Prevalence of Wernicke's Encephalopathy When Receiving Dextrose Before Thiamine: A National Study of Veterans.

Abstract

A commonly taught principle when treating emergency department (ED) patients with acute alcohol intoxication is to give thiamine before dextrose to avoid precipitating Wernicke's encephalopathy (WE). We sought to evaluate the prevalence of WE among a national sample of veterans who presented to the ED with alcohol intoxication and who then received dextrose before thiamine. This is a retrospective, cross-sectional study of patients presenting to all Veterans Affairs (VA) Emergency Departments (ED) over a 10-year period (2010-2019). Data was obtained through a Microsoft SQL (Redmond, WA) query of the VA Corporate Data Warehouse. Inclusion criteria were any ED visit with an alcohol intoxication ICD9/10 code and/or serum ethanol result > 50 mg/dL, as well as administration of any intravenous fluids containing dextrose at any concentration. Exclusion criteria was administration of thiamine before dextrose infusion. The primary outcome was diagnosis of WE by ICD9/10 code or manual chart review in the ED, hospitalization, or follow-up visit within 90 days. Data was analyzed with descriptive statistics. 120 encounters by 114 individual patients met the inclusion/exclusion criteria, with a median age of 59 (IQR 49-64). There were 104 (91%) male patients, 77 (68%) were white, 27 (24%) were Black, and 6 (5%) were Hispanic. Most patients with a recorded AUDIT-C screened positive for AUD (94%) or had a documented history of alcohol abuse (90%). No cases of Wernicke's encephalopathy were identified by ICD code or manual chart review. Our data suggest that alcohol-intoxicated patients are unlikely to develop Wernicke's encephalopathy with acute dextrose administration. While administering thiamine in alcohol-intoxicated patients is low risk and potentially beneficial, we submit that hypoglycemia treatment should not be delayed for this intervention.