Objectives: To find and validate generalizable sepsis subtypes using data-driven clustering. Design: We used advanced informatics techniques to pool data from 14 bacterial sepsis transcriptomic datasets from eight different countries (n = 700). Setting: Retrospective analysis. Subjects: Persons admitted to the hospital with bacterial sepsis. Interventions: None. Measurements and Main Results: A unified clustering analysis across 14 discovery datasets revealed three subtypes, which, based on functional analysis, we termed “Inflammopathic, Adaptive, and Coagulopathic.” We then validated these subtypes in nine independent datasets from five different countries (n = 600). In both discovery and validation data, the Adaptive subtype is associated with a lower clinical severity and lower mortality rate, and the Coagulopathic subtype is associated with higher mortality and clinical coagulopathy. Further, these clusters are statistically associated with clusters derived by others in independent single sepsis cohorts. Conclusions: The three sepsis subtypes may represent a unifying framework for understanding the molecular heterogeneity of the sepsis syndrome. Further study could potentially enable a precision medicine approach of matching novel immunomodulatory therapies with septic patients most likely to benefit.

Improved risk stratification and prognosis prediction in sepsis is a critical unmet need. Clinical severity scores and available assays such as blood lactate reflect global illness severity with suboptimal performance, and do not specifically reveal the underlying dysregulation of sepsis. Here, we present prognostic models for 30-day mortality generated independently by three scientific groups by using 12 discovery cohorts containing transcriptomic data collected from primarily community-onset sepsis patients. Predictive performance is validated in five cohorts of community-onset sepsis patients in which the models show summary AUROCs ranging from 0.765–0.89. Similar performance is observed in four cohorts of hospital-acquired sepsis. Combining the new gene-expression-based prognostic models with prior clinical severity scores leads to significant improvement in prediction of 30-day mortality as measured via AUROC and net reclassification improvement index These models provide an opportunity to develop molecular bedside tests that may improve risk stratification and mortality prediction in patients with sepsis.

The relationships between cytokine responses in septic shock are currently poorly understood. Some studies have pointed to a biphasic model, with an initial proinflammatory phase, followed by a reactive, anti-inflammatory response to explain the pathogenesis of the most severe form of sepsis. However, evidence for the coexistence of both responses has been found. In this study, the plasma levels of 17 cytokines and chemokines, in 20 patients with septic shock, 11 patients with systemic inflammatory response syndrome (SIRS), during the first 24 hours following diagnosis, and 10 healthy controls, were analyzed and compared. Patients with septic shock showed increased levels of IL-6, IL-8, MCP-1, MIP-1β, IFN-γ, GM-CSF and IL-10 compared to healthy controls. Patients with SIRS showed higher levels of IL-6, IL-8, MCP-1, MIP-1β, G-CSF and IL-10 than controls. Patients with septic shock showed higher levels of IL-8, GM-CSF, MIP-1β than those with SIRS. The Spearman test demonstrated a positive association between the pro-inflammatory mediators IL-6, IL-8, MCP-1, MIP-1β, IFN-γ, GM-CSF and the immunomodulatory cytokine IL-10 in septic shock. Consequently, correlation studies supported the notion that secretion of pro- and anti-inflammatory mediators in septic shock occurs as a simultaneous immune response program initiated early in the course of the disease, revealing that both types of cytokine play a role from the very beginning of this life-threatening condition.

Introduction: Host immunity should play a principal role in determining both the outcome and recovery of patients with sepsis that originated from a microbial infection. Quantification of the levels of key elements of the immune response could have a prognostic value in this disease. Methods: In an attempt to evaluate the quantitative changes in the status of immunocompetence in severe sepsis over time and its potential influence on clinical outcome, we monitored the evolution of immunoglobulins (Igs) (IgG, IgA and IgM), complement factors (C3 and C4) and lymphocyte subsets (CD4+ T cells, CD8+ T cells, B cells (CD19+) and natural killer (NK) cells (CD3-CD16+CD56+)) in the blood of 50 patients with severe sepsis or septic shock at day 1, day 3 and day 10 following admission to the ICU. Results: Twenty-one patients died, ten of whom died within the 72 hours following admission to the ICU. The most frequent cause of death (n = 12) was multiorgan dysfunction syndrome. At day 1, survivors showed significantly higher levels of IgG and C4 than those who ultimately died. On the contrary, NK cell levels were significantly higher in the patients who died. Survivors exhibited a progressive increase from day 1 to day 10 on most of the immunological parameters evaluated (IgG, IgA, IgM, C3, CD4+, CD8+ T cells and NK cells). Multivariate Cox regression analysis, including age, sex, APACHE II score, severe sepsis or septic shock status and each one of the immunological parameters showed that NK cell counts at day 1 were independently associated with increased risk of death at 28 days (hazard ratio = 3.34, 95% CI = 1.29 to 8.64; P = 0.013). Analysis of survival curves provided evidence that levels of NK cells at day 1 (> 83 cells/mm³) were associated with early mortality. Conclusions: Our results demonstrate the prognostic role of NK cells in severe sepsis and provide evidence for a direct association of early counts of these cells in blood with mortality.