Introduction

The following exercise is a case study of a girl (‘Rachel’) with cleft palate who was studied by Howard (Reference Howard1993). Rachel has grossly impaired speech and a severely reduced phonological system. Yet, she retains a high level of intelligibility. Her speech disorder has only been minimally responsive to prolonged therapy. The case study is presented in five sections: primer on cleft lip and palate; speech, language and hearing in cleft lip and palate; client history; focus on phonological analysis – part 1; and focus on phonological analysis – part 2.

Primer on cleft lip and palate

Cleft lip and palate is a congenital malformation of the upper lip and gum and hard and soft palates. A cleft of the lip can be complete, extending through the lip and into the nose, or incomplete, involving a variable degree of notching of the lip. A cleft lip may be unilateral or bilateral as is shown in Figure 1.1. A cleft of the palate may also be complete or incomplete. In a complete palatal cleft, the cleft extends the full length of the palate. In an incomplete palatal cleft, the cleft may involve just the uvula and soft palate. A palatal cleft may also be submucous (see Case study 2 for further discussion). Several methods of classifying cleft lip and palate have been proposed. A system that continues to be used in many cleft centres is the one proposed by Kernahan and Stark (Reference Kernahan and Stark1958). This system recognises the embryological division of the primary and secondary palates at the incisive foramen. If a palatal cleft occurs in front of the incisive foramen, it is called a primary palate or prepalate cleft. If a palatal cleft occurs behind the incisive foramen, it is called a secondary palate or simply palate cleft. Primary palate clefts may be unilateral (left or right), bilateral or median. Figure 1.2 shows different types of palatal clefts.

Figure 1.1 Unilateral and bilateral cleft lip and nose.

Reprinted with permission from the Cleft Palate Foundation (www.cleftline.org).

Figure 1.2 Unilateral and bilateral cleft palate.

Reprinted with permission from the Cleft Palate Foundation (www.cleftline.org).

Some forms of cleft are more common than others. Cleft lip and palate is the most common diagnosis, accounting for 46% of cases. Isolated cleft palate and isolated cleft lip account for 33% and 21% of cases, respectively. Unilateral clefts are nine times as common as bilateral clefts. They also occur twice as frequently on the left side than on the right (Hopper et al., Reference Hopper, Cutting, Grayson and Thorne2007). The epidemiology of cleft lip and palate has been extensively investigated. Matthews et al. (Reference Matthews, Oddone-Paolucci and Harrop2015) examined the epidemiology of cleft lip and palate in Canada between 1998 and 2007. The mean birth prevalence was 0.82 per 1,000 live births for cleft lip with or without cleft palate, and 0.58 per 1,000 live births for cleft palate. Cleft lip with or without cleft palate was significantly higher in boys, with a boy to girl ratio of 1.75:1. Cleft palate was significantly greater in girls, with a boy to girl ratio in 2007 of 0.59:1. The incidence of oral clefts varies among different ethnicities. Saad et al. (Reference Saad, Parina, Tokin, Chang and Gosman2014) found that the incidence of any cleft disease was highest in the white (non-Hispanic) population in the state of California at 16.2. Lower incidence rates were reported in the Hispanic population (12.26), Asian/Pacific Islanders (11.57), the African American population (8.9) and the Native American population (8.15).

The exact causes of cleft lip and palate are still unknown. What is clear is that genetic factors and environmental teratogens increase the likelihood that a child will develop a cleft. Several genes have been implicated in the aetiology of orofacial clefts (Simioni et al., Reference Simioni, Araujo, Monlleo, Maurer-Morelli and Gil-da-Silva-Lopes2015). Cleft lip and palate is also a clinical feature of many genetic syndromes. The most common syndrome associated with cleft lip and palate is van der Woude syndrome. Isolated cleft palate is most commonly associated with microdeletions of chromosome 21, resulting in velocardiofacial, DiGeorge, or conotruncal anomaly syndromes (Hopper et al., Reference Hopper, Cutting, Grayson and Thorne2007). Several teratogens and other environmental factors have been implicated in the aetiology of cleft lip and palate. Reduced folic acid levels, alcohol consumption, active and passive smoking, and antiepileptic drugs (e.g. topiramate) have all been associated with non-syndromic cleft lip and palate (Bezerra et al., Reference Bezerra, Oliveira, Soares, Cardoso, Ururahy, Neto, Lima-Neto, Luchessi, Silbiger, Fajardo, Oliveira, Almeida, Hirata, Rezende and Hirata2015; Margulis et al., Reference Margulis, Mitchell, Gilboa, Werler, Mittleman, Glynn and Hernandez-Diaz2012; Sabbagh et al., Reference Sabbagh, Hassan, Innes, Elkodary, Little and Mossey2015).

Speech, language and hearing in cleft lip and palate

Speech-language pathologists and audiologists must assess and treat the speech, language and hearing impairments that occur in cleft lip and palate. The primary speech defect is hypernasal speech related to velopharyngeal incompetence, although abnormal dentition and the presence of fistulae can also have phonetic consequences during speech production. Phonetic anomalies may have an adverse impact on a child's developing sound system (i.e. phonology). For example, the child who adopts a backed pattern of articulation in an effort to achieve closure and a build-up of air pressure in the vocal tract may eventually adopt backing as an organising principle within his system of sound contrasts. Such a child has a phonological disorder as well as a phonetic disorder. In a study of 80 children aged 6–15 years with cleft lip and palate, Albustanji et al. (Reference Albustanji, Albustanji, Hegazi and Amayreh2014) reported speech abnormalities including articulation and resonance deficits in 74% of subjects. Productive phonological processes in these children were consonant backing, final consonant deletion, gliding and stopping.

Although speech can improve following palatal surgery, phonetic and phonological defects may persist for many years. Nyberg et al. (Reference Nyberg, Peterson and Lohmander2014) examined the speech of 69 children who had a one-stage palatal repair at a mean age of 13 months. At 5 years of age, more than mild hypernasality, weak pressure consonants and perceived incompetent velopharyngeal function were present in 19 to 22% of children. This improved to less than 5% at 10 years of age. Audible nasal air leakage was present in 23% at 5 years and did not improve by 10 years. Frequent or persistent compensatory articulation was present in 30% at 5 years of age and in 6% at 10 years. At 5 years, 57% of children gave an impression of normal speech. This increased to 89% at 10 years. A high prevalence of distorted /s/ was present in these children at 5 and 10 years of age.

Children with cleft lip and palate often experience expressive language delay. Morris and Ozanne (Reference Morris and Ozanne2003) reported delayed expressive language in 9 of 20 cleft children aged 3 years. Eight of these children achieved a mean length of utterance (MLU) which was below average for their age. There is considerable evidence of delayed lexical development in children with cleft palate. Hardin-Jones and Chapman (Reference Hardin-Jones and Chapman2014) found that the size of the expressive lexicon of toddlers with cleft palate was significantly smaller than that of a noncleft group at 21 and 27 months of age. Discourse deficits have also been reported in children with cleft palate. Klintö et al. (Reference Klintö, Salameh and Lohmander2015) examined narrative retelling in 29 children with unilateral cleft lip and palate. An information score below 1 standard deviation from the norm value was obtained by 65.5% of these children. This compared with 30% in a comparison group of children. Several studies have found evidence of reading impairments in children with cleft palate. Conrad et al. (Reference Conrad, McCoy, De Volder, Richman and Nopoulos2014) found that subjects with non-syndromic cleft of the lip and/or palate performed significantly worse on a test of word reading than control subjects. Word reading deficits were not associated with measures of speech or hearing, but were correlated with auditory memory impairments.

Hearing loss is commonly found in children with cleft palate. In a retrospective audit of 123 newborns with cleft deformities, Tan et al. (Reference Tan, Hee, Yeoh, Lim, Tan, Yeow and Daniel2014) reported the incidence of hearing loss to be 24.4%. This was significantly higher than the hospital incidence of 0.3%. Hearing loss is most often conductive in nature and is associated with the development of otitis media with effusion. However, sensorineural hearing loss can also occur, particularly in children with syndromic cleft palate. Ventilation tube insertion is beneficial to the recovery of hearing in children with cleft palate and otitis media with effusion (Kuo et al., Reference Kuo, Tsao, Cheng, Lien, Hsu, Huang and Shiao2014). However, even after the placement of tubes, hearing loss may persist. Chen et al. (Reference Chen, Messner and Curtin2008) reported that of 30 newborns who failed hearing screening and had tympanostomy tubes placed, 43% had persistent hearing loss. Factors, which predicted persistent hearing loss, were cleft palate alone, female infants and the presence of an associated syndrome.

Client history

Rachel is 6 years old. She was born 11 weeks prematurely with a central cleft of the hard and soft palates. At 2;2 years, she underwent surgical repair of her palatal cleft. Rachel has a severe speech disorder, although her receptive language and expressive language have developed normally. Rachel has a history of fluctuating, mild to moderate, conductive hearing loss (average 45–55 dB). At 3;0 years, grommets were inserted. These were inserted again at 4;0 years. Auditory ability improved significantly following grommet insertion. At 5;11 years, T-tubes were inserted with reported improvements in hearing levels. Notwithstanding improvements in hearing, Rachel's performance in assessments of auditory discrimination for speech sounds remained inconsistent. There was no evidence of either oral apraxia or developmental apraxia of speech.

Rachel had received speech therapy for approximately three years by the time of the study. However, her speech problems had remained largely resistant to change and little progress had been made in therapy. There was also concern that Rachel had reached a plateau and that any further change in her speech would be difficult for her to achieve. Rachel had deficits across several aspects of speech production. She had difficulty initiating, maintaining and coordinating phonation. Her voice was breathy and she displayed high pitch. In relation to resonance, she exhibited hypernasality, nasal emission and nasal friction. A pharyngoplasty was performed at 5;5 years. However, it had had little effect in reducing her nasal emission. In terms of articulation, Rachel displayed glottalisation of consonants. There was also a lack of alveolar and post-alveolar segments in her speech. Rachel exhibited greater difficulty with the articulation of obstruents than nasals and approximants.

Focus on phonological analysis – part 1

Audio- and video-recordings were made of Rachel's speech production during two clinical sessions over a period of five days. To obtain comprehensive coverage of the entire phonological system, the Sheffield Test of Phonetics and Phonology (Eastwood, Reference Eastwood1981) was used. The nearly 100 words from this test were supplemented by words recorded during spontaneous speech and picture description tasks. A detailed phonetic transcription of all words was completed. Symbols from the IPA and extensions to the IPA were used in the transcription. Following transcription, the PACS framework (Grunwell, Reference Grunwell1985) was used to carry out a phonological analysis of Rachel's speech. The data that was used in this analysis is examined in this unit and in the next unit.

Focus on phonological analysis – part 2

Several other aspects of the manner of articulation were examined in Rachel's speech. They included her ability to signal the stop–fricative–approximant continuum and the stop–affricate continuum. These continua tell us something about Rachel's ability to signal the difference between open and close sounds and, in the case of the stop–affricate continuum, the timing of release of closure. Contrasts of place of articulation and the voicing of segments were also examined.

Introduction

The following exercise is a case study of a girl (‘Louise’) aged 3;8 years with Kabuki make-up syndrome who was studied by Van Lierde et al. (Reference Van Lierde, Van Borsel and Van Cauwenberge2000). Kabuki make-up syndrome is a rare genetic syndrome which is characterised by a dysmorphic face, postnatal growth retardation, skeletal abnormalities, intellectual disability and unusual dermatoglyphic (fingerprint) patterns (Matsumoto and Niikawa, Reference Matsumoto and Niikawa2003). For those with the disorder, speech, language and hearing can be adversely affected. The case study is presented in five sections: history and clinical presentation; clinical assessment; communication and cognition profile; focus on speech production; and clinical intervention.

History and clinical presentation

Louise is the second child of healthy, non-consanguineous parents. Her sister, who is 2 years older, is healthy. After a complicated pregnancy, Louise was born at 37 weeks of gestation weighing 2.610kg. A fetal right chylothorax was detected at 20 weeks of pregnancy. (A chylothorax is the presence of lymphatic fluid in the pleural space secondary to leakage from the thoracic duct or one of its tributaries.) Karyotyping by amniocentesis was undertaken and was found to be normal. At birth, Louise was observed to have a high-arched palate with a submucous cleft. She also exhibited generalised hypotonia. The following postnatal investigations were normal: an electroencephalogram, a computed tomography scan of the brain, electromyography and metabolic screening. An internal strabismus with mild nystagmus was revealed during an ophthalmologic examination. Louise had transtympanic drains fitted at 11 months, 18 months and 2 years. At 2;4 years, a hearing examination revealed pure-tone thresholds within the normal range. This examination was repeated at 3;0 years and again revealed normal hearing. Oto-acoustic emissions were also detected. A team of geneticists and dysmorphologists diagnosed Louise at 3;1 years as having Kabuki make-up syndrome (KMS). The diagnosis was based on the presence of facial characteristics (e.g. arched eyebrows), a high-arched palate with submucous cleft, fingertip pads, a foot deformity, broad thumbs, a mild to moderate delay in motor development, and postnatal growth deficiency. At 3;4 years, an assessment of motor skills showed Louise to be at percentile 1 on the gross and fine motor scales of the Peabody Developmental Motor Scales (Folio and Fewell, Reference Folio and Fewell1983). A slight, general hypotonia was also observed.

Clinical assessment

The Dutch version of the McCarthy Developmental Scales (Van der Meulen and Smrkovsky, Reference Van der Meulen and Smrkovsky1986) was used to assess Louise's cognitive level. Language was assessed by means of the Dutch version of the Reynell Developmental Language Scales (Schaerlaekens et al., Reference Schaerlaekens, Zink and Van Ommeslaeghe1993). Louise's voice was assessed by an otorhinolaryngologist and two voice therapists. The otorhinolaryngologist conducted nasolaryngoscopy. The voice therapists used the GRBAS scale (Hirano, Reference Hirano1981) to assess Louise's voice. In order to determine the fundamental frequency of Louise's voice, she was asked to sustain the vowel /a/ for 4 seconds into a microphone. The instrumentation used in this analysis was the Multi-Dimensional Voice Program (model 4305) from Kay Elemetrics Corporation. A picture naming test which consisted of 135 black-and-white drawings of common objects and actions was used to assess Louise's articulation skills. The speech data obtained from this assessment were analysed independently of their relation to the adult targets as well as in relation to the adult standard forms. Three relational analyses were used: phonotactic analysis; phonetic analysis; and phonological process analysis.

Communication and cognition profile

Louise was found to have normal cognitive functioning. Her mean cognitive score was 98. Louise scored at the 60th percentile for receptive language on the Dutch version of the Reynell Developmental Language Scales. She was able to understand a range of named objects, verbs and adjectives. She was also able to respond correctly to instructions that involved an action–object semantic relation. Certain ‘wh’ questions were understood and there was evidence of emerging comprehension of passive sentences. However, it was still difficult for Louise to understand sentences such as ‘The dog is bitten by the rabbit’. Louise was able to understand utterances such as ‘John pushes the baby. Who is naughty?’, a number of spatial prepositions and some terms relating to the size of objects. She could also comprehend primary colours. Louise's expressive language skills exhibited strengths and weaknesses. She was at the 75th percentile in her ability to produce the names of items in the Reynell and to define concrete words (e.g. soap) and abstract words (e.g. being hungry). Louise performed at the 30–40th percentile on the expressive semantics subtest of the Reynell. She was able to express semantic relations of two elements (e.g. ‘prepare dinner’) during story telling based on pictures (e.g. setting the table). However, she was unable to capture the general theme of the depicted situations. Louise's worst area of expressive language (20th percentile) was her morphosyntactic abilities. Nouns, verbs and personal pronouns were the only word classes produced. She made use of singular and plural nouns, but did not use irregular plural forms. Louise also used some nouns with diminutive endings. Verbs only occurred in infinitive form. There were no examples of third-person singular verbs, past participles or future tense verbs. Compound sentences involving either coordination or subordination were completely absent. A negative sentence was occasionally produced. Louise's expressive output only contained sentences of two or three words, with an average of 2.4 words per sentence.

Nasolaryngoscopy failed to reveal any organic or functional voice disorder. Normal results were obtained on all perceptual and instrumental analyses of the voice. The results of the articulation assessment are examined below.

Focus on speech production

• Phonetic inventory: Louise could correctly produce all Dutch vowels and 68% of Dutch consonants. She could not produce correctly the nasal /ɲ/ and the fricatives /f/, /v/, /ʃ/, /Ʒ/ and /h/.

• Phonotactic analysis: Target syllables were usually retained. A change in syllable structure occurred in only 10% of words.

• Phonetic analysis: Compared to target productions at the segmental level, 55% of Louise's consonants were in error and 21% of her vowels. Consonant errors included omissions, substitutions, distortions and additions. Substitutions were the most common error type. The most common types of distortion errors were dentalisation, labiodentalisation, devoicing, weak articulation, mild to moderate hypernasality and moderate nasal emission. Vowel errors included lowering, backing, neutralisation (replacement by a schwa) and unrounding of a target rounded vowel.

• Phonological process analysis: Syllable structure processes are present including cluster reduction (affecting /s/-, /t/- and /R/-blends), final and initial consonant deletion (the former chiefly affecting final /k/) and deletion of unstressed syllables. The following substitutions were in evidence, some of which are shown in the table below.

1. (a) /p/ → /f/; /b/ → /v/; /k/ → /X/; /k/ → /s/; /t/ → /f/

2. (b) /s/ → /t/; /z/ → /b/

3. (c) /k/ → /t/; /ɣ/ → /p/

4. (d) /f/ → /j/

Clinical intervention

Van Lierde et al. (Reference Van Lierde, Van Borsel and Van Cauwenberge2000) recommend the use of ‘tailor-made’ therapy with children who have KMS. They consider this approach to be warranted by the considerable variation that occurs in communication skills both between children with KMS and within individual children with this syndrome. The latter was particularly evident in Louise's case. She displayed a number of intact skills and areas of performance that were within normal limits. For example, Louise had normal cognitive functioning, good receptive language skills and her production of speech sounds was within normal limits for her age. There were also no vocal or laryngeal abnormalities. However, Louise also had considerable difficulties. For example, she had particularly poor expressive language skills in the area of morphosyntax. Although Louise's hearing was within normal limits, she had a history of otitis media that required the placement of transtympanic drains. She also had a submucous cleft palate, slight general hypotonia and poor gross and fine motor skills. Also, her speech sound production was highly variable, and she displayed persisting normal phonological processes, and processes that are uncharacteristic of normal development. Louise also exhibited hypernasality and moderate nasal emission. According to Van Lierde et al., this pattern of communication abilities and impairments cannot be explained by general developmental delay, structural deviations of the speech apparatus, hearing loss or specific language impairment. This pattern, these authors argue, is ‘somewhat reminiscent’ of a phonologic–syntactic disorder.

Introduction

The following exercise is a case study of a 13-year-old girl (‘CB’) with spastic dysarthria who was studied by Marchant et al. (Reference Marchant, McAuliffe and Huckabee2008). CB has a medical diagnosis of spastic hemiplegic cerebral palsy. She received speech and language therapy for her communication disorder between 6 and 11 years of age. The case study is presented in five sections: primer on developmental dysarthria; client history and communication status; focus on spastic dysarthria; intervention; and speech outcome.

Primer on developmental dysarthria

Hodge (Reference Hodge and Cummings2014) defines developmental dysarthria as ‘a group of speech disorders caused by dysfunction of the immature nervous system that delays speech onset and impairs the strength, speed, accuracy, coordination and endurance of the muscle groups used to speak. Depending on the extent of impairment, one or more of the speech processes of respiration, phonation, resonance, articulation and prosody may be affected’ (26). A child with developmental dysarthria may have reduced breath support for speech (respiration), with the result that only short utterances are possible. The vocal folds may fail to adduct normally during phonation, causing the child to speak with a breathy voice. Closure of the velopharyngeal port may not be adequate during speech production, with the result that the child produces hypernasal speech (resonation). Impairments of the strength, speed and accuracy of articulatory movements may lead to the production of weak and distorted consonants and vowels (articulation). The child with dysarthria may be unable to vary the pitch and loudness of the voice, both of which can compromise intonation (prosody). The resulting speech impairment may be mild in nature, and have few implications for a child's intelligibility. Alternatively, the speech disorder may be so severe that no intelligible speech production is possible. In cases of anarthria, an alternative means of communication may need to be found for the client.

A large range of medical conditions, illnesses and events can give rise to developmental dysarthria. Cerebral palsy is the single largest cause of developmental dysarthria. However, other causes of this motor speech disorder include traumatic brain injury (TBI), infections (e.g. encephalitis), cerebral neoplasms, birth anoxia, brain damage related to metabolic disorders (e.g. phenylketonuria), cranial nerve damage in syndromes (e.g. Möbius syndrome) and neurodegenerative disorders (e.g. Duchenne's muscular dystrophy and Friedreich's ataxia). On account of these diverse aetiologies, it is difficult to obtain figures for the prevalence and incidence of developmental dysarthria. Typically, such figures are reported in relation to particular clinical groups within the dysarthria population. In this way, Morgan et al. (Reference Morgan, Mageandran and Mei2010) reported a low incidence of dysarthria (1.25) in a cohort of 1,895 children following TBI. In children with severe TBI, this incidence figure rose to 205. Mei and Morgan (Reference Mei and Morgan2011) reported the incidence of post-surgical dysarthria in 27 children with posterior fossa tumour to be 30%. Sigurdardottir and Vik (Reference Sigurdardottir and Vik2011) found severe dysarthria in 16% of 152 Icelandic children with congenital cerebral palsy. Developmental dysarthria should be distinguished from acquired dysarthria in childhood. It is only in the former type of dysarthria that the neurological injury has its onset prior to the acquisition of speech skills.

Client history and communication status

CB is a 13-year-old girl with spastic hemiplegic cerebral palsy. She has spastic dysarthria. CB attends a mainstream school, and is a native speaker of New Zealand English. Speech is her primary means of communication. Her vision and hearing are within normal limits, and her cognitive skills are sufficient for study participation. Although CB had not received speech and language therapy (SLT) for her dysarthria for one year prior to the study, she received continuous SLT between the ages of 6 and 11 years. However, her parents reported limited success from this intervention. There was little information available on the nature of this intervention other than that it took place for only 30 minutes once a week. Parental report suggested that it involved sound production drills. CB is currently receiving instruction in the use of an augmentative communication device. However, she is resistant to using it, and wishes to continue using speech as her primary means of communication.

An oromotor analysis of CB was undertaken. CB displayed severely restricted lingual and labial movement. This included inadequate tongue elevation, tongue lateralisation, tongue retraction, lip pursing and lip seal. The Goldman–Fristoe Test of Articulation (Goldman and Fristoe, Reference Goldman and Fristoe1986) was conducted. This revealed significantly impaired consonant accuracy, particularly in fricative and affricate production. Perceptual analysis by a listener experienced in dysarthria research confirmed a severe spastic dysarthria, characterised by excessively slow rate, strained–strangled phonation, and imprecise consonant and vowel articulation. CB's expressive language was assessed by means of the Language Assessment Remediation and Screening Procedure (LARSP; Crystal, Reference Crystal1997), the conversation analysis profile (Fey, Reference Fey1986) and the profile in semantics (Crystal, Reference Crystal1997). These assessments revealed a severe language delay. The results of the LARSP indicated a severe syntactic delay. CB's attempts at more complex sentences were highly unintelligible and could not be analysed. During conversational exchanges, CB's reduced speech intelligibility led to frequent communicative breakdowns. This resulted in the use of simplified sentences, and the repetition and rephrasing of utterances. Communication partners frequently sought clarification of CB's utterances. This increased the occurrence of her responsive utterances, and reduced the frequency of her attempts to communicate.

Focus on spastic dysarthria

Spastic dysarthria is more common than other forms of dysarthria in cerebral palsy (Nordberg et al., Reference Nordberg, Miniscalco, Lohmander and Himmelmann2012). Hodge (Reference Hodge and Cummings2014) describes the pathophysiological signs and speech features of spastic dysarthria in children. Pathophysiological signs include slow movements that are limited in range, muscle weakness, excessive muscle tone, muscle rigidity, persisting primitive oral–pharyngeal reflexes and hyperactivity of reflexes that normally persist into adulthood (e.g. jaw stretch, gag). The speech features of this form of dysarthria include vowel and consonant articulation errors, hypernasality, slow speaking rate and short breath groups. During the production of utterances, there are uncontrolled changes in voice quality. The pitch of the voice is lower than expected for the child's age. Speakers with spastic dysarthria exhibit extended word durations. Abnormal resting postures of the lips, tongue and jaw are common.

These speech features of spastic dysarthria have been confirmed in a number of studies. Platt et al. (Reference Platt, Andrews, Young and Quinn1980) examined the speech intelligibility and articulatory impairment of 50 adult males with cerebral palsy. Spastic cerebral palsy and dysarthria were present in 32 of these subjects. The performance of these subjects on two intelligibility measures – accurate recognition of single words and a prose intelligibility rating – was impaired. Indices of articulatory impairment – DDK syllable rates and percentage of correctly articulated phonemes – were also reduced in these subjects. Specific phonemic features in the dysarthric speech of these subjects included anterior lingual place inaccuracy, reduced precision of fricative and affricate manners, and an inability to achieve the extreme positions in the vowel articulatory space. In a later study, Wit et al. (Reference Wit, Maassen, Gabreëls, Thoonen and de Swart1994) compared the performance of two children with TBI-related spastic dysarthria to that of two children with perinatal-onset spastic dysarthria on a number of maximum performance tasks. The three tasks examined maximum sound prolongation, fundamental frequency range and maximum repetition rate. The performance of the children with perinatal-onset spastic dysarthria on all three tasks was poorer than that of their peers with normal speech. The subjects with TBI-related spastic dysarthria performed within normal limits on maximum sound prolongation and fundamental frequency range. However, their maximum repetition rate was extremely slow.

Intervention

CB received a six-week intervention that consisted of phonetic placement therapy (PPT) and surface electromyography (sEMG)-facilitated biofeedback relaxation therapy. PPT focused on articulation, with five consonant sounds selected for treatment: /t/, /s/, /f/, /ð/ and /ʃ/. The selection of these sounds was based on several factors, including CB's developmental stage, the effect of the sound upon intelligibility, and the results of the Goldman–Fristoe Test of Articulation. All sounds were targeted during each PPT session. However, one sound was the focus of the session and received 30 minutes of treatment. CB was informed of the target sound at the start of the session, and was given an auditory and visual representation of it. A mirror was used to help CB place her articulators. Speech drills and specific feedback were employed. CB was also required to provide feedback about her sound productions. Traditional articulation hierarchies were used, with CB progressing to the next level when an accuracy rate of 80% was achieved for a specific target. For all targeted speech sounds, CB did not progress beyond the sounds-in-words level.

sEMG-facilitated biofeedback relaxation therapy was used with the aim of reducing CB's orofacial spasticity. A portable biofeedback device was used during therapy. In a quiet room at home or at school, CB was seated upright and surface electrodes were applied to the skin. Electrodes were placed in three locations: under the chin, on the left top lip and on the right top lip. These locations corresponded, respectively, to the submental (floor of mouth) muscles, the left superior orbicularis oris muscle, and the right superior orbicularis oris muscle. The submental and orbicularis oris muscles were selected for treatment in order to establish if a reduction in muscle tone could improve the articu-lation of consonants and vowels. The first 20 minutes of treatment focused on the reduction of submental amplitude measures during rest and non-speech postures. The equipment's software provided visual feedback in the form of an animated character. To control this character, CB received the following instructions: ‘I want you to make the man sit on his chair for 10 seconds. Remember, to do this you need to try and stay relaxed.’ The aim was to achieve a consistent response at different thresholds. This process was repeated multiple times for all lingual and labial postures.

Speech outcome

The effects of PPT and sEMG were variable, with positive and negative findings resulting from both therapies. At rest, there were no significant differences in sEMG amplitudes across the treatment phases. However, there was a trend towards reduced submental amplitude post-sEMG treatment. There were also considerably smaller standard deviations for all submental values post-sEMG, indicating greater stability. During non-speech postures, amplitude measures decreased significantly for both tongue protrusion and lip pursing tasks post-sEMG.

These therapies also produced variable perceptual effects. There was a significant increase in single-word intelligibility post-PPT that was maintained following sEMG-facilitated biofeedback. However, on Duffy's perceptual rating scale (Reference Duffy1995), there was no change to any articulatory parameters or to overall intelligibility. Imprecise consonants and overall intelligibility were rated as severely deviant. Vowel distortions were also considered to be markedly deviant. Phonemes were still rated as prolonged following both therapies. The subject's self-perception of her speech impairment remained unchanged following intervention. CB still had moderate concern about her speech disorder.

Acoustic measures of vowel and consonant articulation were also made following intervention. There were significant changes in the second formant (F₂) for /æ/ and /u/ following PPT and sEMG, respectively. However, there were no changes in any other formant values. In terms of consonant articulation, there were no significant differences in CV durational measures across any of the targeted syllables. There was only one significant decrease in alternate motion rates, and that was for /kə/ post-sEMG. There was also a significant decrease in inter-syllable gap durations for /pə/ and /tə/ following both PPT and sEMG.

Introduction

The following exercise is a case study of a boy (‘Zachary’) who was studied by Powell (Reference Powell1996). Zachary's speech delay was first noted at 30 months of age. Subsequent assessment revealed a pattern of speech behaviours which was consistent with a diagnosis of developmental apraxia of speech and oral apraxia. The case study is presented in five sections: primer on developmental apraxia of speech; client history; neurological, adaptive and cognitive evaluation; speech, language, hearing and oral mechanism evaluation; and intervention and outcome.

Primer on developmental apraxia of speech

Developmental apraxia of speech (DAS), which is also known as childhood apraxia of speech (CAS) and developmental verbal dyspraxia (DVD), is a complex motor speech disorder which has its onset in early childhood. A position statement published by the American Speech-Language-Hearing Association (2007) defines CAS as:

Although DAS has been extensively investigated, little is still known about the epidemiology of the disorder. Shriberg et al. (Reference Shriberg, Aram and Kwiatkowski1997) stated that DAS occurs in 1–2 children per 1,000. The prevalence of DAS is considerably higher in certain metabolic and genetic disorders. Shriberg et al. (Reference Shriberg, Potter and Strand2011) reported the prevalence of CAS in the metabolic disorder galactosaemia to be 18 per 100. This is 180 times the estimated risk for idiopathic CAS. DAS is more commonly found in boys than in girls. Hall et al. (Reference Hall, Jordan and Robin1993) found an average male:female ratio of approximately 3:1 in a review of 24 group studies and 11 single-subject studies. DAS has been reported in several chromosomal and genetic syndromes including cri du chat syndrome, Down's syndrome and 7q11.23 duplication syndrome (Kumin, Reference Kumin2006; Marignier et al., Reference Marignier, Lesca, Marguin, Bussy, Sanlaville and des Portes2012; Velleman and Mervis, Reference Velleman and Mervis2011). The presence of DAS in these syndromes and many others confirms the neurogenetic origins of the disorder.

The speech features of DAS are well characterised. Children with DAS produce a range of consonant errors. These errors include the deletion of initial and final consonants, cluster reductions, voicing errors and substitutions (Lewis et al., Reference Lewis, Freebairn, Hansen, Iyengar and Taylor2004; Jacks et al., Reference Jacks, Marquardt and Davis2006). The vowel system in DAS is often severely disordered. Lewis et al. (Reference Lewis, Freebairn, Hansen, Iyengar and Taylor2004) found that 100 per cent of the children with CAS in their study produced vowel errors. Davis et al. (Reference Davis, Jacks and Marquardt2005) charted the vowel inventory and accuracy patterns of three children with suspected DAS over a three-year period. Vowel accuracy was impaired in all children, although accuracy did show a moderate increase from the first data recording to the final data recording. Errors consisted mainly of vowel substitutions and de-rhoticisation. No consistent pattern of errors was found in the substitutions. Prosodic disturbances, including anomalies of rate, intonation and stress, have been reported in children with DAS (Odell and Shriberg, Reference Odell and Shriberg2001). Children with DAS also display reduced diadochokinetic rates and poor sequencing of sounds and syllables (Moriarty and Gillon, Reference Moriarty and Gillon2006). There is often an accompanying oral apraxia (Alcock et al., Reference Alcock, Passingham, Watkins and Vargha-Khadem2000). Language problems are also present, with receptive language skills superior to expressive skills (Aziz et al., Reference Aziz, Shohdi, Osman and Habib2010; Grigos and Kolenda, Reference Grigos and Kolenda2010).

Client history

A developmental and medical history was taken, using Zachary's mother as an informant. Zachary was a full-term baby who had a normal delivery. He developed influenza when he was 8 months old. This caused diarrhoea and a high fever. Zachary became dehydrated during his illness as he would not consume liquids. He developed middle ear infections and chickenpox when he was 3 years old. At 30 months of age, a speech delay was noted. As a result, a communication evaluation was conducted. The speech-language pathologist reported that Zachary's behaviours were consistent with DAS and oral dyspraxia. Zachary was enrolled in a course of speech and language therapy. However, his progress was reportedly slow. At 38 months of age, Zachary underwent a comprehensive multidisciplinary evaluation.

Neurological, adaptive and cognitive evaluation

As part of his neurological evaluation, Zachary had a CT scan and an EEG. The scan results were normal, while the EEG was ‘mildly diffusely slow for age’. The paediatric neurologist concluded that Zachary exhibited ‘developmental delay, language greater than motor, of unknown etiology’. At 47 months of age, Zachary underwent a second neurological examination. His sensory examination was normal. Zachary's motor skills were characterised as ‘generally normal, although he [was] minimally hypotonic and [had] a generalized decrease in coordination with running’. The neurologist concluded that Zachary was ‘afflicted with an organic static, probably prenatal encephalopathy resulting in a developmental expressive aphasia, tongue apraxia, and possibly a mild developmental delay’.

On the Vineland Adaptive Behavior Scales (Sparrow et al., Reference Sparrow, Balla and Cicchetti1984), Zachary performed in the low range relative to same-age peers on ‘daily activities required for personal and social sufficiency’. His intellectual functioning was in the low average to borderline range on standardised testing. At 63 months of age, Zachary's non-verbal intellectual functioning was evaluated. Zachary was required to place blocks into a wooden frame to complete visual problems. These problems increased in complexity from simple concretistic matching to conceptual items that required reasoning and problem-solving skills. Zachary displayed some incoordination on this task, especially when smaller or irregularly shaped blocks were used. When Zachary had to combine two or more smaller blocks into a larger one that could be fitted into the frame, his responses were frequently mirror images of the stimuli.

Speech, language, hearing and oral mechanism evaluation

Zachary's language skills were assessed using standardised tests on four occasions between 4;0 years and 6;1 years. On all occasions, his comprehension was in the extremely low to low average range. At 5;1 years, Zachary's score on the Peabody Picture Vocabulary Test (PPVT; Dunn and Dunn, Reference Dunn and Dunn1981) placed him in the 8th percentile. One year later at 6;1 years, Zachary's score on the PPVT placed him in the 14th percentile. Zachary appeared to follow most verbal commands that did not require a verbal response. His comprehension was facilitated by the use of gestures alongside verbal commands. In relation to expressive language, Zachary had fewer than 10 single words at 4 years of age. He imitated single words inconsistently, and produced car-like noises as he played with a toy car. Zachary had such limited expressive speech skills that an articulation test could not be performed. At 4 years of age, he was only able to produce six consonants ([m], [p], [b], [d], [k], [h]) and four vowels ([i], [o], [ɑ], [u]). Syllable structure was adversely affected, with Zachary producing words with a CV structure or reduplicated CVCV structure. His speech displayed frequent homonymous forms.

An examination of the oral mechanism revealed that Zachary's speech structures were symmetric and functioning. There were no apparent organic abnormalities of the oral structures that might interfere with speech production. However, oral motor functioning was difficult for Zachary. He was able to perform most simple voluntary movements. Exceptions were puffing his cheeks, lateralising and elevating his tongue inside his mouth, and pushing the examiner's finger when it was placed against his cheek. It was unclear to what extent Zachary's comprehension problems compromised his performance on these tasks. The production of alternating nonsense syllables was also difficult for him. Zachary had one episode of otitis media at 3 years of age. He passed pure-tone hearing screening.

Intervention and outcome

For over a year, Zachary received individual speech and language therapy at two different institutions. The content of this therapy was consistent with some of the published literature on DAS. However, it resulted in only modest gains in Zachary's communicative competence at which point Powell (Reference Powell1996) undertook an alternative intervention. This new intervention emphasised the early stimulation of unknown aspects of phonology with a view to encouraging broadening of the phonetic inventory and distribution of sounds. It was delivered in four, one-hour treatment sessions per week (Zachary's previous treatment involved two, 30-minute sessions per week). Therapy sessions were modular in nature. One such module, which is taken from Powell (Reference Powell1996: 325), is shown below.

Introduction

The following exercise is a case study of a man (‘GS’) who was studied by Morrish (Reference Morrish1988). GS underwent a total glossectomy for the treatment of a carcinoma at the base of his tongue. His post-operative speech production was examined in detail. The case study is presented in five sections: primer on oral cancer and glossectomy; speech and swallowing following glossectomy; client history; focus on articulation and intelligibility; and focus on instrumental and acoustic analyses.

Primer on oral cancer and glossectomy

Glossectomy is the surgical removal of whole or part of the tongue (total and partial glossectomy, respectively). The procedure is typically performed to treat tongue cancer, although in a smaller number of cases it may also be used to correct congenital macroglossia (Choi et al., Reference Choi, Kim, Park and Kwon2013). The tongue is the most common intraoral site for oral cancer worldwide (Moore et al., Reference Moore, Johnson, Pierce and Wilson2000). In 2013, there were 13,590 estimated new cases of oral tongue cancer in the United States and 2,070 estimated deaths. The most common type of tongue cancer is a squamous cell carcinoma. Less commonly found carcinomas of the tongue include adenoid cystic carcinoma and mucoepidermoid carcinoma (Leong et al., Reference Leong, Pinder, Sasae and Mortimore2007; Luna-Ortiz et al., Reference Luna-Ortiz, Carmona-Luna, Cano-Valdez, Mosqueda-Taylor, Herrera-Gómez and Villavicencio-Valencia2009). As well as tobacco smoking and alcohol consumption, other risk factors for tongue cancer include certain viruses (e.g. Epstein–Barr virus and human papilloma virus (HPV) 16 and 18), cultural practices – prevalent in parts of India – such as reverse smoking (the burning end of cigars is within the mouth) and dipping (placing a mixture of Khaini tobacco and slaked lime in the lower gingival groove), and sexual behaviours including oral sex (although this is likely to be related to HPV infection) (Heck et al., Reference Heck, Berthiller, Vaccarella, Winn, Smith, Shan'gina, Schwartz, Purdue, Pilarska, Eluf-Neto, Menezes, McClean, Matos, Koifman, Kelsey, Herrero, Hayes, Franceschi, Wünsch-Filho, Fernández, Daudt, Curado, Chen, Castellsaqué, Ferro, Brennan, Boffetta and Hashibe2010; Stich et al., Reference Stich, Parida and Brunnemann1992; Zheng et al., Reference Zheng, Xia, Zheng, Takahashi, Masuda and Takano2010).

On account of the serious implications of glossectomy for speech and swallowing, this operation is performed as a last resort when other treatment options (e.g. radiotherapy) have failed to treat a tumour. The defect that is created by glossectomy is reconstructed with a local closure, a local flap or a free flap. A local closure is used when the defect is small. If the defect is relatively large, then it is closed with the use of a flap. To form local flaps, tissue can be raised from the neck (platysma muscle), chest (pectoralis muscle) and forehead (frontalis muscle). Free flaps can be formed using tissue from the radial forearm, inside of the thigh (gracilis flap) and the abdomen (rectus abdominis muscle). The oral surgeon must consider a range of factors in the choice of flap, including the size of defect to be reconstructed, the aesthetic appearance of the tongue and the functional outcome for the patient in terms of speech and swallowing. The radial forearm free flap has emerged as the standard for partial glossectomy as it provides the desired bulk and contour for reconstruction. For total glossectomy, the more bulky anterolateral thigh flap and rectus abdominis can achieve greater propulsion of food into the pharynx during swallowing, the greater volume of these flaps compensating for the lack of motor function (Allan et al., Reference Allan, Van Haren, Wang, Thaller, Taub, Patel, Buchman and Cohen2015).

Advanced carcinoma of the tongue can necessitate additional surgical procedures including laryngectomy, mandibulectomy and pharyngectomy. These procedures also have serious implications for swallowing and the production of speech and voice. Van Lierop et al. (Reference Van Borsel and Vandermeulen2008) studied eight patients who underwent total glossectomy. Three patients also required a total laryngectomy for a tumour involving the pre-epiglottic space or larynx. Five patients had a marginal mandibulectomy, one underwent segmental mandibulectomy and one required partial pharyngectomy. For patients with advanced squamous cell carcinoma of the tongue, Sinclair et al. (Reference Sinclair, Carroll, Desmond and Rosenthal2011) reported reduced disease recurrence at 12 months postoperatively for patients who underwent total laryngoglossectomy (40%) compared to total glossectomy (61%). Disease-free survival at 12 months was also higher in patients with total laryngoglossectomy (50%) than in patients with total glossectomy (40%). However, intelligible speech was less often achieved by patients with total laryngoglossectomy (10%) than by patients with total glossectomy (30%).

Speech and swallowing following glossectomy

Clients who undergo glossectomy are under the care of a multidisciplinary team. Included in this team are oral surgeons, radiation and medical oncologists, otolaryngologists, prosthodontists and speech-language pathologists. It is the role of the speech-language pathologist to assess and treat swallowing and speech problems in clients with glossectomy. SLP management of the client should begin at the point of diagnosis and continue until the best possible speech and swallowing outcomes have been achieved. However, the duration and intensity of intervention can vary markedly between healthcare systems and treatment centres.

Even the very best surgical results following glossectomy will involve some degree of compromise of swallowing and speech function. Dziegielewski et al. (Reference Dziegielewski, Ho, Rieger, Singh, Langille, Harris and Seikaly2013) examined functional outcome data in 12 patients who underwent total glossectomy with laryngeal perseveration. These investigators also conducted a systematic review of the literature. Fifty per cent of the patients in this study, and 24% with systematic review, were still using gastrostomy tubes at one year post-surgery. In patients who could swallow, swallowing transit times more than doubled, but aspiration did not occur. On average, spoken sentence intelligibility was 66%. The best swallowing and speech functional outcomes occurred in patients who attended over 80% of swallowing and speech rehabilitation sessions. Vega et al. (Reference Vega, León, Cervelli, Pons, López, Fernández, Quer and Masià2011) examined 39 patients who underwent glossectomy (24 total glossectomy, 15 subtotal glossectomy). Oral feeding was resumed in 33 (85%) of these patients. Speech was judged to be good or acceptable in 27 (87%) patients. Improved speech and swallowing outcomes following glossectomy are associated with flap reconstructions which have sufficient bulk and vertical height to allow for contact with the palate (Rigby and Hayden, Reference Rigby and Hayden2014).

Speakers who undergo glossectomy can achieve relatively good intelligibility through the use of unusual labial, mandibular and pharyngeal speech compensations (Kazi et al., Reference Kazi, Prasad, Kanagalingam, Georgalas, Venkitaraman, Nutting, Clarke, Rhys-Evans and Harrington2007). A number of these compensations occurred in the speech of seven German patients who were studied by Barry and Timmermann (Reference Barry and Timmermann1985). These patients were aged 32 to 62 years and each underwent a partial glossectomy. Plosive sounds involving the tongue were substituted by fricatives, glottal stops or bilabial stops. Kaipa et al. (Reference Kaipa, Robb, O'Beirne and Allison2012) examined speech produced by a 31-year-old female in the three-month period following glossectomy. Acoustic analysis revealed improvements in the vowel space area during this time. However, there was deterioration in the perception of this speaker's consonants, with anterior sounds being perceived more correctly than medial and posterior sounds. Bressmann et al. (Reference Bressmann, Jacobs, Quintero and Irish2009) assessed speech acceptability in 22 patients with partial glossectomies. These investigators found that the amount of tongue tissue resected predicted 41 per cent of the variance in post-surgical speech acceptability. Moreover, a defect size of more than 20.4 per cent of tongue tissue was found to be the critical cut-off point for poorer speech acceptability.

Client history

Subject ‘GS’ is a 69-year-old white man who underwent total glossectomy eight years earlier. This procedure was necessitated by a diagnosis of invasive squamous cell carcinoma at the base of the tongue. The epiglottis was preserved and there was no mandibular involvement. Dentition also remained unaffected, although GS's lower teeth had been replaced by dentures. The defect created by glossectomy was reconstructed using a forehead flap. Frontalis muscle was introduced through the cheek. There was no free edge to this flap – it was entirely set in. Following glossectomy, GS received speech therapy for two years. Notwithstanding the radical nature of GS's surgery, a speech and language therapist judged his speech to be good.

Focus on articulation and intelligibility

To determine GS's intelligibility, 16 phonetically naive subjects listened through headphones to an audio-recording of 10 randomly devised sentences. All sentences were played once in full and then singly, with no repetition permitted. Subjects also listened to an extract of conversation with GS. Repetition of the recording was permitted, although subjects were advised not to spend more than five or six minutes on this task. Subjects then orthographically transcribed what they heard. The transcription had to be 100% correct in order for a subject to be said to have understood what GS said. Intelligibility scores across the sentences and conversational passage were high at 65% and 58%, respectively. There was considerable variability between listeners, however, with intelligibility scores on the conversational passage ranging from 15% to 71%. A second intelligibility test was conducted. Because it had been observed that GS used a bilabial place of articulation for all plosive sounds, the focus of this test was on his production of initial plosive consonants. Subjects listened to audio-recordings of GS as he produced the following CVC monosyllables – kit, beat, tick, deep, dip, bill, peat, keep, pit, gill and teak. Although the bilabial plosives /p, b/ were identified at a rate better than chance, the alveolar plosives /t, d/ and the velar plosives /k, g/ did not even reach chance, and were overwhelmingly identified as bilabial plosives.

In the absence of a tongue, GS used a complete bilabial closure to articulate all plosive sounds. To establish if he varied this closure to signal different plosives, a close-up video analysis of GS during speech production was performed. The degree of bilabial protrusion, jaw retraction/protrusion and jaw elevation/lowering was measured in millimetres as deviations from a rest position. The same CVC monosyllables as above were used during this examination. Although there was little difference between alveolar and velar plosives, there was a consistent and statistically significant difference between bilabial and alveolar plosives in terms of jaw lowering. A lower posture was adopted for bilabial plosives. These same plosives also showed a statistically significant difference in terms of jaw protrusion, with bilabial plosives more protruded than alveolar plosives. There was a slight statistically significant difference between bilabial and alveolar plosives in terms of bilabial protrusion. GS used a more protruded posture for /t, d/ than for /p, b/.

Focus on instrumental and acoustic analyses

Instrumental analyses using electromyography (EMG) and a strain gauge apparatus were used to confirm the results of the video analysis of GS's speech. During EMG, surface electrodes were placed on either side of the philtrum from where they measured the activity of the orbicularis oris muscle. GS read CV monosyllables in which the vowels were neural for lip position. There was a high statistical difference in the activity of the orbicularis oris muscle during the articulation of /t, d/ and /p, b/, with greater activity observed during alveolar plosives. Through use of the strain gauge apparatus it was confirmed that jaw lowering was greater for /p, b/ than for /t, d/. These instrumental techniques confirmed that GS was using certain postures consistently during the production of bilabial and alveolar plosives, and that these postures were detectable during video and instrumental analyses, even if they were not always perceived by listeners.

Videofluoroscopy was used to assess GS's articulation of high and low vowels. For high vowels, the pharynx widens and the grafted flap rises slightly. There is slight protrusion of the jaw which retains a close position. The soft palate closes firmly and is raised to a position well above Passavant's ridge. For low vowels, pharyngeal narrowing is achieved by a strong retraction of the anterior wall and the epiglottis. The flap rises and bunches and is carried back with the anterior wall of the pharynx to approximate the posterior wall. The jaw moves downwards and backwards, with downward movement exaggerated in comparison with normal speakers. GS's production of vowels and consonants also underwent acoustic analysis. GS repeated randomly ordered CVC words which were analysed on a sound spectrograph. Formant frequencies were measured and compared with those from a normal speaker. GS was unable to differentiate between high and low vowels acoustically. Spectrograms were also made of all the CVC words used in the intelligibility tests. Several acoustic characteristics were examined. However, none were judged to correspond to GS's articulatory differences between bilabial and alveolar plosives.

Introduction

The following exercise is a case study of a 39-year-old man (‘AB’) who was studied by Grunwell and Huskins (Reference Grunwell and Huskins1979). AB suffered a cerebrovascular accident in December 1972 which resulted in dysarthria and other neurological sequelae. The case study is presented in five sections: primer on acquired dysarthria; client history and clinical presentation; speech evaluation; focus on mixed dysarthria; and assessment issues.

Primer on acquired dysarthria

‘Acquired dysarthria’ is a collective term for a group of motor speech disorders which are caused by damage to the central and/or peripheral nervous systems. Typically, these disorders have their onset in adulthood, although acquired dysarthria may also arise in childhood. The perceptual anomalies that occur in acquired dysarthria reflect impairments in one or more speech production subsystems, namely, articulation, resonation, phonation, respiration and prosody. Acquired dysarthria may minimally disrupt speech intelligibility. Alternatively, the disorder may compromise speech production to such an extent that the client is severely unintelligible and must use an alternative means of communication.

The incidence and prevalence of acquired dysarthria in the general population is largely unknown. To understand the epidemiology of this motor speech disorder, it is necessary to examine its incidence and prevalence in relation to specific clinical populations. Adults who sustain a cerebrovascular accident or stroke are most at risk of acquired dysarthria. Using the Registry of the Canadian Stroke Network's database, Flowers et al. (Reference Flowers, Silver, Fang, Rochon and Martino2013) estimated the incidence of dysarthria in a sample of 221 patients with acute ischaemic stroke to be 42%. Another significant cause of acquired dysarthria in children and adults is traumatic brain injury (TBI). Safaz et al. (Reference Safaz, Alaca, Yasar, Tok and Yilmaz2008) reported the presence of dysarthria in 30.2% of 116 persons with TBI who were followed at a single centre in the five-year period between 2000 and 2006. Aside from stroke and TBI, other neurological disorders associated with acquired dysarthria include multiple sclerosis, Parkinson's disease, and motor neurone disease (also known as amyotrophic lateral sclerosis). The incidence and prevalence of acquired dysarthria in several of these disorders has also been investigated. Danesh-Sani et al. (Reference Danesh-Sani, Rahimdoost, Soltani, Ghiyasi, Haghdoost and Sabzali-Zanjankhah2013) reported the presence of dysarthria in 42.1% of 500 patients aged 11 to 69 years with multiple sclerosis. Perez-Lloret et al. (Reference Perez-Lloret, Nègre-Pagès, Ojero-Senard, Damier, Destée, Tison, Merello and Rascol2012) estimated the prevalence of dysarthria in 419 patients with Parkinson's disease to be 51%.

All aspects of speech production may be compromised in dysarthria. In terms of articu-lation, a speaker with dysarthria may produce weak and distorted consonant and vowel sounds. Problems with resonation may arise on account of mistiming of the closure of the velopharyngeal port. A breathy voice may reflect inadequate glottal valving of the pulmonary airstream during phonation. A speaker with dysarthria may produce short, truncated utterances on account of impaired respiration. In terms of prosody, a speaker with dysarthria may place stress on the wrong syllables within words, and the wrong words within sentences. Each of these speech production difficulties are related to neuromuscular deficits which affect the range, speed, force and timing of movements of the articulators and other organs of speech.

Dysarthria may occur in isolation in clients. However, it is often the case that it is found alongside dysphagia and other neurogenic communication disorders, principally aphasia and apraxia of speech. In their study of 221 clients with acute ischaemic stroke, Flowers et al. (Reference Flowers, Silver, Fang, Rochon and Martino2013) found that the highest co-occurrence of any two impairments was 28% for the presence of dysphagia and dysarthria. Dysarthria in combination with dysphagia and aphasia was present in 10% of these clients. A combination of dysarthria, sialorrhoea (drooling) and dysphagia was present in 136 (33%) of 419 patients with Parkinson's disease examined by Perez-Lloret et al. (Reference Perez-Lloret, Nègre-Pagès, Ojero-Senard, Damier, Destée, Tison, Merello and Rascol2012). These additional disorders are significant in that they can make a differential diagnosis of dysarthria difficult (particularly if apraxia of speech is present), and they provide additional obstacles to the successful rehabilitation of clients.

Client history and clinical presentation

AB is 39 years old. In December 1972, he had a cerebrovascular accident (an intracerebral haematoma). He has a history of hypertension. Following his CVA, he presented with right-sided weakness. He displayed right upper and lower facial weakness, some dysphasia, dysphonia and dysarthria of brainstem origin. AB's wife reported that his speech had been somewhat nasal and difficult to understand prior to his CVA. In January 1974, AB's speech musculature was examined. It was found to be bilaterally affected, with greater impairment on the right. There was spastic weakness of the lips and tongue. The palate was atrophied and foreshortened. The palatal impairment was attributed to flaccid weakness. All articulatory movements, but especially those involving the palate, were impaired. AB was given a speech diagnosis of mixed dysarthria owing to the presence of both bulbar and suprabulbar signs.

Speech evaluation

AB's speech production underwent a comprehensive assessment. The slow movement of the articulators contributed to a general slow speaking rate. There was articulatory imprecision in the production of consonants and vowels which was related to slowness of movement and weakness of articulation. AB frequently failed to reach target positions. This resulted in errors of vowel quality and the use of a fricative stricture in place of a stop. In terms of phonation, AB displayed a breathy voice. The loudness of the voice did not vary. There was also a lack of pitch variation, and the pitch of the voice was abnormally high. AB exhibited audible nasal emission, which was evident on a spectrogram as a high frequency noise trace. Excessive nasal resonance was evident during voiced segments.

The interaction of these deviant features reduced the intelligibility of AB's speech. AB's uneconomical use of air meant that he was only able to speak in short phrases. His slow rate of articulatory movements also contributed to the restricted length of his utterances. The lack of variation in pitch and loudness and poor breath control compromised the use of stress, rhythm and intonation. Breathy voice and excessive nasal resonance limited AB's ability to produce qualitative distinctions between vowels. A reduction in intra-oral air pressure compromised consonant production, particularly plosives and fricatives.

Focus on mixed dysarthria

AB's speech disorder was diagnosed as a mixed flaccid-spastic dysarthria. Individual deviant speech parameters could be attributed to spasticity and/or flaccidity. AB's slow rate of speech and articulatory imprecision could be attributed to spasticity which limits the range, force and speed of articulatory movements. Poor adduction and abduction of the vocal folds, manifested in a breathy voice and audible inspiration, was suggestive of a flaccid condition of the vocal folds. However, an indirect laryngoscopy was not performed and so this impression could not be confirmed. AB's breathiness might also be explained by spasticity. A spastic condition of the vocal folds, manifested in extreme tension of the folds and their reduced elasticity, could also explain AB's problems with high monopitch and monoloudness.

AB's palatal abnormality, which is responsible for his nasal emission and hypernasality, can be attributed to the combined effects of spasticity and flaccidity. The effect of spasticity on the palate is to produce a tendency towards downward movement. Flaccidity makes the palate too weak to counteract this movement. Because a flaccid weakness of the palate tended to predominate in AB, he had an incompetent velopharyngeal sphincter. AB's spasticity was exacerbated by the increased muscular effort needed to counteract the effects of his weak and fatiguing speech musculature. His general body tension rose as he made more effort to speak. As he made more effort to speak, the imbalance created by the asymmetrical nature of his spasticity was also exaggerated. The effect of these factors was heard as increasing inefficiency and weakening of articulations, the loss of control over coordination of breathing and phonation leading to increasing air wastage, and continuous nasal escape related to a breakdown of control over velopharyngeal valving.

Assessment issues

The assessment of AB's speech focused on the functioning of speech production subsystems and the impact of deviant speech parameters on AB's intelligibility. The functioning of physiological mechanisms such as phonation and respiration and the impact of impairment of these mechanisms on speaker intelligibility correspond to the ‘body’ and ‘activity’ levels of the International Classification of Functioning, Disability and Health (ICF; World Health Organization, 2001). Lowit (Reference Lowit and Cummings2014: 400) states that ‘[a]t the body level, the functioning of the various speech subsystems, i.e. respiratory, laryngeal, and velopharyngeal function and oral musculature are investigated. In terms of the activity level, a variety of data elicitation methods are available to assess intelligibility, ranging from evaluations of single words and phrases to reading passages and spontaneous speech.’ The evaluation of these levels typically proceeds by means of published assessment protocols such as the Frenchay Dysarthria Assessment-2 (FDA-2; Enderby and Palmer, Reference Enderby and Palmer2008), which examines speech production under the following 11 sections: reflex; respiration; lips; jaws; palate; larynx; tongue; intelligibility; rate; sensation; and associated factors.

Today, dysarthria assessment is as likely to consider the impact of this speech disorder on an individual's participation in daily activities (the third level of the ICF framework) as it is on a speaker's intelligibility. During an assessment of dysarthria, it is at least as important for clinicians to consider the impact of dysarthria on an individual's self-esteem, quality of life and social participation as it is to assess the effect of respiratory, phonatory and articulatory impairments on a client's intelligibility. The need for assessment of the impact of dysarthria on a client's psychological wellbeing and participation could not be clearer. Piacentini et al. (Reference Piacentini, Mauri, Cattaneo, Gilardone, Montesano and Schindler2014) reported that dysarthria-related quality of life is significantly compromised in patients with multiple sclerosis and dysarthria, while Skolarus et al. (Reference Skolarus, Burke, Brown and Freedman2014) found that dysarthria is a predictor of participation restrictions in stroke survivors. A number of clinical tools have been validated for the purpose of assessing impact and participation, not all of which are designed specifically for clients with dysarthria. (The reader is referred to Lowit (Reference Lowit and Cummings2014) for a list of scales for the assessment of the psychosocial effects of communication disorders.)

Introduction

The following exercise is a case study of a 35-year-old man (‘R’) who was studied by Day and Parnell (Reference Day and Parnell1987). R was diagnosed at 23 years of age with Wilson's disease. He has had a 10-year contact with speech-language pathology since his diagnosis. The case study is presented in five sections: primer on Wilson's disease; client history and presentation; speech intervention: part 1; speech intervention: part 2; and speech outcome.

Primer on Wilson's disease

Wilson's disease is an autosomal recessive disorder that affects copper metabolism, leading to toxic accumulation of copper in the liver, central nervous system and kidneys (Beinhardt et al., Reference Beinhardt, Leiss, Stättermayer, Graziadei, Zoller, Stauber, Maieron, Datz, Steindl-Munda, Hofer, Vogel, Trauner and Ferenci2014). The disorder is also known as ‘hepatolenticular degeneration’, a term which reflects damage to the liver (‘hepato’) and the lenticular nucleus of the basal ganglia (‘lenticular’) as a result of copper deposits. Wilson's disease is caused by mutations in the ATP7B gene (Lorincz, Reference Lorincz2010). The prevalence of the disorder is estimated to be 1:49,500 (Møller et al., Reference Mei and Morgan2011). Lai and Tseng (Reference Lai and Tseng2010) estimated the annual incidence rate to be 0.27 per 100,000. There is evidence of gender differences in Wilson's disease, with more men than women developing the neuropsychiatric form of the disorder, and more women than men developing the hepatic form (Litwin et al., Reference Litwin, Gromadzka and Członkowska2012). There is a good, long-term prognosis for those individuals with Wilson's disease who receive adequate care (Beinhardt et al., Reference Beinhardt, Leiss, Stättermayer, Graziadei, Zoller, Stauber, Maieron, Datz, Steindl-Munda, Hofer, Vogel, Trauner and Ferenci2014).

Neurological and psychiatric symptoms are a common feature of Wilson's disease. In a sample of 126 patients with the disease, Mihaylova et al. (Reference Mihaylova, Todorov, Jelev, Kotsev, Angelova, Kosseva, Georgiev, Ganeva, Cherninkova, Tankova, Savov and Tournev2012) reported neurological signs in 82 subjects. The most frequently observed signs were tremor and dysarthria. Rigidity, bradykinesia and pyramidal signs were found in over 25% of patients. Dystonia, chorea, athetosis, ballismus and epilepsy were rarely observed in these patients. Moores et al. (Reference Moores, Fox, Lang and Hirschfield2012) examined 48 adult patients with Wilson's disease, 21 of whom presented with neurological symptoms. Diagnostic magnetic resonance imaging revealed basal ganglia and brainstem abnormalities and atrophy in 64%, 64% and 36%, respectively. At follow-up, 50% exhibited basal ganglia lesions and 55% displayed atrophy. Individuals with Wilson's disease have an increased lifetime prevalence of psychiatric disorders, including major depressive disorders and bipolar disorders (Carta et al., Reference Carta, Sorbello, Moro, Bhat, Demelia, Serra, Mura, Sancassiani, Piga and Demelia2012).

There has been little systematic investigation of dysarthria in Wilson's disease. This is despite the fact that dysarthria is a common neurological symptom of the disorder. Machado et al. (Reference Machado, Chien, Deguti, Cançado, Azevedo, Scaff and Barbosa2006) reported dysarthria in 91% of a sample of 119 patients with Wilson's disease. An early study of 20 patients with Wilson's disease reported the presence of a mixed dysarthria, with prominent ataxic, spastic and hypokinetic features (Berry et al., Reference Berry, Darley, Aronson and Goldstein1974). The 10 most common speech production errors in these individuals were reduced stress, monopitch, monoloudness, imprecise consonants, slow rate, excess and equal stress, low pitch, irregular articulatory breakdown, hypernasality and inappropriate silences. Speakers with Wilson's disease exhibit an impaired speech rate and impaired control of speech rate (Pernon et al., Reference Pernon, Trocello, Vaissière, Cousin, Chevaillier, Rémy, Kidri-Osmani, Fougeron and Woimant2013). Dysarthria in Wilson's disease has been found to correlate with lesions of the putamen and caudate (Starosta-Rubinstein et al., Reference Starosta-Rubinstein, Young, Kluin, Hill, Aisen, Gabrielsen and Brewer1987).

Client history and presentation

In 1974, when R was aged 23 years, he began to experience dysarthria, drooling, dysphagia, decreased mental function and some dystonia. Subsequently, he went on to develop right foot drop, postural changes, clumsiness of the upper extremities and marked personality changes. These changes manifested as intermittent periods of depression. R was diagnosed as having Wilson's disease. His older sister had been diagnosed two years earlier with the disorder and had died shortly thereafter.

Following diagnosis, R was prescribed Penicillamine (an agent used for the mobilisation and elimination of copper) and was placed on a low-copper diet. However, R did not comply with either his drug or dietary regimen. Between 1974 and 1976, he contacted a number of specialists in psychiatry, internal medicine and speech-language pathology for assistance with his problems. These contacts were of limited duration with the exception of a period of psychiatric treatment for depression which lasted 18 months. R's speech deteriorated significantly during his psychiatric treatment. In April 1976, a psychiatrist remarked that ‘speech (had) deteriorated almost to the point of being completely nonunderstandable…a direct result of the neurological damage associated with the (Wilson's) disease’. The psychiatrist also expressed the view that ‘while (R's) speech might improve slightly, the chances of significant improvement (were) poor’. In the same year, R attempted suicide by means of a drug overdose. He was hospitalised in a mental health facility after a short period in an intensive care unit. His marriage failed during this time.

When R first attended the Columbia Speech and Hearing Clinic at the University of Missouri in 1976, he was already profoundly dysarthric. His speech was described as ‘unintelligible for the most part’. His speech was non-functional, and he relied on written messages and sign language for communication. However, these other modes of communication were also of limited effectiveness. R also presented with drooling and dysphagia, had a fixed, grimace-like expression, and had oral and facial rigidity. On account of his severe speech disorder, R had been unable to work for two years. Accordingly, he received full disability compensation under social security.

Speech intervention: part 1

At the Columbia Speech and Hearing Clinic, R received intervention for his dysarthria during 22 of 28 consecutive semester sessions over a 10-year period. A range of therapeutic techniques were used to improve intelligibility. Progress was monitored through the use of intelligibility ratings performed by familiar and unfamiliar listeners. During the 10-year period of R's intervention, specific areas of emphasis and therapy procedures, and the length and complexity of target utterances were varied to meet R's changing needs and abilities. Medical supervision of R was maintained throughout this entire time. Even though certain drug and dietary controls were prescribed, R displayed poor compliance with them.

Between admission in 1976 and the winter semester in 1977, R's speech intelligibility in clinic increased significantly. Oral-motor exercises which were aimed at improving the flexibility and precision of the oral structures were attempted initially. However, these produced little gain and were consequently abandoned. There was also an early emphasis on increased vocal volume and differentiation of vowel productions. However, this also made little contribution to the improvement of intelligibility. The focus of therapy then turned for several years to the production of single consonants and consonant blends and clusters. The production of multisyllabic words and word-final consonants was particularly difficult for R and made a large contribution to his reduced intelligibility. R was encouraged to engage in self-monitoring of his productions and, with the assistance of the therapist, to achieve consistency in his standards of judgement. His reliance on gestural communication, which was largely confusing and unsuccessful, was discouraged. Throughout this time, R refused to consider the use of an augmentative or alternative communication system.

Speech intervention: part 2

After intervention on the production of individual speech sounds proved to have limited benefit for speech intelligibility, the focus of intervention shifted to the use of suprasegmental guidelines on a continuous basis in contextual speech. The first two of these four guidelines emphasised the use of slowed rate and syllable-by-syllable production. This had a marked, positive effect on speech intelligibility for R. In fact, so considerable was the increase in R's intelligibility that he was able to secure part-time employment in 1980 for the first time since his diagnosis. In addition to these techniques, R was encouraged to make use of adequate respiratory support and appropriate pausing. He also received instruction in the use of stress and intonation patterns. However, these latter techniques were not as effective at improving intelligibility as slowed rate and syllable-by-syllable production. The third guideline encouraged the use of overarticulation within R's compensatory techniques. This required R to exaggerate articulatory movements. The fourth and final guideline, which complemented the third guideline, encouraged R to make use of increased mouth opening. This also increased R's intelligibility considerably. Notwithstanding these gains, R's motor speech skills were still highly vulnerable to the effects of physical fatigue. He was advised to rearrange his schedule in order to obtain sufficient rest, especially prior to therapy sessions. Because this advice was not always acted upon, R's compliance with this recommendation became a precondition for his enrolment in further therapy.

R's intelligibility was also compromised by hypernasality. R's hypernasal speech did not respond to traditional therapy techniques. Accordingly, a palatal lift appliance was recommended, which R used for approximately two years. By 1983, R's hypernasality had improved so much that even on days when he did not wear his palatal lift to therapy, his hypernasal speech had diminished considerably. Eventually, R was able to maintain improvements in his velopharyngeal control in the absence of the device, only resorting to its use during periods of fatigue.

Although R made considerable speech gains in clinic, he failed to generalise the use of his compensatory guidelines to communicative situations outside of clinic. Consistent with the therapeutic goal of helping R achieve optimal intelligibility and functional speech that was appropriate to his communicative needs in a range of settings, R was encouraged to assume ever greater responsibility for self-evaluation of his intelligibility and his use of the guidelines. Daily logs, which included ratings of intelligibility and consistency in use of the guidelines, were used for this purpose. Both R and his communicative partners completed these logs on a consistent basis. There was some deterioration in R's speech performance in the summer semester in 1985 when these logs were not completed. R married for the second time in 1983. R's wife occasionally attended therapy sessions in order to become acquainted with objectives and techniques which she could implement at home. It is expected that with greater involvement of R's wife in providing support, behaviour maintenance and feedback to R, he might be able to be discharged permanently from therapy.

Speech outcome

R received intervention for his dysarthria over a 10-year period. At the outset of therapy, R was judged to have only 5% intelligibility (rated sentence by sentence) in connected speech. By the end of therapy, this had increased to an average of 95% intelligibility. During his 10-year period of intervention, R experienced several episodes when there were significant decreases in his intelligibility. These episodes could be accounted for by factors in R's lifestyle which he did not adequately manage and by R's underlying neurological disorder. When audiotaped samples of R's conversational speech from 1977 and 1985 were analysed and ranked according to the relative prominence of deviant speech features, it was found that whereas reduced intelligibility was first-ranked in 1977, it had dropped to fifth-ranked in 1985. This reflected R's success in using his compensatory techniques to achieve the major goal of treatment, optimal intelligibility and functional speech. The most prominent deviant speech features recorded in 1985 were imprecise consonants and prolonged phonemes. R's improved communicative status has enabled him to engage in full-time employment for the last three years. He has been remarried for two years and is the father of an infant son.

Introduction

The following exercise is a case study of a man aged 71 years who was studied by Peach and Tonkovich (Reference Peach and Tonkovich2004). This client developed apraxia of speech (AOS) following a cerebrovascular accident. His case is somewhat unusual in that his AOS was caused by a subcortical haemorrhage of the basal ganglia. Most cases of AOS reported in the literature arise on account of cortical damage. The client also had stroke-induced aphasia. The case study is presented in five sections: primer on apraxia of speech; medical history; early post-stroke period; language assessment; and focus on speech.

Primer on apraxia of speech

McNeil et al. (Reference McNeil, Robin, Schmidt and McNeil2009: 264) state that ‘apraxia of speech is a phonetic-motoric disorder of speech production’. This definition is extended to include the locus of breakdown in speech production – AOS is caused when well-formed phonological frames are not efficiently translated into the kinematic patterns that are used for carrying out intended movements. This inefficiency is manifested as temporal and spatial segmental and prosodic distortions both within and between articulators. The durations of consonants and vowels and the time between sounds, syllables and words may be extended as a result. The perceptual consequences may be heard as sound substitutions, the mis-assignment of stress, and other phrase- and sentence-level prosodic abnormalities. The location of speech errors within utterances is relatively consistent and the type of errors is invariable. (It should be noted that some researchers (e.g. Staiger et al., Reference Staiger, Finger-Berg, Aichert and Ziegler2012) state that the invariability or consistency of speech errors is contentious at best, and should not be included as a criterion in the diagnosis of AOS.) AOS is not related to deficits of muscle tone or reflexes. It is also not related to deficits of sensory and language processing. AOS rarely occurs in pure or isolated form. When it does, it is not accompanied by deficits of basic motor physiology, perception or language.

Little is known about the epidemiology – prevalence and incidence – of AOS. Duffy (Reference Duffy2005) reported that AOS accounted for 7.6% of all motor speech disorders diagnosed in the Mayo Clinic Speech Pathology practice between 1987 to 1990 and 1993 to 2001. This figure was based on 6,101 evaluations of people with a primary speech pathology diagnosis of a neurologic motor speech disorder. AOS is most frequently caused by cerebrovascular accidents. However, there is also a neurodegenerative form of AOS called primary progressive AOS (Duffy et al., Reference Duffy, Strand, Clark, Machulda, Whitwell and Josephs2015). There is still no consensus on the neuroanatomical location of the lesion(s) that are associated with AOS. However, potential regions include Broca's area, the lateral premotor cortex, subcortical structures and the anterior insula, all in the left hemisphere (Robin and Flagmeier, Reference Robin, Flagmeier and Cummings2014). As described above, pure AOS is a rare occurrence. AOS is much more likely to be found alongside dysarthria, aphasia and/or oral apraxia. Botha et al. (Reference Botha, Duffy, Strand, Machulda, Whitwell and Josephs2014) reported that non-verbal oral apraxia is very common among patients with progressive AOS, and that the severity of AOS is predictive of oral apraxia. The presence of these other conditions makes a differential diagnosis of AOS difficult. However, there is evidence that specialist speech and language therapists display high intra-rater and inter-rater reliability in diagnosing the presence and severity of AOS in clients with communication difficulties following stroke (Mumby et al., Reference Mumby, Bowen and Hesketh2007).

Medical history

The client is a 71-year-old, right-handed man. He presented with an acute loss of speech and right hemiplegia following a cerebrovascular accident. Prior to his CVA, his medical history was unremarkable. Three days post-onset a CT scan was performed. It revealed a large, left frontal haemorrhage that probably originated in the region of the basal ganglia and extended into the frontal white matter. The lower half of the primary motor cortex, the posterior portion of the inferior frontal gyrus, the insula and the postcentral opercular structures were not involved.

Early post-stroke period

Ten days after his stroke, the client was admitted to rehabilitation. He was alert but inconsistently responsive when an initial speech-language evaluation was performed. The client exhibited muteness and occasionally used simple gestures to refer to his inability to speak and understand questions. He used undifferentiated pointing responses to tasks that required auditory word discrimination and the matching of visual words to pictures. The client displayed no appropriate responses to questions in either formal or informal conversational contexts. Occasionally, when functional commands were presented in an appropriate context, he was able to follow them. Writing was impaired, with the non-preferred hand used to produce an unintelligible scribble.

A programme of speech-language intervention was initiated. During the next two weeks, the client exhibited minimal vocalisation, and spontaneously used a few grammatically well-formed sentences. Bucco-facial apraxia compromised his imitation of oral movements. He began to respond successfully to tasks that elicited automatic verbal responses by the end of the fourth week post-stroke. However, his speech remained aphonic. His inability to imitate speech consistently frequently frustrated him. There was moderate impairment of word recognition and auditory comprehension of yes–no questions. By pointing to an alphabet board, he was able to spell one simple word out of three.

His speech output at five weeks post-stroke was much improved from his initial appearance. However, he continued to be dysfluent – there was reduced phrase length and poor grammatical complexity – and he had substantial difficulty initiating speech. He was echolalic and perseverative. In conversation, there were occasional, but more frequent grammatically well-formed sentences of normal lengths. Although he was able to read sentences aloud, he did so with substantial difficulty. Through the use of self-cuing he was able to increase the consistency of vocalisation during speech production. There was mild impairment of auditory comprehension for word recognition. Limb apraxia made it difficult for him to follow commands using parts of his body. The client's naming was poor and was influenced by linguistic variables such as frequency of occurrence, open versus closed class and concrete versus abstract. Verbal paraphasias were also observed.

Language assessment

The first formal assessment of the client's speech and language skills was undertaken at 6 weeks post-onset, when the Boston Diagnostic Aphasia Examination (BDAE; Goodglass and Kaplan, Reference Goodglass and Kaplan1983a) was conducted. The client's speech production was aphonic and effortful. He displayed trial and error behaviour, groping articulatory movements and attempts at self-correction. His speech displayed dysprosody which was unrelieved by extended periods of normal rhythm, stress and intonation. He had difficulty initiating utterances and his speech was perceived to contain phoneme substitutions. His description of the cookie theft picture from the BDAE is shown below:

The client's performance on the various subtests of the BDAE revealed a mild to moderate aphasia. The Rating Scale Profile of Speech Characteristics revealed that articulatory agility and paraphasia in running speech were the two most impaired areas, while repetition and auditory comprehension were areas of relative strength. The client's melodic line, phrase length, grammatical form and word-finding were disrupted, but were not as severely impaired as articulatory ability and paraphasia. This pattern of deficits did not correspond to any of the classical aphasia syndromes. The client's performance on each subtest of the BDAE is shown below:

• Auditory Comprehension:

  • Word discrimination (69/72)

  • Body part identification (17/20)

  • Commands (12/15)

  • Complex ideational material (8/12)

• Oral Expression:

  • Oral agility (10/12)

  • Verbal agility (14/14)

  • Automatised sequences (5/8)

  • Repetition of words (8/10)

  • Repeating phrases (high probability) (8/8)

  • Repeating phrases (low probability) (7/8)

  • Word reading (30/30)

  • Responsive naming (29/30)

  • Visual confrontation naming (102/105)

  • Body part naming (29/30)

  • Oral sentence reading (8/10)

• Reading Comprehension:

  • Symbol and word discrimination (10/10)

  • Word recognition (8/8)

  • Comprehension of oral spelling (7/8)

  • Word–picture matching (9/10)

  • Reading sentences and paragraphs (9/10)

• Writing:

  • Mechanics of writing (1/3)

  • Primer-level dictation (14/15)

  • Written confrontation naming (3/10)

Focus on speech

A detailed assessment of the client's speech was undertaken. The client orally read a series of word lists that sampled 45 voiced and unvoiced stop, fricative, affricate, nasal, semi-vowel and consonant cluster phonemes. Each phoneme or blend was sampled in initial, medial and final positions of words. Single phonemes were examined in words which were one to two syllables in length, while blends were examined in words of one to three syllables. Two experienced judges undertook broad phonemic transcriptions of a total of 1,057 audiotaped responses. Two types of phoneme errors were excluded from the analysis. First, phonemes which were produced in place of a target phoneme in a word were not included in the analysis. Second, substitutions of voiceless for voiced cognate phonemes were also omitted. These exclusions resulted in 77 (25%) of the total errors undergoing analysis. It was found that significantly more substitution errors were produced than other error types, followed by additions, repetitions, the use of intrusive schwa and finally omissions. Significantly more phoneme errors occurred in word initial position than in word medial and final positions. There were similar percentages of errors across different phoneme groupings (consonant cluster, fricative, nasal, etc.).

Introduction

The following exercise is a case study of a 32-year-old, Dutch-speaking woman (‘TDF’) who was studied by Van Borsel et al. (Reference Van Borsel, Janssens and Santens2005). TDF presented at the neurological department of a university hospital with speech problems of sudden onset. Although her history revealed noteworthy neurological events, her speech problems were considered to be of psychogenic origin. The case study is presented in five sections: primer on foreign accent syndrome; client history and medical investigations; language and oral motor evaluation; focus on articulation and prosody; and focus on accent.

Primer on foreign accent syndrome

Kuschmann and Lowit (Reference Kuschmann and Lowit2012) define foreign accent syndrome (FAS) as a ‘motor speech disorder in which a variety of segmental and suprasegmental errors lead to the perception of a new accent in speech’ (738). To date, there have been no epidemiological investigations of this speech disorder, although there is general consensus that the condition is rare. Typically, the disorder has a neurogenic aetiology. FAS has been reported in adults following cerebral infarction (Sakurai et al., Reference Sakurai, Itoh, Sai, Lee, Abe, Terao and Mannen2015; Trax and Mills, Reference Trax and Mills2013), traumatic brain injury (Lippert-Gruener et al., Reference Lippert-Gruener, Weinert, Greisbach and Wedekind2005) and the development of brain tumours (Tomasino et al., Reference Tomasino, Marin, Maieron, Ius, Budai, Fabbro and Skrap2013). The speech disorder has also been documented in adults with neurodegenerative disorders such as multiple sclerosis (Chanson et al., Reference Chanson, Kremer, Blanc, Marescaux, Namer and de Seze2009). Less commonly, a psychogenic aetiology is presumed to underlie the disorder. Verhoeven et al. (Reference Verhoeven, Mariën, Engelborghs, D'Haenen and De Deyn2005) described FAS in the case of a Dutch-speaking adult with conversion disorder. A developmental form of FAS has been described in two child patients by Mariën et al. (Reference Mariën, Verhoeven, Wackenier, Engelborghs and De Deyn2009).

Linguistic investigations of FAS have focused primarily on the production of segments and use of intonation in speakers with the disorder. Verhoeven and Mariën (Reference Verhoeven and Mariën2010) described pronunciation problems at the segmental level in a female speaker of Belgian Dutch who developed a French/German foreign accent after a left hemisphere stroke. Many of this speaker's vowels had clear creaky voice. There was a substantially smaller vowel space and less acoustic differentiation between [i], [ɪ] and [e]. There was a tendency to devoice fricatives and stops. Velar fricatives were pronounced as velar stops, and the glottal fricative was almost consistently realised as a glottal stop in word-initial position. The alveolar trill was systematically pronounced as a uvular trill, which is a typical feature of French learners of Dutch.

Intonation and other aspects of speech prosody have been extensively examined in FAS. Katz et al. (Reference Katz, Garst and Levitt2008) reported inaccurate and highly variable lexical stress assignment in a 46-year-old, monolingual female who presented with FAS of unknown aetiology. This client's sentence-level intonation also showed occasional deviations from typical American English patterns. Kuschmann and Lowit (Reference Kuschmann and Lowit2012) studied the pragmatic use of intonation in four speakers with FAS. Specifically, the use of phonetic parameters (fundamental frequency, intensity and duration) and phonological categories (pitch accents and de-accentuation) to signal information status was examined. Although speakers with FAS were able to use phonetic parameters to differentiate between given and new information, they frequently placed pitch accents on given information instead of de-accenting these elements. The foreign qualities of the Dutch-speaking adult described by Verhoeven et al. (Reference Verhoeven, Mariën, Engelborghs, D'Haenen and De Deyn2005) extended beyond pronunciation to include features of the lexicon, syntax and pragmatics.

Client history and medical investigations

TDF first visited the neurological department of the authors’ university hospital because of speech problems of sudden onset. Her medical history contained several, noteworthy features. Since the age of 6, TDF had a right-sided neurosensory hearing loss with sloping configuration. A head trauma and whiplash injury were sustained nine years earlier. This caused chronic headache with variable lateralisation which was predominant in the occipito-nuchal and parietal region. An otorhinolaryngologist had been consulted one month earlier for a hoarse voice following a minor head trauma. However, a clinical ENT examination was normal: no structural abnormalities were revealed by flexible laryngoscopy and the voice sounded normal to the otorhinolaryngologist. It was shortly after this ENT consultation that TDF's speech problems appeared suddenly during an evening out with friends. TDF was described as being mute for some time after the onset of her speech difficulties. A clinical neurological examination was also normal. There were no motor or sensory impairments. Pathological reflexes were absent, and tendon reflexes were symmetrical. The results of coordination, posture and gait tests were normal. A whole brain CT scan was also normal.

TDF experienced significant psychiatric problems. She was being treated for depression and suicidal ideation at the time of the study. This was related to family problems. Specifically, TDF's husband had rapidly progressing Huntington's disease. Yet, she had been kept unaware of the family history in relation to this condition. TDF had concerns about the implications of this disease for her 10-year-old son who had a learning disorder. TDF took two antidepressants, sertraline and trazodone. However, neither drug is known to have any effect on speech either singly or in combination. A further drug, prazepam, was also administered to TDF. (Prazepam is a benzodiazepine which is used in the treatment of anxiety disorders.) This drug can cause slurred speech when given in high doses. However, TDF did not receive high doses of this drug.

Language and oral motor evaluation

The results of a number of language assessments indicated that TDF did not have aphasia. The maximum score was obtained on the Dutch version of the Token Test (Van Dongen et al., Reference Van Dongen, Van Harskamp and Luteijn1976) and on a Dutch aphasia battery, the Stichting Afasie Nederland test (Deelman et al., Reference Deelman, Koning-Haanstra, Liebrand and Van Den Burg1981). TDF had intact comprehension of written language on the Dutch version of the Aachener Aphasie test (Graetz et al., Reference Graetz, De Bleser and Willmes1992). TDF could recite without error the days of the week, months of the year, and could count numbers from 1 to 20. Word and sentence repetition was also intact. There were no word-finding difficulties during confrontation naming or a word retrieval test. TDF could also read aloud without difficulty. Video-recorded speech samples revealed some grammatical difficulties. Some auxiliaries, articles and prepositions were omitted, giving TDF's speech a telegraphic quality at times. Grammatical errors, including the use of singular for plural nouns, incorrect verb conjugation and incorrect word order, were present but not pervasive. There was no paralysis, paresis or oral apraxia during oral motor testing.

Focus on articulation and prosody

Video-recorded speech samples of TDF were phonetically transcribed and analysed. The samples included recitation of automatic series, word and sentence repetition, reading aloud, confrontation naming, word retrieval, monologue and conversation. TDF displayed no struggle behaviour during speech. The use of auditory masking and delayed auditory feedback during reading aloud did not diminish TDF's accent. A number of articulatory errors were noted, some of which are shown below:

• /ma:ntɑx/ for /ma:ndɑx/ (‘Monday’)

• /ɑltɛi/ for /ɑltɛit/ (‘always’)

• /di/ for /dri/ (‘three’)

• /awaji/ for /hawaji/ (‘Hawaii’)

• /spre:kə/ for /spre:kən/ (‘talk’)

Other, less common error patterns were the production of the trill sound /r/ with insufficient trill, and the substitution of /ɣ/ or /x/ for /h/. Several errors occurred on a single occasion: /ʃ/ for /s/, /j/ for /r/, /ŋ/ for /n/, /r/ for /n/, the addition of /ɣ/ in word-initial position, and vowel changes such as /ə/ for /ɔ/ and /ɛ/ for /u/.

TDF displayed a number of prosodic anomalies. Word and sentence stress were aberrant and TDF displayed a tendency towards scanning speech. Stress was placed on the wrong syllables in words. Incorrect word stress occurred on 18.1% of two-syllable words, 35.1% of three-syllable words, 62.5% of four-syllable words, and 30% of words of five or more syllables. The speech tasks with the highest level of incorrect word stress were confrontation naming (61.1%) and word retrieval (51.7%). Incorrect sentence stress occurred in 6.8% of the sentences produced. Stress was placed on the wrong word in these sentences.

Focus on accent

The perception of a foreign accent in TDF was confirmed by a listener experiment. A total of 33 third-year students in speech-language pathology, none of whom were familiar with foreign accent syndrome, listened to a speech sample of 6 minutes 25 seconds’ duration. The speech sample contained fragments of automatic speech, word repetition, reading aloud, monologue, conversation and confrontation naming. The subjects were asked to decide if the speech sample was from a native speaker of Dutch and, if not, which language was the speaker's mother tongue. All subjects expressed the view that TDF's speech was unusual, with 69.7% stating that Dutch was not the speaker's native language. Linguistic features which were judged to be significant in this assessment were prosodic abnormalities (improper word stress, scanning, strange intonation), grammatical errors (deletion of articles, improper word order, errors of pluralisation), use of short sentences and the occurrence of ‘strange sounding’ sounds.

Although nine different accents were perceived in total, most subjects indicated that TDF was likely to be a native of an Eastern European country. TDF's personal history revealed no source for this accent. TDF was born in the Dutch-speaking part of Belgium to parents who were native speakers of Dutch. TDF was raised in a Dutch-speaking environment and attended Dutch schools. She had never lived abroad, and had only ever undertaken short journeys, mainly to Spain. She studied French at introductory level at school. TDF's husband was also a native Dutch speaker.

TDF was seen one year later when she accompanied her husband to a neurological consultation. Her speech was nearly normal at this stage. Her foreign accent had gradually resolved within five months of onset. A second listener experiment was conducted with 13 students in speech-language pathology, none of whom had participated in the first listener experiment. Although all 13 subjects judged TDF's speech to be unusual, only one stated that TDF was not a native speaker of Dutch. However, this subject was unable to identify TDF's mother tongue.