Transcatheter mitral valve replacement (TMVR) has become a valuable treatment option in patients with severe mitral regurgitation (MR). Previous aortic valve replacement is considered a relative contraindication for the procedure due to the risk of left ventricular outflow tract (LVOT) obstruction and interactions between the two prostheses and anchoring mechanism. Moreover, in transapical redo surgery, complications at the access site may occur. In this case, the authors describe the successful TMVR as a tertiary cardiac surgery using the Tendyne® system.

An 83-year-old male patient presented with dyspnoea and angina under cardiac decompensation (NYHA III). He had previously been treated with coronary artery bypass grafting for severe three-vessel disease at the age of 64 and transapical transcatheter aortic valve replacement (TA-TAVR, 29 mm S3) for aortic valve stenosis at 79 years. Coronary treatment was complemented by percutaneous coronary intervention and stenting of the PL branch, also at 79 years. Moreover, he was known to have arterial and pulmonary artery hypertension, chronic obstructive lung disease and peripheral artery disease with a history of bilateral femoropopliteal bypass surgery (Table 1). Before the start of the complaint, he lived at home independently.

On admission, blood pressure was elevated (180/100 mmHg); heart rate (71 bpm) and room air oxygen saturation (98%) were normal. Pulmonary auscultation revealed reduced breathing sounds, in line with pulmonary oedema in the thoracic x-ray. The ECG showed a normal frequency sinus rhythm with left axis deviation and bifascicular block. Echocardiography revealed severe secondary mitral valve regurgitation (MR) with an effective regurgitation orifice area (EROA) of 29 mm² with a vena contracta width of 7 mm, and a regurgitation volume of 50 ml. The left ventricular function was moderately impaired (ejection fraction 36%, left ventricular end diastolic volume 215 ml) and there was moderate tricuspid regurgitation. Laboratory analysis on admission revealed elevated of NT-proBNP (6,417 ng/L). Due to elevated troponin T with an increasing trend, coronary angiography was performed, excluding renewed coronary stenoses.

The patient was discussed by a multidisciplinary heart team and initially primed for mitral valve transcatheter edge-to-edge repair, but the procedure was not successful due to strong tethering of the posterior leaflet. Considering the high-risk profile (STS-Score 7.5%, EuroScore II 35.4%) the patient opted for TMVR with the Tendyne® system.

Pre-procedural computed tomography (CT) simulation was used to determine the ideal access and exclude interactions between the mitral and aortic valve prosthesis. Special attention was paid to the neo-LVOT, which was predicted to have 10.6 mm (Figure 1A).

TMVR was performed under echocardiographic and fluoroscopic guidance, using the previous anterolateral thoracotomy for transapical access. The partially pulmonary adhesion was carefully dissected from the thoracic wall and the cardiac apex, revealing the pledges of the previous transapical procedure. For pucturing of the left ventricle with optimal angulation to the mitral position, a site posterolateral to the previous access was chosen. After apical pre-dilation with a 21F sheath, a 26F sheath was placed, through which the implantation was carried out without complications. Fluoroscopy by the end of the procedure confirmed optimal valve positioning (Figure 1B). The patient was extubated in the operating room and was admitted to normal ward on the first postoperative day.

Postoperative magnetic resonance imaging (MRI) showed optimal positioning of the prosthesis in the mitral anulus without any sign of paravalvular leakage, which was confirmed by echocardiography (Figures 1C,D). LV function remained moderately impaired. As predicted, there was no obstruction of the neo-LVOT. Haemoglobin was stable postoperatively with no need for transfusions.

The patient was discharged home seven days postoperatively. He subsequently participated in a three-week inpatient cardiac rehabilitation programme. There were no postoperative complications according to the MVARC criteria and no further inpatient admission was necessary in the year that followed. Echocardiography excluded paravalvular leakage or relevant transvalvular gradients (< 5 mmHg). LV function was only lightly impaired. The patient continued to live independently at home with improved exercise capacity (NYHA II).

Multi-valvular disease in elderly, multimorbid patients is an important issue to deal with. With more than 1,700 procedures performed worldwide, TMVR with the Tendyne® system (Abbott) is an emerging treatment strategy for severe mitral valve regurgitation in high-risk patients.

As it is potentially fatal, LVOT obstruction is of major concern in patients with TMVR. Implantation of the prosthesis leads to the formation of the so called neo-LVOT, confined by the native anterior mitral leaflet fixed in an opened position, the valve stent and the septum (1). LVOT obstruction may lead to acute hemodynamic deterioration or chronic heart failure due to an increased left ventricular afterload (1, 2). As a consequence, pre-procedural computed tomography simulation and measurement of the predicted LVOT has gained in importance (1).

An existing aortic valve prosthesis may add complexity as it influences aortomitral angulation. Moreover, LV hypertrophy as a consequence of long-standing aortic valve stenosis may narrow the LVOT. We therefore relied on CT simulation in this complex case to visualise the planned implantation and ensure there was an adequate LVOT area.

The feasibility of TMVR in the presence of a surgical or transcatheter aortic valve prosthesis has previously been described previously (3, 4). Similarly, simultaneous transapical implantations of an aortic and mitral valve prosthesis has been reported (5). However, in both aortic and mitral valve surgery, little is known about transapical re-intervention. Despite the fear of apical fragility, individual reports describe good results of redo transapical TAVR within a week, three or seven years after the initial procedure (6–8). As in our case, postoperative adhesions were easily controllable and the apical tissue was rated unexpectedly normal (6). In the cases mentioned of redo transapical TAVR, the same apical access site was used. For optimal angulation in our patient, we had to use a new apical access site posterolateral of the previous position. Transapical access may cause myocardial damage and scarring. However, the existence of two adjacent accesses did not lead to restriction of the left ventricular function in our patient. He showed moderately impaired LV function in both pre- and postprocedural echocardiography and MRI (ejection fraction 36% pre- and 37% post-procedurally). There was even an improvement to a lightly impaired left ventricular function in the one-year follow up.

In conclusion, this case demonstrates that a redo transapical access for TMVR as a tertiary cardiac procedure can be easily performed with good clinical results. Accurate pre-operative screening is crucial.