Breathing exercises for patients with COPD: to teach or not to teach.

To the Editor:

The article by Jones et al titled "Comparison of the Oxygen Cost of Breathing Exercises and Spontaneous Breathing in Patients With Stable Chronic Obstructive Pulmonary Disease" in the May 2003 issue adds valuable evidence to aid our decision on when to teach (or not to teach) breathing exercises to patients with chronic obstructive pulmonary disease (COPD). Their methods were clear and straightforward, and the patients were representative of the age (mean age = 68.5 years) and severity of COPD in patients seen in many pulmonary rehabilitation programs (mean percentage of predicted forced expiratory volume in 1 second [[FEV.sub.1]] = 39%). However, the article did raise some concerns.

First, much of the introduction discusses the topics of work of breathing (WOB) and ventilatory muscle efficiency and how these 2 Factors may be affected in patients with COPD; however, WOB and ventilatory muscle efficiency are not adequately defined in relation to each other. The authors correctly define WOB, but not the rate of WOB (WOB), which is arguably more important; WOB is the product of WOB (for single breath) and respiratory rate (RR) (WOB = P*VT*RR, where P = inspiratory pressure and VT = tidal volume). Ventilatory muscle efficiency, or mechanical efficiency of breathing, was not defined, but is directly related to WOB by the following equation: ventilatory muscle efficiency = WOB/[VO.sub.2]*100%, where [VO.sub.2] is oxygen consumption. (1,2) Therefore, interventions such as pursed lip breathing (PLB) and diaphragmatic breathing (DB) that decrease RR also will decrease WOB, and if ventilatory muscle efficiency does not change, then [VO.sub.2] will decrease, which is what Jones and colleagues' results indicate.

Second, given the wide range of body mass index (BMI) values (13.7-28.1 kg/[m.sup.2]; anorexic to overweight), [VO.sub.2] values would be more conventionally and clearly expressed as mL/min/kg, rather than mL/min. (3,4) Using data from 3 hypothetical patients with similar means and standard deviations for [VO.sub.2] (mL/min) as well as mean BMI values (see Table), it is apparent that when body mass is accounted for in presenting [VO.sub.2] results, the difference can be dramatic: subject 1's [VO.sub.2] in mL/min is 25% lower than that of subject 3, but is 53% higher when expressed as mL/min/kg. Because [VO.sub.2] results are not expressed relative to body mass, clear conclusions regarding the relevance of [VO.sub.2] changes cannot be made. (Also, in Tab. 1 in the article by Jones et al, BMI is improperly defined as kg*[m.sup.2], rather than kg/[m.sup.2]).

Last, my main concern is Jones and colleagues' questioning of the effectiveness of PLB and their suggestion that it is mechanically inefficient, despite the fact that RR decreased over 25% and [VO.sub.2] (mL/min) decreased by up to 5.5% using PLB. I would argue that, under these conditions (supine, at rest), with these patients (relatively underweight), the clinical significance of the change in [VO.sub.2] is irrelevant (maybe even statistically insignificant if body mass is accounted for). To start with, their basal [VO.sub.2] (2.91 mL/min/kg, assuming mean body weight of 60 kg [132 lb], mean height of 1.75 m [5 ft 9 in], and mean BMI of 19.6) is lower, not higher, (5) than the expected norm (3.5 mL/min/kg), which is surprising given their low [FEV.sub.1]. In addition, if [VO.sub.2] (mL/min) is converted to kilocalories expended per day (kcal/d), then the difference between spontaneous breathing and PLB is only 69 (1,256 versus 1,187 kcal/d, respectively). Thus, I believe their subjects did not use PLB at rest, because there was little energetic rationale for them to do so, not because, as they hypothesized, it was mechanically inefficient.

We must keep in mind that these measurements were made at rest, whereas many physical therapy interventions occur with exercise. I have seen, in agreement with others, (6,7) patients with COPD, new to pulmonary rehabilitation, who have not been taught PLB, but who do use it spontaneously; their pulse oximetry values and dyspnea almost universally improve. (6) Therefore, I would argue that PLB, while not energetically more favorable at rest, is an important intervention for patients with COPD during exercise or during dyspneic episodes, when it may be more energetically favorable.

I congratulate the authors for helping us understand the effects of different breathing patterns on the oxygen cost of breathing in patients with COPD. Their results help clarify the circumstances under which we should or should not teach our patients to use PLB. I hope that future research will investigate the effect of various breathing patterns on [VO.sub.2] in different positions and during exercise, as well as test interventions to alleviate dynamic hyperinflation.

John D Lowman, Jr, PT, MS, CCS

Graduate Student

Department of Physiology

School of Medicine

Medical College of Virginia Campus

Virginia Commonwealth University

Box 980551

Richmond, VA 23298-0551

Part-time Staff Physical Therapist

Department of Physical and Occupational

Therapy

Duke University Hospital

Durham, NC

Table.
Data for 3 Hypothetical (a)

Subject   Mass   Height   BMI              [Vo.sub.2]   [Vo.sub.2]
No.       (kg)   (m)      (kg/[m.sup.2])   (mL/min)     (mL/min/kg)

1         47     1.85     13.7             150.0        3.19
2         58     1.85     16.9             174.5        3.01
3         96     1.85     28.0             200.0        2.08
X                         19.6             174.8        2.80
SD                         7.5              25.0        0.60

(a) BMI = body mass index, [Vo.sub.2] = oxygen consumption.

References

(1) West JB. Respiratory Physiology: The Essentials. 6th ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2000.

(2) Otis AB. The work of breathing. In: Rahn H, ed. Respiration. Vol 1. Washington, DC: American Physiological Society; 1964:463-476.

(3) Velloso M, Stella SG, Cendon S, et al. Metabolic and ventilatory parameters of four activities of daily living accomplished with arms in COPD patients. Chest. 2003;123:1047-1053.

(4) Arzt M, Harth M, Luchner A, et al. Enhanced ventilatory response to exercise in patients with chronic heart failure and central sleep apnea. Circulation. 2003;107:1998-2003.

(5) Donahoe M, Rogers RM, Wilson DO, Pennock BE. Oxygen consumption of the respiratory muscles in normal and in malnourished patients with chronic obstructive pulmonary disease. Am Rev Respir Dis. 1989;140:385-391.

(6) Tiep BL, Burns M, Kao D, et al. Pursed lips breathing training using ear oximetry. Chest. 1986;90:218-221.

(7) Mueller RE, Petty TL, Filley GF. Ventilation and arterial blood gas changes induced by pursed lips breathing. J Appl Physiol. 1970;28:784-789.

Author Response:

We appreciate the opportunity to respond to Mr Lowman's comments on our article, and we will respond to his 2 main points. We are unable to address his first point--that we "correctly define WOB [work of breathing], but not the rate of WOB (WOB)"--because this latter term actually was not used in our article. Nonetheless, because Mr Lowman has raised the issue of power, we shall take this opportunity to elaborate on the fact that we chose to frame the work in a clinical context, and thus defaulted to commonly used concepts and nonmenclature. To date, the concept of the WOB, rather than the WOB or power of breathing, tends to be the focus clinically and in research. For example, in a recent state-of-the-art review on respiratory muscles, (1) power was not mentioned. The fact that power is not the clinical measure most used in the assessment of both peripheral muscles and respiratory muscles at this time probably reflects the fact that such a measure is not yet readily accessible, and that it does not have documented validity and reliability.

With respect to the first of his 2 main points, Mr Lowman attempted to reconstruct our data; however, his theoretical data do not approximate the actual data, which precludes much meaningful discussion. The mean value for oxygen consumption ([VO.sub.2]) in our study was 3.55 mL [O.sub.2]/kg/min rather than the 2.9 as calculated theoretically by Mr Lowman. As a point of clarification, the value of 3.5 mL [O.sub.2]/kg/min (resting oxygen consumption in individuals with no cardiovascular/ cardiopulmonary impairment) is comparable to other physiologic measurements in that it is associated with some inherent variability and potential systematic measurement differences even with adherence to strict calibration procedures. In our study, therefore, the pretesting conditions were strictly standardized across subjects, and the measurements were obtained using the same equipment, calibrated in the same way, by the same individuals, in the same place, under the same ambient conditions (see the "Method" section of our article).

Mr Lowman's second main point queried the clinical significance of our findings, given that we studied subjects m a supine position and that the body mass index of our subjects ranged from relatively overweight to underweight. First, we refer him to the "Method" section of the article in which we explain that the subjects were studied in a supine position purposefully so that we could use the least intrusive [VO.sub.2] measurement method. This was important because, to examine oxygen cost with out introducing ventilatory measurement artifact in this patient population, we needed to use a canopy that requires the patient to lie quietly in a supine position (and not in a sitting position, let alone exercising) so the canopy could be positioned over the head, secured for possible air leaks, and safely supported by a plinth.

Alternative methods alter breathing. For example, a mouthpiece and nose-clip alter dead Space and breathing pattern and may not be tolerated by patients with impaired pulmonary function. In addition, a face mask needs to fit snugly. This can also alter normal breathing and can contribute to claustrophobia in patients who are prone to dyspnea and distort the breathing pattern further. Both alternatives are likely to alter energy demands.

We anticipate that other investigators will help extend our work and that of others, with the development and use of artifact-free technology, in the study of breathing patterns adopted spontaneously by patients experiencing dyspnea and in the study of breathing energetics. With respect to body mass, the patients we studied represented a range of body masses, which was desirable experimentally to enhance the generalizabilily of the results to the population of patients with chronic obstructive pulmonary disease (COPD) in the real world. To control the contribution of intersubject variation, we chose to use a robust, within-subject experimental design with a judicious sample size (see the "Data Analysis" section of our article) to avoid the between-group differences that may have occurred had we used a between-subject design, where age, body weight, and inherent variability of resting [VO.sub.2] could have introduced marked systematic sampling error.

We believe Mr Lowman's observation of and purported concurrence with the literature (2,3) regarding improvement in status with breathing exercises in "patients with COPD ... who have not been taught PLB [pursed lip breathing], but who do use it spontaneously" is somewhat overstated. Although we cannot comment on Mr Lowman's personal experience, we are familiar with the work of Tiep and colleagues (2) and Mueller and colleagues, (3) whom he cites. Tiep and coworkers stated, "None of the subjects had experienced pursed lip breathing training within a year of the study, and they were neither naturally nor by training pursed-lips breathers." (2)

With respect to their subject selection, Mueller and colleagues stated that their subjects "had to have received at some time in the past adequate instruction in PLB." (3) That their subjects used PLB spontaneously cannot be presumed, because this was not stated explicitly. Whether patients spontaneously adopt PLB, and under what conditions and to what degree, should neither be overstated nor understated, because this is the crux of the matter in understanding the normal determinants of breathing pattern in patients with chronic lung conditions.

We do not dispute the literature or Mr Lowman's observation that the status of some patients improves when they adopt PLB. In fact, we would proffer that this observation is what gave physical therapists the idea of using PLB as all intervention in the first place. However, we would argue that physical therapists have taken the observation of a naturally occurring phenomenon in some patients and applied it therapeutically, believing that the abnormal breathing pattern in patients with COPD is the cause of the ventilatory distress in their patients and that changing Ibis pattern is the solution. This notion is dispelled by the literature reported in our article supporting variable effects of breathing exercises, including detrimental effects. Therefore, we would argue that to prescribe the most efficacious intervention, we need to understand when and why some patients adopt this breathing pattern spontaneously and to explain this phenomenon based on an analysis of respiratory mechanics and energetics and subjective perception of dyspnea. By understanding the underlying mechanism, we can direct interventions specifically to the underlying cause of ventilatory distress.

It was a pleasure to be able to share our work and engage in intellectual discussion regarding this important issue of breathing exercises, which have been a mainstay of time-honored--rather than evidence-based--conventional chest physical therapist practice. To conclude, we would like to emphasize the original focus of our study and its clinical relevance. Our data lend support for viewing mechanical efficiency and energetic efficiency of respiration as clinically distinct. Given that imposed breathing patterns associated with reduced oxygen cost are not adopted routinely or sustained when taught to patients with COPD who are prone to ventilatory distress, such intervention is counterintuitive considering the established indications for breathing exercises. This topic, therefore, is of considerable clinical interest and importance. Viewing abnormal breathing patterns in patients with COPD as the effect rather than the cause of their problem is a novel approach. This shift in orientation should help elucidate physical therapy interventions that can be targeted to the cause rather than the effect, with the promise of superior outcomes.

Detailed studies of spontaneous breathing patterns in patients with COPD at rest and during varying intensities of exercise are warranted to elucidate the mechanical impairments of ventilation and their energy consequences. These studies then could provide a cogent rationale, if any, for breathing exercises or provide support for evidence-based alternatives. If sufficient evidence for breathing exercises such as PLB is demonstrated, we will need to prescribe this intervention in a targeted manner, that is, for particular patients and for particular conditions. In light of contemporary evidence related to a primary focus on oxygen transport overall in patients with COPD rather than a narrow focus on airways and lungs, the relevant question is: "What effect do breathing exercises have on lung function and dyspnea over and above that resulting from mobilization/exercise and body positioning?"

Last, we concur that the development of clinical tools that measure the power as well as the work of breathing would be a major advance in physical therapy.

Letters to the Editor should relate specifically to material published in the Journal or to research/clinical issues of relevance to the physical therapy profession.

To be considered for publication, letters commenting on published articles must be received within 8 weeks of publication of the article. Letters should be submitted via e-mail to [e-mail] or mailed to the Editorial Office in hard copy (double-spaced). Receipt of Letters to the Editor is not acknowledged; however, correspondents will be notified if the letter has been accepted for publication.

Authors of the article in question will be invited to respond to the letter. Accepted Letters to the Editor will be printed with the author response whenever possible.

The Journal reserves the right to copyedit Letters to the Editor. Unless extensive editing is required, correspondents will not be sent a copy of the edited version to review.

Alice YM Jones, PT, PhD, FACP

Associate Professor

Department of Rehabilitation Science

The Hong Kong Polytechnic University

Hung Hom, Kowloon, Hong Kong

Elizabeth Dean, PT, PhD

Professor and Coordinator of the

Advanced Graduate Programs

School of Rehabilitation Sciences

T-325, 2211 Wesbrook Mall

Vancouver, British Columbia, Canada

V6T 2B5

Cedric CS Chow, PT, MSc

Physical Therapist

Caritas Medical Centre, Kowloon, Hong

Kong

References

(1) Laghi F, Tobin MJ. Disorders of respiratory muscles. Am J Crit Care Med. 2003;168:10-48.

(2) Tiep BL, Burns M, Kao D, et al. Pursed lip breathing training using ear oximetry. Chest. 1986;90:218-221.

(3) Mueller RE, Petty TL, Filley GF. Ventilation and arterial blood gas changes induced by pursed lips breathing. J Appl Physiol. 1970;28:784-789.