Download Left Ventricular Filling Pressure Assessment Using Left Atrial Transit

Left Ventricular Filling Pressure Assessment Using Left
Atrial Transit Time by Cardiac Magnetic
Resonance Imaging
Jie J. Cao, MD, MPH; Yi Wang, ScD; Jeannette McLaughlin, RN; Elizabeth Haag, RN;
Peter Rhee, MD; Michael Passick, RDCS; Rena Toole, RDCS; Joshua Cheng, RT;
Andrew D. Berke, MD; Justine Lachman, MD; Nathaniel Reichek, MD
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Background—Left atrial (LA) size and function reflect left ventricular (LV) hemodynamics. In the present study, we
developed a novel method to determine LA circulation transit time (LATT) by MRI and demonstrated its close
association with LV filling pressure.
Methods and Results—All subjects were prospectively recruited and underwent contrast-enhanced MR dynamic imaging.
Mean LATT was determined as the time for contrast to transit through the LA during the first pass. In an invasive study
group undergoing clinically indicated cardiac catheterization (n⫽25), LATT normalized by R-R interval (nLATT) was
closely associated with LV early diastolic pressure (r⫽0.850, P⫽0.001), LV end-diastolic pressure (r⫽0.910,
P⬍0.001), and mean diastolic pressure (r⫽0.912, P⬍0.001). In a larger noninvasive group (n⫽56), nLATT was
prolonged in patients with LV systolic dysfunction (n⫽47) (10.1⫾3.0 versus 6.6⫾0.7 cardiac cycles in normal control
subjects, n⫽9; P⬍0.001). Using a linear regression equation derived from the invasive group, noninvasive subjects were
divided into 3 subgroups by estimated LV end-diastolic pressure: ⱕ10 mm Hg, 11 to 14 mm Hg, and ⱖ15 mm Hg.
There were graded increases from low to high LV end-diastolic pressure subgroups in echocardiographic mitral medial
E/e⬘ ratio: 9⫾5, 11⫾4, and 13⫾3 (P⫽0.023); in B-type natriuretic peptide (interquartile range): 44 (60) pg/mL, 87 (359)
pg/mL, and 371 (926) pg/mL (P⫽0.002); and in N-terminal pro–B-type natriuretic peptide: 57 (163) pg/mL, 208 (990)
pg/mL, and 931 (1726) pg/mL (P⫽0.002), demonstrating the ability of nLATT to assess hemodynamic status.
Conclusions—nLATT by cardiac MR is a promising new parameter of LV filling pressure that may provide graded
noninvasive hemodynamic assessment. (Circ Cardiovasc Imaging. 2011;4:130-138.)
Key Words: MRI 䡲 left atrium 䡲 circulation transit time 䡲 hemodynamics 䡲 natriuretic peptides
T
using modern technologies such as nuclear scintigraphy,10,11
computed tomography,12 and magnetic resonance (MR) angiography.13 To date, studies of segmental circulation time
have been limited to measurements of transit time between
pulmonary artery and left atrium (LA),8 pulmonary artery and
ascending aorta13,14 or to femoral artery,15 or antecubital vein
and peripheral tissues.16 –18 In contrast, the clinical correlates
of LA mean transit time have not been explored.
The timing and extent of LA emptying are closely related to
LV diastolic filling pressure. In the presence of increased LV
filling pressure, the deceleration time of early diastolic transmitral inflow is decreased as is the inflow during atrial contraction.19 –21 Hence, we hypothesized that prolongation of LA
circulation transit time (LATT) was a marker of increased LV
filling pressure. We have developed a novel method for determination of LATT using dynamic cardiac MRI and examined its
relationship to LV filling pressure in a group of patients
he major cause of congestive heart failure is an interrelated set of hemodynamic derangements including elevated left ventricular (LV) filling pressure. Although invasive
hemodynamics remain the gold standard for assessment of
hemodynamic and circulatory function in patients with heart
failure, a variety of noninvasive techniques, including Doppler echocardiography1–3 and biomarkers,4 – 6 have proven
helpful. However, none offers a graded direct assessment of
hemodynamic or circulatory impairment.
Clinical Perspective on p 138
Circulatory transit time measures the time taken for blood
to flow between or within circulatory structures. It was
recognized more than half a century ago, using invasive
methods, that global circulation time, from systemic vein to
systemic artery, is prolonged in patients with heart failure.7–9
More recent studies have confirmed earlier observations
Received July 20, 2010; accepted January 13, 2011.
From St Francis Hospital, Research Department; State University of New York, Stony Brook, Division of Cardiology and Department of Biomedical
Engineering, NY.
Correspondence to Jie Jane Cao, MD, St Francis Hospital, State University of New York, Stony Brook, 100 Port Washington Blvd, Roslyn, NY 11576.
E-mail [email protected]
© 2011 American Heart Association, Inc.
Circ Cardiovasc Imaging is available at http://circimaging.ahajournals.org
130
DOI: 10.1161/CIRCIMAGING.110.959569
Cao et al
Figure 1. Schematic illustration of the recruitment that included
an invasive and a noninvasive cohort. All subjects underwent
research cardiac MRI and were tested for biomarkers. Additionally, the invasive cohort underwent clinically indicated cardiac
catheterization and the noninvasive cohort underwent research
echocardiography.
undergoing invasive heart catheterization. This relationship was
then used to characterize the association of established noninvasive markers of hemodynamics with LATT in a larger noninvasive cohort with LV systolic dysfunction (LVSD).
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Methods
Participant Recruitment
The study protocol was approved by the St Francis Hospital
Institutional Review Board, and written informed consent was
obtained from all participants. Participants were recruited prospectively and divided into invasive and noninvasive groups. Whereas all
subjects underwent cardiac MR (CMR) and blood testing for
biomarkers, those in the noninvasive arm also had an echocardiogram. The detailed recruitment scheme is illustrated in Figure 1. The
invasive group consisted of patients who were undergoing clinically
indicated left and right heart catheterization and consented to a
research CMR study on the same day. Subjects with severe mitral
regurgitation were excluded. In the noninvasive group, normal
control subjects were recruited if they were free of hypertension,
diabetes mellitus, and cardiovascular disease, whereas patients with
LVSD were recruited from those referred for clinical CMR for
evaluation of cardiomyopathy. Additional subjects with LVSD were
recruited from the cardiac catheterization laboratory. Consistent with
the invasive group, subjects with severe mitral regurgitation were
excluded. The inclusion criteria for LVSD subjects were LV ejection
fraction ⬍50% with New York Heart Association functional class I
to III with or without clinical history of decompensated heart failure.
Exclusions were the same for all participants, including impaired
renal function with glomerular filtration rate ⬍45 mL/min/1.73 m2;
claustrophobia; pacemaker/defibrillator implantation; or other metallic hazards. All participants completed a questionnaire for demographic information and medical history. Clinical charts were reviewed to confirm the cardiovascular history in all subjects.
Cardiac MRI
All participants underwent CMR in a 1.5-T Avanto scanner (Siemens, Malvern, PA) with an 8-element phased array surface coil.
Cardiac volumes and systolic function were assessed using balanced
steady-state free-precession (SSFP) cine imaging with retrospective
ECG gating. Images were acquired during breath-hold at inspiration
in a stack of short-axis planes (8-mm thickness with 2-mm gap) and
3 long-axis planes (2-, 3-, and 4-chamber views) with the following
parameters: echo time, 1.3 ms; repetition time, 3.1 ms; flip angle,
70°; and average in-plane resolution of 1.3⫻1.3 mm2. To determine
LATT, dynamic images were acquired every cardiac cycle over at
least 50 and up to 100 cardiac cycles in a sagittal plane, where the
main pulmonary artery and LA were well defined (Figure 2) during
the infusion of gadopentetate dimeglumine (Magnevist, Schering
AG, Berlin, Germany) (0.01 mmol/kg) at 6 mL/s followed with a
15-mL saline flush. An ECG-gated saturation recovery SSFP sequence was used with an inversion time of 90 ms; field of view,
500 mm; echo time, 0.92 ms; imaging acquisition window, 160 ms
LA Transit Time and LV Filling Pressure
131
Figure 2. MR contrast-enhanced dynamic images were acquired
from a patient with left ventricle systolic dysfunction in a sagittal
view in which main pulmonary artery (PA), right ventricle (RV), and
left atrium (LA) were well defined. The sequential appearance of
the gadolinium contrast is shown with the peak signal intensity in
the PA first (left panel), followed by the peak signal intensity in LA
(right panel).
per slice; slice thickness, 15 mm; and flip angle, 50°. Parallel
imaging was applied with an acceleration factor of 2.
Echocardiography
A transthoracic color Doppler echocardiogram with tissue Doppler
imaging was obtained by experienced research technologists in 9 normal
control subjects and 38 patients with LVSD in the noninvasive group
within 2 hours of cardiac MRI using Philips IE33 echocardiographs
(Andover, MA). Six patients declined echocardiographic examination.
Echocardiograms were not obtained in the invasive cohort. Mitral
inflow velocities of the peak E and A waves were measured using
pulsed Doppler. Tissue Doppler imaging was acquired from the medial
and lateral mitral annulus and E/e⬘ ratios calculated using the mitral
inflow E and the peak velocity of the tissue e⬘ wave of the medial and
lateral mitral annulus.22 E/e⬘ ratios were not available in 2 individuals in
whom Doppler acquisitions were incomplete. The peak pulmonary
artery systolic pressure was estimated using the peak velocity of the
tricuspid regurgitant jet and right atrial pressure estimates based on
phasic respiratory changes in the inferior vena cava.23 There were 5
cases in which tricuspid regurgitation was not present and pulmonary
artery systolic pressure could not be estimated.
BNP Assays
A blood sample was collected for B-type natriuretic peptide (BNP)
or N-terminal proBNP (NT proBNP) before cardiac MRI for all
study subjects (n⫽81). BNP was measured using a chemiluminescent microparticle immunoassay in EDTA plasma (ARCHITECT
BNP assay by Abbott Laboratory, Abbott Park, IL). NT proBNP was
measured using an electrochemiluminescence immunoassay (Roche
Diagnostics, Indianapolis, IL).
Invasive Hemodynamic Testing by Right and Left
Heart Catheterization
The right and left heart catheterization were performed under
fluoroscopic guidance via a femoral vein and artery, respectively,
following standard clinical protocols. Hemodynamic measurements
included pressure recordings in the right atrium, main pulmonary
artery, LV, and pulmonary capillary wedge (PCWP) positions. In
addition, blood oxygen saturation in the pulmonary artery and aorta
were measured, and oxygen content was estimated so that cardiac
output and cardiac index by the assumed Fick method could be
calculated and pulmonary vascular resistance determined. Hemodynamic tracings were recorded and stored electronically. Two experienced cardiologists reviewed tracings and the values of each
hemodynamic measurement were determined by consensus. Three
variables were derived from LV filling pressure tracings: early LV
diastolic pressure (LVDP), LV end-diastolic pressure (LVEDP), and
mean LV filling pressure, taken as the average of early and end LV
filling pressure. Inspiratory hemodynamic values were used because
MR dynamic imaging was acquired during an inspiratory breathhold. The average time between invasive catheterization and CMR
132
Circ Cardiovasc Imaging
March 2011
Table 1.
Characteristics of invasive group (Nⴝ25)
Hemodynamic Variables
Mean right atrial pressure (mmHg)
Mean⫾SD
7.1⫾.6.0
Pulmonary artery systolic pressure (mmHg)
28.8⫾13.3
Pulmonary artery diastolic pressure (mmHg)
13.4⫾9.5
Mean pulmonary artery pressure (mmHg)
18.5⫾10.5
Pulmonary artery oxygen saturation (%)
71.3⫾7.3
Arterial oxygen saturation (%)
94.8⫾3.4
Mean pulmonary capillary wedge pressure (mmHg)
10.2⫾5.8
Left ventricular systolic pressure (mmHg)
127⫾16
Left ventricular early diastolic pressure (mmHg)
7.4⫾5.3
Left ventricular end diastolic pressure (mmHg)
13.7⫾6.9
Left ventricular mean diastolic pressure (mmHg)
10.6⫾5.9
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Cardiac output (L/min)
5.9⫾1.3
Cardiac index (L/min/m2)
3.0⫾0.6
Pulmonary vascular resistance (dyn 䡠 s/cm5)
128⫾120
Cardiac Volumes and Systolic Function by CMR
Figure 3. Mean circulation transit time in the left atrium (LATT)
was determined from the first pass portion of the time intensity
curves. A, Normal individual with R-R–normalized LATT (nLATT)
5.2 cardiac cycles. B, Patient with LVSD with nLATT 14.7 cardiac cycles.
was 5 hours and 28 minutes. There was no hemodynamic instability
between studies among the participants.
LVEDV (ml/m2)
89⫾40
LVESV (ml/m2)
62⫾72
LV stroke volume (ml)
72⫾19
LVEF (%)
46⫾16
2
LV mass (g/m )
62⫾23
RVEDV (ml/m2)
65⫾19
RVESV (ml/m2)
34⫾20
RV stroke volume (ml)
63⫾14
RVEF (%)
52⫾15
RV mass (g/m2)
18⫾6
LA volume (ml/m2)
45⫾17
Circulation Transit Times
Image Analysis
Left and right ventricular end-diastolic volume (EDV), end-systolic
volume (ESV), ejection fraction (EF), and myocardial mass were
assessed from cine CMR images using MASS software (Medis,
Leiden, The Netherlands). LA volume was assessed using the
biplane area-length method as 0.85⫻A1⫻A2/L, where A1 and A2
were the areas by planimetry in the 4- and 2-chamber views and L
was the left atrial length perpendicular to the mitral annular plane in
the 4-chamber view.24 All measurements were normalized to body
surface area. The CMR contrast-enhanced dynamic image series were
analyzed to evaluate LATT using MASS software. Small circular
regions of interest approximately 5 mm in diameter were placed at the
center of the LA at a phase in which the anatomic structure was best
defined by the contrast enhancement, then propagated through the image
series to assess serial contrast signal intensity and a graph of signal
intensity over time constructed. LATT was measured only during the
first pass (Figure 3). The calculation of LATT was based on indicator
dilution theory25 using the following equation:
冕
冕
⬁
tS(t)dt
LATT ⫽
0
⬁
S(t)dt
0
where S(t) is the signal from the time-intensity curve and t is the time
at which contrast signal is measured passing through the LA. A
derivative between 2 adjacent data points on the time-intensity curve
was calculated to determine the starting and ending points of the
LATT (seconds)
8.8⫾2.9
nLATT (cardiac cycles)
9.8⫾3.5
Biomarkers
BNP (pg/ml)
NT proBNP (pg/ml)
77 (140)*
185 (516)*
RV indicates right ventricle.
*Nonnormally distributed continuous variables are shown as median (interquartile range).
contrast signal curve during first pass. All data points before the
starting point were averaged to obtain a baseline signal, which was
subtracted from the signal at each point on the time intensity curve.
The signals beyond the ending point were attributed to recirculation
and discarded from analysis. All data points were then fitted to the
equation to calculate LATT using a custom Matlab (MathWorks Inc,
Natick, MA) program. A normalized LATT (nLATT) was calculated
as the measured LATT divided by the R-R interval in seconds.
Statistical Analysis
Statistical analyses were performed using SPSS 11.0 for Windows
(SPSS Inc, Chicago, IL). The Student t test and ␹2 test were used to
compare the continuous and categorical variables, respectively,
between normal control subjects and patients with LVSD. Probability values of t tests were determined assuming equal variances if the
Levene test for equality of variance was not statistically significant
(Pⱖ0.05). Otherwise, probability values were determined without
assuming equal variances. Continuous variables with nonnormal
distributions were compared as medians using the Mann-Whitney
Cao et al
LA Transit Time and LV Filling Pressure
133
Table 2. Correlation of LA Transit Times With Invasive
Hemodynamic Measurements
Hemodynamic Variables
(n⫽25)
LATT
r
Mean right atrial pressure
0.624
0.01
0.359
0.93
PA systolic pressure
0.588
0.02
0.512
0.09
Mean PA pressure
0.620
0.01
0.528
0.07
PA oxygen saturation
⫺0.647
0.01
⫺0.605
0.01
Arterial oxygen saturation
⫺0.594
0.02
⫺0.463
0.23
Mean PCWP
0.613
0.01
0.783
⬍0.01
LVDP
0.607
0.01
0.850
⬍0.01
LVEDP
0.764
⬍0.01
0.910
⬍0.01
Mean LV diastolic pressure
0.718
⬍0.01
0.912
⬍0.01
⫺0.775
⬍0.01
⫺0.717
⬍0.01
0.576
0.03
0.394
Cardiac index
Pulmonary vascular resistance
P Value*
nLATT
r
P Value*
0.560
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
PA indicates pulmonary artery.
*P values shown are adjusted for multiple testing by Bonferroni correction.
test. Pearson correlations were performed to assess the associations
of hemodynamic indices and LATT. Probability values were further
adjusted by Bonferroni correction for multiple testing in Pearson
correlations. Using the linear regression equation relating nLATT to LV
filling pressure in the invasive group, LVEDP was estimated in the
noninvasive group. Predicted LVEDP was then divided into 3 subgroups with LVEDP ⱕ10 mm Hg, 11 to 14 mm Hg, and ⱖ15 mm Hg.
Cardiac structure and function and echocardiographic indices were
compared across the subgroups to test for trends by comparing means
using polynomial linear comparisons in ANOVA. Biomarker levels,
which did not have a normal distribution, were compared using the
Kruskal-Wallis test. Probability values ⬍0.05 were considered statistically significant for all analyses. The limits of agreement between
estimated LV filling pressure from nLATT, measured LV filling
pressure, and estimated PCWP from E/e⬘ were calculated in pairwise
Bland-Altman analyses. To determine interobserver and intraobserver
variability, nLATT was analyzed in 9 randomly selected cases by 2
experienced analysts blinded to subjects’ clinical status. Intraclass
correlation coefficients and paired t tests revealed strong agreement
between readers (intraclass correlation⫽0.97, P⫽0.455 for paired t test)
and within readers (intraclass correlation ⫽0.99, P⫽0.362 for paired t
test). The average difference was 0.11 seconds (1.4%) between readers
and 0.08 seconds (1.1%) within readers.
Results
Relationship of LA Circulation Transit Time and
Invasive Hemodynamic Measurements
In the invasive group, there were 25 patients undergoing
clinically indicated invasive hemodynamic measurements.
The average age was 58⫾16 years, and 13 were female. All
had clinically suspected congestive heart failure. Eight had at
least moderate valvular dysfunction including 3 with moderate mitral regurgitation, 2 with moderate tricuspid regurgitation, and 3 with moderate to severe aortic stenosis. The
average hemodynamic indices, ventricular volumes, and systolic function and biomarker levels are shown in Table 1.
There were 8 patients with LVEF below 50%. The Pearson
correlation demonstrated closer association of LV filling
pressure with nLATT than with LATT (Table 2), despite lack
of a significant correlation between LATT and heart rate
(P⫽0.214). There were strong correlations between nLATT
and LV early diastolic pressure (r⫽0.850, P⫽0.001), LVEDP
(r⫽0.910, P⬍0.001), mean LV diastolic pressure (r⫽0.912,
Figure 4. Correlations between nLATT and early LVDP (A),
LVEDP (B), and mean LVDP (C). Corresponding linear regression
lines and equations are shown.
P⬍0.001), and mean PCWP (r⫽0.783, P⬍0.001). Probability values remained consistently ⱕ0.01 or lower for nLATT
correlations, with all LV diastolic pressure indices and mean
PCWP after Bonferroni correction for multiple testing was
performed. There were also strong associations between
nLATT, cardiac index, and pulmonary artery oxygen saturation in Bonferroni-corrected comparisons. The correlations
between nLATT and LV early, end, and mean filling pressures respectively are shown in scatterplots (Figure 4).
LV filling pressure was estimated noninvasively using the
following regression equations derived from the invasive data:
134
Circ Cardiovasc Imaging
March 2011
Table 3. Demographics of Patients With LVSD (nⴝ47) in the
Noninvasive Cohort
Variables
Mean⫾SD or n (%)
Age
57⫾14
Male, %
31 (66)
Body mass index, kg/m2
28.5⫾5.5
Body surface area, m2
1.96⫾0.23
Heart rate, bpm
64⫾12
Systolic blood pressure, mm Hg
128⫾22
Diastolic blood pressure, mm Hg
72⫾12
Medical history, %
Hypertension
27 (57)
Diabetes mellitus
11 (23)
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Hyperlipidemia
29 (62)
Ever smoking
25 (57)
History of myocardial infarction
17 (36)
History of coronary revascularization
15 (32)
History of congestive heart failure
21 (47)
agreement with the limits of agreement ⫾4.6 mm Hg, followed
by ⫾5.4 mm Hg for early LVDP and ⫾5.6 mm Hg for LVEDP.
Comparison Between Patients With LV
Dysfunction and Normal Control Subjects
Figure 5. Bland-Altman plots comparing measured early LVDP
with estimated early LVDP from the nLATT (A), LVEDP with estimated LVEDP from nLATT (B), and mean LVDP with mean LVDP
estimated from nLATT (C). Corresponding linear regression lines
and equation are shown.
early LVDP⫽1.27nLATT⫺5.05; LVEDP⫽1.78nLATT⫺3.78,
and mean LVDP⫽1.52nLATT⫺4.40. Invasively measured and
estimated LV filling pressures from nLATT were then compared
in Bland-Altman plots (Figure 5), demonstrating a mean difference of 0 mm Hg for all 3 filling pressure indices. Overall,
estimated and measured mean LV filling pressure had the best
The noninvasive group included 56 individuals, 47 with
LVSD of various etiologies. The demographics of the LVSD
cohort are shown in Table 3. Compared with the normal
control subjects, patients with LVSD were older (57⫾14
versus 48⫾19 years, P⬍0.001). Most were stable outpatients,
but 2 were inpatients hospitalized for heart failure. The
prevalence of New York Heart Association functional classification I through III was 46%, 23%, and 23%, respectively.
There were no class IV patients. A history of coronary disease
(myocardial infarction or coronary revascularization) was
present in 26 (55%) patients. LV volume and mass were
higher and LVEF and stroke volume lower in patients with
LVSD (Table 4). Significant valvular disease (ⱖ2⫹) among
patients with LVSD included moderate mitral regurgitation in
4 (7%) subjects, moderate tricuspid regurgitation in 2 (4%),
and moderate aortic regurgitation in 3 (5%).
There were significant prolongations in LATT and nLATT
(P⫽0.002 for both) in patients with LVSD when compared
with normal control subjects. Using the linear regression
equation described above, LVEDP was estimated for the
participants in the noninvasive group. Predicted LVEDPs in
the normal control subjects were all below 10 mm Hg, with a
mean value of 8.6⫾1.2 mm Hg. Comparing the predicted
LVEDP subgroups from low to high (ⱕ10 mm Hg, 11 to
14 mm Hg, and ⱖ15 mm Hg), there were graded decreases in
LV and RV systolic function and stroke volume and graded
increases in mitral annular E/e⬘ ratio and BNP level (Table 5),
demonstrating the ability of nLATT to assess hemodynamic
status. The significant associations of nLATT and mitral E/e⬘
ratio were also demonstrated by the Pearson correlation,
comparing them as continuous variables with r⫽0.590
(P⬍0.001) for mitral lateral E/e⬘ (Figure 6A) and r⫽0.495
(P⬍0.001) for mitral medial E/e⬘. A Bland-Altman plot was
Cao et al
Table 4. Comparison of Cardiac MRI, Echocardiographic, and
Biomarker Characteristics Between Normal Subjects and
Patients With LVSD in the Noninvasive Cohort
Normal Control
Subjects
(n ⫽ 9)
Systolic
Dysfunction
(n ⫽ 47)
P Value
LVEDV, mL/m2
88⫾9
104⫾27
0.092
2
LVESV, mL/m
40⫾7
65⫾24
0.003
LV stroke volume
96⫾16
77⫾20
LVEF, %
55⫾5
39⫾8
LV mass, g/m2
57⫾6
65⫾18
0.159
RVEDV, mL/m2
84⫾9
73⫾20
0.095
RVESV, mL/m2
40⫾6
36⫾13
0.301
RV stroke volume, mL
84⫾14
72⫾21
0.027
RVEF, %
52⫾4
51⫾9
0.747
MRI variables
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
2
0.008
⬍0.001
RV mass, g/m
25⫾5
20⫾6
0.013
LA volume, mL/m2
42⫾6.0
45⫾22
0.428
LATT, seconds
6.8⫾1.4
9.0⫾2.4
0.002
nLATT, cardiac cycles
6.6⫾0.7
9.5⫾2.6
0.002
Pulmonary systolic pressure,
mm Hg
25⫾4
28⫾8
0.110
Mitral lateral E/e⬘ ratio
5.0⫾1.8
8.3⫾2.8
0.001
Mitral medial E/e⬘ ratio
7.3⫾2.3
11.5⫾4.1
0.001
BNP, pg/mL (interquartile
range)
26 (24)
86 (134)
0.009*
NT-proBNP, pg/mL
(interquartile range)
29 (44)
200 (702)
⬍0.001*
Echocardiographic indices
Biomarkers
RV indicates right ventricle.
*BNP and NT-proBNP were not normally distributed variables and are shown
as medians (interquartile range) and analyzed using the Mann-Whitney test for
nonparametric comparison.
constructed comparing predicted PCWP, based on the mitral
lateral E/e⬘ ratio using a published regression equation22 and
predicted LVEDP from nLATT, which showed a mean
difference of 0.7 mm Hg and limits of agreement
⫾8.5 mm Hg (Figure 6B).
Discussion
In this prospective study, we developed and applied a novel
strategy for estimation of LV filling pressure using nLATT by
contrast-enhanced dynamic CMR imaging. The nLATT not only
predicted normal LV filling pressures in control subjects but also
differentiated between patients with and without hemodynamic
derangement among those with LVSD.
Early observations as well as more recent studies have
demonstrated a relationship between prolongation of various
indices of circulation time and LV enlargement, LV dysfunction, reduced cardiac output, and congestive heart failure.7–17
To our knowledge, this is the first study to explore the
hemodynamic significance of LA circulation transit time. The
measurement of LATT by CMR is derived from the ratio of
volume to flow, based on the principles of indicator dilution
theory.26,27 The relationship of prolonged LATT to increased
LA Transit Time and LV Filling Pressure
135
LV filling pressure that we demonstrated is consistent with a
large body of echocardiographic literature characterizing the
intricate interplay between reduced mitral inflow velocity and
increased LV filling pressure.28 Comparison of estimated
PCWP from E/e⬘ using a published regression equation to LV
filling pressure estimates from nLATT demonstrated a small
mean difference (0.7 mm Hg), further supporting the feasibility of assessing LV filling pressure by nLATT. A potential
advantage of MR dynamic imaging over Doppler echocardiography is the ability of MR to assess blood flow kinetics
within the LA through the entire first pass, providing a more
stable and comprehensive evaluation of LV filling and
circulatory function than assessment of the filling period in a
single heart beat by Doppler.
LA enlargement is a powerful marker for prediction of
cardiovascular events in population-based studies.29 –32 However, it lacks specific value in guiding diagnosis and therapy
in individual patients. In our LVSD cohort, there was only a
trend suggesting an association of larger LA volume with
hemodynamic derangement. In contrast, nLATT directly
assesses hemodynamic status, which has major significance
for patient evaluation and treatment. Although nLATT correlated well with all 3 LV filling pressure variables measured,
it appeared to best estimate the mean filling pressure as
shown in Bland-Altman analysis. This finding is in agreement
with observations made in the catheterization laboratory
when LA pressure and LVEDP were measured simultaneously.33 Because the effect of LV filling pressure on LA
contrast kinetics is expected to span the entire period of
diastolic filling, it is physiologically plausible to find the
mean rather than early or end-diastolic LV filling pressure
that can best be predicted using integrated LA measures such
as nLATT, although early, end, and mean LV filling pressures are closely interrelated.
Assessment of LV filling pressure can be challenging in
patients with atrial fibrillation because variable systolic ejection and diastolic filling durations give rise to variable
diastolic volume and pressure during invasive evaluation.
Similar challenges exist in applying noninvasive parameters
such as E/e⬘, which estimates LV filling pressure using
single-beat evaluations. Our findings suggest that nLATT may
have added advantages in patients with atrial fibrillation because
LV filling pressure assessment is obtained through all cardiac
cycles in the entire first pass, thereby providing an average value
over time. In patients with atrial fibrillation, we did not observe
any recognizable qualitative differences in the shape or duration
of time-intensity curves when compared with those with normal
sinus rhythm and similar filling pressure.
The development of fast single-beat MR imaging makes it
possible to measure subintervals of the global circulation time
with high spatial and temporal resolution. Most contemporary
noninvasive strategies such as first-pass nuclear scintigraphy11
and MR angiography13,14 allow circulation time to be measured
only between central veins and the aorta due to limited visualization of the cardiac chambers. This makes assessment of
LATT difficult with those techniques. The use of a planar
saturation recovery SSFP dynamic sequence in our study enabled us to obtain images with well-defined anatomy and to use
minute amounts of gadolinium contrast to acquire dynamic
136
Circ Cardiovasc Imaging
March 2011
Table 5. Comparison of Predicted LVEDP From the Normalized LA Transit Time With Cardiac
Structure, Systolic Function, Echocardiographic Indices, and Biomarkers in the Noninvasive Cohort
LVEDP ⱕ10
mm Hg
(n⫽25)
LVEDP ⬎10 and
⬍15 mm Hg
(n⫽18)
LVEDP ⱖ15
mm Hg
(n⫽13)
P Value
for Trend
LVEDV, mL/m2
97⫾22
104⫾29
109⫾29
0.170
LVESV, mL/m2
52⫾18
64⫾24
73⫾29
0.011
LV stroke volume, mL
86⫾17
79⫾21
71⫾20
0.023
LVEF, %
47⫾9
40⫾9
35⫾9
LV mass, g/m2
58⫾10
66⫾19
72⫾22
0.020
RVEDV, mL/m2
78⫾16
67⫾17
79⫾26
0.891
2
RVESV, mL/m
37⫾10
32⫾12
41⫾15
0.366
RV stroke volume, mL
79⫾17
68⫾14
75⫾32
0.587
RVEF, %
53⫾8
53⫾7
46⫾11
0.031
LA volume, mL/m2
42⫾12
47⫾24
47⫾28
0.481
LATT, seconds
7.3⫾1.2
8.9⫾1.8
11.0⫾3.0
⬍0.001
nLATT, cardiac cycles
7.0⫾1.1
9.2⫾1.0
12.6⫾2.2
⬍0.001
Cardiac structure and function
⬍0.001
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Circulation transit time
Echocardiographic indices
24⫾5
30⫾7
30⫾10
Mitral lateral E/e⬘ ratio
Pulmonary systolic pressure, mm Hg
6⫾2
8⫾2
11⫾3
⬍0.001
0.033
Mitral medial E/e⬘ ratio
9⫾5
11⫾4
13⫾3
0.023
Biomarkers
BNP, pg/mL (interquartile range)
40 (46)
86 (144)
160 (437)
0.002*
NT-proBNP, pg/mL (interquartile range)
51 (163)
195 (579)
818 (1193)
0.017*
RV indicates right ventricle.
*BNP and NT-proBNP were not normally distributed variables and are shown as medians (interquartile range) and
analyzed using the Kruskal-Wallis test.
images adequate for transit time measurement. This technique,
with a single acquisition, can be easily added to any research or
clinical CMR protocol. In addition, the very small contrast doses
required make repeated measurement of nLATT feasible both in
a single study session and serially over time.
To date, invasive hemodynamic evaluation remains the gold
standard for LV diastolic filling pressure assessment, but it is not
without risk. Noninvasive strategies such as assessment of the
relationship between transmitral inflow and mitral annular diastolic velocities by echocardiography have been used widely to
assess LV filling pressure.1–3 However, a major limitation of
E/e⬘ ratio is its largely dichotomized assessment of LV filling
pressure, which limits flexible clinical use. We found that graded
increases in average E/e⬘ correlated well with estimated LVEDP
Figure 6. Scatterplots comparing nLATT with mitral annular lateral E/e⬘ ratio (A) and Bland-Altman plots comparing LVEDP estimated
by nLATT and PCWP estimated by mitral lateral E/e⬘ ratio (B).
Cao et al
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
subgroups from low to high values. However, the linear correlation was only modest. This observation is consistent with
published work comparing estimated LVEDP by E/e⬘ with
invasively measured LVEDP.3 BNP and NT proBNP are used
clinically for diagnosis and treatment of patients with heart
failure but, in addition to heart failure, many other conditions
may be associated with small increases in BNP level, making it
nonspecific in the intermediate range.34 Therefore, there is a
need to develop new noninvasive indices with high specificity in
detecting hemodynamic derangement. Our observations provide
very promising initial results for a new approach to noninvasive
assessment of LV filling pressure.
We recognize the limitations of our study. The cohort size is
small and does not include a large number of patients having
heart failure with preserved EF, nor does it permit assessment of
effects of age, sex, and ethnicity. Thus, larger studies of more
diverse populations are needed to confirm and extend our
findings. Our CMR dynamic imaging protocol was to acquire
images over 50 cardiac cycles at the early phase of the study, but
we soon recognized that longer acquisition spans up to 100
cardiac cycles were warranted in patients with severely elevated
LVEDP. Future research should evaluate free-breathing acquisitions validated with invasive hemodynamics. Furthermore, LA
pressure often exceeds LVEDP in the presence of severe mitral
regurgitation, which probably alters blood transit through the LA
and may lead to impaired assessment of LVEDP. Therefore, we
excluded subjects with severe mitral regurgitation, which needs
to be addressed in a cohort with large number of patients having
this condition. Last, the current study is a cross-sectional
evaluation. The serial use and prognostic significance of nLATT
measurement are yet to be defined.
In conclusion, nLATT by CMR is a very promising new
parameter of LV filling pressure that may provide graded
noninvasive assessment of abnormalities of circulatory and
hemodynamic function in patients with LVSD.
Sources of Funding
This project was funded in part by the Helen L. Deblinger Foundation.
Disclosures
None.
References
1. Dokainish H, Zoghbi WA, Lakkis NM, Al-Bakshy F, Dhir M, Quinones
MA, Nagueh SF. Optimal noninvasive assessment of left ventricular
filling pressures: a comparison of tissue Doppler echocardiography and
B-type natriuretic peptide in patients with pulmonary artery catheters.
Circulation. 2004;109:2432–2439.
2. Nagueh SF, Sun H, Kopelen HA, Middleton KJ, Khoury DS. Hemodynamic determinants of the mitral annulus diastolic velocities by tissue
Doppler. J Am Coll Cardiol. 2001;37:278 –285.
3. Ommen SR, Nishimura RA, Appleton CP, Miller FA, Oh JK, Redfield
MM, Tajik AJ. Clinical utility of Doppler echocardiography and tissue
Doppler imaging in the estimation of left ventricular filling pressures: a
comparative simultaneous Doppler-catheterization study. Circulation.
2000;102:1788 –1794.
4. Kazanegra R, Cheng V, Garcia A, Krishnaswamy P, Gardetto N, Clopton
P, Maisel A. A rapid test for B-type natriuretic peptide correlates with
falling wedge pressures in patients treated for decompensated heart failure: a pilot study. J Card Fail. 2001;7:21–29.
5. Maeda K, Tsutamoto T, Wada A, Hisanaga T, Kinoshita M. Plasma brain
natriuretic peptide as a biochemical marker of high left ventricular enddiastolic pressure in patients with symptomatic left ventricular dysfunction. Am Heart J. 1998;135:825– 832.
LA Transit Time and LV Filling Pressure
137
6. Maisel AS, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc
P, Omland T, Storrow AB, Abraham WT, Wu AH, Clopton P, Steg PG,
Westheim A, Knudsen CW, Perez A, Kazanegra R, Herrmann HC,
McCullough PA. Rapid measurement of B-type natriuretic peptide in the
emergency diagnosis of heart failure. N Engl J Med. 2002;347:161–167.
7. Blumgart HL, Weiss S. Clinical studies on the velocity of blood flow: IX.
The pulmonary circulation time, the velocity of venous blood flow to the
heart, and related aspects of the circulation in patients with cardiovascular
disease. J Clin Invest. 1928;5:343–377.
8. de Freitas FM, Faraco EZ, Nedel N, Deazevedo DF, Zaduchliver J.
Determination of pulmonary blood volume by single intravenous
injection of one indicator in patients with normal and high pulmonary
vascular pressures. Circulation. 1964;30:370 –380.
9. Morris LE, Blumgart HL. Velocity of blood flow in health and disease.
Circulation. 1957;15:448 – 460.
10. Jones RH, Sabiston DC Jr, Bates BB, Morris JJ, Anderson PA, Goodrich JK.
Quantitative radionuclide angiocardiography for determination of chamber to
chamber cardiac transit times. Am J Cardiol. 1972;30:855–864.
11. Slutsky RA, Bhargava V, Higgins CB. Pulmonary circulation time: comparison of mean, median, peak, and onset (appearance) values using
indocyanine green and first-transit radionuclide techniques. Am Heart J.
1983;106:41– 45.
12. Muller HM, Tripolt MB, Rehak PH, Groell R, Rienmuller R, Tscheliessnigg KH. Noninvasive measurement of pulmonary vascular resistances by assessment of cardiac output and pulmonary transit time. Invest
Radiol. 2000;35:727–731.
13. Shors SM, Cotts WG, Pavlovic-Surjancev B, Francois CJ, Gheorghiade
M, Finn JP. Heart failure: evaluation of cardiopulmonary transit times
with time-resolved MR angiography. Radiology. 2003;229:743–748.
14. Francois CJ, Shors SM, Bonow RO, Finn JP. Analysis of cardiopulmonary transit times at contrast material-enhanced MR imaging in patients
with heart disease. Radiology. 2003;227:447– 452.
15. Freis ED, Schnaper HW, Johnson RL, Schreiner GE. Hemodynamic alterations in acute myocardial infarction, I: cardiac output, mean arterial
pressure, total peripheral resistance, central and total blood volumes, venous
pressure and average circulation time. J Clin Invest. 1952;31:131–140.
16. Kopelman H. The circulation time as a clinical test. Br Heart J. 1951;
13:301–308.
17. Gargill S. The use of sodium dehydrocholate as a clinical test of the
velocity of blood flow. N Engl J Med. 1933;209:1089 –1093.
18. Ziegler RF. Circulation time determinations from the right ventricle.
Circulation. 1951;4:905–908.
19. Cecconi M, Manfrin M, Zanoli R, Colonna P, Ruga O, Pangrazi A, Soro
A. Doppler echocardiographic evaluation of left ventricular end-diastolic
pressure in patients with coronary artery disease. J Am Soc Echocardiogr.
1996;9:241–250.
20. Giannuzzi P, Imparato A, Temporelli PL, de VF, Silva PL, Scapellato F,
Giordano A. Doppler-derived mitral deceleration time of early filling as
a strong predictor of pulmonary capillary wedge pressure in postinfarction patients with left ventricular systolic dysfunction. J Am Coll
Cardiol. 1994;23:1630 –1637.
21. Rossvoll O, Hatle LK. Pulmonary venous flow velocities recorded by
transthoracic Doppler ultrasound: relation to left ventricular diastolic
pressures. J Am Coll Cardiol. 1993;21:1687–1696.
22. Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA.
Doppler tissue imaging: a noninvasive technique for evaluation of left
ventricular relaxation and estimation of filling pressures. J Am Coll
Cardiol. 1997;30:1527–1533.
23. Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation of right
atrial pressure from the inspiratory collapse of the inferior vena cava.
Am J Cardiol. 1990;66:493– 496.
24. Messika-Zeitoun D, Bellamy M, Avierinos JF, Breen J, Eusemann C,
Rossi A, Behrenbeck T, Scott C, Tajik JA, Enriquez-Sarano M. Left atrial
remodelling in mitral regurgitation: methodologic approach, physiological determinants, and outcome implications: a prospective quantitative
Doppler-echocardiographic and electron beam-computed tomographic
study. Eur Heart J. 2007;28:1773–1781.
25. Meier P, Zierler KL. On the theory of the indicator-dilution method for
measurement of blood flow and volume. J Appl Physiol. 1954;6:731–744.
26. Hamilton W, Riley RL. Comparison of the Fick and dye injection
methods of measuring the cardiac output in man. Am J Physiol. 1948;
153:309 –321.
27. Hamilton W. The Lewis A. Connor memorial lecture: the physiology of
the cardiac output. Circulation. 1953;8:527–543.
138
Circ Cardiovasc Imaging
March 2011
28. Oh JK, Seward J, Tajik AJ. The Echo Manual. 3rd ed. Philadelphia, PA:
Lippincott Williams & Wilkins; 2006.
29. Benjamin EJ, D’Agostino RB, Belanger AJ, Wolf PA, Levy D. Left atrial
size and the risk of stroke and death: the Framingham Heart Study.
Circulation. 1995;92:835– 841.
30. Gottdiener JS, Kitzman DW, Aurigemma GP, Arnold AM, Manolio TA.
Left atrial volume, geometry, and function in systolic and diastolic heart
failure of persons ⱖ65 years of age (the cardiovascular health study).
Am J Cardiol. 2006;97:83– 89.
31. Pritchett AM, Jacobsen SJ, Mahoney DW, Rodeheffer RJ, Bailey KR,
Redfield MM. Left atrial volume as an index of left atrial size:
a population-based study. J Am Coll Cardiol. 2003;41:10361043.
32. Pritchett AM, Mahoney DW, Jacobsen SJ, Rodeheffer RJ, Karon BL,
Redfield MM. Diastolic dysfunction and left atrial volume: a
population-based study. J Am Coll Cardiol. 2005;45:87–92.
33. Yamamoto K, Nishimura RA, Redfield MM. Assessment of mean left
atrial pressure from the left ventricular pressure tracing in patients with
cardiomyopathies. Am J Cardiol. 1996;78:107–110.
34. Mohammed AA, Januzzi J. Natriuretic peptides in the diagnosis
and management of acute heart failure. In: Cohen-Kligerman B, ed.
Heart Failure Clinics. 2009 ed. Philadelphia, PA: Saunders;
2009:489 –500.
CLINICAL PERSPECTIVE
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Assessment of left ventricular filling pressure is of clinical importance in guiding therapy and evaluating prognosis in
patients with systolic dysfunction. In this prospective study, we developed and applied a novel strategy for estimation of
left ventricular filling pressure using mean circulation transit time in the left atrium by contrast-enhanced dynamic MRI.
The normalized left atrial transit time not only predicted normal left ventricular filling pressures in control subjects but also
differentiated between patients with and without hemodynamic derangement among those with left ventricular systolic
dysfunction. Cardiac MRI is increasingly used in diagnosis and follow-up of patients with systolic dysfunction of various
etiologies. This simple technique of single imaging acquisition with minute amount of gadolinium contrast can be
complementary to any clinical or research study for comprehensive evaluation of patients with systolic dysfunction.
Left Ventricular Filling Pressure Assessment Using Left Atrial Transit Time by Cardiac
Magnetic Resonance Imaging
Jie J. Cao, Yi Wang, Jeannette McLaughlin, Elizabeth Haag, Peter Rhee, Michael Passick, Rena
Toole, Joshua Cheng, Andrew D. Berke, Justine Lachman and Nathaniel Reichek
Downloaded from http://circimaging.ahajournals.org/ by guest on May 10, 2017
Circ Cardiovasc Imaging. 2011;4:130-138; originally published online January 24, 2011;
doi: 10.1161/CIRCIMAGING.110.959569
Circulation: Cardiovascular Imaging is published by the American Heart Association, 7272 Greenville Avenue,
Dallas, TX 75231
Copyright © 2011 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-9651. Online ISSN: 1942-0080
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circimaging.ahajournals.org/content/4/2/130
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Imaging can be obtained via RightsLink, a service of the Copyright Clearance
Center, not the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Cardiovascular Imaging is online at:
http://circimaging.ahajournals.org//subscriptions/