P2737
The assessment of SOFA score predicts mortality and neurogenic outcome in Post-Cardiac Arrest Syndrome patients

Background: The prognosis of survival in patients with out-of-hospital cardiac arrest remains poor. Post-cardiac arrest syndrome includes sepsis-like status and often exhibit multi organ failure in intensive care units (ICU). Some studies reported Sequential Organ Failure Assessment (SOFA) score predicts clinical outcome of patients who need intensive care such as sepsis. However, few reports that evaluated the usefulness of SOFA score for Post-cardiac arrest patients have been published.

Purpose: We sought to assess SOFA score as an indicator of multi organ failure for prediction of mortality and neurogenic outcome in post-cardiac arrest patients.

Methods: 654 out-of-hospital cardiac arrest patients (Ventricular fibrillation (VF)/ pulseless Ventricular tachycardia (VT) 113 (17.3%), Pulseless electrical activity (PEA) 205 (31.3%), Asystole 336 (51.4%)) were transferred to our critical care center. We included 146 cardiogenic arrest patients (126 patients got return of spontaneous circulation (ROSC)) from the institutional consecutive database from Jan 2016 to Dec 2017. To predict mortality in hospital and neurogenic outcome (cerebral-performance-category (CPC)) a month later, we investigated basic characteristics, out-of-hospital information, SOFA score on admission and 48 hours later in ICU.

Results: VF survivor (P=0.031), ROSC on ambulance (P<0.001), emergent coronary angiography (P=0.047), hypothermia therapy (P=0.019) and low SOFA score on admission (P<0.001) have low mortality in hospital a month later. In multivariate analysis, SOFA score on admission (OR 0.71; 95% CI 0.61–0.82; P<0.001) and ROSC on ambulance (OR 4.41; 95% CI 1.74–11.2; P<0.001) was predictive for mortality. On the other hand, ROSC on ambulance (P<0.001) and low SOFA score on admission (P<0.001) had good neurogenic outcome (CPC=1 or 2) a month later. SOFA score on admission (OR 0.79; 95% CI 0.69–0.91; P<0.001) and ROSC on ambulance (OR 9.17; 95% CI 3.20–26.3; P<0.001) was also predictive for CPC=1 or 2. Furthermore, 87 patients survived more than 48 hours in hospital. Among these patients, increase in SOFA score 48 hours later was related to mortality (P<0.001). Increase in SOFA score was most predictive for mortality after admission (OR 1.79; 95% CI 1.36–2.36; P<0.001). Cut-off-value of the change of SOFA score 48 hours later to predict mortality was 1.0 point worse than that on admission (AUC 0.843; 95% CI 0.76–0.93).

Conclusions: The evaluation of SOFA score in ICU is useful to predict mortality and neurogenic outcome in post-cardiac arrest patients.