1Department of Biochemistry, LN Medical College, Bhopal, India
2Department of Pathology, LN Medical College, Bhopal, India
*Corresponding author: Shubho S Biswas, Professor, Department of Biochemistry, LN Medical College, Bhopal, India. E-mail: drssbis@gmail.com, Tel: 091-9302066808, Fax: 0755-4049610
Online published on 27 May, 2015.
The reduction of the fasting plasma glucose cutoff for diagnosis of diabetes to 126 mg/dl in 1997 improved the sensitivity of the test, but still many diabetic cases diagnosed by the standard 75 grams 2 hour post load plasma glucose were being missed. In the Indian population, it has been suggested that the cutoff should further be lowered to 118mg/dl. Studies to verify this have been few as the oral glucose tolerance test is cumbersome to perform. Laboratory data on fasting and post prandial plasma glucose is abundant. Hence our study tested different fasting plasma glucose cutoffs against the 2 hour post prandial plasma glucose instead of the standard 75 grams post load plasma glucose.
1497 test reports of fasting plasma glucose and postprandial plasma glucose were retrospectively analysed by ROC and cross tabulation.
The fasting plasma glucose cutoffs of 118mg/dl showed sensitivity 87.72%, specificity 75.79%, PPV 60.74%, NPV 93.53%, accuracy 79.36%, LR+ve 3.62, LR-ve 1.16 and false positive rate 24.27%. Percentage of subjects diagnosed diabetic by the FPG cutoffs 140mg/dl, 126mg/dl and 118mg/dl were 24.9%, 33.8% and 43.2% respectively.
Lowering the fasting plasma glucose to 118mg/dl improved the sensitivity of the test. However, the variability in the meals of the subjects resulted in higher false positive rate, lower LR +ve and lower specificity.
Fasting Plasma Glucose, Post Prandial Plasma Glucose, Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value
