Low-Flow Aortic Stenosis After Transcutaneous Aortic Valve Replacement: It’s All About the Flow

Along the spectrum of aortic stenosis (AS) lie patients with severely reduced valve area (<1.0 cm² or <0.6 cm²/m² when indexed to body surface area) but low cardiac stroke volume index (SVI; generally considered to be <35 mL/m²). Such patients have a worse natural history than patients with normal SVI, and while their survival is improved by surgical or transcutaneous aortic valve replacement (SAVR or TAVR, respectively), it remains worse than those without low flow.¹ To better understand the determinants of outcomes in these patients, in this issue of JAMA Cardiology Anjan et al² have analyzed 984 patients with SVI less than 35 mL/m² from the Placement of Aortic Transcatheter Valves (PARTNER) trial who underwent TAVR, assessing which baseline and immediate postprocedural characteristics are predictive of 1-year mortality.

Unsurprisingly, having a higher Society of Thoracic Surgeons score, atrial fibrillation, and moderate or severe mitral regurgitation after the procedure were predictive of mortality, but there were some surprises. For example, male sex was a strong risk factor with a hazard ratio of 1.59 (95% CI, 1.18-2.13) in multivariate analysis, but neither age nor left ventricular ejection fraction were significant. Patients with the most severe AS (higher gradients, smaller area) also had a survival advantage following TAVR compared with patients with less severe AS.

The main focus of the analysis is on the changes in SVI following TAVR and how that predicts survival. Patients were stratified based on their predischarge echocardiogram into tertiles of SVI that roughly corresponded into those who had normalized their stroke volume (50% increase in SVI on average), those who remained severely depressed (an average fall in SVI of 11%), and those who improved (an average of 12%) but did not normalize SVI. Over the year of follow-up, the severe low-flow tertile showed greater than 6% excess absolute mortality over the other 2 groups, a difference that held up in a variety of subgroup analyses detailed nicely in the Supplement.

These results, while clearly significant, do need a bit of perspective, however. First, it should not surprise us that among patients with depressed cardiac output, those who improve their SVI early would do better than those who do not. Ideally, we would like evidence of what measures can be taken to improve outcome in these patients, and the observational nature of this study limits insight into this aspect.

Furthermore, these data are not completely relevant to the larger issue of low-flow, low-gradient AS. All of these patients had to fulfill the inclusion criteria for the PARTNER trial, which required a peak aortic valve velocity more than 4.0 m/s and/or a mean pressure gradient more than 40 mm Hg. For patients with depressed ejection fraction, dobutamine echocardiography could be used to generate such findings, but for the larger group of patients with “paradoxical” low-flow (normal ejection fraction), which composed 57% of the study patients here, all presumably met the set inclusion criteria without dobutamine supplementation. We thus have few data from this study regarding the patient with truly low-gradient severe AS, though the fact that larger valves and lower gradients were associated with worse survival suggests that not all of these patients may benefit from aortic valve replacement.

From a methodologic perspective, it is worth examining the fine points of flow determination pre- and post-TAVR. Though time honored and theoretically sound, the Doppler continuity equation requires some assumptions that may not always be true in the TAVR population. In general it assumes a flat velocity profile across a region where both the cross-sectional area and the stroke distance (velocity time integral of the Doppler tracing) are known,³ all of which may be violated in the current analysis. As stated in the article,² and consistent with contemporary recommendations,^4,5 Anjan et al used “the velocity time integral of the distal LV [left ventricular] outflow tract and its diameter in midsystole of the aortic annulus in the parasternal long axis view at baseline.^17,19 After implant, both Doppler and outflow tract measurements were obtained just below the edge of the valve stent.” Note that pre-TAVR, it appears that the area and velocity measurements were made in different places along the left ventricular outflow tract (LVOT), and given common distortions in the LVOT (eg, upper septal bulge and the known noncircularity of the annulus and LVOT), this may lead to either over- or underestimation of flow. Post-TAVR, the LVOT should be more circular, but in the entry region to the prosthesis, there is an abrupt acceleration of flow across a short distance that makes the positioning of the pulsed Doppler sample volume particularly critical. None of this should negate the observations, as consistent methods were used, but it should challenge the echocardiography community to strive continually for more accurate methods in the future, perhaps using 3-dimensional flow calculations. Also, at least some of the flow increase seen was due to the larger number of patients with more than mild aortic regurgitation in the group with normalized flow post-TAVR. Fortunately, reanalysis after excluding such patients yielded the same survival impact for SVI normalization.

Finally, it is worth acknowledging the ongoing contribution of the PARTNER family of trials to our understanding of aortic valve disease. When the valve guidelines were written in 2006, virtually all of the recommendations carried a level of evidence C (expert consensus). By the 2014 guidelines,⁶ there were several level A recommendations, reflecting randomized data from good clinical trials like PARTNER. Such efforts should be applauded and further trials encouraged, so that going forward, more and more of our recommendations in valvular heart disease will indeed be evidence based.

Corresponding Author: James D. Thomas, MD, FESC, Center for Heart Valve Disease at Bluhm Cardiovascular Institute of Northwestern Memorial Hospital, and Feinberg School of Medicine of Northwestern University, 676 N St Clair St, Ste 600, Chicago, IL 60611 (jthomas8@nm.org).

Published Online: June 15, 2016. doi:10.1001/jamacardio.2016.1206.

Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Thomas has received honoraria from Edwards, Abbott, and GE Medical.