gms | German Medical Science

15. Deutscher Kongress für Versorgungsforschung

Deutsches Netzwerk Versorgungsforschung e. V.

5. - 7. Oktober 2016, Berlin

Lerneffekte bei der Einführung der kathetergestützten Aortenklappenimplantation (TAVI): deutsche Erfahrungen aus mehr als 32.000 Fällen

Meeting Abstract

  • Holger Reinecke - Universitätsklinikum Münster, Abteilung für Angiologie, Department für Kardiologie und Angiologie, Münster, Deutschland
  • Eva Freisinger - Universitätsklinikum Münster, Abteilung für Angiologie, Department für Kardiologie und Angiologie, Münster, Deutschland
  • Florian Lüders - Universitätsklinikum Münster, Abteilung für Angiologie, Department für Kardiologie und Angiologie, Münster, Deutschland
  • Jochen Reinöhl - Universitäts-Herzzentrum Freiburg - Bad Krozingen, Kardiologie und Angiologie I, ISAH - Intervention bei strukturellen und angeborenen Herzkr, Freiburg, Deutschland
  • Klaus Kaier - Universitätsklinikum Freiburg, Klinische Epidemiologie, Department für Medizinische Biometrie und Medizinische Informatik, Freiburg, Deutschland

15. Deutscher Kongress für Versorgungsforschung. Berlin, 05.-07.10.2016. Düsseldorf: German Medical Science GMS Publishing House; 2016. DocFV43

doi: 10.3205/16dkvf050, urn:nbn:de:0183-16dkvf0504

Published: September 28, 2016

© 2016 Reinecke et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



Background: Transcatheter aortic valve implantation (TAVI) is a rapidly evolving but extremely expensive technique for therapy of aortic valve stenosis. Previous studies reported learning curves with respect to in-hospital mortality and clinical complications; however, it is not clear to what extent this development is due to a change in patient population characteristics.

Aims: We intended to determine whether observed improvements in outcomes were the result of improved procedures or due to a change in the patient population. In addition, we investigated whether these trends differed between the transfemoral (TF) and the transapical (TA) approach.

Methods: We performed a retrospective observational study of all patients undergoing isolated TAVI in Germany (N=32,436) using the so-called §21-data data from the German Federal Office of Statistics. Data were analyzed using risk-adjusted regression analyses in order to track the development of clinical outcomes of all isolated TAVI procedures performed in Germany from 2008 to 2013. The targeted main outcome measures were in-hospital mortality, stroke, bleeding and mechanical ventilation which were compared firstly regarding the development during the mentioned time period and secondly comparing the differences between TA- and TF-TAVI.

Results: Unadjusted mortality rates showed that reductions in mortality were smaller for TA-TAVI (from 9.0 to 8.0%) than for TF-TAVI procedures (from 9.0 to 4.5%). In addition, risk-adjusted standardized mortality rates for TA- (around 7% in 2013) and TF-TAVI (around 4% in 2013) demonstrated that these differences were independent of baseline risk factor compositions (see Figure 1 [Fig. 1]). Standardized stroke rates were converging towards an identical stroke risk for TF-TAVI and TA-TAVI procedures of around 2% in 2013. Bleeding complications decreased for both TA- (from 15.5 to 9.5%) and TF-access (from 10.0 to 4.8%) routes. Both, unadjusted and standardized bleeding rates were substantially higher for TA-TAVI. In addition, TA-TAVI procedures were associated with an increased likelihood of requiring more than 48 hours of mechanical ventilation (9% in 2013) compared to TF-TAVI (3% in 2013).

Discussion: TAVI-related in-hospital mortality decreased dramatically between 2008 and 2013, and these improvements were not result of a change in the patient population or risk constellation. With respect to in-hospital mortality only, the results indicate the superiority of a TF-first approach.

Practical implications: For the first time, complete data from an entire industrialized nation could be used to assess the outcomes and complications regarding the very new and rapidly developing TAVI technology. This analysis shows that even after adjustment for many of patients’ characteristics and co-variates, the outcome after TF-TAVI developed much more positive compared to that after TA-TAVI. Moreover, the reduction of the observed mortality compared to predicted mortality (data not shown in the abstract) was also significantly better for TF-TAVI compared to TA-TAVI procedures. In summary, these data demonstrate the learning effect of the investigators improving their abilities and, moreover, clearly support TF access for TAVI patients in the future.