The Effect of High Intensity Interval Exercise and Myofunctional Therapy on Obstructive Sleep Apnea

1. introduction Obstructive sleep apnea (OSA) is a sleep disorder that is characterized by apneas and hypopneas during sleep due to repetitive collapse of the upper airway. The prevalence is 3-7% in men and 2-5% in women. Daytime sleepiness and insomnia are associated with both decreased physical activity and increased obesity, increased OSA severity is also associated with decreased physical activity, resulting in a vicious cycle. Accumulating evidence shows that OSA is associated with multiple cardiovascular disorders, such as hypertension, type 2 diabetes mellitus and coronary artery disease. The treatment of OSA may reduce the risk of these cardiovascular disorders. The apnea-hypopnea index (AHI) has been widely used to quantify the level of OSA by the number of apnea episodes per hour. Based on the definition of American Academy of Sleep Medicine (AASM) and an article about epidemiology of OSA, AHI range from 5~15 is defined as mild OSA, and AHI 15~30 is defined as moderate OSA. AHI over 30 can be regarded as severe OSA. Epworth sleepiness scale (ESS) is a questionnaire to measure subject general level of daytime sleepiness. OSA patients often revealed higher score which refers to higher possibility to get sleep disorder. Recently, some articles showed that body and tongue fat deposition have certain relation with OSA, the higher the fat, the more possible to get OSA. Body fat can be quantified through waist and neck circumference. In order to solve OSA, myotherapy (MT) might be an effective method. MT is composed of isotonic and isometric exercise related to oral, oropharyngeal, and respiratory exercise. Based on previous review and meta-analysis, MT can decrease not only AHI score, but also ESS score. It means that MT is beneficial to release the symptom of sudden apnea and hypopneas through decreasing the amount of fatty deposition of the tongue and increasing oral and oropharyngeal muscle tone. Recent article also supports that weight and fat loss of soft tissue of tongue is important for treat OSA[12]. Some articles also mentioned that the snoring situation during sleep time has declined. That is, the sleep interruption to people sleeping beside the OSA patient are alleviated. In addition, the sleep and life quality can be ameliorated by myotherapy. Inconsistent evidence also manifested that oxygen saturation has been improved after myotherapy. In short, MT can be regarded as optimal alternative therapy to mild to moderate OSA patient. Exercise training is an effective intervention in reducing the severity of OSA and improving symptoms of sleepiness and quality of life. According to previous systematic review and meta-analysis, exercise such as aerobic exercise and resistance exercise has been proved effective on improving AHI, ESS. However, these articles didn't focus on body fat that is highly associated with OSA. According to recent article, decreasing body fat is essential to deal with OSA. Hence, high intensity interval training (HIIT) has been proved its same effect on lower down body fat as moderate intensity continuous training (MICT). HIIT also shows greater effectiveness compared with (MICT) on cardiovascular and metabolic function in both healthy populations and populations with cardiovascular disease. In addition, HIIT has the additional benefit of being a time-efficient program which can increase exercise adherence and the participants were more likely to intend to continue. Previous two studies show that HIIT could improve the severity of OSA in both obese adults and obese children, but it still needs evidence to prove the effectiveness among different aspects. According to recent literature review, there is no standard physical therapy model for OSA. Furthermore, less articles discussed about the effect of supervised verse unsupervised treatment and the effect of mix-model treatment. Hence, the purpose of the study is investigating the comparison between supervised HIIT plus myotherapy and unsupervised home exercise plus myotherapy. It is hypothesized that HIIT plus myotherapy might revealed better outcomes on AHI, ESS, and body fat.
2. Method Procedure The subjects included in this study were randomized assigned to experimental group and control group. The randomization orders were decided by computer, and all the contents were concealed into a dark color envelop. Before first treatment, the envelops were opened to determine which treatment protocol were adopted. The treatment of experimental group was composed of myotherapy and resistance exercise plus high intensity interval training (HIIT). The treatment of control group was composed of general stretch exercise plus myotherapy. Subsequently, initial measurement was conducted including BMI, Body fat, waist and neck circumstanced, apnea hypopnea index (AHI), Epworth Sleepiness Scale (ESS), oxygen desaturation index (ODI) and body fat. Then, the treatment protocols were both executed for 30 minutes, sixteen times in two months. After completing the treatment process, the final measurement was conducted as the initial treatment. All the outcomes were collected and analyzed by statistically method.

Treatment Experimental group is composed of myotherapy and high intensity interval training (HIIT) plus resistance exercise. Myotherapy was based on the studies by Lequeux et al. and Cláudia et al. Subjects were instructed to perform tongue slide, tongue force, tongue press, tongue reach, swallowing exercise, smiling exercise, jaw press exercise, chewing exercise, breathing exercise and buccinator exercise. Exercise training would be implemented in the form of high-intensity interval training and resistance exercise. High-intensity interval training intensity of the target heart rate (THR) was calculated as follows: THR = (HRmax - HRrest) × 80-90%Intensity + HRrest. The HIIT program included four 3-min bouts at high-intensity (80-90%HRR), separated by 3-min of active recovery and total for 4 cycles of 24-min HIIT intervention. The HIIT exercise options were running on a treadmill. Polar heart rate monitors were used to monitor exercise intensity. Subsequently, the 20-min resistance exercise program was conducted using a theraband. The different colors of band indicate different levels of resistance. The four limbs were mainly exercised with yellow, red, green and blue colored elastic bands. In this regard, the theraband was color-coded in the following order of increasing resistance: blue, green, blue and dark. Each subject began the program using the blue theraband. As the subjects performed the ability to complete 15 repetitions without difficulty, the intensity of the thraband was progressed. The volume of the resistance was 10 repetitions per set, and 3 set in a training session. Exercises included shoulder press, shoulder lateral raise, biceps curl, butterfly, seat row, leg press(squat), calf raise, latissmus pulldown, abdominal curl and bridge. Five minutes of stretching exercises was conducted as a warm-up and cool down period before and after training sessions. The physiotherapist with abundant experience related to myotherapy and high intensity interval training (HIIT) plus resistance exercise supervised all the treatment process to confirm the quality of execution. This exercise was performed 12~15 repetition depends on patient's condition.

Statistical analysis The primary variable of the study were body fat, neck and waist circumstance, AHI, ODI and ESS. The secondary variable of the study were body weight and BMI. Descriptive statistics for the categorical variable were documented as frequency counts and percentages. The continuous variables were reported as mean + SD, if they were normal distribution, or they were recorded as median and range.

The Wilcox signed test was adopted to analyze the treatment before and after the treatment sessions (time effect). The Mann-Whitney U was applied for the difference before and after treatment between two groups (group effect), and the baseline of two groups was also analyzed by this method. The significant level was set as p value< 0.05. All the statistical data analysis was performed by SPSS version 22.