ANAHEIM, CA — No advantages, clinical or otherwise, were seen from liberal use of red blood cell (RBC) transfusions to keep hemoglobin levels up in patients undergoing on-pump cardiac surgery, in a randomized noninferiority trial with more than 4800 patients.
"We've definitely shown that you can transfuse more sparingly and maintain patient safety and patient outcomes, while saving blood and its associated costs," Dr C David Mazer (University of Toronto and St Michael's Hospital, Toronto, ON) told theheart.org | Medscape Cardiology.
The rate of the primary clinical composite end point, which included new-onset renal failure, was 11.4% in the patients managed according to a "restrictive" strategy that allowed transfusions only at hemoglobin levels that were lower than for liberal transfusions, for which the rate was 12.5% (P<0.0001 for noninferiority).
The international third Transfusion Requirements in Cardiac Surgery (TRICS-3) trial is noteworthy for "the remarkable consistency of the results through various subgroup and sensitivity analyses," Mazer said. "I think that strengthens the message."
Mazer presented the study here today at the American Heart Association (AHA) 2017 Scientific Sessions and is lead author on its coinciding publication in the New England Journal of Medicine.
"Expensive and Toxic"
Dr Daniel Sessler (Cleveland Clinic, OH) agreed that the analyses are internally consistent as well as in agreement with many studies in recent years pointing in the same direction in such patients, that "with rare exceptions, we shouldn't be transfusing above a hemoglobin of 7 [g/dL]." That applies even to sicker patients, older patients, even critical-care patients, said Sessler, who isn't connected to TRICS-3.
"Blood is expensive and blood is toxic," he said. "You don't want to give more than necessary, and there's more and more evidence that, roughly speaking, the less the better."
The guidelines recommend transfusion for extreme anemia and recommend against it at hemoglobin >10 mg/dL, according to Mazer. "In between, they don't provide good guidance."
Practice has varied widely from center to center and even among different practitioners at the same center, he said. When TRICS-3 began, "there was overall equipoise, but transfusion is an emotional issue for people, and many people felt they knew the right answer, even though the evidence wasn't there."
With the release of TRICS-3, he said, "I suspect that people who use restrictive strategies will feel that their practices are substantiated by this study and that the people who transfuse liberally will look at the data and reassess what they're doing."
Although there's been no formal cost-effectiveness analysis, Mazer said at a special media briefing on the trial, it costs an estimated $1100 to infuse one unit of blood, taking into account all processes involved getting it to the patient. So the total cost-difference for the restrictive vs liberal transfusion strategy in the trial amounted to "somewhere in the range of $3 million."
Pilot Study Contributed Some Patients
The trial enrolled 5243 adult patients slated for any form of cardiac surgery with planned cardiopulmonary-bypass support (CPB) support, who were randomized to one of the two transfusion strategies and followed through day 28.
They included 5035 patients from 19 countries enrolled specifically for TRICS-3 and 208 patients who had been in an earlier TRICS pilot study that had "very similar entry criteria," according to Mazer.
After randomization, according to the report, some patients didn't undergo the surgery, some didn't actually use CPB support, and others dropped out for other reasons, leaving a modified intention-to-treat cohort of 4860 patients.
The surgery was CABG only in about 26% of the cohort, CABG plus valve surgery in about 19%, isolated valve surgery in about 29%, and other non-CABG surgery in about 17%, at similar rates in the two arms.
The restrictive transfusion strategy called for transfusion if hemoglobin dropped to less than 7.5 g/dL during the procedure or postoperatively. The liberal strategy called for transfusion for hemoglobin <9.5 g/dL during the procedure or after in the intensive-care unit or to less than 8.5 g/dL after discharge from the ICU and for the rest of the hospital stay.
Of patients in the restrictive arm, 52.3% received RBC transfusion, compared with 72.6% in the liberal arm; platelets and plasma were transfused at about the same rate in both arms.
Rates, Odds Ratios (OR) or Rate Ratios (RR) for Restrictive vs Liberal RBC Transfusion Strategies
|End points||Restrictive (n=2430)||Liberal (n=2430)||OR RR (95% CI)|
|Primary end point* (%)||11.4||12.5||0.90 (0.76–1.07)|
|Prolonged low-output state (%)||40.9||40.6||1.01 (0.90–1.14)|
|Delirium (%)||12.6||10.9||1.18 (0.99–1.41)|
|ICU length of stay (d)||2.1||1.9||0.89 (0.84–0.94)|
In addition to the restrictive group's primary end point being noninferior to that of the liberal group at 28 days, mortality was similar at 3.0% and 3.6%, respectively; nor were the other primary-end-point components significantly different.
In a prespecified subgroup analysis, the only significant interaction (P=0.004) for the primary end point was age, with the restrictive-strategy group showing a significant advantage compared with the liberal strategy in patients at least 75 years old (unadjusted OR 0.70, 95% CI 0.54–0.89). The countering increase in risk for those younger than 75 didn't reach significance.
"Many people believe that older you are, the higher your hemoglobin should be or the more liberally you should transfuse. And the outcome that we saw was opposite to that. And that was a bit surprising," Mazer said when interviewed.
At the media briefing on TRICS-3, Mazer proposed that older patients may have fared better in the restrictive-strategy arm due to a preselection effect, in that surgeons may have preferred to take "good"—that is, healthier—patients aged 75 or older.
"Or it could be that elderly patients are more sensitive to the inflammatory or immune or volume effects of the transfusion, compared with younger patients."
Subgroup analyses have their limitations, but the study's hypothesis-generating age finding at least questions beliefs about transfusions in older cardiac surgical patients, suggesting that "restrictive transfusion may be safe in those patients."
But the finding also doesn't rule out that some cardiac surgical patients might benefit from RBC transfusion or that some might fare more poorly for receiving them, Sessler said. The study says simply that "you shouldn't generally be doing it."
Mazer said the findings are likely transferable to some nonsurgical clinical settings, including patients in the ICU.
The one setting in which the best strategy remains an unknown, he said, is acute coronary syndromes. "And there is a randomized trial going on now, called the [Myocardial Ischemia and Transfusion] MINT trial, that will address that particular clinical scenario."
Mazer reports grants from the Canadian Institutes for Health Research, personal fees from Amgen, Boehringer-Ingelheim, Octapharma, and Pharmascience, and fees and grants from Fresenius Kabi. Disclosures for the coauthors are listed on the journal website. Sessler reports he has no relevant financial relationships.
Medscape Medical News © 2017
Cite this: TRICS-3 Disputes 'Liberal' RBC Transfusion Strategy at Cardiac Surgery - Medscape - Nov 12, 2017.