Foreign accent syndrome (FAS) is a rare speech disorder characterised by the emergence of a new accent, perceived by listeners as foreign. FAS has usually been described following focal brain insults, such as stroke. We describe the unusual case of a woman presenting with FAS as the earliest symptom of progressive degenerative brain disease. At presentation, she showed no language or other cognitive impairment, and functional and structural brain imaging were normal. Follow-up 1 year later revealed the emergence of mild expressive language problems. Repeat functional neuroimaging showed mild hypoperfusion of the perisylvian speech area of the left hemisphere, and structural imaging showed mild left perisylvian atrophy. We interpret the case as an unusual presentation of primary progressive non-fluent aphasia. The case provides further evidence of the variable and circumscribed nature of the clinical presentation of focal cerebral degeneration.
Statistics from Altmetric.com
Foreign accent syndrome (FAS) is an intriguing disorder which has been recognised for almost a century yet continues to be poorly understood. The emergence of the “foreign accent” is most commonly associated with focal brain damage caused by stroke,1 3 although it has also been reported in association with multiple sclerosis4 and head trauma.5 Patients with FAS are often aphasic or FAS may emerge in the course of recovery from a non-fluent agrammatic aphasia.1 2 Nevertheless, most patients with aphasic do not sound like foreign speakers, and FAS can occur in isolation in the absence of aphasia. There is debate regarding the critical locus of pathology. The left hemisphere, particularly Broca’s area, has been implicated by some authors,1 2 although others suggest that subcortical structures,6 such as the basal ganglia or left corona radiata, are critical.3 7 Some cases of FAS have been attributed to non-organic factors,8 highlighting the importance of clinical recognition and understanding of the underlying substrate for patient management.
To our knowledge, FAS has never been reported as an early sign of degenerative brain disease. Of particular interest in this regard is the syndrome of primary progressive non-fluent aphasia,9 10 which shares some similarities to Broca’s aphasia.
We describe a 12 month longitudinal investigation of a patient presenting with a 3 year history of progressive, circumscribed FAS in the absence of abnormalities detectable by neuroimaging.
Clinical history, neurology and investigations
A 64-year-old right-handed Italian woman, with 8 years of education, came to our attention complaining of a progressive change in her native accent. Her family members and an accompanying friend reported that she sounded Spanish. They had noticed the accent change evolving gradually over about 3 years. Her personal history revealed “no source” for her foreign accent. She was born in a village of the Marche region. She spoke only Italian, had never travelled overseas nor studied any foreign language at school. She had worked from home as a seamstress and had never been in contact with Spanish speaking persons. There was no history of behavioural change or cognitive impairment. Her ability to sing was reported to be unchanged. Family history and past medical history were unremarkable. She did not drink alcohol and had never smoked. The only illness of note was a mild reactive depression following the death of her husband 3 years before although this did not require antidepressant medication. She lived alone and was totally independent in activities of daily living.
On examination, she looked healthy, was well kempt and had a pleasant, sociable manner. Her spontaneous speech was well articulated. Her accent sounded Spanish. Neither lexical–semantic nor syntactic errors were noted. No features of speech apraxia were detected. No difficulties were demonstrated in her understanding of spoken language. Neurological examination was entirely normal.
Assessment of language, using the Milan Standard Language Scale,11 the Pyramid and Palm Tree Test of semantic association,12 the Token Test of verbal comprehension13 and category and letter fluency (table 1)14 confirmed the absence of aphasia. Spontaneous speech was well articulated, grammatically correct and without phonological or semantic errors. There was, however, a slightly prolonged pause between words. Fragments of the patient’s conversation were recorded on tape and presented to 10 native Italians subjects who were asked to judge whether the speaker was Italian and, if not, the nature of the speaker’s accent. All 10 subjects reported the accent to be foreign: 7/10 gave the response “Spanish”, 2/10 “South American” and 1/10 “strange, Spanish-like”. It was not possible to perform a formal acoustic analysis of the patient’s speech, making it difficult to specify fully the basis of the accent change. Nevertheless, her way of linking one word with the next, together with alterations in vowel sounds (eg, Italian “il” resembling Spanish “el”) and consonants (“t” resembling “d”), contributed to the “foreign” impression. There were no obvious errors in stress application. She performed within the normal range on a variety of non-language tasks (table 1).
Structural and functional brain neuroimaging, using MRI and single photon emission computed tomography, was normal.
We were able to re-evaluate the patient after an interval of approximately 12 months. She complained of the emergence of linguistic difficulties. On clinical examination, her spontaneous speech was more hesitant, with occasional phonological paraphasias. Her comprehension appeared normal. She continued to sound Spanish. Neurological examination was normal.
Her spontaneous speech was mildly agrammatic with sporadic omission of functors, such as articles and prepositions. Some phonological paraphasias were noted, mainly substitutions and deletions.
Formal language testing (table 1) confirmed the presence of mild to moderate impairments in naming, reading, writing and repetition. By contrast, spoken and written comprehension were normal. Qualitative analysis of errors revealed the presence of agrammatism and phonological errors. The language profile was in keeping with a mild non-fluent aphasia.
She performed all non-verbal tasks very well with no significant differences from her baseline scores.
An MRI scan of the brain showed mild left perisylvian atrophy and single photon emission computed tomography scan showed left frontotemporal hypoperfusion (fig 1).
We have documented the case of an Italian patient presenting with a 3 year history of FAS in the absence of cognitive or neuroimaging abnormalities. The patient’s accent was systematically identified as Spanish or Spanish-like, even by people who had no knowledge of her history.
Follow-up 1 year later revealed the emergence of a non-fluent aphasia and neuroimaging findings of atrophy and hypoperfusion in the language areas of the left hemisphere. It is likely that the patient has a degenerative brain disease. The circumscribed language disorder, associated with asymmetric atrophy of the left perisylvian region, is suggestive of a diagnosis of non-fluent primary progressive aphasia, a variant of frontotemporal lobar degeneration,10 although without longer term follow-up and neuropathological confirmation this can only be presumptive. A focal “language presentation” of Alzheimer’s disease is an alternative possibility, although the absence of other cognitive deficits, particularly of memory and the asymmetric distribution of atrophy, make this diagnosis less likely. The striking and novel feature of the case is that the presenting symptom was of a pure FAS, a rare speech disorder, most commonly arising following stroke and in the context of demonstrable aphasia. FAS has not hitherto been described as an isolated presenting symptom of neurodegenerative disease. The present case highlights the selective and variable presentation of focal brain degeneration. Focal degenerative disease provides a natural model for increasing our understanding of the neuropsychological functions and their neural substrate.
FAS has been interpreted by some as a form of speech apraxia,22 based on the fact that patients show a “failure to produce smooth and fully integrated speech movements”. Speech characteristics may include alterations in vowel and consonant sounds, as noted in the present case. However, it is worthy of note that our patient did not show conventional features of bucco-linguo-facial apraxia. Although speech apraxia and oral apraxia can dissociate,23 most tasks used to elicit bucco-linguo-facial apraxia involve pharyngeal and laryngeal muscles that are also involved in speech production. This suggests that any putative alterations in the motor regulatory aspects of the patient’s speech production must indeed be circumscribed and subtle. It is of interest that the patient’s accent was identified as Spanish or Spanish-like, a language related to Italian. Although a wide variety of accent changes have been recorded in the literature, they frequently bear some relationship to the native language (eg, English or Australian accent in an American speaker; Korean accent in Japanese speaker), suggesting the presence of altered yet recognisable sounds. In the case of our patient, FAS preceded the onset of more obvious phonological disturbances of progressive non-fluent aphasia.
There continues to be debate as to the anatomical site critical for FAS. Some reports implicate the premotor and motor cortex,1 2 and others the basal ganglia.3 6 7 In the present case, the available data would seem to favour a cortical basis. The patient showed no neurological signs, and the evolution of disease revealed increased cortical symptomatology, consistent with findings on neuroimaging.
An additional complication in the interpretation of FAS is that some cases appear to have a non-organic basis.8 In the present case, FAS emerged during a period of mourning when the patient showed mild depressive symptoms. The normal clinical and imaging findings might have invited an initial interpretation of functional disorder. Yet, the FAS persisted despite resolution of her mood disturbance and proved to be the harbinger of a more widespread language disorder associated with degenerative change within the perisylvian language area. The case reinforces the need for caution in assuming a functional substrate on the grounds of normal imaging findings and encourages clinicians to assess patients longitudinally before making definitive clinical judgements. Such longitudinal assessment is not only relevant for the clinical management of the patient, but provides the potential to shed further light on the intriguing condition of FAS.
Competing interests: None.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.