Windischberger Christian, Langenberger Herbert, Sycha Thomas, Tschernko Edda M, Fuchsjäger-Mayerl Gabriele, Schmetterer Leopold, Moser Ewald
Institute for Medical Physics, University of Vienna, Austria.
Magn Reson Imaging. 2002 Oct;20(8):575-82. doi: 10.1016/s0730-725x(02)00563-5.
BOLD-based functional MRI (fMRI) can be used to explicitly measure hemodynamic aspects and functions of human neuro-physiology. As fMRI measures changes in regional cerebral blood flow and volume as well as blood oxygenation, rather than neuronal brain activity directly, other processes that may change the above parameters have to be examined closely to assess sensitivity and specificity of fMRI results. Physiological processes that can cause artifacts include cardiac action, breathing and vasomotion. Although there has been substantial research on physiological artifacts and appropriate compensation methods, controversy still remains on the mechanisms that cause the fMRI signal fluctuations. Respiratory-correlated fluctuations may either be induced by changes of the magnetic field homogeneity due to moving organs, intra-thoracic pressure differences, respiration-dependent vasodilation or oxygenation differences. The aim of this study was to characterize the impact of different breathing patterns by varying respiration frequency and/or tidal volume on EPI time courses of the resting human brain. The amount of respiration-related oscillations during three respiration patterns was quantified, and statistically significant differences were obtained in white matter only: p < 0.03 between 6 vs. 12 ml/kg body weight end tidal volume at a respiration frequency of 15/min, p < 0.03 between 12 vs. 6 ml/kg body weight and 15 vs. 10 respiration cycles/min. There was no significant difference between 15 vs. 10 respiration cycles/min at an end tidal volume of 6 ml/kg body weight (p = 0.917). In addition, the respiration-affected brain regions were very similar with EPI readout in the a-p and l-r direction. Based on our results and published literature we hypothesize that venous oxygenation oscillations due to changing intra-thoracic pressure represent a major factor for respiration-related signal fluctuations and increase significantly with increasing end tidal volume in white matter only.
基于血氧水平依赖性功能磁共振成像（fMRI）可用于明确测量人类神经生理学的血流动力学方面及功能。由于fMRI测量的是局部脑血流量、血容量以及血液氧合的变化，而非直接测量神经元的脑活动，因此必须仔细研究其他可能改变上述参数的过程，以评估fMRI结果的敏感性和特异性。可能导致伪影的生理过程包括心脏活动、呼吸和血管运动。尽管已经对生理伪影及适当的补偿方法进行了大量研究，但对于导致fMRI信号波动的机制仍存在争议。与呼吸相关的波动可能是由移动器官引起的磁场均匀性变化、胸腔内压力差、呼吸依赖性血管舒张或氧合差异所诱发。本研究的目的是通过改变呼吸频率和/或潮气量来表征不同呼吸模式对静息人脑回波平面成像（EPI）时间进程的影响。对三种呼吸模式下与呼吸相关的振荡量进行了量化，仅在白质中获得了具有统计学意义的差异：在呼吸频率为15次/分钟时，终末潮气量为6与12 ml/kg体重之间，p < 0.03；在12与6 ml/kg体重以及15与10次呼吸周期/分钟之间，p < 0.03。在终末潮气量为6 ml/kg体重时，15与10次呼吸周期/分钟之间无显著差异（p = 0.917）。此外，在EPI读出的前后和左右方向上，受呼吸影响的脑区非常相似。基于我们的结果和已发表的文献，我们推测由于胸腔内压力变化导致的静脉氧合振荡是与呼吸相关信号波动的主要因素，并且仅在白质中随着终末潮气量的增加而显著增加。
