Study Population

The study cohort comprised 249 consecutive patients with severe symptomatic aortic stenosis who underwent TAVI from January 2009 to August 2014.

Seventy-nine patients underwent TAVI without BPD, and 170 patients underwent TAVI with BPD. All patients were evaluated by a multidisciplinary team and were considered to be at high risk or to have contraindications to surgical treatment. Before the procedure, patients underwent TEE assessment of the aortic valve. All patients gave written informed consent for TEE imaging and TAVI, in accordance with a protocol approved by the institutional review committee.

Two-Dimensional (2D) and Three-Dimensional (3D) TEE Imaging

The equipment used was an iE33 xMATRIX echocardiographic system (Philips Ultrasound, Bothell, WA), with 3D TEE capabilities. All images obtained were processed with the quantitative analysis software QLAB (Philips Ultrasound). Different operators in our lab performed preprocedural TEE imaging. In every case, image acquisition and postprocessing of the 3D images were supervised, performed, and rechecked by our most experienced operator in TEE imaging related to TAVI.

The initial preimplantation TEE study should confirm the severity of aortic stenosis by 2D and 3D imaging, as well as by Doppler interrogation of the valve. Our TEE protocol, through the use of both 2D and 3D imaging, offered all details of the aortic valve features needed to determine whether patients were optimal candidates for TAVI without BPD.

Measurement of the aortic annulus was obtained at 120° to 140°, ensuring a perpendicular position of the aorta relative to the left ventricle. By means of QLAB’s multiplanar tool, coaxial planes were used to generate an orthogonal plane that allowed the measurement of the 3D cross-sectional area and diameter of the aortic annulus and root. Thickness of the aortic leaflets, mobility, and calcium distribution were assessed in 45° and 120° to 140° views in end-diastole using both 2D and 3D tools ( Figure 1 ). Distribution and degree of valve calcification were classified as mild (1), moderate (2), or severe (3). To classify valve calcification, the thickness of each leaflet was measured; the mean value of the three leaflets together was obtained as well. Calcification was considered mild when the mean leaflet thickness was <3 mm, with an absence of nodules at the cusps and annular calcification; it was classified as moderate when mean leaflet thickness was 3 to 5 mm with small calcium nodules at the cusps and localized calcification at the aortic annulus and severe when leaflet thickness was >5 mm, with large nodules and diffuse calcification of the aortic annulus. In addition, calcium distribution on the aortic valve was designated as regular or irregular, with irregular distribution considered to be asymmetric presence of calcium at the commissures or leaflets.

(A) Two-dimensional orthogonal views of the aortic valve. (B) Three-dimensional multiplanar reconstruction of the aortic valve confirming a central orifice ( asterisk ) and regular calcification of the valve.

The presence of calcium nodules near the coronary ostia or at the left ventricular outflow tract (LVOT) was described to evaluate its potential effect on prosthesis deployment and function.

The mobility of aortic cusps was visually classified as slightly restricted (1), moderately restricted (2), or severely restricted (3). For this purpose, we took into account commissure fusion. We considered mobility to be slightly restricted when all commissures seemed to be open, moderately restricted when there was one fused commissure, and severely restricted when there were two or more fused commissures.

Aortic valve residual orifice was classified into two groups by using a 3D multiplanar tool, obtaining a short-axis plane of the valve: central or eccentric and irregular orifice ( Figure 2 ).

(A) Heavy calcification of the aortic leaflets and commissures. (B) Irregular and bulky calcification of the aortic valve in a patient with valve area < 0.4 cm ² . (C) Central valve residual orifice with fusion of one commissure. (D) Aortic valve with no commissure fusion and regular calcium distribution. (A,B) Examples of aortic valves with unfavorable features for TAVI without balloon predilation. (C,D) Examples of aortic valves with favorable features for this technique.

Aortic valve area was estimated by the continuity equation as well as with 3D planimetry. Aortic regurgitation before TAVI was classified into one of four groups, absent (0), mild (1), moderate, (2) or severe (3), according to the European guidelines on valvular heart disease.

Our criteria for favorable or unfavorable TAVI without BPD are presented in Table 1 . After echocardiographic data were reviewed with the interventional team, patients who fulfilled all favorable criteria (valve area > 0.4 cm ² , central orifice, calcification ≤ grade 2, mobility ≤ grade 2, no LVOT calcification, absence of calcium nodules in the “landing zone” of the prosthesis, and aortic regurgitation ≤ grade 2) underwent TAVI without BPD ( Figure 3 ). The interventional team defined the type and size of the prosthesis according to the manufacturer’s recommendations.

Selection of patients with a favorable profile for performing TAVI without BPD. Calcification was considered to be mild (1) when the mean leaflet thickness was <3 mm, with absence of nodules at the cusps and annulus calcification; it was classified as moderate (2) when mean leaflet thickness was 3 to 5 mm with small calcium nodules at the cusps and localized calcification at the aortic annulus and severe (3) when leaflet thickness was >5 mm, with large nodules and diffuse calcification of the aortic annulus. Mobility of the aortic cusps was visually classified as slightly restricted (1) when all commissures seemed to be open, moderately restricted (2) when there was one fused commissure, and severely restricted (3) when there were two or more fused commissures.

Within the procedure, 3D TEE imaging was used in almost all steps of TAVI. Once the device was delivered into the native aortic valve, and by using the 3D multiplanar tool, we ensured that 50% to 60% of the total device height was at the LVOT; after confirming proper positioning, the device was deployed just below the aortic annulus in a subcoronary location, and its function was immediately evaluated.

Procedural success was defined as successful delivery and deployment, correct positioning of the device in the proper anatomic location with adequate performance of the prosthetic heart valve, and a final mean transaortic gradient ≤ 20 mm Hg, without aortic regurgitation ≥ grade 2 and no valve embolization or need to implant a second valve or conversion to surgery. Postprocedural aortic regurgitation degree was classified in accordance with Valve Academic Research Consortium 2 criteria.

After the procedure, patients were closely observed for periprocedural and 30-day major adverse cardiac and cerebrovascular events (stroke, myocardial infarction, aortic valve reintervention or surgery, need for new permanent pacemaker implantation (PPI), procedure-related mortality, and all-cause mortality). Subsequent follow-up was done in an outpatient setting.

Statistical Analysis

Continuous variables are reported as mean ± SD. They were compared by using two-tailed Student t tests. Categorical variables are expressed as frequencies and percentages and were compared by using χ ² tests and Fisher exact tests as appropriate. All tests were two sided, and differences were considered to be statistically significant at P < .05. Statistical analysis was performed with Stata/IC version 12.1 (StataCorp LP, College Station, TX).