Second-wave SARS-CoV-2 reinfection rate in Manaus may have reached 31%, study suggests | AGÊNCIA FAPESP

Second-wave SARS-CoV-2 reinfection rate in Manaus may have reached 31%, study suggests An Anglo-Brazilian research center analyzed samples from blood donors tested for antibodies against the novel coronavirus (patients had to be transferred to other parts of Brazil from Manaus when healthcare services in the capital of Amazonas state were overwhelmed during the second peak of the epidemic; photo: Brazilian Air Force)

Second-wave SARS-CoV-2 reinfection rate in Manaus may have reached 31%, study suggests

June 09, 2021

By Karina Toledo  |  Agência FAPESP – A study conducted by scientists affiliated with the Brazil-UK Center for Arbovirus Discovery, Diagnosis, Genomics and Epidemiology (CADDE) suggests that up to 31% of COVID-19 cases in the Amazonian city of Manaus reported since January 2021, when the city was hit by a second wave of the epidemic, were cases of reinfection by novel variant P.1.

The researchers tested samples from blood donors to detect antibodies against SARS-CoV-2. Their findings are described in an article posted to the preprint platform medRxiv and not yet peer-reviewed.

“We first selected 3,655 people who repeatedly donated blood in 2020 and early 2021. We then whittled these down to donors who hadn’t been vaccinated and had donated at least three times in the period, at least once before July 2020 and once after January 2021 [when P.1 was already dominant in the region],” Ester Sabino, last author of the article, told Agência FAPESP

Sabino is a professor at the University of São Paulo (USP) and principal investigator for CADDE, which is supported by FAPESP

The 238 samples included in the study were submitted to a laboratory test that detects immunoglobulin G (IgG) antibodies, which typically appear some two weeks after the onset of symptoms during the acute stage of the infection, and then decay, eventually becoming undetectable in many cases. The researchers assumed that in the event of reinfection the amount of antibodies would rise again in the most recent blood samples.

Based on this analysis, the donors were divided into four groups. In the first, all three samples tested negative. In the second, samples donated in 2020 tested positive but those donated in January 2021 tested negative or displayed antibody level decay. The third group comprised samples from subjects infected only by P.1, meaning they tested negative in 2020 but positive in 2021. The fourth group comprised cases in which antibody levels described a V shape – up in the first sample, down in the second, and up again in the third.

“These cases in the fourth group were the ones that most clearly characterized reinfection,” Sabino said.

If the antibody level detected in 2021 was higher than in 2020, the second sample was classified as a “probable reinfection” despite the long period since the first sample was donated. If the 2021 level was lower than the 2020 level but higher than expected after normal decay, the sample was classed a “possible reinfection”.

Statistical analysis showed that the chances of donors who tested positive in 2020 contracting P.1 ranged from 9.5% to 18% (depending on whether cases considered possible or probable reinfections were included). For those who tested negative in 2020, the probability of infection by P.1 was 40%.

The reinfection rate they calculated varied from 16% to 31%, depending on whether cases considered possible or probable reinfections were included.

According to Sabino, the findings match the estimates produced by the CADDE group with the aid of mathematical models in a study reported in Science in April 2021 (more at: 

“It’s very difficult to estimate the reinfection rate with official data alone,” she said. “The right way would be to follow a large group of people in the first wave of the epidemic, wait for them to display symptoms, do the RT-PCR test, keep the samples, and repeat everything in the second wave with the same volunteers. In Manaus that’s impossible. Most people weren’t tested by the molecular method in the first wave. For this reason, we’re pursuing alternative methods to estimate the risk of reinfection by the novel variant P.1.”

According to Carlos Prete, a PhD candidate at the University of São Paulo’s Engineering School (POLI-USP) and first author of the article, a limitation of the study was that some cases classified as infections by P.1 may actually have been unobserved reinfections.

“There’s a gap of months between successive samples from any one donor, so a donor can be infected during the period between donations and nevertheless test negative on the date of the donation made after the infection because of antibody decay,” Prete said, adding that subnotification of cases creates a false impression that the reinfection rate is low.

“Owing to under-reporting and the large proportion of people who are infected but asymptomatic, the probability of an infected person being notified as a confirmed case is low, so even if the probability of reinfection is considerable the proportion of people with two confirmed infections will always be small,” he said.

The article “Reinfection by the SARS-CoV-2 P.1 variant in blood donors in Manaus, Brazil” can be accessed at:




Agência FAPESP licenses news reports under Creative Commons license CC-BY-NC-ND so that they can be republished free of charge and in a straightforward manner by other digital media or by print media. The name of the author or reporter (when applied) must be cited, as must the source (Agência FAPESP). Using the button HTML below ensures compliance with the rules described in Agência FAPESP’s Digital Content Republication Policy.

Topics most popular