Post-mortem sepsis diagnosis through vitreous humour biomarkers: A retrospective forensic cohort study

Background: Sepsis remains a challenging post-mortem diagnosis due to the nonspecificity of autopsy findings. This study evaluates five sepsis biomarkers (urea, high-sensitivity C-reactive protein, procalcitonin, lactic acid, and glucose) detected in vitreous humour. Methods: In this retrospective study, 127 vitreous humour samples were analysed. Septic deaths were defined by clinical, radiological, and autopsy evidence, serving as reference standard for case/control group identification. Concentrations of urea, hs-CRP, PCT, lactic acid, and glucose were measured using ELISA and compared between septic/non-septic groups. Marker stability was tested across five post-mortem interval groups. A systematic literature review following PRISMA guidelines was conducted to contextualise the findings within the existing forensic evidence on sepsis biomarkers. Results: Urea and hs-CRP levels were significantly higher in septic cases (p < 0.001). hs-CRP demonstrated superior discriminatory ability (AUC = 0.729) and specificity (88.4%) compared to urea (AUC = 0.686), which showed higher sensitivity (85.7%) in ROC analyses. Combined analysis yielded an AUC of 0.781. All biomarkers except glucose and lactic acid remained stable during post-mortem up to 187 h. Conclusions: Vitreous humour represents a promising matrix for post-mortem sepsis diagnosis. Urea and hs-CRP in vitreous humour demonstrated both discriminatory ability and post-mortem stability, making them valuable forensic tools, particularly when conventional matrices are unavailable. However, as a proof-of-concept study conducted in a selected forensic cohort with documented post-mortem intervals, these findings require validation in broader forensic scenarios. The demonstrated stability of urea, hs-CRP and PCT up to 187 h suggests potential applicability when the exact post-mortem interval is unknown.