DIABETICMedicine DOI: 10.1111/dme.12516

Research: Metabolism Serum bilirubin concentrations are positively associated with serum C-peptide levels in patients with Type 2 diabetes J. O. Chung, D. H. Cho, D. J. Chung and M. Y. Chung Division of Endocrinology and Metabolism, Department of Internal Medicine, Chonnam National University Medical School, Gwangju, Republic of Korea Accepted 3 June 2014

Abstract Aims To investigate the relationship between physiological serum total bilirubin concentrations and serum C-peptide levels in Korean patients with Type 2 diabetes. Methods A total of 588 patients with Type 2 diabetes were investigated in this cross-sectional study. Fasting C-peptide level, 2-h postprandial C-peptide level and DC-peptide (postprandial C-peptide minus fasting C-peptide) level were measured in all patients.

Fasting C-peptide level, postprandial C-peptide level and DC-peptide level tended to be higher in patients with higher bilirubin concentrations. Partial correlation analysis showed that serum bilirubin levels were significantly correlated with fasting C-peptide level (r = 0.159, P < 0.001), postprandial C-peptide level (r = 0.209, P < 0.001) and DC-peptide level (r = 0.186, P < 0.001) after adjustment for other covariates. In the multivariate model, the association between serum bilirubin concentrations and serum C-peptide levels remained significant after adjustment for confounding factors including age, gender, familial diabetes, hypertension, hyperlipidaemia, BMI, HbA1c, duration of diabetes and associated liver function tests (fasting C-peptide level: b = 0.083, P = 0.041; postprandial C-peptide level: b = 0.106, P = 0.005; DC-peptide level: b = 0.096, P = 0.015, respectively). Results

Conclusions Serum bilirubin concentrations within the physiological range were positively associated with serum C-peptide levels in patients with Type 2 diabetes.

Diabet. Med. 31, 1316–1322 (2014)

Introduction Type 2 diabetes is a progressive disorder. Pancreatic b-cell dysfunction may be implicated in the development, as well as the progression of Type 2 diabetes [1], but the natural history of b-cell function might be heterogeneous among people with Type 2 diabetes [1]. A previous study showed the heterogeneous nature of b-cell failure in patients with Type 2 diabetes who had different rates of an ongoing decline in b-cell function, suggesting that multifactorial determinants are involved in its progression [2]. Bilirubin, the natural end product of heme metabolism, has been recognized as a potent endogenous antioxidant [3]. Although severe hyperbilirubinaemia causes neurotoxicity in neonates, normal to modestly elevated bilirubin concentrations have been reported to be protective against oxidative Correspondence to: Min Young Chung. E-mail: [email protected]

1316

stress-related disorders [4]. Type 2 diabetes is related to oxidative stress [5]. Several clinical studies have shown the inverse correlation between serum bilirubin levels and the risk of diabetes [6,7]. In addition, experimental studies have reported that bilirubin inhibits oxidative stress-induced b-cell damage [8–10]; therefore, bilirubin might influence b-cell function in Type 2 diabetes; however, the relationship between bilirubin and b-cell function has not been fully understood in a clinical setting. Because C-peptide, which is formed during cleavage of insulin from proinsulin, is secreted in equal amounts to insulin and is not removed by the liver; the clearance rates of C-peptide are constant over the range of C-peptide levels. Peripheral C-peptide levels more accurately reflect portal insulin secretion than peripheral insulin levels [11]; therefore, in practice, serum C-peptide levels are widely used as an indirect measure of b-cell function and for selecting treatment regimens for hyperglycaemia [12–14]. In addition,

ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK

Research article

What’s new? • Although several clinical studies have reported an inverse correlation between serum bilirubin levels and the risk of diabetes, the relationship between serum bilirubin levels and serum C-peptide levels in patients with Type 2 diabetes is unknown. • Our results show that serum total bilirubin concentrations within the physiological range are positively associated with serum C-peptide levels in patients with Type 2 diabetes.

DIABETICMedicine

a semi-structured interview using a questionnaire. Alcohol consumption was defined as consumption of any type of alcoholic beverage at least once a week in the past 1 year. Hypertension was diagnosed if the patient had blood pressure > 140/90 mmHg or used anti-hypertensive drugs. Hyperlipidaemia was defined as serum concentrations of total cholesterol ≥ 6.5 mmol/l and/or a fasting triglyceride level of ≥ 2.3 mmol/l, or a history of taking lipid-lowering agents. The study was approved by the local ethics committee. All the participants gave informed consent.

Measurements

C-peptide has been found to be a bioactive peptide in several studies [15]. In the present study, we investigated the relationship between physiological serum total bilirubin concentrations and serum C-peptide levels in Korean patients with Type 2 diabetes.

Patients and methods Patients

This study was conducted from July 2012 to June 2013. A total of 588 patients with Type 2 diabetes were randomly selected using a random number table from patients who visited the diabetes clinic at our hospital for the whole study period. Type 2 diabetes mellitus was diagnosed either according to the Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus [16]: fasting plasma glucose ≥ 7.0 mmol/l, 2-h plasma glucose ≥ 11.1mmol/l during a 75-g oral glucose tolerance test, or where patients were on anti-diabetic medication. Patients with positive glutamic acid decarboxylase autoantibodies, a history of glucocorticoid use, renal impairment (plasma creatinine > 106 lmol/l), chronic liver disease, including hepatitis B or C virus infection, liver cirrhosis, pancreatitis, infection, heart failure or haematological disease were excluded from the study. Patients with a continuous requirement of insulin within 1 year of diagnosis were also excluded. Patients with serum bilirubin concentrations above the upper normal limit of 22.2 lmol/l and patients with abnormal liver function tests, defined as serum aspartate aminotransferase or alanine aminotransferase levels greater than two times the upper limit of the normal range (i.e. > 74 U/l) were excluded. History and physical examinations, including measurements of blood pressure, height and body weight, were conducted. BMI was calculated as the weight (kg) divided by the square of the height (m2). Waist circumference was measured to the nearest 0.1 cm at the narrowest point between the lower limit of the ribcage and the iliac crest. Clinical data on diabetes duration and alcohol consumption and family history data were acquired through

ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK

Blood samples were taken from the antecubital vein without insulin or oral hypoglycaemic agents having been taken between 8:00 h and 10:00 h after an overnight fast, but the anti-diabetic drugs were continued until 1 day before the blood test to prevent hyperglycaemia. After an overnight fast, blood samples were taken, and patients were allowed to eat a standardized meal, calculated on the basis of body weight (10 kcal/kg; carbohydrate 60%, protein 20%, and fat 20%) according to the Korean Diabetes Association recommendations [17]. Blood samples for the measurements of glucose and C-peptide levels were taken 2 h after a standardized meal. DC-peptide level was calculated as the postprandial serum C-peptide level minus the fasting C-peptide level. Plasma glucose was measured using the hexokinase method (Daiichi, Tokyo, Japan). Serum insulin levels were measured using an IRMA kit (Dainabot, Tokyo, Japan) with an intra- and inter-assay coefficient of variation of 1.5 and 6.1%, respectively. Indices of insulin resistance were calculated as follows [18]: homeostasis model assessment of insulin resistance (HOMA-IR) = [insulin (lU/ml) xglucose (mmol/l)]/22.5 and quantitative insulin sensitivity check index (QUICKI) = 1/[log insulin (lU/ml) + log glucose (mmol/l)x18]. Serum C-peptide levels (Biosource Europe SA, Nivelles, Belgium) were measured by radioimmunoassay with an intra- and inter-assay coefficient of variation of 3.3 and 7.1%, respectively. HbA1c levels were measured using ion exchange liquid chromatography with HLC-723-GHbV apparatus (Tosoh, Tokyo, Japan). Total cholesterol, HDL cholesterol, LDL cholesterol and triglycerides were also measured (AU5400; Olympus, Tokyo, Japan). Serum bilirubin levels were measured using an enzymatic method with bilirubin oxidase on an automatic analyzer (AU5407; Olympus). The normal range of serum total bilirubin levels in the present study was 3.8–22.2 lmol/l. Serum levels of aspartate aminotransferase and alanine aminotransferase were determined using an automated analyser (AU5407; Olympus). Free fatty acid levels were measured using a colorimetric method with a NEFA-HR kit (Wako, Kyoto, Japan). Plasma malondialdehyde level was measured using an enzyme-linked immunosorbent assay kit (OxiSelectTM malondialdehyde adduct; Cell Biolabs, San Diego, CA, USA).

1317

DIABETICMedicine

Statistical analyses

Data are expressed as the mean  SD values or median (interquartile ranges), unless otherwise stated. The chi-squared test was used for categorical variables, while one-way ANOVA or a Kruskal–Wallis test was used for continuous variables. Bilirubin levels were divided into tertiles. Normal distribution was tested using the Kolmogorov–Smirnov test. Variables with skewed distributions were log-transformed before analysis. Partial correlation analysis was used to analyse the relationship between serum C-peptide level and serum bilirubin concentration. Multiple linear regression models were used to determine the association between serum bilirubin concentration and serum C-peptide level with identified independent variables and factors that were previously reported to have independent associations. Highly intercorrelated (r > 0.2) confounding factors were not used in the same models. Because of the high correlation between aspartate aminotransferase and alanine aminotransferase, aspartate aminotransferase was excluded from the covariates. Fasting and postprandial glucose concentrations were highly correlated with HbA1c and fasting and postprandial glucose levels were therefore also excluded from the covariates. Statistical analysis was performed using SPSS software (SPSS version 17.0, SPSS, Inc., Chicago, IL, USA). A P value of < 0.05 was taken to indicate statistical significance.

Results The mean  SD age of the subjects was 56.4  13.6 years. The median (interquartile range) duration of diabetes was 5.0 (9.0) years and the mean  SD HbA1c level was 73  26 mmol/mol (8.8  2.4%). The mean  SD BMI was 24.2  4.1 kg/m2, and the median (interquartile range) fasting C-peptide concentration, postprandial C-peptide concentration and DC-peptide concentration were 0.67 (0.50) mmol/l, 1.40 (1.20) mmol/l, and 0.67 (0.89) mmol/l, respectively (Table 1). BMI and fasting C-peptide, postprandial C-peptide, DC-peptide, total cholesterol and LDL cholesterol levels tended to be higher in patients with higher total bilirubin levels (Table 2). Age, duration of diabetes, malondialdehyde level and HbA1c level tended to be lower in subjects with higher total bilirubin levels. The Pearson partial correlation coefficients between serum C-peptide level and the clinical variables including bilirubin, BMI, HbA1c, malondialdehyde, and duration of diabetes, adjusted for age, gender, familial diabetes, hypertension and hyperlipidaemia, are shown in Table 3. Serum bilirubin and BMI were positively correlated with fasting C-peptide level, postprandial C-peptide level, and DC-peptide level. In addition, HbA1c and malondialdehyde levels and duration of diabetes were negatively correlated with serum C-peptide levels. Serum bilirubin concentration was inversely related to malondialdehyde level (r = 0.315, P < 0.001).

1318

Bilirubin and C-peptide  J. O. Chung et al. Table 1 Characteristics of patients with Type 2 diabetes No. patients, n Mean  SD age, years Men, n (%) Hypertension, n (%) Hyperlipidaemia, n (%) Family history of diabetes, n (%) Alcohol consumption, n (%) Median (IQR) duration of diabetes, years Mean  SD BMI, kg/m2 Mean  SD waist circumference, cm Mean  SD systolic blood pressure, mmHg Mean  SD diastolic blood pressure, mmHg Mean  SD HbA1c, mmol/mol Mean  SD HbA1c,% Median (IQR) fasting glucose, mmol/l Median (IQR) postprandial glucose, mmol/l Median (IQR) fasting C-peptide, mmol/l Median (IQR) postprandial C-peptide, mmol/l Median (IQR) DC-peptide, mmol/l Mean  SD total cholesterol, mmol/l Median (IQR) triglycerides, mmol/l Median (IQR) HDL cholesterol, mmol/l Mean  SD LDL cholesterol, mmol/l Median (IQR) free fatty acid, nmol/l Median (IQR) malondialdehyde, pmol/mg Mean  SD total bilirubin, lmol/l Median (IQR) AST, U/l Median (IQR) ALT, U/l Treatment, n (%) No medication Oral hypoglycaemic agent Insulin Oral hypoglycaemic agent + insulin

588 56.4 337 261 238 196 113 5.0 24.2 85.5 123.4 76.8 73 8.8 7.5 11.0 0.67 1.40 0.67 4.6 1.4 1.1 2.7 593.0 28.3 12.1 23.0 22.0 57 365 93 73

 13.6 (57.3) (44.4) (40.5) (33.3) (19.2) (9.0)  4.1  10.7  15.5  10.7  26  2.4 (4.1) (6.5) (0.50) (1.20) (0.89)  1.2 (1.1) (0.5)  1.0 (534.5) (3.2)  4.1 (13.0) (21.0) (9.7) (62.1) (15.8) (12.4)

IQR, interquartile range; ALT, alanine aminotransferase; AST, aspartate aminotransferase.

Linear regression models were used to determine the association between serum bilirubin concentration and serum C-peptide level (Table 4). There was a significant association between serum bilirubin concentrations and serum C-peptide levels, with adjustments for variables including age, gender, BMI, alanine aminotransferase, free fatty acid and HbA1c levels, familial diabetes, duration of diabetes, hypertension and hyperlipidaemia. As b-cell secretory function might be related to insulin resistance, we further investigated the relationship between serum bilirubin levels and insulin resistance in patients without insulin use to avoid cross-reactivity of insulin immunoassays between exogeneous and endogeneous insulin. The median (interquartile range) was not significantly different for HOMA-IR [2.1 (2.4), 2.1 (1.9), 2.5 (2.3); P = 0.102] and QUICKI [0.34 (0.05), 0.34 (0.04), 0.33 (0.04); P = 0.102] according to the tertile of serum bilirubin concentrations. The significant associations were retained between serum bilirubin level and serum C-peptide concentration in subjects without insulin use (fasting C-peptide level: b = 0.084, P = 0.021; postprandial C-peptide level: b = 0.111, P = 0.012; DC-peptide level: b = 0.113, ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK

Research article

DIABETICMedicine

Table 2 Characteristics of patients with Type 2 diabetes stratified by serum total bilirubin tertile Serum total bilirubin tertile

No. patients, n Mean  SD age, years Men, n (%) Median (IQR) diabetes duration, years Mean  SD BMI, kg/m2 Mean  SD waist circumference, cm Hypertension, n (%) Hyperlipidaemia, n (%) Family history of diabetes, n (%) Alcohol consumption, n (%) Mean  SD systolic blood pressure, mmHg Mean  SD diastolic blood pressure, mmHg Mean  SD HbA1c, mmol/mol Mean  SD HbA1c,% Median (IQR) fasting glucose, mmol/l Median (IQR) postprandial glucose, mmol/l Median (IQR) fasting C-peptide, mmol/l Median (IQR) postprandial C-peptide, mmol/l Median (IQR) DC-peptide, mmol/l Mean  SD total cholesterol, mmol/l Median (IQR) triglycerides, mmol/l HDL cholesterol, mmol/l LDL cholesterol, mmol/l Median (IQR) free fatty acid, nmol/l Median (IQR) malondialdehyde, pmol/mg Mean  SD total bilirubin, lmol/l Median (IQR) AST, U/l Median (IQR) ALT, U/l Treatment, n (%) No medication Oral hypoglycaemic agent Insulin Oral hypoglycaemic agent + insulin

Tertile 1:

Serum bilirubin concentrations are positively associated with serum C-peptide levels in patients with Type 2 diabetes.

To investigate the relationship between physiological serum total bilirubin concentrations and serum C-peptide levels in Korean patients with Type 2 d...
76KB Sizes 0 Downloads 5 Views