cardiac imaging for assessing low-gradient severe aortic...

J A C C : C A R D I O V A S C U L A R I M A G I N G VO L . 1 0 , N O . 2 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 1 9 3 6 - 8 7 8 X / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 7 . 0 1 . 0 0 2

Cardiac Imaging for AssessingLow-Gradient Severe Aortic Stenosis
Marie-Annick Clavel, DVM, PHD,a Ian G. Burwash, MD,b Philippe Pibarot, DVM, PHDa
ABSTRACT

Fro

Qu

is t

Hi

Lif

rep

Ma

Up to 40% of patients with aortic stenosis (AS) harbor discordant Doppler-echocardiographic findings, the most common of

which is the presence of a small aortic valve area (#1.0 cm2) suggesting severe AS, but a low gradient (<40mmHg) suggesting

nonsevere AS. The purpose of this paper is to present the role of multimodality imaging in the diagnostic and therapeutic

management of this challenging entity referred to as low-gradient AS. Doppler-echocardiography is critical to determine the

subtype of low-gradient AS: that is, classical low-flow, paradoxical low-flow, or normal-flow. Patients with low-flow, low-

gradient AS generally have aworse prognosis comparedwith patientswith high-gradient or with normal-flow, low-gradient AS.

Patientswith low-gradient AS and evidence of severe AS benefit fromaortic valve replacement (AVR). However, confirmation of

the presence of severe AS is particularly challenging in these patients and requires a multimodality imaging approach including

low-dose dobutamine stress echocardiography and aortic valve calcium scoring by multidetector computed tomography.

Transcatheter AVR using a transfemoral approach may be superior to surgical AVR in patients with low-flow, low-gradient AS.

Further studies are needed to confirm the best valve replacement procedure and prosthetic valve for each category of low-

gradient AS and to identify patientswith low-gradient AS inwhomAVR is likely to be futile. (J AmColl Cardiol Img 2017;10:185–

202) © 2017 by the American College of Cardiology Foundation.

C alcific aortic stenosis (AS) is the most preva-lent valvular heart disease in developedcountries and is responsible for approxi-

mately 85,000 valve replacement procedures and15,000 deaths per year in North America. Currently,surgical or transcatheter aortic valve replacement(AVR) remain the only effective treatments for severeAS. The diagnosis and staging of AS are determinedprimarily on an assessment of the hemodynamicseverity and left ventricular (LV) systolic functionby Doppler echocardiography, and the presence ofsymptoms. Typically, AS is considered severe if thepatient has a mean transvalvular gradient $40mm Hg, a peak aortic jet velocity $4 m/s, an aorticvalve area (AVA) #1.0 cm2, and an indexedAVA #0.6 cm2/m2 (1–3). However, up to 40% of thepatients with AS harbor discordant Doppler-echocardiographic findings, the most common ofwhich is the presence of a small AVA (#1.0 cm2) sug-gesting severe AS, but a low gradient (<40 mm Hg)

m the aInstitut Universitaire de Cardiologie et de Pneumologie de Québec

ébec, Québec, Canada; and the bUniversity of Ottawa Heart Institute, Unive

he Canada Research Chair in Valvular Heart Diseases, Canadian Institutes

s research program is funded by research grant # FDN-143225 from CIHR. D

esciences and Medtronic for echocardiography core laboratory analyses i

orted that they have no relationships relevant to the contents of this pap

nuscript received November 28, 2016; revised manuscript received Decem

suggesting nonsevere AS (Figure 1) (4–6). This discor-dant grading pattern, often referred to as “low-gradient AS,” raises challenges and uncertainties asto the “actual” severity of the valve disease and theappropriate therapeutic decision making. Thepurpose of this paper is to provide a start-of-the-artreview and some future perspectives on theetiology, pathophysiology, diagnosis, and therapeuticmanagement of low-gradient AS. In particular, wewill underline the important role of multimodalityimaging for the characterization and management ofthe different types of low-gradient AS, including clas-sical and paradoxical low-flow, low-gradient; andnormal-flow, low-gradient AS.

CLASSIFICATIONOFASANDDIFFERENT TYPES OF

LOW-GRADIENT AS

The American College of Cardiology/American HeartAssociation and European Society of Cardiology/

(Québec Heart and Lung Institute), Laval University,

rsity of Ottawa, Ottawa, Ontario, Canada. Dr. Pibarot

of Health Research (CIHR), Ottawa, Ontario, Canada.

r. Pibarot has received a research grant from Edwards

n transcatheter heart valves. All other authors have

er to disclose.

ber 26, 2016, accepted January 5, 2017.

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jcmg.2017.01.002&domain=pdf

http://dx.doi.org/10.1016/j.jcmg.2017.01.002

ABBR EV I A T I ON S

AND ACRONYMS

AS = aortic stenosis

AVA = aortic valve area

AVR = aortic valve

replacement

HFrEF = heart failure with

reduced left ventricular

ejection fraction

LF-LG = low-flow,

low-gradient

LVEF = left ventricular ejection

fraction

Clavel et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 0 , N O . 2 , 2 0 1 7

Cardiac Imaging for Assessing Low-Gradient AS F E B R U A R Y 2 0 1 7 : 1 8 5 – 2 0 2

186

European Association for Cardio-ThoracicSurgery guidelines have proposed similarclassification systems for the different stagesof AS according to the hemodynamic severityof AS (mild, moderate, or severe), the state ofthe LV systolic function (normal vs.depressed), the patient’s symptomatic status(asymptomatic vs. symptomatic), the flowstate (normal vs. low), and the level of thegradient (high vs. low) (Figure 1) (1,3). In pa-tients with moderate AS, defined as a meangradient between 20 and 40 mm Hg and anAVA between 1.0 and 1.5 cm2 (“stage B” inAmerican College of Cardiology/American

Heart Association guidelines), there is currently noindication for AVR unless the patient undergoes car-diac surgery for another reason (e.g., coronary arterybypass grafting). Patients with high-gradient (meangradient $40 mm Hg) AS have a Class I indication forAVR if they have a depressed left ventricular ejectionfraction (LVEF) in the absence of symptoms (stage C2)or if they are symptomatic (stage D1). In patients witha low gradient (<40 mm Hg) but small AVA (#1.0cm2), several subtypes can be defined according tothe LVEF and flow status (Figure 1, CentralIllustration).

CLASSICAL LOW-FLOW, LOW-GRADIENT AS. Theclassical low-flow, low-gradient (LF-LG) AS entity isdefined in the guidelines as a LVEF <50%, AVA #1.0cm2, and mean gradient <40 mm Hg (Figures 1 and 2).This entity is found in about 5% to 10% of the ASpopulation, is more prevalent in men, and is veryoften associated with coronary artery disease (7).Patients with classical LF-LG AS generally have anenlarged LV cavity with depressed LV systolicfunction, and a substantial proportion also havefunctional mitral regurgitation (Figure 2, OnlineVideos 1 and 2). The depressed LVEF may becaused by afterload mismatch related to the pres-ence of severe AS and/or concomitant intrinsic dis-ease of the myocardium, the most frequent being anischemic cardiomyopathy. The classical LF-LG ASentity is identified as the stage D2 in the guidelines,and there is a Class IIa recommendation for AVR inpatients with confirmed severe AS (Figure 1, CentralIllustration).

PARADOXICAL LF-LG AS. As opposed to patientswith classical LF-LG AS, those with paradoxical LF-LGAS have a preserved LVEF. This entity is defined as anLVEF $50%, the presence of a low flow (stroke vol-ume index <35 ml/m2), an AVA #1.0 cm2, an indexedAVA #0.6 cm2/m2, and a mean gradient <40 mm Hg(Figures 1 and 3). The paradoxical LF-LG pattern is

observed in 5% to 15% of AS patients and is moreprevalent in women and elderly individuals. Typi-cally, these patients have a small LV cavity withpronounced LV concentric remodeling and a restric-tive physiology, leading to a decrease in stroke vol-ume despite a preserved LVEF (Figure 3, OnlineVideos 3 and 4). Other factors may also lead to a lowflow state including significant mitral regurgitation,mitral stenosis, tricuspid regurgitation, and atrialfibrillation. The paradoxical LF-LG entity is identifiedas stage D3 in the guidelines, and AVR is indicated(Class IIa recommendation) if the patient is symp-tomatic and has true severe stenosis (Figure 1, CentralIllustration). In essence, classical LF-LG AS representsthe heart failure with reduced left ventricular ejectionfraction (HFrEF) form of AS, whereas paradoxical LF-LG AS represents the heart failure with preservedLVEF form of AS (Figures 2 and 3).

NORMAL-FLOW, LOW-GRADIENT AS. Normal-flow,low-gradient AS is defined as an LVEF $50%, a normalflow state (stroke volume index $35 ml/m2), anAVA #1.0 cm2, an indexed AVA #0.6 cm2/m2, and amean gradient <40 mmHg (Figures 1 and 4) (8–10). It isobserved in up to 25% of AS patients and is thus themost common subset of low-gradient AS. This entityand recommendations for therapeutic managementare not addressed in the valve guidelines (1,3). How-ever, recent studies suggest that approximately 50% ofthese patients may have a severe stenosis and thusmay benefit from AVR when symptomatic (Figure 1).The HAVEC (Heart Valve Clinic International database)group proposed to label this entity as “stage D4” (10).

MODERATE AS WITH LOW LVEF. Patients withconcordant AVA (>1.0 cm2) and gradient (<40mm Hg) findings have moderate AS, and in general,conservative management with serial follow-up isrecommended (Figures 1 and 5, Central Illustration).However, several studies have suggested that amoderate AS may have a detrimental effect on out-comes in patients with depressed LVEF, which raisesthe prorogating hypothesis that AVR may be benefi-cial in such patients (11,12). This hypothesis iscurrently being tested in the TAVR UNLOAD (Trans-catheter Aortic Valve Replacement to UNload the Leftventricle in patients with ADvanced heart failure)trial where patients with HFrEF and moderate ASconfirmed by resting and/or dobutamine stressechocardiography are randomized to optimized heartfailure therapy alone versus optimized heart failuretherapy plus transcatheter AVR (NCT02661451) (13).The moderate AS, low-LVEF entity could be labeled as“stage B2” by analogy with stages C2 and D2 (Figures 1and 5).

http://jaccimage.acc.org/video/2017/1551_VID1.avi






https://clinicaltrials.gov/ct2/show/NCT02661451

FIGURE 1 Classification and Characterization of the Different Types of AS According to AVA, Gradient, LVEF, and Flow

The classification of types of AS, including only the categories associated with symptoms and/or depressed LVEF. It does not include stage C1 (i.e., patients with high-

gradient AS, no symptoms, and preserved LVEF). Question mark indicates stage labels or indications for AVR that are proposed by the authors but are not included in

the guidelines and will need to be further tested and validated. AS ¼ aortic stenosis; AVA ¼ aortic valve area; AVAi ¼ indexed aortic valve area; AVR ¼ aortic valve

replacement; LVEF ¼ left ventricular ejection fraction; MG ¼ mean gradient; RCT ¼ randomized controlled trial; SVi ¼ stroke volume index.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 0 , N O . 2 , 2 0 1 7 Clavel et al.F E B R U A R Y 2 0 1 7 : 1 8 5 – 2 0 2 Cardiac Imaging for Assessing Low-Gradient AS

187

TECHNICAL REASONS LEADING TO

LOW-GRADIENT AS

The first step when confronted with a patient with adiscordant AVA (small) and gradient (low) situation isto ascertain the accuracy of the Doppler-echocardiographic measurements of stroke volume,AVA, and gradient. Measurement errors may lead tothe erroneous conclusion that the patient has a low-gradient AS entity, and can result in either an un-derestimation of severe AS or an overestimation ofmoderate AS.

TECHNICAL ISSUES UNDERESTIMATING AS SEVERITY.

Accurate measurement of velocity and gradientsby Doppler-echocardiography requires optimalalignment of the continuous-wave Doppler beam withthe direction of the aortic flow jet. If not, this mayresult in an underestimation of the gradient, anoverestimation of the AVA, and an underestimationof AS severity. Apical windows are usually the mostintensively interrogated for aortic jet velocity mea-surements using continuous-wave Doppler; however,the apical window reveals the maximum velocity in

only 40% of patients, whereas the maximum velocityis obtained from the right parasternal window in up to50% of patients (14). A multiwindow interrogationthat includes not only the apical window, but also theright parasternal and suprasternal windows, is thusessential to obtain an accurate measure of the veloc-ity and gradient.

TECHNICAL ISSUES OVERESTIMATING AS SEVERITY. Themost common technical pitfall that may lead to anerroneous diagnosis of low-flow state and over-estimation of AS severity is an underestimation of theleft ventricular outflow tract (LVOT) diameter mea-surement. The effective AVA is determined by thecontinuity equation method, where the numerator isthe stroke volume measured at the LVOT and thedenominator is the time-velocity integral (TVI) of theaortic transvalvular flow:

AVA ¼ LVOT area� LVOT TVIAortic valve TVI

Given that the LVOT diameter is squared in thecontinuity equation, a small error in this measure-ment may result in an important error in the

CENTRAL ILLUSTRATION Algorithm for the Management of Low-Gradient AS

Clavel, M.-A. et al. J Am Coll Cardiol Img. 2017;10(2):185–202.

This figure presents a 4-step algorithm for the diagnostic and therapeutic management of low-gradient AS. AVC ¼ aortic valve calcification; AVCd ¼ aortic valve

calcification density; AVR ¼ aortic valve replacement; CMR ¼ cardiac magnetic resonance; MDCT ¼ multidetector computed tomography; RCT ¼ randomized

controlled trial; TEE ¼ transesophageal echocardiography; TTE ¼ transthoracic echocardiography; other abbreviations as in Figures 1 to 3.



188


189

calculation of the stroke volume and AVA. Hence, anunderestimation of the LVOT diameter may lead tothe false conclusion that the patient has low-flow,low-gradient severe AS when, in fact, the patienthas normal-flow and/or moderate AS.

The LVOT dimension that is measured with stan-dard 2-dimensional (2D) echocardiography is theanteroposterior diameter at peak systole and thestroke volume is calculated by assuming that theLVOT cross-section is circular (Figure 6). However,the LVOT is a dynamic 3-dimensional (3D) structurethat is often elliptically shaped, with the ante-roposterior dimension representing the smaller minoraxis diameter, as compared with the generally largersagittal diameter. Hence, 2D echocardiography mayunderestimate the LVOT area compared with 3D im-aging modalities such as 3D echocardiography, mul-tidetector computed tomography (MDCT), or cardiacmagnetic resonance (CMR) (15–19). Some studiessuggest that the underestimation of the LVOT area by2D echocardiography using the circular assumptionmay be more important if the LVOT diameter ismeasured 5 to 10 mm below the aortic annuluscompared with at the aortic annulus (20).

To overcome the potential underestimation of theLVOT diameter and thus stroke volume and AVA by 2Dechocardiography, the use of a hybrid approach hasbeen suggested, where the LVOT area is measured byMDCT or 3D echocardiography (Figure 6) and the LVOTand aortic flow velocities are measured by Dopplerechocardiography (18,19). This hybrid approach isinteresting and may help to ascertain the accuracy offlow and AVA measurements. However, it is subject tosignificant pitfalls. First, MDCT has been shown tooverestimate the LVOT area compared with CMR im-aging or direct anatomic measurement at the time ofsurgery (21,22). Hence, the use of MDCT may result inan overestimation of “actual” effective AVA. Second,the AVA cut-point value generally used to define se-vere AS, 1.0 cm2, has been established and validated byoutcome studies in which AVA was measured bystandard Doppler-echocardiography (1,3). This cut-point cannot be directly extrapolated to the hybridmethods that systematically measure larger AVAscompared with Doppler-echocardiography. A recentstudy reported that AVA calculated by a hybridapproach (MDCT-Doppler) was not superior to AVAdetermined by standard Doppler-echocardiographyto predict transvalvular gradients or patient out-comes. Moreover, the best discriminative cut-pointof the hybrid AVA to predict mortality in patientswith AS under medical treatment was larger (i.e.,1.2 cm2) versus the Doppler-echocardiographic AVA(1.0 cm2) (19).

To optimize the accuracy of the measures of strokevolume and AVA, one may consider the followingsteps (Central Illustration):

Step 1. The clearest on-axis image of the LVOTproviding the largest LVOT diameter should be used(Figure 6). The 2009 European Association of Echo-cardiography/American Society of Echocardiographyguidelines suggest measuring the diameter and ve-locity 5 to 10 mm below the aortic annulus (2). How-ever, recent studies suggest measuring the LVOTdiameter inner-edge-to-inner-edge from the base ofthe right coronary cusp anteriorly to the commissureposteriorly, as originally described by Skjaerpe et al.(20,23) (Figure 6).

Step 2. The measures of LVOT diameter and strokevolume should be corroborated with other methods.LVOT diameter generally relates to body surface area,and 90% of patients have an LVOT diameter within2 mm of predicted for their body surface area usingthe following formula (24):

LVOTdiameter ¼ ð5:7� body surface areaÞ þ 12:1

If the measured diameter is more than 2 mmsmaller or larger than the predicted value, oneshould suspect a technical error and reassess the2D-echocardiographic LVOT measurement and/orcorroborate it with 3D imaging modalities (Figure 6).It should be noted that patients with a bicuspidvalve may have an LVOT diameter larger than thatpredicted by this formula (24). Moreover, in obesepatients, the formula may overestimate the actualLVOT diameter.

Another approach to corroborate the LVOT diam-eter measurement in a patient with low-gradient ASwith a small AVA is to calculate the Doppler velocityindex (i.e., the ratio of the LVOT to aortic flow time-velocity integrals) (25,26). If the index is <0.25, theresults are consistent with AVA and indirectlycorroborate the validity of the LVOT measurement(Figure 3). If the index is >0.25 and thus suggestive ofnonsevere AS, one should suspect an error in theLVOT diameter measurement.

Several 2D or 3D-echocardiographic methods canbe used to corroborate the stroke volume measure-ment (Central Illustration). The ASE guidelines do notrecommend using the standard Teichholz method tomeasure LV volumes and LVEF (27). However, amodified Teichholz method may be considered as afirst approach to rapidly corroborate the LVOT strokevolume (Figures 3 and 4). The LV end-diastolicdiameter should be measured apical to the septalbulge to obtain the largest diameter and avoid

FIGURE 2 Case of Classical Low-Flow, Low-Gradient AS With True-Severe Stenosis

An 83-year-old man with New York Heart Association functional class III dyspnea. LVEF was markedly depressed at rest but improved substantially with dobutamine stress

echocardiography (DSE) (A) (Online Videos 1 and 2). There was a significant increase in stroke volume (SV) and mean transvalvular flow rate (Qmean ¼ SV/LV ejection time),

but the increase in stroke volume was not sufficient to completely normalize flow, and the AVA/gradient discordance at rest persisted at DSE (B). The AVA projected at a

normal flow rate (250 ml/s) was thus calculated to confirm stenosis severity (C). First, the valve compliance (VC) is calculated by dividing the absolute increase in AVA by

the increase in Qmean during DSE: VC ¼ DAVA/DQmean. In this case, VC is equal to 0.015 cm2/100 ml/s. Then the projected AVA is calculated with the formula: projected

AVA¼ rest AVAþ VC� (250� Rest Qmean). The projected AVA is equal to 0.82 cm2 in this patient, thus confirming the presence of true-severe AS. The patient underwent

transcatheter AVR. Dob ¼ dobutamine; MG ¼ mean gradient; Qmean ¼ mean transvalvular flow rate; SV ¼ stroke volume; other abbreviations as in Figure 1.



190

potential underestimation of the LV end-diastolicvolume by the Teichholz formula. The end-diastolicvolume is then multiplied by the LVEF measuredby biplane Simpson method to estimate the LVOTstroke volume (28). The biplane Simpson methodgenerally underestimates stroke volume due tofrequent foreshortening of the LV cavity, and carefulconsideration of image quality is required whencorroborating data using this technique. If imagequality is adequate, 3D echocardiography can beused to estimate LV end-diastolic and -systolic vol-umes and thus derive stroke volume (Figure 3,Central Illustration). One of the advantages of 3Dechocardiography is that it does not rely ongeometrical assumptions (27). The 2D- or 3D-volu-metric methods measure the “total” LV strokevolume and should not be used to corroborate

LVOT stroke volume (i.e., forward stroke volume)in presence of more than moderate mitralregurgitation.Step 3. Continuity equation AVA may be corroboratedby other methods such as planimetry of the anatomicorifice using 2D or 3D transthoracic or trans-esophageal echocardiography (29,30). An AVA <1.0cm2 on planimetry confirms the presence of severeAS; however, an AVA slightly >1.0 cm2 does notdefinitively exclude severe AS, because the effectiveorifice area is usually slightly smaller than theanatomic AVA. As previously discussed, hybrid(MDCT-Doppler) method may be used to corroborateAS severity (18,19). However, a larger cut-point valuefor AVA (<1.2 cm2) should be used to define severe ASto avoid an over-reclassification of severe AS tomoderate AS (19).

http://jaccimage.acc.org/video/2017/1551_VID1.mp4

http://jaccimage.acc.org/video/2017/1551_VID2.mp4

FIGURE 3 Case of Paradoxical Low-Flow, Low-Gradient AS

An 82-year-old woman with New York Heart Association functional class III dyspnea and recent hospitalization for heart failure. This patient has a small LV cavity with

pronounced concentric remodeling, severe diastolic dysfunction, impaired global longitudinal strain (�13%), and dilated left atrium (A) (Online Video 3). The valve

appears to be severely thickened and calcified with restricted opening (Online Video 4). The stroke volume measured in the LV outflow tract is 53 ml (B and C), and this

patient has a low-flow state (stroke volume index <35 ml/m2). The LV end-diastolic diameter measured apical to the septal bulge (where the diameter is the largest)

was 42 mm. The total LV stroke volume estimated with modified Teichholz formula is 51 ml (LV end-diastolic volume by Teichholz: 79 ml � LVEF by Simpson: 65%),

which is consistent with the LVOT stroke volume. The total stroke volume measured by 3D echocardiography was 56 ml (not shown), also consistent with the LVOT

stroke volume. It is also important to perform multiwindow interrogation with continuous-wave Doppler (D and E). In this patient, a slightly higher velocity was

obtained at the right parasternal border window (E). In this patient, the AVA, AVAi, and DVI are in the severe range, but the mean gradient is low (26 mm Hg).

The presence of severe stenosis was corroborated by the presence of very high aortic valve calcium score at MDCT (see Figure 7, right panel). The patient underwent

transcatheter AVR. AVAi ¼ indexed aortic valve area; diam. ¼ diameter; DVI ¼ Doppler velocity index; LVEDV ¼ left ventricular end-diastolic volume; LVOT ¼ left

ventricular outflow tract; other abbreviations as in Figure 2.


191

PHYSIOLOGICAL REASONS LEADING TO

LOW-GRADIENT AS

LOW-FLOW STATE. The presence of a low-flow stateis one of the most frequent causes of low-gradient AS.Mean gradient and peak aortic jet velocity are highlydependent on the magnitude of transvalvular flow,and a small decrease in LVOT stroke volume and thustransvalvular flow rate may result in an importantdecrease in transvalvular gradient, a phenomenonknown as gradient “pseudo-normalization” (Figure 2).Although AVA is much less flow-dependent thangradient (31), a decrease in transvalvular flow is

nonetheless associated with a reduction in forcesapplied to the valve cusps, a decrease in valveopening and AVA, and a phenomenon known as“pseudo-severe” AVA (Figure 5). Hence, in the pres-ence of low flow, the gradient and peak aortic velocitymay underestimate AS severity, whereas AVA mayoverestimate AS severity. A low-flow state is definedin the guidelines as a stroke volume index <35 ml/m2

(1,3). The clinical relevance and prognostic value ofthe cut-point of 35 ml/m2 has been confirmed byseveral studies and meta-analyses (4,32–37).

A low-flow state may be related to depressed LVsystolic function, a condition named “classical” low-



FIGURE 4 Case of Normal-Flow, Low-Gradient AS

A 65-year-old symptomatic (New York Heart Association functional class II) man with AVA-gradient discordance, preserved LVEF, and normal flow (A to C). Both AVA and

AVAi are small, but the gradient is 32 mm Hg (D and E). The stroke volume estimated by the modified Teichholz method is similar to that obtained in the LVOT (A to C).

This patient had an aortic valve calcium score (2,430 AU) suggesting the presence of severe AS. The patient underwent surgical AVR. Abbreviations as in Figures 2 and 3.



192

flow AS (Figures 1, 2, and 5). Alternatively, a low-flowstate may be present in patients with preserved LVEF,a condition named “paradoxical” low-flow AS (Fig-ures 1 and 3). This low-flow state is predominantly aresult of pronounced LV concentric remodeling withadvanced diastolic dysfunction, impaired LV filling,and reduced longitudinal systolic function. Otherfactors may contribute to reduce LV outflow in thesetting of preserved LVEF including atrial fibrillation,mitral regurgitation, mitral stenosis, right ventriculardysfunction, tricuspid regurgitation, and constrictivepericarditis.

Whereas current guidelines use stroke volume in-dex to define the presence of a low-flow state, mea-sures that also account for the systolic ejection periodare physiologically more appropriate and may betterreflect the effect of flow on transvalvular gradient andAVA (38,39). The mean transvalvular flow rate iscalculated by dividing stroke volume by the LV ejec-tion time, and a value <200 ml/s is consistent with alow-flow state. In fact, stroke volume index and meanflow rate are complementary measures of LV outflowin the sense that the former is likely more important

from the standpoint of prognosis, whereas the latteris more relevant from the standpoint of diagnosis.

SMALL BODY SIZE. Body surface area is the mainphysiological factor determining the magnitude ofstroke volume and cardiac output in a normal subject.Given that a small body size is associated with a lowertransvalvular flow rate, it may lead to a discordantsmall AVA and a low-gradient situation. Indeed, asmall AVA in a small patient may actually correspondto moderate AS and be associated with a low gradient.In patients of small body size, it is thus recommendedto calculate the indexed AVA, and a value >0.6 cm2/m2

would rule out the presence of severe AS. Indexationof AVA for body surface area may, however, over-estimate the severity of AS in obese individuals. Twoapproaches may be considered to overcome this lim-itation: 1) use a lower cutoff value of indexed AVA(<0.5 cm2/m2) to define severe AS in patients with abody mass index $30 kg/m2, similar to what has beenproposed for defining prosthesis-patient mismatch(40); or 2) use other types of indexation such as AVAindexed to height (<0.45 cm2/m) (41). Of note, similar

FIGURE 5 Case of Classical Low-Flow, Low-Gradient AS With Pseudo-Severe Stenosis

A 73-year-old symptomatic (New York Heart Association functional class III) woman. In this patient, there is an improvement of LVEF and

complete normalization of flow rate with DSE. The peak stress mean gradient is <40 mm Hg and the AVA is >1.0 cm2, thus suggesting the

presence of pseudo-severe AS. The patient was managed conservatively with heart failure therapy. Abbreviations as in Figures 2 and 3.


193

limitations apply to the definition of low flow usingstroke volume indexed to body surface area.An alternative in obese patients is to use strokevolume indexed to a 2.04 power of height to definelow flow (<22 ml/m2.04) (42).

REDUCED ARTERIAL COMPLIANCE AND HYPERTENSION.

The vast majority of patients with calcific AS havereduced arterial compliance and systolic

hypertension (43,44). This may further increase LVafterload (already augmented because of AS) and leadto a decrease in LV outflow, thus contributing to thedevelopment of a low-gradient AS pattern (43,45–47).The total LV afterload resulting from the addition ofthe valvular and arterial loads may be estimated bythe valvuloarterial impedance (Zva), which is calcu-lated by dividing the sum of the mean gradient andsystolic blood pressure by the stroke volume index

FIGURE 6 Pitfalls in the Measurement of the LVOT Diameter

(A) MDCT image showing the elliptical shape of the LVOT with the anteroposterior diameter (green arrow) that is smaller than the sagittal

diameter (blue arrow). (B) Anatomic view of the aortic aspect of the aortic valve illustrating the optimal plane for 2-dimensional (2D)

echocardiographic measurement of the LVOT. The plane should bisect the right coronary cusp anteriorly and the commissure between the left

and noncoronary cusp posteriorly (green arrows). A plane bisecting a cusp posteriorly provides a truncated section of the LVOT and thus may

underestimate the LVOT diameter (red arrow). (C) The parasternal zoomed view providing the largest diameter of the LVOT should be used.

The LVOT diameter should preferably be measured at—or very close to—the aortic annulus (i.e., base of cusp/commissure) (green arrows)

rather than 5 to 10 mm below the annulus, as proposed in the European Association of Echocardiography/American Society of Echocardi-

ography guidelines (yellow arrows). (D) The LVOT area estimated by 2D echocardiography assuming a circular shape of the LVOT

underestimates the actual LVOT area (and thus the AVA) measured by 3D echocardiography by 16% in this case. Abbreviations as in

Figures 2 and 3.



194

(48). Typically, patients with LF-LG AS have abnor-mally high valvuloarterial impedance (>4.5 mm Hg/ml$m�2); if not, the low flow state is likely not relatedto afterload excess, but is rather an intrinsic cardio-myopathy such as ischemic disease.

Moreover, reduced arterial compliance leads to afaster arterial wave reflection from the periphery. Theearly reflection of the arterial wave at the end ofsystole may dampen the transvalvular gradients, in-dependent of transvalvular flow (49,50). This phe-nomenon may, at least in part, explain the small AVAand low gradient discordance observed in patientswith normal-flow, low-gradient AS (Figures 1 and 4)(46,50). In patients with low-gradient AS andconcomitant systolic hypertension, it is recom-mended to reassess the echocardiographic

parameters of AS severity, as well as the symptomaticstatus, after the normalization of blood pressure.

INCONSISTENCIES IN GUIDELINES CRITERIA

FOR SEVERE AS

The majority of patients with discordant grading atechocardiography or catheterization actually havenormal flow (i.e., stroke volume index $35 ml/m2)(Figures 1 and 4) (5,6,8). Besides reduced arterialcompliance, another cause of normal-flow, low-gradient AS is the inherent inconsistency, ormisalignment, of the AVA and gradient cut-points forthe definition of severe AS in the valve guidelines (6).Indeed, in a patient with a normal flow rate, an AVAof 1.0 cm2 does not correspond to a mean gradient of


195

40 mm Hg or a peak AS velocity of 4 m/s, but rathercorresponds to a mean gradient of 30 mm Hg and apeak velocity of 3.5 m/s (5,8). To overcome thispitfall, some investigators have suggested decreasingthe AVA cut-point for defining severe AS to 0.8 cm2.However, patients with an AVA between 0.8 and 1.0cm2, but with a low gradient, are at increased risk ofmortality if treated medically and receive an impor-tant survival benefit with AVR (5,35). Hence, the bestapproach, for now, appears to be to maintain thecurrent cut-points: AVA #1.0 cm2 as a sensitivecriteria and mean gradient $40 mm Hg as a specificcriteria. In symptomatic patients with discordantgrading and low-gradient AS, additional tests andparameters should be used to confirm the stenosisseverity.

MANAGEMENT OF LOW-GRADIENT AS

Patients with LF-LG AS, especially those with classicalLF-LG, have worse outcomes following surgical ortranscatheter AVR compared with patients with high-gradient AS (33,51–53). Nonetheless, several studiesand meta-analyses have reported that patients withclassical LF-LG; paradoxical LF-LG; or normal-flow,low-gradient severe AS display a significant survivalbenefit with AVR compared with conservative man-agement (32,35,54–60). Hence, the crucial step todetermine the need for AVR in a symptomatic patientwith low-gradient AS is to confirm the presence of asevere stenosis, which may be particularly chal-lenging in the context of low-gradient AS (CentralIllustration). In these patients, there is frequently un-certainty with respect to the “actual” severity of ASgiven that some echocardiographic or catheterizationparameters (i.e., AVA, Doppler velocity index) suggesta severe stenosis, whereas others (gradients and peakvelocity) suggest nonsevere AS.

CONFIRMATION OF AS SEVERITY TO DETERMINE

THE NEED FOR AVR. Among patients with smallAVA, low-gradient AS, 30% to 50% have nonsevereAS; these patients should probably be treatedconservatively because they would not be expectedto receive a benefit from AVR (Figure 1) (7,11,50,61,62).Stress echocardiography and/or MDCT may be used toidentify those patients with true-severe AS whoshould be referred for valve replacement (CentralIllustration).

Low-dose dobutamine stress echocardiography. Inpatients with classical LF-LG AS, it is difficult todifferentiate true-severe AS, which generally requiresAVR, from pseudo-severe AS, in which there isincomplete opening of a valve that is only moderately

or mildly stenotic. Low-dose (up to 20 mg/kg/min)dobutamine stress echocardiography (or dobutaminestress cardiac catheterization) is useful to distinguishtrue-severe from pseudo-severe stenosis in thesepatients with a low LVEF (Figures 2 and 5) (63).Various parameters have been proposed to definetrue-severe AS (Table 1). However, the guidelinessuggest a peak stress mean gradient $40 mm Hg inthe presence of an AVA #1.0 cm2 as the main criteriato confirm the presence of severe AS in patients withclassical LF-LG AS (1,3).

The main limitation of this test is that the flowresponse to dobutamine and thus the magnitude oftransvalvular flow rate at peak stress varies exten-sively from one patient to the other. Most patientsnormalize or even over-normalize their flow rate, andif they have true-severe AS, they will generallydisplay a high gradient at peak stress. On the otherhand, many patients exhibit a significant increase inflow with dobutamine; however, it is not sufficient toreach the normal flow range, and a discordant smallAVA, low-gradient pattern persists at the end of thedobutamine stress test (Figure 2, Online Videos 1 and2). In these patients, it is useful to calculate the pro-jected AVA at a normal flow rate using the methodpresented in Figure 2. On the basis of the AVA-flowbehavior, this parameter projects what the AVAwould be at a standardized normal flow rate of 250ml/s. This parameter, standardized for flow, has beenshown to better predict the actual hemodynamicseverity of the valve stenosis and the clinical outcomeof patients with classical or paradoxical LF-LG AS, ascompared with standard stress echocardiographyparameters (11,61,64). A projected AVA <1.0 cm2

confirms the presence of true severe AS (Figure 2,Table 1). It should be noted that many patients do nothave an increase in stroke volume with dobutamineinfusion, and indeed, may have a decrease in strokevolume; however, transvalvular flow rate will almostalways increase because of a shortening of the ejec-tion time. Importantly, a minimum 15%(ideally $20%) increase in flow rate is required toobtain a reliable estimate of the projected AVA (11). Inpatients with no or minimal increase in flow rate,dobutamine stress echocardiography remains incon-clusive, and other tests such as aortic valve calciumscoring may be used to corroborate AS severity asdescribed in the following section (Figure 7).

Although the vast majority of patients with lowLVEF and low gradient (i.e. classical LF-LG) have alow-flow state at rest, some may nonetheless have anormal mean transvalvular flow rate (>200 ml/s) (39).These patients generally have enlarged LV cavitieswith moderately depressed LVEF and no or minimal



TABLE 1 Multimodality Imaging for Identification of Low Flow and Confirmation of Stenosis Severity in Low-Gradient AS

Imaging Modality Imaging Parameter and Criteria Advantages Limitations

Low Flow

Doppler-echocardiography Stroke volume index<35 ml/m2*

Good marker of LV pump functionand prognosis

Does not account for the effect of ejectionduration on the gradient. May overestimatethe prevalence of low-flow state inobese patients.

Mean transvalvular flow rate<200 ml/s

Better determinant of gradientthan stroke volume index

Potentially inferior to stroke volume index topredict prognosis.

Severe AS

Doppler-echocardiography Peak aortic jet velocity $4 m/s*Mean gradient $40 mm Hg*

Less subject to measurement errorthan the AVA

Highly flow-dependent. May underestimateAS severity in low-flow states.

AVA <1.0 cm2*Indexed AVA <0.6 cm2/m2*(<0.5 cm2/m2 if BMI $30 kg/m2)Doppler velocity index <0.25

Less flow-dependent than thegradient or peak velocity

Subject to measurement error. Mayoverestimate AS severity in low-flowstates.

Severe valve leaflet thickening andcalcification. Severely reducedleaflet mobility.

Reduced leaflet mobility may overestimateAS severity in low-flow states.

Often difficult to assess by TTE; betterassessed by TEE.

Dobutamine-stressechocardiography

Peak aortic jet velocity $4 m/sMean gradient $40 mm Hg*Mean gradient $30 mm Hg

Less subject to measurement errorthan the AVA

Highly flow-dependent. May underestimateAS severity if flow rate remains belownormal with dobutamine stress oroverestimate AS severity if supra-normalresponse to dobutamine stress.

AVA<1.0 cm2*AVA<1.2 cm2

Doppler velocity index <0.25Increase in AVA<0.3 cm2

Less flow-dependent than thegradient or peak velocity

Subject to measurement error. Mayoverestimate AS severity if flow rateremains below normal with dobutaminestress

Projected AVA <1.0 cm2

Indexed Projected AVA <0.55 cm2/m2Standardized for flow Subject to measurement error. Not

measurable if increase in flow rate <15%with dobutamine stress.

MDCT Aortic valve calcium score>2,000 AU in men>1,200 AU in womenAortic valve calcium density>500 AU/cm2 in men>300 AU/ cm2 in women

Highly accurate and reproducible.Independent of flow andhemodynamics. Does not requireadministration of stress orcontrast agent.

Reflects anatomic rather than hemodynamicseverity. Does not take into accountvalvular fibrosis and therefore mayunderestimate AS severity.

*Criteria and parameters recommended in the guidelines.

AS ¼ aortic stenosis; AVA ¼ aortic valve area; BMI ¼ body mass index; LV ¼ left ventricular; MDCT ¼ multidetector computed tomography; TEE ¼ transesophageal echocardiography; TTE ¼ transthoracicechocardiography.



196

mitral regurgitation, and are therefore able togenerate a normal stroke volume. In these patients,the AVA at rest may actually be a good predictor ofthe presence of true severe stenosis, and dobutaminestress may not be required (Table 1) (39).

Dobutamine stress echocardiography (symptom-atic patients) or exercise stress echocardiography(patients with no or equivocal symptoms) may also beuseful to confirm the presence of severe AS in pa-tients with paradoxical LF-LG or in those withnormal-flow, low-gradient AS (61). However, animportant proportion of patients with paradoxical LF-LG AS have small thick left ventricles with restrictivephysiology, which may preclude a significant increasein flow during dobutamine stress and can lead to sideeffects (LVOT dynamic obstruction, hypotension,etc.) (Figure 3, Online Videos 3 and 4). Hence,assessment of the valve morphology by transthoracicor transesophageal echocardiography and

quantitation of aortic valve calcification by MDCTmay be the preferred approach to corroborate ASseverity in these patients (Figure 7). In patients withnormal-flow, low-gradient AS, the stroke volume is bydefinition normal at rest, and the utility of stressechocardiography is thus limited in these patients.However, some of these patients may actually have alow-flow state with a reduced mean transvalvularflow rate as a result of prolongation of the ejectiontime, and therefore may benefit from dobutaminestress testing. On the other hand, some patients withclassical or paradoxical LF-LG may have a normalmean flow rate (>200 ml/s) despite a reduced LVEFand/or stroke volume index if their ejection time isshort. In such cases, the resting AVA measured undernormal flow rate conditions likely reflects the trueseverity of AS, and DSE is not needed to confirmstenosis severity (39). The documentation of mark-edly abnormal valve leaflet morphology/mobility and



FIGURE 7 Quantitation of Aortic Valve Calcification by MDCT to Differentiate True Versus Pseudo-Severe Stenosis in Low-Gradient AS

Abbreviations as in Figures 1 to 3.


197

the presence of severe valve calcification at MDCTmight be considered to trigger AVR in symptomaticpatients with normal-flow, low-gradient AS (Figure 7).However, further studies are needed to develop andvalidate imaging markers of AS severity in patientswith low-gradient AS, and particularly those withnormal-flow and those with paradoxical low-flow AS.Assessment of valve leaflet morphology by transthoracicor transesophageal echocardiography. Transthoracic ortransesophageal 2D or 3D echocardiography can beused to assess the degree of valve leaflet thickeningand calcification and can therefore corroborate the“anatomic” severity of AS (Figure 3, Online Video 4).The limitations of this method are: 1) it is subject tointerobserver and intraobserver variability; and 2) itis semiquantitative with limited gradations (mild,moderate, and severe calcification) (65). Hence, thisassessment may need to be confirmed by quantitativemethods such as MDCT (Figure 7). The presence ofreduced leaflet mobility at echocardiography is also acriterion supporting the presence of severe AS.However, this feature, as with AVA, may overestimateseverity in the presence of low flow (pseudo-severephenomenon due to incomplete opening of thevalve).

Quantitation of aortic valve calcification by MDCT. MDCTis a highly accurate and reproducible method toquantitate aortic valve calcification and it has theadvantage of being independent of hemodynamics/flow and does not require the administration of anystress or contrast agents. This method is thus appli-cable to all patients with low-gradient AS. Aorticvalve calcium scoring by MDCT is useful to corrobo-rate AS severity in patients with low-gradient ASwhen dobutamine stress echocardiography is notindicated, not feasible, or inconclusive. Hence, MDCTis particularly helpful in patients with classical LF-LGAS with no or minimal increase in flow with dobut-amine stress; in patients with paradoxical LF-LG; andin those patients with normal-flow, low-gradient AS(Figures 2, 3, 4, and 7). For the quantitation of calci-fication, a noncontrast MDCT scan during trainedend-inspiration breath-hold is performed, and theradiation exposure for such an examination is <3mSV. The amount of calcification in the region of theaortic valve is quantitated using the modified Agat-ston method, in which calcification is defined as 4adjacent pixels with a density >130 Hounsfield units(66). Importantly, MDCT assesses the “anatomic”rather than the “hemodynamic” severity of AS.




198

Nonetheless, the aortic valve calcium score measuredby MDCT strongly correlates with hemodynamicseverity, the progression rate, and the clinical out-comes of AS patients (19,50,67,68). It is important tohighlight that women reach the same hemodynamicseverity as men with less calcification, and therefore,different cut-point values of valve calcification scoreneed to be applied to differentiate severe versusnonsevere AS in men ($2,000 AU) compared withwomen ($1,200 AU) (Table 1, Figure 7) (50,69). Thesame approach should be applied when using cut-point values for aortic valve calcium density (i.e.,calcium score divided by aortic annulus area): $500AU/cm2 in men versus $300 AU/cm2 in women.

MDCT has important pitfalls that should be appre-ciated. In particular, this method does not capturevalvular fibrosis, which also contributes to the hemo-dynamic severity of the valve stenosis. This issue may,at least in part, explain the discrepancy between menand women with regard to the cut-point values foraortic valve calcium score and density to identify se-vere AS. Recent studies suggest that for a given degreeof hemodynamic severity, women have less valvularcalcification butmorefibrosis comparedwithmen (70).Furthermore, younger patients with bicuspid valvesmay have hemodynamically severe AS with no or littlevalve calcification (71). However, these patients rarelyexhibit a low-gradient AS pattern.Indication for AVR in low-gradient AS. In patients withlow-gradient AS who have symptoms and/orLVEF <50%, it is essential to confirm the stenosisseverity (Central Illustration). AVR should be consid-ered (Class IIa) in those patients with evidence oftrue-severe AS on the basis of the assessment of theaortic valve leaflet morphology/mobility by trans-thoracic or transesophageal echocardiography,confirmation of hemodynamic severity by dobut-amine stress echocardiography, and/or quantitationof aortic valve calcification by MDCT (Table 1). Pa-tients with nonsevere AS on the basis of dobutaminestress echocardiography or MDCT should a priori betreated conservatively (Figures 5 and 6) (72). Howev-er, nonsevere AS does not imply the absence of AS,and most of these patients have moderate ormoderate-to-severe AS. Therefore, patients shouldreceive close clinical and echocardiographic surveil-lance to evaluate for both disease progression andprogressive symptoms. In patients with a reducedLVEF, guideline-based heart failure therapies shouldbe optimized (72). Although moderate AS may be welltolerated in patients with preserved LV systolicfunction, it may represent a severe hemodynamicburden and have a detrimental effect on outcomes inpatients with a depressed LVEF (11,12). In the TOPAS

(Truly or Pseudo Severe Aortic Stenosis) study, thecut-point for the projected AVA that best predictedmortality in patients with classical LF-LG AS treatedmedically was >1.2 cm2, which is larger than the cut-point traditionally used to define severe AS (11).These findings raise the hypothesis that AVR might bebeneficial in patients with moderate AS and HFrEF(Figures 1 and 6, Central Illustration). The TAVR UN-LOAD trial may provide guidance in the managementof such patients (13).

INDIVIDUALIZED RISK STRATIFICATION TO SELECT

THE TYPE OF AVR. Once the indication for AVR isestablished, the next step is to decide on the best typeand access for AVR in the individual patient (CentralIllustration). The choice between surgical versustranscatheter AVR is essentially determined by thepredicted surgical risk that is assessed by combiningthe Society of Thoracic Surgeons risk estimate, pa-tient’s frailty, major organ system dysfunction, andprocedure-specific impediments (1). TranscatheterAVR is recommended in patients who are at prohibi-tive surgical risk (Society of Thoracic Surgeonsscore $10%) and a predicted post-TAVR survival >1year. Transcatheter AVR is a reasonable alternative tosurgical AVR in patients at high (>8%) or intermedi-ate (>4%) risk for surgical AVR. Besides the surgicalrisk scores, it is also important to consider the type oflow-gradient AS to select the type of AVR (CentralIllustration).

Classical low-flow, low-gradient AS. Depressed LVEF is awell-known and powerful risk factor for operativemortality following surgical AVR. In contrast, there isno or only a weak association between LVEF andprocedural outcomes following transcatheter AVR(32,33,73). Nonrandomized studies have also sug-gested that transcatheter AVR is associated withbetter and faster recovery of LV systolic function andbetter survival compared with surgical AVR (53,74).Although randomized studies such as the PARTNER(Placement of Aortic Transcatheter Valves) trialshowed no difference in outcome between trans-catheter versus surgical AVR, patients with very lowLVEF as well as those with no flow reserve ondobutamine stress were excluded (32,75). Theabsence of flow reserve, defined as an increase inLVOT stroke volume <20% during dobutamine stressechocardiography or catheterization, is a strong pre-dictor of high/extreme surgical risk for AVR in pa-tients with classical LF-LG AS (55). However, a recentstudy from a large, multicenter, international registryrevealed that patients with classical LF-LG AS haveexcellent procedural and 1-year outcomes withtranscatheter AVR (76). Furthermore, the absence of


199

flow reserve had no effect on the 30-day or 1-yearmortality. Transcatheter AVR with transapical accesshas been associated with more important myocardialinjury and less recovery of LVEF compared withtransfemoral or transaortic access and should prob-ably be avoided in patients with classical LF-LG AS(77). These findings in their entirety would suggestthat transcatheter AVR via a transfemoral approachmay be superior to surgical or transapical AVR inpatients with classical LF-LG AS, especially in thosepatients with no flow reserve on dobutamine stress(Central Illustration).

Coronary artery disease is highly prevalent amongpatients with classical LF-LG AS, and the presence ofsevere coronary artery disease (SYNTAX score >22) isindependently associated with reduced survivalfollowing AVR (78). Furthermore, patients with sig-nificant coronary artery disease and incompleterevascularization have worse outcomes after AVR.Therefore, revascularization strategies are an impor-tant component of the management of patients withclassical LF-LG AS who are undergoing surgical ortranscatheter AVR.Paradoxical LF-LG AS. Because of their particular clin-ical, anatomic, and hemodynamic features (higherprevalence of women, small LV cavity with restrictivephysiology, small aortic annulus, and so on), patientswith paradoxical LF-LG AS may be at higher surgicalrisk compared with patients with high-gradient AS(Figure 3) (52). Furthermore, they may be more proneto develop prosthesis-patient mismatch followingsurgical AVR, which may negatively affect their clin-ical outcome (79). Compared with surgical AVR,transcatheter AVR is associated with less frequentand less severe prosthesis-patient mismatch, partic-ularly in the subset of patients with a small aorticannulus (80,81). On the other hand, transcatheterAVR is associated with a higher prevalence of para-valvular regurgitation compared with surgical AVR,and even mild aortic regurgitation may be potentiallydetrimental in patients with paradoxical LF-LG AS.The post hoc analysis of the PARTNER 1A trial none-theless suggests that transcatheter AVR may be su-perior to surgical AVR in these patients (32). Furtherstudies are needed to confirm which treatment mo-dality is best in these patients.Normal-flow, low-gradient AS. In contrast to patientswith classical or paradoxical LF-LG AS, those withnormal-flow, low-gradient AS do not appear to be atincreased surgical risk. Hence, the strategy for thesepatients should be to select surgical or transcatheterAVR depending on the comorbidities and expectedsurgical risk as recommended for patients with high-gradient AS (Central Illustration). However, there is

no specific recommendation for patients with normal-flow, low-gradient AS in the guidelines, and furtherstudies are needed to determine the best timing andtype of AVR for these patients.Futility of AVR. Although the vast majority of symp-tomatic patients with low-gradient severe AS benefitfrom AVR in terms of longevity and quality of life,unfortunately, some have poor clinical and functionaloutcomes despite successful intervention (82). Animportant research priority in this field is to developand validate imaging and blood biomarkers to predicttreatment futility, that is, identify patients who areunlikely to benefit from AVR. Multimodality imagingmay be helpful in this regard. Very low baseline meangradient (<20 mm Hg), moderate/severe mitralregurgitation, severe right ventricular dysfunction,and severe tricuspid regurgitation are among theimaging factors that have been associated with anincreased risk of poor outcome and lack of functionalimprovement following AVR in patients with low-gradient AS (73,83–87). Quantitation of myocardialfibrosis by CMR is also a very promising approach toidentify patients with low-gradient AS who maypotentially have irreversible myocardial impairmentand thus a low likelihood to benefit from AVR (88).Further studies are needed to determine which CMRmethods, parameters, and cut-point values should beapplied to identify the presence of extensive irre-versible fibrosis that would potentially compromisethe utility of AVR. An approach on the basis ofmultiple complementary blood biomarkers of car-diovascular stress may be helpful to predict treatmentfutility in low-gradient AS (89). Pending furtherstudies, the presence of these emerging biomarkers ofhigher risk of treatment futility should not be used atthis time to preclude the consideration of AVR in low-gradient AS patients with evidence of severe AS.Indeed, such patients often have an even worseprognosis with conservative management.

CONCLUSIONS

Low-gradient AS is found in up to 40% of AS pa-tients with a small AVA. Multimodality imaging isessential for optimal diagnosis, risk stratification,and therapeutic management of this entity. Doppler-echocardiography is critical to determine the type oflow-gradient AS: classical low-flow, paradoxical low-flow, or normal-flow AS (Central Illustration). Hybridmethods fusing MDCT and Doppler imaging may beused to corroborate the measurement of flow andAVA, but larger cut-point values of stroke volumeindex and AVA should be applied to define a low-flow state and severe AS, respectively. Patients



200

with LF-LG AS generally have a worse prognosiscompared with patients with high-gradient or withnormal-flow, low-gradient AS. Patients with low-gradient AS and evidence of severe AS benefit fromAVR; however, confirmation of the presence ofsevere AS is challenging in these patients andrequires a multimodality imaging approachincluding dobutamine stress echocardiography andaortic valve calcium scoring by MDCT. TranscatheterAVR using a transfemoral approach may be superiorto surgical AVR in patients with LF-LG AS. Further

studies are needed to confirm the best type of AVRfor each category of low-gradient AS and to identifypatients with low-gradient AS in whom AVR is likelyto be futile.

ADDRESS FOR CORRESPONDENCE: Dr. PhilippePibarot, Institut Universitaire de Cardiologie et dePneumologie de Québec (Quebec Heart and LungInstitute), 2725 Chemin Sainte-Foy, #A-2075, Québec,Québec, Canada, G1V 4G5. E-mail: [email protected].

RE F E RENCE S

1. Nishimura RA, Otto CM, Bonow RO, et al. 2014AHA/ACC guideline for the management of pa-tients with valvular heart disease: a report of theAmerican College of Cardiology/American HeartAssociation Task Force on Practice Guidelines.J Am Coll Cardiol 2014;63:e57–185.

2. Baumgartner H, Hung J, Bermejo J, et al.Echocardiographic assessment of valve stenosis:EAE/ASE recommendations for clinical practice.J Am Soc Echocardiogr 2009;22:1–23.

3. Vahanian A, Alfieri O, Andreotti F, et al. Guide-lines on the management of valvular heart disease(version 2012). Eur Heart J 2012;33:2451–96.

4. Hachicha Z, Dumesnil JG, Bogaty P, Pibarot P.Paradoxical low flow, low gradient severe aorticstenosis despite preserved ejection fraction isassociated with higher afterload and reducedsurvival. Circulation 2007;115:2856–64.

5. Berthelot-Richer M, Pibarot P, Capoulade R,et al. Discordant grading of aortic stenosisseverity: echocardiographic predictors of survivalbenefit associated with aortic valve replacement.J Am Coll Cardiol Img 2016;9:797–805.

6. Minners J, Allgeier M, Gohlke-Baerwolf C,Kienzle RP, Neumann FJ, Jander N. Inconsistentgrading of aortic valve stenosis by current guide-lines: haemodynamic studies in patients withapparently normal left ventricular function. Heart2010;96:1463–8.

7. Clavel MA, Fuchs C, Burwash IG, et al. Pre-dictors of outcomes in low-flow, low-gradientaortic stenosis: results of the multicenter TOPASStudy. Circulation 2008;118:S234–42.

8. Dumesnil JG, Pibarot P, Carabello B. Paradoxi-cal low flow and/or low gradient severe aorticstenosis despite preserved left ventricular ejectionfraction: implications for diagnosis and treatment.Eur Heart J 2010;31:281–9.

9. Lancellotti P, Magne J, Donal E, et al. Clinicaloutcome in asymptomatic severe aortic stenosis:insights from the new proposed aortic stenosisgrading classification. J Am Coll Cardiol 2012;59:235–43.

10. Dulgheru R, Pibarot P, Sengupta PP, et al.Multimodality imaging strategies for theassessment of aortic stenosis: viewpoint of theHeart Valve Clinic International database(HAVEC) group. Circ Cardiovasc Imaging 2016;9:e004352.

11. Clavel MA, Burwash IG, Mundigler G, et al.Validation of conventional and simplified methodsto calculate projected valve area at normal flowrate in patients with low flow, low gradient aorticstenosis: the multicenter TOPAS (True or PseudoSevere Aortic Stenosis) study. J Am Soc Echo-cardiogr 2010;23:380–6.

12. Samad Z, Vora AN, Dunning A, et al. Aorticvalve surgery and survival in patients with mod-erate or severe aortic stenosis and left ventriculardysfunction. Eur Heart J 2016;37:2276–86.

13. Spitzer E, Van Mieghem NM, Pibarot P, et al.Rationale and design of the Transcatheter AorticValve Replacement to UNload the Left ventricle inpatients with ADvanced heart failure (TAVR UN-LOAD) trial. Am Heart J 2016;182:80–8.

14. Thaden JJ, Nkomo VT, Lee KJ, Oh JK. Dopplerimaging in aortic stenosis: the importance of thenonapical imaging windows to determine severityin a contemporary cohort. J Am Soc Echocardiogr2015;28:780–5.

15. Jilaihawi H, Doctor N, Kashif M, et al. Aorticannular sizing for transcatheter aortic valvereplacement using cross-sectional 2-dimensionaltransesophageal echocardiography. J Am CollCardiol 2013;61:909–16.

16. Chin CW, Khaw HJ, Luo E, et al. Echocardiog-raphy underestimates stroke volume and aorticvalve area: implications for patients with small-area low-gradient aortic stenosis. Can J Cardiol2014;30:1064–72.

17. Khalique OK, Kodali SK, Paradis JM, et al.Aortic annular sizing using a novel 3-dimensionalechocardiographic method: use and comparisonwith cardiac computed tomography. Circ Car-diovasc Imaging 2014;7:155–63.

18. Kamperidis V, van Rosendael PJ, Katsanos S,et al. Low gradient severe aortic stenosis withpreserved ejection fraction: reclassification ofseverity by fusion of Doppler and computedtomographic data. Eur Heart J 2015;36:2087–96.

19. Clavel MA, Malouf J, Messika-Zeitoun D,Araoz PA, Michelena HI, Enriquez-Sarano M. Aorticvalve area calculation in aortic stenosis by CT andDoppler echocardiography. J Am Coll Cardiol Img2015;8:248–57.

20. LaBounty TM, Miyasaka R, Chetcuti S, et al.Annulus instead of LVOT diameter improvesagreement between echocardiography effective

orifice area and invasive aortic valve area. J AmColl Cardiol Img 2014;7:1065–6.

21. Tsang W, Bateman MG, Weinert L, et al. Ac-curacy of aortic annular measurements obtainedfrom three-dimensional echocardiography, CT andMRI: human in vitro and in vivo studies. Heart2012;98:1146–52.

22. Wang H, Hanna JM, Ganapathi A, et al. Com-parison of aortic annulus size by transesophagealechocardiography and computed tomographyangiography with direct surgical measurement.Am J Cardiol 2015;115:1568–73.

23. Skjaerpe T, Hegrenaes L, Hatle L. Noninvasiveestimation of valve area in patients with aorticstenosis by Doppler ultrasound and two-dimensional echocardiography. Circulation 1985;72:810–8.

24. Leye M, Brochet E, Lepage L, et al. Size-adjusted left ventricular outflow tract diameterreference values: a safeguard for the evaluation ofthe severity of aortic stenosis. J Am Soc Echo-cardiogr 2009;22:445–51.

25. Jander N, Hochholzer W, Kaufmann BA, et al.Velocity ratio predicts outcomes in patients withlow gradient severe aortic stenosis and preservedEF. Heart 2014;100:1946–53.

26. Michelena HI, Margaryan E, Miller FA, et al.Inconsistent echocardiographic grading of aorticstenosis: is the left ventricular outflow tractimportant? Heart 2013;99:921–31.

27. Lang RM, Badano LP, Mor-Avi V, et al. Rec-ommendations for cardiac chamber quantificationby echocardiography in adults: an update from theAmerican Society of Echocardiography and theEuropean Association of Cardiovascular Imaging.J Am Soc Echocardiogr 2015;28:1–39.

28. Dumesnil JG, Dion D, Yvorchuk K, Davies RA,Chan K. A new, simple and accurate method fordetermining ejection fraction by Doppler echo-cardiography. Can J Cardiol 1995;11:1007–14.

29. Ng AC, Delgado V, van der Kley F, et al.Comparison of aortic root dimensions and geom-etries before and after transcatheter aortic valveimplantation by 2- and 3-dimensional trans-esophageal echocardiography and multislicecomputed tomography. Circ Cardiovasc Imaging2010;3:94–102.

30. Nakai H, Takeuchi M, Yoshitani H, Kaku K,Haruki N, Otsuji Y. Pitfalls of anatomical aortic

mailto:[email protected]

mailto:[email protected]

http://refhub.elsevier.com/S1936-878X(17)30004-9/sref1




















































































































































201

valve area measurements using two-dimensionaltransoesophageal echocardiography and the po-tential of three-dimensional transoesophagealechocardiography. Eur J Echocardiogr 2010;11:369–76.

31. Burwash IG, Hay KM, Chan KL. Hemodynamicstability of valve area, valve resistance and strokework loss in aortic stenosis: a comparative anal-ysis. J Am Soc Echocardiogr 2002;15:814–22.

32. Herrmann HC, Pibarot P, Hueter I, et al. Pre-dictors of mortality and outcomes of therapy inlow flow severe aortic stenosis: a PARTNER trialanalysis. Circulation 2013;127:2316–26.

33. Le Ven F, Freeman M, Webb J, et al. Impact oflow flow on the outcome of high risk patientsundergoing transcatheter aortic valve replace-ment. J Am Coll Cardiol 2013;62:782–8.

34. Capoulade R, Le Ven F, Clavel MA, et al.Echocardiographic predictors of outcomes inadults with aortic stenosis. Heart 2016;102:934–42.

35. Dayan V, Vignolo G, Magne J, Clavel MA,Mohty D, Pibarot P. Outcome and impact of aorticvalve replacement in patients with preserved LVejection fraction and low gradient aortic stenosis:a meta-analysis. J Am Coll Cardiol 2015;66:2594–603.

36. Eleid MF, Sorajja P, Michelena HI, Malouf JF,Scott CG, Pellikka PA. Survival by stroke volumeindex in patients with low-gradient normal EFsevere aortic stenosis. Heart 2015;101:23–9.

37. Eleid MF, Goel K, Murad MH, et al. Meta-analysis of the prognostic impact of strokevolume, gradient, and ejection fraction aftertranscatheter aortic valve implantation. Am JCardiol 2015;116:989–94.

38. Kadem L, Pibarot P, Dumesnil JG, et al. Inde-pendent contribution of the left ventricular ejec-tion time to the mean gradient in aortic stenosis.J Heart Valve Dis 2002;11:615–23.

39. Chahal NS, Drakopoulou M, Gonzalez-Gonzalez AM, Manivarmane R, Khattar R, Senior R.Resting aortic valve area at normal transaorticflow rate reflects true valve area in suspected lowgradient severe aortic stenosis. J Am Coll CardiolImg 2015;8:1133–9.

40. Kappetein AP, Head SJ, Généreux P, et al.Updated standardized endpoint definitions fortranscatheter aortic valve implantation: the ValveAcademic Research Consortium-2 consensusdocument. Eur J Cardiothorac Surg 2012;42:S45–60.

41. Tribouilloy C, Bohbot Y, Maréchaux S, et al.Outcome implication of aortic valve area normal-ized to body size in asymptomatic aortic stenosis.Circ Cardiovasc Imaging 2016;9:e005121.

42. Cramariuc D, Cioffi G, Rieck AE, et al. Low-flowaortic stenosis in asymptomatic patients: valvulararterial impedance and systolic function from theSEAS substudy. J Am Coll Cardiol Img 2009;2:390–9.

43. Briand M, Dumesnil JG, Kadem L, et al.Reduced systemic arterial compliance impactssignificantly on left ventricular afterload andfunction in aortic stenosis: implications for

diagnosis and treatment. J Am Coll Cardiol 2005;46:291–8.

44. Rieck AE, Cramariuc D, Boman K, et al. Hy-pertension in aortic stenosis: implications for leftventricular structure and cardiovascular events.Hypertension 2012;60:90–7.

45. Eleid MF, Nishimura RA, Sorajja P, Borlaug BA.Systemic hypertension in low gradient severeaortic stenosis with preserved ejection fraction.Circulation 2013;128:1349–53.

46. Kadem L, Dumesnil JG, Rieu R, Durand LG,Garcia D, Pibarot P. Impact of systemic hyperten-sion on the assessment of aortic stenosis. Heart2005;91:354–61.

47. Little SH, Chan KL, Burwash IG. Impact ofblood pressure on the Doppler echocardiographicassessment of aortic stenosis severity. Heart2007;93:848–55.

48. Hachicha Z, Dumesnil JG, Pibarot P. Useful-ness of the valvuloarterial impedance to predictadverse outcome in asymptomatic aortic stenosis.J Am Coll Cardiol 2009;54:1003–11.

49. Kadem L, Garcia D, Durand LG, Rieu R,Dumesnil JG, Pibarot P. Value and limitations ofthe peak-to-peak gradient for the evaluation ofaortic stenosis. J Heart Valve Dis 2006;15:609–16.

50. Clavel MA, Messika-Zeitoun D, Pibarot P, et al.The complex nature of discordant severe calcifiedaortic valve disease grading: new insights fromcombined Doppler-echocardiographic andcomputed tomographic study. J Am Coll Cardiol2013;62:2329–38.

51. O’Sullivan CJ, Stortecky S, Heg D, et al. Clinicaloutcomes of patients with low-flow, low-gradient,severe aortic stenosis and either preserved orreduced ejection fraction undergoing trans-catheter aortic valve implantation. Eur Heart J2013;34:3437–50.

52. Clavel MA, Berthelot-Richer M, Le Ven F, et al.Impact of classic and paradoxical low flow onsurvival after aortic valve replacement for severeaortic stenosis. J Am Coll Cardiol 2015;65:645–53.

53. Ben-Dor I, Maluenda G, Iyasu GD, et al. Com-parison of outcome of higher versus lower trans-valvular gradients in patients with severe aorticstenosis and low (<40%) left ventricular ejectionfraction. Am J Cardiol 2012;109:1031–7.

54. Monin JL, Quere JP, Monchi M, et al. Low-gradient aortic stenosis: operative risk stratifica-tion and predictors for long-term outcome: amulticenter study using dobutamine stress he-modynamics. Circulation 2003;108:319–24.

55. Quere JP, Monin JL, Levy F, et al. Influence ofpreoperative left ventricular contractile reserve onpostoperative ejection fraction in low-gradientaortic stenosis. Circulation 2006;113:1738–44.

56. Clavel MA, Dumesnil JG, Capoulade R,Mathieu P, Sénéchal M, Pibarot P. Outcome ofpatients with aortic stenosis, small valve area andlow-flow, low-gradient despite preserved leftventricular ejection fraction. J Am Coll Cardiol2012;60:1259–67.

57. Eleid MF, Sorajja P, Michelena HI, Malouf JF,Scott CG, Pellikka PA. Flow-gradient patterns insevere aortic stenosis with preserved ejection

fraction: clinical characteristics and predictors ofsurvival. Circulation 2013;128:1781–9.

58. Mehrotra P, Jansen K, Flynn AW, et al. Differ-ential left ventricular remodelling and longitudinalfunction distinguishes low flow from normal-flowpreserved ejection fraction low-gradient severeaortic stenosis. Eur Heart J 2013;34:1906–14.

59. Ben-Dor I, Dvir D, Barbash IM, et al. Outcomesof patients with severe aortic stenosis at highsurgical risk evaluated in a trial of transcatheteraortic valve implantation. Am J Cardiol 2012;110:1008–14.

60. Maes F, Boulif J, Pierard S, et al. Natural his-tory of paradoxical low gradient “severe” aorticstenosis. Circ Cardiovasc Imaging 2014;7:714–22.

61. Clavel MA, Ennezat PV, Maréchaux S, et al.Stress echocardiography to assess stenosisseverity and predict outcome in patients withparadoxical low-flow, low-gradient aortic stenosisand preserved LVEF. J Am Coll Cardiol Img 2013;6:175–83.

62. Clavel MA, Côté N, Mathieu P, et al. Para-doxical low-flow, low-gradient aortic stenosisdespite preserved left ventricular ejection fraction:new insights from weights of operatively excisedaortic valves. Eur Heart J 2014;35:2655–62.

63. deFilippi CR, Willett DL, Brickner ME, et al.Usefulness of dobutamine echocardiography indistinguishing severe from nonsevere valvularaortic stenosis in patients with depressed leftventricular function and low transvalvular gradi-ents. Am J Cardiol 1995;75:191–4.

64. Blais C, Burwash IG, Mundigler G, et al. Pro-jected valve area at normal flow rate improves theassessment of stenosis severity in patients withlow flow, low-gradient aortic stenosis: the multi-center TOPAS (Truly or Pseudo Severe AorticStenosis) study. Circulation 2006;113:711–21.

65. Rosenhek R, Binder T, Porenta G, et al. Pre-dictors of outcome in severe, asymptomatic aorticstenosis. N Engl J Med 2000;343:611–7.

66. Agatston AS, Janowitz WR, Hildner FJ,Zusmer NR, Viamonte M Jr., Detrano R. Quantifi-cation of coronary artery calcium using ultrafastcomputed tomography. J Am Coll Cardiol 1990;15:827–32.

67. Clavel MA, Pibarot P, Messika-Zeitoun D, et al.Impact of aortic valve calcification, as measured byMDCT, on survival in patients with aortic stenosis:results of an international registry study. J Am CollCardiol 2014;64:1202–13.

68. Dweck MR, Jenkins WS, Vesey AT, et al. 18F-NaF uptake is a marker of active calcification anddisease progression in patients with aortic steno-sis. Circ Cardiovasc Imaging 2014;7:371–8.

69. Aggarwal SR, Clavel MA, Messika-Zeitoun D,et al. Sex differences in aortic valve calcificationmeasured by multidetector computed tomographyin aortic stenosis. Circ Cardiovasc Imaging 2013;6:40–7.

70. Simard L, Côté N, Dagenais F, et al. Sex-relateddiscordance between aortic valve calcification andhemodynamic severity of aortic stenosis: is valvularfibrosis the explanation? Circ Res 2016 Nov 22[E-pub ahead of print].







































































































































































































202

71. Shen M, Tastet L, Capoulade R, et al. Effect ofage and aortic valve anatomy on calcification andhaemodynamic severity of aortic stenosis. Heart2017;103:32–9.

72. Fougères É, Tribouilloy C, Monchi M, et al.Outcomes of pseudo-severe aortic stenosis underconservative treatment. Eur Heart J 2012;33:2426–33.

73. Baron SJ, Arnold SV, Herrmann HC, et al.Impact of ejection fraction and aortic valvegradient on outcomes of transcatheter aortic valvereplacement. J Am Coll Cardiol 2016;67:2349–58.

74. Clavel MA, Webb JG, Rodés-Cabau J, et al.Comparison between transcatheter and surgicalprosthetic valve implantation in patients with se-vere aortic stenosis and reduced left ventricularejection fraction. Circulation 2010;122:1928–36.

75. Elmariah S, Palacios IF, McAndrew T, et al.Outcomes of transcatheter and surgical aorticvalve replacement in high-risk patients with aorticstenosis and left ventricular dysfunction: resultsfrom the Placement of Aortic Transcatheter Valves(PARTNER) trial (Cohort A). Circ Cardiovasc Interv2013;6:604–14.

76. Ribeiro H, Lerakis S, Cavalcante J, et al.Transcatheter aortic valve implantation in patientswith low-flow, low-gradient aortic stenosis: theprospective multicenter TOPAS-TAVI registry.Paper presented at: TCT 2016; October 29 toNovember 2, 2016; Washington, DC.

77. Ribeiro HB, Nombela-Franco L, Munoz-Garcia AJ, et al. Predictors and impact ofmyocardial injury after transcatheter aortic valvereplacement: A multicenter registry. J Am CollCardiol 2015;66:2075–88.

78. O’Sullivan CJ, Englberger L, Hosek N, et al.Clinical outcomes and revascularization strategies inpatients with low-flow, low-gradient severe aorticvalve stenosis according to the assigned treatmentmodality. J Am Coll Cardiol Img 2015;8:704–17.

79. MohtyD,Boulogne C,Magne J, et al. Prevalenceand long-term outcome of aortic prosthesis-patientmismatch in patients with paradoxical low-flow se-vere aortic stenosis. Circulation 2014;130:S2–31.

80. Pibarot P, Weissman NJ, Stewart WJ, et al.Incidence and sequelae of prosthesis-patientmismatch in transcatheter versus surgical valvereplacement in high-risk patients with severeaortic stenosis—a PARTNER trial cohort A analysis.J Am Coll Cardiol 2014;64:1323–34.

81. Zorn GL 3rd, Little SH, Tadros P, et al.Prosthesis-patient mismatch in high-risk patientswith severe aortic stenosis: a randomized trial of aself-expanding prosthesis. J Thorac CardiovascSurg 2016;151:1014–23.e3.

82. Lindman BR, Alexander KP, O’Gara PT,Afilalo J. Futility, benefit, and transcatheter aorticvalve replacement. J Am Coll Cardiol Intv 2014;7:707–16.

83. Dahou A, Magne J, Clavel MA, et al. Tricuspidregurgitation is associated with increased risk ofmortality in patients with low-flow low-gradientaortic stenosis and reduced ejection fraction: re-sults of the multicenter TOPAS study (True orpseudo-severe aortic stenosis). J Am Coll CardiolIntv 2015;8:588–96.

84. Dahou A, Clavel MA, Capoulade R, et al. Rightventricular longtitudinal strain for risk stratifica-tion in low-flow, low-gradient aortic stenosis withlow ejection fraction. Heart 2016;102:548–54.

85. Arnold SV, Reynolds MR, Lei Y, et al. Pre-dictors of poor outcomes after transcatheter aorticvalve replacement: results from the PARTNER(Placement of Aortic Transcatheter Valve) trial.Circulation 2014;129:2682–90.

86. Cavalcante JL, Rijal S, Althouse AD, et al.Right ventricular function and prognosis in pa-tients with low-flow, low-gradient severeaortic stenosis. J Am Soc Echocardiogr 2016;29:325–33.

87. O’Sullivan CJ, Stortecky S, Butikofer A, et al.Impact of mitral regurgitation on clinical outcomesof patients with low-ejection fraction, low-gradient severe aortic stenosis undergoing trans-catheter aortic valve implantation. Circ CardiovascInterv 2015;8:e001895.

88. Herrmann S, Stork S, Niemann M, et al. Low-gradient aortic valve stenosis: Myocardial fibrosisand its influence on function and outcome. J AmColl Cardiol 2011;58:402–12.

89. Lindman BR, Breyley JG, Schilling JD, et al.Prognostic utility of novel biomarkers of cardio-vascular stress in patients with aortic stenosisundergoing valve replacement. Heart 2015;101:1382–8.

KEY WORDS aortic stenosis, computedtomography, echocardiography, low flow,low-gradient

APPENDIX For supplemental videos andtheir legends, please see the online version ofthis article.


























































































cardiac imaging for assessing low-gradient severe aortic...

Documents