The literature on the physiological response (vasodilation, neuromuscular fatigue, and muscle oxygenation) following the application of different dosages of oxygen therapy in patients with Chronic Respiratory Failure (CRF) and Long-Term Oxygen Therapy (LTOT) during exercise is scant. The evaluation of these aspects can allow the clinicians and the rehabilitation staff to correctly dose the oxygen therapy at rest and during exercise and to reach a higher level of improvement after training. For this purpose, we will recruit 20 patients admitted to the Pulmonary Unit of the ICS Maugeri in Lumezzane (BS) with the presence of CRF defined as PaO2 at room air less than 60 mmHg, the need for LTOT since 3 months, and with a stable clinical condition. This is a crossover study and will last 3 days. We will test the same subject, randomly, in the following three conditions: A) CONDITION ROOM AIR: patient will breathe room air through the Venturi mask (Vmask FiO2 21%) and will be considered as "sham condition" B) CONDITION FiO2 30%: the subject will breathe through a Venturi mask with a FiO2 of 30%. C) CONDITION FiO2 60%: the subject will breathe through a Venturi mask with a FiO2 of 60%. During each condition, we will evaluate: a) oxygen saturation (SatO2), transcutaneous paCO2 value (tcCO2), BORG fatigue and dyspnea, blood gas analysis; b) mitochondrial function through the Near Infra-Red Spectroscopy and c) vascular function by Single Passive Leg Movement (sPLM) technique; d) central and peripheral neuromuscular fatigue after a submaximal intermittent isometric contraction. The present project will help to understand the best doses of oxygen therapy to allow patients to achieve a higher level of vasodilation and mitochondrial function and a lower level of neuromuscular fatigue. We could apply these results to the rehabilitation program in order to get a greater level of improvement in exercise tolerance.

Advanced Chronic Obstructive Pulmonary Disease (COPD) with Chronic Respiratory Failure (CRF) and long-term oxygen therapy (LTOT) need is a condition with a poor prognosis that causes symptoms such as dyspnea and fatigue that dramatically reduce the quality of life of the person with COPD. Typically, the advanced phase of COPD is characterized by a fluctuating pattern with recurrent hospitalizations, and a vicious cycle in which dyspnea increases and exercise tolerance and physical activity are reduced, which in turn lead to depression, social isolation, and low quality of life and increased risk of death.

Muscle dysfunction in these patients contributes, together with dynamic hyperinflation, to the increase of fatigue and dyspnea during exercise, leading to the early cessation of effort and the decrease of maximum aerobic capacity. In patients with CRF, the increase of arterial oxygenation could have a beneficial result through the direct inhibition of the stimulation of the carotid body receptors by reducing ventilation and respiratory rate.

The increase in arterial oxygenation could also promote the increase in muscle oxygenation by i) reducing the production of lactates during efforts at iso-work and ii) increasing the pulmonary vasodilation and consequently the cardiac output and therefore oxygen delivery to the exercising muscle. Interestingly, the increase in muscle oxygenation is likely delaying the onset of diaphragm fatigue.

In the clinical setting, studies have shown the acute administration of oxygen to be useful in reducing the ventilatory demand of the COPD patient by reducing minute ventilation and respiratory rate and improving exercise tolerance. In COPD patients with and without chronic respiratory failure, acute oxygen administration appears to improve endurance capacity, maximal exercise capacity, dyspnea, and minute ventilation. In patients with moderate/severe non-hypoxic COPD at rest and during exercise, training with oxygen supplementation provides greater benefit to exercise tolerance and respiratory pattern. Instead, in patients with normoxia at rest and exercise-induced desaturation, the exercise tolerance evaluated with the six-minute walking test progressively improves with the addition of oxygen, although there is enormous variability within the individual groups.

In this population, the addition of oxygen during exercise leads to marginal effects linked only to a slight benefit on dyspnea. Therefore, to date, this treatment regimen only aims at patients with exercise-induced desaturation with associated severe dyspnea. Therefore, there is little support from the literature in offering oxygen therapy extensively during physical training to patients with COPD. Besides, there is a lack of solid studies that lead to firm conclusions on the use of oxygen therapy with particular reference to the benefits on functional outcomes, symptoms and quality of life.

While most of the studies concern normoxic patients with COPD at rest, there is little or no literature devoted to COPD with CRF and LTOT. To our knowledge, there is a lack of physiological studies that investigate the response (vasodilation, neuromuscular fatigue and muscle oxygenation) of the application of different dosages of oxygen therapy in patients with COPD associated with CRF and LTOT.

This information would allow to better describe the origin of effort intolerance and guide the clinician on the most appropriate oxygen therapy dosage to obtain the best physiological response to exercise.

PROTOCOL In this study, neuromuscular fatigue at three different FiO2 concentrations will be evaluated.

The study will take place over 5 days Day 1: familiarization Day 2: exhaustion test (Tlim) of submaximal isometric contractions at 30% MVC with a repeated cycle of 3" ON and 3" OFF will be performed Day 3,4 and 5 pre and post-fatigue neuromuscular assessments will be performed at different FiO2. The fatigue protocol will involve performing submaximal isometric contractions intervalled for a duration equal to 80% of the Tlim. Each condition was evaluated after stabilization of at least 10 minutes.

The days will be performed in a randomized trial.

The three conditions vary by FiO2 as follows:

A) CONDITION ROOM AIR: patient will breathe room air through the Venturi mask (Vmask FiO2 21%) and will be considered a "sham condition" B) CONDITION FiO2 30%: the subject will breathe through a Venturi mask with a FiO2 of 30%.

C) CONDITION FiO2 60%: the subject will breathe through a Venturi mask with a FiO2 of 60%.

MEASURES At the beginning of the protocol (T0), anthropometric measurements (BMI), comorbidities measured with CIRS scale, anamnestic fatigue evaluated by the Fatigue Severity Scale (FSS), anamnestic dyspnea with Barthel scale dyspnea, Quality of life by CAT and MRF scores, blood gas analysis value in air, lung function by global spirometry

1. Evaluation of central and peripheral neuromuscular fatigue according to interpolated twitch protocol. Subjects will be seated upright with back support. The hip and knee will be flexed to 90° and the force will be measured by a force transducer.

   Evaluation of the isometric force: Maximum Voluntarily Contraction (MVC). Electromyographic evaluation: M waves will be recorded from the vastus lateral. Evaluation of the electrically stimulated resting force (Qtpot) and of the maximum voluntary activation (MVA). The electrical stimulation used will consist of single square wave pulses of 0.1 ms duration, delivered by a constant current stimulator (DS7AH, Digitimer). The intensity of the stimulus used will be defined as follows: the current will be progressively increased from 0 mA to the value beyond which there will be no further increase in the amplitude of the M wave. The stimulus used for the study will be set at 125% of the intensity required to produce a maximum M wave response. Qtpot will be measured 5 seconds after MCV measurements. The MVA will be calculated during the electrically stimulated MVC: a single contraction superimposed on the MVC will be compared with the force produced during Qtpot. The protocol provides for the repetition of 2 measurements of MVC and Qtpot before and after the fatiguing task.

2. Measurement of pulsed oxygen saturation (SatO2), the transcutaneous paCO2 value (tcCO2), BORG fatigue and dyspnea.

3. Evaluation of mitochondrial function in vivo through the Near InfraRed Spectroscopy (NIRS) method by applying a non-invasive probe on vastus lateralis. The muscle oxidative capacity test will measure the relative concentration of deoxy-haemoglobin and oxyhemoglobin in tissues. The total haemoglobin (THb = HHb + HbO2), and the Hb difference (Hbdiff = HbO2 - HHb) will be obtained as derived measures.

4. Assessment of sternocleidomastoid accessory respiratory muscle fatigue by EMG

5. The evaluation of the vascular function will be performed with a Doppler ultrasound at the level of the common femoral artery, in basal conditions and during the application of the Single Passive Leg Movement (sPLM) technique. The sPLM test will be performed on the right common femoral artery, and measurements will be made using a Doppler ultrasound system (Logiq V4-GE, Milwaukee, WI, USA). The sPLM protocol will consist of 60 seconds of baseline data collection at rest, followed by a 1-second passive flexion-extension of the leg. The leg will then be kept fully extended for the remaining 60 s after the movement. For each subject the arterial diameter at rest, the blood flow at rest, the relative changes will be determined (Dpeak) from rest, the peak blood flow and the Area Under the Curve (AUC) of the femoral blood flow during the evaluation were collected. The peak blood flow values, relative changes from rest and AUC after leg movement will be calculated second by second

Inclusion Criteria:

• PaO2 in room air less than 60 mmHg assessed by arterial blood gas analysis
• FEV1/FVC <70%
• FEV1 < 50% of predicted
• need for LTOT for 3 months
• important non-comorbidities

Exclusion Criteria:

• Presence of lung diseases other than COPD, respiratory tract infections in the last 4 weeks, terminality, severe neurological and cardiologic comorbidities.