Systematic review: adipose tissue, obesity and gastrointestinal diseases


Dr M. A. Mendall, Department of Gastroenterolgy, Mayday University Hospital, Croydon CR7 7YE, UK.




Obesity is increasingly being recognized as a risk factor for a number of benign and malignant gastrointestinal conditions. However, literature on the underlying pathophysiological mechanisms is sparse and ambiguous. Insulin resistance is the most widely accepted link between obesity and disease, particularly colorectal cancer. The recognition that intra-abdominal fat is immunologically active sheds new light not only on the pathogenesis of obesity-related gastrointestinal conditions, but also on inflammatory conditions such as Crohn's disease.


To describe the biology of adipose tissue, its impact on the immune system and explores the possible underlying mechanisms linking obesity to gastrointestinal diseases. It also looks at the role of mesenteric fat in determining severity and course of Crohn's disease.


Relevant English-language literature and abstracts cited on MEDLINE database were reviewed.


Our recent finding of an association between obesity and subclinical bowel inflammation suggests that, apart from promoting generalized immune activation, fat also evokes local immune responses. We propose that the proinflammatory milieu promoted by obesity could underlie many of these associations and that the mechanism implicating insulin resistance may merely represent an epiphenomenon. In Crohn's disease, on the other hand, intra-abdominal fat may provide a protective mechanism.


The potential of adipose tissue as a therapeutic target is vast and needs exploration.


Adipose tissue is no longer regarded as an inert repository of stored fat, but is also an active endocrine organ and modulator of immune function.1, 2 These activities explain the relation between obesity and diabetes, and cardiovascular disease. Mesenteric adipose tissue has been implicated in a wide range of gastrointestinal (GI) disorders from fatty liver, to GI cancers to acute pancreatitis (AP), Crohn's disease (CD)1 and even functional bowel disorders. Associations between obesity and GI diseases to date have mainly considered insulin resistance as the pathogenic link.3

The first section of this review will deal with the biology of adipose tissue and will identify mechanisms whereby adipose tissue could influence GI diseases and the second part will consider the associations of adipose tissue and obesity with GI disorders in the light of these mechanisms. Fatty liver disease will not be discussed further as it has been extensively discussed in the world literature.

Abdominal visceral fat, i.e. the intraperitoneal fat component which includes mesenteric and omental fat.4 Accumulation of intra-abdominal fat (central obesity) correlates more strongly with disease states compared with total body fat.5 Computerized tomography scanning and magnetic resonance imaging are the gold standard for quantifying intraperitoneal fat, and agrees reasonably well with waist-hip ratio used in clinical studies.6–9 This review will concern itself mainly with adipose tissue in this location, although many of the comments apply to adipose tissue in general.

Biology of adipose tissue

This section will consider, at first, the histiogenesis of adipose tissue and the other cellular components of intra-abdominal fat, and the regulation of adipocyte size and number. Secondly, the current state of knowledge of the regulation of synthesis and actions of a variety of metabolically and immunologically active mediators by adipose tissue will be described. Finally, in the light of this a novel mechanism for the association of adipose tissue with GI disorders will be proposed.

Histiogenesis of adipose tissue

Adipose tissue consists not only of adipocytes but also of connective tissue matrix, nerve tissue, stromovascular cells and immune cells (macrophages, T/B lymphocytes).10 All these components contribute to the metabolic and immunological derangements observed in the obese.

Adipocytes, chondrocytes, osteocytes and myocytes originate from common embryonic stem cell precursors.11 Current understanding of the process of adipocyte differentiation is based on studies from preadipocyte cell lines and mice models. The process of adipogenesis begins before birth and accelerates rapidly after birth.10 Even in the adult the potential to generate new fat cells persists, although the capacity decreases with age.

The first cell identifiable in the committed adipocytes lineage is the preadipocyte.11 Further differentiation is preceded by a period of growth arrest during which induction of specific genes through the activated transcription factors peroxisome proliferators-activated receptor (PPAR)-γ and CCAAT/enhancer-binding protein-α (C/EBP-α) takes place.11 This is followed by a period of clonal expansion seen in mice but not in human cell lines.12 During adipocyte differentiation, cells convert from a fibroblastic to a spherical shape, and dramatic changes in cell morphology, cytoskeletal components, and the level and type of extracellular matrix components occur.11 During the terminal phase of differentiation adipocytes acquire sensitivity to insulin and increase de novo lipogenesis. This results from an increase in glucose transporter numbers, increased glucose sensitivity, loss of β-1 adrenoceptors, and increased β-2 and -3 adenoceptors.11, 13, 14

In the obese, both hypertrophy and hyperplasia of adipocytes occurs. Hypertrophy occurs during the initial stages of development of obesity. However, the capacity to hypertrophy is limited and once adipocytes reach a critical size neo-adipocyte differentiation takes place under the influence of growth factors secreted by the hypertrophied adipocytes.11

Cells of the adipocyte lineage share a number of antigenic and functional properties with macrophages.15 A large number of genes encoding cytokines, transcription factors and scavenger receptors are expressed in both lineages. Preadipocytes and adipocytes express MOMA-2 but not F4/80 or CD11b, all of which are surface markers for the monocyte-macrophage lineage. Adipocyte precursors exhibit phagocytosis and microbicidal activities similar to macrophages.15

Obesity is characterized by the progressive infiltration of adipose tissue by macrophages.16, 17 Enlarging adipocytes secrete macrophage chemoattractant protein 1 (MCP-1) that attract macrophages. However, an additional source could be the transdifferentiation of preadipocytes to macrophages. This phenomenon has been demonstrated in animal models, and highlights the plasticity of preadipocytes.18, 19 The progressive dedifferentiation of primary adipocytes has also been reported under the influence of tumour necrosis factor (TNF)-α in certain adipocyte cell lines. It is yet unclear as to whether true reversion to a preadipose phenotype does take place. This phenomenon may play a role in regulating adipocyte number and consequently adipose tissue mass.20, 21


Biologically active proteins are secreted both by adipocytes and, the non-adipocyte fraction of adipose tissue.2 These act both locally and at distant sites influencing various metabolic and immune processes. Proteins synthesized and secreted by adipocytes are termed adipokines. Although the term adipokine was initially used to refer to proteins secreted by all elements of the adipose tissue, it is now specific to adipocytes.22 Collectively, they constitute the adipokinome of the adipocyte. In addition to proteins, adipocytes secrete fatty acids, cholesterol, steroid hormones, prostaglandins and prostanoids and retinol. The lipid substances and adipokinome together are said to constitute the secretome of the adipocyte.22

Well over 50 adipokines have been identified so far. They can be classified based on the processes they regulate, which include lipid homeostasis, immune function, insulin sensitivity, control of blood pressure, haemostasis and appetite and energy balance.22 TNF-α was the first identified molecular link between obesity and insulin resistance.23 But then the increased expression of a number of other adipokines [e.g. interleukin-6 (IL-6), resistin, MCP-1, plasminogen activator inhibitor-1 (PAI-1), leptin, adipsin and acylation stimulating protein (ASP)] has been associated with obesity, coronary artery disease (CAD) and non-insulin-dependant diabetes mellitus (NIDDM)24–27 (Tables 1 and 2). Adipose tissue is also the source of growth factors [e.g. insulin-like growth factor-1 (IGF-1), macrophage colony stimulation factor, transforming growth factor-α], chemokines (IL-8, macrophage inflammatory protein-1α) and ADAM proteins, levels of which are dysregulated in obesity. Levels of vascular growth factors [e.g. vascular endothelial growth factor (VEGF), hepatocyte growth factor] are increased in obesity possibly accounting for the increased incidence of metastatic disease seen in obese cancer patients.28–32

Table 1.   Examples of receptors expressed by adipocytes2
Receptor typeReceptors expressed by adipocytes
  1. PPAR, peroxisome proliferators-activated receptor; TNF, tumour necrosis factor; IL, interleukin.

Cytokine receptorsLeptin, TNF-α, IL-6
Nuclear hormone receptorsAndrogen, oestrogen, glucocorticoid, progesterone, thyroid hormone
Other endocrine hormone receptorsInsulin, glucagon, growth hormone, angiotensin 1 and 2, adiponectin
Catecholamineα1, 2; β1, 2, 3
Other receptorsPPAR-α, -γ, -δ; RXR (retinoid × receptor) prostaglandins
Table 2.   Examples of adipokines and associated diseases2, 24–27
AdipokinesMetabolic regulationEffectsAssociated diseases
  1. MCP-1, macrophage chemoattractant protein 1; TNF, tumour necrosis factor; IL, interleukin; PAI, plasminogen activator inhibitor; ASP, acylation stimulating protein; MIF, macrophage migration inhibitory factor; NIDDM, non-insulin-dependant diabetes mellitus; CAD, coronary artery disease.

TNF-α↑ in obesityPromotes insulin resistanceNIDDM, CAD
IL-6↑ in obesityPromotes insulin resistanceNIDDM, CAD
IL-18↑ in obesityProatherogenic
Promotes insulin resistance
Leptin↑ in obesityModulates immune response
Suppresses appetite
Adiponectin↓ in obesityAnti-inflammatory
Promotes insulin sensitivity
Stimulates fatty acid oxidation
↓ levels in CAD, NIDDM
Visfatin↑ in obesityPromotes insulin resistanceNIDDM,CAD
Resistin↑ in obesityPromotes insulin resistanceNIDDM, CAD
MCP-1↑ in obesityEncourages macrophage migration
Promotes insulin resistance
PAI-1↑ in obesityPromotes insulin resistance
Adipsin↑ in obesityModulates glucose, lipid metabolismNIDDM, CAD
ASP↑ in obesityModulates glucose, lipid metabolismNIDDM, CAD
MIF↑ in obesityInhibits macrophage migration 
C-reactive protein↑ in obesityProinflammatory, atherogenicNIDDM, CAD

As the adipocytes hypertrophy they secrete increasing amounts of these adipokines. Adiponectin is, however, an exception, being secreted preferentially by smaller adipocytes.33, 34 Adiponectin has anti-inflammatory, proinsulin and antiatherogenic properties. Circulating levels are decreased in obesity, CAD and NIDDM.35, 36 IL-10 and IL-1 receptor antagonist (IL-1Ra) are other anti-inflammatory cytokines secreted by adipocytes. In contrast to adiponectin and IL-10, IL-1Ra levels are increased in the obese and may actually contribute to the development of insulin resistance.31

Although intra-abdominal adipose tissue accounts for only 10% of the total body fat, it more strongly correlated with metabolic abnormalities and disease states when compared with subcutaneous fat. This is probably because of the direct drainage to the liver and its close proximity to antigens within the GI tract.37

Regulation of adipokine secretion

Various stimuli have been proposed to explain the overproduction of adipokines in the obese. These include hypoxia, endoplasmic reticular (ER) stress and oxidative stress (Figure 1).22, 38, 39Dealing with each of these mechanisms in turn:

Figure 1.

 Overlapping metabolic and inflammatory pathways in adipocytes and macrophages. IKK, Jun N-terminal kinase (JNK) and protein kinase C (PKC) are activated by Toll-like receptor (TLR) signalling in response to microbial products.43 TLR signalling also inhibits liver X receptor (LXR) activity in macrophages promoting cholesterol accumulation.43 Chronic activation of the innate immune system as in chronic infections and obesity can result in diabetes and atherosclerosis.

As the adipose tissue expands the vasculature is insufficient to adequately oxygenate the adipocytes. The resultant hypoxia initiates an inflammatory response that in turn promotes angiogenesis. The adipocyte response to hypoxia is probably mediated through hypoxia-inducible factor-1 (HIF-1).40, 41 A similar mechanism operates in tumour growth.

The metabolic and structural changes generated by obesity also increase the demand on the endoplasmic reticulum. This is particularly relevant to adipose tissue where marked changes in tissue architecture, increased glucose uptake and increase in protein and lipid synthesis take place. ER stress leads to activation of serine/threonine kinases – inhibitor of nuclear factor-kappa B kinase (IKK) and Jun N-terminal kinase (JNK).42 IKK in turn activates nuclear factor-kappa B (NF-κB) by metabolizing its inhibitor.43 This stimulates the production of inflammatory cytokines TNF-α and IL-6.43

Oxidative stress results from the increased glucose delivery to the adipocytes stimulates reactive oxygen species production by mitochondria.39 This oxidative stress in turn activates IKK and JNK promoting the inflammatory response.

Whatever the mechanism behind the synthesis of proinflammatory cytokines by the hypertrophied adipocytes of the obese, this synthesis decreases with weight loss and is hence reversible.44–46

Adipose tissue, obesity and the immune system

The effects of obesity on the immune system are not restricted to local effects within adipose tissue. Elevated levels of proinflammatory cytokines have been noted in the serum of asymptomatic obese individual; levels corresponding to the degree of obesity.47 TNF-α is only present at very low levels in human blood suggesting that TNF-α released by adipose tissue has only autocrine/paracrine actions. IL-6, however, is present at much higher levels. Adipocyte-derived IL-6 has been estimated to comprise 30% of the circulating IL-6 suggesting an endocrine action.48 Furthermore, these elevated levels of IL-6 are associated with increased circulating levels of C-reactive protein (CRP) suggesting that even although the elevation in levels is modest compared with those seen in sepsis, they could be having real effects on innate immune function.

Obesity is also associated with altered functioning of circulating immune cells.47, 49 Decreased T-/B-cell function, increased monocyte and granulocyte phagocytosis and oxidative burst, and an increase in leucocyte count have been described. More recently, circulating mononuclear cells from the obese have been shown to exhibit increased NFkB nuclear binding with decreased levels of NFkB inhibitor, together with increased mRNA expression of IL-6, TNF-α and migration inhibition factor. These are clear signs of an activated state. Furthermore, there is a good correlation between the markers of macrophage activation and plasma levels of free fatty acids.50 It has previously demonstrated that macronutrient challenges in normal subjects increase NFkB nuclear binding in circulating mononuclear cells, raising the possibility that the activated state of mononuclear cells is due to increased circulating levels of free fatty acids found in the obese. Indeed hyperlipidaemia in mice mediates an inflammatory response by the same signalling cascade through which lipopolysaccharhide activates the innate immune system (this engages a receptor complex comprising Toll 4 CD14, CD14 and MD-2).51

The more recently discovered adipokines leptin, resistin and adiponectin may play a role in regulating systemic levels of inflammation. Leptin for instance stimulates myeloid differentiation from bone marrow precursors and could contribute to the leucocytosis observed in the obese.52 Leptin in vitro may promote monocyte diapedesis, but has been found to improve insulin sensitivity in humans and rodents and is therefore an unlikely contributor to the inflammatory response associated with obesity.53 Resistin in humans is mainly secreted by mononuclear cells either circulating or adipose tissue associated. Levels are elevated in obesity and have been shown to be proinflammatory through activation of NFkB pathways in human endothelial cells. This in turn leads to the endothelial dysfunction which plays a role in the pathogenesis of atherosclerosis.54

Adiponectin is the most widely studied adipokine that has anti-inflammatory properties. It inhibits phagocyte activity and TNF-α production by macrophages and also inhibits TNF-α-induced expression of adhesion molecules through NF-κB signalling pathways.55 Hence, the low levels observed in obesity may have a permissive effect on immune activation systemically.

This then leaves open the question as to whether obesity is additionally associated with evidence of immune activation and inflammation at mucosal surfaces, and if so by what mechanism.

Obesity and tissue inflammation

We have recently studied the determinants of whole gut inflammation by measuring levels of calprotectin in faeces in a population of normal late middle-aged subjects. Calprotectin is a calcium-binding protein found only in neutrophils and monocytes. Levels in faeces correlated well with faecal levels of indium-labelled white cells and inflammatory bowel disease activity. A number of environmental factors known to modulate innate immune activity affected levels, of which obesity was one. Exercise was another. The relationship with obesity was graded throughout the range of body mass index (BMI). A quarter of subjects had faecal levels of calprotectin considered abnormal and of a magnitude seen in inflammatory bowel disease with no associated symptoms. This is the first evidence of increased inflammatory activity at the tissue level at distant sites in normal subjects associated with obesity.56 This could represent an important mechanism linking obesity to a variety of diseases and a range of GI disorders in particular.

A number of processes could underlie this association. First, circulating levels of adipokines could influence the immune response at the mucosal level to gut contents as could elevate circulating levels of free fatty acids. Local paracrine effects of serosal adipocytes could also play a role.

Whether by distant or local mechanisms, adipokines can influence gut permeability and hence exposure to luminal antigens which initiate the immune response. Inflammatory mediators are known to increase paracellular and transcellular uptake of proteins from the gut. Patients with histologically proven quiescent CD have increased gut permeability, attributable to the heightened local expression of TNF-α.57 Treatment with infliximab (anti-TNF) is associated with decrease in gut permeability in patients with active CD.58 Furthermore, studies on enterocyte cell lines have supported the role of TNF-α as a potential regulator of endosomal uptake of antigens at physiological concentrations.57, 59 Elevated levels of IL-6 have been associated with increased gut permeability following haemorrhagic shock.60 Activation of macrophages evokes a similar response.61

Whatever the mechanism behind the association between obesity and GI diseases, hyperinsulinaemia is unlikely to play a role as it has been demonstrated that insulin has anti-inflammatory effects. Many of the associations between GI cancer and obesity have hitherto been attributed to insulin resistance. This may, however, merely be an epiphenomenon. These associations and others will now be considered in this new light.

Obesity, adipose tissue and GI disease

Obesity is a known risk factor for oesophageal, gastric cardia and colorectal carcinoma.1 Its role in determining the severity of AP and more recently, its association with functional bowel disorders are well documented.62, 63 The association of these disorders in the light of our knowledge of adipocyte biology will be discussed.

Crohn's disease, on the other hand, whilst not necessarily being associated with obesity, is associated with marked changes in mesenteric adipose tissue behaviour, which may also be illuminated from the above discussion of adipocyte biology.64, 65 This will be considered separately to the other disease associations.


Until recently, the most widely accepted mechanism for the association between obesity and GI cancers was insulin resistance. Our findings of an association between obesity and GI inflammation at the tissue level opens up inflammation as an alternative mechanism. These will be discussed below (Figure 2).

Figure 2.

 Pathophysiological mechanisms linking obesity and GI cancers. Chronic innate immune activation is central to the metabolic effects of obesity and the consequent end organ damage. In addition, the effects on the GI tract could be mediated through paracrine effects of the adjoining mesenteric fat.

Insulin resistance

Obesity is associated with a state of insulin resistance. Candidate molecules implicated include increased circulating levels of free fatty acids, leptin and TNF-α and, decreased levels of adiponectin. Insulin resistance is associated with elevated levels of insulin and IGF-1 (Figure 3). IGF-1 inhibits apoptosis and promotes cell cycle progression.66 An increase in cell turnover enhances the rate at which molecular alterations occur during the development of cancer. Both normal colon epithelia and cancer cells express IGF-1 receptors.67 In colon cancer cell lines IGF-1 induces the production of VEGF, an angiogenic factor that promotes tumour growth.68

Figure 3.

 Regulation of insulin-like growth factor (IGF). IGF-1 and IGFBP are produced in the liver. Part of the IGF-1 is bound to the binding protein and is inactive. Production of both these factors is primarily under control of growth hormone.68 Hyperinsulinaemia inhibits IGFBP production and enhances the effect of GH on the liver by promoting GH receptor proliferation. IGF-1 and insulin act on colonic epithelium promoting cell turnover.69–74

Insulin is known to be anti-inflammatory and its use in the critically ill for stringent control of blood sugar is associated with improved outcome.69 It acts by suppressing three major proinflammatory transcription factors – NF-κB, activator protein-1 (AP-1) and early growth response-1 (EGR-1).70 Although certain epidemiological and cell culture studies have demonstrated a link between hyperinsulinaemia and colorectal neoplasia, this is most likely to be an epiphenomenon, as anti-inflammatory agents are known to reduce the risk of GI cancers. If at all, its tumorigenic potential is likely to be mediated through its effect on increasing IGF-1 levels. However, the evidence linking IGF-1 and colorectal tumorigenesis is equivocal with recent research showing no statistically significant results when a Bonferroni adjustment was applied.71, 72 This questions the validity of this argument.


This alternative mechanism implicates the NF-κB inflammatory pathway. Activated NF-κB induces proinflammatory-related genes inducible nitric oxide synthase (iNOS) and cyclo-oxygenase-2 (COX-2). Expression of these genes promotes cell survival whereas non-expression leads to apoptosis.73, 74 Long-term activation of this pathway as seen in chronic inflammation and obesity promotes tumorigenesis. Activated NF-κB, iNOS and COX-2 are overexpressed in many solid tumours and cancer cell lines.75 iNOS inhibits capases – the proteases involved in apoptosis, directly damages DNA and inhibits DNA repair mechanism.76, 77 Epidemiological studies have shown a decreased risk of colorectal and oesophageal adenocarcinoma (OA) following regular use of non-steroidal anti-inflammatory agents (NSAIDs). Although the mechanism is unclear it is probably related to COX-2 inhibition by NSAIDs. We propose that the proinflammatory effects of obesity act via these mechanisms are at least in part responsible for the associations of obesity with GI cancers.

Obesity and colorectal cancer

Obesity is a risk factor for colorectal cancers, the association being stronger in men than women. In women, this correlation exists at younger ages but is less evident at older ages.78 Type 2 diabetics have a 40–60% increased risk of colorectal cancers, more so of the proximal large bowel.79 The ‘adenoma-carcinoma’ sequence was used to explain occurrences of sporadic colorectal cancer in man.80 It defined stepwise molecular and histological alterations that led to cancer evolution.

The inflammation-dysplasia-carcinoma sequence is used to explain cancer development in patients with long-standing ulcerative colitis.81 Histologically, the sequence starts in the inflamed mucosa as a hyperplastic lesion, which progressed sequentially through dysplasia into adenocarcinoma. Genetic mutations in the APC, p53, SMAD4 and K-ras tend to occur with similar frequency in both sporadic CRC and colitis-associated CRC although their relative order may differ.82 It would thus appear that once initiated both these processes follow a similar course.

These findings suggest the possibility of sporadic colorectal cancer arising in a background of subclinical bowel inflammation. As obesity is associated with elevated levels of asymptomatic bowel inflammation a causal mechanism is plausible. Epidemiological studies and animal models have confirmed the role for IGF-1 in colorectal carcinogenesis.83 Hence, it is possible that obesity could promote cancer through both inflammation as well as insulin resistance (Figure 2).

An attraction of the inflammation hypothesis over the insulin resistance hypothesis was that many other environmental risk factors for colorectal cancer were associated with altered levels of bowel inflammation in a direction to favour carcinogenesis. Lack of exercise, smoking and reduced fibre intake were all associated with increased levels of bowel inflammation.56 The possibility that insulin resistance is merely an epiphenomenon raises itself.

Obesity and gastro-oesophageal adenocarcinoma

Obesity and gastro-oesophageal reflux disease (GERD) are recognized risk factors for OA.84

Histologically, OA disease begins in an area of hyperproliferative epithelium, which then progresses sequentially through metaplasia (Barrett's oesophagus), dysplasia and carcinoma in situ into invasive carcinoma.85 iNOS and COX-2 are overexpressed in Barrett's and OA.86 Several studies have suggested that NSAIDs protect against OA.87–89 This points to an important role of inflammation in the pathogenesis of OA.

The risk of OA associated with obesity may not solely explained by the promotion of reflux. An association between increasing histological, flow cytometric, and genetic abnormalities and increasing abdominal fat distribution in Barrett's epithelium has been observed.90 Moreover, a large epidemiological study has suggested that a high BMI may promote the progression of Barrett's to OA, whilst not playing a role in the initial development of Barrett's oesophagus.91 The promotion of inflammation by immune mechanisms or hyperinsulinaemia could be important contributory mechanisms besides the chemical injury associated with reflux to the development of OA.

Obesity and pancreatitis

Acute pancreatitis is unique in the ability of localized inflammation to cause a massive and inappropriate immune response – otherwise known as the systemic inflammatory response syndrome (SIRS) that can lead to multiorgan failure and death. Pancreatic necrosis rather than interstitial inflammation is important in determining the severity of SIRS as well as local complications.62

A recent meta-analysis has confirmed that obesity is a risk factor for severity of AP.62 It correlated strongly with the risk of systemic and local complications in particular. Pancreatic necrosis and sepsis are the commonest local complications.62

One mechanism to explain the association of severity with obesity is the larger quantity of peripancreatic fat and the resulting microcirculatory disturbances.92 However, the balance between apoptosis and cell necrosis may be important. Acinar cell death is the hallmark for AP, and occurs by cell necrosis and apoptosis. Apoptosis, when compared with cell necrosis is associated with a less severe inflammatory response.93, 94 NF-κB activation is an early feature in acute experimental pancreatitis.73, 74 NF-κB is known to activate antiapoptotic mechanisms and hence could promote acinar death by necrosis. The adverse effect of obesity on pancreatitis could thus be attributed to its promoter effect on NFkB activation.75, 95

Inflammatory mediators play a critical role in its pathogenesis of SIRS and the balance between proinflammatory and anti-inflammatory mediators determines severity and outcome. Proinflammatory mediators involved include TNF-α, IL-6, IL-1b, PAF, intracellular adhesion molecule 1 (ICAM-1), IL-8, MCP and substance P. Anti-inflammatory mediators involved include IL-10, C5a and sTNFR.96–100 Adipose tissue contributes to the proinflammatory pool and could hence worsen the course of SIRS by this mechanism.

Crohn's disease and mesenteric adiposity

Unlike the other diseases mentioned in this review, CD is not associated with obesity. However, it is increasingly becoming clear that mesenteric fat is likely to play an important role in its pathogenesis. Fat wrapping and mesenteric adipose tissue hypertrophy are consistent features recognized on surgical specimens in patients with CD.101–103 Historically, these were attributed to transmural inflammation and the cytokines released from gut lymphoid tissue.104 However, recent research suggests that the hypertrophied fat contributes actively to disease severity and may influence onset of complications.64, 65

In CD, mucosal ulcers is most pronounced along the mesenteric attachments.104, 105 On the other hand, ulcers in intestinal tuberculosis are oriented transversely, and in infectious diseases like salmonellosis and shigellosis the ulcers are located along the antimesenteric border near the Peyer's patches.104–107 This could suggests a causal link between mesenteric adipose tissue and mucosal ulceration.

Histopathological features of the fat in CD include fibrosis, perivascular inflammation, intimal and medial thickening of vessels, significant infiltration of inflammatory cells and an increase in the number but decrease in the size of adipocytes.66, 108, 109 Inflammatory cells are mainly CD68- and CD3-positive T lymphocytes. Massive infiltration by these cells suggests that the inflamed mucosa and its adjacent mesentery share a common inflammation in CD.64, 65 The hypertrophied mesenteric fat is a major source of the increased TNF-α, IL-6 and other circulating proinflammatory cytokines seen in these patients which contribute to the debilitating systemic symptoms seen in this disease.64, 65

It is likely that the mesenteric fat in CD is exposed to gut microbial antigens. Adipocytes express Toll-like receptors (TLRs) and CD14, which can interact with these antigens activating NF-κB pathways.110 By a yet unknown mechanism simultaneous PPAR-γ activation takes place leading to adipocyte proliferation and reduction in size. The decrease in size of adipocytes is a result of redistribution of stored fats. Adiponectin production is increased in CD in contrast to that seen in obesity, because adiponectin production is inversely related to adipocyte size.

Mesenteric adipose tissue hyperplasia is present at disease onset, and is not affected by severity, or duration of disease.64 An inverse relationship between adiponectin levels and both CRP and IL-6 noted in CD supports the anti-inflammatory role of adiponectin and possibly of mesenteric fat. Interestingly, patients with fistulating disease have relatively lower adiponectin expression, the cause of which is unknown.65 These findings are consistent with the hypothesis that the hypertrophied fat serves as a physical and chemical barrier preventing spread of inflammation to the intra-abdominal space. Defects in this barrier may predispose to fistulation and perforation.

Although obesity does not increase the risk of CD, it is associated with increased disease severity and anorectal complications.111 The hepatic steatosis observed in CD maybe the result of prolonged exposure of the liver to elevated levels of inflammatory cytokines.112 It is likely that mechanisms similar to those seen in the other described diseases could contribute to disease severity in the obese.

Obesity and functional bowel disorders

Recent epidemiological studies have shown an association between obesity and functional bowel disorders. A positive correlation was noted between increasing BMI and abdominal pain associated with diarrhoea (>3/day, loose stools, or urgency).113, 114 The underlying pathophysiological mechanisms are unknown, but low-grade inflammation could play a role.115 Further research in this field is warranted.

Therapeutic implications

The recognition of adipose tissue as an organ contributing actively to disease pathology offers a number of potential therapeutic targets. Approaches that reduce adipose tissue depots, including surgical fat removal, exercise, and reduced caloric intake, improve proinflammatory adipokine levels and reduce the severity of their resultant pathologies. Several drugs used to treat insulin resistance, hypertension and hypercholesterolaemia alter circulating adipokine levels favourably.43–45 However, their impact on obesity-associated GI diseases is unclear.

Synthetic ligands of PPAR-γ such as the glitazones are commonly used in the treatment of NIDDM. The beneficial effects on the inflammatory environment noted in obese rats include a decrease in circulating levels of TNF-α, blood sugar and leptin and, an increase in adiponectin.13 Their use as prophylactic agents in obesity warrants investigation. In animal models of intestinal inflammation, glitazones decrease tissue injury associated with immune activation and hence, may be beneficial in the treatment CD.116

Already anti-inflammatory drugs have a role in cancer chemoprevention. The use of statins significantly reduces the risk of colorectal cancer.117 An unanticipated by-product of drugs used in the prevention of cardiovascular disease many of which have anti-inflammatory properties, could be the prevention of GI cancers and amelioration of other GI diseases related to obesity.


The proinflammatory effects of obesity could underlie many of the associations described, with hyperinsulinaemia merely being an epiphenomenon of this process. The behaviour of mesenteric fat in CD highlights the immunological role of adipose tissue upon stimulation even in the lean. It appears that obesity worsens the course of non-obesity-related GI disorders either through systemic effects or possibly through paracrine effects of the adjoining fat. A more useful understanding of the role of obesity and intraperitoneal adipose tissue in GI diseases comes from an appreciation of the proinflammatory and sometimes anti-inflammatory actions of adipose tissue.


No external funding was received for this study.