Rev Esp Quimioter 2018; 31(4): 316-322

Assessment of the severity scores in patients included in a sepsis code in an Emergency Departament

ALBERTO REDONDO-GONZÁLEZ, MARÍA VARELA-PATIÑO, JESÚS ÁLVAREZ-MANZANARES, JOSÉ RAMÓN OLIVA-RAMOS, RAÚL LÓPEZ-IZQUIERDO, CARMEN RAMOS-SÁNCHEZ, JOSÉ MARÍA EIROS

Objectives. The objective of the study is to determine the usefulness of the SOFA (Sequential Organ Failure Assessment), quick SOFA (qSOFA), LODS (Logistic Organ Dysfunction System) and EWS (Early Warning Score) scores to predict in-hospital mortality among septic patients attended in the emergency department; to evaluate what factors are associated with mortality; and develop a predictive model of in-hospital mortality.
Material and methods. Retrospective study including patients over 14 years of age included in the sepsis code of an Emergency Department of a University Hospital between November 2013 and September 2015. Demographic variables, hemodynamic and analytical variables, and in-hospital mortality were collected to obtain qSOFA, SOFA, LODS, EWS scores. Receiver operating characteristic curves were constructed for each score. Logistic regression was used to evaluate the probability of in-hospital mortality.
Results. A total of 349 patients were analyzed, median age 72.7 (range 86), males: 54.4%. The in-hospital mortality was 21.8%. AUC obtained: LODS: 0.73 (IC 95% 0.67-0.80; p<0.001), EWS: 0.73 (IC 95% 0.65-0.81; p<0.001), SOFA: 0.72 (IC 95% 0.65- 0.78; p<0.001), qSOFA: 0.67 (IC 95% 0.58-0.76; p<0.001). After the multivariate analysis, these were the independent factors associated with in-hospital mortality: Oxygen saturation ≤92%, Glasgow coma score <14, lactate ≥2mmol/L (p<0.05). Two prognostic models were generated: MPRO1: age, oxygen saturation ≤92% and Glasgow coma score <14, AUC: 0.78 (IC 95% 0.72-0.84; p<0.001) and MPRO2 formed by the previous ones and lactate ≥2mmol/L, AUC: 0.82 (IC 95% 0.76-0.87; p<0.001)
Conclusions. SOFA score and the new developed scores could be useful in asses the risk of in-hospital mortality in patients included in the sepsis code.

Rev Esp Quimioter 2018; 31(4): 316-322 [Texto completo PDF]