Identifying molecular phenotypes in sepsis: an analysis of two prospective observational cohorts and secondary analysis of two randomised controlled trials
Authors' conclusion
Molecular phenotypes previously identified in ARDS are also identifiable in multiple sepsis cohorts and respond differently to activated protein C.
Study context
- Population
- Critically ill adults with sepsis (VALID and Early Assessment of Renal and Lung Injury (EARLI) cohorts; PROWESS-SHOCK and Vasopressin and Septic Shock Trial (VASST) trials)VALID N=1140
- Comparison
- Hyperinflammatory vs hypoinflammatory molecular phenotype
- Primary endpoint
- Number of molecular phenotypes (classes) in each sepsis cohort by latent class analysis (LCA)
Methodological detailslatent class analysis, retrospective application of validated classifier models
- Latent class analysis
- Retrospective application of validated classifier models
Primary result
As reported in the abstract
Phenotype classification (two-class model best fit both VALID and EARLI)
| Group | Value | Cohort |
|---|---|---|
| Hypoinflammatory | 70.5% | VALID |
| Hyperinflammatory | 29.5% | VALID |
| Hypoinflammatory | 64.8% | EARLI |
| Hyperinflammatory | 35.2% | EARLI |
Value
Hypoinflammatory70.5%
Hyperinflammatory29.5%
Hypoinflammatory64.8%
Hyperinflammatory35.2%
Cohort
HypoinflammatoryVALID
HyperinflammatoryVALID
HypoinflammatoryEARLI
HyperinflammatoryEARLI
- Two-class model best fit both VALID and EARLI
- P < 0.0001
- Higher pro-inflammatory cytokines, more vasopressor use, more bacteraemia, lower protein C and higher mortality in hyperinflammatory vs hypoinflammatory
- P < 0.0001 for all
Further findings
PROWESS-SHOCK: phenotypes and response to activated protein C · interaction P
- PROWESS-SHOCK: phenotypes and response to activated protein C — Response to activated protein C differed by phenotype
- p=0.0043
- PROWESS-SHOCK: phenotypes and response to activated protein C: of 1680 patients
Concordance of sepsis phenotypes with acute respiratory distress syndrome (ARDS) phenotypes (classifier models)
- AUC, depending on the model
- 0.87-0.96
- Findings similar excluding participants with both sepsis and ARDS
VASST: treatment interaction with vasopressor type
- Phenotype proportions similar to other cohorts; no treatment interaction with vasopressor type
- P = 0.72
Original publication
Identifying molecular phenotypes in sepsis: an analysis of two prospective observational cohorts and secondary analysis of two randomised controlled trials. Lancet Respir Med. Published 23 August 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →