This study had a prospective, non-concurrent control comparative design. Among patients with dysphagia admitted to Seoul National University Hospital between January 2006 and July 2007, twenty-eight individuals (26 stroke and two traumatic brain injury patients) were enrolled who met the following criteria: (1) VFSS showing reduced laryngeal elevation during swallowing (judged by three rehabilitation physicians), (2) VFSS showing supraglottic penetration or subglottic aspiration when swallowing 2 ml of 50% liquid barium (Solotop® suspension 140) solution, and (3) no clinical evidence of vagus or glossopharyngeal nerve palsy in addition to the inclusion criteria from the VitalStim® manual.8

Only conventional dysphagia management (CDM) without electrical stimulation therapy was applied to patients admitted between January 2006 and March 2007 as a control group. Electrical stimulation of suprahyoid muscles (ESSM) was performed on patients between April 2007 and July 2007. Patients who received electrical stimulation also underwent CDM.

The protocol of this study was approved by the institutional review board of Seoul National University Hospital. The VDS9 and ASHA NOMS swallowing scale (ASHA level)10 were used as parameters for evaluating swallowing function. The VDS (from 0 to 100) was determined using videofluoroscopic analysis, and consisted of several factors such as subglottic aspiration and food residues. This scale was reported to be sensitive and specific for detecting supraglottic penetration and subglottic aspiration, and was considered useful for predicting persistent dysphagia after stroke.11 The ASHA level is a measurement of both the supervision level required and diet level that intuitively reflects a patient's functional status.10

Treatments were performed for 30 minutes per day, 5 days per week for 4 weeks by three experienced occupational therapists who shared treatment protocols. Three rehabilitation physicians with considerable experience in the management of dysphagia (a senior professor, clinical fellow, and second-grade resident physician) determined the VDS and ASHA level before and after treatment with consensus. The VFSS was performed in a manner previously described9 and the VDS was determined while the patients swallowed 2 ml of a 50% liquid barium solution.

Conventional dysphagia management consisted of ice cube training, double swallowing, effortful swallowing, Mendelsohn maneuver, and thermal tactile stimulation. If a patient could not swallow an ice cube without overt aspiration symptoms, ice cube training was excluded from CDM. All patients were sufficiently conscious to perform effortful swallowing and the Mendelsohn maneuver. STIMPLUS DP200® (Cybermedic Corp., Iksan, South Korea) was used for suprahyoid NMES. The stimulation parameters used were 60 Hz of 500 microsecond duration with intermittent stimulation (1 second on and 1 second off) to avoid laryngeal spasms and to simulate normal pharyngeal swallowing. The active electrodes were placed at midpoints between the bilateral edges of the hyoid and chin, whereas the reference electrodes were placed at midpoints between the bilateral mandibular angle and chin (Fig. 1). The locations of active electrodes were at the points considered near to the motor point of the anterior belly of the digastric muscle. This was because we had previously shown that surface electrical stimulation moved the hyoid bone anteriorly and superiorly during stimulation using this electrode geometry in normal volunteers.7 Stimulus intensity was started at 3 mA and increased by 1 mA. When a volunteer expressed discomfort but not pain, we administered the stimulus at this intensity three times to allow adaptation to the electrical stimulus. When a volunteer continued to complain about discomfort or pain, we defined the stimulus intensity used prior to the patient's complaint as the maximal tolerable intensity. We administered stimulation of the maximal tolerable intensity for 30 minutes using a 1 second on/1 second offcycle.

SPSS version 12.0 (Chicago, IL, USA) for Windows was used for statistical analysis. The Mann-Whitney U test was used to confirm that initial swallowing function, patient age, and post-brain injury dysphagia onset were similar between the CDM and ESSM groups. To compare the degree of VDS score changes between the two groups, the Mann-Whitney U test was also used. Fisher's exact test was performed to compare the prevalence of ASHA level improvement in the CDM and ESSM groups. p-values less than 0.05 were considered statistically significant.

The VDS score decreased from 29.8±13.5 to 17.9±7.8 in the ESSM group, and from 29.2±15.5 to 16.6±11.8 in the CDM group (Fig. 2). There was no significant difference between groups (p=0.796, Mann-Whitney U test). One patient in the ESSM group (patient No. 7) showed an increase in VDS score after treatment. This paradoxical VDS increase was caused by a newly developed premature bolus loss during liquid swallowing due to a dental prosthesis, and there were no findings such as motor weakness or dysarthria due to a new brain lesion.

Six patients (85.7%) in the ESSM group and 14 patients (66.7%) in the CDM group showed improved ASHA levels (Fig. 2), but there was no significant difference observed between the ESSM and CDM groups (p=0.633, Fisher's exact test).