Recurrent bioprosthetic mitral valve thrombosis treated with anticoagulation

Abstract

Bioprosthetic valve thrombosis (BPVT) is more common than previously thought and likely underreported. BPVT can be accurately diagnosed with cardiac imaging and treated successfully with anticoagulation, thus preventing reoperation. We hereby report a case of recurrent BPVT in the mitral position successfully treated with anticoagulation along with review of literature.

1 |. INTRODUCTION

The mechanism of bioprosthetic valve (BPV) dysfunction is predominantly due to structural deterioration. Bioprosthetic valve thrombosis (BPVT) has traditionally been considered a rare event. However, recent data suggest that BPVT is not uncommon and in fact an increasingly recognized entity. The true incidence of BPVT may be underestimated as most cases remain underdiagnosed. The diagnosis can be challenging due to low general awareness of the condition. A high degree of clinical suspicion and formal echocardiographic criteria can accurately diagnose BPVT. Early diagnosis remains crucial as it can be successfully treated with anticoagulation.

2 |. CASE REPORT

A 72-year-old man presented to the emergency room with one-week history of progressive exertional dyspnea and lower extremity edema. His past medical history included mitral valve replacement (MVR) eight years ago [27-mm Carpentier-Edward (CE) Magna] for severe mitral regurgitation (MR) due to myxomatous mitral valve disease, paroxysmal atrial fibrillation (pAF), and nonischemic cardiomyopathy with LVEF 30%-35%, s/p dual-chamber implantable cardioverter-defibrillator (ICD). Two years prior to the current admission, he presented with multiple ICD shocks secondary to atrial tachycardia. Workup revealed multiple echo densities on the mitral bioprosthetic valve with BPV stenosis and a transmitral mean gradient of 7 mm Hg at heart rate of 61 beats per min (bpm). Pressure half-time was 259 ms After a detailed negative workup for infective endocarditis, a possible diagnosis of nonbacterial thrombotic endocarditis (NBTE) (Figure 1, Movies S1 and S2) was considered. He was treated with unfractionated heparin (UFH) for two weeks. However, due to lack of improvement, he underwent redo MVR with a similar valve (27-mm CE Magna) and was discharged on warfarin with international normalized ratio (INR) goal of 2-3. One month prior to the current presentation, warfarin was interrupted perioperatively for total right hip replacement surgery.

FIGURE 1.

Echocardiogram demonstrating mitral BPV echo densities (previous admission). Mitral BPV thickening and echo densities seen on (A) TTE (red arrow); (B) TEE (blue arrow); (C) 3DTEE; and (D) TEE transmitral gradients

At the time of his current admission, physical examination revealed a chronically ill appearing man with blood pressure 121/94 mm Hg, irregular rhythm with heart rate of 84 bpm, normal S1 and S2 without murmurs, jugular venous pressure (12 cm of water), bibasilar crackles, and bilateral 4 + pitting pedal edema. He was started on intravenous (IV) diuretics and admitted to the cardiology inpatient service. Laboratory evaluation included normal blood counts and metabolic panel, elevated B-type natriuretic peptide (2751 pg/mL), and INR (2.3). Transthoracic echocardiogram (TTE) showed LVEF 30%-35%, mitral BPV leaflet thickening with mean gradient of 10 mm Hg at heart rate of 65 bpm, and multiple echo densities suspicious for vegetation or thrombus (Figure 2A,B, Movie S3). Pressure half-time was 212 ms Transesophageal echocardiogram (TEE) revealed severe BPV leaflet thickening with restricted motion, a large echo density encompassing both leaflets with a mobile component measuring 1.4 × 0.4 cm and a mean gradient of 9.2 mm Hg at heart rate of 67 bpm (Figure 2C,D, 3A,B Movies S4, S5 and S6). A detailed laboratory workup including infectious, rheumatologic, immunologic, and allergic (for bovine pericardial valve) tests was unremarkable. Given echocardiographic findings of BPV stenosis and echo densities and an extensive negative workup for other etiologies, he was treated empirically for BPVT with UFH and eventually transitioned to warfarin with an increased INR goal of 2.5-3.5. After three months of uninterrupted anticoagulation, a repeat TEE showed complete resolution of BPV thickening and echo density with significant reduction in the mean transvalvular gradient to 4 mm Hg at heart rate of 65 bpm (Figure 4, and Movies S7 and S8), indicating that the cause of the patient’s initial presentation was likely BPVT. The patient is currently asymptomatic and is followed clinically and with regular surveillance TTE.

FIGURE 2.

Echocardiogram demonstrating mitral BPV thrombosis (current admission). Mitral BPV thickening and echo densities seen on (A) TTE (red arrow); (B) TTE transmitral gradients; (C) TEE (blue arrow); (D) TEE transmitral gradients

FIGURE 3.

Echocardiogram demonstrating mitral BPV thrombosis (current admission). Mitral BPV thickening and echo densities seen on (A & B) 3DTEE

FIGURE 4.

Echocardiogram demonstrating complete resolution of mitral BPV thrombosis after anticoagulation. Resolution of mitral BPV thickening and thrombosis after anticoagulation (A) TEE; (B) 3DTEE; and (C) TEE transmitral gradients. BPV = bioprosthetic valve; LA = left atrium; LV = left ventricle; RA = right atrium; RV = right ventricle; TEE = transesophageal echocardiogram; TTE = transthoracic echocardiogram

3 |. DISCUSSION

The true incidence of BPVT remains uncertain; estimates range from less than 0.5% to over 6% of BPV recipients depending on the mode of diagnosis (pathology vs imaging) and length of follow-up.¹ BPV dysfunction due to thrombosis is commonly mistaken with “valve degeneration,” leading to underreporting. Though rare, BPVT is a clinically important entity and can occur in all four valve locations. BPVT is distinguished from BPV degeneration based on various echocardiographic criteria. BPVT presents with increased cusp thickness, reduced cusp mobility, and less severe regurgitation, whereas BPV degeneration is associated with calcified cusps, reduced mobility, and significant regurgitation.

The mechanisms underlying BPVT are incompletely understood, but involve blood flow perturbation resulting in high wall shear stress and increased blood stasis, plasma protein adsorption leading to activation of hemostatic factors, and patient-related factors including hypercoagulable states, anemia, renal insufficiency, obesity, diabetes mellitus, smoking, low cardiac output, and periprocedural trauma. Suboptimal anticoagulation in patients taking oral anticoagulation (OAC) and pAF are additional risk factors for BPVT.¹

BPVT presentation may vary from incidental imaging findings without symptoms to syncope, dyspnea, heart failure, or cardiogenic shock from valve obstruction. Diagnosis is typically made by echocardiography. A model consisting of three echocardiographic predictors increased the sensitivity and specificity of diagnosing BPVT in the setting of concordant clinical features: (a) 50% increase in transvalvular gradients compared with baseline within five years of surgery, in the absence of a high cardiac output state; (b) increased cusp thickness (>2 mm), especially on the downstream aspect of the valve; and (c) abnormal cusp mobility.² Although TTE is helpful, performing TEE is strongly recommended when clinical suspicion is high to facilitate expeditious diagnosis.

Although current 2017 ACC/AHA and ESC guidelines recommend oral anticoagulation for only the first three months following surgical BPV replacement in the absence of risk factors,^3,4 our case highlights the fact that risk of BPVT is not limited to the first three months after implantation. Several studies have reported that BPVT may occur late after initial implantation and should be suspected in the appropriate clinical scenario. Among 149 patients who underwent mitral BPV implantation at a single center, a retrospective review of TEEs identified nine patients (6%) with BPVT and median time from implantation to diagnosis was 12 months.⁵ Another study identified 46 cases (11.6%) of histologically proven BPVT among 397 patients who underwent BPV explantation. The median time to explantation was 24 months, with 15% of cases occurring more than 5 years after valve placement.²

Anticoagulation is the mainstay treatment for BPVT in hemodynamically stable patients. Early diagnosis of BPVT is crucial as most patients respond to anticoagulation and BPVT resolves completely, thus avoiding need for repeat valvular intervention. Several studies have reported successful resolution of BPVT when treated with VKA (vitamin K antagonist).^5–8 A prospective evaluation of warfarin showed that anticoagulation was effective in 83% of patients with suspected BPVT, and most responded within 3 months.⁷ A recent study by Petrescu et al reported long-term outcomes of anticoagulation in 83 patients treated with warfarin for suspected BPVT. Echocardiographic parameters normalized in 75% of patients within three months. However, warfarin-treated patients had significant higher rates of major bleeding compared with matched controls. Additionally, BPVT recurred in 23% of warfarin responders after a median of 23 months, and all but one patient with recurrent BPVT responded to anticoagulation. Thus, longer term or even indefinite anticoagulation with warfarin could be considered after an initial BPVT episode while balancing bleeding risks.⁸ In retrospect, we believe that our patient’s presentation of BPV stenosis two years prior to the current presentation was likely BPVT. At that time, an extended trial of anticoagulation may have obviated the need for redo MVR.

During the initial presentation two years prior to the current admission, the patient underwent extensive workup including serial negative blood cultures and nondiagnostic rheumatologic and allergic testing. A diagnosis of NBTE was subsequently considered, and the patient was started on anticoagulation with intravenous heparin. During surgery, the bioprosthetic mitral valve was noted to be degenerated with calcification of the leaflets and fusion of the commissure due to pannus overgrowth. Pathology of the explanted valve revealed a degenerated bioprosthetic mitral valve with calcification but without any vegetations.

NBTE is described typically in native valves, is difficult to diagnose, and requires a strong clinical suspicion. Typical clinical features include a triad of (a) disease process known to be associated with NBTE such as carcinomatosis with diffuse intravascular coagulation, connective tissue disease, or antiphospholipid syndrome; (b) the presence of a heart murmur; and (c) evidence of multiple systemic emboli. Echocardiographically, NBTE vegetations are small, often measuring a few millimeters. Pathologically, they are characterized by vegetations on the cardiac valves which consist of fibrin and platelet aggregates without inflammation or bacteria.⁹ Except for a single case report,¹⁰ to our knowledge, NBTE has never been described in prosthetic valves.

In contrast, although BPVT is more common during the first few months after valve implantation, it can present as a late manifestation up to 5 years, as noted in our patient. This diagnosis also necessitates a strong clinical suspicion, especially considering that patients are often asymptomatic and a gradual increase in transvalvular gradients in the absence of high-flow states can be the first manifestation of BPVT during routine surveillance. Echocardiographic features of BPVT include thickened, immobile cusps or mobile echo densities as noted in our patient. Intervention at this stage with anticoagulation could prevent further progression and resolve the thrombus. Typical symptoms include progressive exercise intolerance and heart failure, similar to other causes of prosthetic valve dysfunction. Definitive diagnosis is often made after valve explantation and pathological examination.

While current 2017 ACC/AHA guidelines do not recommend routine surveillance with TTE until after 10 years of bioprosthetic valve implantation in the absence of symptoms, our case highlights that early diagnosis and management of BPVT are critical. Thus, consideration should be given to regular TTE monitoring for the development of BPVT in high-risk cohorts. Determining the optimal frequency of TTE monitoring for BPV recipients requires further investigation.

4 |. CONCLUSIONS

Bioprosthetic valve thrombosis may occur late after valve implantation and should not be confused for valve degeneration. BPVT should be suspected in patients with clinical or echocardiographic evidence of BPV dysfunction, especially when presenting within five years after valve implantation. Elevated transvalvular gradients can be the first clue in diagnosing subclinical BPVT. Anticoagulation should be instituted in hemodynamically stable patients without contraindications before pursuing repeat valve replacement. Surgery should be reserved for nonresponders to anticoagulation or patients in whom hemodynamic status precludes further delay. Although the optimal duration of anticoagulation is not known, long-term anticoagulation should be considered. In the event of cessation of oral anticoagulation, antiplatelet therapy and frequent echocardiographic surveillance can be considered.

Supplementary Material

Movie S2

Movie S2. 3DTEE mid-esophageal four-chamber view showing mitral BPV thickening with mobile echo density.

Movie S1

Movie S1. TEE mid-esophageal four-chamber view demonstrating mitral BPV thickening with mobile echo densities.

Movie S4

Movie S4. TEE mid-esophageal four-chamber view showing mitral BPV thickening with mobile echo density.

Movie S3

Movie S3. TTE apical four-chamber view showing mitral BPV thickening with mobile echo densities.

Movie S5

Movie S5. 3DTEE mid-esophageal four-chamber view demonstrating mitral BPV thickening with mobile echo density.

Movie S6

Movie S6. 3DTEE mid-esophageal four-chamber view demonstrating mitral BPV thickening with mobile echo density.

Movie S8

Movie S8. 3DTEE mid-esophageal four-chamber view demonstrating complete resolution of mitral BPV thrombosis after anticoagulation.

Movie S7

Movie S7. TEE mid-esophageal four-chamber view demonstrating complete resolution of mitral BPV thrombosis after anticoagulation.

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study (case report).