Do Sepsis-3 Criteria Facilitate Earlier Recognition of Sepsis and Septic Shock? A Retrospective Cohort Study.

BACKGROUND: New Sepsis-3 criteria are supposed to "facilitate earlier recognition … of patients with sepsis." To test this, we performed novel and direct comparisons of Sepsis-1 vs. Sepsis-3 criteria with respect to time differences of sepsis onset. METHODS: In a cohort of intensive care unit (ICU) patients prospectively diagnosed with severe sepsis or septic shock according to Sepsis-1 criteria between 01/2010 and 12/2015, the time differences between meeting Sepsis-1 vs. Sepsis-3 criteria as time of sepsis onset and the corresponding differences in illness severity were tested. Similar comparisons were performed for septic shock subset meeting different Sepsis-1 vs. Sepsis-3 criteria. Patients with non-ICU-acquired sepsis and patients with sepsis onset more than 48 h postadmission (ICU-acquired sepsis) were analyzed separately to account for differences in availability of routinely collected organ dysfunction data. RESULTS: A total of 10,905 ICU patients were screened; 862 patients met Sepsis-1 criteria, of whom 834 (97%) also met Sepsis-3 criteria. In patients, admitted to the ICU with sepsis, Sepsis-3 criteria compared with Sepsis-1 criteria were more frequently fulfilled within the first 3 h (84% vs. 75%, P < 0.001).In patients with ICU-acquired sepsis, sepsis onset was in 50% at least 1 day earlier after application of Sepsis-3 (P = 0.011). These patients were systemic inflammatory response syndrome negative at the earlier sepsis onset, but suffered already from organ dysfunction. Sepsis-3 criteria were timely in 86% and 1 day delayed in 7%. Only 7% (8 patients) did not meet Sepsis-3 criteria in this group. These patients had already an increased SOFA score and did develop neither a further increase nor the new septic shock criteria. Classification according to Sepsis-3 reduced the proportion of septic shock (51% vs. 75%, P < 0.001).Twenty-eight-day mortality was 38% for new septic shock compared with 33% of Sepsis-1 septic shock (P > 0.05). Patients not detected by Sepsis-3 had a 28-day mortality of 11%. CONCLUSIONS: Sepsis-3 criteria facilitate an earlier and more predictive recognition of sepsis and septic shock in patients with non-ICU and ICU-acquired sepsis primarily diagnosed by Sepsis-1 criteria. These results require further validation with prospectively collected data.