779
Cardiovascular and Respiratory Effects of
Adenosine in Conscious Man
Evidence for Chemoreceptor Activation
Italo Biaggioni, Bjarki Olafsson, Rose Marie Robertson,
Alan S. Hollister, and David Robertson
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
The cardiovascular and respiratory effects of intravenous adenosine were studied in conscious normal
volunteers. Bolus injections of adenosine increased systolic and diastolic pressures initially (+ IS and
+13 mm Hg after 100 jug/kg) followed by a subsequent reduction in systolic and diastolic pressures
(—12 and —16 mm Hg). Heart rate increased during trough blood pressure (R-R interval shortening
of 298 msec after 100 /xg/kg). Adenosine steady-state infusions increased heart rate ( + 30 beats/min
during 140 /xg/kg/min), systolic pressure ( + 16 mm Hg), and pulse pressure ( + 21 mm Hg) but
decreased diastolic pressure slightly ( - 5 mm Hg), resulting in no significant change in mean arterial
pressure. Adenosine stimulated respiration, resulting in decreased Paco2 (41 to 31 mm Hg), increased
Pao2 (101 to 113 mm Hg), and increased pH (7.42 to 7.50). The increased ventilation was not explained
by bronchoconstriction, hypotension, or hypoxia. The observed pressor and tachycardic effects are
mediated through reflex autonomic mechanisms since they are completely abolished in patients with
severe autonomic failure. These autonomic mechanisms probably involve chemoreceptor activation
since adenosine is pressor when infused in the aortic arch proximal to the origin of the carotid arteries
but depressor when infused in the descending aorta. It is concluded that the hemodynamic and
respiratory effects of adenosine observed in normal volunteers are in part due to chemoreceptor
stimulation. These findings raise the possibility that adenosine is an endogenous, modulator of
respiration in man. (Circulation Research 1987;61:779-786)
T
he physiologic relevance of adenosine, an
intermediate product in the metabolism of
adenosine triphosphate (ATP), has been increasingly recognized. Commonly used drugs may act
at least partially through adenosine receptor blockade,
e.g., methylxanthines like theophylline and caffeine,
or adenosine uptake inhibition, e.g., dipyridamole. It
has been postulated that endogenous adenosine is
involved in the regulation of local blood, flow, renin
release, and neurotransmission among other physiologic processes. Adenosine, given as a rapid bolus
injection, is an effective agent in the treatment of
supraventricular tachyarrhythmias by slowing atrioventricular (AV) nodal conduction' and may become
the treatment of choice for this arrhythmia.2 The
possible role of adenosine as an antiarrhythmic agent
From the Departments of Medicine and Pharmacology, Divisions
of Clinical Pharmacology, and Cardiology, Vanderbilt University
School of Medicine, Nashville, Tenn.
This work has been partially presented in abstract form (J Am Coll
Cardiol 1987;9:137A).
Supported by grants HL-34021, HL-31419, and RR 00095 from
the National Institutes of Health and by a Grant-in-Aid from the
American Heart Association contributed in part by the Tennessee
Affiliate.
Dr. Biaggioni is a Clinical Associate Physician of the Elliot V.
Newman Clinical Research Center. Dr. R.M. Robertson is an
Established Investigator of the American Heart Association. Dr. D.
Robertson is a Burroughs Wellcome Scholar in Clinical Pharmacology.
Address for correspondence: Italo Biaggioni, MD, Department of
Pharmacology, Vanderbilt University, Nashville, TN 37232.
Received March 3, 1987; accepted June 26, 1987.
underscores the need to study further the cardiovascular
effects of this agent in man. Accordingly, one of the
purposes of this study was to define the dose-response
relation for the cardiovascular effects of bolus injections of adenosine in man.
Previous in vitro studies and animal experiments
suggest that adenosine may also be a useful hypotensi ve
agent. Adenosine is a potent vasodilator both in vitro
and in vivo, acting on specific vascular receptors.
Adenosine also produces bradycardia due to AV nodal
conduction delay and slowing of intrinsic sinoatrial
(SA) node firing, inhibits norepinephrine release
through presynaptic receptors, and inhibits renin re- '
lease. With these characteristics, adenosine would
theoretically produce hypotension with little or no
reflex sympathetic activation, tachycardia, or
tachyphylaxis secondary to renin activation. Indeed, in
anesthetized patients, adenosine induces a significant
and sustained reduction in blood pressure ( - 35 mm
Hg) with only a mild increase in heart rate ( + 9
beats/min).3 However, in contrast to the observed
effects of adenosine in anesthetized patients, the
authors have previously reported that in conscious man,
adenosine causes a dose-related increase in heart rate
and in systolic blood pressure, while producing only a
slight decrease in diastolic blood pressure.4 It was
hypothesized that these effects were the result of
activation by adenosine of a reflex autonomic mechanism that is probably blunted during anesthesia. The
present study tested this hypothesis by determining the
effects of adenosine in patients with severe autonomic
failure. Since these patients are functionally devoid of
780
autonomic reflexes, the intrinsic effects of adenosine
should be apparent in them.
Finally, it has been observed that the intravenous
administration of adenosine is associated with an urge
to breathe deeply,4 suggesting that adenosine stimulates
respiration in man. 5 A final purpose of this study was
to evaluate the respiratory effects of adenosine.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Subjects and Methods
Normal Volunteers
Healthy male volunteers, age 28 ± 2 years, were
asked to abstain from methylxanthine-containing beverages (coffee, tea, chocolates, or soft drinks) for at
least 3 days prior to the study. Plasma levels of caffeine
were measured to assure compliance. Throughout the
study, blood pressure was monitored continuously
through a radial artery catheter with a pressure transducer connected to a Hewlett-Packard 8805c pressure
amplifier (Hewlett-Packard Co., Waltham, Mass.).
Heart rate was monitored by surface ECG connected to
an HP 8812A beat-to-beat rate computer. Thoracic
excursion and respiratory rate were monitored with a
corrugated tube placed around the chest and connected
to a pressure transducer. Intravenous lines were placed
in an antecubital vein for adenosine or saline administration. After instrumentation, subjects were allowed
to rest in a quiet room for a period of 30 minutes before
any study was begun.
Nine subjects received increasing doses of adenosine
as rapid intravenous bolus injections into the left
antecubital vein, starting at 20 /Ag/kg with increments
of 20 /u,g/kg to a maximum dose of 200 /-ig/kg or until
intolerable side effects developed (as defined by the
subject) or until cardiac arrhythmias appeared. Similar
injections of saline were given before, after, and
randomly during adenosine administration in a singleblind fashion. The concentration of adenosine used was
either 1 or 10 mg/ml so that total volume injected was
less than 5 ml.
Eight subjects received in a single-blind fashion
intravenous infusion of saline followed by increasing
doses of adenosine (80 to 180 //,g/kg/min) of 15 minutes
duration each, using a Harvard syringe pump (model
2720, C.R. Bard MedSystems Inc., North Reading,
Mass.; recalibrated for 35-cc plastic syringes). Near
the end of each infusion, blood samples from the
arterial catheter were obtained for catecholamines and
arterial blood gases determinations. Spirometry was
also obtained during saline and at the maximal adenosine dose for forced vital capacity (FVC) and forced
expiratory volume at 1-second (FEV1) measurements.
(Spirometer 2400, Breon Laboratories Inc., New
York). After the adenosine infusions, saline was
infused again. Hemodynamic and respiratory functions
were monitored during this recovery period.
Patients With Severe Autonomic Failure
To determine the cardiovascular effects of bolus
injections of adenosine in the absence of autonomic
reflexes, adenosine was administered to 4 patients with
severe autonomic failure (age 66 ± 5 years, 3 females).
Circulation Research
Vol 61, No 6, December 1987
Three patients had primary autonomic failure (idiopathic orthostatic hypotension), and 1 had multiple
system atrophy (Shy-Drager syndrome).6 All patients had severe orthostatic hypotension (average
fall in systolic blood pressure on standing of 104 ± 18
mm Hg and in diastolic blood pressure of 59 ± 6 mm
Hg) without adequate compensatory heart rate response. The infusion protocol was similar to the one
described in the preceding section. Because of the
known hypersensitivity of these patients to a variety of
pressor and depressor stimuli, initial doses of 2.5
Mg/kg and 5 jitg/kg/min were used, with gradual
increments until a blood pressure change of 25 mm Hg
was observed.
Patients Undergoing Diagnostic Arteriography
To determine if the site of injection would alter its
cardiovascular actions, adenosine was administered to
4 patients undergoing diagnostic coronary arteriography. After the routine cardiac catheterization, adenosine boluses were injected through a catheter placed
sequentially at 2 different sites: in the aortic arch
proximal to the origin of the carotid arteries and in the
descending aorta distal to the origin of the renal
arteries. An initial dose of 1 /ug/kg was given, followed
by sequential increments of 2.5 /u-g/kg until a change
in blood pressure of 25 mm Hg was observed. All
protocols were reviewed and approved by the Vanderbilt University Institutional Review Board.
Measurements
To determine the effects of bolus injections, baseline
blood pressure prior to each bolus injection was
measured from the arterial tracings by averaging the
normal respiratory fluctuations of blood pressure. The
maximal increase or decrease seen after bolus injections was compared with individual baseline values.
Baseline heart rate was measured by averaging 10 R-R
intervals before each injection. To determine if adenosine induced transient bradycardia, the single longest
R-R interval after bolus injections was measured. Five
R : R intervals were also averaged around the maximal
decrease in blood pressure where significant tachycardia was observed. These measurements were compared
with individual baselines and expressed as changes.
To determine the effects of adenosine infusions, an
average of 12 blood pressure determinations during the
last minute of saline infusion was taken as baseline and
compared to similar averages obtained during adenosine infusions. Mean arterial blood pressure was
calculated ('/3 systolic + ¥3 diastolic). Heart rate was
measured from the rate computer tracings.
Blood for catecholamines was collected into iced
tubes containing reduced glutathione and EGTA and
assayed radioenzymatically as previously described.7
Caffeine levels were determined in 100 fx.\ of plasma
purified with a 1-ml Bond Elut C18 column (Analytichem International, Harbor City, Calif.), using /3hydroxyethyltheophylline as an internal standard and
analyzed by high performance liquid chromatography
(HPLC), with a 2% acetic acid and 6% acetonitrile
Biaggioni et al
Cardiovascular and Respiratory Effects of Adenosine
mobile phase at a flow rate of 1.5 ml/min and ultraviolet
absorbance monitored at 280 nm.
Results are presented as mean ± SEM and evaluated
by analysis of variance (CLINFO System, Clinical Research Center, BNN Software Products Corp., Cambridge, Mass.). When the analysis of variance revealed
a difference, the residual mean square was applied in
a Dunnett's test to characterize which values were
different from baseline. Group differences were analyzed by unpaired t tests. All null hypotheses were
two-tailed, and the criterion of significance was
200
BLOOD
PRESSURE
mmHg
HEART
RATE
bpm
Adenosine for human use was obtained from Sigma
Chemical Company^ St. Louis, Mo. It was dissolved
in normal saline under sterile conditions and tested for
sterility and pyrogenicity. Both the concentration and
purity of the compound was assessed by HPLC and by
HPLC-mass spectrometry.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Results
Bolus Injections of Adenosine in Normal Volunteers
A typical cardiovascular and respiratory response to
an adenosine bolus is shown in Figure 1. Adenosine
consistently produced a biphasic blood pressure response. At 27 ± 0 . 4 seconds, adenosine increased
systolic (/?<0.001) and diastolic (p<0.001) blood
pressures, maximal increases being 17 and 13 mm Hg,
respectively (Figure 2, left panels). Following this
initial pressor response, at 39 ± 0.4 seconds adenosine
produced a decrease in systolic (NS) and diastolic
(/?<0.001) blood pressures (Figure 2, right panels).
Maximal changes were —9 and — 12 mm Hg, respectively. All these changes were dose-dependent, the
threshold dose being 40 /u,g/kg.
Five out of nine subjects developed transient rhythm
disturbances after bolus injections of adenosine. In all
cases, these occurred at the peak of blood pressure and,
781
100
l0
°
50
o
THORACIC
EXCURSION
ADENOSINE
IV
10
Sec
FIGURE 1. Representative response to an intravenous bolus
injection of adenosine (100 ng/kg) on blood pressure, heart
rate, and thoracic excursion.
in 4 subjects, were related to AV conduction delay. One
subject developed a slow ectopic atrial rhythm (Table
1). However, on average, no definite bradycardia was
observed (Figure 2, left panels). In contrast, a dosedependent increase in heart rate was observed during
the trough of blood pressure (p<0.002, Figure 2),
maximal R-R interval shortening being 298 msec.
Adenosine infusions (80 to 140 /ug/kg/min) produced dose-dependent increases in heart rate
(p<0.001) and systolic blood pressure (p<0.003) and
a small decrease in diastolic blood pressure (NS).
FIGURE 2. Effects of bolus injection ofadenosine (x
axis) on heart rate (upper panels, expressed as
change in R-R interval compared with saline),
systolic blood pressure (middle panels), and diastolic blood pressure (lower panels) in 9 normal
volunteers. Measurements were done at peak blood
pressure (27±04 seconds after injection, left panelsj and at trough blood pressures (39 ± 04 seconds
after injection, right panels). The p values shown are
from ANOVA. *Significant differencesfrom baseline
(Dunnett's test).
20 40 60 80
AOENOSINE ug/kg
100
20 40 60 80
ADENOSINE ug/kg
100
782
Circulation Research
Table 1. R-R Interval Change Observed After 100 jug/kg
Bolus Injection in Normal Volunteers
Subject
1
2
3
4
5
6
7
8
9
R-R
PR prolongation (0.35 sec)
PR prolongation (0.22 sec)
PR prolongation (0.23 sec)
Sinus rhythm
Sinus rhythm
II AV block 2:1
Sinus rhythm
Ectopic atrial rhythm
Sinus rhythm
258
-132
52
94
-80
+ 752
-120
+ 450
-130
30
Duration of
arrhythmia
Heart rhythm
Vol 61, No 6, December 1987
flHR
20
bp m
10
5 beats
2 beats
3 beats
0
80
100
120 140 0 3 10
20
flSBP
mm Hg
5 beats
15
10
5
1 beat
p<0.005
0
80
100
120 140 03 10
80
100
120
10
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Changes during the 140 /xg/kg/min dose were +30
beats/min, + 16 mm Hg, and - 5 mm Hg, respectively
(Figure 3). As a result of the divergent effects on
systolic and diastolic blood pressures, mean arterial
blood pressure remained unchanged, whereas pulse
pressure widened (p<0.001, 21 mm Hg for the 140
/Ag/kg/min dose). Adenosine infusions also increased
thoracic excursion (p<0.001, Figure 4) without significantly affecting respiratory rate (14 ±1 resp/min
during 140 /xg/kg/min adenosine infusion compared
with 15 ± 1 resp/min at baseline). This increase in
ventilation resulted in respiratory alkalosis, with a
decrease in Paco2 (p<0.001) and increases in Pao2 (NS)
and pH (p<0.00l). Maximal changes were — 11 mm
Hg, 12 mm Hg, and 0.08, respectively (Figure 4). No
change was seen on the P(A-a)o2 difference during
adenosine infusions (4.0 mm Hg during the 140
/Ag/kg/min dose versus 6.0 mm Hg at baseline).
Adenosine infusions produced no change in FEV1
(4.14 ±0.24 1), FVC (5.12 ±0.35, 1), or the
FEV1: FVC ratio (0.82 ±0.02) compared with saline
infusion (4.28±0.25, 5.25±0.35, and 0.82±0.02,
respectively). Threshold dose for respiratory stimulation was 80 /tg/kg/min, while threshold dose for
cardiovascular effects was 100 /u,g/kg/min. These
effects were dose dependent and were reversed shortly
after discontinuation of the infusion (right side of
Figures 3 and 4).
**
300
THORACIC
EXCURSION
% INCREASE 200
0*' 80
100
7.46
7.44
7.42
[/
FIGURE 3. Cardiovascular effects of steady-state intravenous
infusions of adenosine (x axis) on heart rate and systolic and
diastolic blood pressures in 8 normal volunteers. Time course
after adenosine discontinuation is shown at right. The p values
shown are from ANOVA. *Significant differences from baseline
(Dunnett's test).
Adenosine infusions were associated with a dosedependent increase in plasma catecholamines. Plasma
norepinephrine rose from 222 ± 26 pg/ml during saline
infusion to 323 ±34 pg/ml during 140 /xg/kg/min
adenosine infusion (41% increase,p<0.05). Similarly,
plasma epinephrine rose from 77 ±11 pg/ml to
148 ±29 pg/ml (80% increase, p = 0.08).
Cardiovascular Effects of Adenosine in Patients With
Autonomic Failure
Bolus injections of adenosine were given to 4
patients with severe autonomic failure with both
parasympathetic and sympathetic involvement, as assessed by loss of sinus arrhythmia, lack of normal blood
100 120 140 0 3 10
PoCOz
mm Hg
y
40
36
32
p < 0.001
p< 0.001
I
_J
0
140 0 3 10
ADENOSINE ftg/Vg/min min
44
*
•
0
100
80
*
7.48
-10
no
120 140 0 3 10
7.50
pH
0
-5
90
•4
p< 0.001
mm Hg
• tu
PoO 2
mmHg
100
5
80 100 120 140 03 10
ADENOSINE*ig/kg/min min
0
80 100 120 140 0 3 10
ADENOSINEpg/kg/min min
FIGURE 4. Respiratory effects of steadystate intravenous infusions of adenosine (x
axis) on thoracic excursion, pH, Pao2, and
Paco2 in 8 normal volunteers. The time course
after adenosine discontinuation is shown at
right. The p values shown are from ANOVA.
*Significant differences from baseline (Dunnett's test).
Biaggioni et al
Cardiovascular and Respiratory Effects of Adenosine
783
200 r
200
BLOOD
PRESSURE
100
mm Hg
100
HEART
RATE
bpm
FIGURE 5. Representative responses to
intravenous bolus injections of adenosine (7.5 and 10 /Jig/kg) on blood pressure, heart rate, and thoracic excursion
in 2 patients with severe autonomic
failure.
100
50[ _
50
THORACIC
EXCURSION
ii:i ii : :U i'.r. I
ADENOSINE
7.5/ig/kg IV
|
ADENOSINE
10/Lig/kg IV
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
pressure overshoot during phase IV of Valsalva maneuver, absence of heart rate response to Valsalva, and
lack of pressor response to sustained handgrip or cold
exposure. Furthermore, supine plasma norepinephrine
was low (130 ±21 pg/ml), considering their ages, and
failed to increase on assuming the upright posture
(149 ±24 pg/ml), especially considering their fall in
blood pressure on standing. Although individual responses to adenosine were variable, these patients were
more sensitive to its actions than normal subjects.
Individual doses that met the predetermined hemodynamic endpoints (see "Subjects and Methods") were
7.5, 10, 20, and 80 /u,g/kg. Representative results are
shown in Figure 5. The effects of maximal doses of
adenosine observed in patients with autonomic failure
are compared in Figure 6A with the effects seen in
normal volunteers. The initial pressor effect of adenosine seen in normal volunteers was greatly attenuated
in patients with autonomic failure. Furthermore, even
though the subsequent decrease in blood pressure was
accentuated in autonomic failure patients, the tachycardia observed in normal volunteers during the trough
blood pressure was totally abolished in autonomic
failure patients. Instead, significant bradycardia occurred (R-R interval prolongation of 176±68 msec).
The ventilatory response to adenosine, as assessed by
the increase in thoracic excursion, was also blunted in
these patients compared with normals (Figure 5).
Two patients also received continuous infusions of
adenosine (60 and 120 /u,g/kg/min). Again, in contrast
to the effects seen in normal volunteers, dramatic
decreases in both systolic and diastolic blood pressure
were observed, with no change in heart rate, in contrast
to the increase in heart rate and systolic blood pressure
observed in normal volunteers (Figure 6B).
Differential Effects of Adenosine According to
Site of Infusion
When adenosine (5 /ig/kg) was infused into the
aortic arch proximal to the origin of the carotid arteries,
it produced an initial increase in heart rate (R-R interval
shortening of 114 ± 48 msec at 8 ± 0.4 seconds) and in
blood pressure (+19 ± 5 and +12 ± 3 mm Hg for
systolic and diastolic pressures, respectively). At
18 ±0.8 seconds, there was- a decrease in blood
pressure ( - 1 0 ± 3 / - 5 ± l mm Hg for systolic/diastolic pressure, Figure 7). In contrast, when adenosine (5
/xg/kg) was infused in the descending aorta, neither the
initial tachycardia nor the increase in blood pressure
was observed (/><0.05 and /?<0.01, respectively,
-300
-200
ARR
msec
3O
n
HR
bpm
|0
- °
o
1
r
20
|Q
.
^.
I UP3
100
L
200
L
1^
20
20 •
10
10 •
ASBP °
I
-30
,
•
-20
•
T
HO-
mmHg _2Q
-30 -40 •
-50 •
SBP
08P
FIGURE 6. A: Cardiovascular effects of intravenous bolus
injections of adenosine in normal volunteers (open bars, 100
yiglkg, n = 9) and in patients with severe autonomic failure
(closed bars, 29 ±17 ]xglkg, n = 4)., Heart rate responses
(expressed as changes in R-R interval) are shown in the upper
panel and change in systolic blood pressure in the lower panel.
Measurements were done at peak blood pressure (left-hand
columns) and trough blood pressures (right-hand columns). B:
Cardiovascular effects of intravenous steady-state infusions of
adenosine in normal volunteers (open bars, 140 ixglkglmin,
n = 8) and in each patient with severe autonomicfailure (hatched
and closed bars, 60 and 120 ixg/kg/min). Changes in heart rate
are represented in the upper panel and changes in systolic and
diastolic pressures in the lower panel.
784
Circulation Research
Vol 61, No 6, December 1987
200 r
BLOOD
PRESSURE
100
mmHg
0
100
HEART
RATE
bpm
FIGURE 7. Representative responses to bolus
injections of adenosine (5 fxg/kg) infused into the
aortic arch (left panel) and descending aorta
(right panel) in a patient undergoing routine
diagnostic catheterization.
50
0
THORACIC
EXCURSION
ADENOSINE
5/ug/kg
AORTIC ARCH
«?FC
ac
ADENOSINE
5 M g/kg
DESCENDING AORTA
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
compared with the ascending aorta injections). Furthermore, the subsequent decrease in blood pressure
observed at 18 ± 0 . 8 seconds after injection in the
descending aorta ( — 2 6 ± 2 / — 9 ± 1 mm Hg systolic/
diastolic pressures) was significantly greater than that
observed when adenosine was injected in the aortic
arch ( - 10 ± 3 / - 5 ± 1 mm Hg, Figure 8).
Side Effects
High-dose adenosine administration in normal volunteers was accompanied by the appearance of side
effects (an urge to breathe deeply, nervousness, head
and neck flushing, and sometimes headache). These
side effects were dose-dependent. Responses to bolus
injections greater than 100 /ag/kg and infusions greater
-150
-100
A RR
msec
-50
0
50
30
20
10
A SBP
mmHg
0
u
• AORTIC
ARCH
D OES.
AORTA
I
-10
-20
-30
8±0,4
18+0.8
sec
FIGURE 8. Cardiovascular effects of intravenous bolus injection of adenosine (5 ng/kg) infused into the aortic arch (closed
bars) or the descending aorta (open bars) in 4 patients
immediately after routine diagnostic catheterization. Measurements were done at8 ±0.4 and at J8±0.8 seconds. *Significant
differences between groups by unpaired t test.
5
SEC
than 140 /xg/kg/min in normal volunteers were not
analyzed because of the dropout rate at higher doses.
Discussion
The results reported here are similar to our previous
observations that in conscious man, adenosine infusions increase heart rate and systolic blood pressure.4
Moreover, it has been found that bolus injections of
adenosine produce an initial increase in both systolic
and diastolic blood pressures, an observation not
previously reported. It is unlikely that this increase is
the result of a direct vascular action of adenosine since
vasodilatory effects of adenosine in most vascular beds
have been well described. Furthermore, adenosine is
clearly hypotensive when given to anesthetized
patients.3 Therefore, it was postulated that increases in
heart rate and blood pressure could be due to adenosinemediated activation of an autonomic reflex mechanism.
The data obtained in patients with severe autonomic
failure support this hypothesis. Increases in heart rate
and blood pressure seen after adenosine infusion or
bolus injections in normal volunteers were totally
abolished in patients with autonomic failure. Since
these patients are selectively devoid of autonomic
reflexes, the observed cardiovascular effects more
likely represent the direct actions of adenosine. Thus,
the effects seen in normal volunteers are not due to the
direct actions of adenosine but to an adenosine-induced
activation of a reflex autonomic mechanism.
It is possible that a central sympathetic activation
might be elicited as a reaction to unpleasant side effects
of adenosine. This could explain some of the observed
changes in plasma catecholamines and cardiovascular
responses. However, it is unlikely that reaction to
distress could explain, in the highly reproducible
fashion observed, the discordant effects on systolic and
diastolic blood pressures during adenosine infusion or
the complex and concerted respiratory and cardiovascular responses occurring after bolus injections of
adenosine.
Biaggioni et al Cardiovascular and Respiratory Effects of Adenosine
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Results in the present study also show that adenosine
administration stimulates respiration in man. Although
methodologic limitations of our measurement make
the observed increase in thoracic excursion supportive
but not conclusive of increased ventilation, our conclusion of respiratory stimulation is based on the
changes observed in arterial blood gases. Increased
ventilation was evidenced by a significant reduction of
Paco2 and the development of respiratory alkalosis.
Before it can be concluded that adenosine stimulates
respiration, the possibility that the increase in ventilation is an epiphenomena associated with other effects
of this substance must be ruled out. Inhaled adenosine
may produce bronchoconstriction in asthmatics.8 Although bronchoconstriction is not found in normal
subjects, it could theoretically produce the hyperventilation observed in our study as a compensatory
mechanism. However, this possibility can be excluded
since there was no change in spirometric data during
adenosine infusions. Likewise, adenosine-induced hyperventilation was not secondary to hypotension since
adenosine infusions had no effect on mean arterial
blood pressure; on the contrary, systolic blood pressure
increased. Neither can it be explained by hypoxia or
diffusion abnormalities, e.g., secondary to ventilation/
perfusion inequalities or pulmonary congestion, since
Pao2 increased and P(A-a)o2 remained unchanged.
Finally, a direct central activation is unlikely since
intravenously administered adenosine does not enter the
central nervous system in animals9 and, furthermore,
adenosine applied directly into the central nervous
system depresses respiration10 and reduces blood
pressure." Therefore, it is concluded that adenosine
administration produces direct respiratory stimulation.
To explain both the increase in blood pressure and
respiratory stimulation seen during adenosine administration, it was postulated that chemoreceptor activation was involved since adenosine has been shown to
produce carotid body chemoreceptor activation in
animals.1213 The effects of adenosine are mediated
through specific cell surface receptors. The magnitude
of its action, therefore, depends on the extracellular
concentration achieved. Because of the rapid uptake
and degradation of adenosine by blood cells in man,'4
its half-life in the circulation is extremely short; it has
been postulated as being less than 10 seconds or even
less than 1 second. Therefore, the site of injection and
mode of administration will critically influence the
concentration of adenosine that reaches the relevant
functional sites responsible for its actions. This very
short half-life of adenosine permitted the hypothesis of
adenosine-induced chemoreceptor action to be tested
by infusing adenosine proximal and distal to the origin
of the carotid arteries. Indeed, the cardiovascular
effects of adenosine were totally different at these two
sites. When infused into the ascending aorta, adenosine
increased blood pressure and heart rate. However,
when adenosine was infused into the descending aorta,
these initial responses were completely abolished.
Rather, a profound reduction in blood pressure was
observed (Figure 8).
785
It has also been shown that intrarenal infusions of
adenosine produce an acute increase in blood pressure
in dogs through activation of afferent renal sympathetic
fibers. '5 Activation of this mechanism could not explain
the respiratory stimulation observed in our study but
could contribute to the pressor effect of adenosine.
However, intrarenal infusion of adenosine does not
increase blood pressure acutely in conscious rats16 nor
chronically in dogs.17 In our patients undergoing
diagnostic catheterization, bolus injections above the
origin of the renal arteries did not produce consistent
increments in blood pressure (data not shown). Thus,
this mechanism does not explain the results observed
in our study. However, its occurrence in man cannot be
completely excluded.
Taken together, our results support the hypothesis
that the direct vascular effects of adenosine (which can
be readily observed in patients with autonomic
failure) are masked in normal volunteers by activation
of autonomic reflex mechanisms, most likely chemoreceptor activation. Preliminary results from other
investigators also suggest that adenosine activates
carotid body chemoreceptors in man.18 Adenosine
increased minute ventilation when administered as a
constant infusion in the ascending aorta but not when
infused distal to the origin of the carotid arteries.
Furthermore, when infused constantly at subhemodynamic doses, adenosine enhances the ventilatory
response to hypoxia, implying an action at the
chemoreceptor level." Adenosine-induced carotid
body chemoreceptor activation has been demonstrated in animals,20 and although this is the most
likely site of action of adenosine in man, a contribution of aortic chemoreceptors cannot be excluded.
Observed differences in cardiovascular response to
adenosine between conscious volunteers and anesthetized patients may be related to a blunting of
chemoreflexes during anesthesia.21
Pure chemoreceptor activation results in increased
blood pressure, bradycardia, and respiratory stimulation. However, unless ventilation is maintained constant, pulmonary stretch receptors are activated secondary to the hyperventilation, and they tend to blunt
the increase in blood pressure and to produce
tachycardia.2223 Thus, in conscious man, the final
hemodynamic effects of adenosine are the result of a
complex interplay of different phenomena: direct
cardiovascular effects of adenosine (bradycardia and
vasodilatation), adenosine-induced activation of chemoreflexes (bradycardia and increased blood pressure),
and secondary activation of pulmonary stretch receptors (tachycardia).
It has been shown in different organs that adenosine
levels are greatly increased during hypoxia, and in this
situation, adenosine may be an important modulator of
physiologic process. It is obvious that conclusions on
the role of endogenous adenosine cannot be drawn from
the present study. However, our finding that adenosine
is a direct respiratory stimulant raises the possibility
that adenosine is an endogenous modulator of respiration in man.
786
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
In summary, in conscious normal human subjects,
adenosine infusions increase heart rate, systolic blood
pressure, and pulse pressure, while producing only a
mild decrease in diastolic blood pressure and no change
in mean arterial blood pressure. Adenosine bolus
injections produce a biphasic response, with an initial
increase in both systolic and diastolic blood pressure
followed by a fall in blood pressure. Heart rate also
increases during bolus injections. Adenosine stimulates respiration in man, producing a respiratory
alkalosis that is not the result of bronchoconstriction,
hypoxia, or hypotension. These changes are dose
related and reflect adenosine-induced activation of
reflex mechanisms, probably chemoreceptor activation, since initial increases in heart rate and blood
pressure seen after intravenous bolus injections in
normal volunteers were abolished in patients with
autonomic failure and were not seen when adenosine
was injected in the descending aorta in patients
undergoing diagnostic catheterization. Although these
results suggest that adenosine is not an effective
hypotensive agent in conscious man, the physiologic
role of adenosine in respiratory homeostasis and its
therapeutic potential as a respiratory stimulant deserve
further study.
Acknowledgments
The authors would like to thank Cheryl Stewart, RN,
Mr. Behret Patel, and Mrs. Suzanna Lonce for their
superb technical assistance and Mrs. Dorothea Boemer
for typing this manuscript.
References
1. DiMarco JP, Sellers TD, Berne RM, West GA, Bellardinelli
L: Adenosine: Electrophysiologic effects and therapeutic use
for terminating paroxysmal supraventricular tachycardia. Circulation 1983;68:1254-1263
2. Belhassen B, Pelleg A: Acute management of paroxysmal
supraventricular tachycardia: Verapamil, adenosine triphosphate or adenosine? Am J Cardiol 1984;54:225-227
3. Sollevi A, Lagerkranser M, Irestedt L, Gordon E, Lindquist C:
Controlled hypotension with adenosine in cerebral aneurysm
surgery. Anesthesiology 1984;61:400-405
4. Biaggioni I, Onrot J, Hollister AS, Robertson D: Humoral and
hemodynamic effects of adenosine infusion in man. Life Sci
1986;39:2229-2236
5. Watt AH, Routledge PA: Adenosine stimulates respiration in
man. Br J Clin Pharmacol 1985;20:503-506
6. Onrot J, Goldberg MR, Hollister AS, Biaggioni I, Robertson
RM, Robertson D: Management of chronic orthostatic hypotension. Am J Med 1986;80:454-464
Circulation Research
Vol 61, No 6, December 1987
7. Peuler JD, Johnson GA: Simultaneous single isotope radioenzymatic assay of plasma norepinephrine, epinephrine, and
dopamine. Life Sci 1977;21:625-636
8. Cushley MJ, Tattersfield AE, Holgate ST: Adenosine-induced
bronchoconstriction in asthma: Antagonism by inhaled
theophylline. Am Rev Respir Dis 1984; 129:380-384
9. Berne RM, Rubio R, Curnish RR: Release of adenosine from
ischemic brain: Effect of cerebral vascular resistance and
incorporation into cerebral adenosine nucleotides. Circ Res
1974;35:262-271
10. Eldridge FL, Millhorn DE, Kiley JP: Antagonism by theophylline of respiratory inhibition induced by adenosine. J Appl
Physiol 1985;59:1428-1433
11. Tseng CJ, Appalsamy M, Biaggioni I, Robertson D: Adenosine
causes hypotension and bradycardia following injection into
the nucleus tractus solitarii of the rat (abstract). Clin Res
1987;35:382A
12. McQueen DS, Ribeiro JA: Effect of adenosine on carotid
chemoreceptor activity in the cat. Br J Pharmacol 1981;
74:129-136
13. McQueen DS, Ribeiro JA: On the specificity and type of
receptor involved in carotid body chemoreceptor activation by
adenosine in the cat. Br J Pharmacol 1983;80:347-354
14. Klabunde RE: Dipyridamole inhibition of adenosine metabolism in human blood. Eur J Pharmacol 1983;93:21-26
15. Katholi RE, Hageman GR, Whitlow PL, Woods WT: Hemodynamic and afferent renal nerve responses to intrarenal
adenosine in the dog. Hypertension 1983;5(suppl I):I-1491-154
16. Ohnishi A, Biaggioni I, Deray G, Branch RA, Jackson EK:
Hemodynamic effects of adenosine in conscious spontaneously
hypertensive and normotensive rats. Hypertension 1986;
8:391-398
17. Hall JE, Granger JP: Renal hemodynamics and arterial pressure
during chronic intrarenal adenosine infusion in conscious dogs.
Am J Physiol 1986;250:F32-F39
18. Watt AH, Reid PG, Stephens MR, Routledge PA: Adenosineinduced respiratory stimulation depends on site of infusion
(abstract). Br J Pharmacol 1986;89:110
19. Maxwell DL, Fuller RW, Nolop EB, Dixon CMS, Hughes
JMB: The effects of adenosine on ventilatory responses to
hypoxia and hypercapnea in humans. J Appl Physiol 1986;
61:1762-1766
20. McQueen DS, Ribeiro JA: Pharmacological characterization of
the receptor involved in chemoexcitation induced by adenosine.
BrJ Pharmacol 1986;88:615-620
21. Zimpfer M, Sit SP, Vatner SF: Effect of anesthesia on the canine
carotid chemoreceptor reflex. Circ Res 1981;48:400-406
22. Gronblad M: Function and structure of the carotid body. Med
Biol 1983;61:229-248
23. Vatner SF, McRitchie RJ: Interactions of the chemoreflex and
the pulmonary inflation reflex in the regulation of coronary
circulation in conscious dogs. Circ Res 1975;37:664—673
KEY WORDS • adenosine • hemodynamics
carotid body chemoreceptor • respiration
• human
Cardiovascular and respiratory effects of adenosine in conscious man. Evidence for
chemoreceptor activation.
I Biaggioni, B Olafsson, R M Robertson, A S Hollister and D Robertson
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Circ Res. 1987;61:779-786
doi: 10.1161/01.RES.61.6.779
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1987 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/61/6/779
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/