Education of Patients

Arguably, over the last 10 years, nothing has changed more fundamentally in diabetes care than the emphasis on lifestyle intervention. For decades, physicians and patients paid lip service to the notion that lifestyle intervention is important. Now we have significant clinical trial evidence that each component of lifestyle intervention, when appropriately administered, can contribute to improved outcomes.^20,21 Furthermore, since the Balanced Budget Act of 1997 and the passage of complementary legislation by most state governments, lifestyle intervention has become a covered benefit for most patients with diabetes in the United States.

Diabetes is a lifelong disease, and health care providers have almost no control over the extent to which patients adhere to the day-to-day treatment regimen. The appropriate role of the health care provider is to serve as a coach to the patient, who has primary responsibility for the delivery of daily care. Thus, it is essential that health care professionals understand the context in which patients are taking care of their disease. Using a prescriptive approach in which patients are told what to do can work on occasion but fails more often than not because of unrecognized barriers to the execution of a particular plan.

As defined by the ADA,²² diabetes self-management education is the process of providing the person with diabetes the knowledge and skills needed to perform self-care, manage crises, and make lifestyle changes. As a result of this process, the patient must become a knowledgeable and active participant in the management of his or her disease. To achieve this rather daunting goal, patients and providers work together in a long-term, ongoing process. Comprehensive diabetes education should be individualized, with emphasis on the issues highlighted in Table 22-2. There are many more specialized topics relevant to almost all patients, such as how to adjust therapy when eating out or during travel, as well as how to access available local health care resources and negotiate the complexity of health care financing in the United States. Although only limited studies are published to date, as a body of work, they do provide support for the concept that diabetes education can be cost effective and can improve outcomes.^21–23

A team of providers is generally required to optimally implement the process of diabetes self-management education, because the amount of information that needs to be exchanged is large and the range of expertise required is broad. It is generally impossible to cover the recommended content fully in the context of several or even many brief encounters with a physician in an office setting. Potential providers in a team care approach could include nurses, dietitians, exercise specialists, behavioral therapists, pharmacists, and other medical specialists including diabetologists or endocrinologists, podiatrists, medical subspecialists, obstetrician-gynecologists, psychiatrists, and surgeons. In the diabetes self-care process, the potential role of the community where the patient lives and works is enormous. At a minimum, this community includes family, friends, employers, health care systems, and health care insurers. Each member of the team has a role to play in the process, and it is useful to review these roles frequently (Table 22-3). The primary role of the providers in this process is to provide guidance in goal setting to manage the risk of complications, suggest strategies to achieve goals and techniques to overcome barriers, provide training in skills, and screen for complications. For this process to be a success, the patient must commit to the principles of self-care, participate fully in the development of a treatment plan, make ongoing decisions regarding self-care from day to day, and communicate honestly and with sufficient frequency with the team.

Fortunately, barriers to providing team care are becoming less daunting, in large measure because of the rapidly expanding number of diabetes education programs and improved insurance coverage for services. The American Association of Diabetes Educators (800-TEAM-UP4; www.diabeteseducator.org) and the ADA (800-DIABETES; www.diabetes.org) can provide information regarding diabetes educators and education programs nationwide.

For diabetes care to be effective, communication and mutual respect among the patient-centered team is critical. Unfortunately, in many communities, the full benefit of the consultation and ongoing care with diabetes educators, nurses, dietitians, pharmacists, medical consultants, and primary care providers is not achieved because of overly hierarchical approaches to care. Non–health care professionals, such as lay or peer supporters, can provide benefit to patients in developing effective diabetes self-care behaviors. Key functions of such efforts have been identified and include assistance in managing and living with diabetes in daily life, social and emotional support, and linkage to clinical care.²³

Perhaps some of the most overlooked contributors to ineffective care in the setting of type 2 diabetes are the relatively common barriers created by psychiatric, neurocognitive function, and adjustment disorders, which are largely responsive to psychosocial therapies.²⁴

Nutrition

The ADA has published technical reviews that exhaustively document the literature regarding the effect of medical nutrition therapy and specific advice on diabetes-related outcomes such as A1C and weight, as well as a position statement.25–27 These are summarized in Table 22-4. An individually negotiated nutrition program in which each patient’s circumstances, preferences, cultural background, and the overall treatment program are considered is most likely to result in optimal outcomes. Ideally, a registered dietitian with specific skill and experience in implementing nutrition therapy in diabetes management should work collaboratively with the patient and other health care team members in providing medical nutrition therapy. For optimum outcomes, this counseling should be performed over a series of visits initially, with intermittent follow-up thereafter. Analogously, physicians and other members of the health care team need to support the nutritional plan developed collaboratively.

Individualized dietary advice can be developed by a physician from a brief diet history obtained by asking: “What do you eat for breakfast? … lunch? … supper? Do you have snacks between breakfast and lunch? … lunch and supper? … supper and bedtime? What do you drink during the day?” Ideally, this information should be obtained at each visit, with specific suggestions for change that both patient and provider agree are important in the context of the overall treatment plan as well as both achievable and sustainable. Easy issues to address include caloric beverages, which tend to elevate glucose levels dramatically and can generally be replaced relatively easily with artificially sweetened alternatives. Juices are generally perceived as healthy but can significantly affect glycemic control and total caloric intake. Portion control and recipe modification are excellent dietary techniques, particularly for meats and fried foods. Substituting low-fat products for higher-fat foods is often useful but needs to be done with the recognition that they are generally higher in carbohydrates. It is important that patients recognize that “fat-free” and “sugar-free” foods are not “free” and that attention to both total carbohydrate and calorie content are critical.

Eating approximately every 4 hours while awake is the most practical dietary plan for most overweight people. Frequent small meals have been shown to be of benefit when used in a controlled inpatient setting, but in general when overweight patients are encouraged to eat more frequently they often overeat more frequently. At a minimum, avoiding high-calorie snacks is reasonable advice for most people with diabetes. If all health care providers repeatedly obtain a diet history every few weeks to months, effective assessment of whether previously agreed-to changes were enacted, reinforcement of the importance of dietary efforts, and the patient’s gradual evolution to a more healthful diet through further dietary modification can be accomplished.

In general, the critical nutrient for glycemic control is carbohydrate. Essentially all carbohydrates consumed are converted to glucose in the gut and require the action of insulin to be cleared from the circulation. A dietary technique called carbohydrate counting can be used in patients with type 2 diabetes to facilitate consistent carbohydrate intake or to allow insulin dose adjustment in response to changes in carbohydrates consumed.²⁸ Although this technique requires less insulin than fixed meal dosing and may help curb weight gain, it has not been shown to improve glycemic control or reduce the rate of hypoglycemia in patients with type 2 diabetes.²⁹ Whereas the β cell in type 2 diabetes has generally lost its responsiveness to glucose, the second phase of insulin secretion is largely spared in type 2 diabetes and is in part driven by amino acids and fatty acids. Therefore, including some protein and fat in each meal and snack may be useful.

Dietary fat is the nutrient most closely associated in epidemiologic studies with the risk of developing type 2 diabetes. Although dietary fat clearly has a major impact on total caloric intake, as well as on circulating lipids, it has a minimal acute impact on glycemia. It is recommended that people with diabetes, if they are overweight, consume a diet modestly restricted in calories, with less than 10% of total calories as saturated fat, less than 10% as polyunsaturated fat, and total avoidance of trans fats. Some advocate substituting foods high in monounsaturated fatty acids—seeds, nuts, avocado, olives, olive oil, and canola oil—for carbohydrate, but most patients do not find adequate variety in the monounsaturated fatty acid category and often overeat these high-caloric-density foods.

Dietary protein similarly has a minimal impact on glucose levels, although as mentioned, amino acids do promote insulin secretion. Metabolism of protein results in the formation of acids and nitrogenous waste that may result in bone demineralization and glomerular hyperfiltration. At least 0.8 g of high-quality protein per kilogram is generally recommended. Protein restriction in the setting of kidney disease has been recommended and is more fully discussed in Chapter 28. There is no evidence that protein intake materially effects the risk of developing kidney disease in patients with diabetes.

The role of vitamins, trace minerals, and nutritional supplements in the treatment of diabetes is poorly understood. There are some patients and providers who are absolutely convinced of the utility of soluble fiber, magnesium, chromium, zinc, folic acid, pyridoxine, cyanocobalamin, vitamin A, vitamin C, vitamin E, vanadium, selenium, garlic, and others. Clinical trial data to support their safety and efficacy are inconclusive. Many patients are convinced that nutritional supplementation is healthful, and it is often counterproductive to engage in scholarly discussion of the nature of the evidence base for their decision. At a minimum, discussion should include the documented efficacy of lifestyle and pharmacologic interventions and the idea that these efforts should not be left by the wayside when budgetary constraints affect potentially more effective interventions.^30,31 A multivitamin/mineral preparation may be reasonable for most patients with diabetes. A recent randomized control trial in patients with diabetes demonstrated fewer self-reported infections and related absenteeism.³² Studies demonstrating the benefits of B-vitamin supplementation on restenosis after angioplasty³³ have recently been called into question; the possibility has been raised that such therapy could increase rates of restenosis after stent placement.³⁴

Although there are proponents of a wide range of dietary composition, there are few data to support these recommendations from long-term outcome studies of prescribed diets. Mixed meals containing 10% to 20% of calories from protein, no more than 10% of calories from saturated fat, no more than 10% from polyunsaturated fats, and the remainder largely from monounsaturated fats (seeds, nuts, avocados, olives, olive oil, canola oil) and carbohydrates, particularly whole grains, fruit, vegetables, and low-fat milk, are probably most reasonable. High-carbohydrate, low-fat diets, although historically recommended by many health organizations, have been shown to increase postprandial blood glucose and triglyceride levels, elevate fasting triglyceride levels, and decrease high-density lipoprotein (HDL) cholesterol levels in insulin-resistant people, including those with type 2 diabetes. Several studies have demonstrated improved lipid levels and blood glucose control in both short- and intermediate-term studies in which total fat intake approaches 45% of calories and carbohydrate intake is as low as 40% of calories. Reducing fat or carbohydrate intake in obese individuals will not necessarily lead to reduced calories. Since weight loss will occur only in the setting of caloric restriction, arguably the most appropriate approach is to limit intake of both fat and highly processed, easily digestible carbohydrates. The treatment of obesity is discussed in Chapter 2; the principles discussed are appropriate when type 2 diabetes is complicated by obesity. To date, short-term studies of medical nutrition therapy, physical activity, and comprehensive lifestyle approaches have been shown to improve the control of classic CVD risk factors, as well as intermediate markers of CVD risk such as C-reactive protein; no long-term, large-scale study of intentional weight loss has been powered to examine CVD endpoints. Look AHEAD (Action for Health in Diabetes) will examine CVD events for up to 11.5 years in a study in which patients with type 2 diabetes 45 to 74 years of age with a body mass index = 25 kg/m² will be recruited. Patients will be randomized to a 4-year intensive weight loss program (calorie restriction and physical activity) or to diabetes support and education. With planned recruitment of 5000 patients at 16 centers over 2.5 years, the study is designed to provide a 0.90 probability of detecting an 18% difference in major CVD event rates between arms.³⁵

Exercise

There is a substantial body of literature supporting exercise as a modality of treatment in type 2 diabetes, including a recent technical review by the ADA.^36,37 The recommendations of this technical review are summarized in Table 22-5. Exercise is perhaps the single most important lifestyle intervention in diabetes, because it is associated with improved glycemic control, insulin sensitivity, cardiovascular fitness, and cardiac remodeling. Aerobic exercise and resistance (strength) training both have a positive impact on glucose control. Improvements in glycemic control are generally apparent immediately, become maximal after a few weeks of consistent exercise, but only persist for 3 to 6 days after the cessation of training. To maintain effects on glycemia, a minimum of three exercise sessions a week is suggested, with no more than 2 days rest between sessions.

The recommended approach to promoting an increase in physical activity is analogous to that discussed for diet. Goals, methods, intensity, and frequency have to be negotiated with patients, with great sensitivity to recognizing barriers and helping patients discover solutions. The role of educators, exercise specialists, physical therapists, and social supports in this process is critical. The major role for the physician is to screen for complications (neuropathy, nephropathy, retinopathy, vascular disease) and discover ways for patients to be able to exercise safely. Exercise in the presence of uncontrolled diabetes, hypertension, retinopathy, nephropathy, neuropathy, and cardiovascular disease can occasionally result in devastating problems. Vigorous exercise should be avoided in the setting of proliferative or severe nonproliferative diabetic retinopathy; waiting 3 or more months after successful laser photocoagulation is probably prudent. In the setting of severe peripheral neuropathy, non-weight-bearing activities such as bicycling are probably appropriate. In the setting of known ischemic heart disease, an initial period of exercise under monitoring as provided in cardiac rehabilitation programs is appropriate. These issues can all be addressed creatively and should never provide an insurmountable barrier to increasing physical activity.

Recommendations regarding stress testing before initiating an exercise program in previously sedentary people are provided in some detail in Table 22-5. The utility of such treadmill tests in potentially low-risk populations is limited by their poor sensitivity and specificity.^38,39 It should also be noted that in an older asymptomatic population with normal electrocardiograms, adenosine technetium-99m sestamibi single-photon emission-computed tomography myocardial perfusion imaging revealed that over 20% of patients studied had silent ischemia. However, in 5-year follow-up, there was a spontaneous remission of abnormalities in many, generally very low event rates, and no difference in outcomes based on whether stress testing was performed or not.⁴⁰ If the planned exercise program does not involve more strenuous (both intensity and duration) activity than the patient has engaged in recently but merely involves more frequent activity, screening cardiovascular stress testing is unlikely to be particularly useful.

Even if the stress test does not demonstrate characteristics of ischemia, it is appropriate to counsel patients to avoid overexertion and seek medical follow-up for exertional symptoms such as chest, jaw, or arm discomfort or for worsening dyspnea. Improving exercise tolerance should be viewed as a measure of improving cardiorespiratory function.

For the average patient with type 2 diabetes starting an exercise program, these levels of moderate exertion will generally be reached with quite low-level activity initially, such as walking at a pace of 2 miles an hour. Initially, it may even be necessary to negotiate once-weekly or shorter-duration exercise sessions and proceed from there, picking up the pace as tolerated and increasing the duration and frequency of exercise sessions slowly to avoid overuse injuries.

Self-Monitoring of Blood Glucose

Self-monitoring of blood glucose (SMBG) has not been demonstrated in clinical trials to substantially change outcomes in non-insulin-treated type 2 diabetes when evaluated in isolation and may be associated with decreased quality of life.^41,42 However, many diabetes self-management programs have been associated with improved glycemic control. In all of these, SMBG was integral to the overall process, suggesting that SMBG is at least a component of effective therapy. The frequency and type of glucose monitoring should be determined in collaboration with the patient, taking into account the overall treatment plan and goals, the patient’s abilities, and the stage of diabetes. SMBG can theoretically improve the safety of treatment with insulin or sulfonylureas, since it allows the identification of minimal or asymptomatic episodes of hypoglycemia. Coupled with appropriate education, dose adjustments, or modest changes in the lifestyle plan, SMBG can minimize risk of severe hypoglycemia or weight gain. Although severe hypoglycemia is relatively rare in type 2 diabetes, it is more common in the elderly and can have devastating consequences, such as physical injury as a result of trauma or a change in the perceived ability of a patient to continue to live independently as a result of confusion or loss of consciousness. In general, it is optimal to have patients use glucose monitoring to assess the nature of any hypoglycemic symptoms they experience. Many patients are fearful of hypoglycemia and routinely consume extra calories in response to a variety of life’s circumstances—such as when they are hungry, sweaty, nervous, or upset—without documenting hypoglycemia. Most “hypoglycemic” symptoms in patients with type 2 diabetes are not related to hypoglycemia and do not need to be treated with calorie consumption. Counseling patients to carry commercially available glucose tablets and glucose monitoring equipment at all times and to take a 15-g dose of carbohydrate for documented mild to moderate hypoglycemia helps patients avoid excessive calorie intake and recreational consumption of sweets for treatment of hypoglycemia.

Optimal timing of SMBG will vary depending on individual characteristics and treatment. It is important to advise patients to vary the time of the day at which blood glucose levels are checked. For some patients, the highest blood glucose of the day is the morning glucose, whereas for others the highest is before bed. Particularly in early diabetes, gestational diabetes, and well-controlled diabetes, monitoring 1 to 2 hours after meals allows patients to assess the effect of the combined lifestyle and pharmacologic efforts in controlling postprandial glucose levels, which often are the only glycemic abnormality present.

When glucose control is poor, having patients concentrate on premeal glucose levels is perhaps most productive. Once the premeal glucose levels reach the low 100s, many advocate that patients switch to checking 1- to 2-hour postprandial glucose levels, because it amplifies the observed effect of diet on glycemic control and enables patients to see that moderate changes in meal plan, activity, and medications have a significant impact on glucose levels. Even after substantial inappropriate changes in food intake, activity, or timing or dose of medication, blood sugar values often return to near-normal levels overnight or by the time of the next meal.

The frequency of glucose monitoring needs to be matched to individual patient needs and treatment. Many clinicians ask patients to monitor at least once a day, varying among breakfast, lunch, supper, bedtime, and mid-sleep, as well as with symptoms. Others ask patients to monitor with intensity similar to that described for patients with type 1 diabetes (four times per day before meals and bedtime, with occasional postprandial and mid-sleep checks). Some ask for sets of glycemic readings more infrequently (e.g., fasting and 1 hour after the biggest meal). In the subset of patients who achieve stable blood glucose levels without significant hypoglycemia, it is generally appropriate to decrease the frequency of SMBG to a few times a week. The important characteristic of the timing of SMBG is that it be frequent enough to optimally inform the treatment plan and that both patient and provider have a good understanding of both the adequacy and the stability of glycemic control.

It has been assumed that the benefits of SMBG stem from facilitating self-management. Theoretically, if patients are aware of the glycemic targets associated with the outcomes they seek to achieve, SMBG enables them to evaluate their response to therapy and make adjustments or seek help as needed to achieve goals. It is useful for patients to keep a daily diary of their SMBG results so they can assess their results periodically and share them with the health care team. Unfortunately, many patients faithfully perform daily or more frequent SMBG, record the results as instructed, and discuss them with their health care team only at quarterly visits. Unless SMBG results are within the agreed-to targets, they should be communicated and reviewed regularly with a member of the health care team by telephone, fax, mail, or e-mail or at an interim visit to trigger changes in therapy as the need arises. Unfortunately, such services are generally not reimbursed, which places an unsustainable burden on many health care teams.

Finally, one of the most difficult areas for health care providers to remain current is in the area of available equipment and supplies, particularly for glucose monitoring. Diabetes educators often have demonstration models and a robust understanding of patient characteristics that match well or poorly with particular devices; they can be exceptionally helpful to patients in selecting appropriate equipment. A useful resource in this regard is the annual Resource Guide, which comes out as the January issue of Diabetes Forecast, a magazine for lay people with diabetes and their families. It is available online at http://forecast.diabetes.org/magazine/archive to find the most recent January issue.

Insulin Sensitizers with Predominant Action in the Liver: Biguanides

Metformin is the only biguanide available in the United States and the subject of a recent review.⁴⁵ Phenformin was removed from the United States market in the 1970s because of deaths associated with lactic acidosis. Metformin activates adenosine monophosphate–activated protein kinase (AMPK), a critical molecule in insulin signaling with effects on metabolism and energy balance. Metformin reduces hepatic insulin resistance, gluconeogenesis, and glucose release.⁴⁶

Metformin has a half-life of 2 to 6 hours and reaches peak concentrations in approximately 1 hour when taken with meals. It is generally administered at least twice daily, although sustained-release formulations, which are particularly effective when administered with the evening meal, are now available. The glucose-lowering efficacy and the prevalence of adverse gastrointestinal effects increase proportionately in the dose range 500 to 2000 mg/d. The maximal dose of 2550 mg does not generally provide additional benefit beyond that seen at 2000 mg daily. Since metformin does not increase insulin levels, it is not associated with a significant risk of hypoglycemia. The most common adverse events are gastrointestinal: nausea, diarrhea, crampy abdominal pain, and dysgeusia. About a third of patients have some gastrointestinal distress, particularly early in their course of treatment. These adverse effects can be minimized by starting with a low dose (500 mg) once daily with a meal and titrating upward slowly (over weeks) to more effective doses (500 to 1000 mg twice daily). Sustained-release metformin is associated with less frequent and less severe upper gastrointestinal symptoms but can increase the frequency of diarrhea, a much less common adverse effect overall. The majority of patients have no treatment emergent complaints during metformin initiation, and at least 90% tolerate it adequately with long-term use. Perhaps as a result of clinical or subclinical gastrointestinal effects, metformin is associated with less weight gain than other available oral antihyperglycemic agents and in some studies has even been shown to produce modest mean weight loss.

The issue of greatest concern regarding metformin therapy is lactic acidosis, which is quite rare and occurs almost exclusively in patients who are otherwise at high risk of developing lactic acidosis.⁴⁷ When used prudently, the risk of lactic acidosis is virtually zero. Patients at increased risk of developing lactic acidosis due to baseline medical conditions should avoid taking metformin.⁴⁸ The package insert suggests that metformin is absolutely contraindicated in patients with renal insufficiency, because the drug is cleared renally.⁴⁹ The drug should not be used in males with a serum creatinine greater than or equal to 1.5 mg/dL or in females at 1.4 mg/dL. There have been recent suggestions that metformin dosing should be based on estimated glomerular filtration rate (eGFR), with contraindication to use in stage 4 to 5 chronic kidney disease (eGFR < 30 mL/min/1.73 m²) and caution with eGFR less than 50 mL/min/1.73 m².⁵⁰

Metformin was originally contraindicated in patients with congestive heart failure requiring treatment. However, recent reports suggest improved morbidity and mortality in patients with stable compensated heart failure.⁵¹ This prompted the U.S. Food and Drug Administration (FDA) to recommend a change in labeling in 2005 to allow its use in patients with stable compensated heart failure without other contraindications to its use.⁵²

Metformin is also contraindicated in patients with hepatic insufficiency, acidosis, hypoxia, and in those with a history of binge drinking or heavy alcohol use. Caution is required in elderly people, and specific recommendations to measure creatinine clearance with a timed urine collection prior to initiating therapy in patients over the age of 80 are provided in the package insert. Patients with acute illness, poorly controlled chronic illness, surgical indications, dehydration, or simultaneous treatment with nephrotoxic drugs (e.g., iodinated contrast dye) should have metformin withheld until the clinical course is stabilized and adequate renal function is confirmed.

Metformin has the strongest record of achievement of antidiabetic agents in outcome studies in patients with type 2 diabetes. In the UKPDS, among overweight subjects, those randomly assigned to metformin not only had improvements in microvascular complications similar to those of subjects randomly assigned to insulin and sulfonylurea but also exhibited a reduction in diabetes-related deaths and myocardial infarction.¹⁰ The validity of this observation has been challenged because of unusual responses in a subsequent subrandomization. Beneficial effects of metformin on macrovascular complications through glucose-independent mechanisms are plausible and suggested by consistent observations, such as metformin-associated reductions in low-density lipoprotein (LDL) and procoagulant factors.⁵³

Insulin Sensitizers with Predominant Action in Peripheral Insulin-Sensitive Tissues: Thiazolidinediones

The thiazolidinedione class of drugs, often termed TZDs or glitazones, has engendered great enthusiasm and controversy since the first agent, troglitazone, was approved in 1997.^54–55 Troglitazone was withdrawn from the U.S. market in the year 2000 as a result of rare cases of fulminant hepatic necrosis, and because pioglitazone and rosiglitazone appear to be safer. This class of oral antidiabetic agents is believed to work by binding to and modulating the activity of a family of nuclear transcription factors termed peroxisome proliferator–activated receptors (PPARs), particularly PPAR-γ. They are associated with slow improvement in glycemic control over weeks to months in parallel with an improvement in insulin sensitivity and reduction of free fatty acid (FFA) levels. Pioglitazone and rosiglitazone are generally well tolerated, with weight gain and fluid retention the most common adverse effects. There is no substantial evidence that these newer agents are associated with hepatotoxicity, but this record of safety has been established in the setting of careful liver function test monitoring. Therefore, it is important to continue to recommend that glitazones not be used in patients with active hepatocellular disease or with unexplained serum alanine aminotransferase levels greater than 2.5 times the upper limit of normal; routine monitoring of liver function tests with these glitazones is no longer recommended.^56,57

A promising attribute of the glitazones that has generated great interest is an effect to improve insulin secretory dynamics in subjects with diabetes and impaired glucose tolerance. Initial reports suggest that these agents provide for more durable glycemic control than other antidiabetic agents.⁵⁸ These latter observations provide hope that pioglitazone and rosiglitazone may be useful not only in diabetes prevention but also in halting the progression of established diabetes, thus reducing the need for additional drug therapy over time.⁵⁹

Early studies from small, short-term, randomized controlled trials suggested that the glitazone agents have important cardiovascular benefits, including reduced progression of carotid intimal medial thickness, decreased rates of in-stent restenosis in the setting of percutaneous coronary intervention, normalization of vascular endothelial function, improvements in dyslipidemia, reduction of blood pressure, improvements in fibrinolytic and coagulation parameters, and reduction in inflammatory markers.^54,55 Data from large multicenter trials examining hard clinical endpoints have revealed potential differences in cardiovascular effects between the two agents. Pioglitazone did not significantly reduce the primary composite endpoint of all-cause mortality, nonfatal myocardial infarction, stroke, acute coronary syndrome, endovascular or surgical intervention in the coronary or leg arteries, and amputation above the ankle over a period of nearly 3 years.⁶⁰ Although the clinical relevance is debatable, the secondary endpoint, all-cause mortality, nonfatal myocardial infarction, and stroke was significantly decreased. Meta-analyses and interim analyses of ongoing studies indicate either harmful or neutral cardiovascular outcomes from rosiglitazone and either neutral or beneficial outcomes from pioglitazone.^54–55 As a result, the FDA requested that the prescribing information for rosiglitazone be updated with a black box warning indicating the possible induction of myocardial ischemia with this drug and advised against co-administration with insulin or in patients with heart disease who are taking nitrates, both instances associated with particular risk.⁶¹ Very recently, a randomized trial of rosiglitazone versus metformin or sulfonylurea demonstrated no difference in cardiovascular deaths or hospitalizations.⁶² The major difference between these two glitazones is related to lipid effects; pioglitazone leads to greater increases in HDL cholesterol and smaller increases in LDL cholesterol than rosiglitazone and lowers triglycerides, while rosiglitazone tends to raise triglycerides.⁶³ A recent large observational study suggests that among people older than age 65, pioglitazone is associated with a significantly lower risk of heart failure and death than rosiglitazone.⁶⁴

The adverse effect that has engendered the most concern regarding this class of drugs is fluid retention and its manifestations: peripheral edema, a decrease in hemoglobin concentration, and heart failure.⁶⁵ In the registration trials for rosiglitazone and pioglitazone, patient withdrawals for edema or heart failure were less than 1%. The patients most likely to experience edema are those treated with insulin and those with preexisting edema. Women, obese patients, and those with known ischemic heart disease, heart failure, diastolic dysfunction, or renal insufficiency also seem to be at increased risk. It is prudent to start high-risk patients with the lowest marketed dose (pioglitazone 15 mg daily or rosiglitazone 2 mg daily). Increasing edema can be treated with restricted sodium intake or diuretics, and limited data suggest particular benefit from amiloride or spironolactone.⁶⁶ It should be noted that fluid retention to the point of congestive heart failure requiring hospitalization has been confirmed in large randomized controlled trials.^62,67 Findings such as these have prompted the FDA to recommend strengthening the product labeling of pioglitazone and rosiglitazone to caution against use in patients with any degree of heart failure. However, no study has demonstrated worsening myocardial structure or function with these drugs.

Another issue that engenders substantial clinical concern regarding glitazone therapy is weight gain. Careful study indicates that the weight gain is a result of subcutaneous and not visceral fat accumulation; in fact, there is a reduction in visceral, hepatic, and intramyocellular fat. Therefore, the weight gain observed with glitazones, while potentially having negative consequences from a cosmetic standpoint, is perhaps less likely to cause adverse cardiovascular effects. The weight gain with glitazones is minimized with concomitant metformin therapy and greatest with insulin and sulfonylurea therapy. Both weight gain and fluid retention are more common and severe in patients with the greatest glycemic responses, making expectant management of these adverse effects mandatory.

Finally, increased fracture rates were reported in clinical trials of both rosiglitazone and pioglitazone, mainly in women.^58,68 Distal sites were primarily affected in these studies, but small randomized controlled trials identified the loss of bone density (using DEXA scanning) at the lumbar spine as well.⁶⁸ Preclinical studies suggest that activation of PPAR-γ inhibits bone formation by diverting stem cells from the osteogenic to the adipocytic lineage and may stimulate the development of osteoclasts. Currently, no data are available with respect to the prevention or management of thiazolidinedione-related bone loss, but prudent measures would include at a minimum an assessment of risk factors and appropriate bone density screening.

Insulin Secretagogues

Currently available insulin secretagogues all bind to the sulfonylurea receptor (SUR1), a subunit of the adenosine triphosphate (ATP)-sensitive potassium channel (K_ATP) on the plasma membrane of pancreatic β cells. The SUR1 subunit regulates the activity of the channel and also binds ATP and adenosine diphosphate (ADP), effectively functioning as a glucose sensor and trigger for insulin secretion. The K_ATP channel closes upon sulfonylurea binding, as well as with increases in intracellular ATP and decreases in ADP, as a result of fuel metabolism. The membrane depolarization that ensues causes the opening of voltage-dependent L-type calcium channels. Subsequent calcium influx results in an increase in intracellular calcium, which leads to insulin secretion. Differences in pharmacokinetic and binding properties of the various insulin secretagogues result in the specific responses that each agent produces. The major clinical differences between them seem to be related to duration of action and to fairly subtle variations in their hypoglycemic potential.

Sulfonylureas

The sulfonylureas have been available since the 1950s. They have a relatively quick onset of action and variable duration of action. There are numerous choices available (Table 22-7) which can be divided into first- and second-generation agents. In general, the second-generation agents (glipizide, glyburide, and glimepiride) are more potent and as a result have fewer adverse effects and drug-drug interactions. Extended-release glipizide and glimepiride are preferred agents because they can be dosed once daily in the vast majority of patients and involve a relatively low risk of hypoglycemia and weight gain. Glyburide has a higher risk of hypoglycemia and has been associated with an abrogation of ischemic preconditioning, a protective autoregulatory mechanism in the heart, mediated by interaction with the vascular and cardiac SUR2 receptors.⁶⁹

There has also been concern regarding the potential for these agents to accelerate the progression of β cell failure, a key feature in the pathogenesis of type 2 diabetes.62,70–71 Human data are difficult to acquire because only indirect approximations of β-cell mass are available. Metformin and thiazolidinediones do appear to preserve glycemic control better than glyburide, but it is unknown whether this is a class effect.⁵⁸ Nevertheless, the long-term safety and low cost continue to justify a solid role for sulfonylureas in the medical management of type 2 diabetes.

The maximum marketed dose is generally two to four times higher than the maximally effective dose, likely because sulfonylureas were originally the only oral agents available for diabetes. To minimize the risk of hypoglycemia, it is probably most appropriate to employ relatively low doses of sulfonylureas. In general, limiting the dose to one-fourth of the maximum (e.g., glipizide-ER 5 mg, one tablet daily or glimepiride 4 mg, half a tablet daily) minimizes costs and adverse events. Even smaller doses of sulfonylureas are remarkably effective, particularly in patients on concomitant insulin-sensitizing therapy, and well-tolerated. Small doses are also indicated in treating elderly patients, who are more susceptible to sulfonylurea-induced hypoglycemia, particularly in the setting of declining renal function or erratic meals and activity.

Side effects other than hypoglycemia and modest weight gain are uncommon, particularly for the second-generation agents, but include nausea, vomiting, rashes, purpura, and pruritus. Rare patients may develop hematologic reactions such as leukopenia, thrombocytopenia, hemolytic anemia, and cholestasis with or without jaundice.

Repaglinide

Repaglinide is a member of the meglitinide family of insulin secretagogues, distinct from the sulfonylureas. It is quickly absorbed and has a short half-life and a distinct SUR1-binding site. As a result, it produces a generally faster and briefer stimulus to insulin secretion. It is dosed with each meal and provides better postprandial control, as well as less hypoglycemia and weight gain than glyburide. Repaglinide does seem to have a long residence time on the sulfonylurea receptor and as a result has a more prolonged effect on fasting glucose than would be expected based on its short pharmacologic half-life. Repaglinide is available in 0.5-mg, 1-mg, and 2-mg tablets. The maximal dose is 4 mg with each meal. As is the case with the sulfonylureas, there is a minimal glucose-lowering advantage of high doses versus moderate doses of repaglinide.

Nateglinide

Nateglinide is a derivative of phenylalanine, structurally distinct from both sulfonylureas and the meglitinides. It has a quicker onset and shorter duration of action than repaglinide. Its interaction with SUR1 is fleeting. As a result, its effect to lower postprandial glucose is quite specific, and it only modestly lowers fasting glucose. This provides advantages (less hypoglycemia) and disadvantages (less overall glucose-lowering effectiveness). Therefore, nateglinide is most appropriately used when fasting glucose levels are modestly elevated in the setting of early diabetes or in combination with insulin sensitizers or long-acting evening insulin. Nateglinide is available as 120-mg tablets and is dosed with each meal. A 60-mg tablet is available but is reserved for use in patients with minimal hyperglycemia.

The rationale for stimulating insulin secretion in a way that minimizes fasting hyperinsulinemia and maximizes postprandial control is compelling.⁷² These agents theoretically would be advantageous in comparison to sulfonylureas for patients with erratic eating patterns, renal dysfunction, or who otherwise are at increased risk of hypoglycemia. Clinical studies demonstrate that repaglinide leads to a reduction in endothelial dysfunction⁷³ and carotid intima media thickness⁷⁴ relative to sulfonylureas, observations that are attributed to their specificity for postprandial glucose. In addition to these potential cardiovascular advantages, these newer “glinide” agents demonstrate little binding to the vascular smooth muscle and cardiac SUR2 receptors. However, the use in the United States of these newer agents has been modest, in part because of the need for multiple daily doses, greater expense than with sulfonylureas, and lack of head-to-head comparative studies that demonstrate superiority over newer sulfonylureas, which have similarly been shown to have low potential for producing hypoglycemia and weight gain.

Carbohydrate Absorption Inhibitors: α-Glucosidase Inhibitors

α-Glucosidase inhibitors (AGIs) inhibit the terminal step of carbohydrate digestion at the brush border of the intestinal epithelium. When administered with the first bite of a carbohydrate-containing meal, carbohydrate absorption is shifted more distally in the intestine and is therefore delayed, allowing the sluggish insulin secretory dynamics characteristic of type 2 diabetes to catch up with carbohydrate absorption. There are two currently marketed agents in the United States: acarbose and miglitol. Acarbose is largely not absorbed from the intestine, whereas miglitol is. Their use has been limited by a number of factors, including frequent gastrointestinal complaints, the need to administer the medication at the beginning of each meal, and modest reductions in fasting glucose and A1C. These concerns should be balanced against the ability of the AGIs to lower glucose in virtually everyone, without hypoglycemia or weight gain. The major adverse effects are flatulence, abdominal distress or distension, and diarrhea. These result from excessive blockade of carbohydrate absorption in the small bowel, leading to fermentation and gas production in the colon. To maximize the likelihood that these agents will be tolerated, start with a low dose, such as one-fourth of the maximum dose once daily, and increase over a period of weeks to months to one-fourth to half the maximal doses with each meal. A recent study in prediabetic patients demonstrated a statistically significant reduction in cardiovascular endpoints and new hypertension associated with treatment with acarbose, suggesting that modulating postprandial excursions or perhaps nonglycemic effects of the drug may provide for benefits vis-à-vis cardiovascular disease.⁷⁵ This notion is further supported by the observation in a single-center subgroup analysis from the same study that progression of intima media thickness is also slowed by acarbose.⁷⁶ The AGIs are contraindicated in patients with chronic intestinal conditions, particularly inflammatory bowel disease.

Incretin-Based Therapies

Incretins are peptide hormones released by the gut that have a multitude of beneficial effects on glucose metabolism. Several recent reviews of these therapies are available.77–79 Incretins promote insulin secretion in response to an oral carbohydrate load in excess to that observed from intravenous carbohydrate exposure, a phenomenon called the incretin effect. Glucose-dependent insulinotropic peptide (GIP) and glucagon-like peptide 1 (GLP-1) account for the majority of the incretin effect, the latter being most critical. GLP-1 exhibits glucose-lowering effects through stimulation of insulin secretion, inhibition of glucagon production, slowing gastric emptying, and improving satiety. The insulin secretory effect is glucose-dependent, meaning that the β-cell stimulatory effect is abolished as the glucose level approaches hypoglycemia. Despite the inhibitory effect on glucagon, the counterregulatory hormone response to hypoglycemia is preserved with GLP-1 activity. GLP-1 has important β-cell protective effects in preclinical studies, suggesting that therapeutic approaches may prevent diabetes or slow the progression of β-cell failure, but this requires confirmation in humans. Frankly, many authorities believe that clinically significant benefits on preservation or augmentation of β-cell mass is unlikely. GLP-1 also has binding sites in the appetite centers of the hypothalamus and confers an important regulatory effect on food intake and weight. The enzyme, dipeptidyl peptidase IV (DPP-IV) rapidly inactivates GLP-1 and GIP. Therefore, therapeutic approaches involve agents that are resistant to degradation by DPP-IV (such as exenatide) or that potentiate GLP-1 levels through inhibition of DPP-IV (DPP-IV inhibitors).

Exenatide

Exenatide is a synthetic form of a naturally occurring GLP-1 receptor agonist that is found in the saliva of the Gila monster and exhibits all of the effects of naturally occurring GLP-1. Exenatide is approved as monotherapy or for add-on therapy to sulfonylureas, metformin, or thiazolidinediones.⁸⁰ It is not currently approved or in combination with insulin, because limited data are available. Exenatide effectively reduces postprandial and fasting glucose and exhibits sustained reductions in HbA_1c of approximately 1.0% and weight loss of 2.6 to 3.6 kg over 6 months and about 5 kg over 2 to 3 years, at least in those who continue therapy. It is available in a prefilled pen and dosed twice daily subcutaneously up to 60 minutes before the main meals. The most common side effects of exenatide are nausea, vomiting, and diarrhea and are minimized through slow titration. Gastrointestinal side effects generally diminish with continued use. Patients should initiate therapy with the 5-mcg dose in the 60 minutes before breakfast and supper. Pre-lunch dosing may be substituted for the pre-breakfast dose, so long as lunch and supper are separated by at least 6 hours. The dose should be increased to 10 mcg after 1 month, provided that nausea is minimal. Because of increased risk of side effects, contraindications include gastroparesis and severe renal disease (creatinine clearance <30 mL/min). Postmarketing cases of acute pancreatitis have been reported in association with exenatide, but a causal link has not been proven, nor has a mechanism been established. Nevertheless, it is recommended that exenatide not be restarted following any case of pancreatitis. Hypoglycemia is uncommon except in combination with sulfonylureas. In summary, exenatide has important therapeutic advantages, including weight loss, lack of hypoglycemia, and relatively few contraindications. However, the injectable route, side-effect profile, and expense ultimately limit its use as a first-line agent. Long-acting investigational GLP-1 receptor agonists (liraglutide and exenatide in a once-weekly preparation) are under review by the FDA; they lower fasting glucose to a greater extent and are associated with greater reduction in A1C and better tolerability.

Dipeptidyl Peptidase IV Inhibitors

DPP-IV inhibitors share similar features with exenatide, the major exception being weight neutrality. Furthermore, these agents do not have dramatic effects on gastric emptying and therefore produce a lesser degree of gastrointestinal side effects. As with exenatide, DPP-IV inhibitors exhibit β-cell preservation properties in preclinical studies. The DPP-IV enzyme has numerous other substrates, but studies of dual-receptor knockout mice for GLP-1 and GIP demonstrate that augmentation of GLP-1 and GIP levels are primarily responsible for the metabolic effects. Specificity appears to be crucial, inasmuch as less specific inhibitors have demonstrated adverse effects on immune function and cancer growth in animal studies. While highly selective inhibitors currently marketed or under late stages of development for diabetes have not shown such properties, continued long-term surveillance will be of particular interest.

Currently, only two DPP-IV inhibitors, sitagliptin and saxagliptin, are approved for use, but numerous DPP-IV inhibitors are in late stages of development, including vildagliptin and alogliptin. The currently available agents are approved for use as monotherapy and combination therapy, though not with insulin.81,81a The usual dose is the maximum dose, which is administered without titration. Owing to pharmacokinetic considerations, lesser-sized tablets are available and recommended in patients with estimated GFR less than 50 to 60 mL/min/1.73 m². This class of drugs is very well tolerated and has very few contraindications. Side effects are uncommon but may include nasopharyngitis and headache. Serious hypersensitivity reactions have been reported in aftermarket surveillance, including anaphylaxis, angioedema, and exfoliative skin conditions with sitagliptin, but causality is difficult to substantiate because of the rarity of events. The weight neutrality, lack of hypoglycemia, broad applicability, tolerability, and ease of use create a unique niche for DPP-IV inhibitors. However, as with exenatide, the precise role of these agents in diabetes management has not been established. Cost, modest efficacy, and uncertain long-term safety remain barriers to use.

Pramlintide

Pramlintide is a synthetic soluble form of the naturally occurring hormone, amylin, which is cosecreted with insulin from the β cell. The pharmacologic effects of amylin are complementary to insulin in that they inhibit the inappropriate release of postprandial glucagon and slow gastric emptying. The primary consequence is reduction in postprandial glucose, with modest reductions in overall glycemia (HbA_1c −∼0.5%). Pramlintide also induces modest weight loss through control of appetite centers in the brain. As with GLP-1, the counterregulatory hormone response to hypoglycemia is unaltered, and in the absence of other therapies, pramlintide does not cause hypoglycemia. That being said, early clinical trials that did not make anticipatory reductions in insulin dosage did demonstrate severe hypoglycemia in patients with type 1 diabetes.

Pramlintide is currently only approved for use in conjunction with basal-bolus insulin regimens in patients with type 1 or type 2 diabetes who are not meeting glycemic targets.⁸² Pramlintide was originally only available by vial and syringe but recently, a multi-use pen with several dose increments that facilitate titration became an option. For type 2 diabetes, the starting dose is 60 mcg (10 units) subcutaneously before major meals (those exceeding 30 g of carbohydrates), increasing to 120 mcg (20 units) if a patient has been nausea-free for 3 to 7 days. The major side effect is nausea, but it appears to be less intense than that observed with exenatide and also appears to be minimized with slow titration. Patients who have difficulty with 60 mcg may benefit from the lower-dose pen intended for the treatment of type 1 diabetes, which starts at a dose of 15 mcg and increases to 30 mcg. It is advisable that the prandial insulin dose be decreased by 50% at the initiation of pramlintide, although ultimately, most patients need a smaller decrement. Postprandial glucose levels generally peak later and at a smaller amplitude following pramlintide administration than with prandial insulin alone. This may lead to early hypoglycemia (in the first 30 minutes) followed later by hyperglycemia when administered in conjunction with typical rapid-acting insulins. Consequently, the timing of the prandial insulin dose may also require adjustment. The drug is contraindicated in patients with gastroparesis, history of severe hypoglycemia, or hypoglycemia unawareness. Because of the complexity of the regimen—resulting from multiple injections and the need for active concomitant titration of insulin—pramlintide is only recommended for highly motivated patients. Furthermore, its use is only advisable for patients in need of modest refinement of glycemic control. Additional study is needed to determine whether there is utility for monotherapy or concomitant use with basal insulin alone or oral agents.

Insulins

Insulin has been commercially available since the early 1920s and is arguably still the mainstay of therapy for the majority of people with type 2 diabetes worldwide. Subcutaneous injection of insulin in type 2 diabetes can be used to supplement endogenous production of insulin, both in the fasting state to modulate hepatic glucose production and in the postprandial state to facilitate glucose clearance into muscle and fat for storage. Currently, the vast majority of insulin used worldwide is of recombinant human origin; animal-source insulin is difficult to obtain and more expensive. The available formulations largely differ in their pharmacokinetics, as reviewed in Table 22-8.

There are now three rapid-acting insulin analogs—lispro, aspart, and glulisine—each with onset of action in 5 to 15 minutes, near-peak activity from approximately 45 minutes to 2 hours, and a duration of activity of approximately 4 hours, with slightly more prolonged activity at higher doses. These three molecules have a different structural modification that disrupts the “tail structure” of the insulin molecule; as a result, they do not exhibit as great a tendency to form dimers and hexamers at high concentration. Marketing suggests that there are differences between agents, but on the background of endogenous insulin secretion in the setting of type 2 diabetes, clinically relevant differences in glycemic control or adverse effects have not been demonstrated. These rapid-acting analogs are associated with better postprandial glucose control than regular human insulin, as well as a reduction in the risk of hypoglycemia in some studies. They should be administered up to 15 minutes before meals or even immediately after meals. Dosing after meals in combination with techniques such as carbohydrate counting may be helpful in overweight patients with type 2 diabetes who often have no idea what they will eat, effectively until they are done eating. Carbohydrate counting may be cumbersome for some patients, and a fixed-dose prandial insulin regimen that is adjusted weekly may result in similar improvements in glycemic control.⁸³ However, insulin requirements are lower with carbohydrate counting, and weight gain tends to be less.

Regular human insulin is approximately half as fast as the rapid-acting analogs, with onset in 10 to 30 minutes, peak activity at 2 to 5 hours, and duration of action of 5 to 8 or more hours. Regular insulin should ideally be administered 30 to 60 minutes before meals, which certainly has issues with respect to adherence and the potential to produce hypoglycemia if the meal is subsequently missed. Intermediate-acting insulin, neutral protamine Hagedorn (NPH), is approximately twice as slow as regular insulin, with an onset of action in 1 to 2 hours, a fairly broad peak of activity from 6 to 12 hours, and a duration of action of 16 to 24 or more hours. Human regular and NPH insulin have a dramatic dose dependence to their profile of activity: the higher the dose, the broader the peak and longer the duration of action. The manufacturing of other human insulins, Lente and Ultralente, has been discontinued.

Insulin glargine is a novel long-acting insulin analog with distinctive properties. It is administered as a clear, colorless, acidic solution and therefore cannot be mixed with other forms of insulin. Upon subcutaneous injection, the solution is neutralized, and the insulin glargine precipitates nearly instantaneously. It provides a flat, essentially peakless profile of activity, with a duration of action of more than 24 hours in most people with type 2 diabetes. Injection discomfort is related to the acidity of the solution and may be reduced by warming the insulin to room temperature.

Detemir is the most recently developed insulin analog. It is relatively rapidly absorbed but has a prolonged duration of action, attributable in large part to its acylation at position B29 with a 14-carbon fatty acid (myristic acid). As a result of this structure, detemir is highly protein bound in interstitial fluid and plasma. Like glargine, detemir exhibits a nearly peakless profile, but the duration of action is several hours shorter, at least in patients with type 1 diabetes.⁸⁴ In patients with type 2 diabetes, insulin-clamp studies demonstrate that the duration of both basal insulin analogs is dose dependent, lasting more than 24 hours in most patients.⁸⁵ This indicates that detemir can be dosed once daily in most patients if they require over 0.4 units/kg. In the only head-to-head clinical trial in patients with type 2 diabetes, detemir exhibited similar improvements in glycemic control as glargine when either was added to existing oral agents.⁸⁶ However, about half of patients on detemir received twice-daily dosing. Although larger doses were required to attain glycemic control, detemir was associated with slightly less weight gain than glargine. This has been attributed to the lipophilic properties of detemir, which may facilitate entry into the hypothalamic appetite centers in the brain. Detemir has also been shown in carefully controlled conditions to display less day-to-day variability in insulin action than NPH or glargine.^84,87

Insulin may be indicated in patients with type 2 diabetes who are not meeting glycemic targets despite two oral agents, in patients with contraindications to or inability to afford intensification with oral agents, in symptomatic patients (weight loss, fatigue), or in patients with poor glycemic control (HbA_1c >10%). In the latter case, it may be possible to wean insulin over time as glucotoxicity resolves. Though sulfonylureas are generally maintained as insulin is initiated and titrated, they generally can be discontinued later, with a modest further increase in the insulin dose (10 to 30 units) without a deterioration in glycemic control. Metformin in particular may prevent excessive weight gain on insulin therapy. Patients receiving concomitant thiazolidinediones are at increased risk of edema and heart failure, and discontinuation or dose reduction should be considered. Specifically, the combination of rosiglitazone with insulin is not recommended in the prescribing information because of concerns regarding increased risk of myocardial infarction in this setting. Pioglitazone is approved for use in combination with insulin in the United States. Arguably, the combination of insulin, metformin, and pioglitazone is the most powerful combination of glucose-lowering therapies available. Several strategies for initiating insulin therapy follow.

Strategies

It is important that both patient and health care provider agree on how to reach the goals of therapy. Therefore, biases and concerns of the patient should be addressed when trying to determine which agent should be prescribed. These biases can be elucidated in interviews with patients through discussions of various strategies, some of which follow.

Guidelines

It is a disappointing reality that despite multiple effective agents for treating hyperglycemia, only 57% of patients with diabetes reach minimum treatment goals for A1C specified by professional organizations.⁹⁹ The reasons for this are multiple, but the ever-increasing range and number of agents, as well as the increasing complexity of care can contribute to therapeutic inertia among providers and patients. The ADA and the European Association for the Study of Diabetes (EASD) published a consensus statement for step-wise, goal-directed management in order to assist patients and providers with managing hyperglycemia.⁴³ The choice of individual agents is based upon efficacy in lowering A1C, long-term safety and durability, side effects and contraindications, ease of use, and cost.

Given the high frequency of failure of diet and exercise alone, the consensus statement recommends metformin in conjunction with lifestyle measures at the time of diagnosis. Metformin is inexpensive, effective, weight-neutral and has a long safety profile in patients for whom there are no contraindications or intolerable side effects.

In general, the statement recommends combination regimens that incorporate therapies with complementary mechanisms of action. The ADA-EASD statement supports either insulin or sulfonylureas as “tier-1” options for add-on therapy in patients who inadequately respond to metformin. Both are rapidly effective, have a long safety profile, and are relatively inexpensive. Insulin is presented as an early option because of its cost effectiveness and efficacy, in direct contrast to traditional practice, in which initiation is often inappropriately delayed. Pioglitazone and GLP-1 agonists are deemed to be less well validated “tier-2” options in the ADA-EASD algorithm for addition to metformin and lifestyle therapy, particularly for those in whom hypoglycemia is a major consideration and in the case of GLP-1 agonists when weight is of major concern. Alternatives to these preferred two tiers of drugs, including α-glucosidase inhibitors, glinides, pramlintide, DPP-IV inhibitors, and colesevelam are certainly effective glucose-lowering therapies, but for various reasons—including side effects, cost, modest A1C reduction, or lack of long-term safety data—they were not included in the ADA-EASD consensus statement. It should be noted that the ADA-EASD consensus panel does not recommend the use of rosiglitazone, glyburide, or chlorpropamide, owing to lingering safety concerns and the availability of alternatives with fewer concerns (pioglitazone, glimepiride, glipizide, gliclazide).

The guideline advocates the addition of insulin at any time, particularly if patients are symptomatic, severely hyperglycemic (A1C > 10%), or have ketonuria. Intermediate or long-acting basal insulin at bedtime, titrated to fasting glucose goals, is the preferred means of initiating insulin. If fasting blood glucose is in target range and a patient is not reaching A1C goals, the addition of premeal insulin or a trial of premixed intermediate- and short-acting insulin is advised. A1C levels should be obtained every 3 months during treatment, and if patients are not meeting targets, adjustment in therapy is indicated.

Guidelines from the American Association of Clinical Endocrinologists (AACE) also recommend early goal-directed therapy, but stress that the choice of agents should be individualized, particularly with respect to the risk of hypoglycemia.⁴⁴ In this guideline, initial monotherapy is preferred only for those patients with an A1C between 6% and 7%. If the initial A1C is between 7% and 8%, the guidelines recommend starting combination therapy. Insulin therapy should be added when the A1C exceeds 10% or in patients whose A1C exceeds 6.5% on two non-insulin drugs.

1. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care. 2009;32:S13–S61.

2. Engelgau, MM, Geiss, LS, Saaddine, JB, et al. The evolving diabetes burden in the United States. Ann Intern Med. 2004;140:945–950.

3. Wild, S, Roglic, G, Green, A, et al. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27:1047–1053.

4. Narayan, KM, Boyle, JP, Thompson, TJ, et al. Lifetime risk for diabetes mellitus in the United States. JAMA. 2003;290:1884–1890.

5. Bjork, S. The cost of diabetes and diabetes care. Diabetes Res Clin Pract. 2001;54(Suppl 1):S13–S18.

6. Gu, K, Cowie, CC, Harris, MI. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971–1993. Diabetes Care. 1998;21:1138–1145.

7. American Diabetes Association. Economic costs of diabetes in the US in 2007. Diabetes Care. 2008;31:596–615.

8. CDC Cost-Effectiveness Group. Cost-effectiveness of intensive glycemic control, intensified hypertension control and serum cholesterol level reduction for type 2 diabetes. JAMA. 2002;287:2542–2551.

9. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837–853.

10. UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352:854–865.

11. Ohkubo, Y, Kishikawa, H, Araki, E, et al. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract. 1995;28:103–117.

12. Action to Control Cardiovascular Risk in Diabetes Study GroupGerstein, HC, Miller, ME, Byington, RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545–2559.

13. ADVANCE Collaborative GroupPatel, A, MacMahon, S, Chalmers, J, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358:2560–2572.

14. Duckworth, W, Abraira, C, Moritz, T, et al. Intensive glucose control and complications in American veterans with type 2 diabetes. N Engl J Med. 2009;360:129–139.

15. Holman, RR, Paul, SK, Bethel, MA, et al. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359:1577–1589.

16. Ray, KK, Seshasai, SR, Wijesuriya, S, et al. Effect of intensive control of glucose on cardiovascular outcomes and death in patients with diabetes mellitus: a meta-analysis of randomised controlled trials. Lancet. 2009;373:1765–1772.

17. American College of Endocrinologists. American College of Endocrinology consensus statement on guidelines for glycemic control. Endocr Pract. 2002;8(Suppl 1):5–11.

18. Skyler, JS, Bergenstal, R, Bonow, RO, et alAmerican Diabetes AssociationAmerican College of Cardiology FoundationAmerican Heart Association. intensive glycemic control and the prevention of cardiovascular events: implications of the ACCORD, ADVANCE, and VA diabetes trials: a position statement of the American Diabetes Association and a scientific statement of the American College of Cardiology Foundation and the American Heart Association. Diabetes Care. 2009;32:187–192.

19. Rohlfing, CL, Wiedmeyer, HM, Little, RR, et al. Defining the relationship between plasma glucose and HbA_1c: analysis of glucose profiles and HbA_1c in the Diabetes Control and Complications Trial. Diabetes Care. 2002;25:275–278.

20. Klonoff, DC, Schwartz, DM. An economic analysis of interventions for diabetes. Diabetes Care. 2000;23:390–404.

21. Norris, SL, Engelgau, MM, Narayan, KM. Effectiveness of self-management training in type 2 diabetes: a systematic review of randomized controlled trials. Diabetes Care. 2001;24:561–587.

22. Funnell, MM, Brown, TL, Childs, BP, et al. National standards for diabetes self-management education. Diabetes Care. 2009 Jan;32(Suppl 1):S87–S94.

23. Fisher, EB, Earp, JA, Maman, S, et al. Cross-cultural and international adaptation of peer support for diabetes management. Fam Pract. 2009 Mar 10. [[Epub ahead of print]].

24. Delamater, AM, Jacobson, AM, Anderson, B, et al. Psychosocial therapies in diabetes: report of the Psychosocial Therapies Working Group. Diabetes Care. 2001;24:1286–1292.

25. American Diabetes AssociationBantle, JP, Wylie-Rosett, J, Albright, AL, et al. Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2008 Jan;31(Suppl 1):S61–S78.

26. Sheard, NF, Clark, NG, Brand-Miller, JC, et al. Dietary carbohydrate (amount and type) in the prevention and management of diabetes: a statement by the American Diabetes Association. Diabetes Care. 2004;27:2266–2271.

27. Klein, S, Sheard, NF, Pi-Sunyer, X, et al. Weight management through lifestyle modification for the prevention and management of type 2 diabetes: rationale and strategies: a statement of the American Diabetes Association, the North American Association for the Study of Obesity, and the American Society for Clinical Nutrition. Diabetes Care. 2004;27:2067–2073.

28. Gillespie, SJ, Kulkarni, KD, Daly, AE. Using carbohydrate counting in diabetes clinical practice. J Am Diet Assoc. 1998;98:897–905.

29. Bergenstal, RM, Johnson, M, Powers, MA, et al. Adjust to target in type 2 diabetes: comparison of a simple algorithm with carbohydrate counting for adjustment of mealtime insulin glulisine. Diabetes Care. 2008 Jul;31(7):1305–1310.

30. Egede, LE, Ye, K, Zhang, D, et al. The prevalence and pattern of complementary and alternative medicine use in individuals with diabetes. Diabetes Care. 2002;25:324–329.

31. Ernst, E. Complementary medicine: its hidden risks. Diabetes Care. 2001;24:1486–1488.

32. Barringer, TA, Kirk, JK, Santaniello, AC, et al. Effect of a multivitamin and mineral supplement on infection and quality of life. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 2003;138:365–371.

33. Schnyder, G, Roffi, M, Pin, R, et al. Decreased rate of coronary restenosis after lowering of plasma homocysteine levels. N Engl J Med. 2001;345:1593–1600.

34. Lange, H, Suryapranata, H, De Luca, G, et al. Folate therapy and in-stent restenosis after coronary stenting. N Engl J Med. 2004;350:2673–2681.

35. Ryan, DH, Espeland, MA, Foster, GD, et al. Look AHEAD (Action for Health in Diabetes): design and methods for a clinical trial of weight loss for the prevention of cardiovascular disease in type 2 diabetes. Control Clin Trials. 2003;24:610–628.

36. Boule, NG, Haddad, E, Kenny, GP, et al. Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. JAMA. 2001;286:1218–1227.

37. Sigal, RJ, Kenny, GP, Wasserman, DH, et al. Physical activity/exercise and type 2 diabetes. Diabetes Care. 2004;27:2518–2539.

38. Inzucchi, SE. Noninvasive assessment of the diabetic patient for coronary artery disease. Diabetes Care. 2001;24:1519–1521.

39. American Diabetes Association. Consensus development conference on the diagnosis of coronary heart disease in people with diabetes: 10–11 February 1998, Miami, Florida. Diabetes Care. 1998;21:1551–1559.

40. Young, LH, Wackers, FJ, Chyun, DA, et alDIAD Investigators. cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial. JAMA. 2009 Apr 15;301(15):1547–1555.

41. Towfigh, A, Romanova, M, Weinreb, JE, et al. Self-monitoring of blood glucose levels in patients with type 2 diabetes mellitus not taking insulin: a meta-analysis. Am J Manag Care. 2008 Jul;14(7):468–475.

42. Farmer, AJ, Wade, AN, French, DP, et alDiGEM Trial Group. blood glucose self-monitoring in type 2 diabetes: a randomised controlled trial. Health Technol Assess. 2009 Feb;13(15):iii–iv. [ix–xi, 1–50].

43. Nathan, DM, Buse, JB, Davidson, MB, et alAmerican Diabetes AssociationEuropean Association for Study of Diabetes. Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2009;32:193–203.

44. Rodbard, HW, Blonde, L, Braithwaite, SS, et alAACE Diabetes Mellitus Clinical Practice Guidelines Task Force. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the management of diabetes mellitus. Endocr Pract. 2007;13:1–68.

45. Setter, SM, Iltz, JL, Thams, J, et al. Metformin hydrochloride in the treatment of type 2 diabetes mellitus: a clinical review with a focus on dual therapy. Clin Ther. 2003;25:2991–3026.

46. Leverve, XM, Guigas, B, Detaille, D, et al. Mitochondrial metabolism and type-2 diabetes: a specific target of metformin. Diabetes Metab. 29, 2003. [6S88–6S94].

47. Misbin, RI. The phantom of lactic acidosis due to metformin in patients with diabetes. Diabetes Care. 2004;27:1791–1793.

48. Stades, AME, Heikens, JT, Erkelens, DW, et al. Metformin and lactic acidosis: cause or coincidence? A review of case reports. J Int Med. 2004;255:179–187.

49. Glucophage/Glucophage XR [package insert], Princeton, NJ, Bristol-Myers Squibb Company, January 2009. http://packageinserts.bms.com/pi/pi_glucophage.pdf. [accessed August 25, 2009].

50. Shaw, JS, Wilmot, RL, Kilpatrick, ES. Establishing pragmatic estimated GFR thresholds to guide metformin prescribing. Diabet Med. 2007 Oct;24(10):1160–1163.

51. Tahrani, AA, Varughese, GI, Scarpello, JH, et al. Metformin, heart failure, and lactic acidosis: is metformin absolutely contraindicated? BMJ. 2007 Sep 8;335(7618):508–512.

52. Inzucchi, SE, Masoudi, FA, McGuire, DK. Metformin in heart failure. Diabetes Care. 2007;30:e129.

53. Wulffele, MG, Kooy, A, de Zeeuw, D, et al. The effect of metformin on blood pressure, plasma cholesterol and triglycerides in type 2 diabetes mellitus: a systematic review. J Intern Med. 2004;256:1–14.

54. McGuire, DK, Inzucchi, SE. New drugs for the treatment of diabetes mellitus: part I: thiazolidinediones and their evolving cardiovascular implications. Circulation. 2008;117:440–449.

55. Ceriello, A. Thiazolidinediones as anti-inflammatory and anti-atherogenic agents. Diabetes Metab Res Rev. 2008;24:14–26.

56. ACTOS (package insert). Takeda Pharmaceuticals America. Lincolnshire, IL, December 2003. Available at http://www.actos.com/pi.pdf. [Accessed September 12, 2004].

57. Avandia (package insert). GlaxoSmithKline. Research Triangle Park, NC, August 2004. Available at http://us.gsk.com/products/assets/us_avandia.pdf. [Accessed September 12, 2004].

58. Kahn, SE, Haffner, SM, Heise, MA, et al. ADOPT Study Group. Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med. 2006;355:2427–2443.

59. DREAM (Diabetes REduction Assessment with ramipril and rosiglitazone Medication) Trial Investigators. effect of rosiglitazone on the frequency of diabetes in patients with impaired glucose tolerance or impaired fasting glucose: a randomised controlled trial. Lancet. 2006;368:1096–1105.

60. Dormandy, JA, Charbonnel, B, Eckland, DJ, et al. PROactive investigators. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366:1279–1289.

61. http://www.fda.gov/Cder/drug/InfoSheets/HCP/rosiglitazone200707HCP.htm [Site accessed 12/2/08].

62. Home, PD, Pocock, SJ, Beck-Nielsen, H, et alRECORD Study Team. rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes (RECORD): a multicentre, randomised, open-label trial. Lancet. 2009 Jun 20;373(9681):2125–2135.

63. Goldberg, RB, Kendall, DM, Deeg, MA, et al. A comparison of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes Care. 2005;28:1547–1554.

64. Juurlink, DN, Gomes, T, Lipscombe, LL, et al. Adverse cardiovascular events during treatment with pioglitazone and rosiglitazone: population based cohort study. BMJ. 2009 Aug 18;339:b2942.

65. Nesto, RW, Bell, D, Bonow, RO, et al. Thiazolidinedione use, fluid retention, and congestive heart failure. A consensus statement from the American Heart Association and American Diabetes Association. Circulation. 2003;108:2941–2948.

66. Karalliedde, J, Buckingham, R, Starkie, M, et al. Effect of various diuretic treatments on rosiglitazone-induced fluid retention. J Am Soc Nephrol. 2006;17:3482–3490.

67. Ryden, L, Thrainsdottir, I, Swedberg, K. Adjudication of serious heart failure in patients from PROactive. Lancet. 2007;369:189–190.

68. Grey, A. Skeletal consequences of thiazolidinedione therapy. Osteoporos Int. 2008;19:129–137.

69. Riddle, MC. Editorial: Sulfonylureas differ in effects on ischemic preconditioning—is it time to retire glyburide? J Clin Endocrinol Metab. 2003;88:528–530.

70. Del Prato, S, Pulizzi, N. The place of sulfonylureas in the therapy for type 2 diabetes mellitus. Metabolism. 2006;55:S20–S27.

71. Sawada, F, Inoguchi, T, Tsubouchi, H, et al. Differential effect of sulfonylureas on production of reactive oxygen species and apoptosis in cultured pancreatic beta-cell line, MIN6. Metabolism. 2008;57:1038–1045.

72. Blicklé, JF. Meglitinide analogues: a review of clinical data focused on recent trials. Diabetes Metab. 2006;32:113–120.

73. Manzella, D, Grella, R, Abbatecola, AM, et al. Repaglinide administration improves brachial reactivity in type 2 diabetic patients. Diabetes Care. 2005;28:366–371.

74. Esposito, K, Giugliano, D, Nappo, F, et al. Regression of carotid atherosclerosis by control of postprandial hyperglycemia in type 2 diabetes mellitus. Circulation. 2004;110:214–219.

75. Chiasson, JL, Josse, RG, Gomis, R, et al. Acarbose treatment and the risk of cardiovascular disease and hypertension in patients with impaired glucose tolerance: The STOP-NIDDM trial. JAMA. 2003;290:486–494.

76. Hanefeld, M, Chiasson, JL, Koehler, C, et al. Acarbose slows progression of intima-media thickness of the carotid arteries in subjects with impaired glucose tolerance. Stroke. 2004;35:1073–1078.

77. Riddle, MC, Drucker, DJ. Emerging therapies mimicking the effects of amylin and glucagon-like peptide 1. Diabetes Care. 2006;29:435–449.

78. Inzucchi, SE, McGuire, DK. New drugs for the treatment of diabetes: part II: incretin-based therapy and beyond. Circulation. 2008;117:574–584.

79. Drucker, DJ. Dipeptidyl peptidase-4 inhibition and the treatment of type 2 diabetes: preclinical biology and mechanisms of action. Diabetes Care. 2007;30:1335–1343.

80. Exenatide prescribing information. http://pi.lilly.com/us/byetta-pi.pdf. [Accessed 12/3/08].

81. Sitagliptin prescribing information. http://www.merck.com/product/usa/pi_circulars/j/januvia/januvia_pi.pdf. [Accessed 12/3/08].

81a. Saxagliptin prescribing information. http://packageinserts.bms.com/pi/pi_onglyza.pdf. [Accessed 8/25/2009].

82. Pramlintide prescribing information. https://www.symlin.com/pdf/SYMLIN-pi-combined.pdf. [Accessed 12/3/08].

83. Bergenstal, RM, Johnson, M, Powers, MA, et al. Adjust to target in type 2 diabetes: comparison of a simple algorithm to carbohydrate counting for adjustment of mealtime insulin glulisine. Diabetes Care. 2008;31:1305–1310.

84. Owens, DR, Bolli, GB. Beyond the era of NPH insulin—long-acting insulin analogs: chemistry, comparative pharmacology, and clinical application. Diabetes Tech Ther. 2008;10:333–349.

85. Klein, O, Lynge, J, Endahl, L, et al. Albumin-bound basal insulin analogues (insulin detemir and NN344): comparable time-action profiles but less variability than insulin glargine in type 2 diabetes. Diabetes Obes Metab. 2007;9:290–299.

86. Rosenstock, J, Davies, M, Home, PD, et al. A randomised, 52-week, treat-to-target trial comparing insulin detemir with insulin glargine when administered as add-on to glucose-lowering drugs in insulin-naive people with type 2 diabetes. Diabetologia. 2008;51:408–416.

87. Heise, T, Nosek, L, Ronn, BB, et al. Lower within-subject variability of insulin detemir in comparison to NPH insulin and insulin glargine in people with type 1 diabetes. Diabetes. 2004;53:1614–1620.

88. Riddle, MC, Rosenstock, J, Gerich, J. Insulin Glargine 4002 Study Investigators: The Treat-To-Target Trial: Randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care. 2003;26:3080–3086.

89. Gerstein, HC, Yale, JF, Harris, SB, et al. A randomized trial of adding insulin glargine vs. avoidance of insulin in people with type 2 diabetes on either no oral glucose-lowering agents or submaximal doses of metformin and/or sulphonylureas. The Canadian INSIGHT (Implementing New Strategies with Insulin Glargine for Hyperglycaemia Treatment) Study. Diabet Med. 2006 Jul;23(7):736–742.

90. Bretzel, RG, Nuber, U, Landgraf, W, et al. Once-daily basal insulin glargine versus thrice-daily prandial insulin lispro in people with type 2 diabetes on oral hypoglycaemic agents (APOLLO): an open randomised controlled trial. Lancet. 2008;371:1073–1084.

91. Holman, RR, Thorne, KI, Farmer, AJ, et al4-T Study Group. addition of biphasic, prandial, or basal insulin to oral therapy in type 2 diabetes. N Engl J Med. 2007;357:1716–1730.

92. Garber, AJ, Wahlen, J, Wahl, T, et al. Attainment of glycaemic goals in type 2 diabetes with once-, twice-, or thrice-daily dosing with biphasic insulin aspart 70/30 (The 1-2-3 study). Diabetes Obes Metab. 2006;8:58–66.

93. Robbins, DC, Beisswenger, PJ, Ceriello, A, et al. Mealtime 50/50 basal + prandial insulin analogue mixture with a basal insulin analogue, both plus metformin, in the achievement of target HbA_1c and pre- and postprandial blood glucose levels in patients with type 2 diabetes: a multinational, 24-week, randomized, open-label, parallel-group comparison. Clin Ther. 2007;29:2349–2364.

94. Monnier, L, Lapinski, H, Colette, C. Contributions of fasting and postprandial plasma glucose increments to the overall diurnal hyperglycemia of type 2 diabetic patients: variations with increasing levels of HbA(1c). Diabetes Care. 2003;26:881–885.

95. Rosenstock, J, Ahmann, AJ, Colon, G, et al. Advancing insulin therapy in type 2 diabetes previously treated with glargine plus oral agents: prandial premixed (insulin lispro protamine suspension/lispro) versus basal/bolus (glargine/lispro) therapy. Diabetes Care. 2008;31:20–25.

96. UKPDS Study Group. (UKPDS 16), Overview of six years’ therapy of type 2 diabetes—a progressive disease. Diabetes. 1995;44:1249–1258.

97. Del Prato, S, Wishner, WJ, Gromada, J, et al. Beta-cell mass plasticity in type 2 diabetes. Diabetes Obes Metab. 2004;6:319–531.

98. BARI 2D Study GroupFrye, RL, August, P, Brooks, MM, et al. A randomized trial of therapies for type 2 diabetes and coronary artery disease. N Engl J Med. 2009 Jun 11;360(24):2503–2515.

99. Ong, KL, Cheung, BM, Wong, LY, et al. Prevalence, treatment, and control of diagnosed diabetes in the U.S. National Health and Nutrition Examination Survey 1999–2004. Ann Epidemiol. 2008;18:222–229.

100. Brunton, S. Insulin delivery systems: reducing barriers to insulin therapy and advancing diabetes mellitus treatment. Am J Med. 2008;121:S35–S41.

101. Nathan, DM. Finding new treatments for diabetes: how many, how fast … how good? N Engl J Med. 2007;356:437–440.

102. Gaede, P, Lund-Andersen, H, Parving, H-H, et al. Effect of a multifactorial intervention on mortality in type 2 diabetes. N Engl J Med. 2008;358:580–591.

103. Narayan, KM, Gregg, EW, Engelgau, MM, et al. Translation research for chronic disease: the case of diabetes. Diabetes Care. 2000;23:1794–1798.