M. Malig1, C. N. Jenne1, D. J. Roberts1, C. G. Ball1, Z. Xiao1, A. W. Kirkpatrick1 1University Of Calgary,Calgary, AB, Canada
Introduction: To date no single biomarker of systemic inflammation clearly predicts either outcomes or monitors therapy responses. However, biomediator panels may provide a more comprehensive model of patient variation, and may reveal novel underlying relationships between mediators. Selection of open abdomen (OA) management of critically ill patients practically designates a critically ill population with markedly increased complexity and potentially catastrophic complications. The randomized-controlled human trial (Peritoneal-VAC) examining active negative pressure peritoneal therapy provided a rich repository of biomediator samples in such an OA population, although no individual biomarker assessed was predictive of survival. Thus, a novel biomediator panel was created and tested using multivariate analysis techniques in an attempt to examine predictive values in this extreme cohort.
Methods: Partial least squares regression with discriminant analysis (PLS-DA) was used to develop a biomediator panel modeled at baseline, 24, and 48 hours after OA management using Peritoneal-VAC plasma samples. Eight cytokines/acute phase proteins (APP) were used to generate an eight-component biomediator panel to predict overall patient outcome. Components for the panel were selected based on literature research and included: TNF-α, IL-1β, IL-4, IL-6, IL-7, IL-8, IL-10, and fibrinogen.
Results: The generated models had R2 values ranging from 0.213-0.541, whereas Q2 ranged from 0.007-0.13. The biomediator panel modeled at 24 hours was most predictive for patient outcome (Q2 = 0.13), and accounted for over 50% of the observed experimental variance.
Conclusion: The proposed biomediator panel was more predictive of patient outcome when modeled at certain time points than others; however, overall the panel was a poor predictor of patient outcome. Thus, more research is required to determine more informative biomediators and the most appropriate time frame in which the biomediator storm may best predict outcome in the critically ill/injured.