Brain and Language 85 (2003) 185–189 www.elsevier.com/locate/b&l Thalamic stuttering: A distinct clinical entity? John Van Borsel,* Sandie Van Der Made, and Patrick Santens Ghent University Hospital, UZ Gent, 2PI, Spraakrevalidatie, De Pintelaan 185, Gent B-9000, Germany Accepted 15 January 2003 Abstract A 38-year-old right-handed male with no history of speech or language problems presented with neurogenic stuttering following an ischaemic lesion of the left thalamus. He stuttered severely in propositional speech (conversation, monologue, confrontation naming, and word retrieval) but only slightly in non-propositional speech (automatic speech, sound, word and sentence repetition, and reading aloud). It is suggested that thalamic stuttering may constitute a distinct clinical entity. Ó 2003 Elsevier Science (USA). All rights reserved. Keywords: Neurogenic stuttering; Acquired stuttering; Thalamic stuttering; Thalamus 1. Introduction Although stuttering most often has its onset in childhood, usually somewhere between the ages of two and five (Johnson & Associates, 1959), dysfluent speech may appear also for the first time later in life, beyond the typical childhood period. A dysfluency that originates after the typical childhood period has been called acquired stuttering, or sometimes also late-onset stuttering or adult onset stuttering. Acquired stuttering may have various causes. It can be of psychogenic origin, drug-induced or malingered. Most frequently, however, acquired stuttering is of neurogenic origin, i.e., associated with damage to the central nervous system. In recent years, several cases of neurogenic stuttering have been reported (see Ringo & Dietrich, 1995 and Van Borsel, 1997, for a review). At first sight, it would seem that neurogenic stuttering is not linked to any specific lesion site. The damage that lies at the origin of the dysfluencies in neurogenic stuttering may be bilateral or unilateral, focal or diffuse, cortical or subcortical, situated in the right hemisphere or the left hemisphere. Within one hemisphere the lesion may be localized in the frontal lobe, the temporal lobe, or the parietal lobe. Only in conjunction with damage to the occipital lobe neurogenic stuttering has not been * Corresponding author. Fax: +32-9-240-4993. E-mail address: john.vanborsel@rug.ac.be (J. Van Borsel). observed. This should not mean, however, that neurogenic stuttering has no localizing significance at all. There is still the possibility that within neurogenic stuttering the symptoms vary according to the lesion site and that different types of neurogenic stuttering can be distinguished accordingly, just like is the case in other neurogenic disorders of speech and language, such as dysarthria and aphasia. A few proposals in that direction with regard to neurogenic stuttering have already been formulated (Ackermann, Hertrich, Ziegler, Bitzer, & Bien, 1996; Koller, 1983). The present paper reports a case of neurogenic stuttering due to a thalamic stroke. An analysis of the dysfluency pattern prompted the suggestion that thalamic stuttering may be a distinct clinical entity. 2. Case history The subject of this study was a right-handed male businessman who had no history of speech or language problems. At the age of 38 he suffered a stroke. An initial CT-scan on admission showed no abnormalities. Magnetic resonance imaging (see Fig. 1) 2 weeks later revealed an ischemic lesion in the left ventrolateral thalamus. The initial neurological testing demonstrated aphasic symptomatology and a very mild right hemiparesis with a decreased sensation in the right hemisoma. Six months later these symptoms were still present, although less 0093-934X/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0093-934X(03)00061-0 186 J. Van Borsel et al. / Brain and Language 85 (2003) 185–189 Fig. 1. MRI showing an infarct of the left lateral thalamus. prominent. At that time a marked cognitive dysfunction was demonstrated by neuropsychological testing. Attention and episodic memory (Digit span from the Wechsler Memory Scale; Rey Auditory Verbal Memory Test) were clearly deficient. Mild executive dysfunction (Verbal Fluency; Stroop Color and Word Test; Wisconsin Card Sorting Test; Trail Making Test) and oral apraxia was found. A neurolinguistic examination revealed aphasic symptoms consistent with a thalamic aphasia. The patient showed good auditory comprehension. He obtained a score of 55/61 on the Dutch version of the Token test (Van Dongen, Van Harskamp, & Luteijn, 1976) and a score of 45/45 on the sentence comprehension test of the S.A.N., a Dutch aphasia battery by Deelman, Koning-Haanstra, Liebrand, and Van De Burg (1981). Word and sentence repetition, as assessed by a subtest from the Aachener Aphasie Test (A.A.T.) (Graetz, De Bleser, & Willmes, 1992) appeared to be preserved too. When presented the subtest ‘‘benoemen’’ (naming of pictures of objects) from the A.A.T., however, the patient demonstrated manifest word finding problems and on the subtest ‘‘diernamen noemen’’ (naming animals) from the S.A.N., a word retrieval test, he scored on percentile 0. Moreover, there were frequently pauses during conversation. In addition to language problems the patient also demonstrated fluency failures, which were further analysed from videotaped speech samples (Sony Video Hi 8 camera). Modalities included in the speech samples were automatic speech (days of the week, months of the year, counting from 1 to 20), repetition of sounds, words, and sentences (subtest from the A.A.T.), reading aloud (the IPA text ‘‘The north wind and the sun. . .,’’ International Phonetic Association, 1974), conversation (on the patientÕs job and leisure activities), monologue (describing action and situation pictures from the AAT), confrontation naming (the subtest ‘‘benoemen’’ from the AAT) and word retrieval (subtest ‘‘diernamen noemen’’ from the S.A.N.). A possible adaptation effect was tested by having the patient read aloud the IPA text three times in a row. Finally, for three of the above speech types namely automatic speech, repeating words, and reading aloud, the patient was also tested under time pressure by having him perform the tasks again, at the end of the session, as fast as he could. The recorded speech samples were subjected to a consensus transcription (ordinary spelling) and a consensus analysis of the occurrence and nature of dysfluencies by two of us (JVB and SVDM). Overall, across all speech samples gathered, a dysfluency index (number of stuttered syllables relative to the total number of syllables produced) of 12% was found. Interjections, word repetitions, and part-word repetitions were the most frequent type of dysfluency accounting for 35.5, 24.5, and 24.5%, respectively, of all dysfluencies. Less frequently the patient showed phrase repetitions (6.5%), revisions (5.5%), incomplete phrases (2%), sentence repetitions (1.5%), and prolongations (1%) (see Fig. 2). A further analysis of the word level repetitions showed that whole-word repetitions consisted manly of a single repetition of a word (e.g., 36 out of 51 instances during conversation) and that part-word repetitions consisted most of the time of a single repetition of part of the word (e.g., 33 out of 51 instances during conversation). The most striking characteristic of the patientÕs dysfluency, however, was a discrepancy in Fig. 2. Relative frequency of different types of dysfluency (across various speech tasks). J. Van Borsel et al. / Brain and Language 85 (2003) 185–189 Fig. 3. Percentage of dysfluencies in different speech modalities. frequency of occurrence of dysfluencies in the less propositional speech modalities versus frequency of occurrence in more propositional speech modalities. Whereas in automatic speech, sound, word and sentence repetition, and reading, frequency of occurrence of dysfluencies ranged from 0 to 6%, conversation, mono- 187 logue, confrontation naming, and word retrieval yielded percentages between 18.5 and 24 (see Fig. 3). As one can also see from Fig. 3, performance under time pressure did not give rise to a higher number of dysfluencies. As to a possible adaptation effect, the number of dysfluencies during the reading task appeared to be too limited for any such effect to happen. A second evaluation, one year and two months later, revealed the same patterns. Again interjections were the most frequent type of dysfluencies, followed by word repetitions and part-word repetitions, and again there was a striking difference between the occurrence of dysfluencies in more propositional speech modalities and the occurrence in less propositional speech modalities. Follow-up neuropsychological testing at that time indicated unchanged memory and attention with improved executive function. Functional neuro-imaging by means of perfusion-SPECT (see Fig. 4) revealed hypoperfusion of the left thalamus, right cerebellum and left frontal and anterior temporal regions. 3. Discussion While it is well-known that the lesions causing neurogenic stuttering may have different localizations, cases Fig. 4. Perfusion SPECT scan demonstrates areas of hypoperfusion, indicated by arrows. 188 J. Van Borsel et al. / Brain and Language 85 (2003) 185–189 of neurogenic stuttering due to thalamic lesions are rare. In fact, the only reported cases, as far as we could ascertain, seem to be the ones of Andy and Bhatnagar (1992) and of Abe, Yokomay, and Yorifuji (1993). Andy and Bhatnagar (1992) reported stuttering in four neurosurgical patients who also suffered from chronic pain, seizures and somatosensory disorders. Dysfluencies observed were sound, syllable and word repetitions, prolongations, and hesitations. The presence of mesothalamic discharges in each of these patients, as assessed by EEG, and the fact that therapeutic (left) mesothalamic self-stimulation through implanted brain stimulation electrodes ameliorated the stuttering in all four patients, lead the authors to conclude that the stuttering in these subjects resulted from mesothalamic perturbation. Abe et al. (1993) described what they called ‘‘stuttering like repetition’’ in a 38-year-old man resulting from infarcts in the paramedian thalami and the midbrain. In this patient the repetitions were restricted to syllables. Repetition of words or phrases was not observed. In addition, there are a few studies that reported dysfluencies induced by electrical stimulation of the ventrolateral thalamus in parkinson patients. Ojemann and Ward (1971) mentioned repetition of the first syllable during a naming task on stimulation of the ventrolateral thalamus. Schaltenbrand (1975) reported compulsory speech with a pattern like that in stammering, stuttering, and palilalia. Furthermore, Andy and Bhatnagar (1991) observed dysfluencies elicited by mechanical perturbation of the thalamus rather than electrical stimulation. In a 29-year-old patient with a history of seizures they observed dysfluencies that consisted of repetition of an initial syllable during electrode insertion in the intralaminar centromedian nucleus in the left thalamus, preparatory to therapeutic lesion placement for chronic pain. The present case can be considered as further evidence of the importance of the thalamus in the production of fluent speech. In our patient, who had never stuttered before, a lesion in the left ventrolateral thalamus gave rise to the frequent occurrence of dysfluencies. Interjections, word repetitions, and part-word repetitions were especially observed, but also phrase repetitions, revisions, incomplete phrases, sentence repetitions, and prolongations occurred. Canter (1971) claimed that in neurogenic stuttering dysfluency and level of propositionality are in an inverse relationship. Choric speaking and repetition would be more difficult than oral reading, and self-formulated speech would be the easiest task. Ringo and Dietrich (1995), however, in a critical review of 97 published cases of neurogenic stuttering, found no evidence supporting the idea of an inverse relationship between propositionality and dysfluency. They concluded that in neurogenic stuttering dysfluencies commonly occur across speech tasks and that in this respect neurogenic stuttering differs from developmental stuttering, where it is often observed that dysfluencies tend to decrease under conditions of reduced propositionality. The findings of the present study are neither compatible with the idea of an inverse relationship between propositionality and dysfluency nor with the idea of a tendency in neurogenic stuttering for dysfluencies to occur across speech tasks. The patient of the present study was definitely less fluent in more propositional speech. A similar observation was reported by Abe et al. (1993). Repetition of syllables in their patient ‘‘was exclusively observed in spontaneous speech or in reply to a question and was rarely observed when he repeated words spoken to him, or when he read sentences’’ (p. 1024). More in particular, a frequency of stuttering of 5 and 9% were reported for repeating and reading and a frequency of 28 and 58% for naming and spontaneous speech, respectively. Clearly, there are exceptions to the pattern of an inverse relationship between propositionality and dysfluency or to the pattern that dysfluencies occur irrespective of the propositionality of speech in neurogenic stuttering. Neurogenic stuttering resulting from a thalamic infarction seems to be one of these exceptions. It has been suggested before that in neurogenic stuttering different lesion sites may yield different dysfluency patterns. Koller (1983) suggested that a distinct dysfluency syndrome is associated with extrapyramidal disease, in particular parkinsonÕs disease. This syndrome would be characterized by the rare occurrence of dysfluencies in choral and repetitive speech and the frequent presence of an adaptation effect (i.e., a reduction of the number of dysfluencies on consecutive readings of the same text). However, the notion of a distinct extrapyramidal dysfluency syndrome was not corroborated in a study by Hertrich, Ackermann, Ziegler, and Kaschel (1993). The patient of this study, a 65-year-old female with parkinsonÕs disease, did not show a positive adaptation effect and she did not present with a higher frequency of iterations during self-formulated speech than during repetitive speech. Ackermann et al. (1996) in turn put forward the suggestion that neurogenic stuttering due to mesiofrontal lesions might be associated with a specific pattern of dysfluency. They reported the case of a 53-year-old male with transcortical motor aphasia following an ischemic infarction of the mesiofrontal cortex. Apart from being aphasic the patient was also dysfluent, showing mostly prolongations and repetitions. According to Ackermann et al. (1996) the pattern of dysfluency was different from that usually seen in neurogenic stuttering. The dysfluencies were restricted to word-initial sounds and syllables and were absent during sentence repetition and during reading aloud. This constellation, according to Ackermann et al. (1996), might be characteristic of supplementary motor involvement. Van Borsel, Van Lierde, Van Cauwenberge, Guldemont, and Van Orshoven J. Van Borsel et al. / Brain and Language 85 (2003) 185–189 (1998), however, reported the case of a 69-year-old male who suffered a hemorrhage involving the supplementary motor region and who did not conform to pattern outlined by Ackermann and colleagues. In their patient the occurrence of dysfluencies was not restricted to wordinitial position and there were not less dysfluencies during sentence repetition and while reading aloud than in other modalities. To what extent the pattern of a greater dysfluency in more propositional speech observed in our patient is typical of thalamic involvement and whether or not it is justified to speak of a distinct clinical entity of thalamic stuttering must become clear from additional cases of stuttering resulting from thalamic lesions. Some caution is in order, however. It must be reminded that the thalamus serves as a relay station for many pathways, connecting brainstem, subcortical and spinal nuclei with the cerebral cortex. Therefore, a thalamic lesion may disrupt functions classically attributed to other centers of the nervous system. Dysfunction remote from a known lesion, or diaschisis as it was called by Von Monakow (1905), was also encountered in the present case. Perfusion-SPECT revealed a large hypoperfused area in the left frontal region, even at a late stage of recovery. It can therefore not be excluded that the clinical picture, at least in part, is the consequence of left frontal dysfunction. Perhaps that the higher degree of stuttering associated with propositional speech was a reflection of the executive dysfunction. In conclusion, neurogenic stuttering following thalamic lesions should be considered as the consequence of disruption of cerebral networks involving cortical as well as subcortical structures, in which the thalamus serves as an important relay structure. In how far lesions of different parts of this network can yield different dysfluency patterns and in how far a thalamic involvement gives rise to a distinct dysfluency pattern remains to be determined. At any rate, it is evident that neurogenic stuttering is not a uniform disorder. 189 References Abe, K., Yokomay, R., & Yorifuji, S. (1993). Repetitive speech disorder resulting from infarcts in the paramedian thalami and midbrain. Journal of Neurology, Neurosurgery, and Psychiatry, 56, 1024–1026. Ackermann, H., Hertrich, I., Ziegler, W., Bitzer, M., & Bien, S. (1996). Acquired dysfluencies following infarction of the left mesiofrontal cortex. Aphasiology, 10, 409–417. Andy, O. J., & Bhatnagar, S. C. (1991). Thalamic-induced stuttering (surgical observations). Journal of Speech and Hearing Research, 34, 796–800. Andy, O. J., & Bhatnagar, S. C. (1992). Stuttering acquired from subcortical pathologies and its alleviation from thalamic perturbation. Brain and Language, 42, 385–401. Canter, G. J. (1971). Observations on neurogenic stuttering: A contribution to differential diagnosis. British Journal of Disorders of Communication, 6, 139–143. Deelman, B. G., Koning-Haanstra, M., Liebrand, W. B. G., & Van De Burg, W. (1981). S.A.N. Test, een afasietest voor auditief taalbegrip en mondeling taalgebruik. Lisse: Swets & Zeitlinger. Graetz, P., De Bleser, R., & Willmes, K. (1992). De Akense afasietest. Nederlandstalige versie. Lisse: Swets & Zeitlinger. Hertrich, I., Ackermann, H., Ziegler, W., & Kaschel, R. (1993). Speech iterations in Parkinsonism: A case study. Aphasiology, 7, 395–406. International Phonetic Association (1974). The principles of the international phonetic association. London: IPA. Johnson, W., & Associates (1959). The onset of stuttering: Research findings and implications. Minneapolis: University of Minnesota Press. Koller, W. C. (1983). Dysfluency (stuttering) in extrapyramidal disease. Archives of Neurology, 40, 175–177. Ojemann, G. A., & Ward, A. A. (1971). Speech representation in ventrolateral thalamus. Brain, 4, 669–680. Ringo, C. C., & Dietrich, S. (1995). Neurogenic stuttering: An analysis and critique. Journal of Medical Speech-Language pathology, 3, 111–122. Schaltenbrand, G. (1975). The effect of speech and language of stereotactical stimulation in thalamus and corpus callosum. Brain and Language, 2, 70–77. Van Borsel, J. (1997). Neurogenic stuttering: A review. Journal of Clinical Speech and Language Studies, 7, 16–33. Van Borsel, J., Van Lierde, K., Van Cauwenberge, P., Guldemont, I., & Van Orshoven, M. (1998). Severe acquired stuttering following injury of the left supplementary motor region: A case report. Journal of Fluency Disorders, 23, 49–58. Van Dongen, H. R., Van Harskamp, F., & Luteijn, F. (1976). Tokentest. Handleiding. Nijmegen: Berkhout. Von Monakow, C. (1905). Gehirnpathologie. Vienna: H€ older.