Design
This study is a randomized controlled clinical trial with two parallel arms involving 40 obese adolescents. To assess the effectiveness of four weeks of intermittent hypoxia added to a traditional exercise and diet intervention on inducing short (after the four-week intervention) and longterm (at two months follow-up) weight loss in obese adolescents, we will recruit 20 adolescent boys and 20 adolescent girls (aged 11 to 15 years) from our summer weight loss camp. They will be stratified according to gender and randomly assigned to two groups: the control group who will exercise, eat a balanced diet, and sleep in a normobaric condition, and the hypoxia group, who will exercise, eat a balanced diet, and sleep in normobaric hypoxia chambers (‘sleep high and train low’). Outcome assessment and data analysis will be performed by trained professionals who will be blinded to the group assignment of subjects. The flow of participants through the trial is shown in Figure 1.
Sample size estimation
Sample size estimation in this RCT is based on the expected fat loss following four weeks of hypoxia exposure plus exercise and dieting. The data of our preliminary experiment showed that the means of fat percentage decrease in the normoxia and hypoxia groups were 3.1 and 6.0% respectively, and the standard deviation of the change was about 3%. It is estimated that a sample size of 17 participants per group will be required to observe a similar result with a power of 80%. Considering a 15% dropout and exit rate, we will recruit 40 subjects with 20 in each group.
Ethical approval and consent
The study will be conducted according to the principles expressed in the Declaration of Helsinki. The study protocol has been approved by the institutional review board at the Shanghai University of Sport (reference number: 2014 Ethics Approval Note 1). Participants in this study are volunteers. None of the measurements or the intervention are known to entail any significant health risk. The study has its own physician to ensure the eligibility and safety of all participants. All data will be handled and archived confidentially. The benefits and associated risks of the study will be carefully explained and the voluntary nature of participation will be emphasized. Informed consent and assent will be obtained from all participants and their parents or legal guardians. Participants and their parents or legal guardianswill have the option to end the participation at any stage if they so wish. If the physician and the principal investigator believe that there are risks of serious adverse events in the study the trial will be stopped. The trial is registered with Chinese Clinical Trial Registry as ChiCTR-TRC-14004106.
Subject recruitment
Obese adolescents aged 11 to 15 years will be recruited from the children registered for the 2014 summer weight loss camp at the Shanghai University of Sport. A public health nurse will assess the Tanner stage of each subject using the Tanner grading system [25, 26]. Obesity will be defined based on body-mass index (BMI), calculated as weight in kilograms divided by height in meters squared (kg/m2). Although a BMI of ≥ 25 kg/m2 and ≥ 30 kg/m2 are international cutoff points for overweight and obesity, for many Asian populations individuals with a BMI of ≥ 23 kg/m2 are considered to be at increased risk and those with a BMI of ≥ 27.5 kg/m2 at high risk [27]. Thus, we chose a BMI of ≥ 25 kg/m2 as the criterion to recruit obese adolescents. Adolescents will be excluded if they have concomitant renal, hepatic, or cardiac disease, and/or are being treated with drugs that could affect body weight and appetite (such as orlistat, lorcaserin, and phentermine-topiramate, as well as appetite suppressants).
Randomization and blinding
After a participant is confirmed to be eligible and written informed consent has been obtained, she/he will be randomly assigned to the hypoxia or normoxia group. The randomization procedure, stratified according to gender, will be conducted by an independent statistician using a computerized randomization program. In order to minimize the potential bias, the exercise physiologist and the dietitian who manage the exercise training and the diet intervention will be blinded as to whether a subject will sleep in the hypoxia chamber or not. The hypoxia chamber will be prepared by an independent researcher. The researcher performing outcome assessments will be blinded to the subjects’ intervention allocation.
Intervention
All the subjects, including both the normoxia (control) and the hypoxia groups, will undergo four weeks of aerobic exercise training and dieting (eating a balanced diet). In addition, the normoxia group will sleep in normal conditions while the hypoxia group will sleep in a normobaric hypoxic chamber every night. All the measurements will be conducted two days prior to, and post-intervention, and repeated two months later after the intervention (two months follow-up). Fasting blood samples will be collected in the morning of the testing day. In order to maximize the compliance and avoid the occurrence of any accident the intervention will be closely supervised by a physician, a dietitian, and an exercise physiologist.
Exercise training
The same aerobic exercise training will be applied to both the normoxia and the hypoxia groups. Participants will exercise for six days per week, twice daily, for one hour per session. The intensity of the exercise will be estimated using the Metabolic Equivalent of Task (MET) score. MET, the unit of energy expenditure, will be obtained by dividing oxygen uptake values (ml · kg-1 · min-1) by 3.5 (1 MET is defined as resting metabolic rate that is 3.5 ml · kg - 1 · min - 1). VO2 will be measured using Cosmed K4b2 Portable Metabolic Measurement System (Cosmed, S.r.l., Rome, Italy) according to the manufacturer’s instructions.To promote participants’ interest, the exercise training will consist of three different activities including swimming (intensity: 6 MET), aerobic exercise (intensity: 7.5 MET), and basketball (intensity: 6 MET).
Diet modification
All participants will receive well-defined and balanced daily meals during the four-week intervention. Dietary recommendations will be individualized based on the individual’s basal metabolic rate and will range from 1,600 to 2,000 kcal/day. The basal metabolic rate will be calculated using the Mifflin equation [28]. The caloric intake will be calculated based on the Chinese food chart. Each day, three well-balanced meals will be provided with the following calorie allocations: breakfast 35%, lunch 40%, and dinner 25%. Each meal comprises 30% protein, 20% fat, and 50% carbohydrate by energy. Animal and vegetable oil and starch-rich food will be minimized, while the intake of vegetables, fruits, bean products, rabbit meat, beef, pork, and cellulose will be increased. This prescribed diet includes pivotal nutrients such as vitamins, minerals, essential amino acids, fiber, and polyunsaturated fatty acids.
Hypoxia exposure
To test the hypothesis that hypoxia exposure would ameliorate the compensatory increase in appetite elicited by exercise and diet, the experimental group (hypoxia group) will sleep in a normobaric hypoxia environmental chamber every night during the four- week intervention. The advantage of using an environmental chamber compared to real altitude situations is that the effects of hypoxia can be isolated from the influence of other confounding factors present in a real altitude situation, such as the influence of temperature, humidity, and physical activity levels. The normobaric hypoxia will be designed to mimic an altitude of 2700 m. We will use the large hypoxic training system (Low Oxygen System, Dubai, Germany)) of the hypoxia test laboratory of Shanghai Oriental Oasis Training Base to simulate a hypoxic environment (14.7% O2; approximately 2700 m). After a one-day hypoxia acclimation period, participants of the hypoxia group will sleep in the hypoxia training laboratory for 10 hours every night (from 21:00 to approximately 7:00 the next day), seven times per week, for four weeks.
Attrition and compliance
Due to the voluntary nature of the enrollment of the summer weight loss camp, attrition rates will be very low. In a pilot trial involving 50 overweight male Chinese adolescents, 47 children completed a four-week diet and exercise intervention, and only three children (6%) dropped out of the study due to loss of interest. Considering the addition of a two-month follow-up, we expect a higher attrition rate (15%) in this study. Our recruitment plan was made based on this consideration.
Outcome assessments
Baseline assessments will be conducted two days before the onset of the four-week intervention. Post-intervention and two month follow-up assessments will be conducted two days and two months after completion of the intervention, respectively. The primary outcome of the study will be body composition and the secondary outcome measures will include appetite score and blood levels of gastrointestinal hormones including leptin, ghrelin, PYY, CCK and GLP-1, as well as circulating levels of IL-6. All the assessments will be performed in the laboratory of exercise physiology at the Shanghai University of Sport.
Primary outcome
Body composition
Body mass and height will be measured using a digital scale (Yaohua Weighing System Co., Shanghai, China) and a wall-mounted stadiometer (TANITA, Tokyo, Japan), respectively. Body composition and fat distribution will be measured using dual energy X-ray absorptiometry (DXA) (GE Lunar Prodigy, Madison, Wisconsin, United States)). The software (ENCORE, version 10.50.086, GE, Madison, Wisconsin, United States)) will be used to analyze total lean mass (TLM), total fat mass (TFM), total body fat percentage (%TBF), android fat percentage (%AF), and gynoid fat percentage (%GF).
Secondary outcomes
Cardiorespiratory fitness
Peak oxygen consumption (VO2peak) for all participants will be measured using the Cosmed (K4b2) portable metabolic system. Expired respiratory gases will be collected on a breath-by-breath basis during a submaximal treadmill (H/P/Cosmos Pulsar 4.0, Cosmos Sports & Medical Ltd., Nussdorf- Traunstein, Germany) test. Exercise will start at a speed of 2 km/h, which will be increased every 2.5 minutes by 1 km/h until 8 km/h is reached. The criterion of exercise termination will be 80% of the maximum heart rate (HRmax). Trained research assistants will record heart rate and power output data at the end of each stage. Heart rate will be measured using a polar heart rate monitor (Polar Electro, Kempele, Finland).
Blood analyses
Fasting blood samples (2 mL, 12-hour fasting) will be obtained at baseline, post-intervention, and at two months follow-up. Serum levels of gastrointestinal hormones including leptin, ghrelin, PYY, CCK and GLP-1, as well as cytokine IL-6, will be measured using commercially available ELISA kits (R&D Systems, Minneapolis, Minnesota, United States). As reported, the lower limit of sensitivity of the assays will range between 0.8 pg/mL and 10 pg/ml. The intra-assay coefficient of variation (CV) will be less than 5% and the inter-assay CV will be less than 10%. Absorbance will be read at 450 nm wavelength using a microplate reader (Bio-Rad 550, Bio-Rad, Hercules, California, United States).
Appetite assessment
Children’s appetite at baseline, during, and post-intervention, and during follow-up will be assessed via a simple eight-item appetite questionnaire (Additional file 1). The questionnaire was developed by modifying the eight-item Council on Nutrition Appetite Questionnaire (CNAQ). The CNAQ is a short, simple, appetite assessment tool, which has been validated and proved to be able to predict weight loss in community-dwelling adults and nursing home residents [29]. We modified the CNAQ to reflect the nature of our intervention.
Data management and statistical analysis
Data generated in the study will be collected and summarized using mean (standard deviation) values. Group differences in demographical and clinical characteristics at baseline will be tested using a two-sample t-test for quantitative data and chi-square test for qualitative data. Repeated measures analysis of variance (RM ANOVA) will be used to analyze the main effect of treatment and potential time-by-treatment interactions. For these analyses, parametric assumptions of the normality and homoscedasticity will be checked using standard tests and graphical methods. If those assumptions are violated, data transformation and non-parametric procedures will be considered sequentially. Multivariate regression analysis will be performed to analyze the relationship between changes in body composition, appetite score, and gastrointestinal hormones following intervention and during follow-up. Multiple models will be developed and tested for different sets of outcome measures and regressor variables. All statistical analyses will be performed using the Statistical Package for the Social Sciences (SPSS, Inc., Chicago, Illinois, United States) for Windows version 18.0, and a significance level of 0.05 will be used.