Can HbA1c Replace OGTT for the Diagnosis of Diabetes Mellitus Among Chinese Patients With IFG?

Overview

This is a cross-sectional study that aims to assess the sensitivity and specificity of using HbA1c as a diagnostic test for detecting the presence of diabetes mellitus (DM) when compared to the use of oral glucose tolerance test (OGTT) among Hong Kong Chinese adult who have impaired fasting glucose (IFG). The investigators will recruit around 1000 non-diabetic adult participants who have impaired fasting glucose (i.e. fasting glucose level between 5.6 to 6.9mmol/L) AND without symptoms of hyperglycaemia to undergo both HbA1c test and oral glucose tolerance test after obtaining their informed consent. A diagnosis of DM is confirmed when both the fasting glucose level and 2-hour-post challenge plasma glucose level fall into diabetic range (i.e. fasting glucose level ≥7.0mmol/L and 2-hour-post challenge plasma glucose level ≥11.1mmol/L) after a standard 75g OGTT. Participants with only 1 plasma glucose value within the diabetic range will be invited to repeat an oral glucose tolerance test for confirmation of their diagnosis as recommended by American Diabetes Association and World Health Organization. Demographic data of the participants including age, gender, smoking status, medical history, diet and activity level will be collected. Lipid profile, blood pressure, waist circumference and body mass index will be checked and the total cardiovascular risk in 10 years will be calculated for each participant using the Joint British Societies 2005 and Framingham 2008 equations to evaluate the baseline cardiovascular risk of the participants. These data will be analyzed using SPSS. The primary outcomes are the sensitivity and specificity of HbA1c in detecting diabetes mellitus diagnosed by oral glucose tolerance test (the gold standard). A receiver operating characteristic (ROC) curve will be obtained by plotting sensitivity against (1-specificity) for each cutoff value for identification of an optimal cutoff point. Diagnostic accuracy will be assessed by the area under the curve (AUC). The differences in characteristics among participants diagnosed to have diabetes mellitus by oral glucose tolerance test or HbA1c will be compared using independent t-test or χ2 test for comparison between the 2 groups.

Objectives: 1. To calculate the sensitivity and specificity of HbA1c, using different cut-off values compared to the use of OGTT as the gold standard in the diagnosis of the presence of DM among Chinese adult patients with IFG 2. To identify the optimal cut-off value of HbA1c for diagnosing DM among the studied population 3. To compare the prevalence of DM among Hong Kong Chinese non-diabetic adult patients with IFG using OGTT and/or HbA1c criteria as recommended by the 2010 ADA for the diagnosis of DM 4. To compare the characteristics (i.e. demographics, presence of risk factors for DM and cardiovascular risk) of Hong Kong Chinese patients who are diagnosed to have DM by OGTT criterion alone, HbA1c criterion alone, or both criteria Hypotheses: 1. The prevalence of DM is higher among Chinese adult patients with IFG compared to the general population of Hong Kong. The figure was estimated to be as much as 25% from clinical observation. The overall prevalence of DM would be even higher if both OGTT and HbA1c criteria are used. 2. The use of HbA1c, with the cut-off threshold of ≥ 6.5% recommended by the 2010 ADA criteria for the diagnosis of DM, will pick up at least 70% of the diabetes cases identified by OGTT among this high risk group, which is superior to its use among the Chinese general population. 3. The sensitivity of using HbA1c ≥ 6.5% for the diagnosis of DM among this high risk group is superior to the use of FG ≥ 7.0mmol/L. 4. The optimal cut-off threshold of HbA1c in diagnosing DM for this high risk group may not be 6.5% but lower. 5. A subset of patients with impaired glucose regulation (IFG +/- IGT) will be identified as diabetic according to the HbA1c criterion alone. This group of patient may have a different cardiovascular risk profile compared to the group of patients who are diagnosed to have DM according to plasma glucose criteria alone. Methods: Research Design: This is a cross-sectional diagnostic test study to be carried out at various general out-patient clinics (GOPCs) across Hong Kong. Target population and Sampling method: Around 1000 Chinese adult subjects who are classified to have IFG will be recruited from various GOPCs from the Hong Kong West Cluster and New Territories East Cluster. Convenience sampling will be employed. Sample Size Calculation In this diagnostic test study, the sample size calculation was considered using the disease prevalence and test characteristics represented by the sensitivity and specificity of HbA1c to diagnose DM. The use of HbA1c is assumed to have an expected sensitivity and specificity of 80%, an approximate estimate according to previous studies, compared to OGTT for the diagnosis of DM among subjects with IFG and/or IGT. With an estimated prevalence of DM of 25% in the high risk Chinese population in Hong Kong based on epidemiological survey in the Chinese population with similar risk profiles, a sample of 204 subjects with DM and 612 subjects without DM will be recruited to obtain the lower limit of the 95% confidence interval greater than 0.7 with a probability of 0.9530. A sample of 816 will be required. To account for 20% attrition rate, a total sample of 1020 will be required, with 255 subjects with DM and 765 subjects without DM.