Left ventricular volume shifts and aortic root expansion during isovolumic contraction.

Abstract

Aortic valve opening involves conformational changes of the aortic root, including the ventricular-aortic junction (VAJ), sinotubular junction (STJ), and cusps. Moreover, the aortic root is contiguous with the left ventricular outflow tract (LVOT), which changes diameter throughout the cardiac cycle. Aortic root expansion prior to valve opening facilitates outward displacement of aortic cusp attachments, which helps flatten the cusps, thereby reducing cusp stress and fatigue, ultimately enhancing functional valve durability. The mechanisms underlying aortic root expansion prior to valve opening, however, remain incompletely characterized. The study aim was to establish a link between such aortic root expansion and intraventricular volume shifts into the LVOT during isovolumic contraction (IVC).
Miniature radiopaque markers were implanted on the left ventricle, VAJ, STJ, and aortic cusps of six sheep. After one week, 3-D marker coordinates were obtained using biplane videofluoroscopy (60 Hz). Triangular areas at the VAJ and STJ were calculated; LV main chamber (non-LVOT) and LVOT volumes were calculated using multiple tetrahedra. End-diastole was defined as the peak of the electrocardiogram R-wave, and end-IVC when aortic cusp separation began.
During IVC, blood within the left ventricle was redistributed to the LVOT: mean LVOT volume was increased (+0.2 +/- 0.1 ml, p = 0.009) as non-LVOT volume fell (-0.8 +/- 0.4 ml, p = 0.006). Concomitantly, the aortic root expanded as both VAJ and STJ areas increased (+0.23 +/- 0.12 cm2 (p = 0.005) and +0.25 +/- 0.14 cm2 (p = 0.007), respectively) prior to aortic cusp separation.
Aortic root expansion prior to valve opening is closely related to intraventricular volume shifts into the LVOT during IVC. Such volume shifts may 'prime' the aortic valve for ejection. These findings expand our understanding of cardiac dynamics by showing that blood acts as a coupling link between various cardiac units. Preservation of these normal aortic root dynamics may enhance the efficacy and durability of aortic surgical interventions.

Full text

Page 1

Left Ventricular Volume Shifts and Aortic Root
Expansion during Isovolumic Contraction
Filiberto Rodríguez1, G. Randall Green1, Paul Dagum1, J. Francisco Nistal1, Katherine B.
Harrington1, George T. Daughters1,2, Neil B. Ingels1,2, D. Craig Miller1
1Department of Cardiothoracic Surgery, Stanford University School of Medicine, Stanford, California, 2Laboratory of
Cardiovascular Physiology and Biophysics, Palo Alto Medical Foundation Research Institute, Palo Alto, California, USA
The aortic valve has classically been assumed to be
an inert structure that opens passively as the forward
transvalvular pressure gradient initiates ejection of
blood into the aorta. Valve opening, however, is a com-
plex and dynamic process involving three-dimension-
al (3-D) conformational changes of the aortic root,
including the aortic annulus (more correctly termed
the ‘ventricular-aortic junction’ (VAJ), which is coro-
net-shaped and not planar), the sinotubular junction
(STJ), the sinuses of Valsalva, and the aortic cusps (1-5).
Moreover, the aortic root at the VAJ is contiguous with
the left ventricular outflow tract (LVOT), which
changes its diameter throughout the cardiac cycle (3).
Prior to ejection, the aortic root expands during early
systole before valve opening, and this expansion
reduces shear stress on the aortic leaflets (4-7).
Outward displacement of the aortic cusp attachments
facilitates flattening of the cusps as the valve opens,
rather than simply bending on a hinge, which reduces
cusp fatigue and ultimately enhances the functional
durability of the native valve.
The mechanisms underlying aortic root expansion
prior to valve opening, however, remain incompletely
characterized. Dagum et al. previously reported the
deformational dynamics of the aortic root throughout
Presented in part at the American Heart Association Scientific
Sessions, 1999, Atlanta, GA, and the American College of
Cardiology, 2003, Chicago, IL, USA
Address for correspondence:
D. Craig Miller MD, Department of Cardiothoracic Surgery, Falk
Cardiovascular Research Center, Stanford University School of
Medicine, Stanford, CA 94305-5247, USA
e-mail: dcm@stanford.edu
© Copyright by ICR Publishers 2006
Background and aim of the study: Aortic valve open-
ing involves conformational changes of the aortic
root, including the ventricular-aortic junction (VAJ),
sinotubular junction (STJ), and cusps. Moreover, the
aortic root is contiguous with the left ventricular out-
flow tract (LVOT), which changes diameter through-
out the cardiac cycle. Aortic root expansion prior to
valve opening facilitates outward displacement of
aortic cusp attachments, which helps flatten the
cusps, thereby reducing cusp stress and fatigue, ulti-
mately enhancing functional valve durability. The
mechanisms underlying aortic root expansion prior
to valve opening, however, remain incompletely
characterized. The study aim was to establish a link
between such aortic root expansion and intraventric-
ular volume shifts into the LVOT during isovolumic
contraction (IVC).
Methods: Miniature radiopaque markers were
implanted on the left ventricle, VAJ, STJ, and aortic
cusps of six sheep. After one week, 3-D marker coor-
dinates were obtained using biplane videofluo-
roscopy (60 Hz). Triangular areas at the VAJ and STJ
were calculated; LV main chamber (non-LVOT) and
LVOT volumes were calculated using multiple tetra-
hedra. End-diastole was defined as the peak of the
electrocardiogram R-wave, and end-IVC when aortic
cusp separation began.
Results: During IVC, blood within the left ventricle
was redistributed to the LVOT: mean LVOT volume
was increased (+0.2 ± 0.1 ml, p = 0.009) as non-LVOT
volume fell (-0.8 ± 0.4 ml, p = 0.006). Concomitantly,
the aortic root expanded as both VAJ and STJ areas
increased (+0.23 ± 0.12 cm2(p = 0.005) and +0.25 ± 0.14
cm2(p = 0.007), respectively) prior to aortic cusp sep-
aration.
Conclusion: Aortic root expansion prior to valve
opening is closely related to intraventricular volume
shifts into the LVOT during IVC. Such volume shifts
may ‘prime’ the aortic valve for ejection. These find-
ings expand our understanding of cardiac dynamics
by showing that blood acts as a coupling link
between various cardiac units. Preservation of these
normal aortic root dynamics may enhance the effica-
cy and durability of aortic surgical interventions.
The Journal of Heart Valve Disease 2006;15:465-473

Page 2

the cardiac cycle in detail (5). These authors noted that
both the aortic annulus (i.e., the VAJ) and the STJ
underwent rapid circumferential expansion during
isovolumic contraction (IVC), with the VAJ reaching its
maximal area at end-IVC. More recently, Timek et al.
studied simultaneous aortic and mitral annular
dynamics using data from this same experiment (8);
mitral annular and aortic annular areas were noted to
change in reciprocal fashion during late diastole and
early systole, with peak aortic annular area again seen
just prior to LV ejection. Timek and colleagues sug-
gested that this aortic annular expansion prior to ejec-
tion may be related to LV volume redistribution during
IVC (8), whereas concomitant mitral annular area
reduction occurs predominantly prior to ventricular
systole being strongly influenced by atrial contraction
(9).
In order to further elucidate the mechanisms under-
lying aortic root expansion prior to valve opening, the
relationship between aortic root dynamics and volume
shifts within the ventricle during IVC was assessed. In
a new analysis using experimental animal data previ-
ously reported by the present authors’ laboratory (5,8),
intraventricular volume shifts were measured and
temporally related with VAJ and STJ expansion during
IVC. The aim was to establish a link between aortic
annular expansion and intraventricular volume shifts
into the LVOT prior to LV ejection.
Materials and methods
The experimental methods have been reported pre-
viously in detail (5,8), and are summarized below. All
animals received humane care in compliance with the
principles of laboratory animal care formulated by the
National Society for Medical Research and the Guide
for Care and Use of Laboratory Animals, prepared by the
National Institutes of Health. This study was approved
by the Stanford Medical Center Laboratory Research
Animal Review Committee and conducted according
to Stanford University policy.
Animals and surgical preparation
Six adult, Suffolk, male sheep (mean body weight 70
± 7 kg) underwent a left thoracotomy while under gen-
eral anesthesia (inhalational isoflurane, 1-2.2%). Four
miniature radiopaque markers were inserted into the
subepicardium to silhouette the left ventricle along
four equally spaced longitudinal meridians defining
the anterior, lateral, posterior, and septal LV walls at
the equatorial level; a single marker was placed at the
LV apex (#1-5; Fig. 1). The septal marker was placed
using epicardial echocardiographic
Epicardial color Doppler echocardiography document-
ed aortic and mitral valve competence. Using car-
guidance.
diopulmonary bypass (mean time 157 ± 18 min) with
cardioplegic arrest (mean time 107 ± 18 min) and
working through a left atriotomy, six markers were
sutured around the mitral annulus (#6-11; Figs. 1 and
2B). Atransverse aortotomy was made 5 mm above the
top of the commissures, and nine markers were
sutured around the aortic root and leaflets (#12-20;
Figs. 1 and 2). The aortotomy and atriotomy were
closed, the heart de-aired, the cross-clamp removed,
and the heart defibrillated. An implantable micro-
manometer pressure transducer (PA4.5-X6; Konigs-
berg Instruments, Inc., Pasadena, CA, USA) was
placed in the LV chamber through the apex and exteri-
orized. The incisions were closed, and the animals
allowed to recover.
466 LV volume shifts during IVC
F. Rodriguez et al.
J Heart Valve Dis
Vol. 15. No. 4
July 2006
Figure 1: Myocardial marker array. Four radiopaque
markers were inserted in the subepicardium to silhouette
the left ventricle along equally spaced longitudinal
meridians defining anterior, lateral, posterior, and septal
walls at the equatorial level (#1-4); a single marker was
placed at the left ventricular (LV) apex (#1). On the mitral
annulus, markers were placed at the mid-septal ‘saddle
horn’ (#6), the lateral annulus (#9), the left (#7) and right
(#11) fibrous trigones, and the anterior (#8) and posterior
(#10) commissures. On the aortic root, markers were placed
at the level of the ventricular-aortic junction at the nadir of
the belly of each cusp: the non-coronary (NC) cusp (#12);
the left coronary (LC) cusp (#13); and the right coronary
(RC) cusp (#14). The white area represents a 2-D
representation of the left ventricular outflow tract (LVOT).
The shaded area represents a 2-D representation of the non-
LVOT region of the left ventricle.

Page 3

Data acquisition
Following a recovery period of 8 ± 2 days, studies
were conducted in the cardiac catheterization labora-
tory. Animals were premedicated with ketamine, intu-
bated, and mechanically ventilated (Veterinary
anesthesia ventilator 2000; Hallowell EMC) with 100%
oxygen. In order to minimize reflex sympathetic and
parasympathetic responses, autonomic blockade was
accomplished with low-dose intravenous infusions of
esmolol (20-50 µg/kg/min, titrated to reduce the heart
rate below 110 beats per min) and atropine sulfate (0.01
mg/kg/min). A micromanometer-tipped catheter
(SPC-500; Millar Instruments, Inc., Houston, TX, USA)
zeroed in a 37ºC water bath was introduced through a
left carotid sheath and advanced to the aortic arch.
A Philips Optimus 2000 biplane Lateral ARC 2/poly
DIAGNOST C2 system (Philips Medical Systems,
Pleasanton, CA, USA) was used to record videofluoro-
scopic images at 60 Hz. Simultaneous biplane video-
fluoroscopic and hemodynamic data were acquired
with the animal in the right lateral decubitus position.
Animals were studied in normal sinus rhythm after
autonomic blockade with ventilation arrested at end-
expiration during data acquisition. Two-dimensional
images from the two X-ray views were digitized using
custom software (10). These data were merged to yield
3-D coordinates for each radiopaque marker every 16.7
ms, the accuracy of which was 0.1 ± 0.3 mm compared
with known marker-to-marker 3-D lengths (11).
Analog pressure and ECG signals were digitized
simultaneously and recorded on videotape along with
marker images during data acquisition. At the end of
the experiment, correct marker position was confirmed
in all animals at post-mortem examination.
Data analysis
Hemodynamics and cardiac cycle timing
For each cardiac cycle, the time of end-systole (ES)
was defined as 16.7 ms (one videofluoroscopic frame)
before the time of peak negative LV rate of pressure fall
(-dP/dtmax). End-diastole (ED) was defined as the vide-
ofluoroscopic frame containing the peak of the electro-
cardiogram r-wave. The beginning of IVC was defined
as the time of ED, and end-IVC (or beginning ejection)
was defined as the first videofluoroscopic frame show-
ing aortic cusp separation (Fig. 3).
Aortic root deformations
The VAJ area was calculated as the area of the trian-
gle in 3-D space between markers 12, 13, and 14 (Fig.
2). Similarly, the STJ area was calculated as the area of
the triangle defined by markers 15, 16, and 17 at the
top of each commissure (Fig. 2). Aortic valve opening
dynamics were measured by cusp edge separation cal-
culated as the distance between two markers (#18, 19,
LV volume shifts during IVC
F. Rodriguez et al.
467
J Heart Valve Dis
Vol. 15. No. 4
July 2006
Figure 2: Aortic root and mitral valve marker arrays. A)
The aortic root is shown as a frustum of a cone with its
base at the ventricular-aortic junction (V AJ) and a plane
transecting the cone parallel to its base at the sinotubular
junction (STJ). Markers were placed at the level of the V AJ
at the nadir of the belly of each cusp: the non-coronary
(NC) cusp (#12); the left coronary (LC) cusp (#13); and the
right coronary (RC) cusp (#14). Similarly, markers #15,
16, and 17 were placed at the level of the STJ at the top of
each commissure (left to right, right-NC, NC-left,
respectively). The sinuses of Valsalva are circumscribed by
markers #15, 14, and 16; markers #16, 12, and 17; and
markers #17, 13, and 15. B) On the mitral annulus,
markers were placed at the mid-septal ‘saddle horn’ (#6),
the lateral annulus (#9), the left (#7) and right (#11)
fibrous trigones, and the anterior (#8) and posterior (#10)
commissures. The fibrous area of continuity between the
aortic and mitral valves, which is flanked by the left and
right fibrous trigones, represents the aortomitral curtain.
On the aortic root, markers were positioned on the aortic
annulus at the level of the V AJ (#12-14) and at the level of
the STJ (#15-17). Markers were placed on the thickened
nodule at the center of the free edge of each aortic cusp, i.e.,
the nodulus of Arantius (#18-20).
A
B

Page 4

or 20; Fig. 2B) on the nodules of Aranti.
Left ventricular volumes
Instantaneous LV volumes were calculated every
16.7 ms from the epicardial LV, mitral annular, and VAJ
markers using multiple tetrahedra reconstructed from
the 3-D marker coordinates. The LV chamber was
divided into LVOT and non-LVOT regions (see Fig. 1).
The main LV (non-LVOT) volume was calculated from
the epicardial LV markers and mitral annular markers
(#1-11; Figs. 1 and 2B); LVOT volume was calculated
using the septal LV marker (#2; Fig. 1), the three mark-
ers at the VAJ (#12-14; Figs. 1 and 2), and the markers
along the fibrous aortomitral continuity, which includ-
ed the mid-septal mitral ‘saddle horn’ marker (#6; Figs.
1 and 2B) and the left and right fibrous trigone mark-
ers (#7 and #11; Figs. 1 and 2B).
Total LV volume was calculated as the sum of LVOT
volume plus non-LVOT LV volume. Although epicar-
dial LV volume calculated in this manner overesti-
mates true LV chamber volume (because it
incorporates LV muscle mass), the change in epicardial
LV volume is an accurate measurement of the relative
change in LV chamber volume (12). Stroke volume
(SV) was calculated as the difference between end-
diastolic total (LVOT + non-LVOT) LV volume (EDV)
and end-systolic total LV volume (ESV); that is, SV =
(EDV - ESV). The calculated ejection fraction (EF)
using this method, however, would underestimate
true EF because the measured LV volumes include
myocardial wall volume (volume calculated from epi-
cardial markers). This causes the denominator (EDV)
of the EF calculation [(EDV - ESV)/EDV] to be artifi-
cially large, thereby yielding a smaller LVEF.
The rate of volume change (∆V/∆t) was calculated
every 16.7 ms to assess changes in both LVOT (∆Vot/∆t)
468 LV volume shifts during IVC
F. Rodriguez et al.
J Heart Valve Dis
Vol. 15. No. 4
July 2006
Figure 3: Group mean aortic root temporal dynamics. Upper panel: Left ventricular pressure (LVP, mmHg, open circles) and
aortic pressure (AoP, mmHg, open squares) on the left ordinate with aortic leaflet separation (AoL, cm, solid squares) on the
right ordinate versus time (ms) on the abscissa. Lower panel: LVP (mmHg, open circles) and AoP (mmHg, open squares) on
the left ordinate with ventricular-aortic junction area (V AJ, cm2, solid circles) and sinotubular junction area (STJ, cm2, solid
triangles) on the right ordinate versus time (ms) on the abscissa. A 600-ms time window spanning the cardiac cycle for all six
animals is shown. Dashed lines define end-diastole (ED), end-isovolumic contraction (end-IVC), maximal aortic cusp
separation (AoV Max), and end-systole (ES). Data are expressed as mean values for all animals; error bars represent 1 SEM.

Page 5

and non-LVOT (∆Vnot/∆t) regions during IVC (Fig. 4).
The sum of the differences in volume between each
frame was then used to approximate the area under
the rate of volume change curve to obtain the volume
changes during IVC.
Septal wall motion
Septal marker position was determined using a
right-handed Cartesian coordinate reference system
with the origin at the mid-septal annulus (marker #6;
Fig. 1), positive lateral axis passing through the mid-
lateral annulus (marker #9), positive posterior axis
directed towards the right fibrous trigone (marker
#11), and positive apical axis directed towards the LV
apex (marker #1). Total septal marker displacement
from ED through end-IVC was resolved into lateral,
posterior, and apical components.
Statistical analysis
All data were reported as mean ± SD, unless other-
wise stated. Hemodynamic and marker-derived data
from two consecutive steady-state beats in sinus
rhythm were time-aligned at ED, and data from the
two beats were averaged for each animal. Changes in
cusp separation, VAJ area, STJ area, LVOT volume, and
non-LVOT volume at ED, end-IVC, and maximal aor-
tic cusp separation were compared using one-way
repeated measures ANOVA, with the Student-
Newman-Keuls post-test for multiple comparisons
applied when a significant F-value (p <0.001) was
detected. Comparisons of changes in LVOT and non-
LVOT ventricular volumes and septal wall motion
from ED to end-IVC were also made using a one-sam-
ple t-test versus zero at ED. A p-value <0.05 was con-
sidered to be statistically significant.
LV volume shifts during IVC
F. Rodriguez et al.
469
J Heart Valve Dis
Vol. 15. No. 4
July 2006
Figure 4: Intraventricular volume shifts. Upper panel: Left ventricular pressure (LVP, mmHg, open circles) on the left
ordinate with rate of left ventricular outflow tract (LVOT) volume change (∆Vot/∆t, solid circles) on the right ordinate versus
time (ms) on the abscissa. Lower panel: LVP (mmHg, open circles) on the left ordinate with rate of main left ventricular
chamber (non-LVOT) volume change (∆Vnot/∆t, solid triangles) on the right ordinate versus time (ms) on the abscissa. A 600-
ms time window spanning the cardiac cycle for all six animals is shown. Dashed lines define end-diastole (ED), end-
isovolumic contraction (end-IVC), maximal aortic cusp separation (AoV Max), and end-systole (ES). Data are expressed as
mean values for all animals; error bars represent 1 SEM.