The classic triad of Wernicke encephalopathy (WE) includes encephalopathy, ocular motor dysfunction, and gait ataxia. Often forgotten is that, in 1881, Wernicke also reported optic disc edema in two of his patients and disc hyperemia in all cases (1). We describe a patient with WE who primarily presented with vision loss and optic disc edema.
A 48-year-old woman sought medical attention because of a 2-week history of painless, progressive, symmetric vision loss in both eyes. Three weeks before evaluation, she was admitted to the hospital for intractable vomiting due to severe gastritis. Ten days into her stay, she noticed painless, bilateral visual changes. Initially, she stated that only the superior half of her vision was affected. Over the next several days, she claimed that her vision deteriorated to the point that she could only see “lights and shadows” in both eyes. On discharge, she saw an ophthalmologist who detected light perception vision in both eyes and right optic disc edema. This prompted a referral to our medical center for further evaluation. She had a medical history of schizoaffective disorder. The patient denied exposures to alcohol, moonshine, or illicit substances. She had previously lived in a group home due to poor self-care; but, in the months leading up to admission, she had lived independently with the help of case workers. The patient described her diet as “1 to 2 meals per day” and “mostly junk food.”
On examination, the patient was oriented and conversant. Her vision was light perception in both eyes. Pupils measured 7 mm and were sluggish to light without a relative afferent pupillary defect. She had full extraocular movements without pain; there was no nystagmus. The right optic disc was edematous and the left disc was normal. The ophthalmic and neurologic examinations were otherwise normal, although the patient's gait was not tested due to poor vision.
On admission to the hospital, MRI of the brain and orbits with contrast and fat suppression was unremarkable (Fig. 1A, B). Lumbar puncture revealed a cerebrospinal fluid glucose of 65 mg/dL (normal 40–75 mg/dL), protein of 80 mg/dL (normal 18–53 mg/dL), and 2 white blood cells/mL. The opening pressure was 14 cm of water. Hematologic studies that were either normal or negative included erythrocyte sedmentation rate, C-reactive protein, drug screen, AQP4, angiotensin-converting-enzyme, venereal disease research laboratory, HIV, toxoplasmosis, hepatitis panel, T4, and an electrolyte panel. The vitamin B12 level was on the lower side of the laboratory's normal range at 440 ng/mL. The patient was placed on a standard hospital diet and given 1 gram of intravenous methylprednisolone for presumed bilateral inflammatory optic neuropathy. After her first dose of intravenous steroids, the patient's speech and thoughts became incoherent. The steroids were discontinued. Over the next several days, she became progressively more restless and encephalopathic. Although the psychiatry service did not believe this was related to her schizoaffective disorder, they nevertheless changed her psychotropic medications from ziprasidone to olanzapine. Six days into her admission, MRI of the brain revealed interval development of increased T2 signal in the mammillary bodies, periaqueductal gray matter, and hypothalamus with contrast enhancement of the mammillary bodies (Fig. 1C, D). These findings were consistent with WE. Serum thiamine levels were undetectable. Other abnormally low laboratory values included folate (2.1 ng/mL; normal >5.9 ng/mL), potassium (2.9 mM/L; normal 3.0–4.5 mM/L), magnesium (1.6 mg/dL; normal 1.7–2.5 mg/dL), and protein (5.0 g/dL; normal 6.0–8.3 g/dL). She had developed a fine horizontal gaze-evoked jerk nystagmus present in both directions of lateral gaze. The patient was started on thiamine (1.5 g daily intravenous for 2 days, followed by 500 mg intravenously for 5 days, followed by 100 mg oral daily indefinitely), folic acid (1 mg oral daily), a daily multivitamin, electrolyte repletion, and a high calorie diet. Within 1 week, her mental status improved, as did her visual acuity to 20/100 in each eye.
In our patient with WE, optic neuropathy was the prominent presenting clinical feature. Initially, she had neither the classic clinical triad nor a diagnostic MRI of WE. Refeeding and administration of corticosteroids were associated with unmasking of the encephalopathy. The diagnosis of WE was made 1 week later with characteristic MRI findings, undetectable serum thiamine levels, and clinical improvement with proper nutrition and thiamine supplementation.
Although optic neuropathy with disc edema has been well described in association with WE, it is exceedingly rare for WE to manifest with isolated vision loss. In nearly all reported cases, visual dysfunction constitutes a single feature in a constellation of other clinical findings that variably include confusion, gait ataxia, ophthalmoplegia, vestibular dysfunction, polyneuropathy, nystagmus, and characteristic brain MRI abnormalities. Any of these changes can be present or absent. Indeed, in one report of autopsy-confirmed cases, only 17% of patients had the full clinical triad and 19% had no features of the triad (1). This suggests that WE can be easily overlooked and underdiagnosed on the basis of classic clinical symptoms alone. Given its variable clinical presentation, the only pathognomonic feature of WE is the rapid improvement of signs and symptoms after thiamine supplementation (2).
Another unusual aspect of our case was the patient's neuroimaging. Her MRI initially was normal, but a repeat study 6 days later showed classic evidence of WE (Fig. 1). WE typically demonstrates symmetric increased T2 signal in the periaqueductal gray matter, periventricular regions of the third ventricle, medial thalami, tectal plate, and mammillary bodies and contrast enhancement of the mammillary bodies (2). One study found these MRI findings to be 93% specific for WE (3). Another report documented that 80% of patients with clinically diagnosed WE demonstrated stereotypical MRI changes in the thalamus and periventricular areas (4). These regions are believed to have a particularly high concentration of thiamine-dependent cellular channels that maintain an osmotic gradient. With thiamine depletion, reversible cytotoxic edema accumulates in the interstitial space. Our patient demonstrated that radiographically appreciable edema may not manifest until later stages of the disease.
A likely cause of our patient's worsening clinical course was refeeding syndrome, a complication of feeding a chronically malnourished patient (5). When total body vitamin and mineral stores are low, the sudden administration of glucose leads to electrolyte shifts. Cells internalize nutrients, such a thiamine, and the serum is left with critically low levels of vitamins and minerals. Refeeding syndrome underlies the classic teaching to give thiamine before dextrose to an alcoholic patient. In our patient's case, her history of vomiting and a “junk food diet” contributed to chronic malnutrition. Refeeding syndrome contributed to acute malnutrition and thiamine deficiency, which could explain her worsening course, electrolyte values, and MRI after she received a standard hospital diet.
Although clinical, laboratory, and neuroimaging studies strongly suggest that WE was the primary contributor to her optic neuropathy, the patient's level of residual impairment was atypical for WE and other coincident optic neuropathies should therefore be considered. Folate deficiency rarely can contribute to an optic neuropathy. Refeeding syndrome causes an entire spectrum of nutritional deficiencies that also may have contributed. Unusual toxic exposures cannot be ruled out. Moreover, without a baseline eye examination, any preceding optic neuropathies also cannot be excluded.
Our case demonstrates that optic neuropathy may rarely be a presenting feature of WE. In addition, WE should be considered even in the absence of typical MRI abnormalities, and precautions should be taken to avoid exacerbating clinical symptoms with refeeding syndrome.
STATEMENT OF AUTHORSHIP
Category 1: a. conception and design: A. A. Sura and L. B. Kline; b. acquisition of data: A. A. Sura and J. K. Cure; c. analysis and interpretation of data: A. A. Sura, J. K. Cure, and L. B. Kline. Category 2: a. drafting the manuscript: A. A. Sura and L. B. Kline; b. revising it for intellectual content: A. A. Sura, J. K. Cure, and L. B. Kline. Category 3: a. final approval of the completed manuscript: A. A. Sura, J. K. Cure, and L. B. Kline.
1. Harper CG, Giles M, Finlay-Jones R. Clinical signs in the Wernicke-Korsakoff complex: a retrospective analysis of 131 cases diagnosed at necropsy. J Neurol Neurosurg Psychiatry. 1986;49:341–345.
2. Sechi G, Serra A. Wernicke's encephalopathy: new clinical settings and recent advances in diagnosis and management. Lancet Neurol. 2007;6:442–455.
3. Antunez E, Estruch R, Cardenal C, Nicolas JM, Fernandez-Sola J, Urbano-Marquez A. Usefulness of CT and MRI imaging in the diagnosis of acute Wernicke's encephalopathy. Am J Roentgenol. 1998;171:1131–1137.
4. Zuccoli G, Santa Cruz D, Bertolini M, Rovira A, Gallucci M, Carollo C, Pipitone N. MR imaging findings in 56 patients with Wernicke encephalopathy: nonalcoholics may differ from alcoholics. AJNR Am J Neuroradiol. 2009;30:171–176.
5. Mehanna HM, Moledina J, Travis J. Refeeding syndrome: what it is, and how to prevent and treat it. BMJ. 2008;336:1495–1498.