Free
Correspondence  |   January 2014
In Reply
Author Affiliations & Notes
  • Steven M. Frank, M.D.
    The Johns Hopkins Medical Institutions, Baltimore, Maryland. sfrank3@jhmi.edu
  • Accepted for publication publication September 16, 2013.
    Accepted for publication publication September 16, 2013.×
Article Information
Correspondence
Correspondence   |   January 2014
In Reply
Anesthesiology 01 2014, Vol.120, 241-242. doi:10.1097/ALN.0000000000000029
Anesthesiology 01 2014, Vol.120, 241-242. doi:10.1097/ALN.0000000000000029
I appreciate the comments by Dexter and Epstein regarding our publication on optimizing preoperative blood ordering.1  In their comment, they pose the question of whether our recommended maximum surgical blood order schedule would be changed if we applied the criteria published in their own study.2 
The answer is that I do not know whether our recommendations would be changed because Dexter and Epstein’s methods are somewhat complicated and difficult to understand. Our primary goal was to develop an algorithm that was simple and easy for other institutions to apply using their own data from an anesthesia information management system. In addition, I believe that our methods are more reliable because our algorithm does not rely as heavily on estimated blood loss (EBL), a parameter that most clinicians recognize as a crude measure that is fraught with error. In our algorithm, EBL was complimented by two other measures—percentage of patients receiving erythrocyte transfusion and the average number of erythrocyte units per patient—two variables that are much more objective and easy to determine from electronic anesthesia records.
The second question posed in their letter to the editor is why we used EBL greater than 50 ml in our algorithm to indicate the need for type and screen, whereas their study used EBL less than 50 to indicate no need for type and screen. This question is really about what to do with cases where EBL is 50 ml. We recognized that EBL is almost always reported as rounded values, and we made the decision to put cases with EBL of 50 ml in the “no type and screen” category, as long as the percentage of patients transfused was less than 5% and the transfusion index (average units/patient) was less than 0.3. This decision was based on the observation that many anesthesia providers enter “50” when EBL is minimal, because the electronic anesthesia records do not allow a text entry for EBL. In Dexter and Epstein’s proposed algorithm, the cases with EBL of 50 would be more likely to have a type and screen ordered, because they used EBL less than 50 as a criterion not to order a type and screen.
In summary, it is difficult to compare our maximum surgical blood order schedule with Dexter and Epstein’s study because our algorithms have more differences than similarities. In addition, our publication included the actual maximum surgical blood order schedule as an appendix, whereas theirs did not, making the comparison even more difficult. I can report, however, that our type and crossmatch to transfusion ratio has decreased by 29% since the release of the maximum surgical blood order schedule at our institution. This is the evidence that we have effectively reduced unnecessary blood orders, which will decrease cost, and perhaps improve patient safety, because the blood bank personnel can now focus on completing the blood orders for patients who may actually need transfusion.
Steven M. Frank, M.D., The Johns Hopkins Medical Institutions, Baltimore, Maryland. sfrank3@jhmi.edu
References
Frank, SM, Rothschild, JA, Masear, CG, Rivers, RJ, Merritt, WT, Savage, WJ, Ness, PM Optimizing preoperative blood ordering with data acquired from an anesthesia information management system.. A. (2013). 118 1286–97
Dexter, F, Ledolter, J, Davis, E, Witkowski, TA, Herman, JH, Epstein, RH Systematic criteria for type and screen based on procedure’s probability of erythrocyte transfusion.. A. (2012). 116 768–78