Performance of Sepsis Scoring Systems in a developing country ­ a prospective cohort study

ABSTRACT

Background: Sepsis and its sequelae poses a significant socio-economic burden on health care systems globally. Risk stratification plays a fundamental role in emergency department management, since early and aggressive management in high-risk cohorts leads to improved outcomes. Several risk stratification tools exist but in the local setting (developing country with high chronic disease burden) there is no standardised recommendation for beside utilisation. Objectives: We aimed to compare the ability of the quick Sepsis-related Organ Failure Assessment (qSOFA) score with the Systemic Inflammatory Response Syndrome (SIRS) criteria and National Early Warning Score (NEWS) to detect and risk stratify patients with presumed sepsis outside of the intensive care unit (ICU). Methods: A prospective observational cohort study was conducted at a public tertiary hospital during the period May to June 2017. Ethical and institutional approval was secured and informed consent was sought from study participants aged eighteen (18) years and older. Demographic and clinical data were collected via a data collection instrument and statistical analysis was undertaken using IBM SPSS v23. Results: 304 patients were treated for presumed sepsis. The primary outcomes of in-hospital death or intensive care unit admission were seen in 14.8%. Discrimination for the primary outcome was highest for NEWS (AUROC 0.88 [95% CI 0.83-0.94]) followed by qSOFA (AUROC 0.82 [95% CI 0.74-0.89]) and SIRS (AUROC 0.69 [95% CI 0.61-0.77]). A NEWS value of ≥4 resulted in a sensitivity of 93.3%, and negative predictive value of 98.3% (p<0.001). A qSOFA score ≥2 demonstrated a specificity of 94.6 % and a negative predictive value of 91.4% (p<0.001). A SIRS criteria score ≥2 resulted in a sensitivity of 88.9%, and a negative predictive value of 95.0% (p=0.001). Univariate analysis showed that need for supplemental oxygen, an oxygen saturation less than 91%, a Glasgow Coma Scale <15 and non-selfpresentation were associated with the highest odds ratios for death in-hospital or ICU admission. Conclusions: Urgent identification of high-risk patients with presumed infection is critical in achieving a positive outcome. NEWS was superior to both qSOFA AND SIRS in predicting in-hospital mortality and need for ICU admission A qSOFA score ≥2 demonstrated a high specificity but poor sensitivity, thus limiting its use as a bedside tool. The findings of this study are consistent with the Sepsis-3 guidelines, which recommend qSOFA as being superior to SIRS criteria. However, we found that NEWS had a superior predictive value to both. Its role in the identification of high-risk subjects should be further evaluated.