BACKGROUND: In patients with chronic liver disease (CLD), glycated haemoglobin (HbA(1C)) levels have been shown to be apparently lower than real values, whereas serum glycated albumin (GA) levels are apparently higher. The present study was aimed to examine whether both glycaemic indices are influenced by hepatic function. METHODS: Subjects consisted of 82 patients with CLD. Various indicators for hepatic function as well as HbA(1C) and GA were also measured. Estimated HbA(1C) values were calculated from the mean plasma glucose levels. Two hundred and two type 2 diabetic patients without CLD were studied as controls. RESULTS: Although GA was strongly correlated with HbA(1C) in patients with CLD as well as diabetic patients, GA levels in patients with CLD were relatively higher than those in diabetic patients. In patients with estimated HbA(1C) < or = 5.8%, GA levels significantly increased but HbA(1C) levels decreased as a function of decreasing hepaplastin test (HPT). The ratio of GA/HbA(1C) (G/H ratio) increased as a function of decreasing HPT. In patients with estimated HbA(1C) > 5.8%, in contrast, GA levels were independent of HPT levels. In the patients with CLD, GA and HbA(1C) were associated with mean plasma glucose levels and some indicators for hepatic function. The multivariate analysis revealed a significant association of G/H ratio with HPT, cholinesterase and direct bilirubin. The G/H ratio was not associated with the mean plasma glucose but with HPT and cholinesterase levels. CONCLUSIONS: The G/H ratio correlates with hepatic function but not with plasma glucose levels. Therefore, CLD should be suspected for diabetic patients with an elevated G/H ratio.
BACKGROUND: In patients with chronic liver disease (CLD), glycated haemoglobin (HbA(1C)) levels have been shown to be apparently lower than real values, whereas serum glycated albumin (GA) levels are apparently higher. The present study was aimed to examine whether both glycaemic indices are influenced by hepatic function. METHODS: Subjects consisted of 82 patients with CLD. Various indicators for hepatic function as well as HbA(1C) and GA were also measured. Estimated HbA(1C) values were calculated from the mean plasma glucose levels. Two hundred and two type 2 diabeticpatients without CLD were studied as controls. RESULTS: Although GA was strongly correlated with HbA(1C) in patients with CLD as well as diabeticpatients, GA levels in patients with CLD were relatively higher than those in diabeticpatients. In patients with estimated HbA(1C) < or = 5.8%, GA levels significantly increased but HbA(1C) levels decreased as a function of decreasing hepaplastin test (HPT). The ratio of GA/HbA(1C) (G/H ratio) increased as a function of decreasing HPT. In patients with estimated HbA(1C) > 5.8%, in contrast, GA levels were independent of HPT levels. In the patients with CLD, GA and HbA(1C) were associated with mean plasma glucose levels and some indicators for hepatic function. The multivariate analysis revealed a significant association of G/H ratio with HPT, cholinesterase and direct bilirubin. The G/H ratio was not associated with the mean plasma glucose but with HPT and cholinesterase levels. CONCLUSIONS: The G/H ratio correlates with hepatic function but not with plasma glucose levels. Therefore, CLD should be suspected for diabeticpatients with an elevated G/H ratio.