A multiplex analysis of sepsis mediators during human septic shock: a preliminary study on myocardial depression and organ failures

Abstract

Background: The mechanisms of organ failure during sepsis are not fully understood. The hypothesis of circulating factors has been suggested to explain septic myocardial dysfunction. We explored the biological coherence of a large panel of sepsis mediators and their clinical relevance in septic myocardial dysfunction and organ failures during human septic shock.

Methods: Plasma concentrations of 24 mediators were assessed on the first day of septic shock using a multi-analyte cytokine kit. Septic myocardial dysfunction and organ failures were assessed using left ventricle ejection fraction (LVEF) and the Sequential Organ Failure Assessment score, respectively.

Results: Seventy-four patients with septic shock (and without immunosuppression or chronic heart failure) were prospectively included. Twenty-four patients (32%) had septic myocardial dysfunction (as defined by LVEF < 45%) and 30 (41%) died in ICU. Hierarchical clustering identified three main clusters of sepsis mediators, which were clinically meaningful. One cluster involved inflammatory cytokines of innate immunity, most of which were associated with septic myocardial dysfunction, organ failures and death; inflammatory cytokines associated with septic myocardial dysfunction had an additive effect. Another cluster involving adaptive immunity and repair (with IL-17/IFN pathway and VEGF) correlated tightly with a surrogate of early sepsis resolution (lactate clearance) and ICU survival.

Conclusions: In this preliminary study, we identified a cluster of cytokines involved in innate inflammatory response associated with septic myocardial dysfunction and organ failures, whereas the IL-17/IFN pathway was associated with a faster sepsis resolution and a better survival.

Keywords: Biomarkers; Mortality; Myocardial depression; Septic shock.

Fig. 1

Hierarchical clustering of all sepsis mediators. The parameters were reordered using computerized hierarchical clustering with the corrplot package of R statistical environment. Hierarchical clustering is a statistical method for finding comparatively homogenous clusters of cases based on measured characteristics. The algorithm uses a set of dissimilarities or distances between cases when constructing the clusters and proceeds iteratively to join the most similar cases. Distances between clusters were recomputed by the Lance–Williams dissimilarity update formula according to the complete linkage method

Fig. 2

Forest plot for odds ratios (a) and focused principal component analysis (FPCA, b) for the association between sepsis mediators and septic myocardial dysfunction. FPCA is a simple graphical display of correlation structures focusing on a particular dependent variable. The display reflects primarily the correlations between the dependent variable and all other variables (covariates) and secondarily the correlations among the covariates. The dependent variable (septic myocardial dysfunction, SMD) is at the center of the diagram, and the distance of this point to a covariate faithfully represents their pairwise Spearman correlation coefficient (using ranked values of continuous variables). Green covariates are positively correlated with the dependent variable. For parsimony, only covariates significantly correlated with the dependent variable (with a p value < 0.05 and inside the red circle) are displayed. The diagram also shows relationships between covariates as follows: Correlated covariates are close (for positive correlations, allowing identification of clusters) or diametrically opposite vis-à-vis the origin (for negative correlations), whereas independent covariates make a right angle with the origin. See text for sepsis mediators’ abbreviations

Fig. 3

Prevalence of septic myocardial dysfunction according to the number of increased sepsis mediators (above the median value) from the first cluster

Fig. 4

Forest plot for odds ratios (a) and focused principal component analysis (FPCA, b) for the association between sepsis mediators and intensive care unit mortality. b Correlation of ICU mortality (dependent variable at the center) with sepsis mediators at day 1, patient’s severity (as assessed by SAPS II score), organ dysfunction (as assessed by SOFA score at septic shock onset) and lactate clearance (relative difference between lactate concentration at septic shock onset and after 24 h of resuscitation) as a surrogate of early sepsis resolution. See Fig. 2 legend for details on FPCA. Variables positively and negatively correlated with ICU mortality are in green and yellow, respectively. See text for sepsis mediators’ abbreviations