[Cut-off points of fasting fingertip capillary blood glucose for detecting both undiagnosed diabetes and pre-diabetes].

Objective: To determine the efficient cut-off points of fasting fingertip blood glucose test for undiagnosed diabetes mellitus (DM), impaired glucose tolerance (IGT), and impaired fasting glucose (IFG) in community-based residents aged above 45 years old.

Methods: A cluster-randomized study was conducted from May 2008 to January 2009. A total of 3250 subjects aged above 45 years in two communities of Baoding city received questionnaire investigation and tested for fingertip blood glucose. Those subjects whose capillary blood glucose level ≥ 5.1 mmol/L were subjected to 75 g oral glucose tolerance test. Undiagnosed diabetes mellitus and pre-diabetes were identified by fasting plasma glucose and OGTT. In this study, the cut-off points of fasting capillary blood glucose for detecting undiagnosed diabetes and pre-diabetes were evaluated, using receiver operator characteristic curve (ROC).

Results: Of 1351 subjects that having had oral glucose tolerance test, 230 cases were diagnosed as diabetes mellitus (7.3%), 166 cases (5.2%) as IFG, and 204 (6.7%) as IGT under fasting capillary blood glucose as test variable and state variables according to the following criteria. (1) FPG≥7.0 mmol/L or/and 2hPG≥11.1 mmol/L (2) FPG<5.6 mmol/L (3) FPG<7.0 mmol/L and 7.8 mmol/L≤2hPG≤11.1 mmol/L, areas under three ROC curves were 0.905, 0.633 and 0.719, respectively. The cut-off values of screening for undiagnosed DM, IGT and IFG were 6.0 mmol/L, 5.7 mmol/L, and 5.7 mmol/L, respectively. When cut-off value of screening for undiagnosed DM was 6.0 mmol/L, the maximal sensitivity was 78.0% and specificity was 89.3%. But there were both lower sensitivity and specificity in screening for IFG and IGT according to the best predicting value (5.7 mmol/L) from the ROC curves (50.3% and 28.0% vs. 60.8% and 28.0%).

Conclusion: Fasting capillary blood glucose with the lower cut-point of 6.0 mmol/L in screening for undiagnosed diabetes mellitus alone, was relatively reliable, whereas for both IFG and IGT the fasting fingertip blood glucose tests were fallible. It was convenient and could be used in screening the DM at the community level.