Author + information
- Azeem Latib, MD†,‡,⁎ (, )
- David Messika-Zeitoun, MD, PhD§,
- Francesco Maisano, MD∥,
- Dominique Himbert, MD§,
- Eustachio Agricola, MD‡,
- Eric Brochet, MD§,
- Ottavio Alfieri, MD∥,
- Antonio Colombo, MD†,‡ and
- Alec Vahanian, MD‡
- ↵⁎EMO-GVM Centro Cuore Columbus, Via Buonarroti 48, 20145 Milan, Italy
To the Editor: Transcatheter aortic valve implantation (TAVI) continues to penetrate into daily practice supported by increasing data demonstrating efficacy and safety on the short- to medium-term but the long-term durability of transcatheter heart valves (THV) remains unknown. In this report, we present a case series of reversible Edwards Sapien XT dysfunction (Edwards Lifesciences Inc., Irvine, California), which could have been falsely interpreted as early THV failure.
The first case was an 83-year-old man referred for severe symptomatic aortic stenosis (AS) (mean gradient 75 mm Hg, aortic valve area 0.6 cm2). He had a history of stroke, coronary artery bypass grafting, and moderate renal insufficiency. Ejection fraction was normal. Conventional surgery was considered high risk (EuroSCORE [European System for Cardiac Operative Risk Evaluation] 30%, Society of Thoracic Surgeons [STS] score 5%) and a transfemoral TAVI was successfully performed (Sapien XT 26 mm). Discharge transthoracic echocardiography (TTE) showed a normally functioning prosthesis (mean gradient 11 mm Hg) with no paravalvular leakage. The patient was discharged on aspirin and clopidogrel. At 1 month, he was asymptomatic (New York Heart Association [NYHA] functional class I) and TTE was unchanged. He came back 6 months later for a scheduled follow-up, still on dual antiplatelet therapy, and surprisingly the mean gradient had dramatically increased to 47 mm Hg. He complained of dyspnea since several weeks (NYHA functional class II) and was admitted 2 weeks later for further investigation. An even higher mean gradient (68 mm Hg) was observed during TTE and transesophageal echocardiography showed mildly thickened leaflets with a severe restrictive motion (Fig. 1). Despite the absence of direct visualization of thrombus and due to the rapid gradient increase, we added a vitamin K antagonist. Twenty-five days later, the mean gradient decreased to 37 mm Hg and almost normalized in 2 months (Fig. 1). Eight months later, the patient remained on aspirin, clopidogrel, and vitamin K antagonist, and was asymptomatic (mean gradient 11 mm Hg).
Our heart team decided to perform a TAVI in an 81-year-old man with hypertension, renal insufficiency, and severe AS (mean gradient 82 mm Hg, valve area 0.9 cm2) predominantly because of frailty (EuroSCORE 11%, STS score 3.9%). Discharge TTE showed a normally functioning prosthesis (mean gradient 8 mm Hg), preserved left ventricular function, and a trivial paravalvular leak. He was discharged on aspirin and clopidogrel and stopped the clopidogrel after 3 months. He was seen again 3 and 9 months after the procedure, and was asymptomatic with normal transaortic gradients. He presented 15 months later with worsening exertional dyspnea (NYHA functional class II/III). TTE showed a 45 mm Hg mean gradient without worsening of the paravalvular leak. Based on the previous case, we opted for an attempt of vitamin K antagonist therapy in addition to aspirin, which resulted in rapid resolution of symptoms and gradient (19 mm Hg).
An 83-year-old male with hypertension, dyslipidemia, previous stroke, and severe symptomatic AS (mean gradient 40 mm Hg, valve area 0.8 cm2) considered at high risk for conventional surgery (EuroSCORE 20%, STS score 17%) underwent a 26 mm Sapien XT transfemoral implantation. Discharge TTE showed a 9 mm Hg mean gradient, normal ventricular function, and mild posterior paravalvular leak. He remained well (NYHA functional class I) with mean transaortic gradients below 20 mm Hg at 1, 6, and 12 months. At 2 years, he presented with shortness of breath (NYHA functional class III) and elevated aortic gradients (mean gradient 37 mm Hg) without worsening of the paravalvular leak. Vitamin K antagonists were also added and resulted in gradient normalization (13 mm Hg) and marked symptomatic improvement.
In this report, we present the first series of symptomatic bioprosthetic dysfunction presenting within 1 to 2 years after transfemoral implantation of a Sapien XT valve. The THV dysfunction was rapidly reversible with anticoagulation, which resulted in normalization of transaortic gradients within 2 months, suggesting probable valve thrombosis. Although there have been reports of reversible dysfunction of surgical bioprostheses with anticoagulation, the optimal antithrombotic therapy after surgical or transcatheter bioprosthesis implantation is still uncertain (1).
The progressive gradient increase of a prosthesis that was functioning normally after implantation may have first suggested an early THV structural degeneration and we even considered valve-in-valve implantation. However, the timing, clinical onset, and absence of valve calcification were unusual for early bioprosthesis degeneration even if there are only limited case reports of THV failure in the literature and none with the Edwards Sapien valve (2,3). We first performed an attempt of anticoagulation using vitamin K antagonist before proceeding to a second procedure. Initial therapy with heparin may be considered based on the thrombosis/hemorrhagic risk balance. It is worth noting that we could not identify any clinical trigger, sign of myocardial ischemia, or changes in ejection fraction in our 3 patients. In the current THV literature, we only found 1 report of thrombotic THV dysfunction 4 months after a transapical 26 mm Sapien valve implantation, which presented acutely with non–ST-segment elevation myocardial infarction, heart failure, and gradient increase (4). In this case, a thrombus was visible during echocardiography and the patient responded well to anticoagulant therapy within 1 month. Interestingly, no thrombus was visible on the aortic valve in any of our 3 cases during TTE and transesophageal echocardiography but all responded well to the anticoagulation regimen. Although this strategy was very successful, the follow-up in 2 patients is still limited and a longer follow-up is desirable. We thus recommended lifelong anticoagulation.
This case series has important clinical implications as it identifies a potentially reversible cause of THV dysfunction. In patients presenting with prosthetic valve stenosis within the first few years after the procedure, an attempt of anticoagulation should be performed before considering a second procedure or valve-in-valve implantation, particularly as the response to therapy appears to occur fairly rapidly (within 1 to 2 months). It also highlights that further research is needed on the optimal antithrombotic strategy after TAVI.
Please note: Dr. Latib is a member of the Medtronic advisory board, and has consulted to Edwards Lifesciences. Dr. Messika-Zeitoun is a consultant to, and has received lecture fees from, Edwards Lifesciences, Valtech, and Symetis. Dr. Maisano is a consultant to Valtechcardio, Medtronic, and St. Jude Medical. Dr. Himbert is a proctor for Edwards Lifesciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Latib and Messika-Zeitoun contributed equally to this manuscript and are joint first authors.
- American College of Cardiology Foundation
- Vahanian A.,
- Alfieri O.,
- Andreotti F.,
- et al.
- Kefer J.,
- Astarci P.,
- Renkin J.,
- et al.