Short-term Mortality Higher in Transfused TAVR Patients

Patients who receive red blood cell transfusions after TAVR fare worse in the first 30 days compared with others, results from the TRITAVI registry show.

In a propensity-matched analysis, transfusion was associated with about double the risk of early mortality (6.4% vs 3.1%), as well as greater risks of cerebrovascular accident and stage 2 or 3 acute kidney injury, according to researchers led by Marco Zimarino, MD, PhD (“SS. Annunziata” Hospital – Chieti, Italy).

Transfusion was independently associated with 30-day mortality on multivariable analysis, along with major vascular complications and major bleeding, they report in a study published recently online in Circulation: Cardiovascular Interventions.

“Physicians should refrain from liberal transfusion to patients. They should refrain from using red blood cell transfusion as a fluid repletion means or to correct a lower-than-expected hemoglobin value,” Zimarino told TCMTD, adding that these results show that patient outcomes should be considered along with issues about the availability of blood products. “Our study supports really a restrictive use of such resources.”

Francesco Costa, MD, PhD (University of Messina, Italy), who was not involved in the study, pointed out that there is a lack of randomized data to guide use of transfusion after TAVR. He agreed that available evidence backs a more-conservative approach.

“I think that we should try to save this kind of resource as much as possible because it is scarce, and actually, if we are not sure that this could be beneficial in any way, we should not do it,” Costa said.

TRITAVI Registry

Though improvements in technology and technique have reduced periprocedural bleeding in patients undergoing TAVR, it remains common, and red blood cell transfusions are frequently administered. Prior studies have shown associations between transfusion and worse outcomes, including acute kidney injury and mortality.

It’s unclear, however, whether transfusion is a marker of poor outcomes or a causal driver of these events. Also, there’s no consensus around the best transfusion strategy in the setting of TAVR, an issue not directly addressed in clinical guidelines.

I think that we should try to save this kind of resource as much as possible because it is scarce, and actually, if we are not sure that this could be beneficial in any way, we should not do it. Francesco Costa

The TRITAVI registry, which retrospectively enrolled patients treated at six hospitals in Italy and one in Spain between 2012 and 2018, was set up to provide insight into transfusion’s impact on patients undergoing TAVR. The current analysis included 2,587 patients who had their procedure via the transfemoral approach. Transfusions were administered by physician discretion in 16% of cases, with an average of 1.83 units used.

Factors that were independently associated with receipt of transfusion were female sex, low body mass index, NYHA class III/IV disease, a lower baseline hemoglobin level, valve-in-valve procedures, major bleeding, major vascular complications, permanent pacemaker implantation, and use of dual antiplatelet therapy.

Propensity-matching resulted in 842 patients who received transfusions and an equal number who did not. Transfusion was associated with greater risks of the following outcomes in the first 30 days:

• Mortality: 6.4% versus 3.1% (HR 2.07; 95% CI 1.06-4.05)
• Cerebrovascular accident: 2.6% versus 0.7% (HR 3.75; 95% CI 1.04-13.50)
• Stage 2 to 3 acute kidney injury: 10.5% versus 2.6% (HR 4.35; 95% CI 2.21-8.55)

Risk of MI was 0.2% in each group (HR 1.00; 95% CI 0.06-16.00).

The investigators attempted to tease out whether transfusion is a marker or causal driver of increased risk but performing an analysis confined to patients without major vascular or bleeding complications after TAVR. Transfusion was significantly associated with 30-day mortality before, but not after, propensity-matching in this cohort.

However, in the full cohort, transfusion was independently associated with mortality on multivariable adjustment (HR 1.62; 95% CI 1.01-2.52), as were major vascular complications (HR 1.85; 95% CI 1.04-3.49) and major bleeding (HR 1.69; 95% CI 1.02-2.65).

“While awaiting trials dedicated to blood management after transcatheter aortic valve replacement, use of red blood cell transfusion should be strictly limited to lifesaving situations,” Zimarino et al conclude.

Strategies to Reduce Transfusions

Costa, who wrote an accompanying editorial with Mauricio Cohen, MD (University of Miami, FL), pointed out that this study cannot resolve the debate over whether transfusions are simply associated with poor outcomes—that is, transfused patients are sicker and more likely to die anyway—or are directly leading to increases in acute kidney injury and mortality. “This actually is a question that remains unanswered after this study, and for this reason we still need a randomized controlled trial,” he urged.

Nevertheless, evidence available to date supports more-restrictive use of transfusions, Costa said. He and Cohen lay out five strategies for ways to avoid transfusion:

• Use a restrictive hemoglobin threshold, and consider transfusion if hemoglobin is less than 7 or 8 g/dL rather than less than 10 g/dL.
• Avoid excessive hemodilution.
• Use less-intense antithrombotic therapy.
• Address a patient’s anemia with iron supplementation or medication before the procedure.
• Use more-careful planning and procedural technique to prevent access-site complications.

“A careful planning of the procedure is absolutely paramount to reduce the risk of bleeding and, consequently, the need for transfusion,” Costa said. He added that “a minimalistic approach to antithrombotic therapy after TAVR seems to be the best answer, and this would also help to reduce the risk of periprocedural and postprocedural bleeding complications.”

It is challenging to sort out the best transfusion threshold in the setting of TAVR, “a procedure performed in frail older adults with multiple comorbidities, many of whom have anemia and bleeding diathesis secondary to acquired Von Willebrand disease,” Costa and Cohen write in their editorial.

“Only a randomized clinical trial comparing the impact of a restrictive versus a liberal strategy could solve this issue,” they say. “Such trial should enroll patients with anemia at baseline and test noninferiority of a restrictive versus a liberal transfusion approach with a composite outcome measure including death, stroke, and acute kidney injury. Considering the large number of anemic patients undergoing TAVR and the high event rate in this population, a randomized trial would be very feasible.”

Photo Credit: Marco Zimarino