A multiplex cytokine score for the prediction of disease severity in pediatric hematology/oncology patients with septic shock

• We develop a multiplex cytokine model named PMCS to evaluate septic shock.
• PMCS integrates four inflammatory cytokines: IL-6, IL-10, TNF-α and IFN-γ.
• PMCS correlates well with hemodynamic disorder and mortality in septic shock.
• PMCS improves severity and mortality prediction compared with individual cytokines.
• PMCS presents similar discriminative power on fatal outcome with PIM2.

Although many inflammatory cytokines are prognostic in sepsis, the utility of cytokines in evaluating disease severity in pediatric hematology/oncology patients with septic shock was rarely studied. On the other hand, a single particular cytokine is far from ideal in guiding therapeutic intervention, but combination of multiple biomarkers improves the accuracy. In this prospective observational study, 111 episodes of septic shock in pediatric hematology/oncology patients were enrolled from 2006 through 2012. Blood samples were taken for inflammatory cytokine measurement by cytometric bead array (CBA) technology at the initial onset of septic shock. Interleukin (IL)-6 and IL-10 were significantly elevated in majority of patients, while tumor necrosis factor (TNF)-α and interferon (IFN)-γ were markedly increased in patients with high pediatric index of mortality 2 (PIM2) score and non-survivors. All the four cytokines paralleled the PIM2 score and differentially correlated with hemodynamic disorder and fatal outcomes. The pediatric multiplex cytokine score (PMCS), which integrated the four cytokines into one score system, was related to hemodynamic disorder and mortality as well, but showed more powerful prediction ability than each of the four cytokines. PMCS was an independent predictive factor for fatal outcome, presenting similar discriminative power with PIM2, with accuracy of 0.83 (95% CI, 0.71–0.94). In conclusion, this study develops a cytokine scoring system based on CBA technique, which performs well in disease severity and fatality prediction in pediatric hematology/oncology patients with septic shock.