Endocrine Abstracts

Aim: Diabetes mellitus is defined by the World Health Organization (WHO) as fasting plasma glucose (FPG) ≥7.0 mmol/l (mM) or 2-hour post-load glucose (2HPG) ≥11.1 mM in the 75-gram oral glucose tolerance test (OGTT). However, reported FPG cut-off levels that correspond to this 2HPG level are below 7.0 mM in Asian studies. Our study thus aims to find the optimal FPG cut-off that corresponds to 2HPG of 11.1 mM in an Asian population in Singapore.

Methods: Seven hundred and eighty-seven subjects were screened for diabetes with a 75-gram OGTT at the outpatient clinics of a Singapore hospital from 2001 to 2007. Plasma glucose levels were measured using a Beckmann-Coulter analyser. Regression models and receiver operating characteristic (ROC) curves in SPSS 16.0 were used to define the optimal FPG cut-off. Stratified analyses were performed for age and sex.

Results: The mean age of our patients (393 males, 49.9%) was 50.0±15.4 years (range 14–93). Their average FPG was 6.5±2.8 mM (range 3.0–24.5), and mean 2HPG was 11.0±5.6 mM (range 3.2–36.4). Exponential regression models were the best fit (higher R² value than linear, quadratic and logarithmic) for the whole population. The FPG level corresponding to 2HPG 11.1 mM using the exponential model was 6.1 mM. Patients aged 50 years and above had lower FPG cut-off (6.1 mM) corresponding to 2HPG 11.1 mM than younger patients (6.2 mM). The FPG cut-off derived from ROC analysis of the whole population was 6.0 mM (sensitivity 81.5%, specificity 82.0%, area under curve 0.90). Both age groups had FPG cut-off of 6.0 mM in the ROC analyses. Men and women had similar FPG cut-offs in both quadratic models and ROC analyses.

Conclusion: The FPG cut-off for diagnosis of diabetes in our Asian population is lower than the current WHO criteria.