Blood Transfusion Associated with Acute Kidney Injury in TAVR Patients


In patients undergoing transcatheter aortic valve replacement (TAVR), the need for blood transfusion is strongly linked to acute kidney injury (AKI), a complication that in turn predicts both short- and long-term mortality. The findings, published online October 9, 2012, ahead of print in Circulation: Cardiovascular Interventions, suggest that transfusion should be used judiciously but unfortunately fail to shed light on the mechanism involved.

Peter P. de Jaegere, MD, PhD, of Erasmus Medical Center (Rotterdam, The Netherlands), and colleagues examined the relationship between blood transfusion and AKI in 995 patients who underwent TAVR at 7 centers worldwide between November 2005 and January 2012. Device types included both CoreValve (Medtronic, Minneapolis, MN) and Sapien (Edwards Lifesciences, Irvine, CA).

One in 5 TAVR Patients Develop AKI

Valve Academic Research Consortium-defined AKI (absolute increase in serum creatinine ≥ 0.3 mg/dL or relative increase ≥ 50% within 72 hours) occurred in 206 patients (20.7%), of whom 31 (3.1%) received dialysis. Both cardiac and noncardiac mortality rates were higher in patients with AKI at 30 days and over a median follow-up of 12 months (interquartile range, 4-23 months).

Multivariate analysis identified several independent predictors of AKI including transfusion of 3 or more units of red blood cells (RBCs), maximum leukocyte count, congestive heart failure, peripheral vascular disease, and surgical risk (table 1). Notably, these did not include potential triggers of transfusion such as baseline anemia, bleeding-vascular complications, or perioperative blood loss.

Table 1. Independent Predictors of AKI After TAVR

 

OR

95% CI

P Value

RBC Transfusion ≤ 24 hrs
1-2 Units
3-4 Units
≥ 5 Units

 
1.47
3.05
4.81

 
0.98-2.22
1.24-7.53
1.45-15.95

 
0.064
0.015
0.010

Peripheral Vascular Disease

1.48

1.05-2.10

0.026

Congestive Heart Failure

1.43

1.01-2.03

0.042

Maximum Leukocyte Count ≤ 72 hrsa

1.05

1.02-1.09

0.001

Logistic EuroSCOREb

1.02

1.00-1.03

0.006

a Per 109 cells/L increase.
b Per % increase.

Predictors of 30-day mortality included perioperative life-threatening bleeding, post-TAVR aortic regurgitation, leukocyte count within 72 hours, and logistic EuroSCORE. Predictors of mortality beyond 30 days included patient-related variables and blood transfusion. AKI heightened the odds of death in both the short- and long-term (table 2).

Table 2. Association Between AKI and Mortality

 

OR

95% CI

P Value

30 Days

3.15

1.56-6.38

0.001

Beyond 30 Days

1.57

1.13-2.17

0.007


Overall, the findings support “a direct harmful effect of transfusion on the kidneys,” the researchers conclude. “It is known that preserved RBCs suffer structural or functional changes including reduced deformability and increased aggregability, all of which—particularly in older patients with impaired renal function—might induce (further) renal dysfunction. Also, the coadministration of proinflammatory molecules may play a role either directly or indirectly by inducing inflammation.”

Transfusion Merely a Marker?

But in an interview with TCTMD, Philippe Généreux, MD, of Columbia University Medical Center (New York, NY), remained skeptical that transfusion alone could be responsible for AKI, despite having recently coauthored another paper that identified a similar, though nonsignificant, connection (Généreux P. Am J Cardiol. 2012;Epub ahead of print).

The findings by de Jaegere et al are in line with what has been shown before in smaller studies, he said, but there are 2 problems with the current study. For one, it started enrolling patients as early as 2005. “This was the early experience, where we had a lot transfusions and maybe more bleeding,” Dr. Généreux pointed out. “Right now, if we go with more experienced [operators] and a less sick population, the [incidence] of AKI would probably be lower.”

In addition, while the study draws attention to an important issue, Dr. Généreux stressed, it does not clearly show a causal relationship between transfusion and AKI. “From my point of view, what is lacking from this study and [those that came before] is the reason for transfusion,” he noted. “What they said is that bleeding did not necessarily trigger transfusion. Maybe there is another reason more important than [bleeding, such as unidentified] baseline characteristics.”

Transfusion may merely be serving as a marker of a sicker population, Dr. Généreux suggested. Moreover, even if there is a causal relationship, it is unclear what steps can be taken to avoid AKI when transfusion occurs.

“To me, there are 2 triggers of AKI. There are very complicated procedures [in general,] with bleeding and vascular complications, and there is the other group that [lacks that complexity],” he said. In the latter group, “there probably is an interaction with some other factor that we cannot identify.”

Therefore, it is difficult to know what steps can be taken to prevent AKI in this context, Dr. Généreux commented.

In addition to advising “more careful use” of transfusions during TAVR, Dr. de Jaegere and colleagues call for “unequivocal criteria” that can be used in deciding whether to administer RBCs. While patient selection can also be helpful, they point out that high operative risk—often due to the presence of comorbidities like peripheral vascular disease—is what steers many patients toward a less invasive option than surgical valve replacement in the first place. Finally, “[i]t remains to be seen how additional pre- and postoperative care [such as optimal hydration] may avoid AKI,” they conclude.

Study Details

Numerous characteristics differed between patients who did and did not experience AKI. On univariate analysis, AKI was associated with congestive heart failure, peripheral vascular disease, higher operative risk, use of the transapical approach, and vascular complications. However, AKI was not associated with perioperative blood loss or contrast media volume. Severe anemia, thought to be a factor that would encourage transfusion, was associated with 2.4 times less blood loss vs. no anemia at baseline and, conversely, with 2.3 times more units of blood transfused.

 


Source:
Nuis R-J, Rodés-Cabau J, Sinning J-M, et al. Blood transfusion and the risk of acute kidney injury after transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2012;Epub ahead of print.

 

 

Related Stories:

Caitlin E. Cox is News Editor of TCTMD and Associate Director, Editorial Content at the Cardiovascular Research Foundation. She produces the…

Read Full Bio
Disclosures
  • Dr. de Jaegere reports serving as a physician proctor for Medtronic.
  • Dr. Généreux reports receiving speaker’s fees from and serving as a consultant to Edwards Lifesciences.

Comments