Skip to main content
. Author manuscript; available in PMC: 2012 Sep 26.
Published in final edited form as: Arch Intern Med. 2011 Sep 26;171(17):1587–1589. doi: 10.1001/archinternmed.2011.423

TABLE.

Relative Odds of Developing a Urinary Tract-related Bloodstream Infection By Type of Blood Component Received

Blood Component n Volume, Median [IQR], La Unadjusted
Adjustedb
OR 95% CI OR 95% CI
 RBCs (yes/no) 969 1.5 (0.7 – 3.2) 6.87 4.66, 10.12 4.91 2.91, 8.27
 RBCs without PLT or FFP (yes/no)c 695 0.7 (0.7 – 1.4) 3.76 2.23, 6.34 4.11 2.13, 7.92
 RBCs length of storage, wk 449 1.5 (0.7 – 3.2) 1.36 1.13, 1.63 1.40d 1.02, 1.93
 PLTs (yes/no) 969 0.7 (0.4 – 2.2) 5.94 4.05, 8.72 1.09 0.59, 2.01
 PLTs, L 969 0.7 (0.4 – 2.2) 1.74 1.45, 2.07 1.16 0.996, 1.35
 FFP (yes/no) 969 1.5 (0.6 – 3.1) 3.79 2.61, 5.51 1.52 0.86, 2.71
 FFP, L 969 1.5 (0.6 – 3.1) 1.41 1.24, 1.60 1.13 0.9995, 1.29

Abbreviations: CI, confidence interval; FFP, fresh-frozen plasma; IQR, interquartile range; OR, odds ratio; PLTs, platelets; RBCs, red blood cells.

a

Median volume and interquartile range for those who received the blood component.

b

Models included RBC, PLT, FFP, age (centered on mean), gender, race, medications (antibacterials, antimicrobials, antifungals, antivirals, immunosuppressants, insulin, statin), diabetes, cardiovascular disease, cancer, and surgery (cardiovascular, digestive tract, other).

c

Patients who received PLT or FFP were excluded.

d

Adjusted model also included volume (liters) of RBC units. Only subjects receiving RBCs were included.