Blood thiamine and thiamine phosphate ester concentrations in alcoholic and non-alcoholic liver diseases.

Thiamine state was investigated in patients with alcoholic liver disease, patients with various non-alcoholic liver diseases, and controls using a direct technique (thiochrome assay) to measure thiamine, thiamine monophospate, and the active coenzyme thiamine pyrophosphate in whole blood after isolating the fractions by ion exchange chromatography. Overall nutrition was similar in all groups as assessed by anthropometry, and no patient had clinical evidence of thiamine deficiency. There was no significant difference among the groups in mean concentration of any form of thiamine. The scatter was much greater in patients with alcoholic liver disease but only 8.7% had biochemical thiamine deficiency (defined as a blood concentration of the active coenzyme greater than 2 SD below the mean control value). An unexpected finding was of abnormally high total thiamine concentrations (greater than 2 SD above the mean control value) in 17.4% of patients with alcoholic liver disease, the highest concentrations being found in two patients with severe alcoholic hepatitis and cirrhosis. The ratio of phosphorylated to unphosphorylated thiamine was calculated as an index of phosphorylation and, although the mean did not differ significantly among the groups, the range was greatest in alcoholic liver disease. The lowest ratios occurred in the two patients with severe alcoholic hepatitis, but neither had evidence of thiamine pyrophosphate deficiency. Contrary to studies using indirect assay techniques, these results suggest that thiamine deficiency is unusual in well nourished patients with alcoholic liver disease. The new finding of unexpectedly high thiamine concentrations in some patients may be due to abnormalities of hepatic storage or release in liver disease, particularly in severe alcoholic hepatitis. There was no convincing evidence of impaired thiamine phosphorylation in any patients with liver disease. Conclusions from studies using indirect assays on the prevalence and mechanisms of thiamine deficiency in liver diseases may not be valid.