Abstract-Using PET with the opioidergic ligand [11 C]diprenorphine, the authors demonstrate decreased tracer binding in the pineal gland of cluster headache patients vs healthy volunteers. Opioid receptor availability in the hypothalamus and cingulate cortex depended on the duration of the headache disorder. Therefore, the pathophysiology of cluster headache may relate to opioidergic dysfunction in circuitries generating the biologic clock. NEUROLOGY 2006;66:1108-1110 T. Sprenger, MD; F. Willoch, PhD; M. Miederer, MD; F. Schindler, MD; M. Valet, MD; A. Berthele, MD;M.E. Spilker, PhD; S. Förderreuther, MD; A. Straube, MD; I. Stangier, MS; H.J. Wester, PhD; and T.R. Tölle, MD, PhDCluster headache (CH), a lateralized headache disorder comprising periorbital pain and ipsilateral autonomic symptoms, is characterized by circadian and seasonal periodicity. The hypothalamus has been implicated in the pathogenesis of CH because this region displayed increased brain activity during attacks 1 and structural changes have been evidenced in the same area by voxel-based morphometry.2 Neuroendocrinologic alterations point to participation of biologic clock generators in the pathophysiology. The secretion of melatonin, which is mainly produced by the pineal gland and which plays a major role in the synchronization of the sleep/wake cycle, has been shown to be altered in CH, 3 and oral administration has been suggested for the treatment of CH. Regarding the opioidergic system, the circadian rhythm of beta-endorphin serum levels is lacking in patients with CH, 4 and opioids can alter the secretion of melatonin by the mammalian pineal gland. 5 We investigated CH patients by opioidergic ligand PET.Methods. Seven men with left-sided CH according to International Headache Society guidelines (six episodic and one chronic CH; mean age 46.9 years) and eight healthy men (mean age 45.3 years; group difference in age not significant in two-tailed t test: p ϭ 0.85) were investigated with [ 11 C]diprenorphine (DPN) PET. [ 11 C]DPN is a competitive nonselective opioid antagonist. All patients were in bout (i.e., in an active cluster period) but out of an acute attack at the time of the scanning. Disease duration varied from 1 to 19 years (mean 7.3 years; median 7 years). Clinical MRI was unremarkable in all patients. A pain-free period of 6 hours was mandatory, and all patients were in bout for at least 2 weeks at the time of scanning. The headache frequency was between one and three attacks per day in all patients, and patients were allowed to continue their prophylactic medication. Five patients were taking verapamil (with one of them additionally taking methysergide), one was taking a combination of ergotamine and valproate, and one was not taking any drugs. All participants gave written consent, and the study was approved by the local ethics committee and the radiation authorities.PET scans were acquired with a CTI ECAT EXACT scanner (Knoxville, TN) with a total axial field of 16.2 cm. The tracer was injected as an IV bolus (mean dose 19.2...