Changes in functional coupling of oscillatory brain activity in primary dysmenorrhea revealed by magnetoencephalography

• Main Authors: Pin-Shiuan Lee, 李品萱
• Other Authors: Li-Fen Chen
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/6t3g58

博士 === 國立陽明大學 === 生物醫學資訊研究所 === 106 === Primary dysmenorrhea (PDM), menstrual pain without identifiable pelvic pathology, is a common problem in women of reproductive age. The prevalence of PDM is high and about 20-90 % females have suffered from the cyclic menstrual pain, even with 15 % of whom have asked for leave from school or work. However, its underlying neural mechanism is still unclear. This dissertation focuses on changes in functional coupling of brain oscillations in primary dysmenorrhea and the effect of brain-derived neurotrophic factor (BDNF) Val66Met polymorphism on menstrual pain using resting-state magnetoencephalography. Theta oscillation is well documented for its association with neural processes of memory. Increase of theta activity is commonly observed in patients with chronic neurogenic pain. Hence the goal of the first study was to investigate encoding sensory and emotional information of long-term menstrual pain revealed by source imaging of theta activity in PDM. Forty-six young women with PDM and 46 age-matched control subjects underwent resting-state magnetoencephalography study during menstrual and periovulatory phases. Our results revealed increased theta activity in brain regions of pain processing in women with PDM, including the right parahippocampal gyrus, right posterior insula, and left anterior/middle cingulate gyrus during the menstrual phase and the left anterior insula and the left middle/inferior temporal gyrus during the periovulatory phase. The correlations between theta activity and the psychological measures pertaining to pain experience (depression, state anxiety, and pain rating index) implicate spontaneous theta activity may reflect emotional and sensory processing of pain and encode long-term pain experience in young women with PDM. Besides investigating theta rhythm, the interaction of slow and fast oscillations have been proposed to relate to different brain functions. Cross-frequency phase-phase coupling (n:m CFC) at the given cortical region has been proposed as a potential mechanism to regulate neural communication between different temporal scales. The aim of the second study was to propose a quantitative assessment of n:m CFC values in resting-state brain signals using MEG data. Simulation experiments with different frequency pairs were conducted to evaluate the effect of SNR on phase estimation and CFC calculation. The values of n:m CFC between each pair of frequencies (delta, theta, alpha, beta, and gamma) were estimated from eye-open resting-state MEG data. The results of simulation data showed increased accuracy of phase and n:m CFC estimation for balanced power of two targeted frequencies. The results of real MEG data exhibited that CFC maps of low-alpha activity presented more broadly distributed coupling with high frequencies (including beta and gamma) in the frontal regions related to ventral attention and frontoparietal control networks, whereas high-alpha activity coupling with the gamma band in the posterior part of frontoparietal control network. Such quantitative assessments of n:m CFC features provide as an indicator of network coordination and functional integration in spontaneous oscillations asssociated with PDM. Therefore, the third study aimed at examining association of n:m CFC functionally with long-term cyclic pain experience in PDM. The resting-state MEG recordings were obtained in fifty PDM subjects and fifty healthy subjects during menstruation and peri-ovulatory phases. The n:m CFC among five frequencies (theta, alpha, beta, gamma) was used as an indicator of intra-regional functional coupling between oscillations. The PDM group exhibited hypo-synchronization between low-alpha and low-beta band for right caudal middle frontal and inferior parietal areas, and between low-alpha and gamma band for right postcentral areas during painful state (menstruation phase); whereas they exhibited hypo-synchronization between multiple frequency pairs at the right rostral middle frontal and precentral areas during pain-free state (peri-ovulatory phase). Anxiety scale was positively correlated with low-beta/gamma phase-phase coupling in the left lateral occipital regions during the menstruation. The results in the third study revealed intrinsic dysfunctional changes of regional functional coupling in the right central and frontal areas in PDM. The n:m CFC might act functionally different depending on pain-related processing of sensory, cognition, and emotion. Recently there has been a growing interest in BDNF that may be implicated in the etiology of stress-related phenotypes. BDNF plays a key role in the formation of neural proliferation, growth, and survival, yet little is known about whether the BDNF Val66Met polymorphism contributes susceptibility to PDM and is associated with dysfunctional coupling of menstrual pain revealed by n:m CFC features. In the final study, we hypothesized that the BDNF Val66Met variant would be associated with changes in CFC during pain processing. Our results showed that there were significant interactions between group (PDM and CON) and gene (Val/Val, Val/Met, and Met/Met) via CFC analysis. Decreased low-alpha/gamma coupling in PDM with Met/Met was observed in left inferior frontal, central and superior parietal areas. The most increased n:m CFC was showed in CON with Met/Met than all the other gene groups within CON. Moreover, anxiety scale was negatively linked to low-alpha/gamma phase-phase coupling of CON with Met/Met in the left superior parietal region during the menstruation. Compared to CON with Met/Met, our results implicated that PDM with Met/Met revealed dysfunctions in pain-related adaptation and alleviation of anxiety. In summary, our results demonstrate that women with PDM exhibit dysfunction of resting-state network by theta activity and CFC, which was associated with encoding of pain experience. The neural processes involved in long-term pain experience were mediated by emotion, perception, and cognition in PDM during MENS and POV phases. These findings provide neuroimage evidences of neuroplasticity after long-term menstrual pain, which would help better understanding of underlying neural mechanisms in PDM.