The effect of anxiety on brain activation patterns in response to inspiratory occlusions: an fMRI study

Abstract

Respiratory sensations such as breathlessness are prevalent in many diseases and are amplified by increased levels of anxiety. Cortical activation in response to inspiratory occlusions in high- and low-anxious individuals was found different in previous studies using the respiratory-related evoked potential method. However, specific brain areas showed different activation patterns remained unknown in these studies. Therefore, the purpose of this study was to compare cortical and subcortical neural substrates of respiratory sensation in response to inspiratory mechanical occlusion stimuli between high- and low-anxious individuals using functional magnetic resonance imaging (fMRI). In addition, associations between brain activation patterns and levels of anxiety, and breathlessness were examined. Thirty-four (17 high- and 17 low-anxious) healthy non-smoking adults with normal lung function completed questionnaires on anxiety (State Trait Anxiety Inventory - State), and participated in a transient inspiratory occlusion fMRI experiment. The participants breathed with a customized face-mask while respiration was repeatedly interrupted by a transient inspiratory occlusion of 150-msec, delivered every 2 to 4 breaths. Breathlessness was assessed by self-report. At least 32 occluded breaths were collected for data analysis. The results showed that compared to the low-anxious group, the high-anxious individuals demonstrated significantly greater neural activations in the hippocampus, insula, and middle cingulate gyrus in response to inspiratory occlusions. Moreover, a significant relationship was found between anxiety levels and activations of the right inferior parietal gyrus, and the right precuneus. Additionally, breathlessness levels were significantly associated with activations of the bilateral thalamus, bilateral insula and bilateral cingulate gyrus. The above evidences support stronger recruitment of emotion-related cortical and subcortical brain areas in higher anxious individuals, and thus these areas play an important role in respiratory mechanosensation mediated by anxiety.

Cortical Sources of Respiratory Mechanosensation, Laterality, and Emotion: An MEG Study

“…The lateral frontal cortex in the middle frontal gyrus (MFG) was also found to be activated bilaterally. The present results are consistent with some past RREP and fMRI studies [ 4 , 20 , 21 , 42 ]. For example, von Leupoldt et al (2010) suggested the right lateral frontal cortex as an area of origin of cortical sources for RREP [ 42 ].…”

“…Using a high-density 128-electrode EEG recording system, von Leupoldt et al (2010) reported a negative band in the frontal area for the Nf peak and a positive band in the centroparietal area for the P1 peak [ 42 ]. Recent fMRI reports also suggested that bilateral sensorimotor cortices were consistently activated by transient 150 ms inspiratory occlusions [ 20 , 21 ].…”

“…In contrast, Davenport and Vovk (2010) reported that load elicited cortical activations primarily in the inferior frontal gyrus and opercula triangularis, and they identified MFG as a major area associated with chemoreception [ 4 ]. However, later fMRI studies consistently reported MFG as one major area of cortical neural activation by single and paired inspiratory occlusions [ 20 , 21 ]. Our experimental setup of transient inspiratory occlusion stimulus involved changing blood gas, which activated chemoreceptors for homeostasis.…”

“…Therefore, functional magnetic resonance imaging (fMRI) is often used to compensate for the shortcomings of EEG measurements. Past fMRI studies have revealed that short inspiratory occlusion elicits brain neural activation in brain substrates, including cortical (sensorimotor cortices, frontal lobes, supramarginal gyrus, cingulate cortex) and subcortical (thalamus, hippocampus, and caudate) areas [ 20 , 21 , 22 ]. Other neuroimaging studies have also shown that dyspnea induced by resistive load is associated with neural activation in the amygdala, cingulate cortex, and insular cortex [ 7 , 9 , 23 , 24 ].…”

“…Some researchers have discovered that RREP peak amplitudes tend to be greater in the right cortex compared to the left [ 19 ], while others have not found much difference between the two sides [ 10 , 25 ]. Some past attempts examining the effect of negative emotion (e.g., anxiety) on RREP have not found significant differences in terms of peak amplitudes between the two hemispheric recordings [ 26 , 27 ]; however, with the fMRI technique, a recent study did notice an association between anxiety level and activation of the precuneus and inferior parietal gyrus in the right hemisphere in response to short inspiratory occlusion [ 21 ]. Additionally, results in the literature regarding emotional influences on hemispheric lateralization are mixed [ 28 , 29 , 30 ].…”