Back

Sequence Proven Reinfections with SARS-CoV-2 at a Large Academic Center

Morris, C. P.; Eldesouki, R. E.; Fall, A.; Gaston, D. C.; Norton, J. M.; Gallagher, N.; Luo, C. H.; Abdullah, O.; Klein, E. Y.; Mostafa, H. H.

2022-05-19 infectious diseases
10.1101/2022.05.17.22275210 medRxiv
Show abstract

BackgroundIncreased reinfection rates with SARS-CoV-2 have recently been reported, with some locations basing reinfection on a second positive PCR test at least 90 days after initial infection. MethodsWe identified cases where patients had two positive tests for SARS-CoV-2 and evaluated which of these had been sequenced as part of our surveillance efforts, and evaluated sequencing and clinical data. Results750 patients (920 samples) had a positive test at least 90 days after the initial test. The median time between tests was 377 days, and 724 (79%) of the post 90-day positives were collected after the emergence of the Omicron variant in November 2021. Sequencing was attempted on 231 samples and successful in 127. Successful sequencing spiked during the Omicron surge and showed higher median days from initial infection compared to failed sequences (median 398 days compared to 276 days, p<0.0005). A total of 122 (98%) patients showed evidence of reinfection, 45 of which had sequence proven reinfection and 77 had inferred reinfections (later sequence showed a clade that was not circulating when the patient was initially infected). Children accounted for only 4% of reinfections. 43 (96%) of 45 infections with sequence proven reinfection were caused by the Omicron variant, 41 (91%) were symptomatic, 32 (71%), were vaccinated prior to the second infection, and 6 (13%) were Immunosuppressed. Only 2 (4%) were hospitalized, and both had underlying conditions. ConclusionSequence proven reinfections increased with the Omicron variant but generally caused mild infections.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.