Despite improvements in available techniques and prostheses, the long-term evolution of patients after transcatheter paravalvular leak closure remains poorly understood. The objective of this study is to evaluate the long-term outcome of patients after transcatheter paravalvular leak closure and to identify factors predictive of survival.

The investigators designed a retrospective multicentre observational registry entitled FFPP (e.g. "Fermeture des Fuites Para-Prothétiques" in French). Seventeen centres from France, Poland, Turkey and Belgium enrolled retrospectively at least one PVLc procedure on aortic, mitral, tricuspid or pulmonary valves without limit in its delay of performance.Follow-up of at least one year was expected

Paravalvular leak (PVL) of cardiac valve prostheses is a frequent complication after surgical valve replacement. The incidence varies according to the type of prosthesis and location, ranging from 2-15% for aortic prostheses and 7-17% for mitral prostheses, with an increasing incidence over time. PVL may also be observed after percutaneous valve replacement. After transcatheter aortic valve implantation (TAVI) the incidence of PVL at 1 year is 6%, regardless of the type of prosthesis placed.

The majority of these leaks do not lead to symptoms but they are sometimes responsible for hemolytic anemia, heart failure and infectious endocarditis, thus posing an indication for closure. The reference technique for PVL closure (PVLc) is redo cardiac surgery despite significant morbidity and mortality.

Percutaneous PVLc techniques have been developed since 1992. Several small case series have been published since then, highlighting the different closure techniques and the modalities for evaluating PVL by 3D ultrasound and CT scan. Several multicenter registries have also been published that trace current practices and also identify prognostic factors for success in the short to medium term.

However, despite improvements in available techniques and prostheses, the long-term evolution of patients after percutaneous PVLc remains poorly understood. The objective of this study is to evaluate the long-term outcome of patients after percutaneous PVLc and to identify factors predictive of survival.

The investigators designed a retrospective multicentre observational registry entitled FFPP (e.g. "Fermeture des Fuites Para-Prothétiques" in French). Seventeen centres from France, Poland, Turkey and Belgium enrolled retrospectively at least one PVLc procedure on aortic, mitral, tricuspid or pulmonary valves without limit in its delay of performance.Enrolment started on January 1st, 2017. Study database was closed on december 31st, 2019. The study was financially supported by Marie Lannelongue Hospital, without any financial support from the industry. This research was conducted according to Declaration of Helsinki. The study protocol was approved by an independent review board (CCTIRS, 6th January 2017, n°16-622). An electronic case report form was created (easy-crf®). Each local investigator entered data using a secure access. Data quality control was performed by Marie-Lannelongue research assistants.

Past medical history, cardiac surgery details, clinical, biological and echocardiographic parameters were collected. EUROSCORE II risk scale (http://euroscore.org) was assessed with the assumption that a surgical valve replacement would be performed. Both technical success and clinical success were defined according to the Paravalvular Leak Academic Research Consortium. Technical success was implantation of at least one device within the leak, with an at least 1-grade decrease in the echocardiographic leak-severity grade, no valve dysfunction, and no surgical conversion of the procedure. Clinical success was survival at one month without re-admission for heart failure, blood transfusion, or open-heart valvular surgery. The investigators distinguished per procedural complications, in-hospital complications (index hospitalization) and complications occurring during follow-up. Complication's severity was graded using a five levels scale.For each patient, the latest follow-up data at the time of the inclusion period was collected. Follow-up of at least one year was expected. New valvular surgery, new PVLc procedure, death and its causes were collected through the patients themselves or their cardiologist by each local investigator.

Inclusion Criteria:

Seventeen centres from France, Poland, Turkey and Belgium enrolled retrospectively at least one PVLc procedure on aortic, mitral, tricuspid or pulmonary valves without limit in its delay of performance.

Exclusion Criteria:

• Patients under 18 years of age,

An electronic case-report form (eCRF, Easy-crf.com) was completed for each patient. The medical history (notably regarding heart surgery), symptoms, physical findings, laboratory test results, and echocardiographic parameters were collected. The EUROSCORE II risk score (http://euroscore.org) was determined, since surgical valve replacement was among the treatment options. Creatinine clearance below 60 mL/min defined renal failure and below 30 mL/min severe renal failure. Heart failure was defined according to the European Society of Cardiology as the presence of symptoms (dyspnoea, orthopnoea, asthenia) with physical signs of heart failure (limb oedema, hepato-jugular reflux, crackles on pulmonary auscultation) linked to the valvular disease. Anaemia was defined as haemoglobin <13 g/dL in men and <12 g/dL in women and haemolytic anaemia as anaemia with lactic dehydrogenase elevation and/or schistocytes and/or low haptoglobin, in the absence of other causes of anaemia.