More Evidence

Irl B. Hirsch, MD, Editor

In 1989, the American Diabetes Association (ADA) published a consensus statement reviewing the evidence to date for the role of cardiovascular risk factors in prevention and treatment of macrovascular disease.¹ Because of the lack of prospective data at the time, it was concluded that "should glycemic control be demonstrated to prevent, retard, or ameliorate microvascular or macrovascular disease, then euglycemia should be the goal of therapy . . . and until such data are available, a prudent course of action is to achieve the best glucose control possible."¹

The passionate pleas of some physicians to treat glucose levels aggressively were largely ignored because, up to that time, the only completed clinical trial was the University Group Diabetes Program (UGDP), which failed to show any benefit of glucose control in those with type 2 diabetes.² Indeed, in that study, treatment with tolbutamide (Orinase, a first-generation sulfonylurea) and phenformin (a biguanide that is no longer on the market) resulted in worse cardiovascular outcomes.

Despite an almost two-decade interval, clinicians may now feel safe that aggressive diabetes treatment is indicated for both major forms of diabetes. The Diabetes Control and Complications Trial (DCCT), the Stockholm Intervention Diabetes Study (SIDS), and the Kumomoto study have all shown improved microvascular outcomes, with suggestions of macrovascular benefits, from meticulous glucose control.

However, these studies have been ignored by many, since they do not represent the more common population in the United States, that of obese patients with type 2 diabetes. Furthermore, the pilot study for the Veterans Affairs Cooperative Study on Glycemic Control failed to show a benefit of glycemic control on cardiovascular events.

The announcement of the results of the United Kingdom Prospective Diabetes Study (UKPDS) last September has helped to clarify some of this confusion. The UKPDS was a 20-year study that recruited 5,102 newly diagnosed patients with type 2 diabetes. The study design was extremely complex, and the amount of data generated is massive and can be overwhelming. It is also interesting that the interpretation of the data by different people has been so inconsistent.

Because of the importance and complexity of these data, we are publishing in this issue the ADA's position statement describing the implications of the results of this study (p. 5).

There are several ways to review the data. Those who are extremely interested will want to read the five articles published immediately after the data were released. Many will want to read the summary from the ADA position statement (p. 6) on implications.

However, there is a danger that the complexity of all of the data analysis, particularly the subgroup analysis, will confuse readers and detract from the major messages. These include:

• Glucose control in type 2 diabetes appears to have a similar beneficial effect to that found in type 1 diabetes in retarding the development of microvascular complications.
• The lower the HbA_1c, the lower the risk of microvascular complications.
• There is a strong suggestion that improved glucose control can decrease the risk of fatal and nonfatal myocardial infarction and sudden death.
• In obese patients, as first-line therapy, metformin (Glucophage) appears to result in better outcomes than sulfonylureas or insulin.
• In contrast to the results of the UGDP, there does not appear to be any harm associated with the use of the long-acting sulfonylureas chlorpropamide (Diabinese) or glyburide (Micronase and DiaBeta).
• Insulin therapy does not lead to a greater incidence of cardiovascular events.
• Lowering blood pressure with either the angiotensin-converting enzyme (ACE) inhibitor captopril (Capoten) or the -blocker atenolol (Tenormin) to a mean blood pressure of 142/82 significantly improves the risk of diabetes-related deaths, heart failure, stroke, and microvascular complications, particularly retinopathy.

Several issues about the data need to be addressed in more detail, however. For example, the substudy of metformin added to those individuals failing sulfonylureas is of concern because there was an increase in diabetes-related deaths. One must be careful not to overinterpret this finding, although any explanation at this point would be speculative. I have no plans to stop my patients on this combination, although further studies would be welcomed.

I also wonder about the subjects randomized to glipizide (Glucotrol). In 1987, there was a protocol change, and patients were randomized to this shorter-acting sulfonylurea. Is it at least possible that inhibition of the potassium channel as seen with tolbutamide will be seen with this shorter-acting agent?³

Data for glipizide will only be a small part of the information to follow. There was also a substudy of the addition of acarbose (Precose) to either diet therapy, monotherapy, or combination therapy. Although not yet published, these results were presented with the other data. At 3 years, only 39% of subjects remained on the drug (compared to 58% on placebo) because of side effects, and for those remaining on the drug, HbA_1c decreased 0.5%. Although many could accept an HbA_1c decrease of this magnitude, for the intention-to-treat analysis, the decrease was 0.2%. This is probably the more appropriate statistical test, since this is how the other major analyses were performed.

It will be interesting to see how long it takes to get these results to the primary care physicians, who see the vast majority of patients with diabetes in this country. Hopefully, the results will encourage physicians to be more aggressive in treating both hyperglycemia and hypertension. I continue to be amazed at how many physicians have not heard of the DCCT. Thus, our challenge will be to spread the word about UKPDS and the DCCT.

Is this enough? Is simply understanding the major messages of these landmark studies enough to change the way we manage diabetes? Although extremely important, I maintain that prescribing aspirin therapy, treating hypertension, and even the counseling and treatment of hypercholesterolemia is an order of magnitude simpler than treating diabetes. Our major barrier is that, for most of us, our systems for diabetes management are not ideal. We simply don't have enough time to do everything required to benefit our patients the most.

What we really need is a shift in our thinking about the treatment of diabetes. Nurses and diabetes educators predominantly managed diabetes treatment in the DCCT. Because of reimbursement issues, most physicians in this country can't afford this approach. Yet data continue to emerge that our colleagues from other diabetes health care disciplines can do a better job than doctors.⁴

Although the reasons for this are multifactorial, I believe the most important one is simply the issue of time. Giving a physician-supervised nurse, dietitian, or pharmacist primary responsibility for the treatment of diabetes and surveillance of its complications has always seemed to me the most efficient way to manage this chronic disease. This is the greatest potential advantage of managed care organizations,⁵ an institution otherwise best known as the scapegoat for everything wrong with American medicine.

I am optimistic that, in time, physicians will be able to embrace this different treatment philosophy. I'm not sure that we have any other choice.

References

¹American Diabetes Association: Consensus statement: Role of cardiovascular risk factors in prevention and treatment of macrovascular disease in diabetes. Diabetes Care 12:573-79, 1989.

²University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. Diabetes 19 (Suppl. 2):747-830, 1970.

³Engler RL, Yellon DM: Sulfonylurea K-ATP blockade in type II diabetes and preconditioning in cardiovascular disease: time for a reconsideration. Circulation 94:2297-2301, 1996.

⁴Aubert RE, Herman WH, Waters J, Moore W, Sutton D, Peterson BL, Bailey CM, Koplan JP: Nurse case management to improve glycemic control in diabetic patients in a health maintenance organization. Ann Intern Med 129:605-12, 1998.