Diabetes mellitus is characterised by hyperglycaemia, which is associated with long-term damage, dysfunction and failure of various organs. Several studies30,31 have confirmed relationships between hyperglycaemia and the risk of developing such micro- and macrovascular complications as retinopathy, neuropathy, nephropathy and cardiovascular disease. However, many have compared the rates of each condition in subjects already classified according to the diagnostic criteria as having diabetes or not. Few studies consider whether the current diagnostic glucose levels represent the best level for predicting an increased risk of such complications, and no formal statistical threshold for any complication has been consistently demonstrated.
The relationships of FPG and 2 h PG with the development of retinopathy were evaluated in a study undertaken in the Pima Indian population over a wide range of plasma glucose cutpoints23. Both variables were similarly associated with retinopathy, indicating that by this criterion, each could work equally well for diagnosing diabetes. The authors concluded that both measures were equivalent in terms of the properties previously used to justify diagnostic criteria. These findings were confirmed in a similar study in Egypt, in which the FPG and 2 h PG were each strongly and equally well associated with retinopathy27. For both the FPG and the 2 h PG, the prevalence of retinopathy was markedly higher above the point of intersection of the two components of the bimodal frequency distribution (7.2 mmol/l and 2 h PG 11.5 mmol/l).
Using Receiver Operating Characteristic (ROC) curves, it is possible to determine the value of a diagnostic test which provides maximum sensitivity and specificity for predicting the occurrence of a given complication associated with diabetes32. A ROC curve is a graphical representation of the relationship between sensitivity and specificity for any diagnostic test. It is constructed by plotting the true positive rate (sensitivity of the test) against the false positive rate (1 - specificity) for a series of possible thresholds. Such values have been calculated for the Pima Indian population in relation to retinopathy, and the optimum cut-off values among those over 24 years of age were 7.2 mmol/l for a fasting plasma glucose test and 13.0 mmol/l for a 2 h postload glucose test. These values differ from the cut-off values suggested as diagnostic of diabetes from the bimodal distribution of blood glucose results, particularly for the fasting value. The use of ROC analyses for predicting nephropathy among the Pima population indicated that the various measures of hyperglycaemia were poor at predicting this complication. No study has reported ROC curves for glucose values predicting cardiovascular disease.
Was this article helpful?