These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

Introduction
Charles M. Clark, Jr., MD, Guest Editor

In this issue, we use a combination of invited articles and commentaries to present the challenges of non-insulin-dependent diabetes mellitus (NIDDM). We begin with the articles by Dr. Maureen Harris (p. 26) and Dr. David Marrero (p. 30). Dr. Harris summarizes data from her encyclopedic compendium of diabetes epidemiology, Diabetes in America, 2nd Edition, which she and her co-editors recently have revised and updated. Dr. Marrero’s contribution then examines the current adherence to practice guidelines in primary care practitioners.

Dr. Harris summarizes the increasing scope of NIDDM. Since 1958, the pre- valence of diabetes has increased fivefold to a currently estimated 8 million diagnosed (90–95% of whom have NIDDM) with another 8 million having undiagnosed NIDDM. Although in- creased detection may underlie some of this increase, three major factors sustain the notion that this increase is real and that it will continue. The first is the changing ethnic mix of the U.S. population brought about by increased immigration of populations genetically more susceptible to NIDDM. The second is the changing lifestyle of Americans. We have adopted a sedentary lifestyle, with decreased caloric expenditures without a concomitant decrease in caloric intake. This has led to an increase in obesity now reaching epidemic proportions. In the Great Lakes area of the Midwest, more than 30% of adults are obese, and the incidence of childhood obesity in the United States has doubled in the past decade. The third factor is the aging of the population.

Dr. Harris goes on to describe the troubling statistics regarding the morbidity and mortality from diabetes and its comorbidities. Each year, one in every 400 patients with diabetes develops end stage renal disease (ESRD), now accounting for one-third of such cases. One in 150 have amputations. Diabetes continues to be the leading cause of adult blindness. About 20% of patients with NIDDM have evidence of retinopathy at diagnosis, and 5–15% will develop proliferative retinopathy.

Although microvascular and neuropathic complications of NIDDM are responsible for the lion’s share of the morbidity of diabetes, it is macrovascular disease that is the killer of people with NIDDM. For coronary heart disease, the protective effect of female sex is lost. Diabetes is an equal opportunity killer. The incidence of myocardial infarctions, reinfarctions, and death from infarctions are all two- to fourfold higher in men and more substantially increased in women. Similar statements can be made for strokes and peripheral vascular disease.

Dr. Harris also discusses the differing effects of ethnic background on the prevalence of diabetic complications. Microvascular disease is much more common in minorities. ESRD is fourfold more common, and retinopathy is twice as common in African Americans as it is in whites. Among Latinos, retinopathy and proteinuria appear to be twice as common as in a comparable white population. In contrast, survival rates on dialysis are worse among whites, and, in contrast to microvascular disease, coronary heart disease is less common in minorities. The major risk factor for microvascular disease is hyperglycemia, and this persists when patients are stratified by age or insulin treatment. These relationships hold in whites and in Mexican Americans.

We then turn from a description of the scope of NIDDM in America to a description of how we are currently managing NIDDM in primary care settings. In this summary of recent research, Dr. Marrero reveals those areas where we need to improve the screening for and treatment of the microvascular complications of NIDDM. The recommended initial evaluations and screening for hypertension in such patients appear widely acted upon. In contrast, foot examinations in these potentially high-risk patients is less common, as is routine screening for retinopathy. Laboratory screening for lipids again appears widely acted upon, but using glycated hemoglobin for the assessment of glycemic control still has not found wide acceptance, nor has the use of microalbuminuria screening.

The data that Dr. Marrero summarizes define the challenges of widely implementing screening and treatment standards. It appears from these data that the national programs directed at increasing recognition of hypertension and hyperlipidemia have resulted in increased awareness and screening. In contrast, those screening and follow-up examinations specific to diabetes have a long way to go.

We then turn to examine the data supporting the idea that NIDDM presents a public health problem that can be addressed by treatments currently available. The first issue is the relationship between the microvascular complications of NIDDM and hyperglycemia. In his commentary (p. 35), Dr. Frederick Ferris reviews the most recent publication in the landmark Wisconsin Epidemiologic Study of Diabetic Retinopathy, began a decade ago by Drs. Ronald and Barbara Klein. In this study, patients were followed for a decade, and the incidence of diabetic background and proliferative retinopathy and proteinuria was measured in three groups of patients as a function of their baseline glycated hemoglobin.

These data add substantially to the proposition that the level of hyperglycemia, per se, predicts the development of microvascular disease in all patient groups. The groups studied were younger patients on insulin, primarily with insulin-dependent diabetes mellitus (IDDM); older patients on insulin, primarily with NIDDM; and older patients not on insulin, all with NIDDM. In all outcomes measured, there was a strong correlation between microvascular complications and glycated hemoglobin. Such data sustain the proposition that there is no such thing as "mild diabetes" and that in NIDDM, as in IDDM, microvascular complications are a function of chronic hyperglycemia.

Until recently, there were few studies relating the incidence of macrovascular disease to diabetes control. This has been partially corrected by the Scandinavian studies of Kuusisto and associates. The Finnish study of elderly patients is summarized in the commentary by Dr. William Duckworth and Dr. Ward Chambers (p. 37).In this study, HbA1c was a powerful predictor of both coronary events and coronary deaths. The data are striking, with the mean odds ratios in women with NIDDM versus nondiabetic individuals being 11.7 for death and 4.7 for nonfatal myocardial infarction. Similar data for men were 0.43 and 1.4, respectively. Only HbA1c and duration of diabetes were significant predictors; the increased risk was seen at an HbA1c above 7%. There was also observed a correlation between type of treatment and the incidence of events, with diet-treated patients having the lowest incidence, insulin-treated the highest, and oral-agent-treated being intermediate. Perhaps the most striking finding in these elderly patients was the lack of significance in predicting coronary events of the traditional risk factors of smoking, excess weight, high blood pressure, and high cholesterol. As Duckworth and Chambers point out, it is not that these risk factors should be ignored, but rather that increased emphasis of glycemic control is necessary.

We then move from epidemiologic, hypothesis-generating studies to clinical trials, which attempt to guide us further into what beneficial and/or harmful effects we can anticipate from interventions on those factors identified in the epidemiologic studies. We only briefly review the findings of the Diabetes Control and Complications Trial (DCCT), as they are widely known, if not acted upon. The question of the applicability of these findings to patients with NIDDM has sparked considerable debate in the diabetes community and among primary care physicians.

In his commentary (p. 42), Dr. Ronald Klein points out both the strengths and weaknesses of the recently published prospective trial of intensive management of a group of Japanese patients with NIDDM. In a design closely mimicking the DCCT, Ohkubo and associates found a reduction in retinopathy, nephropathy, and neuropathy similar to that found in IDDM patients in the DCCT. The study was small, and the patients were thin and had, presumably, insulin-sensitive NIDDM from the low doses of insulin required for metabolic control. Nevertheless, this study is one more piece of evidence supporting the case for lowering HbA1c in patients with NIDDM.

In a commentary on the analysis of the Multiple Risk Factor Intervention Trial (MRFIT) data (p. 44), Dr. Trevor Orchard notes the striking lack of data on the combined effects of traditional macrovascular risk factors and diabetes on the incidence of cardiovascular disease. The MRFIT data on more than 5,000 patients excluded from the intervention but being treated for diabetes directly address this issue. In this study, it is clear that diabetes, per se, dramatically increases the risk of the traditional risk factors.

The study is unique both for its size and for its 12-year follow-up. In this population, the absolute risk for cardiovascular death was increased threefold, and remained significant even after adjustment for other risk factors (age, race, income, high cholesterol, high systolic blood pressure, and cigarette smoking). In the presence of diabetes, these additional risk factors accounted for 94% of the cardiovascular deaths. Thus, the traditional cardiovascular risk factors in the nondiabetic population play at least the same, if not a greater role in cardiovascular deaths in diabetes.

Dr. Steven Edelman in his commentary (p. 47) summarizes and reviews the recently completed Veterans Affairs feasibility study. This trial was designed to determine the feasibility, or lack thereof, of a trial of intensive insulin therapy in older patients with NIDDM. The trial demonstrated that such a study was feasible. There was a significant and sustained difference in HbA1c between the experimental and standard groups, and the side effects of the

intensive insulin therapy were minimal. The two most common side effects of intensive insulin therapy in the DCCT were hypoglycemia and weight gain. Neither were significant problems in the VA study.

If this was a feasibility study, where are we with the main study? Apparently the investigators are still trying to convince the VA and the National Institutes of Health, among others, to support this important study, which was the number one priority of the recently published Regenstrief Conference.1

In her commentary (p. 50), Linda Haas reviews a recent clinical trial comparing the benefits of urine versus blood glucose monitoring in a group of older patients with poorly controlled NIDDM. In this study on patients treated by diet only and seen monthly, equal improvement was seen in both groups, an approximate 2% reduction in glycated hemoglobin. Given the difference in cost between these two groups, the authors conclude that urine monitoring is more cost-effective.

As Ms. Haas points out, this might not be true if one were to substitute telephone consultation and self adjustments for the monthly visits. In a similar study with patients with diabetes and other chronic diseases, Wasson was able to use telephone consultation to reduce outpatient visits in a VA patient population.2 Nevertheless, it seems that the combination of diet, urine monitoring, and frequent visits (whether by telephone or in the clinic) has more potential for reducing hyperglycemia than it is generally credited with.

In a second commentary on the effectiveness of intensive diet and exercise programs in NIDDM, Kellie Antinori-Lent reviews the recent report from the Pritiken group (p. 52). This study involved patients in a combination of extremely low-fat (10%), calorie-restricted diet and a standard exercise program.

Although it might seem unrealistic to generalize from unselected patients in a cloistered environment to the general population with NIDDM, the study is of value because it gives us an idea of the potential of intensive diet and exercise to reduce glucose, lipids, blood pressure, and dependence upon pharmacologic therapy. In all of these there were significant reductions, particularly in patients who were not taking insulin at the onset of the program. A 2% fall in HbA1c was again observed.

The challenge to devise programs to more effectively take advantage of the potential of diet and exercise is the subject of the contribution of Martha Funnell (p. 55).

References
¹Colwell JA, Clark CM: Forum Two: Unanswered research questions about metabolic control in non-insulin-dependent diabetes mellitus. Ann Intern Med 124 (Suppl 1, Part 2):178-79, 1996.

²Wasson GC, Whaley F, et al: Telephone care as a substitute for routine clinic follow-up. JAMA 267:1788-93, 1992.

Return To American Diabetes Association Home Page

Copyright © 1996 American Diabetes Association

Last updated: 5/15/96
For ADA Related Issues contact CustomerService@diabetes.org

For Technical Issues contact webmaster@diabetes.org