Ex Vivo Endotoxin Stimulation of Blood for Predicting Survival in Patients With Sepsis: A Systematic Review

Abstract

Background: Sepsis is a syndrome characterized by host immune dysfunction, with the extent of immunoparalysis differing among patients. Lipopolysaccharide (LPS) is used commonly to assess the immune function of critically ill patients with sepsis. However, the reliability of this ex vivo diagnostic test in predicting clinical outcomes remains uncertain.

Research question: Does LPS-induced tumor necrosis factor (TNF) production from the blood of patients with sepsis predict mortality? Secondary outcomes included ICU and hospital stay durations, nosocomial infection rate, and organ recovery rate.

Study design and methods: Human sepsis studies from various databases through April 2023 were evaluated. Inclusion criteria encompassed LPS-stimulated blood assays, English language, and reported clinical outcomes. Bias risk was evaluated using the Newcastle-Ottawa scale (NOS). Relationships between TNF production and mortality were analyzed at sepsis onset and during established sepsis, alongside secondary outcomes.

Results: Of 11,580 studies, 17 studies (14 adult and three pediatric) were selected for analysis. Although 15 studies were evaluated as moderate to high quality using the NOS, it is important to note that some of these studies also had identifiable biases, such as unclear methods of participant recruitment. Nine studies detailed survival outcomes associated with LPS-induced TNF production at sepsis onset, whereas five studies explored TNF production's relationship with mortality during established sepsis. Trends suggested that lower LPS-induced TNF production correlated with higher mortality. However, heterogeneity in methodologies, especially the LPS assay protocol, hindered definitive conclusions. Publication bias was highlighted using funnel plot analysis. Concerning secondary outcomes, diminished TNF production might signify worsening organ dysfunction, although the link between cytokine production and nosocomial infection varied among studies.

Interpretation: For functional immune profiling in sepsis, streamlined research methodologies are essential. This entails organizing cohorts based on microbial sources of sepsis, establishing standardized definitions of immunoparalysis, using consistent types and dosages of immune stimulants, adhering to uniform blood incubation conditions, and adopting consistent clinical outcomes.

Keywords: clinical outcomes; endotoxin; ex vivo stimulation; mortality; sepsis.

Figure 1 –

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart depicting the study selection process. LPS = lipopolysaccharide.

Figure 2 –

A, B, Forest plots illustrating the correlation between lipopolysaccharide-induced tumor necrosis factor concentrations (measured in picograms per milliliter) and mortality at the time of sepsis onset (A) and during established sepsis (B). SMD = standardized mean difference.

Figure 3 –

A, B, Funnel plots illustrating the distribution of study effect sizes in relationship to their SEs, used to evaluate potential publication bias: studies describing the relationship between lipopolysaccharide-induced tumor necrosis factor concentrations and mortality at the time of sepsis onset (A) and during established sepsis (B). Filled circles represent observed studies with real data points, and hollow circles represent hypothetical missing studies that would need to be included to give an adjusted estimate that is potentially less biased than the original.