Aim
The main purpose of this study was to compare the recovery degree, prognosis and safety of swallowing function in PSD patients with early rehabilitation measures at different times, so as to provide a basis for exploring the best time of early swallowing function rehabilitation in PSD patients.
Study Design And Setting
Our study was a prospective, multicenter, subject and evaluator blind, practical randomized controlled trial. Study subjects will be recruited, treated and evaluated at three medical facilities. Informed consent will be obtained when screening is appropriate, followed by baseline assessment prior to randomization. Randomly generated sequences were assigned to a control group and three experimental groups in a ratio of 1:1:1:1. After randomization, the experimental group will start 4-week of continuous care regimen (neuromuscular electrical stimulation therapy and rehabilitation exercise) at different times, while the control group will only receive routine rehabilitation (neuromuscular electrical stimulation therapy). All four groups will be offered other conventional ischemic stroke treatments. The main outcome was to compare the recovery of swallowing function and efficacy of the four groups after the completion of 4-week rehabilitation training. Our protocol presentation is in accordance with the Standard Protocol Items: Recommendations for Interventional Trials 2013 (SPIRIT 2013) Guidelines [31]. The study flow-chart is presented as Fig. 1. The study enrollment, intervention, and evaluation schedule is show as Table 1.
Participants
The inclusion criteria are as follows: to be eligible, subjects must (1) Age 18 years or older; (2) It should comply with the Main Diagnostic Points of Various Cerebrovascular Diseases in China 2019 formulated by the Cerebrovascular Group of the Chinese Society of Neurology [32], and confirmed by CT or MRI examination; (3) Dysphagia was detected by the drinking water test and standard swallowing function score scale; (4) The time of admission was less than or equal to 3 days after onset; (5) The neurological Impairment Scale (National Institutes of Health Stroke Scale, NIHSS) was used to assess clear consciousness, no cognitive impairment, able to understand and execute simple instructions, and able to complete rehabilitation training; (6) After discharge, the patient lives alone and does not use smart phones; (7) Informed consent and voluntary participation in this study.
The exclusion criteria are as follows: (1) Progressive stroke; (2) Patients with thyroid disease and throat infection affecting local swallowing activities; (3) Patients with severe organ dysfunction, malignant tumor and other critical conditions; (4) Participate in other intervention researchers at the same time; (5) Withdrawal or loss of follow-up and incomplete participation in this study.
Table. 1 Participant timeline

Ethical Issues
Our protocol complies with the principles of the Declaration of Helsinki, and has been approved by the Ethics Committee of Lanzhou University Second Hospital. We also completed registration at Chictr.org.c (ChiCTR2000029149).
Recruitment Strategies And Informed Consent
First, all patients will be enrolled at the neurology inpatient units of three medical institutions to complete 4-week swallowing rehabilitation. The details of our study, including trial objectives, features, possible benefits and risks, other treatment options available, and the rights and obligations of subjects stated in the Declaration of Helsinki should then be made clear to patients by researchers with their grouping codes. After obtaining written informed consent, subjects will be enroll in the study. During the trial, if new ideas about research ethics emerge, the ERC will deal with written amendments to informed consent and, after approval, the individual's consent will be requested again. In the event that patients drop out; their available data will be retain for final analysis. We will strictly protect the personal information of research subjects and only use it for data analysis.
Randomization And Allocation Concealment
The sample size was allocated according to the proportion of monthly admission to the department of neurology of each hospital, and data were collected at the same time. In order to prevent contamination between the experimental group and the conventional group, the data of the control group were collected first, and the data of the experimental group were collected after all the patients in the control group were discharged. Subjects were randomly and equally assigned and recorded by computer-generated randomization grouping codes. Subjects were placed in sealed, opaque envelopes, which were opened upon admission and agreed to join the study to ensure secrecy of the assignment.
Double blindness was used. The evaluators and data analysts kept group assignments blind, patients were informed that they would begin rehabilitation intervention after assessment, baseline and post-intervention data were collected by two study investigators; Grouping is not recorded in the hospital's electronic medical record. In the last case, the blind was unblinded, and the sponsor, evaluator, implementer and data analyst were present to open the sealed envelope in order
Interventions
Control group
The control group will only receive clinical routine rehabilitation therapy -- neuromuscular electrical stimulation therapy on the basis of routine care for ischemic stroke, from the second day of admission to discharge, once a day, 20 minutes each time. Due to the rapid bed rotation, the average length of stay of PSD patients was about 7 days.
Experimental group
The experimental group was given rehabilitation exercise on the basis of the control group. According to the different times of starting rehabilitation exercise training after stroke onset, the experimental group was divided into 0–3 days group (group A), 4–7 days group (group B) and 8–14 days group (group C). Each group lasted for 4 weeks, once every morning and afternoon, about 30 minutes each time. One to one by the rehabilitation specialist nurses to teach patients and their families, and issued manuals, training videos and needed items (tongue depressors, whistles, molars), explain the purpose of rehabilitation exercises, significance and simple mechanism of action. Based on evidence-based medicine, the rehabilitation exercise was modified and perfected by Delphi method, and constructed to be suitable for PSD patients. The content mainly includes facial training, training the lips, tongue, throat, jaw movement training, the training well, Masake training method, Shaker training method, presented in the form of manuals, and with practice video, and music as the background, the stroke rehabilitation training swallowing visualization, visualization, simplification, details are shown in Table. 2.
Patients practice rehabilitation exercise for the first time, first open the rehabilitation exercise manual, teach rehabilitation exercise training action, after the basic understanding, in the ward of the TV broadcast rehabilitation exercise video, guide patients to follow the video training, for patients do not understand the place to explain, until patients understand the whole set of rehabilitation exercise. During rehabilitation training, caregivers of patients are required to participate in the training, because each patient is in a different state. When patients cannot complete certain movements themselves, caregivers can assist patients in training. In the rest of the training time, patients follow the video for training, caregivers for auxiliary training, if there is a problem with rehabilitation nurses for movement guidance.
Table 2
Post-Stroke Dysphagia rehabilitation exercise
Designation
|
Content
|
Methods
|
Ⅰ. Facial training
|
1.Cheek massage
|
Place your hands on both sides of your cheeks and massage clockwise for 4 beats and counterclockwise for 4 beats.
|
2.TMJ massage
|
Place both hands on the bilateral TMJ and massage 4 beats clockwise and 4 beats counterclockwise for 4 consecutive 8 beats.
|
3. Frown
|
Frown 4, relax 4, hold 2 8 beats.
|
4. The gills
|
Drum gills 4 beats, air movement 2 beats left, air movement 2 beats right, lasting 4 eight-beats.
|
5. Smile
|
Smile 4 beats, relax 4 beats, and continue for 4 octaves.
|
6. Show
|
Grin on 4 beats, relax on 4 beats, and continue for 4 octaves.
|
7. Open and close your mouth
|
Open mouth for 4 beats, close mouth for 4 beats. It lasts 4 octaves.
|
Ⅱ. Labial ministry training
|
1. Zipped lips pronounce "um"
|
Press your lips together and say "um" for 4 beats, relax for 4 beats. It lasts 4 octaves.
|
2. Close your mouth and say "woo"
|
Press your lips together, say "woo" 4 beats, relax 4 beats. It lasts 4 octaves.
|
3. "Yi" and "woo" conversion
|
Pronounce "yi" 2 beats, immediately pronounce "woo" 2 beats, relax 4 beats. It lasts 4 octaves.
|
4. Keep your lips closed
|
Close your lips, press the 4, relax the 4. It lasts 4 octaves.
|
5.Resistance training
|
Tongue depressor in the middle of lips, lips hold tongue depressor, forcibly close, pull out tongue depressor, and lips resistance 2 beat, relax 2 beat;With lip resistance 2 beats, relax 2 beats. It lasts 4 octaves.
|
Tongue depressor on the right half of the lips, lips hold tongue depressor, forcibly close, pull out tongue depressor, and lips resistance 2 beat, relax 2 beat;With lip resistance 2 beats, relax 2 beats. It lasts 4 octaves.
|
Tongue depressor on the left half of the lips, lips hold tongue depressor, forcibly close, pull out tongue depressor, and lips resistance 2 beat, relax 2 beat;With lip resistance 2 beats, relax 2 beats. It lasts 4 octaves.
|
6.Whistle training
|
Whistle 4 beats, relax 4 beats. It lasts 4 octaves.
|
7.To suck
|
Wrap your lips around the molar stick. It lasts 2 octaves.
|
Ⅲ. Tongue
training
|
1.Slip the tongue
|
Extended tongue for 2 beats, contracted tongue for 2 beats. It lasts 4 octaves.
|
2. Move your tongue up and down
|
The tongue touches the upper lip for 4 beats, and the tongue touches the lower lip for 4 beats. It lasts 4 octaves.
|
3.Move your tongue from side to side
|
The tongue touches the left corner of the mouth 4 beats, and the tongue touches the right corner of the mouth 4 beats. It lasts 4 octaves.
|
4.Tongue on top of palate
|
Tongue top palate 4 beats, relax 4 beats. It lasts 4 octaves.
|
5.Licking your lips with the tip of your tongue
|
Lick your lips with the tip of your tongue 1 time, 8 beats. It lasts 4 octaves.
|
6. Pronounce "T, D"
|
Pronounce "t" 2 beats, relax 2 beats; pronounce "d" 2 beats, relax 2 beats. It lasts 4 octaves.
|
7. Pronounce "CH" and "C"
|
Pronounce "CH" 2 beats, relax 2 beats; Pronounce "C" 2 beats, relax 2 beats. It lasts 4 octaves.
|
9. Pronounce the "K, G" sound
|
Pronounce "K" 2 beats, relax 2 beats; Pronounce "G" 2 beats, relax 2 beats; It lasts 4 octaves.
|
Ⅳ.Pharyngeal
training
|
1. Hold your breath and sound training
|
The patient sat on the chair, supported the chair face with both hands and did pushing and pressing action 2 beats at the same time, held his breath, kept the upper body upright, closed the glottis 2 beats, suddenly opened the glottis, exhaled "ah" 2 beats, relaxed 2 beats. It lasts 2 octaves.
|
2. Subglottic swallowing exercises
|
The patient placed his finger on the upper edge of the thyroid cartilage, took a deep breath in the nasal cavity for 2 beats, held his breath for an empty swallowing, and felt the thyroid cartilage move upward for 4 beats. Immediately after swallowing, he coughed for 2 beats. It lasts 2 octaves.
|
3. Pronunciation training
|
Open mouth pronounce "ah" 4 beats, relax 4 beats;The tongue tip on top of the palate pronounce "yi" 4 beats, relax 4 beats. It lasts 2 octaves.
|
Ⅴ. Mandibular exercise training
|
1.Teething stick training
|
Place the molar stick between the patient's incisors, and let the patient bite. It lasts 2 octaves.
|
2.Move your jaw from side to side
|
The jaw moves 2 beats to the left and 2 beats to the right. It lasts 4 octaves.
|
3.Tap your teeth up and down
|
Tap teeth up and down for 2 consecutive eight beats.
|
VI.Respiratory muscle training
|
1.Abdominal breathing
|
Inhale for 2 beats, hold for 2 beats, exhale for 2 beats, and relax for 2 beats. It lasts 4 octaves.
(Abdominal breathing take supine or comfortable meditative sitting position, relax the whole body, the right hand on the belly button, the left hand on the chest.As you inhale, expand your abdomen outward as much as possible, keeping your chest still.As you exhale, pull your stomach in as much as possible, keeping your chest still.)
|
2.Lip constriction
|
Inhale for 2 beats, hold for 2 beats, and exhale for 4 beats. It lasts 4 octaves.
(Breathe in through your nose and exhale with your lips closed. Hold the exhale for as long as possible.)
|
Ⅶ. Masake training method
|
Place the tip of your tongue between your teeth for 2 beats, empty swallow for 4 beats, relax for 2 beats. It lasts 4 octaves.
(When swallowing, the tongue tip later between the teeth of a small part of the tongue or pull out a small part of the tongue by hand, and then the patient to swallow, so that the throat wall forward movement and tongue root close to increase the pressure of the pharynx.)
|
Ⅷ. Shaker training
|
Let the patient lie supine on the bed, raise the head to watch the toe 2 beats (note that the shoulder should not leave the bed surface), relax 2 beats. It lasts 4 octaves.
(Have the patient lie on the bed on their back with their head as high as possible, but their shoulders should not leave the bed surface, and their eyes should be on their toes.)
|
Outcome Measures
Primary outcome measures
Degree of improvement in swallowing function: Swallowing function was assessed on the day of enrollment and the day of the end of the intervention. The Standard Swallowing Function Assessment Scale (SSA) was used, with a score ranging from 18 to 46, and the higher the score, the worse the patient's current swallowing function.
Efficacy assessment: Swallowing function was evaluated on the day of enrollment and the end of intervention. The Water Swallow Test (WST) of the patient was grade 1, and the swallowing disorder basically disappeared Using the WST rating, the patient was WST grade 1, and the swallowing disorder basically disappeared and was considered to be cured. The WST of patients increased by 2 grade or more, and the swallowing function improved significantly. The WST of the patient increased by 1 grade and the swallowing function improved. Patients with no significant changes in WST and swallowing function were considered ineffective. Effective rate = (cured + effective + effective) cases/total cases ×100%.
Secondary Outcome Measures
(1) Incidence of choking: During the intervention, the patient's family members recorded The Times of choking reaction during autonomous feeding or nasal feeding at any time, and The Times of choking for each patient/the total times of choking for all patients in the group multiplied by 100%.
(2) Incidence of pneumonia: Patients diagnosed with pneumonia were recorded by their competent doctors during the intervention. New or progressive pulmonary infiltrate appeared after stroke, with 2 or more clinical symptoms of infection: the patient's body temperature was greater than or equal to 38℃; Lung imaging (X - ray or CT) suggests recent inflammatory lesions or lung texture changes; New respiratory catarrh symptoms, cough, expectoration or aggravation of the original respiratory symptoms, purulent sputum or sputum volume increased significantly, with or without chest pain; New wet rales or changes in respiratory sounds were detected in the lungs. The total number of white blood cells is greater than or equal to 10×109/L, and there is a history of aspiration, it is caused by food, oral and pharyngeal secretions, stomach contents and other irritant substances into the lower respiratory tract pneumonia.
(3) Quality of life: Assessment was performed at enrollment and on the day of the end of the intervention. The Swallowing Disorders Specific Quality of Life Scale (SWAL-QOL) included 11 items, which were divided into 44 related items. Each item included 5 grades, with a score ranging from 1 to 5. The higher the score, the higher the patient's quality of life was.
(4) Compliance rate: the number of rehabilitation exercises completed by patients or their family members during the intervention process is recorded, and the number of training completed by patients/the total number of rehabilitation exercises training multiplied by 100%.
Follow-up
Due to the limitations of clinical practice, PSD patients generally stay in hospital for 1 week, and then practice at home for another 3 weeks. Follow-up was conducted after the 4-week intervention period. The degree of recovery of swallowing function and quality of life of the patients were evaluated by video, and the patients or their caregivers were asked to collect the required information and give feedback in the form of photos.
Data Management And Quality Control
(1) The research group of director of nursing Department of Lanzhou University Second Hospital is responsible for the data management of the project. Data supervision is completed by the director's research group of the Nursing Department of Lanzhou University Second Hospital, which is independent of the sponsor and has no conflict of interest. If any content is not included in this plan, it shall be decided by the sponsor, evaluator and data supervisor through consultation. Interim results are available to the data analyst, final results are available at the end of the study, and the sponsor may terminate the study. At the stage of project design, experts in medical statistics, neurology and rehabilitation medicine will guide and determine the final research scheme.
(2) Intervention and data collection stage: subjects were selected strictly according to the inclusion criteria; Data will be obtained from medical records, direct interviews with patients or their caregivers. All rehabilitation nurses received centralized and unified training before the implementation of the study, to help them get familiar with and master the content of rehabilitation exercises, and to ensure the homogenization of operation modes of different personnel; The intervention time of each subject should be as consistent as possible and the intervention environment should be the same. During home training after discharge, patients were provided with daily feedback of training photos, and patients who failed to complete training without reason were provided with video supervision. Three graduate students to ensure the homogeneity of data collection methods completed data collection.
(3) Data entry and analysis stage: During data entry, two researchers input data separately for proofreading. After the input, 10% of the data will be randomly select for verification. If the check rate is greater than 90%, it is qualified.
(4) The audit team, independent of the investigator and sponsor, reviews the progress and materials of the study every three months.
Sample size considerations
We assumed a two-tailed comparison and set the type I error rate at 0.05 with 90% power. Based on previous studies [33, 34] of rehabilitation training for swallowing after stroke, we plan to screen approximately 180 patients with PSD. After screening, subjects were randomly divided into three experimental groups and a control group. For a conservative estimate (dropout rate = 10%), we assume that 164 subjects will complete the study. In order to reduce the lost follow-up rate, we will use regular communication via phone or social media to keep participants engaged.
Statistical analysis
In this study, Excel2010 was used for data entry, SPSS25.0 statistical software package was used for analysis, measurement data were expressed by Analysis of variance(ANOVA) for single-factor analysis, multivariate linear regression analysis for multi-factor analysis. Multiple fillings if missing data exists. Χ2 test was used for univariate analysis and Logistic regression was used for multivariate analysis. P < 0.05 was considered as statistically significant difference. SSA score, WST score and SWAL-QOL score were completed at enrollment and at 4 weeks of intervention, and assessments at different time points were analyzed by repeated measures of ANOVA. Adverse events in each group will be recorded as the percentage (%) that was assessed for safety using the Chi-square test or Fisher's exact test..