• Stacey L Valentine, Jenifer R Lightdale, Chau M Tran, Hongyu Jiang, Steven R Sloan, Monica E Kleinman, and Adrienne G Randolph.
• 1Critical Care Division, Department of Anesthesiology, Perioperative and Pain Medicine, Boston Children's Hospital, Boston, MA. 2Harvard Medical School, Boston, MA. 3Department of Pediatrics, Division of Pediatric Critical Care, University of Massachusetts Children's Medical Center, Worcester, MA. 4Department of Medicine, Boston Children's Hospital, Boston, MA. 5Clinical Research Center, Boston Children's Hospital, Boston, MA. 6Department of Laboratory and Transfusion Medicine, Boston Children's Hospital, Boston, MA. 7Program for Patient Safety and Quality, Boston Children's Hospital, Boston, MA.
• Pediatr Crit Care Me. 2014 Feb 1;15(2):e89-94.

ObjectiveResults of a large multicenter randomized clinical trial published in 2007 demonstrated no benefit in using a liberal versus conservative RBC transfusion threshold in stable critically ill children. Using the conservative threshold decreased the number of RBC transfusions without increasing adverse outcomes. We aimed to determine if wide dissemination of this evidence altered the hemoglobin threshold used for RBC transfusions in our pediatric medical-surgical ICU.DesignBefore-after retrospective cohort study using multiple administrative databases and chart review.SettingPICU serving medical and surgical patients.PatientsAll potentially stable children receiving a RBC transfusion in the PICU in 2006 (prepublication) and in 2009-2010 (postpublication). Children were considered unstable and excluded if they were severely hypoxic, receiving renal replacement therapy, hemodynamically unstable, or bleeding.InterventionsPhysician education on evidence supporting hemoglobin transfusion thresholds in teaching conferences, staff meetings, and via e-mail.Measurements And Main ResultsIn 2006, 14.6% of patients (n = 285/1,940) received a RBC transfusion. In 2009-2010, 12.1% of patients (n = 551/4,542) received a RBC transfusion. We evaluated patients transfused when they were potentially clinically stable, including 145 children in 2006 (191 transfusion days) and 266 children in 2009-2010 (369 transfusion days). We found no significant differences in age, sex, race, diagnoses, postoperative status, illness severity scores, mortality, or length of stay between these two groups. The median hemoglobin transfusion threshold decreased significantly from 8.0 g/dL (interquartile range 7.3, 8.6 g/dL) in 2006 to 7.5 g/dL (interquartile range 6.9, 8.1 g/dL) in 2009-2010 (p = 0.001). The percentage of transfusion days using a hemoglobin threshold more than 7 g/dL decreased from 81% (n = 154) in 2006 to 71% (n = 261) in 2009-2010.ConclusionAlthough transfusion thresholds in potentially stable critically ill children in our PICU significantly decreased after dissemination of best available evidence, 71% of patients were transfused at a hemoglobin threshold more than 7 g/dL.

36 keywords...

hide…