A quantitative method for the assessment of

Journal of the Neurological Sciences 377 (2017) 42–46

Contents lists available at ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

A quantitative method for the assessment of dysarthrophonia in myasthenia gravis Kostas Konstantopoulos a, Yiolanda-Panayiota Christou b, Paris Vogazianos a, Eleni Zamba-Papanicolaou b, Kleopas A. Kleopa b,⁎ a b

European University Cyprus, Nicosia, Cyprus Neurology Clinics, The Cyprus Institute of Neurology and Genetics, Nicosia, Cyprus

a r t i c l e

i n f o

Article history: Received 21 November 2016 Received in revised form 27 March 2017 Accepted 28 March 2017 Available online 31 March 2017 Keywords: Myasthenia gravis Speech acoustics PRAAT Phonation Voice Electroglottography

a b s t r a c t Speech and voice symptomatology (dysarthrophonia) are often reported by patients with myasthenia gravis (MG). However, they have been poorly investigated despite their significant impact on quality of life. Quantitative methods for the assessment of dysarthrοphonia could facilitate the evaluation of these common MG symptoms. The goal of this study was to investigate the phonatory (sustained phonation and reading) and speech (diadochokinesis) function in MG patients using quantitative measures. The voice/speech of 12 MG patients (7 with anti-AchR and 5 with anti-MuSK antibodies) and 24 age-matched healthy controls was recorded and analyzed using electroglottography (EGG) and speech acoustics. For the analysis of voice, the variables that were found to distinguish MG patients compared to healthy controls were a higher average fundamental frequency (P b 0.05), a higher standard deviation of the average fundamental frequency (P b 0.001), a higher mean fundamental frequency of the vibrating vocal folds (P b 0.005) and a higher fundamental frequency range (P b 0.005). The analysis of diadochokinesis showed that MG patients had a higher mean duration of the silent interval between a series of repetitive /pa/ syllables (P b 0.05), of the sound /t/ (P = 0.05) and of the silent interval between a series of repetitive /ka/syllables (P b 0.05). No statistical differences were found in any of these variables between the MG subgroups with anti-AchR or anti-MuSK antibodies. This study demonstrates that non-invasive physiological methods (EGG and speech acoustics) offer essential tools for the assessment of dysarthrophonia in MG patients. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Symptoms of dysarthrophonia resulting from bulbar muscle weakness are common manifestations of myasthenia gravis (MG). They may appear at any stage of the disease, often in association with dysphagia and have been reported as initial disease symptoms even before the neurological diagnosis in up to 15% of MG patients [1–3]. Dysarthrophonia may be more prominent in patients with antibodies to muscle specific kinase (MuSK), 40% of whom exhibit bulbar involvement at disease onset, including weakness in the face, pharynx, and tongue [4]. During the course of MG, the disease progresses to involve the bulbar muscles in two-thirds of all patients including patients with antibodies to acetylcholine receptor (AchR). Perceptually, the dysarthrophonia of MG is predominantly of the flaccid type including ⁎ Corresponding author at: Neurology Clinics, The Cyprus Institute of Neurology and Genetics, 6 International Airport Avenue, P.O. Box 23462, 1683 Nicosia, Cyprus. E-mail addresses: [email protected] (K. Konstantopoulos), [email protected] (Y.-P. Christou), [email protected] (P. Vogazianos), [email protected] (E. Zamba-Papanicolaou), [email protected] (K.A. Kleopa).

http://dx.doi.org/10.1016/j.jns.2017.03.045 0022-510X/© 2017 Elsevier B.V. All rights reserved.

breathiness or harshness of voice, hypernasality, imprecision of consonants, monopitch, monoloudness, and short phrases [5–11]. Research on dysarthrophonia is increasing in neurological diseases such as stroke [12–16], multiple sclerosis [17–19], Parkinson's disease [20], and amyotrophic lateral sclerosis (ALS) [21]. The assessment of dysarthrophonia is based mostly on perceptual methods that are prone to inter-examiner differences. In contrast to perceptual methods, objective methods including speech acoustics and objective voice analysis based on signals produced by speech motor control offer the advantage of quantification in the analysis of dysarthric speech [7]. These methods have not been tested in MG as part of oropharyngeal muscle assessment although they could facilitate the diagnostic evaluation and monitoring of the disease [22]. Specific acoustic measures of vocalic and consonantal segment durations have been proposed to distinguish dysarthric from normal speech [23]. The acoustic method involves analysis of the speech signal during reading of a standard passage and/or during diadochokinesis (DDK) (maximum repetition of the syllables/pa/ /ta/ /ka/), testing that is analogous to strength, range, or speed tests in clinical neurology [24]. Acoustic analysis of diadochokinetic rate has been found to distinguish

K. Konstantopoulos et al. / Journal of the Neurological Sciences 377 (2017) 42–46

Huntington's disease mutation carriers at pre-clinical stages [25], as well as Parkinsonian patients [26], when compared to healthy controls. In flaccid dysarthria, typical acoustic findings include slow lingual or bilabial diadochokinesis DDK (speech) and increased jitter, increased mean fundamental frequency and reduced pitch range (voice) [7]. The assessment of speech and phonatory symptomatology using objective speech/voice analysis could help to quantify these common MG manifestations avoiding the inter-examiner differences that are inherent in the perceptual methods of analysis. The aim of this pilot study was to identify distinct differences in phonation and speech using electroglottography (EGG), a physiological method for voice analysis, and speech acoustic methods in MG patients compared to healthy controls. 2. Methods 2.1. Study design This was a case control pilot study including a group of 12 MG patients (7 with anti-AchR antibodies and 5 with anti-MuSK antibodies) exhibiting dysarthrophonia and 24 healthy control participants. The study complied with the Helsinki Declaration and the study protocol was approved by the Cyprus National Bioethics Committee (EEBK/EΠ/ 2014/02) according to National and European Law. All participants signed an informed consent form before entering the study.

43

The MG group in this study consisted of 3 males and 9 females and was compared to 24 matched healthy controls for age (P N 0.05), gender and education (P N 0.05), mostly selected from spouses of patients and clinic personnel. No differences in MMSE score were found between the two groups (P N 0.05). MGFA classification scores at worse disease stage were: V (n = 1, 8.33%), IV (n = 6, 50%), III (n = 4, 33.33%) and II (n = 1, 8.33%); whereas MGFA classification scores at last clinic visit and at the time of the study were IV (n = 1, 8.33%), III (n = 1, 8.33%), II (n = 7, 58.33%) and I (n = 3, 25%). Most patients (n = 10, 83%) were on regular treatment with pyridostigmine, among other treatments for MG. Table 1 shows age, education, MMSE score, disease duration and QMG score of the participants in the clinical and control groups. Following neurological assessment, a history form specialized for speech/voice and the Frenchay Dysarthria Assessment (FDA) [30] was used by the speech therapist to measure the severity of dysarthria among participants. In the history form, all patients reported speech and/or voice abnormalities. Likewise in the FDA all participants in the MG group exhibited mild to moderate dysarthrophonia. The voice handicap index (VHI) was also administered to all participants and was used to measure the impact of voice problems on patient's quality of life [31]. Finally, the recording of voice/speech took place. The same speech therapist specialized in the quantitative analysis of speech/ voice administered the testing in all participants.

2.3. Recording procedure 2.2. Participant selection and clinical dysarthrophonia assessment All MG participants with dysarthrophonia selected to enter the study were recruited from the neurology clinics at the Cyprus Institute of Neurology and Genetics during a period of one year (2014–2015). Before admission to the study, all patients were assessed clinically for the severity of MG by experienced neurologists in MG care using the quantitative MG score (QMG score) [27]. All participants in the patient group had a definite diagnosis of myasthenia gravis based on typical clinical presentation and symptoms, positive electrophysiological confirmation of postsynaptic neuromuscular junction dysfunction, and MG-related antibody status [28], according to the Myasthenia Gravis Foundation of America (MGFA) [22]. Finally, all MG patients had at least a 3-month period of clinical stability prior to the study without MG exacerbations, medication changes or other complications to avoid significant changes in disease status between the neurological and dysarthrophonia assessment. Exclusion criteria included the following: impaired cognition as measured by the Mini Mental State examination (MMSE) [29], the appearance of speech/language problems during childhood and visual/ hearing problems severe enough to interfere with reading/listening, as well as illiteracy; a vocal disorder and/or laryngeal pathology prior to the appearance of neurological symptoms; previous larynx microsurgery; recent episode of endotracheal intubation; primary or metastatic tumor of the larynx, lung or mediastinum; diagnosis of respiratory disease (acute or chronic). Finally, for linguistic consistency only speakers whose native language was Greek were included.

A combination of tasks (sustained phonation and reading of a standard text passage) was used to measure voice while diadochokinesis (DDK) was used to measure speech [7]. The order of recording involved sustained phonation, reading and the DDK. In sustained phonation, every participant was asked to sustain the vowel “a” as long as he/she could (maximum duration) while in reading, every participant read aloud a standard passage. Finally, in DDK, every participant produced in isolation the syllables /pa/, /ta/, /ka/as quickly as possible and as long as she/he could (maximum repetition rate). Sustained phonation is a standard method of voice assessment. The reading of the Greek text passage provides consistent subject matter for comparison and uniform samples [17,32]. The DDK shows an indication of strength, range and speed of the articulators (lips and tongue movement) [7]. The vocalic and consonant durations of the syllables/pa/, /ta/, and/ka/ (DDK) were used to analyze speech. In voice assessment, electroglottography (EGG) as an indirect and non-invasive imaging technique, measures changes in electrical resistance between electrodes placed over the thyroid cartilage. EGG waveforms measure the duration of the relative vocal fold contact patterns within the glottal cycle and they are produced when the contact of the vocal folds increases as electrical impedance decreases [33]. In vocal pathology EGG has been used in combination with stroboscopy and high speed imaging to analyze the pathological voice and alone to describe irregularities in lower vocal fold vibration in patients diagnosed with vocal fold nodules, vocal fold cysts and glottal cancers.

Table 1 Biographical and neurological information in MG participants and healthy pair-matched controls.

Age (yrs) Education (yrs) Cognition (MMSE) Disease Duration (yrs) QMG score

Myasthenia gravis (n = 12)

Controls (n = 24)

Mann–Whitney U test

54.17 ± 18.55 10.97 ± 4.67 28.5 ± 1.31 AchR-Ab+ (n = 7) 9.0 ± 8.3 12.85 ± 4.3

54.53 ± 5.28 11.92 ± 4.45 28.58 ± 0.93

P N 0.05a P N 0.05a P N 0.05a

MMSE: mini mental status examination score. QMG: quantitative myasthenia gravis. a Comparing MG with controls. b Comparing AchR-Ab+ with MuSK Ab+.

MuSK-Ab + (n = 5) 5.8 ± 7.6 12.2 ± 10.2

P N 0.05b P N 0.05b

44


Using EGG, computerized statistics were used for period measurements in the vibrating vocal folds [34,35]. Numerical data can be extracted and measures of central tendency (mean, median, mode, and range) can be shown as a distribution of frequencies (DFx) derived from a period by period basis of the vibrating vocal folds (Fx = 1/Tx where Fx is the instantaneous frequency of vocal fold vibration and Tx is the time of vocal fold contact) [35]. In the present study we measured the central tendency of the fundamental frequency of the vibrating vocal folds (Fo). In sustained phonation, the average fundamental frequency (average Fx), the standard deviation of the fundamental frequency (SDFx) and the jitter (period by period variability as a measure of vocal stability) were used. In reading, the mean fundamental frequency (DFx1) (derived from the measurement of the time distance Tx, between successive glottal cycles) and the frequency range (90% range DFx1) (range as the frequency values above and below the average between which 90% of the observed frequencies fall) were used. In all tasks, the speech of each participant was digitally-recorded at habitual conversational pitch and loudness levels in a quiet examination room. The EGG recordings were obtained using a laptop SONY VAIO that was connected to an electroglottograph processor (PCLX) (Laryngograph Ltd., London, UK) and analyzed as described in Supplementary methods.

articulation (5/12, 42%) associated with fatigue during the day. Furthermore, half of the patients (6/12, 50%) reported difficulties with swallowing, while a small number of them (2/12, 17%) reported associated hypernasality. 3.2. Clinical assessment of dysarthrophonia The results of the FDA showed that most of the participants in the MG group (7/12, 58%) exhibited reduced time for sustained phonation, reduced control of loudness of voice, and breathiness. Regarding speech, FDA showed that the MG participants exhibited slowness in the elevation of tongue (6/12, 50%), in the lateral movement of tongue (5/12, 42%), in the protrusion of tongue (5/12, 42%), and in the alternate movement of tongue (6/12, 50%). The mean score of the VHI showed that MG participants perceived their dysphonia to affect their quality of life as compared to healthy controls. The self-reported symptoms in the history form and the qualitative analysis of the FDA showed that the subgroup with anti-AchR antibodies exhibited more frequently problems with phonation (time for sustained phonation and breathiness) while the participants with anti-MuSK antibodies exhibited more frequently dysphagia, drooling, articulation and hypernasality. However, the subgroup numbers were too small for further statistical analysis.

2.4. Voice and diadochokinetic speech (DDK) analysis 3.3. Electroglottographic evaluation and acoustic diadochokinetic findings During recording, the EGG waveforms were band-pass filtered (Kemo type VBF8 filter) in the range of 10–5 KHz to eliminate low-frequency variations that may be produced by the laryngeal movement, such as in swallowing and high frequency noise as indicated by the instrument's specifications and used in other studies [34]. Using a computer program (Speech Studio, Laryngograph, Ltd.), the voice samples were digitized, analyzed with a 22,050-Hz sampling rate, displayed on the computer monitor and saved as computer files. A 5 s segment derived from the second trial in the middle of the signal (steadiest airflow) to eliminate onset and offset irregularities was selected for analysis in sustained phonation. A similar analysis was made for the reading of the text passage (demarcation points at the beginning and at the end of the passage). In DDK, the participants were instructed to repeat the syllable series/ pa/, /ta/, and /ka/ as rapidly and precisely as possible on a single exhalation (challenge testing) [6,7]. With the completion of /pa/, the test was repeated for/ta/and /ka/. According to the literature, the second trial is elicited for acoustic analysis because some participants may exhibit difficulty to produce DDK in a single trial [7]. So, the first trial has been used for reasons of practice [7]. The PRAAT software for acoustic analysis [36] was used for the analysis of the DDK. The sound files were annotated and segmented according to the guidelines of the software. Each syllable of/pa/, /ta/, and/ka/was then analyzed by measuring the durations of the consonant, the vowel and the silent periods (distance between each syllable measured in milliseconds). The mean duration of the aforementioned variables in 15 consecutive repetitions (maximum DDK production) was used to compare the two groups of participants according to specific criteria (Supplementary methods). 2.5. Statistical analysis All variables were analyzed using non-parametric Mann–Whitney test for independent samples. Probability values of P b 0.05 were considered as statistically significant. The statistical analysis was carried out using SPSS version 20.0 (SPSS Inc., Chicago). 3. Results 3.1. Reported dysarthria and dysphonia in MG patients All of the MG participants (12/12, 100%) reported breathiness and hypophonia regardless of the antibody status, followed by problems in

Electrophysiological data consisting of EGG recordings of vocal fold function and acoustic DDK findings were used to compare the MG group with the healthy controls analyzing a number of different variables. In sustained phonation, the average fundamental frequency of the vibrating vocal folds (Average Fx) (P b 0.05) and the standard deviation of fundamental frequency (P b 0.001) were higher in the MG group as compared to the control group. In reading, the mean fundamental frequency (DFx1 mean) (P b 0.005) and the fundamental frequency range (DFx1 90% range) (P b 0.005) were also higher in the MG group. In DDK, the MG group of participants showed a higher mean duration of the silent interval between a series of repetitive /pa/ production (P b 0.05), a higher mean duration of the release burst of the consonant/ t/ (P b 0.05), and a lower mean duration of silent interval between a series of repetitive /ka/production (P b 0.05). Table 2 shows the means, standard deviations and P values for all variables measured and Fig. 1 Table 2 Distribution of voice and speech variables in MG participants and pair-matched healthy controls. Variable

Myasthenia gravis n = 12 Mean ± SD

Controls n = 24 Mean ± SD

P value

Voice Handicap Index (VHI) Sustained phonation Average Fx (Hz) SDFx (Hz) Jitter (%) Reading DFx1 mean (Hz) DFx1 90% range (Oct) Alternating Motion Rates (AMRs) (ms) Silent interval before /pa/ Release burst of the sound /p/ Vowel duration of /pa/ Silent interval before /ta/ Release burst of the sound /t/ Vowel duration of /ta/ Silent interval before /ka/ Release burst of the sound /k/ Vowel duration of /ka/

21.750 ± 19.027

0.000 ± 0.000

0.000⁎

248.301 ± 95.765 19.934 ± 24.123 15.798 ± 26.626

173.571 ± 34.348 1.628 ± 1.368 0.751 ± 0.390

0.029⁎ 0.000⁎ 0.072

213.148 ± 37.630 1.225 ± 0.804

171.699 ± 31.450 0.588 ± 0.333

0.002⁎ 0.003⁎

0.081 ± 0.034 0.006 ± 0.004 0.138 ± 0.105 0.076 ± 0.034 0.005 ± 0.002 0.205 ± 0.281 0.010 ± 0.036 0.010 ± 0.003 0.204 ± 0.284

0.053 ± 0.027 0.010 ± 0.013 0.100 ± 0.013 0.065 ± 0.024 0.027 ± 0.047 0.101 ± 0.017 0.071 ± 0.050 0.020 ± 0.043 0.114 ± 0.036

0.024⁎ 0.067 0.934 0.090 0.035⁎ 0.212 0.049⁎ 0.224 0.280

⁎ Significant results (Mann–Whitney test).


45

Fig. 1. Electroglottographic demonstration of dysarthia in a patient with MG. These are images showing characteristic electroglottographic recordings of the production of the syllable/pa/ in a pair-matched control participant (A, above) and an MG patient (B, below). The boundaries of the syllable /pa/are defined in pink color. The MG patient (B) shows an uneven duration of the silence between the series of /pa/ production accompanied with two/pa/ productions without any silence between them. Her speech is characterized by flaccid dysarthria and marked hypernasality. In contrast, the control participant (A) exhibits an even duration of the silence and consecutive/pa/production. Si = Silence between the production of /pa/syllables. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

shows the DDK analysis of the syllable/pa/in one MG participant and its pair-matched control demonstrating the differences. No significant differences were found in any of the EEG variables between MG patients with anti AchR antibodies when compared to the patients with MuSK antibodies. In summary, in sustained phonation and reading, the MG group showed more phonatory variability (higher values for average fundamental frequency and standard deviation of fundamental frequency), as well as more variation in the vibrating vocal folds (higher values for mean fundamental frequency and fundamental frequency range) as compared to the control group. In DDK, the MG group showed longer silence intervals during a series of syllable /pa/and /ka/productions and a higher duration of the sound/t/when produced the syllable/ta/. 4. Discussion In this study we used EGG and acoustic methods for voice/speech analysis in MG, a novel approach to quantitatively assess this important but often difficult to clinically quantify manifestation of the disease. We showed that a number of quantitative acoustic/EGG variables provide sensitive measurements to identify speech/voice dysfunction in MG, and could potentially be offered as a valuable tool in the diagnosis and clinical monitoring of the disease. Our results are in agreement with those previously found in flaccid or mixed (flaccid and spastic) dysarthria caused by other neurological diseases [6,17], including increased

standard deviation of fundamental frequency (sustained phonation), as well as increased mean fundamental frequency and fundamental frequency range (reading). This is the only study to our knowledge which used objective methods to measure voice/speech in MG following a careful matching of the participants. Therefore, it offers a highly needed quantification method for oral and voice muscle function that can be used in addition to the basic quantitative QMG score, as previously proposed [22]. The observed differences between patients with MG and controls most likely result from the manifestations of neuromuscular junction dysfunction. Increased muscle weakness and fatigue may result in incoordination of muscle synergies and reduced force of vocal fold movement. Inability to sustain the smooth adduction of the vocal folds and the proper adductory forces for phonation may lead to higher values in all aforementioned voice variables. The use of EGG as an objective method of voice assessment in MG is preferred because it is noninvasive, generates signals that are not affected by supraglottal influences, it is immune to acoustic noise and does not interfere with other variables such as airflow. In diadochokinesis, muscle weakness and fatigue may explain the higher mean duration in silence before the production of the syllable/ pa/ and the lower mean durations for the consonants/p/and/t/. An interesting finding in the present study is that fatigue in speech is expressed as increased duration of silence between syllables and lower duration of the consonants but not as increased or reduced vowel duration. Thus, it

46


appears that fatigue forces the patient to take more time for rest when reading, expressed in increased silence intervals, as observed in the daily neurological practice. Although we could not find any significant differences in the EEG findings between the AchR-Ab and the MuSK-Ab subgroups, this is likely due to the small number of participants. Future studies may help to identify distinct EGG/acoustic markers for each antibody subgroup of the disease and relate it also to disease duration and QMG scoring. Based on our results, variables that may have a high potential to discriminate patient subgroups include the average fundamental frequency during sustained phonation, the standard deviation of fundamental frequency, the mean fundamental frequency and the fundamental frequency range during reading as well as the silent durations between the syllables during DDK. As this was a pilot study, there was no blinding of the examiner to the status of the patients during the voice/speech assessment. Further research is needed to blindly test this method in a larger cohort of MG patients during or immediately after the neurological diagnosis in order to quantify reliably also mild degrees of dysarthrophonia not necessarily perceivable to the human ear. Another limitation of this study is the lack of stringency in the statistical significance by not using statistical correction for the multiple analyses employed. Because this is a pilot study with a small number of patients this was omitted. EGG/acoustics may also help in the differential diagnosis of the disease, if compared to other groups of patients with dysarthria, especially with other neuromuscular disorders such as bulbar ALS which could mimic myasthenia [37]. Along the same lines, a short computerized assessment protocol that could be easily used by the clinician including voice/speech variables sensitive to change during the disease process may be valuable not only in the assessment of MG but also in other neurological diseases exhibiting dysarthrophonia. Such a protocol may also be important for measuring the effectiveness of speech therapy and other therapeutic interventions during the disease process. Conflict of interest All authors declare no conflict of interest. Author contribution Kostas Konstantopoulos, Eleni Zamba-Papanicoloau and Kleopas A. Kleopa designed and conceptualized the study, Kostas Konstantopoulos performed the speech analysis, Yiolanda-Panayiota Christou, Eleni Zamba-Papanicoloau and Kleopas A. Kleopa evaluated patients participating in the study, Paris Vogazianos performed statistical analysis of the results, all authors analyzed and interpreted the results, Kostas Konstantopoulos and Kleopas A. Kleopa wrote the manuscript, all authors critically reviewed and approved the final manuscript. Acknowledgments We would like to thank the Cyprus MG Association for supporting the study and all participants for their participation. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jns.2017.03.045. References [1] A. Vincent, J. Palace, D. Hilton-Jones, Myasthenia gravis, Lancet 357 (2001) 2122–2128. [2] D. Grob, N. Brunner, T. Namba, M. Pagala, Lifetime course of myasthenia gravis, Muscle Nerve 37 (2008) 141–149.

[3] A. Vincent, L. Clover, C. Buckley, J. Grimley Evans, P.M. Rothwell, U.K.M.G. Survey, Evidence of underdiagnosis of myasthenia gravis in older people, J Neurol Neurosurg Psychiatry 74 (2003) 1105–1108. [4] N.E. Gilhus, J.J. Verschuuren, Myasthenia gravis: subgroup classification and therapeutic strategies, Lancet Neurol 14 (2015) 1023–1036. [5] J.P. Bouwyn, P. Magnier, A.L. Bedat-Millet, P. Ahtoy, D. Maltete, R. Lefaucheur, AntiMuSK myasthenia gravis with prolonged remission, Neuromuscul Disord 26 (2016) 453–454. [6] F.L. Darley, A.E. Aronson, J.R. Brown, Motor Speech Disorders, Saunders, 1975. [7] J.R. Duffy, Motor Speech Disorders: Substrates, Differential Diagnosis, and Management, Elsevier Health Sciences, 2013. [8] M. Montero-Odasso, Dysphonia as first symptom of late-onset myasthenia gravis, J Gen Intern Med 21 (2006) C4–C6. [9] S. Ramalho, S. Pereira, P. Oliveira, H. Morais, N. Lima, A. Condé, Dysphagia as a presenting symptom of myasthenia gravis—case report, Int. J. Otolaryngol. Head Neck Surg. 3 (2014) 23–25. [10] H.R. Sharp, A. Degrip, D.B. Mitchell, A. Heller, Bulbar presentations of myasthenia gravis in the elderly patient, J Laryngol Otol 115 (2001) 1–3. [11] T. Yamamoto, N. Chihara, M. Mori-Yoshimura, M. Murata, Videofluorographic detection of anti-muscle-specific kinase-positive myasthenia gravis, Am J Otolaryngol 33 (2012) 758–761. [12] A. Beliavsky, J.J. Perry, D. Dowlatshahi, et al., Acute isolated dysarthria is associated with a high risk of stroke, Cerebrovasc. Dis. Extra 4 (2014) 182–185. [13] P.P. Urban, H.C. Hopf, P.G. Zorowka, S. Fleischer, J. Andreas, Dysarthria and lacunar stroke: pathophysiologic aspects, Neurology 47 (1996) 1135–1141. [14] P.P. Urban, J. Marx, S. Hunsche, et al., Cerebellar speech representation: lesion topography in dysarthria as derived from cerebellar ischemia and functional magnetic resonance imaging, Arch Neurol 60 (2003) 965–972. [15] P.P. Urban, R. Rolke, S. Wicht, et al., Left-hemispheric dominance for articulation: a prospective study on acute ischaemic dysarthria at different localizations, Brain 129 (2006) 767–777. [16] P.P. Urban, S. Wicht, H.C. Hopf, S. Fleischer, O. Nickel, Isolated dysarthria due to extracerebellar lacunar stroke: a central monoparesis of the tongue, J Neurol Neurosurg Psychiatry 66 (1999) 495–501. [17] K. Konstantopoulos, M. Vikelis, J.A. Seikel, D.D. Mitsikostas, The existence of phonatory instability in multiple sclerosis: an acoustic and electroglottographic study, Neurol Sci 31 (2010) 259–268. [18] A.V. Feijo, M.A. Parente, M. Behlau, S. Haussen, M.C. de Veccino, B.C. Martignago, Acoustic analysis of voice in multiple sclerosis patients, J Voice 18 (2004) 341–347. [19] L. Hartelius, E.H. Buder, E.A. Strand, Long-term phonatory instability in individuals with multiple sclerosis, J Speech Lang Hear Res 40 (1997) 1056–1072. [20] I. Midi, M. Dogan, M. Koseoglu, G. Can, M.A. Sehitoglu, D.I. Gunal, Voice abnormalities and their relation with motor dysfunction in Parkinson's disease, Acta Neurol Scand 117 (2008) 26–34. [21] D. Robert, J. Pouget, A. Giovanni, J.P. Azulay, J.M. Triglia, Quantitative voice analysis in the assessment of bulbar involvement in amyotrophic lateral sclerosis, Acta Otolaryngol 119 (1999) 724–731. [22] A. Jaretzki 3rd, R.J. Barohn, R.M. Ernstoff, et al., Myasthenia gravis: recommendations for clinical research standards. Task Force of the Medical Scientific Advisory Board of the Myasthenia Gravis Foundation of America, Neurology 55 (2000) 16–23. [23] J.M. Liss, L. White, S.L. Mattys, et al., Quantifying speech rhythm abnormalities in the dysarthrias, J Speech Lang Hear Res 52 (2009) 1334–1352. [24] R.D. Kent, J.F. Kent, J.C. Rosenbek, Maximum performance tests of speech production, J Speech Hear Disord 52 (1987) 367–387. [25] S. Skodda, W. Gronheit, C. Lukas, et al., Two different phenomena in basic motor speech performance in premanifest Huntington disease, Neurology 86 (2016) 1329–1335. [26] S. Skodda, Aspects of speech rate and regularity in Parkinson's disease, J Neurol Sci 310 (2011) 231–236. [27] R.S. Bedlack, D.L. Simel, H. Bosworth, G. Samsa, B. Tucker-Lipscomb, D.B. Sanders, Quantitative myasthenia gravis score: assessment of responsiveness and longitudinal validity, Neurology 64 (2005) 1968–1970. [28] M.N. Meriggioli, D.B. Sanders, Autoimmune myasthenia gravis: emerging clinical and biological heterogeneity, Lancet Neurol 8 (2009) 475–490. [29] M.F. Folstein, S.E. Folstein, P.R. McHugh, "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician, J Psychiatr Res 12 (1975) 189–198. [30] P. Enderby, R. Palmer, Frenchay Dysarthria Assessment, Pro-Ed, Austin, TX, 2008. [31] M.E. Helidoni, T. Murry, J. Moschandreas, C. Lionis, A. Printza, G.A. Velegrakis, Crosscultural adaptation and validation of the voice handicap index into Greek, J Voice 24 (2010) 221–227. [32] R.D. Kent, J.F. Kent, Task-based profiles of the dysarthrias, Folia Phoniatr Logop 52 (2000) 48–53. [33] C.R. Krausert, A.E. Olszewski, L.N. Taylor, J.S. McMurray, S.H. Dailey, J.J. Jiang, Mucosal wave measurement and visualization techniques, J Voice 25 (4) (2011) 395–405. [34] A.J. Fourcin, Laryngographic Assessment of Phonatory Function, The American Speech-Language-Hearing Association, ASHA Reports, 1981 116–127. [35] A. Fourcin, Aspects of voice iregularity measurement in connected speech, Folia Phoniatr Logop 61 (2009) 126–136. [36] P. Boersma, Praat, a system for doing phonetics by computer, Glot Int. 5 (2001) 341–345. [37] M.G. Huijbers, E.H. Niks, R. Klooster, et al., Myasthenia gravis with muscle specific kinase antibodies mimicking amyotrophic lateral sclerosis, Neuromuscul Disord 26 (2016) 350–353.