AUTHOR=Chen Chen , Parthasarathy Sairam , Leung Jacqueline M. , Wu Michelle J. , Drake Katherine A. , Ridaura Vanessa K. , Zisser Howard C. , Conrad William A. , Tapson Victor F. , Moy James N. , deFilippi Christopher R. , Rosas Ivan O. , Prabhakar Bellur S. , Basit Mujeeb , Salvatore Mirella , Krishnan Jerry A. , Kim Charles C. 

TITLE=Distinct temporal trajectories and risk factors for Post-acute sequelae of SARS-CoV-2 infection

JOURNAL=Frontiers in Medicine

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1227883

DOI=10.3389/fmed.2023.1227883

ISSN=2296-858X

ABSTRACT=Background: The understanding of Post-acute sequelae of SARS-CoV-2 infection (PASC) can be improved by longitudinal assessment of symptoms encompassing the acute illness period. To gain insight into the various disease trajectories of PASC, we assessed symptom evolution and clinical factors associated with the development of PASC over three months, starting with the acute illness period. 

Methods: We conducted a prospective cohort study to identify parameters associated with PASC. We performed cluster and case control analyses of clinical data, including symptomatology collected over three months following infection. 

Results: We identified three phenotypic clusters associated with PASC that could be characterized as remittent, persistent, or incident based on the three-month change in symptom number compared to study entry: remittent (median; min, max: -4; -17, 3), persistent (-2; -14, 7), or incident (4.5; -5, 17) (p=0.041 remittent vs. persistent, p<0.001 remittent vs. incident, p<0.001 persistent vs. incident). Despite younger age and lower hospitalization rates, the incident phenotype had a greater number of symptoms (15; 8, 24) and a higher proportion of participants with PASC (63.2%) than the persistent (6; 2, 9 and 52.2%) or remittent clusters (1; 0, 6 and 18.7%). Systemic corticosteroid administration during acute infection was also associated with PASC at three months (OR [95% CI]: 2.23 [1.14, 4.36]).

Conclusions: An incident disease phenotype characterized by symptoms that were absent during acute illness and the observed association with high dose steroids during acute illness have potential critical implications for preventing PASC.