Comparison of Effects of Endurance and Strength Training Programs in Patients with COPD

ABSTRACT

Comparisons between endurance and strength training in chronic obstructive pulmonary disease (COPD) patients have produced equivocal findings when examining physical function and health-related quality of life (HRQL). One reason for these differences may be due to individual patient responses to the different training modalities. PURPOSE: To compare changes in physical function and HRQL in a group of COPD patients completing both an endurance and a strength training program. METHODS: Eleven mildly diseased patients completed a three month endurance training program and, approximately 5 years later, completed a three month strength training program. Changes in 6 minute walk distance (6 MW), time to rise from a chair five times (CRT), and the total score and subscores from the SF-36 and Chronic Respiratory Disease Questionnaire (CRQ) were examined. RESULTS: The forced expiratory volume as a percent of predicted remained relatively constant over the 5 years (61.1 ± 5.9 vs. 60.0 ± 10.3). Endurance and strength training increased 6 MW by 48.2 ± 11.2 (p = 0.008) and 39.8 ± 9.8 (p = 0.001) meters, respectively. Endurance and strength training decreased CRT by 4.8 ± 0.7 (p = 0.001) and 1.3 ± 1.2 (p = 0.056) seconds, respectively. Endurance training resulted in greater improvements in HRQL as compared to strength training. CONCLUSION: These results show that walk distance improves as a result of participating in either an endurance or a strength training program. However, an endurance training program leads to greater improvements in both general and disease specific measures of HRQL.

KEYWORDS:

• Aerobic training
• exercise
• health-related quality of life
• physical function
• resistance training

Introduction

Chronic obstructive pulmonary disease (COPD) is projected to be the third leading cause of death worldwide in 2020 and fifth leading cause of disability-adjusted life years (1). Once thought of as a disease confined to the lungs, COPD is now recognized as having both pulmonary and skeletal muscle impairments (2). While both lung and skeletal muscle impairments contribute to this disability, skeletal muscle exhibits notable plasticity in response to exercise training. As such, exercise interventions aimed at improving skeletal muscle function have the potential to improve the physical function and health related quality of life (HRQL) of those afflicted with COPD (3).

Endurance training, with the aim of improving cardiorespiratory system function, has been the traditional mode of exercise training used in pulmonary rehabilitation programs (4). Previous research has shown that this form of exercise improves physical function and HRQL in patients with COPD, although individual patient responses are quite variable (5). The mechanisms of improvement have not been clearly defined, as improvements in lung function or cardiovascular function, i.e., peak oxygen consumption, have not been universally reported (6), although improvements in the oxidative capacity of skeletal muscle have been suggested (7,8).

In addition to endurance training, strength training has also been recommended as an exercise modality for improving physical function in patients with COPD. Skeletal muscle mass and strength have been shown to be significantly reduced in patients with COPD when compared to age-matched healthy controls, and there is evidence that skeletal muscle myopathy contributes to the severe disability experienced by patients with COPD (9–11). Therefore, most guidelines on exercise for patients with COPD now recommend a combination of endurance and strength training (4,5,12,13). A recent meta-analysis by Iepsen and colleagues examined studies comparing the effects of combining strength training with endurance training against endurance training alone and concluded that both programs showed similar improvements in physical function and HRQL (14). In contrast, a meta-analysis by Liao et al. concluded that a combined strength and endurance training program may provide greater improvements in HRQL (15).

Very few studies have compared the effects of an endurance only training program to a strength only training program, and those doing so have produced mixed results. Some studies have shown improvements in physical function with endurance training but not strength training (16–18), with strength training but not endurance training (19,20), whereas others have shown improvements with both (21,22). Discordant results are also noted when examining improvements in HRQL, and the quality of evidence from studies supporting one type of training over the other is often poor (23). Reasons for these conflicting results may be due to individual patient responses to the types of training, differences in the disease severity, personal preferences for a given type of training, and subsequent adherence problems. Thus, for patients who are only able to or choose to engage in one exercise modality, it would be important to know if one is more effective. Identifying the type of exercise that yields the greatest improvements in physical function and HRQL may help promote long-term adherence to exercise behaviors. Therefore, to control for the effects of individual patient differences, this investigation compared improvements in physical function and HRQL in a cohort of patients with COPD who completed both an endurance training program, and, subsequently a number of years later, a strength training program.

Methods

Design and overview

Patients in this investigation were originally part of a larger trial that compared the effectiveness of a traditional three-month center-based exercise program with a behavioral lifestyle exercise program that transitioned patients from a center-based to a home-based exercise program (24). Both the center-based and the home-based exercise programs in this previous trial were endurance training programs. Approximately five years later, all patients who had participated in the original endurance-based exercise trial were mailed an invitation to participate in a strength training trial. Eleven of the patients who had participated in the original endurance study responded, agreed to participate in, and completed a three-month strength training study. For both the endurance training and strength training studies, participants were screened for eligibility, and, if eligible, baseline assessments were obtained. In both exercise programs, patients exercised three days/week for three months after which follow-up assessments were made.

Patients

This study included eleven male patients (two African Americans). All patients who agreed to participate signed an informed consent approved by the University's Institutional Review Board prior to participation in each of the exercise studies. Inclusion and exclusion criteria for both exercise studies were similar. Participants were eligible for inclusion if they had an expiratory airflow limitation such that the FEV₁/FVC was <70% and the FEV₁ was >20% of predicted and reported difficulty in performing at least one of the following activities due to dyspnea: walking a city block, grocery shopping, doing household chores, lifting objects chest height or higher, walking upstairs or getting out of a chair. Patients were excluded if they had severe cardiovascular or peripheral vascular disease, were undergoing active treatment for cancer, had uncontrolled hypertension or diabetes, or had participated in a pulmonary rehabilitation or exercise program during the previous three months. Once enrolled in the strength training study, we queried the patients regarding their physical activity levels since their participation in the endurance study. None of the eleven reported that they had regularly exercised or engaged in any form of structured physical activity since their participation in the endurance study. Determination of medical diagnoses that would preclude participation in the study was done through a medical history, physical exam, and a graded exercise test.

Exercise programs

Endurance training

Patients completed a one-hour exercise session thrice weekly for 12 weeks. Each session consisted of a brief warm-up, 30–35 minutes of walking at a rating of perceived dyspnea of 3–5 (moderate to somewhat hard) on the Borg categorical scale, 10–15 minutes of upper extremity endurance training using light elastic resistance bands, and a brief cool-down as suggested by guidelines at the time (25).

Strength training

Patients completed a one-hour exercise session thrice weekly for 12 weeks. Each session consisted of a brief warm-up followed by high-intensity progressive strength training of the following major muscle groups: abductors (overhead press) and horizontal adductors (bench press) of the shoulders; flexors (bicep curls) and extensors (triceps extensions) of the elbows; trunk flexors; hip extensors; and the extensors and flexors of the knee. The exercise program progression was based on American College of Sports Medicine guidelines such that the participants completed three sets of eight repetitions for each exercise at 80% of the maximal load that they could lift one time (1RM) for that given exercise (26). Once a patient was able to complete eight repetitions for all sets, the resistance (weight) was increased by 10%. In addition to increasing the weight by 10% with the completion of eight repetitions across all prescribed sets, the 1 RM was assessed at the end of weeks 2 and 7 so that further adjustments to the intensity could be made. This regimen has been shown to produce significant increases in the strength of older adults (26).

Compliance with each intervention was calculated as the percent of possible exercise sessions (36) completed.

Outcomes

Pulmonary function testing

Pulmonary function tests were performed using a Medical Graphics Corporation 1085D. Spirometry measurements were performed such that reproducibility and acceptability were met for each test according to the American Thoracic Society recommendations (27).

Physical function

Physical function was assessed using several different tests. These included a six-minute walk and timed chair rise. Additionally, self-reported physical function was determined based on the physical function scale of the Medical Outcomes Study 36 Item Short Form (SF-36). The six-minute walk was performed in a dedicated gymnasium with a rectangular walking track that measured 8 meters by 14 meters according to the guidelines of the American Thoracic Society (28). The chair rise test was performed using a straight-backed chair placed with its back against a wall. Participants were first asked to stand from a sitting position without using their arms. If they could perform this task, they were then asked to stand up and sit five times as quickly as possible. The time to complete the task was recorded. No feedback or encouragement was provided for any of the tests.

Health-related quality of life

HRQL was measured using a generic instrument (SF-36) (29) and a disease specific instrument (Chronic Respiratory Disease Questionnaire (CRQ)) (30).

Appendicular composition

Right arm and right leg bone mineral density, fat and lean masses, and percent fat were determined using dual energy X-ray absorptiometry (DEXA) with fan beam technology. The manufacturer's (Hologic, Delphi A) recommendations for patient positioning, scan protocols, and scan analysis were followed (31). The arms and legs were defined using the method of Visser et al. such that the legs were identified by placing a horizontal line at the lowest point of the ischial tuberosity, and the arms by a line through the head of the humerus and the scapula (32). Baseline scans were uploaded prior to analysis of three month scans to ensure consistency in the analysis. Scans were performed and analysed by an International Society for Clinical Densitometry certified technician and reviewed by a board certified radiologist, both of whom were blinded to the subjects' intervention assignment.

Data analyses

Differences between baseline and three-month follow-up scores for both the endurance and the strength training interventions were assessed using dependent t-tests. Change scores (three-month follow-up minus baseline) were calculated for each patient. Differences in change scores between the endurance versus the strength training intervention were assessed using Analysis of Covariance (ANCOVA) with baseline scores as the covariate for any of the outcomes shown to significantly improve from baseline to three months. Additionally, for outcomes where the minimal clinically important difference (MCID) has been determined, the portion of patients within each group that exceeded the MCID for physical function (33,34) and HRQL measures (35,36) was determined. Chi-square analysis was used to determine if there was a difference in the proportion of patients within each training program that experienced an improvement that exceeded the MCID. Effect sizes were calculated using Cohen's d. Significance was set as a p value < 0.05.

Results

Patient characteristics

Patient characteristics prior to completing each of the interventions are shown in (Table 1). Other than age and height, there were no significant differences in baseline characteristics of patients prior to the start of each intervention. On average, the time between participation in each of the exercise interventions was 5 years and ranged between 3 and 7 years.

Table 1. Patient characteristics prior to the start of the endurance and strength interventions.

	Endurance	Strength	Difference
	Mean ± SEM	Mean ± SEM	p
Age, yr	65.0 ± 2.6	70.0 ± 2.7	<0.001
Weight, kg	93.5 ± 7.3	92.3 ± 9.2	0.691
Height, cm	175.3 ± 2.3	172.0 ± 2.0	0.001
BMI, kg/m^2	30.4 ± 2.0	31.5 ± 2.7	0.256
FEV1, L	1.9 ± 0.2	1.4 ± 0.5	0.289
FEV1, % predicted	61.1 ± 5.9	60.0 ± 10.3	0.894
FEV1/FVC, %	57.4 ± 3.5	52.4 ± 7.1	0.417

BMI – body mass index. FEV₁ – forced expiratory volume in one second. FVC – forced vital capacity.

Adherence

Patients attended 85.4 ± 9.7 percent of endurance training sessions and 94.2 ± 10.1 percent of the strength training sessions (p = 0.03).

Outcomes

Comparisons of physical function, HRQL, and appendicular composition outcomes between baseline and three-month follow-up scores for each of the interventions are shown in (Figure 1) and (Tables 2 and 3), respectively. Comparisons of the proportion of patients achieving the MCID in physical function and HRQL measures following each of the interventions are shown in (Table 4).

Figure 1. Physical function scores at baseline and three months for the endurance and strength training interventions are shown. p values for the individual comparison between the baseline and the three month values are presented. SF36-PFS is the physical function scale of Medical Outcomes Study 36 Item Short Form (SF-36) with scores ranging from 0 to 100. All values are mean ± SEM.

Table 2. Health-related quality of life outcomes by group at baseline and three month follow-up.

				p^*	Effect size	Change score	p^†
SF-36 Physical component score	Endurance	Baseline	35.1 ± 3.2	0.049	0.77	6.5 ± 1.3	0.910
		Three month	40.8 ± 2.2
	Strength	Baseline	33.4 ± 2.3	0.822	0.08	1.3 ± 2.9
		Three month	34.1 ± 2.5
SF-36 Mental component score	Endurance	Baseline	57.7 ± 1.5	0.224	0.44	NA	NA
		Three month	59.3 ± 1.9
	Strength	Baseline	56.6 ± 3.6	0.190	0.48	NA
		Three month	60.2 ± 1.9
CRDQ dyspnea	Endurance	Baseline	4.6 ± 0.3	0.125	0.53	NA	NA
		Three month	5.1 ± 0.4
	Strength	Baseline	4.3 ± 0.2	0.114	0.56	NA
		Three month	5.0 ± 0.3
CRDQ emotion	Endurance	Baseline	5.6 ± 0.2	0.066	0.66	NA	NA
		Three month	6.0 ± 0.2
	Strength	Baseline	5.4 ± 0.4	0.171	0.47	NA
		Three month	5.9 ± 0.2
CRDQ mastery	Endurance	Baseline	5.1 ± 0.4	0.199	0.44	NA	NA
		Three month	5.7 ± 0.4
	Strength	Baseline	5.6 ± 0.4	0.301	0.35	NA
		Three month	5.9 ± 0.3
CRDQ fatigue	Endurance	Baseline	4.1 ± 0.3	0.023	0.88	0.8 ± 0.3	0.898
		Three month	5.0 ± 0.3
	Strength	Baseline	3.9 ± 0.4	0.073	0.64	0.9 ± 0.3
		Three month	4.8 ± 0.3
CRDQ total	Endurance	Baseline	4.9 ± 0.2	0.045	0.74	0.6 ± 0.1	0.626
		Three month	5.5 ± 0.2
	Strength	Baseline	4.9 ± 0.3	0.072	0.66	0.4 ± 0.1
		Three month	5.4 ± 0.2

Baseline and three month values are means ± SEM. Change scores are estimated marginal means ± SEM.

^* This p value is from the comparison between the baseline and three month follow-up score within each intervention group.

^† This p value is from the comparison of change scores (baseline – 3 month) from each intervention group. NA – not applicable to calculate change scores or to compare change scores, since they did not change significantly from baseline to three months. Effect sizes are for changes in scores from baseline to three months.

Table 3. Appendicular composition by group at baseline and three month follow-up.

				p^*	Effect size	Change score	p^†
Arm bone mineral density (g/cm^2)	Endurance	Baseline	0.82 ± 0.02	0.633	0.18	NA	NA
		Three month	0.82 ± 0.01
	Strength	Baseline	0.82 ± 0.02	0.793	0.27	NA
		Three month	0.82 ± 0.02
Arm lean mass (kg)	Endurance	Baseline	3.9 ± 0.3	0.243	1.28	NA	NA
		Three month	3.8 ± 0.2
	Strength	Baseline	3.5 ± 0.3	0.848	0.20	NA
		Three month	3.5 ± 0.3
Arm fat mass (kg)	Endurance	Baseline	1.2 ± 0.2	0.288	1.15	NA	NA
		Three month	1.1 ± 0.1
	Strength	Baseline	1.1 ± 0.1	0.288	1.15	NA
		Three month	1.1 ± 0.1
Arm percent fat	Endurance	Baseline	22.3 ± 2.0	0.550	0.63	NA	NA
		Three month	21.9 ± 1.5
	Strength	Baseline	21.8 ± 1.8	0.278	1.18	NA
		Three month	23.0 ± 2.0
Leg bone mineral density (g/cm^2)	Endurance	Baseline	1.12 ± 0.02	0.839	0.21	NA	NA
		Three month	1.12 ± 0.02
	Strength	Baseline	1.13 ± 0.02	0.8063	0.26	NA
		Three month	1.13 ± 0.02
Leg lean mass (kg)	Endurance	Baseline	9.4 ± 0.6	0.320	1.07	NA	NA
		Three month	9.3 ± 0.6
	Strength	Baseline	8.7 ± 0.6	0.490	0.73	NA
		Three month	8.8 ± 0.6
Leg fat mass (kg)	Endurance	Baseline	3.2 ± 0.4	0.435	0.83	NA	NA
		Three month	3.1 ± 0.3
	Strength	Baseline	3.0 ± 0.4	0.586	0.57	NA
		Three month	3.0 ± 0.3
Leg percent fat	Endurance	Baseline	24.0 ± 1.8	0.809	0.25	NA	NA
		Three month	23.8 ± 1.4
	Strength	Baseline	23.8 ± 1.9	0.602	0.55	NA
		Three month	24.2 ± 1.6

All values are mean ± SEM.

^* This p value is from the comparison between the baseline and three-month follow-up score within each intervention group.

^† This p value is from the comparison of change scores (baseline – 3 month) from each intervention group. NA – Not applicable to calculate change scores or to compare change scores, since they did not change significantly from baseline to three months.

Table 4. Proportion of patients in each group exceeding MCID.

	Endurance	Strength	p†
Physical function
Six minute walk	67	63	0.86
Chair rise	100	57	0.04
SF36-PFS	78	40	0.10
HRQL
CRDQ – dyspnea	50	60	0.65
CRDQ – emotion	30	30	1.0
CRDQ – mastery	70	60	0.64
CRDQ – fatigue	60	60	1.0

Data are proportion of patients within each intervention arm whose improvements from baseline to three months exceeded the minimally clinically important difference (MCID).

Physical function

Six-minute walk distance was found to increase significantly from baseline to the three-month follow-up after both the endurance and the strength training interventions. Effect sizes for these differences were 1.23 and 2.18, respectively. When comparing the improvements in six-minute walk distance from baseline to three-month follow-up between the endurance and the strength interventions (48.2 ± 11.2 and 39.8 ± 9.8 meters, respectively), there was no significant difference (p = 0.473). There was no significant difference in the proportion of patients that experienced an improvement in six-minute walk distance that exceeded the MCID following each of the interventions. Chair rise time was found to increase significantly from baseline to the three-month follow-up in the endurance group, and there was a trend for improvement in the strength group. Effect sizes for these differences were 1.79 and 0.90, respectively. When comparing the changes in chair rise times from baseline to three-month follow-up between the endurance and the strength interventions (–4.8 ± 0.7 and–1.3 ± 1.2 seconds, respectively), there was no significant difference (p = 0.238). There were a significantly greater proportion of patients that experienced an improvement in chair rise time that exceeded the MCID after endurance training. Self-reported physical function was found to increase significantly from baseline to the three-month follow-up only in the endurance group. Effect sizes for this difference and the difference for the strength group were 0.81 and 0.13, respectively. When comparing the changes in self-reported physical function from baseline to three-month follow-up between the endurance and the strength interventions (3.8 ± 0.7 and 0.5 ± 1.2 units, respectively), there was no significant difference (p = 0.223). There was no significant difference in the proportion of patients that experienced an improvement in self-reported physical function that exceeded the MCID following each of the interventions.

Health-related quality of life

The physical component score of the SF-36 was found to increase significantly from baseline to the three-month follow-up only in the endurance group. When comparing the changes in the physical component score from baseline to three-month follow-up between the endurance and the strength interventions, there was no significant difference. The mental component score of the SF-36 did not change significantly following either the endurance or the strength interventions. The fatigue score from the CRQ was found to improve significantly from baseline to the three-month follow-up only in the endurance group. When comparing the changes in the fatigue score from baseline to three-month follow-up between the endurance and the strength interventions, there was no significant difference. Dyspnea, mastery, and emotion scores from the CRQ did not change significantly following either the endurance or the strength interventions. The CRQ total score was found to improve significantly from baseline to the three-month follow-up in the endurance group, and there was a trend for improvement in the strength group. When comparing the changes in the total score from baseline to three-month follow-up between the endurance and the strength interventions, there was no significant difference. When comparing the proportion of patients that experienced an improvement in any of the CRQ domains that exceeded the MCID, there were no significant differences between the two interventions.

Appendicular composition

None of the appendicular composition outcomes for either the arm or the leg were found to change significantly following either the endurance or the strength interventions.

Discussion

Results from this investigation show that both endurance training and strength training programs have the potential to improve physical function in patients with COPD. Additionally, endurance training appears to result in greater improvements in HRQL. Following the endurance training program, patients had improvements in all measures of physical function (both objectively measured and self-reported); whereas strength training only resulted in improvements in six-minute walk distance. And while there was a trend for an improvement in the chair rise time with strength training, there was no improvement in self-reported physical function. When comparing the proportion of patients that experienced improvements in physical function that exceeded the MCID, it appears that endurance training had a beneficial effect.

Numerous studies have independently examined the effects or either strength training or endurance training on walking performance; however, few have made direct comparisons within the same study, and those doing so have produced conflicting results. Similar to the results of the present investigation, Spruit et al. (21) and Zanini et al. (22) found six minute walk distance to increase following either an endurance or strength training program. Additionally, Skumlien et al. found treadmill endurance time, a task they equated with activities of daily living, to increase in response to both endurance and strength training; however, they only reported increases in six minute walk distance following endurance training (17). Würtemberger and colleagues also reported increases in six minute walk distance following endurance training, but not strength training (18). In contrast to studies finding improved walk performance following endurance training, Ortega et al. (20) and Dourado et al. (19) found improvements following strength training but not endurance training. One reason for these differences may be related to the disease severity of the patients. Würtemberger and colleagues reported that severely diseased patients respond better to strength training, whereas moderately diseased patients respond better to endurance training (18). While the results of Ortega et al. (20) and Skumlien et al. (17) support this contention, results from the present investigation and others cited above do not (19,21,22). Interestingly, we have previously found that all patients engaging in an exercise program emphasizing endurance training will increase walking performance; however, those moderately diseased appear to benefit the most from such a program (37). Further research is needed to determine if patients with differing levels of disease severity are more responsive to different training modalities.

While the six-minute walk test is one of the most commonly used tests of physical function in patients with COPD, studies have examined other outcomes of physical function, such as chair rise time. One reason for this is the six-minute walk test is considered a test of endurance; whereas the chair rise test is considered a test of strength. We found significant improvements in chair rise time following endurance training, and a trend for an improvement following strength training. The improvement following endurance training exceeded the MCID (1.7s) for the five-repetition sit-to-stand chair rise test, whereas the improvement following strength training was very close (34). Both groups had improvements greater than those reported by Jones et al. for a group of patients completing a multi-disciplinary pulmonary rehabilitation program (34). Zanini et al. found that both endurance only and strength only training improved the 30 second sit-to-stand test (22). While the sit-to-stand test is purported to be a measure of lower limb strength in older adults, Jones et al. (34) and Zanini et al. (22) found only modest relationships between sit-to-stand test results and quadriceps strength. These results, along with the greater improvements we found with endurance training versus strength training, suggest that factors other than strength, such as skeletal muscle endurance, lung function, or balance, may influence sit-to-stand performance in patients with COPD.

It is important to measure both objective and self-reported measures of physical function because of differences between these measures (38). While the measures are related to one another, research has shown that they each tap into different constructs, with differences between the two attributable to a patient's psychological well-being and self-efficacy beliefs (39). Both training modalities resulted in improvements in objectively measured physical function tasks; however, only endurance training resulted in self-reported improvements in physical function. This 3.8 point improvement surpassed the MCID of 1.0 point, and there was a trend for a greater proportion of endurance trained patients to achieve the MCID when compared to strength trained (36). Benton and Wagner compared a traditional pulmonary rehabilitation exercise program emphasizing endurance training to one with strength training added, and found both groups improved SF-36 physical function scores post intervention (40). Others have also demonstrated that a comprehensive pulmonary rehabilitation program consisting of both endurance and strength training leads to increases in self-reported physical function (41). Thus, it appears that it is important to include endurance training to improve self-reported physical function in patients with COPD.

Consistent with previous investigations comparing endurance training and strength training, this investigation found similar improvements in overall scores for disease specific measures of HRQL following both types of training (21). When examining the individual domains of the CRQ, we found a significant improvement in the fatigue domain and a trend for improvement in the emotion domain with endurance training. We also found a trend for improvement in the fatigue domain with strength training. While other studies have noted improvements in most domains with both endurance training (37,42) and strength training (20,43), there are several reasons we may not have seen significant improvements. The first may be related to the fact that the interventions in the current investigation were exercise only. Previous studies included exercise as part of a comprehensive pulmonary rehabilitation program. In addition to exercise training, these programs included education and psychosocial and behavioral interventions. Another reason for the lack of statistical significance may be related to the relatively small sample size. While we did not see statistically significant improvements in some of the CRQ domains, it is interesting to note that large proportions of patients within each group had improvements that exceeded the MCID and the magnitude of the improvements was similar to those previously reported (20,42). We also found that endurance training resulted in improvements of the physical component score of the SF-36, but not the mental component score. These results are similar to the findings of Limsuwat et al. who reported that participation in a pulmonary rehabilitation program emphasizing endurance training yielded improvements in the physical component score but not the mental component score (44). Benton and Wagner compared a traditional pulmonary rehabilitation exercise program to one with strength training added and found both groups improved physical component scores of the SF-36 post intervention (40).

We saw no changes in the appendicular composition with either training modality. This is in agreement with previous studies that also failed to find significant increases in lean body mass following strength training (19,45) or endurance training (46) in patients with COPD. Studies which have combined exercise training with nutritional or anabolic support have shown increases in lean body mass, and it has been suggested that nutritional or anabolic support is needed if gains in lean mass are to be achieved (45–47). Our results suggest that short-term exercise training alone, whether it be endurance or strength training, is not a sufficient stimulus to increase lean body mass in patients with COPD.

A unique aspect and strength of this investigation was the fact that we used the same patients in both exercise interventions, therefore, controlling for between subject variability. Despite the fact that the time between the patients' participation in the endurance and strength training intervention was five years, the patients' disease status remained relatively stable. While the patients in this investigation had mild to moderate airflow limitations, none were smoking at the time of interventions nor did they report smoking between the interventions. We have previously shown that COPD patients across the spectrum of disease severity can realize improvements in physical function and HRQL following participation in an exercise program (37). However, we did note that the gains achieved by those more severely diseased were less than those with moderate to mild disease. By having patients whose disease severity remained stable between the two exercise interventions, we controlled for the possibility of greater improvements as a result of being less severely diseased. Additionally, none of these patients were physically active prior to the start of either intervention, and they had similar levels of physical function and HRQL at the start of each intervention.

The study did have several limitations. Only eleven subjects participated in both training interventions which may have resulted in decreased statistical power. This was partially offset by the use of within subject analyses; however, there were several outcomes that only approached statistical significance. A larger sample size would have been beneficial for these analyses. Additionally, the results may have been affected by selection bias, specifically volunteer bias, where the individuals that volunteered to participate in the training programs may be systematically different from those who did not volunteer. Unfortunately, we did not have any follow-up data from those in the original endurance trial other than the eleven that volunteered for the strength training trial. Another limitation was the lack of female patients. A final limitation of the study was a lack of strength measures. One repetition maximum strength measures were available from the strength training intervention; however, they were not available from the endurance training intervention.

In summary, the results of this study show that walking distance improves as a result of participating in either an endurance or a strength training program. However, an endurance training program also improves subjective measures of physical function, as well as producing greater improvements in both general and disease specific measures of HRQL. As medicine becomes more personalized, so too should rehabilitation. For patients with COPD with a reduced HRQL, endurance exercise may be a better training modality. For those patients whose HRQL is not compromised, either training modality appears to be beneficial. Additionally, individual preferences for one type of training modality, as well as the likelihood of adhering to one training program over the other, should be taken into consideration when prescribing a training modality.

Declaration of interest

MJB, KLS, and NEA declare no conflicts of interest in relation to this article. This work was supported by National Institutes of Health grants HL53755 and AG21332.

Additional information

Funding

National Institutes of Health(HL53755 and AG21332)