The care of patients with severe traumatic brain injury (TBI) is complex and confounded by uncertainty in prognoses. Studies have demonstrated significant unexplained variation in mortality between centers. Possible explanations include differences in the quality and intensity of care across centers, including the appropriateness and timing of withdrawal of life-sustaining therapies. We postulated that centers with a preponderance of early deaths might have a more pessimistic approach to the TBI patient, which would be reflected in an increased hospital TBI-related mortality.
This is a retrospective cohort study. Time to death was used as a proxy for time to withdrawal of life-sustaining therapies. Centers were classified as early or late based on when the majority (75th percentile) of their TBI-related deaths occurred. We evaluated the association between adjusted mortality and center classification using a hierarchical multivariable model. Two hundred trauma centers contributing data to the American College of Surgeons Trauma Quality Improvement Program from 2010 through 2013 were involved. The cohort included 17,505 patients with severe isolated TBI.
One hundred eight centers were classified as early centers. The 75th percentile for time to death was 4 days among early centers versus 7 days in late centers. Mortality was 34% and 33%, respectively. After adjustment for case mix, care in an early center was not associated with increased odds of death (adjusted odds ratio, 0.95; 95% confidence interval, 0.83–1.09). Higher odds of death were independently associated with age, Glasgow Coma Scale (GCS) score, head Abbreviated Injury Scale (AIS) score, multiple comorbidities, traumatic subarachnoid hemorrhage, intracerebral mass lesions, brainstem lesions, and signs of compressed or absent basal cisterns.
Centers rendering early decisions related to withdrawal of life-sustaining therapies in TBI patients, as measured by time until death, do not have worse outcomes than those making later decisions. How and when these decisions are made requires further exploration to balance an opportunity for clinical improvement with appropriate resource use.
Prognostic and epidemiologic study, level III.
From the Department of Critical Care Medicine (V.A.M., G.D.R., B.H.C., D.C.S.), Sunnybrook Research Institute (A.S.A., W.X., D.G.R., B.H.C., D.C.S., A.B.N.), and Department of Surgery (D.C.S.), Sunnybrook Health Sciences Center; and Department of Anesthesia (B.H.C.), University of Toronto, Toronto, Ontario, Canada.
Submitted: June 22, 2015, Revised: September 30, 2015, Accepted: October 16, 2015, Published online: November 21, 2015.
Parts of this study were presented as a poster at the Critical Care Canada Forum, October 29 to November 1, 2014, in Toronto, Ontario, Canada.
The opinions, results, and conclusions reported in this article are those of the authors. The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal's Web site (www.jtrauma.com).
Address for reprints: Victoria A. McCredie, MBChB, Room D108, Department of Critical Care Medicine, Sunnybrook Health Sciences Center, 2075 Bayview Ave, Toronto, Canada M4N 3M5; email: Victoria.McCredie@sunnybrook.ca.