Kelly BJ, et al. - Researchers compared nine biomarkers as predictors of mortality in subjects with clinically suspected bacterial sepsis. They found that 14-day and total mortality risk prediction was achieved by combined biomarkers. In the cohort studied, the best discriminatory ability was shown by serum amyloid P and tissue plasminogen activator.

Methods

• This cohort study was performed at the medical and surgical ICUs at an academic medical center including 139 subjects who met two or more systemic inflammatory response syndrome (systemic inflammatory response syndrome) criteria and received new broad-spectrum antibacterial therapy.
• At onset of suspected sepsis and 24, 48, and 72 hours thereafter, researchers assayed nine biomarkers (α-2 macroglobulin, C-reactive protein, ferritin, fibrinogen, haptoglobin, procalcitonin, serum amyloid A, serum amyloid P, and tissue plasminogen activator).
• Based on both 14-day and total in-hospital mortality, biomarkers between groups were compared.
• They also assessed the predictive validity of single and paired biomarkers via area under the receiver operating characteristic curve.

Results

• Data showed 14- day mortality was 12.9%, and total in-hospital mortality was 29.5%. 
• In the 14-day mortality group, they found significantly lower serum amyloid P (4/4 timepoints) and significantly higher tissue plasminogen activator (3/4 timepoints), and the same pattern held for total in-hospital mortality (Wilcoxon p ≤ 0.046 for all timepoints). 
• The best individual predictive performance for mortality was shown by serum amyloid P and tissue plasminogen activator, and greater predictive performance by achieved by combinations of biomarkers including serum amyloid P and tissue plasminogen activator (area under the receiver operating characteristic curve > 0.76 for 14-d and 0.74 for total mortality).