Open-heart transseptal transcatheter mitral valve implantation with mitral annular calcification

Abstract

Mitral valve replacement (MVR) using a prosthesis of transcatheter aortic valve implantation for mitral annular calcification (MAC) is an alternative procedure which is very different from a conventional MVR and is safe. Acceptable results including mortality and morbidity have been reported. We report the case of a male patient treated with this procedure. There was a risk of left ventricular rupture due to extensive decalcification, so we performed transcatheter mitral valve implantation (TMVI). Since he had a history of surgical aortic valve replacement (AVR), we chose open-heart transseptal approach so as not to interfere with the aortic prosthesis. TMVI has a risk of an iatrogenic left ventricular outflow tract (LVOT) obstruction. Using open-heart transseptal approach, we could easily check an anterior commissure of mitral annulus to prevent LVOT obstruction.

<Learning objective: Transcatheter mitral valve implantation (TMVI) using a transcatheter aortic prosthesis is safe for mitral valve disease with mitral annular calcifications. This patient had a history of surgical aortic valve replacement, so we performed open-heart transseptal TMVI to prevent interference with the aortic prosthesis and the iatrogenic left ventricular outflow tract obstruction.>

Introduction

Decalcification of mitral annular calcification (MAC) sometimes leads to complications such as intractable hemorrhage and left ventricular rupture. Prior aortic valve replacement (AVR) immobilizes the mitral annulus [1]. Herein, we report open-heart transseptal transcatheter mitral valve implantation (TMVI). This procedure has some risks, but has many advantages compared with percutaneous TMVI [2].

Case report

An 80-year-old man (101 kg, 175 cm) had a history of surgical AVR with a bio-prosthesis and a maze procedure, and mitral stenosis was diagnosed without an indication for surgery. Before this operation, mitral infective endocarditis occurred. The infection was healed with antibiotic agents, but the mitral complex broke. He　displayed dyspnea on effort with moderate mitral stenosis and severe mitral regurgitation. His New York Heart Association class was Ⅱ, his Society of Thoracic Surgeons score was 6.57%, and his Euro SCORE Ⅱ was 4.69%. At first, our heart team considered percutaneous TVMI, but the risks of left ventricular outflow tract (LVOT) obstruction and prosthesis migration were high. Therefore, open-heart TMVI was performed. Echocardiography showed moderate mitral stenosis with a mean pressure gradient of 7 mmHg, a mitral valve area of 1.49 cm², and MAC. Mitral regurgitation was severe due to prolapse of the P3. Tricuspid valve regurgitation was trivial with a dilated annulus (44 mm). The aortic valve prosthesis was functional and had no evidence of infection. The left ventricular ejection fraction was 70% with dilated left atrium. Cardiac catheterization showed a pulmonary artery wedge pressure of 70 mmHg. Computed tomography (CT) showed MAC, and the annulus perimeter was 112 mm (Fig. 1B).

Fig. 1.

(A) Preoperative echocardiography showed moderate mitral stenosis, MAC, and severe mitral regurgitation due to P3 prolapse. (B) Preoperative computed tomography showed MAC and mitral annular perimeter was 112 mm.

MAC, mitral annular calcification.

After a median sternotomy on cardiopulmonary bypass with cardioplegic cardiac arrest, we made an extended transseptal incision. The A2 valve was resected to avoid LVOT obstruction due to TMVI. A 27-mm Epic sizer (St. Jude Medical, St Paul, MN, USA) was passed through the annulus, and we selected a 29-mm Edwards Sapien (Edwards Lifesciences, Irvine, CA, USA). For anchoring, eight 2-0 braided polyester sutures with pledgets were placed at soft part around the annulus (Fig. 2A). After washing the prosthesis, we secured the prosthesis and a felt strip using a single 5-0 polypropylene suture covering the polyethylene terephthalate portion of the prosthesis (Fig. 2B). The prosthesis was mounted on the balloon upside down compared with standard fashion. Using the Safari XS (Boston Scientific, Natick, MA, USA) guidewire to avoid injuring the left ventricle, the balloon was inflated with 2 mL more than a nominal, and the prosthesis was deployed (Fig. 2C). This overinflation was performed to avoid prosthesis migration. The volume of the balloon inflation depends on each case. We sutured the felt strip of the prosthesis with anchoring sutures to avoid paravalvular leak (PVL) and valve embolization; the sutures were tied down using a Cor Knot (LSI Solutions, Victor, NY, USA) to avoid interference with the prosthesis (Fig. 2D). After closing the septum, the Kay annuloplasty was performed at tricuspid valve. The total cross-clamp time was 62 min. Transesophageal echocardiography showed no significant PVL and no LVOT obstruction. One week after the operation, an echocardiography showed mild PVL at mitral prosthesis.

Fig. 2.

(A) Anchoring sutures around the mitral annulus. (B) A felt strip was sewn to the Sapien 3 to cover its polyethylene terephthalate portion. (C) The balloon was inflated, and the valve was deployed under direct visualization. (D) Anchoring sutures were tied with felt strip of the prosthesis using a Cor Knot device.

Discussion

It is often that we decide to perform this procedure during the operations. This patient was taken to CT before his operation, but we do not take patients’ CT scan before the operation usually. So, we estimate whether we can perform surgical MVR or not during the operations. Many patients who undergo this procedure have heavy calcification at more than 75% of the circumference of the annulus. If we take patients’ cardiac CT scan, we can estimate the perimeter of the mitral annulus, also we can estimate neo-LVOT [3].

Various methods to manage the MAC have been reported. Bito et al. reported the half and half technique to overcome a posterior MAC during MVR. In that operation, non-everted horizontal mattress sutures were placed on the anterior annulus and everted sutures on the left atrium wall close to the posterior MAC to fix the prosthesis [4]. Guerrero et al. reported a multicenter global registry of TMVI. They reported open transatrial, percutaneous transseptal, and transapical delivery, and open transatrial delivery was superior because of a high technical success rate, low mortality, and low complications [2]. However, their follow-up data showed a high mortality rate. Thirty-day and one-year all-cause mortality was 25% and 53.7%. They concluded that LVOT obstruction was the most important and independent predictor of 30-day and 1-year mortality [5]. With open-heart TMVI, surgeons can resect the mitral anterior leaflet to prevent LVOT obstruction and suture the prosthesis and annulus to prevent PVL and valve embolization [2]. Flannery et al. reported a patient who had previous transcatheter aortic valve implantation and underwent MVR using a transseptal approach to avoid interference with the aortic prosthesis. They implanted an aortic bio-prosthesis upside down at the mitral position because of a small mitral annulus [1]. To prevent procedure-related LOVT obstruction, preoperative and intraoperative measurements are important. Aorto-mitral angle which can be obtained with echocardiography even without CT scans is important in addition to the neo-LVOT. Aorto-mitral angle <120゜is considered to be at risk for LVOT obstruction [6]. A percutaneous prevention method for LVOT obstruction was reported. A percutaneous incision in the anterior mitral valve leaflet before TMVI is performed, called ‘LAMPOON’ [7]. This procedure may be considered if the patient is too fragile and it is difficult to open the chest.

Open-heart transseptal TMVI is useful in patients with MAC who have a history of prior AVR, and we can also prevent LVOT obstruction, PVL, and valve embolization.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest

Dr L. Leroux is proctor for Edwards Lifescience.

Other authors declare that there are no conflicts of interest.