JCPSLP Vol 20 No 2 July 2018

anatomical pictures of the targeted facial muscles (see Dalla Toffola, 2012; Schumann, Bongers, Guntinas-Lichius, & Scholle, 2010; Stepp, 2012; Tassinary, Cacioppo, & Geen, 1989;). Placement of the “ground” electrode was close to the muscles of detection, e.g. forehead, shoulder with the two “differential” electrodes placed so that the electrode axis was parallel to the underlying muscle fibres and spaced about 0.5–1 cm (approximately one quarter of the length of the muscle fibre). Data analysis The non-parametric Wilcoxon matched paired test (two-tailed) was used to determine the significance of the mean change between pre- and post-test scores for the Nottingham System, House Brackman Facial Nerve Grading System and the FaCE Scale for the group. Analysis of the FaCE data was undertaken for the whole group and then repeated with the exclusion of one outlier to examine potential influences. Individual change was examined using Fisher Exact (two-tailed). Effect sizes were calculated using Cohen’s d . Results All 12 participants completed the program. Significant within group gains were seen for all outcome measures (see Table 2). A significant increase in facial symmetry was seen for the group ( p = 0.018; medium, d = 0.701) on the Nottingham System, with an average increase in facial symmetry of 22.74 points on the measure, and a medium effect size (See example in Figure 2). This increase was carried by significant gains for nine of the individuals (see Table 3). One participant did not progress from initial baseline level (#11, this was seen for all three measures), one participant decreased significantly (#5, p = 0.022), while the final participant’s numerical gains did not reach significance. A significant group reduction in severity of facial nerve weakness was seen on the House Brackman Facial Nerve Grading System ( p = 0.006; medium, d = 0.558), with an average reduction in grading of 0.92 of a point (range 1–6 points), and medium effect size. While one participant did not complete the post-treatment retest on the FaCE scale due to unavailability at the time of testing, significant improvement was still seen at a group level ( p = 0.02; medium, d = 0.652) with an average increase of 6.58, and a medium effect size seen. On this final scale, only three individuals showed significant improvement at the individual level although a positive trend was seen in a further six participants. Table 2. Group performance on the three outcome measures showing mean change from pre- to post-intervention

only), consistent with usual practice within the RITH program; total sessions ranged between 20 and 25 sessions. Therapy was conducted in the participants’ homes and involved a series of routine unilateral and bilateral exercises using the participant’s own mirror and with the portable sEMG, laptop and BioGraph Infiniti software program (http://thoughttechnology.com/) brought into the home by the clinician for treatment sessions. The participants were encouraged to carry out daily home practice following the clinician-monitored session a further two to three times each day, including weekends, using the mirror for biofeedback and focusing on the exercises practised in their session. Mime therapy/neuromuscular re-education procedure Targeted facial exercises were individually graded according to the participant, focusing on participant-specific target muscles, with education of facial anatomy/physiology using anatomical diagrams. Resistance movements of “hold” and “stretch” exercises were included with monitoring and training of the participant to observe and adjust movement when over activation of the unaffected side of the face occurred. The participant was taught to aim for symmetry of movement when bilateral movement was attempted. Maintenance of normal status for the unaffected side of the face was encouraged during unilateral movements of the affected side of the face. Sets of three movements for target muscles were interspersed with rest periods of five seconds. Increased time for endurance “holds” (five and then 20 seconds) were gradually introduced according to participant fatigue, with the sEMG also used to determine fatigue points via drop off amplitude of the visual trace. During the later stages of treatment, speech sounds and words were included with bilateral sEMG biofeedback to encourage use of the affected side musculature and with appropriate symmetry. sEMG biofeedback procedure Sensitivity to electrode placement was assessed on the initial trial, with the sEMG readings determining potential muscle activation and guiding the selection of muscle targets. sEMG readings were taken with the MyoTrac Encoder and the corresponding BioGraph Infiniti software program (see Figure 1). The software allowed for use of the laptop to amplify the visual feedback for older clients and clearer biofeedback. The Cleartrode ECG electrodes (Ref no. 1720-003) were placed in accordance with recommendations from the manual and literature using

Significance

Outcome measures

n

Mean change

95% confidence interval

Nottingham (post–pre)

12 22.74 (7.05, 38.436)

p = .018

House Brackman (post–pre)

12 −.92 (−1.44, −0.40)

p = .006

FaCE

11 7.18 (2.17, 12.19)

p = .02

Figure 1. sEMG readings taken using the MyoTrac Encoder and BioGraph Infiniti graph (Photographs reproduced from http:// thoughttechnology.com/)

Wilcoxon matched pairs test (two-tailed).

66

JCPSLP Volume 20, Number 2 2018

Journal of Clinical Practice in Speech-Language Pathology