A simple prediction rule and a neural network model to predict pancreatic beta-cell reserve in young adults with diabetes mellitus.

In the present study we developed and assessed the performance of a simple prediction rule and a neural network model to predict beta-cell reserve in young adults with diabetes. Eighty three young adults with diabetes were included in the study. All were less than 40 years old and without apparent secondary causes of diabetes. The subjects were randomly allocated to 2 groups; group 1 (n = 59) for developing a prediction rule and training a neural network, group 2 (n = 24) for validation purpose. The prediction rule was developed by using stepwise logistic regression. Using stepwise logistic regression and modification of the derived equation, the patient would be insulin deficient if 3(waist circumference in cm) + 4(age at diagnosis) < 340 in the absence of previous diabetic ketoacidosis (DKA) or < 400 in the presence of previous DKA. When tested in the validation set, the prediction rule had positive and negative predictive values of 86.7 per cent and 77.8 per cent respectively with 83.3 per cent accuracy while the ANN model had a positive predictive value of 88.2 per cent and a negative predictive value of 100 per cent with 91.7 per cent accuracy. When testing the performance of the prediction rule and the ANN model compared to the assessment of 23 internists in a subgroup of 9 diabetics whose age at onset was less than 30 years and without a history of DKA, the ANN had the highest ability to predict beta-cell reserve (accuracy = 88.9), followed by the prediction rule (accuracy = 77.8%) and assessments by internists (accuracy = 60.9%). We concluded that beta-cell reserve in young adults with diabetes mellitus could be predicted by a simple prediction rule or a neural network model. The prediction rule and the neural network model can be helpful clinically in patients with mixed clinical features of type 1 and type 2 diabetes.

Diabetes mellitus in young Thai adults.

The purposes of the present study were to 1) find the prevalence of various types of diabetes; 2) determine the prevalence of glutamate decarboxylase autoantibody (anti-GAD) and 3) identify clinical characteristics which may help in predicting insulin deficiency in young Thai adults with diabetes. Subjects consisted of 93 adults with diabetes mellitus aged 15-40 years. In each subject, basal and post glucagon C-peptide levels were determined by radioimmunoassay. Anti-GAD was measured by radioimmunoassay and mitochondrial 3243 tRNA(Leu(UUR)) gene mutation was detected by PCR-RFLP. Data were expressed as mean +/- SEM. The mean age of subjects was 31.0 +/- 0.7 years with age at diagnosis of 25.6 +/- 0.9 years. Thirty nine (41.9%) were males and 54 (58.1%) were females. Pancreatic calcification was found in 7 (7.5%) of the patients while 2 (2.2%) were identified as having Wolfram syndrome. Four (4.3%) had nonketotic diabetes with affected family members in multiple generations consistent with MODY. Mitochondrial 3234 tRNA(Leu(UUR)) gene mutation was detected in only one patient. After excluding 14 subjects with pancreatic calcification, Wolfram's syndrome, MODY or mitochondrial gene mutation, 45 (57.0%) were found to be insulin-deficient and 34 (43.0%) were insulin-sufficient based on post-glucagon C-peptide levels. Using stepwise logistic regression analysis, it was found that younger age at diagnosis (p<0.001), smaller waist circumference (p<0.01), previous history of DKA (p<0.01) was significantly associated with insulin deficiency. After excluding patients with DKA, younger age at diagnosis of diabetes (p<0.05) and lower BMI (p<0.01) were related to insulin deficiency. Concerning the role of autoimmunity, it was found that 13 (28.3%) of insulin-deficient subjects were positive for anti-GAD while 4 (11.8%) of those who were insulin-sufficient had positive results. Of the 54 patients currently on insulin, 42 (77.8%) are insulin deficient and 14 (25.9%) have positive anti-GAD. There were 10 (18.5%) who were both insulin sufficient and negative for anti-GAD suggesting that insulin therapy may not be required. We concluded that about half of young Thai adults with diabetes are not insulin-deficient and treatment with insulin may be unnecessary. The prevalence of glutamate decarboxylase antibody and mitochondrial 3234 tRNA(Leu(UUR)) gene mutation is low and as yet undefined factors are accountable for insulin deficiency in a significant number of patients.

Correlation of apolipoprotein E gene polymorphism to serum lipid concentrations in healthy Thais.

This study determined the genotype distribution of apolipoprotein E (apo E) gene and its relation to serum lipids in 217 healthy Thais consisting of 79 males and 138 females. Serum total cholesterol (TC), HDL-cholesterol (HDL-C) and triglyceride (TG) concentrations were determined by enzymatic-colorimetric methods, while serum LDL-cholesterol (LDL-C) levels were calculated using Friedewald formula. Apo E genotypes were determined by PCR-RFLP. Out of 217 subjects, apo E genotype frequencies were 5.5 per cent for E2/E2, 12.4 per cent for E2/E3, 81.1 per cent for E3/E3 and 0.9 per cent for E4/E4. In men, advancing age was associated with increased serum TC (r = 0.28, P < 0.05) and LDL-C (r = 0.27, P < 0.01). Subjects having the E2 allele had lower TC (r = -0.27, P < 0.05) and LDL-C. (r = -0.25, P < 0.05). Age and apo E genotypes were not associated with HDL-C and TG in men. In women, increasing age was related to higher serum TC (r = 0.45, P < 0.001), LDL-C (r = 0.44, P < 0.001), TG (r = 0.40, P < 0.001) and lower HDL-C (r = -0.36, P < 0.001). The presence of E2 allele was related to lower TC (r = -0.24, P < 0.001), LDL-C (r = -0.26, P < 0.001), TG (r = -0.15, P < 0.05) and higher HDL-C (r = 0.20, P < 0.01) independent of age and menopausal status. We concluded that the epsilon 4 allele of apo E gene is rare in Thais. The presence of the epsilon 2 allele is associated with a more favorable lipid profile and there is a sexual dimorphism concerning the effect of apo E genotype on serum HDL-C and TG.

Correlation between serum insulin and features of metabolic syndrome in Thais.

Several clinical and metabolic abnormalities, i.e. central obesity, hypertension, impaired glucose tolerance or diabetes and dyslipidemia often cluster together and are commonly found in patients with atherosclerotic cardiovascular disease. Hyperinsulinemia and insulin resistance are often evident in subjects with these metabolic abnormalities, so called insulin resistance or metabolic syndrome. In the present study, we looked into the correlations between serum insulin or index of insulin sensitivity and various clinical and metabolic abnormalities. Subjects consisted of 103 males and 118 females. Oral glucose tolerance test was performed on all subjects. Homeostasis model assessment of insulin sensitivity (HOMA-S) was used to determine insulin sensitivity. In males, HOMA-S was found to be significantly correlated with BMI, plasma glucose, insulin, triglycerides and waist circumference. Male subjects in the highest quartile of HOMA-S also had significantly higher systolic blood pressure compared to those in the lowest quartile. In females, HOMA-S was significantly correlated with BMI, blood pressure, plasma glucose, insulin, triglycerides, HDL-cholesterol, waist circumferences and waist-hip ratio. However, after adjustment for BMI, correlation between HOMA-S and blood pressure in women was no longer statistically significant. We, therefore, concluded that correlations between serum insulin or index of insulin sensitivity with certain metabolic abnormalities also existed in Thai subjects. Some of these correlations seem to be at least in part dependent on obesity.

Serum leptin concentrations in relation to body fat, gender, sex hormones and metabolic covariates in Thais.

In the present study, the relation of serum leptin to adiposity, gender and metabolic covariates in normal Thais was examined. Subjects consisted of 224 individuals aged between 20-79 years. Eighty two were men while 142 were women. Data were expressed as mean +/- SEM. Serum leptin was associated with total body fat assessed by dual-energy X-ray absorptiometry in both men (r = 0.80, P < 0.0001) and women (r = 0.73, P < 0.0001). Compared to women, serum leptin concentrations was lower in men (P < 0.0001). The difference still persisted after controlling the adiposity. Compared to premenopausal women, postmenopausal women had higher serum leptin independent of adiposity (P < 0.0001). In men, serum free testosterone was negatively associated with serum leptin (r = -0.36, P < 0.001) while there was no association between serum estradiol and leptin. The relation between serum FT and leptin in men no longer persisted after controlling for adiposity. Body fat was associated with fasting insulin levels in both men (r = 0.26, P < 0.05) and women (r = 0.18, P < 0.05). However, the association between fasting insulin levels and body fat in both men and women no longer existed after adjusting for leptin. We concluded that serum leptin concentrations are associated with total body adiposity and serum leptin may mediate the effect of body fat on insulin sensitivity. There appears to be a sexual dimorphism of serum leptin unrelated to sex hormone status and the amount of body fat.