Valve-in-valve (ViV) trans-catheter aortic valve implantation is a minimally invasive treatment option whereby an aortic bioprosthetic valve is percutaneously implanted inside a degenerated surgical tissue valve. ViV TAVI accounts for 3–4% of TAVI and is an alternative to redo surgical aortic valve replacement (SAVR), particularly in patients at higher surgical risk.1,2 However, patients with a small (≤ 21 mm in diameter) surgical bio-prosthetic valve undergoing ViV TAVI are at increased risk of having a persistent residual transvalvular gradient and may have a reduced long-term survival.3 We present the case of a patient in whom ViV TAVI resulted in failure of clinical improvement and a high residual transvalvular gradient requiring redo SAVR, and discuss options for prevention of this complication.
A 78-year-old woman (height 173cm, weight 67kg, body mass surface (BMS) 1.79m²) underwent SAVR in 2004 using a size 21mm Perimount pericardial tissue valve (model 2900) for severe aortic stenosis. Preoperatively the mean aortic valve gradient was 55mmHg, effective valve orifice area (EOA) 0.5cm², indexed EOA 0.28cm2/m2. A postoperative transthoracic echocardiogram showed a mean gradient across the 21mm aortic valve prosthesis of 21mmHg (Table 1).
Table 1: Echocardiography parameters before and after surgical aortic valve replacement, valve-in-valve trans-catheter aortic valve implantation and redo surgical aortic valve replacement with enlargement of the aortic root.
Abbreviations: AV, aortic valve; EOA, effective valve orifice area; IEOA, indexed effective valve orifice area; SAVR, surgical aortic valve replacement; ViV TAVI, valve-in-valve transcatheter aortic valve implantation.
In 2015 she developed increasing shortness of breath, light-headedness and exercise intolerance. Within six months her walking distance had decreased from 500 to 50 metres. She underwent coronary angiography that demonstrated no haemodynamically significant coronary disease. A transthoracic echocardiogram on 12 February 2016 showed mild concentric left ventricular hypertrophy, normal LV systolic function (estimated LV ejection fraction of 60–65%), an aortic valve mean gradient of 30mmHg and an EOA of 0.8cm², an indexed EOA of 0.45cm²/m², consistent with significant aortic stenosis. Significant thickening/calcification and restriction of the valve leaflets was noted from 2D imaging. She was scheduled for redo SAVR. However, in February 2016 she developed acute gangrenous appendicitis with appendiceal perforation and peritonitis, for which she underwent a laparotomy. After hospital discharge, she experienced dizziness and unsteadiness on her feet, and had difficulty mobilising around her home. Because of patient’s severe deconditioning and a EuroSCORE II estimated mortality of 6.93%, she was considered at intermediate-to-high risk for redo surgery, and was offered a ViV TAVI.
In April 2016 she underwent transfemoral ViV TAVI with a manufacturer-recommended 23mm Edwards Sapien 3 bioprosthesis. A TTE performed one day following this procedure demonstrated a persisting mean gradient across the aortic valve of 28mmHg. The Doppler Dimensionless Index, the ratio of left ventricular outflow tract velocity time integral (VTI) to aortic VTI, was normal at 0.46 and the study was therefore considered to be consistent with normal bioprosthetic valve function, recognising that there may be technical difficulties with measurement of LV stroke volume.
Postoperatively she continued to have symptoms of breathlessness and was re-evaluated. A TTE on 22 April 2016 showed normal LV size and systolic function and mild concentric LVH. The leaflets of the 23mm percutaneous AVR appeared unremarkable and no aortic regurgitation was detected, but the mean aortic valve gradient was increased at 35mmHg. An EOA was calculated as 1.6cm2 (with the previous reservation that the LV stroke volume might be incorrect), indexed EOA of 0.9cm2/m2 (Table 1). After discussion with the patient and her family, redo SAVR was recommended, recognising that her general condition was better than it had been immediately after her episode of peritonitis.
On 16 June 2017 she underwent a redo sternotomy, cardiopulmonary bypass, and removal of the original Perimount bovine tissue valve and ViV bio-prosthesis as one unit. The findings were of redundancy and crowding of the leaflets inside a constrained TAVI valve frame (Figure 1). The 23mm Sapien 3 bioprosthesis was well opposed to the 21mm Perimount valve, with no inter-valvar space, and was free ofpannus formation. Once the surgical aortic valve sewing cuff sutures were removed and the double valve bioprosthetic complex removed, the aortic root was divided down the non-coronary sinus and across the annulus and widened with a pericardial patch using the Nicks technique. A size 23mm Perimount Magna Ease aortic valve was then sutured into position.
Figure 1: Redundancy and crowding of the leaflets inside a constrained TAVI valve frame.
The early postoperative TTE showed normal LV systolic function with a mean aortic valve gradient of 16mmHg. A calculated EOA was 2.7cm², indexed EOA 1.5cm2/m2. Her postoperative course was uncomplicated. At four months cardiology follow-up, she was not limited by breathlessness in her everyday life activities. During a telephone call follow-up in July 2018, the patient indicated a considerable improvement in her excise tolerance since the last surgery.
Aortic stenosis has a prevalence of 12% in the elderly population.4,5 American Heart Association and American College of Cardiology guidelines recommend SAVR with a biological bioprosthesis from the age of 70 years,6 and there is increasing use of tissue valves in patients younger than 70 years.7 This change of clinical practice is because of increased long-term durability of tissue valves, the patient’s personal, occupational and lifestyle preferences, the bleeding risk of anticoagulation with a mechanical valve in the elderly, and the development of a percutaneous ViV option for structural valve deterioration.8
The outcomes of ViV TAVI have improved over time with increased operator experience and improved TAVI technology. However, survival is worse in those in whom the TAVI valve is implanted in a small (≤ 21mm in diameter) surgical bioprosthesis and in those who have a high post-TAVI residual trans-valvar gradient.3 One recently described option for treating such patients is to deliberately fracture the surgical bioprosthetic valve ring by high-pressure balloon inflation.9 This increases the surgical valve EOA and allows fuller expansion of the implanted TAVI valve that, in turn, decreases the residual trans-valvular gradient. Although this does not appear to increase the risk of complications, in particular annular rupture, only a small number of patients have been reported to date. Valve fracturing does not completely overcome the problem of a small aortic annulus.
In the case described, ViV TAVI resulted in a persistently high aortic valve gradient associated with symptoms because the percutaneous device was not adequately expanded within the previously implanted small (21mm) bioprosthesis. We obtained better haemodynamics by an open surgical procedure, including enlargement of the aortic annulus and implantation of a larger aortic valve bioprosthesis.
While echocardiographic assessment of prosthetic valve stenosis is beyond the scope of this paper, the study does also highlight some of the difficulties that may be experienced.
In younger patients with aortic stenosis and a small annulus undergoing SAVR, consideration should be given to enlarging the aortic annulus to allow implantation of a larger valve bioprosthesis and a better future ViV TAVI option.