An uncommon case of bioprosthetic mitral valve thrombosis with subsequent prosthesis dehiscence

Abstract

Bioprosthetic valve thrombosis (BVT) is uncommon. An 82-year-old gentleman presented following a cardioembolic stroke due to a mitral valve infective endocarditis. The patient underwent bioprosthetic mitral valve replacement. Following discharge, he re-presented with weakness due to a new left-sided stroke and right-sided occipital intracerebral haemorrhage. Both transthoracic and transoesophageal echocardiography revealed BVT on the anterior portion of the prosthesis. Following a multidisciplinary team discussion, it was felt that thrombolytics and anticoagulation would be detrimental. The patient continued to worsen and eventually succumbed to congestive cardiac failure. BVT is uncommon, and patients often present with signs and symptoms of heart failure, shock or embolism. There are no guidelines available on managing BVT in patients with concurrent intracerebral haemorrhage, highlighting the importance of multidisciplinary efforts and patient inclusivity in decision making.

Introduction

Bioprosthetic valve thrombosis (BVT) is uncommon and is defined as any thrombosis unrelated to infection, which occludes the path of blood flow, interferes with valve function, or is sufficiently large to warrant treatment. ¹ We report a rare case of BVT with subsequent prosthetic dehiscence leading to fatal congestive cardiac failure.

Case report

An 82-year-old gentleman presented to our hospital due to sudden onset of dysphasia, homonymous hemianopia and lethargy. Initial computed tomography (CT) imaging of the brain revealed a large left-sided parieto-occipital ischaemic stroke. Magnetic resonance imaging with angiography of the brain further confirmed this, with no features suggestive of amyloid angiopathy or vasculitis (Figure 1). Carotid ultrasound was unremarkable for stenosis. His pulse was of regular rhythm, and electrocardiogram (ECG) confirmed sinus rhythm, with no evidence of atrial fibrillation. Examination had revealed a loud pansystolic murmur, prompting transthoracic echocardiography (TTE) which demonstrated evidence of a large 2.0 cm×1.5 cm mitral valve mass on the anterior leaflet, suspicious of infective endocarditis, which was further supported by raised inflammatory markers and blood cultures which grew Streptococcus salivarius (Figure 1). Alongside intravenous antibiotics, the patient subsequently underwent mitral valve replacement using a Carpentier–Edwards PERIMOUNT Magna-Ease (Edwards Lifesciences, Irvine, CA) bioprosthetic valve. Pre-discharge TTE revealed a stable mitral valve bioprosthesis, with no evidence of thrombus or dehiscence. The patient had remained in sinus rhythm throughout his admission, both pre- and postoperatively.

Figure 1.

Imaging of the brain on initial admission, including (a) T1- and (b) T2-weighted magnetic resonance imaging (MRI) showing hyperintensity on the left parieto-occipital regions suggestive of infarction, and (c) MRI angiography which did not demonstrate any abnormalities suggestive of vasculitis. Transthoracic echocardiography in (d) apical three-chamber, (e) parasternal long-axis and (f) apical four-chamber views demonstrating a large 2.0 cm×1.5 cm homogenous mass on the anterior leaflet of the mitral valve, suggestive of a vegetation, causing (g) severe mitral regurgitation.

Unfortunately, he presented again within a week with new right-sided hemiparesis (Medical Research Council Scale 0/5). There were no new murmurs audible on precordial examination, and his vital signs were unremarkable, including a temperature of 36.7°C, blood pressure of 135/78 mmHg and pulse rate of 65 bpm. CT imaging of the brain revealed both a new left-sided temporal infarct and a right-sided occipital intracerebral haemorrhage (Figure 2). Again, ECG was of sinus rhythm. A repeat TTE was performed, revealing a BVT measuring 2.0 cm×1.8 cm on the anterior portion of the prosthesis, which was later confirmed on transoesophageal echocardiography (TOE) (Figure 3). Blood investigations performed at this juncture did not reveal any evidence of new infection, and surveillance blood cultures were negative (Table 1). Although initial TOE did not reveal any evidence of paravalvular leak or dehiscence, the patient had further deteriorated following bouts of acute pulmonary oedema, and serial TTEs later revealed evidence of paravalvular leak, coupled with rocking motion of the prothesis suggestive of dehiscence (Figure 4). Following a multidisciplinary team discussion, it was felt that thrombolytics and anticoagulation would be risky for the patient due to the intracerebral bleeding. Furthermore, the patient and family had declined any repeat surgical interventions in view of the high perioperative risk. The patient was eventually referred for palliative care.

Figure 2.

Computed tomography imaging of the brain on the second admission revealed hypodensity in the left temporal region suggestive of new infarction, as well as a new hyperintense lesion in the right occipital region suggestive of haemorrhage.

Figure 3.

Transthoracic echocardiography in (a) diastole and (b) systole demonstrating a 2.0 cm×1.8 cm homogenous mass located in the anterior portion of the bioprosthetic mitral valve, with no evidence of paravalvular or transvalvular regurgitation. A transoesophageal echocardiogram was also performed, revealing in (c) diastole and (d) systole the same homogenous mass, suspicious of a thrombus.

Figure 4.

Transthoracic echocardiography in (a) apical four-chamber and (b) apical three-chamber views demonstrating paravalvular regurgitation jets in systole, associated with rocking motion of the anterior portion of the bioprosthetic mitral valve, suggestive of valvular prosthesis dehiscence.

Table 1.

Blood investigations.

Test	Results	Test	Results
Haemoglobin (g/L)	11.5	Hepatitis B serology	Negative
Mean corpuscular volume (1015/L)	79	Hepatitis B serology	Negative
White cell count (109/L)	7.3	Human immunodeficiency virus serology	Negative
Platelet (109/L)	220	C-reactive protein (mg/L)	<0.5
Sodium (mmol/L)	135	Erythrocyte sedimentation rate (mm/h)	3
Potassium (mmol/L)	3.1	Troponin-T (<14 ng/L)	85
Urea (mmol/L)	10.5	Prothrombin time (s)	13.0
Creatinine (mmol/L)	112	International Normalized Ratio	1.0
NT-pro brain natriuretic peptide (pg/mL; <220)	245	Activated prothrombin time (s)	45.3

Discussion

BVT is uncommon, occurring in 6.0–12.7% of patients with bioprosthetic valves. ² Although the majority of cases remain subclinical and are only discovered incidentally on routine TTE, patients can present with symptoms and signs consistent with heart failure or cardiogenic shock due to mechanical obstruction. Patients can also present with signs suggestive of embolism, such as in our case, possibly due to delayed thrombus organisation. ³ In our patient, an important differential diagnosis would have been prosthetic valve endocarditis (PVE), especially in view of very recent surgical intervention. ⁴ However, the mobile mass was of soft echo density with an irregular shape and, in our opinion, fairly homogenous. Despite being on the atrial side (which is the site with higher pressure side of the prosthesis), the lack of clinical features and positive blood cultures to support a diagnosis of infection made PVE unlikely. A multi-modality imaging approach may have been helpful in differentiating PVE from BVT through the use of 18-fluorine-fluorodesoxyglucose positron emission tomography, multi detector-row computed tomography and three-dimensional echocardiography. ⁴

Although the exact pathophysiology remains elusive, proposed mechanism for BVT include abnormal haemostatic activation through artificial surface contact, perturbations in blood flow with high endothelial wall shear stress and patient risk factors favouring a hypercoagulable state. ³ The development and presence of thrombi further leads to valve degeneration when left untreated, and often manifest as dehiscence (rocking motion of the prosthesis) and paravalvular leaks on echocardiography, as demonstrated in our patient. ⁵ However, the role of prophylactic anticoagulation post-valvular replacement remains contentious, though some guidelines recommend anticoagulation use up to 90 days postoperatively. ³

Management of BVT includes using anticoagulation, thrombolytics, surgical replacement and transcatheter valve-in-valve therapy, with the majority of evidence having derived from studies on aortic valve replacement (both surgical and transcatheter). ³ At present, there are no guidelines available on managing BVT in patients with concurrent intracerebral haemorrhage, such as in the case of our patient, highlighting the importance of multidisciplinary efforts and patient inclusivity in decision making.