Abstract Alcohol use disorders (AUDs) have been associated with both an increased incidence and worsened outcomes in sepsis. Thiamine (Vitamin B1) deficiency is common in patients with AUDs and can lead to devastating sequela, including neurological, metabolic, and cardiovascular complications. European guidelines recommend parenteral thiamine for the prevention of neurologic complications in hospitalized patients with known AUDs, however there is little focus on cardiovascular complications or the potential for death. Objective data is lacking as to whether patients with AUDs and sepsis are at increased risk of death without the provision of thiamine. In the present proposal, we seek to characterize practice patterns surrounding thiamine administration amongst patents with septic shock and concomitant AUD. Moreover, we hypothesize that septic patients with AUDs who do not receive intravenous thiamine will have increased hospital mortality when compared with patients who do receive thiamine. We support this with the following preliminary data: 1) Patients with AUDs are well-known to have thiamine deficiency, 2) Patients with thiamine deficiency disorders develop profound life-threatening acidosis that is rapidly responsive to thiamine, 3) Single-center data shows a surprisingly high number of septic shock patients with AUDs not receiving thiamine, 4) There is a high incidence of thiamine deficiency in septic shock patients independent of AUDs, and 5) A prospective randomized trial illustrated that the administration of thiamine reduces mortality in patients with sepsis and thiamine deficiency. For this study, we will leverage a large data registry containing information from over 480 health-care facilities across the Unites States. Specifically, we will investigate the proportion of patients with AUD who are admitted to the intensive care unit with septic shock and do/do not receive thiamine. Further, clinical outcomes in patients who received thiamine will be compared to those who did not receive thiamine, using a time-dependent propensity based analysis with risk set matching. If our hypotheses are proven true, we anticipate that this study will lead to a shift in management of septic patients with AUDs with an increased focus on early thiamine administration. As AUDs are common in septic populations and are effectively a proxy for thiamine deficiency, demonstrating improved outcomes from thiamine administration may spark the widespread provision of thiamine to the larger septic population. Finally, the epidemiology of untreated thiamine deficiency amongst septic patients with AUD and its potential effects on hospital outcomes represents an unexplored avenue of investigation.

Project Narrative Patients with alcohol dependence often suffer from thiamine (vitamin B1) deficiency which is a key component of metabolism which may protect against cardiovascular collapse and death. However, while thiamine is recognized as a neuroprotective agent (i.e., preventing Wernicke's Encephalopathy), the cardiovascular protections are often not considered and remain unproven. In addition, provision of thiamine is currently not the standard of care for septic patients with alcohol use disorders. In this large-scale data analysis, we will evaluate the potential for thiamine to prevent death in subjects admitted to hospitals with alcohol use disorders. This work has the potential to change the standard of care in management of those with alcohol use disorders who develop sepsis.