Download Transcutaneous electrical stimulation applied to the infratrochlear

Clinical Science ( 1990) 78,457-462
457
Transcutaneous electrical stimulation applied to the
infratrochlear nerve induces a homatropine-resistant miosis
in humans
B. M. FUSCO, M. ALESSANDRI, V. CAMPAGNOLO AND M. FANCIULLACCI*
Institute of Internal Medicine and Therapeutics IV, and *Department of Preclinical and Clinical Pharmacology, University of Florence,
Florence, Italy
(Received 25 July/8 December 1989; accepted 5 January 1990)
SUMMARY
1. Both high- and low-intensity transcutaneous electrical stimuli were applied to the emergence of the infratrochlear nerve in 18 healthy subjects. The effect on the
size of the homolateral pupil was investigated. The width
of the pupil was also measured when high-intensity transcutaneoiis electrical stimulation was applied to the
contralateral side.
2. The high-intensity pulse resulted in constriction of
the pupil when the stimulation was homolateral. The
miosis was slow in onset (120 s latency) and long-lasting
(80 s). No pupillary changes were detected after either
ipsilateral low-intensity or contralateral high-intensity
stimuli.
3. In 11 healthy subjects, the pupillary response to
transcutaneous electrical stimulation was evaluated during iris parasympathetic blockade induced by homatropine eyedrops. The disappearance of the light reflex
due to homatropine was considered an index of the parasympathetic blockade. Afterwards, a high-intensity pulse
was transcutaneously delivered to the emergence of the
infratrochlear nerve and the ipsilateral pupil size was
measured.
4. A reduction in the pupillary size followed the electrical stimulation, still under the effect of homatropine
which abolished the light reflex. The time course of this
pupillary constriction was similar to that seen without the
influence of homatropine.
5. The findings suggest that homolateral miosis,
observed after unilateral high-intensity stimulation of the
infratrochlear nerve, does not stem from cholinergic
activation, It has been suggested that miosis induced by
transcutaneous electrical stimulation may be due to an
antidromic activation of the iris sensory fibres.
Correspondence: Professor Marcello Fanciullacci, Department of Preclinical and Clinical Pharmacology, vide Morgagni
65,50134 Florence, Italy.
Key words: axon reflex, homatropine, iris sensory
neurons, light reflex, miosis, pupil, transcutaneous electrical stimulation.
Abbreviations: SP, substance P; TES, transcutaneous
.lectrical stimulation.
INTRODUCTION
The size of the pupil is determined by a balance of the
innervation between the autonomically supplied sphincter
and the radially arranged dilator muscles of the iris; the
sphincter muscle plays the major role. The commonest
stimulus for pupillary constriction, which involves cholinergic pathways, is the exposure of the retina to light.
Moreover, reflex pupillary constriction allows for the act
of convergence and accommodation to nearby objects. As
a rule, noxious stimuli induce a mydriatic response [l].
However, such stimuli, when directed to the ocular area,
can result in a pupillary constriction. In particular, the socalled ‘oculo-trigeminal’ reflex, elicited by sustained
stimulation of the cornea and conjunctiva or by pinching
or pricking the skin of the eyelid, consists of a brief
bilateral dilatation followed by a prolonged constriction
of the pupil. The miotic response is noticeable in both
eyes, even though the pupil of the stimulated side is
usually narrowed to a greater extent. The structures which
are involved in this response are unknown. However, it
has been suggested that impulses run through the trigeminal afferents and the trigeminal sensory nucleus; thus, in
some unknown way, pupillary contraction is mediated by
the activation of a cholinergic mechanism [ 2 ] .
Hypotheses have been advanced that the trigeminal
sensory neurons arising from the iris may themselves be
involved in the mechanism of iris constriction through an
antidromic activity, especially after ocular injury. In 1910,
Bruce [3] extended to the iris the concept inferred by the
observation of the antidromic sensory vasodilatation [4]
and explained the ocular response to local injury, including miosis, by an axon reflex. The role of the capillary
458
B. M. Fusco et al.
dilatation, induced by an axon reflex, was emphasized in
1943 by Longworthy & Ortaga [5]to explain the changes in
pupillary size during ocular irritation. In 1957, Perkins [GI
found that electrical stimulation of the ophthalmic trigeminal branch of the rabbit resulted in a miosis similar to
that provoked by local injury. He hypothesized that an
unidentified substance, capable of contracting the pupil
and which he called ‘iridin’, might be released after an
antidromic activation of the iris trigeminal neurons. In
1979, Bill et al. [7] showed that miosis after electrical
stimulation of the trigeminal branch is related to an
increase in substance P (SP), one of the putative neurotransmitters of sensory neurons [8], in the anterior
chamber of the eye. Jampol et al. [9], Butler & Hammond
[lo]and Butler ef al. [ lI ] stated that sensory denervation
prevents miosis caused by eye exposure to both chemical
and physical noxious stimuli. They have concluded that
sensory fibres mediate the pupillary response. Holmdhal
et al. [12] showed that the iris constriction occurring
during the ocular inflammatory response is inhibited by
an antagonist of SP. Moreover, SP and neurokinin A, which
are found in certain sensory neurons of various
mammalian species [13-151, have a potent rniotic effect
on the rabbit eye [lG]. In a recent preliminary study, we
showed that in man too, locally applied SP induces
pupillary constriction [17].
The present study was designed to evaluate, in healthy
volunteers, the pupillary response to transcutaneous electrical stimulation (TES) of an area included in the
ophthalmic trigeminal distribution. TES was applied
alternatively to both homolateral and contralateral sides
of the pupil being examined. The emergence of the infratrochlear nerve was chosen as the stimulation site,
because of the anatomical closeness of the infratrochlear
nerve fibres to the sensory fibres which arise from the iris.
In fact, both bundles of fibres run within the nasociliary
trunk of the trigeminal ophthalmic branch. Moreover, the
response to both low- and high-intensity pulses was
assayed to verify whether the pupillary variations are
related to the intensity of the electrical stimulus. Finally,
the cholinergic involvement in the pupillary response to
electrical stimulus was studied by administering highintensity TES after conjunctival instillation of homatropine in the eye being examined.
METHODS
The study was carried out on 29 healthy subjects, who
were either medical students or members of the medical
staff of our Institute. Informed consent was obtained from
each subject before the start of the study, which had been
fully approved by the Supervisory Committee of our Institute. Before admission to the study, all subjects were
screened in order to exclude any eye disease. A single eye
from each volunteer was chosen at random for the investigation.
Pupillary response to TES
This part of the study involved 18 subjects (10 men
and eight women, mean age +. SEM 37.4 zk 4.1 years). The
effects of low- (10 mA) and high- (40 mA) intensity TES
on the area of the pupil homolateral to the side of the
stimulation were evaluated. Moreover, the response to
high-intensity TES, administered to the opposite side of
the pupil being measured, was also assayed. Each test was
performed on 3 different days, always starting at 09.00
hours. Pupil area was measured by using a TV monocular
electronic pupillometer (Hamamatsu, Japan) consisting of
a TV camera, a chin and forehead rest table, a control unit
and a monitor. An eye-fix lamp, producing a standard
green light, was included in the TV camera.
Measurements were carried out in infrared light conditions and were started after a 5 min period of adaptation
to the darkness. First, subjects were requested to focus on
the green light, which was considered the standard reference point for accommodation. Then the measurement of
the pupil area was performed. After the basal
measurement (pre-stimulation value), an electrical
stimulus, consisting of a single square wave pulse of 0.8
ms duration, was applied to the cutaneous area corresponding to the emergence of the infratrochlear nerve.
The pulse was administered by means of a monopolar
electrolyte-gel sponge electrode connected to a pulse
generator (Neuroton Siemens, Erlangen, West Germany).
During the first two study sessions, the area of the pupil
was monitored every 20 s for 4 min after the administration of electrical pulse on the same side. In one of these
two sessions, chosen at random, subjects received a highintensity stimulus, whereas during the other session a lowintensity pulse was given. In a third session, the
high-intensity stimulus was applied to the side contralateral to the pupil being examined and the response was
evaluated for 4 min at the time intervals mentioned above.
Pupillary response during hornatropine-induced blockade
of the light reflex
This study was performed on 11 volunteers (six men
and five women, mean age 1- SEM 40.3 +. 5.1 years). In
these subjects the effects of high-intensity TES applied to
the emergence of the infratrochlear nerve were assayed in
the pupil of the ipsilateral eye, which had been previously
treated (90 min before) with homatropine hydrobromide
eyedrops (1% aqueous solution). The disappearance of
the light reflex was assumed to be an indication of homatropine-induced blockade of parasympathetic activity. To
evaluate this block, before and during the homatropine
effect, a wcll-tolerated brief light, stimulus (time 0.5 s,
intensity l o o + 10 nW) was administered while the pupil
area was measured. The light-induced miosis was directly
detected by the pupillometer which displayed the values
of the pupil areas measured before ( A,) and after ( A , )the
light stimulation and then the percentage difference
between the two areas ( A3 ) was calculated. Immediately
after the evaluation of the homatropine-induced block of
the reflex to light, the response of the pupil to homolateral
high-intensity TES was evaluated by the above-described
procedure. A control session was performed 1 week later,
during which homatropine was re-administered in the
same eye of each subject, and the measurement of the
Miosis induced by electrical stimulation in humans
pupil area was repeated by following the entire test procedure, without, however, administering any electrical
stimulation.
Statistical analysis
Analysis of variance for repeated measures was used to
test differences in average pupil areas detected during the
first part of the study; the factors were: the stimuli (homolateral low- and high-intensity and contralateral highintensity pulses), the time, and the interaction between
them (type of stimulus x time). Afterwards, point and level
of significance were analysed by applying post-hoc tests.
Comparison between the values measured before and
after each stimulation was made by using Dunnett's test.
Duncan's test allowed point by point comparison of the
values of the pupil area which were observed during the
three kinds of electrical stimulation.
In the second part of the study, the homatropineinduced block of pupillary response to the light stimulus
was analysed by comparing the values of A , and A ,
detected before eyedrop instillation with those obtained
after drug administration. For this purpose, the f-test for
paired data was used. The f-test for paired data was also
employed to compare the pupillary values detected under
the effect of homatropine after TES with those recorded
in the same condition but without TES.
Values of P smaller than 0.05 were set as the significance level for hypothesis testing.
459
t
3
c
!
3
+I
+I
s+I
2
N
8
N
0
3
v!
x
u
!
3
+I
+I
The pupil areas measured after both homolateral application of low-intensity TES and contralateral administration of high-intensity TES did not show any statistical
difference when compared with the respective baseline
values (Table 1).The homolateral high-intensity stimulus
induced a reduction in pupil width which was statistically
significant at 120, 140, 160, 180 and 200 s (P<0.01)
when compared with pre-stimulation values (Table 1).
By comparing the three groups of data, the baseline
values appeared statistically matched. A significant reduction in the pupil area measured after the homolateral
high-intensity electrical stimulation was observed when
compared with those detected after homolateral lowintensity and contralateral high-intensity TES. The reduction in the pupil size was statistically significant 100
(P<0.05), 120, 140, 160, 180 (P<0.01), and 200
( P < 0.05) s after the stimulation (Fig. 1).
Pupillary response to TES under the effect of hornatropine
+I
N
8
3
N
c
!
3
x
3
u
!
3
+I
+I
N
P
3
m
t
3
c
!
3
u
!
3
+I
+I
:
0
+I
2
2
9
3
N
N
3
x
+I
+I
+I
*
N
0
t
2
N
t
3
<
W
c!
3
+I
+I
2
N
't:
3
+I
v!
3
o!
I
3
2
I
v!
3
+I
9
3
N
c
!
3
Y
3
+I
+I
2
N
2
N
t
3
-
'9
3
+I
+I
3
2
+I
x
2
2
v!
3
T
3
v!
+I
+I
N
N
d
;
+I
2
-
c!
c
!
3
+I
v!
*
+I
2
2
v!
N
N
N
0
v!
p!
u!
N
3
+I
3
t
N
0
+I
3
+I
+I
2
N
8
v!
c
!
3
u
!
3
20
+I
+I
3
0
2
c!
c!
8
N
3
N
N
N
N
3
Y
3
+I
+I
+I
2
2
t
3
N
Ninety minutes after application of homatropine eyedrops, a marked mydriasis and disappearance of the light
reflex was obtained in the treated eye (Table 2). Under the
effect of homatropine, the values of the pupil area
detected before the high-intensity TES were not statistically different from the basal values detected during the
3
2
8
RESULTS
Pupillary response to TES
t
N
N
N
0
B. M. Fusco et al.
460
23 r
22 21
-
h
-E
E
20-
n 19-
2
..!
n
2
18-
1716 -
15 -
** **
,
Before
TES
20 40 60 80 100 120 140 160 180 200 220 240
Time after TES (s)
Fig. 1. Pupil areas measured before and after homolateral low- ( 0 ) and high- (A ) intensity TES and contralateral high-intensity TES )(. applied transcutaneously to
the emergence of the infratrochlear nerve. Analysis of
variance for repeated measures revealed significant
effects of the high-intensity stimulus ( F = 6.75, P < 0.01)
and time ( F = 10.72, P<0.001), as well as a significant
stimulus x time interaction ( F = 13.82, P<0.001). Statistical significance (Duncan’s test): * P < 0.05; **P< 0.01 vs
both homolateral low-intensity and contralateral highintensity TES. All data are means? SEM (tz = 18).
control session (Fig. 2). A reduction in the homatropineinduced mydriasis was present after high-intensity TES. It
was statistically significant at 100 ( P < 0.05) and from 120
to 200 s (P<O.01), as compared with the corresponding
control values (Fig. 2).
DISCUSSION
TES, applied to the cutaneous emergence of the infratrochlear nerve, resulted in a long-lasting and sustained
miotic response which occurred only in the ipsilateral
pupil. Pupillary constriction was achieved by administering a single high-intensity pulse, whereas low-intensity
stimulus was incapable of evoking pupillary change. Since
a recruitment of the thinner afferent neurons is obtained
by increasing the intensity of electrical stimulation [ 18,
191, our findings suggest that an activation of the smaller
sensory neurons is needed to elicit the pupillary effect.
The occurrence of miosis exclusively in the stimulated
side indicates that thc response does not necessarily
involve the central nervous structures. In fact, an activation of the brainstem nucleus should be followed by a
mutual narrowing of the contralateral pupil. Moreover,
pupillary constriction, with a similar time course,
occurred even after the conjunctival instillation of a dose
of homatropine capable of disabling the cholinergic reflex
to light; therefore, the miosis induced by the high-intensity
stimulus does not seem to depend on a parasympathetic
mechanism.
The activation of larger diameter afferent fibres produces an inhibition of the sympathetic neurons [20]; thus
miosis after electrical stimulation might stem from a
reduction in the iris sympathetic tonus through the activation of these afferents. However, it seems unlikely, since
no pupillary variation was observed after applying lowintensity stimulation which is capable of exciting the
larger afferent neurons.
The hypothesis that trigeminal fibres contribute to control of pupil size suggests an interesting interpretation of
the present findings. The time course of the pupillary
response to high-intensity TES is slow in onset and longlasting: it is not comparable with that occurring in autonomic-mediated reflexes. On the contrary, it correlates
with the response found in the isolated rabbit iris after
transmural electrical stimulation, which results in a noncholinergic, non-adrenergic constriction that is inhibited
by SP antagonists [21]. Moreover, it is also similar in
onset and duration to the pupillary constriction induced
in the rabbit by direct stimulation of the ophthalmic nerve
[7]. Most of the evidence on the dual ‘sensory-efferent’
function of sensory fibres, in the iris particularly, stems
from experimental studies in animals. In man, SP-like
immunoreactivity is found in ocular structures [22],
including the iris sphincter muscle. The peptide, instilled
in the conjunctival sac of healthy volunteers, produces
consistent miosis [ 171.
The proposal that miosis induced by TES is mediated
by an anterograde discharge of iris sensory fibres leads to
speculation as to the mechanisms through which antidromic activation may be elicited. It has been suggested
that TES may antidromically activate the sensory fibres.
The increased survival of the ischaemia musculocutaneous flap in the rat, observed after local application
of high-intensity, high-frequency TES, provides evidence
for an antidromic activation of the sensory fibres [23]. In
fact, the effect is abolished by pharmacological sensory
denervation and reproduced by local application of calcitonin-gene-related peptide [24], a vasoactive polypeptide
[25]which is contained in the sensory fibres and released
after antidromic stimulation [26]. Therefore, miosis
induced by TES may be due to a preganglionic short loop,
acting through an axo-axonal reflex [27], between two
groups of sensory fibres, the fibres arising from the area
innervated by the infrathrochlear nerve and those coming
from the iris. In fact, both bundles of fibres join in the
same trunk (nasociliary) of the ophthalmic branch. Otherwise an intraganglionic integration through an interTable 2. Blockade of the light reflex during the homatropine-induced mydriasis
The pupil area was measured before ( A,) and after ( A , )
light stimulus. A , represents the percentage difference
between A , and A , . Values are m e a n s k ~( n~=~11).
Statistical significance (paired t-test): * P< 0.001 compared with before homatropine.
Before
A , (mm?)
A (mmz)
A , (“4
homatropine
After
homatropine
20.64 k 1.71
11.2f1.23
45.74 k 5.52
35.6 k 1.2*
34.7+ 1.17*
2.58 k 0.30*
Miosis induced by electrical stimulation in humans
44
-
43.
-
- 42
“E 41.
2 403
39-9 38.
I
a
37 36 -
35t
**
1
**
I
Before
stimulation
20 40 60 80 100 120 140 160 180 200 220 240
Time after TES (s)
Fig. 2. Reduction of the homatropine-induced mydriasis
by homolateral high-intensity TES ( 0 ) applied to the
infratrochlear nerve shown in comparison with the values
detected during the control session under the effect of
homatropine without TES (A ). Statistical significance
(paired t-test): * P < 0.05; **P < 0.01 vs control. All data
are means +EM ( n= 11).
somatic connection, acting like an axon reflex [28], may
occur. Finally, a spreading of the electrical pulse directly
from the stimulated cutaneous area to the iris sensory
fibres must be taken into account as a possible mechanism
of the antidromic activation.
In conclusion, the application of a single high-intensity
electrical stimulus on the cutaneous emergence of the
infratrochlear nerve elicits a non-cholinergic miotic
response only in the pupil of the stimulated side. The
onset, duration and unilaterality of the pupillary constriction d o not relate this phenomenon to the commonly
known pupillary reflexes. Though the interpretation is
uncertain, miosis induced by TES appears to be a
potential tool with which to assay the function of the iris
innervation. The results obtained by using this testing
method on patients affected from diseases, such as cluster
headache, which presumably involve iris sensory neurons,
seem to support its clinical usefulness [29].
ACKNOWLEDGMENTS
This study was supported by the Research National
Council (Rome, Italy), Medicina Preventiva e Riabilitativa, sottoprogetto ‘Controllo del Dolore’ grant no.
87.00272.56, and by the Ministry of Education of Italy.
We thank Ms Sherry Saehlenou for revision of the
English.
REFERENCES
1. McSwiney, B.A. & Suffolk, S.F. Segmental distribution of
certain afferent neurons on the pupillodilator reflex in the
cat. J. Physiol. (London) 1938; 93, 104-21.
2. Alder, F.H. Physiology of the eye, clinical application. 2nd
ed. St Louis:The C.V. Mosby Company, 1953; 169.
46 1
3. Bruce, A.N. Uber die Cieziehung der sensiblen Nervendigungen zum Entzundungsvorgang. Naunyn Schmiedebergs
Arch. Exp. Pathol. Pharmacol. 1910; 63,424-33.
4. Bayliss, W.M. On the origin from the spinal cord of the
vasodilator fibres of the hind limb, and the nature of these
fibres. J. Physiol. (London) 1901; 26,.173-209.
5. Longworthy, 0. & Ortaga, L. Innervation of the iris of the
albino rabbit as related to its function; theoretical discussion
of abnormalities of the .pupils
. observed in man. Medicine
1943; 22,287-98.
6. Perkins. E.S. Influence of the fifth cranial nerve on the intraocular pressure of the rabbit eye. Br. J. Ophthalmol. 1957;
6,93-133.
7. Bill, A., Stjernschantz, J., Mandhal, A., Brodin, E. &
Nilsson, G. Substance P: release on trigeminal nerve stimulation, effects in the eye. Acta Physiol. Scand. 1979; 106,
37 1-3.
8. Lembeck, F. & Gamser, R. Substance P in the peripheral
sensory process. In: Porter, R. & O’Connor, M., eds. Substance P in the nervous system. vol. 1. Ciba Foundation
Symposium, London: Pitman Books, 1982; 35.
9. Jampol, L.M., Axelrod, A. & Tessler, H. Pathways of the
eye’s response to topical nitrogen mustard. Invest. Ophthalmol. 1976; 15,486-9.
10. Butler, J.M. & Hammond, B.R. The effects of sensory
denervation on the responses of the rabbit eye to prostaglandin E, bradykinin and substance P. Br. J. Pharmacol.
1980; 69,495-502.
1 1 . Butler, J.M., Unger, W.G. & Cole, D.F. Axon reflex in ocular
injury: sensory mediation of the response of the rabbit eye
to laser irradiation of the iris. Q. J. Exp. Physiol. 1980; 65,
261-9.
12. Holmdhal, G.R., Hakanson, S., Leander, S., Rosell, K.,
Folkers, K. & Sundler, F. A substance P antagonist, [o-Pro,
~-Trp]-sp,inhibits inflammatory responses in the rabbit eye.
Science 1981;214, 1029.
13. Hokfelt, T., Kellerth, J.O., Nilsson, G. & Pernow, B. Experimental immunohistochemical studies on the localisation and
distribution of substance P in cat primary sensory neurones.
Brain Res. 1975; 100,235-52.
14. Nawa, H., Doteuchi, M., Inouye, K. & Nakanishi, S. A novel
mammalian tachykinin that differs from substance P in its
pharmacological profile. Life Sci. 1983; 34, 1 153-60.
15. Hua, X.Y., Saria, A., Gamse, R. et al. Multiple tachykinins
(neurokinin A, neuropeptide Y , and substance P) in
capsaicin-sensitive neurons in the guinea-pig. Regul. Pept.
1985; 133,1-19.
16. Ueda, N., Muramatsu, I., Taniguchi, T., Nakanishi, S. &
Fujiwara, M. Effects of neurokinin A, substance P, and electrical stimulation on the rabbit iris sphincter muscle. J.
Pharmacol. Exp. Therm. 1986; 239,494-9.
17. Fusco, B.M., Alessandri, M., Rovero, P., Maggi, C.A.,
Geppetti, P. & Fanciullacci, M. Miotic effect of substance P
eyedrops in humans. Regul. Pept. 1988; 22,7 1.
18. Woolf, C.J. Transcutaneous electrical nerve stimulation and
the reaction to the experimental pain in human subjects.
Pain 1979; 7, 1 15-27.
19. Chung, J.M., Lee, K.H., Hori, Y., Endo, K. & Willis, W.D.
Factors influencing peripheral nerve stimulation produced
inhibition of primate spinothalamic tract cells. Pain 1984;
19,277-85.
20. Rowell, L.B. Active neurogenic vasodilatation in man. In:
Vanhoutte, P.M. & Leusen, J., eds Vasodilatation. New
York: Raven Press, 198 I; 1.
21. Leander, R.S., Hakanson, S., Rosell, K., Folkers, F., Sundler,
F. & Tornqvist, K. A specific substance P antagonist blocks
smooth muscle contractions induced by non-cholinergic,
non-adrenergic nerve stimulation. Nature (London) I98 1;
294,467-9.
22. Fanciullacci, M., Pietrini, U., Geppetti, P., Nicolodi, M.,
Curradi, C. & Sicuteri, F. Substance P in the human iris:
possible involvement in echothiophate-induced miosis
in cluster headache. Cephalalgia 1988; 8,49-53.
462
B. M. Fusco et al.
23. Kjartansson, J., Lundeberg, T., Samuelson, U.E. & Dalsgaard, C.J. Transcutaneous electrical nerve stimulation
(TENS) increases survival of ischaemic musculocutaneous
flaps. Acta Physiol. Scand. 1988; 134, 95-9.
24. Kjartansson. J., Lundeberg, T., Samuelson, U.E., Dalsgaard,
C.J. & Heden, P. Calcitonin gene-related peptide (CGRP)
and transcutaneous electrical nerve stimulation (TENS)
increase cutaneous blood flow in a musculocutaneous flap
in the rat. Acta Physiol. Scand. 1988; 134,89-94.
25. Brain, S.D., Williams, T.J., Tippins, J.R., Morns, H.R. 22
Macintyre, 1. Calcitonin gene-related peptide is a potent
vasodilator. Nature (London) 1985; 313,54-6..
26. Gamse, R. & Saria, A. Potentiation of tachykinin-induced
plasma extravasation by calcitonin gene-related peptide.
Eur. J. Pharmacol. 1985; 114,6 1-6.
27. Szolcsanyi, J. Antidromic vasodilatation and neurogenic
inflammation. Agents Action 1988; 23,4- 1 1.
28. Kayahara, T., Takimoto, T. & Sakashita, S. Synaptic junctions in the cat spinal ganglion. Brain Res. 1981; 216,
277-90.
29. Fanciullacci, M., Fusco, B.M., Alessandri, M., Campagnolo,
V. & Sicuteri, F. Unilateral impairment of pupillary response
to trigeminal nerve stimulation in cluster headache. Pain
1989;36,185-91.