Comparison of the diagnostic accuracies of various biomarkers and scoring systems for sepsis: A systematic review and Bayesian diagnostic test accuracy network meta-analysis

Abstract

Purpose: Sepsis affects approximately 50 million people worldwide, resulting in 11 million deaths annually. Conflicting results and insufficient evidence comparing performance biomarkers exist. The study aimed to comprehensively compare available biomarkers and clinical scores for detecting sepsis since its redefinition in 2016 with this systematic review and Bayesian diagnostic test accuracy network meta-analysis.

Materials and methods: We conducted searches in the PubMed, EMBASE, and Scopus databases between January 2016 and December 2023. Eligible studies assessed the diagnostic accuracies of biomarkers, the quick Sequential Organ Failure Assessment (qSOFA) score, or Systemic Inflammatory Response Syndrome (SIRS) criteria in detecting sepsis. Bivariate hierarchical random effects arm-based beta-binomial models were used for quantitative synthesis (PROSPERO Registration Number: CRD42018086545).

Results: We included 78 studies representing 34,234 patients and compared qSOFA score, SIRS criteria alongside seven of the most studied biomarkers: procalcitonin, C-reactive protein (CRP), interleukin-6 (IL-6), presepsin (cluster of differentiation 14 subtypes), CD64, soluble triggering receptor expressed on myeloid cells-1 (sTREM-1), and lipopolysaccharide-binding protein (LBP). CD64 demonstrated the highest superiority index, followed by sTREM-1 and presepsin (diagnostic odds ratio: 20.17 vs 18.73 and 10.04, 95 % credible interval [CrI]: 8.39-38.61 vs 1.31-83.98 and 6.71-14.24; quality of evidence: moderate vs low and low). Multivariable meta-regression analysis identified significant sources of heterogeneity, including study design, proportion of sepsis, sample size, and the risk of bias (patient selection).

Conclusions: The best diagnostic accuracy for sepsis was shown by CD64, with a moderate quality of evidence. Compared to CD64, sTREM-1 and presepsin provided suboptimal and low evidence. These biomarkers were more effective at identifying updated sepsis than clinical scores. We recommend re-considering the addition of biomarkers in screening for sepsis or sepsis-related conditions, as this could lead to more accurate and timely decisions for future clinical interventions.

Keywords: Biomarkers; Diagnostic; Network meta-analysis; SIRS; Sepsis; Systematic review; qSOFA.