J Neurol (1993) 240 : 216-218

Journal of

Neurology

© Springer-Verlag 1993

Insula and aphasia
Jeffrey Shuren*
Department of Neurology, New England Medical Center, Boston, Massachusetts, USA
Received: 30 January 1992 / Received in revised form: 1 June 1992 / Accepted: 7 July 1992

Abstract. A 59-year-old w o m a n d e v e l o p e d i m p a i r e d
speech i n i t i a t i o n as the result of a left a n t e r i o r i n s u l a r
infarct. F i n d i n g s f r o m this case suggest that d o m i n a n t
h e m i s p h e r e a n t e r i o r i n s u l a r lesions i m p a i r the speech
i n i t i a t i o n loop.

Key words: I n s u l a - N o n f l u e n t a p h a s i a - I n i t i a t i o n

Introduction
Studies of B r o c a ' s a p h a s i a reveal that m a n y p a t i e n t s
have d a m a g e to the i n s u l a [6, 9]. O t h e r investigators
f r o m W e r n i c k e [10] to date have suggested that i n s u l a r
lesions p r o d u c e a c o n d u c t i o n aphasia, n o l a n g u a g e dist u r b a n c e , or according to M a r i e [5], a n a r t h r i a . H o w ever, b e c a u s e isolated i n s u l a r lesions with good docum e n t a t i o n are rarely r e p o r t e d , the role of the i n s u l a in
l a n g u a g e has n o t b e e n clearly defined. F u r t h e r m o r e ,
analysis of i n s u l a r l a n g u a g e f u n c t i o n in the past has n o t
t a k e n into full a c c o u n t the architectonics of this region.
T h e following case serves to illustrate the p r e s e n c e of
speech i m p a i r m e n t caused b y a n infarct restricted to the
a n t e r i o r insula.

Case report
A 59-year-old, right-handed black woman with a history of childhood rheumatic fever, hypertension, and cigarette use awoke one
morning in July 1990 with difficulty in speaking. On admission her
blood pressure was 160/96 mmHg with a normal cervical vascular
and cardiac examination. She was fully oriented and attentive but
demonstrated difficulty in initiating speech and pronouncing
words. She displayed aberrant linguistic prosody with occasional
phonemic and rare semantic paraphasias in speech. She made similar paraphasic errors in reading. She displayed difficulty with the
use of function words in her writing. Her verbal comprehension
was impaired on sentences whose meaning was dependent on syntax. She had trouble naming low-frequency words and parts of objects. Repetition was poor. However, during the interview when
* Present address and address for correspondence: Department of
Neurology, University of Florida College of Medicine, Box J-236,
JHMHC, Gainesville, FL 32610, USA

she became angry at a family member, speech fluency, initiation,
pronunciation, and prosody were almost normal. Praxis, memory,
constructions, right-left orientation, finger naming, geography,
and executive function testing were normal. Her cranial nerves
were normal except for slight right lower facial weakness, and her
motor system was normal. Reflexes were brisker in the right lower
extremity than in the left with a right extensor plantar response.
Sensory testing, coordination, and gait were normal.
An admission computed tomographic scan of her head was
normal. A left carotid angiogram showed slight atherosclerotic
changes of the left cavernous internal carotid artery and M1 segment. An echocardiogram showed slight left ventricular hypertrophy, 3+ aortic insuffiency; a bubble study showed no evidence
of a right-to-left shunt. Holter monitoring was normal. MRI of the
head obtained 7 days after admission showed a small infarct deep
in the left anterior insula with pressure on the extreme capsule and
claustrum and bilateral putaminal lacunar infarcts (Fig. la). We
concluded that she had a small vessel infarct or branch occlusion
involving insular branches of the superior division of the left
middle cerebral artery.
Over the following week all her language deficits improved.
Speech initiation, however, remained affected, with mild difficulty
in speech pronunciation and prosody. The Western Aphasia Battery [4] was administered at this time. Subtest scores included spontaneous speech = 19.0 (20), fluency = 9.0 (10), comprehension
= 10.0 (10), repetition = 7.8 (10), naming = 8.5 (10). (Normal
scores on the latter four subtests are 10, 9.9, 9.8, and 9.5, respectively). Her score on the repetition subtest was the result of failure
to repeat the last test sentence. The remainder of her performance
was consistent with preserved repetition. On the Naming subtest
she was able to name all of the 20 objects. Her score reflects her
poor performance on the word fluency subsection. She produced
the names of only 5 animals in lmin. Her aphasia quotient was
90.6 (100, normal score = 98.4), reading and writing was 15.3 (20).
She made no paralexic errors on this subsection. Her performance
was felt to be consistent with her pre-stroke baseline. Praxis was
9.7 (10), and construction was 6.4 (10). The last score was secondary to a low performance on the calculation section. She had always had difficulty with mathematics, and we felt her performance
reflected her pre-stroke baseline. Overall, other than impaired
speech intiation and subsequently impaired verbal fluency, this patient demonstrated no language disturbance. However, though she
produced grammatically correct sentences, an adequate evaluation
of her syntactic processing was not performed. Therefore, this case
cannot shed light on the insula's role, if any, in syntax.
She was discharged to her home with instructions to take one
aspirin a day. A follow-up neurologic examination 1 month later
revealed she continued to have difficulty with speech initiation
without evidence of any language disturbance. Speech pronunciation and prosody were normal. MRI of the head performed 5
months after her stroke demonstrated a deep left anteroinferior

217

Fig. 1. a MRI scan of the head performed 7 days after the patient's
stroke. T2-weighted images demonstrate areas of high signal intensity in the left anterior insula and right putamen consistent with
infarcts. Bar: 5 cm. b MRI scan of the head performed 5 months
after the patient's stroke. Tl-weighted images show a low signal
intensity in the left anterior insula. T2-weighted coronal images
demonstrate an area of high signal intensity in the left anteroinferior insula. Bar: 5 cm

insular infarct (Fig. lb) unchanged from the previous study. Her
difficulty with speech initiation remained unchanged, as well. She
died unexpectedly 7 months after her stroke from unrelated
causes.

Discussion

The aphasia of this patient was characterized by difficulty
in speech fluency, initiation, pronunciation, linguistic
prosody, comprehension of syntax-dependent sentence
meaning, naming, and repetition, which dramatically
improved over the following days, leaving her with a
relatively isolated deficit in speech initiation. In addition
to the left anterior insular infarct, MR! showed small
bilateral putaminal lacunes. Although basal ganglia
infarcts can produce aphasia, they tend to involve the
anterior and superior aspects of the putamen, often in
conjunction with damage of the anterior limb of the
internal capsule [2]. They also tend to be larger in size
than lacunar infarcts. Alexander et al. [1] have argued
that isolated putaminal lesions produce word-finding
difficulty or no language disturbance at all. The consensus, therefore, is that whereas putaminal injury may or
may not result in language disturbances, lacunar infarctions of the putamen do not. In addition, based on the
patient's initial examination and MRI findings of a new
infarct in the insula, the lacunar infarcts are old. Therefore, the putaminal lacunes in this patient are likely to
be unrelated to her aphasia, as they do not explain her
clinical findings. It is also unlikely that pressure on the
extreme capsule and claustrum produced these speech
and language deficits, as difficulty with initiation still
persisted 5 months after the acute event when whatever
edema had been present should have resolved.

The insula's role in language may be better understood by examining its histology and connections. The
insula has been divided into three regions based on cytoarchitectonic studies in Old World monkeys and in man
[7, 8]. In an anterior-ventral to posterior-dorsal gradient,
the insula changes from agranular cortex with three cellular layers to six-layered dysgranular cortex with clustered granule cells in layer 4 and then to granular cortex
with layers 2 and 4 both granular and fully demarcated.
Based on connectivity, the insula may be divided into an
anteroventral (anterior) and a posterodorsal (posterior)
region, as well. Connections of the anterior insula, where
this patient's lesion was located, include agranular orbitofrontal, frontal opercular, prepiriform, ventral temporopolar, anterior cingulate cortex, amygdala, thalamic
ventroposterior medial, medial dorsal, centromedianparafascicular, midline, and reticular nuclei.
The initiation of behavior, including speech, depends
on both appropriate motivation and arousal-activation.
Because the anterior insula has reciprocal projections
with the limbic (cingulate and amygdala) and reticular
(thalamic centromedian-parafascicular and reticular nuclei) systems, as well as with the perisylvian speech
area (frontal opercular region), it would seem to be an
important area in the speech initiation process. In accordance with this hypothesis is the observation that lesions
in other regions of this system (i.e., cingulate, frontal
opercula, thalamus) produce deficits in speech initiation.
During a moment of anger the patient's aphasia nearly completely resolved. Although the mechanism underlying this improvement remains speculative, it is possible
that the limbic-reticular-cortical networks that mediate
emotion may have input into the speech cortex that does
not depend on the insula or at least the portion of insula
damaged in this patient.
Alexander et al. [1] have reported two patients with
infarcts of the insula and adjacent extreme/external capsule region. Both of these patients had a mild fluent
aphasia with word-finding difficulty, rare phonemic
paraphasias, and agraphia. The discrepancy between their
patients and this patient may be related to the location
of their patients' lesions. In their paper, a CT scan of one
of these patients shows an infarct involving anterior and
posterior insula that relatively spared the anteroinferior
region, which had been injured in the patient reported

218
here. The aphasia seen in the patients of Alexander et
al. may have been secondary to dysgranular/granular
insular cortex damage or damage to white matter.
The anatomic distinction between the location of
this patient's infarct and those of Alexander's may relate
to the vascular supply involved. The insular gyri are
supplied by branches of the cortical divisions of the
middle cerebral artery as they travel over this region.
The short insular gyri are fed by the superior division,
while the long insular gyri and transverse temporal gyri
receive branches from the inferior division. In some instances the superior division may send branches to the
entire insula with no contribution from the inferior division [3], or the superior division may feed the short and
long gyri while the inferior division supplies the transverse temporal gyri (personal observation).
Although this case provides evidence that the anterior insula should be included in the anatomic and
functional network that mediates speech, further examples of isolated insular lesions will not only help to
increase our confidence that the insula is a speech area,
but may also better define its role.

References
1. Alexander MP, Naeser MA, Palumbo C (1987) Correlations of
subcortical CT lesion sites and aphasia profiles. Brain 110:
961-991
2. Damasio AR, Damasio H, Rizzo M, Varney N, Gersh F
(1982) Aphasia with nonhemorrhagic lesions in the basal ganglia and internal capsule. Arch Neurol 39 : 15-20
3. Kaplan HA (1972) Anatomy and embryology of the arterial
system of the forebrain. In: Vinken P, Bruyn G (eds) Vascular
disease of the nervous system. (Handbook of clinical neurology, part I). North-Holland, Amsterdam, pp 1-23
4. Kertesz A (1982) The Western Aphasia Battery. Grune &
Stratton, Orlando
5. Marie P (1906) Revision de la question de l'aphasie: la
troisieme circonvolution frontale gauchene joue aucun role
special dans la fonction du language. Semin Med 21 : 241-247
6. Mohr JP, Pessin MS, Finkelstein S, Funkenstein HH, Duncan
GW, Davis KR (1978) Br0ca aphasia: pathologic and clinical.
Neurology 28: 311-324
7. Mufson t~J, Mesulam MM (1982) Insula of the Old World
monkey I, II, and III. J Comp Neurol 212 : 1-52
8. Mufson E J, Mesulam MM (1985) The insula of Reil in man
and monkey. In: Peter A, Jones EG (eds) Cerebral cortex, vol
4. Plenum Press, New York, pp 179-226
9. Vignolo LA, Mazzocchi F (1979) Localization of lesions in
aphasia: clinical-CT scan correlations in stroke patients. Cortex 15 : 627-654
10. Wernicke C (1874) Der aphasische Symptomencomplex.Cohn &
Weigert, Breslau