Progressive Non-Fluent Aphasia in a Bilingual Subject: Relative Preservation of “Mother Tongue”

To the Editor: Although bilingualism is common globally, cases of progressive aphasia due to frontotemporal lobar degeneration have rarely been reported. These cases are of potential interest for understanding the neuropsychological substrates of language.

We present a bilingual patient with progressive nonfluent aphasia to illustrate the differential effect of disease on linguistic output.

The study of the differential effects of brain injury and recovery in bilingual persons has a long history, with two “rules” or “laws” being formulated in the 19th century, by Ribot 1883 and Pitres, 1895 suggesting, respectively, that the first learned language or the most-used language is best preserved or recovers most after the development of aphasia.¹ Empirical studies confirm that bilingual aphasic individuals may have different severity or type of involvement in their two languages.² The question as to what effect a neurodegenerative process, as opposed to the more common cerebrovascular cause for aphasia, might have on the dissolution of language in bilingual patients remains an open one, with few previous case reports.^3,4

Case Report

A 78-year-old, right-handed woman presented to the Cognitive Function Clinic with her son, with whom she lived. He gave a 2-year history of progressive decline in his mother's speech output, initially affecting the names of people, but progressing to involve objects, and eventually becoming largely restricted to “Yes/No” answers. Despite this linguistic decline, her son's impression was that his mother understood the spoken word and could remember past events. Moreover, she was still able to attend to all basic and instrumental activities of daily living, including preparing her son's meals. Her past medical history was unremarkable, and there was no family history of cognitive disorder.

The patient was bilingual. Welsh was her mother tongue, but English had been the usual language of her adult household, as her husband was monoglot English, and this was her son's mother tongue, although he was schooled in Welsh. At consultation, it was evident that the patient talked to her son in Welsh, and he confirmed that his mother had increasingly used Welsh rather than English during the 2 years of decline in speech output.

Neurological examination was normal. Because of the patient's marked nonfluent aphasia, only limited neurolinguistic testing could be undertaken. She was unable to complete the MMSE. Nevertheless, she was able to perform certain elements of the Addenbrooke's Cognitive Examination–Revised,⁵ administered in English, suggesting relatively preserved comprehension. On object-naming, some elements could be named (e.g., pencil, watch, hippopotamus, crown), but superordinate categories were used for others (e.g., “animal” for kangaroo, camel, crocodile; “bird” for penguin). However, with phonemic cueing, these names could be accessed. Despite her anomia, reading of irregular words was normal. She was also unimpaired on tests of visuospatial ability (dot-counting, fragmented letters), and also on the Clock-Drawing Test.

MR brain-imaging showed evident left insula and temporal lobe, mostly anterior, atrophy, with no other focal abnormality. The clinical, neuropsychological, and neuroradiological findings were consistent with a diagnosis of progressive nonfluent aphasia (PNFA).⁶

Discussion

Reports of primary progressive aphasia (PPA) in bilingual subjects are few. The report by Filley et al.³ was of a highly educated woman with a fluent aphasia, with only mildly impaired speech output, conforming to the logopenic progressive aphasia variant of PPA,⁶ and, hence, of dissimilar phenotype to the current case. She was noted to have comparable impairments in her two languages. Conversation, repetition, and naming were reported to be worse in her mother tongue, but these changes were not sufficient to support a definitive conclusion.³ The case of Machado et al.⁴ is more similar to the current case, having the PNFA phenotype (nonfluent, agrammatic, anomic).⁶ In this patient, the loss was overwhelmingly greater in the second learned language, which was infrequently used at the time of disease onset. The authors suggested that it was not the acquisition order of languages, but how recently the language was used that determined the pattern of dissolution.⁴

The current case would seem to contradict this assertion (albeit the fact that quantitative formal analyses could not be undertaken), since the less-frequently-used, but first-learned, language seemed better preserved. It may be that in PPA, as in stroke-related aphasia, Ribot's and Pitres' laws are “honored as much by being broken as by being followed.”⁷ The current findings would be in keeping with a contemporary understanding of bilingualism, in which the neural representation of a second language (L2) converges with the representation of the first learned language (L1).⁸ Focal lobar atrophy might interrupt the neural network responsible for simultaneous activation of L2 (English, in this case) and inhibition of L1 (Welsh).