Download the Frank-Starling Mechanism

959
The Failing Human Heart Is Unable to Use
the Frank-Starling Mechanism
Robert H.G. Schwinger, Michael Bohm, Andrea Koch, Ulrich Schmidt, Ingo Morano,
Hans-Joachim Eissner, Peter Uberfuhr, Bruno Reichart, Erland Erdmann
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Abstract There is evidence that the failing human left
ventricle in vivo subjected to additional preload is unable to
use the Frank-Starling mechanism. The present study compared the force-tension relation in human nonfailing and
terminally failing (heart transplants required because of dilated cardiomyopathy) myocardium. Isometric force of contraction of electrically driven left ventricular papillary muscle
strips was studied under various preload conditions (2 to 20
mN). To investigate the influence of inotropic stimulation, the
force-tension relation was studied in the presence of the
cardiac glycoside ouabain. In skinned-fiber preparations of the
left ventricle, developed tension was measured after stretching
the preparations to 150% of the resting length. To evaluate the
length-dependent activation of cardiac myofibrils by Ca'+ in
failing and nonfailing myocardium, the tension-Ca2' relations
were also measured. After an increase of preload, the force of
contraction gradually increased in nonfailing myocardium but
was unchanged in failing myocardium. There were no differences in resting tension, muscle length, or cross-sectional area
of the muscles between both groups. Pretreatment with oua-
bain (0.02 ,umol/L) restored the force-tension relation in
failing myocardium and preserved the force-tension relation in
nonfailing tissue. In skinned-fiber preparations of the same
hearts, developed tension increased significantly after stretching only in preparations from nonfailing but not from failing
myocardium. The Ca2 sensitivity of skinned fibers was significantly higher in failing myocardium (EC50, 1.0; 95% confidence limit, 0.88 to 1.21 ,umol/L) compared with nonfailing
myocardium (EC50, 1.7; 95% confidence limit, 1.55 to 1.86
gmol/L). After increasing the fiber length by stretching, a
significant increase in the sensitivity of the myofibrils to Ca2
was observed in nonfailing but not in failing myocardium.
These experiments provide evidence for an impaired forcetension relation in failing human myocardium. On the subcellular level, this phenomenon might be explained by a failure of
the myofibrils to increase the Ca 2+ sensitivity after an increase
T o improve myocardial contractility, several compensatory mechanisms can be activated; one of
these is the Frank-Starling mechanism, in
which an increased preload enhances contractile force.
In the nonfailing heart, an increase of the diastolic
volume of the heart is associated with an improvement
in cardiac contractility.12 In numerous studies using
isolated cardiac muscle preparations of animals, the
developed tension was found to increase in parallel with
an increase in muscle length.3 However, there is evidence that the failing left ventricle in vivo subjected to
additional load is unable to use the Frank-Starling
mechanism to improve myocardial contractility.45 The
present study was aimed at investigating whether the
tension-related increase in force of contraction in failing
human myocardium is preserved and, if not, which
mechanisms could be involved.
In vivo, adrenergic reflex mechanisms may mask
changes in contractility in response to different preload
conditions. Therefore, the length-tension relation in
nonfailing myocardium and in human myocardium ter-
minally failing because of dilated cardiomyopathy was
examined in vitro on electrically driven papillary muscle
strips and on skinned-fiber preparations from the same
hearts. The relation between muscle length and developed tension in cardiac muscle represents the basis of
Starling's law of the heart.3 The dependence of developed tension on muscle length arises from several
causes, the most important of which is supposed to be a
length-dependent activation of cardiac myofibrils by
Ca'+.3,6-8This effect is reported to be due to a change in
the Ca'+ sensitivity of the myofibrils, which may be the
result of an alteration in the Ca'+ affinity of tropinin
C.3,7,9 However, there may also be a substantial contribution from a change in the amount of Ca 2+ supplied to
the myofibrils during each contraction.3'10-'2 To study
whether the length-dependent changes in contractility
may result from changes in the Ca2 sensitivity of the
myofibrils and/or from changes in the amount of Ca2
supplied to the contractile apparatus, we performed
experiments with skinned-fiber preparations at a different fiber length and experiments with papillary muscle
strips subjected to increasing diastolic loads after inotropic stimulation with the cardiac glycoside ouabain
and under control conditions. Ouabain blocks the membrane-bound NaX,K+-ATPase, leading to an increase in
the intracellular Na+ that activates the Na+-Ca2+ exchange mechanism to increase intracellular Ca'+.""l
Consequently, the Ca 2+ content of the sarcoplasmatic
reticulum is increased, leading to an enhanced Ca2+
release during depolarization and, thus, to a change in
Ca2+ supplied to the myofibrils during contraction.
Received April 13, 1993; accepted February 10, 1994.
From the Universitat zu Koin (Germany), Medizinische Klinik
III (R.H.G.S., M.B., A.K., U.S., E.E.); the Universitat Munchen
(Germany), Herzchirurgische Klinik (P.U., B.R.) and Institut fur
Medizinische Informationsverarbeitung, Biometrie, und Epidemiologie (H.-J.E.); and the Universitat Heidelberg (Germany), Physiologisches Institut II (I.M.).
Reprint requests to Dr Robert H.G. Schwinger, Universitat zu
Koln, Medizinische Klinik III, Joseph-Stelzmannstr. 9, 50924
Koln, Germany.
of the sarcomere length. (Circ Res. 1994;74:959-969.)
Key Words * human myocardium * heart failure Frank-Starling mechanism * cardiac glycosides
960
Circulation Research Vol 74, No 5 May 1994
TABLE 1. Clinical and Hemodynamic Data for the Patients Studied
Age, y
Sex, M/F
LVEDP, mm Hg
LVEDV, mL
EF, %
Cl, L* m-2* min-
NYHA IV (n=27)
Mean+SEM
57.5+1.8
23/4
25.6+2.7
298.9±38.0
25.5+1.4
2.5+0.1
...
Range
30-64
1.2-3.4
10-45
100-520
15-31
NP
...
3/2
NP
NP
NF(n=6)
NP
LVEDP indicates left ventricular end-diastolic pressure; LVEDV, left ventricular end-diastolic volume; EF, ejection fraction; Cl, cardiac
index; NYHA IV, terminal heart failure (according to New York Heart Association classification); NF, nonfailing myocardium; and NP, no
hemodynamic measurement performed.
Materials and Methods
Myocardial Tissue
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Experiments were performed on isolated, electrically stimulated papillary muscle strips from human left ventricular
myocardium. Dilated cardiomyopathic tissue was obtained
during cardiac transplantation (n=27; 4 females and 23 males;
age, 57.5± 1.8 years; range, 30 to 64 years) (Table 1). Patients
suffered from heart failure clinically evaluated as New York
Heart Association (NYHA) class IV on the basis of clinical
symptoms and signs as judged by the attending cardiologist
shortly before the operation. All patients gave written informed consent before surgery. Medical therapy consisted of
diuretics, nitrates, angiotensin-converting enzyme inhibitors,
and cardiac glycosides. Patients receiving catecholamines or
,8-adrenergic receptor or Ca 2+ antagonists were withdrawn
from the study. Drugs used for general anesthesia were
flunitrazepam and pancuronium bromide with isoflurane. Cardiac surgery was performed with patients subjected to cardiopulmonary bypass with cardioplegic arrest during hypothermia. Nonfailing human myocardium was obtained from six
donors (Table 1) who were brain dead as a result of traumatic
injury. To identify these as control hearts, the clinical situation
before death was considered. These data indicated to the
attending cardiologist and to the cardiac surgery team explanting the heart that the heart was useful as a donor. There was
no evidence of left ventricular dysfunction by echocardiography. These hearts could not be transplanted for technical
reasons. The cardioplegic solution (a modified Bretschneider
solution) contained (mmol/L) NaCl 15, KCl 9, MgC12 4,
histidine 180, tryptophan 2, mannitol 30, and potassium dihydrogen oxoglutarate 1. The papillary muscle strips were taken
from the failing and nonfailing hearts immediately after explantation (maximum time, 10 minutes).
Contraction Experiments
Immediately after excision, the papillary muscles were placed
in ice-cold preaerated modified Tyrode's solution (for composition, see below) and delivered to the laboratory within 10
minutes. The experiments were performed on isolated, electrically driven (1-Hz) muscle preparations. Muscle strips of uniform
size with muscle fibers running approximately parallel to the
length of the strips (diameter, 0.4 to 0.7 mm; length, 5 to 9 mm)
were dissected in an aerated bathing solution (for composition,
see below) under microscopic control with sharp scissors at room
temperature. Connective tissue, if visibly present, had to be
carefully trimmed away. For control and myopathic muscles,
mean lengths were 7.6±0.3 (n=25) and 8.0±0.3 (n=49) mm, and
mean weights were 5.3±0.13 and 5.2±0.12 mg, respectively.
After the experiment, the length and the diameter of the
cylindrical preparations were determined. The papillary muscle
strips were than blotted dry (10 g for 1 minute) and weighed.
Cross-sectional area was estimated assuming the geometry of a
cylinder with a specific gravity of 1.0 and was determined from
papillary muscle diameter (D) by the following formula: crosssectional area=3.14159x (D/2)2. The corresponding values for
papillary muscle strips from failing and nonfailing hearts were
0.60±0.07 and 0.56±0.06 mm2, respectively. Cross-sectional area
and weight did not differ in papillary muscle strips from nonfailing and failing myocardium in each experimental condition
studied. Force of contraction has been given in millinewtons as
well as in millinewtons per square millimeter. The preparations
were attached to a bipolar platinum stimulating electrode and
suspended individually in 75-mL glass tissue chambers for recording of isometric contractions. The bathing solution used was
a modified Tyrode's solution containing (mmol/L) NaCi 119.8,
KCl 5.4, CaCl2 1.8, MgCl2 1.05, NaH2PO4 0.42, NaHCO3 22.6,
Na2-EDTA 0.05, ascorbic acid 0.28, and glucose 5.0. It was
continuously gassed with 95% 02/5% CO2 and maintained at
37°C; its pH was 7.4. Isometric force of contraction was measured
with an inductive force transducer (W. Fleck) attached to a
Hellige Helco Scriptor or Gould recorder. The rate of tension
change was determined by differentiation of the force signal.
Resting tension was constant throughout the experiments. The
preparations were electrically paced at 1 Hz with rectangular
pulses of 5-millisecond duration (Grass stimulator SD 9); the
voltage was 20% above threshold. All preparations were allowed
to equilibrate at least 90 minutes in a drug-free bathing solution
(1 Hz) until complete mechanical stabilization. After 45 minutes,
the solution was changed. After complete mechanical stabilization, the force-frequency relation was studied, starting with a rate
of 0.5 Hz. The force-tension relation was investigated, increasing
preload from 2 to 20 mN (1 Hz). Inotropic intervention was
performed at 1 Hz with a 10-mN preload, and afterward, the
force-tension relation was studied in the same way as under basal
conditions. To test mechanical performance, after each experiment the positive inotropic effect mediated by the elevation of
[Ca2`]0 (15 mmol/L) was measured. There was no difference
between groups. Control strips studied in Tyrode's solution with
a composition identical to that of the original experiments
revealed maximally a 20% reduction of baseline isometric tension
over the period necessary to complete testing. At the end of each
experiment, a test for adequate oxygenation15 was performed by
changing the carbogen to 80% 02/5% C02/15% N2 at the
stimulation rate, at which the force of contraction was maximal.
This procedure lowered oxygen tension significantly in the bathing solution, and all muscle preparations in which the force of
contraction declined more than 10% during 30-minute exposure
were discarded from the evaluation.
Skinned Fibers
In all instances, muscle pieces =10 mm in length and 1 mm
in diameter were excised while being incubated in ice-cold
Bretschneider solution from the left ventricular papillary
muscle. Samples were taken from the same location of each
heart to prevent an influence of different origins of the tissue
on the results. The muscle pieces were incubated in a solution
containing 50% glycerol and (mmol/L) imidazole 20, NaN3 10,
ATP 5, MgC12 5, EGTA 4, and dithioerythritol (DTE) 2 (pH
7.0 at 4°C for 1 hour) and then were divided with tweezers into
smaller fiber bundles. For skinning, the fiber bundles were put
in the same solution also containing 1% Triton X-100 (Merck)
for 24 hours. Afterward, the fibers were stored at -20°C in the
first solution without detergent.
The chemically skinned fiber bundles were prepared under
the microscope and then mounted isometrically and connected
Schwinger et al Heart Failure and the Frank-Starling Mechanism
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
to a force transducer (AME 801, SensoNor). In relaxation
solution, the fiber length was adjusted to an extent where
resting tension was just threshold (slack position). Fiber
diameter and length were the same in all preparations studied
(125 to 175 gtm and 7 to 8 mm, respectively). The mean
sarcomere length measured by laser diffraction was 2.0+0.02
gm of the skinned fibers in relaxation solution before starting
the experiment with the first concentration. The relaxation
solution contained (mmol/L) imidazole buffer 40, ATP 10,
MgC12 12.5, creatine phosphate 10, NaN3 5, EGTA 5, and DTE
1 (Sigma), along with 350 U/mL (pH 7.0) creatine kinase
(Boehringer-Mannheim). The contraction solution had the
same composition as the relaxation solution except that CaCl2
was added. The desired [Ca'+] was obtained by mixing the
relaxation and contraction solution in the appropriate proportions. The contraction solution had a [Ca2+] of 12.9 ,umol/L.
All experiments were performed at 21°C. A "precontraction"
was performed at maximum [Ca'+] to take into consideration
a phenomenon observed in experiments with cardiac fibers. It
takes a long time at the first contraction at Ca's of 12.9 gmol/L
until maximum force is reached, and in some cases it is
somewhat smaller than the force obtained in the following
measurements. After the precontraction at 12 gmol/L Ca2',
the skinned fiber preparations produced a reproducible Ca'`activated tension. The actual [Ca2+] values were calculated by
a computer program according to that reported by Fabiato and
Fabiato.6 From each heart, several skinned-fiber preparations
of ventricle were investigated. Then, a mean value of Ca'+
sensitivity of each heart was calculated by using the Hill
equation. Except when otherwise indicated, results are expressed as mean+SEM.
Materials
Ouabain was obtained from Boehringer-Mannheim. All
other chemicals were of analytical grade or the best grade
commercially available (Sigma). For studies with isolated
cardiac preparations, stock solutions were prepared and applied to the organ bath. All compounds were dissolved in
twice-distilled water. Applied agents did not change the pH of
the medium.
Statistics
The data shown are mean±SEM.
a: Papillary Muscle Strip Preparations
The length-tension relationship was compared between
following groups: (1) failing and nonfailing human myocardium under basal conditions and (2) presence and absence of
pretreatment with ouabain.
b: Skinned-Fiber Preparations
Comparisons between failing and nonfailing myocardium
were performed regarding (1) the Ca'+-induced tension generation and (2) the length-tension relation.
To decide whether the data differed between groups a
repeated-measures ANOVA was calculated for each comparison. The developed tension or the change of the developed
tension was used as an outcome measure determined at
different conditions, eg, preload and Ca'2-activated force
generation. The log-transformed data were compatible with a
normal distribution (Shapiro-Wilk test) and therefore used in
the statistical analysis. The interaction of the grouping and the
repeated-measures factor (ie, the conditions tested) were
tested by Hotelling's T-squared statistic. In the case of significance (P<.05), it has been concluded that the shape of the
considered relation differs between the compared groups, and
two sample t tests at the different levels of the repeatedmeasures factor were done to locate the differences. All tests
were two tailed. The programs of the SAS Institute were used
for analysis.16
Z 4
E
O
961
E Nonfailing
3.5
A
NYHA IV
a)
-2
G)
O
2.5
1 2
1.5
1
Ih
0
-14--- 2
4
6
8
10
'I
-Ie
15
20
Preload (mN)
FIG 1. Graph showing developed tension (ordinate in millinewtons) plotted as a function of preload (abscissa in millinewtons,
logarithmic scale) in human papillary muscle strips from nonfailing and terminally failing myocardium (dilated cardiomyopathy).
Basal force of contraction is given in Table 2. NYHA indicates
New York Heart Association classification.
Results
Papillary Muscle Strip Preparations
To functionally characterize the human tissue used,
we examined the force-frequency relation17 of nonfailing and terminally failing human myocardium.18 After
an increase in stimulation frequency, the force of contraction increased in nonfailing hearts but not in human
myocardium failing because of dilated cardiomyopathy
(not shown). The myocardium used showed the forcefrequency behavior as that typically described in previous studies involving nonfailing and terminally failing
(dilated cardiomyopathic) human myocardium.18-20
To examine the length-tension relation in nonfailing
and failing (dilated cardiomyopathic) human myocardium, the force development of electrically driven papillary muscle strip preparations after variations of preload was measured. With this experimental approach,
additionally influencing factors like frequency can be
excluded. Preload at a stimulation frequency of 1 Hz
(1.8 mmol/L Ca2, 37°C) was increased stepwise from 2
to 20 mN. Fig 1 shows the force of contraction in
relation to the preload applied. The statistical analysis
reveals a significantly different shape between graphs
(P<.05) (Fig 1). In the nonfailing myocardium, developed tension increased in parallel with enhanced preload. In contrast, in papillary muscle strips from terminally failing human myocardium, developed tension did
not change significantly. This holds true for data in
millinewtons as well as in millinewtons per square
millimeter. In addition, the change in force of contraction after an increase in preload (6 mN) was significantly different between groups (P<.05). Table 2 gives
developed tension relative to various preload conditions
for the two groups.
Under basal conditions, ie, preload of 2 mN (1 Hz,
37°C, and 1.8 mmol/L Ca2), parameters of isometric
force of contraction-developed tension, peak rate of
tension rise, peak rate of tension decay, time to peak
tension, and time to half-relaxation -were not different
between nonfailing myocardium and failing (dilated
cardiomyopathic) myocardium (Table 2). After an increase in preload, the peak rate of tension rise changed
Circulation Research Vol 74, No 5 May 1994
962
TABLE 2. Influence of Preload on Contractile Properties
of Isolated Papillary Muscle Strips From Terminally
Failing Versus Nonfailing Human Myocardium
Preload, mN
Nonfailing Hearts
NYHA IV DCM
1.8±0.2
4.1±0.3
12.1± 1.5
1.6±0.1
2
DT, mN
DT, mN/mm2
+T, mN/s
-T, mN/s
RT, mN
3.6+0.2
11.3-+1.0
8.4+0.6
2.4+0.1
8.5+0.8
2.2±0.2
6
2.2±--0.3 *
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
DT, mN
DT, mN/mm2
+T, mN/s
-T, mN/s
RT, mN
1 .7±0.2
3.9+0.3
12.5--1.1
9.1 ±0.7
4.7±0.1
5.2+0.4*
16.0+z1.7
10.9+0.9
4.6+0.1
10
DT, mN
1.7±0.2
2.9+0.4*
6.6+0.5*
DT, mN/mm2
3.9+0.3
+T, mN/s
19.9+2.4
12.9±1.2
-T, mN/s
15.4+1.8
9.8+0.8
RT, mN
7.4+0.2
7.6±0.1
20
DT, mN
3.7±0.5*
1.7±0.2
4.0±0.4
DT, mN/mm2
8.5±0.6*
12.4+1.2
+T, mN/s
23.4±3.6
-T, mN/s
16.4+2.8
9.8±0.8
12.1±0.4
RT, mN
13.8±0.3
NYHA IV DCM indicates dilated cardiomyopathy (New York
Heart Association class IV); DT, developed force of contraction;
+T, maximal rate of tension rise; -T, maximal rate of tension
decrease; and RT, resting tension. Values are mean±SEM.
*P<.05 vs preload of 2 mN.
z
E
0
.C
0
0
0
0
z
E
25
L]
A
20 k
Nonfailing
cu
0
19
NYHA IV
,o
0c:
0
CO
a
Nonfailing
A
NYHA IV
15
13
15 F
0
0
0
10
-0
11 _
9
7
0
0
F
17
-90
CO
0
CL
from 12.1±+1.5 to 23.4+3.6 mN/s in the control myocardial tissue (Fig 2, left). There was no change of peak
rate of tension rise in failing hearts (11.3±0.9 versus
12.4±1.2 mN/s). The values for peak rate of tension
decay increased only in nonfailing myocardium but were
unchanged in failing human myocardium (Fig 2, right).
In nonfailing myocardium, both peak rate of tension rise
and peak rate of tension decay increased when preload
was enhanced from 2 to 20 mN (P<.05), whereas in
failing myocardium, there was no significant change
either for peak rate of tension rise or peak rate of
tension decay. The preload-peak rate of tension rise/
decay curves from left ventricular papillary muscle
strips of failing and nonfailing myocardium are shown in
Fig 2. To exclude the possibility that these differences
are not due to a different content of contractile and
elastic components, we measured resting tension at
each condition tested. The resting tension was not
different between either nonfailing or failing human
tissue (Table 2).
To study whether the length-dependent changes in
contractility may result from changes in the amount of
Ca2' supplied to the myofibrils, we performed experiments after inotropic stimulation. The effects of the
cardiac glycoside ouabain at 0.02 ,umol/L on the forcetension relation of human myocardial tissue were studied. Ouabain was added to the bathing solution before
examination of the force-tension relation. In previous
studies, it has been shown that 0.02 ,mol/L ouabain
exerts only marginal inotropic effects in electrically
driven papillary muscle strip preparations of human
myocardium. After the addition of ouabain, the force of
contraction increased only marginally. After pretreatment with ouabain, the increase in preload was accompanied by an increase in developed tension assessed as
NYHA class IV (Table 3). Therefore, ouabain was
effective in restoring the force-tension relation of failing
human myocardium (Fig 3). The statistical analysis
reveals a significant difference in the preload-tension
relation with and without pretreatment with ouabain
(P<.05). This holds true for data in millinewtons as well
as in millinewtons per square millimeter. Maximal force
development was significantly different between both
groups at the 15- and 20-mN preload condition
,j1
2
4
6
8
Preload (mN)
10
15
20
y
0
2
4
6
8
10
15
20
Preload (mN)
FIG 2. Graphs showing peak rate of tension rise (left, ordinate in millinewtons per second) and peak rate of tension decay (right,
ordinate in millinewtons per second) plotted as a function of preload (abscissa in millinewtons, logarithmic scale) in human papillary
muscle strips from nonfailing and terminally failing myocardium (dilated cardiomyopathy). Basal values are given in Table 2. NYHA
indicates New York Heart Association classffication.
Schwinger et al Heart Failure and the Frank-Starling Mechanism
TABLE 3. Influence of Pharmacologic Intervention on
Developed Force of Contraction of Isolated Papillary
Muscle Strips From Failing Human Myocardium
Ouabain, 0.02 jxmol/L
Control
Preload,
DT,
DT,
mN
DT, mN
mN/mm2
DT, mN
mN/mm2
2
1.6±0.2
3.5+0.5
1.6+0.1
3.6±0.2
1 .7+0.2
6
5.2±1.4
2.3±0.6
3.9±0.3
10
2.6±0.8
5.9±1.6
1.7±0.2
3.9±0.3
3.2±0.7*
20
7.2±1.6*
1.7±0.2
4.0±0.4
DT indicates developed force of contraction; control, forcetension relation without drug added. Values are mean±SEM.
*P<.05 vs control.
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
(P<.05). Treatment with ouabain did not influence the
force-tension relation in nonfailing myocardium. However, there was a trend toward an increased developed
tension in the presence of ouabain in nonfailing tissue.
Skinned-Fiber Preparations
To study the tension generation at subcellular levels
in failing and nonfailing myocardium, we investigated
chemically skinned fiber preparations. Fig 4 shows a
typical recording of an original skinned-fiber experiment from a heart of a patient without heart failure (Fig
4, left) and from a patient with terminal heart failure
(NYHA class IV) (Fig 4, right). Incubation of chemically skinned fibers with increasing concentrations of
free Ca2' caused an increase in isometric tension. A
mean sarcomere length of 2.0+0.02 gum of the skinned
fibers in relaxation solution before starting the experiments was determined. Fiber diameter and length of the
preparations were the same in either group studied. In
addition, there was no evidence of "fall-off' of maximum force at 12.9 ,umol/L; hence, tension-Ca2+ curves
could be repeated without difference in force development obtained at the different [Ca2+] levels. The curves
summarizing the mean values of the Ca2'-tension relation in skinned fibers from patients with terminal heart
failure and in nonfailing control subjects are shown in
963
Fig 5. The threshold concentration of activation was
0.65 gmol/L (pCa 6.19) in ventricular myocardium of
nonfailing and failing hearts. In failing and nonfailing
myocardium, the [Ca2+]-response curve had a sigmoid
shape. At slack length, Ca2' sensitivity as judged from
EC50 values was higher in failing than in nonfailing
tissue (Table 4). The cooperativity indexes given by the
Hill coefficient were similar in failing (2.3+±0.4) and
nonfailing (2.0+±0.4) myocardium (Table 4). Values
between 2 and 3 in mammalian and human cardiac
skinned fibers are in accordance with other reports.21,22
Maximal tension development was 585±+ 135 and
512±+76 ,uN in nonfailing and failing myocardium, respectively. No difference in maximal tension development was observed between fibers isolated from nonfailing and failing human myocardium. However, the
concentration-response curve for Ca2' was significantly
shifted to the left in terminally failing human tissue
when compared with control tissue (P<.01). To gain
insight into the length-dependent activation of cardiac
myofibrils, skinned-fiber preparations were studied after increasing the sarcomere length by stretching. Only
in preparations taken from nonfailing hearts did tension
increase significantly after an increase in sarcomere
length by stretching (P<.05). Fig 6 illustrates the lengthtension relation in both groups. Stretching of the fiber
preparations (120% and 130% of resting length) did not
affect Ca21 sensitivity in skinned-fiber preparations of
myopathic hearts (Table 4). In contrast, in preparations
from nonfailing myocardium an increase of the sarcomere length by stretching the skinned-fiber preparations was accompanied by a significant increase in the
sensitivity to Ca21 (Table 4). The sensitivity to Ca21 in
nonfailing preparations never exceeded the sensitivity
to Ca21 observed in skinned-fiber preparations from
failing hearts. After increasing the sarcomere length by
stretching (constant [Ca 2+]) in skinned-fiber preparations from nonfailing myocardium, a significant increase
in the sensitivity of myofibrils to Ca2' was observed.
This does not hold true for preparations taken from
failing hearts. Fig 7 illustrates the length-dependent
Ca21 sensitivity after stretching from 100% to 150% of
resting length in skinned-fiber preparations from nonfailing (Fig 7, left) and failing (Fig 7, right) myocardium.
Discussion
Z
4
C 3.5
C
3
C 2.5
c] 2
0
'
2
4
6
8
10
15
20
Preload (mN)
FIG 3. Graph shows developed tension (ordinate in millinewtons) plotted as a function of preload (abscissa in millinewtons,
logarithmic scale) in human papillary muscle strips from terminally failing myocardium (dilated cardiomyopathy) in the presence of ouabain (0.02 ,umol/L) and in the control condition. Basal
values are given in Table 3. NYHA indicates New York Heart
Association classification.
In left ventricular muscle strip preparations taken
from nonfailing myocardium, the force of contraction
increased parallel to an increase of preload. Parameters
of isometric contraction were similar to those observed
in previous studies.2324 Additionally, the peak rate of
tension rise and the peak rate of tension decay were
enhanced when preload was increased. These in vitro
findings using papillary muscle strips from nonfailing
hearts are compatible with the improvement in cardiac
performance in vivo after increasing diastolic volume.25
In healthy subjects at low-exercise work loads, when
end-systolic volume changes little, both end-diastolic
and stroke volume increase and contribute to the increase in cardiac output. These observations indicate
that the healthy left ventricle follows Starling's law of
the heart.1'2 Also, in muscle strip preparations from
various healthy animals (cat, dog, and frog) a positive
length (preload)-tension curve was observed.1126 In
contrast, papillary muscle strips from human hearts
964
Circulation Research Vol 74, No 5 May 1994
NYHA IV
Nonfailing
Slack-position
z
z
l 600
.2COC)500
3-
Slack-position
600
c
._
U) 500
a)
,a
D 400
) 400
0.
0
> 300
0
> 300
()
0
200[
200
[
100
100
0
0
12.9
0.01
0.01
0.95
3.4
0.01
0.01
0.65
0.01
12.9
1.7
0.95
3.4
0.01
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Concentration of calcium (umol/l)
Concentration of calcium (umol/l)
FIG 4. Original recordings illustrating developed tension of a chemically skinned fiber (human left ventricular papillary muscle) after
exposure to increasing concentrations of Ca2I. Fibers were prepared from the heart of a patient with severe heart failure (New York Heart
Association [NYHA] class IV) (right) as well as from a patient without heart failure (left).
terminally failing because of dilated cardiomyopathy
failed to generate an increase in force development
when preload was enhanced (the present study). Thus,
Starling's mechanism fails to improve myocardial contractility in failing human heart muscle. However, from
the present study a response in the left ventricle to
changes in end-diastolic volume can be suggested only
indirectly. From clinical studies, there is also evidence
that the chronically dilated left ventricle loaded with
additional volume may be unable to use the FrankStarling mechanism. Diuretic treatment accompanied
by a reduced preload, ie, diastolic fiber length, was not
necessarily accompanied by a decrease in stroke volume.27 In addition, the failing heart has been reported
Ei
oo
NonfbNn
*
0
NYHA
20
20
0.01
0.03
0.1
0.3
1
3
10
30
Concetffion of calium ymo/)
FIG 5. Concentration-response curve for Ca2+ (abscissa, logarithmic scale) in skinned-fiber preparations of nonfailing and
terminally failing human myocardium (left ventricular papillary
muscle) under basal conditions. Experiments were performed in
26 fibers from five hearts in the nonfailing group and in 25 fibers
from nine hearts in the failing group. Maximal developed tension
was 585±135 ,iN in nonfailing myocardium and 512+76 ,N in
failing myocardium. NYHA indicates New York Heart Association
classification.
to eject a subnormal stroke volume when end-diastolic
volume is elevated.28 Several studies have suggested that
the response of the failing heart to increased volume is
probably reduced. In consequence, the cardiac output
of the dilated failing human left ventricle is less influenced by changes of preload rather than by changes of
afterload, with an increase in the latter leading to a
decrease in cardiac output. Therefore, myocardial
stroke volume in failing human myocardium may not be
dependent on the filling pressure or the preload after a
certain degree of myocardial failure. In addition, in
spontaneously hypertensive rats with chronic pressure
overload and normal left ventricular end-diastolic pressure, the left ventricular ejection fraction is depressed.2930 Chronic left ventricular volume overload
created by a arteriovenous fistula caused depression of
various indexes of contractility.31,32 Thus, both chronic
pressure and volume overload depress contractility. The
generally accepted Starling's law of the heart may not be
applicable in the failing human myocardium. The same
seems to be true in certain animal models of heart
failure.4,33
Starling's law of the heart is suggested to be based on
the myocardial length-active tension relation,34'35 in
which the force of contraction and the extent of shortening at any given tension depend on the initial muscle
length. Sarcomere overstretching has been thought to
be responsible for the decreased myocardial contractility in heart failure. However, subsequent studies have
shown that chronic volume loading is followed by a
progressive recruitment of sarcomeres stretched to the
length at which maximal active tension develops.3 In
myocardium removed from overloaded dilated hearts
and fixed at the elevated filling pressures that exist
during life, sarcomere length was similar to that in
normal cardiac muscle.33 Also, in preparations from
human myocardium terminally failing because of dilated cardiomyopathy, sarcomere length was not different from that in nonfailing control preparations.36 Mean
sarcomere lengths in preparations from diseased and
nonfailing myocardium were similar. Thus, an overstretch of sarcomeres or a different length in failing
Schwinger et al Heart Failure and the Frank-Starling Mechanism
TABLE 4. Ca2+ Sensitivity of Skinned Fibers From
Patients With Dilated Cardiomyopathy and From
Patients With Nonfailing Myocardium
Nonfailing
Myocardium
Slack position, n
ECw (mean), ,umol/L
95% confidence limit, ,umol/L
Hill coefficient
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Slack position, n
EC50 (mean), ,umol/L
95% confidence limit,
Hill coefficient
120% of resting length
EC50 (mean), ,umol/L
95% confidence limit,
Hill coefficient
26
1.69
1.55-1.86
2.0+0.4
9
1.90
g.mol/L
ALmol/L
1.74-2.10
2.3+0.5
9
1.43
1.25-1.64
2.3±0.5
NYHA IV
DCM
25
1.03
0.88-1.21
2.3+0.4
9
1.07
0.76-1.50
2.5+0.5
9
0.81
0.70-1.27
2.5+0.5
4
4
Slack position
0.84
1.97
EC50 (mean), gmol/L
0.61-1.20
1.68-2.32
95% confidence limit, ,umol/L
2.6±0.5
2.2±0.5
Hill coefficient
4
4
130% of resting length
1.46
0.80
EC50 (mean), grmol/L
1.32-1.62
0.52-1.23
95% confidence limit, gmol/L
2.3±0.4
2.2+0.6
Hill coefficient
NYHA IV DCM indicates dilated cardiomyopathy (New York
Heart Association class IV).
The degree of cooperativity is given by the Hill coefficients.
Fibers were either stretched to 120% of the resting length (slack
position) or to 130% of resting length.
human myocardium is very unlikely to explain the
different force-tension behavior in failing versus nonfailing human myocardium.
The differences in the length-dependent force development in papillary muscle strips may be due to either
a change of the Ca21 sensitivity of the myofibrils or an
altered amount of Ca'+ supplied to the myofibrils.3'7"1",37
To study whether the lack of tension-related increase in
the force of contraction is due to altered myofibrillar
Ca21 sensitivity, concentration-response curves for Ca2
were performed with skinned-fiber preparations of failing and nonfailing human myocardial tissue. At basal
conditions (slack position), the Ca2 sensitivity of myofibrils from human failing myocardium was significantly
higher compared with nonfailing myocardium (the present study). In addition, skinned-fiber preparations from
dystrophic mice were also more sensitive to Ca21 compared with those from normal mice.38 Moreover, in
skinned-fiber preparations from diabetic rats, maximum
Ca 2-activated force was unchanged compared with that
in healthy animals, but a significant increase in the Ca'+
sensitivity was observed.39 Therefore, myocardium from
diseased animals and from human hearts terminally
failing because of dilated cardiomyopathy shows an
965
increased sensitivity to Ca2`. Gwathmey et a140 studied
the frequency-related force potentiation in control and
myopathic human cardiac tissue. They observed an
attenuated frequency-related force potentiation in failing human tissue despite an increase in resting intracellular Ca2' and in the peak amplitude of the Ca'+
transient as detected with aequorin. They concluded
that these changes could be due to differences in
myofibrillar Ca'+ responsiveness in myopathic and nonfailing human tissue. The Ca' sensitivity as judged
from the pCa values increased when sarcomere length
was enhanced in skinned-fiber preparations from nonfailing human myocardium exclusively. In human myocardium failing because of dilated cardiomyopathy, this
length-dependent increase in force of contraction was
attenuated41 or even absent (the present study). Thus,
the failing human heart might be unable to increase the
myofibrillar Ca2' sensitivity. Consistently, at constant
[Ca+], increasing the fiber length by stretching was not
followed by a significant increase in tension in preparations from failing myocardium. These differences in the
Ca2' sensitivity of myofibrils relative to fiber length
might account for the altered length-tension relation in
the intact preparations of failing human heart. However, in skinned-fiber preparations from myopathic
hearts, changes of sarcomere length could not be measured routinely after stretching of the preparations,
possibly because of a disarrangement of thin and thick
filaments. The Ca'+ sensitivity of the myofibrils was
suggested to explain the altered contraction coupling at
various loading conditions in nonfailing human or animal tissue.37 The Ca ` affinity to troponin C was
suggested to play an important role in determining the
length-tension relation.9'42-45 In skinned ventricular trabeculae from hamster hearts in which the cardiac form
of troponin C was substituted with skeletal troponin C,
length-dependent increases in force of contraction were
suppressed in the skeletal troponin C form.42 The high
Hill coefficient observed (2 to 2.7) indicates that there
may be additional mechanisms positively influencing
cooperativity, such as myosin-actin binding.46 In respect
to the single Ca2'-binding site of troponin C, one would
expect a Hill coefficient of ~1. Consistently, it is not
clear whether or not cardiac troponin C might play an
important role in determining the length-tension relation. In addition, Gwathmey and Hajjar47 demonstrated
that protein kinase C activation may influence the Ca2
sensitivity of myofibrils as well. This effect may be due
to phosphorylation of troponin I and troponin T. As the
contractile proteins seem to be rather unchanged48'49 in
dilated cardiomyopathy, possible mechanisms responsible for the reduced force-tension relation may be due to
alterations of thin filament-regulatory proteins, eg,
troponin T or I isoforms.50'51 Therefore, the change in
myofibrillar ATPase activity has been suggested to be
the result of alterations in the isoform expression of
contractile proteins that regulate the sensitivity of the
myofilaments to Ca`. In a recently published study,
the cooperativity index for the pCa curve in skinnedfiber preparations was unchanged after stretching of
sarcomeres in nonfailing myocardium but decreased in
failing tissue.41 The investigations showed no significant
difference of Ca21 sensitivity between normal and failing myocardium when fibers were stretched. However, a
different skinning procedure was used in these experi-
966
Circulation Research Vol 74, No 5 May 1994
Em
E500
*
W
0
* Non
o
NonfIg 20 % sach
* NYHA V Slack-podffon
O NYHA IV 20 % Shed
E
O
C
100
NoN Sbd-pO
Nofalng 30 % sbtched
* NHA V Sl-poion
cn
0
0
*
o
0
NYHA IV 30 % seched
CD
c80
(0
0
.0
40
0 4
0
0 20
201V
0
0
ffi1
0.01
.030 .1
0.3
1
3
10
30
0.01 0.03
0.1
0.3
1
3
10
30
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Concentration of calcium (rmol/l)
Concentration of calcium (imol)
FIG 6. Concentration-response curves for Ca2+ (abscissa, logarithmic scale) in skinned-fiber preparations of nonfailing and terminally
failing human myocardium (left ventricular papillary muscle) under basal conditions and after stretching to 20% (left) and 30% (right)
above slack length. Data are given in Table 4. NYHA indicates New York Heart Association classification.
ments, allowing the study of Ca2'-release activity of the
sarcoplasmatic reticulum.41 Therefore, the Ca'2-triggered Ca2' release could have influenced the results. In
addition, in this experimental approach the amount of
subcellular structures present after the skinning procedure remains unknown. Most likely, myofibrillar and
intracellular factors may affect the molecular basis of
Starling's law.3'12 This is strengthened by the finding that
the Ca2 sensitivity increases, even at lengths well in the
descending limb of the length-tension relation, where
the number of crossbridges decreases.7
In addition to the Ca2 sensitivity of myofibrils, the
amount of Ca2 supplied to the myofibrils is also length
dependent.3,7 This is supported by the finding that
phasic contraction in skinned cardiac cells triggered by
free Ca2+ was sensitive to length.7 A role for the
Ca2+-induced Ca2' release in the length-dependent
effect on tension has been supposed.750 Changes in
cardiac performance at a given resting length can be
influenced through changes in the effectiveness of
mechanisms that govern excitation-contraction coupling, including Ca21 transport across cell membranes,
Ca2+ loading of the sarcoplasmatic reticulum,52 levels of
Mg 2+_ATPase53,54 or sarcoplasmatic reticulum Ca2'ATPase,55 and phosphorylation of phospholamban in
addition to Ca2+ sensitivity of the contractile filaments.37,56,57 From skinned-fiber studies performed with
nonfailing and failing human myocardium, the gating
mechanism of the Ca2+-release channel of the sarcoplasmatic reticulum was found to be altered in myopathic tissue.41 At increased muscle lengths, the peak of
the intracellular Ca2+ transient increases.7,12,58 Experiments in single cardiac cells have shown that the
sarcoplasmatic reticulum-Ca2+ release is indeed length
dependent.7 It seems to be possible that both inotropic
interventions and changes of muscle length may act
primarily through mechanisms that involve Ca2+ activation37 or Ca2' handling to enhance contractility. This is
supported by the finding that after an increase in muscle
length, the developed tension increased immediately
and then slowly over a period of minutes in human
myocardium.'2 Therefore, changes of the intracellular
Ca 2+ handling could also be responsible for the observed changes. Thus, the influence of enhanced Ca2'
i
.0
Cd
CO
Nonfailing
0
250
CD
c
'S2000
* cai
O
CacIm
0.65mol
.2
NYHA IV
.025
a
* Cc
0.6Sm
o CAu 0.S5,moUI
A Ca~ 1.70pd4
0.SpmoI/1
CalcIum 1.70pmldl
A Cacum 12.9Op0
A
200
50
A
CaIm 12.90pmoLf
0
0
100w
100
100
110
110
12D
130
140
150
h (%)
FiberFibe-ln
FIG 7. Graphs showing developed tension (ordinate, percent basal) plotted as a function of muscle length (abscissa, percent slack
length) in skinned-fiber preparations from nonfailing (left) and terminally failing (right, dilated cardiomyopathy) myocardium (human left
ventricular papillary muscle). Experiments were performed in four or five fibers from five hearts in the nonfailing and in three to six fibers
from five hearts in the failing group. NYHA indicates New York Heart Association classification.
Schwinger et al Heart Failure and the Frank-Starling Mechanism
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
availability after positive inotropic stimulation with ouabain on the force-tension relation was studied.
The cardiac glycoside ouabain was applied in a low
concentration.14 This pretreatment with ouabain restored the positive force-tension relation in failing human myocardium. In human nonfailing tissue, the forcetension relation was not significantly influenced by
ouabain. In both conditions, the force-tension relation
was shifted to the left. This is in accordance with
findings in papillary muscle strip preparations from
nondiseased cats.59 After inotropic stimulation with
Ca'+ or the ,B-adrenergic receptor agonist isoprenaline,
the length-tension curve shifted to the left, and the
length producing optimal contraction decreased.59
Mechanisms responsible for the restored force-tension
relation may include altered [Na+]i or [Ca'+]j,60 altered
activity of the Na+-Ca'+ exchange mechanism, altered
Ca'+-triggered Ca'+ release61 from the sarcoplasmatic
reticulum,40 or differences in Na+,K+-ATPase function
in failing and nonfailing human myocardium. Pharmacologic interventions that increase the time course of
the intracellular Ca'+ transient and twitch contraction
shift the peak force-peak relation in muscle preparations of myopathic human hearts to the left, whereas
interventions that decrease the time course of twitch
contraction and the accompanying intracellular Ca2+
transient shift the peak force versus peak intracellular
Ca'+ relation to the right.62 The observed findings
indicate that the absolute amount of intracellular available Ca'2 rather than mechanisms responsible for Ca'2
movements are determinants of myocardial contractility. This suggests that the dysfunction does not necessarily reside at the level of the myofilaments exclusively.
Consistently, the maximal activated contractions in
skinned-fiber preparations of failing and nonfailing human hearts as well as of multicellular muscle preparations were identical (References 36 and 41 and the
present study). Thus, these observations suggest that
the contractile apparatus in failing human myocardium
may be sufficient to generate a maximal force of contraction similar to that in nonfailing myocardium.
Taken together, the failing human myocardium may
be unable to use the Frank-Starling mechanism because
of alterations in intracellular Ca'2 handling in addition
to the failure to respond to increased preload with
enhanced Ca'2 sensitivity. However, in the terminally
failing human myocardium, the sensitivity to Ca'2 was
higher than in nonfailing tissue, thus suggesting that the
improvement in contractility due to sensitization of the
contractile proteins toward Ca'2 was already maximal.
In a state with almost maximal activation of the FrankStarling mechanism, ie, in terminal heart failure, additional stretching of the sarcomere length may not enhance the Ca'2 sensitivity. In other words, in human
myocardium failing because of dilated cardiomyopathy,
the length-dependent increase in the Ca'+ sensitivity is
shifted to lower sarcomere length, possibly by alterations in thin filament-regulatory proteins.
Limitations of the Study
The absolute values of force generation in the present
study in isolated human papillary muscles seem to be
lower than those reported by Mulieri et al19 at first
glance. However, they do agree quite nicely with those
reported by Phillips et al24 (in trabeculae carneae at
967
30°C, at 2.5 mmol/L Ca2, and at 0.33 Hz, 6.2+0.8
mN/mm2 was found), by Gwathmey et al in Fig 2 of
Reference 62 (in trabeculae carneae at 30°C, at 16
mmol/L Ca2' and at 0.33 Hz, 1.84+0.74 g/mm2
[18.4+ 7.7 mN/mm2] in myopathic muscle and 1.23+±0.49
g/mm2 [12.3 +4.9 mN/mm2] in control muscle was
found), and by Ginsburg et al23 (in papillary muscle and
trabeculae carneae at 37°C, at 2.5 mmol/L Ca'+, and at
0.6 Hz, 2.7+0.8 mN/mm2 [control muscle] and 4.0+0.9
mN/mm2 [cardiomyopathic muscle] was found). When
one also takes into account the different frequency of
stimulation as used in our experiments (1 Hz), the
different [Ca2] value (1.8 mmol/L), and the different
temperature (37°C), these discrepancies may resolve.
Of course, it cannot be excluded that muscle strands as
taken immediately after cardiac arrest from patients
undergoing coronary bypass surgery from the surface of
their normal left ventricle generate more force than do
papillary muscles. Unfortunately, it was not possible to
directly compare muscle strands as used by Mulieri et
al19 and papillary muscles from the same patients
because of the decision of the local ethical committee.
In the present study, only muscle strip preparations
from the left ventricular papillary muscle have been
used. This approach allows us to compare tissue taken
from identical regions of the heart in failing and nonfailing myocardium. However, in the in vivo situation
the effect of volume loading would be a more appropriate technique for studying the ability of the left ventricle
to accommodate an increase in preload. This experimental design has been used in a study using conscious
dogs.63 During the development of congestive heart
failure by rapid pacing followed by a chronic dilatation,
the application of an acute volume load failed to elicit
additional increase in cardiac output and stroke volume.63 These findings may further indicate that the
Frank-Starling mechanism is exhausted in congestive
heart failure states.
In conclusion, the failure of the dilated failing human
heart to respond to an increase of load with an increase
of contractility is the pathophysiological basis of the
ineffective Frank-Starling mechanism in vivo. The pathomechanism is suggested to involve a reduced increase of
myofibrillar Ca2 sensitization after an increase of sarcomere length. An altered intracellular Ca2' homeostasis could also play a role. Therapeutic interventions
targeting this mechanism might be of importance in the
treatment of heart failure.
Acknowledgments
This study was supported by the Deutsche Forschungsgemeinschaft. We thank Heidrun Villena Hermoza for her
excellent technical help.
References
1. Plotnick GD, Becker LC, Fisher ML, Gerstenblith G, Renlund
DG, Fleg JL, Weisfeldt ML, Lakatta EG. Use of the FrankStarling mechanism during submaximal versus maximal upright
exercise. Am J Physiol. 1986;251 :H1 101-H1105.
2. Higginbotham MB, Morris KG, Williams RS, McHale PA,
Coleman RE, Cobb FR. Regulation of stroke volume during submaximal and maximal upright exercise in normal man. Circ Res.
1986;58:281-291.
3. Kenitsch JC. The length-tension relation in the failing myocardium
and its cellular basis. Heart Failure. 1988;4:125-136.
4. Spotnitz HM, Leyton RA, Kelly DT, Beiser GD, Pierce JE,
Epstein SE, Sonnenblick EH. "Overstretched" sarcomeres in
968
5.
6.
7.
8.
Circulation Research Vol 74, No 5 May 1994
subacute volume-pressure loading of dog right ventricles. Circulation. 1972;45/46(suppl II):II-44. Abstract.
Ross J Jr, Sonnenblick EH, Taylor RR, Spotnitz HM, Covell JW.
Diastolic geometry and sarcomere lengths in the chronically
dilated canine left ventricle. Circ Res. 1971;28:49-61.
Fabiato A, Fabiato A. Dependence of the contractile activation of
skinned cardiac cells on the sarcomere length. Nature. 1975;256:
54-56.
Allen DG, Kentish JG. The cellular basis of the length-tension
relation in cardiac muscle. J Mol Cell Cardiol. 1985;17:821-840.
Lakatta EG, Spurgeon HA. Force staircase kinetics in mammalian
cardiac muscle: modulation by muscle length. J Physiol (Lond).
1980;299:337-352.
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
9. Gulati J. Length-sensing function of troponin C and Starling's Law
of the heart. Circulation. 1992;85:1954-1955.
10. Fabiato A. Sarcomere length dependence of calcium release from
the sarcoplasmic reticulum of skinned cardiac cells demonstrated
by differential microspectrophotometry with arsenazso III. J Gen
Physiol. 1980;76:15a. Abstract.
11. Allen DG, Jewell BR, Murray JW. The contribution of activation
processes to the length-tension relation of cardiac muscle. Nature.
1974;248:606- 607.
12. Allen DG, Kurihara S. The effects of muscle length on intracellar
calcium transients in mammalian cardiac muscle. J Physiol (Lond).
1982;327:79-84.
13. Akera T, Brody TM. The role of Na+,K'-ATPase in the inotropic
action of digitalis. Pharmacol Rev. 1978;29:187-220.
14. Schwinger RHG, Bohm M, Rosee KL, Schmidt U, Schulz C,
Erdmann E. Na+-channel activators increase cardiac glycoside sensitivity in failing human myocardium. J Cardiovasc Pharmacol.
1992;19:554-561.
15. Paradise NF, Schmitter JL, Surmitis JM. Criteria for adequate
oxygenation of isometric kitten papillary muscle. Am J Physiol.
1981;241:H348-H353.
16. SAS User's Guide, Version 6 Edition. Cary, NC: SAS Institute Inc;
1989.
17. Bowditch HP. Uber die Eigenthumlichkeiten der Reizbarkeit,
welche die Muskelfasern des Herzens zeigen. Berl Sachs Ges Akad
Wiss. 1871:652-689.
18. Feldman MD, Alderman JD, Aroesty JM, Royal HD, Ferguson JJ,
Owen RM, Grossman W, Mckay RG. Depression of systolic and
diastolic myocardial reserve during atrial pacing tachycardia in
patients with dilated cardiomyopathy. J Clin Invest. 1988;82:
1661-1669.
19. Mulieri LA, Hasenfuss G, Leavitt B, Allen PD, Alpert NR. Altered
myocardial force-frequency relation in human heart failure. Circulation. 1992;85:1743-1750.
20. Schwinger RHG, Bohm M, Erdmann E. Inotropic and lusitropic
dysfunction in myocardium from patients with dilated cardiomyopathy. Am HeartJ. 1992;123:116-128.
21. Ebrecht G, Rupp H, Jacob R. Alterations of mechanical
parameters in chemically skinned preparations of rat myocardium
as a function of isoenzyme pattern of myosine. Basic Res Cardiol.
1982;77:220-234.
22. Hajjar RJ, Gwathmey JK, Briggs GM, Morgan JP. Differential
effect of DPI 201-106 on the sensitivity of the myofilaments to Ca 2+
in intact and skinned trabeculae from control and myopathic
human hearts. J Clin Invest. 1988;82:1578-1584.
23. Ginsburg R, Bristow MR, Billingham ME, Stinson EB, Schroeder
JS, Harrison DC. Study of the normal and failing isolated human
heart: decreased response of failing heart to isoproterenol. Am
Heart J. 1983;106:535-540.
24. Phillips PJ, Gwathmey JK, Feldmann MD, Schoen FJ, Grossman
W, Morgan JP. Postextrasystolic potentiation and the forcefrequency relationship: differential augmentation of myocardial
contractility in working myocardium from patients with end-stage
heart failure. J Mol Cell Cardiol. 1990;22:99-110.
25. Renlund DG, Gerstenblith G, Fleg JL, Becker LC, Lakatta EG.
Interaction between left ventricular end-diastolic and end-systolic
volumes in normal humans. Am J Physiol. 1990;258:H473-H481.
26. Gillebert TC, Sys SU, Brutsaert DL. Influence of loading patterns
on peak length-tension relation and on relaxation in cardiac
muscle. J Am Coll Cardiol. 1989;13:483 - 490.
27. Wilson JR, Reichek N, Dunkman WB, Goldberg S. Effect of
diuresis on the performance of the failing left ventricle in man. Am
J Med. 1981;70:234-239.
28. Starling EH. Linacre Lecture on the Law of the Heart: 1915.
London, England: Longmans; 1918.
29. Mirsky I, Pfeffer JM, Pfeffer MA, Braunwald E. The contractile
state as the major determinant in the evolution of left ventricular
dysfunction in the spontaneously hypertensive rat. Circ Res. 1983;
53:767-778.
30. Morgan HE, Backer KM. Cardiac hypertrophy: mechanical neural
and endocrine dependence. Circulation. 1991;83:13-20.
31. Pinsky WW, Lewis RM, Harteley CJ, Entman ML. Permanent
changes of ventricular contractility and compliance in chronic
volume overload. Am J Physiol. 1979;237:H575-H580.
32. Alyono D, Ring WS, Anderson MR, Anderson RW. Left ventricular adaptation to volume overload from large aortocaval
fistula. Surgery. 1984;96:360-367.
33. Ross J Jr, Sonnenblick EH, Taylor RR, Spotnitz HM, Covell JW.
Diastolic geometry and sarcomere length in the chronically dilated
canine left ventricle. Circ Res. 1971;28:49-61.
34. Lakatta EG. Starling's law of the heart is explained by an intimate
interaction of muscle length and myofilament calcium activation.
JAm Coll Cardiol. 1978;10:1157-1164.
35. Lakatta EG. Length modulation of muscle performance. In:
Fozzard HA, ed. The Heart and Cardiovascular System. 2nd ed.
New York, NY: Raven Press Publishers; 1992:1325.
36. Wankerl M, Bohm M, Morano I, Ruegg JC, Eichhorn M, Erdmann
E. Calcium sensitivity and myosin light chain pattern of atrial and
ventricular skinned cardiac fibers from patients with various kinds
of cardiac disease. J Mol Cell Cardiol. 1990;22:1425-1438.
37. Morgan JP. Abnormal intracellular modulation of calcium as a
major cause of cardiac contractile dysfunction. N Engl J Med.
1991;325:625- 632.
38. West JM, Stephenson DG. Contractile activation and the effects of
2,3-butanedione monoxime (BDM) in skinned cardiac preparations from normal and dystrophic mice (129/ReJ). Pflugers Arch.
1989;413:546-552.
39. Khandoudi N, Guo AC, Chesnais M, Feuvray D. Skinned cardiac
fibers of diabetic rats: contractile activation and effects of 2,3butanedione monoxime (BDM) and caffeine. Cardiovasc Res. 1993;
27:447-452.
40. Gwathmey JK, Slawsky MT, Hajjar RJ, Briggs GM, Morgan JP.
Role of intracellular calcium handling in force-interval relationship of human ventricular myocardium. J Clin Invest. 1990;85:
1599-1613.
41. D'Agnolo A, Luciani GB, Mazzucco A, Gallucci V, Salviati G.
Contractile properties and Ca 2-release activity of the sarcoplasmic reticulum in dilated cardiomyopathy. Circulation. 1992;85:
518-525.
42. Babu A, Sonnenblick E, Gulati J. Molecular basis for the influence
of muscle length on myocardial performance. Science. 1988;240:
74-76.
43. Gulati J, Sonnenblick E, Babu A. The role of troponin C in the
length dependence of Ca2`-sensitive force of mammalian skeletal
and cardiac muscles. J Physiol (Lond). 1991;441:305-324.
44. Parmacek MS, Leiden JM. Structure, function, and regulation of
troponin C. Circulation. 1991;84:991-1003.
45. Moss RL, Nwoye LO, Greaser ML. Replacement of skeletal TnC
with cardiac TnC does not alter the length dependence of tension
development in skinned skeletal muscle fiber. Biophys J. 1990;
57:146. Abstract.
46. Bremel RD, Weber A. Cooperation within actin filament in vertebral skeletal muscle. Nature. 1972;238:97-101.
47. Gwathmey JK, Hajjar RJ. Effect of protein kinase C activation on
sarcoplasmic reticulum function and apparent myofibrillar Ca2l
sensitivity in intact and skinned muscles from normal and diseased
human myocardium. Circ Res. 1990;67:744-752.
48. Wiegand V, Schuler S, Figulla H, Warnecke H, Kreuzer H. Contractile proteins in human dilatative cardiomyopathy. J Mol Cell
Cardiol. 1986;18(suppl 2):74. Abstract.
49. Hirzel HO, Tuchschmic CR, Schneider J, Krayenbuehl HP,
Schaub MC. Relation between myosin isoenzyme composition,
hemodynamics and myocardial structure in various forms of human
cardiac hypertrophy. Circ Res. 1985;57:729-740.
50. Anderson PAW, Malouf NN, Oakeley AE, Pagani ED, Allen PD.
Troponin T isoform expression in humans: a comparison among
normal and failing adult heart, fetal heart, and adult and fetal
skeletal muscle. Circ Res. 1991;69:1226-1233.
51. Hunkeler NM, Kullman J, Murphy AM. Troponin I isoform
expression in human heart. Circ Res. 1991;69:1409-1414.
52. Movsesian MA, Bristow MR, Krall J. Ca2` uptake by cardiac
sarcoplasmic reticulum from patients with idiopathic dilated cardiomyopathy. Circ Res. 1989;65:1141-1144.
53. Pagani ED, Alousi AA, Grant AM, Older TM, Dziuban SW Jr,
Allen PD. Changes in myofibrillar content and Mg-ATPase activity
in ventricular tissues from patients with heart failure caused by
Schwinger et al Heart Failure and the Frank-Starling Mechanism
54.
55.
56.
57.
58.
coronary artery disease, cardiomyopathy, or mitral valve insufficiency. Circ Res. 1988;63:380-385.
Alousi AA, Grant AM, Etzler JR, Cofer BR, Van der Bel-Kahn J,
Melvin D. Reduced cardiac myofibrillar Mg-ATPase activity
without changes in myosin isoenzymes in patients with end-stage
heart failure. Mol Cell Biochem. 1990;96:79-88.
Harigaya S, Schwartz A. Rate of calcium binding and uptake in
normal animal and failing human cardiac muscle. Circ Res. 1969;
25:781-793.
Davis BA, Edes J, Gupta RC, Young EF, Kim HW, Steenart NAB,
Szymanska G, Kranias EG. The role of phospholamban in the
regulation of calcium transport by cardiac sarcoplasmic reticulum.
Mol Cell Biochem. 1990;99:83-88.
Limas CJ, Olivari MT, Goldenberg IF, Levine TB, Benditt DG,
Simon A. Calcium uptake by sarcoplasmic reticulum in human
dilated cardiomyopathy. Cardiovasc Res. 1987;21:601-605.
Fabiato A. Myoplasmic free calcium concentration reached during
the twitch of an intact isolated cardiac cell and calcium-induced
59.
60.
61.
62.
63.
969
release of calcium from the sarcoplasmic reticulum of a skinned
cardiac cell from the adult rat or rabbit ventricle. J Gen Physiol
1981;78:457-497.
Bassatt AL, Bartelstone HJ, Hoffman BF. Cardiac muscle: changes
in optimal length during inotropic interventions. Experientia. 1969;
25:153-154.
Beuckelmann DJ, Nabauer M, Erdmann E. Intracellular calcium
handling in isolated ventricular myocytes from patients with
terminal heart failure. Circulation. 1992;85:1046-1055.
Nabauer M, Callwaert G, Cleemann L, Morad M. Regulation of
calcium release is gated by calcium current, not gating charge, in
cardiac myocytes. Science. 1989;244:800-803.
Gwathmey JK, Hajjar RJ. Relation between steady state force and
intracellular (Ca2) in intact human myocardium. Circulation.
1990;82:1266-1278.
Komamura K, Shannon RP, Ihara T, Shen YT, Mirsky I, Bishop S,
Vatner ST. Exhaustion of Frank-Starling mechanism in conscious
dogs with heart failure. Am J PhysioL 1993;265:H1119-H1131.
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
The failing human heart is unable to use the Frank-Starling mechanism.
R H Schwinger, M Böhm, A Koch, U Schmidt, I Morano, H J Eissner, P Uberfuhr, B Reichart
and E Erdmann
Downloaded from http://circres.ahajournals.org/ by guest on April 28, 2017
Circ Res. 1994;74:959-969
doi: 10.1161/01.RES.74.5.959
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1994 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/74/5/959
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/