点击显示 收起
美国糖尿病学会第65届科学年会
65th Scientific Sessions of the American Diabetes Association
2005年6月10－14日
美国加利福尼亚州圣地亚哥
June 10 - 14, 2005, San Diego, California
The Broadening Domain of the Metabolic Syndrome
Anne Peters, MD
Introduction
The definition of the metabolic syndrome has been evolving over the past 10 years. When troglitazone, the first insulin sensitizer, became available in 1996, the era of discussing insulin resistance and its multiple metabolic abnormalities was born, and it became increasingly clear that the treatment of diabetes required reducing the risk for cardiovascular disease (CVD) as well as lowering glucose. In addition, stemming from Gerald Reaven's, MD,[1,2] work in the 1980s and 1990s, it has been recognized that insulin resistance and its associated high risk for CVD precede the diagnosis of diabetes by 10-20 years.[3] Many researchers[4-10] have shown that increased insulin resistance and the metabolic syndrome increase the risk for CVD and the development of type 2 diabetes.
Naming and characterizing this syndrome has been the subject of many symposia and committees around the world. The first published definition was from the World Health Organization (WHO)[11] in 1998, with its related European Group for the Study of Insulin Resistance (EGIR) guidelines.[12] In 2001, at the urging of the American Association of Clinical Endocrinologists (AACE), a new International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code (277.7) for the "dysmetabolic syndrome" was approved.[13] The diagnostic criteria and operational definition are fairly broad, and include most of the key features of the syndrome. However, some are hard to measure in routine clinical settings, such as hypercoagulability and vascular endothelial dysfunction, which are clearly features of the syndrome but generally not measured in practice. The ICD-9-CM, however, allows clinicians to code for the metabolic syndrome, which should become increasingly important as more treatments become available.
The National Cholesterol Education Project Adult Treatment Panel (NCEP-ATP III) included a widely used definition of the metabolic syndrome in the 2001 guidelines.[14] In addition, the American College of Endocrinology (ACE) has published a position statement in collaboration with AACE on "insulin resistance" (their preferred term), which uses body mass index (BMI) rather than waist circumference to measure central obesity, introduces ethnicity as a risk factor, and emphasizes that diagnosis should be based on clinical judgment informed by the evaluation of risk factors.[15] The most recent definition of the metabolic syndrome was released by the International Diabetes Federation at an April 2005 conference in Berlin, Germany.[16] The Table compares the definitions of the WHO, NCEP-ATP III, and the International Diabetes Federation (IDF). In all cases except the IDF definition, the original definitions were written before the American Diabetes Association (ADA) lowered the fasting glucose cutpoint for normal glucose tolerance from 110 mg/dL to 100 mg/dL, so many investigators modify the original definitions to use the lower fasting glucose level.
WHO = World Health Organization; NCEP-ATP III = National Cholesterol Education Project Adult Treatment Panel; IDF = International Diabetes Federation; BP = blood pressure; HDL = high-density lipoprotein; BMI = body mass index; IGT = impaired glucose tolerance
*Adapted from: Kendall DM, Harmel AP. The metabolic syndrome, type 2 diabetes, and cardiovascular disease. Understanding the role of insulin resistance. Am J Manag Care. 2002;8:S635-S653.
Although debate exists as to which definition of the metabolic syndrome is the best, in clinical practice it is important to assess risk for CVD and diabetes in all patients. If a patient is overweight and/or has a family history of type 2 diabetes or CVD, obtaining a fasting glucose level, lipid panel, and blood pressure measurement can be helpful in guiding treatment. And because first-line treatment is based on principles of lifestyle modification, it rarely hurts to have people exercise (as long as they don't have a medical contraindication to doing so), eat well, and stop smoking.
Much of the news at the recent ADA meeting addressed treatment of the metabolic syndrome, which is discussed separately by David Kendall, MD. The most interesting studies on the metabolic syndrome itself involved characterizing the syndrome to fit an ever larger group of individuals, including children, women, and people with type 1 diabetes. Papers were also presented comparing features of the syndrome with outcomes, such as coronary artery disease (CAD) or CVD events, in an attempt to define which components are most useful.
The Nateglinide And Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR) trial is a multinational, randomized, placebo-controlled, prospective study to determine whether CVD or progression to type 2 diabetes can be reduced or delayed in subjects with impaired glucose tolerance by the administration of valsartan and/or nateglinide.[17] In the screening phase for the study, 43,509 subjects who were 50 years of age or older and had a history of CVD or were at increased risk for CVD underwent an oral glucose tolerance test (OGTT). Of the 8853 who had known CVD, the prevalence of the metabolic syndrome (based on the NCEP criteria) was calculated to be 51%. When divided by region (North America, Latin America, Europe, and Asia), rates of the metabolic syndrome were lowest in Asians (36%), suggesting that the current criteria may need modification in this population.
A team of researchers from Finland[18] used their existing database to study the validity of the various definitions of the metabolic syndrome for predicting progression to diabetes. The study included 2485 Finnish subjects aged 45-64 years without diabetes at baseline. During the mean follow-up of 10.4 years, 77 individuals developed type 2 diabetes.
The AACE definition was the most sensitive (.78) (with a fasting glucose level of 100 mg/dL for defining the syndrome) and least specific (.50). The WHO definition identified 69% of patients who progressed to diabetes, and the NCEP definition (with fasting glucose >/= 100 mg/dL) identified 64%. The EGIR definition identified the fewest of the cases of diabetes (58%) but was quite specific (.81). Therefore, all the definitions predict patients who are at risk for developing diabetes, with the AACE and WHO definitions being the most sensitive and the EGIR definition the most specific.
A meta-analysis assessed the relationship between the metabolic syndrome, CVD, and new-onset diabetes.[19] A total of 9 primary studies published between 2002 and 2004 that evaluated more than 50,000 patients were included in the review. The median follow-up time was approximately 5 years. The overall risk for CVD was 2 times greater in patients with the metabolic syndrome compared with those without, and the risk for new-onset diabetes was 3 times greater. This confirms the need to identify patients with the metabolic syndrome in order to make every effort to prevent the development of diabetes and CVD.
To determine whether primary care physicians correctly identify the metabolic syndrome, a group of investigators in Spain used an epidemiologic registry of subjects > 18 years of age with high blood pressure seen in a primary care setting.[20] NCEP-ATP III variables were measured in all 12,954 subjects, and a diagnosis of the metabolic syndrome was made if 3 or more criteria were met. In 6607 patients (51%), the diagnostic criteria for the metabolic syndrome were fulfilled. The primary care physicians identified 56% of the patients who had the metabolic syndrome, and wrongly diagnosed the metabolic syndrome in 13% of patients with no diagnostic criteria. Therefore, the metabolic syndrome is common in patients with hypertension in a general medical population, and primary care providers need to pay more attention to delineating the full syndrome in these high-risk patients.
An elevation in low-density lipoprotein (LDL) cholesterol levels is not a part of the metabolic syndrome, although abnormalities in LDL particle size are. Using the Insulin Resistance Atherosclerosis Study (IRAS) of 1486 individuals, various lipid parameters were measured and compared with the finding of the metabolic syndrome.[21] Measurement of the total number of LDL particles and the number of small LDL particles by nuclear magnetic resonance (NMR) appeared to be more related to most components of the metabolic syndrome than were measurements of ApolipoproteinB. The determination of LDL size by NMR appeared to be superior to LDL size by gradient gel electrophoresis. Therefore, when assessing the effects of the metabolic syndrome on the lipid profile, it is important to know the techniques used in the measurement and characterization of lipid composition and size.
Pima Indians have the highest rates of type 2 diabetes in the United States. Researcher in Arizona assessed the effects of the NCEP-ATP III metabolic syndrome components on the incidence rate of elevated urinary albumin excretion in diabetic and nondiabetic Pima Indians aged >/= 15 years.[22] In this study, 1296 Pima Indians without diabetes and 446 Pima Indians with diabetes and normal urinary albumin excretion at baseline were followed for approximately 4.6 years. Of the 1296 nondiabetic subjects, 128 developed elevated albuminuria (10%), as did 149 of 446 diabetic subjects (33%). In both groups, mean arterial blood pressure was predictive of the risk for developing albuminuria. Other predictive features differed between the nondiabetic and diabetic groups. However, this study underscores the need to follow and treat patients with the metabolic syndrome to lower their risk for the development of albuminuria, which is associated with an increased risk for CVD as well as a risk for renal disease.
Nonalcoholic fatty liver disease is becoming recognized as a component of the metabolic syndrome and insulin resistance. Elevations of ALT, AST, and ALK are common in individuals with type 2 diabetes as well as the metabolic syndrome. In 652 subjects from the IRAS who did not have the metabolic syndrome or diabetes at baseline, liver functions were assessed for their role in predicting the development of the metabolic syndrome.[23] After 5 years, 131 (20%) of subjects had developed the metabolic syndrome. The levels of ALT and ALK were correlated with the number of metabolic disorders at follow-up. Therefore, evolving abnormalities in liver function tests may portend the eventual development of the metabolic syndrome in high-risk individuals.
Jean-Pierre Després, PhD, and colleagues, from Laval University, Quebec City, Quebec, Canada, performed a study to assess the usefulness of fasting triglyceride levels plus waist circumference vs the NCEP III guidelines to evaluate the risk of CAD in women.[24] Two hundred fifty-four women (aged 32-82 years) who had had a coronary angiogram to assess the presence/absence of CAD were studied. The hypertriglyceridemic waist phenotype was defined as having both a waist circumference > 80 cm and elevated fasting triglyceride levels (> 1.5 mmol/L). The NCEP-ATP III criteria were defined as the presence of at least 3 of the 5 features for the metabolic syndrome. There was considerable overlap between the 2 groups; 73.2% of women with the hypertriglyceridemic waist phenotype also met at least 3 of the 5 NCEP-ATP III criteria. However, both criteria were predictive of CAD (hypertriglyceridemic waist phenotype: relative odds ratio [OR] 2.4, 95% confidence interval [CI], 1.3-4.3; P = .004; NCEP-ATP III clinical criteria: relative OR 2.5, 95% CI, 1.4-4.6; P < .003). Therefore, elevated triglyceride levels with central obesity in women are as likely to predict CAD risk in women as is the broader NCEP-ATP III definition.
Women who have had gestational diabetes are at increased risk for the development of diabetes in the future. Researchers in Italy evaluated the prevalence of the metabolic syndrome and associated cardiovascular risk factors in women who had had gestational diabetes in the past (pGDM).[25] One hundred sixty-six pGDM women and 98 matched control women were studied 16 months after delivery. Their average age was approximately 34 years. Homeostasis model assessment of insulin resistance (HOMA-IR) was used to estimate insulin resistance, and the metabolic syndrome was defined by the NCEP-ATP III criteria. The metabolic syndrome was present in 9% (n = 15) of the pGDM and in 1% (n = 1) of the control group. Based on HOMA-IR, the pGDM women were significantly more insulin-resistant than the control women. Therefore, it is important to follow women after they have had gestational diabetes, not only for the development of overt type 2 diabetes but for findings of the metabolic syndrome that may occur first.
Native Americans have high rates of type 2 diabetes. Unfortunately, diabetes is occurring in younger age groups as rates of obesity and inactivity are increasing in children and adolescents. Because the metabolic syndrome often precedes the development of type 2 diabetes in adults, 545 Cherokee adolescents, aged 12-19 years (females, n = 299; males, n = 246) were assessed for the presence of the metabolic syndrome.[26] NCEP guidelines were used, but modified by reverse extrapolation of adult cutpoints to their respective percentiles for age.[27] With these criteria, an increased waist circumference was found in 16% of males and 34% of females, and a low HDL cholesterol level was found in 49% of males and 78% of females. Triglycerides were elevated in 8% of both males and females. These findings are higher than those seen in both the National Health and Nutrition Examination Survey (NHANES) III survey and in the Mexican-American population. Therefore, screening of fasting lipid values and assessment of waist circumference (or possibly BMI alone) may be useful in aiding in early detection and screening for at-risk adolescents.
Patients with diabetes have high rates of CVD, and individuals who have had a myocardial infarction often have undiagnosed disorders of glucose metabolism, often either diabetes or impaired glucose tolerance. In one abstract,[28] 141 patients were studied 1 to 6 months after experiencing an acute MI. OGTTs were performed and various laboratory values obtained. Fasting glucose alone failed to identify more than 90% of the patients with abnormal glucose tolerance. Sixty subjects (43%) had type 2 diabetes, 25% of whom had not been previously diagnosed, and 28% of the subjects had impaired glucose tolerance. Other factors associated with CVD risk, such as lipids, fibrinogen, and C-reactive protein levels, were higher in individuals with abnormal glucose metabolism. Therefore, undiagnosed diabetes and prediabetes need to be considered in patients following a myocardial infarction, and, if found, should be treated aggressively to lower CVD risk factors.
The metabolic syndrome occurs in individuals with type 1 as well as type 2 diabetes.[29] In a study from Australia, 427 patients with type 1 diabetes were assessed for the metabolic syndrome, and 15% fulfilled the WHO criteria. Patients were subgrouped based on duration of diabetes. Differences were found in the 104 patients with a duration of diabetes greater than 20 years. For stroke, compared with individuals without the metabolic syndrome, the OR in those with metabolic syndrome was 22.8 (1.5-104.8, P = .008); for ischemic heart disease the OR was 2.4 (.5-10.9, P = 0.3); for peripheral vascular disease the OR was 7.3 (.99-53, P = .05); and for severe retinopathy the OR was 3.7 (1.2-12.5, P = .01).
This study shows that patients with type 1 diabetes and the metabolic syndrome have an increased risk for stroke, peripheral vascular disease, and severe retinopathy. This is likely to be in part related to the fact that those with the metabolic syndrome had higher levels of systolic and diastolic blood pressure, which is part of the definition of the syndrome. Other features of the syndrome, such as an abnormal lipid profile and the presence of albuminuria, are also likely to contribute. Therefore, identifying individuals with type 1 diabetes and the metabolic syndrome can be helpful in defining those who need aggressive treatment for the risk of CVD and retinopathy.
Another study assessed the rate of the metabolic syndrome and CVD in individuals with type 1 diabetes of varying duration.[30] Four hundred eighty-one individuals with type 1 diabetes and an average age of 19.7 years were divided into 4 groups according to diabetes duration: group 1, < 5 years (n = 232); group 2, 5-10 years (n = 136); group 3, 10-15 years (n = 87); and group 4, > 15 years (n = 39). WHO criteria were used to define the metabolic syndrome. The prevalence of hypertension and the metabolic syndrome increased with the duration of diabetes: 7.6% of subjects had the metabolic syndrome in group 1, but 30.2% and 43.5% had the metabolic syndrome in groups 3 and 4, respectively. Therefore, the metabolic syndrome is common in patients with a longer duration of diabetes, even at a young age. This syndrome therefore must be aggressively sought in individuals with type 1 as type 2 diabetes.
Another study assessed the prevalence of the metabolic syndrome in an Italian group of 978 individuals with type 1 diabetes.[31] With the NCEP definition, 13% had the metabolic syndrome. Patients with the highest hemoglobin A1C levels had the highest prevalence of the metabolic syndrome. In addition, the prevalence of overt nephropathy and proliferative retinopathy was significantly higher in patients with features of the metabolic syndrome. This confirms the other studies, in which patients with the metabolic syndrome and type 1 diabetes were at increased risk for microvascular complications.
Differentiating between type 1 and type 2 diabetes can sometimes be difficult in adults. In a study performed in the United Kingdom,[32] 129 individuals with new-onset diabetes were characterized as to the presence or absence of the metabolic syndrome according to the NCEP criteria. Anti-glutamic acid decarboxylase (GAD) antibodies were measured. Seventy-two percent of patients with diabetes had the metabolic syndrome, whereas 28% did not. Patients without the metabolic syndrome were leaner than those with it. The prevalence of GAD antibody positivity was 10.8% in patients without the features of the metabolic syndrome compared with only 1.1% in those with the syndrome. Therefore, patients with type 2 diabetes without the metabolic syndrome may actually have autoimmune slowly evolving type 1 diabetes, and measurement of anti-GAD antibodies may be helpful in these individuals.
Individuals with schizophrenia are at an increased risk for developing type 2 diabetes. This study assessed the prevalence of the metabolic syndrome in patients with schizophrenia with a meta-analysis of studies published from 2002 to 2004.[33] The prevalence of the metabolic syndrome in the group of 845 patients (mean age, 41.5 years) identified was 39.4% -- double the prevalence in the general population.[34] Therefore, patients with schizophrenia should be screened both for the metabolic syndrome as well as type 2 diabetes.
The gold standard for assessing insulin resistance is the euglycemic hyperinsulinemic clamp method. In this study, 64 subjects with features of the metabolic syndrome (and no diabetes) underwent a clamp study.[35] Although the particulars of the technique have relatively little practical clinical relevance, the investigators found that a modification of the NCEP criteria for the metabolic syndrome -- omission of hypertension, inclusion of BMI, family history, and fasting insulin -- greatly increased the specificity of prediction of insulin resistance. Although insulin assays are still not standardized, and the results may vary, inclusion of family history criteria and BMI may help clinicians more accurately identify patients at risk for the complications of insulin resistance.
It would be useful to predict who is and who isn't evolving toward the metabolic syndrome (although this should not be a license to become a couch potato in those at less risk). Elizabeth Barrett-Connor, MD, and her team at the University of California, San Diego, Medical Center, San Diego, California, used a subset from the Rancho Bernardo Study to assess adiponectin levels in patients who later progressed to develop the metabolic syndrome.[36] (Lower levels of adiponectin increase insulin resistance.) They studied 325 nondiabetic white men (mean age, 65 years; BMI 26 kg/m2) and 241 nondiabetic women (mean age, 70 years; BMI 24 kg/m2). Patients underwent an OGTT between 1984 and 1987, with a follow-up assessment between 1992 and 1996. Age-adjusted total adiponectin concentration at baseline was lower in those who developed the metabolic syndrome. Thus, if measurement of adiponectin levels becomes routinely available, it may be a useful way to test for early risk for the development of the metabolic syndrome.
Repeated analyses of the definitions of the metabolic syndrome lead to several conclusions. First, no matter which definition is used, they all identify individuals at higher risk for CVD as well as type 2 diabetes. This appears to hold true in all populations, from children to women to men to individuals around the world, although the specifics of the diagnostic criteria may need to be adjusted to fit the population at hand. It is easy to recommend lifestyle modification, but difficult to implement in practice. Therefore, these high-risk individuals need to be treated aggressively with medications known to lessen the risk for cardiovascular and microvascular complications. The first step to providing these treatments, however, comes from identification of the metabolic syndrome. It is safe to say that data now demonstrate not only that the definitions of the syndrome are useful, but that the application of those definitions should be mandated to direct our attention to the treatment of individuals who have the most to benefit from preventive health measures.
The metabolic syndrome has emerged as a clinical and public health crisis. The incidence of the metabolic syndrome has reached epidemic proportions -- with more than 1 in 4 adults affected by this disorder in the United States and worldwide. Attention has turned to more aggressive clinical management, and treatment goals, therapeutic approaches, and comprehensive management are now the focus of both clinical care and research. The American Diabetes Association's (ADA's) 65th Annual Scientific Sessions were dominated by novel approaches to the identification and management of the metabolic syndrome. In the accompanying feature by Dr. Anne Peters, a summary of numerous reports on the definition, identification, and features of the metabolic syndrome are reviewed in great detail.
What then of management? Is the entirety of the metabolic syndrome an appropriate target for treatment? Is comprehensive management possible? Which of the component parts should be the focus of treatment? What role does insulin resistance (IR) play in the development and management of the metabolic syndrome? These and other questions are addressed in the following sections.
There is now greater consensus around the components of the metabolic syndrome, which include central obesity, insulin resistance, glucose intolerance, hypertension, and dyslipidemia. Each of these component parts is an appropriate target for treatment, yet newer therapies may treat "core" or central components more effectively. Although traditional approaches to the separate risk factors (risk factors for both cardiovascular disease and type 2 diabetes) have proven effective, increasing attention is now being directed at the management of IR and obesity.
Despite broader acceptance of the metabolic syndrome as a clinical disorder, significant confusion exists regarding the specific approach to management. Although somewhat artificial, this summary divides approaches to management by therapy -- therapies that may target either CVD risk, diabetes risk, or both. Given the 2- to 3-fold increase in CVD risk observed in those with the metabolic syndrome, and the known benefit of risk-reduction efforts in others at high risk for CVD, this area is the focus of much of our therapeutic efforts. However, diabetes prevention remains a critical clinical concern, and successful weight management, treatment of insulin resistance, or both should be considered in all patients with the metabolic syndrome.
IR is now broadly accepted as a core component of both the metabolic syndrome and type 2 diabetes. Recent reports have suggested that simple clinical measures are reasonable surrogates for defining IR.[1] This important contribution to the scientific literature demonstrated that IR can be determined with high sensitivity and specificity with routine clinical measures. Utilizing a large euglycemic clamp data set (more than 2300 individuals from Europe; San Antonio, Texas; and the Pima Indian population), the study authors used recursive partitioning (so-called classification trees) to determine the potential of clinical measures of body mass index (BMI), homeostasis model assessment (HOMA), and lipids to identify those with IR. These results suggest that the diagnosis of IR can be made with measures of obesity, with BMI > 28.9 kg/m2 identifying IR as well or better than measures of serum insulin. The addition of other clinical characteristics to the decision tree (including a family history of diabetes or elevated triglycerides) provided an even more sensitive and specific prediction of IR.
From a practical clinical perspective, IR is best defined as the degree of resistance to insulin's effect, above which there is significantly higher risk for developing type 2 diabetes or CVD when compared with those below that same level of IR. As research continues to better refine both our definition of IR and its clinical identification, the practicing clinician must be aware of the best tools for identifying patients with IR. This knowledge will not only guide clinical care but hopefully improve overall health.
One ADA symposium focused on the question: "Is insulin resistance a relevant treatment target?[2]" David Matthews, DPhil, BM, BCh, FRCP, Baylor College of Medicine, Houston, Texas (who with his colleagues developed the HOMA model for assessing insulin sensitivity) reported on the strengths and weaknesses of classic laboratory measures of insulin sensitivity. Speaking specifically about HOMA, frequently sampled intravenous glucose tolerance testing (FSIVGTT), and clamp methodology, Professor Matthews suggested that each has a potential clinical role. HOMA has been widely used in large populations.[3] As a measure of insulin action in the fasting state, HOMA provides useful information but does little to help clinicians identify individuals at highest risk for IR. Nor does HOMA permit evaluation of insulin action in the dynamic state (such as following food ingestion). Both the FSIVGTT and clamp technology provide insights into insulin action in a dynamic state and in the setting of supraphysiologic insulin concentrations. However, neither is a practical clinical tool. The prior report of Stern suggests that BMI alone may be a reasonably sensitive predictor of IR.
Mary Ann Banerji, MD, State University of New York Downstate Medical Center, Brooklyn, presented a review of the use of clinical markers, such as body weight and adiposity, as determinants of IR. Although much remains to be learned, Dr. Banerji's discussion provided compelling evidence of the role of adipose tissue as an active endocrine organ and a potential source of proinflammatory cytokines that promote IR, result in abnormalities of vascular behavior, and contribute to the atherosclerotic process itself.
David Kendall, MD, provided a review of current pharmacologic and nonpharmacologic approaches to managing IR. Nondrug therapies -- including weight loss and increased activity -- have known insulin-sensitizing effects. However, these effects are limited; insulin sensitivity is improved by only ~20%, and lifestyle change is difficult to both achieve and maintain. Recent studies of weight loss per se have suggested only modest glucose-lowering potential, with deterioration in glucose control after 1 year of aggressive weight-loss interventions.[4,5]
The use of medications in the treatment of IR was also reviewed, with an emphasis on the important clinical effects of both metformin and the thiazolidinediones (TZDs). Although both metformin and TZDs are effective glucose-lowering therapies, only TZD treatment is associated with a broad array of improvements in components of the metabolic syndrome. The role of TZDs in diabetes prevention, early treatment of type 2 diabetes, and CVD risk reduction remains an active area of research interest.
The management of IR with TZDs has resulted in greater attention to activators of the so-called peroxisome proliferator-activated receptors (PPARs). TZDs exert much of their effect on IR via activation of PPAR-gamma. Therefore, TZDs not only improve glucose control but favorably affect both free fatty acid metabolism and insulin action. TZD therapy has been proposed as a treatment for the metabolic syndrome by some, and a number of reports at this year's ADA meeting provided support for such an approach to therapy.
Frederick F. Samaha, MD, Philadelphia Veterans Affairs Medical Center, University of Pennsylvania Health System, Philadelphia, Pennsylvania, and colleagues[6] reported on the effects of rosiglitazone on markers of inflammation and lipids in nondiabetic individuals with the metabolic syndrome. Rosiglitazone therapy (8 mg daily) significantly reduced C-reactive protein levels, although there were modest increases in both total and non-high-density lipoprotein cholesterol (HDL-C) levels.
Philippe Szapary, MD, University of Pennsylvania Health System, and colleagues[7] reported on the effect of pioglitazone in patients with the metabolic syndrome. This randomized trial of 60 subjects demonstrated a significant increase in HDL-C and favorable effects on lipid subfractions without an effect on triglycerides or low-density lipoprotein cholesterol (LDL-C) concentrations.
The impact of PPAR-gamma agonists on lipids was further detailed in reports from a head-to-head trial of rosiglitazone and pioglitazone performed in patients with type 2 diabetes and dyslipidemia. Meng H. Tan, MD, Eli Lilly, Indianapolis, Indiana, and colleagues[8] reported that pioglitazone therapy resulted in greater improvements in the atherogenic index of plasma and lowered triglyceride levels effectively while achieving greater increases in HDL-C when compared with rosiglitazone. Another report on data from the same trial demonstrated that the beneficial effect of pioglitazone on triglyceride concentrations was the consequence of favorable changes in very low-density lipoprotein (VLDL) subfraction concentrations.[9] The benefits of TZD therapy were preserved even in the setting of statin therapy.[10,11]
Several reports discussed the potential role of PPAR-gamma activators for the management of early glucose intolerance and the prevention of diabetes. Of note, early use of TZD therapy in a population at high risk for diabetes demonstrated similar protection from progression to diabetes reported in earlier trials with troglitazone.[12]
Significant clinical interest remains in the use of PPAR-alpha activators (fibric acid derivatives) for the management of lipid disorders in those with the metabolic syndrome as well. The combination of fenofibrate and exercise was effective for the management of the metabolic syndrome by means of decreasing visceral obesity, improving hypertriglyceridemia, and controlling hyperglycemia in a rodent model of this disorder.[13] Although the role of PPAR-gamma agonist therapy for the management of the metabolic syndrome requires further detailed study, these results suggest an ever more important role for pharmacologic therapy in those with the metabolic syndrome.
There was considerable attention given to the newest class of insulin-sensitizing medications at this year's meetings. The so-called dual-PPAR activators (activating both PPAR-alpha and PPAR-gamma -- known as the glitazar class of medications) and pan-PPAR activators are in clinical development. These agents possess significant potential as therapy for both diabetes and the metabolic syndrome, particularly for those with dyslipidemia. Results of a number of clinical trials with these unique agents were reported at this year's scientific sessions.
Muraglitazar is one of several novel dual-PPAR activators in active clinical development. Results of several pivotal, phase 3 clinical trials were included in reports at this year's meetings, including results from dose-ranging studies and in comparison with active pioglitazone therapy. In a presentation by Ralph DeFronzo, MD, results of the comparator trial with pioglitazone were provided. Muraglitazar (5 mg daily) resulted in significantly greater reductions in hemoglobin A1C and greater improvements in triglycerides when compared with pioglitazone at a dose of 30 mg/day.[14]
Other studies with muraglitazar were summarized in the late-breaking clinical trial reports, with muraglitazar demonstrating significant reductions in A1C (as much as 1.2%) and reductions in triglycerides of > 30%.[15] These beneficial effects were maintained for as many as 2 years of treatment, suggesting a sustained improvement in insulin sensitivity, glucose control, and dyslipidemia. Although these studies were performed in patients with type 2 diabetes, a significant majority of subjects in these trials met the criteria for the metabolic syndrome. Given the higher incidence of dyslipidemia and glucose intolerance in such a setting, the dual-PPAR activators possess significant clinical potential for the medical management of these individuals.
The results of treatment with another dual-PPAR activator, tesaglitazar, in patients with the metabolic syndrome were reported.[16] All doses of this novel dual-PPAR activator reduced the incidence of the metabolic syndrome, with the highest dose reducing the incidence by nearly 60% (as compared with a > 20% increase in incidence in untreated individuals). The study authors suggest that such a beneficial change could limit the risk of type 2 diabetes and CVD.
A separate report demonstrated that tesaglitazar treatment retarded the development of atherosclerotic lesions in a rodent model of the metabolic syndrome, suggesting a significant role for dual-PPAR activators in high-risk individuals.[17]
Weight loss is considered to be one of the key therapeutic interventions to limit the risks associated with the metabolic syndrome. Indeed, weight loss has been shown to reduce the incidence of the metabolic syndrome and significantly improve control of blood pressure, lipids, and glucose -- all central features of the syndrome. However, achieving and sustaining weight loss, particularly in obese patients with the metabolic syndrome, has been difficult if not impossible with nonpharmacologic, nonsurgical therapies. As such, medical therapies to assist with weight management have been actively sought. Exciting reports from this year's ADA meeting provided an early look at the potential of several classes of pharmacologic therapies to assist in efforts to achieve sustained weight reductions.
Rimonabant is a unique compound that exerts its effects through the endocannabinoid receptor system. Results from a double-blind, placebo-controlled study with rimonabant in patients with diabetes were presented. In overweight and obese patients with type 2 diabetes, rimonabant therapy resulted in significant weight loss (averaging approximately 5% of body weight) and demonstrated improvements in several cardiovascular and metabolic risk factors characteristic of the metabolic syndrome. Rimonabant therapy resulted in a reduction in A1C levels, blood pressure, waist circumference, body weight, and triglycerides. Modest increases in HDL-C were also observed. Among patients treated with rimonabant, 68% lowered A1C below 7%.[18]
Other compounds known to reduce body weight in the setting of diabetes include the incretin mimetics. Compounds such as exenatide mimic the effects of the native gut peptide glucagon-like peptide (GLP)-1. Exenatide was recently approved as adjunctive treatment for the treatment of type 2 diabetes. The results of open-label extension studies (82-104 weeks of therapy) with this unique GLP-1 agonist demonstrated significant reductions in both triglycerides and diastolic blood pressure and an increase in HDL-C. These changes were accompanied by an average 1.2% reduction in A1C and continued weight loss totaling > 4 kg on average after the 82 weeks of treatment.[19] Whether exenatide or other compounds that work via the GLP-1 system will have a greater role in either weight management or the treatment of the metabolic syndrome remains to be determined.
Hypertension is both a defining characteristic of and a key cardiovascular risk factor in the metabolic syndrome. Early combination therapy for hypertension has become a more broadly accepted standard for patients with diabetes. Whether this approach to treatment is of potential benefit in those with the metabolic syndrome has not been carefully examined. Early use of either angiotensin-converting enzymes inhibitors or angiotensin II receptor blockers has been suggested for those with the metabolic syndrome. Results from a subgroup analysis of the Irbesartan/HCTZ Blood Pressure Reductions In Diverse Patient Populations (INCLUSIVE) trial were reported to determine the efficacy of fixed-dose combinations of irbesartan-HCTZ in patients with the metabolic syndrome as defined by the National Cholesterol Education Program (NCEP).[20] Individuals with uncontrolled systolic blood pressure (SBP) were treated with forced titration to target SBP levels. A total of 386 (46%) of the study cohort of 844 patients met the criteria for the metabolic syndrome. Aggressive titration of irbesartan-HCTZ therapy reduced SBP by -21.0 mm Hg ± 14.3 mm Hg, with 73% of patients with the metabolic syndrome achieving an SBP of < 140 mm Hg or < 130 mm Hg for those with coexisting type 2 diabetes. Such studies support the use of both aggressive titration and early combination therapy for those with the metabolic syndrome with uncontrolled blood pressure.
The metabolic syndrome is increasingly common, and the characteristic components of this disorder are obvious targets for treatment to reduce the risk of both diabetes and CVD. The exciting advances in pharmacologic therapy for this syndrome raise the possibility that both nonpharmacologic and pharmacologic treatments will be increasingly applied to those at greatest risk. With better tools to identify individuals with this syndrome, we as clinicians will be called upon to guide patients through the maze of ever more sophisticated treatments for obesity, glucose intolerance, IR, lipid disorders, and hypertension. Only with these aggressive efforts can we hope to stem the tide of the metabolic syndrome and its inherent risk of type 2 diabetes and CVD.
Cardiovascular disease is the primary cause of death in people with diabetes, accounting for roughly 65% of mortality.[1] Of note, increased cardiovascular risk actually precedes the formal diagnosis of type 2 diabetes by many years.[2-4] In other words, the clock starts ticking years before the onset of clinical diabetes.
This ticking-clock hypothesis is based partly on observations of participants in an 8-year follow-up of the San Antonio Heart Study. Patients who were nondiabetic at baseline examination but went on to develop type 2 diabetes had substantially higher total and low-density lipoprotein cholesterol (LDL-C), triglycerides, body mass index, and blood pressure, accompanied by lower levels of high-density lipoprotein cholesterol (HDL-C), compared with subjects who did not develop type 2 diabetes.[3] Thus, patients who developed diabetes exhibited several risk factors for cardiovascular disease prior to actually having diabetes.
Indeed, several more recent studies also support the conclusion that increased cardiovascular disease risk precedes the formal diagnosis of diabetes. Specifically, a meta-analysis of 95,783 people who had 3707 cardiovascular events over 12.4 years found that the progressive relationship between elevated plasma glucose and cardiovascular disease risk begins at a level well below the cutoff for a formal diagnosis of diabetes.[2] Furthermore, in the Nurses Health Study, an increased risk for myocardial infarction and stroke was evident well before the diagnosis of diabetes. The risk for cardiovascular disease increased further after clinical diagnosis of diabetes, and the highest incidence was found in women with diabetes diagnosed at study entry.[4] Finally, data from the Norfolk cohort of the European Prospective Investigation of Cancer and Nutrition indicate that hemoglobin A1C predicts all-cause, cardiovascular, and ischemic heart disease mortality continuously across the male population, among individuals both with and without diabetes, and at A1C levels below those considered indicative of diabetes.[5] The study found that men with A1C levels that were elevated but still within the normal range (5.0% to 6.0%) had a significantly increased mortality risk compared with men with A1C levels less than 5.0%. Furthermore, every 1.0% increase in A1C was associated with a 28% increase in the risk of death, independent of age, blood pressure, serum cholesterol, body mass index, or cigarette smoking habit.[5,6]
Richard Hamman, MD, DrPH,[7] University of Colorado School of Medicine, Denver, examined the data on retinopathy for 890 people who had participated in the Diabetes Prevention Program (DPP). Among the 302 who were considered prediabetic, 7.6% had early-stage retinopathy, whereas retinopathy was found in 12.5% of the 588 who had progressed to diabetes. Dr. Hamman concluded that these findings suggest that some of the risk factors for long-term complications are operating early in diabetes, and emphasized the need for early blood glucose control.
"Prediabetes" is a condition affecting an estimated 41 million people in the United States aged 40-74 years of age, in which blood glucose levels are high but insufficient to be called diabetes. It is now established that prediabetes carries an increased risk for the development of diabetes, heart disease, and stroke. The following question therefore arises: "Should we further lower the criteria for the diagnosis of diabetes?" The current numbers are:
This is reminiscent of what has transpired with LDL-C. We once considered that an LDL-C of < 160 m/dL in the absence of other risk factors for a cardiovascular event was acceptable. With major primary and secondary prevention studies, the number was lowered -- first to 130 mg/dL and then to 100 mg/dL. The Treating to New Targets (TNT) study lowered the crossbar further to 78 mg/dL, and one began to wonder where this was all going. Indeed, it may be an attempt to return to the womb. Neonates have LDL-C levels of around 25-30 mg/dL, and the endothelial lining of their blood vessels is as smooth as their baby bottoms. Will we need to visit this upon blood glucose levels? However, despite concern with the J curve of LDL-C, wherein some argue that a cholesterol level that is too low is not good, no harm has come from lowering cholesterol to these new targets.
In contrast, attempts to intensify glucose-lowering are fraught with difficulty and a 3-fold increase in the incidence of severe hypoglycemia. For this reason, even in the Diabetes Control and Complications Trial (DCCT), after initial reduction of blood glucose by approximately 1% in the intensively treated group, A1C drifted upward to reach levels equivalent to those in the conventionally treated group within the ensuing 8-9 years. This phenomenon is now referred to by some as the "Nike Curve of Diabetes Control," because the shape of the curve resembles the Nike "swoosh" trademark.
Was all this effort and expenditure of taxpayers' money of no avail? On the contrary. This early intensive treatment cut the risk of cardiovascular disease by about 50% and the risk of a heart attack or stroke by about 57%, reported David Nathan, MD, Harvard University, Boston, Massachusetts, and Saul Genuth, MD, Case Western Reserve University, Cincinnati, Ohio, co-chairs of the DCCT and the Epidemiology of Diabetes and Interventions and Complications study (EDIC).[8]
The results of the DCCT were reported in 1993.[9] The study involved 1375 people with type 1 diabetes treated, on average, for 6.5 years after the study was initiated in 1983. The EDIC study, the observational follow-up to the DCCT, observed a delay between improved glycemic control and slowed progression of complications in the former DCCT conventional treatment group.[10] These results suggest a long-term detrimental impact of early, sustained hyperglycemia on microvascular complications. Indeed, even when normoglycemia is achieved following pancreas transplantation, it can take 5- 10 years for renal lesions to reverse in patients with type 1 diabetes.[11] The EDIC study was the first large-scale clinical trial to demonstrate the persistent benefit among patients who were initially randomized to intensive control compared with patients initially randomized to conventional treatment, despite converging A1C values between the 2 groups over time.[12-14] The new data, more than 10 years after the initial study period, strongly support the notion that early intensive treatment -- even for a short period -- has significant beneficial effects on macrovascular disease. The study authors speculate that this may also apply to the type 2 diabetic population, but this is not at all clear.[8]
During the prodrome that precedes type 2 diabetes, there are many operative factors now collectively referred to as the metabolic syndrome (Figure). Although the standard definitions have varied somewhat, the components have not changed and include hypertension, dyslipidemia, obesity with an increase in waist diameter, microalbuminuria, and evidence of a prothrombotic state and inflammation linked to oxidative and nitrosative stress.[15,16] Because insulin has potent anti-inflammatory properties and combats oxidative and nitrosative stress,[15,17] some of the beneficial effects observed may well have been due to factors other than glycemic control.[16,18]
Figure. From womb to tomb and back.
Andre Scheen, MD, PhD,[19] Academic Hospital of Liege, Belgium, presented the first report on the RIO Diabetes Study. Rimonabant, the first in a new class of drugs called the selective CB1 receptor endocannabinoid blockers, can be taken once daily. In a 1-year study, it effected a 12-lb weight loss, trimmed 2 inches from the waist, dropped A1C levels by about .7%, raised HDL-C levels by 15%, lowered triglyceride levels by 25%, and dropped blood pressure by 3/2 mm Hg. This would constitute the first drug targeting most of the manifestations of the metabolic syndrome and producing enduring effects, certainly for a year. It does come with side effects, including nausea, vomiting, anxiety, and depression, that will require attention. Nonetheless, practitioners can look forward to its availability within the next year.
The peroxisome proliferator-activated receptor (PPAR)-gamma agonists are well known to clinicians in the United States. Differences in the metabolic effects of the different drugs within the class have suggested that minor structural differences could account for major differences in biological effects. For example, troglitazone (now withdrawn because of hepatotoxicity), in addition to its capacity for insulin sensitization, also had profound effects on triglycerides and HDL-C. Although this effect was initially ascribed to the vitamin E quinone moiety of the drug, it seems that it could rather have been ascribed to simultaneous activation of PPAR-alpha in addition to PPAR-gamma.
Subsequent studies with pioglitazone and rosiglitazone[20] have shown major differences in lipid-lowering effects in favor of pioglitazone. Studies demonstrate an antioxidant effect as well as an ability to reduce nitrosative stress,[21] an action that may be attributable to its PPAR-alpha effects at the doses used clinically. Studies investigating 2 new agents that combine PPAR-gamma and -alpha, muraglitazar and tesaglitazar, were reported at the American Diabetes Association (ADA) meeting and confirm that a single pill can lower blood glucose levels by as much as 60 mg/dL, increase insulin sensitivity, raise HDL-C levels by as much as 19%, lower triglycerides by about 28% to 40%, and lower LDL-C by 17% in a dose-dependent manner. Although this is well and good, these compounds still face several hurdles, such as weight gain of 2-10 lb. Philipp E. Scherer, PhD,[22] Albert Einstein College of Medicine, Bronx, New York, in his marvelous Banting Lecture, demonstrated that overexpression of the adiponectin gene in Ob/Ob mice led to a huge increase in body weight and fat-cell mass, but demonstrated that the fat cells were of the small, dynamic, good kind that produced adiponectin and led to normal insulin sensitivity and normal glucose and lipid levels. Perhaps, then, it may not be all that bad to gain a few pounds to normalize your metabolism -- but try telling that to a patient who wants to wear a skimpy bathing suit on the beach this summer!
The clock for the microvascular complications of type 2 diabetes also begins ticking before the diagnosis of diabetes. Data obtained from the United Kingdom Prospective Diabetes Study (UKPDS) demonstrated that substantial microvascular as well as macrovascular abnormalities were present in roughly 50% of patients at the time of diagnosis of type 2 diabetes.[23] Data from long-term follow-up of the UKPDS appear to be consistent with the finding that early metabolic control has enduring beneficial effects. In the neuropathy symposium at the ADA meeting, Professor Rury Holman, FRCP, Oxford University, Oxford, United Kingdom, presented the long-term outcomes of the UKPDS, which substantiated the notion that early aggressive treatment of people with type 2 diabetes can reduce the development of neuropathy -- defined as abnormalities in vibration perception -- by almost 50%.[24] This evidence emphasizes the importance of early diagnosis and aggressive initiation of therapy to prevent further tissue damage and associated morbidity in patients with diabetes.
An important study on the role of protein kinase C (PKC) beta 2 in renal disease suggested that perhaps all is not lost even when blood glucose control has not been optimized.[25] Although the pathogenesis of diabetes-related microvascular complications is complex, several putative pathways have been implicated, including (1) increased formation of nonenzymatic advanced glycosylation end products, (2) increased activation of the PKC pathway, and (3) increased glucose flux through the aldose-reductase pathway.[26] Each of these pathways may be activated by oxidative stress. Another contributor to microvascular pathogenesis is the increased level of free fatty acids present in states of insulin resistance. Free fatty acids may increase the formation of reactive oxidative species, and have demonstrated lipotoxic effects on the endothelium, including induction of apoptosis and delayed cell-cycle progression.[27] Oxidative stress has also been implicated in overactivation of poly(ADP-ribose) polymerase, resulting in cellular dysfunction and necrosis.[28] In addition, as the disease progresses and blood glucose levels rise, hyperglycemia-induced reactive oxidative species may in turn contribute to activation of each of the putative pathways.
Ruboxistaurin (RBX) is a new class of compound that selectively inhibits PKC beta 2 in the eye, kidney, and peripheral nerve. There are now data from phase 2 clinical trials to suggest that it may be effective in reducing macular edema[29] as well as improving the symptoms and objective evidence of peripheral nerve damage in diabetes.
For the first time, data were presented at the ADA meeting that support the notion that the compound can actually decrease microalbumin excretion and improve the glomerular filtration rate (GFR). This was reported based on a randomized, double-blind, placebo-controlled, multicenter, pilot phase 2 trial conducted to determine whether RBX 32 mg/day for 1 year could lower the urinary albumin/creatinine ratio (ACR) in persons (n = 123) with type 2 diabetes and nephropathy.[25] These patients had persistent albuminuria (ACR 200-2000 mg/g) despite treatment with stable doses of angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), or both, for at least 6 months prior to the study. Of importance, these patients continued their renin angiotensin system inhibitors for the entire trial. Baseline characteristics did not differ significantly between groups: ACR 764 mg/g ± 430 mg/g, estimated GFR (eGFR; modified MDRD formula) 70.0 mL/min ± 24 mL/min, systolic blood pressure 135 mm Hg ± 14 mm Hg, diastolic blood pressure 75 mm Hg ± 9 mm Hg, and A1C 7.95% ± 1.16%. After 1 year of RBX treatment, ACR decreased significantly from baseline (-24%, P = .02). ACR did not change significantly from baseline (-9%, P = .33) in the placebo group. The ACR-lowering effect of RBX appeared as early as 1 month, was maximal at 3 months, and was sustained for the 1-year period of the study. Placebo-treated persons lost GFR (-4.8 mL/min/year ± 1.8 mL/min/year, P = .009) over 1 year. In contrast, eGFR did not change significantly in the RBX group (-2.5 mL/min/year ± 1.9 mL/min/year, P = .185). Comparisons for both ACR and eGFR did not differ between treatment groups. However, this pilot phase 2 study was not designed to determine such a difference, and the power for either analysis was less than 20%. Treatment-emergent adverse events were similar between groups. The study authors concluded that the addition of RBX to therapy with ACEIs, ARBs, or both had long-term (1 year) favorable effects on ACR and eGFR in persons with type 2 diabetes and nephropathy. This must be viewed in the context of the remarkable change in the rate of progression in renal disease that has occurred with the advent of blood pressure control and the use of ACEIs and ARBs, and the effect that blood glucose and lipid control have had on the rate of decline of renal function. A decade or so ago, the rate of progression of renal disease was a fall in GFR of 1 mL/min/month; with the intensified therapeutic approach, this has fallen to around .2 mL/min/month, leaving little scope for prospective new therapies. The fact that RBX gave added benefit to established therapies for diabetic nephropathy augurs well for the future of the compound in diabetes therapy.
I would say that things are certainly looking up for the person with diabetes and the metabolic syndrome. We have the tools, or will have many of them in the near future. We need to learn how to modify behavior in order to make a big dent in the incidence of macro- and microvascular complications of diabetes. My problem is that I can lead a horse to water, but I have not yet learned how to make it drink.