Scoping review finds undertreatment of Wernicke encephalopathy in adults with alcohol use

A scoping review of 26 studies involving 44,240 adults found that Wernicke encephalopathy, a potentially reversible neurological emergency associated with thiamine deficiency, remains underrecognized and undertreated in people with alcohol use disorder or clinically significant alcohol use. Across observational studies, clinical audits and implementation reports, investigators repeatedly found delayed treatment initiation, frequent use of oral thiamine in high-risk or suspected cases, and substantial variation in diagnostic criteria, dosing and prescribing protocols. The synthesis, published in Addiction Science & Clinical Practice, did not identify an optimal thiamine regimen, but it mapped where care diverges and which gaps appear most actionable.

Chronic alcohol exposure can disrupt thiamine balance through reduced intestinal absorption, depleted hepatic storage, impaired conversion to thiamine pyrophosphate and inadequate dietary intake. Thiamine is an essential cofactor for enzymes connecting glycolysis with the tricarboxylic acid cycle and pentose phosphate pathway. Acute depletion can therefore impair oxidative ATP production, increase oxidative stress and predispose to excitotoxic neuronal injury, particularly in metabolically vulnerable regions such as the thalamus and mammillary bodies. During alcohol withdrawal, a hyperadrenergic state and increased cerebral glucose use further elevate thiamine requirements. If inadequately treated, Wernicke encephalopathy may progress to Korsakoff syndrome, characterized by severe anterograde amnesia and substantial long-term care needs.

Recognition is difficult in the environments where patients commonly receive care, including emergency departments, general hospitals, psychiatric units and medically assisted withdrawal services. Intoxication, withdrawal, delirium, malnutrition, trauma, infection and other comorbidities can obscure the syndrome. The classical triad of encephalopathy, gait ataxia and ocular motor dysfunction is relatively insensitive: older neuropathological series found all three features in only 16–20% of cases, while more recent meta-analytic evidence placed the full triad in approximately one third of clinically identified cases. Postmortem studies have also reported Wernicke-related lesions in as many as 12% of individuals with alcohol use disorder, including cases not diagnosed during life, indicating substantial underrecognition rather than merely disagreement over treatment.

Expanded diagnostic criteria were designed to improve sensitivity. The Caine criteria support a diagnosis when at least two of four features are present: dietary deficiency, oculomotor abnormalities, cerebellar dysfunction, and altered mental status or mild memory impairment. However, fewer than half of the included studies applied formal diagnostic definitions for either alcohol use disorder or Wernicke encephalopathy. Many relied on clinical impression or risk-based inference, alcohol-related complications or blood alcohol concentrations, while relatively few used structured DSM or ICD criteria or validated screening instruments such as AUDIT-C or CAGE. Magnetic resonance imaging can support diagnosis, but the review did not treat imaging as a prerequisite for treatment. Biochemical markers of thiamine status were also reported inconsistently and generally served as supportive evidence rather than stand-alone diagnostic criteria.

Thiamine regimens differed by a factor of hundreds across the literature. Reported daily doses ranged from 5 mg to 1,500 mg and were given orally, intramuscularly or intravenously. Three broad patterns emerged: low-dose regimens of 50 mg per day or less, generally used for prophylaxis or maintenance; intermediate doses of 100–200 mg per day across preventive, equivocal or less clearly defined contexts; and high-dose intravenous regimens of 500 mg per day or more, most often associated with suspected or clinically established encephalopathy. Once-daily administration was common, whereas three-times-daily regimens appeared more often in suspected or established cases. Short intravenous courses of one to three days were commonly followed by oral maintenance once patients were clinically stable, but the boundaries among prevention, treatment and maintenance were inconsistently defined.

Direct evidence capable of resolving the dosing question was scarce. Only two randomized controlled trials evaluated thiamine regimens, and both were rated at high risk of bias because of limitations involving outcome measurement, attrition or reporting. One compared different intramuscular doses in patients with alcohol use disorder and reported modest improvements in working memory with the higher dose. The other compared three intravenous regimens in symptomatic and at-risk patients and found no clear dose-response differences in cognitive or neurological outcomes. The remaining evidence consisted mainly of observational studies, quasi-experimental studies, clinical audits and implementation-focused reports. These sources illuminated routine practice but were vulnerable to confounding, changing case definitions and inconsistent outcome measures. The reviewers did not perform a formal certainty assessment because the scope was to map the evidence rather than estimate pooled treatment effects.

Prescribing audits showed recurring mismatches between treatment and high-risk clinical circumstances. Therapy was often initiated later than expected, and low-dose oral thiamine was frequently administered even in acute or high-risk presentations. By contrast, studies describing more protocolized and guideline-concordant care more often reported earlier treatment and greater use of parenteral thiamine in suspected or confirmed cases. Implementation measures such as electronic order sets, withdrawal protocols and structured prescribing guidance were associated with better adherence to recommended doses and routes, although substantial variation remained between institutions. These are associations rather than proof that a protocol prevented neurological harm: retrospective designs, confounding and local availability of formulations limited causal interpretation. They nevertheless identified delayed recognition and inconsistent implementation as recurring features of care that could be modified.

The biological relationship between thiamine deficiency and overt encephalopathy adds another layer of uncertainty. The review noted that biochemical deficiency can occur without clinical Wernicke encephalopathy, suggesting that deficiency alone does not invariably produce the syndrome. Susceptibility may be influenced by nutritional status, comorbid illness, genetic differences in thiamine transport and additional vitamin deficiencies, all of which may affect neurological vulnerability. A laboratory indicator of low thiamine therefore cannot simply be equated with clinical disease, while the absence of the classical triad does not reliably exclude it. Because clinicians cannot reliably predict which deficient or high-risk patient will progress, the authors identified early recognition and a low threshold for parenteral treatment in high-risk or suspected cases as priorities supported by the mapped literature.

The review has important limitations. Many included studies were retrospective, single-center investigations with small samples and heterogeneous endpoints, and no included study enrolled an exclusively outpatient population. Long-term cognitive recovery and recurrence were rarely assessed. All studies came from high-income countries, with nine conducted in the United States, while women, older adults and populations in low- and middle-income countries were underrepresented or absent. The synthesis cannot determine an optimal dose, establish a reliable dose-response relationship or prove that any particular protocol reduces neurological harm. Its strongest conclusion is descriptive: diagnostic ambiguity, delayed treatment and inconsistent prescribing persist within a narrow and methodologically weak randomized evidence base. Future work should distinguish biochemical thiamine deficiency, suspected encephalopathy, confirmed encephalopathy and Korsakoff syndrome, then test not only doses but also timing, route, duration and implementation in pragmatic trials. More standardized case definitions and clearer protocols across acute-care, psychiatric and addiction-treatment settings could address current inconsistencies while definitive comparative evidence is developed.

Subject of Research: Diagnosis and thiamine management of Wernicke encephalopathy in adults with alcohol use disorder or clinically significant alcohol use

Article Title: Undertreatment of Wernicke encephalopathy in alcohol-related clinical contexts: a scoping review of diagnosis, thiamine regimens, and clinical management

Article References: Goldani, A. A. S., Feitosa, G., Colchete, E. V., Falleiro, C. F., Pedro, C. W., Kessler, F. H. P., Fidalgo, T. M., & von Diemen, L. (2026). Undertreatment of Wernicke encephalopathy in alcohol-related clinical contexts: a scoping review of diagnosis, thiamine regimens, and clinical management. Addiction Science & Clinical Practice, 21(1), Article 80. https://doi.org/10.1186/s13722-026-00687-x

Image Credits: AI Generated

DOI: 10.1186/s13722-026-00687-x

Keywords: Wernicke encephalopathy, thiamine deficiency, alcohol use disorder, scoping review, undertreatment, parenteral thiamine, clinical diagnosis, prescribing practices, Korsakoff syndrome, Wernicke, encephalopathy, alcohol-related