Thiamine administration to all patients with alcohol use disorder: why not?

In the United States (US), there are almost 5 million alcohol-related visits to Emergency Departments (EDs) per year (1). Many of these visits involve patients with alcohol use disorder (AUD) who have, or are at risk for, thiamine deficiency. These individuals should receive thiamine supplementation (2–4). However, thiamine deficiency is difficult to identify because initial symptoms are often nonspecific, e.g., headaches, fatigue, irritability, and abdominal discomfort (5,6). The lack of thiamine supplementation in patients with thiamine deficiency can contribute to heart failure and sudden death, Wernicke’s encephalopathy, and irreversible brain damage. The latter includes Korsakoff’s psychosis, which is characterized by anterograde amnesia, an inability to remember recent events, and confabulation (5,6). Because Wernicke’s encephalopathy and Korsakoff’s psychosis have a shared etiology and often co-occur, they are often referred to as Wernicke–Korsakoff syndrome (WKS).

The guidelines for the prevention and treatment of Wernicke’s encephalopathy explicitly indicate the need for rapid administration of high doses of parenteral thiamine (100 to 1500 mg per day, possibly in multiple doses per day), until the patient’s improvement is observed (2–4). Unfortunately, there is no specific routine laboratory test available to assess thiamine deficiency; the direct measurement of thiamine and/or its phosphorylated forms (e.g., thiamine diphosphate, which represents about 80% of the total body content of thiamine) do not assess thiamine metabolic function and there are no universally accepted cutoff values for thiamine deficiency (7). Therefore, Wernicke’s encephalopathy remains primarily a clinical-based diagnosis (5). Nonetheless, in addition to the clinical evaluation, magnetic resonance imaging (MRI) may have an important role in confirming the clinical diagnosis of Wernicke’s encephalopathy, as it presents with peculiar neuroimaging findings (8). Perhaps more relevant, and as reviewed in Sullivan and Pfefferbaum (9), once the clinical diagnosis is made, MRI may be helpful in assessing the resolution of Wernicke’s encephalopathy in patients receiving thiamine treatment or its worsening and progression to Korsakoff’s psychosis in those without treatment.

Although thiamine deficiency may also occur in several other medical conditions characterized by nutritional deficiencies, including gastrointestinal surgery, recurrent vomiting, chronic diarrhea, cancer and chemotherapeutic treatments, alcohol misuse is the most common risk factor for thiamine deficiency (5,6,10). In almost 90% of Wernicke’s encephalopathy cases, patients have a history of alcohol misuse, and approximately 80% of cases are not recognized prior to death (11). Indeed, autopsy studies have shown that the frequency of Wernicke’s encephalopathy is over 15 times higher among patients with AUD than in patients without AUD (12).

Based on the clinical considerations and findings summarized above, thiamine should be administered to all patients with AUD admitted to EDs, either prophylactically or to treat thiamine deficiency (2–4). Furthermore, EDs represent an ideal setting where thiamine should be administered via the parenteral route to patients with AUD. Yet, that very often does not happen (10). The low rate of thiamine treatment in patients with AUD has been further documented by Peck et al. (13) in this issue of the American Journal of Drug and Alcohol Abuse. In this study, Dr. Peck and colleagues evaluated the thiamine prescription rates in an ED setting among patients with suspected or diagnosed AUD (13). The study was based on a retrospective chart review conducted in two urban teaching hospitals in New York City with 170,000 total patient visits per year. The authors identified 7,529 patients (1,782, or 23.7% female) visits associated with a primary alcohol-related diagnosis between January 1, 2017 to December 31, 2017, more than 4.4% of the total patients visits per year. Most of these patients had a chart diagnosis of alcohol abuse (91.5%) or alcohol dependence (7.8%), while less than 1% had an unspecified alcohol use. The authors report the alarming finding that thiamine was prescribed only in 2.2% of these alcohol-related visits (13). Thus, despite the guidelines for the prevention and treatment of Wernicke’s encephalopathy recommending supplementation of thiamine to patients with AUD in an ED setting (2–4,10), most individuals at risk of Wernicke’s encephalopathy very often did not receive thiamine supplementation in an ED setting. Furthermore, the study by Peck et al. (13) reported that the mean age of patients with AUD was approximately 45 years, suggesting that they had probably been drinking excessive amounts of alcohol for over 20 years (see e.g., 14) and likely had already developed alcohol-related consequences, including malnutrition and thiamine deficiency.

There are no scientific or clinical reasons why patients with AUD seen in an ED should not receive thiamine supplementation, yet it appears that this often the exception rather than the norm. For patients with thiamine deficiency, thiamine supplementation is the logical state-of-the-art treatment that these patients should receive. Although there is presently not sufficient evidence to provide recommendations on the optimal route of thiamine administration (15), from a practical standpoint, if the clinical status of the patient does not prevent the safe placement of an intravenous line, the intravenous route may be a pharmacologically better choice while the patient is in the ED or hospitalized. This should be followed by oral thiamine supplementation after discharge. Failure to administer thiamine involves severe and irreversible consequences. It represents the gold-standard treatment for patients with thiamine deficiency, just like insulin is for patients with Type I diabetes mellitus. Thiamine administration has virtually no side effects and is a hydrophilic vitamin, therefore there is no risk of overdosing because it does not accumulate in the body. The only risks are nonspecific to thiamine and related to the site of injection, when given intravenously; these risks are, however, rare (16–18).

The results from Peck et al. (13) also reflect a larger issue. That is, patients with AUD and those with other substance use disorders often do not receive proper or state-of-the-art treatment for their substance use disorder. For example, we have effective medications for AUD (i.e., acamprosate, disulfiram, and naltrexone) which are approved by the regulatory agencies like the Food and Drug Administration and European Medicines Agency (19). These medications may help patients reduce alcohol consumption, and thereby decrease the related risk of mortality (20). Yet, these medications are significantly underutilized even though the risk of mortality for patients with AUD is 3 to 4 times higher than general population (21). Of note, the overall rate of thiamine administration in patients with AUD or alcohol misuse, described in the present study by Peck and colleagues (13) was 2.2%, a rate very similar to that of patients who receive medications approved for AUD in the US, for example, 1.6% in a recent report by Han and colleagues (22). Such low prescriptions rates of gold-standard treatments would be unthinkable and unacceptable for other chronic health conditions such as hypertension, diabetes, or hypothyroidism, just to name a few. Reasons that contribute to this issue include the lack of enough well-trained physicians and other health care providers with expertise in addiction (23) and the stigma around AUD, substance use disorders and mental illness (24).

Despite that the retrospective nature of the study is a limitation that needs to be taken into account, it is worth considering that such a design also holds important strengths. For example, the large number of patient data obtained and the fact that the results reflect what is happening in a real-world clinical setting, such as the EDs of two hospitals in one of the largest metropolitan areas in the world in a first world country with state-of-the-art medical systems. The latter aspect, however, is also a reminder that these results should be expanded and replicated by accessing nation-wide, large health care databases. Nonetheless, this study remains of clinical relevance given the enormous number of alcohol-related visits in EDs, a finding that has further increased during the past years (1). Furthermore, given that malnutrition is common also in patients with opioid and other substance use disorders, it is worth noting that the clinical importance of these results may be relevant not only for AUD, but also for other addictions as well, as noted by Peck and colleagues (13). The merit of Dr. Peck and his colleagues lies in having observed what happens to their patients and drawing important conclusions.

We hope that these relevant findings may contribute to increasing awareness of the too frequent and unacceptable underdiagnosis and undertreatment of patients with AUD. To achieve this goal, among other possible actions, overcoming barriers due to stigma and expanding addiction-related training among healthcare professionals are critically needed (19). But the urgent question that the study by Peck and colleagues (13) poses is also what can be done, from a clinical practice standpoint, in the short term to increase the rates of thiamine supplementation in patients with AUD admitted to EDs. A simple strategy is prescribing thiamine supplementation to all patients with AUD, regardless of the presence or not of symptoms suggesting thiamine deficiency. An example of a simple protocol that could be implemented is the one proposed by the European Federation of the Neurological Societies guidelines: 200 mg thrice daily, preferably intravenously, both for the prevention and treatment of Wernicke’s encephalopathy (3). For patients with Wernicke’s encephalopathy, treatment should continue until resolution of the clinical symptomatology (3). For the prevention, thiamine supplementation may be stopped at discharge.

This approach will benefit all patients with AUD, in terms of prevention and treatment. Considering that thiamine deficiency is common among patients with AUD, that the consequences of the lack of thiamine administration are serious and often irreversible, that thiamine supplementation is effective, safe and inexpensive, and that the intravenous route is typically easy to employ in ED settings, we ask . . . why not?