Premium
The effect of blood storage duration on in‐hospital mortality: a randomized controlled pilot feasibility trial
Author(s) -
Heddle Nancy M.,
Cook Richard J.,
Arnold Donald M.,
Crowther Mark A.,
Warkentin Theodore E.,
Webert Kathryn E.,
Hirsh Jack,
Barty Rebecca L.,
Liu Yang,
Lester Connie,
Eikelboom John W.
Publication year - 2012
Publication title -
transfusion
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.045
H-Index - 132
eISSN - 1537-2995
pISSN - 0041-1132
DOI - 10.1111/j.1537-2995.2011.03521.x
Subject(s) - medicine , randomized controlled trial , confidence interval , odds ratio , blood transfusion , surgery
BACKGROUND: Whether the duration of storage of blood has an impact on patient outcomes remains controversial. The objective was to determine feasibility of a comparative effectiveness trial to evaluate duration of storage of blood before transfusion on in‐hospital mortality. STUDY DESIGN AND METHODS: A single‐center randomized controlled trial was performed at an acute care hospital in Canada between June and December 2010, involving consecutive hospitalized patients needing blood transfusion. Patients (n = 910) were randomly assigned in a 1:2 ratio to receive freshest available versus standard‐issue (oldest available) blood. Four feasibility criteria were measured: proportion of eligible patients randomized, contrast in age of blood between treatment groups, real‐time data acquisition, and trial impact on blood outdating. In‐hospital mortality was also reported. RESULTS: A total of 1075 of 1129 patients (95.2%) were eligible and 910 of 1075 (84.7%) were randomized: 309 received freshest available blood (1157 units), and 601 received standard‐age blood (2369 units). Contrast in mean age of the oldest blood transfused between groups was 14.6 days: 12.0 (standard deviation [SD], 6.8) days in the fresh arm and 26.6 (SD, 7.8) days in the standard arm. Weekly recruitment and event reporting were achieved for all patients. The blood outdate rate was 0.10%. In‐hospital mortality was 10.5%: 35 deaths (11.3%) in the fresh arm and 61 deaths (10.1%) in the standard arm (odds ratio, 1.13; 95% confidence interval [CI], 0.73, 1.76). CONCLUSION: It is feasible to conduct a large comparative effectiveness trial comparing the effect of freshest available versus standard‐issue blood on in‐hospital mortality. The wide CI around the estimate for in‐hospital mortality supports the need for a large trial.