Thiamine deficiency in critically ill patients with sepsis

Objective: The objective of the study was to determine the prevalence of absolute thiamine deficiency (TD) in critically ill patients with sepsis and to examine the association between thiamine levels and lactic acidosis.

Design: This was a prospective, observational study.

Setting: The setting was an urban, tertiary care center with approximately 50,000 emergency department visits per year and intensive care units numbering approximately 50 total beds.

Patients: Thirty study patients admitted with clinical suspicion of infection and evidence of tissue hypoperfusion, as defined by a lactic acid level greater than 4 mmol/L or hypotension (systolic blood pressure <90 mm Hg) requiring vasopressor support, were enrolled. A control group of 30 patients presenting to the emergency department with minor emergencies was also enrolled.

Interventions: There were no interventions.

Measurements and main results: Plasma thiamine levels were measured at 0, 24, 48, 72, and 162 hours for patients in the study group. Absolute TD was defined as less than or equal to 9 nmol/L derived from established abnormal ranges per Quest laboratory. In the study group, 3 (10%) of 30 had absolute TD upon presentation; and an additional 3 patients (6/30, 20%) developed TD within 72 hours. None of the 30 controls (0/30, 0%) exhibited absolute TD. Of the vasopressor-dependent population, 7.7% (2/26) displayed TD on presentation. For the group overall, there was no correlation between thiamine and lactic acidosis. However, in patients without liver dysfunction, thiamine was statistically significantly negatively correlated with lactic acidosis (r = -.50; P = .02). The relationship between thiamine and lactic acidosis held after multivariable regression analysis controlling for age, sex, and comorbid disease (P < .02).

Conclusions: These preliminary findings indicate that critically ill patients may present with TD or develop this deficiency during their acute illness. We also identified a potential association between thiamine levels and lactic acidosis in patients without significant liver injury.