall seroconversion rate within 2 months was 32% for IgM: 25% in mild cases and 61% in moderate to critical cases. IgM was generally not detected after 4 months (90th percentile equal to 135 days). The overall seroconversion rate for IgG within 2 months was higher for IgG (90%): 91% in mild patients, 100% in moderate to critical patients, and only 54% in asymptomatic cases. About half of the patients (47%) had experienced IgG seroreversion at 10 months, and rates of antibody loss were almost complete (88%) for asymptomatic patients, around half for mild cases (53%) but only 13% for moderate to severe COVID-19.

Reinfection

Patients were followed up for a median of 10 months (IQR, 6.2–10.4). The reinfection rate was 1.1% (6/546). Cases of reinfection occurred at a median of 9 months (IQR 8.2‒10.2) after the acute onset of the first episode. The median age was 44.5 years (IQR 33‒49) and all were HCWs (Table 2). As reported in Table 2, all patients experienced mild infection (6/370, 1.6%) during the first episode and manifested mild (n = 5) or asymptomatic (n = 1) reinfections. Reinfection rates did not differ significantly in seronegative (2/56, n = 3.6%), seroreverted (2/137, 1.5%), or seropositive (2/353, 0.6%) patients (p = 0.085) but were significantly higher in HCWs than in non-HCWs (6/119, 5.0% versus 0/385, p < 0.001). Only one patient had a high-titer serological response against SARS-CoV-2 at the time of reinfection (Table 2) (Fig. 2). After reinfection, none of the patients developed an IgM response and only two had a transitory boosted IgG immunization response (Fig. 2). After repeating isolation and tracing of close contacts, we found no transmission to other individuals. The serological evolution after the first and second infections is described in Fig. 2.

Table 2 Demographic, clinical, and laboratory characteristics of patients with possible SARS-CoV-2 reinfections
Full size table
Fig. 2
figure 2

Humoral IgM and IgG response of reinfected patients. Patient 1—blue line; patient 2—green line; patient 3—yellow line; patient 4—red line

Full size image

Discussion

Our prospective longitudinal study of an unselected population with COVID-19, acquired during the first wave of the pandemic, with different degrees of severity, shows that there was a very low risk of reinfection after a mean follow-up of 10 months and that the primary serological response was not accurately predictive of reinfection.

,20,1–10,

,1–10,13,21,22].

,1,2,10–12,23,24].

The fact that all our reinfections occurred in HCWs is surprising. This could be due either to a higher grade of exposure to COVID-19 cases or to a more rapid access to SARS-CoV-2 molecular PCR testing compared with the general population [1,2,6,12,18].

,8,10,25,26].

It is worth noting that although there is currently no strong evidence that previous SARS-CoV-2 infection reduces the transmission risk [23,14,23,1,7,10,18,25,1,2,3,4,5,6,7,8,9,10,14,24,27,28

,].

,7,

Conclusions

Most people who have recovered from COVID-19 have a low risk of reinfection; secondary infections may occur mainly in patients with a primary mild COVID-19 infection and with low or absent serological response at the time of reinfection. These findings may suggest that natural humoral protective immunity may be transient and may not confer herd immunity. Given the short supply of vaccine in some countries or settings, public health interventions should be extended as not to prioritize patients with a previous history of SARS-CoV-2 infection except for those at risk for poor outcomes and/or high-risk settings of exposure. Further large-scale standardized longitudinal studies with a longer follow-up focused both on humoral and on cell-mediated adaptive SARS-CoV-2 immunity are needed to determine the longevity of protection after infection for subsequent episodes in order to understand the evolution of the pandemic and to design a vaccination plan, taking into consideration emerging variants of concern.

Supplementary information

The online version contains the supplementary material.