A workshop was convened by the International Diabetes Federation to review the latest information relating to the risks associated with impaired glucose tolerance (IGT) and impaired fasting glycaemia (IFG) for future diabetes and cardiovascular disease (CVD). The workshop sought to address three questions: (i) are the current definitions of IGT and IFG appropriate; (ii) are IFG and IGT risk factors, risk markers or diseases; (iii) what interventions (if any) should be recommended for people with IFG and IGT? The determinants of elevated fasting glucose and 2-h plasma glucose in an oral glucose tolerance test (2-HPG) levels differ. Raised hepatic glucose output and a defect in early insulin secretion are characteristic of the former, and peripheral insulin resistance is most characteristic of the latter. Therefore, it is not surprising that the concordance between the categories of IFG and IGT is limited. In all prevalence studies to date only half or less of people with IFG have IGT, and even a lower proportion (20-30%) with IGT also have IFG. In the majority of populations studied, IGT is more prevalent than IFG, and there is a difference in phenotype and gender distribution between the two categories. IFG is substantially more common amongst men and IGT slightly more common amongst women. The prevalence of IFG tends to plateau in middle age whereas the prevalence of IGT rises into old age. Both IFG and IGT are associated with a substantially increased risk of developing diabetes, with the highest risk in people with combined IFG and IGT. Because IGT is commoner than IFG in most populations it is more sensitive (but slightly less specific) for identifying people who will develop diabetes. In most populations studied, 60% of people who develop diabetes have either IGT or IFG 5 years or so before, with the other 40% having normal glucose tolerance at that time. The limited published data suggest that both isolated IFG (I-IFG) and isolated IGT (I-IGT) are similarly associated with cardiovascular risk factors, such as hypertension and dyslipidaemia, with the highest risk in those with combined IFG and IGT. However, some data have suggested that I-IGT is more strongly associated with hypertension and dyslipidaemia (features of the metabolic syndrome) than I-IFG. In unadjusted analyses both IFG and IGT are associated with CVD and total mortality. In separate analyses for fasting and 2-HPG adjusted for other cardiovascular risk factors (from the DECODE study) there remains a continuous relationship between 2-HPG and mortality, but an independent relationship with fasting glucose is only found above 7.0 mmol/l. Glycated haemoglobin (HbA1c) levels are continuously and positively associated with CVD and total mortality independent of other CVD risk factors. Life style interventions, including weight loss and increased physical activity, are highly effective in preventing or delaying the onset of diabetes in people with IGT. Two randomized controlled trials of individuals with IGT found that life style intervention studies reduce the risk of progressing to diabetes by 58%. The oral hypoglycaemic drugs metformin and acarbose have also been shown to be effective, but less so than the life style measures. Similar data do not yet exist for the effectiveness of such interventions in people with I-IFG. Larger studies are required to evaluate the effects of interventions on cardiovascular outcomes in people with IGT. Cost effective strategies to identify people with IGT for intervention should be developed and evaluated. The use of simple risk scores to assess who should undergo an oral glucose tolerance test is one promising approach, although these will need to be population-specific. In conclusion, IGT and IFG differ in their prevalence, population distribution, phenotype, and risk of total mortality and CVD. The consensus of the workshop was: 1. The diagnostic thresholds for all categories of glucose intolerance should be revisited in the light of the latest evidence. There was no clear consensus (with current evidence) on whether IFG and IGT should be classified as diseases, but they clearly represent risk factors and risk markers for diabetes and CVD, respectively. 2. Both IGT and IFG are similarly associated with an increased risk of diabetes, but IGT is more strongly associated with CVD outcomes. 3. Risks are higher when IGT and IFG coexist. 4. Life style interventions are highly effective in delaying or preventing the onset of diabetes in people with IGT and may reduce CVD and total mortality, but the latter requires formal testing.