Effect of Placement of a Speech Appliance on Levator Veli Palatini Muscle Activity During Speech

Effect of Placement of a Speech Appliance on Levator Veli Palatini uscle Activity During Speech Objective: The purposes of this electromyographic study were to examine whether levator veli palatini muscle activity during speech can be changed with placement of a speech appliance and to clarify whether or not the change is related to the type of speech appliance used. Design: Electromyography (EG) was performed during production of speech samples in two conditions: with placement and with removal of a speech appliance. Speech samples were the vowel / / and consonant-vowel ( ) syllables, including nasal, plosive, fricative, and affricate consonants. Setting: Division for Oral-acial Disorders, aculty of Dentistry, Osaka University Dental Hospital, Japan. Participants: Subjects were six patients with repaired cleft palate and velopharyngeal inadequacy (VPI); three routinely wore a palatal lift prosthesis to correct borderline VPI, and the other three wore a speech appliance hybrid that consisted of a pharyngeal bulb and a palatal lift. Interventions: Each patient was asked to produce each speech sample in both conditions of placement and removal of their respective appliances. ain Outcome easures: Smoothed EG signals of the levator veli palatini muscle were recorded with and without the appliance during five repetitions of each speech sample. The average value of peak smoothed levator EG was compared between placement and removal conditions for all subjects and between the two subject groups in each experimental condition. Results: With a speech appliance in place, the range of levator activity was distributed in a portion lower than that in the removed condition for subjects in both appliance groups. In addition, the intervals were smaller in the placement condition than in the removed condition. Conclusion: A speech appliance may prevent hypernasality, nasal emission of air, or both associated with VPI because of mechanical obturation of the velopharynx and an alteration in velopharyngeal function. TAKASHI TACHIURA, D.D.S., PH.D. KANJI NOHARA, D.D.S. TAKESHI WADA, D.D.S., PH.D. KEY WORDS: cleft palate, electromyography, levator veli palatini, speech appliance, velopharyngeal function Speech appliances, such as a palatal lift prosthesis or a pharyngeal bulb, correct hypernasality or nasal emission associated with velopharyngeal inadequacy (VPI). When a speech appliance is in place, the patient can exhibit adequate airtight separation between the oral and nasal cavities during produc- Dr. Tachimura is Associate Professor and Vice Chair, Dr. Nohara is a postgraduate student, and Dr. Wada is Professor and Chairperson of the Division for Oral and acial Disorders, aculty of Dentistry, Osaka University, Osaka, Japan. Submitted June 999; Accepted October 999. This research was supported by grants-in-aid 083809 and 06785 from the Japanese inistry of Education. Presented at the 54th Annual eeting of The American Cleft Palate Craniofacial Association, San Diego, California, 997. Reprint requests: Dr. Takashi Tachimura, Division for Oral and acial Disorders, aculty of Dentistry, Osaka University 8, Yamada-Oka, Suita, Osaka, 565-087, Japan. E-mail tatimura@dent.osaka-u.ac.jp. tion of pressure consonants or while blowing with variable intensity. That is, it is likely that the velopharyngeal function can be adequately regulated in order to prevent nasal emission or hypernasal speech. or normal speakers, the activity of the levator veli palatini muscle for velopharyngeal closure can be changed in relation to oral air pressure during blowing with variable intensity. However, the activity cannot change for speakers with a repaired cleft palate and VPI (Goto, 977; Kuehn and oon, 994, 995). Tachimura et al. (999) demonstrated that when a speech appliance is in place, levator muscle activity can change in relation to oral air pressure during blowing. urther, they found that the range of levator activity shifts to a lower region relative to the maximum activity level in comparison with the range achieved during the appliance-removed condition. Kuehn and oon (994, 995) defined the discrepancy be- 478

Tachimura et al., SPEECH APPLIANCE/LVP/SPEECH 479 IGURE used by subjects exhibiting borderline VPI. IGURE used by speakers exhibiting substantial VPI. tween activity ranges in the levator veli palatini muscle for speech tasks and blowing at the maximum intensity level as the reserve capacity to drive the velopharyngeal function for speech. urther, they reported that normal speakers have a greater reserve capacity of the levator muscle compared with speakers exhibiting hypernasality. It may be that the reserve capacity is levator activity that can be recruited to supplement the requirement for maintaining adequate velopharyngeal closure during targeted tasks such as speech and blowing. Therefore, the discrepancy between levator activity for targeted tasks and the possible maximum activity may be interpreted as a reserve capacity for targeted tasks such as speech and blowing. Decreased levator activity during blowing in association with placement, as reported by Tachimura et al. (999), can be regarded as increased reserve capacity. That is, the clinical effect of a speech appliance in improving velopharyngeal function during blowing may be caused by an increase in the reserve capacity of the levator muscle. Repeated production or sustained phonation of obstruent sounds requires continuous airtight closure of the velopharynx, which blowing does. It can be speculated that the improvement in velopharyngeal function during speech that is associated with placement of a speech appliance might be caused by an increase in the reserve capacity through decreased activity of the levator muscle. It is possible that placement of a speech appliance could also alter velopharyngeal function for speech so that it is more similar to that of normal speakers. This electromyographic study was conducted () to examine whether levator veli palatini muscle activity for speech can be changed in association with placement of a speech appliance, and () to clarify whether or not the change, if any, is related to the type of speech appliance. Subjects ETHOD The subjects were six patients with either substantial or borderline VPI who had undergone primary palatoplasty using a pushback procedure at.0 to.5 years of age. All of the subjects routinely wore either a palatal lift prosthesis (; see ig. ) or a speech bulb and palatal lift prosthesis hybrid (; see ig. ). The selection of the type of speech appliance depended on the severity of VPI. That is, a and a are used to correct borderline VPI and substantial VPI, respectively. The subjects were classified into two groups according to their respective types of speech appliance: the group and the group. Table shows the sex, age, and type of cleft in each patient in each group. Two experienced speech pathologists independently judged the speech of all of the subjects to be within normal limits and free from articulation error with their speech appliance in place, although they showed hypernasality, or nasal emission of air, or both, but no articulatory placement errors, without their prosthesis (Table ). An interjudge reliability coefficient of 0.97 was obtained for the judgments of hypernasality and audible nasal emission. Speech Task and Electromyography The subjects were instructed to perform speech tasks during two obturator conditions: with the appliance placed and with the appliance removed. The speech tasks included productions of / /, /m /, /p /, /s /, and /ts /. Each speech sample was repeated five times in separated, isolated utterances. TABLE Subject Demographics* Subject Group Sex Age (y) Cleft Type* S K S HK SK 8 5 3 5 CPO CPO UCLP * CPO cleft palate only; UCLP unilateral cleft lip and palate; bilateral cleft lip and palate; palatal lift prothesis; speech bulb and palatal lift prothesis hybrid; female; male.

480 Cleft Palate Craniofacial Journal, September 000, Vol. 37 No. 5 TABLE Listener s Judgment of Hypernasality and Nasal Air Escape in Six Subjects Under the Condition of Removal of a Prothesis S K S HK SK Subject Hypernasality* Nasal Air Escape 5 4 4 * Ratings for hypernasality were performed using a 5-point scale ( normal, mild, 3 moderate, 4 severe, 5 very severe hypernasality). palatal lift prothesis; speech bulb and palatal lift prothesis hybrid. A 3-point scale was used to rate nasal air escape during conversation (0 not audible, sometimes audible, always audible). IGURE 3 Levator veli palatini EG activity (95% confidence interval) for speech tasks with and without an appliance in the group. Open and closed circles represent values of maximum and minimum of 95% Stainless steel wires (80 0 3 mm in diameter) were used to record the electromyographic (EG) activity. As in EG studies related to levator veli palatini muscle activity (Goto, 977; Tachimura, 985; Kuehn et al., 993; Kuehn and oon, 994, 995; Tachimura et al., 995, 999), the electrode insertion site was determined on the basis of the levator dimple appearing in the velum during sustained production of /a/. The site for insertion in the levator dimple in the left side was marked using Gentiana violet (methylrosaniline chloride) so that the site for insertion was not overlaid by the velum section of a speech appliance. Electrodes were inserted in the levator dimple without a speech appliance. The electrode insertion site in the velum was anesthetized with 4% lidocaine topical anesthetic. The electrode wires were inserted perorally in the superior, posterior, and lateral directions. The two electrodes were placed with approximately 5 mm spacing between them and 0 mm deep in the subject s left levator muscle. EG signals, obtained using a bioelectronic amplifier (Nihon Kohden Co., Tokyo, Japan AB65J), were high-pass filtered at 30 Hz, rectified, and smoothed with a time constant of 30 ms through an amplifier (Nihon Kohden Co., EI600G). The peak of the smoothed EG trace in each production was measured. confidence intervals in removal and placement conditions, respectively. The symbols S S5 represent the speech samples of / /, /m /, /p /, /s /, and /ts /, respectively. of 0.98 was obtained in the comparison between the output values obtained from the two observers. RESULTS igures 3 and 4 show the EG data in relation to speech samples that were obtained from subjects in the and Bulb- groups, respectively, when the respective appliances were in place and removed. The EG activity in igures 3 and 4 was expressed as a percentage relative to the maximum value of EG activity observed across the experiment for each subject. Table 3 shows the mean values and standard deviations for levator activity in each condition for each group. The Student s t-test was carried out between the two experimental con- Data Analysis The measurement of EG activity from the recording paper was carried out using a computer (NEC, PC-980RX) and an electromagnetic induction-type digitizer (Oscon Co., Tokyo, Japan SQ-300R) with a technique used in previous studies (Tachimura et al., 995, 997, 999). The reliability of the programs for input and output was confirmed in preliminary studies using a set of records containing known real values. The procedures for confirmation of the program and the interobserver reliability have been reported in a previous paper (Tachimura et al., 995). The interobserver reliability of values was determined for two observers who independently recorded 70 sets of measured values. A Pearson correlation coefficient IGURE 4 Levator veli palatini EG activity (95% confidence interval) for speech tasks with and without an appliance in the group. Open and closed circles represent values of maximum and minimum of 95% confidence intervals in removal and placement conditions, respectively. The symbols S S5 represent the speech samples of / /, /m /, /p /, /s /, and /ts /, respectively.

Tachimura et al., SPEECH APPLIANCE/LVP/SPEECH 48 TABLE 3 Levator Veli Palatini uscle Activity Under Each Experimental Condition Group N 65 75 ean 57.75 68.30 Removed SD 3.08 6.7 Placement ean SD t 4.98 40.4 t.54 4.94* 3.48 8.88 6.78* 7.59* t and t show t values obtained with the Student s t test between two experimental conditions for each group and between two subject groups in each experimental condition, respectively. palatal lift prothesis; speech bulb and palatal lift prothesis hybrid. * Significant at p.00. ditions for each group and between the two subject groups in each experimental condition. In the removal condition, the range of levator activity (95% confidence interval) for all of the subjects in the group was distributed in a relatively higher region than for subjects in the group. However, there was no significant difference in the mean values of levator activity between the group and the group. On the other hand, with the appliance the range of EG activity was smaller than that without the appliance for all subjects in both groups, although the levator activity was significantly greater in the group than in the group. All subjects showed decreased levator activity with the appliance for the same speech sample in comparison with that obtained without the appliance. oreover, the intervals of levator activity with the appliance were also less than those without the appliance, irrespective of the type of speech appliance used. In both groups, levator activity was significantly reduced with the appliance compared with those without the appliance. A one-way analysis of variance (ANOVA) was performed for each group to clarify whether placement of the subject s appliance affected levator activity. Table 4 shows the results obtained for each group. In both groups, the difference between f values was statistically significant. This result suggests that placement of an appliance, irrespective of the type of speech appliance, was associated with a decrease in levator activity during speech. DISCUSSION Previous research findings have suggested that a speaker with VPI can show a correlation between oral air pressure and levator muscle activity during blowing with a speech appliance in place, although there is no relationship between these two variables in the removed condition (Tachimura et al., 999). oreover, it has also been demonstrated that levator muscle activity for blowing is smaller with the appliance than without the appliance (Tachimura et al., 999). Repeated production of obstruent sounds, nonnasal speech, or both requires continuous air-tight closure of the velopharynx, a situation that may be similar to that during continuous blowing. In both conditions, expiratory air should be impounded within the oral cavity in order to maintain oral air pressure at the level required for the TABLE 4 Analysis of Variance for Data of Levator uscle Activity in and Groups Using a One-Way Classification Source of Variation Condition Error Total Condition Error Total Sum of Squares 34,896.8 97,.60 3,007.88 9,5. 77,079.9 06,60.50 Degrees of reedom ean Square Compared f 8 48 34,896.8 758.68 9,5. 50.8 45.996* 56.685* palatal lift prothesis; speech bulb and palatal lift prothesis hybrid. * p.00. target tasks, including blowing or repeated production of obstruent sounds. However, it has not been previously demonstrated that the placement of a speech appliance can alter velopharyngeal function for speech. Clinically, some patients show mild hypernasality, emission of air, or both during connected speech despite the absence of any symptoms of VPI during isolated production of obstruent sounds. Kuehn and oon (994, 995) hypothesized that, with a reduced reserve capacity, the muscle may be activated near to or above the threshold of fatigue so that speech would be expected to deteriorate rapidly and become excessively hypernasal if the threshold of fatigue is exceeded by the muscle activation. Kuehn (99) recommended that continuous positive air pressure (CPAP) therapy be used to strengthen the muscles related to velopharyngeal closure in order to increase the reserve capacity to overcome muscle fatigue. In the results of the current study, the levator activity range for speech with the appliance was smaller than that without the appliance, irrespective of the severity of VPI. This decrease in levator activity in association with placement can be regarded as a relative increase in the reserve capacity of the levator muscle. An increased reserve capacity in levator activity may be effective in preventing fatigue of the muscles related to velopharyngeal function for speech. That is, the clinical effect of a speech appliance in improving velopharyngeal function for speech may be caused by an increase in the reserve capacity. A number of authors have reported clinical evidence that speech appliances stimulate increased movement of the velopharyngeal structures (Shelton et al., 968, 97; Blakeley and Porter, 97; azaheri and azaheri, 976; Beery et al., 983). However, those reports were not based on physiological findings of changes in velopharyngeal function associated with placement and removal of the appliance, but on clinical or subjective impressions of decreased hypernasality, nasal emission, or both. ovement gains of this sort have been attributed to the resistance the speech appliance provides against the muscles; strength is increased by working the muscles against resistance (Rosen and Bzoch, 958). However, if the velopharynx showed a movement gain to resistance provided by a speech appliance, the levator activity level might be greater with the speech appliance in place than with it removed. ore-

48 Cleft Palate Craniofacial Journal, September 000, Vol. 37 No. 5 over, an increased activity level might induce fatigue in the muscles related to velopharyngeal function, which might also result in weakened velopharyngeal closure during conversation, connected speech, or blowing. The results obtained here did not support the above hypothesis but provided a rational explanation regarding the clinical effect of a speech appliance. The decreased levator activity may imply that the velopharynx around the bulb for a or narrowed velopharynx for a could be tightly closed by mechanisms related to velopharyngeal function with relatively little or no muscular fatigue. The speaker could then continue to perform tasks including repeated phonation or connected speech and blowing tasks. When a speech appliance is in place, mechanisms to obturate the velopharynx are required to travel a shorter distance from their rest positions than when the appliance is removed. The decreased levator activity might be attributed to the decreased distance needed to obtain airtight closure of the velopharynx. There might be some system to regulate levator activity in relation to the change in distance that velopharyngeal structures must travel. There are several variables that might be involved in such a system, including nasal air flow (Tachimura, 995), nasal air pressure (Tachimura et al., 994; Tachimura, 985; Kuehn et al., 993), oral air pressure (Goto, 977; Kuehn and oon, 994, 995; Tachimura et al. 994, 995,997), or activity of other muscles related to velopharyngeal function in addition to the levator muscle. These aerodynamic variables, which are generated in association with phonation, might be involved in regulation of the velopharyngeal function during speech. This study did not attempt to identify which of those variables are related to changes in levator muscle activity in association with placement of an appliance. uture studies are warranted to examine these relationships. REERENCES Beery QC, Rood SR, Schramm VL. Pharyngeal wall motion in prosthetically managed cleft palate adults. Cleft Palate J. 983;0:7 7. Blakeley RW, Porter DR. Unexpected reduction and removal of an obturator in a patient with palatal paralysis. Br J Disord Commun. 97;6:33 36. Goto T. Tightness in velopharyngeal closure and its regulatory mechanism. J Osaka Univ Dent Soc. 977;:87 06. Kuehn DP. New therapy for treating hypernasal speech using continuous positive airway pressure (CPAP). Plast Reconstr Surg. 99;88:959 966. Kuehn DP, oon JB. Levator veli palatini muscle activity in relation to intraoral air pressure variation. J Speech Hear Res. 994;37:60 70. Kuehn DP, oon JB. Levator veli palatini muscle activity in relation to intraoral air pressure variation in cleft palate subjects. Cleft Palate Craniofac J. 995; 3:376 38. Kuehn DP, oon JB, olkins JW. Levator veli palatini muscle activity in relation to intranasal air pressure variation. Cleft Palate Craniofac J. 993; 30:36 368. azaheri, azaheri EH. Prosthodontic aspects of palatal elevation and palatopharyngeal stimulation. J Prosthet Dent. 976;35:39 36. Rosen S, Bzoch KR. The prosthetic speech appliance in rehabilitation of patients with cleft palate. J Am Dent Assoc. 958;57:03 0. Shelton RL, Lindquist A, Arndt WB, Elbert, Youngstrom KA. Effect of speech bulb reduction on movement of the posterior wall of the pharynx and posture of the tongue. Cleft Palate J. 97;8:0 7. Shelton RL, Lindquist A, Chisum L, Arndt WB, Youngstrom KA, Stick SL. Effect of prosthetic speech bulb reduction on articulation. Cleft Palate J. 968;5:95 04. Tachimura T. Response of levator veli palatini muscle activity to change in intranasal air pressure during regulation of semiclosure of velopharynx for nasal vowel. J Osaka Univ Dent Soc. 985;30:8 59. Tachimura T, Hara H, Koh H, Wada T. Effect of temporary closure of oronasal fistula on levator veli palatini muscle activity. Cleft Palate Craniofac J. 997;34:505 5. Tachimura T, Hara H, Wada T. Effect of oral and nasal air pressures on LVP muscle activity during CPAP loading. Presented at the 5st annual meeting of the American Cleft Palate Craniofacial Association; ay 6, 994; Toronto, Canada. Tachimura T, Hara H, Wada T. Oral air pressure and nasal air flow rate on levator veli palatini muscle activity in patients wearing a speech appliance. Cleft Palate Craniofac J. 995;3:38 389. Tachimura T, Nohara K, Hara H, Wada T. Effect of placement of a speech appliance on levator veli palatini muscle activity during blowing. Cleft Palate Craniofac J. 999;36:4 3.