Patients with chronic obstructive pulmonary disease (COPD) often experience symptoms of dyspnea and fatigue during daily life activities. The avoidance of these activities induces an inactive lifestyle which is associated with skeletal muscle weakness and decreased exercise tolerance. Comprehensive multidisciplinary pulmonary rehabilitation is a key intervention in symptomatic patients leading to enhanced exercise tolerance and improved health-related quality of life. While pulmonary rehabilitation has established the status of evidence-based treatment in patients with stable COPD, the present thesis has focussed on three areas where uncertainty remains: 1) the implementation of rehabilitation programs in non-stable patients, 2) the reasons why some patients are responding better than others to exercise training and 3) the paradigm of changing physical activity behaviour as a spin-off to the improvement of exercise tolerance and symptoms. Hospitalizations for an exacerbation of COPD are associated with prolonged inactivity and muscle weakness. In a systematic review we describe the systemic consequences of (repeated) hospitalizations. Furthermore, the scientific evidence about a range of interventions during and after the exacerbation, including exercise training, nutritional interventions and self-management strategies, is thoroughly discussed. The most important conclusion is that exercise training is safe, feasible and effective both during and after hospitalization for an exacerbation, if the modality is adapted to the compromised respiratory situation. In severe cases, exacerbations can lead to admission at the intensive care unit (ICU) and mechanical ventilation. The associated immobility leads to detrimental muscle wasting and ICU-related myopathy and polyneuropathy are highly prevalent in patients with a prolonged ICU stay. Consequently, patients report poor functional status several years after the ICU admission. We investigated the feasibility and effectiveness of a daily passive or active bedside cycling session, initiated at the moment that patients became cardiorespiratory stable. Most patients were sedated and mechanical ventilated at the start of the intervention. We showed that daily bedside cycling during ICU stay is safe in selected patients and leads to an enhanced recovery of muscle strength and functional exercise capacity at the moment of hospital discharge.Although the beneficial effects of pulmonary rehabilitation have been well established at a group level, individual patients can show less favorable results. It has been shown that patients with more pronounced deconditioning and a higher ventilatory reserve generally have better effects after rehabilitation, but the explained variability remains low. The development of contractile muscle fatigue during training might be a marker of adequate overload to the muscle. This way it is important that patients can fatigue their muscles in order to obtain significant physiological training effects. We reported that approximately 60% of patients develop quadriceps fatigue during high-intensity exercise training. It was impossible to predict whether a patient would develop muscle fatigue based on lung function, baseline muscle strength, exercise tolerance or any other baseline characteristics. Furthermore, training intensity was similar in patients with and without fatigue. Muscle fatigability decreased during the course of the rehabilitation program. Patients who developed muscle fatigue during training had more pronounced increases in six-minute walking distance and health-related quality of life. The difference between patients with and without fatigue clearly exceeded the proposed minimal important difference for these measures, indicating the importance of developing quadriceps contractile fatigue during exercise training.Studies investigating the influence of pulmonary rehabilitation on physical activity levels report unequivocal results. It is clear that beneficial effects in terms of exercise tolerance do not automatically translate into a higher level of daily physical activity. The lack of change in physical activity levels likely impacts on the long-term maintenance of training effects. We found no changes in daily physical activity behaviour after a six-month pulmonary rehabilitation program, despite a significant increase in exercise tolerance. Remarkably, the addition of an individual activity counseling program, based on techniques of motivational interviewing, did not enhance the physical activity levels. These findings indicate that it is a clear challenge to change the physical activity behavior of patients with severe COPD and that this topic requires further attention in the near future.