Blood Product Components and Nosocomial Infection Risk in Trauma Patients: A Retrospective Study Using the NTDB

Author(s):
Jacob O’Dell; Christopher McCoy; Robert Winfield; Sue Min Lai; Edward Ellerbeck; Christopher Guidry

Background:

Packed red blood cell transfusion has been shown to increase nosocomial infection risk in the injured population; however, the infectious risk profiles of non-PRBC blood products in trauma are less clear.

Hypothesis:

We hypothesized that fresh frozen plasma (FFP), platelet (PLT), and cryoprecipitate administration would not predict greater risk for nosocomial infection.

Methods:

We performed a retrospective, matched, case-control study utilizing the American College of Surgeons National Trauma Data Bank (NTDB) data for 2019. We included all patients which received any volume of PRBC within 4 hours of presentation. Our outcome of interest was any infection. Controls were matched to cases using individual matching with a desired 1:3 case:control ratio. Univariate analysis, bivariate analysis according to infection status, and multi-variable logistic regression modeling the development of infection were then performed upon the matched data.

Results:

1,563 infectious cases were matched to 3,920 non-infectious controls. 1^st 4-hour transfusion volumes for FFP, PLT, and cryoprecipitate in the infection group exceeded those in the control group. The 1^st 4-hour FFP transfusion volume (per unit OR 1.02, 95% CI 0.99 – 1.04, p = 0.28) and cryoprecipitate transfusion volume (per unit OR 1.01 , 95% CI 0.99 – 1.02,  p = 0.43) did not predict infection. Platelet transfusion volume (per unit OR 0.92, 95% CI 0.86 – 0.98, p = 0.01) was protective for infection.

Conclusions:

Adjusting for confounding, we demonstrated that FFP, and cryoprecipiate transfusion volumes were not associated with the development of nosocomial infection in a trauma population. PLT transfusion volume was protective for infection.