Epidemiologic studies show that the incidence and prevalence of IBD vary significantly depending on geographic location and patients’ racial or ethnic backgrounds. IBD occurs worldwide and exhibits a relatively low incidence in Asian, Mediterranean, and Middle Eastern countries and a higher incidence in European countries, the United States, Canada, Australia, and New Zealand. This may reflect racial, ethnic, and genetic factors. Prevalence rates for IBD among non-Caucasians in the United States are lower than rates for Caucasians. In one study, the prevalence of Crohn disease was 43.6 per 100,000 population for Caucasians, 29.8 per 100,000 for African Americans, 5.6 per 100,000 for Asians, and 4.1 per 100,000 for Hispanics (4). A study of African-American children reported a Crohn disease incidence of 7 to 12 cases per 100,000 (5). However, recent data indicate that the incidence of IBD among African Americans and second-generation South Asians is increasing (6). Among ethnic groups, Jews in the United States have the greatest risk for developing IBD compared with non-Jewish Caucasians. The incidence rate is two to four times greater and the prevalence two to nine times greater in this group. Ashkenazi Jews exhibit a particularly high IBD risk, especially those originating in Middle Europe, Poland, or Russia.

Chronic IBD is a common inflammatory disorder of unknown etiology. The intrinsic complexity of IBD and its variable manifestations hampers progress in understanding its pathogenesis. However, currently favored theories implicate genetically determined, immunologically mediated mechanisms of injury. The fundamental pathogenic question could be: Does the chronic recurring inflammatory activity in IBD reflect an appropriate response to a persistently abnormal stimulus (a structural alteration of the intestine or causative agent in the environment) or an abnormally prolonged response to a normal stimulus (aberrant regulation of immune responses) (Fig. 11.1)? It is conceivable that some factors initiate the disease, whereas others sustain the inflammatory process or possibly even reactivate it.

There is considerable evidence that the development of both CD and UC is determined, at least in part, by genetic factors. Overwhelming evidence exists that both ulcerative colitis and Crohn disease cluster within families. In population-based studies, 5% to 10% of individuals with IBD report having an affected family member (7). In fact, having a family member with IBD represents the greatest risk factor for developing the disease. Individuals with a first-degree relative with IBD have a 10- to 15-fold increased risk of also developing the disease compared with those without an affected family member (8). Approximately 75% of families
with multiple affected members show concordance for disease type (i.e., all affected family members have CD disease or all have UC). In the remaining 25%, some members have CD while others have UC (9). This finding suggests that UC and CD may have some common, as well as distinct, susceptibility genes. Twin studies show that monozygotic twin concordance for CD ranges from 42% to 58% (10). In contrast, concordance in dizygotic twins is only 4% (11). Monozygotic twin concordances are significantly lower for UC (10). These findings suggest that although there is a strong genetic component that determines susceptibility to inflammatory bowel disease, there are also environmental factors that play an important role in disease development.

Both CD and UC represent, at least in part, disorders of both innate (macrophage, neutrophil) and acquired (T and B cell) immunity. It is currently believed that the main abnormality responsible for the development of inflammation in these disorders is a loss of tolerance to enteric commensal bacteria or other pathogens (26). In normal individuals, tolerance is mediated by regulatory T cells, B cells, natural killer T cells, and dendritic cells that produce transforming growth factor beta (TGF-β), interleukin (IL)-10, interferons, and prostaglandin J2. In IBD, lamina propria macrophages and dendritic cells are increased in number, demonstrate an activated phenotype, and express many proinflammatory cytokines and chemokines. In addition, expression of costimulatory molecules and adhesion molecules, vital to the extravasation of macrophages
and neutrophils from the vasculature, is also increased in both CD and UC (27).

Crohn disease is associated predominantly with T_H1 cytokine production (28). Ulcerative colitis, on the other hand, does not fit clearly into either the T_H1 or T_H2 category, although an atypical or modified T_H2 response seems to occur in established UC (26). Although the types of cytokines produced in UC and CD differ somewhat (Table 11.2), both diseases are associated with abnormal immune responses to nonpathogenic commensal bacteria within the gut. Cross-reactivity of peripheral blood and colonic lamina propria CD4+ T cells with indigenous flora in patients with UC and CD suggests that abnormal T cell–specific immune responses to the normal flora of the host are important in the pathogenesis of both diseases (29).

In normal individuals, pathogens are recognized by germline encoded pattern recognition receptors on epithelial cells, neutrophils, macrophages, and dendritic cells. These pattern recognition receptors include lectins, mannose receptors, complement receptors, scavenger receptors, Nod proteins, and toll-like receptors (TLRs). At least ten different TLRs have been described, each of which recognizes a different bacterial factor. Activation of TLRs ultimately results in expression and activation of NF-κB (30). NF-κB is activated in the tissues of IBD patients, where it is thought to have proinflammatory activity. NF-κB stimulates expression of many molecules that likely play a role in IBD including IL-1β, TNF, IL-6, IL-8 and other chemokines, ICAM1 and other adhesion molecules, CD40, CD80, CD86, and the T-cell stimulator ICOS (31). NF-κB also stimulates expression of protective molecules including TNF-induced protein 3, CARD15, cyclo-oxygenase 2, β defensins, and peroxisome proliferator-activated receptor (PPAR)-γ(31).

Activated T lymphocytes are regulated by both effector and regulator T-cell subpopulations in healthy gut mucosa. Effector T cells are capable of inducing intestinal inflammation, while regulator T cells are able to control or prevent inflammation. The immunosuppressive function of the regulator cells is mediated through production of IL-10 and TGF-β. These regulator cells are thought to play pivotal roles in mediating tolerance toward luminal antigens. Genetically engineered IL-10–deficient mice develop severe transmural inflammation of the small and large intestine reminiscent of CD (32). In addition, studies suggest that defects in the IL-10 and TGF-β regulatory signaling pathway may exist in humans with UC (33).

Activation of effector cytotoxic T cells and cytokine release result in generation of activated matrix metalloproteinases, enzymes that are mediators of tissue destruction. In addition, cytokines act directly on the microvasculature, up-regulate adhesion molecules, and enhance recruitment of additional effector cells including neutrophils and macrophages, which amplify and perpetuate the inflammatory response and contribute to additional tissue injury.

As a result of these immunologic events, the mucosa becomes heavily infiltrated by inflammatory cells. Soluble inflammatory mediators produced by neutrophils, lymphocytes, monocytes, fibroblasts, mast cells, neuroendocrine cells, and nerves generate many of the functional and histologic changes that characterize the disease (Fig. 11.2). One sees a large number of activated T and B cells, increased immunoglobulin secretion, the presence of anticolon antibodies (34), and aberrant expression of class II HLA molecules (35).

T-cell profiles differ between patients with Crohn disease and those with ulcerative colitis. As previously mentioned, CD is characterized primarily by a T_H1 response. T_H1 responses are mediated by IFN-γ, the production of which is stimulated by IL-12. Patients with CD also exhibit a T_H17 response, associated with IL-17 production. IL-17 expression is stimulated by IL-6, TGF-β, and IL-23. In patients with ulcerative colitis, the T-cell profile has been more difficult to characterize, but may represent an atypical T_H2-type response. This atypical T_H2 response may be mediated by natural killer T cells that secrete IL-13 (36).

Populations of mucosal B cells and plasma cells increase in UC, a finding that initially suggested that the disease was antibody mediated and complement dependent. In addition, patients with UC demonstrate circulating autoantibodies including those directed against human intestinal tropomyosin isoform as well as anticolonocyte antibodies (34,37). This production of antiself antibodies is now thought to represent a phenomenon that is a secondary protective response aimed at clearing apoptotic cells.

Patients with UC also commonly demonstrate the presence of circulating antineutrophil cytoplasmic antibodies (ANCAs) (38,39). Antineutrophil cytoplasmic antibodies were initially described as sensitive and specific markers for active Wegener granulomatosis but are now known to occur in a wide range of diseases. The antigens recognized by ANCAs are by definition cytoplasmic, predominantly localizing to the primary granules of neutrophils. ANCAs exhibit two distinct patterns on ethanol-fixed human neutrophils: (a) a coarse granular cytoplasmic staining (C-ANCA) or (b) a perinuclear staining pattern (P-ANCA). The perinuclear staining pattern results from redistribution of the cytoplasmic antigen to the nucleus. Serine proteinase 3, cathepsin G, and elastase are the antigens to which C-ANCAs react and myeloperoxidase is the usual antigen to which P-ANCAs react in patients with vasculitis. In contrast, a unique subset of P-ANCAs characterizes UC patients. The identity of the antigen is unclear.

The prevalence of a positive P-ANCA in UC patients ranges from 49% to 86% (40,41). The perinuclear ANCA pattern is 93% to 97% specific (39), but only 46% to 60% sensitive for the diagnosis (42). P-ANCAs are also found in up to 25% of patients with CD (40,43). Titers of these antibodies, however, do not correlate with the degree of severity of the associated colitis, and although they may serve as a convenient clinical marker for UC, their role in the pathogenesis of the disease is unclear. Interestingly, a recent report suggests that P-ANCA in UC may represent a cross-reacting antibody to an antigenic target on Escherichia coli and Bacteroides bacterial strains (44).

The absence of P-ANCAs in the serum defines one population of CD patients, whereas a smaller number of patients are positive for P-ANCA. CD patients with serum P-ANCA expression exhibit a UC-like clinical phenotype (45). One hundred percent of CD patients with P-ANCA have symptoms of left-sided colitis with clinical and histopathologic features of UC. The patients also express a rare allele (R241) of the intercellular adhesion molecule-1 (ICAM1) gene, suggesting the possibility that R241 is a marker for a CD–UC overlap syndrome (46).

In normal mucosa, the inflammatory response is terminated by induction of apoptosis in activated T cells once the pathogen has been eliminated. However, in CD, mucosal T lymphocytes are resistant to apoptosis, leading to their accumulation and persistence of the inflammatory response (19,47). In UC patients, T cells are more susceptible to Fas-mediated apoptosis. In addition, Fas ligand is strongly expressed by T cells in active UC, but not in CD, suggesting the Fas-Fas ligand-induced apoptosis contributes to mucosal damage in UC (48).

Numerous data suggest that environmental factors play a role in the development and progression of both forms of IBD. Susceptibility genes for both UC and CD are known to demonstrate incomplete penetrance. As noted earlier, concordance rates for monozygotic twins are >50% for Crohn disease and <10% for UC. This finding suggests that factors other than genotype must be involved in the pathogenesis of IBD. In addition, the incidence of IBD has increased in the
developed parts of the world over the last 50 years, and is now becoming increasingly common in less developed countries as they become more industrialized and the standard of living improves. Environmental changes that might affect development of the mucosal immune system or the indigenous enteric flora include improved hygiene, consumption of sterile or at least noncontaminated foods, childhood vaccinations, and increased age at first exposure to a variety of intestinal pathogens.

Exogenous agents, including diet, infections, smoking, and other environmental factors, all may play an etiologic role in IBD. Environmental influences related to certain forms of industrial pollution may also account for the recent increase in CD incidence in certain countries.

Numerous studies have demonstrated that exposure to food-associated antigens plays an important role in the gastrointestinal inflammation that occurs in patients with CD. In addition, patients treated with simplified or elemental diets containing proteins in the form of amino acids or small peptide fragments improve symptomatically and show decreased endoscopic or serologic evidence of inflammation (49). Rectal exposure of CD patients to a series of food antigens resulted in increased rectal blood flow and lymphocyte proliferation in comparison to non-CD control patients (50), a finding that suggests that patients with CD show gut-specific sensitization to food antigens. Reactions were seen with yeast and citrus antigens, although individual patients reacted also to other antigen groups. Although it is possible that this sensitivity to food antigens is merely a reflection of exposure to antigens through mucosal defects, the absence of similar sensitivities in patients with UC makes this possibility unlikely (50). It is not likely, however, that exposure to food-associated antigens represents the primary abnormality in patients with CD. Instead, exposure to food in the proximal gastrointestinal tract may lead to sensitization and stimulation of the immune system in genetically susceptible individuals.

For many years investigators have been suspicious that IBD may have an infectious etiology. These suspicions are based on several observations. First, CD patients have an increased incidence of childhood infections including pharyngitis, tonsillitis, and rhinitis (51). In addition, gastroenteritis in early infancy has been linked to later development of CD (52). Furthermore, studies have shown that patients with CD tend to have increased serum levels of antibodies directed against nonpathogenic as well as pathogenic enteric organisms (53). Many studies have attempted to link IBD to infections with Mycobacteria, Yersinia, and several viruses. However, no definitive link with any one infectious agent has ever been made.

Current evidence, instead, suggests that the resident bacterial flora of the gut may be a factor in initiating and propagating the inflammation in IBD. In CD patients, T lymphocytes are hyperreactive to bacterial antigens, a factor that suggests that local bacterial tolerance mechanisms may be abnormal in these individuals (54). Patients with both UC and CD show higher numbers of bacteria attached to their intestinal mucosa than do unaffected individuals (55). In addition, bacterial invasion of the mucosa has been reported in both UC and CD patients (56). IBD patients also have increased mucosal production of IgG antibodies directed against a wide range of commensal organisms (57). The clinical observation that, in some patients, disease flares may be ameliorated by antibiotic administration is supportive of a bacterial role. Finally, recent evidence suggests that the NOD2 CD susceptibility gene is involved in regulation of host responses to bacterial organisms (12). Overall, many view inflammatory bowel disease as a disease initiated by a general loss of tolerance for the commensal bacteria of the gut.

The association between tobacco use and development of IBD is well established. Smoking decreases the risk for the development of UC, but exacerbates and aggravates Crohn disease (58,59). Former smokers have a lower risk of UC than do those who never smoked. In addition, exposure to passive smoke also appears to confer a lessened risk of developing UC relative to nonexposed nonsmokers (60). Overall, the effect of smoking appears to be dose dependent (61). Interestingly, nicotine has been shown to have an inhibitory effect on T_H2 lymphocyte function, the type of cells most implicated in UC (62). Indeed, nicotine-based enemas have been demonstrated to be beneficial in patients with milder forms of distal colitis. Nicotine has no effect on T_H1 cells characteristic of the Crohn inflammatory response. Tobacco use is also associated with protection against sclerosing cholangitis and pouchitis in UC patients (61,63).

Increased intestinal permeability may play a role in the pathogenesis of CD. Increased permeability not only occurs in the intestines of patients affected by the disease, but also in their unaffected first-degree relatives (64). It has been suggested that this increased permeability may represent a predisposing factor to the development of CD because a leaky intestinal barrier may intensify antigen absorption, leading to exaggerated systemic immune stimulation.

Appendectomy early in life (before the age of 20) has been shown in several studies to decrease the risk of developing UC (65,66). Interestingly, the risk for UC is reduced only in patients who undergo appendectomy for acute appendicitis,
and not in those whose appendices are removed because of nonspecific abdominal pain or incidentally during surgery for other causes. This finding suggests that the appendicitis that results in appendectomy, rather than appendectomy itself, is protective. Alternatively, there may be other factors among patients destined to develop UC that prevent those individuals from developing appendicitis. A recent report suggests that the risk for Crohn disease may also be decreased in patients who have undergone appendectomy (67).

Epidemiologic data suggest that use of nonsteroidal anti-inflammatory drugs (NSAIDs) can exacerbate existing UC, and may even induce it de novo (68). This effect was initially attributed to the cyclooxygenase (COX)-1 inhibitory effect of the drugs, but recent reports suggest that even COX-2–specific inhibitors also demonstrate this effect (69). Possible mechanisms by which NSAIDs exert these effects include inhibition of protective mucosal prostaglandin production and increased leukocyte migration and adherence. It has been estimated that NSAID use increases the risk of IBD exacerbation by as much as 30%.

As many as 40% of UC patients report that psychological stress represents a trigger for their disease (70). There is evidence to link psychological stress with increased susceptibility to infection and illness through stress-related impairment of the immune system. Some animal models suggest that stress may play a role in the development of colitis. Cotton-top tamarinds, primates that spontaneously develop colitis and serve as a model for human IBD, develop colitis only in long-term captivity (71).

Differences in IBD incidence exist among individuals with different occupations. Occupations involving work in the open air and physical exercise appear to protect against development of IBD, whereas exposure to air-conditioned, artificial working conditions or extended and irregular shifts can confer an increased IBD risk (72).

As noted previously, there has been a steady rise in the incidence and prevalence of CD in Western Europe, Canada, and the United States over the past several decades. The incidence of CD ranges from 3.4 to as many as 14.6 per 100,000 in differing Western countries (1,73). CD affects all ages and both sexes, but its incidence peaks in the 2nd and 3rd decades of life. A second minor peak in incidence occurs in patients aged 50 to 70 years. Crohn disease is more common among Caucasians than among other racial groups, and is more frequent in Jewish than non-Jewish populations (4,5).

Recent epidemiologic evidence suggests that there are two forms of CD: One inherently indolent (nonperforating), which tends to recur slowly, and another inherently aggressive (perforating), which tends to evolve more rapidly. Patients with the relatively aggressive “perforating” type of CD are more prone to develop fistulae and abscesses, whereas the more indolent “nonperforating” type tends to lead to stenotic obstruction. The latter associates with an exaggerated inflammatory response and an exaggerated proinflammatory cytokine response (74). Host responses determine which form of CD becomes manifest in any individual, and differences in cytokine expression between the two forms of CD might have diagnostic, investigative, and therapeutic implications (74).

The signs and symptoms of CD are often subtle, frequently resulting in a delay in diagnosis until months, or sometimes years, after symptom onset. However, the diagnosis can usually be made on the basis of a careful clinical history, physical examination, and diagnostic testing. The presentation of a patient with CD depends in large part on the location, extent, and severity of gastrointestinal involvement. Crohn disease most frequently affects the ileocecal region, followed (in decreasing order of frequency) by the terminal ileum alone, diffuse involvement of the small bowel, and isolated colonic disease (75). Because the clinical manifestations of CD are very diverse, both in the sites of tissue involvement and in the severity of the inflammation, a wide spectrum of clinical manifestations results (Table 11.3).

Patients with ileocolonic disease experience intermittent episodes of crampy, often postprandial, abdominal pain. Pain may be referred to the periumbilical region, especially in children (75). The abdominal discomfort may be accompanied by loose stools. Stools are small, frequent at night, loose to watery, but not usually overtly bloody. Such symptoms are often attributed to dietary factors or irritable bowel disease. The past history commonly includes perirectal or perianal abscesses and fistulae. Physical examination may localize tenderness to the right lower quadrant. Occasionally, an inflammatory mass may be palpable. Patients with diffuse small intestinal Crohn disease present with diffuse abdominal pain, diarrhea, anorexia, and weight loss. Malabsorption may also occur. These patients demonstrate diffuse abdominal tenderness on physical examination. Colonic CD may mimic ulcerative colitis. Patients complain of diarrhea often containing blood and/or mucus and crampy lower abdominal pain that may be relieved with defecation. Crohn colitis is characterized by more extensive bleeding, more perianal disease, and less pain than Crohn ileitis.

Growth retardation occurs in many pediatric patients with CD, and may occur before other disease signs or symptoms develop. The growth failure and malnutrition result from inadequate dietary intake, malabsorption, and increased nutritional requirements, and, in treated patients, from drug therapy, particularly corticosteroids.

Progressive transmural inflammation with scarring and deep ulceration may ultimately lead to symptoms associated with intestinal obstruction, perforation, bleeding, or fistula formation. When obstruction develops, it usually does so in the distal ileum. Extensive mucosal ulceration predisposes the patient to bacterial translocation with all of its complications, including a predisposition to bacterial endocarditis (76). Patients also demonstrate altered small intestinal motility with abnormal receptor-mediated small intestinal contraction (77). Deep linear ulcers or fistulae may sometimes give rise to profound lower GI bleeding.

A sudden worsening of clinical symptoms and/or an unusual disease presentation should alert one to the possibility of ischemia or viral infection superimposed on pre-existing CD. Ischemia may develop secondary to vasculitis, or may occur because of endothelialitis resulting from infection with cytomegalovirus (CMV), particularly if immunosuppressive therapy has been utilized.

The gross features of the bowel reflect the stage of the disease, with the most severe lesions being seen in advanced transmural disease. Patients with severe disease are those most likely to come to resection. Thus, the pathologist is only likely to see the gross pathology of advanced disease.

The complications of CD are listed in Table 11.10. The extraintestinal and neoplastic manifestations are discussed
later in this chapter. Complications also result from therapy. These include steroid-related osteonecrosis often affecting the hip or knee, as well as secondary viral or bacterial infections.

In recent decades, the incidence of UC in the United States and in Europe has risen, whereas in other countries the incidence has plateaued (2,87,109). It is likely that two factors have contributed to this increase: (a) improved survival of incident cases and (b) diagnosis of milder cases due to increased use of sigmoidoscopy and fecal occult blood testing in the community. UC mostly affects young white people, but there is an increasing recognition that the disease affects many ages and many ethnic groups (2). The incidence of UC inversely correlates with smoking and clinical relapses have been associated with smoking cessation (110).

UC demonstrates a wide clinical spectrum affecting all age groups, with a predominance of the disease in young and middle-aged persons. The peak age of incidence is the 3rd decade. Women are more commonly affected than men. Mean age at diagnosis is 32 years (111). The disorder is less common in black persons than white persons. Fifty to sixty percent of patients present with an insidious onset of vague abdominal discomfort. Symptoms vary from those of mild rectal irritation to an acute severe colitis. In children, growth failure, characterized by decreased linear growth and delayed sexual maturation, often represents the first sign of disease. Patients with UC develop this complication less commonly than CD patients (112).

Diarrhea, hematochezia, urgency of defecation, attacks of crampy abdominal pain, and perianal soreness are common in early stages of the disease. Abdominal pain is usually less severe than in CD. Approximately 10% of patients initially present with severe colitis characterized by abdominal pain, more than 40 bloody stools per day, fever, tachycardia, weight loss, and hypoalbuminemia. When UC remains limited to the left colon, constipation rather than diarrhea may result because colonic spasm results in stool retention in the right colon where the normal absorptive capacity is unimpaired.

Disease onset is abrupt in approximately 30% of patients. The disease may be explosive, with a sudden onset of acute diarrhea, frequent bloody stools, continuous abdominal pain, anorexia, rapid weight loss, chronic iron deficiency anemia, and a persistent fever. Acute and massive rectal bleeding affects up to 3% of patients (111). The bleeding originates from large denuded areas of the mucosa in the face of widespread telangiectatic lamina propria vessels. In contrast, the bleeding seen in CD usually originates from a localized source such as an eroded blood vessel in the submucosa or deeper layers.

Characteristically, UC is a chronic mucosal disease with relapses and spontaneous remissions. It usually starts in the rectum and then spreads more proximally. Most patients have distal disease at the time of first attack, but some have more proximal disease, including total colonic involvement. There are several major clinical presentations: (a) acute fulminating colitis, in which the whole colon and rectum are affected with extensive and deep ulcerations; (b) continuous colitis, in which the symptoms persist from the onset and vary in severity (usually only affecting the left side of the colon); and (c) recurrent colitis, in which the attacks are limited and separated by varying periods of remission. Disease distribution usually remains relatively constant, with about 50% of patients in clinical remission in any one time. Relapses are unpredictable except that the disease activity in foregoing years indicates a 70% to 80% probability that the disease will continue the following year (113). The cumulative probability of having a relapsing course is 90% after 25 years of disease. Activity in the first 2 years of diagnosis significantly correlates with having an increased probability of 5 consecutive years of disease activity. The probability of maintaining work capacity after 10 years is 92.8%, with a range of 90.8% to 94.8%.

UC progresses to a more serious form in 53.8% of patients. Factors associated with disease progression include toxic colitis, extent of disease at diagnosis, presence of joint symptoms, younger age at diagnosis, and severe bleeding (111). Various infections (CMV, Salmonella, and Clostridium difficile), medications, and intervening ischemic disease also exacerbate UC (Table 11.11).