Blood transfusion is one of the most commonly performed procedures that occurs during a hospitalization, and is performed during 11% of all hospital stays. Over the last decade, significant advances have been made regarding RBC transfusion thresholds, with improved understanding that a restrictive approach to RBC transfusion results in avoidance of complications associated with over transfusion while doing so at a reduced cost of care.  These results have been demonstrated in a wide variety of patient populations including critically ill ICU patients, orthopedic hip fracture, gastrointestinal bleeding, cardiac surgery, sepsis, pediatric critical care, geriatrics, and traumatic brain injury.

Although these results are significant, there still exists a wide variability in RBC usage between hospitals and individual physicians when performing the same procedure. This is clearly demonstrated by a study that was performed in 1992, in which 114 medical school faculty members were asked to evaluate 24 hypothetical scenarios, using clinical vignettes, and determine appropriate care. Upon reviewing the results, there was a remarkable lack of agreement between who should or should not be transfused. As we have moved forward, we have come to better understand that these decisions are often times not based on clinical evidence but based on the training provided by predecessors, which further leads to variability in care as hospitals often times develop their own culture surrounding practice.

As stated in a recent meta-analysis, it has become clear that a restrictive approach to transfusions in a wide variety of patient populations significantly reduced cardiac events, bleeding, bacterial infections, and mortality.  As these findings are being recognized and placed into clinical practice, significant reductions in blood usage has taken place (some studies suggest an 18% annual reduction over the past 3 years in some centers). This begs the question, has our reduction in the use of transfusion led to undertransfusion resulting in potential harm to the patient?

According to a recent article by Hibbs and colleagues, there are several ways to define undertransfusion.  Undertransfusion can be defined by the evidence compiled from randomized control trials (RTCs) on hemoglobin (Hb) and platelet (PLT) thresholds using a variety of different clinical situations as the failure to provide a transfusion if a patient’s lab variables fall below the set standards of the RTCs.  The other alternative is to define undertransfusion on the basis of outcome, such as “any adverse complication which may have been avoided by blood transfusion”.  This definition then gives us pause to ask the next question, how we might determine that the transfusion would have averted the adverse complications. Of the studies noted to date, Hibbs and coworkers have found that undertransfusion is uncommon and is identifiable only after considerable effort.  For now, clinical vigilance is what will ultimately keep our patients safe.

  1. Waters et al. Transfusion 2015;55(4):700-702.