Inflammatory Bowel Disease: Introduction
In a day-to-day practice, general pediatricians are expected to care for a wide variety of gastrointestinal disorders. Fortunately, most of the gastrointestinal aliments in childhood are not serious and self-limiting in nature and do not usually require a referral to a sub-specialty such as pediatric gastroenterology with few exceptions. Inflammatory bowel disease (IBD) is an exception; consultation with a pediatric gastroenterologist is mandatory for adequate management. Crohn’s disease (CD) and ulcerative colitis (UC) are the two most common chronic IBDs. For decades, CD and UC were considered totally different entities, since the clinical features, treatment options, and natural history of these diseases appeared to be quite different. However, recent developments and findings have made clinicians and scientists begin to think of IBD as one entity, with a spectrum of findings. Recent work has, for example, identified CD and UC occurring in the same families, common susceptibility genes and pathways between these two entities, and similar responses to emerging therapies to both conditions. Classic CD is found at one end of the spectrum and UC at the other, but many patients have features that overlap. In this chapter, we will try to approach CD and UC as one entity but differentiate CD from UC whenever necessary.
Although a diagnosis of IBD can be made at any age, about one in every four new diagnoses of IBD is made before the age of 20 years. Population-based studies suggest that IBD is unevenly distributed throughout the world, with the highest disease rates occurring in Western or industrialized countries. Recent epidemiologic surveys have also suggested that IBD incidence rates have changed over the second half of the twentieth century, with a gradual increase in CD, but reaching a plateau in UC incidence.1 Only a few systematic, population-based studies have been done in North America regarding the true incidence of childhood IBD to date, and estimate the incidence is about 7–12 per 100,000 children. Although the true impact of IBD in children is not entirely known, we estimate that about 100,000 children are suffering from IBD in North America at any given time.2 This makes IBD one of the common chronic gastrointestinal disorders for which pediatricians and pediatric gastroenterologists provide care.
There are differences in the descriptive epidemiology of IBD when pediatric IBD is compared with adult-onset IBD. While CD and UC occur with equal distribution in adults, the incidence of CD is three times higher than UC in pediatric age groups.2 In adult IBD, there is a nearly equal ratio of male to female disease, with a slight female predominance. In contrast, prepubertal children with CD exhibit a male preponderance of 1.5:1.3 Currently there is no molecular explanation why children with IBD differ from adults in regard to gender ratio.
Etiology and Pathogenesis
The etiology of IBD is not completely understood at present, although significant advances have been made over the last decade. It is becoming increasingly more evident that certain individuals have a genetic predisposition and that one or more environmental triggers can lead to the development of disease.
The well-documented increase in the incidence and prevalence of IBD is part of a world-wide emergence of chronic autoimmune and inflammatory diseases, a phenomenon closely linked to social and economic development. Initially noted in Northern Europe and North America, this increase has now been documented in the rest of Europe, Japan, and South America, and most recently in the Asia-Pacific region.4 An explanation for the higher frequency of IBD has been known as the “hygiene hypothesis,” which postulates that there has been a fundamental lifestyle change from one with high to one with low microbial exposure.5 Exposure to fewer microbial antigens early in life would lead to a less robust immune system, ill prepared to tackle challenges later on in life, and mounting immune responses unable to eliminate offending agents resulting in chronic inflammation. Numerous other environmental factors and stimuli are considered risk factors for IBD, including smoking, diet (junk food), drugs, geography and social status, increased stress, the enteric flora, altered intestinal permeability, and appendectomy.1,4
IBD is highly heritable. This concept is strongly supported by family, twin, and phenotype concordance studies and is now confirmed by the discoveries of many susceptibility genes.6,7 A positive family history is a well-known risk factor for development of IBD. Recent research also shows that heritable factors have more influence in CD than in schizophrenia, asthma, and essential hypertension.8 Initial family-based linkage studies of IBD implicated the NOD2 gene in CD and the MHC region in UC for increased susceptibility.7,9 Recently, genome-wide association scans (GWAS), which employ high-density single nucleotide polymorphism (SNP) array technology, have increased the number of possible genetic factors linked to IBD. To date, this method of broad, unbiased screening has provided as many as 30 susceptibility loci in CD and 17 loci in UC. However, the most surprising finding was that the vast majority of the new gene discoveries were common to both CD and UC, further reinforcing our belief that CD and UC stem from same pathogenic process, with a differing clinical spectrum.
The hypothesis that infectious agents cause IBD has been popular for many years. However, a multitude of studies have failed to confirm the presence of infectious agents by histological examination, culture of tissue homogenates, genomic identification, and serum antibodies. More recently an entero-adhesive/invasive strain of E. coli has been described as being associated with ileal CD, but its potential etiological role remains unclear. Instead of specific infectious agents, evidence continues to accumulate that the indigenous commensal gut flora is the target of the chronic immune response in IBD. The majority of IBD patients show an enhanced immunological reactivity against gut bacterial antigens. Bacterial flagellin has been recently reported as a dominant antigen in CD, apparently defining a population of patients with complicated CD. The discovery that CD is associated with mutations of the NOD2/CARD15 gene, whose product is a bacteria-sensing cytoplasmic protein, suggests that the ability of the immune system to normally recognize the gut flora may be genetically altered in IBD.
Inflammation is the most common type of reaction that the body mounts against external or internal offending agents. The gut is particularly susceptible to inflammation because, even under normal circumstances, a baseline “physiological inflammation” is present in the mucosa, representing a controlled immune response against dietary and microbial antigens. When this physiological response becomes excessive and chronic, it leads to injury, resulting in anatomical and functional abnormalities. Major advances during the past few decades have helped our understanding of the cellular and molecular mechanisms mediating mucosal immunity, as well as the alterations that lead to chronic gut inflammation.10
UC and CD are grouped together in this chapter due to many similarities in their clinical and epidemiologic features and therapeutic responses. They are both chronic, inflammatory diseases of the gastrointestinal tract with periods of remission and exacerbation. Although they share many similarities, they each have distinguishing characteristics. UC is chronic inflammation involving only the mucosa of the colon. The inflammation is continuous, starting in the rectum and extending proximally to varying extents. In contrast, CD has transmural inflammation and is not localized solely to the colon. CD can be found anywhere in the GI tract, from mouth to anus. The inflammatory process in CD is patchy, which can be helpful in distinguishing UC from colonic CD. The terminal ileum is the most common site of CD. Because disease activity often occurs in several areas, about 60% pediatric patients have ileocolonic involvement while 20–30% have isolated colonic disease.3 In certain cases, if skip lesions, granulomas, and small bowel involvement, all hallmarks of CD, are absent, the diagnosis between UC and CD may be more difficult. If there is oral or perianal involvement, then CD is more likely. However, in approximately 10% of cases of colonic IBD, it is difficult to distinguish between CD and UC. These patients are given an interim diagnosis of indeterminate colitis or the newly coined term inflammatory bowel disease-undetermined (IBDU).11
The most common presentation of UC is diarrhea and rectal bleeding, while the “classic presentation” of CD consists of abdominal pain, diarrhea, poor appetite, and weight loss. Since CD presentation can be insidious and nonspecific, a delay in the diagnosis of CD is more common, compared with UC. The time from diagnosis to the onset of symptoms in CD can be anywhere from 5 months to a few years. In both UC and CD, approximately 30% of pediatric patients present with moderate signs of systemic illness. Another 10% present with severe colitis, defined as more than five bloody stools per day, anemia, fever, hypoalbuminemia, tachycardia, and weight loss.12
Although abdominal pain is the single most common presenting symptom in IBD, it can have several other less obvious presentations, such as growth failure, arthralgias, rashes, and iron deficiency anemia, without notable GI symptoms. Being aware of these other presentations can aid in early referral and initiation of treatment. In addition, since abdominal pain is a common symptom in a general pediatrician or family practitioner’s office, and most of these children will not have IBD, not every child with abdominal pain needs to be referred to a gastroenterologist. Table 16–1 lists “red flags” in the history and physical examination that should raise suspicion for IBD in a patient and warrant further investigation.
|Abdominal pain distant from umbilicus||Decreased growth velocity|
|Pain that interferes with sleep||Delayed sexual maturation|
|Discrete episodes of pain that are acute in onset||Finger clubbing|
|Pain precipitated by eating||Oral ulcerations|
|Dysphagia, odynophagia||Abdominal tenderness|
|Involuntary weight loss||Abdominal mass|
|Rectal bleeding||Perianal fistula, fissures|
|Unexplained low-grade fevers|
|Erythema nodosum, pyoderma gangrenosum|
|Severe eye pain or persistent conjunctivitis|
|Strong family history of IBD|
Abdominal pain is seen in 60–95% of patients presenting with CD and 30–70% of patients presenting with UC. Sometimes it can be difficult to distinguish between abdominal pain from IBD and functional abdominal pain. Location, chronicity, and severity may help determine if it is from IBD. Abdominal pain in CD can awaken the child at night, cause decreased appetite, and in the case of ileocecal disease, can be localized to the RLQ. Odynophagia and dysphagia can be seen in esophageal CD, which occurs in approximately 10% of CD patients.13
Diarrhea is seen commonly in UC (70–90%). It is seen in 65–75% of CD patients, and can be intermittent, depending on the location of disease.12 Distal colonic disease is typically associated with tenesmus (a sensation of incomplete evacuation) and urgency. Gross blood is usually seen with colonic involvement and may or may not be associated with abdominal pain. Nocturnal stooling is also common in colonic disease.
Between 50% and 90% of UC patients and 20% and 60% in CD present with rectal bleeding.12 The severity is dependent on location and severity of disease, and can be intermittent. In addition, CD can often have occult blood-positive stools without gross blood. Because of this, any child with iron deficiency anemia should have stool tested for blood and may require further GI work-up if positive.
Weight loss is a major problem in CD; over 80% of patients have some degree of weight loss at presentation.14 The cause is probably multifactorial, but is mostly due to poor nutritional intake. Many children have decreased appetite/intake due to pain or increased stools. Adding to this, they can have decreased nutritional absorption due to small bowel disease. A higher metabolic state due to chronic inflammation may have some contribution as well.
Growth failure is a critical concern in childhood-onset IBD. It is important for physicians to know that growth impairment can be the only presenting sign of CD, even before GI symptoms manifest (Figure 16–1).15 In a child with decreased height velocity or lack of growth, IBD should be on the differential diagnosis. About 30% of pediatric CD patients and 6% of UC patients will have growth failure at presentation.16 It has been shown that the majority of early onset CD patients (onset prior to puberty) have reduced adult height.17 Growth failure is more pronounced in boys than girls for unknown reasons.14 The cause of persistent growth failure is multifactorial, including nutritional deficits, malabsorption, increased metabolic demands, medications, and possibly other unknown mechanisms.
Delayed puberty affects twice as many patients with CD compared to UC. Average age of menarche in healthy adolescents is 12.8 years; however, one study found that more than one-half of females with CD had delayed menarche to age 16 years if disease started before puberty. Duration of puberty might also be prolonged in patients due to frequent relapses of disease during puberty.18 The role of endocrinologic mechanisms in the delay is not known. In evaluating a patient for delayed growth and puberty, bone age radiography can aid in determining if they will have time to catch up in those areas.
Perianal or perirectal disease is one of the features distinguishing between UC and CD, since it is not a feature of UC. Perianal disease can vary widely and can include multiple large anal tags (Figure 16–2), perirectal abscesses, nonhealing deep fissures, and fistulas (Figure 16–3). A single noninflamed anal skin tag at the midline position is not considered perianal disease. Most perianal disease is painless unless there is abscess formation. Approximately 80% of fistulas seen in CD are perianal/perirectal. One third of patients will develop a perianal fistula or abscess at some point in their disease course.
IBD is not a single-organ disease, but a systemic disease with many “extra”-intestinal features. Over 130 different EIM have been reported in the literature, with most being very rare. However, between 25% and 30% of patients will exhibit some EIM in their lifetime.19 These cause varying degrees of morbidity and mortality in IBD patients. The exact etiology of these conditions is unknown, but autoimmune reactions to bacteria, induction of immune complexes and associated inflammatory responses, and genetic factors are all possible explanations. EIM often correlate with GI symptoms, but some will be present even in remission. The most common EIM will be covered here.
Joint manifestations are the most common EIM in IBD. Up to 25% of IBD patients are affected by arthralgias or arthritis, and 20–40% patients have more than one episode.20 These manifestations can precede intestinal symptoms, and some children will be referred to a rheumatologist before being seen by a gastroenterologist. Joint disease in IBD can be separated into two categories, axial and peripheral. Axial forms include ankylosing spondylitis and sacroiliitis. Asymptomatic sacroiliitis can be found in 10–52% of patients and is usually revealed by bone scans.20Ankylosing spondylitis occurs in <2% of patients and is also associated with HLA-B27 positivity.21 Axial arthropathy does not usually parallel GI disease symptoms, and treatment consists of physical therapy and exercise. Peripheral joint disease, on the other hand, is less commonly seen in IBD and is often associated with large bowel disease. As the IBD is treated, the peripheral joint symptoms also decrease in most cases.20
Skin involvement has been described in 10–15% of patients with IBD.19 Cutaneous manifestations can be classified into three main groups: granulomatous, reactive, and nutritional. The two most common cutaneous manifestations of IBD are erythema nodosum (Figure 16–4) and pyoderma gangrenosum (Figure 16–5), which are reactive in nature.
Erythema nodosum is seen in CD more than UC with prevalence between 2% and 12%.22 It presents as a single or multiple tender red nodules typically on the extensor surface of the lower extremities that spreads rapidly to adjacent skin and develops into a burrowing ulcer with irregular violaceous edges. Over days to weeks, the nodules flatten and turn brown or gray. In addition, patients can have systemic symptoms such as fever. It usually correlates with disease activity, so treatment should focus on treating the underlying IBD.
Pyoderma gangrenosum is reported in 0.5–5% of patients with IBD but specific prevalence data in children are lacking.22 Pyoderma gangrenosum may develop before bowel symptoms, during quiescent disease, or even following colectomy. About 40% of IBD patients with pyoderma gangrenosum also develop arthritis.21 These ulcerated lesions are multiple in the majority of patients, and most appear below the knee.
Other skin manifestations are more rare. Sweet syndrome, an acute febrile neutrophilic dermatosis, is a rare manifestation of IBD and presents as tender, erythematous plaques or nodules involving the arms, legs, trunk, hands, or face. It may parallel the intestinal disease or may precede the diagnosis of IBD. Skin biopsies reveal a neutrophilic infiltrate. Granulomatous skin lesions can occur as perianal and peristomal ulcers and fistulas, oral ulcers, and cutaneous granulomatous ulcers, all of which are commonly referred to as metastatic CD. The skin lesions have the same histological features as the bowel disease. The most frequent nutritional deficiency presenting with cutaneous manifestations is acrodermatitis enteropathica, caused by zinc deficiency. In addition, associations between autoimmune cutaneous diseases and IBD have been reported, including psoriasis.22
Ocular manifestations occur in approximately 1% of patients with IBD.21 These are seen more frequently in CD patients with colonic disease. Ocular diseases can mimic conjunctivitis, so it is important that the clinician be aware that these patients are at higher risk for serious eye disease. Acute anterior uveitis is an ophthalmological emergency and requires prompt intervention. Treatment of uveitis with systemic or topical steroids is essential to prevent progression to blindness. Patients with IBD can also develop eye disease, such as cataracts and glaucoma, secondary to corticosteroid use.
Bone disease is being studied intensively since the recognition that up to 40% of adult IBD patients have osteoporosis. In one study, the relative risk for fractures in adult IBD patients was 1.4 times that of the general population.23 This is extremely important in the pediatric population, since this is the time of skeletal growth and maturation. Multiple factors could contribute to decreased bone mass density (BMD) in these patients, including medications (corticosteroids), inflammatory responses, decreased physical activity, and nutritional deficiencies. Studies have used the DEXA scan to determine if IBD children do have lower bone mass and increased risk of fractures compared to their peers. It seems that they do have lower bone mass density but this may be in part due to smaller bone size, which may or may not cause an increased risk of fracture.24 Currently, a DEXA scan is not indicated in all pediatric IBD patients. Until more studies are done, the most beneficial thing for these children is disease control (while trying to spare corticosteroids), nutrition (including vitamin D and calcium), and increased activity level. Bisphosphonates can be considered in certain patients, but should not be used without referral to endocrinology.
Oral aphthous stomatitis (Figure 16–6) is seen in at least 5–10% of patients with UC and 20–30% with CD.22 Oral lesions may occur before the onset of intestinal symptoms of IBD or may parallel intestinal disease activity. Symptoms range from mild discomfort to severe debilitating pain. Biopsies of these lesions often reveal lymphedema and granulomata in patients with CD. Other oral lesions can include lip swelling, fissures (Figure 16–7), and gingivitis.
The most common serious hepatobiliary complication among pediatric IBD patients is primary sclerosing cholangitis (PSC), a disorder of both intrahepatic and extrahepatic bile ducts. It affects patients with UC more than CD and males more than females. It is estimated that 3.5% of UC patients will develop cholangitis.21 PSC should be suspected in an IBD patient with pruritus, jaundice, fatigue, and anorexia, although they can be asymptomatic. Laboratories show persistently increased alanine aminotransferase, gamma-glutamyl transpeptidase, and alkaline phosphatase levels. Diagnosis is confirmed with cholangiography and liver biopsy. Treatment with ursodeoxycholic acid improves liver tests but it does not change disease course.21 IBD patients with PSC are also more likely to have other autoimmune manifestations.
Pancreatitis can also be an EIM, but it is more commonly due to duodenal CD, PSC, or drug reactions.22 Cholelithiasis is usually a complication secondary to malabsorption of bile acids in small intestinal disease and is reported in approximately 10% of adults with CD.22