469
Capillary Growth and Changes in Heart
Performance Induced by Chronic
Bradycardial Pacing in the Rabbit
ANDREW J.A. WRIGHT AND OLGA HUDLICKA
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
SUMMARY To determine whether chronic bradycardial pacing could change myoeardial capillary
density and heart performance, we used transvenous right atrial pacing to reduce the heart rate of
rabbits to about 66% of normal for up to 52 days. We compared control animals, animals which had
been chronically paced, and sham-operated animals. The chronically paced animals showed no cardiac
hypertrophy [heart;body weight ratio (H:BR) was unchanged]. All animals paced for longer than 10
days showed increased myoeardial capillary density (CD) when compared with controls of similar
heart weight. The increase in CD was correlated with the duration of pacing (r = 0.830) and reached
levels up to 70% greater than those of controls. The increase In stroke volume seen during acute
bradycardial pacing was more than twice as great in animals which had been chronically paced as in
the controls, which had not previously been paced (84% vs. 38%). Resting minute work and stroke work
were similar in all groups, but maximum values achieved after norepinephrine administration were
much greater in animals that had been chronically paced. Minute work increased by 59.1 ± 9.6% in
controls, by 82.1 ± 18.3% in sham-operated animals, and by 107.1 ± 8% (means ± SE) in paced animals.
Stroke work increased by 84.9 ± 9.8% in controls, by 75.4 ± 19.5% in sham-operated animals, and by
121.8 ± 7.8% in paced animals. Although the greater CD of most of the chronically paced hearts may be
partly responsible for their improved performance, this cannot be the only factor, since animals paced
for short periods showed improved performance but normal CD. Circ Res 49: 469-478, 1981
IT IS well known that heart rate decreases during
postnatal development. At the same time, capillary
density (capillaries/mm2) decreases too (Shipley et
al., 1937; Rakusan et al., 1965; Rakusan, 1971; Henquell et al., 1976) since the muscle fibers grow faster
than capillaries. There are, on the other hand, cases
in which a relatively lower heart rate is connected
with a higher capillary density: a hare, for instance,
has a considerably lower heart rate than a rabbit,
and has a higher myoeardial capillary density
(Poupa et al., 1970). Myoeardial capillary density
also has been reported to increase in chronically
exercised animals, and has been associated with
resting bradycardia (Frank, 1950; Hakkila, 1955;
Tomanek, 1970). We decided to find out whether
the changes in capillary density can be induced by
changes in heart rate, using chronic bradycardial
pacing. The heart rate of rabbits was decreased to
about hah0 of the original frequency by intravenous
right artial pacing for up to 52 days, and heart
performance and myoeardial capillary density were
compared with values for control or sham-operated
rabbits of a similar body weight.
From the Physiology Department, University of Birmingham, Medical
School, Birmingham, England.
This study wag supported in part by British Heart Foundation Grant
No. 808.
Address for reprints: Dr A.J.A. Wright, Physiology Department,
University of Birmuigham, Medical School. Vincent Drive, Birmingham.
B16 2TJ, England.
Received November 12, 1980; accepted for publication February 10,
1981
Methods
Experiments were performed on 63 New Zealand
Red rabbits of 1.65-3.4 kg body weight of either sex.
Out of these, 33 served as controls, 17 were chronically paced, and 13 were sham-operated.
Chronic Bradycardial Pacing
Pacing was achieved through use of right atrial
pacing by an external pacemaker in freely moving
animals for up to 52 days.
Implantation of Electrodes
A pair of ECG electrodes, and a pacing and
indifferent electrode were implanted in each rabbit.
The ECG electrodes were made from thin steel
discs 1 cm in diameter and drilled at three points
on the periphery to allow suturing. The pacing
electrode consisted of a bundle of ten 0.1-mm copper wires soldered to a short length of silver wire,
the tip of which had been formed into a sphere
about 1 mm in diameter by melting in a Bunsen
flame. The electrode was insulated by vinyl cannula
(1.5 mm external diameter) except at the tip. Silastoseal (Midland Silicones Ltd) in the cannula lumen
near the tip prevented any entry of blood. The
indifferent electrode was made by stripping the
insulation from the final 5 cm of a length of 10stranded wire (each strand 0.1 mm in diameter) and
forming the bared portion into a loop. The separate
strands were spread apart to maximize the conducting area available.
470
CIRCULATION RESEARCH
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
All the electrodes were sterilized by soaking overnight in Benzylkonium or formaldehyde in isopropyl alcohol, then washed by immersion in sterile
saline for at least 3 hours.
Under halothane anaesthesia, and aseptic conditions, the ECG electrodes were implanted under
the skin 2-3 cm on either side of the midline of the
thorax.
Their leads were externalized through a subcutaneous tunnel to an incision made between the
scapulae. A small two-pin plug was soldered to the
leads, which were connected by a shielded cable to
the recording system (a Devices ECG pre-amplifier
fed into either a Devices M2 two-channel pen recorder or a storage oscilloscope). The indifferent
electrode then was implanted under the skin in the
back of the animal. The pacing electrode was introduced into the right jugular vein, and both electrodes were connected to a Grass S8 stimulator.
While the ECG was monitored, the pacing electrode was advanced down the jugular vein until the
electrode was shown to be in the heart. This was
found when tachycardia was induced by a pacing
stimulus (pulses of 1 msec, at above 4V, 5-6/sec).
The usual final electrode position seen at autopsy
was in the lower part of the right atrium, well fixed
by fibrous tissue.
Once reliable pacing at a satisfactorily low voltage had been achieved, the pacing electrode was
tied in place at the neck and brought out through
the back incision by a subcutaneous tunnel. It was
sutured firmly to the skin and the position of the
indifferent electrode also was secured firmly. The
animal was fitted with a harness to protect the
leads. The pacing leads were connected to a small
two-pin socket, and this, together with the ECG
plug, was taped to the harness. The animal was
given the first of seven daily doses of antibiotic (18
mg of oxytetracycline hydrochloride. (EngemycirL,
Gist-Brocades N.V.).
Pacing Technique
The rabbits were kept in individual cages allowing full freedom of movement. Rabbit pellets and
water were available ad libitum. Two or 3 days after
the operation, when they were well recovered, they
were connected to an ECG recording system (Devices ECG pre-amplifier and M2 pen recorder). The
heart rate by this time had fallen from rather elevated levels immediately after the operation (~300
bpm) to normal or near-normal values (215-250
beats/min). To induce bradycardia, pacing impulses
1 msec long, at about 3 V, were applied to the right
atrial and indifferent electrodes by use of an external pacemaker connected to the animal by a flexible
lead. By using an appropriate frequency (which was
about 55% of the normal heart rate), we were able
to induce premature depolarizations within the mechanical refractory period of the heart; these premature depolarizations caused no contraction
themselves, but blocked every second endogenous
VOL. 49, No. 2, AUCUST
1981
depolarization and associated contraction (Fig. 1).
The heart rate would be exactly halved, but some
acceleration of the underlying endogenous rhythm
occurs during pacing, making the actual rate ~55%
of normal.
The animals were continuously paced for as close
as possible to 24 hrs/day, 7 days a week, within the
practical limitations of cage cleaning and repair of
connecting leads. The duration of the pacing period
was as long as 52 days.
Initially, the animals were connected continuously to the ECG, with regular display of the recorded activity. It soon become apparent, however,
that the fact of pacing could be confirmed easily
merely by palpating the animal's thorax, since the
slowed heart beat was considerably more vigorous
than normal. In later experiments, the ECG was
checked only every few days, largely to determine
the heart rate when the pacing was stopped briefly.
Because the endogenous heart rate changed over
the course of study, it was necessary, during at least
the first weeks of pacing, to adjust the pacing frequency slightly every 2 or 3 days, using the ECG
and/or palpation.
Sham-operated rabbits had ECG and pacing electrodes implanted but were not paced.
Studies on Heart Performance
Acute experiments were performed in order to
assess changes in heart performance and blood pressure in chronically paced animals as compared to
control and sham-operated ones.
All animals were anaesthetized with sodium pentobarbitone (Sagatal, May & Baker Ltd.) 35 mg/kg,
iv, supplemented as necessary. Cannulae were inserted into the jugular vein for administration of
drugs and dye, into the femoral artery for blood
pressure recording, and a tracheal cannula also was
inserted. The animals were allowed to breathe spontaneously. Rectal temperature was maintained in
the range 37.5-38.5C.
Heart rate was monitored from the blood pressure record. Cardiac output was determined by use
of standard indicator-dilution techniques, either the
dye dilution or the thermodilution method. The
former method involved bolus injection of 0.5 ml
LVP
loo
imraHg)
r
I
1 sec
ECG
'
»
t
t
t
t
t
t
t
i
i
i
t
t
FIGURE 1 Bradycardial pacing. First part of recordnormal; second part—bradycardial pacing. Stimulus
artifacts are shown by the arrows. Once pacing is established, spikes on the left of the stimulus artifacts are
endogenous R-waves, spikes on the right are induced Rwaves. Alternate endogenous R-waves are blocked—e.g.,
at point A.
EFFECTS OF CHRONIC BRADYCARDIAL PACING/ Wright and Hudlicka
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
indocyanin green (Cardiogreen, Hynson, Westcott
& Dunning Inc.) into the jugular vein with blood
sampling from the left femoral artery by use of a
purpose-built sample collector. The dye-dilution
curve for whole blood was plotted by use of spectrophotometric data obtained on a Unicam SP1800
spectrophotometer from the plasma of the sampled
blood; allowance was made for the hematocrit (measured during every determination).
As an index of heart performance we used the
maximum minute work and stroke work values
obtainable in each animal during the positive inotropic effects of increasing doses of norepinephrine.
(From 3-7 doses of noradrenaline in the range of 110 jug/kg were given iv to each animal. The dose
was increased until the minute work began to decline, and the response to the previous dose was
then taken to be maximal). Minute work was calculated as cardiac output times mean blood pressure, and stroke work was obtained by dividing
minute work by the heart rate. Both were standardized per kilogram body weight, and expressed as
joules/min per kg and joules/beat per kg. Stroke
and minute work seemed to be more sensitive indices of heart performance in these experiments
than, e.g. (dp/dt) max.
When the thermodilution method was employed,
bolus 0.7-ml injections of room temperature
(~20°C) saline were made into the cavity of the left
ventricle via a cannula introduced via the left carotid artery. Blood temperature was monitored in
the descending aorta by use of a purpose-built cannula which incorporated an ITT U23US thermistor.
The off-balance voltage of the Wheatstone bridge
circuit in which the thermistor was included was
amplified by a Devices sub 1C preamplifier for
display by a Rikadenki-Mitsui pen recorder. For
both dye and thermodilution methods, correction
was made for recirculation by extrapolation of the
exponential part of the dilution curve downslope.
For the dye-dilution method, cardiac output was
calculated from CO. (ml/min) = (D/A) where D
= dye injected in mg. A = curve area in mg/ml per
min.
For the thermodilution method, CO. was calculated according to the method of Beillin and Bhattacharya (1977). Both methods gave similar values
for CO.
At the end of each experiment, the rabbits were
killed by a pentobarbitone overdose. The hearts
were excised, rinsed in saline, and firmly blotted.
The vessels were trimmed flush, and any large
deposits of fat were removed, before weighing to
the nearest 0.05 g.
Estimation of Capillary Density
Capillary density was estimated using Gomori's
(1939) calcium cobalt stain for alkaline phosphatase
which specifically stains capillary endothelium and
thus reveals all the capillaries present in the heart
muscle. The hearts were excised within a few min-
471
utes of the death of the animal and stored in cool
isotonic saline. In the first trials of the alkaline
phosphatase technique, capillaries were found to be
equally well demonstrated with immediate and
somewhat delayed (up to 5 hours postmortem) sampling and freezing. The hearts underwent a slow
postmortem contraction, which was apparently
complete within 1 hour. Since the degree of contraction of the myocardium will alter the measured
capillary density, it is important that all the hearts
be contracted similarly at sampling. All hearts from
which capillary counts were used were therefore
sampled between 1 and 2 hours after excision.
Tissue samples were taken from two fixed locations, one of which gave good cross-sections of the
subendocardium, and the other which gave good
cross-sections of the subepicardium. Samples were
frozen in isopentane cooled in liquid nitrogen. By
use of a cryostat (SLEE), 12-/un sections were cut
from each sample and stored at -20°C for staining
at leisure. Capillary density was determined for
each heart from counts made in 20 fields, and this
gave a total observed area of 1.13 mm2, containing
1000-4000 capillaries. Only areas cut in good crosssection were counted. Statistical evaluation was
done on the basis of unpaired Student's t-teets,.
Results
Resting Bradycardia
All animals in which ECG electrodes were implanted showed a postoperative reduction in heart
rate over the course of 2-5 days. In addition, the
bradycardially paced animals showed a more prolonged and gradual decline in the heart rate (observed when the pacing was temporarily turned
off). By the end of the pacing period, the animals
paced for 10 days or longer had a mean heart rate
of 190 ± 3.3 (SE) beats/min vs. 209 ± 3.6 beats/min
for sham-operated animals (paced, n = 12; shamoperated, n = 7; P < 0.01 by Student's t-test). The
decline in heart rate developed gradually: We compared the heart rate of the chronically paced animals on the day prior to death with the heart rate
5 days after the operation for electrode implantation (when the initial rapid decline in heart rate
had leveled off) and calculated the percentage fall
in heart rate. When the degTee of bradycardia was
plotted against the intervening period of pacing,
there was a linear negative correlation between the
percentage of decreased heart rate and the period
of pacing up to about 25 days (r = —0.745, P <
0.01). However, no further decrease in resting heart
rate was observed in three animals paced for a
longer period.
Heart Weight-to-Body Weight Relationship
The weight range of animals subjected to longterm pacing was 2.15-3.30 kg. Animal growth had
continued normally during the pacing period. To
investigate the possible occurrence of cardiac hy-
CIRCULATION RESEARCH
472
pertrophy in the paced animals, we compared the
heart weight-to-body weight relationship of this
group with that of a group of 47 control animals of
the same body weight range. As shown in Figure 2,
the relationships were similar; the mean heart
weight to body weight percentage was 0.241 in the
paced animals vs. 0.242 in the controls. [Dry weight
as a proportion of wet weight was determined for
some control and chronically paced hearts, and was
found to be reasonably constant (20.4-23.5%)].
Thus no hypertrophy had occurred due to the pacing.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Capillary Density
We found no consistent differences in capillary
density between our subendocardial and subepicardial samples. (The mean values differed by 3% in
control animals and 1% in chronically paced animals). We therefore used the mean capillary density
found at both sites in a given heart for further
analysis. There was considerable variation in capillary density in the control hearts, much of which
was related to variations in heart weight. As shown
in Figure 3, we found a good inverse relationship
between capillary density and heart weight in shamoperated and unoperated control animals, with
heart weights of 4.2-8.0 g. Hearts heavier than 8.0
g did not show any lower capillary density and were
10-
H«rt
Wetg
(0)
8-
6-
1
1
1
2
3
A
Body
Weight (kg)
FIGURE 2 Heart weight vs. body weight for chronically
paced animals. The continuous regression line (r =
0.783) w for chronically paced animals (M). The dotted
regression line (r — 0.671) is for control animals (data
not shown) of the same body weight range (2.15-3.30 kg.).
Data from sham-operated animals are also shown (O).
VOL. 49, No. 2, AUGUST 1981
therefore treated separately. The correlation coefficient for capillary density against heart weight for
hearts of less than 8.0 g is — 0.829 (significant at P
< 0.0001) whereas, for the hearts of more than 8.0
g, it is -0.063 (nonsignificant).
Capillary density found in the chronically paced
animals plotted against heart weight is shown in
Figure 4, in relation to the regression lines representing the control animals. The duration of pacing
in days is shown next to each of the appropriate
data points. It can be seen that most of the paced
hearts were above the regression line obtained for
control hearts, and some showed a considerably
higher capillary density than predicted for control
hearts of the same weight. The mean change in
capillary density (compared with the regression
line) was + 33.8 ± 5.7% (SE) for the 13 animals
paced for more than 10 days and -3.3 ± 1.8% for
the 13 sham-operated animals. This is statistically
highly significant (P < 0.001).
Figure 5 shows a section from a heart which had
been paced for 52 daya, compared with a section
from a control heart of the same weight. The capillaries appear in cross-section as dark spots when
stained for alkaline phosphatase. The higher capillary density in the chronically paced heart is obvious. When the expected capillary density of control hearts, read from the regression line in Figures
3 and 4, was taken as 100%, we could evaluate the
capillary density differences shown by the paced
hearts. This is done in Figure 6; the result shows
clearly that the deviation of the capillary density
observed in the paced hearts from the value expected on the basis of the control regression lines is
related to pacing duration. The correlation coefficient of 0.83 is highly significant (P < 0.001). Obviously, animals paced for longer periods of time
show greater differences from the expected capillary
density. The four animals paced for 10 days or less
showed no significant change in capillary density.
Heart Performance
Effect of Acute Bradycardial Pacing on Heart
Performance in Control and Chronically Paced
Animals
Chronic pacing was usually discontinued 1-3
days before acute experiments. In a small group of
the chronically paced animals, bradycardial pacing
was reinstituted in acute experiments. The results
were compared with those from control animals
which were being paced for the first time. The
initial heart rate, mean blood pressure, and stroke
volume values were not significantly different between the groups. As Table 1 shows, bradycardial
pacing caused similar great reductions in heart rate,
in both groups, and similar minor changes in blood
pressure. Stroke volume, however, increased by
more than twice as much in the animals that had
been chronically paced than in controls. Greater
ability to increase stroke volume at low heart rates
EFFECTS OF CHRONIC BRADYCARDIAL PACING/ Wright and Hudhcka
473
3000
FIGURE 3 Capillary density us.
heart weight for non-paced animals. The left-hand regression line
is for hearts of less than 8.0 g; it
has r of -0.829 (P < 0.001). The
right hand regression line is for
hearts of more than 8.0 g.; it has r
of -0.063 (NS). (•) = control animals, (D) = sham-operated animals.
2000
C D. t o p t / « c r |
may be of value in maintaining normal cardiac
output during chronic bradycardial pacing.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Heart Performance during Norepinephrine
Administration
Values of heart rate, cardiac output, stroke volume, and mean blood pressure in control, bradycardially paced, and sham-operated animals are shown
in Table 2. Resting values are compared with the
values observed following the dose of noradrenaline
that produced maximum minute work and stroke
work in a particular animal (4,5-9.0 ^g/kg). In each
animal the change after norepinephrine was expressed as a percentage of resting value, and the
means of the percentages are also shown.
The relatively low resting heart rate seen in the
chronically paced animals prior to the acute experiment was no longer evident under anesthesia; their
heart rates were similar to those of the normal
control animals. The sham-operated animals
showed a significantly higher resting heart rate than
the controls. Although the resting cardiac output
and stroke volume were highest in the chronically
paced animals, the differences between the groups
were not statistically significant. The resting mean
5
I
«m,.^¥
8
blood pressure was significantly higher in the shamoperated group than in the paced group.
After the administration of norepinephrine, heart
rate dropped in 27 out of 35 animals; there were no
differences in heart rates between the groups. The
t
V
t t.
9
4 Capillary density us. heart weight for chronically paced animals. The regression lines are for nonpaced animals (not shown) and are the same as in
Figure 3; the number next to each point shows the
duration of pacing.
FIGURE
FIGURE 5 Heart sections stained for alkaline phosphatase. Upper section—control; lower section—chronically
paced (52 days). The hearts were very similar in weight.
474
CIRCULATION RESEARCH
VOL. 49, No. 2, AUGUST 1981
Discussion
Chronic bradycardial pacing in rabbits produced,
(1) a decreased heart rate when pacing was turned
off, (2) no heart hypertrophy as shown by heart to
body weight ratio, (3) increased capillary density in
hearts paced for at least 10 days, which was correlated with pacing duration up to 52 days, and (4)
increased stroke work and minute work in response
to norepinephrine.
20
30
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
FIGURE 6 Percentage capillary density increase vs.
pacing duration. The regression line (r = 0.830, P <
0.001) is for the chronically paced animals and eight
sham-operated animals (represented by one point—
mean ± SD—at zero days pacing duration).
mean cardiac output of the paced group increased
more than that of the other groups, and reached
the greatest absolute value. The mean stroke volume of this group also increased more than that of
the other groups, and was significantly higher. The
increases in mean blood pressure and the absolute
levels attained were similar in all groups.
The cardiac stroke work before and after norepinephrine is shown in Figure 7. Cardiac minute work
at rest was 1.69 ± 0.13 g/kg in control rabbits, 1.92
± 0.13 g/kg in sham-operated rabbits, and 1.81 ±
0.14 g/kg in chronically paced rabbits. It increased
to 2.62 ± 0.18, 3.02 ± 0.37, and 3.77 ± 0.34 g/kg,
respectively, after norepinephrine. The resting values were similar in all three groups, but the increase
in minute and stroke work in the paced group was
considerably greater. This increase resulted from
greater cardiac output and stroke volume being
expelled by these hearts against similar mean blood
pressures.
Perhaps surprisingly, the performance (maximum percentage increase in minute work or stroke
work) was not significantly correlated with pacing
duration (P > 0.30) (Fig. 8); even animals paced for
10 days or less showed performance above that of
controls. This was in distinct contrast to the capillary density changes, which were only seen after 10
days or more of pacing.
TABLE
1 Hemodynomic Changes with Acute Pacing
Percentage change in
Group
n
Heart rate
Control
5
-44.6
±1.1
Chronically paced
7
-44.2
±1 1
Mean BP Stroke volume
-6.2
±2.4
-3.4
±1.3
Values are means ± S E n » number of aruraala
* P < 0.05 vs. control group
+38 2
±11.5
+64.0*
±13.5
Resting Bradycardia
The earliest suggestion of developing differences
between the chronically bradycaxdially paced rabbits and the control animals was the gradual development of resting bradycardia (observed when the
pacing was interrupted) in the paced animals. The
reduction in heart rate below rates observed earlier
in the same rabbits was usually between 9% and
15% when pacing had been employed for more than
2 weeks. Resting bradycardia has been reported
after prolonged physical training. Frank (1950),
Hakkila (1955), and Tomanek (1970) described
mean heart rate reductions of 15%, 18%, and 9%,
respectively, results very similar to ours. The bradycardia of physical training has been shown to be
due to increased vagal tone by Frick et al. (1967). It
is possible that the bradycardia observed in the
current study has the same mechanism, since the
heart rates of chronically paced and control rabbits
were similar under barbiturate anesthesia. Under
these conditions, vagal tone is reduced or abolished
(Halinen et al., 1978) and the heart rate rises well
above the resting level (from -210 to ~280).
Heart Weight-to-Body Weight Relationship
Heart hypertrophy in our experiments was
judged on the basis of heart:body weight ratio. We
used only wet weight since, in a few experiments,
dry weight did not indicate any changes in water
content. We can compare our results with similar
data obtained in other studies in which wet weights
were used as well. Most procedures which have
produced bradycardia (exercise, heart block) and
thus increased stroke volume to maintain normal
cardiac output, also cause an increase in H:B ratio.
Brockman (1965) reported an increase of 10—40% in
H:BR in dogs in which heart rates were controlled
by the ventricular rhythm for up to 14 months and
Turina et al. (1969) saw a 26% increase in dogs in
which heart rates had been approximately halved
for 2-3 months. Bloor and Leon (1970) described an
increase in H:BR of 36% in rats exercised daily.
Swimming produced heart hypertrophy in rats
(Carlsson et al., 1978). To our knowledge, the only
study where no significant change in H:BR was
found while there was an increase in C:F ratio was
that by Leon and Bloor (1968) on rats exercised by
swimming twice a week for 10 weeks. It is therefore
rather surprising that no obvious hypertrophy was
found in hearts with such a dramatic decrease in
EFFECTS OF CHRONIC BRADYCARDIAL PACING/ Wright and Hudlicka
475
TABLE 2 Hemodynamic Effects of Norepinephrine Administration
Heart rate (beats/min)
Groups
Control
n = 10
Paced
n - 14
Sham-operated
n = 8
Rest
NE
Change
Rest
NE
Change
281
±6.3
288
±9.4
316*
±14.4
269
±110
268
±9.4
285
±17.4
-4.0
±4.2
-7.0
±4.4
-9.1
±17.4
141.9
±7.2
165.5
±9.7
152.3
±11.1
162.8
±6.3
234.2'
±19.1
185 9
±23.0
+ 14.4
±4.7
+41.4"
±8.3
+22.2
±10.8
' P <0.05 vs. C,
<0.10>0.05 vs. P
SIG.
Cardiac output (ml/min per kg)
*P <0.01 vs. C
" P <0.O2vs.C
Stroke volume (ml/beat per kg)
Rest
NE
0.505
0.605
±0.032 ±0.033
0 575
0.871*
±0 031 ±0.062
0 487
0.657
±0.040 ±0.073
•P
<
<
"P <
<
* * *P <
Mean hlood pressure (mm Hg)
Change
Rest
NE
Change
+22.4'*'
±8.3
+53.2"
±10.4
+32.7*"
±13.5
86.5
±4.7
82.1
±3.9
95.9*
±5.3
119.5
±7.7
119.9
±4.2
126.9
±8.7
+40 6 "
±9.7
+49 1 "
±5.3
+33 5 "
±9.0
0.01 vs C,
0.05 vs. SO
0.05 vs. C,
0.01 vs. resting
0.05 vs resting
*P <0.05i vs. P
• • P <0.01 vs rest
n — number of animals; NE - norepinephnne; SIG. ™ significances, values are means ± SE.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
heart rate, and increase in stroke volume. We did
not measure fiber diameters, since the lack of section shrinkage associated with our histological techniques did not cause sufficient separation of muscle
fibers, but they seem to be considerably smaller
(Fig. 5.) than in hearts of control animals. A similar
finding—i.e., a decrease in mean fiber diameter—
was reported by Brown et al. (1976) in skeletal
muscles subjected to an intensive work load by
CONTROL
0.015
CHRONICALLY
PACED
SHAM
OPERATED
n
0.010 STROKE
WORK
electrical stimulation, and by Pette et al. (1975) who
suggested the possibility of fiber splitting. It is
possible that a drastic reduction in heart rate and
a great increase in stroke volume represent a very
big load which might cause splitting of myocardial
fibers, but this requires further studies.
Capillary Density
Myocardial capillary density has been studied
extensively in the rat; unfortunately, very few results on rabbits have been reported. Shipley et al.
(1937) found a CD. of 3310 caps/mnA This is
substantially higher than the control values of 14002800 caps/mm2 found in the present study. Some of
this difference may be due to shrinkage of the
•f
Q.0O5 -
150 -
MAXIMUM
STROKE WORK
% INCREASE
( n • 10)
I n -141
I n- 8 )
7 Cardiac stroke work changes after norepinephrine. Upper—plain columns show stroke work at
rest, striped columns show maximum stroke work after
norepinephrine administration; lower—maximum percentage increases in stroke work after norepinephrine.
*P < 0.01 vs. control; P > 0.10, < 0.05 vs sham-operated.
**P < 0.001 vs. control P < 0.02 us. sham-operated, n =
number of animals.
FIGURE
10
20
30
40
50
PACING DURATION IDAYSl
FIGURE 8 Maximum percentage increase in stroke
work vs. pacing duration. The striped bar shows the
mean ± SE for 18 control and sham-operated animals.
476
CIRCULATION RESEARCH
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
tissue, which was extreme with many of the histological methods used in older studies. However,
Shipley et al. (1937) attempted to correct for shrinkage in their study. Poupa et al. (1970) also obtained
a value for CD. which was higher than that found
in our study: 3180 caps/mm2. The major reason for
these differences is likely to be that Poupa et al.
and Shipley et al. cut their hearts at diastole, rather
than in a state of contraction, as in the present
study. This factor does not seem to have received
sufficient attention, and it may account for some of
the variation in the myocardial capillary density
values reported in the literature. For example, a
30% shortening of muscle fiber length due to partial
contraction would be expected to produce a 30%
reduction in CD., due to an increase in the crosssectional area of the muscle fibers. A difference in
length of this degree is quite plausible, and it would
explain the differing results. Gray and Renkin
(1978) showed that contraction of skeletal muscle
could cause muscle fiber areas to increase by as
much as 80%.
Another reason for the discrepancy between our
results and those of Shipley et al. (1937) and Poupa
et al. (1970) might be the fact that the values in the
above-mentioned papers were averaged, rather
than related to a particular heart weight. This is
obviously of great importance, as shown in Figure
3. There was a clear inverse relationship between
the heart weight and capillary density in hearts of
4.15-8 g. A similar conclusion was made by Rakusan
et al. (1967) on the basis of indirect measurements
of capillary density based on the blood content in
small vessels, and capillary diameter.
The extent of the decline in capillary density in
our experiments over the range of heart weights
studied was even greater than expected from considerations of increasing muscle fiber area during
growth, if capillary proliferation does not occur.
The observations strongly support the hypothesis
that capillary growth does not occur in the normal
adult rabbit heart of less than 8.0 g. The rabbit
heart would resemble the rat heart in this respect,
since this species also shows a decline in capillary
density with increasing heart size (Rakusan et al.,
1965). Moderate reduction in capillary density need
not be any disadvantage to the heart, since Henquell et al. (1977) studied the distance between
open capillaries in the subepicardial layers of the
rat heart in vivo, and found a similar ability to
reduce it by capillary recruitment in large and small
hearts, although the total capillary density of the
smaller hearts was greater. These authors calculate
that larger hearts require less open capillaries per
unit area to maintain a given oxygen tension in the
myocardium, because oxygen consumption per
gram of heart drops with size.
In the heaviest hearts in this study (hearts over
8 g) capillary density was relatively constant and
not significantly related to heart weight (r = 0.063).
VOL. 49, No. 2, AUGUST 1981
Because capillary density would be expected to
decline increasingly slowly with increasing heart
weight, it is quite possible that our data would not
reveal a continued gradual decline in capillary density. True constancy of capillary density with increasing heart weight would imply either capillary
proliferation, or alternatively lack of change in muscle fiber area with heart growth (Le., lengthening
only of muscle fibers). With regard to the latter
point, Korecky and Rakusan (1973) found symmetrical rather than linear growth in muscle fibers
of the rat heart, but the possibility of asymmetrical
growth in rabbit muscle fibers cannot be dismissed.
Chronically Paced Animals
The situation apparent in the control animals, in
which myocardial capillary density declined gradually with increasing heart weight over most of the
weight range observed, was completely altered
when the activity pattern of the heart was transformed by bradycardial pacing. The myocardial
CD. of animals bradycardially paced for more than
10 days increased, often markedly, to an extent that
was correlated with the duration of pacing up to 52
days.
With bradycardial pacing, some of the animals
paced for about 3 weeks showed changes in CD. as
great as those observed in rabbits paced for more
than twice as long. It is possible that the greatest
changes observed (50-70%) represent the maximum
change possible with this technique, with different
animals reaching this level with greater or lesser
times of pacing. Capillary density was, for instance,
considerably increased in a rabbit bradycardially
paced for 14 days, but pacing durations of 10 days
or less produced no significant change, and the
increase in some rabbits paced for up to 20 days
was rather small. This delay before an increased
CD. is observed is longer than that found in skeletal
muscle undergoing chronic electrical stimulation (4
days, according to Brown et al., 1976). Earlier investigators reporting evidence for capillary growth
in the heart have applied a stimulus for longer
periods, so the delay before the apparent growth of
new vessels has not been determined previously.
The causes for increased capillary density are not
clear. Unge et al. (1979) found proliferation of capillaries in the heart in rats subjected to swimming
or chronic administration of dipyridamole. Their
results would suggest that capillary growth might
be connected with increased coronary blood flow—
a suggestion which might equally apply to bradycardial pacing, since coronary blood flow is much
higher in diastole than in systole. Increased blood
flow was suggested as one possible factor involved
in capillary growth in subcutaneous tissue by Branemark (1965) and in electrically stimulated skeletal muscles by Hudlicka and Schroeder (1978) and
Myrhage and Hudlicka (1978), but further experiments are necessary to find out whether it is in-
EFFECTS OF CHRONIC BRADYCARDIAL PACING/WrightandHudlicka
volved in capillary growth in bradycardially paced
hearts.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Heart Performance
The low resting heart rate evident in the chronically paced animals before the acute experiment
did not persist under anesthesia; however, since the
sham-operated group showed a significantly higher
resting heart rate than the controls, it is possible
that the introduction of a cannula into the right
atrium has had some effect on the resting heart rate
of this group; if so, the relatively lower heart rate of
the chronically paced animals would indicate that
chronic pacing had negated some tendency for electrode implantation alone to raise resting heart rate.
Similarly, because the resting mean blood pressure
was significantly higher in the sham-operated group
than in the paced group, it may be possible that the
introduction of a pacing electrode per se has some
effect on blood pressure, which is neutralized by
pacing through that electrode.
Bradycardially paced hearts were different from
control hearts in two respects: their stroke volume
was obviously considerably higher, since cardiac
output and blood pressure were maintained. On
transition from the resting state under anesthesia
to acute pacing, the increase in stroke volume was
almost twice as great as the increase in stroke
volume in control rabbits (Table 1). Further, the
ability of chronically paced animals to increase
stroke work and minute work in response to a
similar dose of norepinephrine was considerably
greater than in control rabbits. Norepinephrine was
chosen because it produced the greater increase in
the stroke and minute work than either epinephrine
or isoproterenol.
When the maximum stroke work percentage increase observed in each heart is plotted against
pacing duration (Fig. 8) it is obvious that this factor
increases rapidly during the first 2-3 weeks of pacing. Comparison with the data for increases in capillary density reveals that changes in capillary density were much slower; there was in fact no significant correlation of the ability to increase stroke
work (or minute work) with changes in capillary
density (r < 0.20, P > 0.50), among the paced
animals. At least moderate increases in performance were possible with no increase in capillary
density. The factor (or factors) which limits performance with norepinephrine is not clear at present;
whatever it is, it has obviously changed in the
chronically bradycardially paced hearts. That performance changes apparently unconnected with the
increased capillary density can occur emphazises
the multi-faceted nature of the heart's ability to
adapt to chronic changes in activity. Of course, this
is not to say that the greater capillary density of
the paced hearts, or other changes in their vasculature, might not be more closely related to performance if performance were assessed differently.
All
References
Beillin, LJ, Bhattacharya AJ (1977) The effect of prosUglandin
synthesis inhibitors on renal blood flow distribution in conscious rabbits. J Physiol (Lond) 269: 395-405
Bloor CM, Leon AS (1970) Interaction of age and exercise on
the heart and its blood supply Lab Invest 22: 160-165
Branemark P-I (1965) Capillary form and function. The microcirculation of granulation tissue Bibl Anat 7: 9-28
Brockman SK (1965) Cardiodynamics of complete heart block.
Am J Cardiol 16: 72-83
Brown MD, Cotter MA, Hudlicka O, Vrbova G (1976) The
effects of different patterns of muscle activity on capillary
density, mechanical properties and structure of slow and fast
rabbit muscles. Pfluegers Arch 361: 241-250
Carlsson S, Ljungqviit A, Tomling G, Unge G (1978) The
reaction of the va3cular pattern of the hypertrophied myocardium to increased cardiac volume load. A mtcroangiographical
3tudy. Acta Pathol Microbiol Scand (A) 86: 297-301
Frank A (1950) Experunentelle Herzhypertrophie. Z Gesamte
Exp Med 115: 312-349
Fnck MH, Elovaimo RO, Somer T (1967) The mechanism of
bradycardia evoked by physical training. Cardiologia 51: 4654
Gomori G (1939) Microtechnical demonstration of phosphatase
in tissue sections. Proc Soc Exp Biol Med 42: 23-26
Gray SD, Renkin EM (1978) Microvascular supply in relation to
fiber metabolic tvpe in mixed skeletal muscles of rabbits
Microvasc Res 16": 406-425
Hakkila J (1955) Studies on the myocardial capillary concentration in cardiac hypertrophy due to training. Ann Med Exp
Biol Fenn 33 (suppl 10): 1-82
Halinen MO, Hakumaki MOK, Sarajas HSS (1978) Suppression
of autonomic post-ganglionic discharges by pentobarbital in
dogs, with and without endotoxemia. Acta Physiol Scand 104:
167-174
Henquell L, Odoroff CL, Honig CR (1976) Coronary mtercapillary distance during growth- Relation to Pt 0 2 and aerobic
capacity. Am J Physiol 231: 1852-1859
Henquell L, Odoroff CL, Honig CR (1977) Intercapillary distance
and capillary reserve in hypertrophied rat hearts beating in
situ. CLrc Res 41: 400-4O8
Hudlicka O, Schroeder W (1978) Factors involved in capillary
growth in a normal adult skeletal muscle (abstrf. Fed Proc 37:
314
Korecky B, Rakusan K (1973) Growth of cardiac muscle cells
during the life span of a rat (abstr). Circulation 46: 49
Leon AS, Bloor CM (1966) Effects of exercise and its cessation
on the heart and its blood supply. J Appl Physiol 24: 485-490
Myrhage R, Hudlicka O (1978) Capillary growth in chronically
stimulated adult skeletal muscle as studied by intravital microscopy and histological methods in rabbits and rats. Microvase Res 16: 73-90
Pette D, Ramirez BU, Wuller W, Simon R, Exner GU, Hildebrand R (1975) Influence of intermittent long-term stimulation
on contractile, histochemical and metabolic properties of fibre
populations in fast and slow rabbit muscles. Pfluegers Arch
361: 1-7
Poupa O, Rakusan K, Ostadal B (1970) The effect of physical
activity upon the heart of vertebrates. Medicine and Sport,
vol 4. Basel/New York^pp 202-233
Rakusan K (1971) Quantitative morphology of capillaries of the
heart. Number of capillaries in animal and human hearts
under normal and pathological conditions. Methods Achiev
Eip Pathol 5: 272-286
Rakusan K, Jelinek J, Korecky B, Soukupova M, Poupa 0
(1965) Postnatal development of muscle fibres and capillaries
in the rat heart. Physiol Bohemoslov 14: 32-37
Rakusan K, de Rochemont W, Braash W, Tschopp N (1967)
Capacity of the terminal vascular bed during normal growth,
in cardiomegaly, and in cardiac atrophy. Circ Res 21: 209-215
Shipley RA, Shipley LJ, Wearn JT (1937) The capillary supply
in normal and hypertrophied hearts of rabbits. J Exp Med 65:
29-41
478
CIRCULATION RESEARCH
Tomanek RJ (1970) Effects of age and exercise on the extent of
the myocardial capillary bed. Anat Rec 167: 55-62
Turina M, Bussman WD, Rrayenbuhl HP (1969) Contractility
of the hypertrophied canine heart in chronic volume overload.
Cardiovasc Res 3: 486-495
VOL. 49, No. 2, AUGUST 1981
Unge G, Tomling G, Adollsson J, Ljungqvist A (1979) Capillary
proliferation in heart and skeletal muscles in swimming exercised and dipyndamole treated rats (abstr). Microvasc Res 17:
S22
Functional and Structural Changes in the
Rabbit Ear Artery after Sympathetic
Denervation
ROSEMARY D. BEVAN AND HIROMICHI TSURU
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
SUMMARY We studied the tissue weight, dimensions, contractility, elasticity, and sensitivity to
exogenous norepinephrine (NE) of denervated and innervated segments of the central ear arteries of
white New Zealand rabbits. Three different age groups received unilateral superior cervical ganglionectomies, "growing" at 3-4 weeks, "young adult" at 9-11 weeks, and "mature" at 16-20 weeks. In the
growing group, 8 weeks after ganglionectomy, the denervated arteries showed mean decreases in
tissue weight (11%), total wall thickness (12%), cross-sectional area of media (17%), contractility (16%),
and increases in the tangential modulus of elasticity and sensitivity to NE (2.3-fold) compared to the
contralateral control vessels. The change in medial cross-sectional area was significant in the growing
and young adult but not the mature animals. The other changes, however, although consistently seen,
differed quantitatively among the groups. These results indicate that an intact innervation is necessary
for normal development and maintenance of the artery wall. However, the precise consequences of this
influence vary at different ages. Whether this influence involves a special trophic factor is not known.
CircRes 49: 478-485, 1981
DENERVATION produces remarkable morphological and functional changes in skeletal muscle
(Guth, 1968; Drachman, 1976). In several mammalian species studied, diversification of muscle fibers
occurs during postnatal development and is partially dependent on the innervation (Gutmann,
1976). It also has been reported that the sympathetic nervous system exerts a trophic influence on
skeletal and smooth muscle (Mendez et al., 1970;
Luco and Luco, 1971). However, it is generally
believed that'denervation of smooth muscle produces little morphological change (Thoenen and
Tranzer, 1963; Levi-Montalcini, 1972). More recently, several apparently conflicting observations
have been made regarding morphological changes
associated with denervation of smooth muscle, using different methods or species and tissues. Following sympathetic denervation of the growing rabbit
ear artery, there was less 3H-thymidine uptake
(DNA precursor) and there were fewer labeled
From the Department of Pharmacology. UCLA School of Medicine,
University of California, Los Angeles. California.
Suporledb\ US Public Health Service Grant HL 20581 and American
Heart Association-Greater Los Angeles Affiliate 4O8IG
Dr Tsuru's present address is: Department of Pharmacology, Nagoya
University School of Medicine, Tsuruma, Showa-ku, Nagoya 466, Japan.
Address for reprints Dr Rosemary D Bevan. Department of Pharmacology, UCLA School of Medicine, Los Angeles, California 90024
Received March 26. 1979, accepted for publication February' 10, 1981
smooth muscle cell nuclei in the tunica media than
in the control artery, suggesting that sympathetic
innervation influences proliferation of vascular
smooth muscle (Bevan, 1975). In contrast, Chamley
and Campbell (1976) observed that the presence of
sympathetic nerve fibers or homogenates of sympathetic chain delayed by several days the normal
dedifferentiation and proliferation of isolated medial smooth muscle cells from the aorta and ear
artery of "young" rabbits in tissue cultures. In addition, Campbell et al. (1977) have shown that
sympathetic denervation of chicken extensor secundariorum longus muscle by a total brachial plexus
lesion in the 2-week and adult chicken produced a
significant increase in dry weight, due to hyperplasia of the smooth muscle cells.
In the present paper, effects of sympathetic denervation on contractility, dimensions, elasticity,
weight, and sensitivity to exogenous norepinephrine
were examined in the course of a study of sympathetic nervous influence on the blood vessel wall. A
preliminary report has appeared elsewhere (Bevan
and Tsuru, 1978).
Methods
Three groups of New Zealand white male rabbits
were used: (1) growing rabbits of 3-4 weeks of age,
(2) young adult rabbits of 9-11 weeks, and (3)
Capillary growth and changes in heart performance induced by chronic bradycardial pacing
in the rabbit.
A J Wright and O Hudlicka
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Circ Res. 1981;49:469-478
doi: 10.1161/01.RES.49.2.469
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1981 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/49/2/469.citation
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/