By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: Data from transcatheter aortic valve replacement procedures performed at German hospitals without on-site cardiac surgery shows relatively low rates of major complications and mortality similar to hospitals with full surgical programs.

SOURCE: Eggebrecht H, Bestehorn M, Haude M, et al. Outcomes of transfemoral transcatheter aortic valve implantation at hospitals with and without on-site cardiac surgery department: Insights from the prospective German aortic valve replacement quality assurance registry (AQUA) in 17,919 patients. Eur Heart J 2016 May 17 [Epub ahead of print].

Since its commercial introduction just a few years ago, transcatheter aortic valve replacement (TAVR) has become rapidly integrated into the standard of care for aortic stenosis patients at elevated risk for surgery. The early model encouraged performance of the procedure in a cath lab or hybrid operating room, with full cath lab and surgical teams, and with cardiopulmonary bypass and cardiac surgical equipment at the ready should an emergency arise requiring open surgical intervention. Guidelines from the European Society of Cardiology recommend performing TAVR procedures only at hospitals with cardiac surgery available on site. The situation in the United States is even more restrictive, limiting TAVR to a subset of cardiac surgical centers that meet particular operator- and institution-specific requirements. In Germany, regulations requiring on-site cardiac surgical backup for TAVR have been relaxed as part of an effort to increase access to this procedure. As a prerequisite for performing TAVR, these hospitals were required to evaluate patients as part of a heart team with visiting cardiac surgical specialists from external, collaborating hospitals. The data from 2013-2014 have been published, comparing transfemoral TAVR procedures at institutions with and without cardiac surgery.

Between the beginning of 2013 and the end of 2014, 17,919 patients in Germany underwent transfemoral TAVR. By 2014, 22 hospitals were performing this procedure without on-site surgical backup vs. 75 with surgery. Of these, 1,332 underwent TAVR at hospitals without on-site cardiac surgery. One concern has been that performance of TAVR at non-cardiac surgery centers would lead to a detrimental change in patient selection. Patients undergoing TAVR at non-cardiac surgical hospitals were significantly older (82.1 vs. 81.1 years; P < 0.001), belonged to higher NYHA CHF class, and were more likely to present with a history of coronary disease, peripheral arterial disease, COPD, and neurologic events. As a result, patients undergoing TAVR at non-cardiac surgical hospitals had a higher calculated risk of surgical mortality according to both the GAV-score and the logistic EuroSCORE. Nearly 84% of patients underwent TAVR as elective procedures.

Procedure times were longer in hospitals without on-site cardiac surgery (110.3 ± 48.2 vs. 79.3 ± 44.8 min; P < 0.001), although fluoroscopy times were similar. Total procedural complications were lower in non-cardiac surgery centers (8.4% vs. 11%; P = 0.004), while catastrophic complications, including annular rupture, aortic dissection, and device embolization, were similarly rare (all < 1%) in both groups. While the composite of complications that could potentially benefit from open cardiac surgery were similar between groups (3.4% vs. 3.9%), conversion to open sternotomy was less likely at non-cardiac surgical sites (0.3 vs. 0.7%). In-hospital mortality (3.8 vs. 4.2%; P = 0.396), myocardial infarction, stroke, and vascular complications were all similar between groups.

It is worth noting that, similar to prior reports, in-hospital mortality was very high for all patients requiring emergent cardiac surgery for TAVR complications (50% in non-cardiac surgery hospitals and 62.5% in hospitals with on-site surgery; P = 0.694).

The authors argued that serious complications from TAVR have declined markedly over time with increased experience and better devices, and that their data support the safety of performing this procedure at sites without on-site cardiac surgical backup.


It is remarkable to note that more than 1,300 TAVR procedures were performed in German hospitals without on-site cardiac surgery over the short period from 2013 to 2014. Figures for mortality and life-threatening complications were low and were similar when compared to centers with cardiac surgery. The authors claimed their data support the feasibility and safety of this approach. Does this mean expanding TAVR to non-cardiac surgical centers is a good idea?

To explore this further, let’s look at the data closely. Compared to centers with cardiac surgery, those without had significantly lower institutional procedure volumes. This translated to longer procedure times, as well as higher rates of at-least-moderate aortic insufficiency and higher rates of permanent pacemakers, both of which can be affected by operator and institutional experience. Mortality did not significantly increase, and this is undoubtedly positive. Although the number of procedures analyzed here was substantial, it is likely not large enough to show a difference in low-frequency outcomes such as mortality and emergent cardiac surgery. It should also concern clinicians that patient selection was significantly different at non-cardiac surgery centers, compared with the more experienced hospitals with both cardiology and cardiac surgery on site. Although this did not translate into a statistical difference in mortality, one could wonder whether the heart team model is consistently applied within this paradigm. Although the rates of emergent cardiac surgery with TAVR have fallen to below 1%, and the odds of surviving such a complication — even in a center with on-site surgery — are low, it is difficult to discount the notion that small numbers of lives might be saved by the immediate availability of cardiac surgery. I am not yet convinced that the issue of access to these mainly elective procedures is compelling enough to expand outside of established cardiac surgical centers. In the United States, we might expect steady expansion of TAVR procedures to more hospitals that currently offer cardiac surgery; however, the inclusion of hospitals without cardiac surgery on-site is most likely not on the immediate horizon.