Restrictive or liberal? When it comes to transfusion strategies during cardiac surgery, that’s the question that has divided cardiac anesthesiologists and their surgical partners for decades. Now an international multicenter trial—the largest of its kind ever undertaken—seems to have definitively answered the question.
The Transfusion Requirements in Cardiac Surgery III (TRiCS III) investigation concluded that in patients at moderate to high risk for death, a restrictive strategy regarding red-cell transfusion is not inferior to a liberal strategy with respect to the composite outcome of death from any cause, myocardial infarction, stroke or new-onset renal failure with dialysis (N Engl J Med2017;377:2133-2144). These clinical outcomes came with the added benefit of significantly less blood transfused as well as significant cost savings.
“I have been a restrictive transfuser for much of my career,” said principal investigator C. David Mazer, MD, professor of anesthesia at the University of Toronto, in Ontario. “When the TITRe2 study came out two years ago [N Engl J Med 2015;372:997-1008], there was some concern about whether restrictive transfusion practices were really safe, because that study showed no difference in primary outcomes but a secondary outcome of mortality at 90 days that was just significantly higher in the restrictive group than the liberal group. So we think the community of cardiac anesthesiologists and surgeons has been wondering whether or not restrictive transfusion is, in fact, safe.”
Enter TRiCS III, a multicenter, open-label, noninferiority trial comprising 5,243 adults with a European System for Cardiac Operative Risk Evaluation (EuroSCORE) I score of 6 or higher, all of whom were scheduled for cardiac surgery with cardiopulmonary bypass. (The scale ranges from 0 to 47, with higher scores indicating increased risk for death after cardiac surgery.)
Participants were randomly assigned to one of two red-cell transfusion strategies. The restrictive threshold called for transfusion if patients’ hemoglobin level was less than 7.5 g/dL, starting from induction of anesthesia. The liberal strategy had patients transfused if their hemoglobin level was less than 9.5 g/dL in the operating room or ICU, or was less than 8.5 g/dL in the non-ICU ward.
Patients were enrolled in the study between 2014 and 2017, at 74 sites across 19 countries. Of the 5,243 patients, 4,860 were analyzed in the per-protocol analysis, with 2,430 in each group. At baseline, patients in the two groups were similar in demographic and operative characteristics.
Significantly Fewer Transfusions
Dr. Mazer reported the results (session LBS.01) at the 2017 annual meeting of the American Heart Association (Table). The composite primary outcome occurred in 11.4% of the patients in the restrictive threshold group, compared with 12.5% of those in the liberal threshold group (odds ratio [OR], 0.90; 95% CI, 0.76-1.07; P<0.001 for noninferiority). Mortality was 3.0% in the restrictive transfusion group and 3.6% in the liberal transfusion group (OR, 0.85; 95% CI, 0.62-1.16). New focal neurologic deficit affected 1.9% and 2.0% of patients, respectively (OR, 0.92; 95% CI, 0.61-1.38). Myocardial infarction occurred in 5.9% of patients in each group (OR, 1.00; 95% CI, 0.79-1.27). Renal failure with dialysis occurred in 2.5% of patients undergoing the restrictive strategy, compared with 3.0% of their liberal strategy counterparts (OR, 0.84; 95% CI, 0.60-1.19).
|Table. Primary Outcome by Type of Transfusion|
|Characteristic||Restrictive Threshold, n (%)||Liberal Threshold, n (%)||Odds Ratio (95% CI)|
|Composite||276 (11.4)||303 (12.5)||0.90 (0.76-1.07)|
|Mortality||74 (3.0)||87 (3.6)||0.85 (0.62-1.16)|
|New focal neurologic deficit||45 (1.9)||49 (2.0)||0.92 (0.61-1.38)|
|Myocardial infarction||144 (5.9)||144 (5.9)||1.00 (0.79-1.27)|
|Renal failure with dialysis||61 (2.5)||72 (3.0)||0.84 (0.60-1.19)|
With respect to secondary outcomes, the study found that red cell transfusion occurred in 52.3% of restrictive threshold patients, compared with 72.6% of the liberal threshold group (OR, 0.41; 95% CI, 0.37-0.47). Not surprisingly, restrictive strategy patients also received fewer median transfusions (two; range, one to four) than those in the liberal strategy group (three; range, two to five).
No meaningful differences were found in any of the other secondary outcomes, including ICU or hospital length of stay, days on mechanical ventilation, prolonged low-output state, infection, bowel infarction, acute kidney injury, seizures, delirium or encephalopathy.
“The bottom line is that TRiCS III definitively showed—with results that were consistent across subgroup and sensitivity analyses—that restrictive transfusion was not inferior to liberal transfusion, but it resulted in significantly less blood transfusion,” Dr. Mazer said in an interview with Anesthesiology News. “So important clinical patient outcomes were preserved, with the bonus of reduced transfusions … and its associated costs.”
Indeed, cost savings was another benefit of the restrictive strategy. “Depending on your cost estimate for blood and the currency used, if you factor in the activity cost of somewhere around $1,000 to $1,200 per unit, the amount of money saved just from blood reduction and allogeneic blood was in the millions of dollars range,” he said. “And that was just for the patients in this study itself.”
Although these results came as little surprise to Dr. Mazer, he hoped they would, at a minimum, add some clarity to a critical element of cardiac surgical practice. “Until now, most people’s transfusion practices were based on opinions rather than high-level evidence. So this provides high-quality evidence demonstrating the safety of a restrictive transfusion strategy, with a reduced likelihood of allogeneic blood transfusion.”
Dr. Mazer was quick to add that he’s not an advocate of anemia. “Anemia is not a normal state and is clearly associated with adverse outcomes. So from a clinical perspective, clinicians should do everything they can to keep the patient’s own hemoglobin levels as high as possible. But if the patient becomes anemic to a hemoglobin between 7.5 and 9.5 g/dL, are they better off to be transfused or not? And it would appear that they’re not better off to be transfused.”
Results Still Need Interpretation
Nevertheless, as Victor A. Ferraris, MD, PhD, told Anesthesiology News, it is difficult to make blanket statements about groups of patients given that each person presents with unique circumstances. Indeed, he pointed to a review article that he co-authored (Innovations [Phila]2016;11:157-164), that showed some high-risk patients actually benefit from transfusion.
“There are multiple competing factors that impact the transfusion/anemia balance,” said Dr. Ferraris, the Tyler Gill Professor of Surgery at the University of Kentucky, in Lexington. “It is difficult to completely account for all of these factors.” He acknowledged that the TRiCS III study is a classic that has gained national attention. “But, like everything else, the devil is in the details. There is no question that some patients benefit from transfusion. The problem is picking out which ones.”
For Aryeh Shander, MD, the study has many advantages that bolster the strength of its findings, beginning with the high comorbidity scores of its participants. “When you perform a trial on very sick patients, it makes much more of a definitive statement. Because generally speaking, we don’t like to transfuse healthier patients; we tend to transfuse the sicker ones,” said Dr. Shander, chief of anesthesiology, critical care medicine, pain management and hyperbaric medicine and director of the Team Health Research Institute at Englewood Hospital and Medical Center, in New Jersey.
Dr. Shander was also encouraged by the current trial’s real-world application. “One of the problems that typically happens in a prospective trial is that you’re essentially removing that population from what’s considered to be the standard of care and putting constraints on what type of therapeutic interventions are going to occur,” he said. “But in this particular trial, they tried to give the physician as much freedom as they could, to make it more of a real-life type of situation. And that’s a very important point, because now it starts to resemble what should be done in the field.”
Even though the study ultimately showed noninferiority between the two approaches in the primary outcome, Dr. Shander pointed out one very important caveat: Patients in the restrictive transfusion group received less blood than their liberal counterparts. “If you look at the literature, it strongly suggests that adverse outcomes from blood transfusion are dose-dependent. The TRiCS III study didn’t show that, but if you look at the 95% confidence interval, it may favor the restrictive group in terms of reducing these adverse outcomes.”
Finally, Dr. Shander noted that because previous studies of transfusion strategies have yielded conflicting results, clinicians need to consider the weight of the evidence. “My feeling is that you should never make any seismic changes to your practice based on a single study, especially if it does not reflect real-life practice. But now we have several of these trials showing no inferiority with a restrictive transfusion strategy. In the end, we should remember that the disease we’re treating is anemia, and treatment of anemia includes more than just red cell transfusions.”