Designing Treatment Strategies for the Life Cycle of Diabetes

At an industry-sponsored symposium held in conjunction with the American Association of Clinical Endocrinologists’ 2003 Annual Meeting and Clinical Congress, three leaders in diabetes treatment and prevention utilized a case study to present the latest information on the relationship between hyperglycemia, insulin deficiency and insulin resistance, designing a balanced treatment regimen targeting fasting and postprandial glucose levels as well as addressing cardiovascular disease risk factors and identifying key studies linking early interventions and aggressive treatment with the prevention and or delay of comorbid conditions. A case study was used to illustrate the various concepts.

This program was supported by an unrestricted educational grant from Aventis Pharmaceuticals.

Gestational Diabetes Treatment Options and Perinatal Risk

“In May, 1983, JS was a 29-year-old Hispanic woman with six menstrual periods a year and increased facial hair who presented for infertility evaluation,” said Zachary Bloomgarden, MD, Associate Clinical Professor of Medicine, Mount Sinai School of Medicine in New York City. “Laboratory revealed normal fasting blood glucose (FBG) and total cholesterol with triglycerides of 150, HDL of 45 and LDL of 95.”

At the time, she was diagnosed with infertility associated with polycystic ovary syndrome (PCOS). Dr. Bloomgarden noted, that with current clinical understanding, she had evidence of insulin resistance.

She was treated for infertility with clomiphene and became pregnant. At 24 weeks, she had a 50-gram glucose challenge test with a one-hour plasma glucose of 135mg/dl. Two weeks later, she had a 100-gram oral glucose tolerance test with a glucose of 98 and 210, 200 and 170 mg/dl at zero, one, two, and three hours respectively. She was diagnosed with gestational diabetes mellitus (GDM).

“There is certainly an association between GDM and adverse perinatal outcomes,” said Dr. Bloomgarden. “A fasting glucose >105 mg/dl is associated with increased intrauterine fetal death after 36 weeks. GDM increases risk of fetal macrosomia, neonatal hypoglycemia, jaundice, and maternal hypertension. Increased fetal growth can complicate obstetric management.”

A number of studies show the risks of adverse pregnancy outcomes in patients with diabetes and the importance of early glycemic control to reduce malformation risk. One study by Aberg and colleagues found elevated adverse outcomes in 9.5% of 3864 infants of women with preexisting diabetes, 5.7% of 8688 infants in mothers with GDM and a similar number, 5.7% of 1.2 million Swedish births between 1987 and 1997 (Aberg A, et al. Early Hum Dev. 2001;61:85).

A study by Mills and colleagues looked at 347 diabetic women and controls who were enrolled within 21 days of conception and further compared them to a second group with diabetes who entered later. Major malformations were higher in both diabetic groups compared to controls. Those starting glucose control within the first 3 weeks had lower incidence than those who enrolled later (Mills JL, et al. NEJM. 1988;318:671).

A study published in 1995 enrolled 33 women who were randomized to preprandial glucose targets between 60 and 105 mg/dl. A similar number were randomized into a group that was to maintain 2-hour postprandial glucose < 140 mg/dl. Both groups had an initial hemoglobin A1C of 6.5% (de Veciana M, et al. NEJM. 1995;333;1237).

“Interestingly, using the preprandial goals only lowered the A1C to 8.1%, so they did rather poorly,” said Dr. Bloomgarden. “In contrast, the postprandial patients achieved an A1C of 6.5%.”

The question then became, were there any significant differences in actual outcomes between the two groups? The preprandial group had more caesarean sections, neonatal hypoglycemia, and their babies were larger for gestational age.

“The initial treatment for GDM is lifestyle modification, nutrition, exercise, and calorie restrictions,” said Dr. Bloomgarden. “Such an approach allows you to achieve acceptable goals in many cases.”

Insulin is the pharmacologic therapy most consistently shown to reduce fetal morbidity. A group of 274 women with GDM were equally randomized to receive insulin two (BID) or four (QID) times a day. One hundred and one women in the BID group maintained a mean blood glucose <105 mg/dl, had large-for-gestational-age babies, and other complications when compared with 126 women in the QID group. In those with pre-existing diabetes, the outcomes are clearer. Of 60 in each group, only half of those taking insulin twice a day achieved a mean blood sugar <105 mg/dl (Nachum Z, et al. BMJ 1999;319:1223).

“JS was treated with NPH insulin twice a day in increasing doses,” said Dr. Bloomgarden. “She gained 57 pounds during her pregnancy and gave birth to a 10 lbs, 4 oz. boy with only mild hypoglycemia in the post natal period.”

Designing an Oral Therapeutic Regimen to Treat Type-2 Diabetes

Three major studies have recently been published on ways to prevent diabetes. The Study to Prevent Non-Insulin-Dependent Mellitus examined the progression of impaired glucose tolerance (IGT) to frank diabetes. Only 32% of acarbose-treated patients developed diabetes compared to 42% in the placebo arm (Chiasson JL, et al. Lancet. 2002; 359:2072).

The Diabetes Prevention Program looked at IGT-to-diabetes progression. Lifestyle intervention resulted in a 58% decrease in incidence and a 31% fall using metformin when compared with placebo (Knowler WC, et al. NEJM. 2002;346:393).

Troglitazone in the Prevention of Diabetes study showed a drop in diabetes incidence when women on the medication were compared with placebo (5.4% vs. 12.1% respectively) (Buchanan TA, et al. Diabetes. 2002:51:2796).

“Our patient is now 39 years old,” said Philip Levy, MD, Clinical Professor of Medicine, University of Arizona College of Medicine in Phoenix. “She has three children and had GDM controlled by insulin in all three pregnancies. Her symptoms are weight gain, fatigue, tingling in her feet, and recurrent bladder infections.”

Laboratory studies indicate that she has features of insulin resistance, decreased insulin secretion, and glucose toxicity.

“We are beginning to perceive a shift from starting with only one agent to initially prescribing two,” said Dr. Levy. “You have your choice of insulin secretagogues, insulin sensitizers and alpha glucosidase inhibitors.”

Side effect profiles impact on treatment decisions. Of these, weight gain and hypoglycemia are the ones of most concern clinically.

Bugos and colleagues looked at mean change in weight from baseline during 12 months of treatment with glimepiride, glyburide and glipizide. There was less weight gain with glimepiride, and a trend toward weight neutrality (Bugos C, et al. Diabetes Res Clin Pract. 2000;50 (suppl 2):47).

“When we use combination therapies, we utilize different and complimentary mechanisms of action,” said Dr. Levy. “There’s better A1C efficacy and the side effects/toxic profiles are somewhat less when you use lower doses of each drug.”

There are some early studies indicating glimepiride improves both the first and second phases of insulin secretion in type 2 diabetes. Korytowski and colleagues reported on 11 obese patients who underwent euglycemic and hyperglycemic clamp studies pre- and four months post-glimepiride treatment.
Posttreament results were compared with 7 nondiabetic subjects matched for age and BMI. They found statistically significant changes in FBG, fasting plasma insulin, and both phases of insulin secretion (Korytkowski M, et al. Diabetes Care. 2002;25:1607).

“Fasting, postprandial first phase and second phase insulin secretion went up with glimepiride,” said Dr. Levy. “The ability to augment the first-phase of insulin secretion represents a key attribute for glimepiride.”

Building a Strategy to Prevent Cardiovascular Disease

JS is now 50 and comes to the doctor for a routine evaluation. She complains of increasing blood sugars after meals with fatigue and thirst. She experiences shortness of breath when walking up steps with her laundry, but denies chest pains. In addition to some neuropathy, she has reduced pedal pulses and background diabetic retinopathy. She is currently on insulin glargine, glimepiride and metformin along with an ACE-inhibitor for hypertension.

“When I saw this woman, my first concern was her risk for cardiovascular disease,” said Ramachandiran Cooppan, MB, ChB, Assistant Clinical Professor of Medicine, Harvard Medical School in Boston. “Much of this is related to the enormous risk because she is a woman with diabetes.”

The greatest cause of mortality in type 2 diabetes (T2DM) is atherosclerotic vascular disease. A study by Grundy and others showed that diabetic patients have a 2- to 4-fold greater risk of coronary heart disease (CHD) and have a poor prognosis for survival. They are at greater risk of mortality from stroke and have more risk of permanent brain damage with carotid emboli (Grundy SM, et al. Circulation. 1999;100:1134).

“The best data to bring this home came from Dr. Steve Haffner and the East-West Study,” said Dr. Cooppan. “A patient without diabetes or a history of cardiac disease has a 3.5% risk of having a myocardial infarction (MI) over 7 years compared with 45% in those with both diabetes and a prior MI. A person with diabetes but no cardiac history has the same risk for fatal events as the person who has had one heart attack” (Haffner S, et al. NEJM. 1998;339:229) (Figure 1).

Data from the Framingham Heart Study show diabetic females are at higher risk for cardiovascular disease and mortality. Although dated, the National Health and Nutrition Examination Survey (NHANES) results show CHD mortality was going down among all groups except diabetic females (Kannel WB, et al. Am Heart J. 1990;120:672; Gu K, et al. JAMA.1999;281:1291).

Screening for microalbuminuria is an important test that Dr. Cooppan thinks is not done often enough. Practice guidelines say screening in type 1 diabetes should begin with puberty and after five years of disease duration. In T2DM, it should start at diagnosis. If positive, they suggest a retest at 3 to 6 months and if two of three tests are positive, begin treatment. If the original is negative or two of three retests are, annual rescreening is needed.

“Not only is this a risk for renal disease, but also a marker for cardiovascular disease,” said Dr. Cooppan. “It is second only to smoking as a risk factor for ischemic heart disease.”

A number of studies have been done looking at the efficacy of angiotensin converting enzyme inhibitors (ACE-I) in microalbuminuria. Data from the MICRO-HOPE section of the Heart Outcomes Prevention Evaluation (HOPE) indicates that ACE inhibition, in this instance using ramipril, improves microalbuminuria (HOPE Study Investigators. Lancet. 2000;355:253).

“Looking at current guidelines, one of the important changes taking place in relation to diabetes is its elevation from risk factor to cardiovascular disease risk equivalent,” said Dr. Cooppan. “This means that a patient has more than a 20% chance of having a major event over the next 10 years. This is a totally different implication clinically.”

Although the guidelines say that the goal of treatment should be LDL cholesterol of <100, there is little or no guidance on how much less is acceptable.
Currently, the approach is to get LDL down to target using statins as the first choice and bile acid binding resin or fenofibrate if needed. Then, if the triglycerides are still elevated, look to raising HDL cholesterol through behavioral interventions, better glycemic control or, possibly, fibrates. If triglyceride lowering is still indicated following glycemic control, fibric acid, or statins are suggested (ADA. Diabetes Care. 2000;23 (suppl 1):S57).

The importance of optimal lipid control was reinforced by results of other recent trials. In the Heart Protection Study, there was reduction of 20% in major vascular events in nearly 4,000 patients with diabetes on statin therapy (HPS Study Group. Lancet. 2002;360:7).

Hypertension is another consideration in treating this group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) reemphasized that thiazide diuretics are important as first-line agents in treating hypertension.

They looked at a large number of patients, 36% were diabetic, with high blood pressure plus one other cardiovascular risk factor. They were treated with either chlorthalidone, amlodipine or lisinopril and followed for a mean of 4.9 years. There was no difference found in primary outcomes of fatal CHD or nonfatal MI (ALLHAT Collaborative Research Group. JAMA. 2002;288: 2981).

“I view this study as indicating that while thiazides are useful in treatment, you need more than one drug to treat hypertension effectively,” said Dr. Cooppan. “I think today there is a place for small doses of thiazides in the management of diabetics, especially to enhance the action of ACE-Is.”
The important issue then becomes, the degree of reduction in cardiac events that can be expected with aggressive treatment. The STENO-2 Study probably gives the best data currently available.

They found 50% reductions in both cardiovascular and microvascular disease when patients were treated to goal for glycemia, blood pressure, and cholesterol in addition to ACE-I and aspirin. There were also major increases in the percentage of people hitting their key goals in the intensive therapy group (Gaede P, et al. NEJM. 2003;348:383).

“I think STENO is probably the best data we have that a multifactor approach is important,” said Dr. Cooppan.

Many patients with T2DM need insulin. One of the reasons was seen in data from the United Kingdom Prospective Diabetes Study (UKPDS). They found a progressive loss of beta-cell function regardless of intervention in non-insulin treated patients. (UKPDS Group. Diabetes. 1995;44:1249).

“Epidemiological data indicate the issue of postprandial hyperglycemia is going to be important,” said Dr. Cooppan. “The Hanefeld data from Germany really brought that home showing that if your two-hour postprandial blood glucose was high, there were additional MIs and deaths when compared to those with elevated fasting sugars” (Hanefeld M, et al. Diabetologia. 1996;39:1577).

Ultimately, the best insulin therapy may be to mimic nature. Basal insulin is given to suppress insulin production between meals and overnight, maintain nearly constant levels and provide about half of daily need. This would be supplemented by a bolus of rapid-acting insulin around mealtimes to limit postprandial hyperglycemia and provide 10% to 20% of the daily insulin requirement at each meal.

The first consideration is to establish the goals of therapy. For most patients premeal BGs should be between 90-130 mg/dl in more than 50% of the readings. The 1-2 hour post-meal readings should be 120-170 mg/dl. Other goals might be set for specialized populations such as those with severe or asymptomatic hypoglycemia and in pregnancy.

A regimen should be established that includes basal and bolus doses, an insulin to carbohydrate ratio and sensitivity factors to correct for hyperglycemia. Nutritionists and Certified Diabetes Educators play an important role in developing a successful program. There is also a need to study the effects of exercise and assess pre- and post- exercise glucose levels.

“This is a very flexible approach, said Dr. Cooppan. “The patient and the physician have a lot of latitude to creatively design an insulin program.”