Retrospective observational study: intra-hospital management and prognostic factors of septic patients from access to the emergency room to hospitalization
Background Despite the development of the Surviving Sepsis Campaign (SSC) guidelines, mortality from sepsis and septic shock remains unacceptably high. Sepsis affects 20-30 million people in the world, 1 in 4 does not survive. Aim of study is to identify possible prognostic indicators of severity and mortality and analysis of global intra-hospital mortality.
Methods It is a retrospective observational study: access of 386 adult patients (≥16 years old) to the Emergency Department of the Hospital of Padua between 1 January 2014 and 31 December 2017 with diagnosis of ""sepsis"", ""severe sepsis"" or ""septic shock"" according to ICD-9 coding, and hospitalization in internal medicine (MI-group A) or intensive care unit (TI-group B). Two groups are compared for analysis of vital parameters, qSOFA, history of fever, presentation symptoms, comorbidities, chronic therapies, current antibiotic treatment, execution of blood and / or urinary cultures; administration of antibiotic therapy in DEA. Statistical methods.
Results Age was significant, with a median of older and more comorbid patients (p = 0.0002) in MI. Regardless of the assigned color code (p = 0.0030), fever (p = 0-2620) and the detection of signs and symptoms at the admission, patients admitted to TI were significantly more hypotheses than those admitted to MI (PAS 60, PAM 64, p <.0001) with an SI of 1.2 vs. 0.9 (p <.0001). The evaluation of the vital parameters and laboratory tests, including PCR, Lactates, Procalcitonin, Leucocytes and qSOFA index (p = 0.0198) did not show a statistically significant difference in the choice of the Hospitalization Department. In group B, in the presence of signs of infection and hemodynamic instability, in addition to being required microbiological tests, antibiotic therapy was established early (73.68%, p < .0001) and amines (52.22%, p <.0001). Intra-day and 30-day mortality was not statistically significant (p = 0.2907, p = 0.7210, respectively).
With regard to overall mortality, 104 patients died during the period (26.94%), a value lower than that known in the literature. The multivariate analysis showed that the most significant predictors of mortality are the color code assigned to access in ps (cod / col G OR 2.5, cod / col R OR 10.53, p <.0001), lactic acid (≥4 OR 2.44, p = 0.0184) and the Shock Index (≥0.7 OR 4.0, p = 0.0395). In the univariate analysis, SI (p = 0.0230), early antibiotic therapy (p = 0.2167) and the need for amine administration (p = 0.0006) do not seem to be related to a poor prognosis.
Conclusions Clinical evaluation and indexes of hemodynamic instability (PAS, PAM, SI) are important in the choice of the Department of Hospitalization. Hemodynamic instable patients receive more emo-urocolture, antibiotic therapy and vasoactive amines in Emergency Room before being admitted to intensive care unit. Not statistically significant of mortality difference is consistent with the severity disease in TI and the age of patients in MI. In the study of global mortality we identify that the predictors of mortality (cod / col, lactates, SI) reflect the data from the literature.
