Abstract Objective To analyze the relationship between fasting and 2 hour plasma glucose in oral glucose tolerance test (OGTT) and find out the cut point of fasting plasma glucose (FPG) level for screening and diag...Abstract Objective To analyze the relationship between fasting and 2 hour plasma glucose in oral glucose tolerance test (OGTT) and find out the cut point of fasting plasma glucose (FPG) level for screening and diagnosing of diabetes mellitus. Methods The OGTT was done in 815 subjects without history of diabetes mellitus. With 2h plasma glucose (PG) as a golden standard, sensitivity, specificity and Kappa value were analyzed in different cut off pints, and a receiver operating characteristic (ROC) curve was also made. Results There was good relation between FPG and 2h PG levels. Using the 2h PG≥11 mmol/L as a golden standard, the sum of false positive and false negative rates was lower when the cut points were between 6.44 mmol/L (116 mg/dl) and 7.06 mmol/L (127 mg/dl). The FPG≥7.06 mmol/L with a higher specificity was good for confirming a diagnosis, while the FPG≥6.44 mmol/L with a higher specificity was good for screening. The sensitivity and specificity of each cut off point were not affected by sex, age, body mass index and blood pressure status. Conclusion There is good relation between FPG and 2h PG. According to the ROC curve, the FPG≥7.06 mmol/L with a higher specificity was a good criterion for diagnosing diabetes mellitus, while the FPG≥6.44 mmol/L with a higher sensitivity was a good criterion for screening.展开更多
文摘Abstract Objective To analyze the relationship between fasting and 2 hour plasma glucose in oral glucose tolerance test (OGTT) and find out the cut point of fasting plasma glucose (FPG) level for screening and diagnosing of diabetes mellitus. Methods The OGTT was done in 815 subjects without history of diabetes mellitus. With 2h plasma glucose (PG) as a golden standard, sensitivity, specificity and Kappa value were analyzed in different cut off pints, and a receiver operating characteristic (ROC) curve was also made. Results There was good relation between FPG and 2h PG levels. Using the 2h PG≥11 mmol/L as a golden standard, the sum of false positive and false negative rates was lower when the cut points were between 6.44 mmol/L (116 mg/dl) and 7.06 mmol/L (127 mg/dl). The FPG≥7.06 mmol/L with a higher specificity was good for confirming a diagnosis, while the FPG≥6.44 mmol/L with a higher specificity was good for screening. The sensitivity and specificity of each cut off point were not affected by sex, age, body mass index and blood pressure status. Conclusion There is good relation between FPG and 2h PG. According to the ROC curve, the FPG≥7.06 mmol/L with a higher specificity was a good criterion for diagnosing diabetes mellitus, while the FPG≥6.44 mmol/L with a higher sensitivity was a good criterion for screening.
