Journal of Cardiology 65 (2015) 369–376

Original article

An easy and reproducible parameter for the assessment of the pressuregradient in patients with aortic stenosis disease: A magnetic resonancestudy

Valentina Valenti (MD)a,b,*, Sebastiano Sciarretta (MD)c, Matt Levin (BS)a,Leon Shubayev (BS)a, Sophia Edelstein (BS)a, Mohammad I. Zia (MD)a,d,Speranza Rubattu (MD)c,e, Massimo Volpe (MD, FAHA)c,e, Seth Uretsky (MD)f,Steven D. Wolff (MD)a

a Carnegie Hill Radiology, New York, NY, USAb Department of Radiology, University ‘‘La Sapienza,’’ Sant’Andrea Hospital, Rome, Italyc IRCCS Neuromed, Pozzilli, IS, Italyd Schulich Heart Centre, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canadae Division of Cardiology, Department of Molecular and Clinical Medicine, Sant’Andrea Hospital, University ‘‘La Sapienza’’, Rome, Italyf Department of Cardiovascular Medicine, Morristown Medical Center, Morristown, NJ, USA

A R T I C L E I N F O

Article history:

Received 2 December 2013

Received in revised form 30 June 2014

Accepted 14 July 2014

Available online 22 August 2014

Keywords:

Aortic stenosis

Cardiac magnetic resonance

Pressure gradient

Gorlin’s formula

Hakki’s formula

A B S T R A C T

Aim: Cardiovascular magnetic resonance (CMR) has been increasingly used as an alternative method to

evaluate the severity of aortic stenosis. The aim of our study was to evaluate whether the indirect

measurement of the aortic gradient (Calc-PG), derived from Gorlin’s formula, is a reproducible parameter

for gradient assessment. Then, we evaluated if this parameter is correlated with left ventricular

hypertrophy, considered as a marker of severity of aortic stenosis, better than phase-contrast sequences-

derived pressure gradient (PC-PG) and aortic valve area.

Methods: Forty-one patients with isolated aortic stenosis underwent CMR. Calc-PG was obtained from

the formula (cardiac output/aortic valve area)2, and it was compared to PC-PG.

Results: We found that the Calc-PG has higher correlation with left ventricle mass than PC-PG (r2 0.44,

p < 0.001 vs. r2 0.26, p < 0.01), also after multivariate analysis adjusting for age, gender and

hypertension (p < 0.001). Furthermore, Calc-PG was more reproducible than PC-PG. The receiver

operating characteristic comparison curve analysis showed that Calc-PG has a significantly higher ability

to describe the presence of left ventricular hypertrophy than PC-PG (area under the curve 0.85, 95% CI

0.70–0.94, p < 0.0001 vs. 0.74, 95% CI 0.58–0.87, p = 0.03).

Conclusions: We propose that transaortic gradient indirectly calculated by using the simplified Gorlin’s

equation could be an alternative method to assess the severity of aortic stenosis.

� 2014 Japanese College of Cardiology. Published by Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Journal of Cardiology

jo u rn al h om ep age: ww w.els evier .c o m/lo c ate / j j c c

Introduction

Aortic stenosis is the most common valvular heart disease [1].The 2006 American College of Cardiology/American Heart Associ-ation guidelines for the management of valvular heart diseasesrecommend assessing the severity of aortic stenosis by using bothanatomical and hemodynamic parameters, namely the aortic valvearea and the transaortic gradient, respectively [2]. However, recent

* Corresponding author at: Advanced Cardiovascular Imaging, 170 East 77th

Street, New York, NY 10075, USA. Tel.: +1 212 369 9200; fax: +1 212 369 5048.

E-mail addresses: valevale2012@hotmail.com, v.valenti80@yahoo.it (V. Valenti).

http://dx.doi.org/10.1016/j.jjcc.2014.07.015

0914-5087/� 2014 Japanese College of Cardiology. Published by Elsevier Ltd. All rights

evidence suggests that the evaluation of the transaortic gradientmight be more accurate to distinguish subjects with moderate orsevere aortic stenosis [3–5].

Cardiovascular magnetic resonance (CMR) has been increas-ingly used as an alternative method to echocardiography toevaluate the severity of aortic stenosis [6–8]. Echocardiographygenerally provides reliable measurement of pressure gradient.However, a proper visualization of aortic valve and an accurateestimation of pressure gradient by echocardiographic examinationcan be difficult in some patients with poor acoustic windows, suchas obese subjects, subjects affected by chronic obstructivepulmonary disease, or in patients who underwent major cardiacsurgery. In addition, an accurate estimation of pressure gradient by

reserved.

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376370

echocardiography requires a significant technical expertise of thesonographer [9,10]. With CMR, evaluation of the aortic valve areawith steady-state-free-precession sequence has been demonstrat-ed to be highly reliable and reproducible [8,11,12]. On the otherhand, the assessment of aortic flows and aortic pressure gradientby using the phase-contrast sequences-derived pressure gradient(PC-PG) is subject to several potential sources of error that maycompromise the correct classification of the severity of aorticstenosis and, subsequently, the clinical management of thesepatients [13].

Here we tested the ability of an additional non-invasiveparameter, beyond PC-PG and aortic valve area, for estimatingpressure gradient in aortic stenosis by using CMR. It consists of theindirect calculation of the gradient from the cardiac output andaortic valve area, by using the inverse simplified Gorlin’s formula[14,15]. A potential advantage of this method is that it can be usedto determine the transvalvular pressure gradient without theacquisition or analysis of phase contrast images.

Left ventricular (LV) hypertrophy represents the main marker ofpreclinical organ damage in patients with aortic stenosis. Previousstudies indicated that LV hypertrophy represents a valid surrogatemarker of the severity of the disease, and remarkably, it is anaccurate and strong predictor for the occurrence of major adverseevents in subjects affected by aortic stenosis [16,17].

The aim of our study was to evaluate whether the above-mentioned parameter (hereafter Calc-PG) based on the formula,(cardiac output/aortic valve area)2, is an easy and reproducibleindex for the evaluation of pressure gradient. In addition, we testedwhether this parameter is correlated with LV hypertrophy,considered as a marker of cardiac remodeling caused by aorticstenosis and as a marker of the severity of the disease, better thanPC-PG and aortic valve area.

Methods

This study is a retrospective analysis of patients referred for aclinical CMR evaluation in our center. Inclusion criteria included atleast moderate aortic stenosis, as indicated by aortic valve area,and preserved ejection fraction (LV ejection fraction >50% andright ventricular ejection fraction >40%). Exclusion criteriaincluded regional ventricular wall motion abnormalities, othersignificant associated valve disease, myocardial ischemia or scar,intra-cardiac shunt, and patients with an irregular cardiac rhythm.We collected a total of 99 patients with aortic stenosis, whounderwent CMR study. From these we excluded 58 patients due to

Fig. 1. The aortic localizer view (A) showing a turbulent aortic sten

the presence of atrial fibrillation, concomitant presence of othervalve diseases, myocardial fibrosis detected by contrast delayanalysis, segmental anomalous wall motion, or global LVdysfunction. The remaining 41 patients represent our studypopulation. We used the same protocol as in our previouspublished papers for the evaluation of transvalvular flow byCMR [18,19].

Patients were imaged with a 1.5-T MRI scanner using a 8elements, phased-array cardiac coil (GE Signa, EXCITE, GE MedicalSystems, Milwaukee, WI, USA). Imaging was electrocardiogram-gated, and performed during breath holds. After scout imagesacquisition, short- and long-axis cine images were acquired usinga steady-state free precession pulse sequence (FIESTA) with thefollowing parameters: repetition time (TR) was 3.5 ms; echo time(TE) was 1.5 ms; the flip angle was 608; views per segment (VPS)were 12; field of view (FOV) was 350 mm � 350 mm; matrix sizewas 192 � 160; nominal temporal resolution was 42 ms; breathhold time range was 12–20 s; slice thickness was 8.0 mm, andnominal spatial resolution (voxel size) was 8 mm �1.8 mm � 2.2 mm. With regard to velocity-encoded phase con-trast imaging, a cine localizer was obtained parallel to thedirection of flow in order to ensure that measurements of velocitywere perpendicular to the plane of flow (Fig. 1A). From thislocalizer, PC images were acquired in an imaging planeperpendicular to the jet, from the LV outflow tract through thetips of the aortic valve cusps. Typically, 6–9 contiguous slices wereacquired, each 4 mm thick, extending 16–24 mm proximal to theaortic cusp tips and 4–12 mm distal (Fig. 1B). Valve 0 mmcorresponds to the reference plane at the level of tips of the openaortic cusps, whereas valve �24 mm and valve +12 mm arelocated 24 mm and 12 mm proximal to and distal to this reference,respectively. We used the following nominal scan parameters: TRwas 6.5 ms; TE was 3.8 ms; the flip angle was 208; VPS was 6; FOVwas 480 mm � 360 mm; matrix size was 512 � 224; nominaltemporal resolution was 78 ms; breath hold time range was18–30 s; maximum encoded velocity (VENCmax) was 550 cm/s.Slice thickness was 4.0 mm; and nominal spatial resolution (voxelsize) was 4 mm � 0.94 mm � 2.1 mm. After that the clinical scanwas completed and additional phase contrast images of astationary bottle of water (phantom) were acquired for baselineflow correction [20]. CMR data were analyzed utilizing ReportCard 4.0 software (GE Medical Systems, Waukesha, WI, USA). Leftventricle volumes were determined by manual endocardialborder tracing in short axis, from the base to apex inend-diastolic and end-systolic phases. LV mass was measured

osis jet and imaging slice planes perpendicular to the jet (B).

Table 1Clinical and cardiac magnetic resonance variables in patients with AS.

Patients (n = 41)

Age (years) 72.7 � 10.4

Body surface area (m2) 2 � 0.4

Gender (male) 27

Hypertension 21

Left ventricle ejection fraction (%) 64.1 � 6.6

Left ventricle end-diastolic volume (ml/m2) 71.9 � 16.2

Left ventricle end-systolic volume (ml/m2) 26.2 � 8.6

Left volume stroke volume (ml/m2) 43.9 � 10.9

Left ventricle cardiac output (ml/min) 5.5 � 1.3

Left ventricular mass/body surface area (g/m2) 64.9 � 16.2

Right ventricular ejection fraction (%) 60.7 � 9.4

Right ventricular end-diastolic volume (ml/m2) 70.8 � 15.7

Right ventricular end-systolic volume (ml/m2) 28.7 � 10.8

Aortic valve area (cm2) 1.02 � 0.23

Phase contrast peak pressure gradient (mmHg) 50.4 � 23.9

Calculated peak pressure gradient (mmHg) 35.1 � 23.9

Fig. 2. Relationship between the phase-contrast pressure gradient (PC-PG) and the

calculated pressure gradient (Calc-PG) according to the Pearson’s coefficient (r2

0.577, p < 0.001) (A) and the Bland–Altman analysis (COV 37%; 95% limits of

agreement 17.24–47.78). (B). There is moderate agreement between the 2

parameters.

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376 371

in end-diastolic phase by manually tracing the endocardial andepicardial borders in short axis with exclusion of the papillarymuscles. The ejection fraction, stroke volume, and the cardiacoutput for both ventricles were derived. Severity of aortic stenosiswas estimated using 3 methods:

- The Calc-PG: inverse Hakki’s formula [15,21], which is asimplification of Gorlin’s [14] formula; DP = (cardiac output/aortic valve area) [2].

- The aortic valve planimetry. Manual planimetry of the aorticvalve area was measured at systole, when the valve was the mostopen. Dark signal was not planimetered as these dark areas likelyrepresent calcium.

- PC-PG: indirectly calculated by simplified Bernoulli equation [22](DP = 4 � velocity2) from the transvalvular aortic jet velocity.Then, we investigated the correlation between Calc-PG, PC-PG,and aortic valve area with LV hypertrophy. Two independentmeasurements of all parameters were performed in each patientby two different blinded observers.

All analyses were carried out using a standard statisticalpackage (SPSS Version 16.0, Chicago, IL, USA and Medcalcsoftware). Continuous variables were expressed as mean � stan-standard deviation. Categorical variables were expressed as frequen-cy and percentage. We compared means of continuous variablesbetween two groups by using t-test. All tests were 2-tailed, andstatistical significance was accepted at p < 0.05. Correlation betweenparameters estimating the aortic stenosis severity and LV mass wascalculated by using the coefficient of determination r2. Thecomparison between PC-PG and Calc-PG was performed by usingcoefficient of determination, the Bland–Altman analysis, and thereceiver operating characteristic (ROC) curve analysis. In the ROCanalysis, we used LV mass as a categorical variable. We performed aROC analysis to corroborate our previous results indicating that Calc-PG correlated better with LV mass than PC-PG and aortic valve area. Inorder to perform the ROC analysis we considered LV mass as acategorical variable. Since definite and standardized age-, gender-,and ethnicity-related cut-off values for identifying LV hypertrophy byCMR in the general population are not available yet we categorized LVmass by using the 75 percentile of LV mass/body surface area in ourpopulation (70.46 g/m2). LV hypertrophy was considered as a validindirect surrogate parameter to estimate the functional severity ofaortic stenosis in ROC analyses [16,17]. Interobserver variability wasmeasured using the coefficient of determination, Bland–Altmananalysis, and intraclass correlation coefficient analysis. Univariateand multivariate predictors were analyzed by using bivariate andmultiple linear regression analyses.

Results

Demographic and CMR data of the study population aresummarized in Table 1. Aortic gradient was indirectly calculatedby velocity-encoding phase contrast or it was measured usingHakki’s formula. There was moderate agreement between the PC-PG and the Calc-PG according to the coefficient of determination(r2 0.577, p < 0.001) and as shown in the Bland–Altman analysis(COV 37%; 95% limits of agreement 17.2–47.8; Fig. 2A and B). Wefound that the coefficient of determination of Calc-PG with LVmass was significantly higher (r2 = 0.44, p < 0.001) than thecorrelations of PC-PG (r2 = 0.26, p = 0.01) and aortic valve areawith LV mass (r2 = 0.02, p = ns) (Fig. 3). Of note, the correlationbetween aortic valve area and LV mass was poor and notsignificant. After multivariate analysis adjusting for age, gender,and hypertension, the Calc-PG was confirmed to be the strongestpredictor of increased LV mass (Calc-PG = Beta 0.62, p < 0.001; PC-PG = Beta 0.47, p < 0.001) (Tables 2A and 2B). In addition, we found

that Calc-PG shows a higher area under the curve (AUC) (AUC: 0.85,CI 0.70–0.94, p < 0.0001) when it is associated with the presence ofLV hypertrophy than PC-PG (AUC: 0.74, CI 0.58–0.87, p = 0.03), asindicated by the ROC curve analysis (Fig. 4). Finally, the Calc-PGshowed the highest interobserver reproducibility in comparison to

Fig. 3. Relationship between left ventricle mass and different parameters to assess aortic stenosis according to the Pearson’s coefficient. (A) There is mild agreement between

phase-contrast pressure gradient (PC-PG) and left ventricular mass (r2 = 0.26, p = 0.01). (B) There is no agreement between aortic valve area and left ventricular mass

(r2 = 0.02, p = ns). (C) There is moderate agreement between calculated pressure gradient (Calc-PG) and left ventricular mass (r2 = 0.44, p < 0.001).

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376372

Table 2AMultivariate analysis of PC-PG versus left ventricular mass adjusted for age, gender,

and hypertension.

Left ventricular mass (g/m2)

Beta p-value

Age �0.22 ns

Gender �0.36 <0.01

Hypertension 0.01 ns

PC-PG 0.47 <0.001

PC-PG, measured pressure gradient phase contrast sequence.

Table 2BMultivariate analysis of Calc-PG versus left ventricle mass adjusted for age, gender

and hypertension.

Left ventricular mass (g/m2)

Beta p-value

Age �0.24 0.03

Gender �0.34 <0.01

Hypertension 0.09 ns

Calc-PG 0.62 <0.001

Calc-PG, calculated gradient by cardiac output and aortic valve area.

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376 373

the other parameters for the estimation of aortic stenosis(Figs. 5 and 6), whereas the PC-PG had the lowest one. Theinterobserver reproducibility according to the coefficient ofdetermination was: r2 0.515 for aortic valve area measurement,r2 0.454 for PC-PG and r2 0.608 for Calc-PG (all p < 0.001). TheBland–Altman analysis showed for the aortic valve area a biasof �0.10 (COV 13%; 95% limits of agreement 0.23–0.44), for thePC-PG �18 (COV 46%; 95% limits of agreement 11–47), and for theCalc-PG 11.4 (COV 35%; 95% limits of agreement from 40.2 to 17.5).The intraclass correlation coefficient of the interobserver agree-ment was 0.6901 (95% CI 0.3116–0.8796) for aortic valve area,0.5263 (95% CI 0.05921–0.8044) for PC-PG, and 0.7375 (95% CI0.3962–0.8997) for the Calc-PG.

Discussion

Our study demonstrates that in patients with aortic stenosisassessed by CMR, pressure gradient calculated by the cardiac

Fig. 4. Receiver operating characteristic curve comparison analysis. Calculated

pressure gradient (Calc-PG) shows a higher area under the curve (AUC) with respect

to the presence of left ventricular hypertrophy (AUC: 0.85, CI 0.70–0.94) than

phase-contrast pressure gradient (PC-PG, AUC: 0.74, CI 0.58–0.87).

Fig. 5. Interobserver reproducibility. Relationship between the aortic valve area

(panel A, r2 0.51, p < 0.001), the phase-contrast pressure gradient (panel B, r2 0.44,

p < 0.001), and the calculated pressure gradient (panel C, r2 0.61, p < 0.001)

obtained from 2 different observers, according to the Pearson’s coefficient. PC-PG,

phase-contrast pressure gradient.

Fig. 6. Interobserver reproducibility. Relationship between the aortic valve area

(panel A, COV 13%; 95% limits of agreement 0.23–0.44), phase-contrast pressure

gradient (PC-PG, panel B; COV 46%; 95% limits of agreement 11.11–46.99), and the

calculated pressure gradient (Calc-PG, panel C; COV 35%; 95% limits of agreement

from 40.17 to 17.45) according to the Bland–Altman analysis.

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376374

output and aortic valve area is more reproducible than the pressuregradient measured by using the phase-contrast sequences or theaortic valve area. Moreover, Calc-PG correlated with the LVhypertrophy better than the other parameters. LV hypertrophyrepresents the main marker of preclinical organ damage in patients

with aortic stenosis. Previous studies indicated that LV hypertro-phy represents a valid surrogate marker of the severity of thedisease, and remarkably, it is an accurate and strong predictor forthe occurrence of major adverse events, such as systolic anddiastolic dysfunction, heart failure, syncope, and mortality insubjects affected by aortic stenosis [16,17].

Echocardiography generally provides reliable measurement ofpressure gradient. However, echocardiographic assessment oftransaortic gradient and aortic valve area could be affected by highthoracic impedance and by the inexperience of the sonographer[9,10]. CMR allows the direct visualization and planimetry of theaortic valve area and the flow-dedicated sequences for themeasurement of velocity and pressure gradient. However,although the evaluation of aortic valve area has been shown tobe accurate and reproducible with this imaging technique[6–8,11], the measurement of the PC-PG has some limitationsdue to no perpendicular sampling of the transaortic flow velocitywith respect to the blood flow, or to aliasing and signal artifactscaused by turbulences across the aortic valve [6,14,23–25].Previous studies reported good agreement between PC-PG andechocardiography [26–28]. However, these studies includedpopulations with small samples and with mild or moderatedisease. Caruthers et al. [6] measured the correlation betweenvelocity–time integrals measurements made by CMR and thosemade by Doppler ultrasound in 24 patients with aortic stenosis.They reported a good correlation between CMR and Doppler onlyfor aortic valve area greater than 0.8 mm2. For values less than0.8 mm2, PC-PG underestimated more than half of patients withthe disease. O’Brien et al. [23] studied both in vivo and in vitroaortic stroke volume obtained from the phase contrast sequence inorder to evaluate errors in steady obstructive and non-obstructiveturbulent flows. They concluded that the signal loss correlatedwith turbulent flow was associated with flow errors in stenotic jetsfor severe lesions. The difficulty in measuring turbulent flows byCMR using the phase contrast sequence could represent asignificant limitation to the potential usefulness of this imagingmethod in the evaluation of patients with aortic stenosis. In fact,recent evidence indicates that a correct assessment of transaorticpressure gradient is necessary to perform an accurate evaluation ofpatients with aortic stenosis. First of all, in up to 30% of patientswith aortic stenosis there is a discrepancy between aortic valvearea and pressure gradient in the definition of the degree ofseverity of the valve disease [3,4]. More importantly, thetransaortic pressure gradient was shown to predict the incidenceof adverse outcomes better than aortic valve area in patients withaortic stenosis [5]. Our study could suggest that an indirectcalculation of the transaortic pressure gradient by using a simpleformula which includes the aortic valve area and cardiac outputcould represent an easy and reproducible way to overcome thelimitations of the PC-PG and to more consistently assess theseverity of aortic stenosis. Future studies are needed to validatethese assertions. The Calc-PG does not use phase contrast sequencethat could be a source of error, but only the steady state freeprecession sequence. Importantly, Calc-PG is derived from aorticvalve area and cardiac output, and it is known that CMR allowsprecise and rapid measurement of aortic valve area and representsthe gold standard for the measurement of cardiac output fromventricular volumes [28,29]. Because of all these advantages, webelieve that Calc-PG correlates with LV hypertrophy better thanthe other parameters. In fact, LV hypertrophy indirectly reflects theactual functional severity of aortic stenosis. Of note, the correlationbetween cardiac output and LV mass could also partiallycontribute to the close correlation of Calc-PG with LV mass. Infact, LV mass is strongly dependent on LV chamber size, which is inturn physiologically determined by stroke volume and myocardialcontractility. Therefore, LV mass is also related to stroke volume

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376 375

[30]. Alternatively, it could be possible that among patients withaortic stenosis with preserved cardiac function, a better preservedcardiac output could be associated with more developed LVhypertrophy. In this regard, it is known that in patients with aorticstenosis and reduced cardiac systolic function, the transaorticpressure gradient often underestimates the severity of aorticstenosis [31]. Future studies conducted in subjects with aorticstenosis and cardiac dysfunction are needed to evaluate how Calc-PG is correlated with LV hypertrophy and severity of stenosis withrespect to the other parameters.

Of note, we found no correlation between aortic valve area andLV mass, thus confirming previous evidence indicating a poorprognostic value of this parameter [32]. Indeed, aortic valve area isaffected by anthropometric characteristics, such as body size, andit is not infrequent that reductions of valve area are not paralleledby proportional increases in the pressure gradient in smallerindividuals.

The lack of comparison between Calc-PG and the PC-PG by areference standard represents the main limitation of the study.However, the purpose of our study was to demonstrate that Calc-PG was a reproducible parameter for transaortic gradientevaluation and to assess its correlation with LV hypertrophy.Future studies conducted by using catheterization as a standardreference are needed to validate this parameter. Interestingly,PC-PG was previously shown to underestimate the transvalvularpressure gradient compared to echocardiography [6,23]. Calc-PGmay also estimate lower pressure gradients than echocardiogra-phy. A direct comparison of Calc-PG to echocardiography withthe reference pressure gradient value obtained with catheteri-zation will provide more accurate information regarding thispossibility.

Another limitation is the fact that our study is retrospectivewith a small number of patients. However, since Calc-PG iscalculated from aortic valve area (a widely validated method toevaluate aortic stenosis by CMR) and cardiac output (CMRrepresents the standard reference to assess LV volume), and sinceCMR represents the gold standard for the mass estimation, ourresults were consistent and robust and the variability amongdifferent subjects regarding these parameters was relatively low.This allowed us to observe statistically significant results, despitethe limited sample size of our population.

Finally, we included only patients with normal LV ejectionfraction. The correlation between calc-PC and LV hypertrophy inpatients without normal global systolic function may be lowerthan what we observed.

In conclusion, our study demonstrates the pressure gradientcalculated by the cardiac output and aortic valve area is easierand more reproducible than the other commonly used param-eters to define the degree of aortic stenosis and it correlatesbetter with LV mass. Therefore, on the basis of these results, wepropose that this method to calculate the pressure gradient byCMR could be routinely used in clinical practice to assess theseverity of aortic stenosis together with PC-PG and aortic valvearea.

Funding

None.

Conflict of interestDr. Steven Wolff is an owner of NeoSoft, LLC and NeoCoil, LLC.

Acknowledgment

We thank Dr. Azhar Supariwala for his contribution to thestatistical analysis.

References

[1] Carabello BA, Paulus WJ. Aortic stenosis. Lancet 2009;373:956–66.[2] Bonow RO, Carabello BA, Chatterjee K, de Leon Jr AC, Faxon DP, Freed MD,

Gaasch WH, Lytle BW, Nishimura RA, O’Gara PT, O’Rourke RA, Otto CM, ShahPM, Shanewise JS, American College of Cardiology/American Heart AssociationTask Force on Practice Guidelines. 2008 focused update incorporated into theACC/AHA 2006 guidelines for the management of patients with valvular heartdisease: a report of the American College of Cardiology/American HeartAssociation Task Force on Practice Guidelines (Writing Committee to revisethe 1998 guidelines for the management of patients with valvular heartdisease). Endorsed by the Society of Cardiovascular Anesthesiologists, Societyfor Cardiovascular Angiography and Interventions, and Society of ThoracicSurgeons. J Am Coll Cardiol 2008;52:e1–42.

[3] Minners J, Allgeier M, Gohlke-Baerwolf C, Kienzle RP, Neumann FJ, Jander N.Inconsistencies of echocardiographic criteria for the grading of aortic valvestenosis. Eur Heart J 2008;29:1043–8.

[4] Minners J, Allgeier M, Gohlke-Baerwolf C, Kienzle RP, Neumann FJ, Jander N.Inconsistent grading of aortic valve stenosis by current guidelines: haemo-dynamic studies in patients with apparently normal left ventricular function.Heart 2010;96:1463–8.

[5] Jander N, Minners J, Holme I, Gerdts E, Boman K, Brudi P, Chambers JB, EgstrupK, Kesaniemi YA, Malbecq W, Nienaber CA, Ray S, Rossebø A, Pedersen TR,Skjærpe T, et al. Outcome of patients with low-gradient ‘‘severe’’ aorticstenosis and preserved ejection fraction. Circulation 2011;123:887–95.

[6] Caruthers SD, Lin SJ, Brown P, Watkins MP, Williams TA, Lehr KA, Wickline SA.Practical value of cardiac magnetic resonance imaging for clinical quantifica-tion of aortic valve stenosis: comparison with echocardiography. Circulation2003;108:2236–43.

[7] John AS, Dill T, Brandt RR, Rau M, Ricken W, Bachmann G, Hamm CW.Magnetic resonance to assess the aortic valve area in aortic stenosis: howdoes it compare to current diagnostic standards. J Am Coll Cardiol 2003;42:519–26.

[8] Garcia J, Kadem L, Larose E, Clavel MA, Pibarot P. Comparison betweencardiovascular magnetic resonance and transthoracic Doppler echocardiogra-phy for the estimation of effective orifice area in aortic stenosis. J CardiovascMagn Reson 2011;28:13–25.

[9] Pearlman AS, Adams DB, Kraft CD, Witt SA. Advancing recognition of theresponsibilities of cardiac sonographers. J Am Soc Echocardiogr 2009;12:1414–6.

[10] Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP, IungB, Otto CM, Pellikka PA, Quinones M. Echocardiographic assessment of valvestenosis: EAE/ASE recommendations for clinical practice. Eur J Echocardiogr2009;10:1–25.

[11] Kupfahl C, Honold M, Meinhardt G, Vogelsberg H, Wagner A, Mahrholdt H,Sechtem U. Evaluation of aortic stenosis by cardiovascular magnetic resonanceimaging: comparison with established routine clinical techniques. Heart2004;90:893–901.

[12] Reant P, Lederlin M, Lafitte S, Serri K, Montaudon M, Corneloup O, Roudaut R,Laurent F. Absolute assessment of aortic valve stenosis by planimetry usingcardiovascular magnetic resonance imaging: comparison with transesopha-geal echocardiography, transthoracic echocardiography, and cardiac cathe-terisation. Eur J Radiol 2006;59:276–83.

[13] Chai P, Mohiaddin R. How we perform cardiovascular magnetic resonance flowassessment using phase-contrast velocity mapping. J Cardiovasc Magn Reson2005;7:705–16.

[14] Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenoticmitral valve, other cardiac valves, and central circulatory shunts. I. Am Heart J1951;41:1–29.

[15] Hakki AH, Iskandrian AS, Bemis CE, Kimbiris D, Mintz GS, Segal BL, Brice C. Asimplified valve formula for the calculation of stenotic cardiac valve areas.Circulation 1981;63:1050–5.

[16] Kupari M, Turto H, Lommi J. Left ventricular hypertrophy in aortic valvestenosis: preventive or promotive of systolic dysfunction and heart failure.Eur Heart J 2005;26:1790–6.

[17] Orsinelli DA, Aurigemma GP, Battista S, Krendel S, Gaasch WH. Left ventricularhypertrophy and mortality after aortic valve replacement for aortic stenosis. Ahigh risk subgroup identified by preoperative relative wall thickness. J Am CollCardiol 1993;22:1679–83.

[18] Tanaka K, Makaryus AN, Wolff SD. Correlation of aortic valve area obtained bythe velocity-encoded phase contrast continuity method to direct planimetryusing cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2007;5:799–805.

[19] Uretsky S, Supariwala A, Nidadovolu P, Khokhar SS, Comeau C, Shubayev O,Campanile F, Wolff SD. Quantification of left ventricular remodeling in re-sponse to isolated aortic or mitral regurgitation. J Cardiovasc Magn Reson2010;12:32.

[20] Chernobelsky A, Shubayev O, Comeau CR, Wolff SD. Baseline correction ofphase contrast images improves quantification of blood flow in the greatvessels. J Cardiovasc Magn Reson 2007;9:681–5.

[21] Puymirat E, Chassaing S, Trinquart L, Barbey C, Chaudeurge A, Bar O, BlanchardD. Hakki’s formula for measurement of aortic valve area by magnetic reso-nance imaging. Am J Cardiol 2010;106:249–54.

[22] Yoganathan AP, Cape EG, Sung HW, Williams FP, Jimoh A. Review of hydrody-namic principles for the cardiologist: applications to the study of blood flowand jets by imaging techniques. J Am Coll Cardiol 1988;12:1344–53.

V. Valenti et al. / Journal of Cardiology 65 (2015) 369–376376

[23] O’Brien KR, Cowan BR, Jain M, Stewart RAH, Kerr AJ, Young AA. MRI phasecontrast velocity and flow errors in turbulent stenotic jets. J Magn ResonImaging 2008;28:210–8.

[24] Cawley PJ, Maki JH, Otto CM. Cardiovascular magnetic resonance imaging forvalvular heart disease: technique and validation. Circulation 2009;119:468–78.

[25] Garcia J, Kadem L, Larose E, Pibarot P. In vivo velocity and flow errorsquantification by phase-contrast magnetic resonance imaging. Conf Proc IEEEEng Med Biol Soc 2008;2008:1377–80.

[26] Kilner PJ, Manzara CC, Mohiaddin RH, Pennell DJ, Sutton MG, Firmin DN,Underwood SR, Longmore DB. Magnetic resonance jet velocity mapping inmitral and aortic valve stenosis. Circulation 1993;87:1239–48.

[27] Nayak KS, Hu BS, Nishimura DG. Rapid quantitation of high-speed flow jets.Magn Reson Med 2003;50:366–72.

[28] Sondergaard L, Thomsen C, Stahlberg F, Gymoese E, Lindvig K, Hildebrandt P,Henriksen O. Mitral and aortic valvular flow: quantification with MR phasemapping. J Magn Reson Imaging 1992;2:295–302.

[29] Grothues F, Moon JC, Bellenger NG, Smith GS, Klein HU, Pennell DJ. Interstudyreproducibility of right ventricular volumes, function, and mass with cardio-vascular magnetic resonance. Am Heart J 2004;147:218–23.

[30] Ganau A, Devereux RB, Pickering TG, Roman MJ, Schnall PL, Santucci S, SpitzerMC, Laragh H. Relation of left ventricular hemodynamic load and contractileperformance to left ventricular mass in hypertension. Circulation 1990;81:25–36.

[31] Monin JL, Monchi M, Gest V, Duval-Moulin AM, Dubois-Rande JL, Gueret P.Aortic stenosis with severe left ventricular dysfunction and low transvalvularpressure gradients: risk stratification by low-dose dobutamine echocardiog-raphy. J Am Coll Cardiol 2001;37:2101–7.

[32] Dweck MR, Joshi S, Murigu T, Gulati A, Alpendurada F, Jabbour A, Maceira A,Roussin I, Northridge DB, Kilner PJ, Cook SA, Boon NA, Pepper J, Mohiaddin RH,Newby DE, et al. Left ventricular remodeling and hypertrophy in patients withaortic stenosis: insights from cardiovascular magnetic resonance. J CardiovascMagn Reson 2012;14:50.