One part of the justification for lowering the fasting glucose diagnostic threshold is to identify the same percentage of the population as identified by the 2 h cut-point. This implies that the 2 h cut-point is the criterion that best identifies individuals at risk of developing complications as a result of hyperglycaemia. However, in many populations the tests identify different individuals33,34,35.
The 1985 WHO criteria selected the fasting and 2 h cut-offs on estimates of the thresholds for microvascular disease. After reviewing the statistical relation between the FPG distribution and 2 h PG distribution, it became evident that these criteria effectively defined diabetes by the 2 h PG alone because the fasting and 2 h cut-point values were not equivalent at those levels. Almost all individuals with FPG greater than or equal to 7.8 mmol/l have 2 h PG levels of 11.1 mmol/l or above when given an OGTT. On the other hand, only about one-quarter of those with 2 h PG exceeding 11.1 mmol/l (and without previously known diabetes) have FPG greater than 7.8 mmol/l36. Thus, the cut-point of FPG 7.8 mmol/l defined a greater degree of hyperglycaemia in comparison to the cut-point of 2 h PG 11.1 mmol/l. Understandingly, this discrepancy is undesirable, and therefore the ADA Expert Committee investigated cut-point values for both tests which reflect a similar degree of hyperglycaemia and risk of adverse outcomes.
The decision to change the diagnostic cut-point for the FPG to 7.0 mmol/l was based on the belief that the cut-points for the FPG and 2 h PG should diagnose similar conditions, given the equivalence of the FPG and the 2 h PG in their associations with vascular complications. Based on the populations considered, the summary estimate for the FPG cut-point was chosen at the upper end of the equivalence estimates as the FPG of 7.0 mmol/l is slightly higher than most of the investigated cut-points which would give the same prevalence of diabetes at the criterion of 2 h PG 11.1 mmol/l. Therefore the committee noted that slightly fewer people will be diagnosed with diabetes if the new FPG criterion is used alone than if either the FPG or the OGTT is used and interpreted by the previous WHO and NDDG criteria. This acknowledgement was confirmed by numerous studies showing that an equivalence of prevalence has not yet been achieved with the revised FPG cut-point.
Reviewing the history and science behind the diagnostic criteria shows the tension between the need for international consensus and evidence of heterogeneity in ideal diagnostic levels in different populations. There are large differences in the prevalence of the major forms of diabetes, their determinants and the associated complications among various ethnic groups worldwide. However, studies conducted in only a few single ethnic groups have been considered as evidence for the development of international diagnostic criteria. In particular, the work on Pima Indians has been used extensively to predict at what glycaemic level the risk of developing diabetic complications increases. The Pima Indians have the highest rate of diabetes in the world37, they develop diabetes at a younger age than other groups38 and their blood glucose measurements show a bimodal distribution which may not be apparent in all populations39. Generalising the data from this specific population to other ethnic groups may not be ideal.
IMPAIRED GLUCOSE TOLERANCE (IGT) AND IMPAIRED FASTING GLYCAEMIA (IFG) AS NEW RISK-FACTOR CATEGORIES FOR DIABETES
In the previous classification of diabetes, Impaired Glucose Tolerance was included as a separate class of diabetes. This meant that until 1980, people with a 2 h post-glucose load plasma glucose level between 7.8 mmo/l to 11 mmol/l were diagnosed with diabetes. It is now categorised together with IFG as a stage in the natural history of disordered carbohydrate metabolism with higher than normal fasting (IFG) or 2 h post-oral glucose load (IGT) glucose levels not reaching diabetic thresholds (see Table 2.2). As such, IGT and IFG were regarded in the absence of pregnancy (where they contribute to the class of gestational diabetes) not as clinical entities in their own right but as risk factors for future diabetes and cardiovascular disease40. This new categorisation rescued them from being assigned to a disease that could restrict life insurance and certain jobs and other social penalties in some countries. IGT or IFG can be observed as intermediate stages in any of the disease processes listed in Figure 2.1 under the assumption that persons within this stage are at higher risk than the general population for diabetes41. Individuals with IGT have a raised risk of macrovascular disease42,43 as IGT is associated with other known CVD risk factors including hypertension, dyslipidaemia and central obesity12. The diagnosis of these risk categories, therefore, may have important prognostic implications, particularly in otherwise healthy, ambulatory individuals.
Numerous population-based studies have calculated the risk of progression from IGT to diabetes. Two reviews44,45 including 15 different populations from Europe, the USA, India, Africa and the Pacific island of Nauru, reported an annual rate of development of diabetes varying from 2% to 14%. Not surprisingly, the risk was highest in those populations with a high background prevalence of diabetes. Data on IFG are still scarce, but in two studies there was an annual conversion rate to diabetes of 1% in middle-aged French civil servants46, and 6% in a high-prevalence population in Mauritius47.
Underlying this concept of progression from an intermediate risk state is the assumption that there are three separate glycaemic states: normoglycaemia, diabetic hyperglycaemia and in between the two a non-diabetic hyper-glycaemia, which confer different level of risk for developing diabetes-related complications. However, across the glucose spectrum in the few longitudinal studies of the natural history of increasing glucose intolerance and development of diabetes related complications, no clear threshold has been observed. Indeed, the relationships are curvilinear. In addition, the development of diabetes complications seems to be related both to the duration and degree of glycaemic exposure. Little attention has been paid to the relationship between these two variables, mainly because determination of exposure duration requires sophisticated metabolic surveillance systems. Hence, rarely are comparisons made between individuals with moderate levels of hyperglycaemia for extended periods and individuals with high levels over short time frames.
Was this article helpful?