Abstract
Regarding Dr Shevel’s comments in response to our recent editorial (1) discussing new data on cortical spreading depression (CSD) (2). Dr Shevel raises some important issues.
We highlighted the demonstration by Lambert and colleagues that CSD can activate trigeminal second order neurons (2), and perhaps should have set out the historical context more clearly, as indeed to some extent the new data remain controversial (3). When the original observations related to the question of CSD and pain were published using the Fos protein method (4), the finding was hotly contested (5) and fiercely discussed (6). Interestingly, the Fos technique as a marker of neuronal activation has since been criticized by the original authors (7). Recent data claims to demonstrate categorically, using the more reliable electrophysiological techniques as used by Lambert and colleagues, that CSD does indeed cause significant neuronal activation at the deeper and superficial laminae of the trigeminal nucleus caudalis and cervical region (8), with some very particular anatomical specificity (9). Moving to higher species, such as the cat, it has been difficult to demonstrate CSD-induced trigeminovascular activity, even when localized to the same caudal region of the trigeminal nucleus and cervical extension, and in the superficial laminae (10), shown in the rodent. Moreover, the induction of CSD is very difficult in primates (11) and even harder for practical reasons to demonstrate directly in humans during aura. The only examples of CSD in humans in migraine come from blood flow imaging studies (12,13), and thus by inference. The best direct evidence in humans comes from the injured, compromised brain (14–16), which is hardly typical of migraine with aura. Clearly, the entire subject of CSD-induced trigeminovascular activation is beyond the text of an editorial. Hence, when considering the original work in the context of migraine, we believe it is necessary to address the wider context, as well as the findings of the paper. Undoubtedly, Lambert and colleagues, and indeed others, demonstrate trigeminovascular activation after CSD in rodents; however, there are nuances as regards applying these data to migraine. We remain surprised that some consider the issue of CSD causing pain during migraine, such a complex human condition, a settled issue on the basis of rodent data (3), with so little direct information in humans.
Regarding the comment ‘that this increased discharge does not involve the trigeminal nerve, but that there is instead some other pathway’, we completely agree with Dr Shevel. We struggle with the correspondent’s reasoning, however, that blocking the peripheral input means the ‘exact opposite’. We have, perhaps naively, considered that blocking the afferent pathway actually means blocking the afferent pathway. We are aware of technical criticisms (3) of Lambert and colleagues’ (2) work that to us seem, in the round, somewhat overemphasized, since one of us (PJG) has observed firsthand the technical dexterity and attention to detail of the investigators. Given that CSD is a central event, and that any other pathway is, by definition, in the central nervous system: the central hypothesis seems, at least, attractive: quod erat demonstrandum.