Journal of Neurology, Neurosurgery, and Psychiatry 1990;53:67-71

67

Vertical gaze palsy and selective unilateral
infarction of the rostral interstitial nucleus of the
medial longitudinal fasciculus (riMLF)
J

Bogousslavsky, J Miklossy, F Regli, R Janzer

Case report
A 60 year old housewife who was a smoker had
been investigated for ten years for arterial
hypertension, diabetes mellitus and elevated
blood cholesterol. Three weeks before admission, she experienced transient dysarthria and
dizziness and the day before had been unable to
stand and experienced tingling in the left hand.
One day later, she suffered acute rotatory
vertigo with vomiting, diplopia, dysarthria,
and left-sided weakness and numbness. On
admission, the patient was well-oriented in
time and place but had severe dysarthria. Blood
pressure was 140/85 mm Hg. The visual fields
were normal. Eye movements were normal.
The pupils could not be assessed, because of
previous cataract operations. She had a slight
lower facial paresis on the left. The remainder
of the cranial nerves were normal. She had
a moderate left hemiparesis with decreased
sensation for temperature and pain. The
upper limb showed dysmetria and dysdiadochokinesia. Over 24 hours after admission, the
patient progressively worsened and developed
a complete left hemiplegia, despite intravenous
heparin therapy (30 000 Iu/24 h). Progressive
weakness then developed on the right, and four
days later she had a spastic tetraplegia with
facial diplegia and severe lower cranial nerve
palsy. The clinical picture showed a typical
It is well known that vertical gaze palsy is locked-in syndrome, except for the horizontal
related to mesencephalic dysfunction.'2 More eye movements which were unimpaired when
and pursuit gaze. The
recently, the critical area for the genesis of up- tested for voluntary and
close her eyes on
could
open
patient
and downward saccades in the monkey and in
complete vertical
was
a
there
but
command,
dorsoto
be
located
been
found
has
humans
and
pursuit movefor
voluntary
palsy
gaze
medial to the anterior pole of the red nucleus, in
testing
ments.
Bedside
vertical
optokinetic
an area now called the rostral interstitial
nucleus of the medial longitudinal fasciculus with a drum showed no response, but the
manoeuvre elicited a full
(riMLF)34. Other upper midbrain structures vertical oculocephalic
response. This state
and
downward
upward
the
include
vertical
posterior
gaze
implicated in
four months. The
for
remained
unchanged
commissure, the interstitial nucleus of Cajal
ptosis and
then
palpebral
developed
patient
and the nucleus of Darkschewitsch.5 Classicweakness in the right eye. It became
ally, upgaze palsy has been related to uni- or adduction
bilateral upper midbrain lesions, while a more difficult to communicate with the patient
bilateral lesion is required to give rise to using her preserved eye movements. At the
the
downgaze palsy.56 However, the respective role same time, the tetraplegia became flaccid,
chewing
purposeless
developed
patient
of the riMLF, posterior commissure, nucleus
of Cajal, nucleus of Darkschewitsch, and movements, and her weight dropped by 30 kg.
related structures has remain difficult to assess She died six months after admission.
in humans, in the absence of case reports with
selective involvement of one or the other of Laboratory investigations
Brain CT was checked three times and showed
these structures.
Our report of a patient with combined up- a right occipital, left lenticular and right ponand downgaze palsy in association with an tine infarcts. Bilateral vertebral digital arterial
upper midbrain infarct limited to the riMLF angiography ten days after admission showed
on the right side may provide further insight in progresive stenosis of the basilar artery from
the understanding of the control of vertical the last part of the middle third, with occlusion
of the distal third. Brainstem auditory evoked
gaze.

Abstract
We report a clinico-pathological correlation study in a patient with basilar artery
thrombosis, who developed tetraplegia
and combined up- and downgaze palsy
involving voluntary saccades and
visually-guided movements, but sparing
the oculocephalic responses. At necropsy,
apart from bilateral infarction in the
basis pontis, there was a single unilateral
infarct selectively destroying the rostral
interstitial nucleus of the medial longitudinal fasciculus (riMLF) on the right.
The posterior commissure and its
nucleus, the nucleus of Cajal, the nucleus
of Darkschewitsch and the pontine
tegmentum were spared. We suggest that
the unilateral riMLF lesion may have
disrupted bilateral upgaze excitatory and
inhibitory inputs and unilateral downgaze excitatory inputs. The functional
anatomy of inhibitory and excitatory
vertical gaze circuitry, which remains
speculative, may explain why a unilateral
lesion of the upper midbrain tegmentum
may be sufficient to generate an upgaze
palsy or a combined up- and downgaze
palsy, while an isolated downgaze palsy
requires bilateral lesions.

Department of
Neurology* and
Division of
Neuropathologyt
Centre Hospitalier
Universitaire Vaudois,
Lausanne, Switzerland

Bogousslavsky*
Miklossyt
F Regli*
R Janzert

J

J

Correspondence to:
Dr Bogousslavsky,
Department of Neurology,
CHUV, 1011 Lausanne,
Switzerland.
Received 24 February 1989.
Accepted 6 July 1989

68

Figure 1 Paraffin
sections stainedfor myelin
(Loyez), through the pons
at four different levels (A:
mesencephalic-pontine
junction; B: upper pons;
C and D: midpons),
showing the extent of the
bilateral cystic infarction.
Al, Bl, Cl and Dl are
schematic representations
of the lesion at the same
levels.

. . . .:."

'S°FW*_,x..B

Bogousslavsky, Miklossy, Regli, Janzer

..6

`.k1.

*...S.,
~
~ zE
~ ~, ~

f

Sg

_-

*t fy

|

w...

,:j;
si; 'Sst

J

v

..

:,:

t

°

w-b.,

.}

.N1

we,,
K1

3

x

t

:

1

t

I

':

f

potentials showed no abnormality when per-

formed three and four months after admission.
ECG and standard blood tests were normal.

Neuropathologicalfindings
The brain was fixed in

10%/

formalin for three

weeks. After macroscopical examination, the
brain was cut in 8 mm coronal slices. Blocks
were taken from the left striatum, the left
occipito-temporal, the right occipital region
and from the cerebellum. Ten pm paraffine
sections were stained with hematoxyline-

Vertical gaze palsy and selective unilateral infarction of the rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF)

.s

.-

superior part of the medulla oblongata were
stained alternatively with hematoxyline-eosine
and Loyez stains. Forty gm frozen serial sections from blocks of the thalamus and upper
midbrain on both sides were stained with Oil
red 0 and Sudan black.
The brain weighed 1290 g. There was no sign
of oedema or atrophy. The circle of Willis
showed severe atherosclerosis. The middle
third of the basilar artery was totally occluded
by an organised, recanalised thrombus which
extended into the rostral segment of the artery.
In the examination of the cerebral hemispheres
there was revealed a small old cystic infarct

RiM'vLF
.-

Nk

N
R,

(1-5 cm diameter) in the deep territory of the
left middle cerebral artery destroying the
supero-lateral part of the striatum and involving the anterior-superior part of the internal
capsule, and an old infarct in the territory of the
right posterior cerebral artery destroying the
posterior part of the lingual and fusiform gyri.
The most important lesion was an infarct in
the territory of the basilar artery, the
appearance of which corresponded to the duration of the clinical history. The infarct involved
large parts of the ventral pons, but the tegmentum of the pons was completely spared (fig 1).
The medial longitudinal fasciculus, the
trochlear and abducens nuclei and their fibres,
as well as the paramedian pontine reticular
formation (PPRF) were intact on both sides.
The caudal pons, medulla oblongata and
cerebellum did not show any foci of infarction.
At the meso-diencephalic junction, there was
an infarct at the level of the central mesencephalic grey, and the right rostral interstitial
nucleus of the mnedial longitudinal fasciculus
(riMLF) was damaged (fig 2). The nucleus of
Darkschewitsch, the interstitial nucleus of
Cajal, the nucleus and fibres of the posterior
commissure as well as the nuclei of the pretectal
region were completely spared. On the left side,
the mesencephalic tegmentum was entirely
spared.
A more recent minute focus of infarct was
found in the supero-lateral part of the right
oculomotor nuclear complex, involving the
rostral part of the dorsal, ventral and intermediate columns.

}

..<

69

7\\k

..
4
~ ~~~

~

~~~~~~~~~~~~~~~~~~
4

#.

gpp

..S A~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ g.
..

.

.

j~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....
......

W*Me1

Figure 2 a: Coronal section of the upper midbrain at the level of the posterior
commissure and demonstrating the localisation of the spared left riMLF (dashed line).
Nissl stain. Magnification: 30 x b: Same section on the right side, showing the riMLF
entirely destroyed by the infarct. Arrows point to the inferior limit of the lesion. Same
staining and magnifications as A. c: Coronal section adjacent to B, stained with Sudan
black showing clearly the clearly defined infarct in the right riMLF. Magnification: 30
x riMLF
rostral interstitial nucleus of the medial longitudinalfasciculus; RN = red
nucleus.
.

=

eosine, cresyl violet, luxol fast blue-Van
Gieson, periodic acid-Schiff and Loyez stains.
Forty pm frozen sections were cut from blocks
of the same regions and were stained with Oil
red 0 and Sudan black. Ten pm serial paraffine
sections from the mesencephalon, pons and

Discussion
Our patient had a severe infarct of the ventral
pons due to basilar artery thrombosis. The
pontine tegmentum was completely spared,
which explained the normal horizontal eye
movements. On the other hand, a small focus of
infarction was found in the mid-brain tegmentum on the right, which selectively involved the
riMLF, and corresponded to the vertical gaze
disturbances. A more recent small infarct
involved part of the right oculomotor nucleus
on the right, explaining the palpebral
ptosis and adduction weakness in the right eye,
which developed shortly before death. Thus,
although our patient had other ischaemic
lesions than that in the upper midbrain tegmentum, it is the first case in which vertical
gaze abnormalities can be correlated to selective unilateral riMLF infarction.

70

Bogousslavsky, Miklossy, Regli, Janzer

The syndrome of combined up- and downgaze palsy shown by our patient has usually
been related to bilateral involvement of the
upper midbrain tegmentum.6 Eleven infarct
cases with pathological study have been reported.4 7-16 However, in rare instances,
it appears that a unilateral lesion confirmed
at necropsy at the same level may be sufficient
to generate combined vertical gaze palsy.
Dereux'7 reported a patient with a unilateral
lesion involving the upper midbrain and periaqueductal area. Garcin et al'8 reported the case
of a woman aged 65 with an upper brainstem
infarct that was limited to the left side, who had
a combined vertical gaze paresis, though
"slight upward and downward movements
were still possible". Collier'9 described a
woman aged 49 with up- and downgaze palsy
and a left thalamo-mesencephalic haemorrhage. Molnari0 reported a man aged 72 with
upgaze palsy and "limitation of downgaze"
with a left paramedian thalamo-mesencephalic
infarct. Leigh and Zee2' reported a 38 year old
woman with upgaze palsy and decreased downward saccades velocity and smooth pursuit
with a left thalamic-midbrain haemorrhage.
More recently, Ranalli and Sharpe22 reported a
64 year old man who had a right paramedian
upper brainstem infarct and palsy of upward
saccades, decreased amplitude and velocity
of downward saccades, decreased gain and
amplitude of vertical smooth pursuit and
vestibulo-ocular response, and inverted Bell's
phenomenon on prolonged lid closure. In the
first five cases, the neuropathological study did
not include microscopic details, but the region
of the posterior commissure and riMLF was
involved in all instances. In the sixth case, the
right riMLF, nucleus of Darkschewitsch, and
part of the nucleus of Cajal and ofthe nucleus of
the posterior commissure were involved, but
the posterior commissure was spared. In our
patient, the right riMLF was selectively
involved in the upper midbrain, while the
posterior commissure and its nucleus, the
nucleus of Cajal and the nucleus of Darkschewitsch were completely spared.
Though our study suggests that selective
involvement of riMLF on one side may
generate a combined vertical gaze palsy, the
explanation of this remains unclear. Cases of
unilateral midbrain lesion with upgaze'5 23-30 or
combined up- and downgaze'722 palsy have
been reported, but there is no single case with
pathological verification in which downgaze
palsy was due to a unilateral lesion. In patients
with isolated downgaze palsy and an apparently
unilateral infarction on MRI31 a small contralateral lesion cannot be excluded.
It is possible that inhibitory as well as
excitatory upgaze fibres from the riMLF on
one side may decussate in the posterior commissure and pass through the contralateral
riMLF on their way to the oculomotor nuclear
complex. On the other hand, there is no
experimental or clinical evidence that
downgaze fibres decussate in the posterior
commissure.520 This decussation is probably
lower in the midbrain, close to the oculomotor
nuclear
complex.5 32 The combined

involvement of bilateral upgaze inhibitory
inputs and unilateral downgaze excitatory
inputs from a unilateral riMLF lesion may thus
be sufficient to generate downgaze palsy,
because the uninvolved contralateral downgaze
inputs cannot compensate. Moreover, the
preservation of contralateral downgaze inputs
explain why downgaze was usually only
paretic, in contrast to complete upgaze palsy, in
the previously reported patients with a combined vertical gaze palsy and a unilateral lesion.
On the other hand, this would also explain
why downgaze palsy from a unilateral riMLF
involvement may not occur without associated
upgaze palsy, because bilateral upgaze
excitatory and inhibitory inputs are involved
concomitantly within the riMLF:
The sparing of the vertical oculocephalic
response in our patient suggests that the
riMLF does not interfere with this reflex,
which may be mediated at this level by the
nucleus of Cajal and related structures.2230
1 Henoch. Tuberculose der Corpora quadrigemina. Berlin Klin
Wschr 1864;1:125-7.
2 Parinaud H. Paralyse des mouvements associes des yeux.
Arch Neurol (Paris) 1883;5:145-72.
3 Buttner-Ennever JA, Buttner U. A cell group associated
with vertical eye movements in the rostral mesencephalic
reticular formation of the monkey. Brain Res 1978;151:
31-47.
4 Buttner-Ennever JA, Buttner U, Cohen B, Baumgartner G.
Vertical gaze paralysis and the rostral interstitial nucleus
of the medial longitudinal fasciculus. Brain 1982;105:
125-49.
5 Pierrot-Deseilligny C. Circuits oculomoteurs centraux. Rev
Neurol 1985;141:349-70.
6 Bogousslavsky J, Meienberg 0. Eye-movement disorders in
brain-stem and cerebellar stroke. Arch Neurol 1987;
44:141-8.
7 Muskens LJJ. La base anatomique des positions forcees des
yeux soi-disant paralysies du regard. Rev Neurol 1933;
40:287-96.
8 Schuster P. Beitrige zur Pathologie des Thalamus opticus.
I. Mitteilung: Kasuistik. Gefissgebiet der A. thalamogeniculata, der A. thalamo-perforata, der A. tuberothalamica und der A. lenticulo-optica. Arch Psychiat
Nervenkr 1936;105:358-432.
9 Angelergues R, De Ajuriaguerra J, Hecaen H. Paralysie de la
verticalite due regard d'origine vasculaire: etude anatomoclinique. Rev Neurol 1957;96:301-19.
10 Castaigne P, Buge A, Cambier J, Escourolle R, Brunet P,
Degos JD. Demence thalamique d'origine vasculaire par
ramollissement bilateral, limite au territoire du pedicule
retro-mamillaire. A propos de 2 observations anatomocliniques. Rev Neurol 1966;1 14:89-107.
11 Segarra JM. Cerebral vascular disease and behaviour. I. The
syndrome of the mesencephalic artery (basilar artery
bifurcation). Arch Neurol 1970;22:408-18.
12 Halmagyi GM, Evans WA, Halliman JM. Failure of downward gaze-the site and nature of the lesion. Arch Neurol
1976;35:22-6.
13 Trojanowski JQ, Lafontaine MH. Neuroanatomical
correlates of selective downgaze paralysis. J Neurol Sci
1981;52:91-101.
14 Trojanowski JQ, Wray SH. Vertical gaze ophthalmoplegia:
selective paralysis of downward gaze. Neurology 1980;
30:605-10.
15 Pierrot-Deseilligny C, Chain F, Gray F, Serdaru M,
Escourolle R, Lhermitte F. Parinaud's syndrome. Electro-oculographic and anatomical analyses of 6 vascular
cases with deductions about vertical gaze organization in
the premotor structures. Brain 1982;105:667-96.
16 Lepore FE, Gulli V, Miller DC. Neuro-ophthalmological
findings with neuropathological correlation in bilateral
thalamic mesencephalic infarction. J Clin Neuro-Ophthalmol 1985;5:224-8.
17 Dereux J. Paralysie verticale du regard. These Medecine,
Paris. L. Arnette 1926.
18 Garcin R, Bertrand I, Frumusan P. Etude anatomo-clinique
d'un cas de syndrome de Parinaud et de myoclonies
rythmiques du voile du palais. Rev Neurol 1933;40:
812-20.
19 Collier J. Nuclear ophthalmoplegia, with especial reference
to retraction of the lids and ptosis and to lesions of the
posterior commissure. Brain 1927;50:488-98.
20 Molnar L. Die lokal diagnostische Bedeutung der vertikalen
Blicklahmung. Beitrage zur Symnptomatologie und
Faseranatomie des meso-diencephalen Obergangsgebietes. Archiv fur psychiatre und nervenkrankheiten
vereinigt mit Zeitschrift fur die gesamte neurologie und

psychiatre 1958-59;198:523-34.

21 Leigh RJ, Zee DS. The neurology of eye movements. Philadelphia: FA Davis Corporation 1983.

Vertical gaze palsy and selective unilateral infarction of the rostral interstitial nucleus of the medial longitudinalfasciculus (riMLF)
22 Ranalli PJ, Sharpe JA, Fletcher WA. Palsy of upward and
downward saccadic, pursuit, and vestibular movements
with a unilateral mid-brain lesion: pathophysiologic
correlations. Neurology 1988;38:114-22.
23 Reich F. Herd im thalamus opticus. Nerol Centralblatt
1904;14:662.
24 Freund CS. Zur klinik und antomie der vertikalen blicklihmungen. Neurol Centralblatt 1913;32:1215-29.
25 Van Gehuchten P. Syndrome de Parinaud. Etude
anatomique. J Beig Neurol Psychiatrie 1940;40:126-33.
26 Kyorney S. Blickstorungen bei Vascularen Herden des
mesodiencephalen (bergangsgebietes. Archiv fur psychiatre und nervenkrankheiten vereinigt mit Zeitschrift fur
die gesamte neurologie und psychiatre 1959;198:535-43.
27 Hatcher MA Jr, Klintworth GK. The sylvian aqueduct
syndrome. A clinicopathologic study. Arch Neurol 1966;
15:215-22.

71

28 Christoff N. A clinicopathological study of vertical eye
movement. Arch Neurol 1974;31:1-8.
29 Auerbach SH, De Piero TJ, Romanul F. Sylvian aqueduct
syndrome caused by unilateral midbrain lesion. Ann
Neurol 1982;11:91-4.
30 Bogousslavsky J, Miklossy J, Deruaz JP, Regli F, Assal G.
Unilateral left paramedian infarction of thalamus and
midbrain: a clinicopathological study. J Neurol Neurosurg
Psychiat 1986;49:686-94.
31 Iwasaki Y, Matsuda K, Sugasawa J, Utsumi T. A case of
supranuclear downward gaze paralysis due to infarction at
the rostral interstitial MLF detected by MRI. Jpn J
Neuro-ophthalmol 1987;4:206-10.
32 Bogousslavsky J, Regli F. Upgaze palsy and monocular
paresis of downward gaze from ipsilateral thalamomesencephalic infarction: a vertical "one-and-a-half'
syndrome. J Neurol 1984;231:43-5.