Alleviation of Dyspnea and Changes in Physical Activity Level by Air Flow to the Face With a Fan

Abstract

BACKGROUND:

Dyspnea is an unpleasant subjective symptom and is associated with decreased physical activity level (PAL). Effect of blowing air toward the face has received a great deal of attention as a symptomatic therapy for dyspnea. However, little is known about the duration of its effect and its impact on PAL. Therefore, this study aimed to measure dyspnea severity and changes in dyspnea and PALs with air blasts to the face.

METHODS:

The trial conducted was open-label, randomized, and controlled. This study included out-patients with dyspnea caused by chronic respiratory deficiency. Subjects were provided a small fan and instructed to blow air toward their faces either twice a day or when having trouble breathing. Subsequently, severity of dyspnea and PALs was measured using visual analog scale and physical activity scale for the elderly (PASE), respectively, before and after 3-week treatment. Amounts of changes in dyspnea and PALs before and after treatment were compared using analysis of covariance.

RESULTS:

Overall, 36 subjects were randomized, and 34 were analyzed. Mean age was 75.4 y (26 males [76.5%] and 8 females [23.5%]). Visual analog scale score for dyspnea (SD) before treatment was 33 (13.9) mm and 42 (17.5) mm in the control and intervention groups, respectively. PASE score before treatment was 78.0 (45.1) and 57.7 (38.0) in the control and intervention groups, respectively. No significant difference in changes in dyspnea severity and PAL was observed between the 2 groups.

CONCLUSIONS:

No significant difference was observed for dyspnea and PALs in subjects after blowing air toward their own faces with a small fan for 3 weeks at home. Disease variability and impact of protocol violations were high due to small number of cases. Further studies with a design focused on subject protocol adherence and measurement methods are required to understand impact of air flow on dyspnea and PAL.

Introduction

Dyspnea is a symptom that often develops in the advanced stages of chronic diseases and malignant tumors of the respiratory and circulatory systems. One of the causes is thought to be the discomfort caused by increased work of breathing or effort during inhalation and/or exhalation. Dyspnea can occur even after simple activities of daily living, such as eating and changing clothes, with worsening of the disease state that causes dyspnea. Therefore, patients with dyspnea have reduced physical activity levels (PALs) and impaired self-management behaviors; both have an impact on their quality of life (QOL).^1-4 Where treating the cause of dyspnea is difficult due to advanced chronic respiratory diseases or advanced cancer, blowing air to the face has gained popularity as a symptomatic therapy.

In 1987, Schwartzstein⁵ confirmed through an experiment that applying cold air to the face alleviates dyspnea felt by healthy subjects. Although the exact reason behind how air flow to the face can alleviate dyspnea is still unclear, Capucine⁶ reported that stimulation of trigeminal nerve facial receptors makes the brain believe that ventilation flow is higher than actual values. Therefore, simply having air blown toward the face results in the sensation of ventilation, similar to breathing, which can alleviate dyspnea.

According to a study in which either oxygen or air was administered by nasal cannula to terminal subjects complaining of dyspnea, but who were not hypoxemic⁷ compared to before the start of treatment, QOL was improved for each method after one week. However, no significant difference was observed between the group that had inhaled oxygen and the group that had inhaled air. Additionally, dyspnea was compared immediately after a few minutes of blowing air toward the face or other parts of the body, and dyspnea was alleviated in subjects who had air blown toward their faces.^8-12 Tsai et al¹³ also conducted a meta-analysis based on 8 studies that performed air blowing at different time points. They concluded that short-term air blowing several times during the day is effective for relieving dyspnea symptoms.¹³

However, in previous studies evaluating short-term air flow to the face, changes in dyspnea severity were measured immediately after the application of air flow, and hence there is insufficient verification of how long the effect can last and how it affects patients’ PAL and QOL. Thus, this study aimed to examine the duration of the effect of air flow to the face on dyspnea by measuring the changes in PALs and the severity of dyspnea after a 3-week intervention and to discuss the effectiveness and benefits of air flow during dyspnea.

QUICK LOOK.

Current knowledge

It has been shown that using a fan to blow airflow on the face several times for a short period can improve dyspnea. The long-term effects of air flow and its impact on the amount of physical activity need to be elucidated.

What this paper contributes to our knowledge

The sample size was small, and no statistical differences in dyspnea and physical activity were identified between the 2 groups divided by frequency and timing of air flow to the face. To clarify the changes in physical activity level due to air flow to the face, it is necessary to consider the target disease, increase adherence to the research protocol, and examine methods of measuring physical activity.

Methods

A 3-week, open-label, randomized study (Clinical trial registration: UMIN-ICDR R000045388 UMIN000039821) was conducted on out-patients with chronic dyspnea who underwent treatment at 4 institutions (private university-based hospitals; a public medical center; and a clinic in the Kanto, Kansai, and Kyushu areas) between March 2020–May 2021. Following a questionnaire survey on dyspnea and PAL, they were randomly classified into 2 groups (treatment: control; 1:1) and were provided with small electric fans. The treatment group was instructed to use the fan to blow air toward their faces whenever they were dyspneic, and the control group was instructed to use the fan only twice a day. The second questionnaire was sent to subjects after 3 weeks, and the same questions as before the treatment were asked. The primary outcome of the study was the extent of change in dyspnea severity before and after the intervention. The main secondary outcome was the magnitude of PAL change before and after the intervention based on the analysis of covariance.

This study was conducted in accordance with the ethical principles based on the Declaration of Helsinki, the Ethical Guidelines for Medical and Health Research Involving Human Subjects (February 28, 2017, Ministry of Health, Labor and Welfare), and the Ethical Guidelines for Nursing Research (2004, Japanese Nursing Association). In addition, this study was conducted following the approval of the research ethics committee of the researchers’ affiliated facility (Research Ethics Committee, University of Tokyo Health Sciences: 19–14H) and the ethical review body of each research facility. Written consent was obtained from all subjects at the start for participation in this study.

Subjects age ≥ 20 y with dyspnea caused by chronic respiratory dysfunctions that affected their daily living activities were targeted. The degree of dyspnea of the target population was evaluated as Modified Medical Research Council scale^14,15 grade II (walking slower on flat roads than people of the same age due to shortness of breath or sometimes stopping due to shortness of breath when walking on flat roads at their own pace) or above. Exclusion criteria included subjects who were hospitalized or received additional prescriptions of oral steroids or new antibiotic administration because of an exacerbation of respiratory failure within the past 4 weeks of inclusion and those who were unable to respond appropriately to questionnaire surveys because of problems related to cognitive function.

The treatment consisted of instructing subjects to direct a fan toward the face after exercise and when feeling short of breath, based on previous studies that showed the technique relieved dyspnea.^8-12 To prevent information bias after receiving the equipment, fans were provided to both the intervention and control groups. Differences in effectiveness were measured by varying the frequency of fan use to ≥ 2 times daily. One small rechargeable fan with a blade diameter of approximately 9 cm (Rechargeable Multi Handy Fan with aroma tray, Hanwa Co, Ltd; https://lifeonproducts.co.jp/en/product/pr-f015/?pc-switcher=1) was provided to each subject. The aroma tray, which is the fan accessory, was labeled with a warning that it should not be used throughout the research period.

It was explained to the intervention group via the instruction manual that “the timing to use the fan is 1–2 min after moving the body and when out of breath.” The points to remember when using the fan were described as follows: (1) to blow air toward the face, focusing on the nose; (2) to use the fan after placing it on a table instead of holding it in one hand; and (3) to use it for approximately 2 min or until shortness of breath is relieved after exercising or when feeling short of breath. The purpose of blowing air toward the nose is to focus the air flow on the face, not other parts of the body, such as the neck. The purpose of using the fan on a table is to prevent interference with respiratory muscles that assist breathing during dyspnea; exercise of these accessory respiratory muscles could be hampered if the fan is held with the upper limb, thereby aggravating dyspnea. Subjects in the control group were also given a fan and were told in the instruction manual “to use the fan twice a day, that is, for 2 min each within an hour after waking up and within an hour before sleeping.” The instruction manual also stated that subjects should use the fan after placing it on a table to blow air toward their faces. Subjects in both groups were informed that the fans would not be returned, and they could continue to use the fan after the research period ended.

As the study progressed at multiple facilities concurrently, block allocation was performed (block size: 4) (http://www.randomization.com). Allocation instructions were provided in sealed opaque envelopes that were hand delivered to research participants along with a fan just before they went home. The study was designed in such a way that neither the subjects nor the medical professionals were aware of the allocation. Before allocation, subjects completed the first questionnaire. The medical staff hand delivered the treatment manual to each group in a sealed state. The subjects opened the manual after reaching home and used the fan according to the instructions provided. The second questionnaire was sent by post to the subjects’ residences, and the subjects returned the completed questionnaire to the researchers.

Demographic information on randomized subjects was collected from medical records, including age, sex, presence of home oxygen therapy, cause of dyspnea, comorbidities, percent of predicted FEV₁ or percent vital capacity (VC) measured in the past 6 months, and average breathlessness (on visual analog scale) and physical activity (on physical activity scale for the elderly [PASE]) in the past week. The primary outcome was the amount of change in average dyspnea in the past one week measured using visual analog scale^14,15 before and after treatment, and the key secondary outcome was the amount of change in average PALs in the past one week measured by PASE.

To measure the degree of dyspnea by visual analog scale, the question “Please tell us about the average difficulty in breathing in the past 7 days” was included in the questionnaire. “I had no difficulty in breathing at all” was printed on the left side of a 100-mm line, whereas “I had the worst difficulty in breathing” was printed on the right, and subjects described their level of breathing difficulty by placing a mark on the 10-cm line proportional to the degree of dyspnea they felt at that time based on the aforementioned 2 end limits. The distance from the left of the 100-mm line to the sign marked by the subjects was measured in millimeter and converted into data. The higher the visual analog scale score the worse the dyspnea. The extent of change in dyspnea severity was calculated by subtracting the pre-treatment from the post-treatment visual analog scale scores, with a positive amount of change indicating dyspnea exacerbation.

PASE is a questionnaire-based assessment scale designed to measure the PAL of the elderly.¹⁶ Although the cohort consisted of all age groups rather than only the elderly, PASE was used because Aota et al¹⁷ reported that it is useful as an index for the PAL of subjects with chronic respiratory diseases. The higher the PASE score the greater the PAL. The amount of change in PAL was calculated by subtracting the pre-treatment from the post-treatment PASE scores, and a positive amount of change indicated an increase in PAL. PASE was purchased from New England Research Institutes (Watertown, Massachusetts). However, because a Japanese version was not available, a version translated by the researchers was used.

The sample size was calculated based on previous studies^8,9 that stated the amount of change in the visual analog scale score for dyspnea was 7.6 mm and 24.6 mm. The difference in the average amount of change in the visual analog scale score before and after treatment between the 2 groups was expected to be 10 mm. The SD was expected to be 15 mm. To obtain a power of 80% at a significance level of 0.05 by t test, the sample size of each group was calculated to be 36 cases. Because interrupted cases and protocol violations during research were expected, we planned to recruit 50 subjects for each group. Based on the principle of intention to treat (ITT), the analysis population was reduced by excluding subjects who responded “I have been using a fan during dyspnea since before participating in this study” to the pre-treatment question.

In the primary analysis, the primary outcome was analyzed using analysis of covariance, with the extent of change in dyspnea severity before and after treatment as the objective variable, the value before treatment as the covariate variable, and the group as a factor. Likewise, the main secondary outcome was analyzed using analysis of covariance, with the magnitude of change in PAL before and after treatment as the objective variable and the baseline value for intergroup differences as the covariate variable. Since an ITT analysis would include cases that did not use the fan as planned, we tried to analyze the data separately for subjects who used the fan as per protocol and for those who broke the protocol. Data for subgroups were analyzed using summary statistics to determine changes before and after the intervention; subsequently, subjects were classified into those whose actual fan use was ≤ 2 times per day and those whose actual fan use was ≥ 3 times per day.

In the case of missing data at baseline, imputation by mean was performed using the average values of the group. Furthermore, if the second data set was missing, the baseline-observation-carried-forward technique was used. A significance level of 0.05 was set for the analyses, and all statistical analyses were performed using EZR (Easy R) Ver 1.52 (Saitama Medical Center, Jichi Medical University, Saitama, Japan),¹⁸ which is a graphical user interface for R (R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R Commander designed to include statistical functions frequently used in biostatistics.

Results

In total, 36 subjects were randomized (control group n = 19, intervention group n = 17) (Fig. 1). The number of subjects was lower than planned because of the extended duration of the study owing to the COVID-19 pandemic. Table 1 shows subject characteristics. There were no noteworthy differences between the 2 groups in terms of age, sex, S_pO2 at rest, the presence or absence of home oxygen therapy, or disease-causing dyspnea. The most common causes of dyspnea were respiratory diseases, especially COPD. Pulmonary function test data could not be collected from all the participants. However, no difference in either percent of predicted FEV₁ or percent VC was observed between the 2 groups. The severity of dyspnea was milder in the control group than in the intervention group. PALs were higher in the control group than in the intervention group, with no statistically significant difference.

Fig. 1.

Of the 83 subjects, 36 who met the criteria and provided consent were randomized. Because of block allocation of 4 subjects per pair at 4 sites, the numbers of patients after randomization were 19 in the control group and 17 in the intervention group. There were no interruptions during the study, but 2 participants who had been using fans before joining the study were excluded from the primary analysis. One patient who did not respond to the number of times he/she used a fan was excluded from the subgroup analysis.

Table 1.

Subject Characteristics at Baseline

There was no interruption in treatment, but 2 subjects stated in the first questionnaire that they had been “blowing air toward the face with a fan” as a “coping method when shortness of breath is severe” since before treatment. Therefore, these 2 subjects were excluded according to the research protocol, and 34 subjects were analyzed (Fig. 1). One subject in the intervention group was excluded from the subgroup analysis because they did not report the actual number of times the fan was used, and 16 and 17 subjects in the intervention and control groups, respectively, were analyzed. Regarding compliance with the protocol, 12 participants of 16 (75.0%) in the intervention group and 11 of 17 (64.7%) in the control group complied with the protocol (Table 2). One fourth of the intervention group used the fan < 2 times, which was like the control group, which was instructed to use the fan 2 times. However, the difference is that participants in the intervention group used the fan during daytime activities, whereas the control group used the fan within an hour of waking and within an hour before bedtime.

Table 2.

Number of Blows

The mean visual analog scale score for dyspnea before and after treatment was 33 (13.9) mm and 34 (23.1) mm, respectively, in the control group (a mean difference of 0.88 [20.1]) and 42 (17.5) mm and 38 (21.9) mm, respectively, in the intervention group (a mean difference of −9.12 [25.9]). The least squares (LS) mean of the amount of change in visual analog scale score for dyspnea, which was analyzed using analysis of covariance with the baseline value as the covariate, was −5.8 mm (95% CI −22.5 to 10.9, P = .48) lower in the intervention group than in the control group (Figure 2A); however, there was no significant difference. Since the number of cases was lower than planned, a Wilcoxon rank-sum test was performed to consider the effect of the distribution of the data, but the results were no different from the analysis of covariance. The mean PASE score before and after treatment was 78.0 (45.1) and 71.7 (53.2), respectively, in the control group (a mean difference of −5.4 [36.0]) and 57.7 (38.0) and 69.6 (43.3), respectively, in the intervention group (a mean difference of 12.2 [34.3]). Similarly, analysis of covariance indicated that the LS mean of the amount of change of both groups was 14.3 (95% CI −11.0 to –39.6, P = .26) higher in the intervention group than in the control group (Figure 2B), but no significant difference was observed.

Fig. 2.

The least squares (LS) mean of changes in dyspnea (A) and physical activity levels (PALs) (B) in each group 3 weeks before and after air blowing. (A) The LS mean of dyspnea is −5.8 lower in the intervention group than in the control group (95% CI −22.5 to 10.9, P = .48). (B) The LS mean of PAL is 14.3 higher in the intervention group than in the control group (95% CI −11.0 to 39.6, P = .26). PAL = physical activity level; LS = least squares.

Fan use varied in both groups. As stated in the Methods section, the main analysis prioritized random groupings according to the principle of ITT, but we considered a detailed analysis to be necessary. To evaluate values in each group, subjects were classified into 2 groups based on frequency of fan use (≤ 2 or ≥ 3 times). Figure 3 shows the boxplots of the results. In the control and intervention groups, subjects with a higher frequency of fan use had a higher median PAL. However, no significant differences were found.

Fig. 3.

Changes in dyspnea (A) and physical activity levels (B) before and after air blowing. The subjects were divided into 4 groups according to the frequency of air blowing (ie, the “few” groups blew air 1–2 times daily, and the “many” groups blew air 3–4 times or 5–6 times daily). C-few = control group with lower frequencies of air blowing; C-many = control group with higher frequencies of air blowing; I-few = intervention group with lower frequencies of air blowing; I-many = intervention group with higher frequencies of air blowing.

Discussion

This 3-week interventional study evaluated changes in visual analog scale scores for dyspnea and PALs with the aim of examining whether the efficacy of short-term air flow using a fan to the face was sustained in subjects with dyspnea. The results revealed no significant difference between the control and intervention groups. The sample size could not be obtained as planned, which may have resulted in a lack of power. In addition, there were differences from previous studies in terms of subject disease, measurement methods, and effects of protocol non-compliance.

The type of disease that caused dyspnea differed between the present study and previous ones. Unlike previous studies^8-12 that reported the efficacy of short-term external air flow on terminal subjects admitted to the palliative care ward complaining of dyspnea, subjects in the present study were out-patients with dyspnea. Many subjects in previous studies were diagnosed with lung cancer, but our subject population had chronic lung diseases. Because the causes of dyspnea are quite complex, as described by Spathis et al,¹⁹ the type of disease may also be related to the method of reducing breathlessness. The differences in results between our study and previous studies could be explained by differences in the target populations. Therefore, consideration should be given to limiting the subjects to COPD and classifying them into groups based on their severity. In addition, it was also necessary to examine differences in dyspnea due to diseases, such as cancer, chronic heart failure, and interstitial pneumonia.

In terms of the questioning method, the present study inquired about the average dyspnea for one week before and after fan use, rather than inquiring about the severity of dyspnea soon after fan use, because the present study aimed to examine whether treatment efficacy was sustained. The visual analog scale, which was used in this study, is a widely used tool for evaluating subjective symptoms. Previous studies,^8,9 based on which our sample size was calculated, also used this scale. Given the nature of the current study’s questioning method, which was inquiring about the average dyspnea for one week, a stricter measurement rule should be used to improve the visual analog scale accuracy for dyspnea. The higher-than-expected number of protocol violations also had a significant impact on the results. However, there was not enough information to examine why some subjects failed to adhere to the protocol. For these subjects, a plan was needed that would enable compliance with the protocol through more detailed explanation and monitoring.

The change in PAL due to the intervention is a new finding, but again, no predominant difference could be confirmed. There are 2 types of tools for measuring changes in PAL: objective data using activity meters and subjective measures such as the PASE. The reason why the PASE, a subjective scale questionnaire method, was used in this study was that it is easy to apply to the evaluation of patients in a clinical setting. We thought that the questionnaire method would be a less burdensome measurement method for subjects in this research. Since we did not combine the questionnaire method with the measurement method in this study, we do not know whether the difference between the 2 groups would have been clearer if the measurement method had been used. However, we believe it is necessary to collect objective data together in the future.

The small number of cases was the most important limitation of this study. The study period coincided with the COVID-19 pandemic, and the number of subjects with chronic respiratory failure who presented to medical facilities decreased. The number of cases should be increased, and accurate data including objective measurement should be used to confirm the effectiveness. Furthermore, the basic metabolic panel could be used to identify the patients with tachypnea due to dyspnea. Regarding the background of the subjects, the subjects in the control group were characterized by milder dyspnea and higher levels of physical activity at baseline. Additionally, it was possible that subjects using long-term home oxygen therapy were less likely to feel the effects of air flow as air was constantly being pumped into their nasal passages.

Conclusions

We classified subjects with dyspnea caused by chronic diseases into 2 groups and instructed them to blow air toward their faces using a small fan twice in the morning and evening or after exercising and whenever they experienced dyspnea. We measured the changes in dyspnea severity and PALs before and after treatment. The differences between the 2 groups were unclear because of the small sample size owing to the COVID-19 pandemic. PAL increased in some subjects who blew air frequently, but the number of subjects was too small to perform statistical tests. Further studies will be required to determine whether air flow can contribute to improve dyspnea and PALs.