RESEARCH PAPER Altered hypothalamic functional connectivity in cluster headache: a longitudinal resting-state functional MRI study Fu-Chi Yang, 1,2 Kun-Hsien Chou, 3,4 Jong-Ling Fuh, 5,6 Pei-Lin Lee, 7 Jiing-Feng Lirng, 8,9 Yung-Yang Lin, 1,4,5,6 Ching-Po Lin, 1,3,4,7 Shuu-Jiun Wang 1,4,5,6 Additional material is published online only. To view please visit the journal online (http://dx.doi.org/10.1136/ jnnp-2014-308122). For numbered afliations see end of article. Correspondence to Dr Shuu-Jiun Wang, Department of Neurology, Taipei Veterans General Hospital, No. 201, Section 2, Shipai Road, Beitou 112, Taipei, Taiwan; sjwang@vghtpe.gov.tw; Ching-Po Lin, Institute of Neuroscience, National Yang-Ming University, Taiwan No. 155, Section. 2, Li-Nong St. Beitou 112, Taipei, Taiwan; cplin@ym.edu.tw Received 16 March 2014 Revised 1 June 2014 Accepted 7 June 2014 To cite: Yang F-C, Chou K- H, Fuh J-L, et al. J Neurol Neurosurg Psychiatry Published Online First: [ please include Day Month Year] doi:10.1136/jnnp- 2014-308122 ABSTRACT Background Neuroimaging studies implicate hypothalamic dysfunction in the pathogenesis of cluster headache (CH). Disruptions in non-traditional pain processing areas, including the cerebellum and visual cortex, have also been reported in CH. It is unknown whether the hypothalamus interacts signicantly with these areas, and whether any such interactions vary between the in-boutand out-of-boutperiods in CH. This study aimed to investigate the resting-state functional connectivity (FC) of the hypothalamus of patients with CH. Methods Using 3-T functional MRI, we conducted a seed-based resting-state intrinsic FC analysis of the hypothalamus in 18 episodic CH patients during in-bout and out-of-bout periods, and in 19 healthy controls. Correlations between hypothalamic FC and clinical variables were also assessed. Results Compared to controls, CH patients showed hypothalamic FC changes with the medial frontal gyrus and occipital cuneus during in-bout and out-of-bout periods. Compared to out-of-bout scans, in-bout scans revealed decreased hypothalamic FC with the medial frontal gyrus, precuneus, and cerebellar areas (tonsil, declive and culmen). Additionally, the annual bout frequency correlated signicantly with the hypothalamic FC in the cerebellar culmen (r=-0.576, p=0.02) and cerebellar declive (r=-0.522, p=0.038). Conclusions Our ndings suggest that in CH, FC differences between the hypothalamus and its regional distribution extends beyond traditional pain processing areas, primarily to the cerebellar, frontal and occipital areas. These changes may be important and associated with CH pathophysiology. INTRODUCTION Cluster headaches (CH) are characterised by severe unilateral attacks with autonomic symptoms, such as lacrimation, rhinorrhea, sweating, ptosis and restlessness or agitation. 1 Most cases are episodic (8090%), manifest in bout periodswith a high daily attack frequency that lasts from 1 week to several months (the in-boutperiod), separated by month-long to year-long remissions (out-of-bout period). 1 The episodic nature of CH suggests a potential role of the hypothalamus. Functional imaging studies have revealed ipsilateral hypothal- amic activation during acute CH attacks. 24 An increased grey matter volume (GMV) in the ipsilat- eral inferior posterior hypothalamus has also been shown. 5 Moreover, deep brain stimulation in the hypothalamus can alleviate CH, further indicating hypothalamic involvement. 6 Central pain transmission is mediated by pain- processing areas, which traditionally include the ascending pain processing and descending pain modulation networks. 7 Functional imaging studies have revealed functional changes in the hypothal- amus and traditional pain-processing areas, includ- ing the anterior cingulate cortex (ACC), posterior thalamus, prefrontal cortex and insular cortex during acute CH attacks. 35 8 Sprenger et al 9 described increased metabolism in pain-modulation areas during in-bout periods (but between acute attacks) compared to out-of-bout periods. Moreover, we observed differences in GMV between in-bout and out-of-bout periods, primarily in the frontal pain-modulation areas. 10 These nd- ings suggest that CH may also result from a disor- dered frontal topdown pain-modulatory network, with signicant dynamic changes between the in-bout and out-of-bout periods. Resting-state functional MRI (RS-fMRI) allows functional connectivity (FC) at rest to be probed through correlated low-frequency spontaneous uc- tuations. 11 12 Low-frequency (<0.1 Hz) uctua- tions are thought to be functionally relevant indices of connectivity between brain regions, 12 and this intrinsic or resting-state activity accounts for the bulk (>95%) of brain metabolism. 13 Therefore, it is important to consider brain activity that occurs in the absence of external stimulation to elucidate how brain networks function and adapt in CH patients. Studies of FC can reveal functional disconnec- tions in neurological and psychiatric disorders. 14 15 Using RS-fMRI, Rocca and colleagues 16 found abnormal FC in the primary visual network of CH patients. Recently, Qiu and coworkers 17 reported an RS-fMRI study showing abnormal FC between the hypothalamus and occipital cortex. Cerebellar activation is also implicated in acute CH attacks. 18 This body of evidence suggests that CH patho- physiology may involve the hypothalamus, and also the frontal pain modulation network and brain areas outside the traditional pain-processing areas (eg, occipital cortex and cerebellum). In this study, we investigated whether interac- tions between the hypothalamus and brain areas implicated in CH pathophysiology are associated with the in-bout/out-of-bout period shifts. We Yang F-C, et al. J Neurol Neurosurg Psychiatry 2014;0:19. doi:10.1136/jnnp-2014-308122 1 General neurology JNNP Online First, published on June 30, 2014 as 10.1136/jnnp-2014-308122 Copyright Article author (or their employer) 2014. Produced by BMJ Publishing Group Ltd under licence. group.bmj.com on June 30, 2014 - Published by jnnp.bmj.com Downloaded from