Transcatheter valve-in-valve replacement of degenerated bioprosthetic aortic valves: A single Australian Centre experience

Highlights

• Patients with degenerated surgical bioprosthetic valves may be at high risk for further surgery.

• Twelve patients had valve-in-valve implants for degenerated bioprosthetic aortic valves.

• There were no periprocedural deaths, myocardial infarcts, neurological events or major vascular complications.

• Two died after 3.6 and 4.5 years. Median survival for those remaining has exceeded 19 months.

• The survivors are stable with NYHA functional class I/II status although prosthesis–patient mismatch and prosthetic stenosis are a concern.

• Transcatheter valve-in-valve implantation is a safe and effective treatment for patients with failed bioprosthetic aortic valves for whom reoperation is considered to be unduly hazardous.

Abstract

Background

Patients with degenerated surgical bioprosthetic valves may be at high risk for further surgery because of age, comorbidities and the difficulties of repeat procedures. Percutaneous valve-in-valve implantation offers what may be a simpler and safer procedure.

Methods

From May 2009 to March 2014 at the Prince Charles Hospital 1625 patients underwent surgical aortic valve replacement while 262 underwent transcatheter aortic valve implantation. Twelve patients had valve-in-valve implants for degenerated bioprosthetic aortic valves.

Results

These implants were deployed successfully without major valvular or paravalvular regurgitation. There were no periprocedural deaths, myocardial infarcts, neurological events or major vascular complications. Two patients died after 1624 and 1319 days. Median survival for the remainder is 581 days; they are stable with New York Heart Association class I/II functional status although 4 have a degree of patient–prosthesis mismatch, one has moderate aortic regurgitation and one required surgery for a late aortic dissection.

Conclusion

Transcatheter valve-in-valve implantation is safe and effective treatment for patients with failed bioprosthetic aortic valves for whom reoperation is considered to be hazardous.

Introduction

Bioprosthetic aortic valves have been preferred to mechanical valves for the elderly and those at high risk of bleeding on anticoagulant therapy, but with improvement in design and growing experience there has been a trend to implanting them in younger patients [1], [2], [3]. Unfortunately they may be subject to structural degeneration and failure after 10–20 years [4], [5]. Surgical valve replacement is the standard of care for patients with failed valves but may be hazardous, particularly for elderly patients with multiple comorbidities [4], [6], [7], [8], [9], [10], [11]. Transcatheter aortic valve implantation (TAVI) has come to be seen as an alternative, but experience is still limited to several hundred patients worldwide [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. We report our experience with this procedure over the past five years.

Between May 2009 and February 2014 1625 patients have undergone surgical aortic valve replacement in our institution (Fig. 1), 1162 (71.5%) with bioprosthetic valves and 463 with mechanical valves. These included 126 (7.8%) who had second or subsequent operations for failed bioprostheses. Two hundred and sixty two (13.9% of all aortic valve replacement patients) who were at extremely high risk for surgery underwent TAVI [22], including 12 who had a valve-in-valve (ViV) procedure (Fig. 1, Table 1).

These patients were symptomatic with dyspnœa (NYHA class 3.1 ± 0.6). Six of them had had 2 or more previous thoracotomies, and 6 had significant coronary artery disease including 4 who had undergone coronary bypass grafting. Their mean age was 78 ± 7 years, body mass index 27 ± 5 kg/m², left ventricular ejection fraction 58 ± 11% and Society of Thoracic Surgeons mortality score 6.7 ± 3.7. The interval since their most recent valve surgery was 10 ± 4 (range 3–16) years. Six of the prosthetic aortic valves were severely stenotic (valve area ≤ 0.8 cm²), and 6 had significant (grade 3–4) regurgitation (Table 1).

All the patients had undergone assessment of their symptomatic status, comorbidities and the mode of prosthetic valve failure. Imaging included transthoracic (TTE) and transœsophageal echocardiography (TOE), multi-detector computed tomography (MDCT), selective catheterisation and angiography to define prosthetic valve morphology, hæmodynamics, vascular access and coronary artery disease.

We obtained the design characteristics of the surgically implanted bioprosthetic valves from the operation reports and information supplied by the companies. The valve dimensions were confirmed with echocardiographic and MDCT measurements.

Our multi-disciplinary team which includes cardiologists, cardiothoracic surgeons with extensive experience in valve replacement and anæsthetists evaluated them for further surgery: all were considered to be at unduly high risk. They gave informed consent for the procedure and data collection.

The procedures were undertaken in a hybrid catheterization laboratory. Valve deployment was achieved through femoral, subclavian, transaortic or transapical routes. The techniques of vascular access and approach to the aortic valve were similar to those for TAVI in native valves [23]. TOE guidance was employed in all but #1, #2 and #5 and temporary pacing in all. Rapid (burst) pacing was used at the time of valve deployment in 6 patients (#3, #4, #6, #9, #10 and #12) and balloon dilatation of stenotic bioprostheses in 4 (#2, #3, #4 and #5) and to reduce paravalvar regurgitation of a CoreValve® (#12).

Valves were successfully implanted in all the patients (Table 1). The resultant mean valve gradients were 15 ± 8 mm Hg (median 12, range 5 → 32 mm Hg) and the indexed effective orifice areas 0.93 ± 0.22 cm².m² (median 0.87, range 0.65 → 1.30 cm².m²). The volume of contrast used was 158 ± 182 ml and the dose-area product 17458 ± 16971 Gy*cm².

Following pre-dilatation of the bioprosthetic valve patient #4 developed torrential aortic regurgitation and asystolic cardiac arrest which were managed with cardiopulmonary resuscitation and immediate prosthetic valve implantation.

We inserted a CoreValve® within a Toronto™ stentless bioprosthesis in patient #7. Immediately following its release it became displaced towards the coronary ostia. A gooseneck snare was used to reposition it in the ascending aorta, and a second CoreValve® was deployed with a good hæmodynamic result (Fig. 2).

Three patients developed left bundle branch block following CoreValve® implantation, 2 others required permanent pacemaker implantation and patient #3 with stage 3 chronic kidney disease needed temporary hæmodialysis. There were no neurological events, major vascular complications or deaths in hospital.

All patients were discharged from hospital in NYHA class I or II functional status. None had suffered a neurological event, major vascular complication or myocardial infarction, and had been readmitted or required further intervention (Table 1).

Four patients (#3, #4, #10, #12) had prosthesis–patient mismatch (PPM) [24], [25], [26], [27] with indexed effective orifice area ≤ 0.80 cm².m² and 2 (#3, #11) trans-valvular peak velocity ≥ 4 m.s^− 1 and mean gradient ≥ 22 mm Hg. Three (#1, #5 and #10) had aortic regurgitation ≤ 2/4 and one (#9) valvular and para-valvular leaks ≥ grade 3/4.

Patient #6 had a routine surveillance trans-thoracic echocardiogram 9 months after TAVI which disclosed type A dissection of the aortic root. It is unclear whether this was related to aortic cannulation or to disease of her aorta, which had been noted to be dilated (50 mm) at the time of her initial assessment [28]. She survived following surgical replacement of her aortic root.

Two patients (#1, #2) died after 54 and 44 months: median survival for the 10 remaining has exceeded 19 months.