Resection of a giant thoracic solitary fibrous tumor through two separate thoracotomies

Abstract

tion, and significant residual shunt) was as low as 48% in a series of 24 patients. Common reasons for failure of this demanding intervention are poor imaging and unsuitable anatomy, such as crescent-shaped or multiple leaks and severe calcification. Additional limitations result from inadequate device configuration, because the occluders in use were originally designed for atrial and ventricular septal defects or patent ducts. Consequently, surgery still remains the criterion standard of treatment for paravalvular leaks while the technology of interventional closure is being advanced. Our program comprised 1842 heart valve surgeries in 2006, including 26 patients (0.14%) operated on for paravalvular leaks. Even in this large series, the 3 cases presented here are unusual because of the multiple previous surgeries. These cases were technically challenging and of high risk, and avoiding fourth-time cardiac surgery for reoperative valve exchange by use of catheter-based intervention would have been preferable. We therefore advocate a multidisciplinary approach that includes interventional cardiologists and surgeons. Depending on level of hemolysis, severity of regurgitation, and anatomy of the paravalvular leak, the risk of surgery must be weighed against the likelihood of success of intervention. Both treatment options are clearly explained to the patient so that an informed decision can be made. Surgical backup in case of interventional complications is provided for every intervention as well. Future studies, perhaps with more suitable device technology, may address a hybrid approach, so that outcomes can be improved even for most complex cases of paravalvular leakage.