By Van Selby, MD

Assistant Professor of Medicine, University of California, San Francisco Cardiology Division, Advanced Heart Failure Section

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

SYNOPSIS: In patients with low-flow, low-gradient aortic stenosis, current guideline-based criteria for identifying true severe aortic stenosis did not predict aortic stenosis severity or survival. Calculation of the projected aortic valve area at a normal transvalvular flow rate more accurately identifies true severe aortic stenosis and is a stronger predictor of outcomes.

SOURCE: Annabi MS, Touboul E, Dahou A, et al. Dobutamine stress echocardiography for management of low-flow, low-gradient aortic Stenosis. J Am Coll Cardiol 2018;71:475-485.

Current guidelines define low-flow, low-gradient aortic stenosis (LF-LG AS) as the combination of left ventricular ejection fraction (LVEF) < 50%, aortic valve area (AVA) ≤ 1.0 cm2, and a mean gradient (MG) < 40 mmHg. Dobutamine stress echocardiography (DSE) is recommended in patients with LF-LG AS to identify those with true severe aortic stenosis (TSAS) from pseudo-severe aortic stenosis (PSAS). TSAS is diagnosed when the AVA is ≤ 1.0 cm2 with MG ≥ 40 mmHg during DSE. However, these criteria have not been well validated.

The creators of The Multicenter Prospective Study of Low-Flow Low-Gradient Aortic Stenosis (TOPAS) enrolled 186 patients with LVEF ≤ 40%, aortic valve MG ≤ 40 mmHg, and indexed AVA ≤ 0.6 cm2/m2. All patients underwent DSE. AS severity was confirmed either by macroscopic evaluation of the valve by a surgeon at the time of aortic valve replacement (AVR) using predefined criteria or by quantification of aortic valve calcification using multidetector computed tomography (MDCT). In addition to standard parameters obtained during DSE, the authors calculated the projected aortic valve area (AVAproj) at a normal transvalvular flow rate (250 mL/min).

The recommended combination of peak stress MG and peak stress AVA correctly classified AS severity in only 47% of patients. AVAproj ≤ 1.0 cm2 correctly classified 70% of patients, better than any other marker evaluated (P < 0.0007). In an adjusted analysis of medically managed patients who did not undergo AVR, AVAproj ≤ 1.0 cm2 was a strong predictor of survival (hazard ratio, 3.65; P = 0.0003). Neither AVA or MG at peak stress predicted survival in medically managed patients. The authors concluded that in patients with low LVEF LF-LG AS, the current guideline-recommended criteria of peak stress MG ≥ 40 mmHg and AVA ≤1 cm2 carry limited value for predicting actual stenosis severity and outcomes. The AVAproj may be superior.


Although they only comprise a small minority of AS patients, those with LF-LG AS often present a diagnostic and therapeutic challenge. It is critical to identify patients with PSAS, where the calculated AVA is low because the heart cannot generate enough flow to open the aortic valve leaflets, even in the absence of severe disease. Guidelines from the American College of Cardiology/American Heart Association recommend DSE to discriminate PSAS from TSAS, and even give a Class IIa recommendation for AVR in patients with LF-LG AS who meet criteria for TSAS (MG ≥ 40 mmHg and AVA ≤ 1.0 cm2 during DSE). However, Annabi et al provided strong evidence that these criteria are neither sensitive nor specific for identifying TSAS. Furthermore, because these parameters do not predict survival in medically treated patients, they are not ideal tools for identifying candidates for AVR.

Both MG and AVA are flow-dependent parameters, and dobutamine is administered with the goal of increasing transvalvular flow to normal levels. However, in the Annabi et al study, approximately one-half of patients did not demonstrate normal flow during DSE. Both subnormal and supranormal flow rates would limit the reliability of both AVA ad MG significantly.

On the other hand, the AVAproj is standardized for a transvalvular flow rate that is fixed for all patients. The formula used by the authors to calculate AVAproj was as follows:

AVAproj = AVArest + AVApeak - AVArest × (250 - Qrest)
  Qpeak - Qrest

Q represents flow; peak values are the maximum obtained during DSE and rest are values obtained during the resting echocardiogram. The AVAproj is independent of LV function and flow. The authors showed strong evidence that AVproj, using a cutoff of 1 cm2, identifies TSAS and predicts survival in LF-LG AS much better than AVA, MG, or a combination of the two. There are several limitations to mention. Medically treated patients were thought to have either nonsevere AS or significant comorbidities, and this may introduce selection bias. The gold standard used in this study for diagnosing severe AS was assessment of the valve by the surgeon at the time of AVR. This is not a rigorous, well-defined metric for grading AS severity. The AVAproj will require further validation in other AS cohorts before its widespread use can be recommended. However, in the meantime, we now have an additional metric for discriminating TSAS vs. PSAS among patients with LF-LG AS. It is worth calculating AVAproj for all DSE performed to evaluate LF-LG AS. It may be particularly helpful when standard parameters for differentiating TSAS from PSAS remain discordant despite DSE.