OBJECTIVE: Bilirubin is a potent antioxidant, and serum total bilirubin (STB) concentrations correlate negatively with cardiovascular risk. In adult diabetic patients and in healthy adults, a negative correlation between STB and glycated hemoglobin (HbA1c) has been reported. We investigated whether there is such an association in children and adolescents with type 1 diabetes mellitus. METHODS: The study group included 224 patients with type 1 diabetes duration of more than 12 months. Patients with suspected or confirmed hemolytic anemia or liver dysfunction were excluded. RESULTS: A statistically significant negative correlation was found between STB and HbA1c (R = -0.15; p = 0.024), which retained its significance in multivariate analysis (β = -0.18, p = 0.005). Patients' age and daily insulin dose were positively correlated with HbA1c levels, whereas other variables included in the multivariate analysis [sex, diabetes duration, insulin regimen, C-peptide, hemoglobin, mean corpuscular hemoglobin concentration (MCHC), alanine transaminase (ALT), and aspartate transaminase (AST)] did not correlate with HbA1c. The mean HbA1c level in patients with STB >1.2 mg/dL (>21 µmol/L; the threshold for clinical diagnosis of Gilbert's syndrome) was lower than in patients with STB ≤1.2 mg/dL (≤21 µmol/L), and the mean difference was 0.63% (6.9 mmol/mol; 95% CI: 0.11-1.16%). CONCLUSIONS: These results show that in young patients with type 1 diabetes, STB concentration is an independent factor inversely associated with HbA1c level. Further studies should investigate the background and long-term effects of this association.
OBJECTIVE:Bilirubin is a potent antioxidant, and serum total bilirubin (STB) concentrations correlate negatively with cardiovascular risk. In adult diabeticpatients and in healthy adults, a negative correlation between STB and glycated hemoglobin (HbA1c) has been reported. We investigated whether there is such an association in children and adolescents with type 1 diabetes mellitus. METHODS: The study group included 224 patients with type 1 diabetes duration of more than 12 months. Patients with suspected or confirmed hemolytic anemia or liver dysfunction were excluded. RESULTS: A statistically significant negative correlation was found between STB and HbA1c (R = -0.15; p = 0.024), which retained its significance in multivariate analysis (β = -0.18, p = 0.005). Patients' age and daily insulin dose were positively correlated with HbA1c levels, whereas other variables included in the multivariate analysis [sex, diabetes duration, insulin regimen, C-peptide, hemoglobin, mean corpuscular hemoglobin concentration (MCHC), alanine transaminase (ALT), and aspartate transaminase (AST)] did not correlate with HbA1c. The mean HbA1c level in patients with STB >1.2 mg/dL (>21 µmol/L; the threshold for clinical diagnosis of Gilbert's syndrome) was lower than in patients with STB ≤1.2 mg/dL (≤21 µmol/L), and the mean difference was 0.63% (6.9 mmol/mol; 95% CI: 0.11-1.16%). CONCLUSIONS: These results show that in young patients with type 1 diabetes, STB concentration is an independent factor inversely associated with HbA1c level. Further studies should investigate the background and long-term effects of this association.