Restrictive Transfusion Therapy Safe for Cardiac Surgery Patients

Setting higher hemoglobin (Hb) thresholds for red-cell transfusion therapy in patients undergoing cardiac surgery with a moderate to high risk of death was noninferior to using traditional thresholds in reducing the risk of death, myocardial infarction (MI), stroke, and new-onset renal failure at six months, according to results from a study published in the New England Journal of Medicine.

“[Our results] show that a restrictive transfusion policy can safely be used and accordingly reduces the unnecessary exposure of patients to blood products and risks secondary to transfusion, which range from mild to life threatening,” principal investigator Nadine Shehata, MD, of the University of Toronto, told ASH Clinical News.

This report updates previous research from the randomized, noninferiority Transfusion Requirements in Cardiac Surgery (TRICS) III trial, which demonstrated that a restrictive strategy (transfusion threshold of Hb <7.5 g/dL) was noninferior to a liberal strategy (Hb <9.5 g/dL intraoperatively or postoperatively while in the intensive care unit [ICU], or <8.5 g/dL when the patient was in the non-ICU ward) in terms of patient morbidity and mortality at 28 days after cardiac surgery.

The TRICS III trial included 5,243 patients who were scheduled to undergo cardiac surgery and were at moderate to high risk of death, defined as a European System for Cardiac Operative Risk Evaluation I score of ≥6. People who were unable to receive blood products, declined blood products, were participating in a preoperative autologous blood donation program, were undergoing heart transplantation, or were having surgery solely for the insertion of a ventricular assist device were excluded from the analysis.

This primary per-protocol analysis included 4,860 participants who were assigned to either a restrictive transfusion strategy (n=2,430) or a liberal transfusion strategy (n=2,430). The primary composite outcome at six months included the following components:

  • death from any cause
  • myocardial infarction
  • new focal neurologic deficit (stroke)
  • new-onset renal failure with dialysis

A secondary composite outcome included these factors, as well as emergency department visit, hospital readmission, or coronary revascularization.

The primary outcome, for restrictive versus liberal transfusion strategies, was 11.4 percent versus 12.5 percent (odds ratio [OR] = 0.90; 95% CI 0.76-1.07; p<0.001 for noninferiority).

At six months postsurgery, the primary composite outcome was observed in 402 patients in the liberal threshold group and 402 patients in the restrictive threshold group (17.1% vs. 17.4%; OR=1.02; 95% CI 0.87-1.18; p=0.006 for noninferiority).

Mortality rates also were similar between the groups: 6.4 percent for liberal threshold vs. 6.2 percent for restrictive threshold (OR=0.95; 95% CI 0.75-1.21). In addition, the investigators observed no difference between the two groups in terms of the expanded secondary composite outcome (43.8% vs 42.8%; OR=1.04; 95% CI 0.93-1.17; p value not reported).

In a subgroup analysis, the investigators found that the restrictive strategy actually was superior in patients aged ≥75 years, but in patients aged <75 years, the liberal strategy was associated with a lower risk of the primary composite outcome (p=0.004).

“The findings in patients 75 years of age or older appear to contradict the current practice, in which a liberal transfusion strategy is used in older patients undergoing cardiac or noncardiac surgery who thus receive more red-cell transfusions than do younger patients,” the researchers noted. “One could hypothesize that older patients may have unacceptable adverse effect related to transfusion (e.g., volume overload and inflammatory and infectious complications) or that there may be age-related differences in the adverse-effect profile of transfusion or anemia. … The mechanisms underlying any potential differential effect of age on outcomes of transfusion must be further studied within the cardiac surgery population.”

Dr. Shehata also noted that the study’s findings could have implications for health-care costs. “Transfusion is costly, and although each unit of red blood cells is approximately $300 to $400 [in Canada], this cost does not account for nursing administration costs and potential adverse events that could prolong hospitalization,” she stated. “Some estimate the cost to be well over $1,000 per unit when administration costs are considered, so a restrictive strategy would reduce those costs.”

Adopting a restrictive strategy also would alleviate some of the strain on the nation’s blood supply. “Donors are donating their blood for patients needing life-saving transfusions,” she added. “Using higher thresholds for a transfusion reduces the unnecessary use of blood products, so that they can be directed to those in need.”

The study’s findings were limited by the lack of required follow-up of patients after 28 days or hospital discharge and the fact that outcomes data were obtained from a variety of sources (telephone contact, databases, hospital records), which increased the odds that some follow-up visits were missed.

“TRICS III was a large, multicenter, international study with a high adherence rate to transfusion strategies, thus we believe our results are valid and generalizable,” Dr. Shehata explained when asked about the study’s strengths and weaknesses. “The only potential finding that likely merits additional investigation is whether the age of the patients should be a factor to consider when transfusing cardiac surgery patients.”

The authors report no conflicts of interest.


Mazer CD, Whitlock RP, Fergusson DA, et al. Six-month outcomes after restrictive or liberal transfusion for cardiac surgery. N Engl J Med. 2018;379:1224-33.