Cervicocranial flexion exercise (CCFE) and superficial neck flexor endurance training have been widely implemented in clinical practice for curing chronic neck pain. By means of CCFE, the muscle balance between deep neck flexor and superficial flexor would be optimal during neck movement. In other words, the superficial neck flexor( scalenes, SCM, and trapezius) would not overactive and the fatigue threshold might increase. Superficial neck flexor endurance training is proved to be efficient in reducing superficial cervical flexor muscle fatigue as well as increasing cervical flexion strength.

Reasonably, Cervicocranial flexion exercise (CCFE) and superficial neck flexor endurance training are also beneficial to pulmonary function due to training the respiratory accessory muscle (scalens and SCM). Hence this article hypothesizes that Cervicocranial flexion exercise (CCFE) and superficial neck flexor endurance training combined with common pulmonary rehabilitation will manifest better outcomes (pulmonary function, dyspnea situation, pain and stiffness level of neck) than pulmonary rehabilitation only.

2.1 subject 20 patients were recruited in this article during 2020/03 to 2021/02. The inclusion criteria were 1.) SCI onset in a year; 2.) motor level above T12 and American Spinal Injury Association Impairment Scale (AIS) grade A, B, C, or D; 3.) age from 20~70; 4.) FEV1< 80% prediction value. The exclusion criteria were 1.) ventilation dependence 2.) Tracheostomy 3.) Psychiatric condition 4.) Progressive diseases 5.) infection 6.) cancer 7.)unable to speak Chinese or English. Patient who met the criteria would were asked to provide written informed consent. The IRB were approved by Tao Yuan General Hospital, Ministry of Health and Welfare, IRB committee.(TYGH108045 ) 2.2 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 consist of Cervicocranial flexion exercise (CCFE) and neck flexor endurance training plus normal cardiopulmonary rehabilitation. The treatment of control group was composed of general neck stretch exercise plus cardiopulmonary rehabilitation. Subsequently, initial measurement was conducted including lung capacity test, dyspnea situation, pain and stiffness of neck. Moreover, lung function such as FVC and FEV1, dyspnea situation, pain and stiffness of neck were also recorded one time in a week as a short-term outcome. Then, the treatment protocols were both executed for 30 minutes, ten times in a month. After completing the treatment process, the final measurement were conducted as the initial treatment. All the outcome were collected and analyzed by statistically method.

2.3 Outcome measurement Lung capacity was evaluated by in lung function examination center of Taoyaun general hospital. The parameters revealed by this test. Before the examination, bronchodilator should not be used to avoid interfering the outcome of examination. In addition, portable lung capacity device(ezOxygen, Taipei, Taiwan) were adopted to record lung function parameters such as FVC and FEV1. First, subjects were instructed to hold deep breath and let his mouth firmly contact to the mouthpiece. Second, subjects were asked to exhale the air into the mouth and the process was repeated three times, between each sessions, subjects were allowed to rest at least thirty seconds. Three tests were recorded and the mean value was calculated. The dyspnea situation was evaluated by the questionnaire Baseline Dyspnea Index (BDI) and Transition Dyspnea Index (TDI). The BDI measured the severity of dyspnea at the baseline and the TDI measured the change from baseline. Both questionnaires are composed of three parts: functional impairment, magnitude of task, and magnitude of effort required to evoke dyspnea. Each part in BDI is scored from 0(very severe) to 4(no impairment) and total score from 0~12( the lower the score, the worse the severity of dyspnea) is also recorded. Each component in TDI is rated by seven grades ranging from -3 (major deterioration) to +3 (major improvement). Total score ranged from - 9 to + 9. The lower the score is, the more deterioration of the severity of dyspnea revealed. The minimal clinical important difference(MCID) of TDI is ≥1 unit[31]. The number rating scale NRS is used to define the level of neck pain and stiffness. The MCID of neck pain VAS is 1.5.

2.4 Treatment Experimental group is consist of CCFE and neck flexor endurance training plus normal cardiopulmonary rehabilitation. Subjects were instructed to perform chin in without excessive SCM contraction. CCFE was conducted in supine position. Subjects were asked to maintain chin for 10 seconds each repetition, ten repetitions for a set, 3 sets in a treatment session[30]. The physiotherapist with abundant experience related to CCFE supervised all the treatment process to confirm the quality of execution. If no symptoms revealed, the neck endurance exercise would involve into the treatment sessions. Neck endurance exercise was also conducted in supine position. Subjects were taught to lift their head with a neutral upper cervical spine first. Then, they gradually move the head and neck through as full range as possible without induce discomfort or neck related symptom. This exercise was performed 12~15 repetition depends on patient's condition[33]. If the subjects were unable to achieve the training level, subject's bed were inclined up from horizontal. Hence, the demand of lifting neck and head declined and subjects could perform aimed repetitions. The cardiopulmonary rehabilitation protocol was based on Wu, Y.-T. et al[9], and it was divide into three stages. In first stage, subjects lied in supine position, and they were instructed to do diaphragm breath, pursed-lip breath, lateral coastal breath, cough training, and rib mobilization. The dosage and type of therapy was decided depending on subjects situation everyday. The total cardiopulmonary training was controlled in 20 minutes. In second stage, the incentive spirometer (picture A) was integrated into the therapy. Subjects were asked to inhale the air slowly with eye contact at the chamber of the spirometer for controlling the air flow velocity. Subject were also asked to inhale as many volumes as possible. Back support was inclined to hold subject in semi-sitting position. 15 repetitions was given to subjects and subjects received 20 second rest between each repetition. In addition, some upper extremities training combined with breath exercise were also involved in after incentive spirometer training. 15 repetitions, 3 sets was given to subjects. In stage 3, the upper extremities exercise were combined with spirometer training. 15 repetitions, 3 sets was given to subjects. The experienced physiotherapist decided the stages depends on patient's capability, that is, the stages were not the same for every subject. The control group were composed of neck stretch exercise plus cardiopulmonary rehabilitation. The neck stretch exercises were executed by physiotherapist before cardiopulmonary training. Neck flexion, neck extension, neck right rotation, neck left rotation, neck right side-bending, and neck left side-bending were applied to subjects 5 times for each direction, and neck maintained at end position for 30 seconds each time.

2.5 Statistical analysis The primary variable of the study were vital capacity, respiratory complication ,and dyspnea. The secondary variable of the study were FEV1, pain and stiffness of neck. 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 were performed by SPSS version 22.

Inclusion Criteria:

• SCI onset in a year.
• Motor level above T12 and American Spinal Injury Association Impairment Scale (AIS) grade A, B, C, or D.
• Age from 20~70.
• FEV1< 80% prediction value.

Exclusion Criteria:

• Ventilation dependent
• Tracheostomy
• Psychiatric condition
• Progressive diseases
• infection
• Cancer
• Unable to speak Chinese or English.