Resuscitation in the First 3 Hours of Sepsis-Induced Hypotension Varies by Patient and Hospital Factors

IMPORTANCE:. Patient and hospital factors affects how we resuscitate patients in the first 3 hours of sepsis-induced hypotension. OBJECTIVES:. To evaluate variability in compliance to the 3-hour surviving sepsis campaign (SSC) bundle and explore the association of early compliance with subsequent shock and in-hospital mortality. DESIGN:. Retrospective cohort study between September 2017 and February 2018. SETTING:. Thirty-four academic medical centers. PARTICIPANTS:. A subgroup sepsis-induced hypotensive patients from a larger shock cohort study. MAIN OUTCOMES AND MEASURES:. Compliance to SSC bundle that was defined as receiving appropriate antibiotics, 30 mL/kg of crystalloid or initiation of vasopressors, and lactate, obtained in the first 3 hours following sepsis-induced hypotension. RESULTS:. We included 977 patients with septic-induced hypotension. Bundle compliance was 43.8%, with the lowest compliance to fluid or vasopressor components (56%). Patients with high Sequential Organ Failure Assessment scores and physiologic assessments were more likely to receive compliant care, as were patients with sepsis-induced hypotension onset in the emergency department (ED) or admitted to mixed medical-surgical ICUs. SSC compliance was not associated with in-hospital mortality (adjusted odds ratio, 0.72; 95% CI, 0.47–1.10). The site-to-site variability contributed to SSC compliance (intraclass correlation coefficient [ICC], 0.15; 95% CI, 0.07–0.3) but not in-hospital mortality (ICC, 0.02; 95% CI, 0.001–0.24). Most patients remained in shock after 3 hours of resuscitation (SSC compliant 81.1% and noncompliant 53.7%). Mortality was higher among patients who were persistently hypotensive after 3 hours of resuscitation for both the SSC compliant (persistent hypotension 37% vs not hypotensive 27.2%; p = 0.094) and noncompliant (30.1% vs 18.2%; p = 0.001, respectively). CONCLUSIONS AND RELEVANCE:. Patients with a higher severity of illness and sepsis-induced hypotension identified in the ED were more likely to receive SSC-compliant care. SSC compliance was not associated with in-hospital mortality after adjusting for patient- and hospital-level differences. Higher mortality is seen among those who remain in shock after initial resuscitation, regardless of SSC compliance