Inflammation in Migraine…or Not…: A Critical Evaluation of the Evidence

Andrew Charles MD; Sinifunanya E. Nwaobi MD, PhD; Peter Goadsby MD, PhD


Headache. 2021;61(10):1575-1578. 

In This Article

What do Inflammatory "Mediators" Mean?

Multiple studies report inflammation in animal models of migraine. Cortical spreading depression in rodents is associated with a variety of indicators of inflammation in most[1–4] but not all[5] studies. Stimulation of peripheral sensory fibers can cause characteristic changes in neurogenic inflammation including increased vascular permeability and vasodilation.[6] Stimulation of peripheral trigeminal nerves can also cause changes in the trigeminal ganglion that have been characterized as "neuroinflammation."[7] Administration of nitroglycerin in rodents, another informative migraine model, can increase the expression of genes encoding inflammatory cytokines[8] and activate brainstem microglia.[9] These animal model studies clearly indicate the activation of inflammatory mechanisms, but their relevance to migraine remains uncertain. It is difficult to determine whether the inflammatory indicator is indeed a migraine mechanism or a nonspecific response to an invasive procedure or supraphysiological stimulation.

In humans, it is now abundantly clear that CGRP plays a primary role in migraine,[10] and to the extent that CGRP can be considered an "inflammatory" molecule, this could be considered evidence for inflammation in migraine. However, neurogenic inflammation in animal models is defined by increased vascular permeability that is mediated primarily by substance P, rather than CGRP. Human studies show that CGRP blood levels are elevated during migraine attacks, whereas substance P levels are not.[11] Selective release of CGRP but not substance P during migraine attacks argues against classical neurogenic inflammation as a trigger for CGRP release. Furthermore, multiple antagonists of substance P receptors have proven ineffective as migraine therapies.[12,13]

Prostaglandins have been proposed as chemical mediators of migraine. Elevated levels of prostaglandins and their metabolites have been reported in blood,[14] saliva,[15] and urine[16] during migraine attacks. Administration of some, but not all, prostaglandins to humans induces headache.[17] Interestingly, the proinflammatory but vasoconstricting prostanoid PGF2alpha does not provoke headache,[18] suggesting that inflammatory effects of prostanoids may not be responsible for their migraine-triggering effects.

Multiple studies have investigated cytokines and other inflammatory mediators in migraine, with variable results.[19–21] No single inflammatory molecule has been consistently elevated in the serum or cerebrospinal fluid (CSF) of individuals with migraine. Cowan et al.[22] examined blood and CSF in patients with migraine both during and between attacks versus controls. They found higher CSF albumin and fibrinogen in the migraine group but concluded that this is not a consequence of increased blood–CSF barrier permeability due to the absence of other indicators. Interestingly, they also found no difference between those with migraine versus controls in CSF or plasma levels of multiple inflammatory mediators, with the exception of lower IL-10 plasma levels in the migraine group. These results provide evidence against a role for inflammation in ictal or interictal migraine, including chronic migraine.