Mild Bilateral Internuclear Ophthalmoplegia: The Diagnostic Role of the Vertical Posterior Canal Vestibulo-Ocular Reflex in Acute Brainstem Demyelination, a Clinical-Radiologic Correlation

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Mild Bilateral Internuclear Ophthalmoplegia: The Diagnostic Role of the Vertical Posterior Canal Vestibulo-Ocular Reflex in Acute Brainstem Demyelination, a Clinical–Radiologic Correlation Luis G. Manrique, MD, Xiajoun Zhang, MD, Laurel Kathryn, AuD, Catherine Marie, AuD, Jorge C. Kattah, MD Background: The ocular signs of internuclear ophthalmoplegia (INO) with slow, restricted adduction of one eye and abduction nystagmus of the contralateral eye are easily recognized and have a high localizing/lateralizing value. However, subtle INO is difficult to diagnose. Recent reports identified novel vestibular abnormalities in unilateral and bilateral INO. Frequent findings include decreased horizontal and posterior canal gains, and generally relative sparing of both anterior canals. We studied one patient with a subtle bilateral INO, performed serial quantitative saccade (QS) and video head impulse test (vHIT), and correlated clinical– radiological findings caused by acute demyelination. Methods: Single case study of a 30-year-old man presented with 1 week of painless, binocular, horizontal diplopia in left gaze. We performed 3 serial neurological examination, QS, vHIT, and clinical–MRI correlation (1 pretreatment and 2 post steroid treatment). Results: We found bilateral slow adducting, clinically positive posterior canal HITs, and borderline abducting saccade velocity, without abducting nystagmus. The videonystagmography with fixation block showed bilateral horizontal gaze evoked nystagmus, and vHIT testing confirmed decreased right horizontal and bilateral posterior canal gains. The abnormalities resolved after steroid treatment. MRI showed acute bilateral medial longitudinal fascicle demyelinating lesions. Conclusions: A bilaterally positive, posterior, canal HIT and slow adduction saccades are localizing findings in bilateral INO, even in the absence of abduction nystagmus. Quantitative confirmation of these findings suggest most frequently an ischemic or demyelinating disorder and are a compelling indication for MRI. This case shows value to Department of Neurology, University of Illinois College of Medicine, Peoria, Illinois and the Illinois Neurologic Institute, Saint Francis Medical Center, Peoria, Illinois. The authors report no conflicts of interest. Address correspondence to Jorge C. Kattah, MD, Department of Neurology. University of Illinois College of Medicine, 530 NE Glen Oak Avenue, Peoria, IL 61637; E-mail: [e-mail] Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 testing multiaxial head impulses and performing QS and vHIT in brainstem lesions. Journal of Neuro-Ophthalmology 2022;42:e281–e288 doi: 10.1097/WNO.0000000000001262 © 2021 by North American Neuro-Ophthalmology Society T he ocular motor findings associated with an internuclear ophthalmoplegia (INO) represent a classic and one of the best known ocular motor findings in both Neurology and Ophthalmology (1); typically, adduction weakness on the side of the lesioned middle longitudinal fasciculus (MLF) is associated with a monocular, jerk, horizontal nystagmus of the abducting eye during horizontal versions. It has a high localizing and lateralizing value. Moreover, in young adults, it may be the first presentation of multiple sclerosis (MS) (2–4). Subtle INO variants, on the other hand, may be more difficult to detect, for example, mild adduction weakness without abduction nystagmus, and even isolated adduction lag, with slower adduction saccade velocity but normal adduction range (5). These milder variants potentially represent a diagnostic challenge, particularly because normal convergence in this case would not have the equivalent diagnostic role played when the MLF lesion causes significant adduction weakness or paralysis if the lesion is in the pons (6–10). Recent reports focused on the evaluation of the vertical vestibulo-ocular reflex (VOR) projections within the MLF, primarily originating from the contralateral posterior canal (11). Therefore, in subtle INO cases, testing these vertical VOR abnormalities could contribute to MLF lesion localization. Data from a cumulative number of 59 patients and 4 articles, featuring vascular and demyelinating unilateral and bilateral INO (11–14), identified different patterns of decreased VOR gain. In unilateral INO, the contralateral vertical VOR, specifically in the plane of the e281 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution contralateral posterior canal; these unilateral INO patients also showed the anticipated decreased gain of the ipsilesional horizontal head impulses on the side of the MLF lesion, due to the compromised efferent motor response of the slow adducting eye to the head impulse (12–14). In bilateral INO, the main findings included bilaterally decreased VOR in both horizontal and vertical canals; the anterior canals were the least affected (8). Here, we report one patient with an acute, subtle bilateral INO; he had a mild exophoria in extreme left lateral gaze with slow right eye adduction, and we noted normal adduction range in right gaze with subtle left adduction saccade slowing. Abduction nystagmus was bilaterally absent on initial examination; there was no skew deviation, and he had normal convergence; in addition, he had bilaterally positive vertical head impulse in the plane of both posterior canals, with normal anterior canal impulses. An acute, enhancing, demyelinating lesion involving the bilateral MLF in the pons accounted for the clinical findings, and although he did not have a diagnosis of multiple sclerosis, and he had numerous asymptomatic demyelinating MRI lesions, the bilateral INO was the only clinically eloquent manifestation of MS. CASE REPORT AND METHODS We evaluated a 30-year-old man in the emergency department for 1 week of painless binocular blurry vision on waking up associated with nausea and one episode of emesis. His initial symptoms fluctuated, and eventually, he became aware of binocular horizontal diplopia in extreme left gaze during his first formal examination. Three days before the visual disturbance started, he was unsteady (described as swaying on a boat) and later developed intermittent dizziness when standing, which did not affect his gait. His medical history was remarkable for a diagnosis of temporal lobe epilepsy, made 4 years earlier, well controlled on lamotrigine. His pre and postcontrast brain MRI then was normal. Initial neurological examination showed a normal afferent neuro-ophthalmologic examination; there was no ocular deviation in straight ahead gaze; we found an 8-prism-diopter exotropia in extreme left gaze. He did not have skew deviation or head tilt. We noted slow left adduction saccades to the right without limited adduction, and a normal cross-cover test in extreme right gaze. Vertical saccades were normal. There was no abduction nystagmus in right or left gaze, and he had no vertical nystagmus. Eyelids and pupils were normal, and importantly, he had normal convergence and did not have skew deviation. He had clinically abnormal right horizontal HIT and bilaterally posterior canal head impulses. We also observed mild difficulty performing tandem gait; the rest of the neurologic examination was normal. The patient underwent serial VOR and videonystagmography (VNG) evaluations at 2, 9, and 39 days after the e282 onset of his symptoms in the neurovestibular clinic. In the immediate follow-up examination, and after the 3-day course of intravenous steroids (1 g of methylprednisolone), he had briefly mild horizontal abduction nystagmus of the left eye in left gaze. We utilized the Micromedical Spectrum 6.0 device, System 2.000 (Chatham, IL) to test the saccade main sequence and VNG. The pretreatment saccade main sequence showed bilateral slow adduction saccades (Fig. 1B, C) and borderline abduction velocities (Fig. 1A, D); the saccade test that we utilized involved a variable degree random target paradigm within a range of ±25°. The plot shows the patients’ responses compared with ±1SD from age-matched subjects (Fig. 1). To calculate the horizontalsaccade versional dysconjugacy index (VID) for saccades greater than 5° amplitude, we measured the ratio of the peak abduction divided by peak adduction velocities. It was 1.7 in right horizontal gaze and 1.6 in left gaze (abnormal, .1.2). A follow-up posttreatment 9 days later was normal. The pursuit gain was normal, and the binocular horizontal optokinetic nystagmus showed bilaterally decreased gain. The VNG with fixation block showed bilateral horizontal gaze evoked nystagmus (right beat in right gaze and left beat in left gaze, with a slight downbeat component). He had a 20-dB hearing loss in the left ear, tested with pure tone audiometry; the cervical vestibular evoked response (cVEMP) was absent in the right and normal in the left side. Ocular vestibular evoked response (oVEMP) was bilaterally normal (tested only posttreatment). We tested the multiaxial video head impulse test (vHIT) in the right eye using a portable lightweight device (ICS impulse; GN Otometrics, Taastrup, Denmark). Two outpatient followed the initial inpatient recording. To test the vHIT, the patient sat up and fixated on a center target located 1 m away. Eye position was calibrated using laser targets projected from the goggles. We tested 20 random head impulses as the patient fixated on the central target in the plane of the 6 semicircular canals, horizontal head impulses first. Table 1 summarizes the vHIT gains (the first test before the initiation of 1 g intravenous methylprednisolone, and the second and third tests post treatment). In addition, we selected a number of individual canal response to illustrate the ratios between the head and ocular gain response and the corrective saccade before treatment (Fig. 2) for the horizontal canals and (Fig. 3) for the vertical canals. The right eye adduction corrective saccades to right head impulses had slower amplitude (Fig. 2 black arrow) than those from the right eye abducting saccade in response to left head impulses (Fig. 2 gray arrow). We found bilaterally decreased posterior canal with correction saccades and normal anterior gain. A head, T2 FLAIR brain MRI sequence showed multiple lesions affecting both supratentorial and infratentorial periventricular, juxtacortical/grey matter, and brainstem lesions. In postcontrast T1, some of them enhanced. In addition, the T1 MRI showed lesions compatible with Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. Saccade main sequence bilateral internuclear ophthalmoplegia (INO): The velocity plot shows the patients responses compared with ±1 SD values from normal age-matched subjects. Note decreased left adduction (B) and right adduction (C) velocity. Abduction saccade velocities were moderately decreased (no abduction nystagmus was present). Adduction saccades were hypometric. “black holes.” Evidence of dissemination in time and space provided a final diagnosis of MS. Relevant to this case, clinical findings, we identified 2 lesions: The first, a hyperintense, T2 sequence MRI lesion, involved the bilateral MLF in the midpons, adjacent to the IVth ventricle. Contrast enhancement was more prominent on the right and correlated with the acute clinically symptomatic, bilateral INO (Fig. 4); a second lesion involved the right superior cerebellar peduncle (not shown). Cervical spine MRI showed increased T2 hyperintensity on the lateral aspect of the cervical cord at C6–C7 level with no enhancement. Cerebrospinal fluid analysis revealed 7 oligoclonal bands, 9 white blood cells, 0 red blood cells, a glucose concentration of 58 mg/dL, and total protein content of 59.4 mg/dL. He had an inpatient multiaxial vHIT, followed by VNG, including saccade main sequence. Steroid treatment improved all his symptoms and the ocular motor and vestibular signs (Table 1). Presently, he is followed in the multiple sclerosis center. DISCUSSION We describe pretreatment ocular motor, vestibular and neuroimaging findings in a patient with an acute bilateral MLF demyelinating lesion, including saccade main sequence, VDI and multiaxial high-frequency horizontal and vertical VOR, and recovery during and after treatment with high-dose steroids 1 month later. Previous reports identified patterns of multiaxial VOR abnormalities in patients with established demyelinating bilateral INO (12–14). To our knowledge, this is the first study to show Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 the vHIT contribution to diagnosing an acute, mild, bilateral demyelinating INO. In this case, the combination of positive bilateral posterior canal HIT and bilateral slow adduction saccade suggested an INO, a helpful localization clue in the absence of monocular abduction nystagmus. The initial saccade main sequence and vHIT recording supported the clinical bilateral INO R . L diagnosis, later confirmed radiographically. We summarized the beneficial effect of high-dose steroids and a 2-week oral steroid in Table 1. The MLF is a heavily myelinated white matter tract containing axons that provide binocular connectivity for horizontal saccades (excitatory and inhibitory), horizontal and vertical pursuit eye movements, horizontal and vertical VOR, and utricular projections (10). Accordingly, studying the function of these different tracts yields diagnostic information. Our patient had acute lesions involving the bilateral MLF (Fig. 4). Postcontrast images showed local breakdown of the blood– brain barrier due to an inflammatory demyelinating lesion and progressive conduction block. This mechanism has been previously confirmed with elegant experiments, utilizing electric vestibular stimulation (evoked vestibulo-ocular: e-VOR) onset latency (15). High-dose steroids improved saccade velocities and vHIT findings because of improved axonal conduction. The clinical findings in most INO patients are straightforward with characteristic decreased amplitude and slow adduction on the side of the affected MLF and monocular dysmetria/nystagmus of the abducting eye, best observed with large amplitude saccades (1). Saccade dynamic infrared video-oculography with QS identified isolated slow adducting saccades in several studies (7,8), and these findings e283 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Serial video head impulse test (vHIT) in a patient with bilateral internuclear ophthalmoplegia (INO), RL Date June 3, 2020 Day 2 October 6, 2020 Post 3 days of 1 G Solumedrol and Oral taper Day 9 July 7, 2020 Day 39 Semicircular Canal vHIT Gain Refixation Saccades RL 0.68 ± 0.07 LL 0.68 ± 0.06 LA 1.01 ± 0.08 RP 0.56 ± 0.06 RA 0.85 ± 0.18 LP 0.67 ± 0.07 RL 0.98 ± 0.06 LL 0.98 ± 0.04 LA 0.90 ± 0.04 RP 0.72 ± 0.08 RA 0.96 ± 0.11 LP 0.78 ± 0.06 RL 0.93 ± 0.04 LL 0.83 ± 0.05 LA 0.90 ± 0.04 RP 0.72 ± 0.08 RA 0.95 ± 0.11 LP 0.78 ± 0.11 Present Present None Present None Present Present, ,amplitude None None None None None Minimal None None None None None Multiaxial vestibulo-ocular reflex gain at day 2 (pretreatment) and days 9 and 39 (post treatment). The gain in all affected canals improved after steroid treatment. The anterior canal remained unchanged. LL, Left Lateral; LA, Left Anterior; LP, Left Posterior; RL, Right Lateral; RP, Right Posterior; RA, Right Anterior. occur in subclinical INO, regardless of etiology. Moreover, velocity and accuracy testing may be substantially impaired in cases without limitation of adduction (adduction lag); in our patient, recordings confirmed bilateral adduction saccade slowing and borderline abducting saccade velocity (Fig. 1); he had normal vertical saccade velocity (not shown). Abduction saccade slowing may be present in bilateral INO as well (16). In one well-documented case, one MLF lesion, and a fascicular abducens lesion coexisted without gaze palsy, the abducting saccade velocity in our patient and the MRI findings excluded this consideration (17). The multiaxial vHIT, with monocular recording from the right eye in the plane of all 6 semicircular canals (SCC) shortly after admission, enabled us to evaluate the function FIG. 2. Horizontal video head impulse: the left panel shows 3 representative right eye response to left head impulses, note decreased gain and overt corrective saccades (black arrow). The right panel shows 3 representative right eye response to right head impulse (the side of then internuclear ophthalmoplegia [INO]); note decreased gain and small overt saccade. e284 Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. Vertical head impulse: two representative vertical canal head impulses. Note decreased gain of the bilateral posterior canals and overt corrective saccades (arrows) and normal gain of the anterior canals. of vestibular projections in the MLF and confirmed quantitatively the clinical findings. We found mildly decreased and symmetric right eye adduction and abduction gains (0.63) in response to horizontal left and right head impulses (Fig. 2), (normal horizontal gain: 0.8). We did not find different right-eye adduction compared with right-eye abduction gain, in response to horizontal head impulses, which we had anticipated from the slow adducting velocity during visually triggered saccades and previous studies (12– 14). Moreover, the right abduction eye movement, in response to a left head impulse, had lower gain than normal, a finding not clinically apparent. To explain the greater difference between horizontal saccade dissociation found with visually triggered saccades and the lesser difference with those from vestibular stimulation in bilateral INO patients, one must consider recent hypothesis proposing a different mechanism for visual vs vestibular triggered ocular motor responses in INO (13). In a series studying 19 chronic bilateral INO patients, recorded with magnetic search coils, 2 factors potentially explain this difference. The first, vestibular-induced adduction saccades on the side of the INO receives an ipsilateral, extra-MLF, excitatory, vestibular input from the ipsilateral ascending tract of Dieters (ATD); this tract projects directly to medial rectus (MR) motoneurons and increases ipsilesional horizontal gain (13). The second to explain the lower gain of the abducting eye with contralesional head impulses, here we quote “it is plausible disfacilitation of inhibitory neurons, originating from type II vestibular neurons in the vestibular commissure,” (13); these inhibitory fibers cross near the demyelination focus in the middorsal pons or in the MLF itself. Applying this postulation to our patient with mild bilateral INO, we found lower than normal but symmetric horizontal canal gain with right eye adduction/abduction movement in response to left or right horizontal head impulses (Fig. 2). Whereas, adduction gain increased from Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 the contribution of the ATD, the abduction gain depends not only on excitation of abducens motoneurons but also on simultaneous inhibition of MR motoneurons by intraMLF inhibitory fibers, as a result lack of MR inhibition slowed abduction saccade velocity. The corrective saccade in response to the defective horizontal VOR, merit a brief comment, we found larger amplitude right abducting than right adducting corrective saccades. This finding was predictable because they represent visually triggered saccade corrections responding to impaired rapid horizontal head acceleration (Fig. 2). In a previous study, chronic MLF demyelination showed decreased adduction VOR gain with absent corrective saccades (13). Pertaining the vertical VOR, we found bilaterally decreased posterior canal gain, in contrast to the normal anterior SCC (Fig. 2 and Table 1). The first report of an impaired contralesional, high-frequency posterior canal VOR in unilateral MLF lesion, utilized scleral search coils in a patient with an unilateral vascular INO; in this patient, the contralateral posterior gain was low, and the anterior canal was normal (11). Further studies from the same group of investigators, included 19 bilateral INO patients with MS, utilizing also magnetic search coils showed decreased gain in all 6 canals but more pronounced in the posterior canals (13). Several case series thereafter utilizing vHIT testing in patients with vascular and demyelinating unilateral and bilateral INO (12,14) provided a translational application to the precise data provided with the investigational search coil recordings (11,13). In one of these reports (14), blinded analysis of the VOR obtained in patients with overt unilateral and bilateral INO correctly identified the MLF lesion from vHIT recordings, relaying principally on the decreased contralesional posterior canal and the unilaterally decreased ipsilesional right horizontal canal gains and corrective saccades. The conclusions from these studies e285 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. Multiplanar MRI: The two top panels are 2 adjacent postcontrast T1 axial MRI sections in midpons. The arrow points to a well-circumscribed area of focal disruption of the blood–brain barrier localized to the bilateral middle longitudinal fasciculus (MLF), the postcontrast T1 sagittal MRI in the lower left panel shows the same area of focal enhancement in the pons. The right lower panel shows a coronal section with localized enhancement of both MLF. The rest of the MRI (not shown) demonstrated numerous white matter lesions, some chronic, including black holes, thus providing evidence of dissemination in time and space. illustrate potential value to testing and recording the vertical VOR in brainstem lesions, particularly in cases with adduction lag, potentially related to MLF lesions. We found normal anterior canal gain in our patient; this finding may reflect the partial and acute nature of the lesion, presumably causing conduction block rather than permanent demyelination. Our patient had a second demyelinating lesion in the superior cerebellar peduncle that probably spared a contingent of anterior canal projections to the brainstem from the superior vestibular nucleus. Steroid therapy normalized saccade velocity and VOR gain (Table 1). Of note, we did not find the expected alternating skew deviation or vertical nystagmus often present with bilateral INO. In addition, he had a normal oVEMP, tested during his oral steroid taper; we speculate that the lack of otolith impairment may be potentially explained by the milder nature of the bilateral INO in this case. The patient complained of nausea and had one episode of vomiting probably as result of subradiographic inflammation in the area postrema. In general, the evaluation of the vertical VOR has not received quite the same attention than the horizontal VOR; thus, its clinical diagnostic impact is only beginning to unfold. Practically speaking, each vertical canal should be tested with head movements in the pitch plane, with the head turn 30° toward the right and then toward the left. The right head turn is needed to align the head with the plane of the left anterior and right posterior canal pair during the vertical HIT, and the left head turn plays the same role for the right anterior and left posterior canal pair (1). In e286 addition, the anatomical structure of the 3-neuron arc of the anterior and posterior canals has not only specific canal innervation, and vestibular nuclear neurons, but also specific direct brainstem tracts to final ocular motor nuclei destination (1,18); therefore, each requires an individual VOR test. The anterior canal fibers travel to the brainstem via the superior vestibular nerve, reach primarily the superior vestibular nucleus, with a small contingent to the medial vestibular nucleus (MVN), and project mostly via extra-MLF pathways to the ipsilateral superior rectus and contralateral inferior oblique third nerve subnuclei (1,18). The posterior canal fibers travel to the brainstem via the inferior vestibular nerve, reach the MVN, and travel within the MLF to reach the ipsilateral trochlear and contralateral inferior rectus nuclei (1,18). Diagnostic lateralization and localization information, therefore, depends on comparative clinical vertical canal (HIT) responses, quantitative (vHIT) VOR gain, and presence of corrective saccades unilaterally (anterior vs posterior canal), with the contralateral homologous vertical canals. There is also value in comparing vertical and horizontal VOR gain (19,20). Following this protocol, diagnosis of isolated individual semicircular canal nerve dysfunction (21), superior, inferior, and combined vestibular neuritis, cochlear and vestibular schwannoma, positional vertigo variants, and Meniere disease can be confidently made (19,20). For example, ototoxicity from gentamicin and bilateral Meniere syndrome often spare the anterior canal function differentiating them from the idiopathic bilateral vestibular neuropathy or the cerebellar atrophy, Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution neuropathy, vestibular areflexia syndrome (20,22). In central vestibular lesions, there are instances where the vertical VOR contributes to lesion identification, keeping in mind that the results may mimic a peripheral vestibulopathy (19). Besides the example provided in this report with preferential posterior canal compromise, in demyelinating bilateral INO, similar findings might occur in neuromyelitis optica (23). In Wernicke encephalopathy, the horizontal canal gain is bilaterally decreased, usually sparing the vertical canal gain, due to greater vulnerability of the medial vestibular nucleus to thiamine deficiency (24,25). Vertical nystagmus, regardless of the cause, may be associated with vertical VOR abnormalities (26). Cerebellar lesions in the AICA territory tend to affect primarily the horizontal VOR, sparing the vertical VOR (19). Additional diagnostic clues will predictably develop with further testing experience. In conclusion, our acute bilateral INO patient illustrates the value of testing the multiaxial head impulse as an important part of the standard neurologic examination. Bedside recording with video-oculography is increasingly more available (The Electrocardiogram of the eyes) (27), it adds quantitative, graphic documentation. Incorporating QS analysis is key, particularly in lesions of the brainstem, where evaluation of the efferent arm of the VOR is important to separate primarily vestibular, from ocular motor lesions, and to recognize those from combined deficits. Detailed quantitative analysis of the horizontal vHIT identified the impact of key test artifacts (28,29); similar studies are now in progress for vertical canal function. Design of routine diagnostic bedside quantitative ocular motor and vestibular evaluation protocols may yield diagnostic lesion localization, particularly valuable in subtle lesions possibly with normal neuroimaging. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: J. C Kattah: conceptualize the study design, L. G. Manrique and X. Zhang examined the patient, prepared the first draft and approved the final manuscript: b. Acquisition of data: L. Donaldson and C. Rayburn performed the eye movement testing and vHIT and analyzed the data; c. Analysis and interpretation of data: L. G. Manrique, X. Zhang, L. Donaldson, C. Rayburn, J. C. Kattah. Category 2: a. Drafting the manuscript: L. Manrique prepared the first draft and contributed to subsequent revisions; b. Revising it for intellectual content: J. C. Kattah contributed in all phases of the manuscript for intellectual content. Selected the MRI images and figure tracings, and reviewed the content in relation to existing literature in the final manuscript version. Category 3: a. Final approval of the completed manuscript: L. G. Manrique, X. Zhang, L. Donaldson, C. Rayburn, J. C. Kattah. REFERENCES 1. Leigh RS, Zee DS. The Neurology of Eye Movements. New York, NY: Oxford university Press, 2915. 2. Baloh RW, Yee RD, Honrubia V. Internuclear ophthalmoplegia. I. Saccades and dissociated nystagmus. Arch Neurol. 1978;35:484–489. 3. Meienberg O, Müri R, Rabineau PA. Clinical and oculographic examinations of saccadic eye movements in the diagnosis of multiple sclerosis. Arch Neurol. 1986;43:438–443. Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 4. Müri RM, Meienberg O. The clinical spectrum of internuclear ophthalmoplegia in multiple sclerosis. Arch Neurol. 1985;42:851–855. 5. Keane JR. Internuclear ophthalmoplegia: unusual causes in 114 of 410 patients. Arch Neurol. 2005;62:714–717. 6. Frohman EM, Zhang H, Kramer PD, Fleckenstein J, Hawker K, Racke MK, Frohman TC. MRI characteristics of the MLF in MS patients with chronic internuclear ophthalmoparesis. Neurology. 2001;57:762–768. 7. Frohman EM, Frohman TC, O’Suilleabhain P, Zhang H, Hawker K, Racke MK, Frawley W, Phillips JT, Kramer PD. Quantitative oculographic characterisation of internuclear ophthalmoparesis in multiple sclerosis: the versional dysconjugacy index Z score. J Neurol Neurosurg Psychiatry. 2002;73:51–55. 8. Frohman TC, Frohman EM, O’Suilleabhain P, Salter A, Dewey RB, Hogan N, Galetta S, Lee AG, Straumann D, Noseworthy J, Zee D, Corbett J, Corboy J, Rivera VM, Kramer PD. Accuracy of clinical detection of INO in MS: corroboration with quantitative infrared oculography. Neurology. 2003;61:848–850. 9. Frohman EM, Frohman TC, Zee DS, McColl R, Galetta S. The neuro-ophthalmology of multiple sclerosis. Lancet Neurol. 2005;4:111–121. 10. Frohman TC, Galetta S, Fox R, Solomon D, Straumann D, Filippi M, Zee D, Frohman EM. Pearls and oy-sters: the medial longitudinal fasciculus in ocular motor physiology. Neurology. 2008;70:e57–e67. 11. Cremer PD, Migliaccio AA, Halmagyi GM, Curthoys IS. Vestibulo-ocular reflex pathways in internuclear ophthalmoplegia. Ann Neurol. 1999;45:529–533. 12. Choi SY, Kim HJ, Kim JS. Impaired vestibular responses in internuclear ophthalmoplegia: association and dissociation. Neurology. 2017;89:2476–2480. 13. Sweet TA, Chen L, Todd MJ, Barnett MH, Halmagyi GM. Vestibulo-ocular reflex deficits with medial longitudinal fasciculus lesions. J Neurol. 2017;264:2119–2129. 14. Lee SH, Kim SH, Kim SS, Kang KW, Tarnutzer AA. Preferential impairment of the contralesional posterior semicircular canal in internuclear ophthalmoplegia. Front Neurol. 2017;8:502. 15. Wang C, Paling D, Chen L, Hatton SN, Lagopoulos J, Aw ST, Kiernan MC, Barnett MH. Axonal conduction in multiple sclerosis: a combined magnetic resonance imaging and electrophysiological study of the medial longitudinal fasciculus. Mult Scler. 2015;21:905–915. 16. Zee DS. Internuclear ophthalmoplegia: pathophysiology and diagnosis. Baillieres Clin Neurol. 1992;1:455–470. 17. Frohman EM, Frohman TC. Horizontal monocular saccadic failure: an unusual clinically isolated syndrome progressing to multiple sclerosis. Mult Scler. 2003;9:55–58. 18. Bronstein AM, Patel M, Arshad Q. A brief review of the clinical anatomy of the vestibular-ocular connections-how much do we know? Eye (Lond) 2015;29:163–170. 19. Chen L, Halmagyi GM. Central lesions with selective semicircular canal involvement mimicking bilateral vestibulopathy. Front Neurol. 2018;9:264. 20. Tarnutzer AA, Bockisch CJ, Buffone E, Weber KP. Vestibular mapping in patients with unilateral peripheral-vestibular deficits. Neurology. 2020;95:e2988–e3001. 21. Strupp M, Glasauer S, Schneider E, Eggert T, Glaser M, Jahn K, Brandt T. Anterior canal failure: ocular torsion without perceptual tilt due to preserved otolith function. J Neurol Neurosurg Psychiatry. 2003;74:1336–1338. 22. Tarnutzer AA, Bockisch CJ, Buffone E, Weiler S, Bachmann LM, Weber KP. Disease-specific sparing of the anterior semicircular canals in bilateral vestibulopathy. Clin Neurophysiol. 2016;127:2791–2801. 23. Lee SU, Kim HJ, Choi JH, Choi JY, Kim JS. Comparison of ocular motor findings between neuromyelitis optica spectrum disorder and multiple sclerosis involving the brainstem and cerebellum. Cerebellum. 2019;18:511–518. 24. Kattah JC, Dhanani SS, Pula JH, Mantokoudis G, Tehrani AS, Toker DE. Vestibular signs of thiamine deficiency during the early phase of suspected Wernicke encephalopathy. Neurol Clin Pract. 2013;3:460–468. e287 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 25. Kattah JC, Guede C, Hassanzadeh B. The medial vestibular nuclei, a vulnerable target in thiamine deficiency. J Neurol. 2018;265:213–215. 26. Jorge A, Martins AI, Gouveia A, Lemos J. The use of video-head impulse test in different head positions in vertical nystagmus and ataxia associated with probable thiamine deficiency. Cerebellum. 2020;19:611–615. 27. Newman-Toker DE, Saber Tehrani AS, Mantokoudis G, Pula JH, Guede CI, Kerber KA, Blitz A, Ying SH, Hsieh YH, Rothman RE, Hanley DF, Zee DS, Kattah JC. Quantitative video-oculography e288 to help diagnose stroke in acute vertigo and dizziness: toward an ECG for the eyes. Stroke. 2013;44:1158–1161. 28. Mantokoudis G, Tehrani AS, Wozniak A, Eibenberger K, Kattah JC, Guede CI, Zee DS, Newman-Toker DE. VOR gain by head impulse video-oculography differentiates acute vestibular neuritis from stroke. Otol Neurotol. 2015;36:457–465. 29. Zamaro E, Saber Tehrani AS, Kattah JC, Eibenberger K, Guede CI, Armando L, Caversaccio MD, Newman-Toker DE, Mantokoudis G. VOR gain calculation methods in video head impulse recordings. J Vestib Res. 2020;30:225–234. Manrique et al: J Neuro-Ophthalmol 2022; 42: e281-e288 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.