Search

Background: The coronavirus disease 2019 (COVID-19) pandemic demands reliable prognostic models for estimating the risk of long COVID. We developed and validated a prediction model to estimate the probability of known common long COVID symptoms at least 60 days after acute COVID-19. Methods: The prognostic model was built based on data from a multicentre prospective Swiss cohort study. Included were adult patients diagnosed with COVID-19 between February and December 2020 and treated as outpatients, at ward or intensive/intermediate care unit. Perceived long-term health impairments, including reduced exercise tolerance/reduced resilience, shortness of breath and/or tiredness (REST), were assessed after a follow-up time between 60 and 425 days. Predictors were selected out of twelve candidate predictors based on three reliable methods, namely the augmented backward elimination (ABE) method, the adaptive best-subset selection (ABESS) method and model-based recursive partitioning (MBRP) approach. Model performance was assessed with the scaled Brier score, concordance c statistic and calibration plot. The final prognostic model was determined based on best model performance. Results: In total 2799 patients were included in the analysis, of which 1588 patients were in the derivation cohort and 1211 patients in the validation cohort. The REST prevalence was similar between the cohorts with 21.6% (n = 343) in the derivation cohort and 22.1% (n = 268) in the validation data. The same predictors were selected with the ABE and ABESS variable selection method. The final prognostic model was based on the ABE and ABESS selected predictors. The corresponding model discrimination in the validation cohort was 0.78 (95% CI: 0.75 to 0.81), calibration slope was 0.92 (95% CI: 0.78 to 1.06) and calibration intercept was -0.06 (95% CI: -0.22 to 0.09). A patient’s probability of developing REST symptoms \hat{y} = exp(S) / (1 + exp(S)) can be calculated with S = −4.947 + 0.349 × number of acute COVID-19 symptoms + 0.339 × severity of acute COVID-19 ward + 1.737 × severity of acute COVID-19 intensive or intermediate care + 0.128 × feeling of stress at home + 0.013 × age at presentation + 0.352 × female sex + 0.346 × presence of at least one cardiovascular risk factor − 0.097 × responsible for childcare/family member + 0.022 × body mass index, with feeling of stress at home ranges from 1 (no stress) to 10 (maximum stress) and responsibility for childcare/family member ranges from 1 (no responsibility/not applicable) to 6 (full responsibility). Conclusion: The proposed model is reliable to identify COVID-19 infected patients at risk for REST symptoms. Before implementing the prognostic model in daily clinical practice, the conduct of an impact study is recommended.

Additional file 1. Online Supplementary Material.

Additional file 4. Logistic regression model based on the ABE and ABESS variable selection, calculated on the total cohort.

RATIONALE Coronavirus disease 2019 (COVID-19) can lead to acute respiratory distress syndrome with fatal outcomes. Evidence suggests that dysregulated immune responses, including autoimmunity, are key pathogenic factors. OBJECTIVES To assess whether IgA autoantibodies target lung-specific proteins and contribute to disease severity. METHODS We collected 147 blood, 9 lung tissue, and 36 bronchoalveolar lavage fluid samples from three tertiary hospitals in Switzerland and one in Germany. Severe COVID-19 was defined by the need to administer oxygen. We investigated the presence of IgA autoantibodies and their effects on pulmonary surfactant in COVID-19 using the following methods: immunofluorescence on tissue samples, immunoprecipitations followed by mass spectrometry on bronchoalveolar lavage fluid samples, enzyme-linked immunosorbent assays on blood samples, and surface tension measurements with medical surfactant. MEASUREMENTS AND MAIN RESULTS IgA autoantibodies targeting pulmonary surfactant proteins B and C were elevated in patients with severe COVID-19, but not in patients with influenza or bacterial pneumonia. Notably, pulmonary surfactant failed to reduce surface tension after incubation with either plasma or purified IgA from patients with severe COVID-19. CONCLUSIONS Our data suggest that patients with severe COVID-19 harbor IgA against pulmonary surfactant proteins B and C and that these antibodies block the function of lung surfactant, potentially contributing to alveolar collapse and poor oxygenation. This article is open access and distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives License 4.0 (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Objectives Critically ill coronavirus disease 2019 (COVID‐19) patients are characterised by a severely dysregulated cytokine profile and elevated neutrophil counts, impacting disease severity. However, it remains unclear how neutrophils contribute to pathophysiology during COVID‐19. Here, we assessed the impact of the dysregulated cytokine profile on the regulated cell death (RCD) programme of neutrophils. Methods Regulated cell death phenotype of neutrophils isolated from critically ill COVID‐19 patients or healthy donors and stimulated with COVID‐19 or healthy plasma ex vivo was assessed by flow cytometry, time‐lapse microscopy and cytokine multiplex analysis. Immunohistochemistry of COVID‐19 patients and control biopsies were performed to assess the in situ neutrophil RCD phenotype. Plasma cytokine levels of COVID‐19 patients and healthy donors were measured by multiplex analysis. Clinical parameters were correlated to cytokine levels of COVID‐19 patients. Results COVID‐19 plasma induced a necroptosis‐sensitive neutrophil phenotype, characterised by cell lysis, elevated release of damage‐associated molecular patterns (DAMPs), increased receptor‐interacting serine/threonine‐protein kinase (RIPK) 1 levels and mixed lineage kinase domain‐like pseudokinase (MLKL) involvement. The occurrence of neutrophil necroptosis MLKL axis was further confirmed in COVID‐19 thrombus and lung biopsies. Necroptosis was induced by the tumor necrosis factor receptor 1 (TNFRI)/TNF‐α axis. Moreover, reduction of soluble Fas ligand (sFasL) levels in COVID‐19 patients and hence decreased signalling to Fas directly increased RIPK1 levels, exacerbated TNF‐driven necroptosis and correlated with disease severity, which was abolished in patients treated with glucocorticoids. Conclusion Our results suggest a novel role for sFasL signalling in the TNF‐α‐induced RCD programme in neutrophils during COVID‐19 and a potential therapeutic target to curb inflammation and thus influence disease severity and outcome., This work highlights the role of cell death ambiguity in neutrophils and its contribution to pathophysiology during critical COVID‐19. The inflammatory COVID‐19 environment provoked a TNF‐α‐induced necroptosis‐sensitive neutrophil subpopulation, characterised by elevated release of DAMPs, increased RIPK1 levels and a pivotal role for MLKL. Neutrophil necroptosis through the RIPK1‐RIPK3‐MLKL axis was further confirmed in COVID‐19 thrombus and lung biopsies. Blunted Fas engagement by sFasL was identified to drive elevated RIPK1 levels upon TNF‐α‐induced necroptosis, while decreased sFasL plasma levels in critically ill COVID‐19 patients correlated to elevated disease severity, which was masked by glucocorticoid treatment.

Additional file 1. e-Appendix 1: Specifications on the RISC-19-ICU registry structure and data collection. e-Appendix 2: Missing data handling. e-Table 1: Overall “unmatched” baseline characteristics on Day 0. e-Figure 1: “Love Plot” presenting Standardized Mean Differences between the unmatched and matchedcohort. e-Table 2: Demographics, characteristics at ICU admission, progression of respiratory support and outcome; stratified by survivor status. e-Table 3: Progression of respiratory support stratified by respiratory support strategy at ICU admission. e-Figure 2: Temporal relationship between the period of admission to the intensive care unit and the use of respiratory support strategies or mortality rate. e-Figure 3: Kaplan Meier curves for intensive care unit mortality stratified by the period of admission to the intensive care unit. e-Table 4: Demographics, characteristics at ICU admission, progression of respiratory support and outcome for patients never requiring intubation and invasive mechanical ventilation. e-Figure 4: C-Reactive Protein stratified by respiratory support strategy on Day 0 over the first week of ICU stay. e-Table 5: Mixed Effect Model of C-Reactive Protein stratified by respiratory support strategy on Day 0 over the first week of ICU stay. e-Figure 5: Multivariable adjusted COX regression model for the incidence of intubation. e-Figure 6: Multivariable adjusted COX regression model for overall ICU mortality. e-Figure 7: Multivariable adjusted COX regression model for ICU mortality (intubated patients only). e-Figure 8: Multivariable adjusted COX regression model for ICU discharge (intubated patients only). e-Table 6: Prognostic Model for the identification of patients with lower ICU mortality risk after a failed HFNC or NIV trial. e-Figure 9: Nomogram, Receiver Operating Curves and stratified Kaplan Meier curve for a prognostic model identifying patients with lower ICU mortality risk after a failed HFNC or NIV trial. e-Table 7: Area Under the Receiver Operating Curves (AUROCs) for the Prognostic Score versus classic severity scores.