The dose-dependent relationship between blood transfusions and infections after trauma: A population-based study. Issue 1 (July 2020)

• Record Type: Journal Article
• Authors: Nederpelt, Charlie J.; El Hechi, Majed; Parks, Jonathan; Fawley, Jason; Mendoza, April E.; Saillant, Noelle; King, David R.; Fagenholz, Peter J.; Velmahos, George C.; Kaafarani, Haytham M.A.
• Is Part Of: Journal of trauma and acute care surgery. Volume 89:Issue 1(2020)
• Publication Date: 2020-07
• Subjects: Trauma resuscitation -- blood transfusion -- infectious complications; Surgical intensive care -- Periodicals; Surgical emergencies -- Periodicals; Wounds and injuries -- Surgery -- Periodicals; 617.026
• DOI: 10.1097/TA.0000000000002637
• Languages: English
• ISSNs: 2163-0755

Abstract

Abstract : OBJECTIVE: The relationship between total transfusion volume and infection in the trauma patient remains unclear, especially at lower volumes of transfusion. We sought to quantify the cumulative, independent impact of transfusion within 24 hours of admission on the risk of infection in trauma patients. METHODS: Using the Trauma Quality Improvement Program 2013 to 2016 database, we included all patients who received blood transfusions in the first 4 hours. Patients who were transferred or had incomplete/wrongly coded information on transfusion volume were excluded. Patients were divided into 20 cohorts based on the total blood product volume transfused in the first 24 hours. A composite infection variable (INF) was created, including surgical site infection, ventilator-associated pneumonia, urinary tract infection, central line associated blood stream infection, and sepsis. Univariate and stepwise multivariable logistic regression analyses were performed to study the relationship between blood transfusion and INF, controlling for demographics (e.g., age, sex), comorbidities (e.g., cirrhosis, diabetes, steroid use), severity of injury (e.g., vital signs on arrival, mechanism, Injury Severity Score), and operative and angiographic interventions. RESULTS: Of 1, 002, 595 patients, 37, 568 were included. The mean age was 42 ± 18.6 years, 74.6% were males, 68% had blunt trauma, and median Injury Severity Score was 25 [17–34]. Adjusting for all available confounders, odds of INF increased incrementally from 1.00 (reference, 0–2 units) to 1.23 (95% confidence interval, 1.11–1.37) for 4 units transfused to 4.89 (95% confidence interval, 2.72–8.80) for 40 units transfused. Each additional unit increased the odds of INF by 7.6%. CONCLUSION: Transfusion of the bleeding trauma patient was associated with a dose-dependent increased risk of infectious complications. Trauma surgeons and anesthesiologists should resuscitate the trauma patient until prompt hemorrhage control while avoiding overtransfusion. LEVEL OF EVIDENCE: Retrospective cohort study, Therapeutic IV. Abstract : Supplemental digital content is available in the text.