Published ahead of print on November 29, 2007, doi:10.1164/rccm.200707-1011OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200707-1011OC
Extrapulmonary Effects of Chronic Obstructive Pulmonary Disease on Physical ActivityA Cross-sectional Study1 Pulmonary Research Institute, 2 Center for Pneumology and Thoracic Surgery, Hospital Grosshansdorf, Grosshansdorf, Germany; and 3 Institute of Social Medicine, Medical University Luebeck, Luebeck, Germany Correspondence and requests for reprints should be addressed to Prof. Dr. Helgo Magnussen, M.D., Pulmonary Research Institute at Hospital Grosshansdorf, Center for Pneumology and Thoracic Surgery, Woehrendamm 80, D-22927 Grosshansdorf, Germany. E-mail: magnussen{at}pulmoresearch.de
Rationale: Physical activity is reduced in patients with chronic obstructive pulmonary disease (COPD). COPD has a systemic component that includes significant extrapulmonary effects that may contribute to its severity in individual patients. Objectives: To investigate the association of extrapulmonary effects of the disease and its comorbidities with reduced physical activity in patients with COPD. Methods: In a cross-sectional study, 170 outpatients with COPD (GOLD [Global Initiative for Chronic Obstructive Lung Disease] stages I–IV; BODE [body mass index, airway obstruction, dyspnea, and exercise capacity] score 0–10) underwent a series of tests. Physical activity was assessed over 5 to 6 consecutive days by using a multisensor accelerometer armband that records steps per day and the physical activity level (total daily energy expenditure divided by whole-night sleeping energy expenditure). Cardiovascular status was assessed by echocardiography, vascular Doppler sonography, and levels of N-terminal pro–B-type natriuretic peptide. Mental status, metabolic/muscular status, systemic inflammation, and anemia were assessed by Beck Depression Inventory, bioelectrical impedance analysis, handgrip strength, high-sensitivity C-reactive protein/fibrinogen, and hemoglobin, respectively. Measurements and Main Results: In a multivariate linear regression analysis using either steps per day or physical activity level as a dependent variable, the extrapulmonary parameters that were associated with reduced physical activity in patients with COPD independently of GOLD stages or BODE score were N-terminal pro–B-type natriuretic peptide levels, echocardiographically measured left ventricular diastolic function, and systemic inflammation. Conclusions: Higher values of systemic inflammation and left cardiac dysfunction are associated with reduced physical activity in patients with COPD.
Key Words: pulmonary disease, chronic obstructive • ventricular function, left • acute phase reaction • activities of daily living
Physical activity is a parameter of increasing clinical interest. Lower levels of physical activity are associated with a higher risk for diseases such as cardiovascular disease (1), type 2 diabetes mellitus (2), mental disorders (3, 4), cancer (5), and chronic obstructive pulmonary disease (COPD) (6). In COPD, regular physical activity modifies smoking-related lung function decline and is associated with lower mortality (6, 7). COPD has been defined as a preventable and treatable disease with significant extrapulmonary effects that may contribute to its severity in individual patients (8). Cardiovascular, mental, and musculoskeletal comorbidities, as well as elevated markers of systemic inflammation and anemia, can frequently be found in patients with COPD and have an impact on mortality in this population (9–12). Physical activity has been shown to be reduced in patients with COPD (13, 14). We hypothesized that this reduction in physical activity may be explained by the extrapulmonary effects of COPD and its comorbidities. Thus, the rationale of our study was to assess the association of extrapulmonary effects and comorbidities in patients with COPD with reduced physical activity as measured by an accelerometer. We used multivariate linear regression analysis, the clinical stages of COPD according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) (8), and the BODE (body mass index, airway obstruction, dyspnea, and exercise capacity) index (15) to analyze the data. We previously presented the methodologic aspects of accelerometer measurement in abstract form in this journal (16).
Study Population One hundred seventy outpatients with stable COPD (128 male, 42 female) were studied between February and November 2006 at the Pulmonary Research Institute at Hospital Grosshansdorf, Schleswig-Holstein, Germany. Patients were recruited from the institute's database that is used for clinical trials in COPD; it consists of 691 ambulatory patients with an established diagnosis of COPD (mean FEV1, 54% predicted; mean age, 62 yr). Patients were contacted by a study nurse, and those who expressed their availability and interest in the study were enrolled. Exclusion criteria were an exacerbation of COPD within the past 2 months, clinical signs of acute heart failure, and severe pain syndromes that could interfere with physical activity. The study was approved by the local ethics committee of Schleswig-Holstein, and participants gave their written, informed consent.
Clinical Stages of COPD
Cardiovascular Status and Systemic Inflammation
Mental Status
Metabolic and Muscular Status
Anemia
Physical Activity
Assessment of Potential Confounders of Physical Activity
Statistical Analysis
Multivariate linear regression analyses were performed using either the physical activity level or steps per day as the dependent variable. It was the aim of this analysis to identify extrapulmonary factors that are associated with reduced physical activity independent of the clinical stages of COPD. Therefore, after controlling for possible confounders, a stepwise approach to model building was applied. In a first step, age, sex, obesity (BMI Data analysis was performed with the Statistical Package for Social Science, version 14.0 (SPSS, Chicago, IL).
Characteristics of patients, measurements of physical activity, and extrapulmonary parameters are given in Tables 1 to 3  . A more detailed description of patients' characteristics can be found in the online supplement (Tables E1 and E2).
Twenty-two percent of the patients had a physical activity level of 1.70 or greater (Table 2). Results of the multivariate analysis for the physical activity level are given in Tables 4 and 5: The total explained variance of the physical activity level (total R2) was 37% for the model that included GOLD stages plus extrapulmonary effects and 36% for the model that included BODE score plus extrapulmonary effects. GOLD stages or BODE score accounted for two-thirds of the explained variance of the physical activity level. The extrapulmonary factors that significantly contributed to the model independently of GOLD stage or BODE score were NT–pro-BNP, diastolic left ventricular function, and systemic inflammation. Together, they accounted for another third of the explained variance of the physical activity level. Deceleration time of the early transmitral flow had a stronger effect on the model than did the ratio of the peak velocity of the early E-wave to atrial A-wave (Tables 4 and 5). Fibrinogen had a stronger effect than hs-CRP (Tables 4 and 5). Similar results of the multivariate analysis for steps per day are given in Tables 6 and 7. Figures 1A–1F depict the impact of the extrapulmonary effects that were significant in the multivariate analysis on steps per day according to the clinical stages provided by GOLD and BODE.
Arterial hypertension did not affect echocardiographic parameters of diastolic function of the left ventricle in our cohort (deceleration time of the early transmitral flow, P = 0.72 for hypertension vs. no hypertension; log ratio of the peak velocity of the early E-wave to atrial A-wave, P = 0.6 for hypertension vs. no hypertension). A weak but significant correlation between NT–pro-BNP and measurements of echocardiography was for left atrial size and NT–pro-BNP (r = 0.36, P < 0.001).
The main finding of our study is that higher values of systemic inflammation and left cardiac dysfunction are associated with reduced physical activity in patients with COPD, independent of the clinical stages of COPD according to GOLD or the multidimensional BODE index. Physical activity has previously been shown to be reduced in patients with moderate to severe COPD compared with healthy subjects (13, 14). Airway obstruction, however, correlates only weakly (14) to moderately (13, 29) with physical activity, which implies that there might be additional factors influencing physical activity in patients with COPD. Garcia-Aymerich and coworkers found further determinants of physical activity in a cohort of 346 patients with severe COPD (FEV1, 35% predicted) (28). Using the Minnesota Leisure Time Physical Activity Questionnaire, they found that socioeconomic status, history of diabetes, health-related quality of life, and long-term oxygen therapy were independently associated with a low level of physical activity. Our study extended these observations by objectively quantifying systemic components of COPD (10) and their associations with reduced physical activity. Left heart failure is present in 20% of elderly patients with COPD (30). In about 50% of these patients, left ventricular diastolic dysfunction causes the symptoms of heart failure (30). Even without symptoms of heart failure, diastolic dysfunction of the left ventricle in COPD is a frequently reported echocardiographic observation (31). Here we show for the first time that echocardiographic parameters indicating impaired diastolic filling processes of the left ventricle are associated with reduced physical activity in patients with COPD. Twenty-three percent of our cohort—nearly 30% of the patients with GOLD stage IV disease—had NT–pro-BNP levels greater than 125 pg/ml, which is an accepted threshold that indicates potential heart failure in patients with COPD (17). NT–pro-BNP values showed a weak correlation with left atrial size, which is known in patients with stable chronic left heart failure (32) and may reflect chronic left ventricular filling pressure (33). An increased left ventricular filling pressure leads to an increase in left ventricular wall stretch, which is the predominant pathophysiologic process underlying increased circulating levels of NT–pro-BNP (34). Higher levels of NT–pro-BNP were associated with reduced physical activity in our patients. We did not find a relationship between physical activity and systolic pulmonary artery pressure estimated by Doppler echocardiography. Doppler echocardiography, however, is of limited value in accurately assessing pulmonary hypertension in patients with advanced lung diseases (35). Furthermore, we could only measure systolic pulmonary artery pressure in 72% of our patients (see Table 3 and the online supplement). The ankle brachial index is a noninvasive, reliable measure of peripheral arterial disease (18) and is a strong predictor for cardiovascular morbidity and mortality in the general population (36). Values less than 0.9 indicate peripheral arterial disease (18). In patients with values less than 0.5, physical activity is reduced, whereas mild to moderate peripheral arterial disease (ankle brachial index, 0.5–0.9) is not associated with reduced physical activity (18). On the basis of the values less than 0.9, we observed a frequency of peripheral arterial disease of 25% in our cohort—nearly 40% in patients with GOLD stage IV. Only 3% of our patients had values less than 0.5. The presence of peripheral arterial disease was not associated with reduced physical activity in our cohort. Systemic inflammation is present in patients with COPD and has an impact on mortality in this population (37, 38). Furthermore, higher values of hs-CRP are associated with a reduced exercise capacity in patients with COPD (39, 40). It is known from epidemiologic studies that people who report less physical activity have higher values for markers of systemic inflammation (41). Here we confirm with accelerometer data the association between physical inactivity and systemic inflammation. Depression is a frequent comorbidity in patients with COPD (19). We found depression to be prevalent in nearly 20% of our patients according to the Beck Depression Inventory—42% of patients with GOLD stage IV disease. Our prevalence data are similar to those from a previous study using the same test (19). Depression was not associated with reduced physical activity in our cohort. The frequency of nutritional depletion in our cohort paralleled GOLD stages, with an overall prevalence of 21%. In a previous study, which included patients with GOLD stages II and III COPD, the prevalence of nutritional depletion was 27% (20). Nutritional depletion was not associated with reduced physical activity in our cohort. Peripheral muscle weakness in COPD is considered to be related to loss of fat-free mass (20), muscle atrophy, and deconditioning (42). We found no relationship between physical activity and peripheral muscle strength. This result has to be discussed in the context of existing data concerning peripheral muscle weakness in COPD. Pitta and coworkers showed that both handgrip muscle strength and quadriceps muscle strength correlated moderately with physical activity in a bivariate model (14). We cannot confirm the association of handgrip muscle weakness with reduced physical activity in our multivariate model. In a landmark study investigating the effect of peripheral muscle weakness on exercise limitations, it was isometric quadriceps force, not handgrip force, that remained a significant predictor of exercise limitations in a stepwise multiple regression analysis (43). In a small study of 16 patients with COPD, impaired endurance time of the quadriceps muscle was shown to be related to physical inactivity (44). Therefore, it seems possible that physical activity may be affected by quadriceps muscle strength, which was not assessed in our study due to the lack of adequate equipment. Preliminary evidence suggests that anemia in patients with COPD may be more prevalent than expected and may be related to systemic inflammation (45). Anemia is associated with reduced functional exercise capacity and a higher mortality rate in patients with COPD (12). The prevalence of anemia in our cohort was 5%, which is less than that reported in a cohort of patients with COPD (mean FEV1, 43% predicted) attending a Veterans Administration clinic (12). The prevalence of 14% in our GOLD stage IV patients, however, was comparable to the reported data in patients with COPD receiving long-term oxygen therapy (46). The presence of anemia was not associated with reduced physical activity in our cohort. Physical activity is defined as any bodily movement produced by skeletal muscles that results in energy expenditure beyond resting energy (1). The "gold standard" to assess total energy expenditure under free-living conditions is the doubly labeled water method (22). This methodology, however, is limited by cost and by its unavailability outside specialized centers. Simple accelerometers are inaccurate at estimating total daily energy expenditure from body movement counts only (47, 48). The SenseWear Pro Armband incorporates several other physiologic sensors that, in conjunction with the accelerometer recordings, give a more valid estimate of total daily energy expenditure (49). The physical activity level, which is defined as total daily energy expenditure divided by resting energy expenditure, is the equivalent of any physical activity performed during the day and has been shown to be a predictor of mortality in a healthy elderly population (22). At present, only limited data about the physical activity levels in COPD are available. Slinde and coworkers determined the physical activity levels in 10 underweight patients with COPD using the doubly labeled water method and indirect calorimetry (50). Physical activity levels ranged from 1.15 to 1.8, which are comparable to the range of physical activity levels found in our cohort. Of interest, only 22% of our patients had a physical activity level of 1.70 or greater, which suggests that most patients with COPD have a sedentary lifestyle. This observation confirms a previous finding that patients with COPD are markedly inactive in daily life (14). In our multivariate analysis, the explained variance of physical activity by the BODE index was similar to the explained variance by GOLD stages. Furthermore, extrapulmonary effects remained significant when added to the BODE index. This is a surprising result, as the BODE index incorporates functional exercise capacity and dyspnea, two parameters that are affected by pulmonary and cardiac function (51). A possible explanation could be the fact that our cohort had a median of only 2 BODE points. The BODE index is potentially superior to GOLD stages in predicting physical activity in a cohort of patients with more severe COPD and a higher BODE score. This possibility, however, is subject to further studies. Our study has limitations that need to be addressed. Because it was a cross-sectional study, no causality or directionality of the findings can be inferred. Furthermore, a selection bias has to be considered, as we recruited our patients from an existing database of patients with COPD who were willing to participate in clinical projects. Therefore, the prevalence and magnitude of observed extrapulmonary effects, which clearly affected statistical analysis in our study, have to be seen in context with other studies and larger epidemiologic cohorts. In summary, this study provides cross-sectional evidence that reduced physical activity in patients with COPD is associated not only with clinical stages of COPD severity but also with left heart dysfunction and systemic inflammation.
The authors thank Dr. Stephanie Bitter-Suermann, Dr. Christine Otten, Dr. Margret Jandl, Dr. Frank Kanniess, Ms. Ines Zimmermann, Ms. Jeanette Kotzur, and Ms. Diana Knut-Rehr for their help collecting patients' data. They thank Prof. Dr. Heiner Raspe for the academic discussions about the topic of the study, and Dr. Olaf Holz and Ms. Mary McKenney for the critical review of the manuscript.
Supported by an unrestricted research grant from AstraZeneca. The funding source had no role in the study design, collection, analysis, and interpretation of the data or in the decision to submit the paper for publication. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200707-1011OC on November 29, 2007 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 10, 2007; accepted in final form November 29, 2007
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
2 test (categorical variables), Kruskal-Wallis test (ordinal or nonnormal metric variables), and analysis of variance (normally distributed metric variables). Skewed data were log transformed to yield normal distributions. 
30 kg/m2), and arterial hypertension were included in the model independent of statistical significance. The potential confounders (drinks per week, smoking status, educational status, pain symptoms, and history of diabetes mellitus) were included in the model if they were of statistical significance (P < 0.05). No adjustment for retirement, early retirement, or employment status was made, as these conditions could also be the result of disease severity. As a second step, GOLD stages or BODE index were included in the model. As a third step, extrapulmonary variables (left ventricular ejection fraction, deceleration time of the early transmitral flow or ratio of the peak velocity of the early E-wave to atrial A-wave, systolic pulmonary artery pressure, NT–pro-BNP, ankle brachial index, fibrinogen or hs-CRP, nutritional depletion, muscle strength, and anemia) were added to the model if they were of statistical significance. All extrapulmonary parameters that were shown to be significant at this step of analysis were included in the final multivariate regression model. For each level (confounders, disease severity represented by GOLD stages or BODE index, extrapulmonary factors), the amount of variance (R2) was calculated. Metric variables were not centered. GOLD stages were dummy-coded with GOLD I as a reference category. BODE index was categorized by quintiles and dummy-coded with the first quintile as the reference category. 
