Download CLINICAL INVESTIGATION Sympathetic reflex control

CLINICAL INVESTIGATION
Sympathetic reflex control of skeletal muscle blood
flow in patients with congestive heart failure:
evidence for ,8-adrenergic circulatory control
ELI KASSIs, M.D., TAGE N. JACOBSEN, M.D., FLEMMING MOGENSEN, M.D.,
AND OLE AMTORP, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
ABSTRACT Mechanisms controlling forearm muscle vascular resistance (FMVR) during postural
changes were investigated in seven patients with severe congestive heart failure (CHF) and in seven
control subjects with unimpaired left ventricular function. Relative brachioradial muscle blood flow
was determined by the local 133Xe-washout technique. Unloading of baroreceptors with use of 45
degree upright tilt was comparably obtained in the patients with CHP and control subjects. Control
subjects had substantially increased FMVR and heart rate to maintain arterial pressure whereas patients
with CHF had decreased FMVR by 51 + 1 % (mean ± SEM, p < .02) and had no increase in heart
rate despite a fall in arterial pressure during upright tilt. The autoregulatory and local vasoconstrictor
reflex responsiveness during postural changes in forearm vascular pressures were intact in both groups.
Further investigations were carried out in the patients with CHF. The left axillary nerve plexus was
blocked by local anesthesia in the seven patients. No alterations in forearm vascular pressures were
observed.' This blockade preserved the local regulation of FMVR but reversed the vasodilator response
to upright tilt as FMVR increased by 30 + 7% (p < .02). Blockade of central neural impulses to this
limb combined with brachial arterial infusions of phentolamine completely abolished the humoral
vasoconstriction in the tilted position. Infusions of propranolol to the contralateral brachial artery that
did not affect baseline values of heart rate, arterial pressure, or the local reflex regulation of FMVR
reversed the abnormal vasodilator response to upright tilt as FMVR increased by 42 12% (p < .02).
Despite augmented baseline values, forearm venous but not arterial plasma levels of epinephrine
increased in the tilted position, as did arterial rather than venous plasma concentrations of norepinephrine in these patients. The results suggest a 18-adrenergic reflex mechanism elicited by spinal or
supraspinal neural impulses and probably modulating a cotransmitter release in the patients with CHF.
Circulation 74, No. 5, 929-938, 1986.
±
REFLEX increases in heart rate and vascular tone are
essential for the maintenance of arterial pressure during upright tilt. The stimulus for these circulatory adjustments appears to reside mainly in cardiopulmonary
and arterial baroreceptors exerting afferent restraint on
sympathetic efferent outflow.' Recent studies in normal human beings have indicated that postural changes
causing a threshold increase of vascular transmural
pressure elicit an arteriolar constriction by means of a
local venoarteriolar sympathetic axon mechanism.2-5
This local vasoconstrictor-reflex response has recently
been shown to contribute to the maintenance of arterial
pressure during postural changes.6 7
Patients with congestive heart failure (CHF) have
From the Departments of Cardiology and Anesthesiology, Gentofte
Hospital, University of Copenhagen, Copenhagen, Denmark.
Address for correspondence: Eli Kassis, M.D., Cardiology Laboratory, Rigshospitalet, Blegsdamvej 9, DK-2100 0 Copenhagen, Denmark.
Received Nov. 11, 1985; revision accepted July 10, 1986.
Vol. 74, No. 5, November 1986
demonstrated abnormal vascular responses to orthostatic stress.8'9 These abnormalities have recently been
related to selectively impaired baroreflex control of
forearm vascular tone. '1 A reduced baroreflex afferent
inhibitory influence on brainstem vasomotor centers,
may explain the neurohumoral excitation widely documented in patients with CHF. 1-15 In fact, animal preparations of chronic heart failure have shown impairment of the arterial baroreflex cardiovascular control'6
as well as a chronically reduced afferent restraint and
degenerative changes of cardiopulmonary baroreceptors.17' 1' However, the augmented sympathetic drive
in patients with CHF'1'` is contrasted by depletion of
norepinephrine in cardiac tissue" and attenuated cardiac responsiveness to sympathetic efferent activity."15 19
With the local `33Xe-washout technique for determination of blood flow,20 consistent results have been
obtained in a recent series of studies of different human
tissues.`~' It has recently been confirmed2' that this
929
KASSIS et al.
method
gives reliable estimates of relative blood flow
different
experimental conditions, including
during
those with dilated veins. It is known that the veins of
patients with CHF are less distensible than those of
normal subjects.22,23
Using the same method to estimate relative changes
in forearm muscle vascular resistance (FMVR), we
performed this study to determine circulatory adjustments to stimuli known to induce sympathetic activation in normal human subjects and to elucidate mechanisms underlying these adjustments in patients with
CHF.
Methods
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Patient selection. Seven male patients ranging in age from
43 to 61 years were studied. All were selected on the basis of
severe CHF that was inadequately controlled by conventional
treatment with digoxin and diuretics. The cause of CHF was
ischemic in all patients with angiographically documented coronary artery disease. No patient was studied within 3 months of a
myocardial infarction. For a minimum of 4 weeks before study,
all vasodilators, calcium antagonists, and l3-blocking agents
were withheld while treatment with digoxin and diuretics was
regulated and maintained. All patients were in sinus rhythm and
had normal electrolytes, blood counts, and therapeutic serum
levels of digoxin during the study. Central and peripheral hemodynamic studies were performed before noon on successive
days. The study protocol was carefully explained and informed
consent was obtained from all patients. The protocol was approved by the official ethical committee in Copenhagen.
Procedures
Central hemodynamic measurements. The patients were
studied in the supine postabsorptive state. All catheter procedures were performed under fluoroscopic control. A balloon
tipped, triple-lumen thermodilution catheter (Gould SP 1435)
was placed in the pulmonary artery for pressure measurements
and determination of cardiac output (dilution curves of five
consecutive measurements were recorded). A Cordis pigtail
catheter was introduced percutaneously through the right femoral artery and positioned in the left ventricle for pressure recordings and angiography. All pressures were measured with
reference to midaxillary level and were recorded with Statham
P23Db transducers. Systemic vascular resistance (dyne-seccm- 5) was calculated as 80 x (mean arterial pressure - right
atrial pressure)/(cardiac output).
Half an hour after angiography, control measurements of
aortic, right atrial, and left ventricular filling pressures were
performed while patients rested supine on a tilt table. Thereafter
the patients were passively and slowly tilted to a 45 degree angle
with the feet supported by a base platform. Pressure measurements were repeated within 5 to 7 min after initiation of tilt and
with reference to the right atrium level set for each patient.
Peripheral hemodynamic measurements. On the morning of
testing, the patients were placed horizontally on a tilt table.
Room temperature was 22° C and remained constant during the
entire investigation. The left axillary nerve plexus was percutaneously infiltrated by 40 ml mepivacaine 1% without vasoconstringent. The effectiveness of this proximal nerve blockade
(PNB) to the left forearm was ensured by motor paralysis,
analgesia, loss of sweat secretion, and increase in skin temperature. This PNB was effective during the 4 hr period of investigation, and no complications or decreases in arterial pressure were
930
polyethylene arterial catheter (Viggo
placed percutaneously in the left and right
brachial arteries. An intravenous cannula was placed in a distal
observed. A 20-gauge
FlowSwitch)
was
vein of the forearm. The arterial and local venous pressures
were continuously measured with Statham P23Db transducers
and recorded simultaneously with heart rate on a Siemens Elema
mingograph 81. After placement of the monitoring lines and
induction of PNB, a 30 min rest period was allowed, during
which the patients were familiarized with the protocol.
Skeletal muscle blood flow was measured by the local 133Xewashout technique. Via a thin cannula, 0.1 ml (1 mCi) of 133Xe
dissolved in isotonic saline (Radiochemical Center, Amersham,
England) was injected into the brachioradial muscle of each
forearmn. Measurements were started 20 min after the injection
to avoid influences from the injection trauma on the washout
rate of 133Xe.3' 6, 7, 24 The y-radiation of 133Xe was detected by
Nal (TI) scintillation detectors placed perpendicular to the sites
of injection at a distance of about 10 cm. Care was taken that the
counting geometry remained constant during a single investigation. The recorded activity was fed to a universal printing yspectrometer (Novo Diagnostic System) with a window of acceptance adjusted around the 81 keV photopeak of '33Xe. The
count rates were printed out every 5 sec without time delay.
The investigations consisted of triads of measurements on
both forearms, each part of a triad lasting 5 min: (1) during
reference level (refl), supine position with forearms studied at
midaxillary line; (2) during test conditions (test); (3) after return
to reference level (ref2). Such triads of measurements were
carried out before (refl), during (test), and after (ref2) each of
the following test conditions: elevation of forearms 20 cm above
reference level, lowering of forearms 40 cm below reference
level, and upright tilt (45 degrees) as described above, with
forearms studied at heart level.
With the same radioactive depot, measurements were repeated in all patients during brachial arterial infusions of propranolol
to the right forearm and brachial arterial infusions of phentolamine to the left neurally blocked forearm. Intra-arterial infusions of propranolol at 16.6 gg/min and of phentolamine at 500
,ug/min in isotonic saline were carried out by means of IVAC
531 continuous nonpulsatile pumps. The total volume of infusate was 0.5 ml/min; no measurable changes in forearm blood
flow have been observed during infusion of vehicle into the
brachial artery at these rates.25 Triads of measurements were
repeated as described above before, during, and after the test
conditions of 40 cm forearm lowering and of 45 degree upright
tilt. Measurements were started after 10 min and carried out
during 40 to 50 min of brachial arterial infusions of each drug.
In additional series of experiments, the effect of a single dose
of propranolol injected into the brachial artery or locally into the
brachioradial muscle was studied during upright tilt. A depot of
0.1 ml of the '33Xe solution was mixed with 0.1 ml (10 ng) of a
propranolol solution in saline (0.1 mg/liter) and injected into the
brachioradial muscle. An equal depot of the '33Xe solution
mixed with 0.1 ml saline was injected into the contralateral
brachioradial muscle and served as a control depot. Bilateral
simultaneous measurements were started 20 min after injecting
the radioactive depots. A group of three consecutive measurements, each lasting 5 min, was obtained during upright tilt and
during reference conditions before and after the tilt. The same
triad of measurements was repeated on the control depot within
15 min after a bolus injection of 0.3 mg (0.3 ml) of propranolol
into the brachial artery.
Protocol. The central and peripheral hemodynamic studies
were performed in all patients before noon on successive days
and at a room temperature of 22°C. The peripheral hemodynamic investigations were started after an hour rest period,
during which the patients were familiarized with thetechniques.
CIRCULATION
PATHOPHYSIOLOGY AND NATURAL HISTORY-CONGESTIVE HEART FAILURE
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
The sequence of these investigations was of the same order as
described above in all patients. The responses to upright tilt
before and after a single dose of propranolol injected locally or
intra-arterially were recorded separately for each patient.
Control subjects. Seven age-matched patients who underwent routine cardiac catheterization and had normal left ventricular ejection fraction and central hemodynamics served as a
control group. Brachioradial muscle blood flow regulation was
studied in the sequence described above. No PNB and no intraarterial infusion of vehicle were performed. Control subjects
were not receiving vasoactive drugs before the study.
Methodologic considerations and calculations. When the
washout of a tracer from a homogeneous tissue with homogeneous perfusion is not influenced by recirculation and is only
perfusion dependent, i.e., that diffusion equilibrium is achieved
within a single passage through the capillaries,26 the perfusion
coefficient, f, can be calculated as:
test conditions was secured by consistent results obtained on
different days in the same patient, the same area for the `33Xe
depot being used.
Within the triad of measurements from the same '33Xe depot,
A was assumed to remain constant and relative blood flow during test conditions was calculated as:
f = A k 100 (ml/min/lOOg)
where A and k denote the tissue-to-blood partition coefficient
(mllg) and the fractional washout rate constant (min 1), respectively.
Methodologic aspects of applying the local 1"3Xe-washout
method for measurement of blood flow in skeletal muscle have
been thoroughly evaluated in a number of studies.20' 24, 27, 28
Three factors are of importance for determination of resting
skeletal muscle blood flow. The initial steep washout rate after
intramuscular injection gives an overestimation of blood flow
because of the local influence of the injection trauma lasting
about 10 min.24 The later part of the washout curve, which is
approximately monoexponential, is influenced by two other
factors: (1) recirculation of the 133Xe due to venoarteriolar
shunting by diffusion27 and (2) accumulation of the 133Xe in fat
tissue lining the veins.28 Both factors tend to reduce the washout
rate of 133Xe, giving an underestimation of skeletal muscle
blood flow by about 20% to 60% at low flow rates as during
resting conditions.24 Thus it is not possible to obtain a correct
estimate of resting blood flow in skeletal muscle by the local
'33Xe technique. However, it is possible to obtain reliable estimates of relative blood flow by comparing the slopes of the
approximately monoexponential washout from the same 133Xe
depot during different test conditions if A remains constant as
the relative impact of shunting by diffusion and fat accumulation remains constant within the flow rates considered. Variations in A during the test conditions of venous stasis are less
than 3%.3 By relating the washout rate under test conditions to
reference values obtained just before and after the test, the small
spontaneous decrease in the washout rate is taken into account.3' 6,7, 20 Studies in dog limbs have shown a close correlation between relative changes in resting muscle blood flow
estimated by the local '33Xe washout technique and femoral
blood flow measured by an electromagnetic flowmeter.2'
The sources of error mentioned above were taken into account in the present study. (1) The same single depot in each
brachioradial muscle was used through the entire experiment
during 20 to 120 min after the intramuscular injection. Preliminary measurements on bilateral control depots in both brachioradial muscles were obtained on different days in two patients
during only the supine reference conditions. After an initial
steep slope, a constant shallow slope of the washout curve was
observed during about 15 to 125 min after the intramuscular
injection. (2) Two reference curves just before and after each
test conditions were obtained. (3) To avoid variations in relative
blood flow caused by other mechanisms than the test conditions,
data were analyzed only when the reference washout rate before
and after each test did not differ more than a maximum of 50%.
(4) Reproducibility of the relative changes in blood flow during
(APtest / APref) (kref / ktest)
where APtest is the mean perfusion pressure (Pa - P,) during test
Vol. 74, No. 5, November 1986
ftest / fref
=
ktest / kref
where kt,,, is the washout rate constant obtained during test
conditions, kref is the average value of kref 1 and kref. The k value
was calculated from the regression analysis (least-squares method) of the count rates transformed logarithmically and corrected
for background activity.
Relative vascular resistance during test conditions was calculated as:
Rtest / Rref
=
conditions and APref is the average value of perfusion pressures
obtained during the two reference conditions before and after
the test.
Determination of plasma catecholamine concentrations.
Plasma concentrations of norepinephrine and epinephrine were
determined by the double-isotope derivative technique29 and use
of a high-performance liquid chromatographic system (Waters
Associates, Milford, MA). Normal resting levels of norepinephrine and epinephrine were 0.4 to 1.9 and 0.07 to 0.44
nmol/liter, respectively. Venous and brachial arterial blood
samples were obtained simultaneously from the forearm 1 hr
after resting in supine position and within 9 to 10 min of initiating upright tilt.
Statistics. The results are given as mean ± SEM. The Wilcoxon signed-rank test was used. Comparison between patients
and control subjects was performed by the Wilcoxon rank-sum
test. A .05 level of significance was set.
Results
Central hemodynamics. The seven patients had classic
symptoms of dyspnea, fatigue, orthopnea, and paroxysmal nocturnal dyspnea for a mean of 18 months
before the study. Baseline hemodynamic measurements confirmed the presence of severely impaired left
ventricular performance with low stroke volume and
increased pressures and dimensions (table 1). Control
subjects had an angiographic left ventricular ejection
fraction of 75 + 4% and normal central hemodynamics.
Hemodynamic responses to postural changes (table 2).
During forearm elevation 20 cm above reference levels, the brachial perfusion pressure (Pa - PJ) decreased
by about 14 and 15 mm Hg in patients with CHF and in
control subjects, respectively. Forearm lowering 40
cm below reference levels caused parallel increments
in arterial and venous transmural pressures, the latter
corresponded to 33 + 2 and 30 ± 1 mm Hg in patients
and control subjects, respectively. During 45 degree
upright tilt, both right atrial and left ventricular filling
pressures decreased substantially in patients with CHF
931
KASSIS et al.
TABLE 1
Clinical and hemodynamic characteristics of patients with CHF
HR
(bpm)
LVEF
(%)
LVEDV
Patient
Age
(yr)
Duration
of
symptoms
(mo)
1
2
3
4
5
6
7
Mean
SEM
55
52
59
48
43
48
61
52
3
8
8
22
38
16
18
16
18
4
83
106
78
95
80
76
78
85
5
25
234
277
495
502
289
300
390
355
40
14
24
10
31
27
26
22
3
(ml)
Mean
Mean
LVEDP
RAP
PAP
(mm Hg) (mm Hg) (mm Hg)
16
32
20
26
20
18
20
22
2
8
7
6
8
5
6
6
7
1
Mean
AP
(mm Hg)
SVR
(dyne-seccm- 5)
(I/minIm2)
CI
92
104
104
75
80
95
84
91
4
22
39
22
42
37
27
40
33
3
2.3
1.7
2.1
2.2
2.1
2.2
1.9
2.1
0.1
1493
2282
1760
1191
1539
1780
1387
1633
133
HR = heart rate; LVEF = left ventricular ejection fraction; LVEDV left ventricular end-diastolic volume; LVEDP = left ventricular enddiastolic pressure; RAP = right atrial pressure; PAP = pulmonary arterial pressure; AP = aortic pressure; CI = cardiac index; SVR = systemic
vascular resistance.
=
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
and in control subjects. However, control subjects had
increased heart rate and maintained arterial mean pressure whereas patients with CHF showed no increase in
heart rate in the tilted position.
.02) corresponding to an increase in FMVR of 60
10% (p < .02) in control subjects, whereas in patients
with CHF blood flow increased by 76 ± 24% (p <
.02) corresponding to a decrease in FMVR of 51
11% (p < .02) (figure 1).
crease
Regulation of FMVR during postural changes. The dein brachial perfusion pressure observed in the
elevated forearm elicited no relative change in brachioradial muscle blood flow (0 ± 1%) in patients
Mechanisms underlying vasodilatation during upright
tilt in patients with CHF. An example of the 133Xe-washout curves obtained in one patient is shown in figure 2
with CHF and control subjects whose FMVR decreased by about 16% and 17%, respectively.
The increased venous transmural pressure observed
in the lowered forearm elicited a decrease in muscle
blood flow of 25 ± 3% (p < .02) in control subjects
and of 19 ± 6% (p < .02) in patients with CHF,
corresponding to substantial increments in FMVR that
were similar in both groups (figure 1).
During 45 degree upright tilt with the forearm at
heart level, blood flow decreased by 43 ± 5% (p <
and the compiled data of the `33Xe studies are shown in
figures 3 and 4.
Effects of localf3-adrenoceptor blockade. During 10 to 50
min of brachial arterial infusions of propranolol at 16.6
.g/min, no alterations in the baseline values of heart
rate or arterial and venous pressures were observed.
Measurements obtained during this period revealed a
vasoconstrictor response to venous pressure elevation
of about 32 mm Hg in the lowered forearm (figure 3,
B) that did not substantially differ from the control
TABLE 2
Mean values (±+ SEM) of arterial, venous, and cardiac filling pressures and heart rate during supine position and postural changes in seven
patients with CHF and seven control subjects (C)
Pa (mm
Supine
(bpm)
C
CHF
C
CHF
88
70
7
CHF
95
96
10
7
(7)
(4)
81A
(3)
121A
(1)
(1)
(7)
(2)
9
7
89
71
(1)
(2)
(6)
(1)
43A
37A
87
69
Forearm lowering (40 cm)
(6)
123A
mean
HR
C
80A
a-
P, (mm Hg)
CHF
Forearm elevation (20 cm)
Upright tilt (450)
Hg)
RAP (mm Hg)
C
LVFP (mm Hg)
CHF
C
4
22
9
(l)
(2)
(2)
(2)
(5)
(2)
(3)
(3)
(5)
(1)
73A
96
OA
IA
89
80A
lA
- A
IlA
2A
(6)
(3)
(2)
(2)
(7)
(2)
(2)
(2)
(3)
(2)
brachial arterial pressure; P0
mean
forearm
venous
pressure; RAP = mean right atrial pressure; LVFP
left ventricular filling
pressure.
Ap
932
< .02 vs supine values.
CIRCULATION
PATHOPHYSIOLOGY AND NATURAL HISTORY-CONGESTIVE HEART FAILURE
4c
I
C)
CL
o
G)
m
(n
1
3tC)
UcIL
i0
m
m
I-
1
0
_4
m
I-
0
X
1
Un
W
-L
_ 0
-1-
0
1
U)
1
CD
1
-
A
:(f
1
-4t
_r
_ m
1
1
in
rr
z
Un
0-r--j
14nr-rn
t"w%...
1
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
response (figure 3, A). However, this administration of
propranolol elicited a decrease in blood flow of 46 ±
5% (p < .02) corresponding to an increase in FMVR of
42 ± 12% (p < .02) during upright tilt (figures 2, B,
and 4, B). Despite this vasoconstrictor response, both
mean and systolic arterial pressures decreased from 95
± 6 and 128 ± 10 mm Hg in the supine position to 70
± 7 and 99 ± 9 mm Hg in the tilted position while
heart rate and arterial pulse pressure were maintained
at predrug values.
Brachioradial muscle vascular responses to upright
tilt were reproducible both before and after a single
dose of propranolol injected either intra-arterially (0.3
mg) or locally into the muscle (10 ng). Muscle blood
flow increased by 56 ± 10% corresponding to a decrease in FMVR of 48 ± 10% in the tilted position
before propranolol. Upright tilt after the brachial arterial or intramuscular propranolol injection elicited a
decrease in blood flow of 40 ± 5% or 46 ± 8%
corresponding to an increase in FMVR of 37 ± 4% or
51 ± 6%, respectively.
z
z
IT)
FIGURE 1. Mean relative change ± SEM in blood flow and FMVR
during 40 cm lowering of forearm and 45 degree upright tilt with
forearm at heart level in control subjects (hatched columns) and in
patients with CHF (open columns).
-
W
0
1
1
1
8]
A
(22.4X4.2)
X
7-'
(12.0±1.4)
..- - -
z
D
0
t
1.1)
tn
Ln
Lfl
ref2 0
9
---
(Q910
450
test
0o-'-
ref 1
10
9_
(12.7 ±1.5)
-
1174.±
74. 71
k 1.
.
B
(7.5± 0.9)
5-
I
I-
U
l
C0)
1
0
1
1
(21.8±1.9)
CY
1
1
b
A
i
1
so 4a3
|
FIGURE 2. Example of 133Xe-washout curves obtained in a patient with CHF before (left), during
(middle), and after (right) 45 degree upright tilt during control conditions (A) and after local /3-receptor
blockade (B), proximal nervous blockade ('A), and
proximal nervous blockade combined with local areceptor blockade ('B). Figures in parentheses denote the washout rate constant per minute (k, x 103
--+ 1 SD).
*
{'3.6)
|~~~~~~Q
% ^It
(12.6±2.3)
-
3
I
e^
-
'
i-
I
2]-
3.2)
100 200
--
*-._ -0--.')-.>**
(18.0±2.8)
100 200
|
(
(17.0±2.7)
100 200 300
Time (sec.)
Vol. 74, No. 5, November 1986
933
KASSIS et al.
W
<
X
fin >
c c,
_J
>
U)
X
40
40
>
m
_
;g- 20X
0
20
O
,
-
-
---11w
'B
m
m u)
mU)rxn
-20
>n
rn -
A B
1.
m
-2
1
-40-
Q.
o
-
-20-
c
u}
-40
3C
-m
-
FIGURE 3. Mean relative change ± SEM in blood flow and FMVR
caused by forearm lowering in patients with CHF during control conditions (A) and after local ,3-receptor blockade (B), proximal nervous
blockade ('A), and proximal nervous blockade combined with local areceptor blockade ('B).
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Effects of PNB. Half an hour after PNB of the limb,
motor paralysis, analgesia, and an increase in skin
temperature were present and lasted through the entire
experiment. Baseline values of heart rate and arterial
and venous pressures were preserved. No relative
change in muscle blood flow was elicited by a decrease
of 15 ± 1 mm Hg in perfusion pressure of the elevated
forearm; however, FMVR decreased by about 15%.
No substantial changes were observed in the increase
of vascular transmural pressure of the lowered forearm
or in the consequent vascular responses (figure 3, 'A).
PNB did not affect the control values of vascular pressures and heart rate during upright tilt. However, the
c
100X
100
U
z
1
G)
af 80-
00
3:
Q
iL 60
0
1
1
T
40
3
m
80
M
60
om
-20
O
-
-
-1-
A
-
2
fi
~0:
A
.-i
3-5
v
1
_, 20
U
)
m
40
W -20
vascular responses were reversed as muscle blood flow
decreased by 30 ± 5% (p < .02) corresponding to an
increase in FMVR of 30 + 7% (p < .02) (figures 2,
'A, and 4, 'A).
Effects of PNB combined with local a-receptor blockade.
Additional brachial arterial infusions of phentolamine
at 500 ,g/min to the neurally blocked limb elicited no
change in supine values of heart rate and vascular
pressures. The given dose of the drug completely abolished the vasoconstrictor response to venous stasis in
the lowered forearm as blood flow increased by 7 +
5% and FMVR decreased by 4 + 3% (figure 3, 'B).
The vasoconstriction in the tilted position was also
abolished as blood flow and FMVR decreased by 6 ±
5% and 1 ± 4%, respectively (figures 2, 'B, and 4,
'B).
Plasma concentrations of catecholamines (table 3). Supine values of venous plasma catecholamines were
within the normal range in control subjects; those for
epinephrine were maintained but those for norepinephrine increased during upright tilt. Patients with CHF
had augmented supine values of circulating catecholamines, which for epinephrine were significantly lower
in venous than in brachial arterial plasma. During upright tilt, patients with CHF had an attenuated increment in venous but not arterial plasma levels of norepinephrine; they had an increase in the venous plasma
levels of epinephrine and maintained the arterial levels. The tilt venous and arterial plasma epinephrine
concentrations were not substantially different.
Discussion
Circulating baseline levels of the neurotransmitter,
norepinephrine, and the adrenal hormone, epinephrine, were elevated in the patients with CHF. Despite
TABLE 3
Mean plasma levels of catecholamines (± SEM) during the supine
and tilted positions in control subjects (venous) and patients with
CHF (venous and arterial)
en
-20
c>
-40
2
Norepinephrine
(nmol/l)
Upright
Supine
tilt
Supine
Upright
tilt
0.93
(0.08)
2.70A
(0.19)
0.32
(0.04)
0.33
(0.03)
3.66
(0.74)
3.02
(0.60)
4.33
(0.59)
3.80A
(0.50)
0.82
(0. 18)
1.29B
(0.13)
0.99A
(0.20)
1.15
(0.12)
X
z -40
1
W
m
-60
I%
-60
-
Iq
-40
en
Control subjects
z
rn
Patients with CHF
Venous
n
FIGURE 4. Mean relative change + SEM in blood flow and FMVR
caused by upright tilt in patients with CHF during control conditions (A)
and after local /3-receptor blockade (B), proximal nervous blockade
('A), and proximal nervous blockade combined with local a-receptor
blockade ('B).
934
Epinephrine
(nmol/l)
Arterial
Ap
< .02, tilt vs supine.
Bp < .02, arterial vs venous.
CIRCULATION
PATHOPHYSIOLOGY AND NATURAL HISTORY-CONGESTIVE HEART FAILURE
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
augmented sympathetic efferent outflow to neuroeffector junctional clefts, which may facilitate further
neurohumoral excitation in CHF,"'5 arterial hypotension during simulated orthostatic stress has recently
been observed in a subset of patients with CHF`0 as it
was in our patients. The results of the present study
seem to indicate that the arterial hypotension is not a
primarily effect of upright tilt but is subsequent to a
vasodilator reflex arc intended probably to improve
cardiovascular performance in patients with CHF.
The cardiovascular response normally intended to
maintain arterial pressure during the gravitational
forces of upright tilt is mainly related to a reflex arc
originating from cardiopulmonary and arterial baroreceptors.' Evidence obtained so far suggests that digitalis at a therapeutic dose in man selectively augments
arterial baroreflex-mediated effects30 and augments
cardiopulmonary baroreflex-mediated vasoconstrictor
responses.3' Moreover, short-term administration of
digitalis may normalize vascular responses to orthostatic stress in patients with moderate-to-severe CHF.R0
Despite conventional therapy, our patients presented
classic symptoms and signs of severe CHF and had
abnormal responses to upright tilt. The patients studied
had therapeutic serum levels of digoxin and were not
hypovolemic as judged by the normal hematocrit and
electrolytes. These patients had dilated rather than
constricted resistance vessels of skeletal muscle during
the orthostatic stress.
The observed decreases in cardiac filling pressures
during upright tilt were comparable in the patients with
CHF and control subjects. Since these pressures were
elevated in the patients, cardiopulmonary baroreceptors might have been less sensitive to a pressure decrease from high to less high levels than to a fall to low
levels. Such a possibility may explain an attenuated
and not reversed vasoconstriction in the tilted position.
However, the patients and control subjects had comparable supine values of arterial pressure, but the patients
developed a state of arterial baroreceptor hypotension
during upright tilt and the expected circulatory adjustments failed to occur.", 16 30 Despite an apparently adequate level of reflex stimulus to the baroreceptors, the
responses observed in the tilted position were severely
deranged in these patients. Impairment of the cardiopulmonary and arterial baroreflex control in CHF has
been reported in human'0 and animal'6 studies.
Studies of the baroreflex afferent limb have disclosed degenerative changes in neurofibrillar end
plates of arterial baroreceptors in patients with CHF.32
Similar changes in cardiopulmonary baroreceptors
combined with abnormally reduced afferent activity
Vol. 74, No. 5, November 1986
from these receptors have been disclosed in animal
preparations of heart failure.'7 18 Abnormally reduced
baroreceptor afferent restraint on sympathetic efferent
outflow' may account for the augmented baseline plasma levels of catecholamines in our patients, but the
question is whether it explains the abnormal vasodilator response to upright tilt. Circulatory homeostasis is
a complex supraspinal integration of afferent impulses
from various receptor sites.' Microneurographic studies in normal human subjects have shown the differentiated sympathetic outflow to skeletal muscle and subcutaneous vascular beds to become uniform after
lidocaine blockade of vagus and glossopharyngeal
nerves in the neck.33 Such a temporary baroreceptor
deafferentation has also been shown to cause the appearance of an abnormal sympathetic reflex.33
We did not study the supraspinal restraint of baroreflex parasympathetic afferents but carefully investigated the sympathetic efferent limb effects on FMVR as
an effector organ. It seems difficult to attribute the
decrease in FMVR during upright tilt in the patients
with CHF to nonspecific depression of vascular reactivity.34 The vasodilatation observed in these patients
seems to be specific to the reflex stimulus of upright
tilt, since the patients had a normal vasoconstrictor
response to sympathetic reflex stimulation evoked by
forearm lowering.
Despite the increased sympathetic vasoconstrictor
activity, venous pressure elevation in the 40 cm lowered forearm elicited vasoconstriction in the patients
with CHF as in control subjects. With the '33Xe-washout technique, studies of different human tissues have
shown that a threshold increase in venous transmural
pressure of about 25 mm Hg or more elicits a-adrenergic arteriolar constriction by means of a local sympathetic axon mechanism.27 This vasoconstrictor axon
reflex has also been demonstrated during blockade of
central sympathetic outflow to the limb.6'7' 21 The vasoconstriction during venous pressure elevation has
been abolished after short-term denervation of the entire sympathetic nerve tree of the limb with use of PNB
combined with blockade of vascular a-adrenoceptors.' The observations in our patients are in agreement with those in normal human subjects27: the vasoconstrictor response to venous stasis in the lowered
forearm was preserved after PNB and was completely
abolished after additional intra-arterial infusions of
phentolamine.
The given dose of phentolamine induced no systemic effect and was within the range previously reported
to achieve vascular a-receptor blockade when given
intra-arterially to the human limb.6'3S PNB of the fore935
KASSIS et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
arm did not only abolish but reversed the vasodilator
response to upright tilt in our patients. Brachial arterial
levels of norepinephrine increased during upright tilt in
these patients, and the dose of phentolamine given to
the neurally blocked forearm completely abolished the
vasoconstriction in the tilted position. Taken together,
these observation suggest that the vasoconstrictor response to upright tilt after PNB is a humoral, a-norepinephric effect.
As observed during forearm elevation 20 cm above
heart level, the skeletal muscle vascular bed was not
maximally dilated after PNB. The fall in perfusion
pressure of the elevated forearm elicited a decrease in
FMVR while the blood flow was maintained at a constant level. These circulatory adjustments were of
similar magnitude and direction at baseline and after
PNB. The observations suggest that intrinsic vascular
mechanisms underlying autoregulation of blood flow
were operative in the patients with CHF and were
essentially unaffected by neural vasoconstrictor activity. Autoregulatory mechanisms might have supported
the vascular responses to upright tilt, since brachial
perfusion pressure decreased by about 12 mm Hg in
these patients (table 2). If this were the case, autoregulation of blood flow should also have been observed in
the tilted position after PNB.
Human studies have indicated that vascular ,3adrenoceptors may take part in regulation of skeletal
muscle blood flow in resting forearm during contralateral isometric handgrip.36 Animal studies have shown a
132-adrenergic mechanism operating in various vascular beds during the stress of hemorrhagic shock and
subserving improvement of the tissue perfusion and
cardiovascular performance.37 38 In the patients with
heightened sympathetic tone due to CHF, the sympathetic stimulation of the orthostatic stress might have
activated vascular /3-adrenoceptors. To test this
hypothesis, brachial arterial infusions of propranolol
were carried out and no systemic effects were elicited
by the drug. The infusate volume was at a rate shown
not to alter forearm blood flow,25 and the given dose of
propranolol was similar to that reported to achieve
vascular, 3-adrenoceptor blockade in the human limb.39
Although the drug preserved the a-adrenergic vasoconstrictor response to venous pressure elevation in the
lowered forearm, it elicited vasoconstriction in the tilted position and muscle blood flow decreased by 46%.
This vasoconstriction might have increased the concentration of propranolol delivered to the forearm during upright tilt, since the drug concentrations in blood
and tissue would depend on the blood flow. Therefore
a single intra-arterial dose of propranolol was given
936
and not only attenuated but also reversed the vasodilator response to upright tilt. Furthermore, a local injection of propranolol into the labeled area elicited vasoconstriction in the tilted position. A local anesthetic
action of propranolol (10 mg/liter) has been shown,40
whereas f3-adrenoceptor blocking properties of the
drug have been achieved at a lower plasma concentration of 0.1 mg/liter.41
Propranolol is not known to possess intrinsic sympathomemetic properties. Although it is not a selective
13-antagonist, its effect on the skeletal muscle vascular
bed in the present study seems most likely to be mediated by ,82-adrenoceptors on the blood vessel wall. The
consistent observations made during the propranolol
studies, together with those made during the blockade
of neural efferent impulses to the forearm, suggest that
a central neural mechanism is elicited by upright tilt to
modulate an apparently /2-adrenergic vasodilatation in
the patients with CHF.
The tilt-induced neurogenic vasodilatation in our
patients may involve a cotransmitter role for epinephrine. There is increasing evidence that circulating epinephrine is taken up into adrenergic nerve terminals,
stored, and eventually released with norepinephrine
into the junctional clefts on the blood vessel wall.424
In normal human subjects, a cotransmitter role for
epinephrine has been reported to augment neurogenic
vasoconstriction by activation of primarily prejunctional 032-adrenoceptors that facilitate the release of
norepinephrine.4 The patients with CHF had augmented circulating epinephrine levels that were higher in
arterial inflow than in venous effluents of the forearm.
Such an extraction of epinephrine in the supine resting
position may be ascribed to intact neuronal membrane
uptake system in the human limb.`5 This neuronal uptake is not operating during the time the neurotransmitter release is evoked by central neural impulses, a time
at which prejunctional adrenoceptors on the neuronal
membrane exert negative (a2) or positive (12) feedback
control of transmitter release.42 43
Our data provide no direct evidence for a cotransmitter release of epinephrine from forearm adrenergic
nerve terminals during upright tilt in the patients with
CHF. Despite the tilt-induced increment in muscle
blood flow that would tend to decrease the concentration of the blood-borne catecholamine, plasma epinephrine levels increased in venous outflow from the
forearm during upright tilt while they were almost
maintained in the arterial inflow in these patients. The
sympathetic stimulation of upright tilt in control subjects did not affect the normal plasma concentrations of
this adrenal hormone. It is tempting to speculate that
CIRCULATION
PATHOPHYSIOLOGY AND NATURAL HISTORY-CONGESTIVE HEART FAILURE
long-term exposure to augmented neurotransmitter vasoconstrictor activity in the patients with CHF may
increase the /3-adrenergic responsiveness of skeletal
muscle resistance vessels to catecholamines or that, in
contrast to its effect in normal subjects,44 epinephrine
at higher concentrations might exert predominantly
postjunctional rather than prejunctional /32 effects.
Circulatory homeostasis during the course of CHF
may apparently be maintained by a delicate balance
between a-adrenergic vasoconstrictor and /3-adrenergic vasodilator neural efferent pathways. The net
hemodynamic significance subjected to central integration of responses' seems to be determined by the
severity of impairment of the afferent restraint from
cardiopulmonary and arterial baroreceptors.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
We thank Ole Henriksen, M.D., Ph.D., for valuable discussions and constructive criticism and Bj0rn Larsen, M.Sc., for
very helpful technical support.
References
1. Abboud FM, Heistad DD, Mark AL, Schmid PG: Reflex control of
peripheral circulation. Prog Cardiovasc Dis 18: 371, 1976
2. Henriksen 0, Sejrsen P: Local reflex in microcirculation in human
cutaneous tissue. Acta Physiol Scand 98: 227, 1976
3. Henriksen 0, Sejrsen P: Local reflex in microcirculation in human
skeletal muscle. Acta Physiol Scand 99: 19, 1977
4. Henriksen 0, Sejrsen P: Effect of the vein-pump upon venous
pressure and blood flow in human subcutaneous tissue. Acta Physiol Scand 100: 14, 1977
5. Henriksen 0: Local sympathetic reflex mechanism in regulation of
blood flow in human subcutaneous adipose tissue. Acta Physiol
Scand 450(suppl): 7, 1977
6. Henriksen 0, Skagen K, Haxholdt 0, Dyrberg V: Contribution of
local blood flow regulation mechanisms to the maintenance of
arterial pressure in upright position during epidural blockade. Acta
Physiol Scand 118: 271, 1983
7. Skagen K, Haxholdt 0, Henriksen 0, Dyrberg V: Effects of spinal
sympathetic blockade upon postural changes of blood flow in human peripheral tissues. Acta Physiol Scand 114: 165, 1982
8. Brigden W, Sharpey-Schafer EP: Postural changes in peripheral
blood flow in cases with left heart failure. Clin Sci 9: 93, 1950
9. Goldsmith SR, Francis GS, Levine TB, Cohn JN: Regional blood
flow response to orthostasis in patients with congestive heart failure. J Am Coll Cardiol 1: 139, 1983
10. Ferguson DW, Abboud FM, Mark AL: Selective impairment of
baroreflex-mediated vasoconstrictor responses in patients with
ventricular dysfunction. Circulation 69: 451, 1984
11. Chidsey CA, Braunwald E, Morrow AG: Catecholamine excretion
and cardiac stores of norepinephrine in congestive heart failure.
Am J Med 39: 442, 1965
12. Cohn JN, Levine TB, Francis GS, Goldsmith SR: Neurohumoral
control mechanisms in congestive heart failure. Am Heart J 102:
509, 1981
13. Levine TB, Francis GS, Goldsmith SR, Simon AB, Cohn JN:
Activity of sympathetic nervous system and renin-angiotensin system assessed by plasma hormone levels and their relation to hemodynamic abnormalities in congestive heart failure. Am J Cardiol
49: 1659, 1982
14. Levine TB, Francis GS, Goldsmith SR, Cohn JN: The neurohumoral and hemodynamic response to orthostatic tilt in patients with
congestive heart failure. Circulation 67: 1070, 1983
15. Rose CP, Burgess JH, Cousineau D: Reduced aortocoronary sinus
extraction of epinephrine in patients with left ventricular failure
secondary to long-term pressor or volume overload. Circulation 68:
241, 1983
Vol. 74, No. 5, November 1986
16. Higgins CB, Vatner SF, Eckberg DL, Braunwald E: Alterations in
the baroreceptor reflex in conscious dogs with heart failure. J Clin
Invest 51: 715, 1972
17. Greenberg TT, Richmond WH, Stocking RA, Gupto PD, Meehan
JP, Henry JP: Impaired atrial receptor responses in dogs with heart
failure due to tricuspid insufficiency and pulmonary artery stenosis.
Circ Res 32: 424, 1973
18. Zucker IH, Earle AM, Gilmore JP: The mechanism of adaptation of
left atrial stretch receptors in dogs with chronic congestive heart
failure. J Clin Invest 60: 323, 1977
19. Bristow MR, Ginsburg R, Minobe W, Cubicciotti RS, Sageman
WS, Lurie K, Billingham ME, Harrison DC, Stinson ED: Decreased catecholamine sensitivity and /3-adrenergic receptor density in failing human hearts. N Engl J Med 307: 205, 1982
20. Lassen NA, Henriksen 0, Sejrsen P: Indicator methods for measurement of organ and tissue flow. In Abboud FM, Shepherd JT,
editors: Handbook of physiology III. The cardiovascular system.
Washington, DC, 1984, The American Physiological Society, p 21
21. Henriksen 0, Amtorp 0, Faris I, Agerskov K: Evidence for a local
sympathetic venoarteriolar "reflex" in the dog hindleg. Circ Res
52: 534, 1983
22. Wood JE, Litter J, Wilkins PW: Peripheral venoconstriction in
human congestive heart failure. Circulation 13: 524, 1956
23. Habak PA, Mark AL, Kioschos JM, McRaven DR, Abboud FM:
Effectiveness of congesting duffs ("rotating tourniquets") in patients with left heart failure. Circulation 50: 366, 1974
24. Tonnesen KH, Sejrsen P: Washout of 133xenon after intramusculV'
injection and direct measurement of blood flow in skeletal muscle.
Scand J Clin Lab Invest 25: 71, 1970
25. Eckstein JW, Wendleng MG, Abboud FM: Forearm venous responses to stimulation of adrenergic receptors. J Clin Invest 44:
1151, 1965
26. Kety SS: Theory and application of the exchange of inert gas at the
lungs and tissues. Pharmacol Rev 3: 1, 1951
27. Sejrsen P, T0nnesen KH: Shunting by diffusion of inert gas in
skeletal muscle. Acta Physiol Scand 86: 82, 1972
28. Lindbjerg IF, Andersen AM, Munck 0, J0rgensen M: The fat
content of leg muscle and its influence on the 133Xe clearance
method of blood flow measurement. Scand J Clin Lab Invest 18:
525, 1966
29. Engelman K, Portnoy B: A sensitive double-isotope derivative
assay for norepinephrine and epinephrine: normal resting plasma
levels. Circ Res 26: 53, 1970
30. Ferrari A, Gregorini L, Ferrari MC, Preti L, Mancia G: Digitalis
and baroreceptor reflexes in man. Circulation 63: 279, 1981
31. Imamura T, Takeshita A, Ashihara T, Yamamoto K, Hoka S,
Nakamura M: Digitalis-induced augmentation of cardiopulmonary
baroreflex control of forearm vascular resistance. Circulation 71:
11, 1985
32. Abrdham D: The structure of baroreceptors in pathological conditions in man. In Kezdi P, editor: Baroreceptors and hypertension.
Oxford 1967, Pergamon Press Ltd., p 273
33. Wallin BG: Functional organization of sympathetic outflow in
man. In Christensen NJ, Henriksen 0, Lassen NA, editors: Adrenergic physiology and pathophysiology. Copenhagen, 1985,
Munksgaard, p 9
34. Myers HA, Honig CR: Influence of initial resistance on magnitude
of response to vasomotor stimuli. Am J Physiol 216: 1429, 1969
35. Agerskov K, Henriksen 0: Effect of intra-arterially induced alphareceptor blockade upon resistance in collateral arteries in patients
with occlusion of the superficial femoral artery. Clin Physiol 1:
385, 1981
36. Eklund B, Kaijser L: Effect of regional and adrenergic blockade on
blood flow in the resting forearm during contralateral isometric
handgrip. J Physiol 262: 39, 1976
37. Gustafsson D, Lundvall J: 32-Adrenergic vascular control in hemorrhage and its influence on cardiac performance. Am J Physiol
246: H351, 1984
38. Gustafsson D, Andersson L, Maartensson L, Lundvall J: Microsphere analysis of /82-adrenergic control of resistance in different
vascular areas after hemorrhage. Acta Physiol Scand 121: 119,
1984
39. Agerskov K: Effect of propranolol on the tone of collateral arteries
937
KASSIS et al.
in patients with occlusion of the superficial femoral artery. Clin
Physiol 3: 43, 1983
40. Coltart DJ, Meldrum SJ: The effect of racemic propranolol, dextropropranolol and racemic practolol in the human and canine cardiac
trans-membrane action potential. Arch Intern Pharmacodyn 192:
188, 1971
41. Coltart DJ, Gibson DG, Shand DG: Plasma propranolol levels
associated with suppression of ventricular ectopic beats. Br Med J
1: 490, 1971
42. Vanhoutte PM, Verbeuren TJ, Webb RC: Local modulation of
adrenergic neuroeffector interaction in the blood vessel wall. Physiol Rev 61: 151, 1981
43. Vanhoutte PM: Alpha and beta adrenergic receptors and the cardiovascular system. J Cardiovasc Pharmacol 3(suppl 1): 1, 1981
44. Floras JS, Aylward PE, Mark AL, Abboud FM: Contrasting effects
of epinephrine and isoproterenol on neurogenic vasoconstriction:
evidence for a co-transmitter and facilitatory role for epinephrine in
man. Hypertension 7: 848, 1985
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Erratum
Ehsani AA, Biello DR, Schultz J, Sobel BE, Holloszy JO: Improvement of left ventricular contractile function by
exercise training in patients with coronary artery disease. Circulation 74: 350, 1986.
An error appeared in the abstract. The sentence beginning five lines from the bottom should have read: "Exerciseinduced regional wall motion improved in the training group."
938
CIRCULATION
Sympathetic reflex control of skeletal muscle blood flow in patients with congestive heart
failure: evidence for beta-adrenergic circulatory control.
E Kassis, T N Jacobsen, F Mogensen and O Amtorp
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. 1986;74:929-938
doi: 10.1161/01.CIR.74.5.929
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/74/5/929
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation 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 is online at:
http://circ.ahajournals.org//subscriptions/