Effect of experimental modulation of mood on exertional dyspnoea in chronic obstructive pulmonary disease

Dyspnoea is a debilitating symptom in individuals with chronic obstructive pulmonary disease (COPD) and a range of other chronic cardiopulmonary diseases and is often associated with anxiety and depression. The present study examined the effect of visually‐induced mood shifts on exertional dyspnoea in individuals with COPD.


INTRODUCTION
Dyspnoea is one of the most common and debilitating symptoms in people with chronic obstructive pulmonary disease (COPD) and other clinical conditions such as heart failure, cancer and endocrine disorders. 1,2For those with COPD, chronic dyspnoea is a fearful experience that is often associated with anxiety, depression, reduced exercise tolerance and poor quality of life. 3espite the significant global burden of this symptom, our current ability to clinically manage dyspnoea is limited, partly due to an incomplete understanding of the underlying mechanisms. 1 The preliminary finding from this study was presented at the 2019 Annual Congress of the Thoracic Society of Australia and New Zealand (TSANZ).
][6][7] These studies have used standard visual (International Affective Picture System (IAPS) 8 ) and acoustic (International Affective Digitized Sound System (IADS) 9 ) stimuli to modulate an individual's affective states, an approach that has been well-documented in pain research. 10][13] In individuals with COPD, von Leupoldt and coworkers used a bicycle ergometer to induce exertional dyspnoea and found that a negative affective state induced greater breathlessness for the same intensity of exercise compared to a positive affective state. 14Our team has previously used treadmill exercise to induce exertional dyspnoea in healthy populations and found that experimentally induced positive mood states, by both IAPS and IADS systems, are associated with a lower level of exertional dyspnoea compared to negative mood states. 5,6The present study further examined the effect of visually-induced mood shifts on exertional dyspnoea in individuals with COPD.The use of a treadmill, simulating walking up a modest gradient provides a stimulus more familiar to this population.Furthermore, we have taken the opportunity to explore the effect of mood on leg fatigue, another troublesome symptom in this population.We hypothesised that for an equivalent level of cardiopulmonary stress, an experimentally induced positive mood state would be associated with lower levels of dyspnoea and leg fatigue compared to an experimentally induced negative state.Replication of our earlier findings 6 in a clinical population could have a significant impact on dyspnoea management.In particular, optimizing mood in a pulmonary rehabilitation environment could improve health-related outcomes.

METHODS
We recruited 20 participants with COPD following approval by Griffith University Human Research Ethics Committee.Individuals, aged ≥18 years old were recruited if they were clinically diagnosed with COPD and had completed a pulmonary rehabilitation program.Individuals with muscular, cardiovascular or neurological problems that would limit exercise performance were excluded from the study.We also excluded individuals who were on medication for the treatment of anxiety or depression.
All participants were required to attend four laboratory sessions ($2 sessions a week) which included a familiarization and three experimental visits.The sample size for this study was based on similar studies by our group giving a power of 90% to detect a 1.0 difference in dyspnoea with p < 0.05 of a Type-1 error. 5,6

Dyspnoea and mood modulation
We induced exertional dyspnoea by using 5-min bouts of treadmill exercise tests at an individualized speed and grade that was established on the familiarization visit. 15Treadmill speed was set at $80% of the average 6-min walk distance speed and slope (5%-12%) was titrated to induce a dyspnoea intensity of 4-5 (0-10 numerical rating scale, 0 = no breathlessness, 10 = breathlessness that would force cessation of exercise).All exercise tests were kept below 85% of maximal age-predicted heart rate.
During each exercise test, participants viewed a series of standard mood-modulating positive, neutral or negative images that were sourced from the IAPS. 8Based on the published mood scores, we selected 35 images for each mood state and the sets were different for males and females to induce an equivalent level of mood shifts.The IAPS images were displayed on a screen mounted in front of the treadmill using custom-built software.Each image in a set was displayed for 6-s followed by a 2-s interval before the next image, a protocol that we have previously validated in healthy individuals. 6At 1-min intervals throughout the exercise, there was a 10-s interval in picture viewing during which a rating scale was presented on the screen to allow the self-reported measurement of either dyspnoea or leg fatigue.

Dyspnoea and leg fatigue measurement
Participants rated their dyspnoea intensity or leg fatigue at minute intervals during exercise using a 0-10 numerical rating scale. 5,6Participants used a clicker attached to the handrail of the treadmill to select a number between 0 and 10 that best reflected the intensity of dyspnoea or leg fatigue.
Participants were familiarized that a rating of '0' for dyspnoea would represent no shortness of breath (SOB) while a rating of '10' would represent extreme SOB, a level that would force cessation or easing of exercise.Similarly, a

SUMMARY AT A GLANCE
This study examined the impact of experimentallyinduced mood shifts on exertional dyspnoea in COPD and explored the cause-and-effect relationship between psychological status and dyspnoea.The findings suggest that positive mood shifts can alleviate dyspnoea and may be an effective strategy for reducing the morbidity associated with this symptom in COPD.rating of '0' for leg fatigue would represent no leg discomfort while a rating of '10' would represent extreme leg discomfort that would force cessation or easing of exercise.

Mood assessment
At the end of each exercise test, participants rated their overall mood during 5-min of IAPS exposure whilst exercising.Mood valence was rated on a 9-point pictorial Self-Assessment Manikin (SAM) scale. 8,9On this scale, a rating of '1' represented extremely unhappy and a rating of '9' represented an extremely happy mood, while a rating of '5' represented a neutral mood.

Familiarization session (visit 1)
During this visit, informed written consent, and measurements of height, weight, blood pressure and lung function were obtained.Participants completed a modified Medical Research Council Dyspnoea Scale (mMRC), 16 St.George Respiratory Questionnaire-COPD (SGRQ-C) 17 and COPD Assessment Test (CAT) 18 questionnaires.
Participants then performed a 6-minute walk test (6MWT) followed by training in attending to the IAPS images and rating mood, dyspnoea and leg fatigue.This was followed by the practice exercise tests while viewing dummy images.During these practice tests, we established an individualized treadmill speed and grade for each participant. 15perimental sessions (visits 2-4) At each experimental visit, participants performed two 5-min treadmill exercise tests separated by 30-min rest while viewing IAPS images at the individualized speed and slope established during the familiarization visit.On any single visit, participants viewed only a single set of either positive, neutral or negative images for both tests performed each day, and the ordering of sets was randomized and balanced between participants.During each exercise test, participants rated either dyspnoea or leg fatigue, the order of which was randomized.Measures of heart rate (HR) and peripheral oxygen saturation (SpO 2 ) were obtained at 1-min intervals.ECG was monitored throughout the exercise to detect any exercise-induced arrhythmias.Measurements of mood valence and multidimensional dyspnoea profile (MDP) 19 were made on the completion of each exercise test.For MDP, participants were instructed to reflect on the last 30-s of the exercise.Given that the IAPS images were clearly either pleasant or disturbing, it was not possible to blind participants to the intervention.To minimize any expectation bias that a particular image set would affect dyspnoea or leg fatigue in a predictable way, we falsely advised participants that the speed and/or slope of the treadmill could change from test to test and that they should focus on purely reporting their symptoms.
Before undertaking the exercise test, participants completed Hospital Anxiety and Depression Scale (HADS), 20 Profile Mood Scale (POMS) 21 and Dyspnoea-12 (D-12) 22 questionnaires to assess for any day-to-day variations in the baseline mood and dyspnoea between sessions.

Statistical analysis
Baseline POMS, HADS and D-12 together with mood valence and MDP were analysed using a one-way analysis of variance (ANOVA).HR, SpO 2 , dyspnoea intensity and leg fatigue were analysed using a two-way repeated measure analysis of variance (2-way ANOVA) (time and experimental condition as factors).Condition-related significant differences were examined using Fisher's LSD post-hoc pairwise comparisons.Statistical significance was accepted at p < 0.05.

Study participants
Participants with mild to severe COPD were recruited for this study (Table 1).All the participants successfully completed the 5 min of exercise at moderately inclined slope (5%-15%) safely although five of them showed evidence of arrhythmias during at least one of the exercise tests.Participant characteristics at baseline are listed in Table 1.

Baseline mood and dyspnoea
The POMS components remained similar between the three experimental conditions (Table 2).Additionally, there were no statistically significant differences in the anxiety and depression components of HADS and baseline dyspnoea-12 between conditions (Table 2).

Mood modulation
Analysis of variance revealed that attending to different sets of IAPS images during exercise had a statistically significant condition-related effect on mood valence (Figure 1; p < 0.001).Post-hoc analysis revealed that compared to neutral images, mood valence was significantly higher when viewing positive images (p < 0.01) and was significantly lower when viewing negative images ( p < 0.001).

Physiological measures
Over the 5-min exercise period, there was no statistically significant differences in HR and SpO 2 between the conditions (p > 0.05) (Table 3).The interaction statistics for HR and SpO 2 indicated that there was no interaction between conditions and time ( p > 0.05).

Exertional dyspnoea
Over the 5-min exercise period, there was a conditionrelated effect on dyspnoea intensity ( p < 0.01) (Figure 2A).Post-hoc analysis revealed that compared to the positive and neutral conditions, the negative condition was associated with greater dyspnoea intensity ( p < 0.05).However, we did not observe any statistically significant difference in dyspnoea intensity between positive and neutral conditions.Eighty-five percent of participants (n = 17) met or exceeded the minimal clinically important difference MCID of 1-point difference for dyspnoea intensity between positive and negative conditions.
Moreover, when viewing negative images, dyspnoea intensity increased at a faster rate over the 5-min of exercise compared to viewing positive images (time Â condition interaction, p < 0.01).

Multidimensional dyspnoea profile (MDP)
The end-exercise dyspnoea unpleasantness while viewing negative images was significantly higher when compared to viewing positive (p < 0.01) and neutral images (p < 0.05).However, there was no statistically significant difference in dyspnoea between positive and neutral conditions (Table 3).Ninety percent of participants (n = 18) met or exceeded the MCID of 1-point difference for dyspnoea unpleasantness between positive and negative conditions.
Out of other sensory and emotional components of MDP, only 'muscle work/effort' and 'air hunger' components increased statistically significantly during negative condition compared with positive (p < 0.05) (Table 3) and F I G U R E 1 Mood valence between conditions.Mean (SEM) levels of pre-and post-exercise mood valence (1 = very unpleasant, 9 = very pleasant and 5 = neutral) during different experimental conditions.Compared to neutral images, mood valence was significantly higher when viewing positive images ( p < 0.01) and was significantly lower when viewing negative images (p < 0.001).
only three participants experienced breathing-related emotional sensation of MDP.

Leg fatigue
Over the entire 5-min exercise period there was no condition-related effect on leg fatigue (p > 0.05) (Figure 2B).Examination of the interaction statistics indicated that there was no interaction between conditions.

DISCUSSION
The main finding of this study is that during treadmill exercise, negative mood modulation was associated with higher dyspnoea.More specifically, the current study showed that negative modulation of mood in individuals with COPD is associated with higher dyspnoea intensity and unpleasantness/bother compared to positive mood.In addition, negative modulation of mood was associated with higher levels of dyspnoea intensity and bother compared to the neutral condition.Our findings indicate that exertional dyspnoea in COPD can be influenced by different emotional states induced by visual stimuli.We further found that different mood states had no effect on leg fatigue in contrast to the significantly different dyspnoea ratings.This finding is consistent with the idea that these primary components of the sensory response to exercise (dyspnoea and leg fatigue) are processed differently by the brain, with dyspnoea perception being more influenced by affect than is the case for leg fatigue. 23,24This study confirms that experimental modulation of mood during exercise can be achieved in individuals with COPD through visual stimuli and that this influences dyspnoea perception between the positive and negative affective states to an equivalent degree as previously described by our group in healthy individuals. 6Additionally, the fact that there were no significant differences in baseline affective states and resting dyspnoea, suggests that our findings are not influenced by any day-to-day variations in baseline mood and dyspnoea, which are common in this population.F I G U R E 2 Dyspnoea intensity and leg fatigue between conditions.Mean (SEM) levels of dyspnoea intensity (A), and leg fatigue (B) during identical treadmill exercise bout performed on three separate days.
During each of the exercise bouts, participants viewed either positive (•), neutral () or negative (▼) International Affective Picture System images (IAPS).For dyspnoea, 0 = no shortness of breath, 10 = extreme shortness of breath, and for leg fatigue: 0 = no leg discomfort, 10 = extreme leg discomfort.A statistically significant difference between *positive and negative; †neutral and negative.
Our results are in general agreement with previous studies reporting that experimentally induced positive affective states are associated with lower levels of dyspnoea compared with experimentally induced negative mood states.In individuals with COPD, von Leupoldt et al. 14 using IAPS during a constant loaded bicycle ergometry, reported that positive mood was associated with reduced exertional dyspnoea intensity compared to the negative; however, this group did not find any differences in dyspnoea unpleasantness/ bother between positive and negative affective states. 14Despite the methodological differences, our current findings when taken together with the findings of von Leupoldt and coworkers, demonstrate a clear effect of experimentallyinduced changes in the mood on dyspnoea.Moreover, there are other few studies exploring the effect of visually induced mood modulation on dyspnoea induced by resistive-loaded breathing and slow-paced breathing.Von Leupoldt et al. 7 found an increase in dyspnoea unpleasantness/ bother, induced by resistive loaded breathing in healthy individuals, in response to viewing negative compared to positive IAPS images. 7Additionally, Allen and Friedman, using IAPS images in resting healthy individuals, found that negative compared to positive images increased both dyspnoea intensity and unpleasantness induced by slow-paced breathing. 4e have previously demonstrated that in healthy participants, during exercise, dyspnoea could be manipulated in either direction using positive, neutral or negative visual (IAPS) 6 as well as acoustic stimuli (IADS). 5However, unlike our previous study in healthy individuals, 6 we did not find any differences in dyspnoea intensity and bother between the positive and neutral conditions.In our previous study in healthy individuals, participants rated a mean mood valence (SEM) of 4.9 (0.4) while viewing neutral images confirming they felt a neutral affective state. 6In contrast, a closer inspection of our current results, it is evident that participants with COPD reported a mean mood valence (SEM) of 6.4 (0.4) with neutral images which is, in fact, categorized as a happier affective state. 8The fact that we were not able to induce a 'true' neutral affective state in the current study limits the scope comparison for dyspnoea between 'true' neutral with positive or negative affective states.Nonetheless, our results confirm that experimentally induced negative mood is associated with a significantly higher level of dyspnoea intensity and bother compared to positive mood.It is worth noting that in the current study, participants were advised that mood may change with the images.This advice may have generated a level of expectancy, we did not assess expectancy and hence cannot determine the degree to which expectancy influenced the results.
Although the present study was not designed to explore the neural basis of dyspnoea modulation via changes in mood, examination of the literature provides some potentially important speculative insights.Functional brain imaging studies with fMRI have confirmed activations selectively in the hippocampus, amygdala and visual cortex when viewing negative IAPS images and in the superior and middle frontal gyrus, prefrontal cortex and posterior cingulate gyrus with positive images. 25,26Likewise, neural imaging studies in dyspnoea have also shown activation of the amygdala 24,27 and anterior insula 11,12 within the limbic nuclei, suggesting a possible neural connection between mood state and dyspnoea.Thus, the central processing of neural activity within these brain regions offers a possible neurophysiological basis for the impact of mood state on dyspnoea perception.
In light of the findings from the current and related mood modulation studies, it is interesting to address their potential relevance to the issue of dyspnoea in chronic disease.9][30] There is substantial evidence that the prevalence of anxiety and depression is higher in individuals with COPD compared to the general population.Moreover, a significant positive correlation between depression and dyspnoea severity has been reported in COPD, and there is evidence that both anxiety and depression are causal of dyspnoea and that dyspnoea is a determinant of anxiety. 31,32There is growing evidence that treating depression and anxiety and improving mood state may improve quality of life and exercise tolerance; however, it is still unclear if these improvements are due to the alleviation of dyspnoea.Nonetheless, these studies support the idea that mood-enhancement psychotherapeutic techniques can alleviate dyspnoea in clinical populations.
In conclusion, the findings from this study indicate that standardized mood-modulating IAPS stimuli are effective in altering the perception of concurrent exertional dyspnoea through a change in affective states.The fact that, as with our previous studies, leg fatigue was not affected by different affective states, suggests that these primary exercise-limiting symptoms result from different patterns of neural processing.The findings from the current study warrant further clinical trials aimed at exploring the efficacy of mood enhancements in alleviating chronic dyspnoea in clinical populations.

T A B L E 3 Note:
End-exercise HR, SpO 2 and MDP between conditions.Values are represented as mean ± SEM.Abbreviations: HR, heart rate; MDP, multidimensional dyspnoea profile; SpO 2 , peripheral oxygen saturation.a A statistically significant difference between positive and negative ( p > 0.05).bAstatistically significant difference between neutral and negative ( p < 0.05).
T A B L E 1 Baseline characteristics of participants.Baseline POMS, HADS and D-12 between conditions.
Abbreviations: 6MWD (m), six-minute walk distance in metres; BODE, body mass index, obstruction, dyspnoea, and exercise; CAT, Chronic Obstructive Pulmonary Disease [COPD] assessment test; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; mMRC, modified medical research council dyspnoea scale; SGRQ-C, St. George's respiratory questionnaire for COPD patients.T A B L E 2 Note: Values are represented as Mean (±SEM) during positive, neutral, and negative conditions. A. POMS: Profile Mood Scale (depression, vigour, confusion, tension, anger, and fatigue domains), B. HADS: Hospital Anxiety and Depression Scale (anxiety and depression), and C. Dyspnoea-12 total score: Sum of Dyspnoea-12 components.There were no statistically significant differences in POMS, HADS and Dyspnoea-12 between conditions ( p > 0.05).