Novel imaging strategies for the Detection of Prosthetic Heart Valve Obstruction and Endocarditis

Abstract

Valvular heart disease is accompanied by a high mortality/morbidity and often requires prosthetic heart valve (PHV) replacement in order to improve quality of life and survival. The major drawback of both mechanical and biological PHV implantations is development of dysfunction, which is a life threatening condition. Cardiac imaging has a pivotal role in determining the correct treatment strategy for patients with PHV dysfunction. In clinical practice however the correct detection of PHV dysfunction remains difficult. Diagnostic dilemmas are predominantly encountered in PHV obstruction and endocarditis. In this thesis the diagnostic role of routine imaging techniques (fluoroscopy and two-dimensional transthoracic (TTE) and transesophageal (TEE) echocardiography) for PHV endocarditis and obstruction was determined and compared. Moreover, the additional value of novel imaging techniques on top of routine diagnostics was investigated. Results for PHV endocarditis: Two dimensional TEE is more sensitive than TTE for the detection of both vegetations and peri-annular complications. TEE detects the majority of signs of PHV endocarditis, however it still misses a substantial number of vegetations (18%) and peri-annular extensions (14%) as compared to the reference standard (i.e. surgical inspection/autopsy or clinical follow-up). This thesis provides evidence that the following novel imaging strategies in the field of PHV endcarditis have an additional value: Three Dimensional Transesophageal Echocardiogram (3D-TEE) or Multidetector Computed Tomography (MDCT) and/or 18F-Fluorodesoxyglucose Positron Emission Tomography (FDG-PET) . Results for PHV obstruction: For the detection of biological and mechanical PHV obstruction TTE is an excellent diagnostic tool for determination of the severity of obstruction, however in most cases it is not able to detect the exact origin of the obstruction. The therapeutical consequence of significant obstruction of a biological PHV in a symptomatic patient is re-operation (if clinically possible). Therefore additional diagnostics are not required in biological PHV obstructions as they do not change patient management. In mechanical PHV obstruction however the differentiation between the cause of acquired obstruction (pannus/thrombosis) is very important as it may result in major therapeutical differences. Obstructive thrombosis may require fibrinolysis and/or heparine infusion, whereas this is strictly contra-indicated in obstructive pannus. Clinical parameters, fluoroscopy and TEE are not able to differentiate between thrombosis and pannus reliably. This thesis however provides evidence that MDCT is able to differentiate more reliably than any other imaging tool clinically available now. In conclusion, PHV endocarditis and obstruction remains difficult to diagnose correctly by echocardiography alone. As clinical implications may be major, this thesis suggests that clinicians should have a low threshold to perform additional imaging by novel techniques in the field of PHV dysfunction such as 3D-TEE, FDG-PET/CT and/or MDCT.