N euro-ophthalmology 0 I 65-
Neuro-ophthalmology- 1996, Vol. 16,
No. 2, pp. 99-I06
Dysfunction of pontine omnipause neurons
causes impaired fixation: macrosaccadic
oscillations with a unilateral pontine lesion
© lEolus Press
Buren (The Netherlands) 1996
K. G. Rottach
L. F. Dell'Osso
B. F. Remler
Accepted 5 December 1995
Department of Neurology, University Hospitals and Veterans Affairs
Medical Center, Case Western Reserve University, Cleveland, OH,
form of saccadic dysmetria secondary to cerebellar dysfunction. They are
usually conjugate, horizontal, and symmetric in both directions of gaze. Us
ing magnetic search coils, we studied a patient with MSO that developed
five years following head injury and involved synchronously horizontal, ver
tical, and torsional planes. The MSO were characterized by directional pre
ponderance and were associated with ipsilateral pontine lesion. We propose
a disturbance of fixation mechanisms due to unilateral disinhibition of sac
cadic burst neurons in three planes. This could arise from either primary or
secondary dysfunction of omnipause neurons, due to impaired input from
the contralateral superior colliculus. The delayed onset is suggestive of
denervation supersensitivity as the underlying pathophysiology.
Macrosaccadic oscillations of eyes (MSO) are regarded as a
Saccadic intrusions; omnipause neurons; superiorcolliculus;
Inappropriate saccades that disrupt steady fixation are termed saccadic intru
sions. One example of saccadic intrusions is macrosaccadic oscillations
(MSO). MSO are to-and-fro oscillations of the eyes, consisting usually of
large horizontal saccades that occur in bursts, building up and then decreas
ing in amplitude, with intersaccadic intervals of about 200 msec.I,2 De
scribed originally in cerebellar patients, MSO are thought to reflect saccadic
dysmetria, when patient's saccades are so hypermetric that they overshoot
the target continuously in both directions.3
U sing magnetic search coil technique, we studied a patient with delayed
onset, post-traumatic, large-amplitude MSO in the horizontal, vertical, and
torsional planes, associated with a unilateral pontine lesion. These oscilla
tions provide insight into the neural fixation mechanisms.
Normally, we hold our eyes quite steadily during fixation.
Correspondence to: Dr. R. John
Leigh, Department of Neurology,
University Hospitals of Cleveland,
2074 Abington Road, Cleveland, OH
44106, USA Tel: (216)-421-3224;
Fax: (216)-421 -3040
Acknowledgements: Supported by US
NIH grant EY06717, the Department
of Veterans Affairs and the Evenor
Macrosaccadic oscillations with a pontine lesion
Fig. 1. MRI. Axial T2-weighted
image shows a hyperintense signal in
the right pons. The lesion is to the
right of midline, involving the
tegmentum and basis pontis.
ing of his visual world - oscillopsia - for ten months. Five years pre
viously, he suffered severe head trauma involving a coup injury in the left
frontal region and a contrecoup injury in the right upper brainstem region
as shown by CT scan; no cerebellar lesion was evident. He sustained left
hemiplegia and horizontal double vision. Eleven months after the injury,
he was seen by an ophthalmologist who did not document any abnormal
eye movements except for poor convergence. On current examination, his
general neurological status was unchanged. Right hand coordination was
preserved; he was able to operate a computer 'mouse' accurately with his
right hand. Visual acuity, color perception, visual fields, pupils, fundus,
and slit-lamp examination were all normal. When he attempted to view a
target in the center of his field of vision, large saccades continuously in
truded on steady fixation. These were mainly horizontal, but occasionally
diagonal, and appeared in clusters. They were suppressed when the pa
tient turned his eyes to the extremes of horizontal gaze. There was a full
range of extraocular movements, with concomitant esodeviation of about
8 prism diopters. No convergence could be elicited. MRI demonstrated
hyperintense signals on T2-weighted images in the left posterior frontal
region and the right pons, which were non-enhancing (Fig. 1). The pon
tine lesion extended down to the level of the abducens nucleus and in-
A 34-year-old man presented with the complaint of jump
L. Averbuch-Heller et al.
volved both the tegmentum and basis pontis. No abnormalities of the cer
ebellum were evident.
EYE MOVEMENT MEASUREMENTS
tions of both eyes and of the head were recorded using the magnetic search
coil technique.4 With head stationary, the patient attempted steady fixation,
with each eye in tum, of visual targets located near primary position and at
eccentricities of ± 20 deg horizontally and ± 15 deg vertically at viewing
distances of 1.2 m (far target) or 18 cm (near target). The effects of viewing
a near target binocularly (attempted convergence) were also measured. Hori
zontal and vertical saccades were made between the fixed target locations,
and horizontal and vertical smooth pursuit was measured as the patient fol
lowed a small target moving through ± 15 deg sinusoidally at 0.3 Hz. Visu
ally enhanced vestibulo-ocular reflex (VVOR) was measured while the pa
tient made active horizontal or vertical head rotations, viewing first the far
and then the near target. Data were filtered (bandwidth 0-90 Hz) prior to
digitization at 200 Hz. Analysis was performed using interactive programs
written in the ASYST language.5 The gain of the smooth pursuit response
and the gain of compensatory eye movements during head rotations (VVOR)
were determined as previously described.4
Horizontal, vertical, and torsional rota
nantly horizontal saccadic intrusions. Each saccade took the eye away from
central fixation and was followed by an oppositely directed saccade within
200 msec (Fig. 2). The oscillations were conjugate and of similar amplitude
in both eyes, with synchronous horizontal, vertical, and torsional compo
nents. In the horizontal and torsional planes, the first saccade was always
directed to the patient's right and clockwise; in the vertical plane, it could be
Fixation was frequently disrupted by large-amplitude, predomi
5 II L
Fig. 2. Two segments of the right eye
recording during attempted fixation
of a distant target. Fixation is inter
rupted by bursts of saccadic intru
sions, which are time-locked in the
horizontal, vertical, and torsional
planes. Note that the first saccade is
always to the right or clockwise, but
has no consistent direction vertically.
The return saccade usually overshoots
the central fixation point. Torsional
and vertical tracings have been offset
for convenience of display. Upward
deflections correspond to rightward,
upward, or clockwise eye rotations,
with respect to the patient.
Macrosaccadic oscillations with a pontine lesion
either in the upward or downward direction, though it was always time
locked with the components in the other planes. During attempted fixation of
the target at 1.2 m, the amplitude was maximal in horizontal plane (mean
13.6 deg, range 10-4-15.5 deg); it was small and about equal in the vertical
and torsional planes (vertical, mean 1.2 deg, range 0.9-1.6 deg and torsional,
mean 1.3 deg, range 0.9-1.9 deg). During viewing of the target at 18 cm, the
amplitudes were decreased in all planes of oscillation (horizontal, mean 8.8
deg, range 6-4-I 1.3 deg; vertical, mean 0.9 deg, range 0.3-1.2 deg; torsional,
mean 0.8 deg, range 0-4-1.0 deg). The intersaccadic interval was about 200
msec (range 160-240 msec). The return saccade was usually bigger than the
first one, and often overshot the fixation point by 2-5 deg. The frequency of
the oscillation was about 1.5 Hz, with occasional interruptions for about 2
sec. The oscillation changed little in darkness, with monocular viewing, or
during voluntary saccades, smooth pursuit, and vestibular eye movements.
Gaze holding was preserved during intersaccadic interval in horizontal
plane, maintaining the square-wave appearance of the oscillations. In the
torsional plane, intersaccadic clockwise drifts were evident; in vertical plane,
there were small downward drifts in the left eye only. For voluntary
saccades to visual target, gains (initial saccade amplitude/target amplitude)
were 0.90 for rightward saccades and 1.13 for leftward saccades. The peak
velocity/amplitude relationship was normal for both voluntary saccades and
the saccadic intrusions. The gain of the horizontal VVOR was 1.02. The
gain of the horizontal smooth pursuit was 0.89.
saccadic oscillations that intruded on steady fixation. Each involuntary
movement away from fixation was followed after about 200 msec by a re
turn saccade, often overshooting the fixation point. These movements corre
spond to what has been called 'macrosaccadic oscillations' (MSO), although
the latter are usually induced by a voluntary gaze shift. MSO are tradition
ally interpreted as a sign of saccadic dysmetria due to midline cerebellar
dysfunction.2,3 This explanation may not account for several features charac
terizing the spontaneous MSO in our patient, such as: I) occurrence of the
saccadic intrusions during attempted central fixation; 2) strict laterality con
cerning horizontal and torsional saccades (the first saccade in the burst being
always to the right and clockwise), with no directional preponderance for
vertical saccades; and 3) presence of the intrusions in darkness. These find
ings are consistent with simultaneous disinhibition of the three categories of
the burst neurons: those for horizontal, vertical, and torsional saccades. Such
disinhibition could be produced by impaired function of omnipause neurons.
Omnipause neurons play the role of a gating mechanism for saccades;
they are crucial for suppressing unwanted saccades during fixation and slow
eye movements.6 Located in the caudal pons within raphe interpositus nu
cleus (RIP) adjacent to the abducens nucleus,7 omnipause neurons exert a
tonic inhibition on horizontal saccadic burst neurons in the pontine parame
dian reticular formation (PPRF),8 and on the vertical saccadic burst neurons
in the rostral interstitial nucleus of the medial longitudinal fasciculus
(riMLF).91t has been found that even single omnipause neurons frequently
project to all saccadic generators (horizontal, vertical, and torsional). JO In
puts into omnipause neurons arise in the superior colliculus (SC), frontal eye
fields, and mesencephalic reticular formation, with SC being particularly
Our patient's visual disability was due to frequent, large
L.Averbuch-Heller et al.
important for sustaining steady fixation. 1 1.12 Electrophysiological studies
show that most pause cells cease discharging for saccades in all directions.3
Therefore, omnipause neuron dysfunction should affect all saccadic genera
tors, causing unwanted saccades simultaneously in horizontal, vertical, and
torsional planes, similar to the findings in our patient.
The lesion on the MRI in our patient is localized to the right pons, extend
ing down to the level of the abducens nerve fascicles, involving the tegmen
tum ventral to the MLF. This location is compatible with the right RIP as
well as the site of decussation of the crossing bundle of fibers from the left
RIP to the right riMLF. Saccadic intrusions have been previously docu
mented in association with pontine lesions. Experimental lesions in the re
gion of nucleus reticularis tegmenti pontis (NRTP) in the monkey produce
vertical square-wave jerks.13 Saccadic intrusions have been also reported
with internuclear ophthalmoplegia.14-16 It could be hypothesized that sac
cadic intrusions in these cases were actually produced by damaging the adja
cent omnipause neuron projections (JA B iittner -Ennever, personal communi
RIP is a midline structure, and its projections are usually considered bila
tera1.6.7 However, there is some evidence that the influence exerted by
omnipause neurons upon burst neurons may be lateralized. It was shown that
omnipause efferent projections to riMLF, and its afferents from SC, appear
to be mainly contralateral.ll.12.T7 Saccade generation is based on push-pull
interaction between the 'fixation' cells and the 'saccade' cells within SC. 18
Balance of the activity in favor of the latter will result in a saccade. Unila
teral injections of bicuculline into SC in monkeys produced 'irrepressible
saccades to the side contralateral to the injection' during attempted fixation,
reminiscent of the findings in our patient. 19.20 The effect was probably ex
erted via the contralateral projections from SC to omnipause and saccadic
burst neurons. This further substantiates the possibility of the lateralized
control over the fixation mechanisms.
Horizontal burst neurons in the right PPRF produce rightward saccades;
burst neurons in the right riMLF produce clockwise torsional saccades.21
However, in the vertical plane excitation of the right riMLF can produce
both upward and downward saccades, as the riMLF on each side contains
burst neurons for both up- and downgaze.22.23 Thus, unilateral disinhibition
of all three types of the burst neurons on the right would result in irrepress
ible intrusion of saccades, which would be rightward and clockwise in the
horizontal and torsional planes, respectively, but might be either upward or
downward in the vertical plane. This was the case with our patient. Torsional
components of the oscillation cannot be explained by Listing's law, being
essentially unchanged with constant horizontal and varying vertical compo
nents, and therefore, are centrally produced.24 Hence, we postulate unilateral
disinhibition of all burst neurons as responsible for the initiation of each
'cluster' of MSO in our patient.
Abnormality of omnipause neuron control over saccadic burst neurons has
been previously suggested for saccadic oscillations that lack intersaccadic
intervals, i.e. 'back-to-back' saccadic intrusions of saccadic flutter or opso
clonus.3 On the other hand, saccadic intrusions such as those of our patient
or smaller square-wave jerks were ascribed to different mechanisms.
Square-wave oscillations, such as occur in progressive supranuclear palsy,
large-amplitude MSO, and saccadic back-to-back oscillations might repre-
Macrosaccadic oscillations with a pontine lesion
I Daroff RB, Hoyt WE Supranuclear
disorders of ocular control systems in
man. In: Bach-y-Rita P, Collins CC,
Hyde JE, editors. The Control of Eye
Movements. New York: Academic
sent a continuum of fixation abnormalities with common underlying patho
physiology. In support of this, Tychsen and colleaguesI6 have reported a
patient with stereotypic cycles consisting of tinnitus and square-wave jerks
evolving into opsoclonus, associated with clinical evidence of pontine dys
function. As for the typical absence of the vertical components in MSO, 1,2 in
the majority of the cases reported, only horizontal eye movements were re
corded, so that the vertical and torsional components of the oscillation might
have been overlooked. Whenever the vertical eye movements were mea
sured, a vertical component of MSO was found.25,26
Cerebellar dysmetria probably does contribute to the pattern of the oscilla
tion in our patient, with successive saccades constantly overshooting the fix
ation site. This latter point is also strengthened by the hypermetric saccades
to the left (gain 1.13). Yet, saccadic dysmetria by itself cannot account for
the occurrence of the initial intrusions while fixating a central target, and
would be unlikely to produce the stereotyped, direction-specific nature of the
oscillations. This is supported by the fact that neither patients with cere
bellectomy27 nor monkeys with experimentally produced cerebellar lesions
exhibit MSO.28,29 Moreover, it has been shown that the cerebellum is less
concerned with initiation of a saccade, but more with optimizing its met
rics.30 Therefore, saccadic dysmetria alone does not provide a sufficient ex
planation for MSO occurring during fixation.
Finally, the delay in the development of MSO after the head trauma in our
patient resembles the delayed onset of pendular nystagmus in oculopalatal
myoclonus or see-saw nystagmus with meso-diencephalic lesionsY,32 In
oculopalatal myoclonus, the rhythmic movements appear after the develop
ment of the olivary hypertrophy.33 Such time delay has been attributed to
denervation supersensitivity.34 A similar mechanism might be responsible
for the late-onset MSO in our patient. Delayed development of denervation
supersensitivity could account for the absence of saccadic intrusions imme
diately after experimentally lesioning RIP.35 Similarly, neuronal dysfunction
on a physiological level would explain the normal anatomical appearance of
the omnipause neurons in some patients with opsoclonus, in whom immuno
logical abnormalities have been implicated.36,37 Usually the time between
the initial insult and the subsequent appearance of the oscillation due to
denervation hypersensitivity is months rather than 4-5 years. It is possible,
however, that MSO developed earlier, but the patient became aware of it
only recently, with the functional deterioration of his vision. Oscillopsia is
not a characteristic feature of saccadic intrusions, due to visual suppression
that occurs normally during a saccade.38 Oscillopsia in our patient might
result from the inability to sustain steady fixation, with return saccades tak
ing his eyes across the target. This might become more of a problem with
progressively increasing amplitude of MSO, eventually resulting in oscil
2 Selhorst JB, Stark L, Ochs AL, Hoyt
WE Disorders in cerebellar ocular
motor control. II. Macrosaccadic
oscillation. An oculographic, control
system and clinico-anatomical
analysis. Brain 1976; 99:509-522.
3 Zee DS, Robinson DA. A hypotheti-
cal explanation of saccadic oscilla
tions. Ann Neurol 1979; S:405-4I4.
4 Averbuch-Heller L, Zivotofsky AZ,
Remler BF, Das VE, Dell'Osso LF,
Leigh RJ. Convergent-divergent
pendular nystagmus: possible role of
the vergence system. Neurology
L.Averbuch-Heller et al.
Hary D, Oshio K, Flanagan SD. The
ASYST software for scientific
computing. Science 1987; 236:1128-
Curthoys IS, Markham CH, Furuya
N. Direct projections of pause
neurons to nystagmus related excit-
atory burst neurons in the cat pontine
paramedian reticular formation. Exp
Neurol 1984; 83:414-422.
BUttner-Ennever JA, Cohen B, Pause
M, Fries W. Raphe nucleus of pons
containing omnipause neurons of the
oculomotor system in the monkey,
and its homologue in man. J Comp
Neurol 1988; 267:307-32 I.
Nakao S, Curthoys IS, Markham CH.
Direct inhibitory projections of pause
neurons to nystagmus-related ponto-
medullary reticular burst neurons in
the cat. Exp Brain Res 1980; 40:283-
BUttner-Ennever JA, BUttner U. A
cell group associated with vertical
eye movements in the rostral mesen-
cephalic reticular formation of the
monkey. Brain Res 1978; 151:31-47.
Nakao S, Shiraishi Y, Li W, Inagaki
M. Cat pontine omnipause neurons:
direct inhibitory connection with
Forel's field burst neurons participat-
ing in the genesis of vertical sac-
cades. Acta Otolaryngol 1991; Supp!.
48 I: I 99-204.
BUttner-Ennever JA, BUttner U. The
reticular formation. In: BUttner-Enne-
ver JA, editor. Reviews of Oculo-
motor Research, Vol 2. Neuroanato-
my of the Oculomotor System.
Amsterdam: Elsevier, 1988:119-176.
Langer TP, Kaneko CRS. Brainstem
afferents to the oculomotor omni-
pause neurons in monkey. J Comp
Neurol 1990; 295:413-427.
Gamlin P DR, Mitchell KR. Reversi-
ble lesions of nucleus reticularis
tegmenti pontis affect convergence
and ocular accommodation (abstract).
Soc Neurosci Abstr 1993; 19:346.
Dell'Osso LF, Abel LA, Daroff RB.
'Inverse latent' macro square-wave
jerks and macro saccadic oscillations.
Ann Neurol 1977; 2:57-60.
15 Herishanu UO, Sharpe JA. Saccadic
intrusions in internuclear ophthalmo-
plegia. Ann Neurol 1983; 14:67-72.
16 Tychsen L, Engelken EJ, Austin EJ.
Periodic saccadic oscillations and
tinnitus. Neurology 1990; 40:549-
17 Hom AK, BUttner-Ennever JA,
Wahle P, Reichenberger I. Neuro-
transmitter profile of saccadic omni-
pause neurons in nucleus raphe
interpositus. J Neurosci 1994;
18 Munoz DP, Wurtz RH. Saccade-
related activity in monkey superior
colliculus. II. Spread of activity
during saccades. J Neurophysiol
19 Hikosaka 0, Wurtz RH. Modification
of saccadic eye movements by
GABA-related substances. I. Effect of
muscimol and bicuculline in monkey
superior colliculus. J Neurophysiol
20 Hikosaka 0, Wurtz RH. The basal
ganglia. In: Wurtz RH, Goldberg ME,
editors. Reviews of Oculomotor
Research, Vol 3. The Neurobiology
of Saccadic Eye Movements. Amster-
dam: Elsevier, 1989:257-28I.
21 Leigh RJ, Seidman SH, Grant MP,
Hanna J. Loss of ipsidirectional quick
phases of torsional nystagmus with a
unilateral midbrain lesion. J Vest Res
22 Moschovakis AK, Scudder CA,
Highstein SM. Structure of the
primate burst generator. I. Medium-
lead burst neurons with upward on-
direction. J Neurophysiol 1991;
23 Moschovakis AK, Scudder CA,
Highstein SM, Warren JD. Structure
of the primate burst generator. II.
Medium-lead burst neurons with
downward on-direction. J Neuro-
physiol 1991; 65:218-229.
24 Ferman L, Collewijn H, Van der Berg
AY. A direct test of Listing's law. II.
Human ocular torsion measured
under dynamic conditions. Vision
Res 1987; 27:939-95I.
25 Fukazawa T, Tashiro K, Hamada T,
Kase M. Multisystem degeneration:
drugs and square wave jerks. Neurol-
ogy 1986; 36:1230-1233.
26 Hotson JR. Cerebellar control of
fixation eye movements. Neurology
27 Estanol B, Romero R, Corvera J.
Effects of cerebellectomy on eye
movements in man. Arch Neurol
28 Ritchie L. Effects of cerebellar
lesions on saccadic eye movements. J
Neurophysiol 1976; 39: I 246-I256.
29 Optican LM, Robinson DA. Cerebel-
lar-dependent 'adaptive control of
primate saccadic system. J Neuro-
physiol 1980; 44: 1058-I 076.
30 Vilis T, Hore J. Characteristics of
saccadic dysmetria in monkeys
during reversible lesions of medial
cerebellar nuclei. J Neurophysiol
31 Halmagyi GM, Aw ST, Dehaene I,
Curthoys IS, Todd MJ. Jerk-wave-
form see-saw nystagmus due to
unilateral meso-diencephalic lesion.
Brain 1994; IIT789-803.
32 N akada T, K wee IL. Oculopalatal
myoclonus. Brain 1986; 109:431-44I.
33 Yokota T, Tsukagoshi H. Olivary
hypertrophy precedes the appearance
of palatal myoclonus (letter). J Neurol
34 Matsuo F, Ajax ET. Palatal myoclo-
nus and denervation supersensitivity
in the central nervous system. Ann
Neurol 1979; 5:72-78.
35 Kaneko CRS, Fuchs AF. The effect
of ibotenic acid lesions of the omni-
pause neurons on saccadic eye move-
ments in the monkey. Soc Neurosci
Abstr 1987; 13:393.
36 Ridley A, Kennard C, Scholtz CL,
BUttner-Ennever JA, Summers B,
Turnbull A. Omnipause neurons in
two classes of opsoclonus associated
with oat cell carcinoma of the lung.
Brain 1987; 110: 1699-1709.
37 Nitschke M, Hochberg F, Dropcho E.
Improvement of paraneoplastic opso-
clonus-myoclonus after protein A
column therapy. N Engl J Med 1995;
Macrosaccadic oscillations with a pontine lesion
38 Burr DC, Morrone MC, Ross J.
Selective suppression of the
magnocellular visual pathway during
L. Averbuch-Heller et al.
saccadic eye movements. Nature