Impact of obesity on exercise performance and pulmonary rehabilitation


  • The Authors: Michael Dreher, MD is a senior respiratory physician at the Department of Pneumology, University Hospital Freiburg, Germany. His research focuses on several aspects of mechanical ventilation including its effects at rest and during exercise, and the process of weaning from mechanical ventilation. Hans-Joachim Kabitz, MD works as respiratory specialist in the Department of Pneumology, at the University Hospital Freiburg, Germany, where he also heads the Respiratory Physiology Research Group. His research focuses on the evaluation of respiratory muscle function and fatigue both at rest and during exercise.


Michael Dreher, Department of Pneumology, University Hospital Freiburg, Killianstrasse 5, D—79106 Freiburg, Germany. Email:


While the impact of obesity on respiratory function has been extensively studied, and several definitive conclusions have emerged, its impact on exercise performance is complex, with the available data sometimes providing contradictory or inconclusive information.

Based on the literature discussed, it appears that resting alterations in lung volumes and gas exchange become attenuated during exercise in the obese, while oxygen cost of breathing and dyspnoea are increased. Respiratory muscle function also seems to be impaired, such that inspiratory muscle strength is reduced and respiratory drive is increased. Furthermore, while there is no reduction in the absolute values of maximal oxygen uptake compared with normal-weight subjects, oxygen uptake at a given workload is increased and maximal workload is reduced in the obese, caused by increases in body mass and/or basal metabolic rate. To date, obesity has not been listed as an indication for pulmonary rehabilitation (PR), hence the reason why conclusive data on the impact of obesity per se on PR are lacking. The majority of evidence discussed is based on comparative data from obese versus normal-weight patients, with respiratory disorders currently established as indications for PR. The best evidence currently available regarding the impact of obesity on PR is for patients with chronic obstructive pulmonary disease (COPD); here, it appears that obesity per se has no negative impact on PR. Otherwise, there are no conclusive data on the impact of obesity on PR in respiratory disorders other than COPD, and this remains to be investigated in the future.


According to the body mass index (BMI), individuals can be categorized as underweight (<21 kg/m2), normal weight (21–25 kg/m2), overweight (25–30 kg/m2), obese (>30 kg/m2) or morbidly obese (>40 kg/m2).1,2

Among its wide range of adverse effects, obesity has been shown to impact on respiratory function, with the associated pathophysiological changes outlined in a previous report in this series.2

In addition, it is known that obesity—even in the absence of an underlying respiratory illness—causes increased breathlessness that is further aggravated during exercise.3,4 However, whether these pathophysiological changes in respiratory function and symptoms (mainly dyspnoea) influence exercise performance in the obese is yet to be fully determined.

Although pulmonary rehabilitation (PR) is based on several interventional measures (e.g. treatment of body composition abnormalities, pharmacological and non-pharmacological therapy, self-management education, psychological support), exercise training is widely regarded as the cornerstone of PR.1 Accordingly, PR has been shown to improve exercise performance in chronic respiratory disease patients with impaired exercise tolerance, exertional dyspnoea or fatigue, and/or impairment of daily life activities.1 In addition, PR has been proposed as a useful adjunct in the comprehensive management of obese subjects.1 However, the application of exercise training in the obese is a difficult undertaking given the associated pathophysiological changes in respiratory function, as well as the possibility that there is a pre-existing impairment in exercise performance. Again, these important questions still require elucidation. The aim of the following review is therefore to clarify and discuss the impact of obesity on exercise performance and PR.


There are several variables in the obese and morbidly obese that differ from normal-weight controls which might impact on exercise performance. The impact of obesity on respiratory function at rest has already been outlined in detail both in this review series and elsewhere;2,5 accordingly, while some of these components are of potential importance to exercise performance in the obese, they will only be summarized rather than extensively discussed in the following section. A synopsis of the complex and numerous pathophysiological changes associated with obesity is given in Table 1.

Table 1.  Obesity-related pathophysiological changes that potentially impact on exercise performance
 At restDuring exerciseSupporting evidence in the literatureConclusive results
  1. FEV1, forced expiratory volume in 1 s; N/A, not applicable/data not available; PaCO2, arterial partial pressure of carbon dioxide; PaO2, arterial partial pressure of oxygen; RV, residual volume; TLC, total lung capacity; VC, vital capacity; ↑, slightly, ↑↑, moderately, ↑↑↑, strongly increased compared with healthy normal-weight controls; ↓, slightly, ↓↓, moderately, ↓↓↓, strongly decreased compared with healthy normal-weight controls; ↔, no difference compared with healthy normal-weight controls.

Breathing pattern, compliance and lung volumes    
 Tidal volumeStrongYes
 Breathing frequency↑↑StrongYes
 Minute ventilationStrongYes
 Respiratory system compliance [total]↓↓N/AStrongYes
 Operating lung volumes↓↓↓↓↓ to ↓StrongYes
 Expiratory flow limitation↑↑Intermediate to strongYes
 Airway resistanceIntermediate to strongYes
 TLC, VC, FEV1, RV↔ to ↓N/AStrongYes
Oxygenation, ventilation and gas exchange    
 PaO2↔ to ↓↔ to ↓IntermediateYes
 Alveolar-arterial oxygen difference↑↑↑↑↑ to ↑StrongYes
 Diffusion capacity↔ to ↓↔ to ↓IntermediateYes
 Metabolic cost of exerciseN/A↑↑IntermediateYes
 PaCO2↔ to ↑↔ to ↑IntermediateNo
 Oxygen uptakeStrongYes
 Peak oxygen uptake [absolute values]N/AStrongYes
 Maximal workloadN/AIntermediateYes
Respiratory and limb muscle function    
 Respiratory drive↑↑↑↑IntermediateYes
 Respiratory muscle strengthN/AIntermediate to weakNo
 Respiratory capacityN/AWeakNo
 Respiratory muscle load↑↑Weak to intermediateYes
 Oxygen cost of breathingIntermediateYes
 Work of breathing↑↑Intermediate to weakYes
 Limb muscle strength↔ to ↑N/AWeakNo
 Oxidative muscle performance↓↓WeakNo
Psychological aspects and dyspnoea    
 Quality of lifeN/AIntermediate to weakNo
 Dyspnoea↔ to ↑↑ to ↑↑StrongYes

Given that several obesity-associated changes (e.g. in gas exchange) appear to be influenced by the pattern of obesity (i.e. upper or abdominal vs lower body obesity) and body composition, it is important to consider factors such as waist-to-hip ratio and/or fat-free body mass in addition to BMI.2,6–9 Here, upper compared with lower body fat distribution has been linked to reduced cardiopulmonary endurance during exercise in morbidly obese women.7 Interestingly, lung volume changes in the obese have been attributed to the cumulative effect of increased chest-wall fat, rather than to a specific regional distribution of fat in the chest wall.9 Furthermore, relative fat distribution does not seem to differ between normal-weight and obese subjects.9 The pathological findings associated with obesity are reportedly aggravated when body position is changed from upright (either standing or sitting) to supine.6 Moreover, obesity is known to negatively impact on mobility in older adults (>60 years of age), resulting in compromised walking, stair climbing and chair-rise ability.10

Interestingly, and unlike what is commonly believed, the absolute values of peak oxygen consumption, and therefore cardiovascular conditioning, have been repeatedly demonstrated not to differ in the obese compared with non-obese subjects.6,8,11–13 However, once peak oxygen values are expressed in relation to bodyweight, there is obviously a notable decline in aerobic capacity.6 Therefore, peak oxygen consumption values should be expressed either in relation to lean body mass or as predicted oxygen consumption values adjusted for bodyweight, by the use of prediction equations.13 Similar results were reported in obese adolescents.14 In another comparative study of 17 morbidly obese subjects (BMI 42 ± 9 kg/m2) and 14 normal-weight controls, workload at maximal exercise was significantly lower in the obese.15

A major burden of obesity in terms of exercise performance is the huge metabolic cost required for a given level of exercise; this has been demonstrated in weight-supported exercise (e.g. cycling), whereby the energy expenditure per unit of workload on a bicycle ergometer was markedly increased in the obese.16 These observations imply that obese compared with normal-weight subjects work at a level that is closer to their maximal performance, both during weight-supported and non-weight-supported exercise, even in everyday activities that would normally require submaximal exertion.

Breathing pattern, compliance and lung volumes

In obese and morbidly obese subjects at rest, tidal volume is reduced, but breathing frequency continues to increase, resulting in a higher minute ventilation compared with that of normal-weight controls.2,5 Again, upper versus lower body fat distribution in the morbidly obese had an impact on these changes, with more rapid, shallow breathing (i.e. higher breathing frequency and lower minute ventilation) in the upper body fat distribution cohort.7 While there is inconsistent data on whether or how changes in flow rates or breath timing account for the increased breathing frequency in morbidly obese subjects,2 it appears from the available data that exercise performance in the obese might be impaired due to the fact that a rapid, shallow breathing pattern accompanied by increased dead space ventilation prevents the obese subject from achieving the required increase in minute ventilation during exercise. However, it has been suggested that the rapid breathing pattern is an appropriate compensatory response aimed at minimizing the mechanical effects of elastic loading, and this might help to reduce respiratory discomfort in the obese.5,17,18 Furthermore, there is evidence that compensatory hyperventilation during exercise is insufficient in obese and morbidly obese subjects.6

While there is little, if any, doubt that total respiratory system compliance is reduced in the obese, the question remains if, and to what degree, chest wall (due to thoracic fat mass loading) and/or lung compliance (due to reduced lung volumes resulting in dystelectasis/atelectasis) is impaired, and how this impacts on respiratory muscle overload.2,19–21 Reduced total respiratory system compliance, however, implies that the respiratory muscles of obese persons need to achieve larger changes in pressures in order to ventilate the same amount of air as normal-weight subjects.

Obesity with small airway occlusion and cranial displacement of the diaphragm due to increased abdominal pressures8,11 leads to exponential decreases in expiratory reserve volume and end-expiratory lung volume (functional residual capacity), thus forcing obese subjects to breathe at lower lung volumes.2,22,23

Interestingly, this reduction in operating lung volumes, together with reduced hyperinflation (i.e. improved operating length of the diaphragm), greater static lung elastic recoil and improved ventilatory efficiency in the obese, has been shown to benefit patients with chronic obstructive pulmonary disease (COPD).17,24 In these cases, obese COPD patients showed reduced dyspnoea and increased peak oxygen uptake during cycle ergometry when compared with normal-weight COPD patients.17 Furthermore, it has been reported that a dynamic increase in end-expiratory lung volume accompanied by a reduction in resting (increased) inspiratory capacity during exercise results in optimized operating lung volumes, which aim to meet the increased ventilatory requirements of exercise and attenuate expiratory flow limitation in obese but otherwise healthy women.5,18

Additional changes in the obese include increased airway resistance, while residual volume, total lung capacity, forced expiratory volume in 1 s and vital capacity remain mostly unaltered unless extreme (abdominal) obesity is present.2,5,6 The reduction in operating lung volumes, together with increases each in expiratory flow limitation, airway resistance and inspiratory muscle load (see later), combine to increase the mechanical ventilatory constraints during exercise in the obese,25 which is once again likely to have a negative impact on exercise performance.

Oxygenation, ventilation and gas exchange

Arterial partial pressure of oxygen is at best only slightly reduced in obesity, whereas the alveolar-arterial oxygen difference is increased, most likely as a consequence of a ventilation-perfusion mismatch (occluded small airways in dependent lung zones).2,6,26,27 It has been shown that gas exchange at rest is negatively influenced by a higher waist-to-hip ratio, and because this ratio is normally greater in men, gas exchange is relatively better in morbidly obese women.6

Diffusion capacity, corrected for lung volume, at rest is mostly preserved in obese subjects,2 but is reportedly reduced during exercise in comparison with normal-weight controls.26 Interestingly, the marginal reduction in arterial partial pressure of oxygen and increase in alveolar-arterial oxygen difference are each improved during exercise in the obese, most likely because of the subsequent recruitment of non-ventilated dystelectatic, but perfused, areas in the lung.2,6

Obesity results in a higher basal metabolic rate, and, for any given work rate, obese participants are subjected to higher oxygen consumption during exercise at a given workload; while this is most likely caused by an elevated basal rate, it has alternatively been attributed to the extra energy needed to move the heavier legs during cycling.2,28 Morbidly obese subjects with upper compared with lower body fat distribution had a higher oxygen requirement and lower anaerobic threshold, while peak workload did not differ between the two groups.7

Another interesting finding with regard to the impact of obesity on exercise performance is that obesity has been identified as an important factor in the development of acute mountain sickness, most likely attributable to impaired breathing during sleep.29 In particular, acute mountain sickness scores were higher and arterial oxygen saturation was lower in the nine obese compared with 10 non-obese participants after 24 h in a hypobaric environmental chamber (simulated altitude: 3658 m).29

Obesity hypoventilation syndrome (OHS) is defined as the combination of obesity (BMI >30 kg/m2) and daytime hypercapnia with arterial partial pressure of carbon dioxide >45 mm Hg (provided other possible causes for daytime hypercapnia are excluded).30 OHS has a highly complex pathophysiological background and an estimated prevalence of up to 50% in patients with a BMI >50 kg/m2.30,31

However, in a recent review based on 41 studies of blood gas analyses in morbidly obese subjects, mean arterial partial pressure of carbon dioxide levels were found to be normal (41 mm Hg, 617 subjects) but included pathological readings because of the wide range of values (mean range 32–73 mm Hg, 617 subjects).6 If present in obese subjects, hypercapnic respiratory failure is an obvious surrogate for the negative impact of obesity on exercise performance.

Respiratory and limb muscle function

In their comparably recent review, Scano and co-workers provide a detailed overview of the complex function and regulation of respiratory muscles in obesity and their potential role in dyspnoea.21 Neural respiratory drive, as estimated via the mouth occlusion pressure after 0.1 s (P0.1), or via an electromyogram of the diaphragm, is reportedly increased in obese subjects, while (global) inspiratory muscle strength, as judged via static inspiratory mouth occlusion pressure (PImax) and respiratory capacity (P0.1/PImax ratio; higher values indicate reduced capacity), are reduced.15,32,33 However, in another study of 45 morbidly obese subjects, there was only a tendency for PImax values to be lower than those of normal-weight controls.34 Furthermore, it has recently been demonstrated that respiratory muscle strength, as assessed by sniff pressures, is similar when obese and normal-weight COPD patients are compared.24 Another study revealed that the ratio of elevated peak electromyographic activity in the diaphragm versus changes in transdiaphragmatic pressure was not linear in obese patients,35 suggesting a dissociation between neural and mechanical events.21 Obese subjects present with an overall increase in inspiratory muscle activity both at rest and during exercise, as well as an increased load on the inspiratory muscles (estimated via the specific inspiratory impedance).15 This has been suggested to predispose obese subjects to the development of exercise-related respiratory muscle fatigue.15 Again, this was shown to differ when obese and normal-weight COPD patients were compared, because in this case, the net load-capacity ratio of the respiratory muscles was similar between the two groups.24 However, several of the aforementioned techniques are volitional and thus rely on the subject's own motivation and willingness to cooperate; therefore, pathologically low values, in particular, must be viewed with caution. Objective assessment of these findings by the use of non-volitional techniques which assess respiratory muscle function and fatigue (e.g. magnetic phrenic nerve stimulation) is yet to be established.

Obesity results in a higher ventilatory requirement (at a given workload),8 a higher work of breathing (at a given ventilation level)36 and a greater oxygen cost of breathing, which, even in the moderately obese, can reach values that are three times higher than those measured in normal-weight controls.15,37,38 Obesity-related chest loading with reduced lung volumes and increased mechanical ventilatory constraints during exercise has been suggested as a possible explanation for impaired exercise performance in the obese and morbidly obese.6

Obesity is known to be associated with physical inactivity and can result in profound deconditioning, particularly of the limb muscles, which are otherwise subjected to a kind of chronic strength training (due to the increased bodyweight).14 Obese subjects have been reported to present with an increased proportion of the less efficient, less fatigue-resistant glycolytic type IIb muscle fibres (compared with type I fibres).39 In addition, the skeletal muscles of obese subjects have a reduced oxidative capacity and show a decrease in mitochondrial content.40 In obese COPD patients (BMI 34.9 ± 1.5 kg/m2), compared with healthy controls (BMI 27.0 ± 1.1 kg/m2), increased limb strength but significant exercise-induced limb fatigue (at exercise intensities of only 52% peak work rate) have been reported;41 here, diverse interventions promoting respiratory muscle unloading attenuated peripheral muscle fatigue, whereas the largest portion of the fatigue persisted.41 It was concluded that locomotor muscle fatigue in obese COPD patients might be partly attributable to insufficient oxygen transport, resulting from exaggerated arterial hypoxaemia and/or excessive respiratory muscle work, but the well-documented alterations in intrinsic muscle characteristics also play a critical role.41

A recent study revealed that the gas exchange threshold was significantly lower in obese adolescents compared with normal-weight controls, while the slow component of oxygen consumption kinetics differed between the two groups, with earlier termination of exercise in the obese.14 These findings suggested a negative impact of obesity on exercise performance, where oxidative performance of the skeletal muscle was thought to be impaired.14

Psychological aspects and dyspnoea

A recent meta-analysis confirmed a reciprocal link between obesity and depression; that is obesity increases the risk of developing depression, while depression is predictive for the development of obesity.42 The consequences of these aspects with regard to the impact of obesity on exercise performance were illustrated in a recent study of 103 adolescents, where patients with elevated depressive symptoms displayed worse cardiorespiratory fitness than those without.43

While dyspnoea under exertion is reported frequently in obese subjects, the underlying mechanisms are very complex and much is still open to debate.2,44–46 It has been shown that exertional dyspnoea for any given work rate is increased in obese women, but it is similar to normal-weight controls when any given oxygen uptake or minute ventilation is compared between groups.18 This implies that respiratory mechanical/muscular factors per se contribute only minimally to the increased dyspnoea.5,18,46 However, it has been emphasized that the interactions which are of relevance in this context are complex, and that respiratory muscle function might indeed play a role in exertional dyspnoea in the obese.21 On the other hand, it has been suggested that it is rather the increased oxygen cost of breathing (for a given work rate) that underlies the increase in exertional breathlessness in the obese.18,46 In line with this, it has been shown that when obese women with or without exertional dyspnoea are compared, the oxygen cost of breathing is increased by as much as 70% in the former subgroup.46 Alternatively, increased exertional breathlessness in the obese was suggested to ultimately reflect a normal awareness for increased ventilation and more contractile respiratory muscle effort during weight-supported exercise.5

The optimization of operating lung volumes resulting in attenuated expiratory flow limitation has been suggested to anticipate a further increase in breathlessness in obesity.18 Interestingly, exercise capacity was shown to be similar when obese but otherwise healthy women with dyspnoea were compared with those without dyspnoea.46 Finally, dyspnoea and daytime sleepiness, which is frequently observed in OHS patients,47 negatively impact on the quality of life in obese patients.48


Pulmonary rehabilitation

In the course of evaluating the impact of obesity on PR, the underlying pulmonary pathologies which justify participation in PR programmes (e.g. COPD) have to be taken into account, because obesity per se does not represent an indication for PR.

In COPD patients, PR has been shown to improve important physiological and clinical parameters such as exercise tolerance and dyspnoea, as well as health-related quality of life.1,49 For this reason, several randomized controlled trials and meta-analyses have recommended PR as an important inclusion in the comprehensive treatment programme designed for COPD patients.1,49,50

Interestingly, low bodyweight and muscle wasting have been identified and thoroughly addressed as relevant factors that have a negative impact on the prognosis of COPD patients.51,52 In contrast, there is only limited information available on the impact of obesity on PR in COPD, despite its frequency in these patients.53,54 In a relatively recent study, the impact of obesity on the outcome of PR was retrospectively assessed in 114 COPD patients.53 Apart from lower baseline values for the 6-min walking distance (6MWD) test in obese compared with non-obese COPD patients, the two groups demonstrated comparable improvements in 6MWD following an outpatient PR programme.53 Similar results were recently reported in a prospective study investigating the impact of being overweight or obese on PR in 261 COPD patients.54 In contrast, another recent study investigating overweight COPD patients revealed that a BMI >25 kg/m2 is an independent predictor of rehabilitation efficacy, at least in terms of improvement in the 6MWD.55 The most likely explanation for this finding is thought to be based on the fact that these patients had greater deconditioning and hence more rehabilitation potential compared with normal-weight patients.55 Furthermore, a small retrospective analysis revealed that even in morbidly obese COPD patients, significant functional improvements could be achieved by an institutional rehabilitation programme.56 Based on this data, it appears that obesity per se does not have a negative impact on PR outcome in COPD patients,53,54 and there is even one case report in which the authors concluded that PR should be recommended for obese COPD patients, based on current evidence.57

In respiratory disorders other than COPD, in all of which obesity may or may not be present as a co-morbidity, no formal evidence-based guidelines exist regarding exercise prescription or response to exercise training.1 This is why PR recommendations for these pathological conditions solely rely on expert opinions that are based on knowledge of the underlying pathophysiology and clinical experience.1 The following respiratory disorders other than COPD represent indications for PR: asthma, cystic fibrosis, bronchiectases (non-cystic fibrosis), neuromuscular disorders, interstitial lung diseases, pulmonary hypertension and patients before and after surgical intervention (e.g. lung transplantation).1,58,59

Although obesity may occur in all of these respiratory disorders, several of them are known to be associated—at least in the long term—with low bodyweight or even cachexia (e.g. cystic fibrosis, neuromuscular disorders, interstitial lung diseases). Furthermore, obese patients following lung transplantation are rarely seen in PR because obesity represents a relative contraindication for this procedure.60,61 This might be different for pre-intervention patients in which weight loss might be one of the primary aims of PR programmes and obesity might be commonly observed.

Until today, there is to the best of our knowledge, no conclusive evidence published about whether there is any impact of obesity on PR outcome when PR is indicated for conditions other than COPD. In this respect, it is well worth mentioning that even the most recent systematic reviews on physical training for cystic fibrosis, bronchiectasis, asthma and interstitial lung diseases do not provide information on the impact of overweight/obesity on physical training.62–65

OHS, while not formally listed as an indication for PR, might represent another important candidate for PR programmes due to its high prevalence.30,31 However, it has to be taken into account that OHS patients seem to be poorly motivated to participate in rehabilitation programmes, mainly due to the long travel distances required to reach the rehabilitation centre.66 Furthermore, despite the offer of travel assistance, 41% of these patients would still not attend.66

In general, low-impact exercise and/or water-based exercise is recommended for severely obese subjects.1 Furthermore, PR is supposed to be an ideal setting in which the needs of obese patients with respiratory disorders can be addressed, especially if the obesity itself is contributing to functional limitation.1

Non-Pulmonary rehabilitation

Apart from PR, several other types of rehabilitation programmes (e.g. the so-called ‘interdisciplinary’ rehabilitation programmes) exist which share the inclusion of patients with pulmonary pathologies in combination with overweight/obesity and are therefore discussed in the following section. However, it is not the aim of this review to explicitly discuss the extensive data available on non-PR programmes for obese subjects lacking pulmonary pathologies.

One study reported that among patients undergoing cardiac rehabilitation, the prevalence of smoking ranged from 33% in the non-obese to 48% in the morbidly obese, thus suggesting a relevant fraction of undiagnosed obstructive ventilation disorders in this cohort.67 Here, the prevalence of overweight/obesity (BMI >25 kg/m2) was as high as 88% (397 of the 449 patients investigated).67 After 10 weeks of exercise training based on heart rate reserve (range 45–85%), all groups (normal weight, overweight and obese) improved their exercise capacity, as indicated by significantly increased metabolic equivalent.67

In an observational study investigating the effects of a 1-month, hospital-based, interdisciplinary rehabilitation programme in obese patients suffering from sleep disturbances,68 improvements in the 6MWD and quality of life were achieved by the end of the programme and remained constant for the following 6 months.68

Among obese subjects undergoing a home-based physical fitness programme (walking and skeletal muscle resistance training), the prevalence of sleep apnoea was 35% (18 of the 52 investigated patients) while the smoking prevalence was 10%.69 The programme resulted in improved cardiopulmonary fitness variables (2-km walking time and heart rate reserve).69

In a cohort of 1019 obese children, the prevalence of asthma was 10% (105 of the 1019 investigated children).70 Obese children showed impaired quality of life compared with normal-weight controls, but this was improved by an inpatient rehabilitation programme.70

These data imply that pulmonary co-morbidities play a major role in obese patients undergoing rehabilitation programmes (inpatient and outpatient settings) for different indications. It has been shown that rehabilitation programmes in this cohort are able to improve exercise capacity and health-related quality of life.

Pulmonary rehabilitation potential in obesity

In theory, obesity can have either a negative, a positive or even no influence at all on PR potential. As outlined earlier, several studies have investigated the effect of PR programmes in obese patients and reported positive effects after a particular rehabilitation programme.56,68–72 However, all these studies share one common limitation; namely, they did not compare obese subjects with normal-weight controls. For this reason, it is not possible to determine the impact of obesity per se on PR from this data.

Nonetheless, evidence from the few controlled studies that compare obese with non-obese subjects suggest that obesity does not have a negative impact on rehabilitation outcome in a cohort of patients with cardiac disease and pulmonary co-morbidities.67 This also holds true for PR outcome in obese versus non-obese COPD patients.53,54


Weight loss

Because obesity is associated with several pathophysiological changes during exercise (see earlier), it could be assumed that weight loss is capable of positively influencing these changes. There are several approved strategies for inducing weight loss in the obese, namely bariatric surgery, a low- to very low-calorie diet, physical activity programmes and/or the combination of these strategies. In this context, a relatively recent study has shown that dietary-induced weight loss alone is not capable of improving exercise capacity in obese children,73 but rather needs to be supplemented with exercise training in order to promote improvements in cardiorespiratory fitness (i.e. peak oxygen consumption) and ventilatory efficiency.73

A 3-week, in-hospital bodyweight reduction programme including a low-calorie diet, tailored aerobic-strength exercise, psychological counselling and nutritional education was shown to result in significant weight loss, which remained constant in an 11-month follow-up visit and was suggested to possibly improve physical performance.72 This supports the current recommendation that specific interventions (e.g. nutritional education, restricted calorie meal planning, encouragement for weight loss and psychological support) should be performed in obese subjects during PR.1

Interestingly, otherwise healthy morbidly obese subjects who lost weight after a 6-week very low-calorie diet and behavioural intervention demonstrated a reduction in alveolar-arterial oxygen difference during exercise.74 This effect in the morbidly obese also extended to resting conditions, where weight loss achieved by bariatric surgery was associated with improved gas exchange, comprising a reduction in alveolar-arterial oxygen difference and improved arterial partial pressure of oxygen and arterial partial pressure of carbon dioxide.6 Moreover, weight loss induced by bariatric surgery leads to the adequate restoration of compensatory hyperventilation and improves gas exchange at moderate to peak exercise intensities in the obese and morbidly obese.75

Finally, it was shown in nine obese men that modest weight loss achieved by a combined diet and exercise programme over 12 weeks led to an increase in end-expiratory lung volume and a decrease in gastric pressure at ventilatory threshold during exercise.76 It was concluded that weight loss promotes the amelioration of breathing mechanics during submaximal exercise in the obese; such improvements appear to be related to the cumulative loss of chest-wall fat rather than to changes in any specific region of fat deposition (chest, subcutaneous abdominal or visceral fat).76

Respiratory muscle training

A recent, controlled study used voluntary normocapnic hyperpnea (30 min, 4 days per week) to investigate the effect of respiratory muscle endurance training (RMET) on respiratory muscle capacity and exercise performance in 20 morbidly obese subjects (BMI 45 ± 7 kg/m2) who were undergoing a low-calorie diet and physical activity programme comprising outdoor walking and low-intensity cycle ergometry.77

While RMET did not improve respiratory muscle strength, there was an increase in breathing duration and a greater reduction in dyspnoea both during the respiratory muscle endurance test and at the maximal ventilatory level (sustained for at least 3 min).77 In addition, RMET significantly improved the 6MWD and quality of life whilst simultaneously reducing dyspnoea-related effort limitation, whereas no such improvements were observed in the non-RMET group.77 It was also speculated that RMET potentially prevents respiratory muscle fatigue during exercise, and that exercise tolerance in the obese might be improved by reducing dyspnoea, which, in turn, was suggested to underlie the observed improvement in quality of life.77 It was therefore concluded that application of RMET in the obese and morbidly obese is a promising tool for the promotion of improvements in functional capacity and adherence to physical activities.77 However, further research is needed to conclusively evaluate the effects of the different types of respiratory muscle training in obese subjects.

Non-invasive mechanical ventilation

There is, to the best of our knowledge, only one pre-existing study that addresses the impact of non-invasive ventilation (NIV) applied during exercise in obese subjects.78 This study reported a prolongation in exercise duration and significant reduction in dyspnoea following NIV application.78 Another study investigating the impact of NIV applied during exercise focused on a cohort of severely ill COPD patients, 33% of whom were also obese.79 Here, non-invasive ventilation led to improvements in oxygenation, exercise capacity (6MWD) and dyspnoea.79 For these reasons, NIV applied during exercise might be a useful adjunct during PR in the obese.

Another use for NIV is in the nocturnal treatment of chronic hypercapnic respiratory failure in COPD patients undergoing PR.80,81 NIV has been shown to augment the benefits of PR by improving health-related quality of life, functional status and gas exchange.80,81 Even though mean BMI was around 27 kg/m2 in these particular studies, it can be deduced from the given standard deviations of ∼6 kg/m2 that at least several of the investigated patients were obese.80,81 Furthermore, NIV represents an important therapeutic strategy in OHS patients, as outlined earlier in this review series.82 Interestingly, there is an ongoing trial that is comparing the impact of a combined NIV/PR programme to NIV alone on weight loss and activity levels in OHS patients.83


The impact of obesity on exercise performance is a complex process that mainly comprises: (i) the attenuation of resting alterations in lung volumes and gas exchange; (ii) increases in oxygen cost of breathing and dyspnoea; and (iii) the impairment of respiratory muscle function. Absolute values of maximal oxygen uptake are similar between obese and normal-weight subjects, whereas oxygen uptake at a given workload is increased and maximal workload is reduced in the obese, likely due to the increased body mass and/or increased basal rate.

Based on the evidence that is currently available, obesity per se does not have a negative impact on PR outcome in COPD patients. The impact of obesity on PR in respiratory disorders other than COPD remains to be conclusively investigated and is thus a worthy candidate for future research.


The authors would like to thank Dr S. Walterspacher, Dr D.J. Walker, D. Langer, Professor R. Gosselink and Professor J.A. Dempsey for their support during the literature search and for helpful comments on the manuscript prior to submission. Dr S. Dieni is acknowledged for assisting with the manuscript preparation.