NEW ORLEANS, LA -- For patients at high cardiac risk who undergo major general or vascular surgery, a liberal approach to transfusion is no better than the more commonly accepted restrictive approach when it comes to 90-day mortality or major ischemic events, results of the TOP trial show.
However, there were hints that the liberal strategy reduces the risk of patients having certain cardiac complications.
TOP investigators Panos Kougias, MD, and Sherene Sharath, PhD (both from SUNY Downstate Health Sciences University and VA New York Harbor Healthcare System, Brooklyn, NY), presented their trial together Saturday at the American Heart Association (AHA) 2025 Scientific Sessions.
"Postoperative anemia is common after major [vascular and general surgery] operations," said Sharath. Although clinical guidelines advise transfusion when hemoglobin levels fall below 7 g/dL, "the safety of this strategy after major operations in high-cardiac-risk patients is unclear."
TOP sought to understand whether a more-liberal approach with a transfusion trigger of 10 g/dL would reduce adverse outcomes, she said.
Louise Sun, MD (Stanford University School of Medicine, CA), discussing the trial after its presentation, stressed that acute anemia -- the reason for transfusion -- is bad no matter the clinical setting. "The reason it is bad for clinical outcomes is that it induces a cascade of unfortunate events," she explained. "Those would include hypotension, stasis, [and] sympathetic nervous system activation, along with decreased tissue oxygenation, hypoperfusion of end organs, and ultimately . . . shock."
But transfusion isn't always the answer, Sun added, because it can lead to acute hemolytic transfusion, allergic reactions, or transfusion-related acute lung injury, circulatory overload, graft-versus-host disease, and purpura, as well potential viral and bacterial infections.
Blood products can also be a scarce resource and must be used judiciously, said Sun. "Now, the question is: is less necessarily equal to more? We know that over the recent couple of decades, there's been a shift towards restrictive transfusion in many fields of medicine."
A 2023 consensus statement came to the conclusion, based on best available evidence, that a hemoglobin (Hb) level of 7 g/dL is an acceptable trigger when using a restrictive transfusion strategy among patients without major cardiac comorbidities.
TOP is unique in that it enrolled a previously unstudied population, Sun noted. Its results confirm that 7 g/dL is a reasonable transfusion trigger, specifically when used perioperatively and in patients with major cardiac conditions, she said. At the same time, the findings suggest that a more liberal strategy might reduce new-onset or worsening heart failure and arrhythmias.
Liberal vs Restrictive
The TOP trial recruited 1,428 patients in the Veterans Affairs health system who were undergoing major general or vascular surgery and were at high cardiac risk due to a history of ischemic heart disease, myocardial infarction, peripheral artery disease, stroke, or transient ischemic attack. Researchers randomized these participants to receive one of two transfusion strategies if their Hb dropped below 10 g/dL within 10 days of the index operation:
Data ultimately were available for 1,424 patients (mean age 70 years; 98% male; 75% white). Transfusion protocols were implemented successfully on the whole, said Kougias. "At randomization, mean hemoglobin values were similar, but quickly separated and maintained a steady 2-gram difference throughout the hospital stay."
A one-size-fits-all transfusion strategy may not be the best. Panos Kougias
The primary endpoint of 90-day mortality or major ischemic events occurred at rates of 9.1% and 10.1% in the liberal and restrictive groups, respectively (RR 0.90; 95% CI 0.65-1.24). Most of the endpoint's individual components -- MI, coronary revascularization, acute renal failure, and stroke -- favored the liberal strategy, though the differences weren't significant and, as Kougias noted, the confidence intervals were wide.
Additionally, the rates of infectious complications, 1-year mortality, and hospital length of stay each were similar in the two groups.
Yet, non-MI cardiac complications were much less frequent in the liberal group than in the restrictive group (5.9% vs 9.9%; RR 0.59; 99% CI 0.36-0.98), with the advantage driven by both heart failure and arrhythmias. For these cardiac complications, "the cumulative incidence curves show a steady divergence," said Kougias. Seven days after randomization, "the curves really start to pull apart, and they keep separating all the way out to 90 days."
He reminded the AHA audience: "We have to remember here, this was a secondary outcome that was not adjudicated, . . . so although the difference is large and worth taking notice of, we should interpret this finding with caution."
Insights for Practice
When it comes to the "reality of clinical practice," the message from TOP is nuanced, Kougias stressed. "A one-size-fits-all transfusion strategy may not be the best." And while the trial did not find a liberal strategy to be superior, that approach might still provide benefits related to heart failure and arrhythmias, he added. "This is a finding that should inform patient-physician conversations but will need to be further confirmed in future, well-designed studies."
Sun pointed out that it's still an open question whether the benefits of a liberal approach seen in the acute phase might be outweighed in the long term by infectious or immunologic consequences of receiving pRBC.
In some instances, clinicians are "a little uncomfortable" with the 7-g/dL threshold because they try to personalize their transfusion practices to each patient's needs, she said. Yet "we know that transfusion practices in many clinical settings have become protocolized, so there's very little room for personalization at the moment." Greater understanding of what's going on physiologically at a tissue level with oxygen supply and demand could help tailor transfusion decisions, she suggested.
Olamide Alabi, MD (Emory University, Atlanta, GA), in an AHA press conference, noted that whether to transfuse a particular patient after surgery is a debate that comes up often in real-world practice. "The TOP trial provides valuable insights into blood transfusion strategies and the risk to surgical patients," she agreed, suggesting that the differences seen for the secondary outcomes merit a closer look.
"Precision medicine in surgical care is not the future -- it's here and now," said Alabi. "Examining this further to determine which patients would benefit from the more liberal approach to avoid the development of heart failure, arrhythmia, or cardiac arrest is extremely important."
That said, TOP's finding that the restrictive cutoff is indeed acceptable "portends the use of fewer resources that are needed and there's also a trend towards fewer transfusion-related complications," Alabi concluded.