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: A meta-analysis of studies comparing transcatheter aortic valve replacement (TAVR) to surgical aortic valve replacement over a two-year follow-up in low surgical risk patients exhibits a higher mortality in the TAVR group, prompting the authors to recommend caution in applying TAVR to low-risk patients until randomized trials are completed.

SOURCE: Witberg G, Lador A, Yahav D, Kornowski R. Transcatheter versus surgical aortic valve replacement in patients at low surgical risk: A meta-analysis of randomized trials and propensity score matched observational studies. Catheter Cardiovasc Interv 2018 Feb 1. doi: 10.1002/ccd.27518. [Epub ahead of print].

The use of transcatheter aortic valve replacement (TAVR), which first gained commercial approval in the United States in late 2011, has been growing. Although the initial indication was reserved for patients at prohibitive risk for surgical AVR, the approval expanded quickly to include high-risk patients, and then more-recently to intermediate-risk patients. The most recent data suggest that more than 24,000 TAVR procedures are performed in the United States annually. Can approval for low-risk patients be far behind? Although interest in the less-invasive TAVR option is keen among patients and referring providers, data on outcomes in the low-risk population (generally defined as a Society for Thoracic Surgery [STS] score < 4%) have been sparse. Witberg et al published a meta-analysis that included seven studies (five were randomized, controlled trials) published between 2012 and 2017. A total of 3,484 patients in this risk category were enrolled in these trials. Mean age of the included patients was 80 years, and the average STS score was 3.

In the whole data set, short-term mortality was similar between TAVR and surgical aortic valve replacement (SAVR; 2.2% vs. 2.6%; relative risk [RR], 0.89; 95% confidence interval [CI], 0.56-1.41; P = 0.62). However, intermediate-term data with a median follow-up of two years showed a higher mortality rate for TAVR than for SAVR (17.2% vs. 12.7%; RR, 1.45; 95% CI, 1.11-1.89; P = 0.006).

Essentially, comparative procedural complication rates mirrored those reported by earlier studies of intermediate and high-risk groups. Importantly, there was no significant difference in rates of stroke or myocardial infarction between TAVR and SAVR groups (1.4 vs. 1.5%, and 0.75 vs. 1.26%, respectively). However, TAVR showed an expected advantage regarding major bleeding and acute kidney injury. The authors concluded that although TAVR and SAVR produce similar short-term mortality, their data suggest a higher mortality for TAVR beyond the short term. They suggested that expansion of TAVR to the low-risk population should be restricted, at least until data from dedicated randomized, controlled trials become available.


Since its commercial introduction in the United States, TAVR has become more common and straightforward, with all-percutaneous access serving as the norm, along with shorter lengths of stay and improved outcomes. Use of this technology has continued to increase year over year with the inclusion of lower-risk patients. At our center we are asked on a regular basis to consider TAVR in patients who do not fit the typical elevated risk profile. Of course, patients are interested in options that are less invasive and result in shorter recovery times (so are their physicians, for that matter). Are we properly serving these patients if we steer them toward TAVR as opposed to SAVR?

This meta-analysis sounds a cautionary note in the otherwise seemingly inevitable march toward the less-invasive procedure, suggesting that TAVR in low-risk patients may be associated with greater mortality at a median follow-up of two years. The authors are correct to counsel caution, and to recommend waiting for the randomized trials of low-risk TAVR before reaching firm conclusions. However, there is reason to doubt the mortality conclusion. Individual randomized, controlled trials of intermediate-risk patients have been published in the past two years: PARTNER 2 (2,032 patients) and SURTAVI (1,660 patients). The authors of both reported no significant mortality differences between TAVR and surgery at two years of follow-up, with a similar total number of patients. The possibility of a mortality hazard specific to low-risk patients as opposed to intermediate-risk patients at the still-short two-year mark is not very plausible. Notably, the longer-term outcomes that will be of particular interest for lower-risk patients are lacking and may remain so for quite a few years.

Practically speaking, there are two major concerns limiting TAVR uptake in the low-risk population. First, the longevity of TAVR valves is still an unknown. These valves simply have not been in use long enough to develop longer-term data, and there is reason to think that the process of crimping valves for transcatheter delivery may affect lifespan. Obviously, this is of more concern to a low-risk patient at age 70 compared to a patient at age 85 with similar comorbidities. Second, pacemaker rates remain significantly higher with TAVR than with SAVR, even with the latest-generation devices. Again, this is of greater concern to younger patients for whom a permanent pacemaker will mean more years of surveillance and generator changes, not to mention the potential deleterious effects of right ventricular pacing. For providers considering the choice between TAVR and SAVR for low-risk patients, age may be a better discriminator than low-risk status alone, at least today. For the future, at least three large trials of low-risk TAVR are ongoing (PARTNER 3, Medtronic Transcatheter Aortic Valve Replacement in Low Risk Patients, and NOTION 2), so more definitive answers can be expected in the coming years.