Diabetes mellitus is historically characterized by hyperglycemia. The pathophysiologic processes causing hyperglycemia include insulin deficiency, impaired glucose disposal (insulin resistance), and increased hepatic glucose production. Type 1 diabetes mellitus results from an insulin deficiency state usually caused by immunologic damage to beta cells. Some type 1 diabetic patients do have features of insulin resistance. Type 2 diabetes mellitus results from insulin resistance, often associated with central obesity, increased hepatic glucose production, and a progressive decline in beta cell function. This loss of beta cell function is not immunologically mediated. Secondary forms of diabetes may occur as a result of pancreatectomy (insulin-deficient state), administration of glucocorticoids (although glucocorticoid use may simply be “unmasking” a predisposition to develop diabetes), hemochromatosis, and rare syndromes such as antibodies to the insulin receptor. Gestational diabetes occurs during pregnancy as a result of glucose counterregulatory hormone production, and may also be more common in patients with a genetic predisposition to develop type 2 diabetes mellitus. This disease management document will be limited to the common forms of diabetes mellitus (DM), types 1 and 2 DM. Approximately 20 million people in the United States have DM and one third are not aware of their diagnosis.

The pathophysiologic processes by which hyperglycemia contributes to the complications of diabetes mellitus are not yet established. However, the following are considerations. Hyperglycemia is associated with the glycation of many proteins, including structural proteins. This may result in advanced glycation end products (AGEs), modified protein products that have been associated with many of the diabetes complications. Glycation of low-density lipoprotein (LDL) makes it more susceptible to oxidation. Lipid oxidation is one of the proposed mechanisms for atherosclerosis. Hyperglycemia increases sorbitol accumulation in tissues and has been invoked as a mechanism for neuropathy and retinopathy. Hyperglycemia increases the concentration of protein kinase C β (PKC β) in the retina, which in turn is associated with increased concentrations of vascular endothelial cell growth factor (VEGF). VEGF contributes to both the increased risk for proliferative changes in the eye and loss of endothelial cell integrity and associated risk for macular edema.

Diagnostic criteria

The diagnosis of diabetes is based on several findings. The following criteria have been established by the American Diabetes Association: ¹

1. Fasting glucose level higher than 126 mg/dL on two occasions. This fasting glucose value is based on data showing that blood glucose values above this fasting level are consistently associated with the risk for retinopathy, the diabetes complication essentially unique to diabetes. This cut point value will miss a number of patients who have diabetes based on oral glucose tolerance testing results. Because of simplicity, fasting blood glucose (BG) concentrations are one of the most common ways to diagnose DM. There are recent observational data suggesting that this threshold for the diagnosis DM may be too high, because patients with impaired glucose tolerance develop retinopathy.
2. Random or “casual” glucose higher than 200 mg/dL, with symptoms of DM. This is a common way to diagnose DM. Many patients may not have obvious symptoms, but that should not alter the fact that a random BG level in this range generally establishes the diagnosis of DM. This criterion is not affected by the time of the last meal.
3. Oral glucose tolerance test result after a 75-g oral glucose load, 2-hour value higher than 200 mg/dL.

The oral glucose tolerance test (OGTT) is not generally recommended in clinical practice. Such testing requires 3 days of high carbohydrate intake. False-positive test results may occur in the presence of recent fasting. Oral glucose tolerance testing is not always reproducible. Consequently, OGTT is generally considered a research tool, and often two OGTTs are necessary to establish a diagnosis of DM.

Hemoglobin A_1c (HgbA_1c) values are too insensitive to be used as a screening test for DM. Elevated values (e.g. higher than 6.2%) are usually associated with a diagnosis of DM, but patients may have a diagnosis of DM with values below this range. Thus, elevated HgbA_1c values are a somewhat specific test for the diagnosis of DM, but are not highly sensitive.

Back to Top

Treatment

Medical Nutrition Treatment

Guidelines for medical nutrition therapy ^2–7 have been established by the American Diabetes Association (ADA) and are summarized in Box 1. The primary focus of these guidelines is targeted to outcomes including glycemic control, weight reduction (as appropriate), blood pressure control, and a favorable lipid profile. There is clear evidence that excess saturated fat in the diet has a detrimental effect on lipid profiles and therefore saturated fat restriction is recommended. The data supporting absolute restriction of carbohydrates are not robust, so the ADA guidelines allow flexibility in carbohydrate and non–saturated fat intake. Separate guidelines have been published about the carbohydrate content and composition of the diet. ⁸ The most important variable in prandial glycemic excursion is total carbohydrate intake. Low glycemic index foods consumed alone result in lower prandial glucose excursion than high glycemic index foods. However, in the context of a mixed meal, differences between low and high glycemic index foods are attenuated. The guidelines support the concept that glycemic index may be a consideration in nutrition therapy. Both the amount ^8–10 and source ^10,11 of carbohydrates are important determinants of postprandial glucose. The relative effects of each have been recently studied. Brand-Miller and colleagues ^12,13 have reported that they analyzed the relative impact of the glycemic index and total carbohydrate content of individual foods on glycemic load—the product of glycemic index and total grams of carbohydrate—using linear regression analysis. Carbohydrate content (total grams) alone explained 68% of the variation in glycemic load, and the glycemic index of the food explained 49%. When total carbohydrate and glycemic index were both included in the regression analysis, the glycemic index accounted for 32% of the variation.

Alcohol and sodium restriction is generally advised. Supplements are not necessary in patients who are otherwise consuming a well-balanced diet. Many recommendations for weight management propose caloric restriction based on the degree of obesity as well as 30 to 45 minutes of exercise 3 to 5 days a week. Exercise is an important component of any weight reduction–glycemic control regimen (see later, “Exercise”). Other nutritional guidelines for patients with diabetes are generally consistent with the ADA guidelines. ^5,14–28

Exercise

Guidelines for exercise have not always been specific with regard to exact exercise prescriptions, especially regarding aerobic and resistance exercises. ^14,29,30 The frequently proposed recommendation that 150 minutes of moderate-intensity (or 90 minutes of vigorous) aerobic exercise a week is generally the amount of exercise required to achieve benefits on glycemic control and reduce coronary heart disease (CHD) risk. This amount of exercise has recently been supported by ADA/American Heart Association (AHA) recommendations. ^6,7 Regular exercise is encouraged, with the caveat that complications of diabetes need to be taken into account. Injury to patients with loss of sensation in their feet is a limitation for weight-bearing exercise. Because of risk of CHD in patients with diabetes, appropriate screening for CHD should be performed before patients engage in any rigorous exercise program. ^{14,17,29–32} Benefits of exercise include weight control and improved glycemic control, often due to improvement in insulin resistance.

Back to Top

Complications

The complications of diabetes mellitus include retinopathy, nephrop-athy, neuropathy, and increased risk for atherosclerotic vascular disease. Diabetes mellitus is the leading cause of blindness in young people and is comparable with macular degeneration as a cause of blindness in older adults. Diabetes mellitus is the leading cause of end-stage renal disease requiring renal replacement therapy, dialysis, or transplantation. DM is the leading cause of nontraumatic amputations of the lower extremity, a result of peripheral neuropathy and peripheral vascular disease. DM is associated with a two- to fivefold increased risk for CHD. ³³

Two large trials—the Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS)—have demonstrated that there is a clear relation between glucose control and the risks for retinopathy (onset and progression), nephropathy (as measured by albuminuria), and neuropathy (both clinical and electromyographic measures. ^34–38 The relation between the degree of hyperglycemia and CHD risk has also been established, ^39,40 but this increased CHD risk begins well below the glycemic threshold for the diagnosis of diabetes. ^{23,25–28,41,42}

Screening

Screening should be done for diabetic complications ( Table 1 ).

Diabetic Retinopathy

Dilated eye examinations by ophthalmologists or optometrists should be performed within 5 years of onset in type 1 diabetes and at the time of diagnosis in type 2 diabetes, because the actual date of onset is unknown in type 2 DM. ^14,29 Follow-up examinations should be performed annually in patients with no or minimal background retinopathy. More frequent follow-up examinations should be performed in patients who have more advanced retinopathy. Handheld ophthalmoscopy in the office may be able to detect diabetic retinopathy because of the limited view of the retina and difficulty detecting diabetic macular edema. Macular edema is a significant cause of visual loss in diabetes mellitus. Detection of macular edema is easier to detect with binocular vision and, in difficult cases, IV fluorescein angiography and confocal microscopy are commonly used by ophthalmologists. Technology for screening with fundus photographs obtained in the physician's office and then read by an experienced reader is available. However, these methods are not yet sufficiently well standardized to use them as a routine screening tool.

Diabetic Nephropathy

The hallmark of early diabetic nephropathy is albumin excretion. Sensitive assays to detect very low levels of albumin, or microalbuminuria, have been available for many years. ^14,29 The simplest screening measure is a spot urine test adjusted for the urine creatinine level. Timed overnight collections and 24-hour collections may also be used. In general, microalbuminuria is defined as more than 30 mg albumin/g of creatinine (spot urines) or 30 to 299 mg/24 hours and more than 300 mg/g creatinine (or 24 hours) as albuminuria. Serum creatinine determinations should be performed at least annually in patients with albuminuria; when estimated glomerular filtration rate (GFR) values are declining, more specific measures of GFR (most commonly, creatinine clearance) should be used.

Peripheral Neuropathy

Monofilament testing in the office is the easiest way to test for the insensate foot. ^43–50 The 5.07-mm monofilament should be applied to the bottom of the feet ⁵⁰ (Fig. 1). Any loss of sensation is associated with an increased risk for ulcer formation. Any patient who has had a foot ulcer is at increased risk for further foot ulcers.

Patients should be instructed to look at their feet daily. Patients who have difficulty looking at their feet should have someone else look at their feet, especially if the patient has visual impairment, or be encouraged to use a mirror, such as a magnifying shaving mirror, if they have trouble seeing the bottoms of their feet (see elsewhere in this section, “Prevention and Treatment of Leg and Foot Ulcers in Diabetes Mellitus”).

Coronary Heart Disease

Careful questioning about symptoms of ischemic coronary disease is still one of the most important ways to screen for symptomatic disease. Many patients with diabetes do not have typical chest pain as the manifestation of angina. Consequently, clinicians must ask about reduced exercise tolerance, dyspnea, or exercise-induced nausea.

Various groups and individuals have considered the issue of screening for CHD. ^{14,17,29,51–54} These guidelines and individual recommendations are not entirely concordant. Whereas nearly every group suggests stress tests for patients with symptoms of CHD or electrocardiographic changes suggesting ischemia, recommendations on screening for asymptomatic disease are less consistent. The ADA considers that candidates for a screening cardiac stress test should include those with the following: (1) a history of peripheral or carotid occlusive disease; (2) sedentary lifestyle, age older than 55 years, and plans to begin a vigorous exercise program; and (3) two or more of the risk factors noted above. ²⁹ The American Association of Clinical Endocrinologists (AACE) guidelines state that:

Screening for asymptomatic coronary artery disease is an important consideration in patients with diabetes. An appropriate protocol for such screening has not been adequately tested. Increasing age, gender, cardiovascular risk factors, microalbuminuria, and retinopathy may identify high-risk groups for whom such testing is indicated. ¹⁴

The AHA consensus group has provided a thoughtful approach to screening for CHD in patients with diabetes. They noted that:

Screening is defined here as the detection of disease in asymptomatic persons. Because screening tests are intended for widespread application, they should be rapid and inexpensive. In addition, to be useful, the results of testing should lead to a change in management, and the results of testing should improve outcome. ¹⁷

The American College of Cardiology (ACC)/AHA Guidelines for Exercise Testing give screening by exercise treadmill testing in patients with diabetes a data quality rating of IIb—that is, its usefulness or efficacy is less well established by evidence or opinion. ⁵⁴ They add that exercise testing “might be useful in people with heightened pretest risk.” Most consensus statements and guidelines on diabetes and CHD have suggested that noninvasive cardiac testing be performed in patients with diabetes and one additional criterion—peripheral arterial disease, cerebrovascular disease, rest changes on the electrocardiogram (ECG), or the presence of two or more major coronary vascular disease (CVD) risk factors. According to these guidelines, risk assessment begins with a medical history, including special attention to symptoms of atherosclerotic disease, such as angina, claudication, or erectile dysfunction. Electrocardiographic changes showing left ventricular hypertrophy and ST-T changes suggest increased cardiovascular risk. Of interest, the ongoing DIAD study, which is designed to determine risk factors associated with clinically silent myocardial disease using stress tests with cardiac imaging, has suggested that the presence of neuropathy may be one of the most important predictors of cardiovascular risk. It is not yet clear exactly how noninvasive testing changes risk management strategies in diabetes mellitus, although, because DM is already considered “coronary heart disease risk equivalent.” Thus, noninvasive testing should be targeted as much as possible to detect patients who may have CHD amenable to surgical intervention. Whereas noninvasive screening in asymptomatic patients may detect disease amenable to percutaneous intervention or coronary artery bypass grafting, the cost-effectiveness and determination of how much such screening affects long-term outcomes are still uncertain.

I believe that careful attention to history of changes in exercise tolerance, atypical symptoms that suggest angina, or suggestive electrocardiographic abnormalities are reasons to consider stress testing. In addition to dyslipidemia, obesity, and hypertension, albuminuria and a family history of CHD may be reasons to consider stress testing in patients who do not have clinical symptoms of CHD. This approach is most consistent with the AACE guidelines and should select those patients at highest risk for CHD. In the absence of robust evidence, as noted by the AHA, physicians still need to make decisions about patients who may have silent myocardial disease. In the near-future, newer imaging techniques such as computed tomography (CT) angiography may help characterize patients at risk for an acute event, even in the absence of symptoms.

Treatment of Complications

Retinopathy

Observational studies have demonstrated a relation between the risk for retinopathy and integrated measures of glycemic control (e.g., HgbA_1c level). As noted earlier, two large clinical trials have demonstrated that diabetic subjects randomized to more intensive glucose control have a reduced risk for new-onset retinopathy or progression of established retinopathy. ^34–38,55 In the Diabetes Control and Complications Trial (DCCT), type 1 diabetic patients without diabetic retinopathy at baseline in the intensive control group (mean in-trial HgbA_1c, approximately 7.0%) had a 76% reduction in risk for new-onset retinopathy compared with the conventional control group (mean in-trial HgbA_1c, approximately 9%). Similarly, the progression of retinopathy was much less likely to occur in the intensive group (54% reduction) compared with the conventional group. Analyses of in-trial HgbA_1c levels as a function of length of time in the study confirmed the concept that higher HgbA_1c concentrations and disease duration are associated with greater risk for retinopathy. The United Kingdom Prospective Diabetes Study (UKPDS) evaluated the effects of glycemic control in type 2 diabetic patients. The intensive policy group had a mean in-trial HgbA_1c value of approximately 7%, whereas the conventional policy group had a mean HgbA_1c value of just below 8%. The 0.9% delta HgbA_1c value in the UKPDS resulted in a 21% reduction in diabetic retinopathy.

The observational extension to the DCCT is called the Epidemiology of Diabetes Intervention and Complications (EDIC) study. Results from the EDIC after 7 years of follow-up have demonstrated that the favorable effects of intensive glucose-lowering therapy are durable over time; this observation has been characterized as “metabolic memory” for intensive glycemic control. ^56,57 After the DCCT, the mean HgbA_1c level difference in the two groups diminished to 0.4% by 1 year. Five years after DCCT closure, the mean HgbA_1c value was 8.1% in the original intensive group and 8.2% in the original conventional group. Nevertheless, there is a substantial reduction in the risk for progression of retinopathy in the original group that was treated more intensively.

Based on these observations, guidelines for the HgbA_1c target level have been set at 7.0% by the ADA ²⁹ and AHA ^6,7 and at 6.5% by the AACE. ¹⁴ The data support a target of 7%, but clear demonstration of benefit on retinopathy has not yet been established.

The relation of hypertension to retinopathy onset and progression has been less consistently established, but I believe that the best evidence available supports this concept. ^55,58–60 The best prospective data come from the UKPDS blood pressure arms. ^58,60–62 In the patients whose blood pressure was treated more aggressively, at 7.5 years there was a 34% reduction in the risk for retinopathy as determined by two-step Early Treatment Diabetic Retinopathy Study (ETDRS) progression and similar changes for retinopathy, as determined by number of aneurysms or cotton wool exudates. The blood pressure (BP) in this trial did not reach the currently proposed guidelines for a value of 130/80 mm Hg. However, this BP target should be associated with some reduction in risk for the progression of retinopathy.

There are observational studies that have suggested a relation between dyslipidemia and the risk for retinopathy, especially retinal hard exudates. ^63–65 However, no large intervention trial has yet evaluated or reported any effects of treating the dyslipidemia on diabetic retinopathy. Thus, guidelines for lipid management are based on CHD risk (see later).

For patients who have established diabetic retinopathy, the use of laser photocoagulation therapy has documented efficacy. ^66–69 The Diabetic Retinopathy Study (DRS) has evaluated the effect of scatter laser therapy in patients with proliferative diabetic retinopathy. Eyes treated with argon or xenon laser have a reduced rate of progression of retinopathy and associated visual loss. The ETDRS has evaluated patients at an earlier stage of their disease, nonproliferative retinopathy with or without macular edema. Laser therapy was associated with a reduced risk for visual loss in patients with clinically significant macular edema.

Thus, these strategies have become the standard of care for patients with diabetes. More recent approaches include intraocular glucocorticoids injections (for macular edema), anti-VEGF injections. Oral protein kinase C inhibitors are under investigation, but not yet available for clinical use.

Nephropathy

There are three major strategies to reduce the risk for onset and progression of diabetic nephropathy. These include glycemic con-trol, management of elevated blood pressure, and modification of the renin-angiotensin-aldosterone system (RAAS) with angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs). A summary of these interventions and the associated guidelines will be reviewed. ^{6,7,14,29,70,71}

The DCCT has analyzed the effects of glycemic control on the risk for new onset of microalbuminuria and progression from microalbuminuria to albuminuria. The group whose glucose levels were more intensively treated had a reduction in the risk for new-onset microalbuminuria of 39% and progression to albuminuria of 54%. ³⁴ The DCCT-EDIC have also demonstrated that the effects of intensive glycemic control on the risk for albuminuria are durable during the 7 years of follow-up. ^56,57 New-onset microalbuminuria was reduced by 50% during the EDIC follow-up in the DCCT intensive treatment group and new onset of clinical albuminuria was reduced by 87%. In the UKPDS, the 0.9% delta HgbA_1c level resulted in a 33% reduction in albumin excretion. ³⁸ Thus, the HgbA_1c guidelines established for retinopathy are also applicable for the management of nephropathy risk and progression.

The early studies of the relation of increasing blood pressure, increasing albumin excretion, and declining GFR published by Parving and colleagues ⁷² have demonstrated the importance of blood pressure reduction in reducing albumin excretion rates and attenuating the decline in GFR. Subsequently, modulating the RAAS with ACE inhibitors and ARBs has demonstrated favorable effects on measures of diabetic nephropathy independent of blood pressure lowering. ^73–81 Specifically, the Lewis study in type 1 diabetes compared captopril with other blood pressure–lowering therapies (excluding calcium channel blockers) in subjects who had moderately elevated albumin excretion and normal or mildly abnormal renal function. There was a marked reduction in doubling of serum creatinine levels (the primary end point) as well as progression to end-stage renal disease or death. These observations were affirmed in other ACE inhibitor studies. These include studies by Ravid and associates ^78,79 that indicated decline and loss of renal function and favorable effects on albuminuria using enalapril in type 2 diabetic subjects. Based on these and other studies, Kasiske and coworkers ⁸² have analyzed the effects of both BP and ACE inhibition and reported that both effects contribute about equally to slowing the progression of loss of renal function. For each 10-mm Hg reduction in BP, there was a 3.70-mL/min relative increase in the GFR; an additional favorable effect of ACE inhibitors was a 3.41-mg/min relative increase in the GFR.

One of the largest studies evaluating the effects of ACE inhibition on albuminuria was the Micro-HOPE study with 3498 diabetic subjects randomized to ramipril (vs. placebo). ⁸³ By one year, the subjects randomized to ramipril were found to have a diminished albumin excretion rate compared with those patients randomized to placebo. More recently, three major trials using ARBs (vs. placebo) in a wide variety of type 2 diabetic patients have shown 16% to 68% reductions in renal disease progression, as measured by albuminuria. ^{29,73,77,84,85} One of these studies used two different doses of irbesartan (150 and 300 mg daily) and demonstrated a dose-response effect. Compared with placebo, the 150-mg dose showed a 44% reduction in albuminuria and the 300-mg dose showed a 68% reduction in albuminuria when compared with placebo. Therefore, the ADA, ²⁹ AACE, ¹⁴ National Kidney Foundation, ⁷¹ and Joint National Committee (JNC) VII ⁷⁰ guidelines have suggested that to reduce the risk and/or slow the progression of nephropathy, the initial step is to optimize blood pressure control. The treatment of both micro- and macroalbuminuria includes ACE inhibitors or ARBs, except during pregnancy. In the absence of head-to-head clinical trials, ACE inhibitors or ARBS may be considered appropriate therapy. Although the guidelines do not specify serum creatinine and potassium levels at which ACE inhibitor or ARB therapy should be discontinued, creatinine and potassium levels need to be monitored after starting therapy and appropriate adjustments should be made for laboratory abnormalities.

Strippoli and colleagues ⁸¹ systematic review of ACE inhibitors and ARBs has summarized the effects of these agents on mortality and renal outcomes from 43 published trials. ⁸¹ They concluded that there are favorable effects from both classes of agents on measures of renal disease, but that ACE inhibitor trials are more likely to show a reduction in mortality. ACE inhibitor therapy resulted in the following relative risks (RRs): doubling of creatinine level, RR = 0.60 (95% confidence interval [CI], 0.34 to 1.05), end-stage renal disease, RR = 0.64 (95% CI, 0.40 to 1.03); progression microalbuminuria to macroalbuminuria, RR = 0.45 (95% CI, 0.28 to 0.71); and all-cause mortality, RR = 0.78 (95% CI, 0.63 to 0.99). The point estimates for relative risk in renal disease end points were similar for the ARBs, with the following relative risks: doubling of creatinine, RR = 0.79 (95% CI, 0.67 to 0.91); end-stage renal disease, RR = 0.78 (95% CI, 0.67 to 0.91); and progression of microalbuminuria to macroalbuminuria, RR = 0.49 (95% CI, 0.32 to 0.75). However, the point estimate for all-cause mortality did not favor ARB therapy—RR = 0.78 (95% CI, 0.63 to 0.99).

There are tantalizing observational data that show relationships between dyslipidemia and the risk for diabetic nephropathy. ^86–89 In subjects from the ETDRS, elevated lipid levels at baseline were associated with a future risk for the need for renal replacement therapy. Post hoc analyses from two large lipid-lowering trials have suggested a favorable effect of statins on renal function. ^90,91 However, effects on diabetic nephropathy are not part of lipid-lowering guidelines because of the absence of robust, prospective, intervention trial data.

Neuropathy

Treatment of the insensate foot requires regular observation of the feet of the diabetic patient. This includes a foot examination by health care providers at every regular visit and regular self-inspection by the patient. ^14,29,45 Patients who have evidence of callus formation or foot deformities (e.g., claw toe, Charcot's arthropathy), should be referred for appropriate footwear, usually shoes with adequate width, adequate depth and often with prescription orthotics (see elsewhere in this section, “Prevention and Treatment of Leg and Foot Ulcers in Diabetes Mellitus”). Patients with severe foot deformities may need custom-made shoes. Patients need to be reminded to stay in their prescribed footwear, because the risk for neurotrophic ulcers is high. Painful neuropathy has a wide variety of treatments, including analgesics, antidepressants, antiseizure medications, and topical capsaicin. ^45,48,92 There are no specific guidelines for the management of painful neuropathy.

Coronary Heart Disease

Dyslipidemia

Guidelines for the management of dyslipidemia have been published by the National Cholesterol Education Program, NCEP (several expert panels since 1988), AACE, ACP, ADA, ACC, and AHA ( Table 2 ). They are generally consistent in recommending aggressive lipid-lowering management for diabetic patients who are consider coronary risk equivalent.*

Physicians should note that not all patients with diabetes have a 20% risk of a cardiac event over a 10-year period as determined by the UKPDS risk engine, ³⁹ so some discretion with the guidelines may be considered. The proposed LDL cholesterol level targets are as follows:

1. The LDL cholesterol level is lower than 100 mg/dL for any patient with diabetes mellitus who is CHD risk equivalent.
2. If the LDL cholesterol level is below 100 mg/dL, but triglyceride (and very LDL [VLDL] cholesterol) levels are elevated, then the non–high-density lipoprotein (HDL) cholesterol level should be lower than 130 mg/dL.
3. The optional guidelines for very high-risk patients, such as diabetic patients with a prior myocardial infarction (MI) are an LDL cholesterol level lower than 70 mg/dL (and non-HDL cholesterol level lower than 100 mg/dL).
4. Patients who have an LDC cholesterol level lower than 100 mg/dL without medication should be treated to achieve a more than 30% reduction in their LDL cholesterol level.

These guidelines were developed based on findings from lipid-lowering trials that included diabetic patients and were confirmed by subsequent trials. A brief review of these supporting trials follows.

Post hoc analyses of diabetic patients that were included in lipid-lowering trials have supported the notion that diabetic patients have comparable relative reductions (or perhaps greater absolute reductions) in the risk for CHD events than their nondiabetic counterparts. These data have been summarized as part of the ACP guidelines. ²⁰ Furthermore, the ADA and AHA guidelines ^6,7,29,96,97 have suggested an LDL cholesterol level target of less than 100 mg/dL for patients with diabetes and an optional target of less than 70 mg/dL for patients with DM who already have CHD based on several clinical trials including, the HPS, ASCOT-LLA, and CARDS trials. ^99–102 The CARDS trial, in 2838 type 2 diabetic patients, showed a 37% reduction in cardiovascular events, with a mean in-trial LDL cholesterol level of approximately 80 mg/dL in the atorvastatin group compared with 80 mg/dL in the placebo group. Thus, both the HPS and CARDS studies have shown favorable effects in diabetic patients whose LDL cholesterol levels were lower than 100 mg/dL. In addition, these guidelines have recommended that in patients with elevated triglyceride levels, and a corresponding increase in VLDL cholesterol levels, that the non-HDL cholesterol value (LDL plus VLDL cholesterol level) be set at 30 mg/dL higher than the LDL target—that is, a non-HDL cholesterol level lower than 130 mg/dL, with an optional target of less than 100 mg/dL.

Since these guidelines were written, two major trials have been reported whose results are less convincing about the benefits of the lower LDL cholesterol level target and the effects of triglyceride level reduction. The ASPEN trial studied 2410 type 2 diabetic subjects who were randomized to atorvastatin, 10 mg, versus placebo. ¹⁰³ At baseline, LDL cholesterol levels were 113 mg/dL (atorvastatin group) and 114 mg/dL (placebo group). The atorvastatin group had a mean in-treatment LDL cholesterol level of 79 mg/dL, whereas the placebo group was essentially unchanged (LDL cholesterol level, 113 mg/dL). This difference was associated with a 10% reduction in the primary composite end point (P = not statistically significant, NSS) and a 27% reduction in fatal-nonfatal MI (P = NSS). Although these results were not as robust as the similarly designed CARDS study, the authors concluded that the “present data do not detract from the imperative that the majority of diabetic patients, especially those with existing CHD, are at risk of CHD and deserve LDL cholesterol lowering to the currently recommended targets.” ¹⁰³

The FIELD trial was designed to assess the effects of fenofibrate (vs. placebo) on cardiovascular disease events in 9795 type 2 diabetic subjects. ^104,105 The difference in total cardiovascular events was 11% (P = .035) and in MI plus CHD death 11% (P = .16). This trial was confounded by very high levels of statin drop-in, especially in the placebo arm. The ACCORD trial results should help clarify the effects of fibrates (http://www.ACCORDTRIAL.com). In the lipid-lowering arm of this trial, the participants were all provided simvastatin therapy and then randomized to fenofibrate versus placebo. Results should be available by 2010.

Hypertension and Renin-Angiotensin-Aldosterone System Blockade

Blood pressure control has a greater effect on reducing the risk for stroke than the risk for MI. The largest blood pressure trials in diabetic patients have demonstrated favorable effects on reduction in CVD. Current clinical guidelines recommend BP targets of 130/80 (or 130/85) mm Hg. ^{14,29,70,104,105} Few clinical trials have actually achieved these goals, but there does not appear to be any risk in reaching these targets. Multiple drug regimens (often three or more) are usually required. Based on several studies, especially HOPE, the EUROPA study, which demonstrated favorable cardiovascular effects with the use of the ACE inhibitors ramipril and perindopril, respectively, in diabetic cohorts, these agents should be considered part of initial therapy in hypertensive type 2 diabetic subjects. ^81,106–108 The beneficial effects could not be entirely attributed to blood pressure reduction in these trials.

Aspirin

Aspirin (ASA) therapy is recommended for patients with diabetes in the ADA and other guidelines. ^29,84 There are few data to suggest benefits for patients without established CHD. Because most patients with type 2 DM are at increased risk for CHD compared with their nondiabetic counterparts, aspirin use in high-risk patients is prudent. The guidelines originally recommended 81 to 325 mg daily because there are no outcomes data comparing ASA doses. Recent ADA/AHA guidelines recommend 75 to 162 mg daily. ^6,7 The concept of aspirin resistance is evolving and some patients with diabetes may be aspirin-resistant. If this observation is supported by future studies, then doses above 81 mg may be prudent for diabetic patients.

Smoking

All the DM- and CHD-related guidelines recommend smoking cessation.

Glycemic Control

Intervention trials have shown a somewhat modest relation between glycemic control and CHD risk. In the UKPDS trial, a delta HgbA_1c value of 0.9% was associated with a 14% reduction in the risk for MI (P = .052) in the intention to treat analyses and a 16% reduction for every 1% HgbA_1c level change as a post hoc observational analysis. ^36,38,109 The metformin arm in obese patients in the UKPDS demonstrated a 39% reduction in MI compared with the conventional arm (P = .010). ³⁷

In the DCCT/EDIC study, there was no statistically significant reduction in CHD risk at the end of the DCCT—this was expected, because the trial included a population at low risk for CHD at randomization—but a 42% reduction (P = .016) in risk of any cardiac event during the duration of the DCCT/EDIC study, 20 years in subjects randomized early in the trial. ¹¹⁰ Thus, the annualized effect of glycemic control on CHD risk is less than that generally associated with other interventions, especially lipid lowering. From the trials underway, it appears that the ACCORD trial (http://www.accordtrial.org) has the greatest likelihood of demonstrating the effects of intensive glycemic control on CHD risk.

Thus, to summarize, the watch word of the American Diabetes Association several years ago was “diabetes is serious.” Careful screening for complications, including retinopathy, nephropathy, and neuropathy clearly are associated with opportunities to reduce the risk for disease progression. Aggressive interventions with glycemic control, as well as lipid and blood pressure management, seem to have favorable effects on many complications of diabetes. Aspirin therapy also reduces the risk for CHD risk in patients with DM. These screening and intervention strategies are supported by robust observational and intervention trial data and, in turn, endorsed by the various organizations that have written disease management guidelines.

Summary

• Diabetes mellitus is a leading cause of blindness, end-stage renal disease, and nontraumatic lower extremity amputations.
• Diabetes mellitus increases the risk for coronary heart disease by two- to fivefold.
• Glycemic control is associated with a reduced risk for the microvascular and neuropathic complications of diabetes mellitus.
• Treatment of CHD risk factors, especially dyslipidemia, is associated with a reduced risk for CHD.
• Early detection of microvascular and neuropathic complications and implementation of appropriate treatment strategies, such as laser therapy (retinopathy), use of ACE inhibitors and ARBs (nephropathy), and proper footwear (neuropathy), will reduce the risk for adverse outcomes from these complications.

Back to Top

Suggested Readings

• Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): Prospective observational study. BMJ. 321: 2000; 412-419.
• Standards of medical care in diabetes—2007. Diabetes Care. 30: 2007; S4-S41.
• Nutrition recommendations and interventions for diabetes—2006: A position statement of the American Diabetes Association. Diabetes Care. 29: 2006; 2140-2157.
• Primary prevention of cardiovascular diseases in people with diabetes mellitus: A scientific statement from the American Heart Association and the American Diabetes Association. Diabetes Care. 30: 2007; 162-172.
• The relationship of glycemic exposure (HbA_1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes. 44: 1995; 968-983.
• Retinopathy and nephropathy in patients with type 1 diabetes four years after a trial of intensive therapy. N Engl J Med. 342: 2000; 381-389.
• ACC/AHA guidelines for exercise testing: Executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Exercise Testing). Circulation. 96: 1997; 345-354.
• Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Arterioscler Thromb Vasc Biol. 24: 2004; e149-e161.
• Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. BMJ. 321: 2000; 405-412.
• Effects of angiotensin-converting enzyme inhibitors and angiotensin II receptor antagonists on mortality and renal outcomes in diabetic nephropathy: Systematic review. BMJ. 329: 2004; 828.