Utility of quantitative procalcitonin level and abbreviated MEDS scoring system as prognostic factors of sepsis patients in the emergency department

Choi, Jun Cheol; Yoon, Yoo Sang; Kim, Mi Ran; Kim, Yang Weon

Abstract

PURPOSE
Procalcitonin (PCT) is a newly introduced marker of systemic bacterial infection. The MEDS (Mortality in Emergency Department Sepsis) scoring system has proven to be the most useful and appropriate clinical prediction tool in cases of systemic bacterial infection in the emergency department. There have been no studies comparing the PCT assay with the abbreviated MEDS scoring system (without neutrophil bands). This study aimed to determine the efficacy of quantitative PCT assay and the abbreviated MEDS score in predicting the 28-day mortality and intensive care unit (ICU) admission in emergency department patients with systemic inflammatory response syndrome (SIRS).
METHODS
This study was carried out prospectively on 212 patients >18-years-of-age with the criteria of SIRS who had visited to the emergency department. They were included if they had an emergency department diagnosis of sepsis, two or more SIRS criteria. We checked the initial PCT concentration and calculated abbreviated MEDS score. The primary outcome was 28-day mortality and ICU admission. We evaluated the predictive abilities of the initial PCT assay and abbreviated MEDS score. Receiver operating characteristic (ROC) curves were used to identify the value of PCT and abbreviated MEDS scoring system that maximized the sum sensitivity and specificity.
RESULTS
Among the 212 patients included in this study, the mortality rate was 21%(45 of 212 patients) and the ICU admission rate was 33%(70 of 212 patients). Non-survivor group had a significantly higher PCT concentration (median, interquartile range (IQR), 12.23 ng/mL, 2.33-39.77 vs. 0.58 ng/mL, 0.10-3.60, p=0.05) and a higher abbreviated MEDS score (12.62+/-4.33 vs. 5.53+/-3.57, p<0.05) than the survival group. ICU admission patients had a significantly higher PCT concentration (median, IQR, 7.75 ng/mL, 1.43-36.85 vs. 0.55 ng/mL, 0.08-3.03 p<0.05) and a higher abbreviated MEDS score (10.90+/-4.60 vs. 5.13+/-3.48, p<0.05) than general ward admission patients. ROC analysis showed the discriminative power of the abbreviated MEDS score and PCT assay in predicting ICU admission and 28-day mortality. For ICU admission, areas of under the curves (AUC) of abbreviated MEDS score and PCT were 0.842 and 0.749, respectively (p<0.05). For PCT, AUC of abbreviated MEDS and PCT were 0.888 and 0.784, respectively (p<0.05). The optimal PCT threshold in predicting ICU admission was 1.22 ng/mL (odds ratio (OR), 6.92; 95% confidence interval (CI), 3.51-13.63; p<0.05) and 1.90 ng/mL in predicting 28-day mortality (OR, 11.90; 95% CI, 4.76-29.74). The optimal threshold of abbreviated MEDS score in predicting ICU admission was 8 (OR, 10.95; 95% CI, 5.47-21.90; p<0.05) and 9 in predicting 28 day mortality (OR, 19.03; 95% CI, 8.28-43.78; p<0.05).
CONCLUSION
Both abbreviated MEDS scoring system and single plasma PCT assays can be used as a useful marker in rapidly and accurately predicting poor prognosis in emergency department patients with sepsis.