Dose-dependent association between blood transfusion and nosocomial infections in trauma patients: A secondary analysis of patients from the PAMPer trial.

Abstract

The Prehospital Air Medical Plasma (PAMPer) trial demonstrated a survival benefit to trauma patients who received thawed plasma as part of early resuscitation. The objective of our study was to examine the association between blood transfusion and nosocomial infections among trauma patients who participated in the PAMPer trial. We hypothesized that transfusion of blood products will be associated with the development of nosocomial infections in a dose-dependent fashion. We performed a secondary analysis of prospectively collected data of patients in the PAMPer trial with hospital length of stay of at least 3 days. Demographics, injury characteristics, and number of blood products transfused were obtained to evaluate outcomes. Bivariate analysis was performed to identify differences between patients with and without nosocomial infections. Two logistic regression models were created to evaluate the association between nosocomial infections and (1) any transfusion of blood products, and (2) quantity of blood products. Both models were adjusted for age, sex, and Injury Severity Score. A total of 399 patients were included: age, 46 years (interquartile range, 29-59 years); Injury Severity Score, 22 (interquartile range, 12-29); 73% male; 80% blunt mechanism; and 40 (10%) deaths. Ninety-three (27%) developed nosocomial infections, including pneumonia (n = 67), bloodstream infections (n = 14), catheter-associated urinary tract infection (n = 10), skin and soft tissue infection (n = 8), Clostridium difficile colitis (n = 7), empyema (n = 6), and complicated intra-abdominal infections (n = 3). Nearly 80% (n = 307) of patients received packed red blood cells (PRBCs); 12% received cryoprecipitate, 69% received plasma, and 27% received platelets. Patients who received any PRBCs had more than a twofold increase in nosocomial infections (odds ratio, 2.15; 95% confidence interval, 1.01-4.58; p = 0.047). The number of PRBCs given was also associated with the development of nosocomial infection (odds ratio, 1.10; 95% confidence interval, 1.05-1.16; p < 0.001). Trauma patients in the PAMPer trial who received a transfusion of at least 1 U of PRBCs incurred a twofold increased risk of nosocomial infection, and the risk of infection was dose dependent. Therapeutic/care management, level IV.