Correspondence  |   March 2017
Liberal versus Restrictive Erythrocyte Transfusion Algorithms and Perioperative Outcomes: Statistical Significance and Pulmonary Complications
Author Notes
  • Mayo Clinic, Rochester, Minnesota. warner.matthew@mayo.edu
  • (Accepted for publication October 20, 2016.)
    (Accepted for publication October 20, 2016.)×
Article Information
Correspondence
Correspondence   |   March 2017
Liberal versus Restrictive Erythrocyte Transfusion Algorithms and Perioperative Outcomes: Statistical Significance and Pulmonary Complications
Anesthesiology 3 2017, Vol.126, 569. doi:10.1097/ALN.0000000000001450
Anesthesiology 3 2017, Vol.126, 569. doi:10.1097/ALN.0000000000001450
It was with great interest that I read the recent meta-analysis by Hovaguimian and Myles1  and the accompanying editorial by Beattie and Wijeysundera2  regarding liberal versus restrictive erythrocyte transfusion triggers for surgical patients and those admitted to acute care environments. Their efforts in addressing the importance of clinical context (e.g., surgical type, comorbid disease) when evaluating transfusion algorithms should be congratulated. There are, however, several additional items deserving of mention.
First, Hovaguimian and Myles1  state that in those undergoing cardiovascular procedures, restrictive transfusion strategies increased the risk of mortality (risk ratio [RR], 1.39; 95% CI, 0.95 to 2.04) and events reflecting inadequate oxygen supply (RR, 1.09; 95% CI, 0.97 to 1.22). This statement is also highlighted in the section titled “What This Article Tells Us That Is New” and in the “Perioperative and Acute Care Transfusion Strategies” figure by Wanderer and Rathmell.3  However, an RR crossing a threshold of 1 does not imply statistical significance and should be labeled accordingly as a nonsignificant result.
First Page Preview
First page PDF preview
First page PDF preview ×
View Large