MDCT, MRI, TTE, TEE compared for aortic valve measurement

Mar 4, 2006

VIENNA - MDCT and MRI are emerging as powerful noninvasive cardiac imaging alternatives, but can they accurately estimate valvular morphology and function?

The preliminary answer is yes for both modalities, according to a Belgian study that compared MDCT and MRI to transthoracic and transesophageal echocardiography (TTE and TEE) for estimating aortic valve area in normal and stenotic valves.

In the study, Dr. Emmanuel Coche and colleagues from Cliniques Universitaires St. Luc in Brussels prospectively examined 41 patients (26 men, ages 60 ± 13 years) who were scheduled for mitral or aortic valve surgery. Those with typical contraindications to CT or MRI were excluded from those exams.

"Patients underwent multidetector-row CT, MRI, and transthoracic echocardiography in random order on two days, and transesophageal echocardiogaphy was performed immediately before surgery," Coche said in a presentation today at the European Congress of Radiology (ECR).

MDCT was performed on a 40-slice Brilliance CT scanner (Philips Medical Systems, Andover, MA) following administration of 120 mL of iodinated contrast. Images were acquired at 120 kVp, 600 mAs, and pitch of 0.2-0.24, and retrospectively gated ECG CT images were reconstructed every 10% of the cardiac cycle. The images were resliced into serial short-axis slices across the aortic valve (thickness 5 mm, gap 0 mm), Coche said.

MR images were acquired on a 1.5-tesla Philips Intera CV scanner. First, short-axis slices were acquired across the aortic valve using a cine balance BFFE sequence in 25 phases (thickness 5 mm, spacing 0 mm, TR 3.4, TE 1.7, flip angle 60°). Next, breath-hold phase-contrast images were acquired in 25 phases across the aortic valve (encoding velocity 700 cm/sec, slice width 10 mm, TR 4.7, TE: 2.8, flip angle 15°).

Two-dimensional transthoracic echocardiography is considered a routine gold standard to assess aortic valve stenosis, and it was performed on a Philips Sonos 7500 cardiac ultrasound scanner with a broadband probe using harmonic imaging (3.5 MHz/0.75 MHz harmonic hmaging), Coche said. Aortic valve area was measured by the continuity equation.

"If you measure the cross-sectional area and the mean velocity in the normal aorta, you will obtain the flow rate," he said. "And if you meaure by Doppler the maximum peak velocity, you will determine the stenotic area by this equation."

TEE was also performed on the Sonos scanner, using a multiplane probe to obtain short-axis views of the aortic valve around 45°. The aortic valve area measurement was obtained using a continuity equation with TTE and with planimetry using TEE, MDCT, and MRI, measured and repeated by two observers, Coche said. Planimetry of the aortic valve was performed at the leaflet tips in all systolic images to determine the aortic valve area. Finally, valve morphology was noted as bicuspid or tricuspid.

The results showed close concordance of the measurements. Aortic valve areas measured by MDCT (2.71 ± 1.7 cm²) and MR (2.67 ± 1.7 cm²) were very similar to TEE planimetry measurements (2.59 ± 1.7 cm²), and the differences were not statistically significant.

Bland-Altman plots for both MRI and MDCT showed greater dispersion of values for larger valves compared to smaller ones. And MDCT and MR measurements were larger than those obtained by TTE calculated using the continuity equation (2.21 ± 1.5 cm², p < 0.001).

Planimetric aortic valve measurements at CT and MR imaging correlated strongly with those measured by TTE (r = 0.95 and 0.96, respectively, p < 0.001) and TEE (r = 0.98 for both). As for morphology, 12 of the 41 patients had bicuspid aortic valves, seven of which were missed at TTE. However, all were correctly identified by MDCT and MRI and confirmed during surgery.

"If we perform a semiquantitative analysis and we classify the value to normal, mild to moderate stenosis, and severe stenosis, we had a perfect agreement between MDCT and TTE," Coche said. "The measurements were highly reproducible, with a very high intraclass correlation for intraobserver and interobserver analysis for the four techniques" (TTE 0.96/0.97, TEE 0.99/0.97, MDCT 0.97/0.97, MRI 0.99/0.97, respectively).

Coche concluded that MDCT and MRI accurately estimate aortic valve area compared to TTE and TEE. In addition, planimetry of the aortic valve area with both MDCT and MRI correlates well to planimetry by TEE, and with the continuity equation by TTE. However, all three planimetric methods slightly overestimate aortic valve area derived from the continuity equation.

"Both MDCT and MRI could be alternatives to the transthoracic echocardiography gold standard to asses aortic stenosis severity in patients with poor acoustic windows, and these test results should be useful for the comprehensive evaluation of patients, as they allow to simultaneously assess coronary vessel anatomy, cardiac function, and valve morphology," he said.

By Eric Barnes
AuntMinnie.com staff writer
March 5, 2006

Doppler US accurate for spectrum of aortic valve measurements, January 26, 2006

Long-term anticoagulant users show elevated coronary, aortic calcium, December 2, 2005

Mitral annular calcification strong risk factor for stroke, December 2, 2005