| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Torrance, CA
Dr. Sue is affiliated with the Division of Respiratory and Critical Care Physiology and Medicine, Department of Medicine, Harbor-UCLA Medical Center, and is Professor of Clinical Medicine, David Geffen School of Medicine, University of California at Los Angeles.
Correspondence to: Darryl Y. Sue, MD, FCCP, Department of Medicine, Box 400, Harbor-UCLA Medical Center, 1000 W Carson St, Torrance, CA 90509-2910; e-mail: dsue{at}ucla.edu
Our understanding of the mechanisms of diseases that result in impaired functional capacity, and impair an individual’s ability to work, recreate, and perform activities of daily living, has evolved in a remarkable and surprising manner. In the past, we knew that a patient with heart failure had dyspnea and fatigue because the heart pumped insufficient blood and the lungs became congested. Similarly, we knew that lung disease limited patients because their ability to increase ventilation was inadequate to meet the increased metabolic demands of work. Of course, newer explanations have challenged many of these ideas, largely because reduced function of the primarily involved organ frequently fails to account for the magnitude of reduction in overall capacity (eg, decreased maximum oxygen uptake [
O2]). Decreased left ventricular ejection fraction, cardiac output, and stroke volume in patients with chronic heart failure, for example, do not entirely explain reduced exercise capacity and poorer prognosis. Accordingly, other contributing factors have been sought and identified, including abnormalities of the peripheral circulation, muscles of ambulation, lungs, and the pulmonary circulation.
In patients with COPD, the severity of airway obstruction must play a major role in determining impairment. Reduced FEV1, however, is poorly predictive of the reduction in exercise capacity,1 2 3 indicating that other factors must play a role. Candidates include abnormalities of the pulmonary circulation, impaired lung gas exchange, decreased performance of respiratory muscles, differences in the degree of hyperinflation, abnormal left and right ventricular performance, familial and other differences in ventilatory response, reduced oxygen-carrying capacity of the blood, and dysfunction of skeletal muscles other than the respiratory muscles (ie, the peripheral muscles).
The significance of peripheral skeletal muscle disease as a limiting factor in patients with COPD has been emphasized.1 4 5 6 In fact, some investigators have concluded that muscle dysfunction is seen in a considerable proportion of COPD patients and have suggested inactivity, acidosis, hypoxemia, chronic inflammation, malnutrition, coexisting heart disease, severe deconditioning, and medications (especially corticosteroids) as some of the proposed mechanisms. Controversy remains as to whether or not COPD is associated with a specific myopathic condition or whether peripheral muscle disorders are secondary to the consequences of COPD, resulting in malnutrition, chronic inflammation, or disuse. How often and how much peripheral muscle disorders affect function in COPD is a very important question. This is because we want to know whether the treatment of COPD should be directed only at the lungs, using bronchodilators, corticosteroids, oxygen, and smoking cessation, or whether therapy with exercise training, nutritional interventions, and anabolic agents would be of benefit. Furthermore, some investigators7 have found an association between reduced muscle mass and survival in COPD patients, independent of a reduction of FEV1.
Two studies described in this issue of CHEST (see page 83 and page 75) focus on skeletal muscle abnormalities in patients with COPD. One study, by Debigaré et al, begins with the assumption that muscle wasting is a not uncommon finding in COPD patients and seeks correlative data in an attempt to infer its cause. The other study, by Heijdra and colleagues, evaluated a group of outpatients with COPD to determine how often and to what degree evidence of peripheral muscle dysfunction could be identified. Interestingly, these different approaches yielded results that seem contradictory. Yet, it is likely that the findings might lead us to distinguish subsets of patients who may benefit from more tailored therapy.
Debigaré and colleagues reported data collected from 45 men with COPD in stable condition who had no evidence of current "inflammatory disease" and had participated in a prior study of muscle mass and survival. To estimate the degree of peripheral skeletal muscle abnormality, they looked at mid-thigh muscle cross-sectional area (CSA) by CT scan, choosing a cutoff value for abnormal based on the findings of a previous study of clinical outcome7
in patients with COPD. Because the 45 men were selected from a population of those with a known proportion of abnormally low muscle CSA, and were not chosen at random or sequentially, it is not possible to extrapolate the prevalence of low mid-thigh muscle CSA to the COPD population in general. In this study, 18 of 45 patients had a CSA of < 70 cm2, and this subset was compared to 27 COPD patients with CSA of 
BORDER="0"> 70 cm2 and to 16 healthy nonsmoking men of similar age. These investigators looked primarily at markers of catabolism (ie, interleukin [IL]-6 and cortisol) and anabolism (ie, bioavailable testosterone, dehydroepiandrosterone sulfate [DHEAS], and insulin-like growth factor-1), hypothesizing that peripheral muscle loss would be associated with a shift toward patterns of greater catabolism and lesser anabolism. COPD patients with a mid-thigh CSA of < 70 cm2 had significantly higher ratios of cortisol/DHEAS and IL-6/DHEAS compared to those with mid-thigh CSA of 70 cm2, as well as higher ratios compared to control subjects. While not drawing a conclusion about causality, the authors suggested that these data indicated a shift toward catabolism that may contribute to the development of decreased peripheral skeletal muscle mass. One point that deserves emphasis, however, is that loss of muscle mass may or may not be associated with an ongoing loss of muscle mass. For example, if the inflammatory or catabolic mechanism is interrupted after muscle mass is lost, the correlation between the two may be weakened. Similarly, some patients simply have always had low muscle mass rather than having lost muscle mass secondary to COPD.
Heijdra and colleagues asked how often evidence for a peripheral muscle disorder could be identified in a set of stable COPD patients attending an outpatient clinic. They surmised that peripheral muscle dysfunction might be important in those with malnutrition, a recognized cocondition in COPD patients, but they suggested that evidence of a "systemic myopathy" might be uncommon in this population of COPD patients. That is, Heijdra et al hypothesized that stable COPD patients with normal fat-free mass (FFM) would not have evidence of clinically significant peripheral muscle dysfunction. In COPD patients and control subjects, they compared anthropomorphic evaluation and skeletal muscle strength (ie, maximal inspiratory and expiratory pressures for respiratory muscles and handgrip for peripheral muscles). Because the slow kinetics of O2 has been attributed to factors intrinsic to the peripheral muscles rather than to impaired oxygen delivery8
9
10
to the muscles, they also looked at the time constant for O2 during low-level, constant-work exercise. Their subjects were 32 stable patients with severe COPD (mean [± SD] FEV1, 0.97 ± 0.29 L) who had been recruited from an outpatient clinic and 36 control subjects of similar age. Importantly, they estimated FFM by bioelectrical impedance, and all 36 control subjects and 31 of 32 COPD patients had a FFM index (FFMI) at or above normal (defined as > 16 kg/m2 for men and > 15 kg/m2 for women). These investigators found that the mean handgrip force, the maximal expiratory pressure, and the time constant for O2 (COPD patients, 72 ± 34 s; control subjects, 78 ± 37 s) were not different between COPD patients and control subjects. They concluded that COPD patients who had been identified from a clinic population, had been stable for at least 6 months, and had a normal FFMI showed no evidence of peripheral muscle dysfunction in the group as a whole. Because the multiple regression analysis of Heijdra and colleagues showed that FFMI accounted for 42% of the variance in peak O2, they suggested that those patients with decreased muscle mass could be identified and perhaps targeted for specific therapy.
Therefore, Debigaré and colleagues demonstrated that COPD patients with decreased mid-thigh muscle mass had findings in the blood suggesting a catabolic state, while Heijdra et al found that their COPD patients had little evidence of a clinically significant peripheral muscle dysfunction. Are these results in conflict? If so, what might be the reasons for these differences?
The most evident differences are in the study populations. The patients in the study by Debigaré et al were chosen to compare a subset of patients with decreased mid-thigh CSA to a group of COPD patients with a CSA > 70 cm2 (ie, below normal but previously associated with better prognosis). On the other hand, almost all of the patients recruited by Heijdra et al had a normal FFMI value. In a study by Bernard et al,11
skeletal muscle weakness was related to peripheral muscle atrophy as estimated by thigh muscle CSA. Although quadriceps strength was reduced in COPD patients compared to healthy subjects, the ratio of quadriceps strength to muscle CSA was not different. Therefore, reduced muscle mass correlated with decreased muscle strength. Similarly, Engelen et al12
used dual-energy radiograph absorptiometry to estimate the whole-body and upper extremity FFM and compared these values to handgrip strength in COPD and control subjects. Reduced handgrip strength was seen in COPD patients (in both those patients with chronic bronchitis and those with emphysema), but the ratio of handgrip strength to FFM was not different from that of control subjects, and patients’ whole-body FFM strongly correlated with extremity FFM. Again, these data suggest that reduced strength (ie, handgrip strength) is seen if and only if there is reduced FFM, due presumably to reduced muscle mass. Thus, one explanation of the difference between the two studies being discussed is that the patients in the study of Heijdra et al had normal or near-normal FFM, meaning that they had normal upper extremity muscle mass and subsequently normal handgrip strength, and normal lower extremity muscle mass, resulting in no difference in O2 kinetics during steady-state exercise. Perhaps these patients, who had been selected from an outpatient clinic but had not been enrolled in a pulmonary rehabilitation program, were better nourished or less deconditioned than were other COPD patients.
A second way of comparing the two studies also helps to reconcile the results. Whole-muscle atrophy might be assumed to be due to the atrophy of individual muscle fibers (ie, if the number of muscle fibers remains constant). Gosker et al13 showed a strong correlation between FFM determined by bioelectrical impedance and mean fiber CSA (determined from a biopsy specimen of the lateral part of the quadriceps femoris) in COPD patients. Extrapolating from data of Gosker et al, the mean FFM of the COPD patients presented by Heijdra et al would translate into a mean muscle fiber CSA that would be well into the normal range. Similarly, the data supplied by Debigaré and colleagues showed that the mean mid-thigh CSA for their group of COPD patients with a CSA of < 70 cm2 was 60% of the values of control subjects (58 vs 96 cm2, respectively). Thus, the patients in the study by Debigaré et al matched well with the COPD patients presented by Gosker et al, who had a mean fiber CSA that was 62% of that of the control subjects (3,839 vs 4,647 µm2, respectively). Again, this analysis supports the idea that different subsets of COPD patients are being contrasted in the two studies being compared.
Finally, the study by Schols and colleagues14 sheds some additional light on potential differences. They reported data from 30 patients with COPD and 26 healthy age-matched control subjects. Eight of 30 COPD patients had elevated C-reactive protein (CRP) levels, and in those 8 patients the levels of IL-8 and two soluble tumor necrosis factor receptors were increased. In addition, the patients with elevated CRP levels had lower FFM, as determined by bioelectrical impedance. Thus, patients having markers of inflammation or acute-phase reactants were those with lower FFM. The patients in the study by Heijdra et al had normal FFM, so they would not have been expected to have these markers associated with inflammation. The patients in the study by Debigaré et al with low mid-thigh CSA, on the other hand, might be similar to the subset of patients in the study by Schols et al who had low FFM and elevated CRP levels.
What should we conclude about the role of peripheral muscle dysfunction in COPD patients? The therapy for COPD is quite limited, with a focus on bronchodilators, corticosteroids, oxygen, and smoking cessation. But, if peripheral muscle dysfunction is important, then therapy with, for instance, exercise training, nutritional intervention, and anabolic hormones may be beneficial. For example, exercise training during a rehabilitation program has been shown to lessen ventilatory requirements and increase exercise performance independent of the degree of airflow obstruction,15 and pulmonary rehabilitation is effective in reducing symptoms.16 Importantly, it appears that much of the improvement with exercise training is due to its effect on peripheral skeletal muscle function.8 9 Could measures of peripheral muscle size or function help to determine which COPD patients would profit additionally from therapies such as anabolic hormones, growth factors, or nutritional intervention?
At a minimum, we need to understand better the frequency, severity, and mechanisms of peripheral muscle dysfunction in COPD patients. If not all patients have clinically evident peripheral muscle dysfunction, then investigators studying COPD therapy must describe their patients in sufficient detail to allow us to understand the state of the peripheral muscles, if relevant. In particular, studies of exercise training and pharmacologic intervention that are directed at the peripheral muscles, whether positive or negative, need to describe whether the patients studied had normal or reduced muscle mass and need to provide other information about the baseline function and subsequent function of the skeletal muscles.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
