Prosthetic mitral valve thrombosis associated with new-onset hyperthyroidism due to Graves’ disease: a case report

Abstract

Introduction

Surgical valve replacement with biological prostheses is associated with a number of complications, including early thromboembolic events as well as endocarditis. Bioprosthetic mitral valve thrombosis has an incidence of 0.64% observed in a 2-year follow-up.¹ The aetiology and pathogenesis remain poorly understood even though a number of risk factors are identified.² However, late thrombosis with subsequent valve failure of bioprosthetic heart valves is a rare phenomenon with unclear pathogenesis.

We report a case of bioprosthetic mitral valve thrombosis occurring 6 years post-implantation linked to new-onset hyperthyroidism.

Summary figure

Open in new tabDownload slide

Case summary

A 59-year-old male Caucasian was referred by his cardiologist due to a significantly elevated transmitral gradient six years after surgival mitral valve replacement (MVR) for severe primary mitral regurgitation (MR) with a Carpentier-Edwards Perimount Magna (29 mm) bioprosthesis. An anticoagulation regimen with a vitamin K antagonist (VKA) was initiated³ for the first 3 months after implantation and thereafter changed to aspirin 100 mg/day due to known coronary artery disease.⁴ His medical history, moreover, includes coronary artery disease and bronchial asthma.

In February 2023, he presented with palpitations, fatigue, weight loss, and subfebrile temperatures. Clinical examination revealed a temperature of 37.9°C, blood pressure of 112/65 mmHg, and a heart rate of 105 bpm, with no signs of heart failure.

Notably, there were no new heart murmurs, Osler’s nodes or other signs suggestive of infective endocarditis (IE) in the clinical examination. Repetitive ECG showed tachycardic sinus rhythm. Laboratory tests indicated severe hyperthyroidism (elevated FT4 and FT3, suppressed TSH) and positive TSH-receptor antibodies, suggesting Graves’ disease.⁵ Coagulation tests showed elevated fibrinogen, D-Dimer, and von Willebrand factor (vWF) (Table 1).

An ultrasound of the thyroid gland showed enlargement without nodules, a diffusely heterogeneous echotexture, and diminished blood flow, consistent with the diagnose of Graves’ disease. Treatment with carbimazole and a non-selective beta-blocker was initiated. Given the patient's symptoms, and the presence of a bioprosthesis with the possibility of endocarditis, the patient was referred to his cardiologist. Subsequent TTE in the outpatient clinic revealed a preserved LV function (LVEF 59%) and an increased transprosthetic mean gradient of 12 mmHg at 75 bpm, without MR or vegetations. Due to artefacts and the elevated gradient, he was referred to our centre for further evaluation with transoesophageal echocardiography (TOE).

TOE revealed a 1.1 × 1.3 × 1.7 cm homogenous, echo dense nodular mass on the ventricular side of the medial cusp of the bioprosthetic valve, causing immobilisation of the same cusp. The lateral and posterior cusps functioned more or less normally with degenerative, age-appropriate changes. The transprosthetic mean gradient was 9 mmHg at 76 bpm. No other masses, vegetations, or abscesses were found (Figure 1A, Supplementary material online, Video S1; Table 2).

Figure 1

Baseline TOE: initial baseline transesophageal echocardiography (TOE) reveals a thrombotic mass on the ventricular side of the medial neo-cusp (indicated by red arrow) in both 2D and 3D reconstructions. (A) One-week follow-up: TOE performed 1 week later showing an unaltered thrombotic mass on the medial neo-cusp (red arrow). A new thrombotic mass is observed on the Cor-Knot on the atrial side (blue arrow). (B). One-month follow-up: TOE after 1 month showing remission of the previous thrombotic mass on the neo-cusp (red arrow). This is accompanied by a decrease in the transvalvular mean gradient (dPmean) and an increase in the mitral valve area. (C) Three-months Follow-up: TOE at 3 months shows no further thrombotic alterations (red arrow) with residual structural valve degeneration. (D).

Open in new tabDownload slide

Given all aforementioned findings and based on the ESC 2023 criteria for IE, IE was possible (1 major and 1 minor criterion).⁶ After obtaining multiple blood culture samples, empiric intravenous antibiotics (ceftriaxone 2 g daily) were started. Therapeutic anticoagulation was withheld due to possible IE and haemorrhagic risk. A follow-up TOE showed an unaltered mass on the medial cusp and a new smaller thrombotic mass on the Cor-Knot on the left atrial side (Figure 1B,  Supplementary material online, Video S2).

Serial blood cultures yielded negative results, and laboratory results showed no signs of infection. Furthermore, the patient remained afebrile, and his general condition improved overall. Given the course of the disease, we assumed that the prosthetic valve thrombosis was most likely linked to a hypercoagulable state caused by Graves’ disease. As a result, antibiotic therapy was discontinued, and therapeutic anticoagulation with VKA was initiated, supplemented by bridging therapy with LMWH until a therapeutic INR was attained. The patient was discharged with adjusted thyreostatic treatment and INR-guided anticoagulation with VKA (INR 2.5–3.5). A follow-up TOE 1 month later showed remission of the thrombotic mass on the medial neo-cusp, resolution of the mass on the mitral annulus, and a reduced transmitral gradient to 4.8 mmHg (Figure 1C, Table 2).

Laboratory results showed normalized thyroid function. The patient fully recovered and remained symptom-free with continued VKA therapy. A 3-month echocardiographic follow-up showed no new thrombotic changes and a slightly elevated transmitral gradient (5 mmHg) (Figure 1D, Supplementary material online, Video S3, Table 2), attributed to structural valve degeneration 6 years post-implantation.

Discussion

We present a case of bioprosthetic mitral valve thrombosis occurring 6 years post-implantation, likely linked to new-onset hyperthyroidism. To our knowledge, no prior reports have documented this association. The coincidence of new-onset hyperthyroidism and sudden prosthetic valve thrombosis influenced therapeutic decisions. The patient, previously stable post-MVR, presented with signs of IE. However, the simultaneous diagnosis of hyperthyroidism and symptom improvement with thyreostatic therapy made it difficult to pinpoint the cause.

Although TOE findings suggested valvular thrombosis (sessile round mass, location on ventricular side of cusps) rather than vegetations, early-stage IE couldn’t be ruled out. Consequently, an empirical antibiotic treatment without anticoagulation was initiated. Despite clinical improvement and normal inflammation markers, the initial mass persisted, and a new mass appeared on the mitral ring. This led us to abandon the IE diagnosis and stop antibiotics after 1 week. Known risk factors for valve thrombosis include hypercoagulable conditions, history of thromboembolic events, reduced left ventricular function, and atrial fibrillation^1,2; however, none of these factors were present in our patient. Instead, we considered the prosthetic valve thrombosis a complication of Grave’s disease and treated it accordingly with continuation of the thyreostatic treatment and initiation of oral anticoagulation. The patient’s favourable clinical course proved our therapeutic decisions right.

The endocarditis team initially contemplated surgical intervention due to the mass size exceeding 10 mm on the existing bioprosthetic valve. However, in the absence of heart failure, uncontrolled infection, signs of septic embolisation, as well as no indications of local complications and stable hemodynamics, it was decided to refrain from surgery given the patient's stable condition.

The initial decision to refrain from early anticoagulation was based on clinical symptoms and findings supporting IE. Considering the lack of evidence for concomitant anticoagulants in bioprosthetic valve endocarditis and the associated risk of serious bleeding complications, this approach was considered reasonable. However, discussing early add-on anticoagulation might have been prudent given the initial echocardiographic findings.

Echocardiography is the cornerstone evaluation in cases of suspected prosthetic valve thrombosis. Assessment of several parameters is suggested by the latest American guidelines on assessment of prosthetic valves.⁷ Serial measurements of multiple parameters are however, prone to measurement errors. In addition, our case highlights the difficulty of assessing the degree of valve degeneration or stenosis using different measurement techniques. While inconsistencies between calculated and measured parameters are a common problem in valvular assessments, they are pronounced in a hyperdynamic state (such as hyperthyroidism), as flow-dependent parameters (velocities and gradients) will overestimate the severity of the prosthetic stenosis. Furthermore, there is a conceptual difference between an effective orifice area calculated based on flow dynamics compared with an anatomically measured valve orifice area, typically resulting in a slightly larger orifice area with the latter technique. In prosthetic valve cases, we prefer multi-planar reconstruction of the anatomic prosthetic valve opening area by three-dimensional TOE, as it is much less flow-dependent and quite reproducible given good image quality. Further diagnostic evaluation with cardiac-CT or FDG-PET-CT was not performed, as we did not observe any signs of paravalvular lesions in TOE or symptoms suggesting extracardiac complications. Retrospectively, comprehensive imaging, including cardiac as well as brain and whole-body scans, could have helped distinguish between thrombosis and IE, detect peripheral lesions, and assess embolism and haemorrhagic risks, ensuring better timing for anticoagulant treatment initiation; hence, due to the patient`s stable condition and the overall low likelihood of IE in this setting, we chose an initial conservative approach.

Hyperthyroidism increases and enhances the activity of procoagulant factors such as factor VIII, factor IX, fibrinogen, and vWF.^8,9 This contributes to a hypercoagulable and hypofibrinolytic state and higher thrombosis risk.¹⁰ Our patient had elevated fibrinogen and vWF levels. Interestingly, an increased sensitivity to VKA in Graves’ disease has been documented,¹¹ highlighting the need for cautious anticoagulation therapy and regular monitoring to prevent thromboembolic and bleeding events.

In addition, hyperthyroidism increases heart rate, contractility, ejection fraction, and cardiac output. At the same time, it raises the risk of mortality and cardiovascular events like atrial fibrillation, thromboembolism, and stroke.^12,13 If untreated, it can lead to cardiomyopathy and heart failure.¹⁴ However, our patient did not show any atrial fibrillation during ECG monitoring. The young age and structurally normal atrium, as well as the initiated beta-blocker treatment, may have counterbalanced these effects.

This case highlights the complexity of managing new-onset prosthetic valve failure with symptoms of both IE and overt hyperthyroidism. A collaborative, multi-disciplinary approach involving cardiologists, infectious disease specialists, haematologists, and endocrinologists is essential for appropriate diagnostic work-up, thyroid disease treatment, and careful anticoagulation management, alongside close clinical and echocardiographic follow-up.

Conclusions

This case suggests that prosthetic valve thrombosis due to new-onset hyperthyroidism needs to be taken into consideration as a differential diagnosis in patients presenting with clinical signs of IE. An interdisciplinary approach with an immediate diagnostic work-up and appropriate treatment of both the underlying disease as well as the prosthetic valve thrombosis are crucial for a favourable outcome.

Lead author biography

Dr. Hasan Hadzalic graduated from the Medical University in Graz, Austria, in 2018. He is now a specialist in internal medicine with broad experience within both internal medicine and critical care. Currently, Dr. Hadzalic is completing a cardiology residency at the Heart Clinic Hirslanden in Zurich, Switzerland.

Supplementary material

Supplementary material is available at European Heart Journal – Case Reports online.

Consent: The authors confirm that written consent for submission and publication of this case report, including imaging and associated text, has been obtained from the patient’s family in line with COPE guidance.

Funding: No funding source was used in this work.

Data availability

The data underlying this article are available in the article and in its online Supplementary material or from the corresponding author on reasonable request.

References

Author notes