Neuromuscular electrical stimulation (NMES) during swallowing training is a technique aimed at strengthening the suprahyoid muscles in particular. This is intended to enhance hyolaryngeal elevation and thereby promote a more efficient and safer pharyngeal swallowing physiology (Humbert, 2012). However, it remains unclear whether and under which conditions the application of electrical stimulation provides added value, and as a result, its use remains limited.

Summary of Findings

PICO (1.1)

Population: Stroke patients

Intervention: NMES+TDT

Comparator: TDT

_{^1. Risk of Bias: serious. Inadequate/lack of blinding of participants and personnel, resulting in potential for performance bias; Imprecision: serious. Left boundary of the 95% CI cross the threshold of clinical relevance;}

_{^2. Risk of Bias: serious. No information about concealment of allocation during randomization process in one study; Inconsistency: serious. Point estimates vary widely; Imprecision: serious. The lower boundary of the 95% CI overlaps with clinical relevance threshold;}

_{^3. Risk of Bias: serious. Inadequate sequence generation/ generation of comparable groups, resulting in potential for selection bias, Deviation from the intended interventions; Imprecision: very serious. Wide confidence intervals (threshold of clinical relevance overlaps with both boundaries of the 95% CI);}

_{^4. Risk of Bias: serious. Inadequate/lack of blinding of participants and personnel, resulting in potential for performance bias, Other bias ; Imprecision: serious. Wide confidence intervals (left boundary of the 95% CI crosses the border of clinical relevance);}

_{^5. Risk of Bias: serious. Inadequate concealment of allocation during randomization process, resulting in potential for selection bias, Deviation from the intented intervention; Imprecision: serious. Wide confidence intervals (left boundary of the 95% CI crosses border of clinical relevance);}

_{^6. Risk of Bias: serious. Inadequate/lack of blinding of participants and personnel, resulting in potential for performance bias, Other bias; Imprecision: serious. Wide confidence intervals (left boundary of 95% CI crosses the threshold of clinical relevance);}

_{^7.Risk of Bias: serious. Inadequate sequence generation/ generation of comparable groups, resulting in potential for selection bias; Imprecision: serious. Wide confidence intervals (left boundary of the 95% CI crosses threshold of clinical relevance).}

PICO (1.2)

Population: head and neck cancer patients with dysphagia

Intervention: NMES+TT or NMES only

Comparator: sham NMES+exercises or exercises only

_{^1. Risk of Bias: very serious. Bias arising from the randomization process; Imprecision: serious. Both boundaries of the 95% CI cross the threshold of clinical relevance;}

_{^2. Risk of Bias: serious. Inadequate sequence generation/ generation of comparable groups, resulting in potential for selection bias; Imprecision: serious. Wide confidence interval;}

_{^3. Risk of Bias: serious. Missing intention-to-treat analysis; Imprecision: serious. Left boundary of the 95% CI crosses the threshold of clinical relevance;}

_{^4. Risk of Bias: serious. Imprecision: very serious. Low number of patients (<200), Wide confidence intervals (left boundary of the 95% CI crosses the threshold of clinical relevance).}

PICO (1.3)

Population: Parkinson’s Disease patients

Intervention: NMES+exercises

Comparator: sham NMES+exercises

_{^1. Risk of Bias: serious. Imprecision: very serious. Wide confidence intervals, Low number of patients (<200), Only data from one study.}

Description of studies

A total of five studies were included in the analysis of the literature (n=2 for post-stroke patients, n=2 for head and neck cancer (HNC) patients, and n=1 for Parkinson’s Disease patients). Important study characteristics and results are summarized in table 2. The assessment of the risk of bias is summarized in the risk of bias tables (under the tab ‘Evidence tabellen’).

Post-stroke dysphagia patients and traumatic brain injury

Wang (2023) conducted a systematic review (SR) and meta-analysis (MA) to compare the efficacy of neuromuscular electrostimulation (NMES) on top of swallowing therapy (ST) with ST alone in patients with post-stroke dysphagia. Eight databases were systematically searched up to 9 June 2022 (CNKI, Wanfang, VIP, SinoMed, PubMed, Embase, Cochrane Library, and Web of Science). Fourty-six randomized controlled trials (RCTs) with patients diagnosed with post-stroke dysphagia (no other neurological disease) and the same ST intervention performed in the experimental and control groups except the experimental group that received NMES, were eligible for inclusion. Outcomes relevant for the purpose of this guideline included swallowing-related quality of life (measured by the Swallowing Quality of Life questionnaire [SWAL-QOL]), swallowing function (measured by FOIS and FDS), and aspiration (measured by PAS).

Additionally, Banda (2023) conducted a network meta-analysis to evaluate the comparative effectiveness of combined and single neurostimulation and traditional dysphagia therapies (TDT) for post-stroke dysphagia. In august 2022, a systematic search was conducted across six databases (Embase, PubMed, Ovid-Medline, CINAHL, Cochrane Library, Web of Science). Additionally, reference lists of published systematic reviews and meta-analyses were searched. Eligible studies were RCTs involving adults with post-stroke dysphagia comparing the effect of different neurostimulation therapies on swallowing function and aspiration. Both standalone and combination therapies were analyzed.

HNC dysphagia patients

Paim (2023) conducted a SR and MA to evaluate the effectiveness of NMES in the rehabilitation of dysphagia following treatment for HNC. A comprehensive search was performed in MEDLINE/PubMed, Embase, Cochrane CENTRAL, SciELO, and PEDro, from database inception to September 2021. Clinical and quasi-clinical trials involving adult patients (>18y) with objectively confirmed dysphagia after radiotherapy for HNC, comparing NMES (either as standalone treatment or combined with swallowing exercises) with sham stimulation, usual care or exercises alone, were included. Outcomes relevant for the purpose of this guideline included functional oral intake of food as experienced by the patient (measured by the American speech-language-hearing association national outcome measurement system swallowing level scale, ASHA NOMS), swallowing-related quality of life (measured by the Head and Neck Cancer Inventory [HNCI], MD Anderson Dysphagia Inventory [MDADI], and the functional assessment after cancer therapy-nasopharyngeal [FACT-NP]), functional swallowing ability (measured by the functional dysphagia scale [FDS]), and aspiration (measured by the Penetration-Aspiration-Scale [PAS]).

Ku (2023) conducted an RCT to compare the effects of transcutaneous neuromuscular electrical stimulation (TNMES) and exercise-based swallowing training (EBST) when administered early after radiotherapy for nasopharyngeal carcinoma (NPC). Eligible patients were adults (>18y) of Chinese ethnicity with a histological diagnosis of NPC who had completed primary radiotherapy or chemoradiotherapy ≤3 months prior and were confirmed to be in remission by negative biopsy. Patients with a history of other head and neck cancers treated with radiotherapy/chemotherapy/surgery, synchronous malignancies, residual or recurrent NPC, other causes of dysphagia, preexisting dysphagia before radiotherapy, contraindications to electrical stimulation, neurological disorders, poor cognitive function, inability to attend training or follow-up were excluded. A total of 111 eligible patients were randomized to receive either TNMES (n = 50) or EBST (n = 57). Both interventions consisted of 12 supervised 40-minute sessions over four weeks, without home-based exercises during the study period. Swallowing function was assessed at baseline, immediately after treatment, and at 6 and 12 months. Relevant outcomes for the purpose of this guideline included PAS, FOIS and QoL (measured by FACT-NP).

Patients with Parkinson’s Disease

Cheng (2023) conducted a SR and MA to evaluate the effectiveness of dysphagia interventions in patients with Parkinson’s Disease. Comprehensive searches were performed in PubMed, EMBASE (Ovid), CINAHL, Web of Science, and Cochrane Library from database inception to April 2022. Reference lists of relevant papers were also screened. Eligible studies were RCTs with parallel or crossover designs, involving adults with Parkinson’s Disease and clinically confirmed dysphagia, comparing any form of dysphagia treatment (including behavioral therapy, stimulation treatment, or pharmacological therapy) with sham, no treatment or conventional dysphagia therapy. Nine RCTs involving 286 patients with Parkinson’s Disease met the inclusion criteria, of which one study covered NMES and was relevant for the purpose of this guideline. Relevant outcomes included PAS.

Table 2. Characteristics of included studies

_{*For further details, see risk of bias table in the appendix}

Results

Stroke patients     

Comparison: NMES+TDT vs. TDT only

Functional oral intake of food as experienced by the patient (crucial)

Thirteen studies reported the outcome measure functional oral intake of food as experienced by the patient.

Five RCTs included in the SR of Wang (2023) reported the outcome using the Functional Oral Intake Scale (FOIS) (scale 1-7, higher is better) (Huang, 2014; Lee, 2014; Simonelli, 2019; Sproson, 2018; Zhang, 2016). Five RCTs included in the NMA of Banda (2023) were added (Bucyana, 2019; Cakmak, 2022; Cola, 2021; Matos, 2022; Permsirivanich, 2009). In total, 153 patients received NMES on top of TDT, and 145 patients received TDT only (Figure 1). The MD was 0.87 points higher (95% CI 0.41 higher to 1.33 higher), in favor of the NMES+TDT group and clinically relevant.

_{Figure 1. Functional oral intake (measured with the functional oral intake scale, FOIS) in stroke patients receiving NMES+TDT versus TDT}

_{Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation}

Two RCTs included in the NMA of Banda (2023) reported the outcome using the Mann Assessment of Swallowing Ability (MASA) up to four weeks posttreatment (scale 38-200, higher is better) (Carnaby, 2020; Umay, 2017). Carnaby (2020) reported a mean of 172.7 (SD 16.7) points in the NMES+TDT group (n=17) vs. 171.7 (SD 17.3) points in the TDT only group (n=16). This corresponded to a calculated MD of 1.00 point higher (95% CI, 10.61 lower to 12.61 higher) in favor of the NMES+TDT group, but not clinically relevant. Umay (2017) reported a mean of 181.27 (SD 20.66) points in the NMES+TDT group (n=58) vs. 157.81 (SD 33.58) points in the TDT only group (n=40). This corresponded to a calculated MD of 23.46 points higher (95% CI, 11.77 higher to 35.15 higher) in favor of the NMES+TDT group and clinically relevant.  

One RCT included in the NMA of Banda (2023) reported the outcome using the Dysphagia Outcome and Severity Scale (DOSS) up to two weeks (scale 1-7, higher is better) (Meng, 2017; Zhang, 2021). Meng (2017) reported a mean of 5.2 (SD 1.4) points in the NMES+TDT group (n=10) vs. 4.9 (SD 1.1) points in the TDT only group (n=10). This corresponded to a calculated MD of 0.30 points higher (95% CI, 0.80 lower to 1.40 higher) in favor of the NMES+TDT group, but not clinically relevant.

Swallowing-related quality of life (crucial)

Eighteen studies reported the outcome measure swallowing-related quality of life.

Sixteen RCTs included in the SR and MA of Wang (2023) reported the outcome using the Swallowing Quality of Life Questionnaire (SWAL-QOL) (scale 0-220, higher is better) (Chang, 2020; Chen, 2017; Chen, 2021; Cui, 2018; Dong, 2017; Gu, 2021; Guo, 2019; Lei, 2020; Li, 2021; Mo, 2022; Shi, 2021; Sproson, 2018; Wang, 2009; Wang, 2021; Wang, 2021(b), Wen, 2020). One RCT included in the NMA of Banda (2023) was added (Cakmak, 2022). In total 709 patients received NMES on top of TDT, and 705 patients received TDT only (Figure 2). The MD was 23.31 points higher (95% CI 17.23 higher to 29.39 higher), in favor of the NMES+TDT group and clinically relevant.

_{Figure 2. Swallowing-related quality of life (measured with the swallowing quality of life questionnaire, SWAL-QOL) in stroke patients receiving NMES+TDT versus TDT Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation}

Cakmak (2022) also reported the outcome dysphagia symptom severity using the Eating Assessment Tool-10 (EAT-10) up to three months (scale 0-40, lower is better). The EAT-10 is used as a surrogate outcome of swallowing-related quality of life in the current guideline. Cakmak (2022) reported a mean of 7.7 (SD 7.9) points in the NMES+TDT group (n=17) vs. 11.5 (SD 8.8) points in the TDT only group (n=17). This corresponded to a calculated MD of 3.80 points lower (95% CI, 9.42 lower to 1.82 higher) in favor of the NMES+TDT group, but not clinically relevant.  

Functional swallowing ability (important)

Three RCTs included in the SR and MA of Wang (2023) reported the outcome using the Functional Dysphagia Scale (FDS) (scale 0-100, lower is better) (Huang, 2014; Lim, 2014; Park, 2016). In total, 53 patients received NMES on top of TDT, and 53 patients received TDT only (Figure 3). The MD was 8.81 points lower (95% CI 16.48 lower to 1.15 lower), in favor of the NMES+TDT group, but not clinically relevant.

_{Figure 3. Functional Swallowing Ability (measured with the Functional Dysphagia Scale, FDS) in stroke patients receiving NMES+TDT versus TDT}

_{Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation}

Number or percentage of patients who return to a normal diet (important)

The outcome measure number or percentage of patients who return to a normal diet was not reported.

Aspiration (important)

Six RCTs included in the SR and MA of Wang (2023) reported the outcome aspiration with the use of the Penetration-Aspiration Scale (PAS) (scale 1-8, lower is better) (Arreola, 2021; Huang, 2014; Lim, 2014; Park, 2016; Simonelli, 2019; Sproson, 2018). One RCT included in the NMA of Banda (2023) was added (Cakmak, 2022). In total, 128 patients received NMES on top of TDT, and 124 patients received TDT only (Figure 4). The MD was 0.52 points lower (95% CI 1.07 lower to 0.04 higher), in favor of the NMES+TDT group, but not clinically relevant.

_{Figure 4. Aspiration (measured with the Penetration-Aspiration Scale) in stroke patients receiving NMES+TDT versus TDT}

_{Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation}

HNC patients

Comparison: NMES+TT or NMES only vs sham NMES+exercises or exercises only

Functional oral intake of food as experienced by the patient (crucial)

Two studies reported the outcome measure functional oral intake of food as experienced by the patient.

One RCT reported the outcome using the FOIS (scale 1-7, higher is better) four weeks posttreatment (Ku, 2023). Ku (2023) reported a mean of 6.52 points in the NMES group (n=31) vs. 6.38 points in the EBST group (n=34). This corresponded to a MD of 0.14 points higher in the TNMES group, which is not clinically relevant. SDs were not reported, and a GRADE assessment could not be performed.

One RCT included in the SR and MA of Paim (2023) reported the outcome using the American speech-language-hearing association national outcome measurement system swallowing level scale (ASHA NOMS) (scale 0-7, higher is better) posttreatment (Ryu, 2009). Ryu (2009) reported a mean ASHA NOMS score of 3.86 (SD 1.80) in the NMES+TDT group (n=14) vs. 3.67 (SD 2.10) in the sham NMES+TDT only group (n=12). This corresponded to a calculated MD of 0.19 points higher (95% CI, 1.33 lower to 1.71 higher) in favor of the NMES+TDT group, but not clinically relevant.

Swallowing-related quality of life (crucial)

Three studies included in the SR and MA of Paim (2023) reported the outcome swallowing-related quality of life (Lin, 2011; Langmore, 2016; Ryu, 2009). One study used the Head and Neck Cancer Inventory (HNCI, scale 0-100, higher is better) (Langmore, 2016) and two studies used the MD Anderson Dysphagia Inventory (MDADI, scale 20-100, higher is better) (Lin, 2011; Ryu, 2009). One additional study published after the search date of the Paim (2023) was added (Ku, 2023). Ku (2023) reported the outcome using the functional assessment after cancer therapy-nasopharyngeal (FACT-NP, scale 0-172, higher is better). In total, 141 patients received TNMES alone or on top of TDT and 90 patients received sham NMES on top of TDT or TDT only (Figure 5). The calculated SMD was 0.09 points higher (95% CI, 0.20 lower to 0.39 higher), which is not clinically relevant.

_{Figure 5. Quality of life (measured with the HNCI, MDADI and FACT-NP) in HNC patients receiving TNMES alone or on top of TDT versus sham NMES on top of TDT or TDT only}

_{Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation; HNC, head and neck cancer}

Functional swallowing ability (important)

One RCT included in the SR and MA of Paim (2023) reported the outcome functional swallowing ability using the FDS (scale 0-100, lower is better) posttreatment (Ryu, 2009). Ryu (2009) reported a mean FDS score of 22.4 (SD 13.4) in the NMES+TDT group (n=14) vs. 35.3 (SD 17.7) in the sham NMES+TDT only group (n=12). This corresponded to a MD of 12.90 points lower (95% CI, 25.13 lower to 0.67 lower), in favor of the NMES+TDT group, and clinically relevant.

Number or percentage of patients who return to a normal diet (important)

The outcome measure number or percentage of patients who return to a normal diet was not reported.

Aspiration (important)

Two studies included in the SR and MA of Paim (2023) reported the outcome aspiration using the PAS (scale 1-8, lower is better) (Lin, 2011; Langmore, 2016). One additional study published after the search date of the Paim (2023) also reporting on the PAS was added (Ku, 2023). In total, 131 patients received NMES on top of TDT or TNMES only and 79 patients received sham NMES on top of TDT or EBST only (Figure 6). The calculated MD was 0.30 points lower (95% CI, 1.25 lower to 0.64 higher), in favor of the NMES+TDT or TNMES only group, but not clinically relevant.

_{Figure 6. Aspiration (measured with the Penetration-Aspiration Scale) in HNC patients receiving NMES on top of TDT or TNMES only versus sham NMES on top of TDT or EBST only}

_{Abbreviations: NMES, neuromuscular electrical stimulation; TDT, traditional dysphagia therapy; SD, standard deviation; HNC, head and neck cancer}

Parkinson's disease patients

Comparison: NMES + exercises vs. sham + exercises

Functional oral intake of food as experienced by the patient (crucial)

The outcome measure functional oral intake of food as experienced by the patient was not reported.

Swallowing-related quality of life (crucial)

The outcome measure swallowing-related quality of life was not reported.

Functional swallowing ability (important)

One RCT included in the SR and MA of Cheng (2023) reported the outcome functional swallowing ability using the Videofluoroscopic Dysphagia Scale (VDS) (scale 0-100, lower is better) four weeks posttreatment (Park, 2018). Park (2018) reported a mean VDS score of 54.11 (SD 5.62) in the NMES+exercises group (n=9) vs. 53.28 (SD 10.21) in the sham NMES+exercises group (n=9). This corresponded to a MD of 0.83 points higher (95% CI, 6.78 lower to 8.44 higher), in favor of the sham NMES+exercises group, but not clinically relevant.

Number or percentage of patients who return to a normal diet (important)

The outcome measure number or percentage of patients who return to a normal diet was not reported.

Aspiration (important)

One RCT included in the SR and MA of Cheng (2023) reported the outcome aspiration using the PAS (scale 1-8, lower is better) four weeks posttreatment (Park, 2018). Park (2018) reported a mean PAS score of 2.33 (SD 0.71) in the NMES+exercises group (n=9) vs. 3.67 (SD 1.32) in the sham NMES+exercises group (n=9). This corresponded to a MD of 1.34 points lower (95% CI, 2.32 lower to 0.36 lower), in favor of the NMES+exercises group, and clinically relevant.

A systematic review of the literature was performed to answer the following question(s):

What is the effectiveness of neuromuscular electrical stimulation and conventional swallowing therapy as treatment for patients with dysphagia compared to placebo electrical stimulation or no electrical stimulation and conventional swallowing therapy?

Table 1. PICO

Relevant outcome measures

The guideline panel considered the functional oral intake of food as experienced by the patient and swallowing-related quality of life as a critical outcome measure for decision making; and functional swallowing ability, the number or percentage of patients who return to a normal diet and aspiration as an important outcome measure for decision making.

A priori, the guideline panel did not define the outcome measures listed above but used the definitions used in the studies.

The guideline panel used the GRADE standard limits of 25% (a relative risk (RR) of <0.80 and >1.25) for dichotomous outcomes and 10% for continuous outcomes as a minimal clinically (patient) important difference. For dichotomous outcomes with very few events, a risk difference of 25% was used as a minimal clinically (patient) important difference.

Search and select (Methods)

A systematic literature search was performed by a medical information specialist using the following bibliographic databases: Embase.com and Ovid/Medline. Both databases were searched from 2015 to the 4th of April 2025 for systematic reviews, RCTs and observational studies. Systematic searches were completed using a combination of controlled vocabulary/subject headings (e.g., Emtree-terms, MeSH) wherever they were available and natural language keywords. The overall search strategy was derived from two primary search concepts: (1) dysphagia; (2) electric stimulation. Duplicates were removed using EndNote software. After deduplication a total of 1481 records were imported for title/abstract screening. Studies were selected based on the criteria described in Table 1. Titles and abstracts were screened using the ASReview software version 1.6.2. The settings TF-IDF and Naïve bayes were used. One study was used as prior knowledge for inclusion (Terré, 2015; Gandhi, 2022). One study was used as prior knowledge for exclusions (Chen, 2022; Baroni, 2024).

The first 10% of references were screened by the working group and the guideline methodologist. The remaining articles were subsequently screened by the guideline methodologist, using the following stopping rule: stop after 200 subsequent exclusions.

Initially, 60 studies were selected based on title and abstract screening. After reading the full text, 41 studies were excluded for post-stroke patients, four for head and neck cancer (HNC) patients, four for neurodegenerative patients, and seven for other patient groups (see the exclusion table under the tab ‘Evidence tabellen’). In total, five studies were included.

The included studies are described separately per disease category in the results section. These categories comprise post-stroke patients, patients with Parkinson’s Disease, and patients with head and neck cancer (HNC).