Author + information
- Marc R. Dweck, MD, PhD∗ (, )
- Calvin Chin, MD and
- David E. Newby, MD, PhD
- ↵∗Reprint requests and correspondence:
Dr. Marc R. Dweck, Centre for Cardiovascular Science, Chancellor’s Building, University of Edinburgh, Little France Crescent, Edinburgh, Scotland EH16 4SB, United Kingdom.
- aortic valve calcification
- aortic valve stenosis
- Doppler echocardiography
- multidetector computed tomography
Current guidelines for the assessment of aortic stenosis severity are based upon echocardiographic measurements of the peak velocity, the mean gradient, and the aortic valve area (AVA) (often indexed to body surface area). Thereafter, the decision to proceed to aortic valve replacement is largely governed by the presence of symptoms in those with severe disease. Although these guidelines are useful in the majority of patients, an important limitation to this model has been elegantly described in this issue of the Journal.
In this large multicenter study, Clavel et al. (1) assessed 646 patients with moderate or severe aortic stenosis and preserved ejection fraction, all of whom underwent comprehensive echocardiographic evaluation. This revealed that, in almost one-third of patients (n = 186), there was a discordance in the classification of aortic stenosis severity: most commonly (n = 172) characterized by a tight AVA (<1.0 cm2) in the context of a relatively low mean gradient (<40 mm Hg). This would suggest that, in a significant proportion of patients, uncertainty exists as to whether they have moderate or severe stenosis, with such a distinction potentially crucial in determining appropriate clinical management and the need for surgery.
So what is the explanation for this discrepancy? Data from the current study provide some clues. It is unlikely to reflect the study population or an effect of patient size, because values for AVA were indexed to body surface area. We can be reasonably reassured that it is unlikely to reflect suboptimal echocardiographic imaging, given that the results were compiled in 3 experienced centers. Indeed, discrepant findings occur frequently in all cardiac centers and are a common experience for most clinical cardiologists. Low-flow severe aortic stenosis has been previously described in patients with low stroke volumes with and without impaired systolic function (2). However, all patients in this study had a normal ejection fraction (>50%), and low flow (defined as an SVi <35 ml/m2) was observed in just 25 subjects, leaving the vast majority (n = 147) with a normal flow status.
As the authors indicate, reduced aortic compliance might have a role. However, the simplest explanation might instead lie in the severity thresholds chosen for each of the different echocardiographic parameters. These are largely historical and derived from cardiac catheterization data that are not directly comparable to echocardiographic measurements. Moreover, the thresholds selected in the guidelines have inherent inconsistencies, with theoretical models indicating that an aortic valve area of 1 cm2 relates more to a mean gradient of 30 to 35 mm Hg rather than the 40 mm Hg used in the guidelines.
Alternatively, the answer might relate to the tendency of echocardiography to underestimate the left ventricular outflow tract (LVOT) diameter. Alongside the pre- and post-valve Doppler data, this represents an integral component of the continuity equation and therefore a key influence on AVA measurements. With the advent of transcutaneous aortic valve replacement, greater focus has been placed on how best to size the LVOT. Contrast-enhanced computed tomography has clearly demonstrated that this structure is not circular but oval, so that echocardiographic measurements on parasternal long-axis views frequently underestimate its true size (3). As a consequence, AVA measurements might also be underestimated in patients with a particularly elliptical LVOT.
Whatever the explanation, perhaps the more important question is what should we do to resolve this discrepancy? The authors of the current study adopted an interesting approach: turning to computed tomography (CT) calcium scoring as a second method of disease severity assessment. First, they studied the patients in whom the classification of aortic stenosis severity was not in doubt to determine the CT calcium score that best differentiated moderate from severe aortic stenosis. Male patients with severe disease had significantly higher AV calcium scores than female patients, so that sex-specific thresholds were required (AVC ≥1,275 AU in women and ≥2,065 AU in men). These provided specificities and sensitivities for severe aortic stenosis of between 80% and 89%. Subsequently, the authors then used these optimal thresholds to determine whether patients with discordant echocardiographic measures had moderate or severe stenosis according to CT. With this approach, roughly one-half were in the moderate (47%) and one-half were in the severe group (53%), indicating that CT might provide useful discrimination in this difficult subgroup of patients.
Follow-up studies are clearly required to assess whether the categorization provided by CT accurately identifies those at high-risk of adverse events; however, this strategy holds several potential advantages. Firstly, CT calcium scoring is relatively simple in its approach, easy to conceptualize, and independent from the confounding effects of volume status, aortic compliance, and left ventricular function. Put simply, the more calcium in the valve, the more severe the aortic stenosis (4). This is in sharp contrast to some of the novel echocardiographic measures of severity that seem to involve a large number of variables and assumptions. Secondly, aortic valve calcium scoring, even when performed in a semi-quantitative manner, has already demonstrated good prediction of disease progression and adverse events, out-performing more conventional measures of disease severity (5).
This approach does have some clear limitations. These include both the availability of CT and the radiation dose associated with these scans (1 to 3 mSv), although the former is increasing, and the latter is of lesser relevance to elderly patients with aortic stenosis. In addition, there are concerns about the reproducibility of calcium scoring at the very high levels observed in the valve (this was not addressed in the study) (6), and this approach does not take into consideration the hypertrophic response of the left ventricle, which is highly variable and also likely to contribute to the development of symptoms and adverse events (7). Finally, and as discussed, there is a lack of follow-up data to validate the thresholds in calcium score proposed in this study.
What should we do when confronted with discrepant echocardiographic findings in clinical practice? In the absence of a gold-standard arbitrator, a pragmatic approach would be to carefully revisit the symptomatic status of the patient, given that this is the key determinant of the need for surgery. If necessary, this might include a cautious graded exercise tolerance test, which can unmask symptoms and provide a functional assessment of the cardiac response to physical stress. However, an alternative assessment of their aortic stenosis severity will often be required, given that elderly patients often have multiple potential causes for their symptoms. Calcification clearly underlies much of the pathophysiological stiffening of the valve and is the best predictor of disease progression (5). Therefore, CT or echocardiographic calcium scoring would seem to be the prime candidates with which to gain further complementary noninvasive information about the state of the valve.
Further studies are required to investigate other potential biomarkers, such as brain natriuretic peptide (8), troponin, the strain pattern on the electrocardiogram (9), and the presence of myocardial fibrosis on cardiovascular magnetic resonance (10). However, what is clear is that physicians should be alert to the potential for discrepancy with standard echocardiographic assessments and the potential for the burden of valvular calcification to help identify those at most risk. Beware the hard hearted…
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Drs. Dweck and Newby are supported by the British Heart Foundation. Dr. Chin has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Clavel M.A.,
- Messika-Zeitoun D.,
- Pibarot P.,
- et al.
- Pibarot P.,
- Dumesnil J.G.
- Jabbour A.,
- Ismail T.F.,
- Moat N.,
- et al.
- Dweck M.R.,
- Boon N.A.,
- Newby D.E.
- Bergler-Klein J.,
- Klaar U.,
- Heger M.,
- et al.
- Greve A.M.,
- Boman K.,
- Gohlke-Baerwolf C.,
- et al.
- Dweck M.R.,
- Joshi S.,
- Murigu T.,
- et al.