Preliminary evidence of the effects of high-frequency repetitive transcranial magnetic stimulation (rTMS) on swallowing functions in post-stroke individuals with chronic dysphagia
Abstract
Background
There is growing evidence of potential benefits of repetitive transcranial magnetic stimulation (rTMS) in the rehabilitation of dysphagia. However, the site and frequency of stimulation for optimal effects are not clear.
Aims
The aim of this pilot study is to investigate the short-term effects of high-frequency 5 Hz rTMS applied to the tongue region of the motor cortex on swallowing functions and the quality of life of post-stroke individuals with dysphagia.
Methods & Procedures
Two male and two female participants were assigned randomly to active and sham groups. The participants in the active group received 10 sessions of active rTMS for 2 weeks, whereas the sham participants received 10 sessions of sham rTMS for 2 weeks. Each participant received a total of 3000 pulses of 5 Hz active or sham rTMS per day for 10 days. Outcome measures were taken at baseline, 1 week and 1 month post-rTMS.
Outcomes & Results
Participants who received active rTMS had improved swallowing functions and swallowing-related quality of life at 1 week and 1 month post-stimulation.
Conclusions & Implications
The study showed that excitatory rTMS applied over the tongue motor cortex is a feasible approach in individuals with chronic post-stroke dysphagia. Further investigation with larger sample population is warranted to support the benefit of this stimulation protocol.
What this paper adds?
What is already known on the subject?
rTMS has been a focus in rehabilitation of post-stroke dysphagia in recent years. However, the optimum protocol for rTMS for this population remains unknown due to various methodologies that studies in the literature employed. The current pilot study aimed to investigate the feasibility of using 5 Hz rTMS on the tongue motor cortex in improving swallowing functions of individuals with post-stroke dysphagia.
What this paper adds?
The current study showed that 5 Hz rTMS can improve swallowing functions and quality of life of individuals with chronic post-stroke dysphagia. This provides new insight into the stimulation target of rTMS application in improving swallowing functions. Further investigation with larger sample size is warranted to ensure the benefits of the current rTMS protocol.
Introduction
Dysphagia, which refers to the difficulties in preparation, control and/or transportation of food from the oral cavity to the stomach (Logemann 1998), is a common health problem after stroke with a reported incidence as high as 78% (Martino et al. 2005, Teasell et al. 2002) and is associated with chest infection, dehydration, malnutrition (Foley et al. 2009), as well as anxiety and depression (Eslick and Talley 2008).
Given the importance of tongue movements and coordination in swallowing, which is for food bolus control and propulsion, improvements in tongue motor functions should lead to improvements in swallowing functions. Traditional therapy for dysphagia requires the ability to follow instructions in doing exercises. This makes individuals who have language and cognitive impairments difficult to do the rehabilitative exercises which subsequently limit the recovery from dysphagia.
As such, researchers have investigated alternative treatment options for dysphagia in recent years, including repetitive transcranial magnetic stimulation (rTMS). rTMS is a non-invasive brain stimulation technique that can modulate activity in the cerebral cortex using electromagnetic field induced by an alternating current (Bailey et al. 2001, Butler and Wolf 2003). The induced current depolarizes nerve cells, which consequently disrupts or stimulates brain activity (Barker 1991). Positive results have been found in using rTMS to improve limb motor functions (Di Lazzaro et al. 2006, Quartarone et al. 2005) in post-stroke individuals.
There are two major paradigms in the application of rTMS. High-frequency (higher than 1 Hz) rTMS excites whereas low-frequency (1 Hz or lower) rTMS inhibits neural or neuronal activities. Michou et al. (2014) applied 5 Hz rTMS in chronic post-stroke dysphagic patients and found increased cortical excitability in the unaffected hemisphere after real rTMS. Improved swallowing function was observed following the application of both approaches on unilateral post-stroke individuals (Khedr et al. 2009, Khedr and abo-El Fetoh 2010, Park et al. 2012, Verin and Leroi 2009). Park et al. (2012) investigated the effect of high-frequency rTMS (5 Hz) over pharyngeal motor cortex of the unaffected hemisphere, whereas Khedr et al. (2009) and Khedr and abo-El Fetoh (2010) investigated the use of high-frequency rTMS (3 Hz) on oesophageal cortical area of the affected hemisphere of post-stroke dysphagic individuals. All studies found significant clinical recovery from dysphagia. Although the studies by Khedr et al. (2009), Khedr and abo-El Fetoh (2010) and Park et al. (2012) employed randomized controlled procedures, their results have to be interpreted with caution. These studies recruited individuals with acute stroke (less than 6 months post-onset) as well as chronic stroke participants. Spontaneous recovery from acute stroke might have contributed to the positive outcomes post-rTMS. In summary, the efficacy of using rTMS as treatment for dysphagia is still uncertain.
Given the importance of tongue movements and coordination in swallowing, the target stimulation site in the current study is the tongue region of the affected motor cortex. High-frequency rTMS is used in the current study to increase cortical excitability of the target region because significantly reduced bilateral cortical activation was found in hemispheric and brainstem stroke individuals with dysphagia (Teismann et al. 2011). This pilot study aims to investigate the short-term effects of 5 Hz rTMS on swallowing functions of individuals with chronic post-stroke dysphagia. It contributes to the understanding of possible new management option for post-stroke dysphagia.
Methods
Participants
Two men and two women (mean age = 71 years; median = 71 years; range = 67–76 years) with chronic dysphagia after stroke were included in the study. Each underwent a screening procedure and magnetic resonance imaging (MRI), in which the site of brain lesion was confirmed by a radiologist, prior to the baseline assessment. All participants were at least 2 years post-stroke, medically stable, cognitively capable to follow instructions and able to sit upright without support. All participants had impaired tongue and swallowing functions as revealed by oromotor assessment and clinical swallowing assessment carried out by a speech and language pathologist. The four participants had no history of neurological disease other than stroke, pre-existing dysphagia before stroke or oromaxillofacial surgery that involves the lips and/or tongue. The participants reported no personal or family history of seizures or epilepsy, no implanted devices and no serious medical conditions such as heart disease. They were not on medication that lower neural thresholds. This study was approved by the Institutional Review Board of the University of Hong Kong and informed consent was obtained from all participants.
Procedures
The current study employed a randomized placebo-controlled research design. The study did not have a formal power calculation, but its results will be used for power calculation of a larger scale study in future. Participants were randomly assigned to receive 10 sessions of 5 Hz active stimulation (n = 2) and to receive 10 sessions of sham stimulation (n = 2). The participants were blinded to the group assignment. The sham stimulation was performed by placing the same stimulation coil as for the active stimulation group on the scalp which was disconnected from the stimulator such that no magnetic stimulation was applied. All participants’ swallowing function was assessed at 1 week pre-, 1 week post- and 1 month post-stimulation. The two participants who received sham stimulation were given active stimulation upon completion of the 1-month post-assessment and were reassessed at 1 week and 1 month after receiving active stimulation. Thus, this is a crossover study in these two participants.
Motor evoked potentials (MEPs)
A neuronavigational system, Brainsight™ (Rogue Research, Montreal, QC, Canada), was used to provide real-time visual feedback on the positioning of the stimulation coil over the head during stimulation sessions. This software, together with the participant's MRI markings, were used to locate the target area, which was the tongue area of the primary motor cortex of the affected hemisphere.
To ensure that the stimulation was targeted on the desired tongue motor cortex and that the neuropathway from the cortex to the tongue muscles was intact, the MEPs of the tongue were obtained before the first stimulation session. The tongue motor cortex was first marked on the scalp by measuring the point that was 9 cm lateral and 3 cm anterior to the vertex (Rodel et al. 2003). Single TMS pulses starting with 50% power were used to elicit the MEPs. The TMS coil was placed tangentially on the scalp, over the earlier marked tongue cortex area. MEPs of the tongue were recorded by the custom-made bipolar surface electromyography (EMG) electrodes that were connected to the Brainsight (Rogue Research). A single EMG electrode was placed on the upper surface of the tongue, on the side that was contralateral to the stimulated hemisphere and approximately 1 cm from the tongue tip. Non-toxic edible glue (Cyano-Veneer Fast, Hager and Werken, Duisburg, Germany) was used to attach the electrode to the tongue surface. The site of stimulation and the minimum intensity of cortical (TMS) pulse used to elicit the tongue MEP were recorded and used for later stimulation.
Repetitive transcranial magnetic stimulation (rTMS)
Each participant received a total of 3000 pulses 5 Hz rTMS per session for 2 weeks (10 sessions). The duration of each session was approximately 30 min. Active stimulations were carried out using a Magstim Rapid2 stimulator and an air-cooled double 70 mm stimulating coil (Magstim, Carmarthenshire, Wales, UK). The stimulation intensity was at 90% of the resting motor threshold obtained. Sham condition was introduced as a control by placing the same stimulation coil, which was disconnected to the stimulator such that it did not give out magnetic stimulation on the scalp. A separate figure of eight, 70 mm diameter stimulation coil (Magstim) was connected to the TMS system and used to produce the audible click. All participants were not able to see the coils, the TMS system or the BrainsightTM systems.
Outcome measurements
Each participant undertook one baseline assessment (1 week before stimulation) and two post-stimulation assessments (1 week and 1 month post-stimulation respectively). The three components of each assessment were: tongue pressure assessment; swallowing-related quality of life questionnaire; and videofluoroscopic swallowing study.
Tongue pressure assessment
The Iowa Oral Performance Instrument (IOPI; IOPI Medical, Washington state, USA) was used to measure the peak tongue pressure (Pmax) under voluntary control. The participants were asked to press an air-filled bulb, which was placed behind the participants’ incisors, as hard as possible for three trials. The maximum pressure in each trial was recorded. The higher the peak tongue pressure, the greater the force the tongue can exert.
Swallowing-related quality of life questionnaire
The Swallowing Activity and Participation Profile (SAPP) is a validated self-report questionnaire that investigates the quality of life of dysphagic individuals (Chan et al. 2011). A total of 38 items are included, each rated on an equal-appearing interval scale. The scale is based on the International Classification of Functioning, Disability and Health (ICF; World Health Organization) framework which evaluates the impact of disability from physical, emotional and social perspectives. The maximum score is 340 points. A higher score indicates a higher impact on the individual.
Videofluoroscopic swallowing study (VFSS)
VFSS, which utilizes X-ray to visualize the swallowing process, was conducted. Each participant was asked to swallow three trials of 10 ml thin liquid, 10 ml honey-thick fluid and 10 ml paste, all mixed with barium sulphate, upon signals given by the clinician. The barium-coated food appears as opaque mass that runs through the oral cavity towards the stomach of the participants under X-ray exposure. The lateral views of the oropharynx and larynx were examined. From the VFSS video, the oropharyngeal swallow efficiency (OPSE) (Logemann et al. 1989) was calculated. It is a global measure for swallowing functions, which quantifies the ability to move food bolus from oral cavity and pharynx to the oesophagus safely and efficiently (Rademaker et al. 1994). The formula is listed in appendix A. An increase in OPSE reflects a more efficient and safer movement of food bolus from oral cavity into the oesophagus. The OPSE of non-dysphgic individuals ranges from 78 to 79 (Logemann et al. 1989, Rademaker et al. 1994).
Results
All participants tolerated the 10-day stimulation without any physical or mental discomfort.
Case 1
This is a 72-year-old man who had right cerebellar stroke 9 years before the study. He was on nasogastric tube for 2 years and on percutaneous endoscopic gastrostomy feeding tube for 1 year after discharge from hospital. At the time he participated in the study he was on pureed meat soft rice diet orally, and drank medium thick fluid.
Baseline assessment revealed that his maximum tongue strength was 21 kPa. Clinical bedside swallowing assessment revealed that he showed dripping of food, prolonged oral transit and multiple swallow. Analysis of VFSS video showed that his OPSE score for thin liquid, honey-thick fluid and paste were 19%, 2% and 13% per second respectively. The SAPP total score was 277 points.
He was assigned to receive 10 sessions of active 5 Hz rTMS on the left tongue motor cortex at 55% of maximum stimulator output. At 1 week and 1 month post-stimulation follow-up assessment, his maximum tongue strength was 17 kPa (decreased by 19% compared with the baseline) and 19 kPa (decreased by 10% compared with the baseline) respectively. Clinical bedside assessment revealed that he had similar dysphagia symptoms compared with the baseline. At 1 month post-stimulation, his OPSE scores for thin liquid, honey-thick fluid and paste were 8% per second (decreased by 58% compared with the baseline), 14% per second (increased by 600% compared with the baseline) and 25% per second (increased by 92% compared with the baseline) respectively. The SAPP total scores at 1 week and 1 month post-stimulation were 121 points (decreased by 56% compared with baseline) and 101 points (decreased by 64% compared with the baseline) respectively.
Case 2
This is a 76-year-old man who had right parietal and left frontal lobe stroke 5 years before the study. At the time he participated in the study he was on regular oral diet.
Baseline assessment revealed that his maximum tongue strength was 25 kPa. Clinical bedside swallowing assessment revealed that he showed gurgly voice and coughing after swallow and oral residue after swallow. The SAPP total score was 68 points.
He was assigned to receive 10 sessions of active 5 Hz rTMS on the right tongue motor cortex at 55% of maximum stimulator output. At 1 week and 1 month post-stimulation follow-up assessment, his maximum tongue strength was 26 kPa (increased by 4% compared with the baseline) and 36 kPa (increased by 40% compared with the baseline) respectively. Clinical bedside assessment revealed that he had similar dysphagia symptoms compared with baseline. The SAPP total scores at 1 week and 1 month post-stimulation were 12 points (decreased by 82% compared with the baseline) and 44 points (decreased by 35% compared with the baseline) respectively.
Case 3
This is a 67-year-old woman who had left upper medulla stroke 3 years before the study and a history of nasopharyngeal cancer. She received radiotherapy in 1979 and suffered from xerostomia after the treatment. At the time she participated in the study she was on soft diet orally and drank thin liquid.
Baseline assessment revealed that her maximum tongue strength was 18 kPa. Clinical bedside swallowing assessment revealed that she showed prolonged oral transit, multiple swallow and gurgly voice after swallow. The SAPP total score was 133 points.
She was assigned to received 10 sessions of sham rTMS on the left tongue motor cortex at 55% of maximum stimulator output. At 1 week and 1 month post-stimulation follow-up assessment, her maximum tongue strength was 21 kPa (increased by 20% compared with the baseline) and 20 kPa (increased by 11% compared with the baseline) respectively. Clinical bedside assessment revealed that she had similar dysphagia symptoms compared with baseline. The SAPP total score at 1 week and 1 month post-stimulation were 160 points (increased by 20% compared with the baseline) and 152 points (increased by 14% compared with the baseline) respectively.
She was given 10 sessions of active 5 Hz rTMS upon completion of 1 month post-stimulation. rTMS was applied on her left tongue motor cortex at 55% of maximum stimulator output. Baseline assessment revealed that her maximum tongue strength was 20 kPa. Clinical bedside assessment revealed that she had prolonged oral transit, reduced laryngeal elevation, multiple swallow and oral residue after swallow. Analysis of VFSS video showed that her OPSE score for thin liquid, honey-thick fluid and paste were 17%, 21% and 7% per second respectively. The SAPP total score was 189 points.
At 1 week and 1 month post-stimulation follow-up assessment, her maximum tongue strength was 24 kPa (increased by 20% compared with the pre-active stimulation baseline) and 21 kPa (increased by 5% compared with the pre-active stimulation baseline) respectively. Clinical bedside assessment revealed similar dysphagia symptoms as in baseline assessment before active stimulation. At 1 month post-stimulation, her OPSE score for thin liquid, honey-thick fluid and paste were 67% per second (increased by 294% compared with the pre-active stimulation baseline), 43% per second (increased by 105% compared with the pre-active stimulation baseline) and 8% per second (increased by 14% compared with the pre-active stimulation baseline) respectively. The SAPP total scores at 1 week and 1 month post-stimulation were 42 points (decreased by 78% compared with the pre-active stimulation baseline) and 131 points (decreased by 31% compared with the pre-active stimulation baseline) respectively.
Case 4
This is a 70-year-old woman who had bilateral parietal and right thalamic stroke 6 years before the study. At the time she participated in the study she was on soft diet orally and drank thin liquid.
Baseline assessment revealed that her maximum tongue strength was 24 kPa. Clinical bedside swallowing assessment revealed that she showed throat clearing and oral residue after swallow. Her OPSE score for thin liquid, honey-thick fluid and paste are 44%, 29% and 11% per second respectively. The SAPP total score was 47 points.
She was assigned to the sham group and received 10 sessions of sham rTMS on the left tongue motor cortex at 55% of maximum stimulator output. At 1 week and 1 month post-stimulation follow-up assessment, her maximum tongue strength was 28 kPa (increased by 17% compared with the baseline) and 30 kPa (increased by 25% compared with the baseline). Clinical bedside assessment revealed similar dysphagia symptoms as in the baseline. At 1 month post-stimulation, her OPSE score for thin liquid, honey-thick fluid and paste were 22% per second (decreased by 50% compared with the baseline), 17% per second (decreased by 41% compared with the baseline) and 6% per second (decreased by 45% compared with the baseline) respectively. The SAPP total score at 1 week and 1 month post-stimulation were 44 points (decreased by 6% compared with the baseline) and 114 points (increased by 143% compared with the baseline) respectively.
She was given 10 sessions of active 5 Hz rTMS upon completion of 1 month post-stimulation. rTMS was applied on her left tongue motor cortex at 55% of maximum stimulator output. Baseline assessment revealed that her maximum tongue strength was 30 kPa. Her OPSE score for thin liquid, honey-thick fluid and paste were 22%, 17% and 6% per second respectively. The SAPP total score was 114 points.
At 1 week and 1 month post-stimulation follow-up assessment, her maximum tongue strength was 37 kPa (increased by 23% compared with the pre-active stimulation baseline) and 36 kPa (increased by 20% compared with the pre-active stimulation baseline) respectively. Clinical bedside assessment revealed that she had similar dysphagia symptoms compared with baseline before active stimulation. At 1 month post-stimulation, her OPSE score for thin liquid, honey-thick fluid and paste increased by 83% per second (increased by 278% compared with the pre-active stimulation baseline), 66% per second (increased by 288% compared with the pre-active stimulation baseline) and 43% per second (increased by 617% compared with the pre-active stimulation baseline) respectively. The SAPP total scores at 1 week post-stimulation were 182 points (increased by 60% compared with the pre-active stimulation baseline) and 84 points (decreased by 26% compared with the pre-active stimulation baseline) respectively.
Discussion
The present study investigated the short-term effects of high-frequency rTMS on swallowing functions and quality of life of four patients with post-stroke dysphagia. All the participants had similar to baseline swallowing performance at the 1 week and 1 month follow-up assessment when the clinical bedside assessment was employed, but when instrumental assessments were employed the results showed that neurostimulation benefited the patients. The results showed improvements in OPSE and quality of life in individuals who received active stimulation, compared with no improvement in those who received sham stimulation. The findings suggest that the application of 5 Hz rTMS in individuals with post-stroke dysphagia may help to improve swallow functions and quality of life. Moreover, these improvements were maintained up to 1 month after stimulation.
One of the major findings of this pilot study was improvement in the OPSE in participants who received active stimulation but not those who received sham stimulation. The improved OPSE suggest that the participants were able to swallow more efficiently and safely after active stimulation. Moreover, the improvements in swallowing functions were observed across bolus consistencies (thin liquid, honey-thick fluid and paste), which is clinically important as one may encounter different food consistencies in a meal.
Another positive finding from the current study was the improvements in quality of life of the participants who received active stimulation. This is evidenced from the reduction in the SAPP total scores in the participants who received active stimulation but not those who received sham stimulation. The changes in self-perception of the swallowing impairments may be a result of improved swallowing functions after application of rTMS. This is an encouraging finding because not only did the instrumental measurement show functional improvements after active stimulation, but the participants also perceived an improvement in the quality of life after treatment via rTMS. Improving one's quality of life is as important as improving one's swallowing functions in rehabilitation of post-stroke dysphagia.
Unexpectedly, there is a lack of observable differences in changes in maximum tongue strength between participants who received active stimulation and those who received sham stimulation. The dissociation between performances in swallowing functions and tongue strength may suggest that the increase in cortical excitability after 5 Hz rTMS may result in better tongue movements, but not increasing tongue strength. It may be possible that increasing tongue strength is better achieved by building up of tongue muscles through isometric exercises. It was found that increased tongue strength is associated with increased tongue volume (Robbins et al. 2007). Michou et al. (2014) proposed an alternative treatment method, paired associative stimulation (PAS), in which electrical stimulation is applied to the muscles and TMS is applied to the corresponding cortical representation. Their results showed more significant improvements in penetration-aspiration scores after PAS than after rTMS. This suggests that swallowing treatment may be more effective when combining both neuromodulation techniques and peripheral stimulation than using neuromodulation techniques alone. Thus, the peripheral input may also be crucial in swallowing rehabilitation. Future research direction may thus be on whether combination of tongue isometric exercise and rTMS can maximize oropharyngeal dysphagia rehabilitation.
Recommendations for further studies
The current study suggested that the tongue motor cortex is a feasible target for application of 5 Hz rTMS in individuals with post-stroke dysphagia. Although the present study demonstrated encouraging positive results in improving swallowing functions and quality of life in post-stroke dysphagic patients, the particular mechanism by which neuromodulation that leads to the observed improvements are unknown. One possible explanation for the improvements in swallowing functions is the changes in cortical excitability subsequent to the application of excitatory, high-frequency rTMS. Increases in cortical excitability by application of 5 Hz rTMS may increase stimulation to the motor neurons in the corticobulbar and corticospinal tracts, which enhances the synaptic innervations that project to the tongue muscles, leading to more coordinated tongue movements, and subsequently more efficient swallowing coordination and functions. Further investigations on neurological changes after stimulation are needed to delineate the underlying neuromechanism of the observed improvements. Functional imaging techniques may provide information about neurological activities of the cortex during swallowing. Further, measurements of tongue MEPs subsequent to rTMS may also provide valuable evidence to elucidate the effect of neuromodulation of excitability of the motor cortex.
Limitations
The major limitation of the current study is the small sample size. Only four participants were included in the study, in which only two of them received sham stimulation. The participants recruited in the study had different site and size of lesions (e.g., subcortical or brainstem stroke). The current study cannot draw any conclusions on the effects of rTMS on different types of stroke. Other studies have found positive results in improving swallowing functions when rTMS is applied to cortical, subcortical or brainstem stroke population. This may suggest that rTMS can be applied to non-cortical stroke population (Khedr et al. 2009, Khedr and abo-El Fetoh 2010, Park et al. 2012, Verin and Leroi 2009). Another limitation was that the stimulated site was not chosen based on bilateral motor mapping, which would have indicated whether the affected or unaffected neural pathway was more appropriate for stimulation. In future studies, the MEPs may first be obtained from both hemisphere such that the side with lower MEP obtained will be selected for stimulation. Moreover, the duration of follow-up assessment is limited in the current study. The maintenance of the observed improvements beyond 1 month post-stimulation remains undefined. The last limitation concerns the equipment used in the study. In the current study, sham coil was not used to give sham stimulation such that the sham condition in terms of sensation over the patients’ scalp and the audible clicks given out by the coil might not be comparable to that of the experimental condition. Therefore, use of sham coil would be recommended in further studies.
Conclusions
The results of the current preliminary study suggest that the application of 5 Hz rTMS over the tongue motor cortex could improve swallowing functions and quality of life for post-stroke dysphagic individuals. The present study provides new insight and preliminary positive evidence to support the application of rTMS to the tongue area of the primary motor strip in the treatment of chronic post-stroke dysphagia.
Acknowledgements
The study was supported by the Seed Funding for Basic Research, The University of Hong Kong (Grant no. 201202159004).
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Appendix A: Formula of oropharyngeal swallow efficiency (OPSE)
