Journal of the Neurological Sciences 360 (2016) 100–101

Contents lists available at ScienceDirect

Journal of the Neurological Sciences
journal homepage: www.elsevier.com/locate/jns

Letter to the Editor
Isolated cerebellar-type hemiataxia in a
thalamic infarction
Keywords:
Isolated hemiataxia
Thalamic infarction
Ventrolateral nucleus
Rubral tremor

Dear Editor,
Dejerine–Roussy syndrome presenting with hemiataxia due to loss of
proprioception is a classic symptom of a thalamic stroke. In contrast, it
has been reported that cerebellar ataxia, which was not directly related
to the loss of proprioception, might occur due to a thalamic lesion. Recent
studies using modern imaging demonstrated that a cerebellar-type
hemiataxia may occur in a thalamic infarction, and the ventral lateral
(VL) nucleus, the destination of the dentato-rubro-thalamic tract, has
been postulated as the responsible lesion [2,7,10]. However, most of the reported cases showed either sensory disturbance or hemiparesis, or both, in
addition to hemiataxia [2,7,10]. In such cases, the ataxia may be explained
by concomitant sensory deficits, or as a partial symptom of the so-called
ataxic hemiparesis [3]. Here we report two patients with isolated
cerebellar-type hemiataxia as the sole manifestation of the VL infarction.
Patient 1: A 77-year-old woman with a history of hypertension and
smoking visited us on the next day of sudden-onset clumsiness of the
right hand and gait disturbance, and was emergently admitted to our
hospital. Neurological examinations on admission revealed limb ataxia
on the right side, i.e. dysdiadochokinesis in the hand pronation–supination test and dysmetria with terminal oscillation in finger-to-nose and
heel-to-knee tests, which did not change with eye closure. Hemiparesis
was not observed, reflexes were symmetrical with negative Babinski's
sign, and sensation was normal for all modalities (i.e. pain, thermal,
touch, vibration and position senses; same for Case 2). Brain MRI revealed a lacunar infarction in the lateral part of the left thalamus corresponding to the VL nucleus (Fig. 1, upper). We treated her with ordinary
antiplatelet therapy. Her hemiataxia gradually improved and disappeared in 2 weeks.
Patient 2: A 73-year-old woman with a history of hypertension suddenly noticed clumsiness in her right extremities upon awakening. She
visited us next day, and was emergently admitted. On admission, neurological examinations revealed mild right dysmetria and decomposition
in finger-to-nose and heel-to-knee tests, which did not change with
eye closure. Motor paresis was not observed, reflexes were symmetrical
with negative Babinski's sign, and sensation was normal for all modalities. Brain MRI revealed a small lacunar infarction in the lateral part of
the left thalamus corresponding to the VL nucleus (Fig. 1, lower). She

http://dx.doi.org/10.1016/j.jns.2015.11.051
0022-510X/© 2015 Elsevier B.V. All rights reserved.

received ordinary antiplatelet therapy. Her hemiataxia gradually improved and disappeared in 4 weeks.
The present cases showed a pure cerebellar-type hemiataxia without hemiparesis or sensory deficits. A few authors previously reported
that ataxia persisted after the hemiparesis and hypesthesia subsided
[2,7], although isolated cerebellar ataxia from the onset following a thalamic infarction has been described in only one case report written in
French [1]. Ours seems to be the second report. A few authors reported
that an isolated hemiataxia may occur following capsular infarction [4,
6]. However, some of the figures presented in these reports may suggest
that the lesions were likely to be localized within thalamus, rather than
in the internal capsule. In both of our patients, the lesions were localized
within thalamus, not involving the internal capsule (Fig. 1). Furthermore, comparison with human anatomical charts [9] revealed that
they exactly correspond to the ventral lateral posterior nucleus (VLp),
the posterior part of the VL nucleus, which is the major relay nucleus
for cerebello-thalamo-cortical pathways [9]. This finding supports our
hypothesis that the isolated ataxia was caused by a lesion at the VL nucleus by disrupting the cerebellar efferent pathway.
Damage to the dentato-rubro-thalamic tract is usually considered to
be responsible for the so-called rubral tremor. However, violent tremors
were not observed in our patients, nor in other reported patients with
VL infarction. Rubral-type tremors have been reported in patients with
thalamic lesions sparing the VL nucleus [8]. This interesting difference
could be explained if the VL nucleus is actually the generator of the
tremors [8], and the loss of input to this nucleus would cause the violent
tremors: whereas the VL lesion itself does not cause tremors but only
cerebellar ataxia. This hypothesis would be supported by the fact that
stereotaxic surgery on the ventral intermediate nucleus, which is another term for VLp, alleviates the tremors. It is also reported that this surgery may cause transient cerebellar ataxia [5].
When we find a patient presenting with isolated cerebellar type
hemiataxia, the VL nucleus should be considered as a candidate location
for the responsible lesion.
Conflict of interest
No conflict of interest.
References
[1] A. Augier, Y. Ould Hmeidi, F. Neves, et al., Isolated cerebellar syndrome due to ventrolateral thalamic nucleus infarction, J. Radiol. 91 (2010) 732–734.
[2] J. Boiten, J. Lodder, Ataxic hemiparesis following thalamic infarction, Stroke 21
(1990) 339–340.
[3] C.M. Fisher, Ataxic hemiparesis. A pathologic study, Arch. Neurol. 35 (1978)
126–128.
[4] V. Kapina, R. Sztajzel, I. Momjian-Mayor, Isolated hemiataxia of the cerebellar type
after a small internal capsular infarct, Cerebrovasc. Dis. 25 (2008) 594–596.
[5] T.E. Kimber, B.P. Brophy, P.D. Thompson, Ataxic arm movements after thalamotomy
for Parkinsonian tremor, J. Neurol. Neurosurg. Psychiatry 74 (2003) 258–259.
[6] Luijckx, Isolated hemiataxia after supratentorial brain infarction, J. Neurol.
Neurosurg. Psychiatry 57 (1994) 742–744.
[7] T.P. Melo, J. Bogousslavsky, T. Moulin, et al., Thalamic ataxia, J. Neurol. 239 (1992)
331–337.

Letter to the Editor

101

Fig. 1. Diffusion-weighted images (DWI) and fluid-attenuated inversion recovery (FLAIR) images of the two patients are presented. In particular, the DWI of patient 1 and the FLAIR image
of patient 2 (placed between DWI and FLAIR) clearly depict the boundaries of each structure, which have been indicated by dotted lines. Such localizations document that the lesions (highsignal spots in both procedures) are confined within thalamus, and not involving the internal capsule. Furthermore, comparison with a chart (reproduced with permission from Morel,
et al. [9]) suggests that the lesions accurately correspond to the VLp nucleus (black arrow).

[8] H. Miwa, K. Hatori, T. Kondo, et al., Thalamic tremor: case reports and implications
of the tremor-generating mechanism, Neurology 46 (1996) 75–79.
[9] A. Morel, M. Magnin, D. Jeanmonod, Multiarchitectonic and stereotactic atlas of the
human thalamus, J. Comp. Neurol. 387 (1997) 588–630.
[10] D.H. Solomon, R.J. Barohn, C. Bazan, et al., The thalamic ataxia syndrome, Neurology
44 (1994) 810–814.

Keiichi Hokkoku, Kiyoshi Matsukura, Junpei Yamamoto, Midori
Kuwabara, Takashi Chiba, Yuki Hatanaka, Masahiro Sonoo⁎
Department of Neurology, Teikyo University School of Medicine, Kaga 2-111, Itabashi-ku, Tokyo 1738605, Japan
E-mail addresses: k1-hokkoku@hotmail.co.jp (K. Hokkoku),
matsukura.kiyoshi@gmail.com (K. Matsukura), junpei.teikyo@
gmail.com (J. Yamamoto), doricho@gmail.com (M. Kuwabara), tk-chiba@med.teikyo-u.ac.jp (T. Chiba), y-hata@med.teikyo-u.ac.jp (Y.
Hatanaka), sonoom@med.teikyo-u.ac.jp (M. Sonoo).
⁎Corresponding author.
20 October 2015