Severe Aortic Stenosis Treated with Three Self-Expandable Valves: Embolization of the First Two and Successful Implantation of a Larger One

Abstract

Transcatheter aortic valve embolization is a serious complication of transcatheter aortic valve replacement (TAVR). We present the case of a patient who required the implantation of three transcatheter aortic self-expandable valves (SEVs) owing to the embolization of two of them.

1. Introduction

Transcatheter aortic valve replacement (TAVR) is, nowadays, the preferred treatment option for severe aortic stenosis in older patients with severe comorbidities due to its excellent outcomes and low complication rate, thanks to device development and the operators’ experience [1,2].

Device embolization is defined as the movement of the device after initial placement leading to the complete detachment of the valve from the aortic annulus and movement into either the aorta or the left ventricle [3]. TAVR embolization is a rare but serious complication with a reported incidence of 0.3%, associated with higher mortality and stroke rates [4]. Most embolization cases (90%) take place intraprocedurally, offering the opportunity to fix the problem immediately. The most common locations for embolization are the ascending aorta (38% [4]–52% [5]) and the left ventricle (31% [4]–48% [5]), followed by the descending aorta (23%) and the aortic arch (8%). Embolization can occur with either self-expanding (28%) or balloon-expandable TAVR (72%) [4]. Optimal strategies to manage the embolization of TAVR devices are unclear; valve-in-valve (TAV-in-TAV) is often used [5,6].

2. Case Summary

A high-risk 84-year-old man with severe aortic stenosis and poor functional status (NYHA III-IV) was referred to our hospital for TAVR. He had a history of hypertension, dyslipidemia, chronic coronary disease, chronic renal failure grade IV and paroxysmal atrial fibrillation. Previously, he had a long history of chronic coronary disease (1996: aortocoronary by-pass from left internal mammary artery [LIMA] to left anterior descending artery [LAD], saphenous vein grafts [SVG] to marginal branch [MB] and right coronary artery [RC]. In 2011, the SVGs to the MB were occluded, and at that moment, a coronary stent was implanted in the RC through by-pass). Given that medical stabilization was not achieved, invasive management was needed. The STS score for mortality at 30 days was 9%.

Complementary assessment:

Electrocardiography revealed normal sinus rhythm without conduction system delays, and left ventricular hypertrophy.

The echocardiogram showed preserved left ventricular function (FEVI 60%), a native aortic valve peak velocity of 4.81 m/sg, gradients of 92/60 mmHg (maximum/mean), with an aortic valvular area of 0.56 mm² according to a continuity equation, and a pulmonary systolic pressure of 60 mmHg.

Coronary angiography demonstrated the patency of the LIMA and SVGs to the RC.

All laboratory values were normal, except for serum creatinine clearance (20 mL/min/1.73 m²) and the N-terminal pro-brain natriuretic peptide level (6487 pg/mL; normal range: 125 pg/mL).

Preprocedural computed tomography (CT) measurements are shown in Figure 1. The aortic root size and measurements were considered favorable to accommodate a TAVR. The three-cuspid aortic valve (AV) was heavily calcified (Agatston score: 3508 UH). The aortic annulus was elliptical in shape with moderate calcification penetrating into the left ventricular (LV) outflow tract 7 mm under the left coronary sinus. Regarding vascular access, the ileofemoral axis showed good caliber, but it was moderately tortuous with significant calcification, including at the femoral puncture site.

The heart team decided to implant an SEV Evolut R^® nº 29 (Medtronic, Inc., Minneapolis, MN, USA) through an open-surgical right femoral artery approach under deep sedation and local anesthesia. Pre-dilatation was performed with an 18 × 40 mm semicompliant balloon, and the prosthesis was implanted at a 3 mm target depth at first attempt. Unfortunately, the removal of the delivery system (DS) caused the dislodgement of the valve due to the collision of the nosecone with the prosthesis inflow frame, resulting in embolization to the ascending aorta (Figure 2).

Hemodynamic collapse, caused by aortic regurgitation, occurred immediately after embolization. Therefore, a second prosthesis inside of the first one was implanted (another SEV Evolut R^® nº 29). Unfortunately, the hemodynamic status did not improve. As this second valve was in place and moderate aortic regurgitation was still present, we decided to post-dilate it (25 × 40 balloon). In order to progress the balloon, different maneuvers were attempted, with the last one being to progress a smaller balloon (18 × 40 mm balloon) (Figure 3a). However, when we retrieved the deflated balloon, the second prosthesis was also displaced up to the ascending aorta. Both valves fit together perfectly, one inside the other, in the aortic arch. At the time, there was no aortic regurgitation. Finally, a 31 mm Core Valve^® (Medtronic, Inc., Minneapolis, MN, USA) prosthesis was successfully implanted (Figure 3b). The depth of implantation was 11 mm from the annulus with no aortic regurgitation (Figure 3c) and a residual aortic gradient of just 9 mm. The patient developed a complete A-V block, remaining pacemaker-dependent.

An echocardiogram at discharge and follow-up showed preserved left systolic function, a transprosthetic gradient of 10 mmHg and moderate systolic pulmonary hypertension (40–45 mmHg). The patient remained stable without sequelae at a ten-year follow-up, with NYHA functional class II.

3. Discussion

Transcatheter aortic valve (TAV) embolization usually occurs during or shortly after valve deployment for several reasons; careful procedure planification to avoid sizing errors is needed. Predisposing factors to embolization include the following: the use of SEVs, bicuspid aortic valves and annular eccentricity. Procedural factors contributing to embolization are as follows: post-dilation, fast-rate pacing failure, inadequate prosthesis release height, the entrapment of the prosthetic strut with prosthesis release material and the underexpansion of the prosthesis [4,5,6,7]. The need for three TAVs in the same procedure is truly extraordinary. To date, only one case has been published of a patient with severe aortic stenosis who required three valves in the same procedure. As in our case, two self-expanding valves were embolized, and finally, a balloon-expandable valve was successfully implanted [6].

In our patient, we selected an SEV due to the elliptical shape and the penetrating calcium. In this educational case, we can observe three reasons that could explain the two-TAVR embolization. Firstly, encouraged by previous rather positive results with repositionable SEVs, we attempted first a very high implant in order to avoid atrioventricular block [8]. Secondly, when one TAV-in-TAV is implanted in this context, snaring and retrieving the first prosthesis to the ascending aorta are recommended for two reasons: to achieve the correct expansion of the second valve and to avoid coronary obstruction [9]. Thirdly, in border-line annulus dimensions, a larger prosthesis should be considered in case of TAV embolization in the ascending aorta [10].

SEV embolization in the aorta can be managed effectively with the implantation of a second device, leaving the dislocated one safely in the aorta. When we fix it, we will have to pay attention to avoiding damage to the aortic wall and also avoiding the occlusion of the aortic arch branches with the pericardium skirt. If the procedure is successful, the reported clinical outcomes in the medium and long term are good [10,11]. In a series of eight patients followed by CT, in three of them, thick leaflets in the embolized prostheses are described. There was no evidence of valve migration, strut fracture, prosthesis-associated aortic complication and thrombosis associated with the embolized valve for all patients with TAVs repositioned in the aorta [11].

Mortality following bail-out TAV-in-TAV is higher than in regular TAVR procedures, with two-thirds of such patients surviving beyond a 1-year follow-up. Concerns in periprocedural TAV-in-TAV include permanent pacemaker implantation, coronary obstruction and an increased risk of endocarditis related to a long procedure and the number of devices used [12]. Long-term concerns would be valve degeneration or coronary re-access, and the last one might be particularly challenging since cuspid alignment is rarely achieved with both devices (fortunately, our patient had a previous patent coronary by-pass) [5].

4. Conclusions

TAVR embolization could be treated with TAV-in-TAV as a bail-out procedure; however, this increases the rate of complications. Training in preventing and treating such complications is needed. Long-term follow-up for our patient was successful.