Adiposopathy: sick fat causes high blood sugar, high blood pressure and dyslipidemia

It is well known amongst clinicians that genetically predisposed patients who acquire excess body fat and maintain a sedentary lifestyle are at increased risk of developing a number of disease states [1], many which constitute important atherosclerotic coronary heart disease (CHD) risk factors that may lead to CHD events (Figure 1). Among the more common metabolic disorders often associated with excessive body fat are abnormalities of glucose metabolism (such as diabetes mellitus), hypertension and dyslipidemia. However, the relationship between excessive body fat and adverse metabolic consequences is not absolute. Obesity alone does not always result in clinical metabolic disease, and not all patients with these metabolic diseases are overweight. Recent and ongoing research into adipose function and dysfunction seem to be confirming these basic and common clinical observations. Studies are now supporting the concept that it is not the presence of excess adipose tissue alone, adiposity, that is causally related, but rather it is the dysfunction of adipose tissue, here termed ‘adiposopathy’, that should be identified as the root mechanistic aetiology of disorders of glucose metabolism, blood pressure and lipid metabolism. The implications of this alteration in emphasis is certainly of significance for mechanistic research, and is also of great practical significance for the clinician.

Adiposopathy: replacing the term ‘metabolic syndrome’

Many authors and scientific organizations have characterized and applied terminology to the constellation of metabolic abnormalities often associated with the accumulation of excessive body fat. Unfortunately, not everyone agrees on any particular characterization, nor do they always agree upon the definition.

For example, the most common current term that refers to the cluster of metabolic abnormalities associated with disorders in glucose metabolism, hypertension and dyslipidemia is the ‘metabolic syndrome’. A similar clinical presentation has also been termed atherothrombogenic syndrome, beer-belly syndrome, cardiovascular metabolic syndrome, chronic cardiovascular risk factor clustering syndrome, deadly quartet (obesity, hyperinsulinemia, hypertension and dyslipidemia), disharmonious quartet, dysmetabolic syndrome, dysmetabolic syndrome X, insulin resistance syndrome, insulin resistance-dyslipidemia syndrome, metabolic cardiovascular syndrome, metabolic syndrome, metabolic syndrome X, multiple metabolic syndrome, plurimetabolic syndrome, Reaven’s syndrome, and syndrome X [2–4]. The variety of different terms for a similar or related condition reflects the difficulty in applying one name to a constellation of related, and sometimes unrelated metabolic abnormalities, and reflects an unsatisfying attempt to identify one root pathophysiologic cause of all of these.

Futher complicating the matter is that differing organizations have different definitions for the same condition. The National Cholesterol Education Program, Adult Treatment Panel III (NCEP ATP III) [5] and the

World Health Organization (WHO) [6] have established the most widely used definitions of the metabolic syndrome (Table 1). The criteria differ in many respects; for example, while waist circumference is a criteria included in the NCEP ATP III definition, waist:hip ratio or body mass index (BMI) ratio is used in the WHO definition. The NCEP ATP III definition does not include glucose intolerance or insulin resistance, while these are important criteria in the WHO definition. The NCEP ATP III does not include albuminuria, while this is a criteria recommended by WHO.

The differing nomenclature and criteria for diagnosis is not without consequence. Obesity is the most common metabolic disease in developed nations. According to the WHO, over 1 billion adults are overweight on a worldwide basis, with at least 300 million being obese [101]. In the USA, the unabated epidemic of obesity is now such that over 30% of adults are obese and over 60% of adults are either overweight or obese [7]. Of equal concern is that the high prevalence of obesity among children and adolescents continues to increase – a trend that suggests that the obesity epidemic will continue to increase in the future [8]. Largely as a consequence of the obesity epidemic, the increased prevalence of the metabolic syndrome might also be considered

an epidemic in that it has been estimated that at least 47 million adults (22% of the population) in the USA are affected [9].

Given the enormity of this disease burden and the known public health consequences, it would seem reasonable that extraordinary efforts be made towards preventing, treating, and potentially curing the metabolic syndrome. However, the conduct and reporting of clinical trials to demonstrate efficacy in the treatment of the metabolic syndrome have been hampered by the lack of consistent terminology, the lack of uniform diagnostic definitions and the different criteria reported in different clinical studies. For example, in recognition that increased body fat is often directly related to the development of the metabolic syndrome, investigators have often substituted BMI for measurement of waist circumference, which is a listed criteria according to the NCEP ATP. Sometimes, the substitution cutoff value has been greater than 30 kg/m², while at other times greater than 25 kg/m² has been used [2]. This is because from a practical standpoint, BMI is routinely obtained in clinical trials, while waist circumference is not.

Further complicating the matter is that not all clinicians are clear on the similarities and differences between the metabolic syndrome and other associated conditions. Not all clinicians are aware that patients with Type 2 diabetes may or may not have metabolic syndrome, and patients with metabolic syndrome may or may not have diabetes. Moreover, even researchers are not certain whether insulin resistance syndrome and metabolic syndrome are the same, or have the same treatment goals [3].

In addition, the criteria to define the metabolic syndrome is largely based on epidemiological and cross-sectional observations mainly obtained in European descent population, and were subsequently generalized to various ethnic groups. Subsequent studies have revealed that aspects of the metabolic syndrome profoundly differ among ethnic groups. Even within the European descent populations, there are additional features of the metabolic syndrome that could be a better predictor of risk of metabolic and CHD risk.

It is also noteworthy that the scientific organizations that have defined metabolic syndrome have not required that the components of the metabolic syndrome be due to any unifying and underlying metabolic process. Yet an increase in abdominal girth, hypertriglyceridemia, low high-density lipoprotein cholesterol (HDL-C) levels, high blood pressure and elevated glucose can all individually be the result of diseases and conditions that are entirely unrelated to one another, and thus not reasonably be connected to any one causality or syndrome. This may help to explain why the diagnosis of the metabolic syndrome may not be a predictor of 11 year CHD mortality among patients with Type 2 diabetes mellitus, and its diagnosis may not provide further predictive value compared with knowledge of its single components [10]. In fact, even in patients without diabetes mellitus, diagnosis of the metabolic syndrome appears to be inferior to established prediction models for either Type 2 diabetes or CHD, such as the Diabetes Predicting Model and Framingham Risk Score [11].

The bottomline is that most clinicians know from their clinical practice experience that select patients with the metabolic syndrome are at high risk for development of atherosclerotic CHD and diabetes mellitus. To some extent, these consequences are supported in the medical literature [12]. Clinicians also have the intuitive sense that a long-term commitment towards treating these multiple metabolic risk factors improves the health of their patients, which is again supported by the medical literature [13]. However, while regulatory agencies such as the US Food and Drug Administration (FDA) recognize the existence of the metabolic syndrome, they have yet to establish criteria in order to grant an indication for its treatment. This lack of indication criteria for the treatment of the metabolic syndrome impairs the development of pharmaceutical agents to treat the underlying root cause and/or pathophysiology, and instead promotes continued research efforts focusing on drug treatment for the components of the metabolic syndrome. In other words, while specific regulatory criteria exists for granting approval for an indication for treatment of the components of the metabolic syndrome (such as Type 2 diabetes mellitus, hypertension and dyslipidemia), no such approvable ‘indication’ criteria exists for treatment of the metabolic syndrome itself, that often reflects underlying metabolic processes that may promote these conditions.

Accumulating scientific data suggest that in most patients, it is the dysfunction of adipose tissue (adiposopathy) that is the root mechanistic aetiology of abnormalities of glucose metabolism, blood pressure and lipid metabolism (Figure 1). These scientific data are substantiated by the common clinical observation that many patients who gain weight develop these metabolic abnormalities, and if these same patients subsequently lose weight, then these metabolic abnormalities improve or may potentially resolve. If it becomes generally accepted that dysfunction of adipose tissue is the root mechanistic pathophysiologic aetiology of the metabolic abnormalities that compose the metabolic syndrome in the majority of patients, then it may be time to replace the term metabolic syndrome (a term largely reflective of sometimes unrelated consequence), and replace it with ‘adiposopathy’ (a term more focused on unified underlying causality). It may then be time to focus on adiposopathy as a primary treatment target, and to establish clear criteria for its diagnosis, and for granting an indication for its treatment. It may be time to acknowledge that adiposopathy is not only a disease, but a genetic disease state of a specific organ that:

•	Can be worsened by poor dietary and lifestyle habits, concurrent diseases, and perhaps certain drugs, such as corticoid steroids and some psychotropic drugs [1];
•	A disease that leads to significant morbidity and mortality;
•	A disease that can be improved with favorable dietary and lifestyle habits and pharmaceutical agents; and
•	A disease whose correction will decrease morbidity and mortality

Subsequently, clinical trials can be better designed and directed towards establishing a more rational direction of treatment compared with the current focus upon the often unrelated components of the metabolic syndrome.

Sick fat (adiposopathy) causes Type 2 diabetes mellitus

Overall, only approximately 12% of US adult patients with a BMI ≥27 kg/m² have Type 2 diabetes mellitus [102]. Conversely, 67% of US patients diagnosed with Type 2 diabetes mellitus have a BMI ≥27 kg/m², while 46% have a BMI greater than 30 kg/m²[103]. Thus, not all patients who are overweight have Type 2 diabetes mellitus, in fact only a minority do, (Figure 2) and not all patients with Type 2 diabetes mellitus are overweight, although the majority are. Therefore, while excessive body fat clearly increases the risk of Type 2 diabetes mellitus, excess body fat alone is not sufficient towards development of Type 2 diabetes mellitus.

Admittedly, adiposopathy is not the only cause of Type 2 diabetes mellitus. Other causes would include metabolic disorders that affect pancreatic function, such as some cases of hemochromatosis and chronic pancreatitis. Known metabolic abnormalities that might directly promote insulin resistance include hypercortisolism, excessive growth hormone and hyperthyroidism. Furthermore, certain populations exist that may have an inherent decrease in pancreatic function.

However, for the majority of patients it is reasonable to conclude that it is the combination

of adiposity, genetic predisposition and perhaps a sedentary lifestyle that leads to fat dysfunction, and it is this adiposopathy that results in Type 2 diabetes mellitus (Figure 1).

While in the past adipose tissue was thought to mainly function as an inert storage organ, adipose tissue is now known to be a very active organ from many metabolic standpoints (Table 2) [14]. Several fat-derived metabolites, hormones, enzymes, cytokines and other factors may have varying degrees of effects upon the activity of insulin, including the induction of insulin resistance (Table 3)[15,16].

One of the most well-described consequences of adiposopathy, if not one of the most well-described results of abnormal fat function, is the abnormal increase in the fasting and postprandial release of nonesterified or free fatty acids (FFAs) [15]. Short-term (2 to 6 h) elevations in FFA levels may enhance insulin secretion. However, longer term FFA exposure is detrimental, and may result in insulin resistance in the liver and muscle, as well as diminished pancreatic β-cell insulin production – which has sometimes been termed ‘lipotoxicity’ [15]. The excessive release of FFA from fat cells appears to be more prevalent in patients genetically predisposed to development of Type 2 diabetes mellitus, and a reduction in FFA levels has been shown to improve insulin activity [15].

In addition to the abnormal release of FFA, adiposopathy is also manifest by the abnormal release of cytokines and other factors that contribute to diminished insulin activity (Table 3). Finally, adiposopathy often appears to be manifested by not only abnormalities in fat function, but also abnormalities in fat distribution. Patients with visceral, upper body fat (android) distribution tend to have more insulin resistance and hyperinsulinemia compared with those with more lower body fat (gynoid) [15]. In fact, it might be argued that the accumulation of visceral and subcutaneous truncal fat reflects a dysfunctional adipose organ, and is a gross anatomical manifestation of adiposopathy.

Just as there is great variance in the functionality and dysfunctionality of adipose tissue, there is also variance in the distribution of body fat among individuals and populations, and adipose tissue function and distribution are related.

For example, some obese patients are metabolically healthy, and have been characterized as ‘metabolically healthy, but obese’ (MHO). [17]. These obese patients appear to be ‘protected’ or ‘resistant’ [17] to the typical metabolic consequences often associated with excess body fat. It has been suggested that MHO is not an unusual presentation, accounting for as much as 20% of the obese population [17]. While it may initially seem counterintuitive that so many obese patients could present with the absence of impaired glucose intolerance, hypertension or dyslipidemia, one might speculate that those with MHO simply represent individuals who have little to no genetic predisposition to adiposopathy. In other words, irrespective of their increased body fat, MHO patients are likely to maintain normally functional adipose tissue as their fat mass increases. In fact, the additional adipose tissue in these individuals may provide additional ‘healthy’ fat functionality, and thus account for a protective effect. As such, it has been suggested that weight loss in these patients might be ‘counterproductive’ and potentially harmful [17,19].

By maintaining favorable fat function and avoiding the adverse metabolic consequences of adiposopathy, MHO patients do not develop high blood pressure, or abnormalities of glucose and/or lipid abnormalities. With regard to fat distribution, it is interesting that MHO patients have been shown to have less visceral fat than obese patients with obesity-related metabolic abnormalities, such as those found with the metabolic syndrome [17]. Thus, a significant percentage of the general population appears not to be predisposed to adiposopathy, irrespective of the onset of obesity. This lack of adiposopathy predisposition is manifested by a lack of obesity-related pathological fat dysfunction, a lack of obesity-related high blood pressure and metabolic abnormalities of glucose and lipid metabolism and a relative lack of accumulation of visceral fat.

Another subgroup of patients that underscores the need to focus on fat function and/or dysfunction includes those who are ‘metabolically obese, but normal weight’ (MONW). These individuals are often young, ‘normal’ weight individuals with premature signs of insulin resistance, hyperinsulinemia and dyslipidemia [17]. It has been suggested that the fat dysfunction of these individuals correlates to an increase in intra-abdominal, or visceral fat, and that increased visceral fat (relative to subcutaneous peripheral fat) is dysfunctional and metabolically harmful [17,20,21]. It has also been suggested that fat dysfunction may correlate with subcutaneous truncal, but not subcutaneous

peripheral fat accumulation [20–23]. Thus, adiposopathy appears to be not only associated with abnormalities in secretion of adipocyte hormones, cytokines, factors, and other molecules, but also associated with an increase in intra-abdominal or possibly subcutaneous truncal fat which all leads to an increased risk of. high blood pressure, and abnormalities of glucose and lipid metabolism.

Taken together, MHO and MONW patients demonstrate the point that it is not simply the abnormal increase in normal adipose tissue that reliably predicts subsequent adverse metabolic consequences of adiposopathy. Instead, it is the sufficient presence of or sufficient increase in inadequately functional or blatantly dysfunctional adipose tissue that is the essential, underlying pathology. Emerging from this paradigm are two other concepts: the ‘obesity metabolic cycle’ and ‘adipose autodysharmonia’.

As noted above, the predisposition to dysfunctional adipose tissue is largely genetic, and in many cases exacerbated by adiposity and sedentary lifestyle (Figure 1). One of the more important consequences of adiposopathy is the promotion of insulin resistance and hyperinsulinemia. In an environment where adipose tissue remains relatively sensitive to insulin (a growth factor) while other organs such as skeletal muscle and liver become more and more insensitive to insulin, hyperinsulinemia may ensue, which in turn may lead to increased adipose tissue and subsequent worsening of adipose function. This worsening adiposopathy may cause yet further insulin resistance among other body organs. Adiposopathy, followed by insulin resistance and hyperinsulinemia, followed by worsening adiposity and adiposopathy, followed by even greater insulin resistance has been termed the ‘obesity metabolic cycle’ [1]. The fact that increased insulin levels may increase adipose tissue mass may help to explain why antidiabetes drugs that lower blood sugar through increased insulin levels may further increase body weight, while antidiabetes drugs that lower blood sugar through increasing insulin sensitivity (without increase or perhaps decrease in insulin levels) are often associated with no weight gain, or perhaps modest weight loss [1].

Also as noted above, the location of adiposity may provide clues as to when adipose dysfunction is most likely to occur. Indeed, studies of human fat depots (abdominal subcutaneous, mesenteric and omental) have shown that even after multiple doublings, preadipocytes retain the characteristics of the fat depots from which they had originated [24]. This suggests that subcutaneous and visceral fat may be genetically predisposed to have significantly different functions. Thus, within the same individual, various fat depots may exhibit functional conflicts in which well-functioning subcutaneous peripheral fat may be providing favorable metabolic benefits, while at the same time, malfunctioning subcutaneous truncal and visceral fat may be inciting detrimental metabolic effects, a fracas situation that might be termed ‘adipose autodysharmonia’.

An extreme clinical example on how dysfunction of adipose tissue may result in diabetes mellitus, even in absence of obesity, is demonstrated in patients with lipodystrophy, a disease where lack of adipose tissue and leptin deficiency results in excessive caloric intake, increased plasma nonesterified free fatty acids and elevated triglycerides concentrations. In many individuals who lack sufficient functional fat, ‘ectopic’ fat is deposited in skeletal muscle and liver, with postreceptor defects in insulin action at the level of skeletal muscle [26]. In other words, lipodystrophy is an example of how the metabolic abnormalities of fat metabolism in absence of obesity may result in similar consequences on glucose disposal typically seen in many obese patients.

Yet another illustrative example of how abnormal function of fat tissue (adiposopathy) results in metabolic consequences leading to an increased predisposition to Type 2 diabetes and CHD can be found amongst Asian Indians. Many persons originating from the Indian subcontinent manifest insulin resistance, even in the absence of obesity [27–29]. A study of this population has revealed that FFA levels are higher and insulin-mediated FFA suppression is impaired in Asian Indian men compared with Caucasians [27]. In addition, leptin levels are higher and adiponectin levels are lower in Asian men compared with Caucasians [27]. Furthermore, non-obese Asian Indians typically exhibit high levels of C-reactive protein [29]. From these scientific observations, it has been proposed that the increased genetic predisposition to adiposopathy, coupled with increasing body fat (adiposity), that accounts for the increased prevalence of Type 2 diabetes mellitus and CHD in this ethnic group [27]. Thus, this is an illustrative example of how adiposopathy (not necessarily adiposity) has been shown to be the root mechanistic cause of many of the same metabolic abnormalities found with the metabolic syndrome in a specific ethnic group. In fact, it has even been suggested that the metabolic syndrome may not be so uncommon even among the general US population of mildly overweight adults [32]. Some estimate that the prevalence of the so-called MONW patients may be as high as 13–18% [17].

Sick fat (adiposopathy) causes high blood pressure

Among overweight adult patients with a BMI between ≥25 kg/m² and less than 30 kg/m², 22–33% have hypertension, and among patients with BMI ≥30 kg/m², greater than 40% have hypertension [103]. Conversely, the prevalence of hypertension among adults who are not overweight (BMI less than 25 kg/m²), is less than25%, yet hypertension is still found in these nonoverweight individuals [102,103,105]. So while BMI is directly associated with high blood pressure, particularly among patients less

than 60 years of age [33], not all patients who are overweight have hypertension (although many do) (Figure 2), and not all patients with hypertension are overweight. Therefore, while excessive body fat clearly increases the risk of hypertension, excess body fat alone is not sufficient towards development of hypertension.

Admittedly, adiposopathy is not the only cause of hypertension. Other causes would include pheochromocytoma, primary hyperaldosteronism, hypercortisolism, hyperthyroidism, renal artery stenosis and various kidney diseases. Furthermore, certain familial or genetic syndromes exists that strongly predispose individuals to hypertension.

However, the majority of patients (> 90%) with high blood pressure have what is called ‘essential hypertension’. Adiposity is perhaps the greatest risk factor for essential hypertension. Excessive body fat may lead to promoters of increased blood pressure, such as fat cell secretory products that result in increase mineralocorticoid release [34], and effects upon the renin–angotensin system (Table 2), and sympathetic nervous system, all which represent adipose dysfunction (adiposopathy). Furthermore, independent of adipose function or dysfunction, excess body fat may physically compress the kidney and increase the incidence of sleep apnea associated with obesity – which also may increase the risk of hypertension.

Sick fat (adiposopathy) causes dyslipidemia

Among US adult patients with a BMI between ≥25 kg/m² and 30 kg/m², 19–30% have hypercholesterolemia (≥240 mg/dl). Among patients with a BMI >30 kg/m², greater than 20–30% have hypercholesterolemia [103,105]. Conversely, among US adults who are not overweight (BMI <25 kg/m²), less than 25% have hypercholesterolemia, although hypercholesterolemia is still present in these nonoverweight individuals [102,103,105]. So while BMI is directly associated with dyslipidemia, particularly among patients less than 60 years of age [33], not all patients who are overweight have hypercholesterolemia, although many do (Figure 2), and not all patients with hypercholesterolemia are overweight. Therefore, while excessive body fat increases the risk of hypercholesterolemia, excess body fat alone is not sufficient towards development of hypercholesterolemia. Furthermore, while there does appear to be a linear relationship between adiposity and blood pressure, glucose levels and waist circumference (with an inverse linear relationship with HDL-C levels), this relationship appears to peak within a BMI range of 30–40 kg/m² for alipoprotein B, low density lipoprotein cholesterol (LDL-C) and triglycerides, with subsequent decreased levels with increasing obesity. This suggests that morbidly obese patients may have some lipid risk profiles more favorable than less obese patients [35], and again underscores the complexity of the relationship of adiposity with fat function.

Admittedly, adiposopathy is not the only cause of dyslipidemia. Other secondary causes would include hypothyroidism, diabetes mellitus and certain types of liver or kidney diseases. Furthermore, specific genetic abnormalities may be present, such as familial hypercholesterolemia, which is due to defective or absent LDL particle receptors, and results in severe hypercholesterolemia irrespective of body fat.

Although hypercholesterolemia may be associated with adiposity, the type of dyslipidemia most associated with the metabolic syndrome and Type 2 diabetes mellitus is fasting hypertriglyceridemia, low HDL-C levels (Table 4), and abnormalities of lipoprotein particle size and subclass distribution (Table 5) [2,36]. Figure 3 illustrates how adiposopathy may lead to increased FFA release, fatty liver, and then to fasting hypertriglyceridemia as the result of increased very-low-density lipoprotein (VLDL) particle production. The subsequent exchange of cholesterol with triglycerides between VLDL and HDL particles through cholesteryl ester transfer protein, as well as LDL particles, results in triglyceride-rich HDL and LDL particles. This process also leads to cholesterol-rich VLDL and IDL-atherogenic remnant particles. Once the triglyceride-rich HDL particles undergo lipolysis through various lipases, the HDL particle becomes smaller and more dense, leading to increased renal clearance and decreased HDL blood levels; this is associated with an increased risk of CHD. Once the triglyceride-rich LDL particles undergo lipolysis through various lipases, the LDL particle becomes more small and dense, leading to what is termed ‘pattern B’; this is also associated with an increase risk of risk of CHD (Table 6).

Thus, it is reasonable to conclude that it is the combination of adiposity, genetic predisposition, and perhaps sedentary lifestyle that leads to fat dysfunction, and it is this adiposopathy that results in an atherogenic dyslipidemia [39].

Adiposopathy: treatments

Current treatments for adiposopathy include those that treat obesity (as it is obesity that often exacerbates the root cause of the metabolic syndrome) and those that treat the metabolic

consequences, such as drug treatments for Type 2 diabetes mellitus, hypertension and dyslipidemia.

Antiobesity agents

Adiposity, along with sedentary lifestyle in genetically susceptible individuals, leads to worsening adiposopathy. It has been theorized that enlargement of fat cells alone might largely account for fat cell dysfunction [15]. In fact, it is the enlargement of fat cells that best correlates with insulin resistance as compared with other measures of adiposity, and enlarged fat cells are a strong, independent predictor of the development of Type 2 diabetes mellitus [15]. Current antiobesity agents such as orlistat (Xenical^®) and sibutramine (Reductil^®, Merida^®) have both been shown to reduce body weight and also improve many metabolic abnormalities of glucose and lipid metabolism that are otherwise associated with increased CHD risk [1]. This illustrates an important principle that: ‘the development of any effective antiobesity agent must not only reduce fat mass (adiposity), but must also correct fat dysfunction (adiposopathy) in order to maximize metabolic health’ [1]. Some investigational antiobesity agents in development have already demonstrated reduction in the incidence of the metabolic syndrome [1]. In order to gain approval for clinical use, other antiobesity agents in development will likely have to demonstrate similar improvements in metabolic function.

Antidiabetes drugs

Antidiabetes drugs may also improve glucose metabolism, and thus improve the hyperglycemia consequence of adiposopathy. Current antidiabetes agents have little impact upon blood pressure, and varying effects upon lipid levels (Table 7)[40]. It is with special interest that some antidiabetes agents may improve fat differentiation and function, while at the same time, paradoxically increase body weight. Through a large number of DNA microarrays, it has been shown that large differences exist between lean and obese mice in expression of genes. Specifically, the expression of many genes normally associated with adipocyte differentiation appeared to be downregulated with obesity [41]. Recruitment of functional fat cells through differentiation is one potential treatment option towards improving glucose metabolism through improved fat function and reduction of the metabolic consequences of adiposopathy.

Peroxisome proliferator-activated receptor (PPAR) γ agonists [16], such as thiazolidinediones (TZDs), represent antidiabetes agents that improve glycemic control, enhance hepatic and muscle insulin sensitivity, and improve β-cell function. TZDs are associated with weight gain directly proportional to the reduction in hemoglobin A1c (HbA1c) [15]. Additionally, obese individuals appear to respond better to TZDs than lean subjects. This may be attributable to the fact that PPARγ is a critical transcription factor in the differentiation of preadipocytes into adipocytes [42]. Thus, by recruiting more functional fat cells, TZDs cause a marked reduction in plasma FFA concentration and inhibit lipolysis in patients with Type 2 diabetes mellitus [15]. TZDs may also:

•

Inhibit the expression of the leptin gene in adipocytes with a decline in leptin levels; Table 8. Prevention trials of lipid-altering therapy including patients with diabetes. Trial Diabetic number Total number in study Lipid-altering drug (mg/day) CHD risk versus placebo in diabetic patients (%) Primary prevention         CARDS* 2838 2838 Atorvastatin 10 -37 (p = 0.001) AFCAPS 155§ 6605 Lovastatin 40‡ -44 (p = NS) HPS 2912 7150 Simvastatin 40 -33 (p = 0.0003) ASCOT 2532 10,305 Atorvastatin 10 -16 (p = NS) PROSPER 623 5804 Pravastatin 40 +27 (p = NS) HHS 135 4081 Gemfibrozil 1200 -68 (p = NS) Secondary prevention         4S Re-analysis¶ 202 483 4444 Simvastatin 20–40 -55 (p = 0.002) -42(p = 0.001) CARE 586 4159 Pravastatin 40 -25 (p = 0.05) LIPID** 1077 9014 Pravastatin 40 -19 (p = NS) LIPS 202 1677 Fluvastatin 80 -47 (p = 0.04) HPS 3051 13,386 Simvastatin 40 -18 (p = 0.002) 4D§§ 1255 1255 Atorvastatin 20 -8 (p = NS) VA-HIT‡‡ 769 2351 Gemifbrozil 1200 -32 (p0.004) DAIS§§ 418 418 Fenofibrate 200 -23 (p = NS) Diabetes trials in progress         FIELD 9795 9795 Fenofibrate 200   ASPEN 2200 2200 Atorvastatin 10   Prevention trials of lipid-altering therapy including patients with metabolic syndrome.         Primary         WOSCOPS 1691§§§ 6595 Pravastatin 40 27% Reduction in CHD risk Secondary         4S post hoc analysis 458§§§ 4444 Simvastatin 20–40 52% reduction (p = 0.00009) in major coronary events CDP 563§§§ 8341 Niacin 3000 9% reduction in total 5-year mortality 25% reduction in risk of nonfatal myocardial infarction in metabolic syndrome patients without HDL-C criterion An additional analysis noted that the most pronounced benefits of treatment were seen in metabolic syndrome patients with low HDL-C (< 40 mg/dl) *CARDS was the only primary prevention trial in this group that was prospective. Mean dose was 30 mg. §The figure of 155 in AFCAPS/TexCAPS includes subjects with diagnosis of diabetes (n = 71 placebo and 84 drug treated). However, an alternative figure often cited is a 239 figure that included subjects with diagnosis of diabetes mellitus or fasting glucose ≥ than 126 mg/dL (n = 113 placebo and n = 126 drug treated). ¶The 483 patients represented a reanalysis wherein diabetes was defined according to fasting glucose ≥126 mg/dl. **782 patients identified themselves as having diabetes, with another 295 having probable undiagnosed diabetes based upon fasting glucose levels for a total of 1077. Of the patients on this study, 627 had history of diabetes, with another 142 found to have fasting glucose levels ≥ 126 mg/dl at baseline. This gives the total of 769 patients. §§These were prospective trials, and not post hoc analysis (as were the rest of the studies in this secondary prevention group. §§§ Number of patients with the metabolic syndrome). NS: Not significant (Reprinted from [2] by permission of Cambridge Medical Publications, all rights reserved).

Improve fat distribution with a decrease in intra-abdominal fat;

•

Decrease hepatic fat content associated with an improvement in hepatic insulin sensitivity; and

•

Decrease in intracellular concentration of metabolites of muscle triglycerides that best predicts the improvement in muscle sensitivity to insulin. Table 9. Regulatory considerations for granting approval of a treatment indication for a new metabolic drug [45]. General principles for approval of a new drug Must have reasonable clinical trial data conducted through adequate and applicable methods that demonstrate the drug is safe and effective under the conditions of use when prescribed, recommended or suggested in the proposed labelling. Specific principles for metabolic drugs Although in most cases, no minimum level of efficacy is established for approval of metabolic drug treatments [45,46], in general, these drugs usually require certain objective minimum improvement in target metabolic parameters as weighed against potential risk. Approved antihypertensive drugs have generally achieved > 45 mm/Hg reduction in blood pressure [47,48]. Approved antidiabetes drugs have generally achieved at least about a 1% reduction in hemoglobin A1c. Approved systemic lipid-altering drugs must generally achieve an LDL-C lowering of at least 15%, and perhaps lower (12%) for non-systemic lipid-altering drugs [45]. Approved antiobesity drugs must generally achieve mean placebo-subtracted weight loss ≥ 5% at the end of 1year, with the proportion of subjects who lose ≥ 5% of baseline body weight is greater in drug- versus placebo-treated group [49]. The mechanism of action and known experience of the metabolic drug must be scientifically and reasonably expected to improve patient outcomes, and in cases of new drugs with novel mechanisms of action, demonstrate at least surrogate outcome benefits irrespective of the efficacy on the metabolic treatment target. For example, a novel HDL-raising drug would require at least ≥2 different imaging modalities to obtain an initial approved indication for HDL-raising, possibly with postapproval confirmatory clinical end point studies.

Although some of the weight gain associated with TZDs is due to an increase in fluid retention, it is also true that through the successful recruitment of preadipocytes into adipocytes, some of the resulting fat weight gain is an indicator of the efficacy of the TZD. Other indicators of improvement in adiposopathy with the creation of ‘healthier’ fat with TZDs are:

•	The reduction in FFA levels;
•	Inhibition of resistin, tumor necrosis factor, and plasminogen activator inhibitor (PAI)-1 gene expression in adipocytes and reduction in their circulating levels;
•	Stimulation of adiponectin gene expression in adipocytes and increase in adiponectin levels;
•	Improvement in pancreatic β-cell function with a reduction in islet fat content and preservation of islet cell histology and β-cell mass.

In addition to TZDs, investigational antidiabetes drugs are in development, such as dual PPAR α/γ agents (e.g., tezaglitazar, muraglitazar) [16,43,44] that likewise, have actions upon molecular targets with the potential to improve glucose metabolism, improve dyslipidemia, reduce ‘lipotoxicity’, and generally improve adiposopathy [16,50].

Antihypertensive drugs

Antihypertensive drugs have been shown to reduce CHD events in patients with diabetes mellitus, as well as reduction in other complications of diabetes mellitus [51], with the best choice of first agents being those that might improve endothelial dysfunction (such as with angiotensin-converting enzyme inhibitors and aldosterone receptor antagonists), followed by the liberal use of addition antihypertensive agents in order to maximize blood pressure control [52].

Lipid-altering drugs

Lipid-altering drugs are also one of the few drug treatments that have been shown to reduce CHD outcomes in patients with glucose abnormalities [16](Table 8). Thus, lipid-altering drugs will continue to be recommended for high risk metabolic syndrome patients [5], and will continue to be an important treatment option to reduce CHD events in patients with adiposopathy.

Conclusion

Abnormal fat function, termed adiposopathy, is a major contributing factor in the development of the most common metabolic diseases encountered in the clinical practice of medicine. Through a better understanding of the pathophysiology, and through established criteria

for its diagnosis, the treatment of adiposopathy holds promise for the reduction in morbidities and mortality – particularly through a reduction in CHD, and a reduction in the presence or onset of Type 2 diabetes mellitus, hypertension and dyslipidemia.

Adiposopathy: future perspectives & regulatory considerations

While regulatory indications exist for treatment of the components of metabolic syndrome (e.g., diabetes mellitus, hypertension, and dyslipidemia), an indication for the treatment of the metabolic syndrome itself remains

elusive. The approval of drugs for treatment of adiposopathy and the metabolic syndrome present special challenges. No clinical trial has demonstrated patient outcomes that benefit from a single drug which improves multiple CHD risk factors, and thus regulatory agencies have not found global risk factor reduction as an acceptable criterion to grant drug approval, labelling or promotion [45]. Examples of suggested regulatory criteria that must be met in order to grant approval of a specific indication for a new drug are listed in Table 9.

As can be seen from this table, other metabolic disease drugs (such as antihypertensive drugs, antidiabetes agents and lipid-altering drugs), have fairly clear criteria to obtain an approvable indication. The main reason as to why these drug treatments have specific criteria and metabolic syndrome drugs do not, is because currently, the metabolic syndrome is not itself a treatment target. As a result, the FDA does not recognize ‘a single, universal, measurable, and directly treatable root pathogenic mechanism underlying the syndrome that would establish the metabolic syndrome per se, as a specific treatment target’ [45]. However, if adiposopathy becomes accepted as a directly measurable and treatable root pathogenic mechanism for the vast majority of patients with the metabolic syndrome, this opens the possibility of guidance towards the research and development of drugs that improve fat function and thus treatments for the metabolic abnormalities associated with the metabolic syndrome.

Such an acceptance would allow scientific organizations to first define adiposopathy. Although adiposopathy would likely have a variety of both phenotypic and laboratory criteria for diagnosis that would change over time, this kind of approach and diagnostic criteria is not unprecedented. Systemic lupus erythematosis (SLE) has both phenotypic and serologic criteria for diagnosis, and this classification allows more uniform conduct of clinical trials for drug treatment (Table 10). Similar guidance in the conduct of clinical trials of drug treatment could be achieved through an accepted definition of adiposopathy, that would also likely evolve over time.

So how would adiposopathy best be defined? Table 11 is a proposal for diagnostic criteria of adiposopathy that incorporates both major and minor criteria. As with the SLE criteria described above, clinical, observational and serological findings are included in the diagnostic criteria. One could therefore speculate that a proposed classification of adiposopathy could be based upon analogous criteria. Particularly for the purpose of identifying patients in clinical studies and response to therapy, a person could be said to have adiposopathy if any three or more of the major criteria, with two minor criteria equivalent to one major criteria.

But obviously, much remains to be done to determine which criteria best select for adiposopathy, with particular attention to which aspects of abnormal fat function lead to an increase in CHD risk – the most common cause of mortality in patients so affected, although cancer risk may also be increased. Furthermore, it would be important to discover which consequences of adiposopathy, if appropriately treated with diet, physical exercise or drugs, are most associated with a reduction in clinical morbidity and mortality. But clearly, a greater focus on identifying and evaluating the treatment of fat dysfunction that often leads to the many of components of the metabolic syndrome holds promise as a superior approach in the targeted management of patients. It would allow for a more focused evaluation of the efficacy of interventions such as diet, exercise, and related pharmaceutical agents upon a common, unifying aetiology of the most common metabolic diseases of our time.

In other words, clinical trials directed at correcting fat dysfunction (adiposopathy) would be easier targeted and conducted, compared with clinical trials directed at an array of abnormalities associated with the metabolic syndrome. This is because the trials would then be focused on the underlying root pathophysiologic cause. Once improvement in adiposopathy was shown to improve patient outcomes, then pharmacologic agents might be able to obtain an indication for treatment of adiposopathy, without necessarily having to always prove more hard outcomes benefits for each drug that is developed.

For example, the FDA does not require outcome data for approval of an antidiabetes agent [45]. This is because improving glucose metabolism alone is accepted as being beneficial to patients. Thus, the approved indication of a new antidiabetes agent is most often for the improvement in high glucose levels alone, whose presence may be asymptomatic in many patients. Demonstration of improvement in patient outcomes is not required, such as a reduction in adverse end-organ events attributed to diabetes. Similarly, new antihypertensive drugs can obtain an approved indication for the treatment of high blood pressure alone, despite presenting no clinical symptoms in many patients. Demonstration of improvement in patient outcomes is

not always required, such as proven cardiovascular or renal outcomes benefits. Lipid-altering drugs have often been granted an approved indication for improvement of lipid levels, (despite presenting no clinical symptoms in many patients), without proven benefits towards reduction in CHD.

Just as with the above metabolic diseases, the presence of adiposopathy may also be largely asymptomatic in many individuals. However

•	If it was accepted that adiposopathy was the root pathophysiologic cause of many, if not most cases of patients who express the metabolic syndrome (even without specific symptoms);
•	If it could be shown that adiposopathy was a well-validated predictor of morbidity and mortality;
•	If it can be shown and accepted that improvement in fat dysfunction improves patient metabolic health (such as an improvement in, or a reduction in the onset of Type 2, high blood pressure and dyslipidemia); and
•	If it could be demonstrated that correction of adiposopathy resulted in improved patient hard outcomes, (such as reduction in CHD events);

then it is conceivable that drugs could obtain approval for the indication of treatment of adiposopathy alone. Once the above were established using the model of diabetes mellitus, hypertension and dyslipidemia therapies, drugs with indications for the treatment of adiposopathy would then not necessarily always require very large, difficult and sometimes prohibitively expensive CHD outcomes studies in order for the approval to potentially benefit patients who now number in epidemic proportions.

While the impact upon the clinical and research community of such an approach would be substantial, it would not necessarily be overwhelming. In a practical research example, the future of antiobesity agent development would likely require two parallel programs: one clinical trial programme focused upon treatment of obesity itself through fat reduction (adiposity), and another clinical trial programme focused upon improvement in fat dysfunction (adiposopathy) (Figure 4). To a large extent, this is already being done. Existing antiobesity agents, such as orlistat and sibutramine, have both demonstrated improvement in weight reduction, as well as improvements in many metabolic parameters with reductions in CHD risk factors [1].

It is also noteworthy that rimonabant (currently an investigational selective cannabinoid [CB]-1 receptor antiobesity antagonist [1]) has also essentially followed this parallel development programme approach. Rimonabant has been shown to cause significant weight reduction after 1 year, and thus was effective in reducing adiposity [55]. This weight loss benefit was subsequently found to be extended to 2 years [56]. Rimonabant has also been shown to improve functional parameters associated with adiposopathy:

•	From an adipose tissue organ standpoint, rimonabant decreased waist circumference (presumably resulting in less subcutaneous truncal and visceral fat) [55,56];
•	From an overall metabolic standpoint, rimonabant increased HDL-C levels, reduced triglyceride levels, improved LDL particle size, improved insulin sensitivity (as determined by glucose tolerance testing and homeostasis model assessment), and reduced C-reactive protein levels; and
•	From an adipocyte standpoint, rimonabant increased adiponectin levels, and decreased leptin levels [55,56].

It has been suggested that the metabolic benefits of rimonabant may not totally be explained through weight reduction [56], with the implication that, in addition to its appetite suppressive effects upon the central nervous system [1], CB-1 receptor antagonism may also have direct adipocyte activity. Indeed, animal studies have found that CB-1 receptors are found in adipose tissue [57]. This raises the possibility that the overall efficacy of rimonabant is the result of both body weight reduction through CB-1 antagonism-induced reduction of appetite by its central nervous system effects [1], and direct favorable hormonal and metabolic changes through CB-1 antagonism directly targeted at the adipocyte [57]. Thus, this has been an illustrative example of a parallel antiobesity development programme that has been focused not only on the treatment of adiposity, but also in the treatment of adiposopathy.