Article

Magnetic Resonance Measurement of Turbulent Kinetic Energy for the Estimation of Irreversible Pressure Loss in Aortic Stenosis

Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, California. Electronic address: .
JACC. Cardiovascular imaging (Impact Factor: 7.19). 01/2013; 6(1):64-71. DOI: 10.1016/j.jcmg.2012.07.017
Source: PubMed
ABSTRACT
OBJECTIVES: The authors sought to measure the turbulent kinetic energy (TKE) in the ascending aorta of patients with aortic stenosis and to assess its relationship to irreversible pressure loss. BACKGROUND: Irreversible pressure loss caused by energy dissipation in post-stenotic flow is an important determinant of the hemodynamic significance of aortic stenosis. The simplified Bernoulli equation used to estimate pressure gradients often misclassifies the ventricular overload caused by aortic stenosis. The current gold standard for estimation of irreversible pressure loss is catheterization, but this method is rarely used due to its invasiveness. Post-stenotic pressure loss is largely caused by dissipation of turbulent kinetic energy into heat. Recent developments in magnetic resonance flow imaging permit noninvasive estimation of TKE. METHODS: The study was approved by the local ethics review board and all subjects gave written informed consent. Three-dimensional cine magnetic resonance flow imaging was used to measure TKE in 18 subjects (4 normal volunteers, 14 patients with aortic stenosis with and without dilation). For each subject, the peak total TKE in the ascending aorta was compared with a pressure loss index. The pressure loss index was based on a previously validated theory relating pressure loss to measures obtainable by echocardiography. RESULTS: The total TKE did not appear to be related to global flow patterns visualized based on magnetic resonance-measured velocity fields. The TKE was significantly higher in patients with aortic stenosis than in normal volunteers (p < 0.001). The peak total TKE in the ascending aorta was strongly correlated to index pressure loss (R(2) = 0.91). CONCLUSIONS: Peak total TKE in the ascending aorta correlated strongly with irreversible pressure loss estimated by a well-established method. Direct measurement of TKE by magnetic resonance flow imaging may, with further validation, be used to estimate irreversible pressure loss in aortic stenosis.

Full-text

Available from: Michael D Hope
Magnetic Resonance Measurement of
Turbulent Kinetic Energy for the Estimation of
Irreversible Pressure Loss in Aortic Stenosis
Petter Dyverfeldt, PHD,* Michael D. Hope, MD,* Elaine E. Tseng, MD,†‡
David Saloner, PHD*†‡
San Francisco, California
OBJECTIVES The authors sought to measure the turbulent kinetic energy (TKE) in the ascending
aorta of patients with aortic stenosis and to assess its relationship to irreversible pressure loss.
BACKGROUND Irreversible pressure loss caused by energy dissipation in post-stenotic flow is an
important determinant of the hemodynamic significance of aortic stenosis. The simplified Bernoulli
equation used to estimate pressure gradients often misclassifies the ventricular overload caused by
aortic stenosis. The current gold standard for estimation of irreversible pressure loss is catheterization,
but this method is rarely used due to its invasiveness. Post-stenotic pressure loss is largely caused by
dissipation of turbulent kinetic energy into heat. Recent developments in magnetic resonance flow
imaging permit noninvasive estimation of TKE.
METHODS The study was approved by the local ethics review board and all subjects gave written informed
consent. Three-dimensional cine magnetic resonance flow imaging was used to measure TKE in 18 subjects (4
normal volunteers, 14 patients with aortic stenosis with and without dilation). For each subject, the peak total TKE
in the ascending aorta was compared with a pressure loss index. The pressure loss index was based on a previously
validated theory relating pressure loss to measures obtainable by echocardiography.
RESULTS The total TKE did not appear to be related to global flow patterns visualized based on
magnetic resonance–measured velocity fields. The TKE was significantly higher in patients with aortic
stenosis than in normal volunteers (p 0.001). The peak total TKE in the ascending aorta was strongly
correlated to index pressure loss (R
2
0.91).
CONCLUSIONS Peak total TKE in the ascending aorta correlated strongly with irreversible
pressure loss estimated by a well-established method. Direct measurement of TKE by magnetic
resonance flow imaging may, with further validation, be used to estimate irreversible pressure loss in
aortic stenosis. (J Am Coll Cardiol Img 2013;6:64 –71) © 2013 by the American College of Cardiology
Foundation
From the *Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco,
California; †Department of Surgery, University of California San Francisco, San Francisco, California; and the ‡Veterans
Affairs Medical Center, San Francisco, California. Dr. Dyverfeldt was supported by the Fulbright Commission, the Swedish
Heart-Lung Foundation, and the Swedish Brain Foundation. Dr. Hope was supported by an RSNA Research Scholar grant.
Dr. Tseng was supported by a grant from the Coulter Foundation and an American Heart Association Grant-in-Aid,
administered by the Northern California Institute for Research and Education using resources from the San Francisco VA
Medical Center. Dr. Saloner was supported by a VA MERIT review grant and NIH grant NS 059944. All other authors have
reported that they have no relationships relevant to the contents of this paper to disclose.
Manuscript received February 23, 2012; revised manuscript received July 3, 2012, accepted July 9, 2012.
JACC: CARDIOVASCULAR IMAGING VOL. 6, NO. 1, 2013
© 2013 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 1936-878X/$36.00
PUBLISHED BY ELSEVIER INC. http://dx.doi.org/10.1016/j.jcmg.2012.07.017
Page 1
N
oninvasive determination of irreversible
pressure loss has long been a goal of
cardiovascular imaging. Irreversible pres-
sure loss caused by energy dissipation in
post-stenotic flow is an important marker of the
hemodynamic significance of aortic stenosis. The
left ventricle has to respond with increased work to
overcome this loss of mechanical energy, resulting
in increased stress on the myocardium.
The true irreversible pressure loss (net transval-
vular pressure gradient, TPG
net
) is currently best
estimated by simultaneous catheter-based pressure
recordings in the left ventricle and the distal as-
cending aorta. However, this procedure is invasive
and therefore not used routinely. The current
method of choice in the clinical assessment of
transvalvular pressure differences is noninvasive
echocardiography. Based on an estimation of the
peak velocity in the vena contracta (v
VC
) of the post-
stenotic flow jet, the maximum drop in (static)
pressure across the valve (maximum transvalvular
pressure gradient, TPG
max
) is estimated by the
simplified Bernoulli equation in combination with
the assumption that v
VC
is much greater than the
flow velocity in the left ventricle (1):
TPG
max
4
VC
2
mm Hg
[1]
The degree to which TPG
max
represents TPG
net
depends on the amount of kinetic energy that is
dissipated distal to the vena contracta, where the
flow transitions from a laminar to a turbulent state
during systole. A portion of the kinetic energy that
is not dissipated is converted into static pressure,
resulting in so-called pressure recovery (2–9). Due
to pressure recovery, TPG
max
overestimates TPG
net
and the increased workload imposed on the left
ventricle by pressure loss (2– 6). For example, a
recent study reported that more than 20% of
TPG
max
was recovered in 16.8% of a large patient
population (6). The clinical implications of the
inability of echocardiography to account for pres-
sure recovery are frequently debated (7–9).
A noninvasive approach to the estimation of true
irreversible pressure loss could refine the diagnosis
of aortic stenosis. Consequently, several investiga-
tors have proposed indexes aimed at addressing the
discrepancy between TPG
max
and TPG
net
based on
data that can be obtained by noninvasive imaging
(8,10–13). These indexes typically take into ac-
count the severity of the sudden expansion that
occurs between the valve and the ascending aorta,
which is known to promote transition to nonlami-
nar flow. Despite being implicit and based on
assumptions about standardized transvalvular flow
patterns, such approaches have been shown to
permit noninvasive estimation of irreversible pres-
sure loss in in vitro experiments, animal models,
and specific patient groups (4,11,13,14). For exam-
ple, Garcia et al. (12,13) (see also Akins et al. [8])
added an energy loss term to the Bernoulli equation
to account for its inability to describe pressure losses
and combined that with the momentum equation.
They noted that irreversible pressure loss depends
on the flow rate (Q) and that it increases with
decreasing vena contracta area (A
VC
) and with
increasing aortic area (A
Ao
). When combining their
results for TPG
net
with the widely used approxima-
tion that v
VC
is much greater than the flow velocity
in the left ventricle in patients with aortic stenosis
(Equation 1), the following relationship is obtained
(12,13):
TPG
net
TPG
max
1
A
VC
A
Ao
2
[2]
By also taking the flow rate dependency
into account (12), this can be written as a
pressure loss index (iPL) that can be used
in patients with varying flow rates:
iPL Q
TPG
net
TPG
max
Q
1
A
VC
A
Ao
2
[3]
Direct measurement of the flow effects
responsible for irreversible pressure loss is
now possible with magnetic resonance
(MR) imaging. This potentially offers a more ap-
pealing way than iPLs to correct for gross discrep-
ancies between echocardiography and catheter-
based pressure gradients. In the transitionally
turbulent flow regime distal to the vena contracta, the
kinetic energy can be decomposed into 2 parts: the
mean kinetic energy and the turbulent kinetic energy
(TKE). The dominant cause of irreversible pressure
loss in clinically relevant aortic stenosis is viscous
dissipation of TKE into heat (15).
Recent developments in phase-contrast magnetic
resonance imaging (PC-MRI) permit noninvasive
estimation of TKE (16,17). There is an important
conceptual difference between PC-MRI velocity
and TKE mapping. Whereas conventional PC-
MRI velocity mapping estimates mean velocities
based on the phase-difference between 2 complex-
valued MR signals acquired with different motion
sensitivity, TKE estimation is achieved by exploit-
ABBREVIATIONS
AND ACRONYMS
4D 4-dimensional (3 spatial
dimensions time)
iPL pressure loss index
MR magnetic resonance
PC-MRI phase-contrast
magnetic resonance imaging
TKE turbulent kinetic energy
TPG transvalvular pressure
gradient
VENC velocity encoding
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
65
Page 2
ing the fact that the relationship in signal amplitude
between 2 such signals is related to the distribution
of velocities within a voxel (16). The amount of
motion sensitivity used in TKE mapping is prefer-
ably chosen to obtain about 50% signal drop at the
TKE values of interest (18,19). The PC-MRI TKE
mapping technique has been successfully validated
against particle image velocimetry, as well as com-
putational fluid dynamics, both in vitro and in vivo
(18–21). The feasibility of the technique for time-
resolved 3-dimensional (i.e., 4-dimensional [4D])
measurements of TKE fields in the human cardio-
vascular system has been demonstrated in a wide
range of applications (17,22).
The aim of this study was to characterize TKE in
the ascending aorta of patients with aortic stenosis and
to assess the relationship between TKE and irrevers-
ible pressure loss. TKE measurements were compared
with previously validated pressure loss indexes that can
be obtained by noninvasive imaging.
METHODS
Subjects. The study was approved by the local
ethics review board, and all subjects gave written
informed consent. A total of 18 subjects (14 aortic
stenosis patients with/without ascending aortic di-
lation, 4 normal volunteers) were enrolled (Table 1)
(23). All subjects underwent 4D PC-MRI flow
imaging, and the patients had a clinical computed
tomography and echocardiography investigation
done within 10 and 7 weeks, respectively, of the
MR study. A broad range of aortic stenoses was
represented in the study.
MR flow imaging and estimation of TKE. For the
assessment of TKE, 4D PC-MRI data were ac-
quired during free breathing on a clinical 1.5-T MR
scanner (Siemens Avanto, Siemens, Erlangen, Ger-
many) using a prospectively cardiac-gated gradient
echo sequence with asymmetrical 4-point motion
encoding, where the latter enables TKE estimation
(16). Respiratory effects were suppressed by using
navigator gating with a 7-mm acceptance window.
Other imaging parameters included TR/TE: 4.3 to
4.4/2.7 to 2.9 ms, flip angle: 8°, k-space segmenta-
tion factor: 2, parallel imaging reduction factor: 2.
The 3-dimensional field-of-view (240 to 360 240
to 360 55 to 80 mm
3
) and matrix size (96 to 144
96 to 144 22 to 32) was adjusted depending on
each subject’s anatomy to cover the thoracic aorta
with a sagittal-oblique slab orientation while main-
taining an isotropic voxel size of 2.5 2.5 2.5
mm
3
. Total scan time was about 10 to 25 min,
depending on the navigator efficiency. On the basis
of previous in vivo studies with MR TKE mapping,
we anticipated peak TKE values of about 1,200 to
1,600 J/m
3
in the patients studied here. Assuming
Table 1. Demographics and Clinical Data
Type Age/Sex
Aortic Area at Sinotubular
Junction (cm)
Aortic Valve
Area (cm
2
)*
Max/Mean Pressure
Gradient (mm Hg)†
Normal 1 24/M 5.7 4.4‡ n/a
Normal 2 36/M 6.2 4.2‡ n/a
Normal 3 32/F 4.5 3.2‡ n/a
Normal 4 28/F 4.3 3.6‡ n/a
Patient 1 90/M 6.6 0.4 215/136
Patient 2 71/M 7.1 0.9 62/38
Patient 3 51/M 14.5 0.9 70/38
Patient 4 69/M 6.2 0.82 65/39
Patient 5 72/M 8.0 1.52 62/33
Patient 6 78/M 7.5 0.64 75/48
Patient 7 69/M 9.6 0.95 52/35
Patient 8 67/M 5.7 0.78 114/71
Patient 9 64/M 9.1 1.0 56/30
Patient 10 79/M 8.1 0.79 112/68
Patient 11 74/M 9.1 0.63 56/30
Patient 12 80/M 8.5 0.5 91/57
Patient 13 55/M 11.9 4.7‡ n/a
Patient 14 57/F 8.5 3.6‡ n/a
*Aortic valve area determined by the continuity equation method. †Pressure gradients calculated from echocardiography velocity measurements using the Bernoulli
equation. Pressure gradients were not available in subjects without aortic stenosis. ‡Aortic valve area (AVA) in subjects without known aortic stenosis was estimated
from the relationship: AVA ⫽⫺2.64 0.04 (height in cm) 0.47 w(w 0 for male, w 1 for female) (23).
n/a not applicable.
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
66
Page 3
isotropic turbulence, a velocity encoding (VENC)
value of 280 cm/s provides optimum TKE sensitiv-
ity at TKE 1,400 J/m
3
(18). TKE values consid-
erably higher than the optimal value are underesti-
mated due to the Rician distribution of MR
magnitude data (18). Consequently, the VENC was
set to 280 cm/s in the patients so as to obtain good
TKE sensitivity and avoid underestimation. The fact
that this VENC setting resulted in aliasing of the
highest velocities in some of the patients was not a
concern as the TKE is based on signal amplitude and
is thus insensitive to such effects. A VENC of 200
cm/s was used in the normal volunteers.
Velocity fields were reconstructed on the scanner
using standard phase-difference algorithms and were
corrected off-line for background phase offsets and
velocity aliasing when needed. Offline reconstruction
using a MATLAB script (MathWorks, Natick, Mas-
sachusetts) written in-house was used to reconstruct
the magnitude images of the individual flow encoding
segments as needed to obtain the TKE (16,17).
The TKE per unit volume is defined as (24):
TKE
1
2
i1
3
i
2
Jm
3
[4]
where
is the fluid density and
i
is the turbulence
intensity (intensity of velocity fluctuations) in 3
mutually perpendicular directions i. A 3-directional
PC-MRI measurement carried out with asymmet-
ric 4-point motion encoding, as done here, provides
i
in 3 mutually perpendicular directions, thus
enabling the calculation of TKE (17). The relation-
ship used to compute
i
appears as (16,25):
i
1
k
v
2
S
S
i
[5]
where |S| and |S
i
| denote the amplitude of a
complex-valued MR signal acquired with zero first-
order motion sensitivity and motion sensitivity in
direction i, respectively. k
v
(i.e.,
/VENC) de-
scribes the net motion sensitivity. The fluid density
was assumed to be 1.060 kg/m
3
.
Estimation of pressure loss. The iPL described in
Equation 3 was used to assess the relationship be-
tween TKE and irreversible pressure loss. Each vari-
able of the pressure loss index was measured with the
most reliable modality available: flow rate with PC-
MRI, aortic valve area with echocardiography, and
aortic diameter with computed tomography (26–28).
Aortic area was measured at the site of the sinotubular
junction, as recommended (6,13). In the normal
volunteers, aortic area was estimated from the MR
data. Flow rate was measured at peak flow systole.
Data analysis. Using commercially available soft-
ware (EnSight 9.2, CEI, Apex, North Carolina), a
protocol was defined for visual inspection of velocity
and TKE fields in the ascending aorta of each subject.
Post-stenotic mean velocity fields were assessed by
generating short streamlines throughout the entire
ascending aorta in each time frame. Streamlines rep-
resent instantaneous velocity fields and are always
tangent to the direction of the velocity vector. Visu-
alization of the spatiotemporal distribution of TKE
was achieved by volume rendering of the TKE data in
the ascending aorta at each time frame.
The total TKE in the ascending aorta was
calculated at each time frame by integrating the
TKE over the aortic segment spanning from the
aortic valve to a standardized level midway between
the brachiocephalic and left common carotid arter-
ies where velocity fluctuations had dampened out
and TKE was low in all subjects. Geometrical
constraints were obtained by manual segmentation
of the 4D PC-MRI data. Plots of total TKE over
time were generated for each subject.
Simple linear regression was used to assess the
relationship between TKE and irreversible pressure
loss with total peak systolic TKE as the independent
variable and iPL as the dependent variable. A 2-sample
t test was performed to assess whether TKE in the
patients was higher than in the normal volunteers.
RESULTS
Plots of the total TKE in the ascending aorta of
each subject are shown in Figure 1. Peak total TKE
in the patients (TKE 3 to 52 mJ) was higher than
in the normal volunteers (TKE 3 mJ), p 0.001.
The presentation of hemodynamics was consistent
in all normal volunteers but notably diverse in the
patients. Peak total TKE occurred post-peak sys-
tole, and the primary regions of elevated TKE
appeared to be located in regions of flow jet
deceleration and wall detachment. The total TKE
did not appear to be related to global flow patterns.
Figure 2 shows data for 1 normal volunteer and 3
patients that exemplify the degree of diversity. The
patients shown in Figures 2B (Patient #9, Q
peak
420 ml/s) and 2D (Patient #7, Q
peak
750 ml/s)
both have eccentric flow directed towards the outer
wall of the ascending aorta accompanied by verti-
cally recirculating flow. Peak total TKE in these
subjects was 13 and 38 mJ, respectively. The pa-
tients shown in Figures 2C (Patient #5, Q
peak
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
67
Page 4
920 ml/s) and 2D, by contrast, have similar peak
total TKE (40 vs 38 mJ) but markedly different flow
patterns.
The peak total TKE in the ascending aorta was
strongly correlated with iPL (Fig. 3). The estimated
regression function was iPL 23.2 14.9
TKE
total
,R
2
0.91. The intercept was not signif-
icantly different from zero. The slope was signifi-
cantly different from zero, p 0.001. For patients
only, R
2
was 0.81.
DISCUSSION
The main finding of this study is that noninvasive
MR flow imaging can be used to estimate irrevers-
ible pressure loss in the ascending aorta. The strong
relationship between TKE as measured by nonin-
vasive MR flow imaging and iPL (Fig. 3) suggests
that for aortic stenosis with a given TPG
max
and
flow rate, the amount of TKE reflects the amount
of energy dissipation and thus the hemodynamic
significance of the stenosis. This accords well with
theoretical fluid mechanics considerations of pres-
sure loss in aortic stenosis (8,15).
The finding that TKE reflects irreversible pres-
sure loss has potential clinical implications. Al-
though accessibility and cost will likely keep echo-
Figure 2. Visualization of Flow Patterns and TKE
Visualization of flow patterns and turbulent kinetic energy (TKE) in 1 normal volunteer and 3 patients with aortic stenosis. For each sub-
ject, volume renderings of TKE (red to yellow color scale) at the time point of peak total TKE have been combined with streamline visu-
alization of the instantaneous velocity field at the time of peak velocity (blue color scale). Color settings were the same in all subjects.
Figure 1. Total TKE in the Ascending Aorta Over Time
Plots of the total turbulent kinetic energy (TKE
total
)inthe
ascending aorta over time (time after R-peak) for normal volun-
teers (dotted lines) and patients with aortic stenosis (solid
lines). Subjects are ordered according to peak total TKE along
the second horizontal axis. The peak total TKE in the aortic ste-
nosis patients was significantly higher than in the normal volun-
teers, p 0.001.
iPL
TKE
total
(mJ)
0 102030405060
0
100
200
300
400
500
600
700
800
900
Figure 3. Total TKE Versus Pressure Loss
Total turbulent kinetic energy (TKE
total
) in the ascending aorta
plotted against index pressure loss (iPL). The total TKE was
obtained by integrating the TKE per unit volume across the
entire ascending aorta. The iPL was calculated based on formu-
las validated by Garcia et al. (12,13). Total TKE was strongly cor-
related with iPL (R
2
0.91).
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
68
Page 5
cardiography as the first-line noninvasive method
for the evaluation of aortic stenosis, pressure gradi-
ent estimation based on the simplified Bernoulli
equation (Equation 1) often misclassifies the ven-
tricular overload caused by aortic stenosis (2– 6).
Echocardiography-based methods for estimation of
irreversible energy or pressure loss have shown
promising results but are not yet used routinely
(27). These methods are limited by assumptions
about standardized flow patterns and are reportedly
confounded by the presence of eccentric flow
(29,30). Variability related to the echocardiography
measurement of aortic diameter may be another
limiting factor (27). Catheterization-based pressure
measurements are invasive and not used routinely.
TKE measurements by MR flow imaging are non-
invasive and do not rely on assumptions about
standardized flow patterns. By combining Equation
3 with the estimated regression function, TPG
net
can be calculated from the total TKE according to
TPG
net
TPG
max
Q
2
23.2 14.9 TKE
total
[mm Hg]. In this way, TKE data could be incorpo-
rated into the clinical evaluation of aortic stenosis and
be directly compared to pressure estimates obtained
with echocardiography and catheterization.
This new approach for the direct measurement of
the flow effects responsible for pressure loss is
relatively simple to perform. The only processing
step needed to obtain the total TKE (Figs. 1 and 3)
is segmentation of the ascending aorta. In the
present study, segmentation was done manually,
and processing time was 5 min per patient. Auto-
matic segmentation of 4D PC-MRI is expected to
be realized in the near future (31). Estimation of
TKE requires measurements of turbulence intensity
in 3 mutually perpendicular directions (Equation
4), and the calculation of total TKE in the aorta
additionally requires volumetric TKE measure-
ments. Currently, MR imaging is the only method
that can be used in vivo that is capable of obtaining
such comprehensive data on energy-dissipating flow
effects. Invasive hot film/wire anemometry and
perivascular Doppler ultrasound have the ability to
provide some information on turbulence intensity in
vivo (32,33). Noninvasive echocardiography methods
have also been proposed and may be able to contribute
with first-order estimations of nonlaminar flow effects
associated with aortic stenosis (34,35).
Others have shown that relative pressure fields
can be computed from 4D PC-MRI velocity data
(36,37). Although this is valuable for the investiga-
tion of normal human cardiovascular pressure dy-
namics, the underlying equations (Navier-Stokes
equations) use only the mean velocity, or accelera-
tion, field as input and do not take energy dissipa-
tion into account. By extending TKE measure-
ments to obtain the so-called turbulence stress
tensor, which is the subject of ongoing research
(38), pressure calculations may be extended to take
energy dissipation into account. This would further
enhance the noninvasive imaging armamentarium
for in-depth investigations of energy-dissipating
flows and potentially enable noninvasive pressure
field mapping in aortic stenosis.
Study limitations. A limitation of this study was the
lack of a true reference for transvalvular pressure
loss. Simultaneous recordings of pressure in the left
ventricle and the distal ascending aorta using high-
fidelity pressure catheters is currently considered the
gold standard for the estimation of transvalvular
pressure loss. Although the approach used here for
calculation of the pressure loss index has previously
been validated against catheter-based methods and
is well established (68,12,13), more studies are
needed to further assess the relationship between
TKE and irreversible pressure loss. Catheter-
measured pressure loss was available in 1 of our
patients. This patient had a measured TPG
net
/
TPG
max
ratio of 0.89, which corresponded well to
the estimated TPG
net
/TPG
max
ratio of 0.85. A
limitation of the 4D PC-MRI technique used here
is that relatively long scan times are needed to
obtain adequate spatiotemporal resolution and cov-
erage. However, advances in accelerated 4D PC-
MRI indicate that total scan times of 10 min are
likely in the near future (39,40).
CONCLUSIONS
This study used a novel MR flow imaging method
to measure the total TKE in the ascending aorta of
patients with aortic stenosis and assessed the rela-
tionship between TKE and irreversible pressure
loss. Peak total TKE in the ascending aorta corre-
lated strongly with a iPL calculated based on
well-established methods. Direct measurement of
TKE may, with further validation, be used to
estimate irreversible pressure loss in aortic stenosis.
Acknowledgment
The authors thank Tino Ebbers for sharing his
group’s post-processing tools.
Reprint requests and correspondence: Dr. Petter Dyverfeldt,
IMH/KVM/KlinFys, Linköping University, SE-581 83,
Linköping, Sweden. E-mail: petter.dyverfeldt@liu.se.
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
69
Page 6
REFERENCES
1. Hatle L, Angelsen BA, Tromsdal A.
Non-invasive assessment of aortic ste-
nosis by Doppler ultrasound. Br
Heart J 1980;43:284–92.
2. Ohlsson J, Wranne B. Noninvasive
assessment of valve area in patients
with aortic stenosis. J Am Coll Car-
diol 1986;7:501– 8.
3. Laskey WK, Kussmaul WG. Pressure
recovery in aortic valve stenosis. Cir-
culation 1994;89:116 –21.
4. Schobel WA, Voelker W, Haase KK,
Karsch KR. Extent, determiants and
clinical importance of pressure recov-
ery in patients with aortic valve steno-
sis. Eur Heart J 1999;20:1355–63.
5. Gjertsson P, Caidahl K, Svensson G,
Wallentin I, Bech-Hanssen O. Im-
portant pressure recovery in patients
with aortic stenosis and high Doppler
gradients. Am J Cardiol 2001;88:
139 44.
6. Bahlmann E, Cramariuc D, Gerdts E,
et al. Impact of pressure recovery on
echocardiographic assessment of
asymptomatic aortic stenosis: a SEAS
substudy. J Am Coll Cardiol Img
2010;3:555–62.
7. Rahimtoola SH. Determining that
aortic valve stenosis is severe: back-to-
the-future. J Am Coll Cardiol Img
2010;3:563–6.
8. Akins CW, Travis B, Yoganathan
AP. Energy loss for evaluating heart
valve performance. J Thorac Cardio-
vasc Surg 2008;136:820–33.
9. Baumgartner H, Otto CM. Aortic
stenosis severity: do we need a new
concept? J Am Coll Cardiol 2009;54:
1012–3.
10. Clark C. The fluid mechanics of aortic
stenosis. I. Unsteady flow experi-
ments. J Biomech 1976;9:567–73.
11. Voelker W, Reul H, Stelzer T,
Schmidt A, Karsch KR. Pressure re-
covery in aortic stenosis: an in vitro
study in a pulsatile flow model. J Am
Coll Cardiol 1992;20:1585–93.
12. Garcia D, Pibarot P, Dumesnil JG,
Sakr F, Durand LG. Assessment of
aortic valve stenosis severity: a new
index based on the energy loss con-
cept. Circulation 2000;101:765–1.
13. Garcia D, Dumesnil JG, Durand LG,
Kadem L, Pibarot P. Discrepancies
between catheter and Doppler esti-
mates of valve effective orifice area can
be predicted from the pressure recov-
ery phenomenon: practical implica-
tions with regard to quantification of
aortic stenosis severity. J Am Coll
Cardiol 2003;41:435– 42.
14. Baumgartner H, Steffenelli T, Nie-
derberger J, Schima H, Maurer G.
‘Overestimation’ of catheter gradients
by Doppler ultrasound in patients
with aortic stenosis: a predictable
manifestation of pressure recovery.
J Am Coll Cardiol 1999;33:1655– 61.
15. Clark C. Energy losses in flow
through stenosed valves. J Biomech
1979;12:737–46.
16. Dyverfeldt P, Sigfridsson A, Kvitting
JPE, Ebbers T. Quantification of in-
travoxel velocity standard deviation
and turbulence intensity by generaliz-
ing phase-contrast MRI. Magn Reson
Med 2006;56:850 8.
17. Dyverfeldt P, Kvitting JPE, Sig-
fridsson A, Engvall J, Bolger AF,
Ebbers T. Assessment of fluctuating
velocities in disturbed cardiovascular
blood flow: in vivo feasibility of gen-
eralized phase-contrast MRI. J Magn
Reson Imaging 2008;28:655–3.
18. Dyverfeldt P, Gardhagen R, Sig-
fridsson A, Karlsson M, Ebbers T. On
MRI turbulence quantification. Magn
Reson Imaging 2009;27:913–22.
19. Elkins CJ, Alley MT, Saetran L, Ea-
ton JK. Three-dimensional magnetic
resonance velocimetry measurements
of turbulence quantities in complex
flow. Exp Fluids 2009;46:285–96.
20. Petersson S, Dyverfeldt P, Gardhagen
R, Karlsson M, Ebbers T. Simulation
of phase contrast MRI of turbulent
flow. Magn Reson Med 2010;64:
1039 46.
21. Arzani A, Dyverfeldt P, Ebbers T,
Shadden SC. In vivo validation of
numerical prediction for turbulence
intensity in an aortic coarctation. Ann
Biomed Eng 2012;40:860–70.
22. Dyverfeldt P, Kvitting JPE, Carlhall
CJ, et al. Hemodynamic aspects of
mitral regurgitation assessed by gen-
eralized phase-contrast MRI . J Magn
Reson Imaging 2011;33:582–8.
23. Haimerl J, Freitag-Krikovic A,
Rauch A, Sauer E. Quantification of
aortic valve area and left ventricular
muscle mass in healthy subjects and
patients with symptomatic aortic
valve stenosis by MRI. Z Kardiol
2005;94:173–81.
24. Mathieu J, Scott J. An Introduction to
Turbulent Flow. Cambridge, UK:
Cambridge University Press, 2000.
25. Gao JH, Gore JO. Turbulent flow
effects on NMR imaging: measure-
ment of turbulent intensity. Med Phys
1991;18:1045–51.
26. Baumgartner H, Hung J, Bermejo JB,
et al. Echocardiographic assessment of
valve stenosis: EAE/ASE recommen-
dations for clinical practice. Eur J
Echo 2009;10:1–25.
27. Pennel DJ, Sechtem UP, Higgins CB,
et al. Clinical indications for cardio-
vascular magnetic resonance (CMR):
Consensus Panel report. J Cardiovasc
Magn Reson 2004;6:727–65.
28. Hiratzka LF, Bakris GL, Beckman
JA, et al. 2010 ACCF/AHA/AATS/
ACR/ASA/SCA/SCAI/SIR/STS/
SVM guidelines for the diagnosis and
management of patients with thoracic
aortic disease. A report of the Amer-
ican College of Cardiology Founda-
tion/American Heart Association
Task Force on Practice Guidelines,
American Association for Thoracic
Surgery, American College of Radiol-
ogy, American Stroke Association,
Society of Cardiovascular Anesthesi-
ologists, Society for Cardiovascular
Angiography and Interventions, Soci-
ety of Interventional Radiology, Soci-
ety of Thoracic Surgeons, and Society
for Vascular Medicine. J Am Coll
Cardiol 2010;55:e27–129.
29. van Pelt R, Nguyen H, ter Haar
Romeny B, Vilanova A. Automated
segmentation of blood-flow regions
in large thoracic arteries using 3D-
cine PC-MRI measurements. Int
J Comput Assist Radiol Surg 2012;
7:217–24.
30. VanAuker MD, Chandra M, Shirani
J, Strom JA. Jet eccentricity: a mis-
leading source of agreement between
Doppler/catheter pressure gradients in
aortic stenosis. J Am Soc Echocar-
diogr 2001;14:853– 62.
31. Richards KE, Deserranno D, Donal
E, Greenberg NL, Thomas JD, Gar-
cia MJ. Influence of structural geom-
etry on the severity of bicuspid aortic
stenosis. Am J Physiol Heart Circ
Physiol 2004;287:H1410 6.
32. Nerem RM, Seed WA. An in vivo
study of aortic flow disturbances. Car-
diovasc Res 1972;6:1–14.
33. Nygaard H, Hasenkam JM, Pedersen
EM, Kim WY, Paulsen PK. A new
perivascular multi-element pulsed
Doppler ultrasound system for in vivo
studies of velocity fields and turbulent
stresses in large vessels. Med Biol Eng
Comp 1994;32:55– 62.
34. Isaaz K, Bruntz JF, Da Costa A, et al.
Noninvasive quantitation of blood
flow turbulence in patients with aortic
valve disease using online digital com-
puter analysis of Doppler velocity
data. J Am Soc Echocardiogr 2003;16:
965–74.
35. Thorne ML, Rankin RN, Steinman
DA, Holdsworth DW. In vivo Dopp-
ler ultrasound quantification of turbu-
lence intensity using a high-pass fre-
quency filter method. Ultrasound
Med Biol 2010;36:761–71.
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
70
Page 7
36. Tyszka JM, Laidlaw DH, Asa JW,
Silverman JM. Three-dimensional,
time-resolved (4D) relative pressure
mapping using magnetic resonance
imaging. J Magn Reson Imaging
2000;12:321–9.
37. Ebbers T, Wigstrom L, Bolger AF,
Wranne B, Karlsson M. Noninvasive
measurement of time-varying three-
dimensional relative pressure fields
within the human heart. J Biomech
Eng 2002;124:288 –93.
38. Binter C, Knobloch V, Manka R,
Sigfridsson A, Kozerke S. Bayesian
multipoint velocity encoding for con-
current flow and turbulence mapping.
Magn Reson Med 2012 Jun 14 [E-
pub ahead of print], doi: 10.1002/
mrm.24370.
39. Gu T, Korosec FR, Block WF, et al.
PC VIPR: a high-speed 3D phase-
contrast method for flow quantifica-
tion and high-resolution angiography.
Am J Neuroradiol 2005;26:743–9.
40. Sigfridsson A, Petersson S, Carlhall
CJ, Ebbers T. Four-dimensional flow
MRI using spiral acquisition. Magn
Reson Med 2012;68:1065–73.
Key Words: aortic stenosis y
magnetic resonance imaging y
pressure loss y transvalvular
pressure gradient y turbulent
kinetic energy.
JACC: CARDIOVASCULAR IMAGING, VOL. 6, NO. 1, 2013
JANUARY 2013:64 –71
Dyverfeldt et al.
Turbulent Kinetic Energy in Aortic Stenosis
71
Page 8
  • Source
    • "Recent research efforts provide compelling evidence that the analysis of blood flow dynamics can improve the management of cardiovascular diseases through flow-derived biomarkers. This is demonstrated by the analysis of the vortical flow in the ventricle (Pedrizzetti et al., 2014) or in the aorta (Bissell et al., 2013), the influence of wall shear stress on the endothelial function (Chiu and Chien, 2011), the estimation of flow energetics (Barker et al., 2014) and turbulence (Dyverfeldt et al., 2013), and the extraction of pressure gradients and its components (Lamata et al., 2014). A landmark recent study has provided initial evidence of the suitability of PC-MRI pressure estimation to assess the severity of aortic coarctation (Riesenkampff et al., 2014). "
    Full-text · Dataset · Sep 2015
  • Source
    • "Recent research efforts provide compelling evidence that the analysis of blood flow dynamics can improve the management of cardiovascular diseases through flow-derived biomarkers. This is demonstrated by the analysis of the vortical flow in the ventricle (Pedrizzetti et al., 2014) or in the aorta (Bissell et al., 2013), the influence of wall shear stress on the endothelial function (Chiu and Chien, 2011), the estimation of flow energetics (Barker et al., 2014) and turbulence (Dyverfeldt et al., 2013), and the extraction of pressure gradients and its components (Lamata et al., 2014). A landmark recent study has provided initial evidence of the suitability of PC-MRI pressure estimation to assess the severity of aortic coarctation (Riesenkampff et al., 2014). "
    [Show abstract] [Hide abstract] ABSTRACT: Pressure gradient - or pressure difference over length - is an accepted clinical biomarker for cardiovascular disease conditions such as aortic coarctation. Currently, measurements of pressure gradients in the clinic rely on invasive techniques (catheterisation), prompting development of non-invasive estimates based on blood flow. In this work, we propose a non-invasive estimation procedure deriving pressure difference from the work-energy equation for a Newtonian fluid. Spatial and temporal convergence is demonstrated on in silico Phase Contrast Magnetic Resonance Image (PC-MRI) phantoms with steady and transient flow fields. The method is also tested on an image dataset generated in silico from a 3D patient-specific Computational Fluid Dynamics (CFD) simulation. The performance is compared to existing approaches based on steady and unsteady Bernoulli formulations as well as the pressure Poisson equation. The new technique shows good accuracy, robustness to noise, and robustness to the image segmentation process, illustrating the potential of this approach for non-invasive pressure di↵erence estimation.
    Full-text · Article · Sep 2015 · Medical Image Analysis
  • Source
    • "In addition to the flexible retrospective quantification of conventional flow parameters, 4D Flow CMR allows for the visualization of multidirectional flow features and alterations of these associated with cardiovascular disease50515253 . Previously reported results include the application of 4D Flow CMR for the analysis of blood flow in the ventricles54555657585960616263 and atria64656667 of the heart, heart valves [3, 43, 68, 69], aorta [41,6970717273747576777879808182, main pulmonary vessels83848586, carotid arteries87888990, large intracranial arteries and veins9192939495969798, arterial and portal venous systems of the liver [46, 85,99100101, peripheral arteries [102] and renal arteries [103, 104]. The intuitive flow visualizations that 4D Flow CMR offers have already found utility in several clinical studies. "
    [Show abstract] [Hide abstract] ABSTRACT: Pulsatile blood flow through the cavities of the heart and great vessels is time-varying and multidirectional. Access to all regions, phases and directions of cardiovascular flows has formerly been limited. Four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) has enabled more comprehensive access to such flows, with typical spatial resolution of 1.5×1.5×1.5 - 3×3×3 mm(3), typical temporal resolution of 30-40 ms, and acquisition times in the order of 5 to 25 min. This consensus paper is the work of physicists, physicians and biomedical engineers, active in the development and implementation of 4D Flow CMR, who have repeatedly met to share experience and ideas. The paper aims to assist understanding of acquisition and analysis methods, and their potential clinical applications with a focus on the heart and greater vessels. We describe that 4D Flow CMR can be clinically advantageous because placement of a single acquisition volume is straightforward and enables flow through any plane across it to be calculated retrospectively and with good accuracy. We also specify research and development goals that have yet to be satisfactorily achieved. Derived flow parameters, generally needing further development or validation for clinical use, include measurements of wall shear stress, pressure difference, turbulent kinetic energy, and intracardiac flow components. The dependence of measurement accuracy on acquisition parameters is considered, as are the uses of different visualization strategies for appropriate representation of time-varying multidirectional flow fields. Finally, we offer suggestions for more consistent, user-friendly implementation of 4D Flow CMR acquisition and data handling with a view to multicenter studies and more widespread adoption of the approach in routine clinical investigations.
    Full-text · Article · Aug 2015 · Journal of Cardiovascular Magnetic Resonance
Show more