Resolved External Ophthalmoplegia and Hearing Loss in Wernicke's Encephalopathy With Thiamine Replacement

Wernicke encephalopathy (WE) is classically described by a clinical triad consisting of confusion, ataxia, and ophthalmoplegia, but recent reports emphasize a history of malnutrition along with 2 elements of the WE triad (Caine's criteria) to enhance diagnostic sensitivity. The ophthalmoplegia, vestibular, and auditory expeditious improvement with intrave- nous thiamine usually confirms the diagnosis; serum levels generally provide additional diagnostic certainty.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Resolved External Ophthalmoplegia and Hearing Loss in Wernicke’s Encephalopathy With Thiamine Replacement Maxwell Q. Nyce, DO, Joshua S. Chisholm, DO, Julia A. Szmanda, DO, A. Katharina Boyce, AuD, Caroline M. Boczar, AuD, Jorge C. Kattah, MD Background: Wernicke encephalopathy (WE) is classically described by a clinical triad consisting of confusion, ataxia, and ophthalmoplegia, but recent reports emphasize a history of malnutrition along with 2 elements of the WE triad (Caine’s criteria) to enhance diagnostic sensitivity. The ophthalmoplegia, vestibular, and auditory expeditious improvement with intravenous thiamine usually confirms the diagnosis; serum levels generally provide additional diagnostic certainty. Methods: Here, we discuss the case of a woman with a distant history of gastric sleeve, poor nutrition and protracted vomiting, who developed acute confusion, imbalance, neartotal external ophthalmoplegia (EO), and hearing loss. The baseline thiamine level was 28 pmol/L (Normal: 70–180 pmol/L). We performed serial neurological, vestibular, and audiological examination to document over 5 days, the effect of intravenous (IV) thiamine, and again at 3 months with continued oral supplementation. We provide serial documentation with photographs and video recording of oculomotor abnormalities, audiometric testing, and a video of horizontal head impulse testing, and imaging findings. Results: Over the course of 5 days of IV thiamine supplementation, we demonstrate our patient’s resolution of near complete EO. We assessed vestibular paresis with horizontal head impulse testing, after complete resolution of the EO. The initially positive bilateral h-HIT showed decreased gain and overt corrective saccades, it clinically resolved by day 5, but video h-HIT testing demonstrated persistent decreased horizontal vestibulo-ocular reflex (VOR) gain and covert horizontal saccades, which persisted at the 3-month examination. By contrast, the vertical VOR gain was normal without corrective saccades. Bedside audiometry completed during the acute phase demonstrated severely restricted auditory speech comprehension, which normalized 3 months later. Severe truncal ataxia improved as well. Conclusions: This case is an example of how awareness of the variations in the clinical presentation of WE can be Illinois Neurologic Institute, Saint Francis Medical Center, University of Illinois College of Medicine, Peoria, Illinois.. The authors report no conflicts of interest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www. jneuro-ophthalmology.com). Address correspondence to Jorge C. Kattah, MD; E-mail address: [e-mail] Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 crucial in achieving an early diagnosis and obtaining better outcomes. A history of the poor nutritional status can be an important clue to aid in this early diagnosis. Journal of Neuro-Ophthalmology 2021;41:e655–e660 doi: 10.1097/WNO.0000000000001057 © 2020 by North American Neuro-Ophthalmology Society W ernicke encephalopathy (WE) is an acute neurological condition necessitating early diagnosis and prompt treatment to avoid irreversible neurological damage. WE is caused by a deficiency of vitamin B1 (thiamine). Originally, described by the Prussian neurologist, Dr. Carl Wernicke in 1881 as “superior acute hemorrhagic polioencephalitis,” who characterized the syndrome by a clinical triad, involving confusion, ataxia, and ophthalmoplegia, with diagnostic neuropathological findings (1). In practice, however, the complete triad is present in a minority of cases. The combination of external ophthalmoplegia (EO) and hearing impairment often suggests a mitochondrial disorder (2); however, near-complete EO and auditory loss is an unusual association with WE. Here, we report a woman with a distant history of gastric bypass, poor nutrition and protracted vomiting, who developed acute confusion, neartotal EO, and hearing loss. In a previous review of 232 cases of WE, ophthalmoplegia, specifically abducens paresis was present in 125 cases (54%) and horizontal gaze palsy (hGP) in 44%, but to the best of our knowledge, complete EO was described only in 1 of 3 original patients reported by Wernicke (1,3) and in a second recent case but with preservation of up gaze (4). Reports that are more recent, to diagnose WE emphasize a history of malnutrition along with 2 elements of the WE triad (Caine’s criteria) proposed to enhance diagnostic sensitivity (5). Our patient also met this criteria, by virtue of her long-standing nutritional deficiency and recent protracted vomiting. Awareness of hearing loss in WE published in the last 25 years have been e655 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Horizontal pretreatment and post-treatment vHIT Date Baseline vHIT Follow-up vHIT (3 mo later) Canal Gain RH LH LP RA LA RP RH LH LP RA LA RP 0.71 ± 0.12 0.48 ± 0.1 0.98 ± 0.07 1.2 ± 0.06 0.97 ± 0.08 0.71 ± 0.04 065 ± 0.03 0.53 ± 0.04 0.88 ± 0.04 0.86 ± 0.08 0.72 ± 0.11 0.83 ± 0.06 Normal horizontal vHIT gain: 0.8. Normal vertical vHIT gain 0.7. Baseline before treatment (top row) and post-treatment studies (lower rows). The horizontal canal gain remains lower than normal 3 months after treatment. always concurrent with the classic triad along with bilateral vestibular loss in some instances (6–12). Rapid recovery from WE with timely treatment is the norm (1), and delayed treatment may cause permanent neurologic deficit (3). After diagnosis and initiation of treatment, we performed serial examinations over her 2week in-patient stay, to document the course of the neurologic findings, recovery of extraocular movement function, and improvement of low frequency hearing thresholds and speech comprehension, and, in addition, improvement, albeit partial of vestibular function. We show serial video transition from near-complete EO to horizontal gaze holding failure and eventually permanent bilateral symmetric horizontal vestibular loss with sparing of the vertical VOR, which is typical for WE (13–15). CASE REPORT A 64-year-old woman with a history of vertical band sleeve gastroplasty approximately 25 years before presented to the emergency department with the complaint of generalized weakness after she had been unable to get out the tub. During the preceding 2–3 days, she had been slow to respond and inappropriately answering questions. Two months before admission, she developed a stuck sensation in her throat when swallowing foods followed by persistent nausea and vomiting. One month before admission, she underwent gastric dilatation, but nausea and vomiting quickly recurred. The day of admission, she was awake, had an unstable station, and was verbally unresponsive to her husband and examiner. Initial examination showed encephalopathy and complete horizontal and vertical gaze palsy, sparing the lids and pupils. She had near-complete paralysis of saccades and pursuit but was not alert enough to test the slow vestibulo-ocular reflex (VOR) consistently. In attempted right gaze, she had a few degrees of adduction of the left eye, suggesting greater involvement of motoneurons over interneurons in the left sixth nerve nucleus, and in left gaze, we could not obtain leftward saccades or pursuit, but with the head turned to the right side, there was limited left gaze (Fig. 1). She had normal direct ophthalmoscopy. In addition, the patient’s husband felt that recent hearing loss was contributing to her difficulty following commands. She had no limb ataxia but was not able to sit without support. Based on the history of vomiting, poor nutrition, and features on examination, we suspected WE. MRI of the brain with and without contrast revealed increased T2 and DWI and low ADC map signal in bilateral periventricular thalami and periaqueductal gray of the midbrain and pons (Fig. 2). The baseline thiamine level reported 3 days later was 28 pmol/L (Normal: 70–180 pmol/L). Intravenous thiamine induced rapid, albeit asymmetric gaze recovery initially and confirmed the initial diagnostic impression. The first day after thiamine administration, she demonstrated left gaze paretic nystagmus and improved, albeit partial right gaze palsy (vertical gaze normalized Video first section. See Supplemental Digital Content, Video, https://collections.lib. utah.edu/ark:/87278/s6sj6mv3). On the subsequent days, ophthalmoplegia resolved, but she developed bilateral h-gaze holding failure (video second section). Whereas the horizontal head impulse test (HIT) became bilaterally positive with reduced gain, the vertical VOR was normal (video second section; See Supplemental Digital Content, Video, https://collections.lib.utah. edu/ark:/87278/s6sj6mv3). Five days later, gaze was normal in all directions, and she was orthophoric. The clinical horizontal HIT became normal;; however, the video HIT showed low gain with covert saccades (video third section; See Supplemental Digital Content, Video, https://collections.lib.utah.edu/ark:/87278/s6sj6mv3). Three days after treatment initiation, her improved mental status allowed formal hearing testing, she underwent FIG. 1. The left panel shows the patients’ attempt to look right (A), she had a complete right gaze palsy; the center panel (B), shows normal straight ahead gaze; and the right panel (C), shows the response to a right head turn, with a partial left conjugate gaze. She had vertical gaze palsy (not photographed). Note normal palpebral fissures ; the pupils were 3 mm and reacted to light. e656 Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Axial T2 FLAIR MRI. Note symmetric increased signal in the thalami (arrows).The diffusion-weighted MRI showed hyperintense signal (restricted diffusion) with hypointensity in the adjusted diffusion coefficient (ADC) map not shown. bedside, pure tone audiometry, which demonstrated moderately severe hearing loss across all frequencies and severely restricted speech comprehension (Fig. 3). Her truncal balance improved and she was able to stand and take steps with a walker 1 week after admission. On outpatient, follow-up 3 months after initiation of thiamine replacement, she had persistently reduced gain of horizontal VOR (Table 1 and Fig. 4). Saccades, pursuits, and gaze holding had normalized. She was able to stand and walk with a wide base and had no recollection of any events during her hospital stay; her mental status was otherwise normal. In addition, her bilateral low- frequency sensorineural hearing loss improved, and her speech discrimination normalized. DISCUSSION In general, the association of complete EO and bilateral sensorineural hearing loss is highly suggestive of a mitochondrial disorder, usually developed over a prolonged period (chronic progressive external ophthalmoplegia: CPEO) and generally without encephalopathy (2). In this case, the patient, with the WE triad, who by virtue of her gastric bypass 25 years earlier, fits the Caine’s diagnostic FIG. 3. Sequential pure tone audiometry. The first test, obtained a few days after admission (left panel) showed lowfrequency and high-frequency hearing loss with better hearing in the 1.500–3.000 Hz. Note stip drop in high frequencies. Audiometry 3 months later (right panel), showed improved low-frequency hearing bilaterally by 30 dB; however, higher frequencies remained abnormal (right panel), probably due to pre-existing presbyacusis. Note the significant improvement in speech comprehension. Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 e657 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. A. Video head impulse test (vHIT) of the horizontal canals. The blue tracing is the head movement, and the black is the eye movement. Note decreased VOR gain and corrective overt saccades (arrows). Comparatively the gain of all 4 vertical canals (not shown) was normal (Table 1). This vHIT pattern is frequent in WE. B. Video head impulse test (vHIT) was obtained 3 months after admission. Note decreased, although slightly improved horizontal canal gain. Comparatively the gain of all 4 vertical canals was normal. She compensated effectively with covert saccades a. This vHIT pattern is frequent in WE. criteria as well. We documented hearing loss once her encephalopathy improved on treatment, retrospectively; poor speech comprehension was a likely cofactor in her initial inability to follow commands, as her husband pointed out. Although the pathophysiology of WE has not been fully elucidated; there is a complicated multiple gene–dependent thiamine transporter system that may be responsible for the heterogeneity of the clinical findings (16). For neurotransmitters’, glutamate excitotoxicity (17) and decreased cholinergic signaling have been implicated (18,19). Impaired immune response, impaired tight junction protein expression, and altered antioxidant capacity may be contributing factors (20). Neuropathology shows bilateral hemorrhages, neuronal loss, demyelination, and vascular proliferation and edema principally in mammillary bodies, medial thalami, and periaqueductal gray (1). In the early phase of WE, brainstem nuclei located near the area postrema, a physiologic blood–brain-barrier (BBB) gap, are often affected, particularly the medial vestibular nucleus (MVN) (1,21). The evolution of the ocular motor findings in this case, suggests initial involvement of horizontal (sixth nerve nucleus). By contrast, vertical eye movements recovered rapidly, without gaze holding failure, but the horizontal eye e658 movement improved at a slower pace as noted in a previous report (4). We documented sequential bilateral and asymmetric involvement of the horizontal neural integrator (bilateral prepositus hypoglossi and medial vestibular nuclei). In WE, horizontal gaze and abducens paresis must return to normal, to enable evaluation of vestibular function; fortunately, in this clinical setting, the eye movements (the efferent arc of the vestibulo-ocular reflex: VOR) respond rapidly to thiamine replacement, and bedside accurate vestibular testing is possible within a matter of days. In this case, when we performed the HIT, she had normal ocular ductions and versions. Decreased horizontal VOR gain suggests irreversible MVN lesions, which did not improve at the 3-month follow-up examination; the vertical VOR was normal 5 days after the initial evaluation, which is a frequent finding in thiamine deficiency (13,14). Horizontal nystagmus, while found in Wernicke’s original publication is not listed as a key WE characteristic; however, it was by far the most common finding in previous large series 198 of 232 cases (85%) (1) and in a series of 52 World War II prisoners (22). Victor, Adams, and Collins reported dramatic improvement in sixth nerve palsies and gaze deficits and nystagmus (1,22), usually apparent within 24 hours after Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution administration of parenteral thiamine. Interestingly, in our case, initially, there was relatively greater paresis of ocular abduction (cholinergic motoneurons) than contralateral adduction (glutamatergic interneurons) (23), supporting the cholinergic dysfunction hypothesis, at least in the brainstem nuclei. Vertical gaze, including vertical neural integrator function and VOR, quickly normalized after thiamine administration. The level of cooperation on initial examination did not allow us to test the integrity of the vertical VOR; however, lack of upper eyelid ptosis and pupillary involvement suggest preservation of third nerve nuclear structures. Previous neuropathological examination in WE revealed involvement of the periaqueductal gray and pretectal region, sparing the third nerve nucleus in n = 7/33 cases (1) which may explain the absence of ptosis or pupillary abnormalities in our patient. One additional feature of this case was reversible lowfrequency hearing loss that improved after treatment and severe impairment of speech comprehension exceeding that expected from the severity of the pure tone thresholds, and thus suggestive of a central auditory pathway conduction defect (24). Although our patient did not complain of decreased hearing, her husband felt that new hearing loss occurred (Fig. 3) and prompted formal evaluation. Besides the WE triad, when tested pervious WE hearing loss, patients showed low and high frequency sensorineural hearing loss and coexistent vestibular/ocular motor findings; reported MRI lesions in the inferior colliculi were responsible for severe impairment of speech recognition in several cases (9,11,12) other reports, noted bilateral lesions in the thalami, as found in our patient (Fig. 2). Pathologic evidence of inferior colliculus involvement in WE is infrequent (1). It is however common in primate models of intermittent thiamine deficiency (25). Interestingly, in our case, hearing improved despite partial, chronic loss of horizontal canal vestibular function, which probably related to irreversible medial vestibular nuclei lesions, which was also an early WE target in the experimental model (25). A 30-dB improvement in low frequencies in our patient, suggests coexistent cochlea or cochlear nuclei compromise. She probably had pre-existing high-frequency hearing loss. The significant heterogeneity of the abnormal neurologic, vestibular, auditory, and systemic findings in patients with thiamine deficiency probably reflects the duration of the nutritional deficit at a cellular level. As mentioned above, variability stems from genetic differences; availability of the intestinal, neuronal, and mitochondrial transporters; the integrity of the BBB and optimized thiamine phosphorylation; and adequate intracellular magnesium, an essential reaction cofactor (16). The body stores range between 30 and 50 mg and can effectively deplete within 2–3 weeks, particularly if malnutrition is complicated with recurrent vomiting. In summary, the clinical manifestations of WE are variable and the early diagnosis critical. The initial findings Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 may be nonspecific, hearing loss, and in particular, dissociation of pure auditory tone thresholds and speech comprehension may be an important diagnostic finding in the proper clinical context. In addition, the sequential improvement of the oculomotor findings with intravenous thiamine and the preferential compromise of the horizontal VOR, sparing the vertical VOR, play a diagnostic role, as serum levels are available only retrospectively and may not be always accurate (3). Measuring pretreatment serum levels of thiamine is important to diagnose those patients who fail to improve because of an alternative diagnosis and thus require additional work-up. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: all the authors contributed to the intellectual content of the manuscript and approved the final resubmission; b. Acquisition of data: the authors personally examined the patient, and contributed to the intellectual content of the manuscript and approved the final resubmission; c. Analysis and interpretation of data: the authors had an opportunity to review the data and participated in the analysis and approved this final version. Category 2: a. Drafting the manuscript: the authors participated in the preparation of the first manuscript draft and approved this final version; b. Revising it for intellectual content: the authors participated in the preparation of the first manuscript draft and approved this final version. Category 3: a. Final approval of the completed manuscript: the authors participated in the preparation of the first manuscript draft and approved this final version. REFERENCES 1. Victor M, Adams R, Collins G. The Wernicke Korsakoff Syndrome. Philadelphia, PA: F. A. Davis Company, 1971. 2. Nesbitt V, Pitceathly RD, Turnbull DM, Taylor RW, Sweeney MG, Mudanohwo EE, Rahman S, Hanna MG, McFarland R. The UK MRC Mitochondrial Disease Patient Cohort Study: clinical phenotypes associated with the m.3243A.G mutationimplications for diagnosis and management. J Neurol Neurosurg Psychiatry. 2013;84:936–938. 3. Sinha S, Kataria A, Kolla BP, Thusius N, Loukianova LL. Wernicke encephalopathy-clinical pearls. Mayo Clin Proc. 2019;94:1065–1072. 4. Kim WJ, Kim MM. Wernicke’s encephalopathy presenting with bilateral complete horizontal and downward gaze palsy in a malnourished patient. Korean J Ophthalmol. 2017;31:372–374. 5. Caine D, Halliday GM, Kril JJ, Harper CG. Operational criteria for the classification of chronic alcoholics: identification of Wernicke’s encephalopathy. J Neurol Neurosurg Psychiatry. 1997;62:51–60. 6. Kondo K, Fujiwara M, Murase M, Kodera Y, Akiyama S, Ito K, Takagi H. Severe acute metabolic acidosis and Wernicke’s encephalopathy following chemotherapy with 5-fluorouracil and cisplatin: case report and review of the literature. Jpn J Clin Oncol. 1996;26:234–236. 7. Buscaglia J, Faris J. Unsteady, unfocused, and unable to hear. Am J Med. 2005;118:1215–1217. 8. Flabeau O, Foubert-Samier A, Meissner W, Tison F. Hearing and seeing: unusual early signs of Wernicke encephalopathy. Neurology. 2008;71:694. 9. Zhang SQ, Guan YT. Acute bilateral deafness as the first symptom of Wernicke encephalopathy. AJNR Am J Neuroradiol. 2012;33:E44–E45. 10. Jethava A, Dasanu CA. Acute Wernicke encephalopathy and sensorineural hearing loss complicating bariatric surgery. Conn Med. 2012;76:603–605. e659 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 11. Walker MA, Zepeda R, Afari HA, Cohen AB. Hearing loss in Wernicke encephalopathy. Neurol Clin Pract. 2014;4:511– 515. 12. Nguyen JTT, Franconi C, Prentice A, Wycoco V. Wernicke encephalopathy hearing loss and palinacousis. Intern Med J. 2019;49:536–539. 13. Choi KD, Oh SY, Kim HJ, Kim JS. The vestibulo-ocular reflexes during head impulse in Wernicke’s encephalopathy. J Neurol Neurosurg Psychiatry. 2007;78:1161–1162. 14. Kattah JCDS, Pula JH, Mantokoudis G, Therani AS,Newman Toker DE. Vestibular signs in non-encephalopathic Wernicke’s disease. Neurol Clin Pract. 2013;3:460–467. 15. Lee SH, Kim SH, Kim JM, Tarnutzer AA. Vestibular dysfunction in Wernicke’s encephalopathy: predominant impairment of the horizontal semicircular canals. Front Neurol. 2018;9:141. 16. Ortigoza-Escobar JD, Alfadhel M, Molero-Luis M, Darin N, Spiegel R, de Coo IF, Gerards M, Taylor RW, Artuch R, Nashabat M, Rodríguez-Pombo P, Tabarki B, Pérez-Dueñas B; Thiamine Deficiency Study Group. Thiamine deficiency in childhood with attention to genetic causes: survival and outcome predictors. Ann Neurol. 2017;82:317–330. 17. McEntee WJ. Wernicke’s encephalopathy: an excitotoxicity hypothesis. Metab Brain Dis. 1997;12:183–192. 18. Jankowska-Kulawy A, Bielarczyk H, Pawelczyk T, Wroblewska M, Szutowicz A. Acetyl-CoA and acetylcholine metabolism in e660 19. 20. 21. 22. 23. 24. 25. nerve terminal compartment of thiamine deficient rat brain. J Neurochem. 2010;115:333–342. Nardone R, Holler Y, Storti M, Christova M, Tezzon F, Golaszewski S, Trinka E, Brigo F. Thiamine deficiency induced neurochemical, neuroanatomical, and neuropsychological alterations: a reappraisal. Scientific World Journal. 2013;2013:309143. Wen LM, Feng L, Jiang WD, Liu Y, Wu P, Zhao J, Jiang J, Kuang SY, Tang L, Tang WN, Zhang YA, Zhou XQ. Thiamin deficiency induces impaired fish gill immune responses, tight junction protein expression and antioxidant capacity: roles of the NFkappaB, TOR, p38 MAPK and Nrf2 signaling molecules. Fish Shellfish Immunol. 2016;51:373–383. Kattah JC, Guede C, Hassanzadeh B. The medial vestibular nuclei, a vulnerable target in thiamine deficiency. J Neurol. 2018;265:213–215. De Wardener H, Lennox B. Cerebral beriberi (Wernicke’s encephalopathy); review of 52 cases in a Singapore prisonerof-war hospital. Lancet. 1947;1:11–17. Horn Anaja KE, Adamczyck C. Handbook of Behavioral Neurosciences. Human Nervous System. 3rd ed: Elsevier, 2012. Hoistad DL, Hain TC. Central hearing loss with a bilateral inferior colliculus lesion. Audiol Neurootol. 2003;8:111–113. Witt ED, Goldman-Rakic PS. Intermittent thiamine deficiency in the rhesus monkey. I. Progression of neurological signs and neuroanatomical lesions. Ann Neurol. 1983;13:376–395. Nyce et al: J Neuro-Ophthalmol 2021; 41: e655-e660 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.