Delayed Mitral Prosthesis Dehiscence Following Transcatheter Aortic Valve Replacement

Abstract

Background

The presence of preexisting mitral prosthesis in transcatheter aortic valve replacement (TAVR) raises concerns about potential valve interaction; however, this issue remains insufficiently explored.

Case Summary

A 79-year-old man presented with hemolysis 8 months post-TAVR. He had a history of mitral valve replacement 16 years prior. Further investigation revealed severe paravalvular leakage of the mitral valve, which was attributed to the dehiscence of the mitral prosthesis. A retrospective review of pre-TAVR computed tomography images identified a subtle, previously undetected defect in the mitral–aortic intervalvular fibrosa, likely masked by neointima.

Discussion

We hypothesize that chronic mechanical stress between the TAVR device and the mitral–aortic intervalvular fibrosa causes progressive damage, resulting in delayed mitral valve dehiscence and hemolysis.

Take-Home Message

Careful preprocedural evaluation and prolonged postprocedural surveillance are essential for patients with preexisting mitral prostheses after TAVR due to the potential for delayed long-term complications.

Graphical Abstract

History of Presentation

A 79-year-old man presented to the emergency department with a 1-month history of progressively worsening shortness of breath and recent dark brown urine. Initial assessment revealed hemodynamic stability; however, laboratory findings showed elevated total bilirubin (10.9 mg/dL), decreased hemoglobin (7.9 g/dL), significantly elevated lactate dehydrogenase (3523 IU/L), and increased reticulocyte count (8.6%), suggestive of hemolysis. The patient was admitted for further evaluation.

Take-Home Messages

• •This case highlights the need for heightened vigilance regarding delayed mechanical complications, such as mitral valve dehiscence, after TAVR in patients with preexisting mitral prostheses.
• •Continuous and thorough follow-up, supported by careful preprocedural imaging analysis, is essential for identifying and addressing potential long-term issues that may not be immediately evident.

The patient's medical history included mitral valve replacement (MVR) with a 29-mm Sorin mechanical valve (Sorin Biomedica, S.p.A.) and de Vega tricuspid annuloplasty (TAP) 16 years prior. Additionally, 8 months before the current admission, the patient underwent transcatheter aortic valve replacement (TAVR) for worsening aortic stenosis. Preprocedural computed tomography (CT), performed to assess aortic valve anatomy and suitability for TAVR, revealed an aorto-mitral distance of 5.8 mm, and an annulus area of 522 mm², a short diameter of 22.6 mm, and a long diameter of 30.0 mm (Figures 1A and 1B). 3mensio Medical Imaging BV revealed a valve calcium volume of 629 at 450 Hounsfield units (cutoff: 500) (Figure 1C). According to our center's TAVR protocol, this patient would typically be considered for 5% to 10% oversizing with preballoon valvuloplasty. However, due to the large calcium deposit in the left ventricular outflow tract, and associated risk of rupture, a decision was made to implant a 26-mm Sapien 3 Ultra (Edwards Lifesciences) balloon-expandable (BE) valve at nominal volume (−0.6% downsizing). Preballoon valvuloplasty was performed using a 20-mm Z-MED balloon (B. Braun Interventional Systems Inc), selected to avoid exceeding the annulus short diameter. Postballoon dilatation was not performed (Video 1). Transthoracic echocardiography (TTE) conducted the day after TAVR showed that both the prosthetic aortic and mitral valves functioned well. After antibiotic treatment for a urinary tract infection before TAVR, the patient was discharged on the seventh day after the procedure without significant complications. Follow-up TTE performed 1-month post-TAVR confirmed well-functioning prosthetic aortic and mitral valves without abnormalities.

Figure 1.

Pretranscatheter Aortic Valve Replacement Computed Tomography

Pretranscatheter aortic valve replacement computed tomography measuring aorto-mitral distance (A), short and long annular diameter (B), and valve calcium volume (C).

Differential Diagnosis

Hemolysis caused by paravalvular leakage (PVL) was initially suspected based on the patient's medical history.

Investigations

TTE and transesophageal echocardiography were performed to assess the patient's cardiac valves, particularly the prosthetic mitral and aortic valves. Additionally, CT was performed, and prior imaging studies were reevaluated.

Echocardiography revealed a 3-mm gap with severe PVL at the anteromedial aspect of the mitral valve, with no signs of endocarditis (Figures 2A to 2D, Video 2). Moderate eccentric tricuspid regurgitation was observed, whereas no abnormalities were found in the aortic prosthetic valve.

Figure 2.

Echocardiography Demonstrating the Paravavlular Leakage

Echography demonstrating a gap (arrow) at the anteromedial aspect of the mitral valve and associated regurgitant leakage flow. Two-dimensional (A and B) and 3-dimensional (C and D) views of mitral prosthesis.

Management

To address the mitral valve dehiscence, a redo MVR with TAP was performed under general anesthesia with double-lumen intubation. A right mini-thoracotomy approach was selected because of the previous sternotomy. Cardiopulmonary bypass was established via arterial cannulation of the femoral artery and bicaval venous cannulation of the femoral and internal jugular veins. Severe peri-aortic adhesions prevented aortic cross clamping. Therefore, the procedure was performed under fibrillatory arrest. Intraoperative findings, consistent with preoperative echocardiography, confirmed dehiscence of the mitral prosthesis at the 12 to 3 o'clock position (Figure 3). The previous prosthesis was removed, and MVR was completed using a 27-mm Mosaic tissue valve (Medtronic), along with TAP using the Clover technique. Fibrillation and cardiopulmonary bypass times were 49 and 105 minutes, respectively.

Figure 3.

Dehiscence of the Mitral Valve Prosthesis

Intraoperative view confirming dehiscence of the mitral valve prosthesis at the 12 to 3 o'clock position (arrow).

Outcome and Follow-Up

The patient was transferred from the intensive care unit to the general ward on the third postoperative day. A postoperative TTE on the seventh day revealed mild PVL of the mitral valve, with no evidence of hemolysis. The patient was discharged on the twelfth postoperative day without any significant complications.

Discussion

TAVR is a well-established alternative to surgical aortic valve replacement in patients with severe aortic stenosis who are at intermediate to high surgical risk.¹ Although prior valve replacement significantly increases the surgical risk and may make patients suitable candidates for TAVR, these individuals are often excluded from randomized controlled trials and registries.² In addition, due to the anatomic proximity of the aortic and mitral valves, theoretical concerns exists regarding valve interactions when TAVR is performed in the presence of an existing mitral prosthesis.

Recent literature suggests that TAVR can be safely performed even in patients with an existing mitral prosthesis, provided that it is carefully selected.3, 4, 5, 6 However, serious complications have been reported. Although complications, such as PVL or valve embolization, are generally associated with the TAVR prosthesis due to the presence of a firmly seated mitral prosthesis, some cases involve complications directly affecting the mitral prosthesis itself. Squiers et al⁷ described a case of mitral prosthesis impingement after the reballooning of a TAVR prosthesis to address moderate PVL. Previously, Acar et al⁸ documented a case of abnormal mitral prosthesis function immediately after TAVR deployment, which led to hemodynamic instability and resulted in mortality within 1 hour of the procedure.

However, although previously reported complications occurred during or shortly after the procedure, we think this case is the first report of a direct mechanical long-term complication resulting from valve interactions. In our case, neither the pre- nor post-TAVR echocardiograms showed findings suggestive of PVL, and the pre-TAVR CT scan deemed the patient suitable for TAVR. Considering the absence of clinical symptoms and normal echocardiographic results, identifying any underlying defects in earlier CT images before the onset of hemolysis proves challenging. However, following the onset of symptoms, a closer retrospective review of the previous CT scan revealed a subtle preexisting defect at the mitral–aortic intervalvular fibrosa, which had likely been covered by neointima, thus preventing immediate detection and clinical significance (Figure 4A and 4B). Given the patient's symptoms emerged >6 months postprocedure, we hypothesized that prolonged mechanical interaction between the TAVR device and the mitral–aortic intervalvular fibrosa gradually induced stress, ultimately resulting in delayed mitral prosthesis dehiscence and subsequent hemolysis.

Figure 4.

CT Findings of the Aortomitral Continuity

Comparison of pretranscatheter aortic valve replacement (TAVR) computed tomography (CT) (A) and CT from the index admission (B). The arrow indicates the site of valve dehiscence. It is suspected that a subtle defect was obscured by neointima on the pre-TAVR CT.

An important anatomic factor to consider is the aorto-mitral distance. In a study by Amat-Santos et al,⁴ TAVR device embolization occurred in 6 of 91 patients with prior mitral prostheses, all of whom had an aorto-mitral distance <7 mm. Although embolization differs mechanistically from our case, both may reflect adverse outcomes of valve-valve interaction. In our case, the distance measured in pre-TAVR CT was 5.8 mm, and although the procedure itself was uncomplicated, this may have contributed to the delayed structural failure observed.

Valve selection is another unresolved issue in this context. Although direct comparative studies on the impact of BE and self-expandable (SE) TAVR valves on the mitral valve are limited, SE valves exert lower radial force at deployment but gradually expand beyond their nominal diameter over time, whereas BE valves achieve a fixed diameter on deployment. Therefore, once a BE TAVR has healed in, the pressure on the surrounding structures stabilizes, whereas SE TAVRs continue to exert force against adjacent tissue.⁹ Therefore, the use of an SE valve in this case may have only delayed the interaction with the mitral prosthesis, and a similar outcome would have occurred regardless of valve type.

This case underscores the need for vigilance regarding the potential for delayed complications, such as mitral valve dehiscence, after TAVR in patients with preexisting prosthetic mitral valves. Although the ideal valve type in this setting remains uncertain, minimizing overlap and avoiding direct contact between prostheses may be more critical than the device platform itself. Further research is needed to define clearer guidelines for patient selection, including parameters such as aorto-mitral distance and angle, procedural planning (eg, implantation depth), and device-specific considerations. Until more robust data are available, TAVR in patients with prior mitral prostheses should be approached with heightened caution. Prospective studies with larger patient cohorts and longer follow-up periods are essential to confirm our findings and guide future clinical practice. Close and prolonged postprocedural surveillance is strongly recommended for patients with similar risk profiles to monitor for delayed complications that may arise from chronic prosthetic interaction.

Conclusions

This case highlights the importance of recognizing that previously undetected structural vulnerabilities, including concealed dehiscence, may become clinically significant in patients with an in situ prosthetic mitral valve after TAVR. Careful preprocedural imaging and long-term monitoring are essential in such cases.

Funding Support and Author Disclosures

The authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Appendix

Video 1

The transcatheter aortic valve replacement procedure

Fluoroscopic imaging demonstrating the transcatheter aortic valve replacement procedure, from preballooning of the aortic valve to deployment of the valve prosthesis and the final result.

Video 2

Echocardiographic findings of the paravalvular leakage

Echography demonstrating a gap at the anteromedial aspect of the mitral valve and associated regurgitant leakage flow.

Case Video Summary

Totally endoscopic redo-mitral valve replacement and tricuspid valvuloplasty