By analyzing energetic and biomechanical basis of walking, and the subsequent changes induced by hypoxic vs normoxic training in obese individuals, it may optimize the use of walking in hypoxia to gain perspective for exercise prescription to set up training programs that aim to induce negative energy balance and to deal with weight management. However to the investigators knowledge, the analysis of changes in mechanics, energetics and efficiency of walking after continuous hypoxic training (CHT) has not been performed yet.

The aims of the present study were:

1. Comparing the changes in body composition between continuous hypoxic training (CHT) and similar training in normoxia; e.g. continuous normoxic training (CNT) in obese subjects.
2. Comparing the metabolic and energetics adaptations to CHT vs CNT.
3. Finally, comparing the associated body-loss induced gait modification since walking intensity at spontaneous walking speed (Ss) is lower in CHT than in CNT.

Hypoxic training embraces different methods as "live high - train high" (LHTH) and "live high - train low" (LHTL); sleeping at altitude to gain the hematologic adaptations (increased erythrocyte volume) but training at sea level to maximize performance (maintenance of sea level training intensity and oxygen flux). The LHTL method can be accomplished via a number of methods and devices: natural/terrestrial altitude, nitrogen dilution, oxygen filtration and supplemental oxygen. Another method is the "live low - train high" (LLTH) method including intermittent hypoxic exposure at rest (IHE) or during intermittent hypoxic training sessions (IHT). Noteworthy, all supporting references were conducted with endurance elite athletes (i.e. cyclists, triathletes, cross-country skiers, runners, swimmers, kayakers, rowers) and there is an extensive literature relative both on LHTH and LHTL (Millet et al. 2010). Interestingly, very recently, the investigators research group proposed a new LLTH method (e.g. repeated sprints training in hypoxia; RSH) for team-sports players (Faiss et al. 2013). This lead to modify the nomenclature (Millet et al. 2013) and to divide LLTH method in four subsets; i.e. IHE, CHT (continuous >30 min low intensity training in hypoxia), IHT (interval-training in hypoxia) and RSH, based predominantly on different mechanisms; e.g. increased oxidative capacity (CHT), buffering capacity (IHT) or compensatory fiber-selective vasodilation (RSH). These new nomenclature and hypoxic methods open doors for investigating the use of different LLTH methods with other groups and for other purposes than the oxygen transport enhancement.

Several recent findings support the use of LLTH in obese subjects in terms of weight loss and/or cardiovascular and metabolic improvements (Kayser and Verges 2013). CHT [low intensity endurance exercise for 90 min at 60% of the heart rate at maximum aerobic capacity, 3 d week-1 for 8 weeks; fraction of inspired oxygen (FiO2) = 15%] in overweight subjects [body mass index (BMI) > 27] lead to larger (+1.1 kg) weight loss than similar training in normoxia. However, no difference was observed regarding BMI between the training modalities (Netzer et al. 2008). In a similar way, CHT (low intensity endurance exercise for 60 min at 65% of the heart rate at maximum aerobic capacity, 3 d week-1 for 4 weeks; FiO2 = 15%) induced similar increases in maximal oxygen consumption and endurance but larger improvements in respiratory quotient and lactate at the anaerobic threshold as well as in body composition than similar training in normoxia (Wiesner et al. 2010). Of interest is that the beneficial results were obtained despite lower training workload in hypoxia. This suggests that hypoxic training intensity can be lower in absolute value, at the spontaneous walking speed (Ss), also known as preferred or self-selected speed (e.g. the speed normally used during daily living activities). This appears to be an appropriate walking intensity for weight reduction programs aimed at inducing negative energy balance (Hills et al. 2006). A lower walking intensity is also likely more protective of the muscles/joints in obese patients with orthopaedic comorbidities. Finally, CHT was also shown (Haufe et al. 2008) to lead to larger change in body fat content, triglycerides, homeostasis assessment of insulin resistance (HOMA-Index), fasting insulin and area under the curve for insulin during an oral glucose tolerance test despite the lower absolute running intensity (1.4 and 1.7 W kg-1 in hypoxia and normoxia, respectively).

The net energy cost of level walking (NCw) represents the energy expenditure per distance unit only associated with walking movements. Previous studies reported higher absolute (J·m-1) and relative (i.e., normalized by body mass: J·kg-1·m-1) NCw in obese compared with normal body mass individuals (Browning et al. 2006; Peyrot et al. 2009), suggesting that the body mass is the main, but not the only, determinant of this lower economy of walking in obese subjects and that other factors may be involved in the higher NCw in these individuals(Browning et al. 2006; Peyrot et al. 2010; Peyrot et al. 2009). If body mass loss is an important method for the treatment of obesity and its associated co-morbidities and it may also be an important to investigate the effect of decreased body mass on gait pattern and mechanical external work (Wext) and their consequences on NCw in obese individuals. Walking is a fundamental movement pattern and the most common mode of physical activity. This form of locomotion may contribute significantly to weight management in overweight and obese subjects (Hill and Peters 1998; Jakicic et al. 2003; Pollock et al. 1971). Only one study showed that body mass reduction of 7% over 3 months resulted in gait kinematic changes (i.e., increases in walking speed, stride length and frequency, swing duration and decrease in cycle time, stance and double support time) in healthy adult obese women (BMI = 37 kg·m-2) (Plewa et al. 2007). However, these authors did not measure the NCw. More recently, Peyrot et al. (Peyrot et al. 2010) reported that, in healthy adolescent obese individuals, a 12-wk voluntary body mass reduction program (-6%) induced a reduction in NCw mainly associated with decreased body mass but also with changes in the biomechanical parameters of walking [i.e., a lesser lower limb muscle work required to rise the center of mass (CM) with Wext unchanged after intervention]. The authors hypothesized that the relation between the changes in absolute NCw and the changes in the biomechanical parameters might be explained by an increase in efficiency of muscle mechanical work with body mass loss as previously showed in cycling (Rosenbaum et al. 2003). Others studies (Messier et al. 2005; Messier et al. 2011), investigating only the effect of body mass loss (-3% and -10%, respectively) on biomechanical parameters of walking in non-healthy overweight and obese older adults with knee osteoarthritis, demonstrated that this body mass loss increased walking speed and reduced knee joint forces. Bariatric surgery may induce greater body mass loss (~30-40%) (Chaston et al. 2007) compared with exercise, diet or pharmaceutical interventions (~10%) (Franz et al. 2007) and may be considered as an interesting tool to maximize the effect of body mass loss on Wext and NCw in obese individuals and, thus, investigate the relationship between the gait pattern changes and the extra cost of walking in these subjects. Similarly, it would be of interest to investigate how the metabolic changes and body mass loss induced by CHT, potentially associated with an increased metabolic efficiency, would affect gait pattern and the extra cost of walking in obese subjects.

Inclusion Criteria:

• Healthy and free of clinically significant orthopaedic, neurological, cardiovascular or respiratory conditions.
• BMI > 30 kg/m^2.
• Age > 18 yr.

Exclusion Criteria:

• Age > 40 yr.
• BMI < 35 kg/m^2.
• Diabetes.
• Neurological disorders, orthopaedic injury, history of falls and medications that provoke dizziness.