Relaxed Transfusion Standard No Threat After Cardiac Surgery
Relaxed Transfusion Standard No Threat After Cardiac Surgery
NEW YORK (Reuters Health) - Fears among some doctors that red-cell transfusions pose a higher risk to cardiac surgery patients appear to be unwarranted.
A new study of the question, the largest to date, shows that being liberal with transfusions did not heighten the likelihood of morbidity, did not significantly increase health care costs, and may save lives.
The research reported online March 11 in the New England Journal of Medicine was not large enough to definitively assess any difference in mortality, but it was significant enough that it is likely to spark concern.
"We happened to find a borderline difference in mortality. The interpretation of the mortality findings is what everyone is arguing about," coauthor Dr. Barnaby Reeves of the University of Bristol in the UK told Reuters Health in a telephone interview.
The increased mortality of 1.6 percentage points in the group where transfusions were more restricted "translates to 16 patients per thousand, which is substantial," said Dr. Soon Park, chief of cardiac surgery at UH Case Medical Center in Cleveland, who was not connected with the research.
"This is an important observation we don't want to lose sight of," he told Reuters Health by phone.
He and Reeves both said the results may point to the need for different rules for cardiac surgery patients when it comes to the issue of when to give a red-cell transfusion.
There has long been debate over the best threshold, leading to widely-divergent practices at cardiac centers.
The rates range from 8% to 93% in the U.S. and from 25% to 75% in the UK, based in part on observational studies suggesting that transfusions heighten the risk of infection, kidney injury, low cardiac output and death. Controlled trials, in contrast, haven't documented a transfusion threat.
Dr. Reeves said the findings should give solace to both sides of the controversy. "Despite doing the biggest trial and having very precise answers, each side is going to interpret the findings the way they want to," he predicted.
The new study, known as TITRe2, involved 2,007 patients who received nonemergency cadiac surgery at 17 specialty centers in the UK. All had hemoglobin levels below 9 g/dL or hematocrit levels below 27%.
Patients assigned to the "liberal" transfusion group immediately received one unit of red cells, with additional units whenever their counts fell back below that threshold. Volunteers assigned to the restrictive-threshold group received transfusions whenever their levels dropped below 7.5 g/dL.
Coronary-artery bypass grafting accounted for 40.7% of the surgeries; valve surgery accounted for an additional 30.5%. Most of the others received both operations.
Ultimately, 92.2% in the liberal-threshold group and 53.4% in the restrictive-threshold group received red cells under the study protocol. However, many patients had received at least one unit before they were enrolled, often during surgery; 63.7% of the people in the restrictive group left the hospital having received blood, compared to 94.9% in the liberal group.
Three months after randomization, 33.0% in the liberal-threshold group had developed a stroke, myocardial infarction, acute kidney injury, infarction of the gut or serious infection versus 35.1% in the restrictive-threshold group (p=0.30).
There was no significant difference in length of stay, the risk of significant pulmonary complications or serious post-operative complications. Costs were similar as well, totaling $17,762 in the restrictive group at three months versus $18,059 in the liberal group.
But the death rate was 64% higher in the restrictive group -- 4.2% died from any cause in that three-month period versus 2.6% in cases where transfusions were given liberally (p=0.045).
The risk was even greater when Dr. Reeves and his colleagues accounted for transfusions that were given before the patients were assigned to a treatment group.
The researchers characterized the finding as "a cause for concern."
Dr. Park said that makes sense because, while people undergoing other types of surgery may be able to increase their cardiac output to compensate for a lower red cell count, that can be a problem for heart surgery patients.
"When their heart motion is reduced or the heart rate can't come up, they may not be able to compensate," he said, adding that the result may be higher mortality. "We have to talk about how much oxygen one should deliver to the body to be safe."
Dr. Park said the fact that the risk of acute kidney injury events was also higher in the restrictive-threshold group may also reflect the problem.
"Liberal usage of blood transfusion in cardiac surgery patients," he said, "may actually be a benefit in terms of reducing renal dysfunction and reducing perioperative mortality."
"I think this will relieve some of the pressure on doctors to adopt a restrictive practice," said Dr. Reeves. "Policymakers have been pushing restrictive practice. It's more efficient, helps the blood supply and all of those kinds of things. Now doctors who have been moved that way because of policy have some grounds for saying, 'I want to transfuse more liberally.' The results clearly show transfusion is safe."
The issue is not likely to be resolved soon.
Dr. Reeves said a new trial, known as TRICS-III, is trying to enroll nearly 3,600 cardiac surgery patients in Canada and the United States. But even though death will be part of the primary outcome being assessed, that trial will still not be large enough to definitively determine if restrictive transfusions pose a higher mortality risk for cardiac surgery patients.
SOURCE: http://bit.ly/1FEQsxF
N Engl J Med 2015.
NEW YORK (Reuters Health) - Fears among some doctors that red-cell transfusions pose a higher risk to cardiac surgery patients appear to be unwarranted.
A new study of the question, the largest to date, shows that being liberal with transfusions did not heighten the likelihood of morbidity, did not significantly increase health care costs, and may save lives.
The research reported online March 11 in the New England Journal of Medicine was not large enough to definitively assess any difference in mortality, but it was significant enough that it is likely to spark concern.
"We happened to find a borderline difference in mortality. The interpretation of the mortality findings is what everyone is arguing about," coauthor Dr. Barnaby Reeves of the University of Bristol in the UK told Reuters Health in a telephone interview.
The increased mortality of 1.6 percentage points in the group where transfusions were more restricted "translates to 16 patients per thousand, which is substantial," said Dr. Soon Park, chief of cardiac surgery at UH Case Medical Center in Cleveland, who was not connected with the research.
"This is an important observation we don't want to lose sight of," he told Reuters Health by phone.
He and Reeves both said the results may point to the need for different rules for cardiac surgery patients when it comes to the issue of when to give a red-cell transfusion.
There has long been debate over the best threshold, leading to widely-divergent practices at cardiac centers.
The rates range from 8% to 93% in the U.S. and from 25% to 75% in the UK, based in part on observational studies suggesting that transfusions heighten the risk of infection, kidney injury, low cardiac output and death. Controlled trials, in contrast, haven't documented a transfusion threat.
Dr. Reeves said the findings should give solace to both sides of the controversy. "Despite doing the biggest trial and having very precise answers, each side is going to interpret the findings the way they want to," he predicted.
The new study, known as TITRe2, involved 2,007 patients who received nonemergency cadiac surgery at 17 specialty centers in the UK. All had hemoglobin levels below 9 g/dL or hematocrit levels below 27%.
Patients assigned to the "liberal" transfusion group immediately received one unit of red cells, with additional units whenever their counts fell back below that threshold. Volunteers assigned to the restrictive-threshold group received transfusions whenever their levels dropped below 7.5 g/dL.
Coronary-artery bypass grafting accounted for 40.7% of the surgeries; valve surgery accounted for an additional 30.5%. Most of the others received both operations.
Ultimately, 92.2% in the liberal-threshold group and 53.4% in the restrictive-threshold group received red cells under the study protocol. However, many patients had received at least one unit before they were enrolled, often during surgery; 63.7% of the people in the restrictive group left the hospital having received blood, compared to 94.9% in the liberal group.
Three months after randomization, 33.0% in the liberal-threshold group had developed a stroke, myocardial infarction, acute kidney injury, infarction of the gut or serious infection versus 35.1% in the restrictive-threshold group (p=0.30).
There was no significant difference in length of stay, the risk of significant pulmonary complications or serious post-operative complications. Costs were similar as well, totaling $17,762 in the restrictive group at three months versus $18,059 in the liberal group.
But the death rate was 64% higher in the restrictive group -- 4.2% died from any cause in that three-month period versus 2.6% in cases where transfusions were given liberally (p=0.045).
The risk was even greater when Dr. Reeves and his colleagues accounted for transfusions that were given before the patients were assigned to a treatment group.
The researchers characterized the finding as "a cause for concern."
Dr. Park said that makes sense because, while people undergoing other types of surgery may be able to increase their cardiac output to compensate for a lower red cell count, that can be a problem for heart surgery patients.
"When their heart motion is reduced or the heart rate can't come up, they may not be able to compensate," he said, adding that the result may be higher mortality. "We have to talk about how much oxygen one should deliver to the body to be safe."
Dr. Park said the fact that the risk of acute kidney injury events was also higher in the restrictive-threshold group may also reflect the problem.
"Liberal usage of blood transfusion in cardiac surgery patients," he said, "may actually be a benefit in terms of reducing renal dysfunction and reducing perioperative mortality."
"I think this will relieve some of the pressure on doctors to adopt a restrictive practice," said Dr. Reeves. "Policymakers have been pushing restrictive practice. It's more efficient, helps the blood supply and all of those kinds of things. Now doctors who have been moved that way because of policy have some grounds for saying, 'I want to transfuse more liberally.' The results clearly show transfusion is safe."
The issue is not likely to be resolved soon.
Dr. Reeves said a new trial, known as TRICS-III, is trying to enroll nearly 3,600 cardiac surgery patients in Canada and the United States. But even though death will be part of the primary outcome being assessed, that trial will still not be large enough to definitively determine if restrictive transfusions pose a higher mortality risk for cardiac surgery patients.
SOURCE: http://bit.ly/1FEQsxF
N Engl J Med 2015.
