Sleep
10(3):272-278, Raven Press, New York
© 1987, Association of Professional Sleep Societies
The Cardioacceleratory Response to Arecoline
Infusion During Sleep in Narcoleptic Subjects
and Controls
Heidi L. Baruch, Surendra Kelwala, and Sheldon Kapen
Veterans Administration Medical Center, Allen Park, and Departments of Psychiatry and
Neurology, Wayne State University, Detroit, Michigan; U.S.A.
Summary: Nine narcoleptic and nine control subjects underwent 4 nights of
sleep recordings. On nights 3 and 4, they received continuous intravenous infusions of saline. Additionally, on both nights they received 0.2 mg glycopyrrolate at the end of the first REM period (REMl) and 0.5 mg arecoline or
placebo in random order 20 min after the end of REM 1. Heart rates were
counted for a 40-min period following the end of REMl. There was a significant and similar cardioacceleratory effect after arecoline in both narcoleptic
and normal subjects, beginning at 5 min from the start of the infusion and
peaking at 9 min. Placebo had no effect. Narcoleptic subjects had consistently
higher baseline heart rates than controls on infusion and noninfusion nights,
most likely owing to age differences between the two groups. The results suggest that narcoleptic persons do not have increased cholinergic sensitivity, or
that the canine model of narcolepsy differs from the human model, or that the
muscarinic receptors that play a role in the pathophysiology of narcolepsy
differ in sensitivity from those that regulate heart rate. Key Words: Arecoline
- Heart rate-Narcolepsy-Muscarinic receptors-Sleep.
Several animal studies have demonstrated a cardioacceleratory response to cholinomimetics secondary to increased central muscarinic receptor activity (1-4). Brezenoff
(1) showed that carbachol and neostigmine, when given as microinjections into different hypothalamic sites in rats, produced predictable effects upon the heart rate.
Microinjections into the posterior or ventromedial hypothalamic nuclei increased the
heart rate, whereas the opposite response was observed with injections into the dorsomedial hypothalamic nucleus and premammillary nuclei. Buccafucso and Brezenoff (2)
further showed that these cardiac responses are due to muscarinic effects of cholinergic agents and not nicotinic effects. When rats were injected with the antimuscarinic
agent atropine before receiving carbachol, the rise in heart rate and blood pressure was
blocked. When rats were pretreated with an antinicotinic drug, this blocking effect did
not occur. This cardioacceleratory response of cholinergic agents appears to be meAccepted for pUblication December 1986.
Address correspondence and reprint requests to Surendra Kelwala, M.D. at VA Medical Center (I16A),
Outer Drive and Southfield, Allen Park, MI 48101, U.S.A.
272
:1
CARDIAC RESPONSE TO ARECOLINE IN NARCOLEPSY
'J
-]
273
diated through complex central adrenergic and cholinergic pathway interactions (3).
Engberg and Svensson (4), in studies on rats, found that cholinergic agonists produced
an increase in firing of noradrenergic neurons within the locus ceruleus, and within the
locus ceruleus, the cholinergic receptors appear to be present on the adrenergic cell
bodies and show muscarinic qualities. Muscarinic agonists cause increased firing of
locus ceruleus neurons, whereas nicotinic agents do not.
Similar to effects in animal studies were those of cholinergic agents on the cardiovascular system of humans, Nurnberger et al. (5) found that 8.0 mg arecoline s.c. administered to Huntington's disease patients produced a significant increase in heart rate (6).
Janowsky et al. (7) reported that psychiatric patients, following physostigmine administration, experienced an increase in heart rate 10 min post infusion. An interesting
finding of that study was the presence of a correlation between the baseline mood and
activity states and the increase in heart rate. Physostigmine administered orally (and
without a peripheral muscarinic blocking drug) was associated with increased blood
pressure in a patient with Alzheimer's disease (8). Central cholinergic supersensitivity,
whereby the central effects of a cholinomimetic outweighed the peripheral effects, was
implicated in these changes in cardiovascular functioning (8). A cardioacceleratory response to 0.5 mg arecoline in sleeping human subjects with major depression was reported by Abramson et al. (9), in which heart rate peaked 5-6 min from the beginning
of the 3-min infusion. Additionally, a significant correlation existed between the magnitude of the heart rate increase and the latency of the arecoline infusion to the onset of
REM sleep (9). Induction of REM sleep with cholinergic agents has been used as a
measure of brainstem muscarinic receptor sensitivity (10).
In our laboratory, we have been studying the cholinergic status of narcoleptic individuals. Increased muscarinic receptor sensitivity in the brain stem nuclei of narcoleptic subjects has been implicated as the cause of REM sleep abnormality and other
symptoms of narcolepsy (11-13). In narcoleptic dogs, physostigmine and arecoline
increase the occurrence of cataplexy, while atropine and scopolamine alleviate the attacks (11). Further, in human and canine narcolepsy, prior to cataplectic attacks, there
is a robust increase in heart rate that is independent of changes in muscular tone and
conceivably caused by a common cholinergic mechanism (14,15). Finally, direct measurement of cholinergic receptors has shown an increase in muscarinic receptor density in most brainstem nuclei of narcoleptic dogs when compared with controls (12,13).
We have used a cardioacceleratory response to low-dose arecoline during sleep as a
parameter to study muscarinic receptor sensitivity in human narcolepsy, and we report
our findings here.
METHODS
,;1
Selection of subjects and sleep studies
After informed consent, all subjects underwent a physical examination and lab investigations that included an electrocardiogram (ECG), automated serum chemistry,
complete blood count with differential, and urinalysis. Subjects additionally underwent
a structured psychiatric interview (Schedule for Affective Disorders and Schizophrenia-life-time version) (16) in order to exclude those who suffered from depression or any other psychiatric disorder that might have interfered with their sleep. Narcoleptic subjects were given a multiple sleep latency test (a series of four 20-min naps
given throughout the day) (17) the day following their first night in the study. Two or
Sleep, Vol. 10, No_ 3,1987
H. L. BAR UCH ET AL.
274
more sleep onset REM periods during these naps were required for confirming the
diagnosis of narcolepsy. Subjects participated in a 4-night protocol whereby their sleep
was monitored via electroencephalogram (C3-A2 or C4-Al and OZ-A2), electromyogram (mentalis and left anterior tibialis), electrooculogram (right and left outer canthus),
and ECG (V5-A2) recordings on Grass polygraphs (model 78D). Respiration was
monitored the first night with nasal/oral thermistors to rule out sleep apnea. The presence of any sleep disorder other than narcolepsy led to exclusion from the remainder of
the study. On the third and fourth nights, subjects went to sleep with an intravenous
scalp vein needle placed on their forearm. This was attached to a 10-ft-Iong polyethylene tube that extended through a porthole in the wall, leading from the sleep room to
the monitoring room. A slow, continuous normal saline infusion was maintained
throughout the night. Five minutes following the end of the first REM period (REMl)
[as determined by standard criteria (18)], subjects received 0.2 mg glycopyrrolate, a
peripherally acting anticholinergic drug, administered from outside the subjects' room
through the intravenous setup described above. On alternate nights, in random order,
they also received either placebo (saline) or 0.5 mg arecoline 20 min following the end
of REMI. The drug was infused over 3 min.
Heart rate data analysis procedure
Heart rate was counted manually from polysomnographic ECG tracings. Previous
studies have shown that heart rate changes due to arecoline occur within 15 min of
infusion (5,7,9). Therefore, data collection began immediately after the end of REMI
and was stopped 20 min after the end of the arecoline/placebo infusion (40 min after
REMl period ended).
Baseline heart rate was calculated by averaging the heart rate during the 5-min period between the end of REMl and the beginning of the glycopyrrolate infusion (see
Fig. 1). Correlations between baseline heart rate and age were examined for each group,
followed by an analysis of covariance using age as the covariate. The drug effects on
the heart rate were compared with those of placebo on 11 subjects using multiple
paired t tests at matching time points. Effects of glycopyrrolate and arecoline infusions
were measured by comparing the baseline heart rate with the minute-by-minute heart
rate changes following the infusions. Analyses of variance with repeated measures
GLYCOPYRROLATE
INFUSION
END OF REM 1
BASELINE HEART RATE
FIG. 1. Calculation of baseline heart rate.
-:-:-:.:-:-:-:-:-:
/
.........
........
.......
...
........
.......
.
........
.
....... ...
........
.......
........
.......
....
........
.......
........
.
.......
........
.......
.....
........
.......
........ .
REM1
Sleep, Vol. /0, No.3, 1987
2
3
4
Minutes
5
CARDIAC RESPONSE TO ARECOLINE IN NARCOLEPSY
.:1
275
were done to examine these heart rate changes over time and to note between-group
differences (between narcoleptic subjects and controls). Additionally, analyses of covariance using baseline heart rate as the covariate were done at each time point of the
experiment to compare the narcoleptic and control group heart rate responses to arecoline. Analyses of variance were also done on the individual subject groups, followed by
t tests for correlated samples. Finally, correlations between baseline heart rates and the
magnitude of increase in heart rates post arecoline were calculated within groups for
narcoleptic and control subjects.
RESULTS
~
..,'
Nine narcoleptics, (five women, four men) aged 22-60 years ex = 46 ± 10.78 years)
and nine control subjects (six women, three men) aged 23-56 years (x 36 ± 10.34
years) completed the study. An unpaired t test showed a near-significant difference in
age between the two groups [t(16) = 1.92, p = 0.07]. There was a positive correlation
between age and baseline heart rate for narcoleptic subjects and controls combined (r
= 0.56, p < 0.05). An analysis of covariance with age as the covariate revealed no
significant difference between the two groups on baseline heart rate [F(1,15) = 2.75].
There was a mild cardioacceleratory effect of glycopyrrolate in both narcoleptic subjects and controls. A two-way analysis of variance with repeated measures was significant for the repeated measure [F(1,IO) =4.286, p < 0.01] and for the between-group
factors [F(1,16) = 7.228, p < 0.05]. There was no significant interaction between the
above two factors [F(2,1O) = 1.152]. When treating each group separately, the control
group showed no significant effect of glycopyrrolate upon heart rate [F(10,80) =
1.807]. While narcoleptic subjects, treated separately, showed a significant effect of
glycopyrrolate upon heart rate [F(10,80) = 3.165, p < 0.01], only one t test (between
the baseline heart rate and min 2 from the start of the glycopyrrolate infusion) was
significant [t(8) = 2.721, p < 0.05].
The cardioacceleratory effect after arecoline was clearly significant. A two-way analysis of variance with repeated measures was significant for the repeated measure
[F(1,20) = 13.411, p < 0.01] and for the between-group factor [F(l,16) = 5.686, p <
0.05]. The interaction was not significant [F(2,20) = 0.630]. There were significant effects of arecoline upon heart rate for both narcoleptic subjects [F(20,160) = 7.165 p <
0.01] and controls [F(20,160) = 6.901, p < 0.01] when examining the groups separately. T-tests between baseline heart rate and each of the 20 min following the arecoline infusion showed a significant heart rate response beginning at min 5 from the start
of the infusion for both the narcoleptic subjects [t(8) = 2.671, p < 0.05], and the controls [t(8) = 6.804, p < 0.01]. This heart rate response peaked at min 9 for the narcoleptic subjects [t(8) = 4.621, p < 0.01] and for the controls [t(8) = 6.804, p < 0.01].
While the subsequent heart rates decreased in both groups, they remained significantly above their respective baselines for the duration of the experiment. On multiple
paired t tests between placebo and arecoline (see Fig. 2), arecoline's effect can be
clearly seen. There was a significant increase in heart rate beginning with the fifth
minute from the start ofthe arecoline infusion, with the peak heart rate effect occurring
at min 9 from the start of the infusion. Figure 2 also shows that glycopyrrolate produced virtually no lasting cardioacceleratory response.
The heart rate response of narcoleptic and control subjects to arecoline did not significantly differ, as is shown in Table 1. The magnitude of increase in heart rate was the
Sleep, Vol. 10, No.3, 1987
H. L. BARUCH ET AL.
276
88
Z 86
....
:E
;; 84
....
<
::l82
....'" 80
;:i 78
I;;;
:i::c 76
74
BASE- 1-10 ,I 2 3, 4
LINE MIN. ARECOLINE
POST PLACEBO
GLYC-INFUSION
OPYRROLATE
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
BASELINE,POST-GLYCOPYYROLATE & MINUTES 1-20 FROM
START OF ARECOLINE/PLACEBO INFUSION
FIG. 2. Comparisons of arecoline and placebo effects on heart rate; n = 11 (8 narcoleptic and 3 control
subjects). :j:Placebo infusion was done only on eight narcoleptic and three control subjects. *p < 0.05 on
multiple paired t tests at matching time points.
same for both groups. However, as can be seen, there was a significant difference in
the baseline heart rates between the two groups. Figure 3 shows that this difference
was maintained throughout the experiment. Analyses of covariance with baseline heart
rate as the covariate (performed at each time point in the experiment) statistically confirmed the above. None of the analyses was statistically significant [F(l, 15)], ranging
from 0.001 to 1.36. Finally, no significant correlations were found for either narcoleptic
subjects (r = - 0.18) or controls (r = 0.09) between baseline heart rates and magnitude
of increase in heart rates.
DISCUSSION
We have attempted to demonstrate a heart rate response in narcoleptic and control
subjects to the cholinergic agonist arecoline preceded by the anticholinergic drug glycopyrrolate. Our study shows the cardioacceleratory effects of glycopyrrolate to be
minor and short lived, and similar in narcoleptic and control individuals. This finding is
consistent with previous results (5,9).
While the difference in baseline heart rates found between the two groups is likely to
be due primarily to age, the sleeping heart rate difference between narcoleptic subjects
and controls in carefully age-matched groups should be examined in the future.
TABLE 1. Heart rate differences between narcoleptic and control subjects on arecoline night
X baseline heart rate
x peak heart rate post arecoline
x change in heart rate from baseline to peak
Sleep, Vol. 10, No.3, 1987
Narcoleptic
Control
p
77.09 ± 13.10
90.00 ± 14.48
12.\8 ± 6.69
64.04 ± 7.22
78.89 ± 9.28
14.51 ± 6.08
<0.02
NS
CARDIAC RESPONSE TO ARECOLINE IN NARCOLEPSY
277
90
88
86
,:1
84
82
80
78
76
.74
z
~72
...
Vl
~70
'"
CONTROLS
N-9
;;;'68
...
...;:i66
'"
~64
:t:
62
60~BA~S~E~_-'I_'lnO-'J-02-o3~''4-'5~6C-'7-o8-n9-'lno-'II'-V12"1~3-'1'4-'IT5-'16'-PI7'-1~8-'1~9-o2no------line
POST ARECOLINE
GLYINFUSION
COPYRROLATE
BASE LINE, POST- GLYCOPYYROLATE, & MINUTES
1-20 FROM START OF ARECOLINE INFUSION.
FIG. 3. Effects of arecoline infusion on the heart rate of narcoleptic and control subjects.
•.'
In the last 15 years, many researchers have documented that centrally active cholinergic drugs have cardioacceleratory properties. We have attempted to see if this cardioacceleratory response can be used as a measure of central cholinergic receptor sensitivity in clinical settings. In narcolepsy, where animal models have strongly suggested
increased muscarinic receptor sensitivity in the brainstem (11-14), one would have
expected to observe a greater heart rate response to arecoline than in controls. Our
findings failed to confirm this hypothesis.
Our observations may be influenced by any of the following possibilities: (a) Narcoleptic individuals may not have an enhanced muscarinic receptor sensitivity. (b) The
canine model of narcolepsy may be different from human narcolepsy, at least with
respect to muscarinic receptor sensitivity. (c) Supersensitivity of muscarinic receptors
in narcolepsy is confined to only those cholinoreceptors that playa role in the pathophysiology of narcolepsy, while muscarinic receptors that regulate heart rate have
normal sensitivity. However, we have also reported that the cholinergic REM induction response is not enhanced in narcoleptic individuals compared with control subjects (19). (d) Narcoleptic individuals still might have greater cholinergic sensitivity
than controls in spite of their similar heart rate responses to arecoline. This could have
occurred were there a "ceiling" effect in the cardioacceleratory response to arecoline
and if narcoleptic subjects were on a "steeper" portion of the cardiac response curve
(owing to their higher initial heart rate). This possibility was rejected when we found no
negative correlations between the baseline heart rates of narcoleptic or normal subjects
(within groups) and the magnitude of their heart rate changes induced by arecoline.
Sleep, Vol. 10, No.3, 1987
278
H. L. BARUCH ET AL.
Acknowledgment: This work was supported by VA Medical Research funds. We wish to thank
Mr. Paul Sherwood and Dr. Manfred Greiffenstein for their heip in stalislicai anaiysis and Prof.
Natraj Sitaram for his invaluable comments in the preparation of this manuscript.
REFERENCES
I. Brezenoff HE. Cardiovascular responses to intrahypothalamic injections of carbachol and certain cholinesterase inhibitors. Neuropharmacology 1972; II :637 -44.
2. Buccafusco JJ, Brezenoff HE. Pharmacological study of a cholinergic mechanism within the rat posterior hypothalamic nucleus which mediates a hypertensive response. Brain Res 1979;165:295-310.
3. Brczenoff HE, Giuliano R. Cardiovascular control by cholinergic mechanisms in the central nervous
system. Annu Rev Pharmacol ToxicoI1982;22:341-81.
4. Engberg G, Svensson TH. Pharmacological analysis of a cholinergic receptor mediated regulation of
brain norepinephrine neurons. J Neural Transm 1980;49: 137 -50.
5. Nurnberger 11 Jr, Jimerson DC, Simmons-Alling S, et al. Behavioral, physiological, and neuroendocrine
responses to arecoline in normal twins and "well-state" bipolar patients. Psychiatry Res 1983;9:191200.
6. Nutt JG, Rosin A, Chase TN. Treatment of Huntington's disease with a cholinergic agonist. Neurology
1978;28:1061-4.
7. Janowsky DS, Risch SC, Huey L, Judd LL, Rausch 1. Central physostigmine-induced cardiovascular
and behavioral changes: toward an acetylcholine hypothesis of stress. Psychopharmacol Bull
1983;9:675-81.
8. Cain Jw. Hypertension associated with oral administration of physostigmine in a patient with Alzheimer's disease. Psychiatry 1986; 143:910-2.
9. Abramson LB, Brown AJ, Sitaram N. A cardioacceleratory response to low-dose arecoline infusion
during sleep in patients with major depressive disorder: relationship to REM sleep induction. Psychiatry
Res 1985;16: 189-98.
10. Sitaram N, Moore AM, Gillin JC. Experimental acceleration and slowing of REM sleep ultradian rhythm
by cholinergic agonist and antagonist. Nature 1978;274:490-2.
11. Delashaw JB, Foutz As, Guilleminault C, Dement WC. Cholinergic mechanisms and cataplexy in dogs.
Exp NeuroI1979;66:745-57.
12. Boehme RE, Baker TL, Milford IN. Narcolepsy, cholinergic receptor changes in an animal model. Life
Sci 1984;34: 1825-8.
13. Kilduff T, Bowersox S, Kaitin K, Baker T, Ciaranello R, Dement WC. Muscarinic cholinergic receptors
and the canine model of narcolepsy. Sleep 1986;9: 102-6.
14. Siegel JM, Fahringer H, Tomaszewski KS, Kaitin K, Kilduff T, Dement We. Heart rate and blood
pressure changes associated with cateplexy in canine narcolepy. Sleep 1986;9:216-21.
15. Guilleminault C, Sava MAQ, Mancuso J, Hayes B. Narcolepsy, cataplexy, heart rate, and blood pressure. Sleep 1986;9:222-6.
16. Spitzer RL, Endecott J. Schedule for affective disorders and schizophrenia. New York: New York State
Psychiatric Institute Biometric Research, 1978.
17. Richardson OS, Carskadon MA, Flagg W, Van den Hoed J, Dement WC, Mitler MM. Excessive daytime
sleepiness in man: multiple sleep latency measurement in narcoleptic and control subjects. Electroencephalogr Clin NeurophysioI1978;45:621-7.
18. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for
sleep stages of human subjects. Los Angeles: BIS/BRI UCLA, 1968.
19. Kelwala S, Kapen S, Baruch H, Koshcorek O. Is there a muscarinic receptor sensitivity in narcolepsy?
Sleep Res 1986;15:31.
Sleep, Vol. 10. No.3, 1987