Impact of Clinical Sepsis Phenotypes on Mortality and Fluid Status in Critically Ill Patients.

Abstract

Sepsis is associated with high rates of in-hospital mortality, despite being the focus of medical research and public health initiatives for several years. The primary objective of this study was to determine the influence of septic phenotypes on rates of in-hospital mortality throughout intensive care unit (ICU) admission. Retrospective, single-center cohort study. Medical ICU of an academic medical center. Medical ICU patients admitted between January 2016 and August 2019 with a "sepsis alert" were screened for admitting diagnosis of "sepsis" or "septic shock." Patients were classified into one of four clinical sepsis phenotypes: multi-organ failure (MOF), respiratory dysfunction (RD), neurologic dysfunction (ND), or other patients (OP). An analysis of 320 patients was completed. In-hospital mortality was different between groups (P < 0.001). Patients with the MOF phenotype had the highest rate of mortality (48.4%), followed by the ND phenotype (39.7%), RD phenotype (20.8%), and OP phenotype (13.7%). There were differences in volume balances between phenotypes at 48 h (P = 0.001), 72 h (P < 0.001), and 96 h (P < 0.001) after hospital presentation, with the MOF and ND phenotypes having the largest volume balances at these time points. Ventilator-free days (P < 0.001) and ICU length of stay (LOS) (P = 0.030) were different between groups. There was no difference in hospital LOS (P = 0.479). This data supports the presence of marked intra-disease differences in septic patient presentation and correlation with clinical outcomes including mortality. Additionally, significantly more positive fluid balances were observed between survivors and non-survivors in some patient subsets. Using pragmatic clinical variables readily available to providers to classify patients into septic phenotypes has the propensity to guide treatment strategies in the future.