We can find a lot of signs and symptoms due to the disturbance of autonomic nervous system in patients with migraine. The purpose of our study is to clarify the autonomic function in migraine. The autonomic nervous function in patients with migraine was studied during headache-free intervals. The following observations were made: (1) a decrease in overshoot in Valsalva's maneuver; (2) orthostatic hypotension; (3) low levels of plasma norepinephrine in the steady state; (4) failure in elevation of the plasma norepinephrine level after head-up tilting; (5) dilatation of the pupils after instillation in the eye of 1.25% epinephrine. The above findings suggest that patients with migraine show sympathetic hypofunction with denervation hypersensitivity of the iris. When we evaluated the orthostatic hypotension during headache period in nine patients with migraine, we could not find any orthostatic hypotension. These phenomenon suggests that sympathetic hypofunction disappeared in patients with migraine during headache. As a next step, sweating function in patients with migraine were examined during headache-free intervals. Our results suggest that sweating function are impaired in patients with migraine. When we investigate the cerebral blood flow in patient with migraine during attack, low blood flow in hypothalamus was observed. When headache was disappeared, hypothalamus blood flow was recovered and increased more than other area. Cerebrospinal hypovolemia patients often showed orthostatic headache. Among cerebrospinal hypovolemia patients we can find many patients with postural tachycardia syndrome, who showed low production of cerebrospinal fluid probably due to sympathetic hyperactivity.

Principles of autonomic nervous system (ANS) physiology were systematized by Langley and Cannon, but I am aware that their descriptions contain serious doubts. I describe on alternative theories with those of Langley and Cannon. (1) ANS and emotion: Emotional alteration accompanies change in ANS activity; James (1884) and Lange (1885) argued that emotion was produced in the ANS. Oppositely, Cannon (1927) and Bard (1928) stated that the ANS was under the influence of emotion, which occurred within the diencephalon. The Cannon-Bard theory had long been accepted, because Langley (1898) incorrectly defined ANS as peripheral nerves only. It is, however, proved that the diencephalon was just an upper part of the central autonomic network, and recent psychological experiments supports the James-Lange theory rather than the Cannon-Bard theory. (2) Spinal parasympathetic nerves: During the times of Langley, it was debated whether parasympathetic vasodilator nerves were present. While Langley (1905) denied the presence of spinal parasympathetic nerves, Kuré succeeded to demonstrate them. Kuré et al. (1928) cut the dorsal roots of dogs, and investigated the central stump after keeping dogs alive long. Large myelinated fibers were completely lost, but small myelinated fibers were well-preserved, indicating that the latter fibers were centrifugal (parasympathetic). (3) Denervation supersensitivity: Eppinger and Hess (1909) claimed that patients with sympathetic hyperactivity showed supersensitivity to administered adrenaline. Oppositely, Cannon (1959) proposed the term of denervation supersensitivity. Since Eppinger and Hess’s cases were not autonomic failure, it is incorrect to criticize Eppinger and Hess from the view-point of denervation supersensitivity theory.

Recent advances in cognitive neuroscientific studies of emotion allow us to understand more closely about the underlying mechanisms of human emotion. Current essential research questions for understanding emotion are as follows: 1) what are the functions of the brain areas or networks involved in emotional processing, 2) how the autonomic bodily responses affect emotion perception or production, and 3) what is the significance of clinical studies for the patients with emotional disorders and autonomic nervous disorders. In this article, I first reconsidered some concepts around emotion, reviewed recent development of the relevant studies mainly focusing on the interoception and the functions of insular cortex, and finally highlighted on the importance of understanding “mind-brain-body” interactions.

The sympathetic nervous system is critical to coping with environmental stress such as fear. Many fears are innate and species-specific, and information on fear stimulation is sent directly from the amygdala to the autonomic center of the hypothalamus and brainstem to elicit sympathetic reactions. On the other hand, fear is learnable, and by analyzing the response to such fears, the upper hierarchical structure of the sympathetic nervous system can be examined. Recent research with human functional neuroimaging has reported that brain activities of the anterior cingulate cortex and the anterior insular cortex are related to sympathetic responses, especially the former is related to the cognitive generation and control of emotion. It is becoming clear that the functional linkage between the amygdala and the anterior cingulate cortex plays an important role in combining the fear and the autonomic nervous system.

The arterial pressure reflexly increases in response to noxious mechanical stimulation (pinching) of the unilateral hindpaw in anesthetized rats. The pressor reflex is mediated via the supraspinal structure; however, the precise brain mechanisms have not been elucidated. Since approximately 90% of the projection neurons from spinallamina I terminate in the contralateral side of the lateral parabrachial nucleus (LPBN), the present study aimed to clarify the involvement of the LPBN in the pressor reflex responses to pinching of the hindpaw in anesthetized rats. Arterial pressure was recorded via a catheter inserted into the carotid artery. Muscimol, a widely used neuronal inhibitor, was nanoinjected into the unilateral LPBN. Pinching was applied with a surgical clamp at a force of 3-5 kg to the unilateral hindpaw for 20 s. Administration of muscimol into the LPBN had no influence on the tonic arterial pressure. On the other hand, the pressor responses to pinching of the hindpaw contralateral to the site of the muscimol injection were significantly attenuated after the administration of muscimol. The pressor responses to pinching of the hindpaw ipsilateral to the site of the muscimol injection were slightly attenuated, but the attenuation was not statistically significant. The present results demonstrate that the LPBN is involved in the pressor reflex responses elicited by pinching of the contralateral hindpaw.