The authors have made a big stride toward developing a model to predict intraoperative transfusion requirements for orthotopic liver transplantation. (Image source: Thinkstock)

The authors have made a big stride toward developing a model to predict intraoperative transfusion requirements for orthotopic liver transplantation. (Image source: Thinkstock)

How many of you perform anesthesia for patients undergoing liver transplantation?  For those of you who have been doing this for a while, clearly the number of units of blood needed for these cases is less than in the past.  Some patients, though, need more blood than others.  Is there a way to predict how much blood is needed for the procedure?

Dr. Brian M. Parker and colleagues from the Departments of General Anesthesiology and Transplant Center, Quantitative Health Sciences, and Hepato-pancreato-biliary and Transplant Surgery, Cleveland Clinic, Cleveland, Ohio, retrospectively analyzed 835 adults who underwent orthotopic liver transplantation at their institution between January 1, 2001, and June 30, 2010, to determine which variables predict blood transfusion requirements, particularly high transfusion requirements, and whether the amount of blood transfused would affect survival.  The results of their analysis were published in  Anesthesia & Analgesia in the article  “Prediction of Intraoperative Transfusion Requirements During Orthotopic Liver Transplantation and the Influence on Postoperative Patient Survival.”

Older patients, and patients with low platelet count, high pretransplant INR and bilirubin, and those who received donor livers where expanded criteria were used, needed more blood and received more salvaged cells.  Those patients who received less blood and salvaged cells had hepatocellular cancer and had their surgery performed by Surgeon X.  Those patients who needed more than 20 units of red blood cells (RBCs) and salvaged cells had a higher level of preoperative creatinine and INR, had a lower pretransplant platelet count, underwent abdominal surgery before the transplant, had blood type A, and had their surgery performed by Surgeon Y.  Table 8 in the article shows the 15 most important factors of 66 considered in predicting the usage of a high number of RBCs and salvaged cells for the procedure for this group.  Those patients who needed more than 30 units of red cells and salvaged cells had a higher Model for End-Stage Liver Disease (MELD) score, had a lower platelet count preoperatively, and had their surgery performed by one of 2 different surgeons, either Surgeon Y or Surgeon Z (the same Surgeon Y who was in the >20 units of RBCs and salvaged cells group).  Table 11 in the article shows the 15 most important factors of 66 considered in predicting the use of a high number of RBCs and salvaged cells for the procedure for this group.

Average follow-up was almost 4 years.  Of the 10 patients who died within a day of transplantation, 8 received more than 40 units of red cells and salvaged cells.  Those patients who received more units of RBCs and salvaged cells, had a higher preoperative INR, who received a transplant from an older donor, and whose donor was neither African American nor Caucasian had a greater likelihood of dying within the early period, roughly 9 months after receiving a transplant.  In the late phase, those who had a primary diagnosis of viral hepatitis or liver cancer had a greater likelihood of dying.

This model may not be universally useful, but it is still of interest.  In the end, this is not to say that if the INR is corrected, transfusion requirements would be less, or that patients who are older or who have viral hepatitis or liver cancer should not undergo the procedure.  However, the important message of the article is not what it can predict. Patient’s with risk factors for bleeding are more likely to receive transfusions. Nothing new in that. Of far greater importance is the authors’ clear demonstration of what the model can’t predict. The authors clearly show that some patients will require large transfusions for reasons that cannot be identified prior to surgery. That’s critical information – the best data and mathematical models available still can’t predict that your patient will not need a large transfusion. So don’t deceive yourself that your patient is not at risk.

As Drs. John Blake and Edward C. Nemergut, Department of Industrial Engineering, Dalhousie University, Halifax, Nova Scotia, Canada, Research and Development, Canadian Blood Services, Ottawa, Ontario, Canada, and Department of Anesthesiology, University of Virginia, Charlottesville, Virginia, note in their accompanying editorial “Hope for the Best, but Prepare for the Worst,” also published in this month’s edition of Anesthesia & Analgesia, “Since it is impossible to determine ahead of time which patients will require large transfusions, the only safe strategy is to treat all patients as if they will require massive transfusion. Everyone involved in the surgery, and those providing the blood products to support the surgery, must be prepared for things to go awry, regardless of how reassuring the preoperative characteristics of the recipient or the donor may appear…hope is more likely to prevail when prepared for the worst.”