Review

Neuropeptides in migraine and cluster headache

L Edvinsson, PJ Goadsby1

Department of Internal Medicine, University Hospital, Lund, Sweden; Department of Neurology1, The Prince Henry Hospital,
University of New South Wales, Australia

Cephalalgia

Edvinsson L, Goadsby PJ. Neuropeptides in migraine and cluster headache. Cephalalgia 1994;14: 320-7. Oslo. ISSN
0333-1024

The cerebral circulation is invested by a rich network of neuropeptide Y (NPY) and noradrenaline containing
sympathetic nerve fibers in arteries, arterioles and veins. However, the nerve supply of vasoactive intestinal peptide
(VIP), substance P (SP) and calcitonin gene-related peptide (CGRP) containing fibers is sparse. While noradrenaline
and NPY cause vasoconstriction, VIP, SP and CGRP are potent vasodilators. Stimulation of the trigeminal ganglion in
cat and man elicits release of SP and CGRP. Subjects with spontaneous attacks of migraine show release of CGRP in
parallel with headache. Cluster headache patients have release of CGRP and VIP during bouts. Treatment with
sumatriptan aborts headache in migraine and cluster headache as well as the concomitant peptide release.

L Edvinsson, Department of Internal Medicine, University Hospital S-221 85 Lund, Sweden. Received 20 May 1994,
accepted 25 May 1994

Despite considerable advances in the neurosciences over the last three decades, until comparatively recently
remarkably little new light had been shed on the very distressing and common condition of migraine. Migraine affects
up to 15% of the adult population, while cluster headache is rare. Recorded in clinical writings since the antiquities and,
in its form with aura, well and widely recognized by both the medical profession and the lay public, the condition defies
understanding. It is perhaps a reflection of the enigma that even now argument continues as to whether the condition is
primarily neural or vascular.

The studies by Ray and Wolff (1) showing that large cerebral and meningeal arteries were sensitive to noxious
stimuli directed interest to the walls of the intracranial vessels. Are there sensors? How is the information propagated?
Is there a neuroanatomical link to explain referred pain? Are there connections to the brainstem and the central
neuronal components of the pain system?

Initial investigations of the cerebral vessels focused on the "classical" autonomic transmitters noradrenaline and
acetylcholine (2); however, with the development of immunocytochemical techniques, attention was directed towards
new neurotransmitter candidates such as neuropeptides. There is now ample immunocytochemical evidence that the
cerebral vessels of laboratory animals are supplied by perivascular nerve fibers containing several biologically active
peptides such as neuropeptide Y (NPY) vasoactive intestinal peptide (VIP), substance P (SP), and calcitonin
gene-related peptide (CGRP) (for review see 3). On the other hand, the peptidergic innervation of human cerebral
vessels has in the past received little attention (4-8). Recently we described the human cerebrovascular innervation in
great detail in this journal (9).

The present review highlights the recent contributions from studies of neuropeptides that link the basic
understanding of cerebrovascular physiology to the pathogenesis of migraine.

Definition

The studies analysed in this review are divided in accordance with the operational definition of the Classification
Committee of the International Headache Society. Migraine is thus defined as headache lasting usually from 4 to 72 h,
often unilateral, with a pulsating quality, of moderate or severe intensity, and aggravated by activity. It is episodic and
may often be associated with nausea, vomiting, or photo-phobia. There should be no neurological abnormality of
relevance on either clinical examination or investigation (10). In clinical practice, two varieties of migraine attack are
commonly distinguished: attacks associated with an aura of neurological symptoms (previously called classical
migraine) and headache of a periodic type without aura (common migraine).

Neuroanatomy and neurochemistry

With the advances in the understanding of the neurobiology of cranial pain processing that have been made over the
last few years it is necessary to review important aspects of the innervation of the cranial vessels. The data of most
importance to the understanding of headache mechanisms are those pertaining to the neural innervation of the large
cerebral and extracerebral vessels, including the pain-sensitive intracranial venous sinuses. Broadly speaking,
innervation of the cerebral circulation is divided into intrinsic and extrinsic neural systems. The intrinsic systems
are those that commence within and do not exit the central nervous system, possibly innervating
intra-parenchymal vessels (11).

The extrinsic system consists of neurons whose pathways ultimately commence in the central nervous
system and then exit synapsing in extra-parenchymal ganglia to innervate target vessels. The three systems,
all well described, are the sympathetic system, the parasympathetic system, and the trigeminal system (Fig.
1). Each is defined anatomically and marked by neuropeptides that give clues to their possible normal
function and role in pathology.

The sympathetic system arises in hypothalamic neurons passing to the intermediolateral cell column of
the spinal cord and synapsing before proceeding out to the superior cervical ganglion. Here they again
synapse and give rise to the fibers that innervate the vessels. This system is marked by the transmitters or
modulator substances noradrenaline, neuropeptide Y (NPY), and possibly adenosine triphosphate (ATP). It is
fundamentally a vascoconstrictor pathway (12). The parasympathetic system arises from cell bodies in the
superior salivatory nucleus passing out with fibers of the facial nerve (VII cranial nerve) (13) and synapsing in
the sphenopalatine and otic ganglia (14, 15). In some species there are additional microganglia on the
internal carotid artery (16). This system is marked by the neurotransmitters acetylcholine, VIP, and peptide
histidine isoleucine [methionine; PHI(M)] In addition, pituitary adenylate cyclase activating peptide (17),
helospectin (18), and nitric oxide synthetase (NOS) activity (19) have been found in the parasympathetic
nerves. Physiologically, the system is vasodilatory. Lastly, there is a sensory innervation of the cranial vessels
and dura mater from the trigeminal system. The cell bodies are bipolar and located in the trigeminal
ganglion. They make functional first-order connections with neurons in the trigeminal nucleus caudalis and in
its revealed extension down to the C2 level (20). This system is marked by CGRP, SP, and neurokinin A. It is
also an essentially vasodilatory system as well as having a primary sensory function.

Peptidergic nerves in man

Immunocytochemistry

Different human cerebral vessels have been examined using either autopsy specimens of larger cerebral
arteries or cortical vessels obtained during neurosurgical tumor resections (9). All major arteries belonging to
the circle of Willis (post-mortem tissue) are supplied with a very dense innervation. This is well illustrated
using immunofluorescence. Numerous NPY immunoreactive fibers can be seen, whereas the supply of VIP
and SP/CGRP is less dense. There is a close similarity in the innervation pattern for NPY and TH
immunoreactivity. The supply of CGRP and SP as well as of VIP is sparse. A markedly different degree of
density is seen in whole mount preparations of human cortical arteries and veins obtained during
neurosurgical tumor resections. The density of NPY and TH immunoreactivity is quite marked. There is a
close similarity in the distribution of NPY and TH immunoreactive fibers. Also the human cortical veins are
supplied by a moderate number of NPY and TH immunoreactive fibers. Here, too, there is a close correlation
between NPY and TH immunoreactivity. Only a few scattered fibers were seen to contain VIP, CGRP, or SP
immunoreactive fibers, despite the use of fresh tissue (9).

Characterization of neuropeptides

HPLC extracts of the human cerebral arteries have revealed that NPY, VIP, and CGRP immunoreactiv-
ity elute each in one major component with the same elution volume as the synthetic human NPY, VIP, and human
a-CGRP, respectively. SP elutes as two peaks, one major peak with the same retention volume as SP, and in a
second peak with the same elution volume as the oxidized SP (6). Quantitative measurements of NPY in the
human middle cerebral artery revealed concentrations of 3.0 ± 2.4 pmol/g. The VIP concentration was 2.3 ± 0.7
pmol/g, CGRP 3.9 ± 1.1 pmol/g, and SP 2.3 ± 0.8 pmol/g in the middle cerebral artery.

Vasomotor responses

Administration of NPY elicits a concentration-dependent contraction of cerebral arteries (4). The responses
induced by NPY are somewhat stronger than that of NA, and the concentration of agonist eliciting half maximum
contractions (pD2) markedly lower for NPY than for NA. The contractile response to NA is antagonized in a
competitive fashion by the a1-adrenoceptor antagonist prazosin. The NPY-induced contractions are instead
mediated via the Y1 subtype of NPY receptors.

In precontracted vessel segments, ACh, VIP, or PHM cause dilatation (4, 6). The relative potency for relaxation
in the cerebral arteries is VIP > PHM > ACh. There are significant differences in potency between VIP and
PHM-27, and between VIP and ACh; the amount of relaxation varies between 76 and 87%.

CGRP, SP, and NKA also act as strong vasodilators. The order of potency is CGRP > SP > NKA. CGRP
induces relaxation with the highest Imax (91%) compared to SP (70%) and NKA (80%). The existence of specific
peptide receptors is supported by agonist and antagonist studies.

Animal studies

Before turning to work directly carried out in man during headache, the data from animal studies will be considered
in brief.

Trigeminal ganglion stimulation

It is clear, albeit intuitively so, that the trigeminal system must play a key role in migraine. It provides the only
known pain-sensitive innervation to the cranial vasculature by virtue of the central convergence Of trigeminal and
upper cervical pain inputs. This takes place at the second-order neurons and has been discussed above. Studies
have examined what takes place when these pathways are activated.

Cerebral blood flow.

Although having no apparent tonic effect on resting cerebral blood flow or glucose utilization (21), trigeminal
ganglion stimulation increases both bulk extracerebral (22) and regional cerebral blood flow, specifically in the
frontal and parietal cortex (23). The effect upon extracerebral blood flow is mediated through strong reflex
connections with the facial nerve dilator system, which relays some 80% of the dilatation with the remaining 20%
mediated antidromically (24). The cerebral effects are largely mediated by the facial nerve dilator system (23)
through a classical autonomic nicotinic ganglion (25) employing vasoactive intestinal polypeptide as the final
transmitter in the system (26).

Neurogenic inflammation.

Series of experiments have examined alterations in vascular permeability and ultrastructural changes in the dura
mater of the rat associated with trigeminal ganglion stimulation. It has been demonstrated that trigeminal ganglion
stimulation increases vascular permeability in the dura (27) and that there is coincident mast cell degranulation
(28). Both these effects are aborted by pretreatment with the potent antimigraine drugs dihydroergotamine and
sumatriptan but 5HT is inactive in this model (29, 30).

Peptides.

Trigeminal ganglion stimulation results in local (cranial) release of both CGRP and SP in the cat, both acting as
markers of trigeminal activation (31). In a similar study in rat, CGRP has been found to be elevated in the superior
sagittal sinus (32). The CGRP increase is attenuated by sumatriptan (31, 32). Studies of trigeminal ganglion
stimulation are hampered by the fact that the ganglion contains cell bodies of not only nociceptive neurons but also
proprioceptive, light touch, and others. In short, it has all the sensory modalities of the head and is thus
non-specific in respect of pain. We have therefore turned to the superior sagittal sinus. This structure is pain
sensitive in man (33) and is predominantly innervated by C fibres.

It is clear from electrophysiological studies that most of the neurons excited by sinus stimulation are either
nociceptive-specific or wide dynamic range (including a nociceptive component) (34). Sagittal sinus stimulation
increases regional cerebral blood flow more than trigeminal ganglion stimulation does (61) and leads to the local
release of both CGRP and VIP (35). Notably, there is no change in either SP or NPY.

Sphenopalatine ganglion stimulation

Anatomical and physiological evidence suggests that VIP-containing and NOS-containing nerves on cerebral
vessels arise from the parasympathetic neural innervation of the feline cerebral circulation (36). Specifically, it is
demonstrated that while there is a
strong percentage of NOS-positive cells in the sphenopalatine ganglion, there is only a very sparse amount of
NOS positivity in the trigeminal ganglion. The coexistence of NOS-like and VIP-like immunoreactivity implies
that the same neurons produce several neurotransmitter candidates. Previous studies have revealed that the
same nerve cell body in the sphenopalatine ganglion may contain neurotransmitter candidates such as
acetylcholine, VIP (37), helospectin (18), and pituitary adenylate cyclase activating peptide (17). VIP fibers to
the cerebral circulation originate in the sphenopalatine and otic ganglia and in miniganglia located along the
carotid artery (38). It is conceivable that the NOS-immunoreactive fibers originate in these ganglia, since one
study has revealed a rich supply of NOS-immunoreactive nerve cell bodies in the sphenopalatine ganglion
(19) that also possess VIP immunoreactivity (36). This possibility raises broader questions about NOS and
VIP in the cerebral circulation. Does NOS blockade alter resting cerebral blood flow? Blockade of NOS in rat
(39, 40), cat (41), and dog (42) reduces resting cerebral blood flow.

There is excellent anatomical evidence that describes the parasympathetic innervation of the cerebral
circulation. The fibers exit from the brainstem in the VII nerve and into the internal auditory meatus, pass
through the geniculate ganglion without synapsing and traverse the greater superficial petrosal nerve to
synapse in the sphenopalatine, otic and internal carotid miniganglia. The peripheral distribution of the fibers to
the cerebral vessels from the sphenopalatine ganglion is via the ethmoid nerve as it traverses the ethmoid
foramen. Stimulation at each of these points, including the most recent observations of the chemical
stimulation of the superior salivatory nucleus (43), can increase cerebral blood flow.

Given the distinct probability that activation of the neural input to the cerebral circulation can alter the
traditional limitations of flow/metabolism coupling in the brain, what might be the purpose of such a system?
The most likely role for such a vasodilator system would be in the situation in which a brain region is
underperfused and requires some increase in flow that might not be readily generated by metabolic changes.
Such a situation may be seen during brain ischemia and indeed it has been shown that extirpation of the
major parasympathetic outflow ganglion, the sphenopalatine, increases infarction size in the rat middle
cerebral artery occlusion model of stroke (44). The other most likely situation for a requirement of
vasodilatation may occur in situations of arterial spasm, such as that seen in subarachnoid hemorrhage, or to
counteract in oligemia, such as may be seen in migraine with aura (45). In the latter case, it may be of
relevance that in some patients with migraine with aura increased levels of VIP are found in the external
jugular véin blood early in the headache phase (46) and more recently we have similar results in cluster
headache (47, 48). These observations provide a plausible link from our physiological studies into man and
suggest a direction for future studies.

Human data

Human data will be considered under the three major headache categories that have been studied,
migraine with and without aura and cluster headache. It is important to note, by way of linking the basic
science to the clinical data, some observations of direct trigeminal ganglion stimulation in man. Patients
under treatment for trigeminal neuralgia by thermocoagulation of the ganglion were noted to flush ipsilateral
to the side of stimulation. As this flush occurred, blood samples were taken from the external jugular vein and
compared with immediately precoagulation samples. There was a marked increase in both CGRP and SP in
the patients who flushed compared to the control precoagulation levels (31). It then appears that much of the
work examining the trigeminovascular system in animals is corroborated by these studies in humans.

Migraine with aura (classical migraine)

Early studies of this condition were confined to either peripheral venous blood or cerebrospinal fluid. In
peripheral blood, VIP and SP levels are not altered during headache (49). Furthermore, the often described
premonitory symptoms of altered bladder habit, particularly increased urination (50), have led to the
suggestion that these may reflect hypothamamic disturbance and therefore be a manifestation of an
essentially central nervous system disorder (51). Analysis of vasopressin levels during headache
(unchanged) has not supported this hypothesis, however (49). In an attempt to examine brain function more
directly, cerebrospinal fluid levels of various substances have been determined. Morphine-like substances
have been reported to be reduced during migraine (52), although no change in ß-endorphin can be detected
(53). Interestingly, plasma and platelet met-enkephalin levels are moderately increased during headache
(54, 55).

Considering the contrasting results of these studies and the results of trigeminal ganglion stimulation in
man, the levels of various neuropeptides were re-examined in man during migraine. Blood samples were
drawn from the external jugular vein during headache. Using markers for the sympathetic (NPY),
parasympathetic (VIP), and trigeminovascular systems (CGRP, SP), the patients were assessed. There were
no changes in the peripheral blood in any of the peptides studied or in NPY, VIP, or SP in the
external jugular blood. There were, however, marked increases in CGRP (Fig. 2) during migraine in man from control levels of 40
pmol/1 to 91 ± 11 pmol/l during headache (46). In two patients with some symptoms similar to cluster headache patients, there
were increases in VIP, but these were not significant for the whole group. We have confirmed these observations in a further
series of patients who had blood sampled during headache and then after administration of sumatriptan. Remarkably, the CGRP
levels return to normal after administration of sumatriptan and successful amelioration of the headache (56, 57). The results of
these studies are summarized in Table 1.

Migraine without aura (common migraine)

A similar picture emerges for migraine without aura as has been seen for migraine with aura. Plasma levels of SP and VIP are
normal both during an attack and interictally (49). Again, as with simple migraine with aura, plasma vasopressin levels are
unchanged during headache (49). Conflicting data have emerged concerning the opioids; in two studies met-enkephalin and
ß-endorphin have been shown to be marginally elevated (54, 55), while in other studies no change has been found specifically in
ß-endorphin (53). Studies of the cerebrospinal fluid have been limited to demonstrating that there is no change in either SP or
somatostatin during headache (58).

Table I. Peptide levels during migraine.

Migraine
Aura No aura Treated
Trigeminal
substance P N N -
CGRP - - N
Sympathetic
neuropeptide Y N N -
Parasympathetic
VIP N N N
PACAP ? ? ?
Opioids
ß-endorphin N N ?
met-enkephalin - - ?
Others
ACTH N N ?
vasopressin N N ?

ACTH, adrenocorticotrophin; CGRP, calcitonin gene-related peptide; N indicates the levels were within the normal range for
the population studied; PACAP, pituitary adenylate cyclase activity peptide; VIP, vasoactive intestinal polypeptide; ? data not
available; - increased levels; N, normal levels.

Again, we have studied patients with migraine without aura during headache by sampling blood from the external jugular vein
(Table 1). As was the case with migraine with aura, substantial increases in CGRP levels were found in the jugular blood while no
changes in substance P, VIP, or NPY were seen in either jugular or peripheral blood (46). Also, more recently, Nicolodi has used
the relatively novel method of saliva sampling and peptide analysis in association with migraine without aura. They have seen
increases in both SP and CGRP along with a reduction in VIP during headache (59). The difference in the changes seen with
these techniques probably represents the different field of sampling. While the cerebral circulation has a preferentially CGRP-rich
trigeminal innervation (60) the innervation of the salivary glands is not similarly characterized. The changes in VIP suggest that
some reflex parasympathetically mediated event is also taking place, but this requires further characterization (35, 61).

Cluster headache

Although cluster headache would seem to be an ideal condition to examine, in that it is a well-de-scribed clear-cut clinical
syndrome in which a number of attacks can be sampled, until recently studies of neuropeptide release have not been particularly
helpful. Plasma levels of SP and somatostatin show only very modest change (62), although met-enkephalin and ß-endorphin
levels show a consistent elevation during headache (63, 64). Cerebrospinal fluid changes have been similar with only changes in
met-enkephalin being noted reliably (Table 2) (65). Interestingly, both substance P and VIP levels are elevated
 Table 2. Peptide levels during cluster headache.
Cluster headache
Plasma CSF Saliva
Trigeminal
substance P N N -
CGRP - ? N
Sympathetic
neuropeptide Y N ? ?
Parasympathetic
VIP - N -
PACAP ? ? ?
Opioids
ß-endorphin - N ?
met-enkephalin - - ?
Others
somatostatin N N ?

CGRP, calcitonin gene-related peptide; N indicates the levels were within the normal range for the population studied; PACAP,
pituitary adenylate cyclase activity peptide; VIP, vasoactive intestinal polypeptide, ? data not available, - increased levels; N, normal
levels.

in the saliva during headache. This may well indicate activation of the central and peripheral arms of the trigeminovascular reflex (66),
but further studies are needed to clarify the relationship between salivary peptide changes and the problem of headache.

Cluster headache is a rare severe disorder that is clinically well characterized with a relatively poorly understood pathophysiology. In
a recent study patients with episodic cluster headache fulfilling the criteria of the International Headache Society were examined during
an acute spontaneous attack of headache to determine the local cranial release of neuropeptides (47, 48). Blood was sampled from the
external jugular vein ipsilateral to the pain before and after treatment of the attack. Samples were assayed for CGRP, VIP, SP and NPY.
Attacks were treated with either oxygen inhalation, sumatriptan, or an opiate. Thirteen patients were studied (10M, 3F). All had
well-established typical attacks of cluster headache when blood was sampled. During the attacks, external jugular vein blood levels of
CGRP (Fig. 3) and VIP were raised while there was no change in NPY or SP. CGRP levels rose to 110 ± 7 pmol/l (RR: < 40), while VIP
levels rose to 20 ± 3 pmol/l (RR: < 7). Treatment with both oxygen and sumatriptan (subcutaneous) reduced the CGRP level to normal,
while opiate administration did not alter the peptide levels. These data demonstrate for the first time in vivo human evidence for
activation of the trigeminovascular system and the cranial parasympathetic nervous system in an acute attack of cluster headache.
Furthermore, it is shown that both oxygen and sumatriptan, while aborting the attacks, terminate activity in the trigeminovascular
system.

In summary, CGRP, which marks the trigeminovascular system, and VIP, which marks parasympa

thetic activity, are both elevated in the cranial venous blood of patients with an acute spontaneous cluster headache. The termination of
the attack with either sumatriptan or oxygen causes normalization of the CGRP levels, reflecting cessation of activity in the
trigeminovascular system, whereas pain relief along with an opiate agonist apparently terminates the attack but does not immediately
cause trigeminovascular activity to cease. The finding of both CGRP and VIP in the cranial venous blood suggests that there is
activation of a brainstem reflex, the afferent arc of which is the trigeminal nerve and the efferent the cranial parasympathetic outflow
from the VIIth nerve. Cluster headache is a severe disorder which is clinically well recognized and serves as an ideal situation in which
to examine some important aspects of the neural innervation of the cranial circulation which provides general physiological information
applicable to our understanding of vascular headaches such as migraine.

Conclusions

The study of neuropeptide levels in migraine and cluster headache is now providing the link between the clinic and research work that is
so crucial if the basic pathophysiology if the problem is to be determined. In both migraine with and without aura there are marked
changes in cranial levels of CGRP indicating activation of the trigeminal system. These levels are rendered normal by the new and
highly effective antimigraine agent sumatriptan coincident with the relief of the headache. Similarly, CGRP is released with trigeminal
activation in studies on animals and, at least for the rat, the changes are also inhibited by sumatriptan. These data substantiate the
usefulness of the animal models employed. Future work is underway to explore the neural innervation
of the cerebral and extracerebral vessels which, along with pharmacological studies of the transmitters involved, accordingly, should permit both a better
understanding of the pathogenesis of migraine and better treatment.

Acknowledgments.-This study was supported by the Swedish Medical Research Council (project no. 5958), and by the Faculty of Medicine, Lund University.

References

1. Wolff HG. Headache and other head pain. New York: Oxford University Press, 1963

2. Edvinsson L, Owman C, Sjöberg N-O. Autonomic nerves, mast cells, and amine receptors in human brain vessels. A histochemical and pharmacological
study. Brain Res 1976;115:337-93

3. Edvinsson L. Role of perivascular peptides in the control of the cerebral circulation. Trends Neurosci 1985;8:126-31

4. Edvinsson L, Ekman R, Jansen I, Ottosson A, Uddman R. Peptide containing nerve fibres in human cerebral arteries: immunocytochemistry,
radioimmunoassay and in vitro pharmacology. Ann Neurol 1987;21:431-7

5. Edvinsson L, Ekman R. Distribution and dilatory effect of vasoactive intestinal polypeptide (VIP) in human cerebral arteries. Peptides 1984;5:329-31

6. Jansen I, Uddman R, Ekman R, Olesen J, Ottosson A, Edvinsson L. Distribution and effects of neuropeptide Y, vasoactive intestinal peptide, substance P and
calcitonin gene-related peptide in human middle meningeal arteries: comparison with cerebral and temporal arteries. Peptides 1992;13:527-36

7. Jansen I, Uddman R, Hocherman M, Ekman R, Jensen K, Olesen J, et al. Localization and effects of neuropeptide Y, vasoactive intestinal polypeptide,
substance P and calcitonin gene-related peptide in human temporal arteries. Ann Neurol 1986;20.496-501

8. Allen JM, Todd N, Crockard HA, Schon F, Yeats JC, Bloom SR. Presence of neuropeptide Y in human circle of Willis and its possible role in cerebral
vasospasm. Lancet 1984;ii: 550-2

9. Edvinsson L, Jansen I, Cunha e Sa M, Gulbenkian S. Demonstration of neuropeptide containing nerves and vasomotor responses to perivascular peptides in
human cerebral arteries. Cephalalgia 1994;14:88-96

10. Olesen J. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988;8 Suppl 7:1-96

11. Lou HC, Edvinsson L, MacKenzie ET. The concept of coupling blood flow to brain function: revision required? Ann Neurol 1987;22:289-97

12. Olesen J, Edvinsson L, editors. Basic mechanisms of headache. Amsterdam: Elsevier, 1988:1-492

13. Chorobski J, Penfield W. Cerebral vasodilator nerves and their pathway from the medulla oblongata. Arch Neurol Psychiatry 1932;28:1257-89

14. Hara H. Hamill GS, Jacobowitz DM. Origin of cholinergic nerves to the rat major cerebral arteries: coexistence with vasoactive intestinal polypeptide. Brain
Res Bull 1985;14: 179-88

15. Walters BB, Gillespie SA, Moskowitz MA. CerebrovascuIar projections from the sphenopalatine and otic ganglion to the middle cerebral artery of the cat.
Stroke 1986;17:488-94

16. Gibbins IL, Brayden JE, Bevan JA. Perivascular nerves with immunoreactivity to vasoactive intestinal polypeptide in cephalic arteries of the cat: distribution,
possible origins and functional implications. Neuroscience 1984;13: 1327-46

17. Uddman R, Goadsby PJ, Jansen I, Edvinsson L. PACAP, a VIP-like peptide, immunohistochemical localization and effect upon cat pial arteries and on
cerebral blood flow. J Cereb Blood Flow Metab 1993;13:291-7

18. Uddman R, Goadsby PJ, Jansen I, Edvinsson L. Helospectin-like peptides: immunohistochemical localization and effects on cat pial arteries and on cerebral
blood flow. J Cereb Blood Flow Metab 1993;13 Suppl 1:S206

19. Noziaki K, Markowitz MA, Maynard KI, Koketsu N, Dawson TM, Bredt DS, et al. Possible origins and distribution of immunoreactive nitric oxide
synthase-containing nerve fibers in cerebral arteries. J Cereb Blood Flow Metab 1993; 13:70-9

20. Goadsby PJ, Zagami AS. Stimulation of the superior sagittal sinus increases metabolic activity and blood flow in certain regions of the brainstem and upper
cervical spinal cord of the cat. Brain 1991;114:1001-11

21. Edvinsson L, McCulloch J, Kingman TA, Uddman R. On the functional role of the trigemino-cerebrovascular system in the regulation of cerebral circulation. In
Owman C, Hardebo JE, editors. Neural regulation of the cerebral circulation. Amsterdam: Elsevier, 1986;407-18

22. Goadsby PJ, Lambert GA, Lance JW. Stimulation of the trigeminal ganglion increases flow in the extracerebral but not the cerebral circulation of the monkey.
Brain Res 1986; 381:63-7

23. Goadsby PJ, Duckworth JW. Effect of stimulation of trigeminal ganglion on regional cerebral blood flow in cats. Am J Physiol 1987;253:R270-4

24. Lambert GA, Bogduk N, Goadsby PJ, Duckworth JW, Lance JW. Decreased carotid arterial resistance in cats in response to trigeminal stimulation. J
Neurosurg 1984; 61:307-15

25. Goadsby PJ. Characteristics of facial nerve elicited cerebral vasodilatation determined with laser Doppler flowmetry. Am J Physiol 1991;260:R255-62

26. Goadsby PJ, Shelley S. High frequency stimulation of the facial nerve results in local cortical release of vasoactive intestinal polypeptide in the anesthetised
cat. Neurosci Left 1990;112:282-9

27. Markowitz S, Saito S, Markowitz MA. Neurogenically mediated leakage of plasma protein occurs from blood vessels in dura but not brain. J Neurosci
1987;7:4129-36

28. Dimitriadou V, Buzzi MG, Lambracht-Hall M, Moskowitz MA, Theoharides TC. In vivo and in vitro ultrastructural evidence for stimulation of dural mast cells by
neuropeptides. Proc Soc Neurosci 1990;16:161

29. Buzzi MG, Moskowitz MA. The antimigraine drug, sumatriptan (GR 43175), selectively blocks neurogenic plasma extravasation from blood vessels in dura
mater. Br J Pharmacol 1990;99:202-6

30. Buzzi MG, Moskowitz MA, Peroutka SJ, Byun B. Further characterization of the 5-HT1 receptor mediating blockade of neurogenic plasma extravasation within
dura mater. Br J Pharmacol 1991;103:1421-8

31. Goadsby PJ, Edvinsson L, Ekman R. Release of vasoactive peptides in the extracerebral circulation of humans and the cat during activation of the
trigeminovascular system. Ann Neurol 1988;23:193-6

32. Buzzi MG, Carter WB, Shimizu T, Heath III H, Moskowitz MA. Dihydroergotamine and sumatriptan attenuate levels of CGRP in plasma in rat superior sagittal
sinus during electrical stimulation of the trigeminal ganglion. Neuropharmacology 1991;30:1193-200

33, McNaughton FL. The innervation of the intracranial blood vessels and the dural sinuses. In: Cobb S, Frantz AM, Penfield W, Riley HA, editors. The circulation
of the brain and spinal cord. New York: Hafner, 1966:178-200

34. Zagami AS, Lambert GA. Stimulation of cranial vessels

 excites nociceptive neurones in several thalamic nuclei of the cat. Exp Brain Res 1990;81:552-66

35. Zagami AS, Goadsby PJ, Edvinsson L. Stimulation of the superior sagittal sinus in the cat causes release of vasoactive peptides. Neuropeptides
1990;16:69-75

36. Goadsby PJ, Uddman R, Edvinsson L. Cerebral vasodilatation from parasympathetic nerves involves nitric oxide in the cat. Neuroscience 1994

37. Hara H, Jansen I, Ekman R, Hamel E, MacKenzie ET, Uddman R, et al. Acetylcholine and vasoactive intestinal peptide in cerebral blood vessels: effect of
extirpation of the sphenopalatine ganglion. J Cereb Blood Flow Metab 1989; 9:204-11

38. Suzuki N, Hardebo JE, Owman C. Origins and pathways of cerebrovascular vasoactive intestinal polypeptide-positive nerves in rat. J Cereb Blood Flow
Metab 1988;8:697-712

39. Kozniewska E, Oseka M, Stys T. Effects of endothelium derived nitric oxide on cerebral circulation during normoxia and hypoxia in the rat. J Cereb Blood
Flow Metab 1992; 12:311-7

40. Tanaka K, Gotoh F, Gomi S, Takashima S, Mihara B, Shirai T, et al. Inhibition of nitric oxide synthesis induces a significant reduction in local cerebral
blood flow in the rat. Neurosci Left 1991;127:129-32

41. Kovach AG, Szabo C, Benyo Z, Csaki C, Greenberg JH, Reivich M. Effects of NG-nitro-L-arginine and L-arginine on regional cerebral blood flow in the
cat. J Physiol 1992; 449:183-96

42. Saito S, Wilson DA, Hanley DF, Traystman RJ. Nitric oxide modulates cerebral blood vessel tone and depresses cerebrovascular resistance changes to
carbon dioxide in the anesthetized dog. J Cereb Blood Flow Metab 1993;13 Suppl 1:S153

43. Nakai M, Tamaki K, Ogata J, Maeda M. Parasympathetic cerebrovascular center of the facial nerve. Circ Res 1993; 72:470-5

44. Kano M, Moskowitz MA, Yokota M. Parasympathetic denervation of rat pial vessels significantly increases infarction volume following middle cerebral
artery occlusion. J Cereb Blood Flow Metab 1991;11:628-37

45. Olesen J, Larsen B, Lauritzen M. Focal hyperemia followed by spreading oligemia and impaired activation of rCBF in classic migraine. Ann Neurol
1981;9:344-52

46. Goadsby PJ, Edvinsson L, Ekman R. Vasoactive peptide release in the extracerebral circulation of human during migraine headache. Ann Neurol
1990;28:183-7

47. Goadsby PJ, Edvinsson L. Evidence of trigeminovascular activation in man during acute cluster headache. Cephalalgia 1993;13:30

48. Goadsby PJ, Edvinsson L. Human in vivo evidence for trigeminovascular activation in cluster headache. Neuropeptide changes and effects of acute
attacks therapies, Brain 1994;117:427-34

49. Blegvad N, Jensen K, Fahrenkrug J, Schaffalitzky de Muckadell OB, Olesen J. Plasma-VIP and substance P during migraine attack. Cephalalgia 1986;5
Suppl 2:252-3

50. Lance JW. Mechanisms and management of headache, 4th ed. London: Butterworth Scientific, 1992

51. Lance JW, Lambert GA, Goadsby PJ, Duckworth JW. Brain-stem influences on cephalic circulation: experimental data from cat and monkey of relevance
to the mechanism of migraine. Headache 1983;23:258-65

52. Anselmi B, Baldi E, Casacci F, Salmon S. Endogenous opioids in cerebrospinal fluid and blood in idiopathic headache sufferers. Headache
1980;20:294-9

53. Jensen K, Blegvad N, Fenger M, Jordal R, Olesen 'J. ß-Endorphin and ACTH in plasma during attacks of common and classic migraine. Cephalalgia
1985;5:177-82

54. Ferrari MD, Frolich M, Odkink J, Tapparelli C, Portielje JEA, Bruyn GW. Methionine-enkephalin and serotonin in migraine and tension headache. In: Rose
FC, editor. Advances in headache research. London: John Libbey, 1987: 227-34

55. Mosnaim AD, Wolf ME, Chevesich J, Callaghan OH, Diamond S. Plasma methionine enkephalin levels. A biological marker for migraine? Headache
1985;25:259-61

56. Goadsby PJ, Edvinsson L. Sumatriptan reverses the changes in calcitonin gene-related peptide seen in the headache phase of migraine. Cephalalgia
1991;11 Suppl 11:3-4

57. Goadsby PJ, Edvinsson L. The trigeminovascular system and migraine: studies characterizing cerebrovascular and neuropeptide changes seen in
humans and cats. Ann Neurol 1993;33:48-56

58. Vecchiet L, Geppetti P, Marchionni A, Spillantini MG, Fanciullacci M, Sicuteri F. Cerebrospinal fluid (methionin5)-enkephalin, substance P and somatostatin-like
immunoreactivities in painful and pain-less human diseases. In: Sicuteri F, Vecchiet L, Fanciullacci M, editors. Trends in cluster headache. Amsterdam: Excerpta Medica,
1987:135-43

59. Nicolodi M, Del Bianco E. Sensory neuropeptides (substance P, calcitonin gent-related peptide) and vasoactive intestinal polypeptide in human saliva:
their pattern in migraine and cluster headache. Cephalalgia 1990;10:39-50

60. O'Connor TP, Kooy van der D. Enrichment of a vasoactive neuropeptide (calcitonin gent-related peptide) in the trigeminal sensory projection to the
intracranial arteries. J Neurosci 1988;8:2468-76

61. Lambert GA, Goadsby PJ, Zagami AS, Duckworth JW. Comparative effects of stimulation of the trigeminal ganglion and the superior sagittal sinus on
cerebral blood flow and evoked potentials in the cat. Brain Res 1988;453: 143-9

62. Sicuteri F, Fanciullacci M, Geppetti P, Renzi D, Caleri D, Spillantini MG. Substance P mechanisms in cluster headache: evaluation in plasma and
cerebrospinal fluid. Cephalalgia 1985;5:143-9

63. Appenzeller O, Atkinson RA, Standefer JC. Serum b-endorphin in cluster headache and common migraine. In: Rose FC, Zilkha K, editors. Progress in
migraine research. Bath: Pit-man, 1981:106-9

64. Hardebo JE, Ekman R, Eriksson M, Holgersson S, Rydberg B, CSF opioid levels in cluster headache. Effect of acupuncture. In: Rose FC, editor.
Migraine: clinical and research advances. Basel: Karger 1985:79-85

65. Hardebo JE, Ekman R. Substance P and opioids in cluster headache. In: Sicuteri F, Vecchiet L, Fanciullaci M, editors. Trends in cluster headache.
Amsterdam/New York/Oxford: Elsevier, 145-58

66. Goadsby PJ, Lance JW, Brainstem effects on intra-and extracerebral circulations. Relation to migraine and cluster headache. In: Olesen J, Edvinsson L,
editors. Basic mechanisms of headache. Amsterdam: Elsevier 1988: 413-27