Chronic Obstructive Pulmonary Disease (COPD) is a chronic disease with related exercise intolerance and marked disability due to symptoms such as dyspnea and fatigue. Effort intolerance and exercise-induced symptoms cause marked impairment in completing activities of daily living (ADL). Pulmonary rehabilitation (PR), which has exercise as a major component, is considered a key treatment in the management of COPD since PR is effective in improving exercise tolerance, exercise-induced dyspnea and fatigue, and health-related quality of life. Rehabilitation is also effective in improving the time required to perform ADLs, reducing symptoms and disability. Studies show that rehabilitation protocols with upper limb exercises added to lower limb training are able to give additional benefits in terms of effort tolerance (endurance time at the arm ergometer and oxygen consumption) and reduction of dyspnea at iso-load.

The primary aim of this study is to evaluate whether the combined "arm and leg" training modality, compared to a gold standard protocol -involving only the lower limbs training- is more effective in improving ADL performance in terms of reduction of exercise time for a specific test (GLITTRE test).

Background:

Chronic Obstructive Pulmonary Disease (COPD) is a chronic disease with related exercise intolerance and marked disability due to symptoms such as dyspnea and fatigue. Effort intolerance and exercise-induced symptoms cause marked impairment in completing activities of daily living (ADL). COPD patients often rely on a caregiver to perform some ADLs for intolerable dyspnea and fatigue since the early stages of the disease. In fact, even while performing simple ADLs, COPD patients develop dynamic hyperinflation and reach a ventilatory limit that forces them to stop or slow down. Symptoms and disability worsen the patient's quality of life.

Pulmonary rehabilitation (PR), which has exercise as a major component, is considered a key treatment in the management of COPD since PR is effective in improving exercise tolerance, exercise-induced dyspnea, and fatigue, and health-related quality of life. Rehabilitation is also effective in improving the time required to perform ADLs, reducing symptoms and disability.

Lower extremity cycle ergometer training improves peak oxygen uptake, and reduces heart rate, minute ventilation, dyspnea, and leg fatigue at submaximal work speeds. Moreover, upper limb training has been described to reduce dyspnea, increase respiratory muscle strength, and have an effect on patients' global exercise tolerance.

Studies show that rehabilitation protocols with upper limb exercises added to lower limb training are able to give additional benefits in terms of effort tolerance (endurance time at the arm ergometer and oxygen consumption) and reduction of dyspnea at iso-load.

To our knowledge, no study has investigated training programs in COPD patients with similar timing of intervention comparing selective lower extremity endurance training with combined upper and lower extremity training.

The primary aim of this study is to evaluate whether the combined "arm and leg" training modality, compared to a gold standard protocol -involving only the lower limbs training- is more effective in improving ADL performance in terms of reduction of exercise time for a specific test (GLITTRE test). The secondary aims are to evaluate whether the combined training modality, compared to a gold standard protocol involving only the lower limbs, is more effective in a) increasing the effort tolerance of the lower limbs; b) increasing the effort tolerance of the upper limbs and c) reducing the energy cost of walking.

Methods:

This is a multicenter RCT study with a low-risk intervention.

Patients with a diagnosis of COPD, according to the Global Initiative for Lung Disease (GOLD) criteria, will be recruited for this purpose and randomly assigned to two groups: Lower limbs (LL) group and lower plus upper limbs (L+UL) group. The single list of randomization was created by a person external to the healthcare staff.

The treatment in the two groups is reported in the intervention section. All training sessions will be supervised by a physiotherapist. The initial training intensity both for lower and upper limbs will be determined on the basis of the load value performed in the last step of the maximal incremental effort test; the load value will be equal to 60% of the maximal wattage observed. Subsequently, the training intensity will be gradually increased during the sessions with a progression based on the symptoms, according to the protocol of Maltais et al. with steps of 10 watts for the lower limbs and 5 watts for the upper limbs.

During the rehabilitation period, formal education (alone or in groups) concerning the knowledge and management of the disease, healthy lifestyle, airway clearance, and inhaled drugs, will also be offered to all patients. Where indicated, airway clearance techniques and/or adaptation to non-invasive home ventilation will be administered.

At baseline, the investigators will collect clinical evaluations as follows:

• Demographic and anthropometric data
• Post-bronchodilator spirometry and plethysmography (Forced Expiratory Volume at 1 second, Forced Vital Capacity, Residual Volume, Inspiratory Capacity)
• Blood gas analysis in ambient air (PaO2, PaCO2, pH)
• Cumulative Illness Rating Score (CIRS)

Pre and post evaluations will be:

• Barthel Index
• EuroQoL

The Outcome measures are reported in the dedicated section.

The sample size has been calculated considering an improvement of the LL group of 23±20 seconds on the GLITTRE test corresponding to the MCID (Minimal Clinical Important Difference) for a standard GOLD rehabilitation treatment using a T-test for unpaired data with a= 0.05, b= 0.80. Estimating that patients in the L+UL group improve 22±20 seconds more than the LL group, the estimated sample is 30 patients (15 per group). Furthermore, considering a drop-out rate of 20%, it is estimated that 36 patients will be enrolled (12 patients per center).

The statistical evaluation will be conducted with the STATA 11 program (StataCorp LLC). Continuous variables will be expressed as mean and standard deviation (or median and quartiles) while categorical or binary variables as a percentage.

Statistical evaluation will be conducted with the use of appropriate parametric and non-parametric tests to verify differences between groups in treatment response (T-test or Wilcoxon to evaluate the difference between means or medians and chi-square test to evaluate the difference between frequencies). A secondary evaluation by logistic regression will also be performed to verify the baseline characteristics predictive of improvement in responders. A p value <.05 will be considered significant.

Inclusion Criteria:

• GOLD class 2-3 COPD
• Forced Expiratory Volume in the first second (FEV1) between 30% and 70% of the predicted value
• ability to walk and climb stairs without assistance
• stable clinical condition (pH > 7.35)

Exclusion Criteria:

• chronic respiratory insufficiency on long-term oxygen therapy (LTOT)
• severe orthopedic, neurological or cardiological comorbidities
• cognitive impairment
• recent exacerbation (within 15 days) requiring a change in therapy
• presence of lung disease other than COPD
• terminality