The new formula not only can save hospital resources and staff time, but also can lead to a more rational allocation of banked blood, according to the researchers.
Before a bypass procedure, blood typically is delivered to the operating room in case it will be needed to replace blood lost during surgery. But not just any blood will do. The reserve blood must be matched with the patient's blood type, and it also must be "cross-matched." In cross-matching, a small amount of the patient's blood is mixed with a small portion of banked blood to test for adverse immunological reactions.
In their study, the Duke researchers found that the patients who are most likely to need the most blood during or immediately following their surgery are those who are over the age of 75, those who have impaired kidney function and those who weigh less than 121 pounds.
"At most U.S. hospitals, four units of typed and cross-matched blood are routinely sent to the operating room at the time of surgery," said George Lappas, M.D., a cardiovascular anesthesiology fellow who presented the Duke team's results on May 1, 2006, at the annual scientific sessions of the Society of Cardiovascular Anesthesiologists. "This is the arbitrary amount that has been used for many years in operating rooms nationwide."
"We thought there needed to be a better way to estimate blood needs, since blood is a scarce commodity that takes staff times and resources to prepare," Lappas said. "Also, with blood being held in operating rooms 'just in case,' it is not available for other patients who might need it."
The savings to hospitals and blood banks would be realized primarily in the reduced costs and manpower involved in the cross-matching process, Lappas said. On average, it costs about $55 to cross-match each unit of blood, meaning that $220 is typically spent for every surgery, whether the blood is used or not, he said.
The team's research was supported by Duke's Department of Anesthesiology.
In their study, the researchers analyzed the charts of 5,402 consecutive patients who received a bypass procedure at Duke University Hospital between 1993 and 2002. They correlated blood use during the first 24 hours after surgery with 13 different patient characteristics, such as anemia before surgery, age, gender, weight, heart and kidney function, tobacco use and other accompanying illnesses.
For the first two-thirds of the patients, they determined which characteristics led to greater blood use. They then developed the formula based on those findings and tested its predictive abilities on the last third of patients.
"We confirmed our hypothesis that a model based on a patient's characteristics prior to surgery could predict how much blood would need to be cross-matched prior to surgery," Lappas said. "This approach would appear to be a more rational and scientifically based approach for predicting blood needs."
The simplified scoring system devised by the team would identify risk factors for transfusion. For example, 1.5 units of cross-matched blood would be made available for patients older than 75, weighing less than 121 pounds or having a creatinine level of more than 1.4. Creatinine is a normal byproduct of metabolism; higher-than-normal levels in the blood indicate an impairment of the kidney's filtering ability, because kidneys normally filter creatinine out of the blood and excrete it in the urine.
Other factors, such as age older than 65, female sex or anemia (a hematocrit of less than 36 percent), would need one additional unit of cross-matched blood. Hematocrit is a measurement of the percentage of oxygen-carrying red blood cells in the blood; normal hematocrit ranges between 36 percent and 40 percent. An additional half unit of blood would be supplied for patients who have creatinine levels of greater than 1.2, who weigh less than 187 pounds, or whose heart function is moderately impaired.
To determine how many units of blood to set aside for an individual about to undergo surgery, physicians would add up all of the amounts for that person's applicable characteristics.
"We believe this model could easily be used by any hospital's blood bank to allocate blood in a more safe and cost-effective manner," Lappas said. "The formula is simple and easy to use."
Other members of the Duke team were Barbara Phillips-Bute and Mark Stafford-Smith.
Last reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
Published on PsychCentral.com. All rights reserved.