| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2000;101:1925.)
© 2000 American Heart Association, Inc.
Clinical Investigation and Reports |
From the First Department of Internal Medicine Kobe University School of Medicine, Kobe, Japan.
Correspondence to Hideyuki Takaoka, MD, First Department of Internal Medicine, Kobe University School of Medicine, 7-5-2, Kusunoki-cho, Chuo-ku, Kobe 6500017, Japan.
| Abstract |
|---|
|
|
|---|
Methods and ResultsWe studied 15 patients with heart failure due
to idiopathic dilated cardiomyopathy (mean ejection
fraction 0.33). We examined LV contractility
(Emax, the slope of end-systolic pressure-volume
relation), LV external work (EW), myocardial oxygen consumption
(M
O2), and mechanical efficiency
(measured as EW/M
O2) with the use of
conductance and coronary sinus thermodilution catheters before
and during dobutamine (DOB) infusion via a
peripheral vein (4.8±0.3 µg ·
kg-1 · min-1 IV). Heart rate was kept
constant with atrial pacing. We carried out a similar protocol during
the intracoronary infusion of the NO synthase
inhibitor
NG-monomethyl-L-arginine
(L-NMMA; 200 µmol). DOB increased Emax, EW, and
M
O2 (by 77±17%, 39±5%, and 21±5%,
respectively), leading to an increase in mechanical efficiency
(25.4±3.1% to 29.6±4.1%). L-NMMA alone did not significantly change
these variables. Although the concurrent infusion of DOB with
L-NMMA increased Emax, EW, and
M
O2 (by 140±21%, 64±9%, and 35±5%,
respectively) more than DOB alone, mechanical efficiency did not
increase further (24.3±3.3% to 29.5±4.5%) because EW and
M
O2 increased in parallel.
ConclusionsThese data suggest that in patients with idiopathic
dilated cardiomyopathy, endogenous NO
spares M
O2 through attenuation of LV
contractile response to ß-adrenergic stimulation while maintaining LV
energy-converting efficiency.
Key Words: heart failure nitric oxide contractility oxygen
| Introduction |
|---|
|
|
|---|
Hare et al5 recently reported that the inhibition of NO synthase (NOS) potentiates the positive inotropic response to ß-adrenergic stimulation in patients with dilated cardiomyopathy but not in subjects with normal left ventricular (LV) function. These authors suggested that increased myocardial NOS activity in the failing human heart attenuates ß-adrenergic responsiveness.
Investigation of the pathophysiology of congestive heart failure
requires a metabolic approach to the assessment of
myocardial oxygen consumption
(M
O2) as well as a functional
approach. In addition to its potential to modulate
M
O2 indirectly via attenuation
of the contractile response, NO appears to regulate oxidative
phosphorylation in cardiac myocytes, probably by
binding to heme moieties and iron-sulfur clusters in proteins involved
in mitochondrial respiration and thus possibly regulating
M
O2. Xie et al6
revealed the role of endothelium-derived NO in the
control of cardiac mitochondrial respiration in vitro. Recently, Zhang
et al7 demonstrated that ACE inhibitors
dramatically reduce M
O2 in
isolated myocardial muscle slices through an NO-dependent mechanism.
Conversely, another study has shown that blockade of NO synthesis
reduces M
O2 in
vivo.8 Nevertheless, the way in which NO modulates
M
O2 in patients with heart
failure has not received much attention.
The purpose of the present study was to evaluate the role of NO in
cardiac mechanics and energetics in patients with idiopathic dilated
cardiomyopathy (IDC) through the use of the
relatively load-independent index of contractility,
Emax, and systolic pressure-volume area
(PVA), both of which were proposed as major determinants of
M
O2 by Suga9 and
others.10
| Methods |
|---|
|
|
|---|
|
Catheterization Procedure
Cardiac catheterization was performed via the
femoral approach on patients in a fasting condition without medication.
After routine right and left heart catheterization, a
7F thermodilution Swan-Ganz catheter (Goodtech Inc) was advanced into
the pulmonary artery, and an 8F conductance (volume) catheter
(CardioDynamics) was advanced into the LV through the right femoral
artery. The conductance catheter was attached to a stimulator/processor
(Sigma-5; CardioDynamics), and a 2F Millar Instruments catheter was
advanced into the LV through the lumen of the conductance catheter. A
4F Judkins catheter was advanced to the ostium of the left main
coronary artery via the left femoral artery for
intracoronary drug infusion and was continuously flushed at a
rate of 2 mL/min with saline containing heparin. An 8F coronary
sinus thermodilution catheter (Cordis Webster, Inc) was then advanced
into the coronary sinus through the right jugular vein. The ECG
and hemodynamic parameters were
recorded on a strip-chart recorder. Each measurement was
obtained as the mean value of
8 consecutive sinus beats.
Assessment of LV Cardiac Mechanoenergetics
The coronary sinus blood flow (CSF) was measured with
the thermodilution method.10 Coronary venous blood
was sampled from the distal lumen of the coronary sinus
catheter for oximetry. M
O2 per
minute was calculated as the product of CSF (mL/min) and the
arterial-coronary sinus oxygen content difference
(vol%) and was divided by heart rate to yield
M
O2 per beat (mL
O2/beat), as in previous
studies.10 11
LV volume measurements with a conductance catheter were described in a
previous report.11 Pressure-volume loops were recorded
for the sequence of beats during the transient decrease in preload
through the inflation of a balloon catheter (Baxter) just above the
inferior vena cava. LV contractility,
Emax is the slope of the linear
end-systolic pressure-volume relation (ESPVR) (Figure 1A
).11 We determined the
effective arterial elastance (Ea), a
variable that incorporates the values of Windkessel model elements
and heart rate as the ratio of end-systolic pressure to stroke
volume (Figure 1B
).12 The ratio of effective
Ea to ventricular elastance
(Ea/Emax)
represents ventriculoarterial coupling. We
normalized Emax and Ea
(mm Hg/mL per m2) for body surface area to
permit a comparison among patients in the present study, as
described previously.11
|
External work (EW) was calculated as the area that is bounded by the
pressure-volume trajectory of 1 beat. Systolic PVA (mm Hg/mL)
was calculated as the area bounded by the ESPVR and
end-diastolic pressure-volume relation (EDPVR) and the
systolic pressure-volume trajectory of each beat (Figure 1B
). Potential energy (PE) was calculated by subtracting EW from
PVA. We calculated mechanical efficiency as the ratio of EW (J/beat) to
M
O2 per beat (J/beat), where
1 mm Hg/ml EW and 1 mL O2 of oxygen
consumption correspond to 1.33 · 10-4 and
20 J, respectively.9
Tau was calculated from a plot of -dP/dt versus P (P= P0e-t/T+Pb), where P is LV pressure, t is the time from peak -dP/dt, T is time constant of isovolumic pressure decay, and P0 and Pb are constants determined with the data.13 LV chamber stiffness was addressed with both nonlinear regression and linear regression analyses of the EDPVR values for the same beats as for the ESPVR. LV chamber stiffness was quantified by fitting the exponential function EDP=AeßEDV to EDP and EDV through nonlinear least-squares regression, where EDP is end-diastolic pressure, EDV is end-diastolic volume, A is elastic constant, and ß is the change in the logarithmic function of LV pressure relative to the change in volume and represents an index of ventricular chamber stiffness.14 LV chamber stiffness was also assessed with the slope of the linear regression analysis of the EDPVR values [EDP=S(EDV-V0d)], where EDP is the end-diastolic pressure, S is the slope of the EDPVR, EDV is end-diastolic volume, and V0d is the diastolic volume intercept.
Study Protocol
Control Study
After routine heart catheterization, atrial
pacing was initiated at 90 bpm or at 15 bpm above the baseline heart
rate and continued for the duration of the study (94±2 bpm). After
stabilization of the hemodynamics,
hemodynamic variables, pressure-volume loops, and
CSF were measured and blood gas samples were collected from the
coronary sinus and coronary arteries. ESPVR was
obtained during inferior vena cava occlusion.
Dobutamine Study
After control measurements were made, dobutamine
diluted in saline was infused via a systemic vein and titrated to
achieve a stable increase in peak +dP/dt. Dobutamine
infusion was begun at a rate of 4 µg ·
kg-1 · min-1 for
10 minutes, and if peak +dP/dt did not increase by at least 20%, the
infusion rate was increased to 5 or 6 µg ·
kg-1 · min-1 at
5-minute intervals. After steady hemodynamic and
contractile states were achieved, we made measurements similar to those
in the control study.
L-NMMA (Control) Study
Dobutamine infusion was discontinued and
hemodynamic variables were monitored for at least
20 minutes until they returned to control values. Then, the
intracoronary infusion of
NG-monomethyl-L-arginine
(L-NMMA), an NOS inhibitor, was started through a Judkins
catheter advanced to the left main coronary artery. The
infusion rate was 20 µmol/min for 10 minutes. We repeated the
same measurements.
Dobutamine/L-NMMA Study
L-NMMA infusion was continued (20 µmol/min for 10
minutes), and dobutamine was infused again for 10 minutes
at the same rate as in the first dobutamine infusion. Then,
we performed the same measurements.
Statistical Analysis
Results are presented as mean±SEM values, unless
otherwise indicated. We obtained ESPVR values through linear regression
analysis. The effects of dobutamine, L-NMMA, and
the combination were analyzed independently with a paired
t test with Bonferroni correction. The effect of the
infusion of L-NMMA on the peak +dP/dt and Emax in
response to dobutamine was analyzed with 2-way
repeated measures ANOVA. Differences were considered significant at
P<0.05.
| Results |
|---|
|
|
|---|
O2. Before L-NMMA infusion,
dobutamine increased LV systolic pressure, peak
+dP/dt, and cardiac index (by 20±4%, 48±8%, and 41±8%,
respectively), with a small decrease in systemic vascular resistance
(SVR) (-15±2%). These changes were accompanied by an increase in CSF
and M
O2 (33±7% and 26±7%,
respectively). The mean dose of dobutamine was 4.8±0.3
µg · kg-1 ·
min-1, and in 2 of 15 patients, the increase in
peak +dP/dt did not reach 20% despite a dobutamine
infusion of 6 µg · kg-1 ·
min-1. After cessation of the
dobutamine infusion, L-NMMA alone did not change these
hemodynamic variables and
M
O2 compared with the control
values. During an intracoronary infusion of L-NMMA, the same
dose of dobutamine increased LV systolic pressure,
peak +dP/dt, and cardiac index (by 26±3%, 72±12%, and 38±9%,
respectively), with a minimal decrease in SVR (-5±1%). The response
of peak +dP/dt to dobutamine was enhanced via L-NMMA
(Figure 2A
O2 (an increase of 19±4%
compared with the dobutamine study).
|
|
Effects of Dobutamine and L-NMMA on
Mechanoenergetic Variables
We successfully assessed the LV ESPVR in 10 of 15 patients. In 5
patients, we were not able to evaluate Emax
because of frequent premature ventricular contractions
during the impeding venous return. Table 3
summarizes the effects of
dobutamine and L-NMMA on mechanoenergetic variables.
Before the intracoronary infusion of L-NMMA,
dobutamine increased Emax by 77±17%
without altering Ea, resulting in an improvement
in ventriculoarterial coupling. Dobutamine
increased EW (39±5%) and M
O2
(21±5%), leading to an increase in the EW/PVA ratio (13±2%) and
EW/M
O2 ratio (4.2±1.4%).
During the intracoronary infusion of L-NMMA without
dobutamine, the mechanoenergetic variables remained
unchanged. Then, the intravenous infusion of
dobutamine during the intracoronary infusion of
L-NMMA caused an increase in Emax (140±21%
versus L-NMMA [control]) that was larger than that after
dobutamine alone. Figure 3
shows a family of representative pressure-volume loops
during inferior vena cava occlusion in response to
dobutamine (left) and dobutamine/L-NMMA
(right). This additional increase in Emax in
response to the same dose of dobutamine as in the
dobutamine study occurred via L-NMMA (Figure 4
). Dobutamine did not alter
Ea and improved ventriculoarterial
coupling to a similar level as that before L-NMMA infusion. During the
intracoronary infusion of L-NMMA, dobutamine
increased EW (64±9%) and M
O2
(35±5%), resulting in an increase in the EW/PVA ratio (18±2%) and
the EW/M
O2 ratio (5.2±2.3%).
The EW/PVA and EW/M
O2 ratios,
however, did not differ between the dobutamine study and
the dobutamine/L-NMMA study.
|
|
|
Table 4
summarizes the changes in
diastolic properties. Dobutamine decreased peak
-dP/dt and the time constant of LV pressure decay (tau) (by 28±8%
and 17±3%, respectively) but did not significantly change chamber
stiffness (ß) and the slope of EDPVR (S). After the cessation of
dobutamine infusion, L-NMMA alone did not change these
variables compared with the control value. During the
intracoronary infusion of L-NMMA, dobutamine
decreased peak -dP/dt and tau (by 26±7% and 15±6%, respectively)
and did not significantly change ß and S. These variables did not
different between the dobutamine study and the
dobutamine/L-NMMA study.
|
| Discussion |
|---|
|
|
|---|
O2, resulting in a slight
improvement in mechanical efficiency; (2) that the selective inhibition
of NOS in myocardium did not change LV mechanoenergetics
under baseline conditions; and (3) that the combination of NOS
inhibition with dobutamine enhanced
Emax, EW, and
M
O2 compared with the infusion
of dobutamine alone. This, however, did not provide a
further improvement in mechanical efficiency because of the
proportional increase in M
O2
with contractile augmentation.
NO is synthesized from L-arginine by 3 NOS isoforms: the
constitutive types, brain NOS and eNOS, and the inducible type, iNOS.
Recent studies have shown that iNOS activity is increased in cardiac
myocytes and endocardial endothelium of patients with
dilated cardiomyopathy15 or in
ventricular myocardium of patients with heart
failure regardless of the cause.16 One of the potential
mechanisms for increased iNOS activity in failing
myocardium is provided by the observation that plasma
levels of tumor necrosis factor-
and interleukin-6 and the
expression of tumor necrosis factor-
in the myocardium
are increased in patients with heart failure.4 17 Winlaw
et al17 showed a significantly increased systemic
concentration of plasma nitrate in patients with heart failure and
reported that this concentration correlated with the severity of heart
failure. Therefore, it seems rational to suppose that the concentration
of NO released by iNOS in human heart failure is increased.
In the present study, the acute administration of L-NMMA did not
elicit a positive inotropic effect and did not change
M
O2 in patients with IDC under
basal conditions. The LV contractile response to
dobutamine, however, was enhanced during the
intracoronary infusion of L-NMMA. There are several mechanisms
by which NO inhibited the positive inotropic response to ß-adrenergic
stimulation. NO activates soluble guanylyl cyclase to produce
cGMP, which inhibits cAMP-stimulated slow inward
Ca2+ channels through the activation of protein
kinase G and activates cGMP-dependent
phosphodiesterase.18 NO also has direct
S-nitrosation of cellular proteins and formation of
peroxynitrite.19 Hare et al5 implicated
iNOS in part in the mediation of the myocardial contractile dysfunction
in patients with dilated cardiomyopathy. Drexler et
al20 recently found that the isoproterenol-induced
increase in the force of contraction was inversely related to cardiac
iNOS activity in LV tissue from the failing human heart.
One of the mechanisms by which NO mediates negative inotropic effect
involves a decrease in the calcium sensitivity of contractile element
via cGMP-dependent protein kinase.21 Therefore, we
expected that the reversal of NO-induced hyporesponsiveness to
ß-adrenergic stimulation via L-NMMA might not be accompanied by a
significant increase in M
O2.
In the present findings, however, the enhanced response to
dobutamine with L-NMMA was associated with a parallel
increase in M
O2 to contractile
augmentation and did not change mechanical efficiency.
We must consider LV afterload, as well as its contractile state, as a
determinant of M
O2. In the
present study, the intravenous infusion of
dobutamine during L-NMMA infusion caused small increases in
SVR and Ea compared with dobutamine
study that were not statistically significant (Tables 2
and 3
). A minimal response of the cardiac index to L-NMMA plus
dobutamine dissimilar to the peak +dP/dt response may in
part be attributed to the increase in systemic afterload for the
failing heart. From the framework of PVA and
M
O2, however, an
afterload-dependent increase in
M
O2 should be accompanied by
an increase in PVA, whereas PVA values were not different for the
dobutamine study and the dobutamine/L-NMMA
study. Thus, given that the slope of the
M
O2-PVA relation is not
affected by cardiac NOS inhibition as was reported in previous
studies,22 it seems reasonable to say that an increase in
M
O2 with the combination of
dobutamine with L-NMMA was due to an increase in
PVA-independent M
O2 in
response to the contractile augmentation rather than to changes in
systemic afterload.
There have been no data that document the possible involvement of NO in
the direct control of PVA-independent
M
O2, which relates to basal
metabolism and excitation-contraction coupling, such as the
ATP consumed for Ca2+ cycling and reuptake by the
sarcoplasmic reticulum. Recent investigators have demonstrated in vitro
a modulatory action of NO on calcium channels of the sarcoplasmic
reticulum,18 and it is possible that NO would imply less
energy expenditure at excitation-contraction coupling, in particular at
higher levels of contractility, although we could not
document this. Further studies were warranted to examine the effect of
cardiac NOS inhibition on PVA-independent
M
O2.
We found that mechanical efficiency and overall energy conversion efficiency did not change with cardiac NOS inhibition during ß-adrenergic stimulation without an alteration in LV loading conditions and ventriculoarterial coupling. In consideration of these findings in view of cardiac mechanics and energetics, the oxygen-saving effect of NO attenuation of the contractile response to ß-adrenergic stimulation does not appear to be deleterious for the failing human heart due to IDC.
In isolated ejecting guinea pig hearts, reduced endogenous NO via L-NMMA has been reported to decrease LV diastolic compliance.23 Paulus et al24 showed that the intracoronary infusion of sodium nitroprusside, a spontaneous NO donor, into human subjects caused a modest decline in LV systolic performance and increased end-diastolic distensibility. Under pathological conditions, such as IDC, we did not observe alteration of the diastolic properties, including LV relaxation and chamber stiffness, with the intracoronary infusion of NOS inhibitor under baseline conditions and in response to dobutamine.
Elucidation of the cardiac effect of NO may be complicated by the systemic effect of the NOS inhibitor.25 By using the direct intracoronary infusion of L-NMMA and relatively load-independent indices such as Emax, we tried to avoid changes in loading conditions that might confound the interpretation of the data. The intracoronary infusion of L-NMMA at a rate of 20 µmol/min did not significantly change SVR or Ea, but our preliminary study revealed that a 40-µmol/min infusion of L-NMMA significantly increased SVR and decreased cardiac index during 4 µg · kg-1 · min-1 infusion of dobutamine (data not shown).
We must consider potential error in the assumption of the linear ESPVR.
Kass et al26 suggested that
contractility-dependent curvilinearity of the ESPVR may
exist. In the present study, possible curvilinearity of the ESPVR
may exist because of the relatively narrow range of altered loading
conditions. The estimation of PVA with linear regression
analysis of ESPVR, however, has been reported to be a reliable
predictor of M
O2 under
different contractile states in human hearts.11
In summary, NO-dependent hyporesponsiveness to ß-adrenergic stimulation does not aggravate LV pump performance and does maintain energy transduction efficiency. In view of the pathophysiology of heart failure characterized as altered oxygen metabolism, the present data support the known therapeutic efficacy of organic nitrates and ACE inhibitors that potentially increase cardiac NO in the treatment of this syndrome.7 27
Received July 9, 1999; revision received October 27, 1999; accepted November 15, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. J. Paulus and J. G. F. Bronzwaer Nitric oxide's role in the heart: control of beating or breathing? Am J Physiol Heart Circ Physiol, July 1, 2004; 287(1): H8 - H13. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Mak and G. E. Newton Redox modulation of the inotropic response to dobutamine is impaired in patients with heart failure Am J Physiol Heart Circ Physiol, February 1, 2004; 286(2): H789 - H795. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Feldman The emerging role of pharmacogenomics in the treatment of patients with heart failure Ann. Thorac. Surg., December 1, 2003; 76(6): S2246 - 2253. [Full Text] [PDF] |
||||
![]() |
P.B. Massion, O. Feron, C. Dessy, and J.-L. Balligand Nitric Oxide and Cardiac Function: Ten Years After, and Continuing Circ. Res., September 5, 2003; 93(5): 388 - 398. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Z. Kojic, U. Flogel, J. Schrader, and U. K. M. Decking Endothelial NO formation does not control myocardial O2 consumption in mouse heart Am J Physiol Heart Circ Physiol, June 5, 2003; 285(1): H392 - H397. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Kumar, K. Nguyen, S. Waxman, B. D. Nearing, G. A. Wellenius, S. X. Zhao, and R. L. Verrier Potent antifibrillatory effects of intrapericardial nitroglycerin in the ischemic porcine heart J. Am. Coll. Cardiol., May 21, 2003; 41(10): 1831 - 1837. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Paolocci, T. Katori, H. C. Champion, M. E. St. John, K. M. Miranda, J. M. Fukuto, D. A. Wink, and D. A. Kass From the Cover: Positive inotropic and lusitropic effects of HNO/NO- in failing hearts: Independence from beta -adrenergic signaling PNAS, April 29, 2003; 100(9): 5537 - 5542. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. McNamara, R. Holubkov, L. Postava, R. Ramani, K. Janosko, M. Mathier, G. A. MacGowan, S. Murali, A. M. Feldman, and B. London Effect of the Asp298 Variant of Endothelial Nitric Oxide Synthase on Survival for Patients With Congestive Heart Failure Circulation, April 1, 2003; 107(12): 1598 - 1602. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Brutsaert Cardiac Endothelial-Myocardial Signaling: Its Role in Cardiac Growth, Contractile Performance, and Rhythmicity Physiol Rev, January 1, 2003; 83(1): 59 - 115. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Chen, J. H. Traverse, R. Du, M. Hou, and R. J. Bache Nitric Oxide Modulates Myocardial Oxygen Consumption in the Failing Heart Circulation, July 9, 2002; 106(2): 273 - 279. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Traverse, Y. Chen, M. Hou, and R. J. Bache Inhibition of NO production increases myocardial blood flow and oxygen consumption in congestive heart failure Am J Physiol Heart Circ Physiol, June 1, 2002; 282(6): H2278 - H2283. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. A. Nikolaidis, T. Hentosz, A. Doverspike, R. Huerbin, C. Stolarski, Y.-T. Shen, and R. P. Shannon Mechanisms whereby rapid RV pacing causes LV dysfunction: perfusion-contraction matching and NO Am J Physiol Heart Circ Physiol, December 1, 2001; 281(6): H2270 - H2281. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Post, R. Schulz, P. Gres, and G. Heusch No involvement of nitric oxide in the limitation of beta -adrenergic inotropic responsiveness during ischemia Am J Physiol Heart Circ Physiol, December 1, 2001; 281(6): H2392 - H2397. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Cotton, M. T. Kearney, P. A. MacCarthy, R. M. Grocott-Mason, D. R. McClean, C. Heymes, P. J. Richardson, and A. M. Shah Effects of Nitric Oxide Synthase Inhibition on Basal Function and the Force-Frequency Relationship in the Normal and Failing Human Heart In Vivo Circulation, November 6, 2001; 104(19): 2318 - 2323. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. S. Wittstein, D. A. Kass, P. H. Pak, W. L. Maughan, B. Fetics, and J. M. Hare Cardiac nitric oxide production due to angiotensin-converting enzyme inhibition decreases beta-adrenergic myocardial contractility in patients with dilated cardiomyopathy J. Am. Coll. Cardiol., August 1, 2001; 38(2): 429 - 435. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||