In recent decades, important strides have been made in understanding the pathophysiology of migraine. However, this enterprise still remains a work in progress. During much of the twentieth century, migraine was thought to result primarily from vascular dysregulation. As part of this hypothesis, aura preceding headache was thought to result from hypoxemia related to transient vasoconstriction and migraine pain from rebound vasodilation, which caused primary nociceptive neurons within the walls of engorged intracranial vessels to undergo mechanical depolarization.
This vascular hypothesis agreed with the observed effects of vasodilating drugs, such as nitroglycerin, which caused headache, and vasoconstricting drugs, such as ergotamines, which resolved headache. However, results from functional neuroimaging studies conducted in the 1980s and 90s showed that decreased cortical blood flow occurring during aura was insufficient to result in ischemia (Figure 1) and that headache onset preceded vasodilation.[2-3]
Figure 1. Functional magnetic resonance imaging showing hemodynamic images during left visual field disturbance show decreased flow and volume in the right occipital lobe, with increased mean transit time. Perfiusion-weighted images obtained interictally and at four time points (20, 35, 59, and I80 minutes) after onset of visual aura affecting the left visual field. Reproduced from Cutrer FM, et al. Ann Neurol. 1998;43:25-31.
Current thinking has moved away from vascular dysregulation as a primary cause of migraine. It is now believed that vasodilation and vasoconstriction are probably epiphenomena and that neuronal dysfunction is the possible primary driver in the pathophysiology of the disorder.[1, 4, 5] Specifically, activation of the trigeminovascular system, cortical spreading depression , and neuronal sensitization are seen as playing important roles in migraine pathophysiology.[1,5]
Activation of trigeminovascular system
The major structures involved in activation of the trigeminovascular system are shown in Figure 2. Sensory neurons from the trigeminal ganglion and upper cervical dorsal roots innervate dural-vascular structures (eg, pial vessels, dura mater, large cerebral vessels). Input from dural-vascular structures and from cervical structures through the upper cervical dorsal root ganglia project to second order neurons in the trigeminocervical complex (TCC).
Nerve fibers involved in the localization of pain ascend from the trigeminal nucleus caudalis to the thalamus and then to the sensory cortex. Distribution of headache pain to regions of the upper neck and head can be attributed to the convergence of projections from the trigeminal nerve at the trigeminal nucleus caudalis and upper cervical nerve roots.
Sensory modulation can occur, via both direct and indirect projection, by descending influences such as those from the hypothalamus, midbrain periaqueductal gray (PAG), pontine locus coeruleus (LC), and nucleus raphe magnus onto the TCC. Sensory modulation also occurs from the hypothalamus, LC, and PAG by ascending influences.
Figure 2. Pathophysiology of migraine. Trigeminocervical complex=TCC; periaqueductal gray=PAG; pontine locus coeruleus=LC; nucleus raphe magnus=NRM. Reproduced from Goadsby PJ. Neurol Clin. 2009;27:335-60.
Cortical spreadin depression
Cortical spreading depression, a self-propagating wave of cellular depolarization that slowly spreads across the cerebral cortex and is associated with depressed neuronal bioelectrical activity and altered brain function, has been linked to migraine aura and headache. Cortical spreading depression is thought to activate neurons in the trigeminal nucleus caudalis, leading to inflammatory changes in pain-sensitive meningeal vascular structures, which produces headache via central and peripheral reflex mechanisms. Cortical spreading depression is also thought to alter the permeability of the blood-brain barrier by activating and upregulating brain matrix metalloproteinase.
Neuronal sensitization, the process by which neurons become increasingly responsive to nociceptive and non-nociceptive stimulation, is thought to play a role in migraine attacks. Sensitization results in decreased response thresholds, increased response magnitude, expansion of receptive fields, and development of spontaneous neuronal activity. Peripheral sensitization in the primary afferent neuron and central sensitization of higher-order neurons of the spinal cord and brain have been shown to play an important role in somatic pain. It is likely that many of the symptoms of migraine, including throbbing headache pain, exacerbation of headache by physical activity, and allodynia are linked to sensitization.