Thrombo-inflammatory prognostic score improves qSOFA for risk stratification in patients with sepsis: a retrospective cohort study.

ABSTRACT

Background Both the thrombo-inflammatory prognostic score (TIPS) and the quick sequential (sepsis-related) organ failure assessment (qSOFA) are quick prognostic scores for sepsis during the early phase, while either of two scores has limited prognostic value for sepsis patients. This study aimed to evaluate whether TIPS adds more information of sepsis risk stratification for qSOFA. Methods This was a retrospective cohort study of patients with sepsis in the emergency department (ED). We performed a receiver-operating characteristic curve, integrated discrimination improvement (IDI), net reclassification improvement (NRI) and decision-curve analysis (DCA) analyses to investigate whether TIPS can improve qSOFA for risk prediction in patients with sepsis. The primary endpoint was mortality and the secondary endpoints were mechanical ventilation and admission to the intensive care unit (ICU) during the 28-day follow-up. Results We identified 821 patients with sepsis. We randomly assigned the patients' data to a derivation group (n = 498; n = 112 died during the 28-days follow-up) or to a validation group (n = 323; n = 61). The addition of TIPS to qSOFA (T-qSOFA) improved the area under the curve (AUC) from 0.724 to 0.824 (p < 0.001) for predicting 28-day mortality. The discrimination improvement was confirmed by an IDI of 0.092 (p < 0.001). Addition of TIPS to the qSOFA resulted in a NRI of 0.247 (p < 0.001). The DCA showed that the net benefit of T-qSOFA was higher than that of TIPS or qSOFA for any threshold probabilities. Conclusions The prognostic value of qSOFA for patients with sepsis was enhanced by adding the TIPS score on admission for risk prediction in patients with sepsis during early phases in the ED.