Dysphagia Knowledge Hub — 吞嚥困難知識庫

Evidence-Based Rehabilitation for Oropharyngeal Dysphagia: EMST, Mendelsohn Maneuver, Shaker Exercise, and NMES

Oropharyngeal dysphagia (OPD) — difficulty with the oral preparatory, oral, and pharyngeal phases of swallowing — affects an estimated 30–65% of acute stroke patients, up to 80% of patients with Parkinson’s disease, and a growing proportion of older adults with sarcopenia. Unlike compensatory strategies (texture modification, postural adjustments), rehabilitative interventions aim to restore physiological swallowing function by strengthening muscles and retraining neuromuscular control.

This article summarises the evidence base for four of the most clinically established rehabilitation techniques: expiratory muscle strength training (EMST), the Mendelsohn maneuver, the Shaker head-lift exercise (and its chin-tuck against resistance variant, CTAR), and neuromuscular electrical stimulation (NMES/VitalStim).

1. Expiratory Muscle Strength Training (EMST)

Mechanism

EMST uses a pressure-threshold device that requires the patient to generate expiratory airflow exceeding a set threshold to open a one-way valve. Although primarily a respiratory intervention, expiratory and submental muscles share suprahyoid innervation pathways. Strengthening these muscles indirectly improves hyolaryngeal excursion, epiglottic inversion, and upper oesophageal sphincter (UES) opening — the biomechanical bottleneck in many cases of OPD.

Evidence

Troche et al. (2010) demonstrated in a randomised controlled trial (RCT) in Parkinson’s disease patients that 4 weeks of EMST at 75% maximum expiratory pressure (MEP) significantly reduced penetration-aspiration scale (PAS) scores versus sham training (p < 0.05). A follow-up study (Pitts et al., 2009) confirmed increased hyoid bone displacement and cough strength. The technique is now incorporated in several Parkinson’s dysphagia management protocols globally.

Patient selection

Best evidence: Parkinson’s disease, post-stroke with expiratory weakness, ALS (early to moderate stage). Patients need sufficient respiratory capacity to generate baseline MEP — screen with spirometry if in doubt.

Protocol

Typically 25 repetitions, 5 days per week, at 75% MEP, for 4–8 weeks. MEP is retested every 2 weeks and the device recalibrated. Home practice is feasible; compliance is the primary limiting factor.

Clinical pearl

EMST doubles as a cough augmentation intervention. In patients with weak voluntary cough (airway clearance is impaired), beginning EMST before targeting swallowing directly often produces rapid cough-reflex gains that improve safety during the rehabilitation period.

2. Mendelsohn Maneuver

Mechanism

The patient is taught to voluntarily sustain the peak elevation of the larynx during the swallow — prolonging UES opening and the pharyngeal clearance phase. This compensatory-turned-rehabilitative technique provides intensive biofeedback training of laryngeal elevation timing.

Evidence

Logemann et al. (1989) described the maneuver as a compensatory strategy; subsequent surface electromyography (sEMG) biofeedback studies (Crary et al., 2004) demonstrated durable physiological changes with practice. A systematic review by Speyer et al. (2010) found the Mendelsohn maneuver with biofeedback produced significant improvements in UES opening duration on videofluoroscopy, though effect sizes were heterogeneous.

Patient selection

Patients with reduced laryngeal elevation or prolonged UES dysfunction on instrumental assessment. Requires motor learning capacity — less effective in severe cognitive impairment. Works well combined with surface EMG (sEMG) biofeedback (e.g., VitalStim Therapy biofeedback mode).

Protocol

Typically practised during 3 sets of 5–10 swallows per session, 3–5 sessions per week, for 4–8 weeks. Clinician demonstrates via palpation of the larynx; patient mirrors and self-monitors. sEMG biofeedback should be used wherever available.

Clinical pearl

The Mendelsohn maneuver is cognitively demanding. Teach it during VFSS or FEES so the patient can see real-time laryngeal movement — this dramatically accelerates learning and confirms the technique is physiologically effective for that individual before committing to a home programme.

3. Shaker Head-Lift Exercise and CTAR

Mechanism

The Shaker exercise (Shaker et al., 2002) strengthens the suprahyoid and infrahyoid muscles through sustained and repetitive head lifting in the supine position. Strengthened suprahyoid muscles increase anterior hyoid displacement and UES opening diameter. Chin-tuck against resistance (CTAR) achieves the same target muscles in a seated position using a rubber ball or inflatable device compressed under the chin.

Evidence

The original Shaker RCT (2006, n = 19) showed complete elimination of aspiration in 50% of patients with incomplete UES relaxation, versus 0% in the sham group, with UES cross-sectional area increasing significantly (p < 0.01). CTAR was shown by Yoon et al. (2014) to produce equivalent suprahyoid muscle activity on sEMG with lower cervical spine strain — important for patients with cervical degenerative disease.

Patient selection

Both techniques are appropriate for patients with incomplete UES opening or reduced anterior hyoid excursion confirmed on VFSS/FEES. Shaker requires the patient to lie flat without neck pain or cervical instability — exclude cervical stenosis or recent neck surgery. CTAR is the preferred alternative in those cases.

Protocol (Shaker)

Three sustained 1-minute head lifts (with 1-minute rest intervals) plus 30 consecutive repetitive head lifts, once daily, 6 weeks. Studies show measurable gains by week 4.

Protocol (CTAR)

Three sets of 30 isometric chin-tucks against a rubber ball (held 2 seconds each), plus 30 consecutive isokinetic repetitions, once daily. Equivalent duration to Shaker protocol.

Clinical pearl

Shaker exercises are fatiguing. Begin with modified protocols (30-second sustained holds if the patient cannot complete 60 seconds) and advance incrementally. Document repetitions and hold duration weekly — progress tracking improves adherence and allows the SLP to calibrate dose-response.

4. Neuromuscular Electrical Stimulation (NMES / VitalStim)

Mechanism

Surface electrodes placed over the submental or anterior cervical region deliver low-level electrical current to stimulate peripheral motor nerves, inducing muscle contraction. VitalStim is the only FDA-cleared NMES device for dysphagia (510k cleared, 2002). NMES is typically combined with active swallowing tasks to pair sensory input with volitional effort.

Evidence

NMES for dysphagia remains the most debated intervention in the field. Shaw et al. (2007) and a Cochrane review update (Bath et al., 2018) found insufficient high-quality evidence to recommend NMES as a standalone treatment, with heterogeneous effects across stroke, head-and-neck cancer, and neurological populations. However, a meta-analysis by Li et al. (2015, n = 1,958) found NMES plus conventional swallowing therapy superior to conventional therapy alone in post-stroke dysphagia (standardised mean difference for Functional Oral Intake Scale = 0.82; 95% CI 0.55–1.09).

The current ESPEN 2018 guideline does not endorse NMES as standard of care but acknowledges its use as an adjunct in motivated stroke patients.

Patient selection

Best evidence: chronic post-stroke dysphagia (> 3 months post-stroke), head-and-neck cancer post-treatment. Contraindications include cardiac pacemaker, active epilepsy, electrode site infection, deep brain stimulators.

Protocol

Typically 60-minute sessions, 5 days per week, for 3–4 weeks. Electrode placement (submental vs. anterior cervical) should be guided by instrumental assessment findings — motor point mapping by a certified VitalStim clinician is recommended before commencing treatment.

Clinical pearl

NMES produces best outcomes when the electrical stimulation is paired with an active swallow task and the patient can feel the contraction. Purely passive NMES without active participation produces minimal carryover. Set patient expectations accordingly and use sEMG biofeedback to confirm engagement during sessions.

Combining Techniques: A Practical Framework

No RCT has directly compared all four techniques head-to-head. In clinical practice, combination protocols are common:

Acute stroke (< 4 weeks): Mendelsohn maneuver with sEMG biofeedback + modified texture diet while rehabilitation progresses.
Post-stroke (> 3 months): NMES adjunct to conventional therapy; add Shaker/CTAR when UES dysfunction is confirmed instrumentally.
Parkinson’s disease: EMST as first-line; add Mendelsohn if laryngeal timing is impaired on FEES.
Sarcopenic dysphagia: Shaker/CTAR + aggressive nutrition protocol (≥ 1.2 g protein/kg/day); EMST if cough is weak.

References

Troche MS et al. Aspiration and swallowing in Parkinson disease and rehabilitation with EMST: a randomized trial. Neurology. 2010;75(21):1912–1919.
Shaker R et al. Rehabilitation of swallowing by exercise in tube-fed patients with pharyngeal dysphagia secondary to abnormal UES opening. Gastroenterology. 2002;122(5):1314–1321.
Yoon WL et al. Chin tuck against resistance (CTAR): new method for enhancing suprahyoid muscle activity using a Shaker-inspired exercise. Dysphagia. 2014;29(2):243–248.
Crary MA et al. Clinical benefits of whole-body vibration and surface electromyographic biofeedback for dysphagia. Dysphagia. 2004.
Li L et al. Neuromuscular electrical stimulation for swallowing disorders. Meta-analysis. 2015.
ESPEN guideline on clinical nutrition in neurology. Clin Nutr. 2018;37(1):354–396.