- Email: [email protected]
Vocal Parameters and Self-Perception in Individuals With Adductor Spasmodic Dysphonia
ARTICLE IN PRESS Vocal Parameters and Self-Perception in Individuals With Adductor Spasmodic Dysphonia *Gleidy Vannesa E. Rojas, *Hilton Ricz, †Vitor Tumas, †Guilherme R. Rodrigues, †Patrícia Toscano, and *Lílian Aguiar-Ricz, *†Ribeirão Preto, Brazil Summary: Objective. The study aimed to compare and correlate perceptual-auditory analysis of vocal parameters and self-perception in individuals with adductor spasmodic dysphonia before and after the application of botulinum toxin. Study Design. This is a prospective cohort study. Methods. Sixteen individuals with a diagnosis of adductor spasmodic dysphonia were submitted to the application of botulinum toxin in the thyroarytenoid muscle, to the recording of a voice signal, and to the Voice Handicap Index (VHI) questionnaire before the application and at two time points after application. Two judges performed a perceptualauditory analysis of eight vocal parameters with the aid of the Praat software for the visualization of narrow band spectrography, pitch, and intensity contour. Results. Comparison of the vocal parameters before toxin application and on the first return revealed a reduction of oscillation intensity (P = 0.002), voice breaks (P = 0.002), and vocal tremor (P = 0.002). The same parameters increased on the second return. The degree of severity, strained-strangled voice, roughness, breathiness, and asthenia was unchanged. The total score and the emotional domain score of the VHI were reduced on the first return. There was a moderate correlation between the degree of voice severity and the total VHI score before application and on the second return, and a weak correlation on the first return. Conclusions. Perceptual-auditory analysis and self-perception proved to be efficient in the recognition of vocal changes and of the vocal impact on individuals with adductor spasmodic dysphonia under treatment with botulinum toxin, permitting the quantitation of changes along time. Key Words: Adductor spasmodic dysphonia–Vocal quality–Perceptual voice judgments–Botulinum toxin–Voice handicap. INTRODUCTION Laryngeal dystonia is a vocal disorder of neurologic origin and is characterized as task-dependent focal action-induction dystonia that affects laryngeal motor control.1,2 It is recognized and referred to in the literature as spasmodic dysphonia because of its main vocal signal, ie, the presence of spasms,3 the most frequent being adductor spasmodic dysphonia (AdSD), which is of higher prevalence among women.4 AdSD is characterized by involuntary spasms of the laryngeal muscles observed at the beginning of phonation,5 associated with vocal symptoms such as strained-strangled voice quality, intermittent voice breaks,6 and effort in speech production,7 with a reduction of intelligibility, which compromise communication and affect the quality of life of an individual8 both in functional, emotional, and social terms.9 The application of botulinum toxin (BTX) to the laryngeal muscles is currently considered to be the treatment of choice,10 inducing paresis or paralysis of the laryngeal muscles and reducing the vocal symptoms for a period of time.11 Although the application of BTX for the treatment of AdSD is beneficial, it may also have side effects associated with the Accepted for publication September 30, 2016. From the *Department of Ophthalmology, Otorhinolaryngology and Head and Neck Surgery, School of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, SP, Brazil; and the †Department of Neurosciences and Behavioral Sciences, School of Medicine of Ribeirão Preto, University of São Paulo, Ribeirão Preto, SP, Brazil. Address correspondence and reprint requests to Lílian Aguiar-Ricz, Department of Ophthalmology, Otorhinolaryngology and Head and Neck Surgery, School of Medicine of Ribeirão Preto, University of São Paulo, Av. Bandeirantes, 3900, Monte Alegre, 14048-900 Ribeirão Preto, SP, Brazil. E-mail: [email protected] Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2016 Published by Elsevier Inc. on behalf of The Voice Foundation. http://dx.doi.org/10.1016/j.jvoice.2016.09.029
iatrogenic characteristics of the treatment, such as the presence of voice breathiness, 12 reduction of vocal intensity, hoarseness, alteration of airflow, effort to speak, fatigue, and reduction of the range of fundamental frequency in semitones.13 The duration of the effect of treatment varied from 3 to 4 months.14–16 The effect of treatment with BTX has been demonstrated by different methods, such as vocal acoustic measurements, aerodynamic measurements, laryngoscopy, perceptualauditory analysis performed by judges, 9,17–19 and selfperception of voice quality by the patient.17 However, there is no information that might permit to draw conclusions about a standardized tool or that might permit the quantitation of the time of benefit and of the reduction of voice changes detected in AdSD. In view of the above considerations, it is important to review evidence in the literature,17,18,20 and according to the experience with patients to construct a tool to be used in the clinical setting without the need for high-cost technology to permit the assessment of vocal signs, together with visual analysis of narrow band spectrography and of recordings of pitch intensity for the follow-up of BTX treatment. BTX treatment leads to improvement of vocal symptoms, although persons with AdSD do not consider their voice to be normal.20,21 On this basis, there is concern about determining the impact of treatment on the voice quality of patients with dystonia, with a search for different perspectives for the analysis of the effect of medication on voice quality, interpersonal communication and communicative performance during daily activities, and also with the identification of patient satisfaction regarding the duration and effect of treatment.
ARTICLE IN PRESS 2 At present, one of the questionnaires more frequently used as an indicator of the effectiveness of treatment with BTX regarding the improvement of symptoms and of the quality of life of patients with AdSD is the Voice Handicap Index (VHI),22 which is available in the version validated for the Portuguese language.23 Thus, the objective of the present study was to compare and to correlate the perceptual-auditory analysis of vocal parameters and self-perception in individuals with AdSD before and after the application of type A BTX. METHODS This was a prospective cohort study which was approved by the Research Ethics Committee (Protocol N0 8339/2013). All subjects gave written informed consent to participate. Participants The study was conducted on 16 individuals with a clinical diagnosis of AdSD, 11 (68.75%) were women and 5 (31.25%) were men, ranging in age from 25 to 90 years (mean: 57 years and 2 months). Of these, 81.25% had associated vocal tremor (VT) and other movement disorders associated with AdSD that did not promote changes in vocal production, ie, blepharospasm and cervical dystonia (12.5%), cervical dystonia (6.25%), generalized dystonia (6.25%), hemidystonia (6.25%), oromandibular tremor (6.25%), tremor in the cephalic region, cervical dystonia, and blepharospasm (6.25%). Only patients with a diagnosis of adductor laryngeal dystonia participated in the study, whereas patients with other types of laryngeal dystonia were excluded. The subjects included were individuals who were receiving treatment for the first time or who needed to be submitted to a new BTX injection as the last application had occurred more than 120 days (mean: 223.31 days; range: 161–412 days) before, and there were no clinical signs of the effect of the medication since the last application. The 16 selected individuals had received, on average, 3.9 BTX injections before the current application. The Pearson coefficient was applied for the inclusion of these subjects in order to determine the relationship between the degree of severity of vocal quality at the preapplication time and the time since the last application of the botulin toxin (number of days). A correlation coefficient of 0.41 (P value = 0.11) was observed, permitting us to conclude that time has no influence on vocal severity since last application. This permitted us to consider all patients to be equal even with different time since the last application. The following were the exclusion criteria: individuals with previous laryngeal surgeries, previous dysphonia, and language, speech, or hearing disorders, or dementia reported by individuals themselves or their relatives, or obtained from their medical records. Also excluded were patients with other neurologic changes, such as dysarthria, Parkinson’s disease, ataxia, cerebellar atrophy, sequelae of cerebrovascular accidents, and others. Procedures All patients were submitted to the standardized application of type A BTX (Dysport®) (Wrexham, United Kingdom) by the introduction of a needle/electrode through the cricothyroid mem-
Journal of Voice, Vol. ■■, No. ■■, 2016
brane at 0.5 cm from the median sagittal line at a lateral and superior angle of 30°–45° until reaching the thyroarytenoid muscle guided by electromyography. Twenty units of type A BTX were applied unilaterally on the left side of the larynx to all patients. All participants were submitted to the recording of the vocal signal and responded to the VHI, a questionnaire validated for the Portuguese language,23 before BTX injection, on the first return (R1) held on average 36 days later (range: 28–49 days), and on the second return (R2) held on average 137 days after application (range: 112–189 days). Sound signal capture The voice sound signal was captured and recorded using a professional microphone G-Track (Samson, Technologies, New York) and the Sound Forge 6.0 software (Sony Creative Software Inc, Middleton, Wisconsin, United States) at a sampling rate of 44,100 Hz and 16 bits, and stored in wav format. The procedure was carried out in an acoustically treated room with noise kept at less than 50 dB and monitored with a digital sound level meter. All participants performed prolonged emission of the vowel /a/, in Portuguese, three consecutive times at habitual vocal intensity and frequency at each time of assessment (preapplication, R1, and R2). The more stable sound signal was extracted visually from the vocal sample of the 16 subjects, the beginning and end of emission were discarded because of their irregular characteristics, and the best sound signal with at least 3 seconds of sound production was accepted. Instrument for assessment Eight vocal parameters, described in Table 1, were proposed for the perceptual-auditory assessment of subjects with AdSD. This analysis was performed based on auditory stimuli (sustained emission of the vowel /a/) and on visual stimuli obtained with the free access Praat software, version 53.42,24 permitting the visualization of the pitch and intensity contour of the vocal signal (Figure 1), and with narrow band spectrography considering the following configurations: bandwidth of 43 Hz and 0.03 (30 ms), frequency range from 0 to 5000 Hz, time step of 0.002 seconds, and frequency step of 20 Hz (Figure 2). The vocal parameters were quantified using a visual analogue scale consisting of a 10-cm line with anchors at each end, with 0 on the left end (no change or absence) and 10 on the right end (severe change or permanent presence). A response form was elaborated for data recording using Microsoft Office Excel (Redmond, Washington, United States), permitting the analysis of each of the eight vocal parameters using the cursor to mark the values on a line. The values were then recorded automatically and transferred to a spreadsheet. Perceptual-auditory analysis of vocal parameters The analysis was carried out by two specialized speechlanguage pathologist with a mean of 10 years of clinical experience in voice analysis and without hearing complaints, with judge 1 (J1) being the most experienced. The previously edited 47 voice recordings corresponding to the 16 subjects tested before the application of BTX and at R1, and to 15 individuals tested
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
3
TABLE 1. Description of the Vocal Parameters Assessed by Perceptual-Auditory Analyses Vocal Parameters Degree of severity of vocal quality Strained-strangled voice quality Oscillation of intensity
Voice break
Vocal tremor Roughness Breathiness Asthenia
Concept Overall impression of vocal quality using the analysis of the seven vocal parameters as reference Compressed and choppy sound with fluctuations of voice quality Unpredictable and uncontrollable variations of the level of intensity, changing from explosive to weakened over time Intensity contour: fluctuation of the intensity tracing in an ascending or descending manner, represented by short- or long-duration peaks during emission Narrow band spectrogram: changes in harmonic stability that becomes more obscure during the duration of the oscillation of intensity Abrupt and unexpected fall of phonation Pitch contour: interruption of the pitch tracing during emission, with a descending fall of the tracing being observed before the interruption and with a return to the values (Hz) close to those present where the interruption occurred Narrow band spectrogram: interruption of harmonics throughout the entire frequency range for the duration of the voice break, with background noise filling out until the return of the sound signal Presence of involuntary rhythmic and regular variations in frequency and intensity Narrow band spectrogram: harmonic ondulations throughout emission in all frequency ranges Audible occurrence of acoustic aperiodicity perceived in the sound signal as a rough, crackling, and tense-strangled voice Audible occurrence of respiratory noise or acoustic turbulence perceived in the sound signal Vocal weakness, loss of potency, reduced vocal energy
at R2, were exhibited in a randomized manner using the Praat software, version 53.42.24 To ensure a consistent response on the part of the judges, the process was divided into the following phases: training of the judges when the perceptual-auditory signals were conceptually exposed intra-examiner calibration and inter-examiner calibration. For the training of judges, which had a duration of 24 hours (3 days, with resting intervals), conceptual exposure of the vocal parameters to be assessed was performed based on perceptualauditory analysis and narrow band spectrography, and the contour of pitch and intensity, as shown in Table 1. Intra-examiner calibration consisted of the presentation of voice bank samples of individuals with AdSD who did not participate in the present study, repeated as many times as necessary for the examiners to understand the parameters to be assessed. The inter-examiner calibration consisted of the interaction between judges regarding the parameters, correlating the stimuli in order to reach a consensus about the differences recognized auditorily and visually, with the establishment of criteria in order to reach an acceptable level of agreement in their findings regarding the presence of vocal changes in individuals with AdSD. The second phase, test, consisted of a 6-hour test applied 1 day after training, when each rater received the response form in digital format. The voices were reproduced for the judges, with speakers positioned at a mean distance of 1 m, with no variation in volume. The stimuli were presented in a random manner, with the judges being unaware of the subject or of the time of treatment. The third phase, retest, was performed 10 days after the test to determine intra-rater agreement, with 20% of the total voice sample being presented (10 recordings).
VHI The questionnaire is applied in order to assess the impact and handicap of the voice of the patient. It contains 30 questions divided into three domains—emotional, functional, and physical— each containing 10 questions. Each item is measured with a Likert scale, which consists of 5 points: 0 = never, 1 = almost never, 2 = sometimes, 3 = almost always, and 4 = always. The questionnaire was filled out by the following methods: an interview when patients were unlearned or had visual problems, or completed by the patients themselves, always under the supervision of the investigator. The VHI scores were calculated as the simple sum of all replies, providing the total score and the score for each domain. The analysis of the 30 items of the VHI permitted the determination of the vocal impact. Statistical analysis The intraclass correlation coefficient (ICC) was used to determine the inter-rater agreement at the time of the test and intrarater agreement between test and retest. Comparison of the individuals at the three time points for each variable studied (vocal parameters) considering J1 responses was performed by means of orthogonal contrasts using a mixed-effects linear model (random and fixed effects). The linear model is used in the analysis of data in which the responses of the same individual are grouped and when the assumption of independence between observations in the same group is not adequate. In order to use this model, it is necessary for its residues to have normal distribution with zero mean and constant variance. The model was adjusted with the PROC MIXED procedure of the SAS 9.1 software (Cary, North Carolina, United States).
ARTICLE IN PRESS 4
Journal of Voice, Vol. ■■, No. ■■, 2016
FIGURE 1. Recording of pitch and intensity (Praat software). Example of sustained emission of the vowel /a/ by an individual with adductor spasmodic dysphonia before and 35 days after treatment with botulinum toxin.
The Student t test for paired samples was used to compare VHI scores between individuals at the three time points, with the level of significance set at P ≤ 0.05. The correlation between the total VHI scores and the degree of severity of vocal quality obtained in the perceptual-auditory analysis was determined using the Pearson correlation coefficient. All analyses were carried out with the aid of the SPSS (Statistical Package for the Social Sciences) (Chicago, Illinois, United States) software, version 20.0. RESULTS Inter- and intra-rater agreement The analyses carried out by the two judges demonstrated excellent agreement (ICC = 0.9) for most parameters, except for asthenia, for which the agreement was only satisfactory (ICC = 0.71) (Table 2).
Intra-rater agreement between test and retest determined by the repetition of 20% of the voices previously analyzed showed excellent or satisfactory agreement, with ICC ranging from 0.8 to 0.5 for the following parameters: degree of severity of voice quality, Oscillation of intensity, strained-strangled voice quality, and voice breaks. For the remaining vocal parameters, the agreement of the values obtained at rest was less intense. Perceptual-auditory analysis of vocal parameters Comparison of the vocal parameters of the 16 individuals before BTX injection and at time R1 revealed a reduction of the oscillation of intensity, voice breaks, and VT, an improvement of these parameters. The same vocal signals (oscillation of intensity, voice breaks, and VT) also differed between R1 and R2, being similar to pre-
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
5
FIGURE 2. Narrow band spectrogram (Praat software). Example of sustained emission of the vowel /a/ by an individual with adductor spasmodic dysphonia before and 35 days after treatment with botulinum toxin. application values. No differences were observed between preapplication and R2. These results are presented in Tables 3 and 4.
Table 7 shows the changes in the values obtained for each question in the three domains before and after BTX application.
VHI Two subjects had difficulty in filling out the questionnaire due to hearing and visual deficits in one case and to difficulty in understanding the questions in the other. Thirty days after BTX application, there was a reduction of the total score (P = 0.039) obtained with the application of VHI and also a reduction of the emotional domain score (P = 0.035). However, there was no improvement in the physical and functional domains. These results are listed in Table 5. Individual analysis of the total score for each participant revealed that 8 of the 16 subjects showed a reduction of more than 10 points at R1, and in 3 of them there was an increase in the scores obtained with treatment. Comparison of the scores obtained before application and the scores obtained at R2 revealed a reduction of the score for eight individuals. Table 6 lists the VHI values for each individual before application, at R1, and at R2.
Correlation between the perceptual-auditory analysis of vocal parameters and the VHI There was a positive correlation between the degree of severity of voice quality obtained in the perceptual-auditory analysis by the raters and the total VHI score of the subjects. There was a moderate correlation before BTX application (r = 0.664, P ≤ 0.010), a weak correlation at R1 (r = 0.284, P ≤ 0.325), and again a moderate correlation at R2 (r = 0.642, P ≤ 0.025). DISCUSSION Perceptual-auditory analysis is an established tool in clinical practice for the characterization and quantification of the vocal parameters of individuals with vocal disorders,25 as also observed in AdSD and revealing the efficacy of treatment. Inter-rater agreement for the eight vocal parameters proposed in the present study was adequate. Agreement between
ARTICLE IN PRESS 6
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 2. Inter-Rater Agreement in the Perceptual-Auditory Analysis of Each Vocal Parameter Vocal Parameters Degree of severity Strained and strangled voice quality Oscillation of intensity Voice breaks
Vocal tremor
Roughness
Breathiness
Asthenia
Value
95% CI
Time
CCC
LL
UL
Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2
0.97 0.98 0.94 0.99 0.97 0.92 0.99 0.99 0.99 0.99 0.99 0.99 0.98 0.99 0.9 0.97 0.98 0.97 0.99 0.98 0.98 0.98 0.94 0.71
0.91 0.95 0.83 0.97 0.92 0.79 0.96 0.96 0.96 0.98 0.98 0.96 0.94 0.97 0.74 0.92 0.96 0.9 0.97 0.94 0.94 0.94 0.85 0.33
0.99 0.99 0.98 1 0.99 0.97 0.99 1 1 1 1 1 0.99 1 0.97 0.99 0.99 0.99 1 0.99 0.99 0.99 0.98 0.89
Abbreviations: CCC, concordance correlation coefficient; CI, confidence interval; LL, lower limit; UL, upper limit.
assessments is related to the time and quality of training used to teach the raters to recognize the vocal parameters of individuals with AdSD based on the equalization of the vocal parameters to be evaluated between raters. It is also important to establish a reference point for the degree of severity.6 All of these aspects were satisfied in the type of analysis proposed for the present study, with satisfactory agreement thus being obtained between judges.
This aspect was also considered by Eadie et al,18 who investigated the difference in the assessment of overall severity of dysphonia and vocal effort in 20 individuals with AdSD who had completed a treatment cycle. Three groups were selected for judgment: individuals with AdSD, subjects with no experience, and speech-language pathologist with experience. In the cited study, there was no difference among the three groups, with a strong correlation being observed in the judgment of the vocal parameters, although the strategies used differed from those used by the group with experience. Although there were no differences between these raters, it should be pointed out that few studies have reported their experiences about evaluation. On this basis, it is important to reflect about perceptual-auditory assessment in the analysis of AdSD voices as there is wide variability of vocal changes between patients and between treatment times, suggesting that greater training is necessary for raters due to the marked difficulty in assessing the vocal signal of these patients. When the voice parameters were compared in the perceptualauditory analysis before and during BTX treatment regarding the general degree of severity of vocal quality, no changes were detected, although a discrete tendency to a reduction of mean values was observed between pre-application and R1. These data differ from those reported in other studies in which the degree of severity of vocal quality showed changes in the comparison of before and after BTX treatments.20,26,27 However, this difference may have been due to the form of quantitation of the degree of severity, which in the present study showed the highest value among the remaining seven vocal parameters. VT was the parameter showing the greatest change before BTX application, and roughness was the parameter showing the greatest change at R1. Strained-strangled voice quality did not change in the present sample, although the absolute mean values showed a discrete reduction. A reduction of this voice signal was observed in other studies after BTX treatment.20,26,27 A change in the comparison between preapplication and R1 was observed in voice breaks and oscillation of intensity based on the analysis of the narrow-band spectrography, intensity, and pitch contour. The reduced voice breaks suggest a reduction of
TABLE 3. Vocal Parameters Values Before and After Treatment With Botulinum Toxin Preapplication Vocal Parameters Degree of severity Strained-strangled voice quality Oscillation of intensity Voicebreaks Vocal tremor Roughness Breathiness Asthenia Abbreviation: SD, standard deviation.
First return (R1)
Second return (R2)
Range
Mean ± SD
Range
Mean ± SD
Range
Mean ± SD
0.5–8.50 0.00–8.30 0.00–8.40 0.00–8.20 0.00–8.50 0.5–7.10 0.00–6.70 0.00–7.50
5.5 ± 2.18 3.09 ± 3.08 4.48 ± 2.86 4.71 ± 2.80 5.08 ± 2.46 3.93 ± 2.02 2.37 ± 1.95 1.65 ± 2.41
0.4–8.70 0.00–7.50 0.00–6.30 0.00–6.80 0.00–6.30 0.40–8.70 0.00–7.10 0.00–5.60
4.53 ± 2.24 2.17 ± 2.47 2.36 ± 2.23 2.51 ± 2.34 2.64 ± 2.26 3.98 ± 2.29 3.07 ± 2.32 1.97 ± 1.91
2.9–5.60 0.00–8.90 0.00–9.30 0.00–9.00 0.3–8.90 0.30–6.90 0.00–5.80 0.00–4.20
5.83 ± 1.72 2.81 ± 3.07 4.14 ± 2.92 4.24 ± 2.96 4.37 ± 2.62 3.92 ± 1.71 2.63 ± 1.87 1.76 ± 1.95
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
7
Vocal Parameters and Self-Perception on AdSD
TABLE 4. Comparison of Vocal Signals Obtained at the Various Times of Assessment: Preapplication (PRE), First Return (R1), and Second Return (R2)
Time of BTX Treatment Degree of severity of voice quality PRE × R1 PRE × R2 R1 × R2 Strained-strangled voice quality PRE × R1 PRE × R2 R1 × R2 Oscillation of intensity PRE × R1 PRE × R2 R1 × R2 Voice breaks PRE × R1 PRE × R2 R1 × R2 Vocal tremor PRE × R1 PRE × R2 R1 × R2 Roughness PRE × R1 PRE × R2 R1 × R2 Breathiness PRE × R1 PRE × R2 R1 × R2 Asthenia PRE × R1 PRE × R2 R1 × R2
95% CI
Estimated Difference Between Means
P Value
LL
UL
0.95 −0.249 1.199
0.118 0.682 0.056
−0.257 −1.482 −0.034
2.157 0.984 2.432
0.88 0.171 0.708
0.188 0.799 0.297
−0.455 −1.195 −0.658
2.214 1.537 2.074
1.962 0.361 1.601
0.002* 0.54 0.011*
0.795 −0.834 0.406
3.129 1.556 2.796
2.034 0.56 1.473
0.002* 0.361 0.021*
0.856 −0.677 0.236
3.211 1.798 2.711
2.335 0.847 1.489
0.002* 0.231 0.040*
0.981 −0.571 0.071
3.69 2.264 2.906
−0.093 0.019 −0.112
0.887 0.977 0.866
−1.419 −1.332 −1.464
1.233 1.371 1.24
−0.829 −0.187 −0.643
0.23 0.784 0.35
−2.219 −1.576 −2.032
0.56 1.203 0.747
−0.322 −0.035 −0.287
0.639 0.96 0.688
−1.717 −1.487 −1.739
1.072 1.416 1.165
Abbreviations: LL, lower limit; UL, upper limit. * Indicates the time in which the comparison of the vocal parameters was performed.
adduction spasms of the vocal folds, this being the major vocal parameter presented by individuals with AdSD, affecting speech intelligibility. The results obtained here are similar to those reported by others even though the present study was based only on perceptual-auditory analysis.17,18,27,28 The oscillation of intensity parameter was used, taking into consideration the tool for the visualization of intensity contour offered by the Praat software. Although this parameter is not easy to assess, this analysis was based simultaneously on perceptive-auditory analysis. Thus, the oscillation of intensity parameter was anchored on voice analysis, representing a parameter of the change between preapplication and the first return. These differences can be observed in Figure 1. There were oscillations of intensity and voice breaks also between R1 and R2, although these values increased, returning to the levels observed before BTX application. This is related
to the time of effectiveness of the treatment reported in various studies, which ranges from 3 to 4 months.27,29,30 Of the 16 individuals studied here, 81.25% showed associated VT, which was the most altered parameter in perceptualauditory analysis, with the highest values being obtained using the visual analogue scale. After BTX application, these values were reduced at R1, increasing again when the effect of BTX ended at the R2 assessment, contradicting the results reported by White et al,30 who demonstrated that the subjects with VT did not show improvement of this parameter after the use of BTX. The parameters roughness, breathiness, and asthenia did not change, suggesting that, although BTX treatment is beneficial, it also triggers a series of other vocal parameters considered to be iatrogenic effects, such as paresis of the thyroarytenoid muscle,17 derived from the inhibitory action on muscle contraction resulting from the blockade of acetylcholine at the junction,
ARTICLE IN PRESS 8
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 5. Voice Handicap Index Before and After the Application of Botulinum Toxin Preapplication Mean ± SD
R1 Mean ± SD
Total score 67.79 ± 26.52
58.21± 27.30
Emotional domain 20.43 ± 12.30
R2 Mean ± SD
P Value
60 ± 28.5
PRE × R1 = 0.039* PRE × R2 = 0.108 R1 × R2 = 0.785
16.36 ± 13.0
15.83 ± 14.0
PRE × R1 = 0.035* PRE × R2 = 0.04* R1 × R2 = 0.60
Physical domain 23.36 ± 7.33
20.43 ± 9.12
22.17 ± 22.50
PRE × R1 = 0.14 PRE × R2 = 0.24 R1 × R2 = 0.57
Functional domain 24 ± 10.2
21.43 ± 11.14
19.50 ± 11.38
PRE × R1 = 0.12 PRE × R2 = 0.03* R1 × R2 = 0.36
Abbreviations: PRE, preapplication; R1, first return; R2, second return; SD, standard deviation. * Indicates the time in which the comparison of the vocal parameters was performed.
to be resolved. The response to the instrument permits the establishment of future projections regarding the dose and duration of BTX treatment, the emotional adaptation to the problems faced, and the social impact. The evaluation of the 16 subjects studied here revealed that the total VHI scores were reduced in eight of them when the pretreatment values were compared with those obtained after 30 days of BTX treatment, with the mean general VHI score changing from 67.79 to 58.21 at R1, in agreement with data reported by Morzaria and Damrose,31 who obtained a total score that varied
a mechanism that not only helps reduce the adductor spasms but also causes changes such as breathiness,12,17,20 worsening of the roughness signal with a predominance of hoarseness,18 and reduction of intensity or asthenia26 as a result of BTX action. The assessment of the vocal impact and handicap of voice disorders using questionnaires is a useful tool for learning about the implications of the disease in the life of affected individuals. In this respect, the response to VHI is helpful in the determination of self-perception of both the impact of dysphonia and the effect of treatment, also identifying complications
TABLE 6. Voice Handicap Index: Individual Score Before and After the Application of Botulinum Toxin Total Score
Functional Domain
Emotional Domain
Physical Domain
Individual
PRE
R1
R2
PRE
R1
R2
PRE
R1
R2
PRE
R1
R2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
– 94 66 53 67 80 82 72 41 101 56 16 107 – 30 84
– 101 31 32 56 90 49 47 41 88 66 16 99 – 36 63
– 71 55 40 59 82 – 65 18 105 34 16 110 – 35 –
– 34 24 15 23 29 35 28 11 34 20 6 39 – 11 27
– 35 7 8 23 29 25 24 11 39 24 6 36 – 12 27
– 24 16 9 24 25 – 26 2 36 12 8 39 – 13 –
– 28 18 12 18 30 31 18 13 39 4 2 39 – 6 28
– 30 10 8 9 34 14 8 13 37 4 0 40 – 8 14
– 16 17 9 9 34 – 15 4 39 0 2 39 – 6 –
– 32 24 26 26 21 16 26 17 28 32 8 29 – 13 29
– 6 14 16 24 27 10 15 17 12 38 10 23 – 16 28
– 31 22 22 26 23 – 24 12 30 22 6 32 – 16 –
Abbreviations: PRE, preapplication; R1, first return; R2, second return.
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
9
Vocal Parameters and Self-Perception on AdSD
TABLE 7. Distribution of the Values Obtained for Each Question of Each Domain (Emotional, Organic, and Functional) of the Voice Handicap Index PRE Question
Median
Emotional domain 7 3.0
R1
R2
Mean ± SD
Median
Mean ± SD
Median
Mean ± SD
2.5 ± 1.6
2.0
2.1 ± 1.6
2.0
1.9 ± 1.6
9
1.0
1.2 ± 1.3
1.0
1.5 ± 1.6
1.0
1.2 ± 1.3
15
2.0
1.8 ± 1.4
2.0
1.7 ± 1.3
1.5
1.4 ± 1.5
23
4.0
2.8 ± 1.7
3.0
2.2 ± 1.6
2.5
2.5 ± 1.5
24
4.0
2.7 ± 1.5
2.0
2.0 ± 1.6
1.0
1.7 ± 1.8
25
2.5
2.1 ± 1.8
1.5
1.4 ± 1.5
2.0
1.8 ± 1.8
27
2.0
2.0 ± 1.3
0.5
1.4 ± 1.6
1.0
1.5 ± 1.7
28
1.5
1.7 ± 1.6
0
1.3 ± 1.7
1.0
1.5 ± 1.6
29
0
1.4 ± 1.8
0
0.9 ± 1.6
0
1.0 ± 1.6
30
2.0
1.7 ± 1.7
0
1.4 ± 1.7
0
1.0 ± 1.8
Physical domain 2 1.0
1.4 ± 1.6
2.0
1.1 ± 1.0
2.0
1.9 ± 1.4
4
2.0
2.2 ± 1.2
2.0
2.1 ± 1.5
2.0
1.7 ± 1.6
10
3.0
2.9 ± 1.2
4.0
3.1 ± 1.2
3.0
2.6 ± 1.4
13
2.5
2.4 ± 1.5
4.0
3.0 ± 1.5
2.5
2.5 ± 1.4
14
3.5
2.8 ± 1.4
2.0
1.7 ± 1.4
2.0
2.3 ± 1.4
17
3.0
2.3 ± 1.7
3.0
2.6 ± 1.4
2.5
2.2 ± 1.5
18
2.0
2.2 ± 1.3
2.0
2.2 ± 1.7
2.5
2.4 ± 1.6
P PRE × R1 = 0.290 PRE × R2 = 0.111 R1 × R2 = 0.809 PRE × R1 = 0.486 PRE × R2 = 1.000 R1 × R2 = 0.795 PRE × R1 = 0.752 PRE × R2 = 0.206 R1 × R2 = 0.256 PRE × R1 = 0.241 PRE × R2 = 0.820 R1 × R2 = 0.638 PRE × R1 = 0.127 PRE × R2 = 0.096 R1 × R2 = 0.139 PRE × R1 = 0.127 PRE × R2 = 0.870 R1 × R2 = 0.417 PRE × R1 = 0.179 PRE × R2 = 0.417 R1 × R2 = 0.754 PRE × R1 = 0.054 PRE × R2 = 0.723 R1 × R2 = 0.429 PRE × R1 = 0.169 PRE × R2 = 0.339 R1 × R2 = 1.000 PRE × R1 = 0.218 PRE × R2 = 0.082 R1 × R2 = 0.175 PRE × R1 = 0.302 PRE × R2 = 0.376 R1 × R2 = 0.056 PRE × R1 = 0.890 PRE × R2 = 0.206 R1 × R2 = 0.328 PRE × R1 = 0.336 PRE × R2 = 0.214 R1 × R2 = 0.087 PRE × R1 = 0.273 PRE × R2 = 0.748 R1 × R2 = 0.207 PRE × R1 = 0.019* PRE × R2 = 0.394 R1 × R2 = 0.194 PRE × R1 = 0.486 PRE × R2 = 0.889 R1 × R2 = 0.714 PRE × R1 = 1.000 PRE × R2 = 0.220 R1 × R2 = 0.429 (continued on next page)
ARTICLE IN PRESS 10
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 7. (continued ) PRE
R1
R2
Median
Mean ± SD
Median
Mean ± SD
Median
Mean ± SD
P
20
3.0
2.2 ± 1.5
0.5
1.2 ± 1.7
2.0
2.0 ± 1.7
21
2.0
2.0 ± 1.6
2.0
1.5 ± 1.5
1.0
1.25 ± 1.4
26
3.0
2.5 ± 1.6
2.0
1.4 ± 1.4
3.0
0.5 ± 1.2
PRE × R1 = 0.047* PRE × R2 = 0.220 R1 × R2 = 0.107 PRE × R1 = 0.151 PRE × R2 = 0.060 R1 × R2 = 0.504 PRE × R1 = 0.82 PRE × R2 = 0.95 R1 × R2 = 0.023*
Functional domain 1 2.5
2.7 ± 1.05
2.0
2.1 ± 1.4
2.0
2.0 ± 1.5
Question
3
3.5
3.2 ± 1.6
3.0
3.0 ± 1.0
2.5
2.2 ± 1.3
5
2.5
2.1 ± 1.6
2.5
2.4 ± 1.3
2.0
2.5 ± 1.4
6
4.0
3.0 ± 1.5
3.5
2.7 ± 1.6
2.5
2.5 ± 1.5
8
3.0
2.3 ± 1.7
1.5
1.7 ± 1.8
2.5
1.9 ± 1.7
11
3.0
2.6 ± 1.5
3.0
2.5 ± 1.6
2.5
2.0 ± 1.5
12
3.0
2.9 ± 1.2
2.0
2.2 ± 1.1
2.5
2.5 ± 1.3
16
1.5
1.7 ± 1.6
1.0
1.5 ± 1.6
1.0
1.5 ± 1.7
19
1.5
1.7 ± 1.6
0
1.2 ± 1.7
1.0
1.5 ± 1.6
22
1.0
1.4 ± 1.6
1.0
1.7 ± 1.9
0
0.5 ± 1.2
PRE × R1 = 0.082 PRE × R2 = 0.095 R1 × R2 = 0.870 PRE × R1 = 0.336 PRE × R2 = 0.026* R1 × R2 = 0.136 PRE × R1 = 0.890 PRE × R2 = 0.206 R1 × R2 = 0.328 PRE × R1 = 0.500 PRE × R2 = 0.504 R1 × R2 = 0.896 PRE × R1 = 0.026* PRE × R2 = 0.082 R1 × R2 = 0.191 PRE × R1 = 0.807 PRE × R2 = 0.147 R1 × R2 = 0.230 PRE × R1 = 0.075 PRE × R2 = 0.389 R1 × R2 = 0.504 PRE × R1 = 0.699 PRE × R2 = 0.777 R1 × R2 = 0.339 PRE × R1 = 0.205 PRE × R2 = 0.820 R1 × R2 = 0.795 PRE × R1 = 0.418 PRE × R2 = 0.111 R1 × R2 = 0.085
Abbreviations: PRE, preapplication; R1, first return; R2, second return; SD, standard deviation. * Indicates the time in which the comparison of the vocal parameters was performed.
from 69.9 to 51.1 after application of the toxin. Although there was a reduction in the scores, these values were above the cutoff note of 19 points proposed for this instrument,32 suggesting that self-perception reveals a greater recognition of the impact of the disease on the quality of life of patients and certifies their involvement with respect to treatment. However, it should be pointed out that there is divergence in the response to treatment, perhaps due to the method of toxin application, the technique used, whether the toxin is applied unilaterally or bilaterally, the quantity of the medication used, the
time of assessment after BTX injection, or factors inherent to the individual and to the way of filling out the questionnaire, with consequent changes in the results obtained. The present study disagreed with other investigations that reported lower scores for the effect of BTX application determined by VHI compared with the present data.12,31 However, it is recognized that treatment with BTX is of benefit for both the phonatory function and the quality of life of these individuals.12,17 Detailed descriptions of the questions of each domain permit us to understand the main aspects that can affect the personal
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
and social life of persons with AdSD regarding oral communication and the way BTX treatment can change or reduce this disadvantage. Each domain has a specific objective. The emotional domain involves the feelings of the person regarding his or her phonation and communication with people as the voice provokes and reveals reactions from the listener. In this domain, most of the present patients reported a marked impact on items 7, 23, and 24 of the VHI, showing that subjects with spasmodic dysphonia feel frustrated and discouraged,33 with difficulties in their interpersonal relations even though they try to accept the fact that they have a voice disorder that should not be the reason for a low self-esteem. This was observed in items 29 and 30, which did not change with treatment, showing a frequency of “sometimes” and “never.” The application of BTX according to the established protocol was unable to modify the emotional perceptions of individuals regarding their voice. This emotional domain should be considered for future proposed treatments. The objective of the functional domain is to determine how AdSD changes daily life activities and social interaction. Regarding the work aspect, the patients reported that their voice problem did not cause economic losses (item 22), a result that requires further study as many of the patients are already retired. Other jobs held by the patients involve tasks that do not depend fully on speech. This impairs their quantitation, with the need for further investigation of this question involving a socioeconomic study.17,34 Regarding communication, item 8, which refers to avoiding groups of persons because of one’s voice, revealed that the subjects improve, progressing from an “almost always” frequency to an “almost never” frequency, as there was no change in the frequency of facts that are part of daily life, as shown in items 3, 6, and 11. The physical domain evaluates the auditory and kinesthetic experiences of individuals regarding their voice. There was an evident presence of hoarseness (item 13) before treatment, which worsened after treatment, a result related to the iatrogenic effect of the temporary muscle paresis that occurs after BTX injection,17 and also related to the effort to speak (items 14 and 20) and to voice breaks in the middle of speech (item 26), which were reduced after treatment and were associated with the symptoms reported by patients with AdSD.33–35 Regarding item 18, which explores whether the patients try to change their voice, the predominant result was “at times” and did not change after BTX application. This reply expresses the coping strategies for a positive or negative situation and the reduced occupational and social participation of the persons.33 Before BTX application, the correlation between perceptualauditory evaluation and total VHI score in terms of degree of severity was moderate. Perceptual-auditory analysis revealed a higher degree for VT, voice breaks, oscillation of intensity, and strained-strangled voice quality, suggesting a relationship with the higher score for the physical domain of the VHI. There was a complaint regarding the effort to speak (items 14 and 20) in agreement with the presence of strained-strangled voice, and the perception of voice breaks (item 26), explained by the oscillation of intensity which alters speech intelligibili-
11
ty and consequently affects the communication of persons with AdSD. At R1, the correlation was weak, with contradictory results being observed as, despite the clear presence of positive changes in the vocal quality of individuals and the improvement observed by perceptual-auditory evaluation, the impact and the vocal disadvantage of AdSD along time cause changes in the emotional and functional aspects of the individuals, affecting their relationship with others and with their environment. Another aspect to be considered is that the iatrogenic effects of BTX application, such as hoarseness and breathiness, were relevant for the patients. Thus, after application of the botulin toxin, patients continued to be dysphonic, although the vocal changes observed did not limit the use of the voice and communication, with a reduction of the vocal impact. However, the changes due to the application of the toxin were still important for the patients studied here. In addition, it is important to point out that, in contrast to other vocal diseases such as vocal nodules and polyps, among others, AdSD is a chronic disease that can be treated but not cured, with the patients ultimately adapting to their lifestyle in the presence of the disease.36 This fact was confirmed in the present study, in which the functional domain did not show significant changes after BTX application. A moderate correlation was observed at R2, related to the duration of the effect of BTX treatment and to the onset of the vocal signs present in AdSD. In view of the above data, the effects of 30 days of BTX treatment were clearly observed, with a return to the vocal parameters observed before treatment at 120 days after application of the toxin.
CONCLUSIONS The present results suggest that perceptual-auditory analysis using narrow-band spectrography and pitch and intensity contour can be considered as an alternative for the identification and quantitation of vocal parameters in individuals with AdSD in order to assess the results of treatment in clinical practice. The parameters of oscillation of intensity, voice breaks, and VT were those that most benefited from the application of BTX, although the benefits did not persist for 120 days. BTX treatment had a positive effect on the self-perception of the subjects measured with the VHI, with a reduction of the vocal symptoms of patients with AdSD and an improvement of voice handicap. However, considering the correlation between the general degree of voice quality and the total VHI score, it is necessary to conduct further studies that will establish a relationship among the time of diagnosis, duration of treatment, and the acquisition of auditory and tactile-kinesthetic perception skills that will allow AdSD patients to identify the vocal signals present before and after BTX treatment. Studies with a larger number of participants using chained speech samples and recording of the results 60 and 90 days after BTX injection are needed, contributing even more to the guidance of treatment of subjects with AdSD.
ARTICLE IN PRESS 12 Acknowledgment We wish to acknowledge Diego Espitia Rojas for the elaboration of the reply form for perceptual-auditory assessment using Microsoft Excel software for the quantitation of vocal parameters. REFERENCES 1. Ludlow C. Spasmodic dysphonia: a laryngeal control disorder specific to speech. J Neurosci. 2011;31:793–797. 2. Meyer TK. The treatment of laryngeal dystonia (spasmodic dysphonia) with botulinum toxin injections. Oper Tech Otolaryngol. 2012;23:96– 101. 3. Roy N. Differential diagnosis of muscle tension dysphonia and spasmodic dysphonia. Curr Opin Otolaryngol Head Neck Surg. 2010;18:165–170. 4. Blitzer A. Spasmodic dysphonia and botulinum toxin: experience from the largest treatment series. Eur J Neurol. 2010;17(suppl 1):28–30. 5. Blitzer A, Brin MF, Stewart CF. Botulinum toxin management of spasmodic dysphonia (laryngeal dystonia): a 12 year experience in more than 900 patients. Laryngoscope. 1998;108:1435–1441. 6. Barkmeir J, Case J, Ludlow CL. Identification of symptoms for spasmodic dysphonia and vocal tremor: comparison of expert and nonexpert judges. J Commun Disord. 2001;34:21–37. 7. Watts C, Nye C, Whurr R. Botulinum toxin for treating spasmodic dysphonia (laryngeal dystonia): a systematic Cochrane review. Clin Rehabil. 2006;20:112–122. 8. Bender BK, Cannito MP, Murry T, et al. Speech intelligibility in severe adductor spasmodic dysphonia. J Speech Lang Hear Res. 2004;47:21–32. 9. Branden M, Johns M, Klein A, et al. Assessing the effectiveness of botulin toxin injections for adductor spasmodic dysphonia: clinician and patient perception. J Voice. 2008;24:242–248. 10. Baylor C, Yorkston K, Tanya E. The consequences of spasmodic dysphonia on communication-related quality of life: a qualitative study of the insider’s experiences. J Commun Disord. 2005;38:395–419. 11. Bielamowicz S, Ludlow CL. Effects of botulinum toxin on the pathophysiology in spasmodic dysphonia. Ann Otol Rhinol Laryngol. 2000;109:194–203. 12. Novakovic D, Waters H, D’Elia J, et al. Botulinum toxin treatment of adductor spasmodic dysphonia: longitudinal functional outcomes. Laryngoscope. 2011;121:606–612. 13. Kendall KA, Leonard RJ. Interarytenoid muscle botox injection for treatment of adductor spasmodic dysphonia with vocal tremor. J Voice. 2011;25:114– 119. 14. Ludlow CL, Naunton RF, Sedory SE, et al. Effects of botulinum toxin injections on speech in adductor spasmodic dysphonia. Neurology. 1988;38:1220–1225. 15. Blitzer A. Botulinum toxin A and B: a comparative dosing study for spasmodic dysphonia. Otolaryngol Head Neck Surg. 2005;133:836– 838. 16. Silverman EP, Garvan C, Shrivastav R, et al. Combined modality treatment of adductor spasmodic dysphonia. J Voice. 2012;26:77–86. 17. Denjockere PH, Neumann KJ, Moerman MB, et al. Tridimensional assessment of adductor spasmodic dysphonia pre- and post-treatment with botulinum toxin. Eur Arch Otorhinolaryngol. 2012;4:1195–1203.
Journal of Voice, Vol. ■■, No. ■■, 2016 18. Eadie TL, Nicolici C, Baylor C, et al. Effect of experience on judgments of adductor spasmodic dysphonia. Ann Otol Rhinol Laryngol. 2007;116:695– 701. 19. Murry T. Spasmodic dysphonia: let’s look at that again. J Voice. 2014;28:694–699. 20. Cannito MP, Woodson GE, Murry T, et al. Perceptual analyses of spasmodic dysphonia before and after treatment. Arch Otolaryngol Head Neck Surg. 2004;130:1393–1399. 21. Liu TC, Irish JC, Adams SG, et al. Prospective study of patients’ subjective responses to botulinum toxin injection for spasmodic dysphonia. J Otolaryngol. 1996;25:66–74. 22. Jacobson BH, Johnson A, Grywalski C, et al. The Voice Handicap Index (VHI): development and validation. Am J Speech Pathol. 1997;6:66–70. 23. Behlau M, Alves dos Santos LM, Oliveira G. Cross-cultural adaptation and validation of the voice handicap index into Brazilian Portuguese. J Voice. 2011;25:354–359. 24. Boersma P, Weenink D. Praat: doing phonetics by computer. Institute of Phonetic Sciences, University of Amsterdam, Amsterdam, 1995. Available at: http://www.fon.hum.uva.nl/praat/. Accessed July 4, 2012. 25. Kreiman J, Gerratt BR, Kempster GB, et al. Perceptual evaluation of voice quality: review, tutorial, and a framework for future research. J Speech Hear Res. 1993;36:21–40. 26. Langeveld TP, Luteijn F, van Rossum M, et al. Adductor spasmodic dysphonia and botulinum toxin treatment: the effect on well-being. Ann Otol Rhinol Laryngol. 2001;110:941–945. 27. Tisch SH, Brake HM, Law M, et al. Spasmodic dysphonia: clinical features and effects of botulinum toxin therapy in 169 patients-an Australian experience. J Clin Neurosci. 2003;10:434–438. 28. Boutsen F, Cannito MP, Taylor M, et al. Botox treatment in adductor spasmodic dysphonia: a meta-analysis. J Speech Lang Hear Res. 2002;45: 469–481. 29. Cloud LJ, Jinnah HA. Treatment strategies for dystonia. Expert Opin Pharmacother. 2010;11:5–15. 30. White LJ, Klein AM, Hapner ER, et al. Co-prevalence of tremor with spasmodic dysphonia: a case-control study. Laryngoscope. 2011;121:1752– 1755. 31. Morzaria S, Damrose E. A comparison of the VHI, VHI-10, and V-RQOL for measuring the effect of botox therapy in adductor spasmodic dysphonia. J Voice. 2012;26:378–380. 32. Behlau M, Madazio G, Moreti F, et al. Efficiency and cutoff values of self-assessment instruments on the impact of a voice problem. J Voice. 2016;30:506.e9–506.e18. 33. Smith E, Taylor M, Mendoza M, et al. Spasmodic dysphonia and vocal fold paralysis: outcomes of voice problems on work-related functioning. J Voice. 1998;12:223–232. 34. Baylor C, Yorkstonb M, Eadi T. The consequences of spasmodic dysphonia on communication-related quality of life: a qualitative study of the insider’s experiences. J Commun Disord. 2005;28:395–419. 35. Benninger M, Gardner G, Grywalski C. Outcomes of botulinum toxin treatment for patients with spasmodic dysphonia. Arch Otolaryngol. 2001;127:1083–1085. 36. Hu A, Hillel A, Meyer T. Factors associated with patient-perceived hoarseness in spasmodic dysphonia patients. J Voice. 2015;doi:10.1016/ j.jvoice.2015.11.003. In press, 1–4.
iatrogenic characteristics of the treatment, such as the presence of voice breathiness, 12 reduction of vocal intensity, hoarseness, alteration of airflow, effort to speak, fatigue, and reduction of the range of fundamental frequency in semitones.13 The duration of the effect of treatment varied from 3 to 4 months.14–16 The effect of treatment with BTX has been demonstrated by different methods, such as vocal acoustic measurements, aerodynamic measurements, laryngoscopy, perceptualauditory analysis performed by judges, 9,17–19 and selfperception of voice quality by the patient.17 However, there is no information that might permit to draw conclusions about a standardized tool or that might permit the quantitation of the time of benefit and of the reduction of voice changes detected in AdSD. In view of the above considerations, it is important to review evidence in the literature,17,18,20 and according to the experience with patients to construct a tool to be used in the clinical setting without the need for high-cost technology to permit the assessment of vocal signs, together with visual analysis of narrow band spectrography and of recordings of pitch intensity for the follow-up of BTX treatment. BTX treatment leads to improvement of vocal symptoms, although persons with AdSD do not consider their voice to be normal.20,21 On this basis, there is concern about determining the impact of treatment on the voice quality of patients with dystonia, with a search for different perspectives for the analysis of the effect of medication on voice quality, interpersonal communication and communicative performance during daily activities, and also with the identification of patient satisfaction regarding the duration and effect of treatment.
ARTICLE IN PRESS 2 At present, one of the questionnaires more frequently used as an indicator of the effectiveness of treatment with BTX regarding the improvement of symptoms and of the quality of life of patients with AdSD is the Voice Handicap Index (VHI),22 which is available in the version validated for the Portuguese language.23 Thus, the objective of the present study was to compare and to correlate the perceptual-auditory analysis of vocal parameters and self-perception in individuals with AdSD before and after the application of type A BTX. METHODS This was a prospective cohort study which was approved by the Research Ethics Committee (Protocol N0 8339/2013). All subjects gave written informed consent to participate. Participants The study was conducted on 16 individuals with a clinical diagnosis of AdSD, 11 (68.75%) were women and 5 (31.25%) were men, ranging in age from 25 to 90 years (mean: 57 years and 2 months). Of these, 81.25% had associated vocal tremor (VT) and other movement disorders associated with AdSD that did not promote changes in vocal production, ie, blepharospasm and cervical dystonia (12.5%), cervical dystonia (6.25%), generalized dystonia (6.25%), hemidystonia (6.25%), oromandibular tremor (6.25%), tremor in the cephalic region, cervical dystonia, and blepharospasm (6.25%). Only patients with a diagnosis of adductor laryngeal dystonia participated in the study, whereas patients with other types of laryngeal dystonia were excluded. The subjects included were individuals who were receiving treatment for the first time or who needed to be submitted to a new BTX injection as the last application had occurred more than 120 days (mean: 223.31 days; range: 161–412 days) before, and there were no clinical signs of the effect of the medication since the last application. The 16 selected individuals had received, on average, 3.9 BTX injections before the current application. The Pearson coefficient was applied for the inclusion of these subjects in order to determine the relationship between the degree of severity of vocal quality at the preapplication time and the time since the last application of the botulin toxin (number of days). A correlation coefficient of 0.41 (P value = 0.11) was observed, permitting us to conclude that time has no influence on vocal severity since last application. This permitted us to consider all patients to be equal even with different time since the last application. The following were the exclusion criteria: individuals with previous laryngeal surgeries, previous dysphonia, and language, speech, or hearing disorders, or dementia reported by individuals themselves or their relatives, or obtained from their medical records. Also excluded were patients with other neurologic changes, such as dysarthria, Parkinson’s disease, ataxia, cerebellar atrophy, sequelae of cerebrovascular accidents, and others. Procedures All patients were submitted to the standardized application of type A BTX (Dysport®) (Wrexham, United Kingdom) by the introduction of a needle/electrode through the cricothyroid mem-
Journal of Voice, Vol. ■■, No. ■■, 2016
brane at 0.5 cm from the median sagittal line at a lateral and superior angle of 30°–45° until reaching the thyroarytenoid muscle guided by electromyography. Twenty units of type A BTX were applied unilaterally on the left side of the larynx to all patients. All participants were submitted to the recording of the vocal signal and responded to the VHI, a questionnaire validated for the Portuguese language,23 before BTX injection, on the first return (R1) held on average 36 days later (range: 28–49 days), and on the second return (R2) held on average 137 days after application (range: 112–189 days). Sound signal capture The voice sound signal was captured and recorded using a professional microphone G-Track (Samson, Technologies, New York) and the Sound Forge 6.0 software (Sony Creative Software Inc, Middleton, Wisconsin, United States) at a sampling rate of 44,100 Hz and 16 bits, and stored in wav format. The procedure was carried out in an acoustically treated room with noise kept at less than 50 dB and monitored with a digital sound level meter. All participants performed prolonged emission of the vowel /a/, in Portuguese, three consecutive times at habitual vocal intensity and frequency at each time of assessment (preapplication, R1, and R2). The more stable sound signal was extracted visually from the vocal sample of the 16 subjects, the beginning and end of emission were discarded because of their irregular characteristics, and the best sound signal with at least 3 seconds of sound production was accepted. Instrument for assessment Eight vocal parameters, described in Table 1, were proposed for the perceptual-auditory assessment of subjects with AdSD. This analysis was performed based on auditory stimuli (sustained emission of the vowel /a/) and on visual stimuli obtained with the free access Praat software, version 53.42,24 permitting the visualization of the pitch and intensity contour of the vocal signal (Figure 1), and with narrow band spectrography considering the following configurations: bandwidth of 43 Hz and 0.03 (30 ms), frequency range from 0 to 5000 Hz, time step of 0.002 seconds, and frequency step of 20 Hz (Figure 2). The vocal parameters were quantified using a visual analogue scale consisting of a 10-cm line with anchors at each end, with 0 on the left end (no change or absence) and 10 on the right end (severe change or permanent presence). A response form was elaborated for data recording using Microsoft Office Excel (Redmond, Washington, United States), permitting the analysis of each of the eight vocal parameters using the cursor to mark the values on a line. The values were then recorded automatically and transferred to a spreadsheet. Perceptual-auditory analysis of vocal parameters The analysis was carried out by two specialized speechlanguage pathologist with a mean of 10 years of clinical experience in voice analysis and without hearing complaints, with judge 1 (J1) being the most experienced. The previously edited 47 voice recordings corresponding to the 16 subjects tested before the application of BTX and at R1, and to 15 individuals tested
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
3
TABLE 1. Description of the Vocal Parameters Assessed by Perceptual-Auditory Analyses Vocal Parameters Degree of severity of vocal quality Strained-strangled voice quality Oscillation of intensity
Voice break
Vocal tremor Roughness Breathiness Asthenia
Concept Overall impression of vocal quality using the analysis of the seven vocal parameters as reference Compressed and choppy sound with fluctuations of voice quality Unpredictable and uncontrollable variations of the level of intensity, changing from explosive to weakened over time Intensity contour: fluctuation of the intensity tracing in an ascending or descending manner, represented by short- or long-duration peaks during emission Narrow band spectrogram: changes in harmonic stability that becomes more obscure during the duration of the oscillation of intensity Abrupt and unexpected fall of phonation Pitch contour: interruption of the pitch tracing during emission, with a descending fall of the tracing being observed before the interruption and with a return to the values (Hz) close to those present where the interruption occurred Narrow band spectrogram: interruption of harmonics throughout the entire frequency range for the duration of the voice break, with background noise filling out until the return of the sound signal Presence of involuntary rhythmic and regular variations in frequency and intensity Narrow band spectrogram: harmonic ondulations throughout emission in all frequency ranges Audible occurrence of acoustic aperiodicity perceived in the sound signal as a rough, crackling, and tense-strangled voice Audible occurrence of respiratory noise or acoustic turbulence perceived in the sound signal Vocal weakness, loss of potency, reduced vocal energy
at R2, were exhibited in a randomized manner using the Praat software, version 53.42.24 To ensure a consistent response on the part of the judges, the process was divided into the following phases: training of the judges when the perceptual-auditory signals were conceptually exposed intra-examiner calibration and inter-examiner calibration. For the training of judges, which had a duration of 24 hours (3 days, with resting intervals), conceptual exposure of the vocal parameters to be assessed was performed based on perceptualauditory analysis and narrow band spectrography, and the contour of pitch and intensity, as shown in Table 1. Intra-examiner calibration consisted of the presentation of voice bank samples of individuals with AdSD who did not participate in the present study, repeated as many times as necessary for the examiners to understand the parameters to be assessed. The inter-examiner calibration consisted of the interaction between judges regarding the parameters, correlating the stimuli in order to reach a consensus about the differences recognized auditorily and visually, with the establishment of criteria in order to reach an acceptable level of agreement in their findings regarding the presence of vocal changes in individuals with AdSD. The second phase, test, consisted of a 6-hour test applied 1 day after training, when each rater received the response form in digital format. The voices were reproduced for the judges, with speakers positioned at a mean distance of 1 m, with no variation in volume. The stimuli were presented in a random manner, with the judges being unaware of the subject or of the time of treatment. The third phase, retest, was performed 10 days after the test to determine intra-rater agreement, with 20% of the total voice sample being presented (10 recordings).
VHI The questionnaire is applied in order to assess the impact and handicap of the voice of the patient. It contains 30 questions divided into three domains—emotional, functional, and physical— each containing 10 questions. Each item is measured with a Likert scale, which consists of 5 points: 0 = never, 1 = almost never, 2 = sometimes, 3 = almost always, and 4 = always. The questionnaire was filled out by the following methods: an interview when patients were unlearned or had visual problems, or completed by the patients themselves, always under the supervision of the investigator. The VHI scores were calculated as the simple sum of all replies, providing the total score and the score for each domain. The analysis of the 30 items of the VHI permitted the determination of the vocal impact. Statistical analysis The intraclass correlation coefficient (ICC) was used to determine the inter-rater agreement at the time of the test and intrarater agreement between test and retest. Comparison of the individuals at the three time points for each variable studied (vocal parameters) considering J1 responses was performed by means of orthogonal contrasts using a mixed-effects linear model (random and fixed effects). The linear model is used in the analysis of data in which the responses of the same individual are grouped and when the assumption of independence between observations in the same group is not adequate. In order to use this model, it is necessary for its residues to have normal distribution with zero mean and constant variance. The model was adjusted with the PROC MIXED procedure of the SAS 9.1 software (Cary, North Carolina, United States).
ARTICLE IN PRESS 4
Journal of Voice, Vol. ■■, No. ■■, 2016
FIGURE 1. Recording of pitch and intensity (Praat software). Example of sustained emission of the vowel /a/ by an individual with adductor spasmodic dysphonia before and 35 days after treatment with botulinum toxin.
The Student t test for paired samples was used to compare VHI scores between individuals at the three time points, with the level of significance set at P ≤ 0.05. The correlation between the total VHI scores and the degree of severity of vocal quality obtained in the perceptual-auditory analysis was determined using the Pearson correlation coefficient. All analyses were carried out with the aid of the SPSS (Statistical Package for the Social Sciences) (Chicago, Illinois, United States) software, version 20.0. RESULTS Inter- and intra-rater agreement The analyses carried out by the two judges demonstrated excellent agreement (ICC = 0.9) for most parameters, except for asthenia, for which the agreement was only satisfactory (ICC = 0.71) (Table 2).
Intra-rater agreement between test and retest determined by the repetition of 20% of the voices previously analyzed showed excellent or satisfactory agreement, with ICC ranging from 0.8 to 0.5 for the following parameters: degree of severity of voice quality, Oscillation of intensity, strained-strangled voice quality, and voice breaks. For the remaining vocal parameters, the agreement of the values obtained at rest was less intense. Perceptual-auditory analysis of vocal parameters Comparison of the vocal parameters of the 16 individuals before BTX injection and at time R1 revealed a reduction of the oscillation of intensity, voice breaks, and VT, an improvement of these parameters. The same vocal signals (oscillation of intensity, voice breaks, and VT) also differed between R1 and R2, being similar to pre-
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
5
FIGURE 2. Narrow band spectrogram (Praat software). Example of sustained emission of the vowel /a/ by an individual with adductor spasmodic dysphonia before and 35 days after treatment with botulinum toxin. application values. No differences were observed between preapplication and R2. These results are presented in Tables 3 and 4.
Table 7 shows the changes in the values obtained for each question in the three domains before and after BTX application.
VHI Two subjects had difficulty in filling out the questionnaire due to hearing and visual deficits in one case and to difficulty in understanding the questions in the other. Thirty days after BTX application, there was a reduction of the total score (P = 0.039) obtained with the application of VHI and also a reduction of the emotional domain score (P = 0.035). However, there was no improvement in the physical and functional domains. These results are listed in Table 5. Individual analysis of the total score for each participant revealed that 8 of the 16 subjects showed a reduction of more than 10 points at R1, and in 3 of them there was an increase in the scores obtained with treatment. Comparison of the scores obtained before application and the scores obtained at R2 revealed a reduction of the score for eight individuals. Table 6 lists the VHI values for each individual before application, at R1, and at R2.
Correlation between the perceptual-auditory analysis of vocal parameters and the VHI There was a positive correlation between the degree of severity of voice quality obtained in the perceptual-auditory analysis by the raters and the total VHI score of the subjects. There was a moderate correlation before BTX application (r = 0.664, P ≤ 0.010), a weak correlation at R1 (r = 0.284, P ≤ 0.325), and again a moderate correlation at R2 (r = 0.642, P ≤ 0.025). DISCUSSION Perceptual-auditory analysis is an established tool in clinical practice for the characterization and quantification of the vocal parameters of individuals with vocal disorders,25 as also observed in AdSD and revealing the efficacy of treatment. Inter-rater agreement for the eight vocal parameters proposed in the present study was adequate. Agreement between
ARTICLE IN PRESS 6
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 2. Inter-Rater Agreement in the Perceptual-Auditory Analysis of Each Vocal Parameter Vocal Parameters Degree of severity Strained and strangled voice quality Oscillation of intensity Voice breaks
Vocal tremor
Roughness
Breathiness
Asthenia
Value
95% CI
Time
CCC
LL
UL
Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2 Pre-application Return 1 Return 2
0.97 0.98 0.94 0.99 0.97 0.92 0.99 0.99 0.99 0.99 0.99 0.99 0.98 0.99 0.9 0.97 0.98 0.97 0.99 0.98 0.98 0.98 0.94 0.71
0.91 0.95 0.83 0.97 0.92 0.79 0.96 0.96 0.96 0.98 0.98 0.96 0.94 0.97 0.74 0.92 0.96 0.9 0.97 0.94 0.94 0.94 0.85 0.33
0.99 0.99 0.98 1 0.99 0.97 0.99 1 1 1 1 1 0.99 1 0.97 0.99 0.99 0.99 1 0.99 0.99 0.99 0.98 0.89
Abbreviations: CCC, concordance correlation coefficient; CI, confidence interval; LL, lower limit; UL, upper limit.
assessments is related to the time and quality of training used to teach the raters to recognize the vocal parameters of individuals with AdSD based on the equalization of the vocal parameters to be evaluated between raters. It is also important to establish a reference point for the degree of severity.6 All of these aspects were satisfied in the type of analysis proposed for the present study, with satisfactory agreement thus being obtained between judges.
This aspect was also considered by Eadie et al,18 who investigated the difference in the assessment of overall severity of dysphonia and vocal effort in 20 individuals with AdSD who had completed a treatment cycle. Three groups were selected for judgment: individuals with AdSD, subjects with no experience, and speech-language pathologist with experience. In the cited study, there was no difference among the three groups, with a strong correlation being observed in the judgment of the vocal parameters, although the strategies used differed from those used by the group with experience. Although there were no differences between these raters, it should be pointed out that few studies have reported their experiences about evaluation. On this basis, it is important to reflect about perceptual-auditory assessment in the analysis of AdSD voices as there is wide variability of vocal changes between patients and between treatment times, suggesting that greater training is necessary for raters due to the marked difficulty in assessing the vocal signal of these patients. When the voice parameters were compared in the perceptualauditory analysis before and during BTX treatment regarding the general degree of severity of vocal quality, no changes were detected, although a discrete tendency to a reduction of mean values was observed between pre-application and R1. These data differ from those reported in other studies in which the degree of severity of vocal quality showed changes in the comparison of before and after BTX treatments.20,26,27 However, this difference may have been due to the form of quantitation of the degree of severity, which in the present study showed the highest value among the remaining seven vocal parameters. VT was the parameter showing the greatest change before BTX application, and roughness was the parameter showing the greatest change at R1. Strained-strangled voice quality did not change in the present sample, although the absolute mean values showed a discrete reduction. A reduction of this voice signal was observed in other studies after BTX treatment.20,26,27 A change in the comparison between preapplication and R1 was observed in voice breaks and oscillation of intensity based on the analysis of the narrow-band spectrography, intensity, and pitch contour. The reduced voice breaks suggest a reduction of
TABLE 3. Vocal Parameters Values Before and After Treatment With Botulinum Toxin Preapplication Vocal Parameters Degree of severity Strained-strangled voice quality Oscillation of intensity Voicebreaks Vocal tremor Roughness Breathiness Asthenia Abbreviation: SD, standard deviation.
First return (R1)
Second return (R2)
Range
Mean ± SD
Range
Mean ± SD
Range
Mean ± SD
0.5–8.50 0.00–8.30 0.00–8.40 0.00–8.20 0.00–8.50 0.5–7.10 0.00–6.70 0.00–7.50
5.5 ± 2.18 3.09 ± 3.08 4.48 ± 2.86 4.71 ± 2.80 5.08 ± 2.46 3.93 ± 2.02 2.37 ± 1.95 1.65 ± 2.41
0.4–8.70 0.00–7.50 0.00–6.30 0.00–6.80 0.00–6.30 0.40–8.70 0.00–7.10 0.00–5.60
4.53 ± 2.24 2.17 ± 2.47 2.36 ± 2.23 2.51 ± 2.34 2.64 ± 2.26 3.98 ± 2.29 3.07 ± 2.32 1.97 ± 1.91
2.9–5.60 0.00–8.90 0.00–9.30 0.00–9.00 0.3–8.90 0.30–6.90 0.00–5.80 0.00–4.20
5.83 ± 1.72 2.81 ± 3.07 4.14 ± 2.92 4.24 ± 2.96 4.37 ± 2.62 3.92 ± 1.71 2.63 ± 1.87 1.76 ± 1.95
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
7
Vocal Parameters and Self-Perception on AdSD
TABLE 4. Comparison of Vocal Signals Obtained at the Various Times of Assessment: Preapplication (PRE), First Return (R1), and Second Return (R2)
Time of BTX Treatment Degree of severity of voice quality PRE × R1 PRE × R2 R1 × R2 Strained-strangled voice quality PRE × R1 PRE × R2 R1 × R2 Oscillation of intensity PRE × R1 PRE × R2 R1 × R2 Voice breaks PRE × R1 PRE × R2 R1 × R2 Vocal tremor PRE × R1 PRE × R2 R1 × R2 Roughness PRE × R1 PRE × R2 R1 × R2 Breathiness PRE × R1 PRE × R2 R1 × R2 Asthenia PRE × R1 PRE × R2 R1 × R2
95% CI
Estimated Difference Between Means
P Value
LL
UL
0.95 −0.249 1.199
0.118 0.682 0.056
−0.257 −1.482 −0.034
2.157 0.984 2.432
0.88 0.171 0.708
0.188 0.799 0.297
−0.455 −1.195 −0.658
2.214 1.537 2.074
1.962 0.361 1.601
0.002* 0.54 0.011*
0.795 −0.834 0.406
3.129 1.556 2.796
2.034 0.56 1.473
0.002* 0.361 0.021*
0.856 −0.677 0.236
3.211 1.798 2.711
2.335 0.847 1.489
0.002* 0.231 0.040*
0.981 −0.571 0.071
3.69 2.264 2.906
−0.093 0.019 −0.112
0.887 0.977 0.866
−1.419 −1.332 −1.464
1.233 1.371 1.24
−0.829 −0.187 −0.643
0.23 0.784 0.35
−2.219 −1.576 −2.032
0.56 1.203 0.747
−0.322 −0.035 −0.287
0.639 0.96 0.688
−1.717 −1.487 −1.739
1.072 1.416 1.165
Abbreviations: LL, lower limit; UL, upper limit. * Indicates the time in which the comparison of the vocal parameters was performed.
adduction spasms of the vocal folds, this being the major vocal parameter presented by individuals with AdSD, affecting speech intelligibility. The results obtained here are similar to those reported by others even though the present study was based only on perceptual-auditory analysis.17,18,27,28 The oscillation of intensity parameter was used, taking into consideration the tool for the visualization of intensity contour offered by the Praat software. Although this parameter is not easy to assess, this analysis was based simultaneously on perceptive-auditory analysis. Thus, the oscillation of intensity parameter was anchored on voice analysis, representing a parameter of the change between preapplication and the first return. These differences can be observed in Figure 1. There were oscillations of intensity and voice breaks also between R1 and R2, although these values increased, returning to the levels observed before BTX application. This is related
to the time of effectiveness of the treatment reported in various studies, which ranges from 3 to 4 months.27,29,30 Of the 16 individuals studied here, 81.25% showed associated VT, which was the most altered parameter in perceptualauditory analysis, with the highest values being obtained using the visual analogue scale. After BTX application, these values were reduced at R1, increasing again when the effect of BTX ended at the R2 assessment, contradicting the results reported by White et al,30 who demonstrated that the subjects with VT did not show improvement of this parameter after the use of BTX. The parameters roughness, breathiness, and asthenia did not change, suggesting that, although BTX treatment is beneficial, it also triggers a series of other vocal parameters considered to be iatrogenic effects, such as paresis of the thyroarytenoid muscle,17 derived from the inhibitory action on muscle contraction resulting from the blockade of acetylcholine at the junction,
ARTICLE IN PRESS 8
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 5. Voice Handicap Index Before and After the Application of Botulinum Toxin Preapplication Mean ± SD
R1 Mean ± SD
Total score 67.79 ± 26.52
58.21± 27.30
Emotional domain 20.43 ± 12.30
R2 Mean ± SD
P Value
60 ± 28.5
PRE × R1 = 0.039* PRE × R2 = 0.108 R1 × R2 = 0.785
16.36 ± 13.0
15.83 ± 14.0
PRE × R1 = 0.035* PRE × R2 = 0.04* R1 × R2 = 0.60
Physical domain 23.36 ± 7.33
20.43 ± 9.12
22.17 ± 22.50
PRE × R1 = 0.14 PRE × R2 = 0.24 R1 × R2 = 0.57
Functional domain 24 ± 10.2
21.43 ± 11.14
19.50 ± 11.38
PRE × R1 = 0.12 PRE × R2 = 0.03* R1 × R2 = 0.36
Abbreviations: PRE, preapplication; R1, first return; R2, second return; SD, standard deviation. * Indicates the time in which the comparison of the vocal parameters was performed.
to be resolved. The response to the instrument permits the establishment of future projections regarding the dose and duration of BTX treatment, the emotional adaptation to the problems faced, and the social impact. The evaluation of the 16 subjects studied here revealed that the total VHI scores were reduced in eight of them when the pretreatment values were compared with those obtained after 30 days of BTX treatment, with the mean general VHI score changing from 67.79 to 58.21 at R1, in agreement with data reported by Morzaria and Damrose,31 who obtained a total score that varied
a mechanism that not only helps reduce the adductor spasms but also causes changes such as breathiness,12,17,20 worsening of the roughness signal with a predominance of hoarseness,18 and reduction of intensity or asthenia26 as a result of BTX action. The assessment of the vocal impact and handicap of voice disorders using questionnaires is a useful tool for learning about the implications of the disease in the life of affected individuals. In this respect, the response to VHI is helpful in the determination of self-perception of both the impact of dysphonia and the effect of treatment, also identifying complications
TABLE 6. Voice Handicap Index: Individual Score Before and After the Application of Botulinum Toxin Total Score
Functional Domain
Emotional Domain
Physical Domain
Individual
PRE
R1
R2
PRE
R1
R2
PRE
R1
R2
PRE
R1
R2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
– 94 66 53 67 80 82 72 41 101 56 16 107 – 30 84
– 101 31 32 56 90 49 47 41 88 66 16 99 – 36 63
– 71 55 40 59 82 – 65 18 105 34 16 110 – 35 –
– 34 24 15 23 29 35 28 11 34 20 6 39 – 11 27
– 35 7 8 23 29 25 24 11 39 24 6 36 – 12 27
– 24 16 9 24 25 – 26 2 36 12 8 39 – 13 –
– 28 18 12 18 30 31 18 13 39 4 2 39 – 6 28
– 30 10 8 9 34 14 8 13 37 4 0 40 – 8 14
– 16 17 9 9 34 – 15 4 39 0 2 39 – 6 –
– 32 24 26 26 21 16 26 17 28 32 8 29 – 13 29
– 6 14 16 24 27 10 15 17 12 38 10 23 – 16 28
– 31 22 22 26 23 – 24 12 30 22 6 32 – 16 –
Abbreviations: PRE, preapplication; R1, first return; R2, second return.
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
9
Vocal Parameters and Self-Perception on AdSD
TABLE 7. Distribution of the Values Obtained for Each Question of Each Domain (Emotional, Organic, and Functional) of the Voice Handicap Index PRE Question
Median
Emotional domain 7 3.0
R1
R2
Mean ± SD
Median
Mean ± SD
Median
Mean ± SD
2.5 ± 1.6
2.0
2.1 ± 1.6
2.0
1.9 ± 1.6
9
1.0
1.2 ± 1.3
1.0
1.5 ± 1.6
1.0
1.2 ± 1.3
15
2.0
1.8 ± 1.4
2.0
1.7 ± 1.3
1.5
1.4 ± 1.5
23
4.0
2.8 ± 1.7
3.0
2.2 ± 1.6
2.5
2.5 ± 1.5
24
4.0
2.7 ± 1.5
2.0
2.0 ± 1.6
1.0
1.7 ± 1.8
25
2.5
2.1 ± 1.8
1.5
1.4 ± 1.5
2.0
1.8 ± 1.8
27
2.0
2.0 ± 1.3
0.5
1.4 ± 1.6
1.0
1.5 ± 1.7
28
1.5
1.7 ± 1.6
0
1.3 ± 1.7
1.0
1.5 ± 1.6
29
0
1.4 ± 1.8
0
0.9 ± 1.6
0
1.0 ± 1.6
30
2.0
1.7 ± 1.7
0
1.4 ± 1.7
0
1.0 ± 1.8
Physical domain 2 1.0
1.4 ± 1.6
2.0
1.1 ± 1.0
2.0
1.9 ± 1.4
4
2.0
2.2 ± 1.2
2.0
2.1 ± 1.5
2.0
1.7 ± 1.6
10
3.0
2.9 ± 1.2
4.0
3.1 ± 1.2
3.0
2.6 ± 1.4
13
2.5
2.4 ± 1.5
4.0
3.0 ± 1.5
2.5
2.5 ± 1.4
14
3.5
2.8 ± 1.4
2.0
1.7 ± 1.4
2.0
2.3 ± 1.4
17
3.0
2.3 ± 1.7
3.0
2.6 ± 1.4
2.5
2.2 ± 1.5
18
2.0
2.2 ± 1.3
2.0
2.2 ± 1.7
2.5
2.4 ± 1.6
P PRE × R1 = 0.290 PRE × R2 = 0.111 R1 × R2 = 0.809 PRE × R1 = 0.486 PRE × R2 = 1.000 R1 × R2 = 0.795 PRE × R1 = 0.752 PRE × R2 = 0.206 R1 × R2 = 0.256 PRE × R1 = 0.241 PRE × R2 = 0.820 R1 × R2 = 0.638 PRE × R1 = 0.127 PRE × R2 = 0.096 R1 × R2 = 0.139 PRE × R1 = 0.127 PRE × R2 = 0.870 R1 × R2 = 0.417 PRE × R1 = 0.179 PRE × R2 = 0.417 R1 × R2 = 0.754 PRE × R1 = 0.054 PRE × R2 = 0.723 R1 × R2 = 0.429 PRE × R1 = 0.169 PRE × R2 = 0.339 R1 × R2 = 1.000 PRE × R1 = 0.218 PRE × R2 = 0.082 R1 × R2 = 0.175 PRE × R1 = 0.302 PRE × R2 = 0.376 R1 × R2 = 0.056 PRE × R1 = 0.890 PRE × R2 = 0.206 R1 × R2 = 0.328 PRE × R1 = 0.336 PRE × R2 = 0.214 R1 × R2 = 0.087 PRE × R1 = 0.273 PRE × R2 = 0.748 R1 × R2 = 0.207 PRE × R1 = 0.019* PRE × R2 = 0.394 R1 × R2 = 0.194 PRE × R1 = 0.486 PRE × R2 = 0.889 R1 × R2 = 0.714 PRE × R1 = 1.000 PRE × R2 = 0.220 R1 × R2 = 0.429 (continued on next page)
ARTICLE IN PRESS 10
Journal of Voice, Vol. ■■, No. ■■, 2016
TABLE 7. (continued ) PRE
R1
R2
Median
Mean ± SD
Median
Mean ± SD
Median
Mean ± SD
P
20
3.0
2.2 ± 1.5
0.5
1.2 ± 1.7
2.0
2.0 ± 1.7
21
2.0
2.0 ± 1.6
2.0
1.5 ± 1.5
1.0
1.25 ± 1.4
26
3.0
2.5 ± 1.6
2.0
1.4 ± 1.4
3.0
0.5 ± 1.2
PRE × R1 = 0.047* PRE × R2 = 0.220 R1 × R2 = 0.107 PRE × R1 = 0.151 PRE × R2 = 0.060 R1 × R2 = 0.504 PRE × R1 = 0.82 PRE × R2 = 0.95 R1 × R2 = 0.023*
Functional domain 1 2.5
2.7 ± 1.05
2.0
2.1 ± 1.4
2.0
2.0 ± 1.5
Question
3
3.5
3.2 ± 1.6
3.0
3.0 ± 1.0
2.5
2.2 ± 1.3
5
2.5
2.1 ± 1.6
2.5
2.4 ± 1.3
2.0
2.5 ± 1.4
6
4.0
3.0 ± 1.5
3.5
2.7 ± 1.6
2.5
2.5 ± 1.5
8
3.0
2.3 ± 1.7
1.5
1.7 ± 1.8
2.5
1.9 ± 1.7
11
3.0
2.6 ± 1.5
3.0
2.5 ± 1.6
2.5
2.0 ± 1.5
12
3.0
2.9 ± 1.2
2.0
2.2 ± 1.1
2.5
2.5 ± 1.3
16
1.5
1.7 ± 1.6
1.0
1.5 ± 1.6
1.0
1.5 ± 1.7
19
1.5
1.7 ± 1.6
0
1.2 ± 1.7
1.0
1.5 ± 1.6
22
1.0
1.4 ± 1.6
1.0
1.7 ± 1.9
0
0.5 ± 1.2
PRE × R1 = 0.082 PRE × R2 = 0.095 R1 × R2 = 0.870 PRE × R1 = 0.336 PRE × R2 = 0.026* R1 × R2 = 0.136 PRE × R1 = 0.890 PRE × R2 = 0.206 R1 × R2 = 0.328 PRE × R1 = 0.500 PRE × R2 = 0.504 R1 × R2 = 0.896 PRE × R1 = 0.026* PRE × R2 = 0.082 R1 × R2 = 0.191 PRE × R1 = 0.807 PRE × R2 = 0.147 R1 × R2 = 0.230 PRE × R1 = 0.075 PRE × R2 = 0.389 R1 × R2 = 0.504 PRE × R1 = 0.699 PRE × R2 = 0.777 R1 × R2 = 0.339 PRE × R1 = 0.205 PRE × R2 = 0.820 R1 × R2 = 0.795 PRE × R1 = 0.418 PRE × R2 = 0.111 R1 × R2 = 0.085
Abbreviations: PRE, preapplication; R1, first return; R2, second return; SD, standard deviation. * Indicates the time in which the comparison of the vocal parameters was performed.
from 69.9 to 51.1 after application of the toxin. Although there was a reduction in the scores, these values were above the cutoff note of 19 points proposed for this instrument,32 suggesting that self-perception reveals a greater recognition of the impact of the disease on the quality of life of patients and certifies their involvement with respect to treatment. However, it should be pointed out that there is divergence in the response to treatment, perhaps due to the method of toxin application, the technique used, whether the toxin is applied unilaterally or bilaterally, the quantity of the medication used, the
time of assessment after BTX injection, or factors inherent to the individual and to the way of filling out the questionnaire, with consequent changes in the results obtained. The present study disagreed with other investigations that reported lower scores for the effect of BTX application determined by VHI compared with the present data.12,31 However, it is recognized that treatment with BTX is of benefit for both the phonatory function and the quality of life of these individuals.12,17 Detailed descriptions of the questions of each domain permit us to understand the main aspects that can affect the personal
ARTICLE IN PRESS Gleidy Vannesa E. Rojas, et al
Vocal Parameters and Self-Perception on AdSD
and social life of persons with AdSD regarding oral communication and the way BTX treatment can change or reduce this disadvantage. Each domain has a specific objective. The emotional domain involves the feelings of the person regarding his or her phonation and communication with people as the voice provokes and reveals reactions from the listener. In this domain, most of the present patients reported a marked impact on items 7, 23, and 24 of the VHI, showing that subjects with spasmodic dysphonia feel frustrated and discouraged,33 with difficulties in their interpersonal relations even though they try to accept the fact that they have a voice disorder that should not be the reason for a low self-esteem. This was observed in items 29 and 30, which did not change with treatment, showing a frequency of “sometimes” and “never.” The application of BTX according to the established protocol was unable to modify the emotional perceptions of individuals regarding their voice. This emotional domain should be considered for future proposed treatments. The objective of the functional domain is to determine how AdSD changes daily life activities and social interaction. Regarding the work aspect, the patients reported that their voice problem did not cause economic losses (item 22), a result that requires further study as many of the patients are already retired. Other jobs held by the patients involve tasks that do not depend fully on speech. This impairs their quantitation, with the need for further investigation of this question involving a socioeconomic study.17,34 Regarding communication, item 8, which refers to avoiding groups of persons because of one’s voice, revealed that the subjects improve, progressing from an “almost always” frequency to an “almost never” frequency, as there was no change in the frequency of facts that are part of daily life, as shown in items 3, 6, and 11. The physical domain evaluates the auditory and kinesthetic experiences of individuals regarding their voice. There was an evident presence of hoarseness (item 13) before treatment, which worsened after treatment, a result related to the iatrogenic effect of the temporary muscle paresis that occurs after BTX injection,17 and also related to the effort to speak (items 14 and 20) and to voice breaks in the middle of speech (item 26), which were reduced after treatment and were associated with the symptoms reported by patients with AdSD.33–35 Regarding item 18, which explores whether the patients try to change their voice, the predominant result was “at times” and did not change after BTX application. This reply expresses the coping strategies for a positive or negative situation and the reduced occupational and social participation of the persons.33 Before BTX application, the correlation between perceptualauditory evaluation and total VHI score in terms of degree of severity was moderate. Perceptual-auditory analysis revealed a higher degree for VT, voice breaks, oscillation of intensity, and strained-strangled voice quality, suggesting a relationship with the higher score for the physical domain of the VHI. There was a complaint regarding the effort to speak (items 14 and 20) in agreement with the presence of strained-strangled voice, and the perception of voice breaks (item 26), explained by the oscillation of intensity which alters speech intelligibili-
11
ty and consequently affects the communication of persons with AdSD. At R1, the correlation was weak, with contradictory results being observed as, despite the clear presence of positive changes in the vocal quality of individuals and the improvement observed by perceptual-auditory evaluation, the impact and the vocal disadvantage of AdSD along time cause changes in the emotional and functional aspects of the individuals, affecting their relationship with others and with their environment. Another aspect to be considered is that the iatrogenic effects of BTX application, such as hoarseness and breathiness, were relevant for the patients. Thus, after application of the botulin toxin, patients continued to be dysphonic, although the vocal changes observed did not limit the use of the voice and communication, with a reduction of the vocal impact. However, the changes due to the application of the toxin were still important for the patients studied here. In addition, it is important to point out that, in contrast to other vocal diseases such as vocal nodules and polyps, among others, AdSD is a chronic disease that can be treated but not cured, with the patients ultimately adapting to their lifestyle in the presence of the disease.36 This fact was confirmed in the present study, in which the functional domain did not show significant changes after BTX application. A moderate correlation was observed at R2, related to the duration of the effect of BTX treatment and to the onset of the vocal signs present in AdSD. In view of the above data, the effects of 30 days of BTX treatment were clearly observed, with a return to the vocal parameters observed before treatment at 120 days after application of the toxin.
CONCLUSIONS The present results suggest that perceptual-auditory analysis using narrow-band spectrography and pitch and intensity contour can be considered as an alternative for the identification and quantitation of vocal parameters in individuals with AdSD in order to assess the results of treatment in clinical practice. The parameters of oscillation of intensity, voice breaks, and VT were those that most benefited from the application of BTX, although the benefits did not persist for 120 days. BTX treatment had a positive effect on the self-perception of the subjects measured with the VHI, with a reduction of the vocal symptoms of patients with AdSD and an improvement of voice handicap. However, considering the correlation between the general degree of voice quality and the total VHI score, it is necessary to conduct further studies that will establish a relationship among the time of diagnosis, duration of treatment, and the acquisition of auditory and tactile-kinesthetic perception skills that will allow AdSD patients to identify the vocal signals present before and after BTX treatment. Studies with a larger number of participants using chained speech samples and recording of the results 60 and 90 days after BTX injection are needed, contributing even more to the guidance of treatment of subjects with AdSD.
ARTICLE IN PRESS 12 Acknowledgment We wish to acknowledge Diego Espitia Rojas for the elaboration of the reply form for perceptual-auditory assessment using Microsoft Excel software for the quantitation of vocal parameters. REFERENCES 1. Ludlow C. Spasmodic dysphonia: a laryngeal control disorder specific to speech. J Neurosci. 2011;31:793–797. 2. Meyer TK. The treatment of laryngeal dystonia (spasmodic dysphonia) with botulinum toxin injections. Oper Tech Otolaryngol. 2012;23:96– 101. 3. Roy N. Differential diagnosis of muscle tension dysphonia and spasmodic dysphonia. Curr Opin Otolaryngol Head Neck Surg. 2010;18:165–170. 4. Blitzer A. Spasmodic dysphonia and botulinum toxin: experience from the largest treatment series. Eur J Neurol. 2010;17(suppl 1):28–30. 5. Blitzer A, Brin MF, Stewart CF. Botulinum toxin management of spasmodic dysphonia (laryngeal dystonia): a 12 year experience in more than 900 patients. Laryngoscope. 1998;108:1435–1441. 6. Barkmeir J, Case J, Ludlow CL. Identification of symptoms for spasmodic dysphonia and vocal tremor: comparison of expert and nonexpert judges. J Commun Disord. 2001;34:21–37. 7. Watts C, Nye C, Whurr R. Botulinum toxin for treating spasmodic dysphonia (laryngeal dystonia): a systematic Cochrane review. Clin Rehabil. 2006;20:112–122. 8. Bender BK, Cannito MP, Murry T, et al. Speech intelligibility in severe adductor spasmodic dysphonia. J Speech Lang Hear Res. 2004;47:21–32. 9. Branden M, Johns M, Klein A, et al. Assessing the effectiveness of botulin toxin injections for adductor spasmodic dysphonia: clinician and patient perception. J Voice. 2008;24:242–248. 10. Baylor C, Yorkston K, Tanya E. The consequences of spasmodic dysphonia on communication-related quality of life: a qualitative study of the insider’s experiences. J Commun Disord. 2005;38:395–419. 11. Bielamowicz S, Ludlow CL. Effects of botulinum toxin on the pathophysiology in spasmodic dysphonia. Ann Otol Rhinol Laryngol. 2000;109:194–203. 12. Novakovic D, Waters H, D’Elia J, et al. Botulinum toxin treatment of adductor spasmodic dysphonia: longitudinal functional outcomes. Laryngoscope. 2011;121:606–612. 13. Kendall KA, Leonard RJ. Interarytenoid muscle botox injection for treatment of adductor spasmodic dysphonia with vocal tremor. J Voice. 2011;25:114– 119. 14. Ludlow CL, Naunton RF, Sedory SE, et al. Effects of botulinum toxin injections on speech in adductor spasmodic dysphonia. Neurology. 1988;38:1220–1225. 15. Blitzer A. Botulinum toxin A and B: a comparative dosing study for spasmodic dysphonia. Otolaryngol Head Neck Surg. 2005;133:836– 838. 16. Silverman EP, Garvan C, Shrivastav R, et al. Combined modality treatment of adductor spasmodic dysphonia. J Voice. 2012;26:77–86. 17. Denjockere PH, Neumann KJ, Moerman MB, et al. Tridimensional assessment of adductor spasmodic dysphonia pre- and post-treatment with botulinum toxin. Eur Arch Otorhinolaryngol. 2012;4:1195–1203.
Journal of Voice, Vol. ■■, No. ■■, 2016 18. Eadie TL, Nicolici C, Baylor C, et al. Effect of experience on judgments of adductor spasmodic dysphonia. Ann Otol Rhinol Laryngol. 2007;116:695– 701. 19. Murry T. Spasmodic dysphonia: let’s look at that again. J Voice. 2014;28:694–699. 20. Cannito MP, Woodson GE, Murry T, et al. Perceptual analyses of spasmodic dysphonia before and after treatment. Arch Otolaryngol Head Neck Surg. 2004;130:1393–1399. 21. Liu TC, Irish JC, Adams SG, et al. Prospective study of patients’ subjective responses to botulinum toxin injection for spasmodic dysphonia. J Otolaryngol. 1996;25:66–74. 22. Jacobson BH, Johnson A, Grywalski C, et al. The Voice Handicap Index (VHI): development and validation. Am J Speech Pathol. 1997;6:66–70. 23. Behlau M, Alves dos Santos LM, Oliveira G. Cross-cultural adaptation and validation of the voice handicap index into Brazilian Portuguese. J Voice. 2011;25:354–359. 24. Boersma P, Weenink D. Praat: doing phonetics by computer. Institute of Phonetic Sciences, University of Amsterdam, Amsterdam, 1995. Available at: http://www.fon.hum.uva.nl/praat/. Accessed July 4, 2012. 25. Kreiman J, Gerratt BR, Kempster GB, et al. Perceptual evaluation of voice quality: review, tutorial, and a framework for future research. J Speech Hear Res. 1993;36:21–40. 26. Langeveld TP, Luteijn F, van Rossum M, et al. Adductor spasmodic dysphonia and botulinum toxin treatment: the effect on well-being. Ann Otol Rhinol Laryngol. 2001;110:941–945. 27. Tisch SH, Brake HM, Law M, et al. Spasmodic dysphonia: clinical features and effects of botulinum toxin therapy in 169 patients-an Australian experience. J Clin Neurosci. 2003;10:434–438. 28. Boutsen F, Cannito MP, Taylor M, et al. Botox treatment in adductor spasmodic dysphonia: a meta-analysis. J Speech Lang Hear Res. 2002;45: 469–481. 29. Cloud LJ, Jinnah HA. Treatment strategies for dystonia. Expert Opin Pharmacother. 2010;11:5–15. 30. White LJ, Klein AM, Hapner ER, et al. Co-prevalence of tremor with spasmodic dysphonia: a case-control study. Laryngoscope. 2011;121:1752– 1755. 31. Morzaria S, Damrose E. A comparison of the VHI, VHI-10, and V-RQOL for measuring the effect of botox therapy in adductor spasmodic dysphonia. J Voice. 2012;26:378–380. 32. Behlau M, Madazio G, Moreti F, et al. Efficiency and cutoff values of self-assessment instruments on the impact of a voice problem. J Voice. 2016;30:506.e9–506.e18. 33. Smith E, Taylor M, Mendoza M, et al. Spasmodic dysphonia and vocal fold paralysis: outcomes of voice problems on work-related functioning. J Voice. 1998;12:223–232. 34. Baylor C, Yorkstonb M, Eadi T. The consequences of spasmodic dysphonia on communication-related quality of life: a qualitative study of the insider’s experiences. J Commun Disord. 2005;28:395–419. 35. Benninger M, Gardner G, Grywalski C. Outcomes of botulinum toxin treatment for patients with spasmodic dysphonia. Arch Otolaryngol. 2001;127:1083–1085. 36. Hu A, Hillel A, Meyer T. Factors associated with patient-perceived hoarseness in spasmodic dysphonia patients. J Voice. 2015;doi:10.1016/ j.jvoice.2015.11.003. In press, 1–4.