Although many advances have been made concerning the pathophysiology of cluster headache, the process is still only partially understood 1. Hypothalamic dysfunction and subsequent activation of the trigeminovascular system and the cranial parasympathetic system seem to play a significant role 2. However, other factors may also play a relevant role in the generation and/or perpetuation of cluster pain.

Referred pain arising from myofascial structures has been shown to be involved in many different disorders, mostly of musculoskeletal origin, as it is the case of myofascial pain syndrome 3 but also of visceral origin such as renal calculi, biliar calculi, dysmenorrhea, or myocardial infarction among others 456. Active myofascial trigger points (MTrPs) whose pressure provoked referred pain reproducing the typical patient's headache, have been described both in chronic tension type headache 7 and in migraine 8, indicating that peripheral sensitization may play a role in the pathophysiology of headache. Also Ashkenazi and Young reported the occurrence of cutaneous allodinia in cluster headache, especially prevalent in chronic patients with a longer duration of the disease, a finding which suggests that central sensitization may also be present in this condition 9. As cluster headache and migraine share some common features concerning trigeminovascular system involvement and response to some abortive and prophylactic drugs, we hypothesized that first neuron hypersensitivity could be also present in patients with cluster headache. Also, just at a clinical level, we noticed that some patients experiencing cluster headache reported that sometimes their attacks were triggered by the wind or when combing their hair, data suggesting the existence of peripheral sensitization.

Based on all the above mentioned data, we decided to examine patients with cluster headache in order to ascertain if referred pain of cluster characteristics suggestive of peripheral sensitization could be observed and, if present, to explore the potential usefulness of anaesthetic injections on pain management. We present herein the results of a case series of patients.

The cases we have presented were not selected from a broader sample, but were the first consecutive cluster patients who we had the opportunity to observe. Unexpectedly, active trigger points were found in every subject examined. It seems, therefore, that peripheral sensitization can be a non infrequent associated feature in cluster headache, at least in those patients whose attacks are not easily controlled.

Myofascial pain is believed to play a relevant role in the pathogenesis of tension-type headache and the presence of active MTrPs related with headache severity and duration has been described in chronic tension-type headache patients 7. In migraine patients, where pain is essentially related with the activation of the trigeminovascular system, active MTrPs have also been found 812 and their inactivation by means of repeated anaesthetic injections has been shown to decrease the frequency and severity of attacks in patients experiencing severe and treatment-refractory migraine 13. To our knowledge, however, the possibility that active MTrPs could also been found in cluster headache has not been previously investigated. Nevertheless, we believe that the mechanism underlying the presence of active TrPs in each of these primary headaches is similar: chronic pain or repeated acute pain sensitize muscular nociceptors creating active TrPs which, in turn, contribute to potentiate headache pain. This kind of vicious cycle explains why the number of active MTrPs has been found to be higher in patients with chronic primary headaches than in healthy subjects or in patients experiencing less frequent headache attacks 78.

Needling therapies are a widely accepted method in myofascial trigger points management, although the basis of its potential effectiveness has not been fully clarified 1415. In the treatment of migraine, injection of tender areas during the attack with either lidocaine or saline was shown to subside the pain 16, and the prophylactic use of anaesthetic injections, as mentioned above, was shown to contribute to patients' improvement 13. In relation to cluster headache, several uncommonly used therapeutic alternatives may, at least in part, act by provoking a first neuron afferent blockade; this would be the case of great occipital nerve blockade, great occipital nerve stimulation, sympathetic C7 anaesthetic blockade, and intranasal lidocaine or cocaine administration 1718192021. Our data, although uncontrolled, suggest that active TrPs injections can be also highly effective in cluster attacks management, either as abortive or as preemptive therapy, and may be of help in chronic cluster management. As it is estimated that between 10–20% of cluster patients are refractory to treatment or develop resistance to it 1, the potential role of active TrP blockade should not be underestimated. Trigger points injection is an easy to perform and well tolerated therapy, its main drawback being that it is rarely wholly effective as monotherapy 22. The combination of active TrPs inactivation with prophylactic drug treatment in the treatment of refractory cluster headache remains a therapeutic option worthy to be investigated.

This was just a preliminary and uncontrolled research and, as such, our results should taken with caution. Also, it must be stated that all of the patients attending our pain clinic have been previously diagnosed and treated by other physicians, so that the sample may be not representative of the average cluster headache patient. With these limitations in mind, we think that our data indicate that first neuron hypersensitivity can be present in cluster headache and that active trigger points inactivation can be a useful complementary therapy.

The authors declare that they have no competing interests.

JH, JMGL and ADR diagnosed and treated patients. EPC and FR participated in the design of the study and drafted the manuscript. All authors read and approved the manuscript's final version.