What Happens to Swallowing Muscles after Stroke?: A Prospective Randomized Controlled Electrophysiological Study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.273645
Source of Support: None, Conflict of Interest: None
Keywords: Dysphagia, electrophysiology, stroke, swallow muscles
Neurological dysphagia is common after stroke and can be seen with 65–81% frequency in acute phase. Although 90% of the patients have spontaneous recovery in 2 or 3 weeks, persistent cases (11–50%) up to six months are reported., Recently, studies show that every stroke case is a risk factor for swallowing disorders.
Swallowing is a complex action involving several oral and pharyngeal muscles. Bolus is directed from mouth to stomach by complex, coordinated, and synergistic movements of these muscles. The term dysphagia refers to the difficulty in swallowing safely due to the deficiencies of coordination, strength, and sense of these muscles.
Stroke generally influences the oropharyngeal phase of swallowing. Oral dysphagia is resulted with the deficient manipulation of the solid and liquid aliments due to the tongue discoordination, fatigue of tongue muscles, mastication difficulty, and decreased tone and strength of buccal and labial muscles. However, pharyngeal dysphagia is associated with residue, airway leak, and aspiration, which leads to more serious problems.
Oral phase of swallowing is under control of high cortical areas of the brain. It is associated with the voluntary contraction of mastication, mimic, and tongue muscles, which are innerved by trigeminal, facial, and hypoglossal nerves. However, pharyngeal phase is under control of central pattern generators of the brain stem and is associated with the involuntary contraction of pharyngeal and laryngeal muscles, which are innerved by glossopharyngeal and vagal nerves. Although the swallowing phases are evaluated separately in studies such as oral, pharyngeal and oesophageal phases, these phases are closely related to each other and should be considered as a whole. For example; the tongue sequentially squeezes the bolus toward the pharynx by contact from the front of the hard palate backward, for safe and effective transport to pharyngeal cavity. The failure in oral phase results with delay or absence of the transmission of the critical sensorial input that should go to the related cortical areas for formed pharyngeal reflex response that protects from aspiration before swallowing.
The major result of the post-stroke dysphagia is aspiration and aspiration pneumonia. Frequency of aspiration pneumonia in stroke patients is 43–50% with a mortality of 45% in the first year after stroke, however, 40–70% of these are silent aspiration, which refers to asymptomatic cases. Eventually, aspiration pneumonia results with increased functional disability and decreased quality of life.,
Even though the dysphagia has these dramatic results for patients, there is still no consensus on diagnosing and screening tools. Common used tools are bedside screening tests, including the observation of the patient during the swallowing action of solid and liquid foods and evaluation of some reflexes, videofluoroscopy, barium passage graphy, and fiberoptic endoscopy. There is poor data about electrophysiological, manometric, and ultrasonographic methods.,
Although aspiration may occur in different phases (before, during or after) of swallowing, persistent dysphagia is generally related with pharyngeal phase (during or after), hence diagnostic tools screen this phase of swallowing generally. However, like the example above, an adequate and effective oral phase is required for the formation and control of bolus, triggering of pharyngeal reflex, and even for normal a pharyngeal phase.
We hypothesised that the most effective and accurate diagnostic tool for screening of post-stroke dysphagia should evaluate both oral and phryngeal phases of swallowing.
Electrophysiological screening of swallowing allows to evaluate oral, pharyngeal and even beginning of esophageal phases. It is used experimentally in a lot of studies because it has no radiation risk like videofluoroscopy and barium passage graphy, and it screens the oral phase in advantage of endoscopy. Moreover, the activation of the muscles that are involved in swallowing action can be evaluated separately by electrophysiological screening.,
Structural changes in involved muscles begin to occur in 4 hours after stroke. Muscle weakness was reported to develop unaffected limbs at one week. This muscle atrophy results with reduced muscle strength and endurance, which leads to decreased quality of life, increased morbidity and mortality. Swallowing is muscle process and swallowing muscle is smaller than extremity muscles; therefore we think that it is inevitable that these swallowing muscles are affected.
Recent electrophysiological studies reported stroke-associated changes in these muscles confirming this data. Some of these studies involved muscles which are associated with an oral phase of swallowing process but either invasive needle electrodes were used in, or usual static activities of the muscles by using a surface electrode were evaluated in these studies.,,,, Moreover, most of these studies are cross-sectional and have no follow-up. However, some studies generally evaluated usual static activity by using surface electrode of these small muscles of swallowing process, which is insufficient to determine their exact functional capabilities. Electrophysiological evaluation of muscles with surface electrodes has some difficulties. The characteristics of the electrical activity of the muscles depend on the length of the muscle fibres, electrode positioning in relation to the muscle fibres, area and distance between the electrodes, and the thickness of the fat layer between the skin and muscles.
With the present study, we aimed to find answers to following questions due to the reasons mentioned above:
To the best of our knowledge present study is the first of its kind in literature; we aimed to evaluate the dynamic functional capacities of the mastication, mimic, and intrinsic tongue muscles, which are associated with the oral phase of swallowing process with a non-invasive electrical stimulation in stroke patients.
This study was performed on 102 subjects between January 2014 and December 2016. Fifty-one inpatients with stroke diagnosis who were admitted to our centre and 51 healthy individuals were evaluated for the study.
Patients aged between 55 and 75 years, who had ischemic stroke diagnosis confirmed with magnetic resonance imaging (MRI), who were admitted for any problem such as motor function impairment or disability, within 20–60 days and have swallow disorder diagnosed with bedside screening test, were included.
Exclusion criteria were the history of malignancy, head and neck surgery, previous stroke and respiratory distress, smoking and alcoholism, and hemorrhagic and/or bilateral stroke, as well as in co-operation, incapability of head positioning and sitting less than 30 minutes for electrodiagnostic evaluation.
The randomisation method was applied to all patients who were hospitalized and eligible for the study criteria between 2014 and 2016 years.
The exclusion criteria for the control group were individuals that aged under 55 or over 70, had hypertension and/or cardiac disease, and psychiatric or endocrine and metabolic diseases like diabetes mellitus, which affects swallowing process, as well as neuromuscular disease that affected compliance with commands during electrophysiological evaluation.
Subjects who have a contraindication for electrical stimulation like a cardiac pacemaker or biomedical device usage, local infection, temporomandibular joint or gastrointestinal disorder were excluded from both groups.
Healthy individuals, patients, and their relatives were informed about the study, and their written consents were obtained at the beginning of the study. The approval of the Ethical Board of the hospital was obtained, and the study was conducted in accordance with the principles of the Helsinki Declaration.
One hundred and two subjects were included in this study.
Demographic and disease characteristics
Demographic characteristics including age, gender, dominant hand, and oral hygiene were recorded for all subjects. Disease characteristics including comorbidities, affected side, post-stroke duration, and infarct region for only patients group were recorded.
Assessment of swallowing function
Bedside screening test were used to select patients for the study, including neurological findings that may affect the swallowing function of the patient, water drinking test, and oxygen saturation test.
According to this test; a Bedside Dysphagia Score (BDS) was obtained by scoring from 0 to 6 with a water-drinking test as described in previous studies  involving the assessment of aspiration findings with drinking 10 ml water, and the measurement of oxygen saturation with a second finger pulse oximeter unaffected by the same procedure. The score with 0–2 was considered as “normal swallowing” and between 3 and 6 as “dysphagia.”
Neurological Examination Dysphagia Score (NEDS) including evaluation of cranial nerves associated with swallowing, head positioning and sitting balance was also calculated. Based on the NEDS calculations, a score of 0–3 was considered as “normal swallowing”, and a score of 4–9 as “dysphagia”. Sum of BDS and NEDS scores used to calculate Total Dysphagia Score (TDS), which is considered as “normal swallowing” between 0–3 points and “dysphagia” between 4–15 points.
Flexible fiberoptic endoscopic evaluation of swallowing
Endoscopic evaluation of patients was performed with the same specialist, while the patient was in a vertical sitting position, using a non-ducted fiberoptic nasopharyngoscope of 3.4 mm diameter, a light source, camera, monitor, and DVD recorder (KarlStorz GmbH and Co KG, Tuttlingen, Germany). Local anaesthetics were not used to interfere with oral and pharyngeal functions. A lubricant gel to the tip of the endoscope and an antifog material to the lens were applied before the examination. The tip of the nasopharyngoscope was inserted from the nostril, and general pharyngeal phase was evaluated. To determine residue, aspiration or penetration were used water up to 10 millilitres according to the patients' tolerance in bedside screening, yoghurt, and a piece of biscuit as solid food. Findings were recorded as video images and examined to score dysphagia levels of patients' with Dzeiwas endoscopic evaluation protocol between 1–6. Score 1 is considered as “normal swallowing function,” while scores 2–6 are considered as “dysphagia.”
All subjects were examined electrophysiologically after they had prepared with 2 hours of fasting, cleaning of face with soap, mouth, and teeth cleaning. Affected side was used for all nerve conduction studies. For the evaluation of intrinsic tongue, masseter, and orbicularis oris muscles, the muscle recording and stimulation areas on face were cleaned up with 70% alcohol solution. Measurements were done by the same specialist while the subject was sitting on the chair with a neutral head position by using a Medelec Synergy 10 channel electroneuromyography (Oxford, U.K.) device. Electroneuromyography (ENMG) window (sweep speed/sensitivity) for observed electrophysiological changes were set 0,5–1 mV/1–2 msec/division. Band pass filter frequency was set between 8 Hz–10 kHz frequency. Stimulation was initiated with low intensity and gradually increased to achieve supramaximal activation. After three times of peak muscle action potential (MAP), an average of three values was calculated.
Nerve conduction studies
Trigeminal nerve: Using Liguori method, recording and reference electrodes were placed with a distance of 2 cm along with a long axis of the muscle fibres of the masseter. Ground electrode was placed on the chin. Mandibular branch of the trigeminal nerve was stimulated from posteroinferior margin of zygomatic arch and posteroventral of temporomandibular joint by bipolar stimulation electrode. To avoid volume conduction, masseter contraction was observed visually.
Facial Nerve: Recording electrode is located at the angle of the mouth just lateral to where the upper and lower lips join. Reference electrode for orbicularis oris was placed on the cheek while the ground electrode was placed on chin. Stimulation was applicated from zygomatic branch of the facial nerve, anterior and inferior to the tragus of earlobe.
Hypoglossal Nerve: Recording and reference electrode were placed on tongue depressor with 2 cm distance, and the depressor was placed on the middle of the tongue to evaluate intrinsic tongue muscles. Ground electrode was placed on chin and the stimulation was applicated from the inferior edge of mandibula.
All electrophysiological studies were done by the same specialist using the same device which explained above and was placed active electrode to submental muscles and reference electrode to chin according to report in the past . A laryngeal piezoelectric sensor was fixed between cricoid and thyroid cartilages also. Electroneuromyography records were taken from submental electrodes and laryngeal sensor. Two deflexions were obtained from laryngeal sensor. The previous one referred to the larynx movement upwards, and the following referred to the movement downwards that means the end of the pharyngeal reflex phase. The initial spike of the first deflexion was labelled as “0” and the second deflexion as “2”. “0–2 interval” was referred to the duration of swallowing reflex. Electrical activity of submental muscles was obtained from submental electrodes. The initial point of the activity was labelled as “A,” which referred to the initial voluntary contraction of the submental muscle complex and the endpoint was labelled as “C,” which referred to the end of the voluntary contraction. “A 0-time interval” was accepted as triggering time of swallowing reflex, which starts with the contraction of submental muscles and ends with the initiation of swallowing reflex. “A-C time interval” was accepted as total contraction time of submental muscle complex that refers to the total oropharyngeal phase duration. All electroneuromyography records were filtered (bandpass 100 Hz–10k Hz), amplified, rectified and integrated.
Assessment of motor functional stage and disability
The motor functional levels of the patients were evaluated with Brunnstrom stage for hand, upper and lower extremities, separately. The functional disability of the patients were evaluated with Functional Independence Measure (FIM) and scored as motor, cognitive, and total functional score between 18–126.
Patients were evaluated by a physical rehabilitation medicine specialist before and after the rehabilitation procedure, while electrophysiological evaluations were performed by another blind specialist. Endoscopic evaluations were done by a blind otolaryngology specialist.
While the bedside screening tests, Brunnstrom stage and FIM were performed by the same physical medicine and rehabilitation (PMR) specialist, electrophysiological evaluations were performed by another blind PMR specialist (1st) on the day of hospital admission (1st day) and the end of therapy (21st day). Endoscopic evaluations were done by a blind otolaryngology specialist.
Standard dysphagia and swallowing rehabilitation procedure was performed to all patients by the same physiotherapist. Patients were educated about oral hygiene, swallowing manoeuvres and, head and neck positioning during feeding. Patients and caregivers were checked every day for accuracy of applications by physiotherapist. Cold stimulation was performed to radix of tongue, palate, tonsillar plica, and oral mucosa 10 sets a day with five repetitions. Oral motor strength exercises for labial, intrinsic tongue and mastication muscles were performed three sets a day with ten repetitions. All exercises were performed once daily by a physiotherapist and the other sets were done by caregivers. In addition, intermittent galvanic stimulation was performed to masseter muscles of patients for 30 minutes a day, five days a week for four weeks (Intelect Advanced- Chattanooga, UK), as mentioned in previous study. Total daily rehabilitation duration was between 60–90 minutes 5 days a week during 4 weeks. Additional rehabilitation procedures for cognitive, sensorimotor, and respiratory disabilities of patients were also added to the rehabilitation program.
Electrophysiological findings of the healthy subjects were compared with the initial (1st day) and follow up (21st day) findings of the patient group. In addition, swallowing functional status, including bedside screening test and FEES scores, and functional status including Brunnstrom stages and FIM scores of the patients before and after treatment, were compared.
Statistical analysis was performed using the Statistical Package for the Social Sciences 20.0 (SPSS, Inc.; Chicago, IL, USA) version for Windows. Normality of the continuous variables was assessed by the Kolmogorov Smirnov test. Descriptive statistics were shown as mean ± standard deviation for continuous variables and frequencies and percentages (%) for nominal variables. As none of the continuous variables was normally distributed, the Mann–Whitney U test was used for the comparison of non-normally distributed continuous variables.
On the other hand, the significance of difference for nominal variables was analyzed using Fisher exact test. Statistically significant differences in repeated measurements within the group were evaluated with the Wilcoxon Signed Rank test. The Bonferroni correction was used to control possible Type I errors in intra-group comparisons (P < 0.025). The other results were considered significant for P < 0.05.
One hundred and two subjects (51 patients with stroke and 51 healthy individuals) were taken to the study. Thirty-nine of the subjects (38.2%) were women while 63 (61.8%) were men, and the mean age was 64.50 years. The demographic features of all subjects are given in [Table 1].
We found no significant difference between the demographic features of the patients and healthy subjects (P > 0.05).
The comorbidities of the patients were hypertension in 42 (83.4%), hyperlipidemia in 21 (41.2%), cardiac disease in 13 (25.5%), and hypothyroidism in 5 (9.8%).
Left hemisphere was affected in 42 patients (82.4%). The median of the post-stroke duration of the patients was 29.50 (29.02 ± 10.61) days. In 35 (68.6%) of patients middle cerebral artery (MCA), in 16 (31.4%) posterior inferior cerebellar artery (PICA) watershed infarcts were presented. Median BDS score of the patients was 4.00, NEDS score was 5.00 and TDS score was 9.00. Dysphagia level of the patients was 4.50, according to FFES. In 46 of the patients (90.2%), dysphagia (residue, penetration and/or aspiration) was detected during swallowing.
Distribution and comparison of the electrophysiological findings of both patients and healthy subjects are shown in [Table 2].
Mean swallowing intervals was significantly longer in the patients group, while MAP amplitudes of the masseter, orbicularis oris and intrinsic tongue muscles were significantly lower than the healthy group (P < 0.005).
Distribution and comparison of swallowing functions, motor functional levels and FIM scores of the patient's group before and after treatment are presented in [Table 3].
After four weeks of treatment, significant recovery in swallowing, motor and general functional levels of the patients were provided (P < 0.025). Dysphagia level was 2.00 with FEES evaluation after treatment and residue, penetration and/or aspiration was permanent in 17 of the patients (33.3%).
The distribution and comparison of electrophysiological findings of the patients (after treatment) and healthy subjects are given in [Table 4].
Although significant recovery in swallowing intervals and MAP amplitudes of the swallowing muscles of the patients were provided after treatment, all swallowing intervals were still longer and MAP amplitudes were still lower compared with the healthy control group (P < 0.05).
Dysphagia is common especially in an early period of all stroke patients even after a hemispheric infarct or a brainstem lesion. However, recent studies have reported significant recovery in swallowing functions with reorganization and compensation,, post-stroke dysphagia may arise from structural and functional changes in oral cavity, larynx, pharynx, and esophagus. Therefore, anatomic structural features of swallowing muscles are important as well as the function of swallowing. Electrophysiological evaluation of swallowing muscles is beneficial because it allows screening both muscle activity and the function of swallowing.
The electrophysiological studies performed in stroke patients. It has been reported that the swallowing intervals are affected independently of the stroke duration. Similar to the recent data, we found significant prolonged in all swallowing intervals compared to healthy individuals.,,, Our study showed that dysphagia affects the function of swallowing in post-stroke patients, whether early or late period. Our study showed that dysphagia affects the function of swallowing in patients early or late after stroke.
Swallowing is known as a complex neuromuscular process associated with coordination of nerves and muscles, and a certain time. But although swallowing duration is normal can we consider all the process as “normal”? We have tried to find an answer to this question with the present study by evaluating masseter, orbicularis oris, and intrinsic tongue muscles MAPs.
Some recent studies, including healthy subjects and other dental causes of dysphagia except stroke have reported the importance of perioral, intrinsic tongue and submental muscles in the swallowing process., Some studies in stroke patients examined static EMG activity of submental muscles without electrical stimulation.[15–17]
This is the first study that evaluates orbicularis oris, masseter and intrinsic tongue muscles in as early as a month stroke patients. The major mortal complication of stroke is aspiration and since aspiration is known to be a pharyngeal phase disorder, there has been a first opinion about swallowing disorders in post-stroke patients as being mainly associated with this phase.
Our study emphasise that aspiration can occur before, during, or after swallowing. Moreover, an effective oral phase is essential for suitable neural network activation and for swallowing initiation and survival.,
The perioral muscle weakness or muscle incoordination in neuronal dysphagia results with the leak of the aliment from the mouth, residue in the lateral and anterior sulcus of mouth, the deficit in the formation of bolus and premature passage to the pharynx. Normal masseter muscle helps hyoid elevation and starts to contract at the beginning of the oral phase, assists pharyngeal phase muscles that are mainly active in hyolaryngeal elevation, and its activity continues until the end of swallowing., Masseter muscle weakness results with mastication dysfunction, mandibular instability and aspiration in association with insufficient suprahyoid, which are located at the floor of the mouth, co-ordinated with intrinsic tongue muscles to direct the bolus to pharynx in the oral phase of swallowing, and infra-hyoid muscle activity. And at least intrinsic tongue muscles seem to be important in swallowing process. Previous studies have results showing that intrinsic tongue muscle pressure increases submental muscle activity additively and is related to hyoid elevation.,
In our study, we found lower MAP amplitudes in orbicularis oris, masseter and tongue muscles and longer swallowing intervals compared to the healthy control group. There were only five patients who have level 1 (normal) swallowing according to FEES evaluation in patients group. These findings indicate that muscles involved in the oral phase of swallowing are in association with total swallowing process and their insufficiency may result in neurological dysphagia in post-stroke patients. Shortened swallowing intervals, increased MAP amplitudes of these muscles and improvement in pharyngeal phase abnormalities in endoscopic evaluation after rehabilitation programme are other supportive findings of our study.
There is recent data about the beneficial effects of combined rehabilitation procedures including electrical stimulation as we used in the present study., However, results are in contradiction because of different application sites and procedures. Since the potential mechanisms are muscle inactivity and the lack of neuronal stimulation; tactile, electrical, and thermal stimulations with strengthening exercises may be beneficial to prevent muscle atrophy until the reorganisation and other compensatory mechanisms. We found significant improvement in muscle dynamic activity, swallowing intervals, and functional disability with 20 days of combined treatment protocol supporting this data. Interestingly, although provided functional and electrophysiological improvement, MAP amplitudes of the patients were still lower than healthy subjects. There is no study similar to us in the literature; therefore our results have not been discussed.
Since the minor structural changes start 4 hours after stroke, significant muscle atrophy may be observed in big muscles after one week even in non-affected extremity and at least 16% strength loss may occur after 10 days of immobility, more muscle atrophy may be expected in small swallowing muscles in early post-stroke period (mean 30 days). Moreover, previous studies indicated that long term irreversible changes in muscles such as loss in muscle mass, reduction of fibre cross-sectional area, and increased intramuscular fat deposition arise in three weeks after stroke. Major limitation of our study is the lack of a screening tool to show these structural muscle changes. Further studies may be improved by imaging utilities like magnetic resonance imaging (MRI) and ultrasound (US) to indicate structural muscle changes. Another limitation of this study was the lack of an untreated group to reveal more clearly the benefits of rehabilitation.
However, MAPs electrophysiologically obtained from selected muscles that are stimulated directly from the innerving nerve seem to be a useful tool for follow up diagnosis. Endoscopic and videoflouroscopic imaging are effective tools for the diagnosis of dysphagia but electrophysiological studies including swallowing intervals and MAP amplitudes may be beneficial for the choice of treatment and give an idea about patients' possible swallowing disorders.
In conclusion, swallowing is a complex neuromuscular process and traditionally considered a three-phase process as oral, pharyngeal, and esophageal. Although this physiological separation, these phases are in relation strictly and should be considered totally. In post-stroke patients pharyngeal phase of swallowing is known to be more affected, but our findings indicate that oral phase is as important as the pharyngeal phase in post-stroke patients. Perioral, mastication and intrinsic tongue muscles in stroke patients are getting weak even in a month.
Electrophysiological study is a non-invasive, easy and cost-effective tool which evaluates both function and dynamic activity of muscles associated with the swallowing process, and it may be beneficial for diagnosis and follow up.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3], [Table 4]