Preventing Unnecessary Blood Transfusions and Reducing Multiple-Donor Exposure among Premature Infants


In an initiative to reduce the risk of unnecessary PRBC transfusions, leadership at Mayo Clinic’s Newborn Intensive Care Unit has developed and implemented an evidence-based PRBC transfusion guideline for extremely low-birth-weight infants.

Premature infants often develop anemia as a result of hemorrhage, immature hematopoiesis, and frequent blood sampling. As a result, these infants frequently receive packed red blood cell (PRBC) transfusions, with up to 80% receiving multiple transfusions. Tragically, a number of morbidities have been associated with PRBC transfusions in this population—due to transfusion reactions and other complications unique to neonates.

Marc Ellsworth, M.D.,

“In the past, providers would transfuse infants based on a hemoglobin number that they felt was safe,” says Marc Ellsworth, M.D., neonatology resident at Mayo and senior author of the initiative article (published in Advances in Neonatal Care). “It was very subjective rather than evidence-based because there were no guidelines as to when to transfuse, and so doctors would pick a number, around 11 or 12, as necessary for a transfusion, even sometimes regardless of how the patient was doing.”

Before implementing a new clinical guideline, the Mayo Clinic team performed a retrospective review of all PRBC transfusions given to extremely low-birth-weight (ELBW) infants in 2012 (pre-guideline). The team identified the indication for each transfusion by reviewing physiological and laboratory data and the daily clinical note.

“As we looked at all the evidence from other studies, we found that it’s safe to let the hemoglobin number drop lower in these infants before doing a transfusion,” says Dr. Ellsworth. “So if you ignore the 11, and you ignore the 10, and you ignore the 9, that’s 3 transfusions you just saved a baby from receiving. Therefore, you avoid the cost and the risk that comes with each transfusion. And these babies do fine with this. They may even do better.”

By adopting evidence-based transfusion thresholds, this clinical practice guideline was successful in eliminating an estimated 39 unnecessary PRBC transfusions to ELBW infants in the first six months of 2014. Further, of the 85 transfusions in this time frame, 52 were administered in adherence to the guideline—a remarkable four-fold improvement when compared to the 2012 pre-guideline data. This represents a projected cost savings of $31,000 for Mayo’s Newborn Intensive Care Unit (NICU) in that six-month period. It also reduces costs for the families of these babies while lowering excess risk.

“This is basically direct evidence that the providers were being more judicious in waiting until those hemoglobin numbers dropped lower before doing a transfusion,” says Dr. Ellsworth. “So, by following the new guideline, they were making a cognizant effort to change their practice, even though that’s not what they were doing six months ago.”

Reducing Multiple-Donor Exposure
The team also discovered a way to limit donor exposure within this vulnerable population. Previous to the study, every time a baby received a transfusion, it would get a new bag of blood from a new donor—which meant a different set of risks for each bag. Not to mention the waste: a premature baby was only getting about 10% of each bag; the rest of the blood was discarded.

“We didn’t realize this was happening until we undertook the protocol for this study,” says Dr. Ellsworth. “It made us reevaluate our transfusion process so that babies could get the same blood from the same bag. For example, if a baby needed five transfusions, it used to get five different bags from five different donors. But now, the baby will get those five transfusions from the same bag, which is saved for about two weeks.

“So not only did we decrease the number of transfusions, we decreased the number of donors each baby is exposed to. We were able to implement these changes immediately, in real time, only after vigilantly looking at our transfusion processes from A to Z.”

And perhaps most importantly, the processes the team used to develop and implement transfusion guidelines can be generalized to other neonatal units. Why? The NICU at Mayo Clinic in Rochester, Minnesota, is very similar to the patient population, the patient risks, and patient morbidity of other NICUs.

“The infants receiving PRBC transfusions represent a pretty homogenous population across the country, all born before 30 weeks,” says Dr. Ellsworth. “We hope the new guidelines Mayo is using may be helpful to other NICUs.”

Christoph Bahn (@cjbahnsen)

Christoph Bahn

Christoph Bahn covers emerging research and discovery for Mayo Clinic Laboratories. His writing has also appeared in The New York Times, Los Angeles Times, and Smithsonian Air & Space. He divides his time between Southern California and Northwest Ohio.