https://orcid.org/0000-0003-2139-3985
Figures
Abstract
Background
Pulmonary rehabilitation (PR) for patients with chronic obstructive pulmonary disease (COPD) improves exercise tolerance and COPD assessment test score (CAT). Oxygen supplementation during PR facilitates exercise physiological benefits. This study aimed to assess the feasibility of a trial comparing two oxygen supplementation methods, with the hypothesis that both would be effective and produce distinct outcomes.
Methods
This double-blind, crossover, randomized controlled trial compared two PR programs—Program A (including PR under FiO₂ 0.3) and Program B (including PR under FiO₂ 0.5)—using high-flow nasal cannula oxygen therapy in patients with COPD and exertional dyspnea (n = 6). Data on the 6-minute walk distance (6MWD), CAT, muscle strength, body composition analysis, respiratory function, and joint range of motion were collected. Participants underwent one month of regular PR followed by two months of oxygen-supplemented PR, with data collected again after this period. Statistical significance was set at 0.05 with a power of 0.8, and the required sample size was calculated accordingly.
Results
The required sample size could not be calculated based on the 6MWD. The improvement in CAT by Program A was greater than that by Program B. The improvements in muscle parameters by Program B were greater than those by Program A. The standardized effect size and the corresponding required sample sizes for the CAT, quadriceps muscle power, lower leg circumference, trunk muscle mass, and leg muscle mass were 0.32/81, 0.66/8, 0.17/114, 0.27/88, and 0.24/56, respectively.
Conclusions
Given the small number of participants, the 6MWD and CAT were not appropriate primary endpoints for comparing the effectiveness of the two oxygen supplementations during PR in patients with COPD. However, the quadriceps muscle power was identified as the most suitable primary endpoint among all the investigated parameters.
Citation: Ito A, Morito A, Ishizaka M, Ogawa Y, Kawai Y, Hanawa Y, et al. (2026) Feasibility assessment of double-blind, crossover, randomized controlled trial protocol comparing two oxygen-supplemented pulmonary rehabilitation for patients with chronic obstructive pulmonary disease: A pilot study. PLoS One 21(5): e0348404. https://doi.org/10.1371/journal.pone.0348404
Editor: Jamal Akhtar, The University of Kansas Health Systems St. Francis Campus, UNITED STATES OF AMERICA
Received: September 17, 2025; Accepted: April 14, 2026; Published: May 7, 2026
Copyright: © 2026 Ito et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: AI 24K20446 Grants-in-Aid for Scientific Research（KAKENHI） https://www.jsps.go.jp/j-grantsinaid/ co-first author, co-corresponding author.
Competing interests: NO authors have competing interests.
Introduction
Chronic obstructive pulmonary disease (COPD) is among the top three causes of deaths worldwide, alongside cardiovascular diseases and cancer, with nearly 90% of COPD-related deaths occuring in low- and middle-income countries [1]. The prevalence of COPD varies widely by region, age, and the availability of diagnostic spirometry [1]. It accounts for approximately 5% of respiratory-related deaths and is a leading cause of death in the older population [1]. COPD is caused by smoking, air pollution, biomass exposure, occupational dusts, and host factors (including abnormal lung development and lung aging) and is characterized by reduced alveolar ventilation due to alveolar destruction and ventilation–blood flow imbalance [1,2]. This results in dyspnea, chronic cough, and increased sputum production [3]. Additionally, it causes a decline in physical function, such as exercise tolerance and muscle weakness, and reduces quality of life [4,5].
Pulmonary rehabilitation (PR) plays an important role in the management of patients with COPD by improving exercise tolerance and quality of life and reducing dyspnea and fatigue [6–12]. The PR program for patients with stable COPD typically comprises supervised exercise training lasting approximately 4–12 weeks, generally combining strength training with aerobic exercise [13–15]. Oxygen therapy combined with exercise for patients with COPD has been increasingly evaluated and adopted [16–19]. Evidence is gradually emerging regarding the intervention effects of PR combined with oxygen therapy [20].
Recently, there has been growing interest in the potential benefits of oxygen therapy with a high-flow nasal cannula (HFNC) for COPD. HFNC reduces respiratory muscle load by lowering the arterial partial pressure of carbon dioxide, increasing end-expiratory and end-tidal volumes, and decreasing the respiratory rate, which may improve respiratory patterns [21–23]. Rehabilitation with HFNC has been shown to be superior to conventional rehabilitation using a nasal cannula or Venturi mask in improving the 6-minute walk distance (6MWD) and exercise endurance time in patients with COPD [24,25]. Additionally, exercise training with HFNC may be superior to exercise training with a regular nasal cannula in patients with chronic respiratory failure receiving long-term oxygen therapy [26].
However, most of these studies have focused on in-situ observations with few interventional studies. Additionally, the optimal setting for HFNC has not been adequately investigated. A meta-analysis of HFNC intervention effects found that while some improvements in quality of life and exercise tolerance were observed, the evidence was insufficient [27]. This lack of consistency was partly due to varying HFNC settings across studies.
Therefore, this study aimed to assess the feasibility of a trial comparing HFNC at two FiO2 levels, with the hypothesis that both methods would be effective but differ in their effects. This study may maximize the effects of PR, further improve patients’ exercise capacity and quality of life, and reduce dyspnea. Moreover, identifying the optimal HFNC setting could enhance its clinical utility for patients with COPD.
Materials and methods
Setting and participants
This double-blind, crossover, randomized controlled trial was reviewed and approved by the International University of Health and Welfare Ethics Committee according to the Declaration of Helsinki (approval number: 21-B-4). All patients provided written informed consent to participate in the study. This study is registered at the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (registration number: UMIN000047507). The protocol described in this peer-reviewed article is published on protocols.io (dx.doi.org/10.17504/protocols.io.x54v9bdxml3e/v1) and is included for printing purposes as S1 File.
Patients with COPD with dyspnea on exertion were recruited from the International University of Health and Welfare (IUHW) Shioya Hospital. The diagnosis of COPD was based on forced spirometry showing the presence of a post-bronchodilator forced expiratory volume in 1 s (FEV1)/ forced vital capacity (FVC) < 0.7 [1,28]. Six participants were consecutively recruited for the pilot study from February 2022 to August 2022. At least nine months per participant were required to record the full data.
Protocol of a double-blind, crossover, randomized controlled trial comparing two oxygen-supplemented PR programs
Study participants were randomly assigned to one of the two groups using random number tables: one group started a PR program with oxygen supplementation at FiO₂ 3 (Program A), and the other began a PR program with oxygen supplementation at FiO₂ 0.5 (Program B) (Fig 1). The protocol developed in this study was based on the measurement methods employed in prior research [29,30]. These assignments were concealed from both the participants and the physical therapists responsible for the PR program.
A double-blind, crossover, randomized controlled trial comparing Program A, including 4 weeks of usual PR followed by 8 weeks of PR under FiO2 0.3, with Program B, including 4 weeks of usual PR followed by 8 weeks of PR under FiO2 0.5. The washout period was set for 12 weeks or more. Abbreviations: 6MWD, Six-Minute Walk Distance; CAT, COPD Assessment Test; mMRC, Modified Medical Research Council Dyspnea Scale; LFT, Lung function tests; PR, pulmonary rehabilitation.
The study protocol is illustrated in Fig 1. The participants underwent regular rehabilitation without oxygen supplementation once a week for four sessions (approximately 1 month). They then underwent eight PR sessions once a week (approximately 2 months) with assigned oxygen supplementation. After the completion of the intervention, a washout period of > 3 months was established. Following the washout period, PR was performed once a week for 12 sessions (approximately 2 months) with crossover oxygen supplementation. The washout period was set according to prior research to allow sufficient time for the learning effects of exercise therapy and the intervention effects of aerobic exercise to dissipate [31–33].
A total of six assessments were conducted: at baseline; after the first course of standard rehabilitation; after the first course of O2-supplemented PR; after the washout period; after the second course of standard rehabilitation; and at the end of the second course of O2-supplemented PR. Evaluations included 6MWD, COPD assessment test (CAT), modified Medical Research Council (mMRC) dyspnea scale, quadriceps strength at knee extension, lower leg circumference, and body composition. Respiratory function was assessed on four occasions: at baseline, after the first course of PR, after the washout period, and at the end of the second course of PR.
The 6MWD was measured once, following the protocol from previous studies [34,35]; however, the distance for a walking round trip was set to 15 m in this study. Participants walked at their own pace in comfortable clothing, aiming to cover as much distance as possible in 6 min. Rest breaks were allowed during walking, and the physiotherapist in-charge encouraged participants to rest if subjective fatigue, dyspnea, or a significant drop in SpO₂ (below 90%) were observed. Participants were instructed to walk slowly if SpO₂ fell below 85% and to stop if necessary. Blood pressure, respiratory rate, heart rate, dyspnea, and lower limb fatigue were measured before and after the measurements using the Borg scale.
CAT and mMRC scores were obtained directly from participants using specialized measurement forms [36,37]. Lower scores indicate better conditions for both scores.
The knee extensor strength was measured using a Mobie dynamometer (SAKAI med, Japan). The participant was seated with the knee joint flexed at 90°, and a band was attached to the distal leg. Isometric contraction was used to assess bilateral muscle strength [38]. Measurements were taken twice, and the maximum value was used.
The circumference of the lower leg was measured using a tape at the point of maximum circumference. Measurements were taken three times, and the median value was used [39].
Body composition was measured using a MC-780A-N body composition analyzer (TANITA, Japan). The participants were assessed in a standing position with their feet bare. The measurement periods were set to be the same. Limb and trunk site-specific total muscle mass, body fat percentage, body fat mass, and lean body mass were measured.
Respiratory function was measured using a CHESTAC-8900 respiratory function tester (CHEST, Tokyo, Japan). Lung capacity and forced vital capacity tests were performed using standardized methods [40,41]. The reference values for the FEV1 predicted values were derived from the Japanese standards [42].
The range of motion (trunk flexion and extension) was measured using the methods of the Japanese Orthopaedic Association, Japanese Association of Rehabilitation Medicine, and Japanese Society of Foot Surgery. Measurements were taken from the lateral aspect of the trunk, with the posterior sacrum as the base axis and the line connecting the first thoracic spinous process and the fifth lumbar spinous process as the axis of motion. Care was taken to ensure that hip motion was not included.
Pulmonary rehabilitation under O2 supplementation with HFNC
The O2-supplemented PR protocol consisted of conditioning, resistance exercise (approximately 20 min), resting on a chair, and walking on a treadmill for 20 min under O2 supplementation with HFNC. PR was conducted for approximately 40 min once a week [43,44]. Conditioning focused on the relaxation and stretching of the thoracic and respiratory muscles. Resistance exercises were performed mainly on the lower extremities while monitoring for dyspnea. The treadmill speed was adjusted to target a breathing difficulty level of 4–5 on the revised Borg scale, which is a guideline for aerobic exercise, and the discontinuation criteria were based on the guidelines of the American College of Sports Medicine [45]. Oxygen supplementation was provided using an NKV-330 ventilator (Nihon Kohden), with a flow rate set at 20 L/m unless otherwise indicated. The FiO2 was set at 0.3 or 0.5. HFNC was initiated just prior to the start of conditioning and discontinued after the end of the walking exercise on a treadmill, when the participant was resting in a sitting position [29].
During the rehabilitation intervention, the Transcutaneous Monitor-4 (Radiometer Medical ApS, Denmark) was used to monitor PtcCO2 and PtcO2. Since PtcCO₂ values are typically higher than PaCO2 values, many commercially available instruments have adopted a post-measurement adjustment approach, such as subtracting 4–5 mmHg from the directly measured PtcCO2 values for better estimation of the adult PaCO2. However, we did not adopt such an adjustment; both PtcCO2 and PtcO2 were recorded directly from the tcSensor 84 sensor (Radiometer Medical ApS, Denmark). This approach was applied consistently across all subject runs [29,46,47]. Skin sensors were applied to the forearms of the participants following the manufacturer’s recommendations.
Safety of HFNC settings
Previous studies on respiratory rehabilitation interventions using HFNC have employed flow rates ranging from 20 to 60 L/min, which is inconsistent [27]. Concerns have been raised that both low and high flow rates could lead to issues, such as CO₂ accumulation and over-reduction. Additionally, high flow rates may increase discomfort and should be considered [48]. Based on the results of previous studies, PtcCO₂ and PtcO₂ were monitored during rehabilitation using the Transcutaneous Monitor, with flow rates between 20 and 40 L/min. At lower flow rates (20 L/min), no issues with elevated PtcCO₂ were observed. However, at higher flow rates (40 L/min), PtcCO₂ dropped considerably during and after exercise therapy, and discomfort was reported (Fig 2). Therefore, the flow rate in this study was set at 20 L/min.
(A) O2 loading (Flow rate 40 L/m). Stretching and Resistance training begins. (B) Stretching and resistance training completed. (C) Aerobic exercise begins. (D) Aerobic exercise completed. (a) During aerobic exercise, CO₂ levels were low. Seventeen minutes and 30 seconds after the start of the treadmill, the patient complained of dizziness, and the exercise was stopped. (b) CO₂ levels decreased further after exercise cessation. The subject continued to report dizziness.
Calculation of the number of subjects needed
Power-based sample size calculations for crossover randomized can be performed using the methods described by Grady et al. [31,49]. Calculating the required sample sizes is essential for feasibility assessments in pilot studies. To determine the required sample size, standardized effect sizes were calculated [31,49,50], targeting a significance level of 0.05 and a power of 0.8. We assumed that both methods (Program A and Program B) would be effective and that their effectiveness would differ. Initially, the 6MWD was selected as the primary endpoint; however, as the study progressed, we reconsidered its suitability due to cases where 6MWD showed no improvement after PR. Our aim then shifted to identifying a more appropriate indicator that demonstrated significant gains post-PR to use in calculating the required sample size. A larger standardized effect size corresponds to a smaller sample size, indicating better feasibility [31,49,50]. The change in Program B, including FiO2 0.5 PR, compared to that in Program A, including FiO2 0.3 PR, was defined as the effect size (E), and the standard deviation (S) was calculated from all measurements in the six patients with COPD. The standardized effect size was then calculated as E/S.
Statistical analysis
Data are expressed as mean ± standard deviation unless otherwise indicated. A two-tailed Student’s paired t-test was used to compare between baseline and after walking rehabilitation. Additionally, a sub-analysis was conducted to assess correlations among evaluation indices using Spearman’s rank correlation coefficient. Statistical significance was set at p < 0.05. Commercially available statistical software (BellCurve for Excel; Social Survey Research Information Co., Ltd., Tokyo, Japan) and SPSS version 25 (IBM Corp., Armonk, NY, USA) were used for the statistical analyses.
Results
Profiles of participants
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) grades of the six participants (4 male, 2 female; mean age 76.0 years, 95% confidence interval [71.8–80.2]) in the study were as follows: GOLD 1 (Mild, FEV1 ≥ 80% predicted), 3 cases (2 male, 1 female); GOLD 2 (Moderate, 50% ≤ FEV1 < 80% predicted), 1 case (female); GOLD 3 (Severe, 30% ≤ FEV1 < 50% predicted), 2 cases (1 male, 1 female); GOLD 4 (Very severe, FEV1 < 30% predicted), 0 cases. The other profiles of the participants are presented in Table 1. Three patients initially underwent the FiO2 0.3 PR, and following the washout and crossover, they underwent the FiO2 0.5 PR (Fig 1). The other three patients initially underwent the FiO2 0.5 PR, and after the washout and crossover, they underwent the FiO2 0.3 PR (Fig 1).
Possible primary endpoints before and after PR intervention
Changes in various parameters before and after the PR intervention are shown in S1 File. The 6MWD, CAT, and items showing significant improvement before and after the intervention were evaluated for possible primary endpoints.
The results of the 6MWD and CAT before and after Programs A and B, including PR under O2 supplementation with HFNC, are shown in Fig 3. No significant difference in 6MWD was observed between pre- and post-intervention for Programs A and B (Fig 3A). The change in 6MWD from pre- to post-intervention (Δ6MWD) for Program B, including FiO2 0.5 PR, was −6.5 ± 59.9 m, indicating a decrease from baseline performance (Fig 3B). The CAT score markedly decreased following the intervention, from 16.9 ± 7.9 pre-intervention to 10.0 ± 5.9 post-intervention (Fig 3C). ΔCAT was −8.2 ± 6.5 for Program A, including FiO2 0.3 PR, and −5.7 ± 9.4 for Program B, including FiO2 0.5 PR, indicating greater improvement with the Program A than with the Program B (Fig 3D). Lower scores indicate better CAT scores.
(A) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), did not significantly increase 6MWD (n = 6). (B) The Δ6MWD for Program B, including FiO2 0.5 PR, indicated a decrease from baseline performance (n = 6). (C) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly decreased CAT score (n = 6). (D) The decrease in CAT score after Program A, including FiO2 0.3 PR, was larger than that after Program B, including FiO2 0.5 PR (n = 6). *: p < 0.05, Line graph: ♦Mean. Colors represent each subject.
The left and right knee extensor strength results are presented in Fig 4. Right knee extensor strength (rt quadriceps muscle power) significantly increased from 25.0 ± 8.4 kg to 29.9 ± 10.0 kg (p < 0.05, Fig 4A), and Δ (rt quadriceps muscle power) increased in an O2-dependent manner (Fig 4B). Left knee extensor strength (lt quadriceps muscle power) significantly increased from 25.1 ± 10.3 kg to 29.0 ± 11.2 kg after the intervention (p < 0.05, Fig 4C), and Δ (lt quadriceps muscle power) increased in an O2-dependent manner (Fig 4D). Overall, both right and left quadriceps muscle power showed significant improvement after the intervention (p < 0.01, Fig 4E), with Δ (rt or lt quadriceps muscle power) displaying an O2-dependent increase (Fig 4F).
(A) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased rt quadriceps muscle power (n = 6). (B) O2-supplemented PR increased rt quadriceps muscle power in a dose-dependent manner (n = 6). (C) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased lt quadriceps muscle power (n = 6). (D) O2-supplemented PR increased lt quadriceps muscle power in a dose dependent manner (n = 6). (E) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased (rt or lt) quadriceps muscle power (n = 12 from 6 patients). (F) O2-supplemented PR increased (rt or lt) quadriceps muscle power in a dose-dependent manner (n = 12 from 6 patients). *: p < 0.05, **: p < 0.01, Line graph: ♦Mean. Colors represent each subject.
The left and right calf circumference results are shown in Fig 5. The right calf circumference significantly increased from 32.5 ± 2.5 cm to 33.0 ± 2.8 cm after the intervention (p < 0.01, Fig 5A), and Δ (rt calf circumference) increased in an O2-dependent manner (Fig 5B). The left calf circumference significantly increased from 32.5 ± 2.7 cm to 33.1 ± 2.8 cm after the intervention (p < 0.01, Fig 5C), and Δ (lt calf circumference) increased in an O2-dependent manner (Fig 5D). Overall, both right and left calf circumferences showed significant improvement after the intervention (p < 0.001, Fig 5E), with Δ (rt or lt calf circumference) displaying an O2-dependent increase (Fig 5F).
(A) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased rt lower leg circumference (n = 6). (B) O2-supplemented PR increased rt lower leg circumference in a dose-dependent manner (n = 6). (C) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased lt lower leg circumference (n = 6). (D) O2-supplemented PR increased lt lower leg circumference in a dose-dependent manner (n = 6). (E) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased rt and lt lower leg circumference (n = 12 from 6 patients). (F) O2-supplemented PR increased rt and lt lower leg circumference in a dose-dependent manner (n = 12 from 6 patients). **: p < 0.01, ***: p < 0.001, Line graph: ♦Mean. Colors represent each subject.
The muscle mass results of the trunk and both lower limbs are presented in Fig 6. The trunk muscle mass significantly increased from 20.2 ± 2.6 kg to 20.4 ± 2.5 kg (p < 0.05, Fig 6A), and Δ (trunk muscle mass) increased in an O2-dependent manner (Fig 6B). The right lower limb muscle mass significantly increased from 5.8 ± 1.2 kg to 5.9 ± 1.2 kg (p < 0.05, Fig 6C), and Δ (rt leg muscle mass) increased in an O2-dependent manner (Fig 6D). Overall, both right and left lower limb muscle masses were significantly increased (p < 0.01, Fig 6E), with Δ (rt or lt leg muscle mass) demonstrating an O2-dependent increase (Fig 6F).
(A) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased trunk muscle mass (n = 6). (B) O2-supplemented PR increased trunk muscle mass in a dose-dependent manner (n = 6). (C) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased rt leg muscle mass (n = 6). (D) O2-supplemented PR increased rt leg muscle mass in a dose-dependent manner (n = 6). (E) Program A or B, including PR with O2 supplementation (FiO2 0.3 or 0.5, respectively), significantly increased rt and lt leg muscle mass (n = 12 from 6 patients). (F) O2-supplemented PR dose-dependently increased (rt and left) leg muscle mass (n = 12 from 6 patients). *: p < 0.05, **: p < 0.01, Line graph: ♦Mean. Colors represent each subject.
No parameters in the lung function tests showed significant changes between pre- and post-O2-supplemented PR interventions (S1 File, latter part).
Change in physical function before and after PR with O₂ supplementation and sample size calculation
The required sample size could not be calculated for the change in 6MWD. For the change in CAT score, the effect size (E/S) was 0.32, with a two-tailed sample size of 81 (Table 2). Because all the muscle-related items in Table 3 significantly increased in an O2-dependent manner with O2 supplementation (S1 File), we adopted the one-tailed calculation for these items. For the change in knee extension strength, the sample size (one-tailed) was 70 with an E/S of 0.30 for the right leg only and 7 with an E/S of 1.07 for the left leg only (one-tailed). For combined left and right leg sampling, an E/S of 0.66 required a sample size of 8 (one-tailed). For the change in lower limb circumference, sampling both legs required a sample size of 8 (one-tailed), with an E/S of 0.66. Sampling only the right leg required 38 participants (one-tailed) with an E/S of 0.29, while sampling only the left leg required 310 (one-tailed) with an E/S of 0.035. For combined left and right leg sampling, an E/S of 0.17 required a sample size of 114 (one-tailed). For muscle mass, trunk muscle mass changes had an E/S of 0.27 with a sample size (one-tailed) of 88, and the right lower extremity muscle mass had an E/S of 0.24 with a sample size (one-tailed) of 111. Sampling both lower extremities required a sample size of 56 (one-tailed) with an E/S of 0.24 (Table 3). These sample sizes were calculated for a one-tailed test; for a two-tailed test, the required sample size would be double.
Discussion
We initially attempted to use the 6MWD as the primary endpoint to compare the effects of PR with two different oxygen addition doses; however, the 6MWD did not prove suitable for achieving our objectives. Most patients with COPD are older and may experience limitations from factors such as pain from co-existing musculoskeletal or neurological disorders, as well as breathlessness caused by mild respiratory infections, heart disease, or cancer [51]. These factors make accurate sample size calculation difficult or even impossible. The 6MWD, influenced by many confounding factors, may not accurately reflect the long-term effects of PR. While the CAT is still commonly used to assess PR in COPD, it remains a subjective symptom assessment rather than an objective measure.
Among the items assessed in the present study, those focusing on lower limb muscle strength, which is an objective measure, could serve as direct and accurate indicators of PR achievement. Future PR studies may benefit from using these items as primary endpoints. Between thigh muscle strength and lower leg circumference, thigh muscle strength demonstrated a large effect size in assessing PR achievement and proved to be practical for calculating the required sample size.
It has been reported that quadriceps strength can be improved through conventional exercise therapy [52,53]. Furthermore, a correlation between COPD severity and prognosis has been demonstrated, suggesting that it may be a useful indicator [54].
Both thigh muscle strength and lower leg circumference can be measured in a sitting position, which could be a valid index for patients with COPD with advanced gait difficulties. For patients with COPD, an additional assessment focusing on lower limb muscle strength was considered to allow a more detailed assessment of which rehabilitation plan would be better and to what extent is actually being achieved.
Of the nine items showing significant improvement before and after intervention, the smallest P value was observed for the left and right lower leg circumference. Despite the statistical significance, the difference between the FiO2 0.3 PR group and the FiO2 0.5 PR group was small. The study aimed to compare the two FiO2 groups on various outcomes, and the most pronounced difference was in quadriceps muscle power (standardized effect size), indicating that the level of oxygen supplementation had a larger effect on muscle strength than on other outcomes. We concluded that using thigh muscle strength as the primary endpoint would be the most feasible approach, as it would require fewer participants to achieve statistically meaningful results, thus optimizing the study’s efficiency.
Given that several participants required a higher FiO2 for adequate exercise therapy (i.e., higher than FiO2 0.3), it was reasonable to assume that the improvement in the muscular system index values would be O2 dose-dependent. However, the CAT, the index value for subjective symptoms, did not improve in an O2 amount-dependent manner; FiO2 0.3 PR showed greater improvement than FiO2 0.5 PR. To further investigate this phenomenon, we examined the correlation between the ΔCAT and the change in various index values and found that thigh muscle strength exhibited a marked positive correlation, whereas trunk muscle mass demonstrated a marked inverse correlation. CAT was more indicative of improvement with decreasing values, which suggests that trunk muscle mass is an important indicator. However, improving thigh muscle strength could potentially worsen subjective symptoms, as it may increase the physical load during daily activities. Further research with a larger sample size is required to confirm this hypothesis.
We examined the changes in pulmonary function test scores and found no significant differences between pre- and post-PR, and none of these scores was useful as a primary endpoint. Although some previous studies have reported significant before-and-after differences [55], we believe that pulmonary function tests currently do not provide a useful primary endpoint for a rigorous study design.
Regarding the safety of HFNC, none of the participants complained of any discomfort except for dizziness, numbness, and headache caused by hyperventilation during the preliminary study with a flow rate of 20 L/m. We set 20 L/m as the basic flow rate. When PtcCO2 was high, the flow rate might be increased to 40–50 L/m to prevent CO2 accumulation. When PtcCO2 was low, it was considered best to discontinue HFNC immediately once rehabilitation is complete.
Although this was a pilot study, we found that using thigh muscle strength as the primary endpoint would allow for future studies to be conducted without a substantial increase in participant numbers. However, increasing the sample size would enhance the ability to examine the correlations between ΔCAT and changes in various other indices. We are considering expanding the participant pool and preparing a final report.
There is a consensus that the 6MWD correlates with COPD prognosis, whereas the relationship between the lower extremity muscle strength system, trunk muscle mass, and COPD prognosis remains less clear. Some studies have reported improvements in lower extremity muscle strength with respiratory rehabilitation [56,57]. However, existing data on the relationship between various muscle strength parameters and prognosis in patients with COPD may be insufficient, warranting further in-depth evaluation. Muscle strength, particularly thigh muscle strength, may be useful in designing studies to assess more effective methods of respiratory rehabilitation.
As a pilot study, this research has some key limitations. First, the small sample size restricts the ability to draw definitive conclusions and limits the generalizability of the findings. Additionally, the use of the 6MWD and the CAT, both of which are influenced by a range of confounding factors such as comorbidities and subjectivity, may have compromised the accuracy of assessing the true effects of PR. Future studies should address these limitations by increasing the sample size to improve statistical power and the generalizability of the findings.
Conclusion
By conducting a pilot double-blind, crossover, randomized controlled trial comparing two oxygen-supplemented PR in patients with COPD, we found that quadriceps muscle power seemed useful as a primary endpoint for this study’s feasibility. We conclude that researchers should not focus solely on 6MWD and should thoroughly investigate how quadriceps muscle strength is related to the prognosis of COPD.
Supporting information
S1 File. Step-by-step protocol (also available on protocols.io).
https://doi.org/10.1371/journal.pone.0348404.s001
(PDF)
S2 File. Changes before and after oxygen-supplemented pulmonary rehabilitation.
Data on lung function tests are shown in the latter part of this table. *: p < 0.05, **: p < 0.01, ***: p < 0.001. Abbreviations: 6MWD, Six-Minute Walk Distance; CAT, COPD Assessment Test; mMRC, Modified Medical Research Council Dyspnea Scale; BMI, Body Mass Index; VC, Vital Capacity; ERV, Expiratory Reserve Volume; IRV, Inspiratory Reserve Volume; TV, Tidal Volume; IC, Inspiratory Capacity; FVC, Forced Vital Capacity; FEV1.0, Forced Expiratory Volume in 1 Second; PEF, Peak Expiratory Flow; V75, Expiratory Flow at 75% of Forced Vital Capacity; V50, Expiratory Flow at 50% of Forced Vital Capacity; V25, Expiratory Flow at 25% of Forced Vital Capacity; MMF, Maximal Mid-Expiratory Flow.
https://doi.org/10.1371/journal.pone.0348404.s002
(DOCX)
S3 File. Results of Spearman’s rank correlation coefficients between each preintervention assessment index.
Abbreviations: 6MWD, Six-Minute Walk Distance; CAT, COPD Assessment Test; mMRC, Modified Medical Research Council Dyspnea Scale; Rt, Right; Lt, Left; BMI, Body Mass Index.
https://doi.org/10.1371/journal.pone.0348404.s003
(DOCX)
S4 File. Spearman’s rank correlation coefficient results between each pre-post change measure.
Abbreviations: 6MWD, Six-Minute Walk Distance; CAT, COPD Assessment Test; mMRC, Modified Medical Research Council Dyspnea Scale; Rt, Right; Lt, Left; BMI, Body Mass Index.
https://doi.org/10.1371/journal.pone.0348404.s004
(DOCX)
Acknowledgments
The authors thank all the medical staff at the International University of Health and Welfare Shioya Hospital and the patients who participated in this study.
References
- 1.
Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: 2026 report. https://goldcopd.org/2026-gold-report-and-pocket-guide/
- 2.
Grippi MA, Callen JP. Approach to the patient with respiratory symptoms. Fishman’s pulmonary diseases and disorders. 6th ed. New York: McGraw-Hill Education. 2023. p. 394–432.
- 3.
West JB, Luks AM. Obstructive diseases. West’s pulmonary pathophysiology, the essentials. 10th ed. Philadelphia: Wolters Kluwer. 2022. p. 69–106.
- 4. Attaway AH, Welch N, Hatipoğlu U, Zein JG, Dasarathy S. Muscle loss contributes to higher morbidity and mortality in COPD: An analysis of national trends. Respirology. 2021;26(1):62–71. pmid:32542761
- 5. Liu W, Liu Y, Li X. Impact of Exercise Capacity Upon Respiratory Functions, Perception of Dyspnea, and Quality of Life in Patients with Chronic Obstructive Pulmonary Disease. Int J Chron Obstruct Pulmon Dis. 2021;16:1529–34. pmid:34103910
- 6. Rochester CL, Alison JA, Carlin B, Jenkins AR, Cox NS, Bauldoff G, et al. Pulmonary Rehabilitation for Adults with Chronic Respiratory Disease: An Official American Thoracic Society Clinical Practice Guideline. Am J Respir Crit Care Med. 2023;208(4):e7–26. pmid:37581410
- 7. Spruit MA, Singh SJ, Garvey C, ZuWallack R, Nici L, Rochester C, et al. An official American Thoracic Society/European Respiratory Society statement: key concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Med. 2013;188(8):e13-64. pmid:24127811
- 8. Bourbeau J, Bhatt SP. Constructing Modern Pulmonary Rehabilitation: Another Brick from the Wall. Am J Respir Crit Care Med. 2023;207(7):804–5. pmid:36656552
- 9. Holland AE, Cox NS, Houchen-Wolloff L, Rochester CL, Garvey C, ZuWallack R. Defining modern pulmonary rehabilitation. An official American Thoracic Society workshop report. Ann Am Thorac Soc. 2021;18(5):e12-29.
- 10. Rugbjerg M, Iepsen UW, Jørgensen KJ, Lange P. Effectiveness of pulmonary rehabilitation in COPD with mild symptoms: a systematic review with meta-analyses. Int J Chron Obstruct Pulmon Dis. 2015;10:791–801. pmid:25945044
- 11. Kerti M, Balogh Z, Kelemen K, Varga JT. The relationship between exercise capacity and different functional markers in pulmonary rehabilitation for COPD. Int J Chron Obstruct Pulmon Dis. 2018;13:717–24. pmid:29535512
- 12. McCarthy B, Casey D, Devane D, Murphy K, Murphy E, Lacasse Y. Pulmonary rehabilitation for chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2015;2015(2):CD003793. pmid:25705944
- 13. Higashimoto Y, Ando M, Sano A, Saeki S, Nishikawa Y, Fukuda K, et al. Effect of pulmonary rehabilitation programs including lower limb endurance training on dyspnea in stable COPD: A systematic review and meta-analysis. Respir Investig. 2020;58(5):355–66. pmid:32660900
- 14. Neder JA, Marillier M, Bernard A-C, James MD, Milne KM, O’Donnell DE. The Integrative Physiology of Exercise Training in Patients with COPD. COPD. 2019;16(2):182–95. pmid:31094224
- 15. Pancera S, Lopomo NF, Porta R, Sanniti A, Buraschi R, Bianchi LNC. Effects of Combined Endurance and Resistance Eccentric Training on Muscle Function and Functional Performance in Patients With Chronic Obstructive Pulmonary Disease: Randomized Controlled Trial. Arch Phys Med Rehabil. 2024;105(3):470–9. pmid:37716519
- 16. Garrod R, Paul EA, Wedzicha JA. Supplemental oxygen during pulmonary rehabilitation in patients with COPD with exercise hypoxaemia. Thorax. 2000;55(7):539–43. pmid:10856310
- 17. Emtner M, Porszasz J, Burns M, Somfay A, Casaburi R. Benefits of supplemental oxygen in exercise training in nonhypoxemic chronic obstructive pulmonary disease patients. Am J Respir Crit Care Med. 2003;168(9):1034–42. pmid:12869359
- 18. Nonoyama ML, Brooks D, Guyatt GH, Goldstein RS. Effect of oxygen on health quality of life in patients with chronic obstructive pulmonary disease with transient exertional hypoxemia. Am J Respir Crit Care Med. 2007;176(4):343–9. pmid:17446339
- 19. Moga AM, de Marchie M, Saey D, Spahija J. Mechanisms of non-pharmacologic adjunct therapies used during exercise in COPD. Respir Med. 2012;106(5):614–26. pmid:22341681
- 20. Nonoyama ML, Brooks D, Lacasse Y, Guyatt GH, Goldstein RS. Oxygen therapy during exercise training in chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2007;2007(2):CD005372. pmid:17443585
- 21. Pisani L, Vega ML. Use of Nasal High Flow in Stable COPD: Rationale and Physiology. COPD. 2017;14(3):346–50. pmid:28459282
- 22. Bräunlich J, Beyer D, Mai D, Hammerschmidt S, Seyfarth H-J, Wirtz H. Effects of nasal high flow on ventilation in volunteers, COPD and idiopathic pulmonary fibrosis patients. Respiration. 2013;85(4):319–25. pmid:23128844
- 23. Bräunlich J, Köhler M, Wirtz H. Nasal highflow improves ventilation in patients with COPD. Int J Chron Obstruct Pulmon Dis. 2016;11:1077–85. pmid:27307723
- 24. Vitacca M, Paneroni M, Zampogna E, Visca D, Carlucci A, Cirio S, et al. High-Flow Oxygen Therapy During Exercise Training in Patients With Chronic Obstructive Pulmonary Disease and Chronic Hypoxemia: A Multicenter Randomized Controlled Trial. Phys Ther. 2020;100(8):1249–59. pmid:32329780
- 25. Fang T-P, Chen Y-H, Hsiao H-F, Cho H-Y, Tsai Y-H, Huang C-C, et al. Effect of high flow nasal cannula on peripheral muscle oxygenation and hemodynamic during paddling exercise in patients with chronic obstructive pulmonary disease: a randomized controlled trial. Ann Transl Med. 2020;8(6):280. pmid:32355724
- 26. Chihara Y, Tsuboi T, Sumi K, Sato A. Effectiveness of high-flow nasal cannula on pulmonary rehabilitation in subjects with chronic respiratory failure. Respir Investig. 2022;60(5):658–66. pmid:35644803
- 27. Oltra G, Ricciardelli M, Virgilio S, Fernandez Parmo D, Ruiz A, Liquitay CME, et al. High-flow nasal cannula during pulmonary rehabilitation for people with chronic obstructive pulmonary disease: a systematic review and meta-analysis. Physiother Res Int. 2024;29(2): e2088.
- 28.
Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global strategy for prevention, diagnosis and management of COPD: 2022 report. 2022. https://goldcopd.org/wp-content/uploads/2021/12/GOLD-REPORT-2022-v1.1-22Nov2021_WMV.pdf
- 29. Umeda A, Morito A, Ishizaka M, Ito A, Ogawa Y, Kawai Y, et al. Transcutaneous CO2 and O2 monitoring during walking with a high-flow nasal cannula in patients with chronic obstructive pulmonary disease. Respir Investig. 2025;63(5):887–97. pmid:40669279
- 30. Chihara Y, Tsuboi T, Sumi K, Tachibana H, Sato A. Effect of high fraction of inspired oxygen and high flow on exercise tolerance in patients with COPD and IPF: A randomized crossover trial. Respir Investig. 2025;63(3):431–7. pmid:40174242
- 31.
Grady DG, Cummings SR, Hulley SB. Alternative clinical trial designs and implementation issues. In: Hulley SB, Cummings SR, Browner WS, Grady DG, Newman TB, editors. Designing clinical research. 4th ed. Philadelphia: LIPPINCOTT WILLIAMS & WILKINS, a WOLTERS KLUWER business. 2013. p. 151–70.
- 32. Minoguchi H, Shibuya M, Miyagawa T, Kokubu F, Yamada M, Tanaka H, et al. Cross-over comparison between respiratory muscle stretch gymnastics and inspiratory muscle training. Intern Med. 2002;41(10):805–12. pmid:12413000
- 33. Volaklis KA, Douda HT, Kokkinos PF, Tokmakidis SP. Physiological alterations to detraining following prolonged combined strength and aerobic training in cardiac patients. Eur J Cardiovasc Prev Rehabil. 2006;13(3):375–80. pmid:16926667
- 34. Beekman E, Mesters I, Hendriks EJM, Klaassen MPM, Gosselink R, van Schayck OCP, et al. Course length of 30 metres versus 10 metres has a significant influence on six-minute walk distance in patients with COPD: an experimental crossover study. J Physiother. 2013;59(3):169–76. pmid:23896332
- 35. Agarwala P, Salzman SH. Six-Minute Walk Test: Clinical Role, Technique, Coding, and Reimbursement. Chest. 2020;157(3):603–11. pmid:31689414
- 36. Jones PW, Tabberer M, Chen W-H. Creating scenarios of the impact of COPD and their relationship to COPD Assessment Test (CAT™) scores. BMC Pulm Med. 2011;11:42. pmid:21835018
- 37. Bestall JC, Paul EA, Garrod R, Garnham R, Jones PW, Wedzicha JA. Usefulness of the Medical Research Council (MRC) dyspnoea scale as a measure of disability in patients with chronic obstructive pulmonary disease. Thorax. 1999;54(7):581–6. pmid:10377201
- 38. Katoh M. Test-retest reliability of isometric ankle plantar flexion strength measurement performed by a hand-held dynamometer considering fixation: examination of healthy young participants. J Phys Ther Sci. 2022;34(6):463–6. pmid:35698554
- 39. Kiss CM, Bertschi D, Beerli N, Berres M, Kressig RW, Fischer AM. Calf circumference as a surrogate indicator for detecting low muscle mass in hospitalized geriatric patients. Aging Clin Exp Res. 2024;36(1):25. pmid:38321234
- 40. Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardisation of spirometry. Eur Respir J. 2005;26(2):319–38. pmid:16055882
- 41. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. American Thoracic Society. Am J Respir Crit Care Med. 1995;152(5 Pt 2):S77-121. pmid:7582322
- 42. Kubota M, Kobayashi H, Quanjer PH, Omori H, Tatsumi K, Kanazawa M, et al. Reference values for spirometry, including vital capacity, in Japanese adults calculated with the LMS method and compared with previous values. Respir Investig. 2014;52(4):242–50. pmid:24998371
- 43. Baumann HJ, Kluge S, Rummel K, Klose H, Hennigs JK, Schmoller T, et al. Low intensity, long-term outpatient rehabilitation in COPD: a randomised controlled trial. Respir Res. 2012;13(1):86. pmid:23017153
- 44. O’Neill B, McKevitt A, Rafferty S, Bradley JM, Johnston D, Bradbury I, et al. A comparison of twice- versus once-weekly supervision during pulmonary rehabilitation in chronic obstructive pulmonary disease. Arch Phys Med Rehabil. 2007;88(2):167–72. pmid:17270513
- 45.
American College of Sports Medicine ACSM. Exercise prescription for individuals with cardiovascular and pulmonary diseases. ACSM’s guidelines for exercise testing and prescription. 11 ed. Philadelphia: Wolters Kluwer. 2022. p. 226–75.
- 46. Umeda A, Ishizaka M, Tasaki M, Yamane T, Watanabe T, Inoue Y, et al. Evaluation of time courses of agreement between minutely obtained transcutaneous blood gas data and the gold standard arterial data from spontaneously breathing Asian adults, and various subgroup analyses. BMC Pulm Med. 2020;20(1):151. pmid:32471394
- 47. Umeda A, Ishizaka M, Ikeda A, Miyagawa K, Mochida A, Takeda H, et al. Recent Insights into the Measurement of Carbon Dioxide Concentrations for Clinical Practice in Respiratory Medicine. Sensors (Basel). 2021;21(16):5636. pmid:34451079
- 48. Prieur G, Medrinal C, Combret Y, Dupuis Lozeron E, Bonnevie T, Gravier F-E, et al. Nasal high flow does not improve exercise tolerance in COPD patients recovering from acute exacerbation: A randomized crossover study. Respirology. 2019;24(11):1088–94. pmid:31387158
- 49.
Hulley SB, Cummings SR, Browner WS, Grady DG, Newman TB. Estimating sample size and power: applications and examples. Designing clinical research. 4th ed. Philadelphia: LIPPINCOTT WILLIAMS & WILKINS, a WOLTERS KLUWER business. 2013. p. 55–83.
- 50. Noordzij M, Tripepi G, Dekker FW, Zoccali C, Tanck MW, Jager KJ. Sample size calculations: basic principles and common pitfalls. Nephrol Dial Transplant. 2010;25(5):1388–93. pmid:20067907
- 51. Liu X, Cao Y, Shi Y, Ding H. Association between multimorbidity and disability among elderly with chronic obstructive pulmonary disease in Shanghai, China: a cross-sectional study. BMC Geriatr. 2025;25(1):901. pmid:41239268
- 52. He Z, Cao B, Liu K, Wei Q. Skeletal Muscle Function in Relation to COPD Severity and Its Predictive Significance for Disease Progression. Int J Chron Obstruct Pulmon Dis. 2025;20:389–97. pmid:40008111
- 53. Swallow EB, Reyes D, Hopkinson NS, Man WD-C, Porcher R, Cetti EJ, et al. Quadriceps strength predicts mortality in patients with moderate to severe chronic obstructive pulmonary disease. Thorax. 2007;62(2):115–20. pmid:17090575
- 54. Santin L, Fonseca J, Hirata RP, Hernandes NA, Pitta F. Minimal important difference of two methods for assessment of quadriceps femoris strength post exercise program in individuals with COPD. Heart Lung. 2022;54:56–60. pmid:35390575
- 55. Menier RJ, Talmud J. Benefits of a multidisciplinary pulmonary rehabilitation program. Chest. 1994;105(2):640–1. pmid:8179664
- 56. Maltais F, Decramer M, Casaburi R, Barreiro E, Burelle Y, Debigaré R, et al. An official American Thoracic Society/European Respiratory Society statement: update on limb muscle dysfunction in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2014;189(9):e15-62. pmid:24787074
- 57. De Brandt J, Spruit MA, Hansen D, Franssen FM, Derave W, Sillen MJ, et al. Changes in lower limb muscle function and muscle mass following exercise-based interventions in patients with chronic obstructive pulmonary disease: A review of the English-language literature. Chron Respir Dis. 2018;15(2):182–219. pmid:28580854