Sunteți pe pagina 1din 21

A hypothetical scheme for the brainstem control of vertical gaze

Bhidayasiri, Roongroj MD; Plant, Gordon T. FRCP; Leigh, R. John MD


Author Information
From the Departments of Neurology (Drs. Bhidayasiri and Leigh), Biomedical Engineering (Dr.
Leigh), and Neuroscience (Dr. Leigh), Department of Veterans Affairs Medical Center and
University Hospitals, Case Western Reserve University, Cleveland, OH; and Department of
Neurology (Dr. Plant), Guys & St. Thomas Hospital, London, UK.
Received November 15, 1999.
Accepted in final form March 1, 2000.
Supported by USPHS grant EY06717, the Office of Research and Development, Medical
Research Service, Department of Veterans Affairs, and the Evenor Armington Fund (to R.J.L.).
Address correspondence and reprint requests to Dr. R. John Leigh, Department of Neurology,
University Hospitals, 11100 Euclid Avenue, Cleveland, OH 44106-5040.
Abstract
Objectives: To develop a hypothetical scheme to account for clinical disorders of vertical gaze
based on recent insights gained from experimental studies.
Methods: The authors critically reviewed reports of anatomy, physiology, and effects of
pharmacologic inactivation of midbrain nuclei.
Results: Vertical saccades are generated by burst neurons lying in the rostral interstitial nucleus
of the medial longitudinal fasciculus (riMLF). Each burst neuron projects to motoneurons in a
manner such that the eyes are tightly coordinated (yoked) during vertical saccades. Saccadic
innervation from riMLF is unilateral to depressor muscles but bilateral to elevator muscles, with
axons crossing within the oculomotor nucleus. Thus, riMLF lesions cause conjugate saccadic
palsies that are usually either complete or selectively downward. Each riMLF contains burst
neurons for both up and down saccades, but only for ipsilateral torsional saccades. Therefore,
unilateral riMLF lesions can be detected at the bedside if torsional quick phases are absent
during ipsidirectional head rotations in roll. The interstitial nucleus of Cajal (INC) is important
for holding the eye in eccentric gaze after a vertical saccade and coordinating eyehead
movements in roll. Bilateral INC lesions limit the range of vertical gaze. The posterior
commissure (PC) is the route by which INC projects to ocular motoneurons. Inactivation of PC
causes vertical gaze-evoked nystagmus, but destructive lesions cause a more profound defect of
vertical gaze, probably due to involvement of the nucleus of the PC. Vestibular signals
originating from each of the vertical labyrinthine canals ascend to the midbrain through several
distinct pathways; normal vestibular function is best tested by rotating the patients head in the
planes of these canals.
Conclusions: Predictions of a current scheme to account for vertical gaze palsy can be tested at
the bedside with systematic examination of each functional class of eye movements.

Clinicians have attempted to account for vertical gaze palsies since Parinauds 1 and Spillers 2
classic descriptions. However, a lack of means by which to identify neurons with similar
properties and projections (e.g., saccade-generating neurons projecting to ocular motoneurons)
and the tendency for conventional lesions to involve both neurons and axons of passage have
made interpretation difficult. Furthermore, few clinical case reports combine discrete,
pathologically confirmed lesions with careful description of defects of eye movements.
Modern experimental techniques have clarified the anatomic connections that serve vertical
gaze, defined neuronal populations with similar electrophysiologic properties, and delineated
behavioral changes caused by inactivation of such populations. 3 In this review, we will
summarize these new anatomic and behavioral studies and use them to develop a hypothetical
scheme for vertical gaze. We will then apply this scheme in order to interpret and comment on
clinical disorders of vertical gaze. Finally, we will discuss the examinations that might be useful
in testing this hypothetical scheme at the bedside.
Development of a hypothetical scheme for vertical gaze.
The ocular motoneurons. The oculomotor and trochlear nuclei contain the ocular motoneurons
that control vertical eye movements (figure 1). These ocular motoneurons supply the superior
and inferior rectus muscles and the superior and inferior oblique muscles. Recall that axons
from motoneurons to the superior rectus and superior oblique muscles are crossed, whereas
those to the inferior rectus and inferior oblique are uncrossed. Although this review will mainly
concern disorders of vertical gaze, which are readily evident at the bedside, associated disorders
of torsional gaze are a common accompaniment and sometimes provide the clue to diagnosis.
Indeed, the pulling actions of the extraocular muscles evolved to move the eyes in the planes of
the labyrinthine semicircular canals, which are not strictly vertical or horizontal. For this reason,
it is often useful to test the vestibular responses by rotating the patients head in the planes of
the vertical semicircular canals, 4 or in roll (ear-to-shoulder).

Figure 1. Anatomic schemes for the synthesis of upward, downward, and torsional eye
movements. From the vertical semicircular canals, primary afferents on the vestibular nerve
(vn) synapse in the vestibular nuclei (VN), and ascend in the medial longitudinal fasciculus
(MLF) and brachium conjunctivum (BC) to contact neurons in the trochlear nucleus (CN IV),
oculomotor nucleus (CN III), and the interstitial nucleus of Cajal (INC). (For clarity, only
excitatory vestibular projections are shown.) The rostral interstitial nucleus of the medial
longitudinal fasciculus (riMLF), which lies in the prerubral fields, contains saccadic burst
neurons. It receives an inhibitory input from omnipause neurons of the nucleus raphe
interpositus (rip), which lie in the pons (for clarity, this projection is only shown for upward
movements). Excitatory burst neurons in riMLF project to the motoneurons of CN III and CN
IV, and also send an axon collateral to INC. Each riMLF neuron sends axon collaterals to yokepair muscles (Herings law). Projections to the elevator subnuclei, innervating the superior
rectus and inferior oblique muscles, may be bilateral due to axon collaterals crossing at the level
of the CN III nucleus. Projections of inhibitory burst neurons are less well understood, and are
not shown. The INC provides a gaze-holding signal, and projects to vertical motoneurons via
the posterior commissure. Signals contributing to vertical smooth pursuit and eyehead tracking
reach CN III from the y-group via the brachium conjunctivum and a crossing ventral tegmental
tract. The axons of ocular motoneurons are not shown, but note that those supplying the superior
rectus (sr) and superior oblique (so) muscles are crossed whereas axons supplying the inferior
rectus (ir) and inferior oblique (io) are uncrossed (adapted from reference 3, with permission).
The riMLF: neural substrate for vertical saccades.
The premotor signal for saccadic eye movements is generated by burst neurons that discharge
only during saccades, and which project monosynaptically to ocular motoneurons. Although it

has been known for some time that the reticular formation of the midbrain houses neurons that
are important for generating vertical saccades, defining the location of these burst cells in the
rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF) awaited application of
the technique of single-unit recording and modern anatomy (figure 2). 5 The riMLF is a wingshaped nucleus lying dorsomedial to the red nucleus, rostral to the tractus retroflexus and the
oculomotor nucleus, and ventral to the periaqueductal gray matter. 6 In the past, this structure
was called the nucleus of the prerubral fields, or the nucleus of the field of Forel. In classic
Nissl-stained sections of human midbrain, differentiation of the riMLF from the adjacent
interstitial nucleus of Cajal (INC) is difficult, but by using parvalbumin immunostaining 7 it is
possible to differentiate riMLF neurons as medium-sized elongated cells. Cells in the adjacent
INC are also parvalbumin-positive, but are round and densely packed. The blood supply of the
riMLF is from the posterior thalamosubthalamic paramedian artery, which arises between the
bifurcation of the basilar artery and the origin of the posterior communicating artery and lies at
the dorsomedial border of the riMLF. It penetrates the floor of the third ventricle, sometimes
with a single vessel supplying both riMLF. 8

Figure 2. Sagittal section of monkey brainstem showing location of rostral interstitial nucleus of
the medial longitudinal fasciculus (riMLF), interstitial nucleus of Cajal (INC), posterior
commissure (PC), and other structures important for the control of vertical gaze. Shaded areas
indicate the mesencephalic reticular formation (MRF), paramedian pontine reticular formation
(PPRF), and medullary reticular formation (Med RF). Asterisks indicate the location of cell
groups of the paramedian tracts, which project to the cerebellum. III = oculomotor nucleus; IV
= trochlear nucleus; VI = abducens nucleus; CG = central gray; MB = mammillary body; MT =

mammillothalamic tract; N III = rootlets of the oculomotor nerve; N IV = trochlear nerve; N VI


= rootlets of the abducens nerve; ND = nucleus of Darkschewitsch; NRTP = nucleus reticularis
tegmenti pontis; NPH = nucleus prepositus hypoglossi; TR = tractus retroflexus. The arrow
refers to the Horsley-Clarke plane of section. (Courtesy of Dr. Jean Bttner-Ennever.)
Each riMLF contains both excitatory and inhibitory burst neurons that discharge during vertical
saccades, and torsional quick phases of nystagmus. 9-12 Neurons in the riMLF burst for either
upward or downward eye movements, but only in one direction for torsional quick phases.
Thus, for example, the right riMLF discharges for quick phases that are directed clockwise with
respect to the subject, as the upper pole of each eye rotates toward the patients right shoulder.
9-12 It is uncertain whether different parts of the riMLF contain neurons that burst for
movements in a particular direction, or whether neuron types are intermingled. 6,13,14 The
riMLF receives inputs from omnipause neurons in the pontine raphe nucleus interpositus, which
gate the activity of all burst neurons. 15 Other inputs are from the superior colliculus, the
nucleus of posterior commissure (PC), the fastigial nucleus of the cerebellum, the contralateral
riMLF, and cortical frontal eye fields. 3,5,16
By recording from individual axons of neurons in the riMLF and then injecting them with
tracer, it has been possible to define projections of upward and downward burst neurons to the
ocular motoneurons. 9,10 Based on this survey of riMLF cells, it is possible to draw some
general principles of this organization (see figure 1). Each riMLF projects simultaneously to
motoneurons that move each eye in the same direction. For example, one type of burst neuron
sends axons to the motoneurons innervating the inferior rectus muscle of one eye and the
superior oblique muscle of the other. These may be regarded as a yoke-muscle pair. 17 Because
the same neuron and same transmitters (probably glutamate/aspartate for excitatory burst
neurons) are responsible for activating both sets of motoneurons, 7,17 this may explain why
vertical saccades are more conjugate than horizontal saccades, which depend on the medial
longitudinal fasciculus to coordinate medial and lateral rectus muscle activity. 18 A second
important finding is that the projections of burst neurons in riMLF to motoneurons innervating
elevator muscles (superior rectus and inferior oblique) appear to be bilateral, with axons
collaterals probably crossing within the oculomotor nuclear complex, and apparently not via the
PC, as had been deduced previously from clinicopathologic studies. 9,17 Projections to
motoneurons supplying the depressor muscles (inferior rectus and superior oblique), however,
appear to be ipsilateral. 10,17 This connectivity predisposes partial lesions of the riMLF to
cause predominantly downward saccadic palsy. The riMLF also projects to the INC, which is
important for vertical gaze-holding and to the cell groups of paramedian tract, which relay all
ocular motor signals to the cerebellum. 19
The effects of lesions of the riMLF have been clarified by inactivating, or selectively lesioning,
neurons (but not axons) in this nucleus with injections of the GABA agonist muscimol and
kainic acid. 12,20 With unilateral injections, a mild defect of downward saccades is produced;
this finding is consistent with the bilaterality of riMLF projections to elevator, but not depressor,
motoneurons. More importantly, there is loss of specific ipsitorsional quick phases; for example,
injection of the right riMLF abolishes quick phases with the upper pole of each eye rotating to

the patients right shoulder. 21 Bilateral riMLF lesions abolish all vertical and torsional saccades
and quick phases. Other types of eye movements are preserved.
Although the riMLF is the key structure for vertical saccades, some vertical burst neurons lie
outside its boundaries. Also, the central mesencephalic reticular formation (MRF), which has
connections with the superior colliculus, nucleus of the PC, paramedian pontine reticular
formation, fastigial nucleus, and cortical eye fields, seems to play an important role in the
generation of vertical saccades. 22,23 Experimental inactivation of the central MRF causes
hypermetria of contralateral and upward saccades. 22 Fixation is disrupted by saccadic
intrusions that are directed away from the side of inactivation. Inactivation of more rostral MRF
causes hypometria and slowing of vertical saccades. 23 These studies have suggested several
distinct mechanisms by which the MRF and superior colliculus govern the generation of
saccades; these hypotheses are critically review by Waitzman et al. 22,23
INC: the neural integrator for vertical gaze.
Holding the eye away from the central position in the orbit requires a sustained contraction of
the extraocular muscles. Without such a sustained contraction, elastic forces of the orbital
tissues would pull the eyes back toward the central position. The ocular motoneurons encode the
eye position command that is necessary to produce a sustained contraction of the extraocular
muscles. However, the neural signals from saccadic burst neurons, vestibular afferents, and
neurons mediating visually driven movements such as smooth pursuit are all encoded in terms
of the velocity of the movement. 24 Thus, there is a need to integrate premotor signals. For
horizontal movements, the region of the medial vestibular nucleus and nucleus prepositus
hypoglossi, along with their cerebellar projections, play a key role. 24,25 For vertical eye
movements, a major contributor is the INC. 26,27
The INC lies adjacent and caudal to the riMLF (see figure 2), close to the rostral end of the
EdingerWestphal nucleus, and is separated from oculomotor nuclei by the MLF. 3 In human
brain sections immunostained using parvalbumin, neurons in the INC are round and densely
packed. 7 The INC contains at least two distinct populations of neurons 28 : one set contributes
to the neural integrator for vertical gaze, whereas the other may be concerned with eyehead
coordination in roll. The INC receives inputs from the saccadic burst neurons of the riMLF, the
vestibular nuclei, 29 and the y-group, 30 which caps the inferior cerebellar peduncle, and
contributes to vertical smooth pursuit. The main projection from the INC to ocular motoneurons
in the oculomotor and trochlear nuclei, as well as the contralateral INC, is through the PC. 31 It
also projects rostrally to the MRF, bilaterally to riMLF, 31 and caudally to targets that include
the gigantocellular reticular formation, which is important for head movements during eyehead
gaze shifts, as well as to the anterior horn of the first four cervical segments of the spinal cord.
31
The population of INC neurons that project to ocular motoneurons variably encode vertical eye

position, saccadic bursts, and eye velocity during vertical smooth pursuit. 32 Stimulation of one
INC in a human patient caused torsional nystagmus with contralaterally beating quick phases
and an ipsilateral ocular tilt reaction (see below). 33
The influence of the INC on vertical gaze has been clarified by pharmacologic inactivation
using microinjections of muscimol. 34 The effects of unilateral and bilateral inactivation of INC
differ. Unilateral inactivation produces a contralateral ocular tilt reaction (OTR), which consists
of a contralateral head tilt associated with skew deviation (ipsilateral hypertropia), intorsion of
the ipsilateral eye, and extorsion of contralateral eye. 35 In addition, torsionalvertical
nystagmus is produced; the torsional component is greater in the eye on the side of the
inactivation, with quick phases directed ipsilesionally. 21,36 The downbeat component may
have different vectors in each eye. These vectors partially reflect the planes of the vertical
labyrinthine canals, but orbital factors such as fibromuscular pulleys may have an effect on the
pulling action of the extraocular muscles. 37,38 Unilateral INC injections can also cause
centripetal drifts of the eyes after saccades with a vertical component, consistent with the view
that the INC is important for integration of vertical saccadic signals. It is unsettled whether a
unilateral lesion can elicit seesaw forms of nystagmus. 36,39 Bilateral INC lesions limit the
vertical range of all classes of conjugate eye movements, although saccades do not become
slow. 34 Furthermore, upbeat nystagmus may occur. 34 Thus, the INC exerts important and
complex influences on vertical eye movements. These experimental findings make it possible to
differentiate the lesions in INC from riMLF. INC lesions limit the range of vertical saccades, but
the velocity does not become slow as is seen in riMLF lesions. 34,40 In addition, with unilateral
midbrain lesions that cause torsional nystagmus, ipsilesional quick phases may occur if the
adjacent riMLF is intact; if the riMLF is also involved, either no quick phases or contralesional
quick phases may be observed. 21,34
The posterior commissure: a critical pathway for vertical gaze.
The PC, which marks the transition from midbrain to diencephalon, contains axons that are
important in the control of vertical gaze. Lying rostral to the superior colliculi (see figure 2) at
the junction of the aqueduct and third ventricle, it consists of a series of crossing fibers
intermingled with scattered cells that constitute the nucleus of the PC (NPC). 9,41 Fibers of the
dorsal part of the PC generally ascend to more dorsal and rostral structures, whereas fibers in
the ventral part predominantly interconnect the NPC and supply the INC region. 41 The PC
contains axons from INC projecting to the contralateral oculomotor complex and INC, and also
carries axons from the NPC projecting to the contralateral INC and riMLF, 9,42 as well the Mgroup, which is important for coordinating eyelid movements with vertical gaze. 43
The behavioral deficits due to pharmacologic inactivation of the PC are similar to those caused
by inactivation of the INC, which projects through the PC. Thus, injections of lidocaine cause
impairment of vertical gaze-holding, resulting in vertical gaze-evoked nystagmus and affecting
the temporal (phase) properties of the vertical vestibulo-ocular reflex (VOR). 44 Destructive
lesions of the PC, however, cause more extensive deficits that restrict all types of vertical eye

movements, especially upward movements, although the VOR and Bells phenomenon may be
partially spared. 45 The difference in deficits between inactivation of axons and destructive
lesions may reflect involvement of the NPC, 9 which probably makes important contributions to
vertical gaze; however, these contributions have yet to be clearly defined.
Medial longitudinal fasciculus: ascending pathway for vertical pursuit and vestibular eye
movements.
The MLF is an important pathway by which signals for vertical gaze reach the vertical ocular
motoneurons and the INC. The cells of origin of these axons are located within the vestibular
nuclei, and encode signals for the VOR and smooth pursuit. 46 Thus, axons in the MLF carry,
from medulla to midbrain, neural signals important for vertical vestibular and smooth pursuit
eye movement and, to a lesser extent, the vertical gaze-holding command. 29,46 The MLF is an
important route for these projections, but the brachium conjunctivum (superior cerebellar
peduncle) and other pathways also contribute. 47,48 Thus, a gaze-velocity signal ascends to the
midbrain in the brachium conjunctivum from the dorsal portion of the y-group, 49 a small
collection of cells that cap the inferior cerebellar peduncle and receive afferents from flocculus
Purkinje cells. The gaze-velocity signal from the y-group may cancel out a head velocity signal
on the MLF during combined eyehead tracking, so that gaze stays on target.
These anatomic and physiologic data would suggest that lesions of the MLFinternuclear
ophthalmopelgia (INO)would cause abnormalities of vertical gaze in addition to the
commonly recognized disruption of horizontal movements (paresis of ipsilateral ocular
adduction and dissociated nystagmus). 50,51 It is well recognized that unilateral INO causes
skew deviation, usually an ipsilateral hypertropia, which is a component of an OTR due to
interruption of otolith-ocular projections to the INC, oculomotor and trochlear nuclei. 35 What
has only been recently demonstrated is that an asymmetry of the canal-mediated VOR is also
present in unilateral INO, which reflects different central projections from the posterior
semicircular canal (principally the MLF) and the anterior canal (brachium conjuntivum and
other pathways in addition to the MLF). 29,48 With bilateral INO, both the upward and
downward VOR are hypoactive (reduced gain) 52; vertical smooth pursuit is moderately
impaired, and the VOR is not appropriately cancelled during combined eyehead tracking,
which impairs performance. Vertical gaze-holding is partially affected in bilateral INO, upbeat
nystagmus on upgaze being relatively common. 52 Although the INC plays a major role as the
vertical neural integrator, lesions affecting the MLF or the adjacent cell groups of paramedian
tracts also appear to contribute to vertical gaze holding. 19,53 Vertical saccades and quick
phases are usually intact.
Application of anatomic scheme to interpretation of disorders of vertical gaze.
Here we consider how well the hypothetical scheme that we have developed for vertical gaze,
which is summarized in figure 1, accounts for disorders of vertical gaze that are observed

clinically. These disorders, and common clinical causes of them, are summarized in the table.

Table 1. Summary of defects of vertical gaze reported with discrete lesions** This is not
intended as comprehensive summary; see elsewhere for more details. 3
Vertical saccadic palsies.

Selective defects of vertical saccades are probably under-recognized, going unnoticed unless the
examiner specifically tests for them by asking patients to shift gaze between two stationary
targets located one above the other (e.g., a pencil tip and the examiners nose). Vertical saccades
may be absent or slowed. A restricted range of vertical saccades, but with normal velocity, is
more commonly encountered as a component of a more global vertical gaze palsy.
Vertical saccadic palsies usually involve either downward saccades or both upward and
downward saccades. Usually, the latter deficit can be most easily explained by bilateral
infarctions of the riMLF, due to occlusion of a single posterior thalamosubthalamic paramedian
artery (as part of top of the basilar syndrome). 6 Several explanations for selective palsy of
downward saccades have been offered, on the basis of the distribution of up- and down- burst
neurons within the riMLF, or the pathway taken by their respective axons as they exit this
nucleus. 6,13,14 However, an alternative explanation, supported by the effects of experimental
lesions of the riMLF, 20 is that motoneurons supplying elevator muscles receive a bilateral
innervation, whereas those supplying depressor muscles receive only an ipsilateral innervation.
9,10 Thus, projections to depressor motoneurons would be expected to be disproportionately
affected by partial riMLF lesions. In future studies, testing torsional quick phases with ear-toshoulder head roll might shed light on the extent to which each riMLF is involved in any patient
so affected. 54
Impairment of vertical smooth pursuit.
Impairment of smooth pursuit in both the horizontal and vertical plane is a common,
nonspecific finding in normal elderly subjects, 55 in patients receiving a variety of drugs, and as
a feature of a range of disorders affecting the cerebrum, cerebellum, and brainstem. 3 The
diagnostic value of vertical smooth pursuit often occurs when it is relatively preserved
compared with saccadic or other eye movements. Thus, clinical lesions affecting the riMLF may
relatively spare vertical pursuit, implying that the MLF and other ascending pathways are intact.
In certain metabolic disorders, such as Niemann-Pick type C (S) disease, the finding of normal
vertical pursuit with slow vertical saccades has been presented as evidence to hypothesize
involvement of vertical burst neurons by this disorder. 56 In bilateral INO, however, vertical
saccades are spared, but vertical pursuit may be abnormal. 52 Smooth, combined eyehead
tracking is also impaired in bilateral INO. Future studies comparing the relative effects of
lesions involving the MLF or brachium conjuntivum would help to confirm the experimental
data that have implied separate contributions to smooth eye and eyehead tracking from the
vestibular nuclei and the y-group.
Impairment of the vertical VOR.
Selective defects of both the otolithic and canal-mediated VOR occur as a feature of INO. The
former is more widely recognized when it causes contralateral OTR with a skew deviation
consisting usually of an ipsilateral hypertropia. 35 Recently, an asymmetry of the canal-

mediated VOR has also been reported in unilateral INO. 48 To demonstrate this deficit, it is
necessary to rotate the patients head, using high-acceleration head thrusts in the planes of the
vertical semicircular canals. 4 Thus, in a patient with a right MLF lesion, when his head was
rotated in the plane of the left posterior semicircular canal, the vestibular response was greatly
impaired (gain of 0.3). Conversely, when the same patients head was rotated in the plane of the
left anterior semicircular canal, the response was approximately twice as great (gain of 0.6),
although it was still not normal (gain range 0.7 to 1.0). 4,48 The explanation for this disparity is
that projections from the left posterior semicircular canal to right-sided motoneurons supplying
the inferior rectus and superior oblique depended on the right MLF. Conversely, projections
from the left anterior canal to right-sided motoneurons supplying the superior rectus and inferior
oblique depended only partially on the MLF, some axons ascending in other pathways, such as
the brachium conjuntivum. 29 Bilateral INO causes proportionally greater disturbance of the
vertical VOR, evident when the head is rotated in pitch. 52 Lesions involving the INC
commonly affect the phase (temporal relationship between head and eye) of the vertical VOR,
but do not abolish the response, because of direct projections from the vestibular nuclei to the
oculomotor and trochlear nuclei. 34 Like smooth pursuit, preservation of the vertical vestibular
response in patients with vertical gaze palsies is often a helpful sign in localizing a lesion to the
rostral midbrain.
Impairment of vertical gaze involving all conjugate eye movements.
Some patients with disorders of vertical gaze cannot move their eyes up or down by any means
saccadic, pursuit, or vestibular. (Although elevation of the eyes during forceful eyelid closure
may be preserved, the mechanism for this Bells phenomenon remains to be clarified.) Such
patients with variants of Parinauds syndrome usually have involvement of either the INC or the
PC, or both. 14,57,58 Thus, on the basis of clinical studies, the INC must be regarded as a
critical structure for vertical gaze. Recent experimental studies have confirmed this view and
shown that bilateral lesions of INC not only impair gaze-holding ability (neural integrator
function), but also greatly restrict the range of vertical eye movements. 34
Previous emphasis on how bilateral lesions are required to produce vertical gaze palsies no
longer seems to be a central issue because of the crossed and commissural projections that
subserve vertical gaze. 31 Thus, for example, each INC projects to ocular motoneurons and to
the opposite INC via the PC. Thus, a lesion of one INC may be, in effect, a bilateral lesion, and
so affects inputs to motoneurons on both sides. 57 What seems to deserve more attention in
future studies is why INC lesions restrict the range of vertical eye movements, 34 and why
lesions of the PC have such profound effects on all vertical eye movements, especially upward.
45 Comparatively little is known about the nucleus of the PC, but this structure may exert
profound effects on vertical gaze, and probably on lid movements as well. In addition, the
central MRF, by virtue of its reciprocal connections with the superior colliculus, may exert
substantial influence on vertical gaze, especially saccades. 22,23 Clinical studies of lesions
restricted to this region are needed.

Disjunctive disorders of vertical gazevertical one-and-a-half syndrome and double elevator


palsy.
Most patients with vertical gaze palsies show conjugate defects, unless the lesion also involves
the oculomotor nucleus or fascicles. Two rare exceptions are vertical one-and-a-half syndrome
and monocular elevator paresis. Two variants of vertical one-and-a-half syndrome have been
described. The first consists of bilateral upgaze palsy and monocular paresis of downward gaze
either ipsilateral or contralateral to the side of the lesion; it has been reported with lesions in the
thalamo-mesencephalic area. 59,60 The second variant of vertical one-and-a-half syndrome
consists of impairment of downward eye movements in association with monocular paralysis of
elevation; it was described in association with bilateral mesodiencephalic infarcts. 61,62
Localization in these cases was based on neuroimaging and no neuropathologic data were
available. Disjunctive saccadic palsy affecting downward movements is conceivable if axon
collaterals from downward burst neurons in the riMLF to either inferior rectus motoneurons or
superior oblique motoneurons innervating one eye were selectively involved. However, because
upward burst neurons send bilateral projections within the oculomotor nucleus to superior rectus
and inferior oblique motoneurons, a prenuclear lesion causing disjunctive upward saccadic
palsy seems unlikely. However, variants of the vertical one-and-a-half syndrome usually affect
more than just saccadic eye movements and, in these cases, a variety of combined lesions seems
possible.
The syndrome of double-elevator or monocular elevation palsy is characterized by an inability
to elevate one eye, which is equally affected whether the eye is in abduction or adduction. This
condition has been thought to be due to supranuclear lesions. 59,63,64 A variant of this
syndrome, called crossed vertical gaze paresis, is characterized by monocular elevation paresis
of one eye with contralateral paresis of downward gaze. 65 Because the superior rectus is a
stronger elevator than the inferior oblique, it is difficult to be sure of weakness of the inferior
oblique. If both of these muscles are weak, then a nuclear lesion is unlikely, because the inferior
oblique is supplied by the ipsilateral oculomotor nucleus and the superior rectus is supplied by
the contralateral nucleus. Furthermore, because the principal elevator of the eye, the superior
rectus, receives bilateral saccadic innervation at the level of the oculomotor nucleus, 9 it seems
unlikely that prenuclear lesion could cause this syndrome. An alternative explanation for
monocular elevator palsy, which seems tenable in some reported cases, 66,67 is that it is caused
by a lesion selectively involving the oculomotor fascicles supplying the inferior oblique and
superior rectus muscles, which lie adjacent as the third nerve exits the brainstem. 68 However,
such an explanation cannot account for a patient in whom there was no misalignment of the
visual axes with the eyes in central position. 69
Other disturbances associated with vertical gaze palsies.
We have focused on conjugate vertical eye movements, but other clinical signs often accompany
vertical gaze palsiesabnormalities of the eyelids, the pupils, and horizontal gaze. Some
progress is being made toward understanding these findings, especially disorders of lid
movements. 70 The accompanying disturbances of horizontal gaze, such as convergence

retraction nystagmus and pseudoabducens palsy, remain unexplained. 3 Further studies to


account for this pathologic synkinesis of horizontal and vertical saccadic movements are
needed.
Testing aspects of the hypothetical scheme at the bedside.
The scheme that we have developed must be regarded as hypotheticalits predictions about the
control of vertical gaze require testing, and clinical studies should be an important part of this
effort. At the bedside, the clinician must examine not only the range of movement and static
alignment of the eyes, but also the properties of each functional class of eye movement. As
outlined above, the reason for such testing is that although certain lesions may cause paralysis
of all upward or downward movements or both, selective defects of ocular motility commonly
occur. It is also important to test the torsional VOR, noting whether quick phases of nystagmus
occur in both directions as the patients head rolls, ear-to-shoulder, from side to side. To
effectively test the vertical VOR, the patients head should be rotated not only vertically in
pitch, but also in the planes of the vertical labyrinthine semicircular canals. 4 This may be
conveniently performed by first turning the patients head horizontally 45 degrees to one side,
and then pitching the head forward or backwards. The best stimulus is a brief, high-speed head
rotationa head thrust. Finally, the evaluation is incomplete without testing horizontal gaze,
and looking for abnormalities of eyelid movements and the pupils, as disorders of vertical eye
movements cannot be viewed in isolation, and must be integrated into the overall control of
gaze.
Acknowledgment
The authors are grateful to Dr. Jean Bttner-Ennever for providing figure 2 and to Dr. Pramod
Sethi for assistance with the literature review.
References
1. Parinaud H. Paralysie des mouvements associs des yeux. Arch Neurol (Paris) 1883; 5:145
172. [Context Link]
2. Spiller WG. The importance in clinical diagnosis of paralysis of associated movements of the
eyeballs, especially of upward and downward associated movements. J Nerv Ment Dis
1905;32:417426, 497530. Ovid Full Text SFX [Context Link]
3. Leigh RJ, Zee DS. The neurology of eye movements. 3rd ed. New York, NY: Oxford
University Press, 1999. [Context Link]

4. Cremer PD, Halmagyi GM, Aw ST, et al. Semicircular canal plane head impulses detect
absent function of individual semicircular canals. Brain 1998; 121:699716. SFX [Context
Link]
5. Bttner-Ennever JA, Bttner U. A cell group associated with vertical eye movements in the
rostral mesencephalic reticular formation of the monkey. Brain Res 1978; 151:3147. SFX
[Context Link]
6. Bttner-Ennever JA, Bttner U, Cohen B, Baumgartner G. Vertical gaze paralysis and the
rostral interstitial nucleus of the medial longitudinal fasciculus. Brain 1982; 105:125149. SFX
[Context Link]
7. Horn AKE, Bttner-Ennever JA. Premotor neurons for vertical eye movements in the rostral
mesencephalon of monkey and human: Histologic identification by parvalbumin
immunostaining. J Comp Neurol 1998; 392:413427. SFX [Context Link]
8. Percheron G. The anatomy of the arterial supply of the human thalamus and its use for the
interpretation of the thalamic vascular pathology. J Neurol 1973; 20:113. [Context Link]
9. Moschovakis AK, Scudder CA, Highstein M. Structure of the primate oculomotor burst
generator I. Medium-lead burst neurons with upward on-directions. J Neurophysiol 1991;
65:203217. [Context Link]
10. Moschovakis AK, Scudder CA, Highstein SM, Warren JD. Structure of the primate
oculomotor burst generator. II. Medium-lead burst neurons with downward on-directions. J
Neurophysiol 1991; 65:218229. SFX [Context Link]
11. Henn V, Straumann D, Hess BJM, Hepp K, Vilis T, Reisine H. Generation of vertical and
torsional rapid eye movement in the rostral mesencephalon-experimental data and clinical
implications. Acta Otolaryngol (Stockh) 1991; 481(suppl):191193. SFX [Context Link]
12. Vilis T, Hepp K, Schwarz U, Henn V. On the generation of vertical and torsional rapid eye
movements in the monkey. Exp Brain Res 1989; 77:111. SFX [Context Link]
13. Wang SF, Spencer RF. Spatial organization of premotor neurons related to vertical upward

and downward saccadic eye movements in the rostral interstitial nucleus of the medial
longitudinal fasciculus (riMLF) in the cat. J Comp Neurol 1996; 366:163180. [Context Link]
14. Pierrot-Deseilligny CH, Chain F, Gray F, Serdaru M, Escourolle R, Lhermitte F. Parinauds
syndrome: electro-oculographic and anatomical analyses of six vascular cases with deductions
about vertical gaze organization in the premotor structures. Brain 1982; 105:667696. [Context
Link]
15. Horn AKE, Bttner-Ennever JA, Wahle P, Reichenberger I. Neurotransmitter profile of
saccadic omnipause neurons in nucleus raphe interpositus. J Neurosci 1994; 14:20322046.
SFX [Context Link]
16. Moschovakis AK, Karabelis AB, Highstein M. Structurefunction relationships in the
primate superior colliculus. II. Morphological identity of presaccadic neurons. J Neurophysiol
1988; 60:263302. SFX [Context Link]
17. Moschovakis AK, Scudder CA, Highstein M. A structural basis for Herings Law:
projections to extraocular motoneurons. Science 1990; 248:11181119. [Context Link]
18. Collewijn H, Erkelens CJ, Steinman RM. Binocular co-ordination of human vertical
saccadic eye movements. J Physiol (Lond) 1988; 404:183197. SFX Full Text [Context Link]
19. Bttner-Ennever JA, Horn AKE. Pathways from cell groups of the paramedian tracts to the
floccular region. Ann NY Acad Sci 1996; 781:532540. SFX [Context Link]
20. Suzuki Y, Bttner-Ennever JA, Straumann D, Hepp K, Hess BJM, Henn V. Deficits in
torsional and vertical rapid eye movements and shift of Listings plane after uni- and bilateral
lesions of the rostral interstitial nucleus of the medial longitudinal fasciculus. Exp Brain Res
1995; 106:215232. SFX [Context Link]
21. Helmchen C, Glasauer S, Bartl K, Bttner U. Contralesionally beating torsional nystagmus
in a unilateral rostral midbrain lesion. Neurology 1996; 47:482486. Ovid Full Text SFX
[Context Link]
22. Waitzman DM, Silakov VL, DePalma-Bowles S, Ayers A. Oculomotor effects of reversible
inactivation of the primate mesencephalic reticular formation (MRF): I. Hypermetric goal-

directed saccades. J Neurophysiol 2000 (in press). [Context Link]


23. Waitzman DM, Silakov VL, DePalma-Bowles S, Ayers A. Oculomotor effects of reversible
inactivation of the primate mesencephalic reticular formation (MRF): II. Hypometric vertical
saccades. J Neurophysiol 2000 (in press). [Context Link]
24. Cannon SC, Robinson DA. Loss of the neural integrator of the oculomotor system from
brain stem lesions in monkey. J Neurophysiol 1987; 57:13831409. [Context Link]
25. Kaneko CRS. Eye movement deficits following ibotenic acid lesions of the nucleus
prepositus hypoglossi in monkeys. II. Pursuit, vestibular, and optokinetic responses. J
Neurophysiol 1999; 81:668681. SFX [Context Link]
26. Crawford JD, Cadera W, Vilis T. Generation of torsional and vertical eye position signals by
the interstitial nucleus of Cajal. Science 1991; 252:15511553. SFX [Context Link]
27. King WM, Leigh RJ. Physiology of vertical gaze. In: Lennerstrand EG, Zee DS, Keller EL,
eds. Functional basis of ocular motility disorders. Oxford, UK: Pergamon, 1982: 267276.
[Context Link]
28. Zuk A, Gwyn DG, Rutherford JG. Cytoarchitecture, neuronal morphology and some efferent
connections of the interstitial nucleus of Cajal (INC) in the cat. J Comp Neurol 1982; 212:278
292. SFX [Context Link]
29. McCrea RA, Strassman A, Highstein SM. Anatomical and physiological characteristics of
vestibular neurons mediating the vertical vestibulo-ocular reflexes of the squirrel monkey. J
Comp Neurol 1987; 264:571594. SFX [Context Link]
30. Partsalis AM, Zhang Y, Highstein SM. Dorsal y group in the squirrel monkey. I. Neuronal
responses during rapid and long-term modifications of the vertical VOR. J Neurophysiol 1995;
73:615631. SFX [Context Link]
31. Kokkoroyannis T, Scudder CA, Balaban CD, Highstein M, Moschovakis AK. Anatomy and
Physiology of the primate interstitial nucleus of Cajal I. Efferents projections. J Neurophysiol
1996; 75:725739. SFX [Context Link]

32. Dalezios Y, Scudder CA, Highstein SM, Moschovakis AK. Anatomy and physiology of the
primate interstitial nucleus of Cajal. II. Discharge pattern of single efferent fibers. J
Neurophysiol 1998; 80:31003111. SFX [Context Link]
33. Lueck CJ, Hamlyn P, Crawford TJ, et al. A case of ocular tilt reaction and torsional
nystagmus due to direct stimulation of the midbrain in man. Brain 1991; 114:20692079. SFX
[Context Link]
34. Helmchen C, Rambold H, Fuhry L, Bttner U. Deficits in vertical and torsional eye
movements after uni- and bilateral muscimol inactivation of the interstitial nucleus of Cajal of
the alert monkey. Exp Brain Res 1998; 119:436452. SFX [Context Link]
35. Brandt T, Dieterich M. Vestibular syndromes in the roll plane: Topographic diagnosis from
brainstem to cortex. Ann Neurol 1994; 36:337347. SFX [Context Link]
36. Rambold H, Helmchen C. Unilateral muscimol inactivations of the interstitial nucleus of
Cajal in the alert rhesus monkey do not elicit seesaw nystagmus. Neurosci Lett 1999; 272:75
78. SFX [Context Link]
37. Demer JL, Oh SY, Poukens V. Evidence for active control of rectus extraocular muscle
pulleys. Invest Ophthalmol Vis Sci 2000 (In press). [Context Link]
38. Helmchen C, Rambold H, Bttner-Ennever JA, Bttner U. Vestibular influence on the
binocular control of vertical torsional nystagmus after unilateral lesions of the interstitial
nucleus of Cajal (IC) in the alert monkey. Soc Neurosci Abstr 1999; 25:664. SFX [Context
Link]
39. Halmagyi GM, Aw ST, Dehaene I, Curthoys IS, Todd MJ. Jerk waveform see-saw
nystagmus due to unilateral meso-diencephalic lesion. Brain 1994; 117:789803. SFX [Context
Link]
40. Helmchen C, Fuhry L, Rambold H, Bttner U. The effects of mesencephalic lesions on 3-D
eye movements. In: Fetter M, Haslwanter T, Misslich H, Tweed D, eds. Three-dimensional
kinematics of eye, head, and limb movements. Amsterdam, the Netherlands: Harwood
Academic Publishers, 1997: 237242. [Context Link]

41. Bttner-Ennever JA, Bttner U. The reticular formation. In: Bttner-Ennever JA, ed.
Neuroanatomy of the oculomotor system. Amsterdam: Elsevier, 1988: 119176. [Context Link]
42. Moschovakis AK, Scudder CA, Highstein M. The microscopic anatomy and physiology of
the mammalian saccadic system. Prog Neurobiol 1996; 50:133254. SFX [Context Link]
43. Horn AKE, Bttner-Ennever JA. Premotor control of the upper eyelid: afferent projections
to the M-group in macaque monkey. Soc Neurosci Abstr 1998; 24:1889. SFX [Context Link]
44. Partsalis AM, Highstein SM, Moschovakis AK. Lesions of the posterior commissure
disabled the vertical neural integrator of the primate oculomotor system. J Neurophysiol 1994;
71:25822585. SFX [Context Link]
45. Pasik P, Pasik T, Bender MR. The pretectal syndrome in monkeys. Disturbances of gaze and
body posture. Brain 1969; 92:521534. SFX [Context Link]
46. Tomlinson RD, Robinson DA. Signals in vestibular nucleus mediating vertical eye
movement in the monkey. J Neurophysiol 1984; 51:11211136. [Context Link]
47. Bttner-Ennever JA, Bttner U. Neuroanatomy of the ocular motor pathways. Baillieres
Clin Neurol 1992; 1:263287. SFX [Context Link]
48. Cremer PD, Migliaccio AA, Halmagyi GM, Curthoys IS. Vestibulo-ocular reflex pathways
in internuclear ophthalmoplegia. Ann Neurol 1999; 45:529533. SFX [Context Link]
49. Carpenter MB, Cowie RJ. Connections and oculomotor projections of the superior
vestibular nucleus and cell group Y. Brain Res 1985; 336:265287. [Context Link]
50. Evinger LC, Fuchs AF, Baker R. Bilateral lesions of the medial longitudinal fasciculus in
monkeys: effects on the horizontal and vertical components of voluntary and vestibular induced
eye movements. Exp Brain Res 1977; 28:120. SFX [Context Link]
51. Gamlin PDR, Gnadt JW, Mays LE. Lidocaine-induced unilateral internuclear
ophthalmoplegia: effects on convergence and conjugate eye movements. J Neurophysiol 1989;

62:8295. SFX [Context Link]


52. Ranalli PJ, Sharpe JA. Vertical vestibulo-ocular reflex, smooth pursuit and eyehead
tracking dysfunction in internuclear ophthalmoplegia. Brain 1988; 111:12991317. [Context
Link]
53. Bttner-Ennever JA, Horn AKE, Schmidtke K. Cell groups of the medial longitudinal
fasciculus and paramedian tracts. Rev Neurol (Paris) 1989; 145:533539. SFX [Context Link]
54. Leigh RJ, Seidman SH, Grant MP, Hanna JP. Loss of ipsitorsional quick phase of torsional
nystagmus with a unilateral midbrain lesion. J Vestib Res 1993; 3:115122. [Context Link]
55. Morrow MJ, Sharpe JA. Smooth pursuit initiation in young and elderly subjects. Vision Res
1993; 33:203210. SFX [Context Link]
56. Rottach KG, Von Maydell RD, Das VE, Zivotofsky AZ, DiScenna AO, Gordon JL, et al.
Evidence for independent feedback control of horizontal and vertical saccades from NiemannPick type C disease. Vision Res 1997; 37:36273638. SFX [Context Link]
57. Ranalli PJ, Sharpe JA, Fletcher WA. Palsy of upward and downward saccadic, pursuit, and
vestibular movements with a unilateral midbrain lesion: pathophysiologic correlations.
Neurology 1988; 38:114122. [Context Link]
58. Bogousslavsky J, Miklossy J, Deruaz JP, Regli F, Assal G. Unilateral left paramedian
infarction of thalamus and midbrain: a clinico-pathological study. J Neurol Neurosurg
Psychiatry 1986; 49:686694. [Context Link]
59. Hommel M, Bogousslavsky J. The spectrum of vertical gaze palsy following unilateral
brainstem stroke. Neurology 1991; 41:12291234. Ovid Full Text SFX [Context Link]
60. Bogousslavsky J, Regli F. Upgaze palsy and monocular paresis of downward gaze from
ipsilateral thalamo-mesencephalic infarction: a vertical one-and-a-half syndrome. J Neurol
1984; 231:4345. SFX [Context Link]
61. Deleu D, Buisseret T, Ebinger G. Vertical One-and-a-Half Syndrome. Supranuclear

downgaze paralysis with monocular elevation palsy. Arch Neurol 1989; 46:13611363. SFX
[Context Link]
62. Deleu D, Ebinger G. Vertical one-and-a-half syndrome. Clinical, oculographic and
radiologic findings. Neuro-Ophthalmology 1991; 11:99101. SFX [Context Link]
63. Thomke F, Hopf HC. Acquired monocular elevation paresis: An asymmetric upgaze palsy.
Brain 1992; 115:19011910. SFX [Context Link]
64. Ziffer AJ, Rosenbaum AL, Demer JL, Yee RD. Congenital double elevator palsy: Vertical
saccadic velocity utilizing the scleral search coil technique. J Pediatr Ophthalmol Strabismus
1992; 29:142149. SFX [Context Link]
65. Wiest G, Baumgartner C, Schnider P, Trattnig S, Deecke L, Mueller C. Monocular elevation
paresis and contralateral downgaze paresis from unilateral mesodiencephalic infarction. J
Neurol Neurosurg Psychiatry 1996; 60:579581. SFX [Context Link]
66. Hriso E, Masdeu JC, Miller A. Monocular elevation weakness and ptosis: an oculomotor
fascicular syndrome? J Clin Neuroophthalmol 1991; 11:111113. Ovid Full Text SFX [Context
Link]
67. Gauntt CD, Kashii S, Nagata I. Monocular elevation paresis caused by an oculomotor
fascicular impairment. J Neuroophthalmol 1995; 15:1114. Ovid Full Text SFX [Context Link]
68. Castro O, Johnson LN, Mamourian AC. Isolated inferior oblique paresis from brain-stem
infarction: Perspective on oculomotor fascicular organization in the ventral midbrain
tegmentum. Arch Neurol 1990; 47:235237. SFX [Context Link]
69. Lessell S. Supranuclear paralysis of monocular elevation. Neurology 1975; 25:11341136.
Ovid Full Text SFX [Context Link]
70. Schmidtke K, Bttner-Ennever JA. Nervous control of eyelid function. Brain 1992;
115:227247. [Context Link]
Key words: Saccades; Vestibulo-ocular reflex; Interstitial nucleus of Cajal.

IMAGE GALLERY
Select All

Figure 1

Figure 2
Table 1

Back to Top

S-ar putea să vă placă și