The objectives of this study were to 1) assess patterns of early crystalloid resuscitation provided to sepsis and septic shock patients at initial presentation and 2) determine the association between time to initial crystalloid resuscitation with hospital mortality, mechanical ventilation, ICU utilization, and length of stay.
Consecutive-sample observational cohort.
Nine tertiary and community hospitals over 1.5 years.
Adult sepsis and septic shock patients captured in a prospective quality improvement database inclusion criteria: suspected or confirmed infection, greater than or equal to two systemic inflammatory response criteria, greater than or equal to one organ-dysfunction criteria.
The primary exposure was crystalloid initiation within 30 minutes or lesser, 31–120 minutes, or more than 120 minutes from sepsis identification.
We identified 11,182 patients. Crystalloid initiation was faster for emergency department patients (β, –141 min; CI, –159 to –125; p < 0.001), baseline hypotension (β, –39 min; CI, –48 to –32; p < 0.001), fever, urinary or skin/soft-tissue source of infection. Initiation was slower with heart failure (β, 20 min; CI, 14–25; p < 0.001), and renal failure (β, 16 min; CI, 10–22; p < 0.001). Five thousand three hundred thirty-six patients (48%) had crystalloid initiated in 30 minutes or lesser versus 2,388 (21%) in 31–120 minutes, and 3,458 (31%) in more than 120 minutes. The patients receiving fluids within 30 minutes had lowest mortality (949 [17.8%]) versus 31–120 minutes (446 [18.7%]) and more than 120 minutes (846 [24.5%]). Compared with more than 120 minutes, the adjusted odds ratio for mortality was 0.76 (CI, 0.64–0.90; p = 0.002) for 30 minutes or lesser and 0.76 (CI, 0.62–0.92; p = 0.004) for 31–120 minutes. When assessed continuously, mortality odds increased by 1.09 with each hour to initiation (CI, 1.03–1.16; p = 0.002). We observed similar patterns for mechanical ventilation, ICU utilization, and length of stay. We did not observe significant interaction for mortality risk between initiation time and baseline heart failure, renal failure, hypotension, acute kidney injury, altered gas exchange, or emergency department (vs inpatient) presentation.
Crystalloid was initiated significantly later with comorbid heart failure and renal failure, with absence of fever or hypotension, and in inpatient-presenting sepsis. Earlier crystalloid initiation was associated with decreased mortality. Comorbidities and severity did not modify this effect.
Supplemental Digital Content is available in the text.
1Department of Emergency Medicine, Hofstra-Northwell School of Medicine, Hempstead, NY.
2Emergency Medicine Service Line, Northwell Health, New Hyde Park, NY.
3Icahn School of Medicine at Mount Sinai, New York, NY.
4Department of Medicine, Northwell Health System, Hofstra-Northwell School of Medicine, Manhasset, NY.
5Department of Science Education, Hofstra-Northwell School of Medicine, Hempstead, NY.
6Kransoff Quality Mangement Institute, Northwell Health, New Hyde Park, NY.
7Division of Pulmonary, Critical Care, and Sleep Medicine, Northwell Health System, Hofstra-Northwell School of Medicine, Manhasset, NY.
*See also p. 1773.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal).
Supported, in part, by a grant from the Center for Medicare and Medicaid Innovation to the High Value Healthcare Collaborative, of which the study sites’ parent health system was a part. Although this investigation did not receive extramural funding directly, this grant helped fund the underlying quality improvement program and database used to conduct this study.
Dr. Masick disclosed that he owns a statistical consulting company providing service to students in the form of tutoring, methodological, and statistical coaching; and he receives royalties for a published textbook. Dr. D'Amore’s institution received funding from Dartmouth High Value Care Collaborative. The remaining authors have disclosed that they do not have any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org