Transcatheter Tricuspid Valve-in-Valve Replacement with an Edwards Sapien 3 Valve

Case Report

In February 2015, a 49-year-old woman presented with symptoms of severe right-sided heart failure and with a history of prior polysubstance abuse and of end-stage renal disease necessitating hemodialysis. She also had chronic viral hepatitis C with suspected cirrhosis and had developed infective endocarditis, resulting in severe tricuspid regurgitation. In 2008, she had undergone surgical TV replacement with a 31-mm Carpentier-Edwards Perimount bioprosthesis (Edwards Lifesciences).

The patient had begun to experience signs and symptoms consistent with right-sided heart failure. She underwent multiple hospital admissions for fatigue, severe dyspnea, peripheral edema, anasarca, and recurrent ascites. The current physical examination revealed lethargy, an elevated jugular venous pressure with cannon A waves, a grade 3/6 diastolic murmur heard best at the left lower sternal border, diminished breath sounds at the bilateral bases, abdominal ascites, and pitting edema in the bilateral lower extremities. During her evaluation, transthoracic echocardiography revealed severe bioprosthetic TV stenosis (Fig. 1). Her mean transvalvular gradient was 14.5 mmHg; her peak transvalvular gradient, 23 mmHg; and her peak velocity, 2.42 m/s.

View Figure

• Download as Powerpoint
• Download Figure

The heart team decided to proceed with off-label use of an Edwards Sapien 3 valve because of the patient's overall frailty and significant comorbid conditions: her estimated prohibitive surgical risks (Society of Thoracic Surgeons scores for mitral valve replacement) were 14% for death and 43% for morbidity or death.

After informed consent was obtained, the patient was brought to the cardiac catheterization laboratory in a fasting, nonsedated state. Both of her groins were prepared and draped in the usual sterile fashion. The patient received conscious sedation by means of systemic fentanyl and midazolam administration. With use of a 4F microintroducer kit (Vascular Solutions, Inc; Minneapolis, Minn), access to the right femoral vein was attained via the modified Seldinger technique to place an 11-cm, 7F sheath.

An additional left femoral vein puncture was performed via a micropuncture needle, and a 6F venous sheath was advanced without difficulty. A deflectable quadripolar catheter (St. Jude Medical, now part of Abbott Laboratories; St. Paul, Minn) was advanced into the rather large right atrium and into the ostium of the coronary sinus. A pacemaker tip was initially placed in the anterior intraventricular venous branch; however, because this particular patient's anatomy produced high torque while the leads were seated at the ostium of the coronary sinus, the leads were redeployed more distally into a middle cardiac vein location, and the tips of the leads were extended further into that vein to ensure optimal stability during sheath manipulation and valve deployment. The patient was then given heparin in order to achieve an activated clotting time of longer than 250 s.

Next, a 7F Swan-Ganz monitoring catheter (Cardinal Health; Dublin, Ohio) was advanced into the 7F sheath and across the tricuspid bioprosthesis, to the pulmonary capillary wedge position. Hemodynamic values were obtained upon insertion into the right atrial, right ventricular, pulmonary arterial, and pulmonary capillary wedge positions. The mean right atrial pressure was 28 mmHg, with an A wave of 33 mmHg. A 0.035-in Amplatz Extra-Stiff guidewire (Cook Medical Inc.; Bloomington, Ind) was then advanced into the Swan-Ganz catheter and parked in the distal right lower pulmonary artery for support. The Swan-Ganz catheter was then removed, and the 7F, 11-cm sheath was exchanged over the Amplatz wire for a 16F Edwards eSheath Introducer Set (Edwards Lifesciences) (Fig. 2).

View Figure

• Download as Powerpoint
• Download Figure

Next, balloon valvuloplasty of the tricuspid bioprosthesis was performed, with and without rapid ventricular pacing, by means of a 20-mm × 4.5-cm True Dilatation^® Balloon Valvuloplasty Catheter (BARD Peripheral Vascular, Inc.; Tempe, Ariz) (Fig. 3). The valvuloplasty balloon was then removed. The 29-mm Edwards Sapien 3 transcatheter heart valve with the 29-mm Edwards Commander delivery system (Edwards Lifesciences) was inserted over the Amplatz wire in the inferior vena cava. Under fluoroscopic guidance, the valve was then advanced into the tricuspid bioprosthesis and positioned in the usual manner for Sapien 3 valve deployment, with use of the prior stent's frame as a reference. Under rapid ventricular pacing and precise positioning, the valve was deployed with excellent results (Fig. 4). The delivery system and the Amplatz guidewire were subsequently removed. Post-deployment transthoracic echocardiograms confirmed a well-seated valve with optimal placement and no paravalvular regurgitation. Doppler measurements showed improvement of the transtricuspid gradient. The mean transvalvular gradient improved to 3 mmHg and the peak velocity to 1.5 m/s. Right-sided heart catheterization revealed immediate improvement in the patient's intracardiac hemodynamic values. The mean right atrial pressure improved to 15 mmHg, with an A wave of 19 mmHg. Repeat echocardiographic evaluation of the valve-in-valve bioprosthetic tricuspid function showed a mean transvalvular gradient of 2 mmHg and a peak velocity of 0.69 m/s, suggesting a substantial reduction in gradient severity (Fig. 5).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The patient's venous sheaths were removed, and hemostasis was obtained via 2 figure-of-eight sutures. The patient tolerated the procedure well with no immediate periprocedural sequelae. She was transported to the intermediate care unit in stable condition. She reported immediate improvement in her dyspnea and fatigue. She was discharged from the hospital after an uneventful postoperative stay of 10 days and was seen as an outpatient 4 months later, when she underwent liver biopsy for hepatitis C-related cirrhosis.

Discussion

The concept of using TAVI in repair of degenerated bioprosthetic valves was first presented with a transfemoral CoreValve system (Medtronic) and then was adopted by Edwards.^10–12 Valve-in-valve has been an increasingly appealing strategy, but the data are few on percutaneous treatment of severe TV degeneration with the Sapien 3 device.⁹ Herein we describe a case of a patient with severe bioprosthetic TV stenosis who was successfully treated with transfemoral TAVI via the 29-mm Edwards Sapien 3 Commander system.

This case demonstrates the optimal performance of the Sapien 3 valve in this particular anatomy. We selected the Sapien 3 because of our extensive experience in percutaneous valve replacement with this device. The ability to steer a 16F eSheath and Commander delivery system of relatively low profile seems particularly suitable for use on a transfemoral route in TV replacement.

Our patient had a 31-mm Carpentier-Edwards Perimount bioprosthesis. The industry-reported internal diameter of the stent is 31 mm, but the true internal diameter is closer to 28.5 to 30 mm because of the pericardial tissue sewn inside the stent frame. The effective internal prosthetic area is, therefore, approximately 638 to 707 mm². The 29-mm Edwards Sapien 3 valve, with a height of 22.5 mm and a larger coverage area of up to 683 mm², was the optimal choice in this patient. The development of the polyethylene terephthalate outer skirt on the proximal base of this valve has the added benefit of reducing paravalvular regurgitation after deployment. Transjugular, transatrial, and transfemoral vascular access for this procedure have been described.^13,14 We chose transfemoral access because a small-bore 16F sheath used with the 29-mm Edwards Sapien 3 proved to be well suited to the anatomic challenge of routing this valve into the tricuspid position.

Last, we performed balloon valvuloplasty with rapid ventricular pacing from the middle cardiac vein. Conventional aortic valve deployments are performed while on a rapid ventricular pacing sequence through a transvenous catheter—across the tricuspid annulus, in close contact with the right ventricular septum. Percutaneous TV deployment is particularly challenging because transvenous pacemakers are customarily deployed across the TV into the right ventricular septum. However, given that valve deployment in this instance was essentially in the TV position, the aim of the temporary wire placement in the middle cardiac vein was to enable rapid RV pacing with minimal chance of dislodging pacing leads during manipulation of the Sapien 3 delivery system. To this end, a deflectable quadripolar catheter in the middle cardiac vein location provided reliable ventricular pacing without engaging the tricuspid annulus. Pacing from the left ventricle would have obliged unnecessary aortic-valve-crossing maneuvers. Rapid ventricular pacing was ultimately performed and was shown to provide stabilization and to improve deployment accuracy in this consciously sedated patient who had substantially elevated right atrial pressures.

Transfemoral access and coronary sinus pacing might provide considerable procedural improvement when compared with a prior reported case⁸ of successful transjugular deployment of an earlier-generation device without rapid pacing. The technique described by Mortazavi and colleagues,⁸ although successful, might have required surgical hemostasis after the device was retracted. Further, deployment in the absence of ventricular pacing can be challenging. A transfemoral approach and a pacing sequence on deployment have made this case similar to that of a conventional transcatheter valve implantation and therefore easier to plan and execute.

Particular care was used during the procedure while manipulating catheters, the delivery system, and the support wire, because the Sapien 3 valve and delivery system was originally designed to navigate a more resistant structure (the aorta) and to deploy in the aortic annulus. The Amplatz Extra-Stiff wire provided exceptional support while we navigated the inferior vena cava, and the steering ability of the Commander system minimized trauma to the right atrium and ventricle. Close clinical follow-up, a full surgical team, and extracorporeal membrane oxygenation on standby—with echocardiography both intraoperatively and during early postoperative recovery—have been used to diagnose and treat atrial perforations sooner rather than later.

In conclusion, the Sapien 3 Commander System seems to be suitable for valve-in-valve transcatheter TV replacement because of its low-profile, technically easy deployment. Although ours was an off-label use of this device, it might provide a valid, minimally invasive alternative in poor surgical candidates or in patients in whom comorbidities pose substantial risk. Careful radiologic, angiographic, and anatomic planning remain the mainstays for procedural success as transcatheter valve replacement continues to progress in its application to extreme situations.