Abstract
BACKGROUND: This study assessed the clinical effects of a ventilatory assist (VA) device in addition to supplemental O2 (VA+O2) on exercise endurance in subjects with severe to very severe COPD managed with long-term oxygen therapy (LTOT).
METHODS: This was a crossover clinical feasibility study of the effects of VA+O2 in subjects with severe to very severe COPD managed with LTOT (N = 15). At visit 1, physiologic measures were obtained, and subjects were tested on the cycle ergometer with VA. Peak work rate and flow for continuous supplemental O2/VA+O2 were established. At visit 2, subjects exercised at a constant work rate of 80% peak work rate to maximum endurance after allocation to VA+O2 or O2. Cardiorespiratory variables, work rate, and dyspnea were included to define potential clinical benefits of VA+O2. Data were analyzed using a linear mixed model.
RESULTS: Fifteen subjects with COPD (mean ± SD, age 67.9 ± 9.0 y, FEV1 0.89 ± 0.35 observed) completed the study. Exercise duration in minutes was significantly longer with VA+O2 versus O2 (least squares mean [standard error], 12.0 [2.0] vs 6.2 [2.0], P = .01). VA+O2 versus O2 was also associated with significantly greater isotime improvements in Borg dyspnea scores (3.6 [0.5] vs 5.7 [0.5], P < .001), SpO2 (96.9 [0.9] vs 91.4 [0.9], P < .001), leg fatigue scores (3.8 [0.6] vs 5.2 [0.6], P = .008), and breathing frequency (22.8 [0.9] vs 25.8 [0.9] breaths/min, P = .01). There were no differences in heart rate.
CONCLUSIONS: In symptomatic subjects with severe to very severe COPD, VA+O2 significantly increased exercise time and improved dyspnea, SpO2, breathing frequency, and leg fatigue versus O2 alone.
Introduction
Limited exercise tolerance is a hallmark symptom of COPD and has a significant, negative effect on quality of life.1 Improving exercise tolerance and endurance is an important treatment goal. Widely used strategies for improving exercise intolerance in patients with COPD include pharmacotherapies such as long-acting bronchodilators and pulmonary rehabilitation.1 Supplemental O2 has been studied as a way to ameliorate exercise intolerance, but results have been inconsistent.2 Most studies on supplemental O2 report improvements in exercise duration3-5 and dyspnea,3,4 but others show improvements in exercise duration but no effects on dyspnea, health status, or quality of life.6,7 Bell et al8 conducted a systematic review on studies evaluating short-term O2 therapy in subjects with interstitial lung disease, demonstrating no significant effects on dyspnea9,10 but improved peak work capacity9 and exercise endurance time.11,12
A number of assist devices have been designed to address the limitations of standard O2 therapy.13,-,15 Most are bulky, complex to operate, cumbersome to wear, and have had limited adoption due to these shortcomings as well as cost. Newer noninvasive open ventilation (NIOV) systems (eg, Life2000 Ventilation System, Hillrom, Chicago, Illinois) are more portable than previous systems, allowing patients greater mobility. A previous study on subjects with severe COPD demonstrated improved stationary exercise duration using NIOV.16 However, these devices can be expensive and can function only with O2 cylinders that are heavy, bulky, and detract from portability. Thus, there is an unmet need for portable devices that further increase patient mobility and ability to perform daily activities.
The present study’s intent was to evaluate the clinical benefits of ventilatory assist (VA) + supplemental O2 (VA+O2) in comparison to standard low-flow nasal O2, investigating the effect on exercise endurance, dyspnea, oxygenation, and leg fatigue in subjects with severe to very severe COPD.
QUICK LOOK
Current knowledge
Severe COPD is associated with hypoxemia and reduced exercise tolerance, which impact patients’ quality of life. Despite advances in the modern modalities of O2 therapy, the limitations in the ability to augment activities of daily living present a significant unmet need, which might be addressed by combining O2 therapy with ventilation. This warrants the use of delivery systems that are lightweight and noninvasive, providing patients with greater autonomy.
What this paper contributes to our knowledge
This study used a submaximal physical stress model on a stationary device, and a crossover design in subjects with severe to very severe COPD. When compared to continuous O2 therapy alone, ventilatory assistance combined with O2 significantly improved mean exercise endurance, SpO2, dyspnea, and leg fatigue. The magnitude of the observed effect was nearly a doubling, but there was variability between individual subjects. This exploratory study supports further investigation of the clinical benefits and utility of ventilatory assist + O2 therapy.
Methods
This was an open-label, crossover, clinical feasibility study (ClinicalTrials.gov registration, NCT02278107). The study was conducted at the John Muir Health Center Outpatient Center pulmonary rehabilitation facility in Concord, California. Subjects were recruited and alternatively assigned to the crossover sequence according to the protocol approved by the John Muir Health Concord Center’s Institutional Review Committee (protocol number P-JM-001-2014) and in accordance with good clinical practice guidelines.
VA+O2 was delivered using a VA device (Tidal Assist Ventilator, Inogen, Goleta, California). This NIOV assist device weighs 3.8 ounces (about 108 g) and was designed to be powered by compressed O2 delivered by a cylinder, wall source, or modified higher-pressure stationary O2 concentrator. The device was designed to be worn on the patient’s body and is breath activated. When activated, a bolus of delivered O2 (1–5 L/min O2) entrains additional room air at the nasal interface, delivering it under positive pressure during the patient’s inhalation to assist breathing. The VA device has a control to adjust the flow of synchronized positive inspiratory pressure to one of 5 levels to meet the patient’s perceived need and activity level. O2 alone was delivered via a standard nasal cannula attached to either an E-size O2 cylinder or an O2 concentrator.
Clinically stable volunteer subjects (males and females age ≥ 18 y) with a diagnosis of severe to very severe COPD who required between 2–5 L/min of constant-flow O2 to maintain SpO2 > 90% during exercise were included in the study. Subjects had to be able to use the VA device and cycle ergometer and were willing and able to provide written, informed consent. Patients were excluded if they reported serious epistaxis ≤ 14 d of study start, were currently enrolled or had participated in another clinical trial ≤ 30 d of study start, or showed any condition or abnormality that in the opinion of the investigators may have compromised the participant’s safety or the quality of the study data.
Subjects who met all inclusion criteria were consented and allocated to a treatment sequence. The allocation process was based on a coin flip to determine the treatment order for participant number 1, after which every other subject started with the alternate therapy (1:1 ratio VA+O2 or O2). Neither the subjects nor the investigators were blinded to the therapy administered. No sham controls were utilized. Subjects were provided with scripted encouragement by the investigators.
The study design consisted of 2 visits and used a protocol modified from Porszasz et al.16 During visit 1, baseline physiologic measures, including vital signs and ratings on the Borg dyspnea17,18 and leg fatigue scales, were obtained. Subjects received training on the cycle ergometer (SO1000R Recumbent Bike, SCIFIT, Tulsa, Oklahoma) and the VA device. Peak work rate O2 flow was established by incrementing the subject’s work rate on the cycle ergometer in the range of 2–10 W and titrating the constant flow O2 level to maintain an average SpO2 level of 90–93%, as possible. The flow of the VA device could be adjusted by the subjects to their level of comfort while maintaining SpO2 > 90%; then the subjects exercised at a constant work rate of 80% of the peak work rate.
A week later, subjects returned for visit 2, when they were to first use the VA+O2 or O2 according to their initial allocation and at the respective flows as determined during the baseline study, while exercising for as long as they could at a constant work rate of 80% peak work rate. After a 1-h rest period, subjects switched (crossover phase) to the other mode of treatment and repeated the exercise protocol. No data were gathered during the recovery phase.
Cycling exercise time was determined by stopwatch from the start of cycling at 80% peak work rate until the subject could no longer sustain that work load. Additional measures collected were the Borg scale ratings of dyspnea,17,18 leg fatigue ratings, SpO2 (forehead and finger pulse oximeters, LNCS TF-I adult forehead sensor, Masimo, Irving, California), heart rate (pulse oximeters, RD SET Adt, Masimo), and breathing frequency/min (counted by the investigator). Except for endurance time, end point comparisons were taken at the point of exercise termination for the shorter constant work rate test (VA+O2 or O2) for each subject (referred to as isotime). In addition, heart rate and SpO2 were monitored continuously, with values recorded every 30 s. Dyspnea and discomfort scores were collected every 60 s independently of the isotime measurements. Only the isotime data points were used as dependent measures. Safety and tolerability were assessed by recording all adverse events throughout the study. The study was conducted between October 2014–February 2016.
Statistical Analysis
Continuous data were summarized by means and SD, and categorical data were summarized by counts and percentages. The effect of the intervention on each of the end points was assessed using a linear mixed model fit with restricted maximum likelihood and including treatment, period, and sequence as fixed effects and a random intercept per subject. For all analyses a 2-sided 0.05 level of significance was used. The analyses were conducted using SAS Enterprise Guide software, version 8.3 update 3 (SAS Institute, Cary, North Carolina).
Results
Twenty-seven patients (16 males/11 females) with moderate to very severe COPD were screened and completed visit 1; 15 subjects (mean ± SD, age 67.9 ± 9.0 y) with COPD were enrolled. Twelve subjects were screen failures (Fig. 1). Baseline demographic and pulmonary function characteristics of subjects who completed the study are shown in Table 1. The individual subjects’ baseline VA+O2, O2 flows, and baseline SpO2 with O2 alone and with VA+O2 are presented in the online supplementary materials, Table SM1 (see related supplementary materials at http://www.rcjournal.com).
Baseline Demographics and Pulmonary Function by Sequence
Flow chart. VA = ventiltory assist
There were no significant effects of treatment order or period on exercise endurance, isotime dyspnea scores, SpO2, or leg fatigue scores (online supplementary materials, Table SM2, see related supplementary materials at http://www.rcjournal.com). Exercise endurance times for individual subjects are presented in Figure 2.
Mean exercise times (Table 2) almost doubled during use of VA+O2 versus O2 (least squares [LS] mean [standard error [SE]] 12.0 [2.0] vs 6.2 [2.0], P = .01). However, there was notable variability in individual results (Fig. 2). For treatment effects on blood O2 saturation, the between–treatment-group LS mean (SE) difference in SpO2 was 5.5 (0.9) (P < .001). Breathing frequency was significantly less with VA+O2 versus O2; the between–treatment-group LS mean (SE) difference was −2.9 (1.0) (P = .01). Between–treatment-group LS mean (SE) differences in Borg dyspnea (P < .001) and leg fatigue scores (P = .008) favored VA+O2 over O2. Heart rate values were similar in the 2 conditions (P = .77) (Table 2). VA+O2 was well tolerated, and no adverse events occurred during either treatment period.
Exercise endurance time (min) with VA+O2 vs O2 alone.
Constant Work Rate Exercise Testing
Discussion
In this open-label, crossover study, conducted on clinically stable subjects with severe to very severe COPD enrolled in an out-patient rehabilitation program, VA+O2 improved cycling endurance, dyspnea, SpO2, breathing frequency, and leg fatigue versus O2 alone. For the primary end point, subjects using the VA+O2 increased their cycle endurance by 5.8 min (94%) over O2; and Borg dyspnea scores dropped by 2.1 points, which is considered a clinically meaningful improvement.19 Breathing frequency was significantly lower with VA+O2 versus O2, which is probably related to the higher SpO2 levels observed in the VA+O2 group. Heart rate response was similar during exercise, suggesting that the increased endurance was achieved despite a similar cardiovascular response. In addition, the VA+O2 was well tolerated with no reported adverse events.
These findings are consistent with those previously reported by Porszasz et al16 in a similar population of subjects with severe COPD treated with the similar NIOV device. In that study, there was a 214% increase in cycle endurance time with the NIOV versus standard O2 (17.6 min vs 5.6 min). There also was a significant reduction in dyspnea at isotime but no change in transcutaneous PCO2. SpO2 at isotime was 5.8% higher with the NIOV+O2 (98.5%) compared to standard O2 (92.7%), very close to the 5.5% higher forehead SpO2 observed with the VA+O2 compared to O2. Intercostal, scalene, and diaphragm electromyography (EMG) activities were reduced during NIOV use compared with standard nasal O2, suggesting respiratory muscle unloading. EMG was not measured in the current study, but the significantly lower breathing frequency in the VA+O2 group compared with the standard nasal O2 group is also consistent with reduced respiratory muscle work.
The mechanism for the improved exercise endurance with ambulatory ventilation systems has not been defined. Porszasz et al16 speculated that reduced respiratory muscle activation (as evidenced by the reduced respiratory muscle EMG activity) and enhanced oxygenation were the main mechanisms, with the latter being related to the greater oxygenation efficiency of the NIOV versus standard continuous or pulsed O2 delivery systems. Both the NIOV and VA+O2 devices inject a predetermined volume of O2 timed during inspiration that enriches inspired gas, although the actual FIO2 of the inspiratory gas mixture has not been measured. The substantially greater SpO2 at isotime in the NIOV+O2 group compared to the standard O2 group supports this idea. The injected bolus of O2 during inspiration with the NIOV and VA+O2 devices increases inspiratory flow that could assist inhalation, lowering work of breathing. However, Porszasz et al16 found that when the NIOV injected the same volume of air (NIOV+air) there was less decrease in dyspnea than with NIOV+O2 and no increase in exercise endurance compared to air without NIOV. This suggests that it is the O2 in the bolus more than the augmented gas flow that is primarily responsible for benefit. Furthermore, the novel observation that leg fatigue is reduced with VA+O2 as compared to standard O2 suggests an additional mechanism for prolonged cycle duration, perhaps related to improved O2 delivery.
A concern with these devices is that patients with very severe airway obstruction could develop dynamic hyperinflation during exertion that could delay triggering of a pulsed device, countering its benefit.20 However, the lower breathing frequency (and longer expiratory phase) observed with VA+O2 should mitigate this effect. Better understanding of the mechanisms of benefit as well as potential adverse effects of VA+O2 will require further investigation.
Previous studies have shown favorable effects of noninvasive devices on stationary exercise endurance, including continuous and bi-level positive airway pressure devices, and proportional assist ventilation, all administered via noninvasive masks.13-15 However, these devices are not well tolerated because of the need to use tight-fitting masks, and they are not sufficiently portable to provide benefit during ambulation.
The main limitations of our study are related to the overall small number of subjects, although the study was successful in detecting significant differences in multiple physiologic outcomes. The main advantage of the crossover design is that it permits powerful paired statistical analyses, requiring a much smaller sample size than a randomized parallel design. However, it also raises the concern about potential carry-over effects. In this study, a 1-h rest period before crossover was used to mitigate any carry-over effects. The randomization scheme was applied only to the first subject, with subsequent subjects alternating the order of conditions. However, there were no differences in results attributable to treatment order (see online supplementary materials, Table SM2). Although not a true randomization procedure, this assured control for the order of conditions. Furthermore, the investigators and subjects were not blinded to the condition, so possible bias cannot be excluded.
Conclusions
These findings support the idea that VA+O2 therapy could offer an alternative to O2 therapy alone that could address the unmet need for a portable VA device that improves oxygenation and enhances exercise endurance while decreasing dyspnea and leg fatigue. It is important to bear in mind that the current results are acute, short-term physiologic outcomes obtained using cycle ergometry and that O2 supply necessitated an E-tank or modified stationary concentrator. Thus, the results cannot be extrapolated to performance of activities of daily living during the long term. Future larger and longer-term studies of VA+O2 using more portable O2 sources are needed to better understand the mechanisms of benefit, patient characteristics that predict a favorable response, patient acceptance and tolerability, and effects on activity levels in everyday settings.
Acknowledgments
We would like to thank Bonnie Elbaum for editorial assistance in preparation of the manuscript.
Footnotes
- Correspondence: Nicholas Hill MD, Tufts Medical Center, 800 Washington Street, #257, Boston, MA 02111. E-mail: nhill{at}tuftsmedicalcenter.org
Ms Hilling discloses a relationship with New Aera. Dr Ameo discloses relationships with New Aera and Inogen. Mr Morishige discloses a relationship with New Aera. Dr Glezer discloses a relationship with Inogen. Dr Hill discloses relationships with New Aera, Inogen, Breas, Caire, Fisher & Paykel, Liberate Medical, Philips Respironics, and Vyaire. The remaining authors have disclosed no conflicts of interest.
The study was funded by New Aera, Inogen provided subsequent financial support for data analysis and manuscript preparation.
Supplementary material related to this paper is available at http://www.rcjournal.com.
See the Related Editorial on Page 627
- Copyright © 2024 by Daedalus Enterprises
