Increased mortality in elderly patients with acute respiratory distress syndrome is not explained by host response

Laura R.A. Schouten, Lieuwe D.J. Bos, A. Serpa Neto, L. A. van Vught, M. A. Wiewel, A. J. Hoogendijk, M. J.M. Bonten, O. L. Cremer, Janneke Horn, T. van der Poll, M. J. Schultz, Roelie M. Wösten-van Asperen, the MARS consortium

Research output: Contribution to journalArticleResearchpeer-review

10 Citations (Scopus)


Background: Advanced age is associated with increased mortality in acute respiratory distress syndrome (ARDS) patients. Preclinical studies suggest that the host response to an injurious challenge is age-dependent. In ARDS patients, we investigated whether the association between age and mortality is mediated through age-related differences in the host response. Methods: This was a prospective longitudinal observational cohort study, performed in the ICUs of two university-affiliated hospitals. The systemic host response was characterized in three predefined age-groups, based on the age-tertiles of the studied population: young (18 to 54 years, N = 209), middle-aged (55 to 67 years, N = 213), and elderly (67 years and older, N = 196). Biomarkers of inflammation, endothelial activation, and coagulation were determined in plasma obtained at the onset of ARDS. The primary outcome was 90-day mortality. A mediation analysis was performed to examine whether age-related differences in biomarker levels serve as potential causal pathways mediating the association between age and mortality. Results: Ninety-day mortality rates were 30% (63/209) in young, 37% (78/213) in middle-aged, and 43% (84/196) in elderly patients. Middle-aged and elderly patients had a higher risk of death compared to young patients (adjusted odds ratio, 1.5 [95% confidence interval 1.0 to 2.3] and 2.1 [1.4 to 3.4], respectively). Relative to young patients, the elderly had significantly lower systemic levels of biomarkers of inflammation and endothelial activation. Tissue plasminogen activator, a marker of coagulation, was the only biomarker that showed partial mediation (proportion of mediation, 10 [1 to 28] %). Conclusion: Little evidence was found that the association between age and mortality in ARDS patients is mediated through age-dependent differences in host response pathways. Only tissue plasminogen activator was identified as a possible mediator of interest. Trial registration: This trial was registered at (identifier NCT01905033, date of registration July 23, 2013).

Original languageEnglish
Article number58
Number of pages20
JournalIntensive Care Medicine Experimental
Issue number1
Publication statusPublished - 2019
Externally publishedYes


  • Aging
  • Biomarker
  • Critical care, intensive care, ARDS, host response
  • Mediation
  • Outcome

Cite this