Pericardial-fluid Biomarkers for Post-mortem Sepsis Diagnosis: A Systematic Review of Diagnostic Utility, Matrix Comparisons, and Pre-analytical Influences

Abstract Sepsis is a major cause of natural death in medicolegal practice, yet its postmortem diagnosis remains challenging because macroscopic findings are often subtle, microbiological results may be equivocal, and blood-based biochemical analyses are frequently compromised by hemolysis, redistribution, and decomposition. When peripheral blood is unavailable or unsuitable, pericardial fluid represents a relatively protected compartment that can be reliably sampled at autopsy and may retain biochemical signals related to systemic infection. Although multiple inflammatory and sepsis-related biomarkers have been investigated in pericardial fluid, their diagnostic performance varies across analytes, and the influence of matrix effects and pre-analytical factors remains incompletely defined. A qualitative systematic review was conducted in accordance with PRISMA 2020 and SWiM guidelines to evaluate the diagnostic utility of pericardial-fluid biomarkers for postmortem sepsis, compare their performance with serum and other matrices, and assess relevant pre-analytical influences. MEDLINE (PubMed), Embase, Scopus, Web of Science, and the Cochrane Library were searched from inception without language or date restrictions, supplemented by targeted Google Scholar searches and reference screening. Eligible studies were human medicolegal autopsy investigations that measured at least one sepsis-related biomarker in pericardial fluid using composite reference standards. Risk of bias was assessed using a modified QUADAS-2 framework, and certainty of evidence was summarized narratively using a GRADE-style approach. Eight studies met the inclusion criteria, evaluating seven pericardial-fluid biomarkers: procalcitonin (PCT), C-reactive protein (CRP), sTREM-1, sIL-2R, presepsin, lipopolysaccharide-binding protein, and endocan. Pericardial-fluid PCT consistently demonstrated the most reliable discrimination between septic and non-septic deaths, with sensitivities and specificities approaching 90% and strong correlation with serum levels. CRP, sTREM-1, and sIL-2R showed limited and inconsistent diagnostic utility, while presepsin, lipopolysaccharide-binding protein, and endocan failed to distinguish septic from non-septic deaths. Overall certainty of evidence was moderate for pericardial PCT and low to very low for other biomarkers. Current evidence supports pericardial procalcitonin as the only pericardial-fluid biomarker with sufficient consistency and clinical relevance for postmortem sepsis diagnosis, particularly when blood is unavailable. Other biomarkers should be interpreted only as adjunctive findings, underscoring the need for larger, standardized autopsy studies.