乌鸦传媒

This article is a collaboration between 乌鸦传媒 and:

Three-dimensional CT imaging may be superior to two-dimensional transesophageal echocardiography for measuring the aortic annulus before transcatheter aortic valve implantation (TAVI), a pair of studies concluded.

Of 88 patients who underwent transesophageal-guided echocardiography (TEE), a retrospective analysis showed that CT imaging would have reassigned 45% of them to a different size Sapien valve, Raj R. Makkar, MD, of Cedars-Sinai Medical Center in Los Angeles, and colleagues reported.

Specifically, 26 of 60 patients would have received a 26 mm device rather than the smaller 23 mm device, and 17 of 36 would have had annuli too large for a 26 mm device, the researchers wrote online in the Journal of the American College of Cardiology.

Makkar and colleagues also noted that real-time CT-guided annular sizing resulted in "excellent" hemodynamic outcomes. Of the 40 patients in this cohort, only two had moderate paravalvular aortic regurgitation following TAVI.

One of these cases had severe leaflet calcification and the other had extensive calcification in the left ventricular outflow tract (LVOT).

A univariate analysis of clinical, echocardiographic, CT, and procedural variables revealed that only LVOT calcification and maximum annular dimension assessed by CT were predictive of significant paravalvular regurgitation.

A multivariable model excluding CT-derived maximum dimension but including circular dimension found, again, that LVOT independently predicted regurgitation, as did CT-derived circular dimension.

There were no differences between transthoracic echo (TTE) and TEE assessments, but many CT-derived measurements significantly differed from TEE measurements. These included the following CT versus TEE measurements:

• Maximum dimension -- 25.6 mm versus 27.8 mm (P<0.001)
• Mean dimension -- 23.2 mm versus 24.7 mm (P=0.001)
• Circular dimension -- 23.6 mm versus 25.2 mm (P<0.001)
• Area dimension -- 23 mm versus 24.4 mm (P=0.001)

The delta of difference between CT- and TEE-derived assessments was significant for maximum dimension (1.3 mm versus 3.7 mm) and circular dimension (0.7 mm versus 1 mm). The P value for each was <0.001.

All patients were enrolled in the PARTNER trial at Cedars-Sinai. Baseline characteristics were similar between the echo and CT groups. Those with compromised kidney function did not have contrast CT imaging.

"The main problem with 2D echo is that it fails to appreciate the noncircular geometry of the aortic annulus, whereas 3D CT appears to be superior in that respect," researchers wrote.

In the overall PARTNER trial cohort, 12% of patients in the TAVI arm had moderate or severe paravalvular aortic regurgitation compared with 0.9% in the surgical arm. "Inappropriate sizing is likely to be a major mechanism of paravalvular aortic regurgitation," Makkar and colleagues said.

The current study was limited because it was a single-center retrospective analysis. The results do not necessarily apply to other types of valves. In addition, other three-dimensional imaging techniques such as 3D TEE and MRI might be useful in those with kidney dysfunction.

In the second study, with 102 patients, valve undersizing was predictive of regurgitation following TAVI, Jonathon Leipsic, MD, of St. Paul's Hospital in Vancouver, British Columbia, and colleagues reported online in the Journal of the American College of Cardiology.

These researchers also used the Sapien valve (Edwards Lifesciences), but 90% of patients received the next-generation valve, Sapien XT, which is smaller than the first-generation model.

Those with compromised kidney function did not undergo CT imaging in this trial either. Also, regurgitation caused by malposition of the valve was not included in the final analysis because it is "a distinct and separate cause of paravalvular regurgitation."

Leipsic and colleagues found that age, annular eccentricity, and sex did not predict regurgitation.

Rather, differences between the valve diameter and TEE-derived diameter were moderately predictive (area under the curve [AUC]: 0.70, 95% CI 0.51 to 0.88), whereas differences between the valve size and CT-derived annular size (using diameter, area, and circumference measures) had moderate to good predictive value, they wrote.

For example, the AUC for each CT reader for mean diameter was 0.81, 0.85, and 0.84. For annular area, the AUC was 0.80, 0.79, and 0.86. For annular circumference, the AUC was 0.76, 0.76, and 0.86.

When CT measures of diameter and area revealed an undersized valve, it was associated with moderate to severe regurgitation.

Leipsic and colleagues stopped short of suggesting CT is superior to TEE. Rather, they said that CT measurements "can provide complementary and additive information."

"However, we do believe that the two-dimensional TEE annular diameter may underestimate the 'true' annular size," they added.

Because CT can offer a lot of information, the investigators said that mean diameter and area are the two most reproducible and predictive CT measurements for regurgitation.

The study was limited by an overall low rate of paravalvular regurgitation. The researchers also did not measure severe calcification with CT and this could have an impact on regurgitation severity.

From the American Heart Association: