It's clear that we need to improve glycemic control and there are clear indications that we need a better approach. This has been made all the more important by the recently disclosed follow-up data from the Diabetes Control and Complications Trial (DCCT) and the Epidemiology of Diabetes Interventions and Complications (EDIC) study, in which it has been demonstrated that the attainment of glycemic control early on in persons with type 1 diabetes who are on frequent injections of insulin or continuing insulin infusions with frequent glucose monitoring confers a metabolic memory that, all other risk factors being equal compared with the conventionally treated group, results in significantly fewer macrovascular events years later. This is clear evidence that there is urgency for achieving glycemic targets and that we need a better approach.

When we think about the prevalence data, the prevalence of diabetes is, in fact, rather substantial. When you see the projections that exist from 2000 to 2030 across age strata and across developing as well as developed countries, it's very clear that even in an individual's prime years, what we're expecting is an enormous increase in the prevalence of diabetes going forward. So this disease promises to be with us.

Why do we worry about this disease? It's not simply a matter of the explosion in numbers; it's because of the impact of diabetes. The relationship of the hyperglycemia of diabetes to outcomes like mortality have been demonstrated in studies such as the European Prospective Investigation into Cancer in Norfolk, United Kingdom (EPIC-Norfolk) study, which was not a diabetes study, but a cancer epidemiology study; when you looked at causes of mortality, there was this strong relationship between glucose levels, as reflected in hemoglobin A1C (A1C), and age-adjusted mortality. There was no threshold; there was this continuum of risk. Every incremental increase over normal resulted in increased levels of mortality, and of course the highest being from known diabetes. Each 1% increase in A1C represented a 28% increase in the risk of death.

Those complications that are most well known for their association with hyperglycemia are the microvascular complications, chief among them being retinopathy, which constitutes the leading cause of blindness; nephropathy, the leading cause of end-stage renal disease from diabetes; and neuropathy, an early and significant complication.

In addition to the microvascular complications that are very strongly linked to glucose, other complications are related to acute abnormalities of the metabolic milieu in people with diabetes; you have ketosis, hyperosmolar syndrome, and hypoglycemia. But increasingly, we are recognizing that there are also connections between macrovascular complications and glycemic exposure; that again has been reconfirmed by the follow-up data from EDIC and the DCCT.

When we look at the relationship of risk factors for microvascular complications in diabetes, we have 2 major categories: controllable and uncontrollable. I want to point out a caveat -- among those things that we call uncontrollable, I've highlighted disease duration because it's very clear now that in people who are at high risk, particularly at high risk for type 2 disease, we actually can control the duration of disease because we can prevent the onset and thus delay the development of diabetes in those people. And in that sense, we actually now have some leverage over disease duration. But in every other sense, these uncontrollable things we really can't do very much about. The controllable elements or risk factors are the things that we really pay most attention to because our interventions are targeted towards reducing their contributions to the complications. And we begin with hyperglycemia.

This is the relationship between hyperglycemia, as reflected by A1C, and the microvascular complication of retinopathy. It's clear that the relationship is not a linear one, that there is a much steeper rate of progression of this microvascular complication at the higher levels of A1C, and those relationships are similar whether you are looking at type 1 or type 2 patients. So these are very strong, well established associations.

In the DCCT, we've seen evidence that intensive insulin therapy decreases glycemic exposure, as you see from the degree of separation that was achieved and maintained between the conventionally treated cohort and those in the intensively treated group. And the result of that separation and the reduction of that glycemic exposure is reflected in the impact on the complications. In this case, in the primary prevention cohort, retinopathy rates were significantly reduced and the rates of progression of retinopathy, likewise, were diminished.

This is also true in the Kumamoto study, but now with type 2 patients. In both the DCCT and in the Kumamoto study, when you achieve this kind of separation in terms of glucose exposure, as a result of frequent injections of insulin coupled with frequent glucose monitoring, you achieve the same outcome. In this instance, with these type 2 patients, we see that there was this significant reduction in both the rates of retinopathy and in the rates of progression of this microvascular complication.

What about macrovascular complications -- the leading cause of death in terms of diabetes complications? This is what we fear in the global explosion of diabetes, these kinds of outcomes: ischemic heart disease, cerebrovascular disease, and peripheral vascular disease, accelerated atherosclerosis.

There is clearly a relationship between glycemic exposure, again reflected by A1C, and macrovascular complications; in this case, myocardial infarction (MI). In this observational analysis from United Kingdom Prospective Diabetes Study (UKPDS) data, what you see is that there is this really strong relationship between increasing levels of A1C and the risk or the hazard ratio for MI. In fact, each 1% increase in A1C increases the risk of an MI in this analysis by 14%, and this regression line really shows that the association between A1C and this risk is highly statistically significant.

This is important because we know from studies like the MI Onset Study and the East/West Study from Haffner and colleagues that diabetes itself confers a risk for MIs that is similar to the risk that is seen in people who don't have diabetes but who have had a known prior event. The same observation has been made now in multiple studies, so there's no question that the risk of a macrovascular outcome like MI is a part of the diabetes phenotype.

The important additional observation, however, is that the interventions that are used are like those seen in the UKPDS, where intensive therapy, including glucose lowering, resulted in a significant reduction of all microvascular events, all of the endpoints, and almost achieved a statistically significant reduction in a macrovascular endpoint. This is now what has been seen in the studies, which show a relationship between glycemic exposure and macrovascular events in addition to microvascular events.

Given what we know about these associations between glucose exposure and vascular outcomes, micro or macro, it's distressing that we see some trends that are in the wrong direction. Glycemic control, that is, achieving the conservative goal of an A1C of less than 7%, actually decreased from 44.5% to 35.8% as we went from the previous to the most recent iteration of the National Health and Nutrition Examination Survey (NHANES) data set. More than 37% of people under treatment with diabetes have A1C values of greater than 8%. The mean reported A1Cs increased from 7.7% to 7.9%, perhaps accounted for by the fact that our fatness increased, but I don't think it's just that. We have a lot of tools available to us, our armamentarium of medications, our armamentarium of devices and techniques for monitoring blood glucose levels have all improved tremendously. We simply need a new approach.

These are the data: just over one third of people are achieving this more conservative goal of less than 7%; another third are at greater than 8%; clearly two thirds are at a level greater than 7%. If we think about the outcomes that we worry about most, namely ischemic heart disease and coronary heart disease mortality, and the risk factors that synergize A1C, blood pressure, and high levels of cholesterol, particularly low-density lipoprotein (LDL), those persons achieving these risk factor targets required to reduce the risk of cardiovascular disease are only 7%. This is really quite pitiful.

We need a better approach, especially in light of what's happening globally -- we are in the throes of an explosive epidemic and the projections over the next 2 decades are sobering. In fact, the estimate is that worldwide we're going to see over a 70% increase in the prevalence of this disease in the next 2 decades, including some areas where there is an expectation of a robust 3-figure increase. These kinds of increases are expected to occur in regions where we have some of the densest populations in the world.

So the call to action is an urgent and compelling one. Diabetes affects a large proportion of the population; worldwide, the prevalence of diabetes is increasing explosively. It's associated with microvascular and macrovascular complications; it has a real sting to it as well as economic consequences. Strategies that decrease glycemic exposure have been shown to decrease complications, whether in type 1 or type 2, and now with the addition of metabolic memory to our evidence base, it becomes all the more compelling that we do a better job of achieving glucose targets. So regular assessment of blood glucose levels is an integral component for optimizing glycemic control, and you will see more of that evidence shortly. Success in the attainment of intensive glycemic control depends heavily on timely and precise blood glucose monitoring. These presentations hope to make sure you understand the nature of the evidence for this and how we can translate that into clinical practices.