Atypical response to bacterial coinfection and persistent neutrophilic bronchoalveolar inflammation distinguish critical COVID-19 from influenza
Seppe Cambier, Mieke Metzemaekers, Ana Carolina de Carvalho, Amber Nooyens, Cato Jacobs, Lore Vanderbeke, Bert Malengier-Devlies, Mieke Gouwy, Elisabeth Heylen, Philippe Meersseman, Greet Hermans, Els Wauters, Alexander Wilmer, the CONTAGIOUS Consortium, Dominique Schols, Patrick Matthys, Ghislain Opdenakker, Rafael Elias Marques, Joost Wauters, Jennifer Vandooren, Paul Proost
Seppe Cambier, Mieke Metzemaekers, Ana Carolina de Carvalho, Amber Nooyens, Cato Jacobs, Lore Vanderbeke, Bert Malengier-Devlies, Mieke Gouwy, Elisabeth Heylen, Philippe Meersseman, Greet Hermans, Els Wauters, Alexander Wilmer, the CONTAGIOUS Consortium, Dominique Schols, Patrick Matthys, Ghislain Opdenakker, Rafael Elias Marques, Joost Wauters, Jennifer Vandooren, Paul Proost
View: Text | PDF
Research Article COVID-19 Immunology

Atypical response to bacterial coinfection and persistent neutrophilic bronchoalveolar inflammation distinguish critical COVID-19 from influenza

Abstract

Neutrophils are recognized as important circulating effector cells in the pathophysiology of severe coronavirus disease 2019 (COVID-19). However, their role within the inflamed lungs is incompletely understood. Here, we collected bronchoalveolar lavage (BAL) fluids and parallel blood samples of critically ill COVID-19 patients requiring invasive mechanical ventilation and compared BAL fluid parameters with those of mechanically ventilated patients with influenza, as a non–COVID-19 viral pneumonia cohort. Compared with those of patients with influenza, BAL fluids of patients with COVID-19 contained increased numbers of hyperactivated degranulating neutrophils and elevated concentrations of the cytokines IL-1β, IL-1RA, IL-17A, TNF-α, and G-CSF; the chemokines CCL7, CXCL1, CXCL8, CXCL11, and CXCL12α; and the protease inhibitors elafin, secretory leukocyte protease inhibitor, and tissue inhibitor of metalloproteinases 1. In contrast, α-1 antitrypsin levels and net proteolytic activity were comparable in COVID-19 and influenza BAL fluids. During antibiotic treatment for bacterial coinfections, increased BAL fluid levels of several activating and chemotactic factors for monocytes, lymphocytes, and NK cells were detected in patients with COVID-19 whereas concentrations tended to decrease in patients with influenza, highlighting the persistent immunological response to coinfections in COVID-19. Finally, the high proteolytic activity in COVID-19 lungs suggests considering protease inhibitors as a treatment option.

Authors

Seppe Cambier, Mieke Metzemaekers, Ana Carolina de Carvalho, Amber Nooyens, Cato Jacobs, Lore Vanderbeke, Bert Malengier-Devlies, Mieke Gouwy, Elisabeth Heylen, Philippe Meersseman, Greet Hermans, Els Wauters, Alexander Wilmer, the CONTAGIOUS Consortium, Dominique Schols, Patrick Matthys, Ghislain Opdenakker, Rafael Elias Marques, Joost Wauters, Jennifer Vandooren, Paul Proost

×

Figure 7

Quantification of biomarkers in BAL fluid from patients with severe COVID-19 or influenza, stratified by the timing of a bacterial coinfection.

Options: View larger image (or click on image) Download as PowerPoint
Quantification of biomarkers in BAL fluid from patients with severe COVI...
Multiplex and ELISA technology was used to determine concentrations of (A) IL-15, (B) granzyme B, (C) CCL2, (D) CCL7, (E) CCL8, (F) CXCL1, (G) CXCL10, (H) CXCL11, and (I) CXCL12α in COVID-19 and influenza BAL fluid samples. (J) BAL fluid neutrophil counts. (K and L) ELISA was used to determine (K) SLPI and (L) elafin concentrations in the BAL fluids. All COVID-19 samples were categorized based on the absence of a coinfection (n = 6) or the acute phase (n = 11) or mid/late phase of a bacterial coinfection (n = 12) based on the timing of the BAL sampling relative to the coinfection time course. Influenza patient samples were also categorized in the acute phase (n = 8) and the mid/late phase (n = 5) of a bacterial coinfection. Data are shown as box-and-whisker plots (box: median with interquartile range, whiskers: full data distribution), with each dot representing an individual patient sample. The dashed lines indicate the lower detection limits. In the mid/late groups, unfilled symbols indicate samples taken during the late phase of the bacterial coinfection; the others were taken during the midphase of the bacterial coinfection. Triangles indicate samples with an additional fungal coinfection. Data were statistically analyzed by a linear mixed model with correction for multiple samples per patient using a random intercept model or a Mann-Whitney U test, where appropriate. P values are shown above brackets.