Unilateral internuclear ophthalmoplegia

Unilateral internuclear ophthalmoplegia The lack of inhibitory involvement in medial rectus muscle activity Guntram Kommerell Inability to adduct the left eye beyond the midline was found in a patient after brain stem infarction. Pursuit movements, refixation saccades, and the vestibulo-ocular reflex were restricted to the temporal hemifield. Only near convergence led to an adduction of 15° in the left eye. Nasally directed saccades were slow, and temporally directed saccades showed glissadic back-drift. Electromyography revealed a lack of burst activity in the left medial rectus; however, inhibition of the left (homolateral) and the right (contralateral) medial rectus muscles during off-saccades and in off-positions of gaze was normal. These observations suggest that inhibition for medial rectus motor neurons is not mediated by the medial longitudinal fasciculus. The findings are compatible with the neurophysiological data, which show that reductions of activity in the medial rectus motoneurons are largely caused by disfacilitation via internuclear neurons rather than by inhibition. (INVEST OPHTHALMOL Vis SCI 21:592-599, 198IJ Key words: internuclear ophthalmoplegia, medial longitudinal fasciculus, saccades, electromyography, inhibition of medial rectus

I n previous electromyographic studies of internuclear ophthalmoplegia (INO), incomplete inhibition of the involved medial rectus eye muscle was found.1"4 Furthermore, impaired inhibition in the antagonist medial rectus has been inferred from abnormal abducting saccades.5' 6 However, in the patient to be presented, a marked unilateral INO was not associated with impaired inhibition of either the homolateral or the contralateral medial rectus, and the abnormalities of adducting and abducting saccades were explained exclusively on the basis of insufficient excitation of the involved medial rectus.

From the Universitats-Augenklinik Freiburg, F.R.G. This study was supported in part by the Deutsche Forschungsgemeinschaft, S.F.B. 70, B4. Reprint requests: Prof. Dr. med. G. Kommerell, Universitats-Augenklinik, D-7800 Freiburg, West Germany. 592

Case r e p o r t A 62-year-old woman (T. A. 020516) experienced sudden onset of dizziness, weakness in her right leg, and an apparent 45° tilt of the walls of her room. When seen in neuro-ophthalmological consultation 10 days later, she complained of diplopia in the right field of gaze. At no time had she worn a patch. A left, unilateral INO was found, with stereopsis preserved in the left hemifield of gaze. General medical and neurological examination was unremarkable except for high blood pressure, slight dysdiadochokinesis on the left side, some tendency to deviate to the right on walking with closed eyes, and slight diffuse brain atrophy on computer tomography. Visual acuity, visual fields, pupillary reactions, lid movements, and muscle strength of the four extremities were all normal. A brain stem stroke involving predominantly the left medial longitudinal fasciculus (MLF) was diagnosed. The patient was observed in the hospital for 4 days; during this time her INO did not change. It took her 3 months to gradually re-

0146-0404/81/100592+08$00.80/0 © 1981 Assoc. for Res. in Vis. and Ophthal., Inc.

Volume 21 Number 4

gain normal eye movements (according to the patient's ophthalmologist). Ocular motor findings. Lateral gaze to the right did not induce adduction of the left eye beyond the midline, but upon convergence, adduction of about 15° was observed. In the left half of the field of gaze, the eyes were parallel. After interruption of fusion with a dark red glass in front of the right eye, which fixated a spot light 2.5 meters straight ahead, an exophoria of 9° was apparent in the left eye; this exophoria was promptly overcome after removal of the dark red glass. The eye movements were recorded on motion pictures and videotape. Electro-oculography (10 days after infarction) Methods. Electrodes were placed at the inner and outer canthi as well as above and below each eye. The leads were attached to a DC-amplifier whose filters were set at 70 Hz (—3 dB). Visual stimuli were displayed on a gray screen at a distance of 1 meter. The paretic left eye was occluded during recording sessions to prevent double vision in the right field of gaze. Results Horizontal eye movements SACCADES. Refixations were tested between three targets positioned 30° to the left, straight ahead, and 30° to the right. In addition, 10° saccades were tested along a series of nine targets between 40° on the left and 40° on the right. Of the left eye: Nasally directed saccades were very slow and were restricted to the temporal hemifield (Fig. 1 A and C). Temporally directed saccades were followed by a glissadic back-drift (Fig. 1 B and D). Of the right eye: Temporally directed saccades were followed by a small glissadic back-drift (Fig. I A and C); the initial saccade overshot most of the time but was occasionally on target or even short of the target. Adducting saccades were hypometric and required a correction (Fig. I B and D). DISSOCIATED NYSTAGMUS. Low-amplitude nystagmus of the abducted right eye was not present at gaze deviations less than 40° to the right. Very weak nystagmus of the abducted left eye was not present until the patient looked to the extreme left. When the fixation lights were switched off and the patient looked at the remembered position of targets

Internuclear ophthalmoplegia

593

in the dark, dissociated nystagmus to the right and to the left was still present b u t was weaker than during fixation. PURSUIT MOVEMENTS AND OPTOKINETIC NYSTAGMUS. Pendular pursuit of a target that oscillated at 0.3 H z between 15° on the left and 15° on the right was quite smooth in the right eye and was interrupted by only a few saccades. Movements of the left eye were much smaller and w e r e restricted to t h e temporal hemifields (Fig. 2, A). Optokinetic nystagmus, tested at a speed of 60°/sec, was completely normal in the right eye. In the left eye, nasally directed quick phases h a d a reduced velocity and stopped at the midline. Nasally directed slow phases also stopped at the midline (Fig. 2, B). VESTIBULO-OCULAR REFLEX. T h e p a t i e n t was

oscillated on a swivel chair between 20° on the left and 20° on the right at a frequency of 0.6 H z . W h e n t h e patient fixated a stationary target, the right eye performed perfect compensatory movements. This means that the gain of the vestibulo-ocular reflex plus smooth pursuit was 1.0. W h e n the light was switched off and the patient tried to look at the r e m e m b e r e d position of the target, t h e movements of the right eye became smaller, and t h e gain of t h e p u r e vestibulo-ocular reflex was 0.9. In the left eye, only half-wave rectified cycles were elicited because the left eye did not adduct beyond t h e midline (Fig. 2, C). There was no phase shift b e t w e e n the right and left eyes as measured at the peak left position. Phase shifts b e t w e e n body a n d eye movements could not b e determined because the oscillation of the swivel chair was not recorded. Vertical eye movements. Refixation saccades between 20° upward and the primary position as well as b e t w e e n 20° downward and the primary position showed glissadic overshoot of about 10%, b u t vertical nystagmus was not present except when the direction of gaze was elevated or depressed more than 40°. O n downward gaze, nystagmus was more pronounced in the left than in the right eye. Vertical optokinetic nystagmus was normal.

594

Invest. Ophthalmol. Vis. Set. October 1981

Kommerell

RE LE

1sec

RMR LMR

1 sec

"V-

RMR

LM R

Fig. 1. INO of the left eye. The left medial rectus (LMR) shows lack of bursts during onsaccades. Phasic inhibition of both medial recti during off-saccades is unimpaired. In (B), the reduced activity of the right medial rectus (RMR) immediately after the burst is probably an artefact due to a displacement of the electrode during the saccade. To avoid effects from the mechanical load of electrodes planted in the muscles, eye movement tracings were not recorded simultaneously with the electromyograms, but corresponding samples were mounted together with the same time scale. Eye movements to the right are recorded as upward deflections.

Volume 21 Number 4

Electromyography (12 days after infarction) Methods. Concentric 30-gauge needle electrodes were inserted into both medial recti after topical anesthesia of the conjunctiva. The examiner's index fingers were used asfixationtargets at a distance of about 50 cm in front of the patient. The right eye was usedforfixationand the left eye was occluded. Signals of both medial recti were recorded simultaneously on magnetic tape and later ink-written on a Mingograph (Siemens) at various paper speeds. Results. The electromyographic pattern of the right medial rectus was completely normal (Figs. 1 and 3). Burst activity during onsaccades and phasic inhibition during offsaccades were clearly seen. Tonic inhibition upon gaze to the extreme right was marked, although very few motor units could still be detected. In the left medial rectus, there was a complete lack of burst activity during on-saccades. Tonic discharge gradually developed when the eye was led from the left to the primary position, but it increased only slightly as the fixating contralateral eye passed the midline toward the right. Some convergence input may have intervened during ductions to the right, since the target was presented relatively close (at 50 cm). Inhibition is shown in more detail in Fig. 3. Most of the firing of the left medial rectus ceased abruptly at about the same time, and decreased activity lasted about as long as a burst was seen in the contralateral medial rectus. The firing pattern of the left medial rectus during five off-saccades is depicted in Fig. 3, D. Phasic inhibition was always quite dramatic, and only a few units, mostly low-amplitude, remained active. Braking bursts, as postulated by Feldon et al.6 on the basis of eye movement tracings, were not found. Phasic and tonic inhibition in the paretic left medial rectus resembled that of the clinically normal right medial rectus. Discussion This patient had a unilateral INO that prevented adduction of the left eye beyond the midline. The classic presentation strongly suggests infarction of the left MLF.

Internuclear ophthalmoplegia

595

Electromyography. In the involved medial rectus, a complete lack of bursts during onsaccades was found, although tonic activation was not absent, being merely limited in intensity; earlier findings were thus confirmed. 1"~4' 7 This peculiar pattern of defective activation is surprising because the internuclear neurons in the MLF have been shown to carry tonic as well as phasic signals. 8 " 10 To solve this problem, it has been suggested that partially damaged MLF fibers might transmit low-frequency signals better than those of high frequency.11' 12 Another possibility is that one of the extra-MLF pathways13 might still conduct tonic—but not phasic—input to the medial rectus. Reisine and Highstein14 recently found the ascending tract of Deiters to relay tonic eye position to the medial rectus subnucleus, but not saccadic bursts; preservation of this pathway might indeed explain the remaining tonic modulation in the electromyogram of patients with MLF lesions. Inhibition of the homolateral, as well as of the contralateral, medial rectus was found to be normal in our patient. Most of the firing ceased abruptly during off-saccades and in off-positions of gaze (Fig. 3). The small amount of discharge that persisted during phasic and tonic inhibition must not be regarded as pathological. Phasic inhibition has been shown to be incomplete in predominantly tonic motor units15; in extreme offgaze, the small muscle fibers are still active in the electromyogram of normal subjects. 16 In view of the electrophysiological data available, our finding of unimpaired inhibition in a patient with unilateral INO is not surprising. There is now consensus in the literature that INO is caused by a lesion of internuclear neurons that originate in the abducens nucleus, cross the midline, ascend in the opposite MLF and terminate monosynaptically on medial rectus motoneurons without decussating a second time. 17 The fibers in the MLF that are concerned with horizontal gaze already contain the same phasic and tonic signals that are characteristic of the cells to which they project, i.e., the medial rectus motoneurons. 8 ' 9 Electrical stimulations of

596

Invest. Opkthabnol. Vis, Sci. October 1981

Kommerell

LE RE J4O° Is

LE Fig. 2. A, Pursuit of a target that oscillates in the horizontal plane. The left eye (covered) responds only with small movements in its temporal hemifield. Eye movements to the right are recorded as upward deflections. B, Optokinetic nystagmus elicited by a Rillfieldof 7° black and white stripes. The pattern moves at a velocity of 60°/sec, first to the left, then to the right. In the left eye (covered), adducting quick and slow phases level off at the midline. C, Horizontal vestibulo-ocular reflex in complete darkness. The left eye shows only half-wave rectified cycles.

the MLF17> 1S and of the vestibular nerve 19 produce only excitatory, not inhibitory, postsynaptic potentials in medial rectus motoneurons. Therefore decreased activities of medial rectus motoneurons during off-saccades and in off-positions of gaze seem largely caused by a decrease of the excitatory input from MLF fibers, i.e., by "disfacilitation" rather than by inhibitory postsynaptic potentials. This assumption is supported by intracellular recordings from medial rectus motoneurons during vestibularly induced off-saccades. The sudden stopping of motoneuron firing was found to be due to disfacilitation, whereas inhibitory postsynaptic potentials contributed only later to the hyperpolarization of the cell.20 In addition, field potentials reflecting the extracellular counterpart of inhibitory postsynaptic potentials are lacking in the medial rectus subnucleus, as opposed to the abducens nucleus where such field potentials are present during vestibulary induced off-saccades.18

Defective disfacilitation of the medial rectus in patients with INO, as described by previous authors, 1 " 4 might be due to damage to the inhibitory input of the internuclear cells in the abducens nucleus. These cells project through the MLF to the medial rectus motoneurons. 10 ' Z1 Total abolition of versional modulation in the medial rectus (case 3 of Orlowski et al., 1 case 1 of Loeffler et al., 2 and case 1 in Pierrot-Deseilligny et al.22) would suggest a complete interruption of all the pathways that carry versional signals to the medial rectus subnucleus, i.e., at least of the MLF and of the ascending tract of Deiters. The discharge of the medial rectus that remains could come from the vergence system. In case 3 of Orlowski et al. l and in case 1 of Loeffler et al.,2 the amount of the remaining discharge was rather high. As a speculative explanation for their findings I would consider that the medial rectus motoneurons that are deprived of their versional input develop a denervation supersensitivity to the

Volume 21 Number 4

Internuclear ophthalmoplegia 597

J RMR LMR

LMR

LMR

Fig. 3. INO of the left eye. Electron! ybgrams show abrupt phasic inhibition of the right medial rectus (RMR) in (A) and of the left medial rectus (LMR) in (B) and during five consecutive saccades in (D). Tonic inhibitions of both medial recti during extreme right and left gaze, respectively, are depicted in (C). unimpaired input from the vergence system.

Interpretation of saccadic abnormalities Saccades of the paretic left eye. The reduced velocity of relaxations from the left to straight ahead (Fig. 1, A) can easily be explained by the lack of burst activity in the left medial rectus. Refixations from straight ahead to the left resulted in saccades with a glissadic backdrift (Fig. 1, B). This glissadic back-drift is due to an inadequately small negative step of tension in the antagonistic medial rectus. The saccade is a little slow because of two factors. First, since tonic activity of the medial rectus is too low before the eye movement begins, its reduction to a normal level during the saccade constitutes an inadequately small negative step. Second, the reduced tonic activity of the medial rectus shifts the starting point of the saccade to the temporal side, so that the saccade does not begin from the midline but rather from 10° on the left. This results in some slack of the lateral rectus and, accordingly, in an impairment of its generated force. Refixations from straight ahead to the right

(Fig. 1, C) are barely discernible in the left eye because medial rectus firing has already reached its upper limit. As for refixations from the right to straight ahead (Fig. 1, D), the initial saccade of the left eye can be explained by burst activity of the lateral rectus and by phasic inhibition of the medial rectus. After the saccade, the eye drifts back because tonic innervation in both eye muscles has not changed much. In the lateral rectus, the tonic step is expected to be small when the muscle receives a program for a saccade in its off-field of gaze.22' 23 In the medial rectus, the tonic step is small because the muscle does not start at the appropriate high level of activity. Even though firing drops to a nearly normal level, the change of tonic activity must be small. Thus abnormal on- and off-saccades principally resemble the patterns that have been found in patients with peripheral eye muscle palsies23"26 and can be explained exclusively on the basis of insufficient excitation of the medial rectus. Impaired inhibition need not be invoked. Saccades of the (contralateral) right eye.

Invest. Ophthalmol. Vis. Sci. October 1981

598 Komnierell

Abducting saccades (Fig. 1, A and C) were consistently followed by a glissadic backdrift, regardless of their size and the field of gaze in which saccades were executed. This "glissadic dysmetria" can be interpreted as a so-called pulse-step mismatch in which the pulse (consisting of agonist burst and antagonist inhibition) lasts too long. Thus these findings are compatible with an explanation from Baloh et al.," who assume that the excessive pulse is an adaptive phenomenon with which the central nervous system reacts to the slowness of abducting saccades in the fellow eye. Patients with peripheral eye muscle palsies who use their paretic eye for fixation, exhibit an increase in pulse activity to compensate for the slow saccades that occur when the paretic muscle is the agonist.27 The adaptive programming is bilateral and results in glissadic overshoot of the nonparetic, nonfixating eye. In the present case of INO, the patient did not exclusively use her paretic eye for fixation; rather, she was binocular most of the time. Therefore the adaptive basis of glissadic overshoot in her right eye may still be questioned. If occlusion of the paretic eye could be shown to cause disappearance of the glissadic overshoot, this glissadic overshoot could, indeed, be regarded as a secondary manifestation of central nervous system plasticity. Further ocular motor abnormalities. The impairment of smooth pursuit and of the vestibulo-ocular reflex in our patient's left eye corresponds well with the disturbances found in monkeys after transection of the MLF. 28 The fact that the nystagmus elicited in our patient on vertical gaze was only weak may have been due to the availability of still one MLF. Bilateral interruption of the MLF in monkeys resulted in severe gaze-evoked nystagmus.28 Although our patient had a very severe INO, a continuous nystagmus of the abducted eye appeared only when the direction of gaze was deviated more than 40° from the primary position. This discrepancy lends support to earlier assumptions that dissociated nystagmus, if present in INO, does not occur because of the MLF lesion but

rather constitutes a "gaze-evoked nystagmus" generated elsewhere in the brain stem. 29 " 32 I thank Jean A. Biittner-Ennever, Ph.D., of the Institut fur Hirnforschung, Zurich, and Dr. med. Volker Henn of the Neurologisches Universitatsspital, Zurich, for their helpful criticism of this paper. I gratefully acknowledge the technical assistance of Giinther Schaubele, Ph.D., Ruth Lohr, and Markus Hiittel, and I am grateful to Mrs. Carol Wanske for help in preparing the manuscript. REFERENCES 1. Orlovvski WJ, Slomski P, and Wojtowicz ST: Bielschowsky-Lutz-Syndrome. Am J Ophthalmol 59:416, 1965. 2. Loeffler JD Hoyt WF, and Slatt B: Motor excitation and inhibition in internuclear palsy. An electromyographic study. Arch Neural 15:664, 1966. 3. Gonzalez C and Reuben RM: Ocular electromyography in the syndrome of median longitudinal fasciculus: patterns of inhibition and excitation. Am J Ophthalmol 64:916, 1967. 4. Huber A and Esslen E: Intemucleare Ophthalmoplegie. Ophthalmologies 165:320, 1972. 5. Pola J and Robinson DA: An explanation of eye movements seen in internuclear ophthalmoplegia. Arch Neurol 33:447, 1976. 6. Feldon StE; Hoyt WF, and Stark L: Disordered inhibition in internuclear ophthalmoplegia. Analysis of eye movement recordings with computer simulations. Brain-103:113, 1980. 7. Breinin GM: Electromyography—a tool in ocular and neurologic diagnosis. III. Supranuclear mechanisms. Arch Ophthalmol 59:177, 1958. 8. King WM, Lisberger-SG, and Fuchs AF: Responses of fibers in medial longitudinal fasciculus (MLF) of alert monkeys during horizontal and vertical conjugate eye movements evoked by vestibular or visual stimuli. J Neurophysiol 39:1135, 1976. 9. Pola J and Robinson DA: Oculomotor signals in medial longitudinal fasciculus of the monkey. J Neurophysiol 41:245, 1978. 10. Delgado-Garcia J, Baker R, and Highstein SM: The activity of internuclear neurons identified within the abducens nucleus of the alert cat. In Control of Gaze by Brain Stem Neurons, Baker R and Berthoz A, editors. New York, 1977, Elsevier/North-Holland Biomedical Press, pp. 291-300. 11. Baloh RW, Yee RD, and Honrubia V: Internuclear ophthalmoplegia. I. Saccades and dissociated nystagmus. Arch Neurol 35:484, 1978. 12. Baloh RW, Yee RD, and Honrubia V: Internuclear ophthalmoplegia. II. Pursuit, optokinetic nystagmus, and vestibulo-ocular reflex. Arch Neurol 35: 490, 1978. 13. Biittner-Ennever JA: Organization of reticular projections onto oculomotor neurons. In Reflex Control of Posture and Movement, Granit R and Pompeiano O, editors. Prog Brain Res 50:619, 1979.

Volume 21 Number 4

14. Reisine H and Highstein SM: Eye position and head velocity signals recorded from ascending tract of Deiters neurons in the cat. In Progress in Oculomotor Research, Fuchs A and Becker W, editors. New York, 1981, Elsevier/North-Holland Biomedical Press. 15. Henn V and Cohen B: Eye muscle motor neurons with different functional characteristics. Brain Res 45:561, 1972. 16. Collins CC: The human oculomotor control system. In Basic Mechanisms of Ocular Motility and their Clinical Implications, Lennerstrand G and Bachy-Rita P, editors. Oxford, 1975, Pergamon Press, p. 149. 17. Highstein SM and Baker R: Excitatory termination of abducens internuclear neurons on medial rectus motoneurons; relationship to syndrome of internuclear ophthalmoplegia. J Neurophysiol41:1647,1978. 18. Nakao S and Sasaki S: Excitatory input from interneurons in the abducens nucleus to medial rectus motoneurons mediating conjugate horizontal nystagmus in the cat. Exp Brain Res 39:23, 1980. 19. Baker R and Highstein SM: Vestibular projections to medial rectus subdivision of oculomotor nucleus. J Neurophysiol 6:1629, 1978. 20. Furuya N, Saito A, Ishikawa M, and Suzuki J: Synaptic events in cat medial rectus motoneurons during vestibular nystagmus. In Integrative Control Functions of the Brain, Ito M, Tsukahara N, Kubota K, and Yagi K, editors. Tokyo, 1979, Kodansha, vol. II, pp. 197-199, cited in ref. 18. 21. Grantyn A, Grantyn R, Gaunitz U, and Robine K-P: Sources of direct excitatory and inhibitory inputs from the medial rhombencephalic tegmentum to lateral and medial rectus motoneurons in the cat. Exp Brain Res 39:49, 1980. 22. Pierrot-Deseilligny C, Rigolet MH, and Chain F: Etude electromyographique de deux cas d'ophtalmoplegie internucleaire: deductions physiopathologiques. Rev Neurol 135:143, 1979. 23. Theopold H and Kommerell G: Phasische und tonische Funktion der Augenmuskeln. Untersuchun-

Internuclear ophthalmoplegia 599

24.

25.

26.

27.

28.

29.

30.

31.

32.

gen an Patienten mit Okulomotorius- oder Abducens-Paralyse. Albrecht von Graefes Arch Klin Exp Ophthalmol 192:247, 1974. Bielschowsky A: Die Motilitiitsstorungen der Augen. In Handbuch der Augenheilkunde, GraefeSamisch, editors. Leipzig, 1907, Engelmann, 2. Auflage Bd 8, I.Abt,Kap XI, Nachtrag I, p. 33. Scott AB: Extraocular muscle forces in strabismus. In The Control of Eye Movements, Bach-y-Rita P, Collins CC, and Hyde JE, editors. New York, 1971, Academic Press, Inc., p. 340. Kommerell G: Clinical clues for the organization of horizontal quick eye movements and subsequent periods of fixation. In Basic Mechanisms of Ocular Motility and their Clinical Implications, Lennerstrand G and Bach-y-Rita P, editors. Oxford, 1975, Pergamon Press, pp. 326-329. Kommerell G, Olivier D, and Theopold H: Adaptive programming of phasic and tonic components in saccadic eye movements. Investigations in patients with abducens palsy. INVEST OPHTHALMOL 15:657, 1976. Evinger LC, Fuchs AF, and Baker R: Bilateral lesions of the medial longitudinal fasciculus in monkeys: effect on the horizontal and vertical components of voluntary and vestibular induced eye movements. Exp Brain Res 28:1, 1977. Bender MB and Weinstein EA: Dissociated monocular nystagmus with paresis of horizontal ocular movements. Arch Ophthalmol 21:266, 1939. Jung R and Kornhuber HH: Results of electronystagmography in man: the value of optokinetic, vestibular, and spontaneous nystagmus for neurologic diagnosis and research. In The Oculomotor System, Bender MB, editor. New York, 1964, Harper & Row, Publishers, p. 435. Kommerell G: Die internukleare Ophthalmoplegie. Nystagmographische Analyse. Klin Monatsbl Augenheilkd 158:349, 1971. Pierrot-Deseilligny C and Chain F: L'ophtalmoplegie internucleaire. Rev Neurol 135:485-513, 1979.