European Journal of Neurology 2003, 10: 445–448

SHORT COMMUNICATION

Subcortical aphasia: three different language disorder syndromes?

D. C. Kuljic-Obradovic
Sveti Sava Hospital, Nemanjina, Belgrade, Yugoslavia

Keywords:
aphasia, paraventricular
white matter, striato-
capsular region, subcorti-
cal, thalamus
Received 8 July 2002
Accepted 3 March 2003
The study analyses clinical presentation of language functions of 32 patients with
subcortical aphasia induced by stroke. The patients have been divided into three groups
according to neuroanatomic localization of the lesion, defined by CT and MRI
examination (striato-capsular aphasia, aphasia associated with white matter paraven-
tricular lesions and thalamic aphasia). The following batteries and tests were used: the
neurologic examination, CT scan, MRI, Doppler ultrasound, Mini Mental State
Examination, Boston Diagnostic Aphasia Examination (BDAE), Boston Naming Test
(BNT), Token Test and Verbal Fluency Test. Clinical presentation of subcortical
aphasias is characterized with preserved repetition, however, some groups differ by
certain specific features of language impairment. Striato-capsular aphasia and aphasia
associated with white matter paraventricular lesions are characterized with lack of
speech fluency, occurrence of literary paraphasias, mainly preserved comprehension
and naming. Thalamic aphasia, however, is characterized with fluent output, impaired
comprehension and naming with predominant verbal paraphasias. The specific features
of language impairment suggest that subcortical structures contribute to language
organization. Considering the results of language tests we presume that the most
prominent feature in striato-capsular aphasia is phonetic impairment of language,
opposite to thalamic aphasia where lexical-sematic processing seems to be affected.

group of subcortical aphasias showed certain specific

Introduction
Contradictory opinions about the role of some sub-
cortical structure in language organization, pathogen-
esis, clinical manifestations and classifications of
subcortical aphasia are mostly common in referential
literature. Authors, sometimes even exclude complete
existence of this entity.
Different authors usually highlight the importance of
the lesions in striato-capsular region (head of the
caudate nucleus, putamen, anterior limb of the internal
capsule), thalamus and paraventricular white matter.
Accordingly, subcortical aphasias are classified into
three groups: striato-capsular aphasia, thalamic aphasia
and aphasia associated with white matter paraven-
tricular lesions.
Whilst the majority of authors accentuate variability
of clinical manifestations amongst and within the
groups of subcortical aphasias, others point out that
certain groups are characterized with clearly identified
symptoms (D’Esposito and Alexander, 1995).
Preservation of repetitive speech is widely recognized
as the common feature, suggesting similarity of sub-
cortical and trascortical aphasias (Benson and Ardila,
1996; Kreisler et al., 2000). On the other hand, each
Correspondence: Dragana C. Kuljic-Obradovic, Sveti Sava Hospital,
Nemanjina 2, 11000 Belgrade, Yugoslavia (fax: + 381(11) 642–014;
e-mail: ukisrki@eunet.yu).
disorders of speech and language functions.
Striato-capsular aphasia in spontaneous speech is
characterized with impairment of articulation and pros-
ody. Comprehension is mainly preserved, whilst nomi-
nation is usually affected with occurrence of semantic
paraphrasias (Benson and Ardila, 1996; Demonet, 1997).
Clinical presentation of aphasia caused by lesions of
paraventricular white matter has not been studied in
greater detail, but some studies suggested that produc-
tive components of language are affected primary
(Staudt et al., 2001).
The reduction of spontaneous verbal output, severe
anomia, defective comprehension and occurrence of
verbal paraphrasias are main features of thalamic
aphasia (Benson and Ardila, 1996; Neau and Bogous-
slavsky, 1996; Demonet, 1997; Nadeau and Crosson,
1997).
Materials and methods
Thirty-two aphasic patients, 18 males and 14 females,
age 47–72 years (mean 57.9), 29 predominantly right-
handed and three ambidextrous with left-sided vascular
lesions of subcortical structures admitted in ÔSveti SavaÕ
Hospital for cerebrovascular diseases, were tested 4–
8 days after the onset of the disease. The native lan-
guage for all patients was Serbian. The patients were
divided into three groups: (i) patients with lesion in the
striato-capsular region 15, (ii) patients with lesion of

Ó 2003 EFNS 445

446 D. C. Kuljic-Obradovic
paraventricular white matter eight and (iii) patients
with thalamic lesions nine. Patients have given their
informed consent.
Clinical neurologic examination performed on the
onset of the diseases presented: right hemiplegia (five
patients), mild right hemiparesis (25 patients), right
sited cutaneous sensory loss (two patients), thalamic
pain (one patient) and right sited deep sensory loss
(three patients). The neurological findings were con-
sistent with radiological localization of the lesions,
although three patients with thalamic lesions suffered
from right hemiparesis without sensory loss. Loss of
sensory function may have been so mild and gradual as
to pass unnoticed. Adequate hearing was confirmed by
tuning-fork tests. Mini Mental State Examination was
used for evaluation of cognitive function.
Both, brain CT on Shimadzu X-ray SCT 4800 TF/TE
(Shimadzu Corporation, Kyoto, Japan) sliced 2–5 mm
and MRI on Siemens Magnetom Plus Impact (1 Tesla)
(Siemens Medical Solutions, Erlargen, Germany), sliced
3–6 mm, without contrast, have been performed 1–
4 days after the onset of the disease. The lesions were
estimated visually by two experienced reviewers. Three
patients had haemorrhages (thalamus), one had haem-
orrhagic infarct (thalamus) and 28 had non-hem-
orrhagic infarcts. The striato-capsular lesions were
approximately 7–25 mm, the lesions of paraventricular
white matter 20–35 mm and thalamic lesions 10–35 mm
measured by MRI.
Doppler ultrasound has also been performed, dem-
onstrating normal finding in 28 patients, atheromatous
plaques and 30% stenoses of the left internal carotid
artery in one patient and slower flow of vertebro-basilar
arteries in three patients.
Diagnosis of aphasia was carried out using the fol-
lowing batteries and tests: Boston Diagnostic Aphasia
Examination (BDAE), Boston Naming Test (BNT),
Token Test (Original Version) and Verbal Fluency Test.
Conversational speech was analysed by BDAE rating
scale profile of speech characteristics. Comprehension
was analysed by subtests: word discrimination, body-
part identification, commands, complex ideational ma-
terial, contained in BDAE and token test-original ver-
sion. Repetition was controlled with BDAE subtests:
repetition of words and sentences. Naming was also
analysed with BDAE subtests (Responsive Naming and
Visual Confrontation naming) and Boston Naming
Test (BNT). Speech fluency test based on animal na-
ming and FAS were performed, too. Paraphasias was
also studied with BDAE tests.
The exclusion criteria were: (i) Patients with history
of previous neurological disorders, (ii) patients with
diseases of relevance for the study (e.g. reduced vision,
deafness), (iii) patients with lack of cooperation, (iv)
patients with Mini Mental State Examination score
below 24, (v) patients with carotid occlusion seen on
Doppler ultrasound and (vi) patients with brain atro-
phy, multiple lesions, or other cerebral tissue lesions
than acute stroke, seen on CT/MRI.
Statistics
Statistical analysis of the data was bases on Kruskal–
Wallis chi-square test that was used for non-parametric
variance analysis (ANOVA).
Results
The used batteries and tests provided results to assess
specific functions of language in the studied groups of
subcortical aphasias.
Conversational speech results in patients with striato-
capsular aphasias and aphasia associated with white
matter paraventricular lesions showed a similar profile
of language impairment. Aphasia is characterized with
lack of fluency, short phase length (three to four
words), affected articulation and preserved repetition.
Nomination and comprehension impairments were not
severe.
On the other hand, thalamic aphasia is characterized
with preserved fluency, repetitive speech and grammar,
but with frequent and severe damage of nomination
and comprehension.
Comprehension tests failed to indicate statistically
significant differences amongst the groups. The most
severe aberrations were recorded in the group of tha-
lamic aphasia (Table 1).
Preservation of repetition was illustrated with aver-
age values remaining above the cutoff score in all three
groups.
Naming tests indicated statistically significant differ-
ences amongst the groups. In the group of thalamic
aphasia the impairment of naming was the most severe.
In all three groups high values of SD were noticed
(Table 2).
Speech fluency test based on animal naming failed,
whilst FAS did show statistically significant difference
amongst the groups. FAS test results were lower than in
animal naming test, which is particularly marked in
striato-cortical aphasia: mean value of animal naming
11.93, FAS 5.64, whilst the difference was the least in
thalamic aphasias: 11.3 vs. 5.87.
Paraphasias usually occurred in the group of tha-
lamic aphasia (mean verbal 9.78, neologisms 2.22, lit-
erary 1.78), and less commonly in the group of striato-
capsular aphasias (literary paraphasias 3,6) and the
least common in the group of aphasias with lesion of
paraventricular white matter.
Ó 2003 EFNS European Journal of Neurology 10, 445–448
 Subcortical aphasia 447

Table 1 Comprehension scores of patients with subcortical aphasia

Thalamic Striato-capsular White-matter

aphasia aphasia lesions aphasia

Mean SD Mean SD Mean SD P-value

a
Word discrimination 52.83 23.08 63.27 10.27 54.69 11.95 >0.05
Body-part identificationa 15.00 6.10 17.70 4.00 16.19 3.78 >0.05
Commanda 10.00 4.50 12.40 4.79 11.88 5.00 >0.05
Complex ideational materiala 6.67 3.50 9.13 3.44 7.75 2.05 >0.05
Total answerb 47.00 8.37 48.20 16.35 44.50 15.15 >0.05
a
Boston Diagnostic Aphasia Examination (BDAE); bToken Test (TT).

Table 2 Naming scores of patients with subcortical aphasia

Thalamic Striato-capsular White-matter

aphasia aphasia lesions aphasia

Mean SD Mean SD Mean SD P-value

Responsive naminga 22.12 4.09 24.40 9.66 24.00 9.12 <0.05

Visual naminga 77.89 30.67 96.87 22.68 93.38 7.17 <0.10
Spontan answerb 28.88 8.54 37.33 18.36 36.63 5.06 >0.05
Total answerb 40.25 7.87 45.60 15.01 44.75 1.12 >0.05
a
Boston Diagnostic Aphasia Examination (BDAE); bBoston naming Test (BNT).

Aphasia associated with white matter paraventricular

Discussion
Subcortical aphasia induced by lesions of studied, left-
sided, localizations (striato-capsular region, paraven-
tricular white matter, thalamus) showed unique aphasia
syndrome.
The most consistent common finding was preserva-
tion of repetition in all three groups, coinciding with
reports of other authors (Kreisler et al., 2000; Benson
and Ardila, 1996; Neau and Bogousslavsky, 1996;
Nadeau and Crosson, 1997; Demonet, 1997), suggesting
preservation of Wernicke’s and Broca’s areas and their
links (Ociæ and Stefanova, 1996).
On the other hand, clearly defined characteristics and
differences in severity of affection of analyzed verbal
functions in certain groups of subcortical aphasia were
showed.
Conversational speech profile suggests that striato-
capsular aphasia is characterized with lack of fluency,
usually preserved comprehension, rare occurrence of
paraphasias, mostly literary. As opposed to the above,
thalamic aphasia is characterized with fluent speech,
long sentences, preserved melodic line, articulation
agility and grammar. Comprehension is, however, more
severely affected. Naming is a great deal impaired,
paraphasias are common, usually verbal. The obtained
findings are supported in the reports of many authors
(Benson and Ardila, 1996; Nadeau and Crosson, 1997;
Wallesch et al., 1997), however, in the others they are
opposed (Megas et al., 1992; Neau and Bogousslavsky,
1996).
lesions possesses similar characteristics as striato-cap-
sular aphasia, but with longer phrase length (mean four
words), what is correlated with other study (Staudt
et al., 2001).
Comprehension is the least affected in cases of stri-
ato-capsular aphasias and the most impaired in tha-
lamic aphasia, although we did not find statistically
significant differences amongst the groups. Our findings
are coinciding with reports of many authors (Benson
and Ardila, 1996; Demonet, 1997; Nadeau and Cros-
son, 1997), but opposing to some (Neau and Bogous-
slavsky, 1996).
The most striking difference (statistically significant)
amongst the groups of subcortical aphasia was noted in
naming. It was showed that the naming ability was the
most severely affected in the group of thalamic aphasias
and the least impaired in striato-capsular ones. All
patients with thalamic aphasia have their naming
function affected, especially responsive naming and
visual naming, coinciding with referential studies
(Megas et al., 1992; Benson and Ardila, 1996; Neau and
Bogousslavsky, 1996; Demonet, 1997; Nadeau and
Crosson, 1997; Raymer et al., 1997). Only few authors
failed to identify naming dysfunction in cases of tha-
lamic infarction (Bogousslavsky et al., 1988).
Nevertheless, analysing our results on comprehen-
sion and naming test one can perceive that the values
of SD are high in all groups, resulting from variability
of clinical presentation within the groups. This fact
is correlating with the most of referential reports

Ó 2003 EFNS European Journal of Neurology 10, 445–448

448 D. C. Kuljic-Obradovic
(Benson and Ardila, 1996; Nadeau and Crosson, 1997;
Wallesch et al., 1997). Previous suggests that features
of language disorders should be discussed as symptom
clusters, not as general clinical-anatomical correlates.
Therefore, farther investigation should consider the
question of more precise and define localization of
subcortical lesions associated with speech and lan-
guage disorders.
On fluency tests, lower achievement on FAS test than
in the animal naming test was particularly striking in
the group of striato-capsular aphasias, whilst in the
thalamic aphasia group the difference was the least
prominent, as in other study (Schmand et al., 2000).
Occurrence of paraphasias was most commonly
recorded in the group of thalamic aphasia, and least in
patients with lesions of the paraventricular white mat-
ter. The most literary paraphasias in patients with str-
iato-capsular aphasia were recorded. Our findings do
not correlate with the report of Demonet who found
higher percentage of semantic paraphasias and absence
of literary paraphasias (Demonet, 1997).
Paraphasias were commonly recorded in the group of
thalamic aphasia with verbal paraphasias, as most
common, whilst neologisms and literary paraphasias
were less frequent; as substantiated in numerous papers
(Megas et al., 1992; Benson and Ardila, 1996; Neau and
Bogousslavsky, 1996; Demonet, 1997) and opposed to
other (Kirshner H & Kistler, 1982).
Considering the results mentioned above, one could
presume that the most prominent feature in striate-
capsular aphasia is phonetic disintegration (Taubner
et al., 1999), opposite to thalamic aphasia where lexical-
sematic processing seems to be affected.
Patterns of neural connectivity and clinical observa-
tions testify that cortical regions operate in concert with
subcortical structures which specifically contribute to
language organization (Grabowski et al., 2002). Dif-
ferent parts of subcortical structures are involved in
specific neuronal networks, cortico-subcortical circuits
and complex, feedback loops. The lesions of certain
subcortical structures disrupt cognitive processing in a
definite way, suggesting their important and specific
role in language organization and other cognitive
function. Some authors suppose that many of the def-
icits following subcortical lesions may represent dis-
connection syndromes (Taubner et al., 1999).
A completely different concept is that aphasia
because of acute subcortical stroke can be largely
explained by cortical hypoperfusion (Hillis et al., 2002).
In summary, true language disorders can be observed
in patients with lesions in specific subcortical areas. The
question of the role of those structures on language
organization is still unsolved, but further neuroimaging
studies can provide opportunity to examine and answer
increasingly sophisticated questions about the func-
tional neuroanatomy of language.
We recognize that our study has certain limitation.
Audiometric test have not been performed and tuning
fork tests were not a good substitute.
Acknowledgements
The author thanks Professor M. Dimitrijevic for his
review of statistical analyses and Mrs V. Kostic for her
assistance with preparation of the manuscript.
References
Benson DF, Ardila A (1996). Aphasia, a Clinical Perspective.
Oxford University Press, New York, pp. 166–179.
Bogousslavsky J, Regli F, Uske A (1988). Thalamic infarcts:
clinical syndromes, etiology, and prognosis. Neurology
38:837–848.
D’Esposito M, Alexander MP (1995). Subcortical aphasia:
distinct profiles following left putaminal hemorrhage.
Neurology 45:38–41.
Demonet JF (1997). Subcortical aphasia(s): a controversial
and promising topic. Brain Language 58:410–417.
Grabowski TJ, Anderson SW, Cooper GE (2002). Disorders
of cognitive function. Continuum. Lippincott Williams &
Wilkins, Philadelphia, pp. 29–33.
Hillis AE, Wityk RJ, Barker PB et al. (2002). Subcortical
aphasia and neglect in acute stroke: the role of cortical
hypoperfusion. Brain 125:1094–1104.
Kirshner H, Kistler K (1982). Aphasia after right thalamic
hemorrhage. Arch Neurol 39:667–669.
Kreisler A, Godefroy O, Delmairg C et al. (2000). The
anatomy of aphasia revisited. Neurology 54:1117–1123.
Megas J, van Loon J, Goffin J, Gybels J (1992). Subcortical
aphasia from a thalamic abscess. J Neurol, Neurosurg
Psychiatry 55:319–321.
Nadeau S, Crosson B (1997). Subcortical aphasia. Brain Lang
58:355–402.
Neau JP, Bogousslavsky J (1996). The syndrome of posterior
choroidal artery territory infarction. Ann Neurol 39:779–
788.
Ociæ G, Stefanova E (1996). Neurolingvistièka interpretacija
afazija. Klinie`ka I Eksperimentalna Neurologija 1:91–100.
Raymer AM, Moberg PJ, Crosson B, Nadeau SE, Gonzales-
Rothi LJ (1997). Lexical-semantic deficits in two patients
with dominant thalamic infarction. Neuropsychologia
35:211–219.
Schmand B, de Bie RM, Koning-Haanstra M, de Smet JS,
Speelman JD, van Zomeren AH (2000). Unilateral pallid-
otomy in PD A controlled study of cognitive and behavioral
effects. Neurology 54:1058–1064.
Staudt M, Grodd W, Neimann G (2001). Early left periven-
tricular brain lesions induce right hemispheric organization
of speech. Neurology 57:122–125.
Taubner RW, Raymer AM, Heilman KM (1999). Frontal-
opercular aphasia. Brain Lang 70:240–266.
Wallesch CW, Johansen-Horbach H, Bartels C, Herrmann M
(1997). Mechanisms of and misconceptions about subcor-
tical aphasia. Brain Lang 58:403–409.
Ó 2003 EFNS European Journal of Neurology 10, 445–448