Impact of sleep restriction on the structural brain network Min-Hee Lee a, *, Youngjin Lee c, *, Yoon Ho Hwang a , Areum Min a , Bong Soo Han b and Dong Youn Kim a Sleep restriction (SR) is defined as the condition of not having enough sleep, and it can cause brain injury. In this study, we examined the impact of SR on the structural brain network. We obtained diffusion MRI (dMRI) data for the SR group of fourteen participants who got less than or equal to 5.5 h of sleep for the last 1 month and normal group of the same number of participants who got 7 h of sleep. We constructed the structural brain networks from the dMRI data and analyzed them using graph theoretical approaches. In comparison with the normal group, the SR group showed higher vulnerability to the targeted node attack and alterations of regional efficiency in the brain regions such as the bilateral orbital part of the frontal gyri, superior occipital gyri, left insula, fusiform, right supplementary motor area, and cingulate gyrus. These findings indicate that SR may cause the reduction of the potential alternative neuronal pathways in the brain and rewiring of neuronal fibers in the structural brain networks, which may result in potential functional impairments, as well as alterations of the structural brain connectivity. Therefore, investigating the structural brain network offers new insight into how SR influences the human brain. NeuroReport 27:1299–1304 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. NeuroReport 2016, 27:1299–1304 Keywords: diffusion MRI, graph theory, sleep restriction, structural brain network Departments of a Biomedical Engineering, b Radiological Science, Yonsei University, Wonju and c Department of Radiological Science, Eulji University, Seongnam, Republic of Korea Correspondence to Dong Youn Kim, PhD, Department of Biomedical Engineering, Yonsei University, 1, Yeonsedae-gil, Heungeop-myeon,Wonju 26493, Republic of Korea Tel: + 82 33 760 2434; fax: + 82 33 760 2579; e-mail: dongkim77@yonsei.ac.kr *Min-Hee Lee and Youngjin Lee contributed equally to the writing of this article. Received 24 June 2016 accepted 2 September 2016 Introduction Sleep is considered to play a crucial role in body restitution, thermoregulation, tissue recovery, and integration of memory for cognitive performance [1]. Sleep restriction (SR), defined as the condition of not having enough sleep, has bad influences on human health, such as decreased concentration, reduced immune functioning, raised blood pressure, increased body weight, and hormonal disturbance [1]. In particular, SR can cause an allostatic overload, sup- press brain cell proliferation, and be a stressor on the brain [2]. These conditions lead to increased proinflammatory cytokines, which promote systemic inflammation and may cause damage to the brain structure [2]. It may have harmful effects on mood states such as depression, anger, tension, and cognitive functions including reduced vigi- lance and working memory [3,4]. There have been many studies conducted on the importance of sleep and the impact of the amount of sleep on the brain [5–8]. Yoo et al. [8] reported that the SR group showed decreased functional connectivity between the amygdala and medial–prefrontal cortex, and increased functional connectivity between the amygdala and brainstem. Consequently, they concluded that the SR group did not appropriately modulate the emotional brain response. The previous studies for SR focused on the changes in functional brain connectivities and regio- nal brain abnormalities associated with SR. As far as we know, there have been no studies for the effects of SR on the alterations in the structural brain connectivity. The human brain is able to economically integrate and segregate information from multiple brain regions [9]. Diffusion MRI (dMRI) is a noninvasive technique sen- sitive to white-matter microstructure in the brain and provides information of anatomical connection patterns [9]. By analyzing the dMRI data by graph theoretical approaches, we can measure the architectural features of complex networks of the brain [9,10]. Analysis of the brain network has advantages in quantifying brain net- work topology and interpreting the neurobiological meaning. Therefore, this analysis can be used to find the abnormalities of brain connections in neurological and psychiatric disorders [9–11]. In this study, we hypothesized that structural brain net- works are affected by SR. To verify the hypothesis, we constructed structural brain networks from dMRI data and compared and analyzed them with graph theoretical approaches including path length, clustering coefficient, small-world organization, node betweenness centrality (NBC), and regional efficiency (RE). Materials and methods Participants We used dMRI data of 28 participants selected among 526 healthy participants who participated in the Human Sleep 1299 0959-4965 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/WNR.0000000000000687 Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.