Using an inexpensive drug for every hip or knee replacement since 2013 has helped St Michael’s Hospital in Ontario, Canada reduce its number of red blood cell transfusions performed during these surgeries by more than 40% without negatively affecting patients, according to new research. The drug tranexamic acid, known as TXA, prevents excessive blood loss during surgeries.
TXA had been shown to be effective in orthopaedic, trauma and cardiac patients but less than half of eligible patients at St Michael’s received this drug because of a previous province-wide shortage. The drug was given only to patients at high risk of requiring a blood transfusion. The TXA shortage ended in early 2013 and, in October 2013, St Michael’s anaesthesiologists started giving TXA to every eligible patient undergoing hip or knee replacement. A total of 402 patients received the drug.
“We wanted to optimise TXA’s use in patients undergoing hip or knee replacements because these procedures often result in high blood loss and frequently require transfusions,” said Dr Greg Hare, an anaesthesiologist at St Michael’s. “The drug costs about $10 per patient, while the average cost of transfusing one unit of blood is $1,200.” Making TXA use mandatory for eligible patients undergoing hip or knee replacement reduced the hospital’s transfusion rate for those surgeries from 8.8% to 5.2% (more than a 40% reduction).
Canadian data suggests the 2014 hip and knee arthroplasty rate was about 9%.
Patients who received TXA did not experience any increase in adverse events, such as heart attack, stroke or blood clots. There was no difference in mortality rates or length of hospital stay.
“Other hospitals and surgical centres should consider making TXA mandatory for similar surgeries because it can improve quality of care, decrease the need for blood transfusions and even save money,” said Hare, who is also one of the founders of St Michael’s Centre of Excellence for Patient Blood Management.
“Making TXA mandatory for eligible patients has made care more efficient, ensuring the best possible care for our patients,” he said.
For this study, TXA was not used in patients identified to be at risk of blood clots, stroke or any other cardiovascular issues.
Tranexamic acid (TXA) therapy can reduce red blood cell (RBC) transfusion; however, this therapy remains underutilized in many surgical patient populations. We assessed whether implementation of a protocol to facilitate universal administration of TXA in patients undergoing total hip or knee arthroplasty would reduce the incidence of RBC transfusion without increasing adverse clinical outcomes.
We implemented a quality of care policy to provide universal administration of intravenous TXA at a dose of 20 mg•kg−1 iv to all eligible patients undergoing total hip or knee arthroplasty from October 21, 2013 to April 30, 2014. We compared data from an equal number of patients before and after protocol implementation (n = 422 per group). The primary outcome was RBC transfusion with secondary outcomes including postoperative hemoglobin concentration (Hb) and length of hospital stay. Adverse events were identified from the electronic medical records. Data were analyzed by a Chi square test and adjusted logistic and linear regression analysis.
Implementation of the protocol resulted in an increase in TXA utilization from 45.8% to 95.3% [change 49.5%; 95% confidence interval (CI), 44.1 to 54.5; P < 0.001]. This change was associated with a reduction in the rate of RBC transfusion from 8.8% to 5.2%, (change −3.6%; 95% CI, −0.1 to −7.0; P = 0.043). Pre- and post-protocol mean [standard deviation (SD)] Hb values were similar, including the nadir Hb prior to RBC transfusion [72 (8) g•L−1 vs 70 (8) g•L−1, respectively; mean difference −1 g•L−1; 95% CI, −3 to 5; P = 0.569]. Length of stay was not altered, and no increase in adverse events was observed.
Implementation of a perioperative TXA protocol was associated with both an increase in TXA use and a reduction in RBC transfusion following hip or knee arthroplasty. Adverse events and length of hospital stay were not influenced by the protocol.