You take the perilous sign out of a 55-year-old patient with alcohol use disorder (AUD) who came to the ED intoxicated. “He’s just metabolizing” your colleague tells you. A few hours go by and he is now awake and asking to leave. The nurse road tests him and he can walk, but is a little “shaky.” When you examine him, his gait appears ataxic. Is this normal?
A 30-something guy had a recent stay for EtOH withdrawal, and was discharged home. During your shift, hisis wife brings him back to the ED and says he just “isn’t right.” He doesn’t remember things the way he should and often seems confused. Is this just his new baseline?
A 40-year-old man was sent to the ED from jail with a concern for alcohol withdrawal. He has normal vital signs, no tremor, but he appears confused and has a dysconjugate gaze on exam. Are we missing something?
All three of these patients should make you think of one of my favorite diagnoses – Wernicke’s encephalopathy.
Wernicke’s encephalopathy (WE), also called “dry” beriberi, is a neurologic disorder caused by thiamine (vitamin B1) deficiency, first diagnosed by Carl Wernicke in 3 paitents in 1881.1 Thiamine, found in organ meats, eggs, and green leafy vegetables, is vital to the utilization of cellular energy. In its active form, thiamine diphosphate acts as a coenzyme in the pyruvate dehydrogenase complex. This enzymatic reaction converts pyruvate into acetyl-coA, linking glycolysis to the citric acid cycle (AKA the Kreb’s Cycle). This step is paramount because it acts as the gate-keeper, linking the inefficient process of anaerobic metabolism (glycolysis) to the powerhouse of aerobic metabolism (citric acid cycle). Together, this allows for the production of a whopping 36 moles of ATP for every mole of glucose. In the setting of thiamine deficiency, aerobic metabolism is inhibited, and ATP production plummets. Wimpy glycolysis becomes the only ATP producer, giving up a meager 2 moles of ATP per mole glucose. If the last few sentences made your eyes glaze over, just refer to this graphic:
Thiamine is also a required cofactor for several steps within the citric acid cycle and is instrumental in the pentose phosphate pathway (I’ll mercifully spare you those details). Check this reference for more.2 In addition to those essential roles, thiamine also acts as a neuromodulator, influencing the release of acetylcholine.
Moral of the story: Eat your greens (or organ meats). But don’t cook them too long; thiamine is very water soluble and will leach out after prolonged soaking or cooking in water.
How does this enzymatic inhibition lead to the neurologic injury of WE? The direct mechanism is unclear, but animal models suggest increased glutamate concentrations (possibly resulting from that citric acid cycle defect I promised not to describe) may be involved.3
Administration of 100 mg IV thiamine prophylactically protects against deficiency for approximately one week. Parenteral administration of thiamine is typically recommended over oral administration. It has been classically thought that absorption of thiamine from the GI tract was limited and saturable even in the absence of alcohol,4,5 however more recent data challenge that belief.6 What is clear is that absorption of thiamine is reduced in the presence of alcohol or malnutrition. A study of healthy human volunteers showed a 50% reduction in absorption of oral thiamine in the presence of EtOH, and there have been cases of alcoholics who developed WE despite high dose oral thiamine supplementation.7 While intramuscular administration of thiamine should work, poor muscle mass and possible coagulopathy in alcoholics make it a less desirable, and less reliable route.
Clinical Presentation & Diagnosis
Wernicke’s encephalopathy consists of 3 classic clinical features: ophthalmoplegia, ataxia, and altered mental status/confusion.1 Alas, patients don’t always follow the textbook, so the clinical picture is often incomplete.8 Other features can include hypothermia, decreased deep tendon reflexes, and eventually a peripheral neuropathy. Tachycardia is often an early sign of thiamine deficiency. Labs may show a lactate associated anion gap metabolic acidosis (AGMA). This is secondary to the preferential conversion of pyruvate to lactate, not acetyl-CoA, in the absence of thiamine. Which raises an important point: thiamine deficiency should be on the ddx for AGMA or unexplained elevated lactates. My first line treatment for undifferentiated AGMA often includes thiamine, glucose, and fluids, but that is a different post for a different day. For untreated patients, WE can progress into Korsakoff’s psychosis, an irreversible syndrome characterized by confabulation (see below) and short-term memory impairments/anterograde amnesia. WE and Wernicke-Korsakoff syndromes have been associated with a 10-20% mortality or higher, owing to bacterial infections and cancer, amongst other causes.9
It is estimated that WE is present in up to 2% of the US population, but remains largely undiagnosed. It’s a challenging diagnosis to make. To start, the disorder occurs in a very vulnerable patient population that may receive less scrutiny due to the assumption that they are in the ED just to “metabolize”. The altered mental status can encompass a broad differential, and the ophthalmoplegia is rarely complete. It can manifest as ptosis, gaze palsies, dysconjugate gaze, and papilledema, with the most common finding being nystagmus. The findings of Wernicke’s sound a lot like an intoxicated patient, right? Most patients drunk enough to find him or herself in the ED has some form of altered mental status, can’t walk, and may even have nystagmus or a dysconjugate gaze. But which ones have WE?
There’s a tool (Caine criteria) that can help identify patients with WE consisting of 4 components:
- dietary deficiencies
- oculomotor abnormalities
- cerebellar dysfunction
- altered mental status or mild memory impairment
When 2 or more elements are present, the tool is very sensitive and surprisingly specific.10
When you think back to your last month of ED shifts, how many people with AUD (who likely meet criteria #1) also have one of the other features? How do we figure this out?
Of course, not all alcoholics have Wernicke’s, but the first step (the most often missed step), is simply considering the diagnosis and identifying patients at risk. This includes those with AUD due to poor consumption and absorption of thiamine. Interestingly, chronic ethanol use may also modify genes involved in thiamine absorption, increasing their vulnerability to WE. Others at risk include patients with:
- Fad diets
- Eating disorders
- Bariatric surgery
- Hyperemesis gravidarum
The institutionalized elderly and prisoners may also be at risk due to poor dietary intake. And patients with CHF on high dose loop diuretics are also at risk due to increased elimination of thiamine.11–14
The key to distinguishing intoxication from WE is that in the former, these abnormal findings resolve when the acute intoxication resolves. If an intoxicated patient sobers up and is asking to leave, take a moment to do a good exam, checking for signs of WE. If the patient still has memory issues, an ataxic gait, or abnormal eye movements, this should raise some major red flags for you. Don’t expect to see the full triad. And don’t blow off confusion or ataxia, ascribing it to intoxication, if you think the patient is clinically sober.
Imaging, should you obtain it, classically shows hemorrhage into the mammillary bodies or third and fourth ventricles. Such advanced findings are in no way necessary to make the diagnosis, and absence of these findings should not change your decision making process.
The treatment for suspected WE is thiamine. But please note, it is not the dose you are used to giving to your AUD patients. Different sources recommend somewhat different doses of thiamine for WE, however the most widely accepted dose is 500 mg IV TID x 2-3 days, followed by 250 mg IV daily x 3-5 days.15 The goal is to restore CNS levels of thiamine as rapidly as possible. To do this, we have to take advantage of the passive uptake of thiamine into the CNS (the active transport mechanism is rate-limited, and saturable) by establishing a high plasma to CNS thiamine ratio. Frequent dosing is required as thiamine levels fall precipitously after parenteral administration. Want to delve more deeply into this? CCH Cook’s 1997 paper is an excellent read.4
To say it again: giving a patient a single standard dose of thiamine does not treat WE.
Where does this high dose recommendation come from? It’s not hard science, but it makes mechanistic sense and patients have failed lower dose therapies of thiamine.7,16 To add support, this high-dose regimen has been adopted by several professional organizations in Europe and the United States and is part of the National Institute of Health and Clinical Excellence (NICE) guidelines.17 Importantly, there are very few risks to high dose thiamine (namely anaphylaxis, which is exceedingly rare),18,19 lending a definitive tip to favor the benefit over risk ratio in such a critical condition. Recovery can be rapid, with some patients demonstrating improvement before leaving the ED, although delayed and incomplete recoveries are also reported.7 A noticeable improvement in confusion is typically seen within 1-2 days; other features may resolve more rapidly.4
A relative myth is that administering glucose before thiamine will cause your patient’s head to explode. . .
It is true, however, that a glucose load will increase thiamine requirements. The concern is that in patients that are already thiamine deficient, a dextrose bolus might push them over the edge. There is no clear evidence that this occurs in patients not already exhibiting overt signs of thiamine deficiency. In practice, I typically give thiamine with glucose, but I don’t worry about which one gets administered first. Would I think twice about ordering an isolated glucose load in a patient for whom I was considering WE? Sure. But I would never withhold glucose in a hypoglycemic patient? No way.
Magnesium is a cofactor for key thiamine dependent enzymes, and supplementation has been shown to improve transketolase (see below) activity when given with thiamine. Administering 2 g MgSO4 IV to patients with normal kidney function is recommended by Goldfrank’s, and your local toxicologists.3
Want to prove that your patient is thiamine deficient? You can get an erythrocyte transketolase test which will indirectly measure thiamine activity. This will not come back in a clinically relevant time period (and is not well validated) but may be useful in certain patients in which the diagnosis is unclear.20
And don’t forget the other side of thiamine deficiency: “wet” beriberi. Clinically, this is high output heart failure. Interestingly, most patients will only manifest one of the two beriberi syndromes. Dietary variations may be responsible for which version of beriberi manifests, which may be why Asians tend to develop the cardiac version and Europeans develop WE.
Think about WE.
This is the number 1, 2, and 3 take home point. It happens more than you expect. As EM physicians, this is your diagnosis to make. Don’t expect the full triad. Use the diagnostic tool to raise suspicion. Treat WE with HIGH DOSE PARENTERAL THIAMINE. This means the patient requires admission. And as always, please consult your local toxicologist when suspecting WE.Smile be Happy by Ahmed Zayan
- 1.Thomson A, Cook C, Guerrini I, Sheedy D, Harper C, Marshall E. Wernicke’s encephalopathy revisited. Translation of the case history section of the original manuscript by Carl Wernicke “Lehrbuch der Gehirnkrankheiten fur Aerzte and Studirende” (1881) with a commentary. Alcohol Alcohol. 2008;43(2):174-179. https://www.ncbi.nlm.nih.gov/pubmed/18056751.
- 2.Aleguas A. Thiamine. In: Brent J, Burkhart K, Dargan P, et al., eds. Critical Care Toxicology. 2nd ed. Cham, Switzerland: Springer International Publishing; 2017:2993-2998.
- 3.Hoffman R. Thiamine Hydrochloride. In: Nelson L, Howland M, Lewin N, Smith S, Goldfrank L, Hoffman R, eds. Goldfrank’s Toxicologic Emergencies. 11th ed. New York: McGraw Hill; 2019:1157-1164.
- 4.Cook C, Hallwood P, Thomson A. B Vitamin deficiency and neuropsychiatric syndromes in alcohol misuse. Alcohol Alcohol. 1998;33(4):317-336. https://www.ncbi.nlm.nih.gov/pubmed/9719389.
- 5.Thomson A, Baker H, Leevy C. Patterns of 35S-thiamine hydrochloride absorption in the malnourished alcoholic patient. J Lab Clin Med. 1970;76(1):34-45. https://www.ncbi.nlm.nih.gov/pubmed/4912963.
- 6.Smithline H, Donnino M, Greenblatt D. Pharmacokinetics of high-dose oral thiamine hydrochloride in healthy subjects. BMC Clin Pharmacol. 2012;12:4. https://www.ncbi.nlm.nih.gov/pubmed/22305197.
- 7.Chataway J, Hardman E. Thiamine in Wernicke’s syndrome–how much and how long? Postgrad Med J. 1995;71(834):249. https://www.ncbi.nlm.nih.gov/pubmed/7784292.
- 8.Harper C, 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(4):341-345. https://www.ncbi.nlm.nih.gov/pubmed/3701343.
- 9.Sanvisens A, Zuluaga P, Fuster D, et al. Long-Term Mortality of Patients with an Alcohol-Related Wernicke-Korsakoff Syndrome. Alcohol Alcohol. 2017;52(4):466-471. https://www.ncbi.nlm.nih.gov/pubmed/28340112.
- 10.Caine D, Halliday G, Kril J, Harper C. Operational criteria for the classification of chronic alcoholics: identification of Wernicke’s encephalopathy. J Neurol Neurosurg Psychiatry. 1997;62(1):51-60. https://www.ncbi.nlm.nih.gov/pubmed/9010400.
- 11.Katta N, Balla S, Alpert M. Does Long-Term Furosemide Therapy Cause Thiamine Deficiency in Patients with Heart Failure? A Focused Review. Am J Med. 2016;129(7):753.e7-753.e11. https://www.ncbi.nlm.nih.gov/pubmed/26899752.
- 12.Hung S, Hung S, Tarng D, Yang W, Chen T, Huang T. Thiamine deficiency and unexplained encephalopathy in hemodialysis and peritoneal dialysis patients. Am J Kidney Dis. 2001;38(5):941-947. https://www.ncbi.nlm.nih.gov/pubmed/11684545.
- 13.Aasheim E. Wernicke encephalopathy after bariatric surgery: a systematic review. Ann Surg. 2008;248(5):714-720. https://www.ncbi.nlm.nih.gov/pubmed/18948797.
- 14.Krishnan M, N B, K T, S S. Wernicke’s Encephalopathy: A Rare Complication of Hyperemesis Gravidarum. J Assoc Physicians India. 2018;66(6):87. https://www.ncbi.nlm.nih.gov/pubmed/31331145.
- 15.Galvin R, Bråthen G, Ivashynka A, et al. EFNS guidelines for diagnosis, therapy and prevention of Wernicke encephalopathy. Eur J Neurol. 2010;17(12):1408-1418. https://www.ncbi.nlm.nih.gov/pubmed/20642790.
- 16.Ambrose M, Bowden S, Whelan G. Thiamin treatment and working memory function of alcohol-dependent people: preliminary findings. Alcohol Clin Exp Res. 2001;25(1):112-116. https://www.ncbi.nlm.nih.gov/pubmed/11198705.
- 17.Thomson A, Cook C, Touquet R, Henry J, Royal C. The Royal College of Physicians report on alcohol: guidelines for managing Wernicke’s encephalopathy in the accident and Emergency Department. Alcohol Alcohol. 2002;37(6):513-521. https://www.ncbi.nlm.nih.gov/pubmed/12414541.
- 18.Stephen J, Grant R, Yeh C. Anaphylaxis from administration of intravenous thiamine. Am J Emerg Med. 1992;10(1):61-63. https://www.ncbi.nlm.nih.gov/pubmed/1736919.
- 19.Thomson A, Guerrini I, Marshall E. Incidence of Adverse Reactions to Parenteral Thiamine in the Treatment of Wernicke’s Encephalopathy, and Recommendations. Alcohol Alcohol. September 2019. https://www.ncbi.nlm.nih.gov/pubmed/31565743.
- 20.Blansjaar B, Zwang R, Blijenberg B. No transketolase abnormalities in Wernicke-Korsakoff patients. J Neurol Sci. 1991;106(1):88-90. https://www.ncbi.nlm.nih.gov/pubmed/1779244.