Echo May Not Provide Best Picture for TAVI
By Chris Kaiser, Cardiology Editor,
November 07, 2011
Action Points
Explain that cardiac magnetic resonance imaging (MR) and computed tomography (CT) more accurately and reproducibly measured aortic root dimensions than transthoracic echocardiography (TTE) in patients undergoing transcatheter aortic valve implantation (TAVI).
Note that MR and CT measurements better predicted severity of aortic regurgitation following TAVI than TTE.
Review
Cardiac MR and CT imaging may offer a better and more reproducible assessment of the aortic root than transthoracic echocardiography (TTE) for patients undergoing transcatheter aortic valve implantation (TAVI), a single-center study found.
Both cardiac MR and CT measurements of the aortic valve annulus prior to TAVI were better predictors of the severity of aortic regurgitation following TAVI compared with TTE, reported Raad H. Mohiaddin, MD, PhD, of the Royal Brompton Hospital in the U.K., and colleagues.
Researchers also found that MR- and CT-derived annulus eccentricity indices were not associated with aortic regurgitation following TAVI, according to the study published in the November 15 issue of the Journal of the American College of Cardiology.
In contrast, measurements derived from TTE tended to have higher variability than MR- and CT-derived measurements, and TTE significantly underestimated the aortic valve annulus size.
These results have "clinical implications for predicting outcome after TAVI," Mohiaddin and colleagues wrote in conclusion.
This conclusion is especially pertinent to interventional cardiologists and surgeons in the U.S. as the FDA earlier this month approved the Sapien valve for treatment of patients too frail to be treated surgically.
Mohiaddin and colleagues noted that there is not yet a gold standard imaging modality for measuring aortic valve annulus diameter prior to TAVI.
The problem with TTE, although it is widely used for this application, is that it doesn't account for the ellipsoid nature of the annulus, they said. Cardiac MR and CT are better suited for this measurement because they produce three-dimensional images.
For this reason, both MR and CT in this study of 202 patients had similarly low inter- and intraobserver variability. In addition MR and CT both showed similarly larger annulus diameter measurements compared with TTE.
Although pre-procedural imaging can be predictive of aortic regurgitation and other complications following TAVI, studies examining imaging's role in TAVI have been small or have only included one imaging modality, researchers said.
To help understand potential differences between cardiac MR, cardiac CT, and TTE, researchers prospectively recruited 202 patients from the Royal Brompton Hospital who had severe aortic regurgitation. All patients underwent both cardiac MR and TTE, while 133 also underwent cardiac CT.
All cases involved a multidisciplinary team that comprised cardiothoracic surgeons, cardiologists, and radiologists.
Ultimately, only 87 patients underwent TAVI (79 receiving the CoreValve and eight the Sapien valve).
The mean patient age was 79, 56% were men, and the mean ejection fraction was 59%.
Cardiac MR had the lowest intra- and interobserver variability compared with CT and TTE, as the following examples demonstrate:
Aortic valve annulus -- 1.7% MR, 3.6% CT, and 6.8% TTE intraobserver, and 5.1% MR, 10.6 CT, and 8.9% interobserver
Sinus of ******** -- 0.6% MR, 2.7% CT, and 3.9% TTE intraobserver, and 2.7% MR, 4.7% CT and 4.7% TTE interobserver
Ascending aorta -- 1.4% MR, 4.5% CT, and 10.3% TTE intraobserver, and 2.1% MR, 4.8% CT, and 10.4% TTE interobserver
Researchers noted the close agreement between MR and CT, particularly among the smallest (0.38 mm bias), largest (0.39 mm bias), and average (0.48 mm bias) annulus sizes.
On the other hand, transthoracic echo significantly underestimated the largest aortic valve annulus diameter (4.52 mm bias) compared with both MR and CT (P<0.0001).
Of the 133 patients in whom all three imaging exams were performed, MR and CT agreed on the same TAVI size (based on the manufacturer sizing for the CoreValve) in 67 patients, while MR and TTE agreed in 18 patients and CT and TTE in 25 patients.
"This study demonstrates that aortic root measurements made by both cardiac MR and cardiac CT are highly reproducible and show close agreement," researchers concluded. "In contrast, TTE-derived measures display higher variability and significantly underestimate aortic valve annulus size compared with cardiac MR and cardiac CT."
They noted that optimal coverage of the entire aortic valve annulus helps prevent aortic regurgitation and achieving optimal coverage will be especially important as a greater range of valve sizes comes to market.
The best way to achieve optimal coverage, researchers said, is by using the largest aortic valve annulus diameter. And in this study, MR and TTE agreed on size selection in 14%, while CT and TTE agreed in 19%.
"Given this poor agreement, we suggest a 3-D imaging modality be used to determine the largest annulus size and corresponding appropriate TAVI," they concluded.
A limitation of the study is that the authors did not evaluate transesophageal echocardiography, which is associated with good TAVI-sizing results, or 3-D TTE.
Also, given the differences between MR and CT -- namely a more rapid assessment with CT but with the potential for contrast-induced nephropathy -- researchers suggested that studies be conducted to determine which patients are best suited for each modality.
Another limitation of the study was the small (9%) number of patients who received the Sapien valve. Investigators therefore did not assess between-group differences.