Journal of Neuro-Ophthalmology, December 1987, Volume 7, Issue 4

Midbrain-diencephalic horizontal gaze paresis.

Four patients with predominantly right-sided infarcts at the midbrain-diencephalic junction had an ipsilateral oculomotor palsy and contralateral gaze palsy. The gaze palsy to the left was supranuclear in nature, because the restriction of abduction was overcome by the oculocephalic reflex. However, it was masked on the right eye by the third nerve palsy. In all cases upgaze was impaired bilaterally, and infraduction was absent on the right. The restriction of abducting saccades remained for months, considerably longer than with supranuclear cortical lesions, suggesting that the damage extended beyond involvement of the frontopontine pathways. These cases represent clinical instances of a recently described experimental syndrome of impaired horizontal saccadic eye movements with lesions at the paramedian midbrain-diencephalic junction.

Journal of CliniCilI Neuro-ophthalmology 7(4): 227-234. 1987. « 1987 Raven Press, Ltd., New York Midbrain-Diencephalic Horizontal Gaze Paresis Joseph c. Masdeu, M.D., and Michael Rosenberg, M.D. Four patients with predominantly right-sided infarcts at the midbrain-diencephalic junction had an ipsilateral oculomotor palsy and contralateral gaze palsy. The gaze palsy to the left was supranuclear in nature, because the restriction of abduction was overcome by the oeuloce­phalic reflex. However, it was masked on the right eye by the third nerve palsy. In all cases upgaze was im­paired bilaterally, and infraduction was absent on the right. The restriction of abducting saccades remained for months, considerably longer than with supranuclear cortical lesions, suggesting that the damage extended beyond involvement of the frontopontine pathways. These cases represent clinical instances of a recently de­scribed experimental syndrome of impaired horizontal saccadic eye movements with lesions at the paramedian midbrain-diencephalic junction. Key Words: Eye move­ments- Mesencephalon - Ophthalmoplegia - Sac­cades- Thalamus-Tomography, x-ray computed. From the Departments of Neurology, St. Vincent's Hospital and New York University School of Medicine, New York, New York (J.C.M.) and the Department of Ophthalmology, Rush­Presbyterian Medical College, Chicago, lIIinois (M.R.) Address all correspondence and reprint request to Dr. J. C. Masdeu at the Department of Neurology, St. Vincent's Hos­pital and Medical Center of New York, 153 West 11th Street, New York, NY 10011, U.S.A. 227 Although it is known that mesencephalic lesions can cause disorders of horizontal gaze (1) and some attention has been paid recently to the pres­ence of abducens weakness with fuller adduction (midbrain pseudo-sixth) (2). the more common finding of a horizontal gaze palsy contralateral to a lesion at the midbrain-diencephalic junction has been seldom documented. Recently, Zackon and Sharpe (3) described in detail two cases of tumors in this location resulting in horizontal gaze paresis. In single instances, the presence of hori­zontal gaze paresis with infarcts in this location has been summarily mentioned (4-6). Experi­mental data support the possibility of horizontal gaze deficit originating from lesions at the upper midbrain and posterior-medial thalamus (7- 9). This report describes, in four cases, the clinical syndrome of horizontal gaze paresis with infarc­tion at the midbrain-diencephalic junction. CASE REPORTS Case 1 This 59-year-old man noticed the sudden onset of incoordination of his left limbs and diplopia. He remained fully alert. Temporal orientation, mildly impaired initially, became normal in 1 week. On examination he was found to have slight bilateral lid retraction, more marked on the left (Fig. 1). In room light, pupillary diameter was 5 mm on the right and 3.5 mm on the left. Response to light was brisk on the left but sluggish on the right. Neither pupil constricted on attempts to focus on near objects. In primary position the eyes were skewed, the right eye being exotropic and the left eye, hypotropic. Upward gaze was absent bilater­ally. Downward movement of the right eye was limited to 10°, whereas the left eye moved fully downward. On watching a downgoing optoki­netic tape the left eye developed retraction and oc­casional adducting beats. Vertical eye excursions 228 J. C. MASDEU AND M. ROSENBERG FIG. 1. Case 1. From top to bottom, attempted up­gaze, downgaze, gaze to the right, and gaze to the left. were not enhanced by tracking, optokinetic stimu­lation, or the oculocephalic reflex. Right lateral gaze was normal. Adduction with the right eye was limited to <5°. Abduction with the left eye was more restricted than when the photograph for Fig. 1 was obtained, 1 month after the onset of ophthalmoparesis. Voluntary saccades were ab­normally shortened and slow. Pursuit movements were more readily elicitable than voluntary sac­cades but resulted in a similar range of abduction. This restriction persisted when one eye was cov­ered. No pupillary constriction occurred on at­tempted lateral gaze. On abduction the left eye de­veloped clockwise rotary nystagmoid jerks. Hori­zontal optokinetic nystagmus and nystagmus induced by the application of 25°C air for 60 s to the right ear showed slowing of the corrective beat to the left in both eyes. The limitation of left eye abduction could be overcome by the oculocephalic reflex (Fig. 1). The patient had terminal dysmetria .•4 '.",~ •• "f his ld~ limbs, loss of pin sensation on the left side, including the face, and a left Ba­binski. Facial strength and corneal reflexes were normal bilaterally. Palatal myoclonus was absent. On follow-up examination 4 months after the stroke, right eye adduction had improved to -15°, and left eye abduction was still restricted to -20°. High resolution (5 mm) computed tomography (CT) was performed in a GE 8800 unit 6 weeks after the stroke. It showed a tiny paramedian mid­brain infarct, involving mainly the retronigral teg­mentum on the right, including the right red nu­cleus and possibly part of the oculomotor nuclear complex (Fig. 2). The infarct extended to the most medial portion of the crus cerebri. The rest of the brainstem appeared normal. Case 2 Four months before the photographs in Fig. 3 were obtained, this 64-year-old hypertensive non­alcoholic man awoke disoriented, unable to walk, and seeing double. When examined 1 month later he was still disoriented in time. He had mild ptosis on the right. In the ensuing weeks, ptosis on the right became less pronounced; retraction of both eyelids, more pronounced on the left, became ap­parent. His right pupil was slightly larger than the left, but both reacted well to light. In the primary position the right eye was positioned higher than the left. This relationship was maintained, regard­less of the eye the patient used to fixate with. The eyes were slightly divergent. Upgaze was mark­edly restricted on both sides, but the right eye cov­ered an arc of -10° and became exotropic (Fig. 3). Downgaze, markedly restricted on the right side, was full on the left. Right lateral gaze was normal. Left lateral gaze was restricted; the right eye barely adducted. Leftward saccades were hypometric and slow. Tracking movements were jerky and as limited as the saccades. This limitation persisted when one eye was covered. No pupillary constric­tion occurred on attempted lateral gaze. The re­striction of abduction with the left eye was greater than when the photographs were obtained and could be overcome by the doll's eye maneuver (Fig. 3, bottom frame). The impairment of sac­cades was also evident in the quick phases of op­tokinetic nystagmus and nystagmus induced by cooling of the labyrinth. Convergence could not be elicited. On abduction the left eye developed clockwise rotary nystagmus. The right eye showed clockwise nystagmus on attempted down gaze. Corneal reflexes and facial movements were normal. He had a very mild left hemiparesis and a broadbased gait, but finger-nose and heel-shin MIDBRAIN HORIZONTAL GAZE PARESIS 229 B FIG. 2. Case 1. Thin-section (5 mm) computed tomog­raphy scan (A) shows a right paramedian midbrain in­farct (arrow). vertical hatching in the diagram (B). tests were done well. Palatal myoclonus was ab­sent. The deficit in ocular motility persisted 6 months after onset, with only slight improvement of left eye abduction. Four months after the stroke, thin-section CT, performed in a GE 8800 unit, showed bilateral slit-like rostrocaudal infarcts FIG. 3. Case 2. From top to bottom. attempted up­gaze, gaze to the right. and gaze to the left. Restricted abduction of the left eye persisted despite covering the right eye. Full abduction of the left eye was ob­tained with the oculocephalic reflex (frame at the bottom). in the medial aspect of both thalami, the right­sided one clearly extending into the paramedian preaqueductal tegmentum, at the midbrain-dien­cephalic junction (Fig. 4). Specific structures likely to have been involved included the dorsomedial and the midline thalamic nuclei bilaterally and, at least on the right side, the rostral interstitial nu­cleus of the medial longitudinal fasciculus, the in­terstitial nucleus of Cajal, the nucleus of Darks­chewitsch, and possibly the oculomotor nuclear I (11II N,·//Y"·"""tllal,,,,,I, Vol. ~, Nl) ~, 1987 230 J. C. MASDEU AND M. ROSENBERG FIG. 4. Case 2. Thin-section CT scans, parallel to Reid's baseline, show bilateral slit-like infarcts (arrows) in the paraventricular area of the thalami at a low-thalamic level (plane of the anterior and haben­ular commissures) (A) and at the midbrain-dience­phalic junction (B, arrow). The infarct on the right side is larger, particularly at the level of the lower section. complex. No lesions were noted in the lower mid­brain or pons. Case 3 This 45-year-old hypertensive woman, on birth control pills, was at work when she experienced sudden diplopia, numbness of the left side of her face, and nausea. She remained fully awake and orientpd When t'xamined 10 days after the onset of her diplopia, she had mild bilateral retr~ction of her eyelids, more pronounced on the left SIde. The right pupil was irregular and slightly larger. than the left. Response to light was preserved bilater­ally. Convergence was absent. Upg~ze ~as ma~k­edly restricted. A downgoing optokmetic tape m­duced adducting saccades of the left eye. Down­ward movements were absent on the right, but only minimally restricted on the left. Right lateral gaze was full. Adduction of the right eye was l~m­ited to -15°, and abduction of the left eye was lIm­ited to -30°. Leftward quick phases of optokinetic nystagmus were slower and of smaller amplitude on the right side. Contraction of the left orbicularis oris was weak when the patient smiled, but her mouth moved symmetrically when she talked. Corneal reflexes were symmetrical. There was a slight terminal dysmetria in movements per­formed with the left hand. Computed tomography showed an infarct in the most anterior portion of the dorsomedial nucleus of the right thalamus. The infarct extended caudally through the preru­bral field of Forel and into the tegmentum of the midbrain, at the midbrain-diencephalic junction (Fig. 5). Here, the structures visibly damaged were similar to the one affected in Case 2 but were re­stricted to the right side. Case 4 This 67-year-old hypertensive man had a por­cine aortic value implant and a coronary artery by­pass graft. One day after experiencing a fleeting episode of blindness, he suddenly became unre­sponsive. He regained consciousness over a 2-day period. When examined 1 week after onset, he was disoriented but could follow commands. There was bilateral ptosis, almost total on the right and partial on the left. The pupillary diameter was 8 mm on the right and 6 mm on the left. Neither pupil reacted to light. There was no abnormal dis­placement of the eyes in primary position. Both voluntary saccades and tracking eye movements were absent in the vertical plane. With the help of the vestibulo-ocular reflex, a 20° downward dis­placement of the left eye could be elicited when the patient attempted to look down. Abduction of the right eye was full, but adduction was absent. Both adduction and abduction of the left eye were limited to -30°, but the limitation in abduction could be overcome with the doll's eye maneuver. Convergence was absent. In addition to the ocular motor weakness, the patient had a partial loss of vision in both hemifields and bilateral dysynergia of truncal and distal movements. Computed to- MIDBRAIN HORIZONTAL GAZE PARESIS 231 FIG. 5. Case 3. Computed tomographic scans, 15° to Reid's baseline, show a right-sided infarct (arrow) at a midthalamic level (A) and at the midbrain-dience­phalic junction (B, arrow). mography showed recent infarction affecting both paramedial thalamomesencephalic regions, the left medial occipitotemporal region, and the right medial occipital lobe. DISCUSSION Unlike vertical gaze disorders, the impairment of horizontal eye movements that may follow midbrain infarcts has received little attention (10). The neuro-ophthalmologic examination is often limited, because the patient is comatose and un­able to generate saccades (11). The four patients described here had lesions that were small enough to spare consciousness at the time of the ocular motor examination. High resolution CT performed >1 month after the ictus allowed visualization of the small infarcts in Cases 1 and 2. Recently, Hochman and colleagues (12) reported what they considered to be the second case of mesencephalic infarct shown by CT and emphasized the impor­tance of thin CT section for the visualization of these minute lesions. In 1912, Claude (5) described a patient with a complete right third nerve palsy, weakness of gaze to the left, and a left-sided hemiasynergia and ataxia. In a clinical report of asterixis related to midbrain infarction Bril and co-workers (4) men­tioned that the patient had a complete right third nerve palsy and paralyzed leftward gaze. His left eye abducted fully with oculocephalic stimulation. Under the heading of "nuclear syndrome of the oculomotor nerve," Pierrot-Deseilligny and others (6) described two patients with an incomplete ocu­lomotor paresis on the right side and a paresis of upward gaze on the left side. The resemblance to the cases reported here is remarkable. The second of their patients had impaired abduction of the left eye, which could be overcome by the vestibulo-oc­ular reflex. More recently, Zackon and Sharpe (3) described two patients with unilateral tumors of the paramedian midbrain tegmentum, who had an ipsilateral third and contralateral paresis of sac­cadic eye movements. All the patients reported here shared remark­ably similar neuro-ophthalmologic findings. The right eye exhibited a more or less complete paresis of the muscles innervated by the oculomotor nerve. Elevation of the left eye was impaired and the left eye tended to be tonically deviated down­ward. This deviation results from a paretic supe­rior rectus with normal function of the inferior rectus. The clockwise nystagmoid jerks on abduc­tion of the left eye indicate that the inferior oblique was functional. Thus, except for the superior rectus, all the muscles of the left eye innervated by the oculomotor nerve were spared in our first three patients. According to experimental data in the monkey (13) and cat (14), the superior rectus of the left eye is innervated by fibers originating in the medial cell column of the right oculomotor nu­cleus. The fibers destined to the superior rectus decussate at the level of the nucleus itself. In addi­tion to involving the perikaria innervating the left superior rectus, a right nuclear lesion would affect the decussating fibers destined to the right supe­rior rectus. Pierrot-Deseilligny and others (6) in­voked this anatomic arrangement to ascribe the 232 J. C. MASDEU AND M. ROSENBERG neuro-ophthalmologic findings in their cases to a nuclear lesion of the right oculomotor complex, with sparing of the left oculomotor nucleus. The clinical findings in our patients could be similarly explained. However, the presence of eyelid retrac­tion and of some other findings characteristic of Parinaud's syndrome in their two cases and in three of ours suggests the contribution of a supra­nuclear component. In our patients, although the lesions on CT seemed to affect the area of the ocu­lomotor nucleus on the right side, they probably involved, at least in the last three patients, other structures known to participate in the genesis of vertical saccadic eye movements, chiefly the ros­tral interstitial nucleus of the medial longitudinal fasciculus (15). Most remarkable was the impairment of abduc­tion in the eye contralateral to the lesion. It should not be confused with a pseudoabducens palsy re­sulting from convergence spasm (spasm of the near triad) (16,17). In such cases the pupils be­come smaller as the patient attempts to look sideways. Our patients' pupillary size remained unchanged on lateral gaze, even though their left pupil (except for case 4) reacted well to light. Also, in cases of convergence spasm, covering the ad­ducting eye often results in full abduction. This maneuver failed to increase the excursion of the abducting eye in our patients. The remote possi­bility of a true abducens palsy was discarded, be­cause full abduction could be obtained with the help of the vestibulo-ocular reflex. Several mechanisms could be invoked to explain the weakness of abduction in these patients. Under the designation of pseudo-6th nerve palsy, Fisher and others (18) described an ipsilateral ab­duction weakness that could be overcome by ice­water caloric stimulation. They had found this ab­normality in cases of cerebellar and thalamic­subthalamic hemorrhages (19). Two mechanisms have been postulated to explain the failure of the eye to abduct (2). First, it could be caused by fixa­tion with the hyperconvergent eye, because cov­ering the adducting eye may result in improved abduction. However, covering the adducting eye did not affect the extent of abduction in our cases. Second, it could be due to convergence vectors present in ther abducting eye, which would coun­teract normal abduction. Caplan (2) pointed out that in cases of pseudo-six, if the abducting eye is watched carefully, convergence or adducting jerks are often noted. Careful inspection, both on direct inspection and on extensive videotaping of our pa­tients' eye movements, failed to disclose ad­dL: cting Jerks on ilttp 111ptE'd abduction. Neverthe-less, our first and third patients did show ad­ducting jerks of their left eye on watching a down­going optokinetic tape. This finding might suggest that impaired relaxation of the left medial rectus may have played a role in the genesis of restrict~d abduction on left lateral gaze. The absence of pu­pillary constriction during attempted abduction does not preclude the possibility that conver­gence- retraction spasm, a common finding in pa­tients with rostral midbrain lesions, may have con­tributed to restrict abduction in our patients. In a well-studied case the adducting jerks of conver­gence nystagmus were found to be imperfect sac­cades, rather than true vergence movements (20). Thus, pupillary constriction would not be ex­pected to happen, even if the pupil can react to light. The second of the patients Caplan (2) de­scribed to illustrate the "pseudo-six" phenomenon bears remarkable resemblance to the patients de­scribed here. In addition to a pronounced restric­tion of right eye adduction (21), that patient had minimal abduction of the left eye (2). Neuropatho­logic examination disclosed a small dorsal parame­dian infarct, which affected partially the right ocu­lomotor and trochlear nuclei and extended into the dorsal portion of the right red nucleus (21). Impairment of abduction has been reported with lesions involving the medial longitudinal fas­ciculus (MLF) (22). Although the lesions in our pa­tients may have affected the most rostral portion of the MLF, it is unlikely that their impairment of abduction was related to damage of this structure. The absence of abducting nystagmus speaks against the diagnosis of an internuclear ophthal­moplegia. Impairment of abduction ascribed to le­sions of the MLF may have been related to con­comitant damage of the midbrain or pontine retic­ular formation. Normal abducting saccadic velocities have been recorded with lesions of the MLF (23). Another mechanism could be invoked to explain the weakness of abduction to the left in these pa­tients. It may simply be the expression of left gaze paresis, due to involvement of the pathways that carry information related to saccadic and pursuit eye movements from the cerebral hemispheres to the paramedian pontine reticular formation. Since the restriction of abduction to the left was over­come by the vestibulo-ocular reflex, the disorder might be classified as a unilateral Roth-Biel­schowsky syndrome (24). Failure of the oculoce­phalic maneuver to overcome the limitation of right eye adduction could be ascribed to involve­ment of the MLF or to a nuclear or fascicular me­dial rectus weakness. However, other signs char- MIDBRAIN HORIZONTAL GAZE PARESIS 233 acteristic of the Roth-Bielschowsky syndrome were absent (25). Fixation of gaze in the direction of the slow component of the vestibulo-ocular re­flex did not occur with caloric stimulation. Al­though limited, leftward corrective movements were made by both eyes. Unilateral supranuclear impairment of hori­zontal gaze in humans is generally short-lived. Hemispheric infarction in the distribution of the middle cerebral artery generally causes only a transient impairment of contralateral horizontal saccades (26). However, in our patients the lateral gaze paresis lasted for as long as 6 months, al­though with some degree of improvement. This pattern is consistent with results of experimental lesions of the frontopontine projection system, which cause a greater and more lasting impair­ment of horizontal gaze the closer they are to their pontine target. For instance, in monkey, a 5 mm electrolytic lesion in the region of the fields of Forel on the right produced an impairment of hori­zontal gaze to the left for 48 h (27). More caudally, in the mesencephalic reticular formation, a smaller (2-4 mm) lesion on the right resulted in an even greater and more lasting defect in eye movements to the left (27). In the patients reported here the abnormal sac­cadic and pursuit movements to the left probably resulted from damage at the right midbrain-dien­cephalic junction to the cerebropontine pathways that mediate horizontal gaze. Experimental data suggest that the frontopontine pathway sub­serving saccadic eye movements decussates at the level of the pontomesencephalic junction (8). Both the frontal eye field and the superior colliculus are known to mediate the generation of horizontal saccades, particularly those directed to the contra­lateral side. However, lesions of either of these structures fail to cause lasting impairment of gaze. Combined lesions of both cause a remarkable pau­city of saccadic eye movements (9). In the monkey the projections from the frontal eye fields to the brainstem follow three main trajectories (28). A dorsal, transthalamic bundle reaches the superior colliculus and periaqueductal gray. A ventral bundle courses in the internal capsule and in the most medial portion of the crus cerebri, decus­sating at the rostral extent of the pontine teg­mentum. An intermediate bundle is formed from the other two in the caudal subthalamic region. It is located medial to the ventral fasciculus retro­flexus and rostralmost red nucleus. This interme­diate bundle projects to the rostral interstitial nu­cleus of the medial longitudinal fasciculus and de­scends into the oculomotor nucleus. Convergence of these pathways at the diencephalomidbrain junction may explain why discrete lesions in this location can impair horizontal as well as vertical gaze. Both the midbrain and the caudal portion of the medial thalamus, at the midbrain-diencephalic junction, were affected in our patients. Which of these lesions produced the horizontal gaze palsy is a matter of speculation. In experimental animals contralateral gaze palsy has resulted from lesions in the reticular formation of the midbrain (27). Zackon and Sharpe (3) described horizontal gaze palsy with pathologically proven involvement of the midbrain tegmentum. Although the caudal thalamus was affected in their second patient, the first one had a tumor restricted to the midbrain (3). Among 22 cases of thalamic infarction re­ported by Rousseaux et aI. (29), two with unilat­eral lesions had impairment of contralateral hori­zontal eye movements. Both belonged to the group of infarcts in the territory of the retromam­milIary pedicle. In addition to medial thalamic in­farction, involvement of the midbrain is likely in these cases. Some experimental findings suggest that damage to the medial portion of the posterior thalamus may playa role in the genesis of this syndrome. In monkey, ablation circumscribed to the superior colliculus produced minimal effects on the initiation and accuracy of contralateral sac­cades (7). However, when the lesion extended to the tectum and posterior-medial thalamus, the monkeys showed a long-lasting deficit in the ability to make accurate saccades to eccentric vi­sual targets on the side contralateral to the abla­tion. The posterior medial thalamic region, but not the pretectal area, was affected in our patients. As Albano and Wurtz (7) have proposed, structures in the posterior medial thalamus, at the midbrain­diencephalic junction, may be part of an indirect pathway between the frontal and collicular areas and the paramedian pontine reticular formation. Right-sided damage of this indirect pathway for the generation of horizontal saccades could ex­plain the impairment of leftward saccades in the patients reported here. For pursuit movements, the occipitoparietotem­poral association cortex projects to the ipsilateral pontine reticular formation, but the course of the pathway is unclear (17). A double decussation has been suggested, the lower one at the pontomesen­cephalic junction (24). This arrangement would be compatible with the findings in our patients of a unilateral midbrain lesion causing an impairment of saccadic as well as pursuit movements to the opposite side. A similar syndrome developed in I CI," NCllrLJ-ophtha[mol, '101. 7. No.4. 1987 234 J. C. MASDEU AND M. ROSENBERG the second patient reported by Zackon and Sharpe (3). However, earlier in the course, the patient had better pursuit eye movements to the side contra­lateral to a midbrain metastasis. Their other pa­tient had pursuit eye movements that were bilater­ally abnormal. Pathology in their cases showed no pontine lesion to account for the impairment of horizontal eye movements. The clinical and radiologic findings in the cases reported here reinforce the notion that a clinically obvious horizontal gaze palsy may result from an infarct at the midbrain-diencephalic junction, even when the pons is unaffected. In these cases con­tralateral saccades are impaired and, at least in some instances, contralateral pursuit eye move­ments also. Acknowledgment: Dr. Richard Brannegan referred the first patient. Dr. Glen D. Dobben performed and in­terpreted CT of the first two cases. Dr. James A. Sharpe provided helpful suggestions for the manuscript. REFERENCES 1. Bender MB, Rudolph SH, Stacy CB. The neurologv of the visual and oculomotor svstems. In: Baker AB, Baker LH, eds. Clinical nellrology, Vol 1. Philadelphia: Harper and Row, 1983:73. 2. Caplan LR. "Top of the basilar" syndrome. Nellrology 1980;30:72-9. 3. Zackon DH, Sharpe jA. Midbrain paresis of horizontal gaze. Ann Nell 1'01 1984;16:495-504. 4. Bril V, Sharpe jA, Ashby P. Midbrain asterixis. AIIII Nellrol 1979;6:362-4. 5. Claude H. Syndrome pedonculaire de la region du Novau rouge. Rep Nellrol (Paris) 1912;23:311-3. 6. Pierrot-Deseilligny C, Schaison M, Bousser MG, Brunet P. Svndrome nucleaire du nerf moteur oculaire commun: a p~opos de deux observations cliniques. Rep Nellrol (Pans) 1981;137:217-22. 7. Albano jE, Wurtz RH. Deficits in eve position following ablation of monkey superior colliculus, pretectum, and posterior-medial thalamus J Nellroplll/sioI1982;48:318-37. 8. Bender MB. Brain control of conjugate horizontal and ver­tical eye movements. A survey of the structural and func­tional correlates. Brain 1980;103:23-69. 9. Schiller PH, True SD, Conway jL. Deficits in eye move­ments following frontal eye-field and superior colliculus ablations. J NeliropJIl/sioI1980;44:1175-89. 10. Walsh FB, Hoyt WF. Cl1nical lIelim-ophthallllology. (3rd ed). Baltimore: Williams and Wilkins, 1969: p 231 11. Castaigne P, Lhermitte F, Buge A, Escourolle R~ Hauw J], Lyon-Caen O. Paramedian thalamic and mldbram mfarcts: clinical and neuropathological study. AIIII Nellrol 1981;10: 127-48. 12. Hochman MS, Sowers J), Bruce-Cregorios j. Syndrome of the mesencephalic artery: report of a case with CT and nec­ropsy findings. J Nellrol Nellrosurg Psychiatn/ 1985;48:1179­81. 13. Warwick R. Representation of the extra-ocular muscles in the oculomotor nuclei of the monkey. JComp Neurol 1953; 98:449-503. 14. Biefang DC Crossing axons in the third nerve nucleus. 111­l'est OphtlzalllloI1975;14:927-31. 15. Buttner-Ennever jA, Buttner U, Cohen B, Baumgartner G. Vertical gaze paralysis and the rostral interstitial nucleus of the medial longitudinal fasciculus. Braill 1982;105:126-49. 16. Cogan DC, Freese CG jr. Spasm of the near reflex. Arch Ophthalmol 1955;54:752-9. 17. Leigh jR, Zee DS. Tile lIelirology of eyt' movements. Philadel­phia: FA Davis, 1983. 18. Fisher CM, Picard EH, Polak A, Dalal P, Ojemann RG. Acute hypertensive cerebellar hemorrhage: diagnosis and surgical treatment. J NcrI' Mellt Dis 1965;140:38-57. 19. Fisher CM. Clinical syndromes in cerebral hemorrhage. In: Fields WS, ed. PatllOgellesls alld treatmellt of cere/Jrol'aswlar disease. Springfield, IL: CC Thomas, 1961:331-2. 20. Ochs AL, Stark L. Hovt WF, D'Amico D. Opposed ad­ducting saccades in convergence-retraction nystagmus. BI'lIllI1979;102:497-508. 21. Growdon jH, Winkler GF, Wray SH. Midbrain ptosis. A case with clinicopathologic correlation. Arch Nellrol 1974;30;179-81. 22. Feldon SE, Hovt WF, Stark L. Disordered inhibition in in­ternuclear ophthalmoplegia: analysis of eye movement re­cordings with computer simulations. Braill 1980;103:113­37. 23 Herishanu YO, Sharpe jA. Saccadic intrusions in internu­clear ophthalmoplegia. AIIII NellroI1983;14:67-72. 24. Hovt WF, Daroff RB. Supranuclear disorders of ocular con­trol systems in man: clinicaL anatomical and physiological correlations. In: Bach-v-Rita P, Collins CC, Hvde jE, eds The cOlltrol of t'l/C 11101";11Iellts. New York: Academic Press, 1971:175-235.. 25. Cogan DC. Neurology or the oClllar 111llSdcs. 2nd ed. Spring­field, IL: CC Thomas, 1956:221-3. 26. Steiner I, Melamed E. Conjugate eve deviation after acute hemispheric stroke: delaved recovery after previous con­tralateral frontal lobe damage. AIIII Nell 1'01 1984;16:509-11. 27. Komatsuzaki A, Alpert J, Harris HE, Cohen B. Effects of mesencephalic reticular formation lesions on optokinetic nystagmus. Exp Nellrol 1972;34:522-34. 28. Leichnetz CR. The prefrontal cortico-oculomotor trajec­tories in the monkev , Nellrol SCI 1981;49:387-96. 29. Rousseaux M, Petit H, Hache )C, Devos PH, Dubois F, Warot P. La motricite oculaire et cephalique dans les in­tarctus de la region thalamique. Ret, Nellrol (Paris) 1985; 141:391-403.