(Circulation. 1995;92:458-466.)
© 1995 American Heart Association, Inc.
Articles |
Left Ventricular Function, Twist, and Recoil After Mitral Valve Replacement
From the Department of Cardiovascular and Thoracic Surgery, Stanford (Calif) University School of Medicine (A.D., M.K., D.C.M.); the Cardiac Surgery Section, Department of Veterans Affairs Medical Center, Palo Alto, Calif (D.C.M.); and the Research Institute of the Palo Alto Medical Foundation, Palo Alto, Calif (S.D.N., G.T.D., N.B.I.).
Correspondence to D. Craig Miller, MD, Department of Cardiovascular and Thoracic Surgery, Falk Cardiovascular Research Center, Stanford University School of Medicine, Stanford, CA 94305-5247.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background Preservation of the mitral subvalvular apparatus during mitral valve replacement (MVR) has become more popular, in part because of the clinically and experimentally demonstrated more optimal left ventricular (LV) performance after surgery; the mechanisms responsible for this beneficial influence, however, have not been clearly elucidated.
Methods and Results Fourteen dogs underwent placement of 26
myocardial markers into the LV and septum. One week later, the animals
were studied while awake, sedated, and atrially paced (120 beats per
minute) both under baseline conditions and after inotropic stimulation
(calcium). The animals then underwent MVR and were randomized into
either chord-sparing (MVR-Intact) or chord-severing (MVR-Cut)
techniques. Two weeks later, the animals were studied under the same
conditions. LV systolic function was assessed by the slope of the
end-systolic pressure-volume relation (Ees); early
LV diastolic filling was analyzed by the
pressure-time constant of relaxation (
). The instantaneous
longitudinal gradient of torsional deformation for the LV (twist) was
also calculated, as were the changes in twist with respect to time
during systole and early diastole (LV recoil). Intergroup
comparison showed a trend toward increased
contractility (Ees,
P=.061, before versus after MVR), as well as faster
relaxation for the MVR-Intact group. Concurrent analysis of LV
systolic function and the rate of systolic twist revealed a significant
inverse relation, which disappeared after MVR when the chordae were
severed.
Conclusions These observations suggest that the mitral subvalvular apparatus acts as a modulator of LV systolic torsional deformation into LV pump (or ejection) performance.
Key Words: ventricles • mechanics • valves • torsion
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The most optimal surgical procedure for patients with chronic mitral regurgitation is valve repair, rather than MVR. Mitral valve repair, however, is not always technically feasible or indicated, which prompts the need for MVR in a substantial minority of patients. An unanswered issue has been what the surgeon should do with the mitral chordae tendineae and leaflets (ie, the subvalvular apparatus) during MVR, a controversy initiated by the observations of Lillehei et al1 during the pioneering years of MVR during the early 1960s that preservation of the mitral chordae tendineae resulted in a lower operative mortality rate compared with chord-severing techniques. The importance of the mitral subvalvular apparatus in maintaining LV pump function after MVR has subsequently been the subject of numerous investigations. Two recent editorials have addressed this issue, with both authors stating their belief that preservation of the chordae tendineae results in a better outcome.2 3 Indeed, Carabello3 stated that ". . . there [no] longer is room to doubt the importance of chordal preservation in mitral valve surgery." The acceptance of chord-preserving MVR techniques has progressed to the point where retrospective as well as randomized prospective clinical studies have been initiated in Germany and elsewhere.4 5 6 7
The elements missing are data obtained from conscious, intact animals or human subjects to elucidate the mechanism(s) whereby LV systolic function is improved by retaining the mitral subvalvular apparatus. Such information would strengthen the clinical rationale for preserving the chordae tendineae during MVR. Suggestions have been made that chord-sparing MVR preserves normal LV myocyte function8 as well as reducing global LV systolic afterload,9 10 but conclusive evidence from randomized clinical trials still remains limited. In a previous review11 by our group of the clinical and experimental evidence supporting the importance of the mitral subvalvular apparatus, superior LV performance was seen when the chordae were preserved in both isolated and intact heart models. The link between the mitral chordae tendineae and LV systolic pump function was called "valvular-ventricular interaction,"12 but the exact mechanisms remain unknown.
In this study, LV systolic and diastolic function in intact canine hearts 1 to 2 weeks after MVR with and without preservation of the chordae tendineae were compared with global LV systolic torsional deformation (LV twist) and early diastolic recoil on the premise that twist (the longitudinal gradient of torsional deformation) is a global reflection of regional sarcomere shortening independent of LV loading conditions; ie, twist reflects cardiomyocyte contractility. Our working hypothesis was that chord division perturbs systolic and diastolic LV function because of deleterious alterations in LV twist dynamics after disruption of annular-papillary continuity during MVR.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Surgical Preparation: Placement of LV Myocardial Markers
Fourteen dogs (25±5 kg) underwent left thoracotomy for placement of myocardial markers into the left ventricle. Anesthesia was induced with sodium thiopental (25 mg/kg IV), and the animals were intubated and ventilated with 100% O2 (Ohio Anesthesia Ventilator) with supplemental inhalational isoflurane (1% to 2%). A micromanometer-tipped catheter (MPC-500; Millar Instruments Inc) was placed into the left femoral artery to monitor arterial pressure. Superior and inferior vena caval pneumatic occluders (In Vivo Metric Systems) were placed to create abrupt LV preload reduction. The heart was suspended in a pericardial cradle, and 26 miniature tantalum radiopaque helices (ID, 0.8 mm; OD, 1.3 mm; length, 1.5 to 3.0 mm; some with variable small extensions or "tails" to facilitate subsequent radiographic identification) were inserted into the LV wall and septum. The markers were placed on the obturator of a modified spinal needle (20 gauge), inserted through the epicardial surface, and deposited into the myocardium by withdrawal of the obturator from the sheath. As depicted schematically in Fig 1
,
16 markers were placed into the LV subepicardial
layer along four equally spaced longitudinal meridians around the left
ventricle, including the anterior (from the origin of the left anterior
descending coronary artery to the apex), lateral (obtuse
margin), posterior (inferior wall along the posterior
descending artery), and septal walls. Each LV meridian contained
markers in the apicoequatorial, equatorial, basoequatorial, and basal
short-axis planes (ie, at four levels), and one additional marker
was positioned at the LV apex. This arrangement allowed a
three-dimensional representation of the left ventricle with
simultaneous biplanar 45° RAO and 45° LAO
videofluoroscopic imaging. Nine additional markers were placed into the
LV subendocardial layer at each level (with the exception of the base)
in the anterior, lateral, and posterior LV meridians under epicardial
echocardiographic guidance deep to the corresponding
subepicardial markers. After the marker positions were verified with
fluoroscopy, atrial pacing electrodes were inserted, the pericardium
was closed, and an intercostal block (20 mL 0.25% bupivicaine) was
performed at the fourth, fifth, and sixth intercostal levels. Chest
tubes were placed, and the incision was closed. The animals were then
allowed to recover in the intensive care unit in conjunction with the
Stanford University Department of Laboratory Medicine, with oxymorphone
HCl (Numorphan 0.05 to 0.2 mg/kg IV, DuPont Merck Pharma) used to
minimize discomfort.
|
), endocardial; 
, epicardial). The four
"septal" markers, represented by the open boxes
(
), include an apicoequatorial marker that lies on the LV surface
(since the RV does not extend all the way to the apex of the LV) and
are placed on the RV subendocardial surface. The septal markers do not
have corresponding wall thickness pairs. A indicates anterior; L,
lateral; P, posterior; and S, septal meridians.
Experimental Protocol Before MVR
After a recovery period of 5
to 34 days (average, 9±7 days
[mean±SD]), the animals were taken to the cardiac
catheterization laboratory for baseline
hemodynamic and videofluoroscopic data acquisition
before subsequent MVR (Pre-MVR). For sedation, diazepam (5 mg IV) and
supplemental ketamine (5 mg/kg IV) were administered as needed.
A long micromanometer-tipped catheter (Millar
SPC-350) was zeroed in a 37°C water bath and advanced into the LV
chamber through a left femoral arterial introducer to
measure LV pressure. To minimize reflex sympathetic and parasympathetic
responses in these conscious animals, autonomic blockade was achieved
with intravenous esmolol (0.25 to 0.4
mg · kg-1 · min-1 infusion,
titrated
to reduce the heart rate to below 120 beats per minute) and atropine
(0.02 to 0.04 mg/kg). UL-FS 49 (Boehringer-Ingelheim), a highly
sinoatrial node–specific, negative chronotropic agent that does
not change the QT interval, inotropic state, or systolic or
diastolic blood pressures,13 was administered
as necessary to lower the heart rate to <120 beats per minute (average
dose, 5.6±1.4 mg during the Pre-MVR studies); all dogs were atrially
paced at 120 beats per minute. Baseline hemodynamic and
biplanar videofluoroscopic data recordings were obtained during
steady-state conditions and over a
physiological range of peak LV systolic pressures
during caval occlusion (sufficient to reduce arterial
systolic pressure by at least 40 mm Hg). After release of the
occluders, the dogs were allowed to recover for 3 to 5 minutes. Each
run was then repeated at a higher inotropic state, after administration
of a bolus of calcium chloride (250 mg IV). All data acquisition runs
containing premature ventricular contractions were
discarded, and the run was repeated.
Surgical Preparation: MVR
Seven to 38 days (average,
13±8 days) after myocardial marker
placement, the dogs were returned to the operating room, where they
were anesthetized and intubated as previously described, and a
micromanometer-tipped catheter was introduced
to monitor arterial pressure. The heart was again exposed
by a left thoracotomy and suspended in a pericardial cradle. After
administration of heparin (300 IU/kg), the right femoral artery and
right atrium were cannulated (12F to 16F USCI arterial and
36F two-stage USCI venous cannulas, respectively), and the animal
was placed on cardiopulmonary bypass with a roller pump
(Pemco Corp) and a bubble oxygenator (Harvey H-1300, Bard
Cardiopulmonary) and Ringer's lactate prime. The
pulmonary artery was vented, the ascending aorta
cross-clamped, and the heart arrested with 500 mL of cold (6°C)
crystalloid antegrade cardioplegia. Myocardial temperature was
subsequently maintained at <14°C with topical cold saline. The
mitral valve was exposed via a left atriotomy, and the integrity of the
subvalvular apparatus was verified visually.
The animals were randomized to undergo conventional MVR (MVR-Cut, with
complete excision of the subvalvular apparatus
and leaflets, n=7) or chord-sparing MVR (MVR-Intact, n=7).
During
MVR-Cut, the chordae to the anterior and posterior leaflets were
divided, and any redundant anterior mitral leaflet tissue was excised.
In contrast, in the MVR-Intact group, no chordae were disturbed, and
the anterior and posterior leaflets were left intact except for small
incisions in the anterior leaflet in certain animals to prevent outflow
tract obstruction. A bioprosthetic valve was then sutured
to the mitral annulus with interrupted horizontal mattress sutures;
specific valve sizes were comparable between the two groups. The
atriotomy was closed, air was evacuated from the heart, and the aortic
cross-clamp was released. After satisfactory rewarming and
resuscitation, the animals were weaned from cardiopulmonary
bypass and decannulated. The heparin effect was then reversed with
protamine sulfate, the pericardium loosely reapproximated, and an
intercostal block performed. Chest tubes were then placed and attached
to 20 cm H2O suction, and the chest was closed. Sodium
nitroprusside (up to 5
µg · kg-1 · min-1) was
administered
early after surgery to keep the systolic blood pressure <100 mm Hg.
The animals were weaned off the ventilator and extubated after
normalization of arterial blood gases. Postoperative pain
management was provided with either oxymorphone or butorphanol tartrate
(Torbugesic 0.4 to 1.2 mg/kg IV or SC, Fort Dodge Laboratories, Inc).
The animals were then returned to the intensive care unit. Typically,
the animals were ambulatory and eating within 24 hours of the
procedure.
Experimental Protocol After MVR
Nine to 20 days (14±3
days) after MVR, the animals underwent a
second catheterization procedure (Post-MVR). The
animals were taken to the cardiac catheterization
laboratory and once again mildly sedated with diazepam and supplemental
ketamine. Sympathetic and parasympathetic blockade was achieved
with intravenous esmolol and atropine, and UL-FS 49 was
administered as needed to lower the heart rate to <120 beats per
minute (average dose, 2.4±2.2 mg during the Post-MVR studies), and the
dogs were atrially paced at 120 beats per minute.
Hemodynamic and videofluoroscopic data were then
acquired during steady-state conditions and during transient vena
caval occlusion before and after inotropic intervention (ie, calcium
bolus). At the conclusion of the study, the dogs were euthanatized with
B-Euthanasia (0.2 mg/kg IV; Schering-Plough Animal Health Corp); proper
positioning of the myocardial markers was thereafter confirmed.
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 Academy of Sciences and published by the National Institutes of Health (DHEW [NIH] publication 85-23, revised 1985). The study was approved by the Stanford Medical Center Laboratory Research Animal Review Committee and conducted according to Stanford University policy.
Data Acquisition
All imaging studies were conducted with the
animal in the supine
position with a General-Electric MLX biplanar L-U arm system (General
Electric Medical Division) with the image intensifiers in the 6-in.
boost fluoroscopic mode. The 45° RAO and 45° LAO biplane images
were recorded simultaneously on Sony U-Matic 5800
3/4-in. videocassette recorders with the x-ray pulses
synchronized by a master synchronization oscillator at 60 Hz. The
analog LV pressure signal was digitized and recorded in digital
format on each individual video image with a custom-designed
intelligent video controller (Control Video Corp); the peak ECG R wave
was detected electronically and digitally encoded and recorded as
an end-diastolic timing marker. At the completion of
the study, images of grids containing 1-cm squares and biplanar images
of a three-dimensional spiral phantom of known dimensions were
recorded to determine radiographic distortion and
magnification factors. The two-dimensional coordinates of each
marker in each projection were digitized frame by frame with a
semiautomated, computerized myocardial marker detection system
(Hewlett-Packard RS/20 equipped with Matrox MVP/AT/NP image processing
boards) and custom-designed image processing and digitization
software developed in our laboratory.14 The data from the
two views were corrected for x-ray magnification and distortion,
and the RAO and LAO marker coordinates were merged by use of custom
software to yield the three-dimensional (x,
y, z) coordinates of each marker every 16.7
milliseconds, as previously described.15 With this system,
determination of marker position is accurate and reproducible with a
mean overall error of 0.1±0.6 mm.15
Data Analysis
Hemodynamics. To minimize the
effects
of intrathoracic pressure variation, only end-expiratory beats were
selected for analysis. Global LV volume was calculated for each
videofluoroscopic frame by use of a multiple tetrahedral model (see
"Appendix"). For each cardiac cycle, end-diastole
was defined as the videofluoroscopic frame that contained the ECG
R-wave marker; end-systole was defined as the time of the maximum
LV pressure/volume ratio.16 SV (=EDV-ESV) was
calculated.
LV afterload was estimated as the effective arterial
elastance (EA=ESP/SV). For each cardiac cycle, the time
derivative of the LV pressure signal was computed to determine LV
dP/dtmax and
dP/dtmin.
Global LV systolic function. Global LV systolic performance was assessed with two relatively load-insensitive indexes of contractility: (1) the ESPVR (chamber elastance)17 and (2) the PRSW.18 To define the ESPVR, LV end-systolic pressure (Pes) and volume (Ves) points were determined for each cardiac cycle analyzed during abrupt preload reduction.19 20 By least-squares linear regression, a straight line was fitted to these points, yielding the equation Pes=Ees(Ves-V0), where Ees and V0 are the slope and the volume axis intercept, respectively. To avoid problems with linear extrapolation of a possibly curvilinear relation beyond the range of data recorded (which can lead to negative volume intercept values),21 22 the LV ESV required to generate an ESP of 100 mm Hg (Ees100) was also computed.23 24 This parameter incorporates changes in both the slope and LV volume-axis intercept of the LV ESPVR; therefore, Ees100 can reliably be used as a single variable to quantify LV contractility. It should be noted that an increase in Ees100 occurs when a greater EDV is required to obtain an ESP of 100 mm Hg, namely, impaired LV chamber contractility, whereas a decrease in Ees100 indicates that a smaller EDV is required to obtain an LVESP of 100 mm Hg, or improved LV contractility.
While Ees provides one measure of LV contractility, it assumes an initial volume axis intercept of 0 mL (since end systole was previously defined by the peak LV pressure/volume ratio). A similar and perhaps more accurate approach is to consider the time-varying elastance model: P(t)=E(t)[V(t)-V0]. An initial estimate of V0 is provided from the previous Ees calculations and then iterated until the new value of V0 differs by <1%. We refer to the iterated value of the slope and intercept of this relation as Eiter and Viter, respectively. A physiological LV volume value at a developed pressure of 100 mm Hg can then be determined, ie, Eiter100.
External pressure-volume LV stroke work (SW) was
computed as the
area within each LV pressure-volume loop, ie, the integral of LV
pressure and volume throughout the cardiac cycle:
SW=
P · dV. The
LV PRSW relation was obtained by the following linear regression:
SW=Mw(EDV-Vw), where Mw and
Vw represent the slope and volume axis intercepts,
respectively. Again, to avoid linear extrapolation beyond the range of
the observed data, the EDV required to obtain an SW of 1000
mm Hg · mL (MW1000) was calculated and used to
characterize LV chamber systolic
function.23
LV diastolic function. LV
function in early
diastole was assessed by use of the time constant of
isovolumic relaxation. This was computed for each beat from a
monoexponential equation (with a
zero-pressure asymptote) as
P(t)=P0exp(-t/), where P(t) is the instantaneous
LV pressure, P0 the initial pressure before the onset of
the pressure decay, and the relaxation time constant. Pressure data
points were considered starting from dP/dtmin
through LVPmin. was then compared with a global measure
of diastolic recoil (see below).
Computation of LV twist and recoil. LV twist was
computed
with a modification25 of the method described by Beyar et
al.26 The cartesian coordinates (x, y, z) of
each epicardial myocardial marker were first transformed into a local
cylindrical coordinate system (r, 
, z). On a
frame-by-frame basis, each marker location was translated and
rotated such that the midpoint of the chord connecting the anterobasal
and posterobasal markers was positioned at the origin and the chord
itself was perpendicular to the z axis, with the line
passing through the apical marker and the midpoint of the basal chord
aligned along the z axis. For each marker in each frame, the
change in azimuth (ie, circumferential rotational angle [(t)])
relative to end diastole (arbitrarily assigned as
0=0.0°) was computed; a positive angle change was
defined as counterclockwise rotation as viewed from the cardiac
apex.
Twist was defined as the average longitudinal gradient of
circumferential rotation. The instantaneous torsion, (t), of each
marker (except the apical and basal reference markers) was plotted
against the longitudinal distance (along the z axis) from
the apex; linear regression of on z provided the slope
of this relation in radians per centimeter and was defined as global LV
chamber twist. Twist was determined for each frame throughout the
cardiac cycle and then smoothed by use of a three-point running
average. Negative twist was defined as counterclockwise rotation of the
apex with respect to the base (as viewed from the apex); minimum twist
(Tmin) was defined as the peak positive (clockwise) LV
twist occurring in early systole; and maximum twist
(Tmax) was defined as peak negative
(counterclockwise) LV twist.
A representative temporal relationship of
LV twist is
illustrated in Fig 2. Early in systole, there is
typically a small initial clockwise twist; this probably
represents a predominance of subendocardial fiber shortening
(compared with subepicardial fiber shortening) during the initial
electrical activation of the ventricle.27 This initial
deflection is followed by a linearly increasing, prolonged
counterclockwise twist extending throughout ejection. The first
derivative of twist with respect to time (dT/dt,
rad · cm-1 · s-1) was
calculated;
peak systolic twist rate (-dT/dtmin [greater
negative twist]) was determined, as was the average systolic twist
rate, Msys, defined as the slope (determined by
linear regression) of the twist rate from peak positive twist
(Tmin) to peak negative twist
(Tmax).
|
Diastole was divided into two phases based on the major inflection point of the temporal twist-time relation. This inflection point has been shown in previous studies to occur at 15% to 20% of diastolic filling in both humans and animals25 28 ; our findings were consistent using either of these values. As with the systolic data, the peak untwist (recoil) rate (+dT/dtmax [less negative twist]) during the early, rapid-filling phase of diastole was calculated; early rapid diastolic untwist (Mear-dia, "recoil") was defined as the slope (again, from linear regression) of the twist-time relation from end systole to 15% LV chamber filling.
Statistical Analysis
All data are reported as mean±SD.
Pre-MVR and Post-MVR raw data
before and after inotropic stimulation were compared within each of the
groups by repeated-measures ANOVA. Comparisons between the two
surgical groups were made by univariate ANOVA (with the
animal identification number, surgical group, and study [Pre-MVR and
Post-MVR, with and without calcium] coded as dummy variables);
when indicated by a significant F statistic
(P<.05), specific differences were isolated with post hoc
Student's t tests using the Bonferroni correction.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Hemodynamics
Table 1
summarizes the
hemodynamic
results for both groups (MVR-Cut and MVR-Intact), before and after MVR,
during baseline conditions and after calcium. There was no difference
in any hemodynamic parameters between the
groups before MVR. After MVR, before calcium infusion (baseline
conditions), LV EDV and ESV were significantly lower in both groups. LV
EDV and ESV after calcium infusion were also lower in the MVR-Cut group
but not in the MVR-Intact group. LV ESP was significantly lower
post-MVR after calcium infusion in both groups versus pre-MVR values.
In the postoperative MVR-Intact group (baseline), LV EDP was
significantly lower compared with pre-MVR values.
|
Despite the use of wall thickness markers to estimate LV wall volume, these LV volume calculations are probably overestimates due to the inability to account for papillary muscle mass and endotrabecular obliteration (as seen angiographically). This overestimation leads to an underestimation of LV ejection fraction. This technique to estimate LV volume, however, has been shown to correlate highly with LV endocardial angiocardiographic volumes (r=.92±.07, SEE=4.42±1.68 mL)29 ; the marker-derived stroke volumes have also been demonstrated to correlate well with data derived from aortic flow probe measurements (r=.96±.04, SEE=0.56±0.21 mL).29 There were no significant differences between the two surgical groups in any of the hemodynamic variables during the four studies.
Ejection Performance and Diastolic
Relaxation
Table 2 summarizes other LV
hemodynamic parameters and the time
constant of diastolic relaxation; there were no differences
between the two groups before MVR. As expected, increased inotropic
state (after calcium infusion) increased
dP/dtmax, dP/dtmin,
and SW in both groups. It is interesting that despite significantly
smaller LV EDV and ESV in the MVR-Cut group after MVR,
dP/dtmax and dP/dtmin did not
change, as might be expected for this preload-dependent
variable. In addition, LV dP/dtmax was higher
(P=.068) in the MVR-Intact group after MVR compared with the
MVR-Cut group.
|
After MVR, the MVR-Intact group data during baseline
conditions
demonstrated faster LV relaxation (ie, reduced ) compared with
before surgery; this was not seen in the MVR-Cut group. Increased
inotropic state (Ca2+) reduced in both groups,
as would be expected.
LV Systolic Function
LV systolic function results are
summarized in Table 3. Inotropic stimulation significantly
increased all
measures of contractility. No significant differences
between the two groups were seen either before or after MVR, although
Ees after MVR was slightly higher (P=.095) in
the MVR-Intact group. Intergroup comparison between pre- and post-MVR
data showed a definite trend toward increased
contractility in the MVR-Intact group, with increases
in Ees (P=.061), Eiter
(P=.087), Eiter100 (P=.091), and
Mw1000 (P=.036). In both groups, the intercept
of the PRSW decreased after MVR, but this change was more dramatic and
statistically significant when the chordae were severed (MVR-Cut,
P=.018; MVR-Intact, P=.084).
|
LV Twist and Recoil
Compared with the pre-MVR baseline
studies, Tmin
decreased significantly (P=.013) and the magnitude of
Tmax increased (P=.051) in the MVR-Intact group
(Table 4). Conversely, in the MVR-Cut group, only the
data after Ca2+ infusion revealed significant
differences in Tmax (P=.007) and peak early
diastolic recoil, Tear (P=.043).
Comparison of three LV twist parameters
(Tmin, Tmax, and rate of early
diastolic untwist [recoil], Tear-dia) did not
indicate any significant differences between the groups either before
or after MVR. The rate of change of twist during systole
(Msys) increased significantly after calcium in both groups
both before and after MVR; however, no significant difference between
the MVR-Intact and MVR-Cut groups was evident. A similar response to
inotropic stimulation was seen during early diastole; in
the MVR-Cut group, LV recoil rate increased before versus after MVR
(P=.062), as was also seen in the MVR-Intact group
(P=.081) (Table 4).
|
Our hypothesis was
that a direct relation existed between the degree of
systolic twist and LV function; this hypothesis was assessed by
analysis of the dependence of LV systolic function (as assessed
by Ees) on LV systolic twist (as characterized by
Msys). Fig 3 summarizes the relation between
Msys and Ees during the pre-MVR and post-MVR
conditions, both with and without chord severing. Before MVR, in both
the MVR-Intact and MVR-Cut groups, there was a significant
(P=.001 for MVR-Intact, P=.004 for MVR-Cut
[probability that the slope of the regression is different from
zero]) indirect relation between Msys and
Ees, but no significant difference was apparent
between the two groups. After surgery, in the MVR-Intact group there
was a slight (albeit statistically insignificant) increase in the slope
of this relation, whereas there was no correlation whatsoever in the
MVR-Cut group. Similar findings were observed regarding the
diastolic recoil data: In both the MVR-Intact and MVR-Cut
groups, the LV diastolic recoil rate appeared to increase,
but again, these changes only approached statistical significance. Only
in the MVR-Intact group was this associated with a higher rate of LV
relaxation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Previously, we used an in situ isovolumic canine heart model and calculated the maximum time-varying elastance, Emax, from the ESPVR to analyze changes in LV contractility after chord-severing MVR. One early experiment12 demonstrated a large (47%) and significant (P<.001) reduction in Emax after division of all mitral chordae tendineae, without a significant change in the ESPVR volume intercept, V0. A subsequent study examined the relative contributions of the anterior and posterior leaflets with respect to their chordae tendineae in terms of global LV systolic function in another isovolumic experimental canine model; the chordae subtending both the anterior and posterior leaflets had a quantitatively similar but additive positive contribution to global LV systolic pump function.30
In the present study, LV EDV and ESV fell in both groups after MVR. Such smaller LV volumes are seen in patients with mitral regurgitation early after mitral repair or MVR, especially when operative intervention is undertaken before an irreversible degree of LV "volume-overload" cardiomyopathy has occurred. The experimental model used in this study, however, was normal canine hearts. These decreases in LV volume may be due in part to reduced intravascular volume, although the animals were not clinically hypovolemic. Moreover, these results are similar to those described by David et al31 ; using radionuclide ventriculography, they compared LV systolic function before surgery and 3 to 6 months after surgery in a small number of patients who had conventional MVR (n=15) or chord-preserving MVR (n=12), but this was not a randomized investigation. In both groups, LV ESV, EDV, and SV decreased significantly after MVR; however, ejection fraction decreased in patients who had conventional MVR but did not change significantly in those who underwent chord-preserving MVR.31 Rozich et al9 studied 15 nonrandomized heterogeneous patients with chronic mitral regurgitation 7 to 10 days after MVR with (n=8) and without (n=7) preservation of some or all chords. These patients were assessed on the basis of preoperative and postoperative two-dimensional, M-mode, and Doppler color flow echocardiography studies. LV volumes and ESS were then derived. Compared with the preoperative study, chord-severing MVR was associated with an increase in LV ESV and ESS, whereas chord-sparing MVR resulted in a decrease in LV ESV and ESS. Ejection performance was preserved with chord-sparing MVR but fell with chord-severing MVR, similar to the results of David et al31 and other investigators in the past. Although LV ESV decreased after surgery in these dogs both with and without chord division, the difference between Rozich's results and our findings might be a reflection of early LV remodeling after MVR in human ventricles that had been subjected to long-standing abnormal loading conditions (ie, chronic MR).
In assessment of LV contractility, a trend toward improved LV performance was seen 2 weeks after MVR (compared with before MVR) in the MVR-Intact group, without any comparable increase in the MVR-Cut group. This finding per se is consonant with previous observations from clinical and experimental studies in both intact and isolated heart models; however, this is somewhat counterintuitive because the introduction of an artificial valve (which would alter normal mitral annular geometry) might be expected to have a detrimental effect on LV systolic function. One possibility is that the insertion of a rigid prosthesis did indeed adversely alter normal mitral annular dynamics, which caused a compensatory increase in myocardial contractility to produce the same level of cardiac output. This possibility is debatable but is not without precedent; data from the study by Ishihara et al8 of chord-preserving MVR using a chronic canine mitral regurgitation model demonstrated a trend toward an increasing slope of the end-systolic stress-volume relation (corrected for LV mass) after compared with before MVR, which returned to baseline 3 months after MVR. Mitral annulus substitution may also explain the clinical notion that LV systolic function (at least early after surgery) is superior after mitral valve repair if a flexible annuloplasty ring is used rather than a circumferential rigid ring.32 Nevertheless, previous experimental studies in normal hearts have not corroborated these pathological human findings.33 34 35
Concurrent with the changes in LV contractility in the
MVR-Intact group, perturbations in systolic LV twist
parameters (namely, Tmin and
Tmax) were seen. Increased LV twist associated with
augmented contractility probably reflects a link
between these two variables, as previously noted by Moon et
al36 in human heart transplant patients, but any inherent
cause-and-effect relation is still enigmatic. This putative
coupling, however, appears to pivot on maintenance of mitral
annulopapillary continuity. Our hypothesis predicted that we would see
an uncoupling between systolic LV function and torsional deformation
after MVR with chord division, which, in fact, did occur. Additionally,
this coupling also probably had an influence on diastolic
LV behavior. The implication is that the left ventricle responded to
the operative changes (eg, a rigid prosthesis) with augmented
diastolic LV filling and systolic function, but only
when the mitral subvalvular apparatus was
intact after MVR. Shintani and Glantz37 investigated the
effect of chord division on the diastolic properties of the
left ventricle. Using a unique volume-clamping technique in dogs,
these investigators demonstrated that disruption of the
subvalvular apparatus increases the intercepts
of the LV ESPVR, end-diastolic pressure-volume, and
SW-EDV relations without any significant changes in slopes.
Additionally, they showed an increase in the isovolumic relaxation time
constant (), an important determinant of early diastolic
filling. This well-designed study was not confounded by the
influence of cardiopulmonary bypass and evaluated LV
diastolic function under a variety of conditions.
Thus, we conclude that one mechanism for the improved clinical outcome after chord-sparing MVR is related to maintaining the normal relation between LV systolic function and torsional deformation; the subvalvular apparatus may possibly modulate "valvular-ventricular interaction" in terms of translation of LV twist into augmented LV systolic pump performance. Our data also suggest a coupling of early diastolic recoil and early LV filling (as quantified by early diastolic relaxation).
Study Limitations
One drawback of animal studies is the
difficulty in mimicking the
clinical situation of chronic mitral regurgitation. We
studied normal, healthy dogs before and after excision of a normal
subvalvular apparatus. However, our results are
not really different from those of Ishihara et al,8 who
created an experimental model of chronic mitral
regurgitation using a closed-chest
chord-rupturing technique. Three months after chronic mitral
regurgitation, the animals underwent MVR with
preservation of the subvalvular apparatus. LV
EDV, contractile function, ESS, and ejection fraction normalized 3
months after MVR. Additionally, isolated cardiomyocyte
function (as assessed by viscosity-velocity curves) was similar in
the postoperative ventricles compared with baseline.
Our study used two
separate groups randomized to undergo MVR with
or without sparing of the subvalvular
apparatus. The size of each group limited the power for
some comparisons to 
40% (assuming a desired significance of
P=.05). For the most part, however, we achieved significant
differences despite the relatively small number of subjects in each
group. Ideally, such a study would use a technique whereby each animal
acted as its own control, ie, each animal studied before and then after
MVR (with the chords intact) followed by restudy after the chords had
been severed. The practicality of such a protocol has been investigated
by our group; currently, we do not think that this approach is
experimentally feasible without a prohibitive mortality risk.
Finally,
the measurement techniques used in this study may not be
sensitive enough to detect subtle but real differences between the two
surgical groups. Sources of error would include marker digitization
(the accuracy and reproducibility of which have previously been shown
to be 0.06 and 0.18 mm, respectively) and LV volume derivation. As
noted above, the multiple tetrahedral technique to compute LV volume
has correlated well with both angiographic techniques and aortic flow
probe SV integration. Shintani and Glantz37 used both
ultrasonic crystals and conductance catheter measurements; the
calculated LV ejection fraction from both methods was 19%.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This study was supported by grant HL-29589 from the National Heart, Lung, and Blood Institute and the Veterans Administration Medical Research Service. Drs DeAnda and Komeda are Carl and Leah McConnell Cardiovascular Surgical Research Fellows. Dr DeAnda was also supported by Individual National Research Service Award HL-08928 from the National Heart, Lung, and Blood Institute. We acknowledge the expert technical assistance of Cynthia E. Handen, BA; Erin K. Schultz, BS; Geraldine C. Derby, RN, BS; Joshua Cohen, BA; Mary K. Zasio, BA; and Carol W. Mead, BA, in the performance of this work and Phoebe Taboada for help in preparing the manuscript and figures.
Calculation of LV Volumes
Global LV volume was
calculated by use of a multiple tetrahedral
model. The 17 epicardial markers divided the epicardial surface into 16
regions: 4 between the basal and basoequatorial levels, 4 between the
basoequatorial and equatorial levels, 4 between the equatorial and
apicoequatorial levels, and 4 between the apicoequatorial level and
apex (see Fig 1
). Each region was defined by 4 adjacent markers
and the
midpoints of their respective levels. For example, the equatorial
anterolateral region was defined by the (1) anterior apicoequatorial,
(2) anterior equatorial, (3) lateral apicoequatorial, and (4) lateral
equatorial markers plus the midpoints of the (5) apicoequatorial and
(6) equatorial planes. These 6 points were divided into 3 adjacent
tetrahedra with the vertices of each represented by the
x, y, z coordinates of 4 of the 6 points. The volume of each
tetrahedron was calculated as one sixth of the following
determinant:
 |
Tetrahedral
volume was then multiplied by a correction factor to
yield the volume of a conical section:
/[4xsin(/4)xcos(/4)] for the basal and
equatorial regions
and /[2xsin(/4)xcos(/4)] for the apical
regions. All 16
regions were then combined to yield total LV volume, including the mass
of the LV wall. With the presence of multiple wall thickness markers,
the volume of the LV wall was able to be estimated, and subsequently
corrected for, to yield an approximate LV chamber volume. For the 12
cylindrical segments, this was
volendo=(/4) · H · {2[volepi/( · H)]1/2-WT}2,
where volendo is the corrected volume, H is the average
height of the cylindrical segment, WT is the average wall thickness for
that segment (calculated from 2 to 4 wall thickness pairs), and
volepi is the original volume. In a similar manner, the 4
apical segments were approximated using
volendo=(/6) · [(6 · volepi/)1/3-WT]3.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Lillehei CW, Levy MJ, Bonnabeau RC. Mitral valve replacement with preservation of the papillary muscles and the chordae tendineae. J Thorac Cardiovasc Surg. 1964;47:532-543.
2. Borst HG. Preservation of the tensor apparatus of the mitral valve. J Heart Valve Dis. 1993;2:148-149. Editorial. [Medline] [Order article via Infotrieve]
3. Carabello BA. The mitral valve apparatus: is there still room to doubt the importance of its preservation? J Heart Valve Dis. 1993;2:250-252. Editorial. [Medline] [Order article via Infotrieve]
4. Okita Y, Miki S, Kusuhara K, Ueda Y, Tahata T, Yamanaka K, Higa T. Analysis of left ventricular motion after mitral valve replacement with a technique of preservation of all chordae tendineae. J Thorac Cardiovasc Surg. 1992;104:786-795. [Abstract]
5. Okita Y, Miki S, Ueda Y, Tahata T, Sakai T, Matsuyama K. Comparative evaluation of left ventricular performance after mitral valve repair or valve replacement with or without chordal preservation. J Heart Valve Dis. 1993;2:159-166. [Medline] [Order article via Infotrieve]
6. Feikes HL, Daugharthy JB, Perry JE, Bell JH, Hieb RE, Johnson GH. Preservation of all chordae tendineae and papillary muscle during mitral valve replacement with a tilting disc valve. J Cardiac Surg. 1990;5:81-85. [Medline] [Order article via Infotrieve]
7. Horstkotte D, Schulte HD, Bircks W, Strauer BE. The effect of chordal preservation on late outcome after mitral valve replacement: a randomized study. J Heart Valve Dis. 1993;2:150-158. [Medline] [Order article via Infotrieve]
8. Ishihara K, Zile MR, Kanazawa S, Tsutsui H, Urabe Y, DeFreyte G, Carabello BA. Left ventricular mechanics and myocyte function after correction of experimental chronic mitral regurgitation by combined mitral valve replacement and preservation of the native mitral valve apparatus. Circulation. 1992;86(suppl II):II-16-II-25.
9.
Rozich JD, Carabello BA, Usher BW, Kratz JM, Bell AE,
Zile MR. Mitral valve replacement with and without chordal
preservation in patients with chronic mitral
regurgitation: mechanisms for differences in
postoperative ejection performance.
Circulation. 1992;86:1718-1726.
10.
Tischler MD, Cooper KA, Rowen M, LeWinter MM.
Mitral valve replacement versus mitral valve repair: a
Doppler and quantitative stress echocardiographic
study. Circulation. 1994;89:132-137.
11. Sarris GE, Miller DC. Valvular-ventricular interaction: the importance of the mitral chordae tendineae in terms of global left ventricular systolic function. J Cardiac Surg. 1988;3:215-234. [Medline] [Order article via Infotrieve]
12.
Hansen DE, Cahill PD, DeCampli WM, Harrison DC, Derby
GC, Mitchell RS, Miller DC.
Valvular-ventricular interaction:
importance of the mitral apparatus in canine left
ventricular systolic performance.
Circulation. 1986;73:1310-1320.
13. Schipke JD, Harasawa Y, Sugiura S, Alexander J, Burkhoff D. Effect of a bradycardic agent on the isolated blood-perfused canine heart. Cardiovasc Drugs Ther. 1991;5:481-488. [Medline] [Order article via Infotrieve]
14. Niczyporuk MA, Miller DC. Automatic tracking and digitization of multiple radiopaque myocardial markers. Comput Biomed Res. 1991;24:129-142. [Medline] [Order article via Infotrieve]
15. Daughters GT, Sanders WJ, Miller DC, Schwarzkopf A, Mead CW, Ingels NB. A comparison of two analytical systems for three-dimensional reconstruction from biplane videoradiograms. Proc Comput Cardiol (IEEE). 1988;15:79-82.
16.
Suga H, Sagawa K, Shoukas AA. Load independence
of the instantaneous pressure-volume ratio of the canine left
ventricle and effects of epinephrine and heart rate on the
ratio. Circ Res. 1973;32:314-322.
17. Sagawa K, Suga H, Shoukas AA, Bakalar KM. End-systolic pressure-volume ratio: a new index of contractility. Am J Cardiol. 1977;40:748-753. [Medline] [Order article via Infotrieve]
18.
Glower DD, Spratt JA, Snow ND, Kabas JS, Davis JW,
Olsen CO, Tyson GS, Sabiston DC, Rankin JS. Linearity of the
Frank-Starling relationship in the intact heart: the concept of preload
recruitable stroke work. Circulation. 1985;71:994-1009.
19.
Kono A, Maughan WL, Sunagawa K, Hamilton K, Sugawa K,
Weisfeldt ML. The use of left ventricular
end-ejection pressure and peak pressure in the estimation of the
end-systolic pressure-volume relationship.
Circulation. 1984;70:1057-1065.
20. Alyono D, Larson VE, Anderson RW. Defining end-systole for end-systolic pressure-volume ratio. J Surg Res. 1985;39:344-350. [Medline] [Order article via Infotrieve]
21.
Burkhoff D, Sugiura S, Yue DT, Sagawa K.
Contractility-dependent curvilinearity of
end-systolic pressure-volume relations. Am J
Physiol. 1987;252:H1218-H1227.
22.
Kass DA, Beyar R, Lankford E, Heard M, Maughan WL,
Sagawa K. Influence of contractile state on curvilinearity of
in situ end-systolic pressure-volume
relations. Circulation. 1989;79:167-178.
23.
Toombs CF, Vinten-Johansen J, Yokoyama H, Johnston WE,
Julian JS, Cordell AR. Nonlinearity of indexes of left
ventricular performance: effects on estimation of
slope and diameter axis intercepts. Am J Physiol. 1991;260:H1802-H1809.
24. Yun KL, Niczyporuk MA, Sarris GE, Fann JI, Miller DC. Importance of mitral subvalvular apparatus in terms of cardiac energetics and systolic mechanics in the ejecting canine heart. J Clin Invest. 1991;87:247-254.
25.
Yun KL, Niczyporuk MA, Daughters GT II, Ingels NB Jr,
Stinson EB, Alderman EL, Hansen DE, Miller DC. Alterations in
left ventricular diastolic twist mechanics
during acute human cardiac allograft rejection.
Circulation. 1991;83:962-973.
26.
Beyar R, Yin FCP, Hausknecht M, Wiesfeldt ML, Kass DA.
Dependence of left ventricular twist-radial
shortening relations on cardiac cycle phase. Am J
Physiol. 1989;257:H1119-H1126.
27.
Ingels NB Jr, Hansen DE, Daughters GT II, Stinson EB,
Alderman EL, Miller DC. Relation between longitudinal,
circumferential, and oblique shortening and torsional deformation in
the left ventricle of the transplanted human heart.
Circ Res. 1989;64:915-927.
28.
Buchwalter MB, Weiss JL, Rogers WJ, Zerhouni EA,
Weisfeldt ML, Beyar R, Shapiro EP. Noninvasive quantification of
left ventricular rotational deformation in normal humans
using magnetic resonance imaging myocardial tagging.
Circulation. 1990;81:1236-1244.
29. Moon MR, DeAnda A, Daughters GT, Ingels NB, Miller DC. Experimental evaluation of different chordal preservation methods during mitral valve replacement. Ann Thorac Surg. 1994;58:931-944. [Abstract]
30. Hansen DE, Cahill PD, Derby GC, Miller DC. Relative contributions of the anterior and posterior mitral chordae tendineae to canine global left ventricular function. J Thorac Cardiovasc Surg. 1987;93:45-55. [Abstract]
31. David TE, Burns RJ, Bacchus CM, Druck MN. Mitral valve replacement for mitral regurgitation with and without preservation of chordae tendineae. J Thorac Cardiovasc Surg. 1984;88:718-725. [Abstract]
32. David TE, Komeda M, Pollick C, Burns R. Mitral valve annuloplasty: the effects of the type on left ventricular function. Ann Thorac Surg. 1989;47:524-528. [Abstract]
33. Spence PA, Peniston CM, David TE, Mihic N, Jabr AK, Narini P, Salerno TA. Toward a better understanding of the etiology of left ventricular dysfunction after mitral valve replacement: an experimental study with possible clinical implications. Ann Thorac Surg. 1986;41:363-371. [Abstract]
34. Rayhill SC, Castro LJ, Nizyporuk MA, Ingels NB, Daughters GT, Derby GC, Tye TL, Bolger AF, Miller DC. Rigid ring fixation of the mitral annulus does not impair left ventricular systolic function in the normal canine heart. Circulation. 1992;86(suppl II):II-26-II-38.
35. Castro LJ, Moon MR, Rayhill SC, Niczyporuk MA, Ingels NB, Daughters GT, Derby GC, Miller DC. Annuloplasty with flexible or rigid ring does not alter left ventricular systolic performance, energetics, or ventricular-arterial coupling in conscious, closed-chest dogs. J Thorac Cardiovasc Surg. 1993;105:643-659. [Abstract]
36.
Moon MR, Ingels NB, Daughters GT, Stinson EB, Hansen
DE, Miller DC. Alterations in left ventricular twist
mechanics with inotropic stimulation and volume loading in human
subjects. Circulation. 1994;89:142-150.
37.
Shintani H, Glantz SA. Effect of disrupting the
mitral apparatus on left ventricular function
in dogs. Circulation. 1993;87:2001-2015.
This article has been cited by other articles:
 |
|
 |
|
  S. Sirivella and I. Gielchinsky Clinical outcomes of surgery of mitral valve regurgitation and coronary artery bypass grafting Interactive CardioVascular and Thoracic Surgery, August 1, 2006; 5(4): 392 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Helle-Valle, J. Crosby, T. Edvardsen, E. Lyseggen, B. H. Amundsen, H.-J. Smith, B. D. Rosen, J. A.C. Lima, H. Torp, H. Ihlen, et al. New Noninvasive Method for Assessment of Left Ventricular Rotation: Speckle Tracking Echocardiography Circulation, November 15, 2005; 112(20): 3149 - 3156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Torrent-Guasp, M. J. Kocica, A. Corno, M. Komeda, J. Cox, A. Flotats, M. Ballester-Rodes, and F. Carreras-Costa Systolic ventricular filling Eur. J. Cardiothorac. Surg., March 1, 2004; 25(3): 376 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. De Canniere, J.-L. Jansens, P. Unger, and J.-L. Le Clerc Left Ventricular Outflow Tract Obstruction After Mitral Valve Replacement Ann. Thorac. Surg., December 1, 1997; 64(6): 1805 - 1806. [Abstract] [Full Text]
|
|
|
|
|
|
E. Esper, F. D. Ferdinand, S. Aronson, and R. B. Karp Prosthetic Mitral Valve Replacement: Late Complications After Native Valve Preservation Ann. Thorac. Surg., February 1, 1997; 63(2): 541 - 543. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||