A special feature of cluster headaches is that during the active cluster period, vasodilating substances such as alcohol, histamine and nitroglycerin can very reliably provoke cluster attacks. Interestingly, these agents are completely inactive in provoking cluster attacks during the remission period. The most reliable effect can be achieved with histamine. When 0.3 mg histamine is administered sc, a pulsating, throbbing headache occurs within ten minutes.

With nitroglycerin, the time at which the headache develops corresponds to the maximum vasodilatory effect of the substance. only occurs after about 30 to 45 minutes Since the time of maximum vasodilation already been exceeded , it cannot be assumed that the cluster attack induction is a direct consequence of vasodilation . This is also supported by the fact that a cluster attack cannot be provoked at any time nitroglycerin There is a refractory period for several hours after a spontaneous attack. postponed in time by an experimentally provoked attack .

Similar to histamine, a latency period of 30 to 45 minutes observed before a cluster attack occurs alcohol Alcohol can provoke a cluster attack during the active phase in approximately 50% , there is also a dose effect . While small amounts of alcohol provoke a cluster provocation, large amounts of alcohol can delay and gaps of up to three days between individual cluster attacks in the active period. In some patients, however, a rebound mechanism with a temporarily increased attack frequency.

Hypoxia also considered a possible attack provocation. This observation is initially based on the fact that administering pure oxygen during a cluster attack can quickly reduce cluster headaches. Nocturnal occurrence also been linked to reduced oxygen saturation at night and possible sleep apnea It is also possible that during the cluster attack a reduced supply impaired central autoregulation . This finding is based on the observation that oxygen saturation after nitroglycerin administration is more pronounced and lasts longer in patients with cluster headache attacks than in control subjects.


Evidence for a disorder of phospholipid metabolism comes from a significant reduction in the choline concentration in the erythrocytes by approximately 50% compared to healthy control subjects. Choline is generated by phosphatidylcholine in the membranes. a normalization of the choline content in the erythrocytes occurs during a two-week treatment with lithium The renormalization is made possible by an inhibition of the choline outflow of the erythrocytes, which provides an explanation for the mechanism of action of lithium in the therapy of cluster headaches.


The use of anti-inflammatory substances , such as acetylsalicylic acid or indomethacin, does not major therapeutic effect on cluster headaches. Since these substances belong to the group of prostaglandin synthesis inhibitors, it is unlikely that prostaglandins play a special role in the genesis of cluster headaches. In fact, no significant differences in prostaglandin activity in cluster headaches could be revealed.


The leukotrienes play an important role in the induction of hyperalgesia, in the increase of vascular permeability, and in the reduction of nociceptive reactions when administered with bradykinin . There is evidence that during the remission phase of cluster headache the release of leukotriene B 4 and leukotriene C 4 significantly reduced compared to healthy volunteers .

In patients with cluster headache, an increased number of mast cells in the skin ipsilateral to the painful side, but also on the pain-free side, was revealed compared to healthy controls. The mast cells are the main store of histamine . Histamine can be found in elevated levels during a cluster headache attack Mast cell accumulation is found in patients with cluster headaches during the attack, particularly in the area of ​​the perivascular and cutaneous nerves. increased number of mast cell degranulations can be observed, which can probably be seen in connection with axon reflexes. Such increased degranulation processes can be found both during a cluster headache period and in the free interval.


The monoamine oxidase activity (MAO) of platelets reduced in affected patients both during the attack and between cluster attacks . There is a stronger drop within the cluster attacks. Furthermore, MAO activity is shown to greater thermostability in cluster headache patients compared to healthy controls. The reduced activity of MAO can be interpreted as an indication of impaired membrane function in cluster headache patients.

significantly increased concentration of norepinephrine and epinephrine during the acute cluster period has also been described in cluster headache patients . However, such increased concentrations were only found for the conjugated norepinephrine and epinephrine , but not for the free norepinephrine and epinephrine. An increase in norepinephrine can be observed both during a spontaneous cluster attack and after nitroglycerin-induced attacks. This increase could be interpreted as a normal response to vasodilation . Overall, the changes in monoamines in cluster headaches are not very specific . The changes could be attributed to physical movements, seasonal influences or even treatment.

regarding serotonin uptake into platelets. Both reduced intake and outright activity have been described.

amino acids

The excitatory amino acids , particularly glutamate, aspartate and glycine, exert important neurotransmitter functions related to nociceptive transmission . The behavior of these excitatory amino acids in the platelets is viewed as a model for the behavior in the neurons. It was shown that the glycine concentrations in the platelets significantly reduced . However, glutamate and aspartate did not show any changed concentrations compared to healthy people. The excitatory amino acids do not have different concentrations between the active phase and the remission phase.

Neuropeptides and opioids

In patients with episodic cluster headache, increased concentrations of met-enkephalin were found in plasma during an acute cluster attack. normal levels were found between attacks during a cluster period and during the remission period . In chronic cluster headache plasma met-enkephalin levels were found before the generation of an acute attack compared to the time during an acute attack or the time after an attack It can be assumed that the increase in plasma met-enkephalin is a secondary response to increased sympathetic activity during a cluster attack. The reason for this is that met-enkephalin is stored primarily in the adrenal glands and is released into the circulation when sympathetic activity increases. In studies, reduced concentrations of met-enkephalin in the cerebrospinal measured in cluster headaches. This can be interpreted as an indication of reduced endogenous antinociceptive activity , which could lead to the development of the pain disorder.

When analyzing inflammatory neuropeptides in cluster headache patients, it was found that increased levels of calcitonin gene related peptide (CGRP) and vasoactive intestinal polypeptide (VIP) can be found in the jugular vein during an acute cluster attack Neuropeptide Y and substance P, on the other hand, do not show increased values ​​during the cluster attack. Interestingly, after treatment of the cluster attack - with both oxygen and sumatriptan sc - a significant reduction in elevated CGRP levels. The application of analgesics, on the other hand, does not induce any change in neuropeptide levels. These findings directly indicate that there is activation in the trigeminovascular system and successful treatment may involve direct blocking of this increased activity. It was also observed in individual cases that a patient in whom a pseudoaneurysm of the cavernous sinus had a pain phenomenology in the sense of a cluster headache This finding suggests the importance of the cavernous sinus in the pathophysiology of cluster headache.

Substance P

Substance P are of particular importance in connection with neurogenic inflammation , nociceptive transmission in neurons and in connection with vaso reactions Substance P leads to a direct increase in pain sensitivity and can itself induce pain . For this reason, increased substance P activity in the neurons of the trigeminal and facial nerves was suspected to be the cause of the pain during the cluster attack. The direct vasoactivity of this neuropeptide was also interpreted as the cause of the autonomic disorders during the cluster attack. However, no altered concentration of substance P cerebrospinal fluid during the active cluster period compared to the remission period and compared to healthy subjects. In contrast, there was reduced substance P activity in the plasma . This could increased metabolism during a cluster attack. For the neuropeptide calcitonine gene related peptide (CGRP), however, increased concentrations were found in saliva during a cluster period. The same applies to the vasoactive intestinal polypeptide (VIP).

The inhibitory transmitter somatostatin is able to block the release of substance P. Somatostatin can be found in the area of ​​the sympathetic ganglia. By blocking the release of substance P, somatostatin could be a possible candidate for the acute treatment of cluster headaches. However, there is very pronounced tachyphylaxis, which is why the use of somatostatin for the treatment of cluster headaches little sense . However, it has actually been shown that plasma somatostatin concentrations reduced during the cluster period and somatostatin injection reverse acute cluster headache .