Secondary Logo

Share this article on:

Unpeeling the Evidence for the Banana Bag: Evidence-Based Recommendations for the Management of Alcohol-Associated Vitamin and Electrolyte Deficiencies in the ICU

Flannery, Alexander H. PharmD, BCCCP, BCPS; Adkins, David A. MD, FCCP; Cook, Aaron M. PharmD, BCPS

doi: 10.1097/CCM.0000000000001659
Review Article

Objective: Patients with a chronic alcohol use disorder presenting to the ICU may be deficient in important vitamins and electrolytes and are often prescribed a “banana bag” as a reflexive standard of therapy. The difficulty of diagnosing Wernicke’s encephalopathy in the critical care setting is reviewed. Furthermore, whether the contents and doses of micronutrients and electrolytes in standard banana bags meet the needs of critically ill patients with an alcohol use disorder is assessed based on available evidence.

Data Source: MEDLINE/PubMed (1966 to June 2015) database search, the Cochrane Database of Systematic Reviews, and manual selection of bibliographies from selected articles.

Study Selection and Data Extraction: Articles relevant to Wernicke’s encephalopathy, vitamin and electrolyte deficiencies in patients with alcohol use disorders, and alcoholic ketoacidosis were selected. Articles were narratively synthesized for this review.

Data Synthesis: Of these deficiencies, thiamine is the most important for the practicing clinician to assess and prescribe replacement in a timely manner. Based on a pharmacokinetic assessment of thiamine, the banana bag approach likely fails to optimize delivery of thiamine to the central nervous system. Folic acid and magnesium may also merit supplementation although the available data do not allow for as strong a recommendation as for prescribing thiamine in this setting. There is no available evidence supporting the prescription of a multivitamin.

Conclusions: Based on the published literature, for patients with a chronic alcohol use disorder admitted to the ICU with symptoms that may mimic or mask Wernicke’s encephalopathy, we suggest abandoning the banana bag and utilizing the following formula for routine supplementation during the first day of admission: 200–500 mg IV thiamine every 8 hours, 64 mg/kg magnesium sulfate (approximately 4–5 g for most adult patients), and 400–1,000 μg IV folate. If alcoholic ketoacidosis is suspected, dextrose-containing fluids are recommended over normal saline.

1Department of Pharmacy Services, University of Kentucky HealthCare, Lexington, KY.

2Department of Pharmacy Practice and Science, University of Kentucky College of Pharmacy, Lexington, KY.

3Division of Pulmonary, Critical Care, and Sleep Medicine, University of Kentucky College of Medicine, Lexington, KY.

The authors have disclosed that they do not have any potential conflicts of interest.

This work was performed at the University of Kentucky.

For information regarding this article, E-mail:

Patients with a chronic alcohol use disorder (AUD) who present with acute illness often exhibit deficiencies in essential vitamins and nutrients (1). Clinicians in the ICU are often challenged by identifying these disease states in critically ill patients. Neurologic manifestations of deficiencies are often obscured by the nature of the patient’s critical illness and/or therapies used early in the ICU stay, such as sedation. In the United States, patients with a suspected AUD are often empirically prescribed “banana bags” or “rally packs” at admission to the ICU. This prescription typically consists of 1 L of IV fluid infused over 24 hours and commonly contains some combination of 100 mg of thiamine, 1 mg of folic acid, 1–2 g of magnesium, and a multivitamin formulation in either normal saline (NS) or dextrose in water (D5W) solutions. The multivitamin, along with folic acid, gives the banana bag its characteristic yellow hue.

In this review, we examine the common vitamin and electrolyte deficiencies in patients with AUDs and re-examine whether the current standard of care (i.e., banana bags) meets the needs of critically ill patients with an AUD based on the available evidence. MEDLINE/PubMed (1966 to June 2015) was searched with a combination of the following search terms: ethanol, alcohol, thiamin*, vitamin B1, Wernicke*, encephalopathy, withdrawal, magnesium, electrolyte, folic acid, folate, multivitamin, and ketoacidosis. The Cochrane Database of Systematic Reviews was searched with similar terms. Articles were manually selected from the reference lists of certain articles as applicable.

Back to Top | Article Outline


Wernicke’s encephalopathy (WE) and Korsakoff syndrome (KS) are linked disease states that complicate the care of patients with a chronic AUD. The classic triad of WE symptoms includes mental status changes, ophthalmoplegia, and ataxia. Other symptoms can include hypothermia, vestibular dysfunction, and nystagmus. Approximately 80% of patients with acute WE develop KS, which is typified by persistent memory deficits including anterograde and retrograde amnesia, often accompanied by confabulation (2).

Thiamine is commonly administered for the prevention or treatment of WE/KS (3). If symptoms of WE are not promptly recognized, or if thiamine dosing is inadequate, irreversible structural changes in the brain may result. This may lead to the development of short-term memory loss (4). In the critically ill, determining the actual risk of thiamine deficiency in patients with a vague history of an AUD or those with an acute intoxication can be difficult; the line between prophylaxis and treatment often blurs.

Although autopsy studies suggest a prevalence of WE of 12.5% in alcoholics, it is likely underrecognized in the ICU (5). The most common reported clinical manifestation in a necropsy study of 131 cases of WE/KS was altered mental status, a frequent and nonspecific finding among ICU patients with a myriad of possible explanations (sepsis, hypoxia, stroke, intoxication, encephalopathy, and sedation among others) (6). Only 20% of patients had received a clinical diagnosis despite the fact that the majority of patients had been seen in a hospital toward the end of their life (6). Similarly, only 10% of patients with structural changes associated with WE manifested the complete classic triad of symptoms (4, 6). Although generally considered a rare presentation of WE, one autopsy study revealed 11 of 36 patients with WE presented with coma. None of these patients carried the diagnosis prior to death (7). It has also been suggested the pathogenesis of alcoholic cerebellar degeneration may be similar to WE (8). These findings have led some authors to propose that 1) traditional diagnostic criteria for WE may be too rigid and 2) a subclinical form of WE exists that, with repeated episodes, could result in similar pathology as seen with classic WE/KS (6). The relative lack of availability and prolonged turn-around time of a thiamine assay mandates prompt clinical recognition and treatment of thiamine deficiency (4). Recently, the European Federation of Neurological Societies (EFNS) endorsed diagnostic criteria for WE/KS derived from a retrospective cohort study of alcoholics (3, 9). Two of the following four features must be present: nutritional deficiency, ocular abnormalities, ataxia (cerebellar dysfunction), or mental status changes or memory impairment (9). Nutritional deficiencies and altered mental status occur in virtually all patients with a severe AUD admitted for intensive care.

Determining the extent and impact of WE in modern healthcare, especially among the critically ill, remains a challenge. However, to illustrate the point of an increasing recognition of WE, a MEDLINE/PubMed search from 1990 to 2014 using the terms “Wernicke*” and “encephalopathy” was undertaken. Results were limited to the English language, filtered to case reports only, and categorized by year (Fig. 1). Not all of these cases are due to alcohol, and the number of published case reports is certainly an inaccurate surrogate for prevalence of a disease state, but this does illustrate the increasing recognition of WE as a significant problem in multiple patient populations, including those with an AUD. It has previously been estimated that greater than 95% of WE cases are secondary to AUDs (10). The exact number of missed clinical and subclinical cases of WE is unknown, but based on necropsy studies from past decades, history would suggest that it is underdiagnosed.

Figure 1

Figure 1

The mechanisms by which thiamine deficiency disrupt metabolism are illustrated in Figure 2. In the thiamine-depleted state, the impairment of the thiamine-dependent enzyme transketolase disrupts the pentose phosphate pathway and impairs myelin sheath maintenance, lipid and glucose metabolism, and branched chain amino acid production. The conversion of pyruvate to acetyl coenzyme A is also reduced, increasing lactic acid production. Finally, thiamine is required as a co-factor for the conversion of α-ketoglutarate to succinate in the Krebs cycle, which may result in shunting the cycle to glutamate production, free radical production, and neuroexcitotoxicity. The disruption of the Krebs cycle results in reduced adenosine triphosphate (ATP) production and cell deficit, which in concert with the above mechanisms, ultimately results in cell death (4). These overwhelming metabolic demands on thiamine-depleted cells may be increased in a state of critical illness. Whole-body stores of thiamine are approximately 30–50 mg, with a required daily intake of 1.5 mg (10). Patients with a significant AUD are at risk for thiamine deficiency due to decreased dietary intake, impaired absorption from the gastrointestinal tract, and decreased conversion to its active phosphorylated form (11).

Figure 2

Figure 2

For patients requiring ICU admission, a strong argument can be made against the use of oral thiamine due to its poor bioavailability. In healthy volunteers, the maximum amount of thiamine absorbed approximates 4.5 mg (from a 30-mg dose). Once the dose exceeds 30 mg, no further absorption occurs due to the rate-limited absorption process (10). Alcoholics exhibit reduced thiamine absorption resulting in serum levels 30–98% below the lower limit of normal (10). Given the bioavailability of enteral thiamine is already low, the additional barriers to thiamine absorption in alcoholics would suggest the absorption is even lower in the severely depleted. For example, one group has proposed that a regimen of 100 mg oral tid would result in absorption that is one-third or less that of healthy subjects, or approximately 5 mg in a 24-hour period, which is inadequate for those with severe deficiencies (4). Oral thiamine, when compared with IV thiamine, failed to increase cerebrospinal fluid (CSF) levels of thiamine or transketolase activity in red blood cells (12). As such, enteral thiamine has no role in the treatment of patients with an AUD early in the ICU stay.

The optimal dosing of thiamine in patients at risk of WE/KS has been a subject of substantial debate, in part, due to the lack of randomized, controlled trials. The argument for higher dosing is based on the premise that thiamine is transported into the brain by both an active, saturable mechanism and a passive mechanism, making a steep concentration gradient between the serum and the central nervous system (CNS) an attractive goal of dosing (4). The active rate-limited transfer of thiamine across the blood-brain barrier occurs at a rate of 0.3 μg/hr/g of brain tissue, which is equivalent to the level of brain thiamine turnover (10). Although active transfer may be enough to meet cerebral needs under normal conditions, it may be inadequate in critically ill thiamine-depleted patients. Passive absorption across the blood-brain barrier normally constitutes less than 10% of the total thiamine transport in animal models, necessitating high levels of thiamine in the blood to optimize the diffusion process (13). Genetic predisposition, as well as potential alcohol toxicity to the apoenzyme system, may also lead to the requirement for higher parenteral doses seen in some alcoholics (14).

The 100-mg dose in most banana bags is reported to have been chosen arbitrarily by two authors in the 1950s based on their approximation of what represented a high dose of thiamine (2). For patients with confirmed WE, this dose is typically ineffective. In a necropsy series of WE patients, 10% of patients that died in-hospital were receiving thiamine, arguably at insufficient doses (6). In a landmark series in Boston, treatment of patients with confirmed WE with 50–100 mg of parenteral thiamine resulted in only 16% of patients having a complete recovery, whereas 84% of the cohort went on to develop KS and suffered from significant long-term impairments (15). Therefore, evidence suggests that commonly used doses may be suboptimal and may not adequately replete body stores, improve clinical signs and symptoms, or prevent mortality (6, 8, 16). Some reports document required doses of up to 1 g of thiamine for WE within the first 24 hours (17, 18).

As the use of these time-honored remedies is re-addressed, new data must be taken into account, particularly pharmacokinetic data. Thiamine is rapidly eliminated from plasma into the urine, has a plasma half-life of approximately 1.5 hours, and returns to initial levels within 6–12 hours following a single 50-mg IV dose (19). Extrapolating the pharmacokinetics of thiamine administration in healthy volunteers to the critically ill suggests that dosing intervals of every 8–12 hours may optimize levels for blood-brain barrier transport. Therefore, the banana bag as it is traditionally administered may not allow for passive absorption of thiamine into the CNS at all given the constant rate of infusion.

The minimum dose of thiamine appears to be at least 200 mg daily for patients with suspected or proven WE (Table 1). The Royal College of Physicians report recommends more than 500 mg of thiamine qd or bid for 3–5 days for the prevention or treatment of WE in alcoholic patients, with the acknowledgement that the data for this dose come from uncontrolled trials and empirical clinical practice (4, 8, 20). The EFNS, in contrast, recommends 200 mg tid in suspected or manifested WE (3).



The dose of thiamine may impact not only mortality but also memory function of ICU survivors admitted with severe deficiency. Alcoholics who presented without the clinical triad of WE undergoing detoxification were studied in a randomized double-blind multidose study of thiamine (21). This trial represents the only randomized trial identified by a Cochrane review investigating thiamine in WE (22). The results suggest a relationship between thiamine dose and working memory performance, indicating at least 200 mg of intramuscular thiamine daily may be required to improve symptoms (21). Although one study suggested a more prolonged IV infusion resulted in improved utilization of thiamine throughout the dosing interval when compared with an IV bolus, this study failed to look at CSF levels of thiamine and to optimize the concentration gradient for passive diffusion of thiamine into the CSF (23).

The fear of anaphylaxis associated with parenteral thiamine remains an exaggerated largely unfounded concern. In a prospective study of 989 patients given 100 mg of thiamine as a bolus injection, no occurrences of anaphylaxis were noted (24). The same authors later reported a retrospective series in which over 300,000 patients had received parenteral thiamine without significant issue (25). In the United Kingdom, the rate of anaphylaxis was estimated to be 1 in 250,000 administrations (8). An infusion, as opposed to bolus injection, may further lessen the likelihood and severity of an anaphylactoid reaction (8). As with any medication, rare cases of anaphylaxis may be seen; therefore, appropriate precautions should be taken (26).

A final point of discussion involves the classic teaching to administer thiamine prior to glucose. Theoretically, administering a glucose load in the setting of inadequate thiamine stores may shut down the pyruvate to acetyl coenzyme A conversion, limiting glycolysis and the link to the Krebs cycle, which may precipitate the development of WE (27). For a thorough review, readers are referred to an excellent comprehensive literature review by Schabelman and Kuo (27). In this review of the literature, the authors found that animal data best support this teaching, as thiamine-deficient rats given infusions of glucose demonstrated changes in brain imaging in as little as 40 minutes (28, 29). However, case reports of this phenomenon in humans are less clear with regard to the importance of timing and amount of glucose. The most cited reference for this phenomenon in humans appears to be a case series of four patients published in the early 1980s; however, as pointed out by some critics, none of these cases resulted in an acute conversion/worsening of WE following the acute administration of glucose (30, 31). Case reports describing this phenomenon have generally involved at least 1–2 L of dextrose or more, and occurred 24 hours or more after the administration of glucose (27). Given the substantial harms of hypoglycemia and conflicting literature to date regarding the administration of thiamine prior to glucose, we do not suggest that glucose administration be delayed in a hypoglycemic patient with an AUD while awaiting the preparation and administration of thiamine. As prolonged glucose administration, but less likely acute glucose administration, may be associated with the worsening of WE, thiamine administration should occur as soon as possible following the consideration of WE on the differential diagnosis.

At least one group of authors has questioned whether the intensive care community is giving adequate attention to thiamine as a prophylaxis for delirium, particularly in the trauma population (32). WE may mask itself as a “wolf in sheep’s clothing” within the context of ICU delirium, a ubiquitous condition in critical care, and may delay appropriate treatment (33). Even in the context of modern day care, failures to appropriately identify and utilize appropriate doses of thiamine to treat WE remain documented (34–36). Given the difficulty of diagnosis, the poor prognosis associated with delayed recognition and inadequate treatment of WE, and minimal drug-drug interactions and adverse effects of thiamine, these authors believe the risk-benefit favors taking thiamine out of the banana bag and administering at treatment doses of 200–500 mg every 8 hours for at least 72 hours or until WE is ruled out.

Back to Top | Article Outline


Folate is necessary for DNA synthesis and the production of normal erythrocytes among a variety of other functions. Dietary folate consumption is necessary for adequate folate stores, as in vivo synthesis does not occur in humans. Deficiencies in folate may result in several consequences, including megaloblastic anemia, and many of the neurologic sequelae of alcohol withdrawal such as confusion, sleep disturbances, depression, and psychosis (37). Low folate concentrations may also be associated with hyperhomocysteinemia, which may increase the risk of seizures in alcohol withdrawal (37).

Individuals who consume alcohol chronically may present with a folate deficiency due to poor dietary intake. Folate is typically absorbed in the duodenum via specific transporters. Alcohol intake inhibits folate absorption by reducing association with the folate transporter, as well as decreasing the expression of transporters in the small intestine (38). Even if the patient with a chronic AUD has adequate dietary intake, it is likely that much of the folate ingested will not be effectively absorbed.

Diminished folate concentrations have been associated with alcohol withdrawal syndromes in at-risk individuals and supplementation of folate seems advisable in patients with a chronic AUD (37). However, the optimal dose is not well defined. Although oral supplementation of 400 μg/d of folic acid appears to be adequate for normal individuals, the diminished absorption of folate in alcoholics may necessitate the need for higher doses or parenteral therapy. Of note, folate absorption is also saturable such that in the folate-deficient state, folate is better absorbed (39). In individuals with adequate folate stores, absorption is less, thereby mitigating some of the risk (and potentially benefit) of oral supplementation.

Sparse clinical evidence is available supporting the routine supplementation of folate in these scenarios. However, given the association of low folate concentrations with the neurologic sequelae of alcohol withdrawal and the relatively limited toxicity in the normal dosing range, supplementation of folate should be considered. IV folate administration appears to be the most reliable method of improving endogenous folate stores, particularly due to the diminished absorption of oral folate related to alcohol use. We recommend that if folate is supplemented, practitioners give 400–1000 μg IV for several days after admission (40). Oversupplementation (> 5 mg/d) is not recommended due to the potential risk of neurotoxicity (41). Conversion to oral therapy is reasonable if no concomitant alcohol consumption will occur.

Back to Top | Article Outline


Magnesium is required in multiple steps of aerobic respiration. Key enzymes in the glycolysis pathway require magnesium. Thiamine and magnesium are also interconnected via enzymatic processes. For example, thiamine diphosphate is a coenzyme for a number of mitochondrial enzymes involved in carbohydrate metabolism, including the enzyme transketolase, an integral enzyme in the pentose phosphate pathway. Magnesium is a co-factor for transketolase and is thought to contribute to maintaining the enzyme in its active form (4, 42, 43). Clinical reports have also supported this link between magnesium and thiamine. Patients with WE may be clinically unresponsive to thiamine treatment in the setting of hypomagnesemia. However, once serum magnesium concentrations are corrected, transketolase activity returns to normal and clinical symptoms resolve (44, 45).

Individuals with an AUD have increased magnesium elimination, which leads to magnesium deficiency relatively quickly when use is chronic (46). Although the mechanism of renal magnesium loss due to alcohol intake is not well defined, the increased magnesium elimination may be due, in part, to increased liberation of magnesium from tissues such as skeletal muscle and cardiac tissue (thus providing more circulating magnesium to be eliminated). It also appears that ingested magnesium is not as well absorbed in the presence of alcohol (47).

Supplementation of magnesium may be important in patients with chronic alcohol use although the evidence is low quality. Patients who present with acute withdrawal syndrome or delirium tremens often have hypomagnesemia (48). The association of hypomagnesemia with CNS alterations is likely due to the reduced concentrations of magnesium in the brain of chronic alcohol users (46). Restoration of normal magnesium concentrations often results in a reduction in withdrawal symptoms although usually in a delayed fashion (46, 49). Of note, some investigators have shown a lack of correlation with both serum and CNS magnesium concentrations in patients presenting with delirium tremens (50, 51). Poor-quality clinical data exist regarding the efficacy of magnesium in preventing withdrawal syndrome. A recent Cochrane review was unable to generate significant recommendations in support of magnesium supplementation (52).

In the typical clinical scenario, however, adequate serum magnesium concentrations are also desirable in order to reduce the risk of cardiac arrhythmias such as atrial fibrillation or torsades de pointes, as well as seizures. Limited evidence is available to guide practitioners on the appropriate magnesium concentration target. In fact, many of the patients who respond to magnesium therapy for torsades de pointes have a normal magnesium serum concentration (53). Avoidance of prolonged hypomagnesemia and frequent use of medications known to prolong the QTc interval is advisable in patients at risk of torsades de pointes (54). In patients who require pharmacotherapy with QT-prolonging agents such as haloperidol, periodic ECG monitoring should be considered (55).

The appropriate amount and duration of magnesium supplementation is also not well defined. Tissue magnesium concentrations are not fully replete after single doses of magnesium supplementation, and available evidence suggests that several days of supplementation are necessary to restore magnesium concentrations to near-normal in patients with deficiency (46, 49). Based on previously published studies describing the typical degree of magnesium deficiency in chronic alcohol users, a dose of approximately 1 mEq/kg of magnesium (in divided doses) is recommended for the first day of supplementation, with 0.5 mEq/kg/d in divided doses over the next 3 days (46). This equates to a daily supplementary dose of approximately 4.5 g of magnesium sulfate for a 70-kg adult, followed by 2–3 g/d thereafter. These patients may also be deficient in phosphorus, with up to 24% of patients presenting with hypophosphatemia (56). Careful monitoring of electrolyte administration is advised in these patients with concomitant renal impairment.

Back to Top | Article Outline


Patients with extreme malnutrition due to a chronic AUD may present with less common nutritional deficiencies such as scurvy or pellagra. The basic premise for including a multivitamin in supplemental infusions is that a wide variety of vitamins and micronutrients are included in a multivitamin, which may help to begin supplementation in patients at risk for less common nutritional deficiencies seen in extreme malnutrition (57). It is plausible that multivitamin supplementation, along with a balanced nutritional diet and alcohol abstinence, can aid in replenishing important vitamins and minerals (58). However, no published data exist investigating the efficacy or safety of acute use of multivitamin injection in patients with possible alcohol withdrawal. Although this lack of data does not necessarily negate the potential benefits of IV multivitamin in the acute setting, it is also unlikely to have a substantial treatment effect of patients with severe deficiencies. In these patients, it seems more advisable to evaluate specifically for suspected nutritional deficiencies based on clinically evident symptoms and available laboratory tests. In this way, clinicians can verify the diagnosis and provide more definitive therapy for isolated nutritional deficiencies, rather than using the low doses of vitamins and minerals provided by IV multivitamin.

Back to Top | Article Outline


The banana bag is typically mixed in a liter of NS or D5W. Evaluation of the intravascular volume status and the likelihood of starvation is necessary to determine the optimal IV fluid for critically ill patients presenting with a history of a chronic AUD. Although NS is more advantageous than D5W in patients requiring intravascular volume resuscitation, some patients with an extensive, chronic AUD may present in a starved state that begets alcoholic ketoacidosis (AKA). AKA in this situation is often underrecognized and may represent the cause of mortality in up to 7% of chronic alcohol users (59–61). A comparison of dextrose versus salt-containing fluids suggested that acidosis resolution occurs more quickly in patients receiving dextrose (62). Thus, it is advisable to utilize dextrose-containing fluids in patients who may have AKA to prevent or treat this insidious complication unless relative contraindications to hypotonic fluid administration exist. For clinicians managing volume resuscitation and maintenance fluids with separate IV fluids, the fluid component of a banana bag likely runs in the background as unnecessary liters of fluid in many cases.

Back to Top | Article Outline


Based on the available data, it is apparent that critically ill patients with a chronic AUD may be deficient in a number of vitamins and electrolytes, most importantly thiamine, folate, and magnesium. In most institutions, we suspect that the typical components of banana bags do not adequately address the neurological sequelae of these deficiencies. With the exception of thiamine, limited data exist supporting routine supplementation for most of these vitamins and electrolytes. On the other hand, the lack of evidence does not necessarily mean lack of effect of these therapies. In treating critically ill patients with a history of a chronic AUD, clinicians should consider the potential benefit for individual patients, the risks, and costs of the various ingredients to the typical banana bag, along with the available literature. Due to the difficulty of establishing a diagnosis of WE/KS and the potentially devastating consequences of missed or delayed diagnosis, empiric therapy is often warranted. For patients admitted to the ICU with symptoms that may mimic or mask WE, based on the published literature, we suggest abandoning the banana bag and utilizing the following formula for routine supplementation during the first day of admission: 200–500 mg IV thiamine every 8 hours, 64 mg/kg magnesium sulfate (≈4–5 g for most adult patients), and 400–1,000 μg IV folate. If AKA is suspected, dextrose-containing fluids are recommended over NS.

Back to Top | Article Outline


1. American Psychiatric Association: Substance-related and addictive disorders. Diagnostic and Statistical Manual of Mental Disorders. 2013Fifth Edition. Washington, DC, American Psychiatric Association
2. Donnino MW, Vega J, Miller J, et al: Myths and misconceptions of Wernicke’s encephalopathy: What every emergency physician should know. Ann Emerg Med 2007; 50:715–721
3. Galvin R, Bråthen G, Ivashynka A, et al; EFNS: EFNS guidelines for diagnosis, therapy and prevention of Wernicke encephalopathy. Eur J Neurol 2010; 17:1408–1418
4. Thomson AD, Cook CC, Touquet R, et al; The Royal College of Physicians report on alcohol: Guidelines for managing Wernicke’s encephalopathy in the accident and Emergency Department. Alcohol Alcohol 2002; 37:London: Royal College of Physicians, 513–521
5. Torvik A, Lindboe CF, Rogde S: Brain lesions in alcoholics. A neuropathological study with clinical correlations. J Neurol Sci 1982; 56:233–248
6. 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
7. Lana-Peixoto MA, Dos Santos EC, Pittella JE: Coma and death in unrecognized Wernicke’s encephalopathy. An autopsy study. Arq Neuropsiquiatr 1992; 50:329–333
8. Cook CC, Hallwood PM, Thomson AD: B Vitamin deficiency and neuropsychiatric syndromes in alcohol misuse. Alcohol Alcohol 1998; 33:317–336
9. Caine D, Halliday GM, Kril JJ, et al: Operational criteria for the classification of chronic alcoholics: Identification of Wernicke’s encephalopathy. J Neurol Neurosurg Psychiatry 1997; 62:51–60
10. Thomson AD: Mechanisms of vitamin deficiency in chronic alcohol misusers and the development of the Wernicke-Korsakoff syndrome. Alcohol Alcohol Suppl 2000; 35:2–7
11. Zahr NM, Kaufman KL, Harper CG: Clinical and pathological features of alcohol-related brain damage. Nat Rev Neurol 2011; 7:284–294
12. Baker H, Frank O: Absorption, utilization and clinical effectiveness of allithiamines compared to water-soluble thiamines. J Nutr Sci Vitaminol (Tokyo) 1976; 22:63–68
13. Greenwood J, Love ER, Pratt OE: Kinetics of thiamine transport across the blood-brain barrier in the rat. J Physiol 1982; 327:95–103
14. Thomson AD, Guerrini I, Marshall EJ: The evolution and treatment of Korsakoff’s syndrome: Out of sight, out of mind? Neuropsychol Rev 2012; 22:81–92
15. Victor M, Adams RD, Collins GH: The Wernicke-Korsakoff Syndrome and Related Neurological Disorders Due To Alcoholism and Malnutrition. 1989Second Edition. Philadelphia, PA, F.A. Davis Company
16. Brown LM, Rowe AE, Ryle PR, et al: Efficacy of vitamin supplementation in chronic alcoholics undergoing detoxification. Alcohol Alcohol 1983; 18:157–166
17. Lindberg MC, Oyler RA: Wernicke’s encephalopathy. Am Fam Physician 1990; 41:1205–1209
18. Nakada T, Knight RT: Alcohol and the central nervous system. Med Clin North Am 1984; 68:121–131
19. Tallaksen CM, Sande A, Bøhmer T, et al: Kinetics of thiamin and thiamin phosphate esters in human blood, plasma and urine after 50 mg intravenously or orally. Eur J Clin Pharmacol 1993; 44:73–78
20. Manzanares W, Hardy G: Thiamine supplementation in the critically ill. Curr Opin Clin Nutr Metab Care 2011; 14:610–617
21. Ambrose ML, Bowden SC, Whelan G: Thiamin treatment and working memory function of alcohol-dependent people: Preliminary findings. Alcohol Clin Exp Res 2001; 25:112–116
22. Day E, Bentham PW, Callaghan R, et al: Thiamine for prevention and treatment of Wernicke-Korsakoff Syndrome in people who abuse alcohol. Cochrane Database Syst Rev 2013; 7:CD004033
23. Drewe J, Delco F, Kissel T, et al: Effect of intravenous infusions of thiamine on the disposition kinetics of thiamine and its pyrophosphate. J Clin Pharm Ther 2003; 28:47–51
24. Wrenn KD, Murphy F, Slovis CM: A toxicity study of parenteral thiamine hydrochloride. Ann Emerg Med 1989; 18:867–870
25. Wrenn KD, Slovis CM: Is intravenous thiamine safe? Am J Emerg Med 1992; 10:165
26. Juel J, Pareek M, Langfrits CS, et al: Anaphylactic shock and cardiac arrest caused by thiamine infusion. BMJ Case Rep 2013; 2013
27. Schabelman E, Kuo D: Glucose before thiamine for Wernicke encephalopathy: A literature review. J Emerg Med 2012; 42:488–494
28. Jordan LR, Zelaya FO, Rose SE, et al: Changes in the hippocampus induced by glucose in thiamin deficient rats detected by MRI. Brain Res 1998; 791:347–351
29. Zelaya FO, Rose SE, Nixon PF, et al: MRI demonstration of impairment of the blood-CSF barrier by glucose administration to the thiamin-deficient rat brain. Magn Reson Imaging 1995; 13:555–561
30. Watson AJ, Walker JF, Tomkin GH, et al: Acute Wernickes encephalopathy precipitated by glucose loading. Ir J Med Sci 1981; 150:301–303
31. Hack JB, Hoffman RS: Thiamine before glucose to prevent Wernicke encephalopathy: Examining the conventional wisdom. JAMA 1998; 279:583–584
32. Blackmore C, Ouellet JF, Niven D, et al: Prevention of delirium in trauma patients: Are we giving thiamine prophylaxis a fair chance? Can J Surg 2014; 57:78–81
33. Lemyze M, Favory R, Alves I, et al: Acute delirium in a critically ill patient may be a wolf in sheep’s clothing. BMJ Case Rep 2009; 2009
34. Soler-González C, Sáez-Peñataro J, Balcells-Oliveró M, et al: Wernicke-Korsakoff’s syndrome: Waiting for Godot? Alcohol Alcohol 2014; 49:117–118
35. Isenberg-Grzeda E, Chabon B, Nicolson SE: Prescribing thiamine to inpatients with alcohol use disorders: How well are we doing? J Addict Med 2014; 8:1–5
36. Wijnia JW, Nieuwenhuis KG: Difficulties in identifying Wernicke-delirium. Eur J Intern Med 2011; 22:e160–e161
37. Thornton WE, Pray BJ: Lowered serum folate and alcohol-withdrawal syndromes. Psychosomatics 1977; 18:32–36
38. Wani NA, Thakur S, Najar RA, et al: Mechanistic insights of intestinal absorption and renal conservation of folate in chronic alcoholism. Alcohol 2013; 47:121–130
39. Santhosh-Kumar CR, Bisping JS, Kick SD, et al: Folate sufficient subjects do not accumulate additional folates during supplementation. Am J Hematol 2000; 64:71–72
40. Rice TL: Folic acid in alcohol-withdrawal delirium. Am J Health Syst Pharm 2005; 62:355–356
41. Institute of Medicine (US) Standing Committee on the Scientific Evaluation of Dietary Reference Intakes and Its Panel on Folate, Other B Vitamins, and Choline: Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. 1998, Washington, DC, National Academies Press (US), 196–305
42. Flink EB: Magnesium deficiency in human subjects–a personal historical perspective. J Am Coll Nutr 1985; 4:17–31
43. Marino PL: Magnesium. The ICU Book. 2007, Third Edition. Philadelphia, PA,Lippincott Williams & Wilkins,625–638
44. Traviesa DC: Magnesium deficiency: A possible cause of thiamine refractoriness in Wernicke-Korsakoff encephalopathy. J Neurol Neurosurg Psychiatry 1974; 37:959–962
45. Peake RW, Godber IM, Maguire D: The effect of magnesium administration on erythrocyte transketolase activity in alcoholic patients treated with thiamine. Scott Med J 2013; 58:139–142
46. Flink EB: Magnesium deficiency in alcoholism. Alcohol Clin Exp Res 1986; 10:590–594
47. McDonald JT, Margen S: Wine versus ethanol in human nutrition. III. Calcium, phosphorous, and magnesium balance. Am J Clin Nutr 1979; 32:823–833
48. Sullivan JF, Lankford HG, Swartz MJ, et al: Magnesium metabolism in alcoholism. Am J Clin Nutr 1963; 13:297–303
49. Chutkow JG: Clinical-chemical correlations in the encephalopathy of magnesium deficiency. Effect of reversal of magnesium deficits. Mayo Clin Proc 1974; 49:244–247
50. Hemmingsen R, Kramp P, Rafaelsen OJ: Delirium tremens and related clinical states. Aetiology, pathophysiology and treatment. Acta Psychiatr Scand 1979; 59:337–369
51. Kramp P, Hemmingsen R, Rafaelsen OJ: Magnesium concentrations in blood and cerebrospinal fluid during delirium tremens. Psychiatry Res 1979; 1:161–171
52. Sarai M, Tejani AM, Chan AH, et al: Magnesium for alcohol withdrawal. Cochrane Database Syst Rev 2013; 6:CD008358
53. Keren A, Tzivoni D: Torsades de pointes: Prevention and therapy. Cardiovasc Drugs Ther 1991; 5:509–513
54. Drew BJ, Ackerman MJ, Funk M, et al; American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology; Council on Cardiovascular Nursing; American College of Cardiology Foundation: Prevention of torsade de pointes in hospital settings: A scientific statement from the American Heart Association and the American College of Cardiology Foundation. J Am Coll Cardiol 2010; 55:934–947
55. Sommargren CE, Drew BJ: Preventing torsades de pointes by careful cardiac monitoring in hospital settings. AACN Adv Crit Care 2007; 18:285–293
56. Vandemergel X, Simon F: Evolution of metabolic abnormalities in alcoholic patients during withdrawal. J Addict 2015; 2015:541536
57. Katz KD: Intravenous multivitamins (“banana bags”) for emergency patients who may have nutritional deficits. Ann Emerg Med 2012; 59:413–414
58. Devika Rani K, Suneetha N, Mohanty S, et al: Association of hyperhomocysteinemia to alcohol withdrawal in chronic alcoholics. Indian J Clin Biochem 2008; 23:150–153
59. Thomsen JL, Felby S, Theilade P, et al: Alcoholic ketoacidosis as a cause of death in forensic cases. Forensic Sci Int 1995; 75:163–171
60. Thomsen JL, Simonsen KW, Felby S, et al: A prospective toxicology analysis in alcoholics. Forensic Sci Int 1997; 90:33–40
61. Palmiere C, Augsburger M: The postmortem diagnosis of alcoholic ketoacidosis. Alcohol Alcohol 2014; 49:271–281
62. Miller PD, Heinig RE, Waterhouse C: Treatment of alcoholic acidosis: The role of dextrose and phosphorus. Arch Intern Med 1978; 138:67–72

alcohol; alcohol use disorder; critical care; electrolytes; folic acid; ketoacidosis; magnesium; multivitamin; thiamine; Wernicke’s encephalopathy

Copyright © by 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.