Should cluster headache be associated with cutaneous allodynia?

In an elegantly designed study, using meticulous quantitative sensory testing (QST), Ladda et al. (1) measured skin sensitivity during ongoing cluster headache (CH) and compared it with skin sensitivity during the in-bout and out-of-bout intervals when the patients were pain free. Contrary to their expectation, the authors found no evidence for cutaneous allodynia (skin hypersensitivity) during CH compared with in-bout and out-of-bout pain-free intervals. In fact, skin sensitivity during CH proved normal compared with healthy subjects. This is the first study that can be directly compared with previous studies that looked at skin sensitivity during migraine using a similar QST methodology. It has now become apparent that the phenomenon of migraine-associated cutaneous allodynia (2) does not develop during bouts of CH. If cutaneous allodynia during migraine is mediated by central sensitization of trigeminovascular neurons in lamina V of the dorsal horn (3), it is reasonable to conclude that central sensitization does not develop during CH. What is it, then, that differentiates between the pathophysiology of these types of headache? Several fundamental differences may exist between activation of visceral pain fibres that innervate intracranial structures such as the meninges, as in migraine, and activation of somatic pain fibres that innervate extracranial structures such the eye, muscles, skin, and gums, as in CH. First, it is worth noting that the activation of somatic nociceptors during CH stands in sharp contrast to the activation of visceral nociceptors during migraine as well as the activation of somatic nociceptors associated with nerve injury. Migraine headache is believed to involve local release of inflammatory molecules capable of producing peripheral sensitization in meningeal nociceptors (4), which is manifested as throbbing. Similarly, neuropathic pain resulting from nerve injury is often associated with peripheral sensitization, which is manifested as allodynia and hyperalgesia that is restricted to the innervation territory of the affected nerve (5). In contrast, the lack of allodynia in the Straube et al. study, even within the territory of the maxillary nerve, suggests that peripheral sensitization does not develop in the CH. If CH involves abnormal parasympathetic activity originating in certain brain areas such as the hypothalamus (6), it is tempting to speculate that a local release of noninflammatory molecules can only activate, but not sensitize, the trigeminal nociceptors that mediate CH. Second, somatic nociceptors terminate heavily in laminae I, II and V of the dorsal horn, whereas visceral nociceptors, although terminating heavily in laminae I and V, have few or no terminals in lamina II (7–9). Therefore, activation of lamina II, which contains inhibitory interneurons, is more likely to occur in CH than in migraine. Accordingly, excitation of lamina V neurons by incoming pain signals would be contravened during CH, but not during migraine, through inhibition exerted by the activation of inhibitory lamina II interneurons. The activation of this inhibitory circuit may account for the blockade of central sensitization and cutaneous allodynia in CH. Third, the development of central sensitization during different pain conditions (e.g. inflammatory vs. neuropathic, visceral vs. somatic, acute vs. chronic) may also vary with selective activation of descending modulatory pathways that involve brain areas such as the hypothalamus, amygdala, nucleus accumbens, periaqueductal grey and rostral ventromedial medulla. The presence of allodynia during migraine may be due to decreased activity of inhibitory descending modulatory pathways and/or increased activity of facilitatory descending modulatory pathways. In contrast, the absence of allodynia in CH may be due to increased activity of inhibitory descending modulatory pathways that regulate the output of nociceptive dorsal horn neurons. The absence of cutaneous allodynia in CH and its presence in migraine suggest that central sensitization is not a feature common to all trigeminal pains or even all trigeminovascular headaches. While blocking central sensitization may be a desirable target for treating migraine headache, it may be unsuitable for other types of headache.