Edwards SAPIEN 3 PPI Registry

Aortic Valve Stenosis, AV Block, Bundle-Branch Block

Transcatheter aortic valve implantation, Pacemaker, Aortic valve calcification, PPI

There are patient related risk factors for PPI that can be identified and assessed in retrospective pooling of 1000+ TAVI patient datasets. Retrospective pooling of 6 datasets already available at participating centres (4 sites in Germany, 1 in Zwolle / The Netherlands, 1 in Linköping / Sweden). Additional assessment of calcifications using a CT data core lab. Statistical analysis of the obtained dataset with respect to the objectives of the registry.

Permanent pacemaker implantation is a widely recognized clinical event associated with TAVI becoming evident within a few days after the procedure. While a number of registries have documented the rates of PPI with different valves, much less evidence has been provided for 1) patient based characteristics (e.g. RBBB etc.) affecting the likelihood of PPI and for 2) procedural variables (e.g. implantation depth, valve size etc.) that should be considered to perform an as safe and minimal invasive procedure as possible. Prior research To date there are 7 published reports on pacemaker rates and predictors associated with the use of the Edwards SAPIEN 3 THV (see Table). These studies suggest that the need for pacemaker implantation in single centers ranges between 14.4 and 20.4% derived from patient numbers between 131 and 335. These analyses resulted in the identification of pre-existing conduction disturbance / aortic valve calcification as patient related and implantation depth /oversizing as procedure related variables associated with Edwards SAPIEN 3 THV TAVI. Furthermore a scoring algorithm for patient related factors has been proposed from a dataset of 240 patients to predict the likelihood of PPI after THV . Limitations of prior research Current evidence though is limited by patient numbers versus event rates (with a max. of 62 PPI considered in any of the available datasets) resulting in a limited power in multivariable analyses, the single center design of these ventures, the lack of a consistent definition of variables potentially associated and the unexplained differences in the number and type of variables identified. Aims For this reason the investigators consider it necessary to 1) to retrospectively pool data from all of these 7 datasets to arrive at a more comprehensive picture of patient related variables (retrospective part I), to 2) assess procedural variables in a prospective multicenter registry (prospective part II) and 3) to potentially validate the identified variables in a future prospective cohort.

Germany