—  SHORT COURSE #03  —

Gastrointestinal Pathology: Controversies in Non-neoplastic Biopsy Interpretation
Moderators: Robert D. Odze, Marie E. Robert, Donald A. Antonioli

Section 3 - Diagnostic Problems in Biopsy Interpretation of Colorectal Inflammatory Disorders

Donald A. Antonioli
Beth Israel Deaconess Medical Center and Harvard Medical School,
Boston, MA


The category of "idiopathic inflammatory bowel disease" (IBD) is classically defined as being composed of two disease entities, ulcerative colitis (UC) and Crohn's disease (CD). Since the etiology of neither condition is known, there is no gold standard for their diagnosis. As a result, diagnosis is achieved by eliminating known causes of enteritis and colitis (such as ischemia, infections and allergies) and by recognizing clinical, imaging, endoscopic and histologic features characteristic of these two conditions. Although the classic forms of UC and CD are well described in the pathology literature, recent advances in imaging techniques, the increasing use of upper gastrointestinal endoscopy and colonoscopy to evaluate these conditions, and the use of diagnostic procedures early in the course of disease have led to the recognition of deviations from typical aspects that may confound the diagnosis of UC and CD and even raise the possibility of other types of inflammatory bowel disease. The findings of classic UC and CD in the colon will be reviewed, but the emphasis will be on atypical and newly-described aspects that may cause diagnostic difficulties in pediatric and adult patients.

Histology of the Normal Colonic Mucosa

Since most colonic inflammatory conditions are diagnosed in mucosal biopsy specimens, the normal histology of this layer of the colonic wall will be briefly reviewed, with an emphasis on findings that should not be overinterpreted as representing disease.

Crypt architecture and cells
The colonic crypts are arranged in parallel array within the lamina propria from the cecum to the rectum. Throughout most of the colon, the crypts extend to the top of the muscularis mucosae; however, in the rectum, the bases of crypts end a few millimeters above this landmark. In addition, crypts in the distal rectum and upper anal canal may be dilated or slightly irregular in contour and be accompanied by fibres of smooth muscle extending into the lamina propria from the muscularis mucosae. These findings, perhaps due to mucosal prolapse or prior minor injury, are non-specific and should not be overinterpreted as evidence of chronic inactive colitis, particularly in young patients without a previous history of IBD. A final normal feature is the presence of crypts having an inverted-"V" appearance, with the apex of the "V" at the luminal surface of the mucosa. This configuration is a manifestation of dividing crypts and contrasts with the irregular branching typical of chronic colitis [1]. The crypts are lined predominantly by enterocytes and goblet mucous cells. Endocrine cells and Paneth cells are located at the base of the crypts, and a small number of intraepithelial lymphocytes may be identified along the length of the crypt [2, 3]. Paneth cells normally are present only from the cecum to the proximal transverse colon; their identification distal to this point is considered a manifestation of metaplasia secondary to prior injury such as involvement by IBD [4]. However, a small number of Paneth cells is sometimes identified in otherwise completely normal mucosa of the rectosigmoid colon in children. As in the case of the mild architectural abnormalities in the rectum described earlier, these distal Paneth cells are most likely the result of minor prior injury and, in isolation, should not be used as evidence of IBD unless there are supportive clinical data and/or abnormalities in specimens from more proximal parts of the colon.

Cells of the lamina propria
The lamina propria contains lymphocytes and plasma cells throughout the colon; however, there is typically a decrease in their number from the cecum to the rectum. In addition, these mononuclear cells, as well as macrophages, are concentrated in the upper half of the lamina propria except in the cecum and proximal right colon, where they occupy the full thickness of this layer [2, 3]. Recognition of the full-thickness extent and greater density of inflammatory cells in the lamina propria of the normal cecum/proximal right colon will prevent overinterpretation of a biopsy specimen from this area as "chronic colitis", particularly Crohn's disease, when more distal specimens are unremarkable [5]. In the resolving phase of an acute self-limited colitis (such as bacterial-induced injury), mononuclear inflammatory cells are increased in the lamina propria, but hallmarks of chronic colitis, such as crypt architectural distortion and basal plasmacytosis proximal to the cecum, are absent [6, 7, 8]. As a rule, apparently increased lamina propria mononuclear inflammation in isolation should be interpreted with caution [3]. It may be a normal finding (cecum), represent a resolving acute injury, may be the result of geographic variation in the "normal" number of such cells, or may be an artifact due to abnormally thick sections. Overinterpretation of this minor change as "chronic colitis" or "non-specific chronic inflammation" may result in unnecessary procedures (e.g., imaging studies), heightened anxiety about IBD, or even empiric therapy for UC or CD [5, 8]. Although neutrophils are completely absent or identified only in small numbers in normal colonic mucosa, eosinophils are consistently present [3]. Interpretation of the range of normal numbers and the significance of the eosinophils, however, remain major problems in evaluating colonic mucosal biopsies. Limited published data suggest that the "normal" number of eosinophils in children varies in different parts of the colon: The largest number is in the cecum, with a gradient of decreasing numbers to the rectum. In addition, the numbers is various segments of the colon may vary in different parts of the United States [9, 10]. Finally, eosinophils are increased as part of the overall inflammatory cellular response to injury, especially in chronic forms of colitis. Given these difficulties, when should eosinophils be interpreted as "abnormally present"? Criteria for suggesting an abnormality (allergic condition; drug reaction) include (1) a large number of eosinophils in mucosal biopsy specimens from throughout the colon unaccompanied by an increase in other inflammatory cell types and (2) eosinophils as the sole or predominant inflammatory cell in foci of cryptitis and crypt abscesses. In infants younger than six months of age ultimately shown to have allergic proctitis, eosinophilic cryptitis/crypt abscesses and greater than 60 eosinophils per 10 high-power fields in the lamina propria are criteria that distinguish this condition from other acute colitides such as infections [11]. These criteria for distal allergic colitis, however, have not been validated in older infants and children. Occasional eosinophils infiltrating the surface epithelium are a common finding, the significance of which in isolation is unclear [3].

Lymphoglandular complexes
Lymphoid nodules, with or without follicular centers, are another frequent and normal finding in the colonic mucosa. The nodules and adjacent crypts form so-called "lymphoglandular complexes". The crypts at the periphery of the complexes are typically splayed apart. These complexes are sites of antigen presentation and processing; as a result, the surface epithelium overlying the complex is mucin-depleted and contains increased intraepithelial lymphocytes. The splayed crypts also exhibit goblet cell mucin depletion, epithelial cell damage and increased intraepithelial inflammation (lymphocytes; neutrophils). These changes in lymphoglandular complexes should not be overinterpreted in isolation as "focal active colitis". In problematic cases, examining multiple levels through the biopsy specimen can be helpful. In lymphoglandular complexes, the crypts, although splayed, do not exhibit irregular branching or crypt destruction by active inflammation. Detection of the latter features supports the diagnosis of a true inflammatory condition [3, 12]. Senescent colonic epithelial cells are eliminated at the mucosal surface either by extrusion into the lumen or by apoptosis and removal within the subjacent lamina propria. Therefore, some apoptotic cellular debris may be normally found in the superficial lamina propria [3]. In contrast, apoptosis of more than a rare epithelial cell at the base of the crypt could be an iatrogenic effect of certain bowel cleansing agents (see next section), but it could also represent a pathologic finding secondary to immunodeficiency, certain infections, or graft-versus-host disease.

Artifacts in Colonic Mucosal Biopsies that should not be Confused with Disease



Effects caused by the endoscopic procedure
The colonoscope itself may cause minor injury to the mucosa. At colonoscopy, the entire colon (and, often, the terminal ileum) are first examined, then biopsy specimens obtained as the last part of the procedure. As a result, there has been sufficient time for the endoscope to irritate the mucosa. Therefore, vascular congestion and edema/fresh hemorrhage in the lamina propria when unaccompanied by epithelial damage or an inflammatory infiltrate are best interpreted as iatrogenic in origin. These changes are most often identified in the distal colon [1, 2, 3]. During colonoscopy, air is introduced to dilate the colon. If the colonoscope causes defects in the mucosal surface, or if mucosal erosions or ulcers are already present, air can enter the lamina propria and produce irregular clear spaces and secondary crypt architectural irregularity. This phenomenon has been termed "pseudolipomatosis" because of its superficial resemblance to adipocytes in the lamina propria. Distinction from the latter cell type is easily made by recognizing that the spaces are irregular and lack cell membranes or peripherally located nuclei [13].

Effects of colonic cleansing procedures
A variety of preparations have been used to cleanse the colon for endoscopy. Oral non-absorbable lavage agents and isotonic saline enemas are relatively gentle, and do not damage the colonic mucosa [14]. The process may be prolonged, however, resulting in poor compliance and inadequate cleansing. Therefore, a more vigorous and shorter cleansing protocol using sodium phosphate (taken orally or in the form of enemas), oral magnesium citrate and/or bisacodyl enemas and suppositories may be favored. These agents may produce mucosal lesions. Sodium phosphate preparations (and, to a lesser extent, magnesium citrate) may produce aphthous ulcers. At endoscopy, these are recognized as discrete, sharply defined lesions, 1 to 3 mm in diameter, surrounded by an erythematous halo. Most often detected in the rectosigmoid, they are identified in a background of otherwise unremarkable mucosa with a prevalence of 6% to 24% in recent reports [14, 15, 16, 17]. At histologic examination, the correlate of the aphthous ulcer is either an erosion overlying a lymphoid aggregate or a focal, often superficial, ischemic-type lesion with mucin-depleted crypts, modest active inflammation and fibrinous exudates [17]. In a patient being evaluated for the cause of diarrhea, these findings raise the differential diagnosis of infectious colitis, drug-related (e.g., NSAID) injury and low-grade IBD, especially Crohn's disease. Integration of all clinical data plus recognition of the possible iatrogenic origin of aphthous ulcers should permit correct categorization of the findings. Oral sodium phosphate preparations may also induce increased epithelial cell proliferation and mild abnormalities at the base of crypts [14, 15, 18]. The latter takes the form of apoptosis and a modest infiltrate of neutrophils (mild cryptitis) unaccompanied by crypt destruction (i.e., crypt abscess formation), increased mononuclear inflammation in the surrounding lamina propria, or crypt architectural abnormalities [14, 15, 16, 18]. In clinical practice, this finding raises a dilemma: is it iatrogenic or a manifestation of disease? Such minimal deviations from normal should be interpreted with caution and categorized as "non-specific" in nature rather than an unequivocally disease-related "focal active colitis" with all the implications inherent in the latter diagnosis [5]. As discussed by Xin and colleagues, focal active colitis that is likely caused by disease is characterized by more florid cryptitis (with or without crypt abscesses) surrounded by an increased concentration of lymphocytes and macrophages (or even mucin granulomas] in the lamina propria [19]. The significance of this type of focal active colitis in isolation will be discussed in greater detail later.

Effects due to endoscope cleaning and the biopsy procedure
A variety of substances have been used to disinfect colonoscopes. If these cleansing agents are incompletely removed prior to the next use of the endoscope, the residual disinfectant may injure the mucosa. For example, glutaraldehyde will produce an acute ischemic colitis, and hydrogen peroxide may induce pseudolipomatosis [1, 20, 21, 22]. Fortunately, recognition of these possible complications has resulted in a marked reduction in their incidence due to more thorough cleaning of instruments. Finally, obtaining and processing the biopsy specimens may cause problems that may have an impact on accurate interpretation of the tissue [1, 3]. If the muscularis mucosae is not included in the specimen, the crypts may be artifactually separated from one another and distorted, thus giving a false impression of architectural features of chronic colitis. Tissue at the edge of biopsy specimens that has been compressed by forceps may give the false impression of atrophy and fibrosis because crypts have been crushed or extruded from the lamina propria. Findings limited to distorted edges of specimens, therefore, should be interpreted with caution. Detection of active inflammation and/or granulation tissue in these locations is helpful in determining that the changes are due to intrinsic disease rather than trauma.

Histologic Features of Mucosal Colitis
A major consideration in evaluating colonic mucosal biopsy specimens is the distinction of an acute self-limited process from idiopathic inflammatory bowel disease, given the psychological and medical impacts of the latter group of diagnoses. The problem is compounded by the fact that the usual cases of self-limited colitis (due, for example, to infections and allergies) are diagnosed by a combination of clinical, epidemiological and microbiological findings. As a result, it is only the unusual cases with a prolonged clinical course, atypical laboratory findings and/or negative infectious workups that eventuate in colonoscopy. Biopsy specimens from these patients may contain indeterminate findings that are difficult to categorize with certainty. The histologic features of active and chronic colitis in mucosal specimens will be reviewed, followed by a brief discussion of the distinction between IBD and an acute self-limited disease.

Active colitis
Activity is characteristic of both the acute self-limited colitides and IBD. The features of activity are well known: neutrophils in the lamina propria, cryptitis, crypt abscesses, erosions and ulcers. Neutrophils confined to the endothelial layer of small vessels (marginated neutrophils) may be the result of the bowel preparation or the colonoscopic procedure itself; therefore, they are not included as a manifestation of active colitis. These findings may be focal or diffuse and, in my opinion, their number and distribution cannot reliably be used to distinguish infectious colitis from IBD. However, inflammatory pseudomembranes, heightened eosinophilia of the lamina propria, and mucin-depleted crypts associated with a relative paucity of neutrophils are findings most suggestive of an ischemic injury or infection with toxin-producing bacteria such as Clostridium difficile or enterohemorrhagic E. coli [1, 6, 7, 8, 23].

Chronic colitis
The histologic features of chronic colitis are summarized in Table 1. As discussed earlier, determination of increased mononuclear inflammation in the lamina propria is the most subjective and probably the least reliable criterion of chronic injury [5, 8]. Precision in diagnosis is important because the pathologist's diagnosis of "increased mononuclear inflammation" or "nonspecific chronic inflammation" in colonic mucosa may be translated by the clinicians to mean IBD [5]. In most cases of IBD, other features to support chronicity will be present, particularly if multiple specimens are submitted. Crypt atrophy refers to a decrease in the length and/or number of crypts and crypt distortion to abnormalities in shape; villiform surface change presumably develops secondary to crypt distortion and prominent surface regeneration. Lymphoid aggregates confined to the base of the lamina propria, basal plasma cells in the lamina propria distal to the cecum, and Paneth cells distal to the hepatic flexure are all useful hallmarks of chronic injury [7, 8, 24, 25]. Basal plasmacytosis distal to the cecum is an excellent marker of chronicity, but may be absent in the early phase of IBD and may diminish over time in inactive disease [25]. These changes, however, are not synonymous with IBD; they may develop with the healing of any severe mucosal injury such as that due to ischemia or chronic infections (e.g., schistosomiasis). As a result, the diagnosis of IBD must be supported by appropriate clinical, imaging and endoscopic findings. If the pathologist has no information about the patient, it is probably best to diagnose such biopsy specimens descriptively as "findings compatible with healed injury" rather than "chronic inactive colitis", since the latter phase will typically be interpreted as IBD by the clinicians. As with active colitis, the changes of chronic injury may be focal or diffuse within a specimen or among specimens. Also, finding only a single feature listed in Table 1 is sufficient to support a diagnosis of chronic injury. Multiple abnormalities may be infrequent in IBD, particularly in the earlier phases of disease and/or with limited sampling of the mucosa [26]. In mucosal biopsy specimens of both acute self-limited colitis and IBD, active and chronic inflammation and crypt damage may be focal or diffuse within the specimen. In IBD, active inflammation is distributed throughout the lamina propria whereas it tends to be limited to the upper half of the lamina propria in the self-limited conditions [5]; however, this feature may not be evident if only a single specimen is available for review. The major discriminator between acute non-relapsing colitis and IBD is the presence of features of chronicity in initial or follow-up biopsy specimens [5, 6, 7, 8, 11]. Having established chronicity, clinicians will correlate this finding with the remainder of the patient's profile to exclude known causes of chronic injury (such as Yersinia infections) before establishing a diagnosis of IBD.

Inflammatory Bowel Disease
The classic features of UC and colonic CD will be reviewed, followed by a discussion of newer and, in some cases, controversial findings that may obscure or delay the correct diagnosis. Accurate distinction of these two conditions has become increasingly important for several reasons. Surgery is classically considered curative for UC but not for CD. Patients with UC but not CD are candidates for creation of ileoanal pouches. Finally, specific therapies are now becoming available for these diseases; for example, use of imfliximab ( a chimeric monoclonal antibody to tumor necrosis factor alpha) is a successful management modality in patients with CD [27, 28, 29].

Clinical features
Both UC and CD develop more commonly in Caucasians than in other races, but they have an equal sex distribution and are common causes of colitis in young patients. Approximately 30% of cases are diagnosed before the age of 25 years; about 10% of all cases develop before 15 years of age [30]. The clinical findings of pediatric UC and CD are, in general, similar to those in adults (e.g., varying combinations of diarrhea, rectal bleeding, abdominal pain, weight loss, fever and anemia] [31]. However, features of IBD unique to pediatric patients are growth retardation and delayed puberty. Growth retardation is twice as prevalent in CD as compared to UC, is related more to disease activity than to corticosteroid therapy and is particularly associated with ileal disease [30, 31].

Natural history
IBD is characterized by exacerbations and remissions, but the natural histories of UC and CD are somewhat different [31]. At first presentation, UC is characterized by more extensive endoscopic and histologic disease in children than adults. In recent series, pancolitis has been documented in up to 50% of children compared to 8% to 20% of adults; conversely, proctitis or localized left-sided disease is present in approximately 25% of children but in up to half of adults at initial evaluation [30, 31, 32, 33, 34]. Over time, progression to more extensive disease, including pancolitis, occurs in 50% to 70% of children with initially limited disease (compared to 10% to 30% of adults), especially those with an early age of onset and/or severe disease [30, 35, 36, 37, 38]. At least one relapse will have been documented in about 90% of pediatric UC patients after 10 years of follow-up, and approximately one-third will have had a colectomy, chiefly because of unresponsiveness to medical therapy, prepubertal growth failure or fulminant colitis ("toxic megacolon") [30]. Since by current definition UC is a disease confined to the colon, proctocolectomy is curative, and the patients are excellent candidates for creation of an ileoanal pouch. In contrast, CD first presents as small bowel disease in 40% of patients, as ileocolitis in 30% and colitis in 30%. Early surgery is much more common than in UC: up to 50% of young patients with CD will have surgery for complications of the disease within 5 years of diagnosis and up to 90% will have an operation within 15 years [31]. Surgery is performed for reasons like those for UC and for additional CD-related indications including treatment of intestinal fistulas and abscesses, gut obstruction and intractable perianal disease [30]. Since CD is a condition that may involve the entire gastrointestinal tract, surgery is limited to removal of clinically significant lesions and is usually not curative. Asymptomatic recurrences have been documented in up to 70% of patients within one year of surgery, and symptomatic recurrent disease is common within five years. Recurrences typically occur proximal to sites of previous anastomoses or ostomies. Patients with CD are usually not considered good candidates for ileoanal pouches because of the risk of recurrent disease in the pouch [1, 30, 31].

Pathologic findings

The pathologic features of UC and CD are identical in children and adults and have been codified in many recent reviews [1, 5, 23, 39]. The distinguishing features of classic UC and colonic CD are summarized in Table 2. Classic UC begins in the rectum and, if it progresses, does so in a continuous retrograde fashion. Although by definition UC is confined to the colon, it may be accompanied by superficial mild non-specific mucosal inflammation in the terminal ileum ("backwash ileitis") (see later discussion). At microscopy, the inflammatory process in UC is superficial (confined to the mucosa and submucosa) and is diffuse within the mucosa of the involved segment [1, 39]. In contrast, colonic CD typically begins as localized right-sided or multifocal disease and progresses in a patchy fashion, with "skip areas" of uninvolved mucosa. Characteristic histologic findings include granulomas, deep or transmural inflammation (often characterized by the presence of lymphoid aggregates in the muscularis propria and/or at the muscularis/serosal interface), deep mural fissures or fistulas, and, in a minority of patients, a necrotizing or giant cell vasculitis. In the mucosa, the inflammatory lesions are often focal rather than diffuse [1, 39, 40]. Although diagnostically important, granulomas are not invariably present in otherwise typical cases of Crohn's colitis. Even with serial sectioning, granulomas are detected in only approximately one-third of mucosal biopsy specimens from patients with Crohn's colitis. However, granulomas have been found in 56% to 82% of CD colonic resections. This higher yield is related to the greater amount of tissue available for examination and to the fact that the submucosa is the most common site for granuloma formation [1, 40, 41]. In children with Crohn's colitis, granulomas tend to be more common in the rectosigmoid than elsewhere in the colon. Also, prospective data suggest that the prevalence of granulomas decreases with increasing duration of disease, perhaps due to the effects of medical therapy; thus, granulomas may be more often detected in children than in adults [42]. The sarcoid-like granulomas characteristic of CD must be distinguished from foreign body-type granulomas and from the non-specific mucin granulomas that may be present in both UC and CD. Mucin granulomas are typically adjacent to or in direct contact with inflamed or ruptured crypts, tend to be poorly formed and often contain giant cells [1, 39]. Their true nature can be determined by detection of intracytoplasmic mucin using stains such as the alcian blue-PAS with diastase pretreatment. Also characteristic of CD are so-called "microgranulomas": small collections of macrophages in the lamina propria away from damaged glands or crypts. They can be highlighted by positive immunohistochemical staining with CD68 [43]. In mucosal biopsy specimens from untreated patients, the presence of non-mucin granulomas or microgranulomas and focality of colitis are the best discriminators for CD [5, 23]. Focality of colitis in the rectosigmoid in children, however, should be interpreted with caution since it may be present at the onset of UC in this population (see later discussion). In adults, villiform surface change, crypt architectural distortion and crypt epithelial mucin depletion are said to be more common in UC than CD [5, 23], but I have not found these to be particularly helpful distinguishing characteristics in pediatric IBD. The pathologist can offer the largest amount of useful information if, at the onset of pediatric IBD, colonoscopy (rather than flexible sigmoidoscopy) with protocol sampling of even endoscopically unremarkable mucosa is performed. Of 42 pediatric patients ultimately proven to have IBD in one recent study, 10 had normal rectosigmoid biopsy specimens. Additional, more proximal, sampling confirmed a diagnosis of CD in 60%; the remaining four patients were later classified as either UC or indeterminate colitis [32].

Atypical Features and Controversial Issues in the Diagnosis of IBD



Unusual features in rectosigmoid mucosal biopsies at the onset of pediatric UC
At first presentation and before therapy, the majority of adult patients with UC (>90%) will have diffuse active colitis, usually with features of chronicity, in rectosigmoid mucosal specimens [24, 32]. Initial rectosigmoid specimens in children ultimately shown to have UC, however, demonstrate focal colitis and/or the absence of chronic changes in approximately one-third of patients and are completely normal in 4% to 8% [24, 32, 44]. These atypical findings are not specifically related to the patients' ages at the onset of colitis (although they are predominantly found in patients younger than 10 years), the duration of symptoms before endoscopy, the symptoms themselves, or the ultimate evolution of UC (i.e., development of diffuse distal disease, proximal progression over time) [26, 32, 44]. The reasons for these findings are unknown. One suggestion is that children may be evaluated earlier in the course of UC than adults [24, 32]; however, it is also clear that changes of chronicity may develop within a few weeks or months of symptom onset [8, 26]. This presentation of UC with focal disease and a paucity or lack of features of chronicity in pediatric patients raises several diagnostic possibilities and stresses the need for a complete evaluation of the patient. First, it should be recognized that ulcerative colitis is not excluded by these findings [19, 24, 32, 44]. Second, they may represent a non-relapsing, infectious-type colitis, which often is patchy and may have rectal sparing 44). Crohn's disease also enters the differential diagnosis; detection of focal proximal colitis, granulomas, ileal disease or perianal disease would support that diagnosis. Finally, allergic conditions in children typically exhibit focal colitis but can be excluded by the clinical workup and the absence of numerous eosinophils in tissue specimens [11]; also, in the follow-up of allergic colitis (as of most colonic bacterial infections), mucosal healing is characterized by reversion to a normal architecture with minimal or no features of chronic injury [6, 8, 11]. The predictive value of focal active colitis for development or recognition of CD once the confounding conditions discussed in the preceding paragraph have been eliminated has recently been examined. In a cohort of 29 pediatric patients with focal active colitis, 8 (28%) developed CD; most of the remainder had either infectious colitis or remained idiopathic [19]. In contrast, focal active colitis in adults evolved into a diagnosis of CD over time in fewer than 15% of patients [46, 47]. One possible reason for the difference in outcome between the two populations is that unlike the case in adults, colonoscopy in children is typically performed for evaluation of abdominal pain, diarrhea or hematochezia rather than cancer surveillance, thus creating a bias towards detection of inflammatory diseases. As in children, many cases of focal active colitis (particularly those with minor abnormalities) in adults are likely secondary to effects of bowel cleansing agents such as sodium phosphate [19, 46, 47].

Effects of medical therapy on the histology of UC in colonic mucosal biopsy specimens
The classic teaching has been that quiescent UC heals with fixed morphologic changes that permit continued recognition of the colonic mucosa as injured. In 1993, however, Odze and colleagues demonstrated that medical therapy of left-sided UC with topical 5-aminosalycylic acid caused reversion of colonic mucosa to a normal appearance in 64% of patients [48]. Since that time, several authors have confirmed and extended this observation. The results of these studies document that in patients with established extensive or pancolitis receiving contemporary medical therapy, histologic diffuse disease has become focal within the colon in up to 54% of patients and the rectum has become unremarkable in up to 34% on one or more occasions during follow-up [25, 49, 50, 51, 52]. These results were not related to the duration of disease or the type of therapy employed (systemic, topical, et cetera) [52]. Rectal sparing and focal colitis are typical of CD. Thus, to avoid diagnostic confusion, it is important to know the medication history of patients with presumed UC in whom such findings are documented. In both children and adults with chronic IBD, unfortunately, such information is often not available at the time of biopsy specimen interpretation. In this situation and in the absence of granulomas, focality and rectal sparing should be described but not interpreted, with the comment that prior medical therapy may have affected the histologic findings.

Appendiceal involvement as a "skip lesion" in UC
In several retrospective studies of colectomy specimens, the authors have attempted to determine the prevalence of appendiceal inflammation in patients with UC of various extents. Appendiceal involvement by UC must be distinguished from incidental acute appendicitis and has been defined as a lesion confined to the mucosa with architectural and cellular features of chronicity that may or may not be accompanied by activity [53]. Not unexpectedly, the prevalence of chronic appendicitis in patients with pancolitis (whether due to UC or CD) is high, being in the range of approximately 60% of cases [53, 54]. Of greater interest is the prevalence of this finding in UC patients with less than pancolitis (i.e., with at least cecal/right-sided sparing in the colectomy specimen). In this group, appendicitis has been reported in 15% to almost 100% of pediatric and adult patients, giving rise to the notion of appendicitis as a "skip lesion" in otherwise classic UC [53, 55, 56, 57, 58]. In a follow-up study of ileoanal anastomoses performed on patients with appendicitis as a presumed "skip lesion" of UC, there were no cases with outcomes suggesting that a diagnosis of CD had been missed [59]. A criticism of these studies relates to their retrospective nature [54, 60]. The sampling of the cecum and right colon has not been performed in a standardized fashion, so that the extent of sampling has been either limited or unknown. Also, information about prior medical therapy and its possible effects on damaged mucosa has not been provided. Thus, whether or not this portion of the colon has been truly "normal" at all times is debatable. However, appendiceal involvement has been documented in some cases that clearly represent otherwise localized left-sided UC [57]. At present, an apparent appendiceal "skip lesion" should not be construed as evidence of CD without additional supportive clinical, imaging and/or histologic data.

Cecal periappendiceal orifice inflammation as a "skip lesion" in UC
Another proposed "skip lesion" of UC is endoscopic and histologic colitis confined to the periappendiceal cecal mucosa in patients with left-sided UC. Strict criteria to exclude CD are needed to support this interpretation: no granulomas, perianal disease, ileal disease or other documented colonic skip lesions should be present [60]. The gross periappendiceal mucosal endoscopic findings vary from erythema and tissue granularity and friability to erosions and ulcers [60]. In several recent prospective endoscopic and biopsy surveys of patients with left-sided UC, the prevalence of periappendiceal localized colitis has varied from 54% to 75% [60, 61, 62, 63]. Biopsies examined from the colon between the left side and the periappendiceal area were normal in most instances [60, 61, 62] but on occasion demonstrated microscopic foci of colitis in endoscopically normal mucosa [63]. In the study by Yang and colleagues, the prevalence of focal periappendiceal disease was the same in patients with and without prior medical therapy [60], and in an 8-year follow-up of their patients, d'Haens et al detected no features of CD [61]. Although there are some issues with the interpretation of the histology in some of these series, the weight of evidence from these prospective studies supports the idea that the endoscopic periappendiceal cecal patch in children and adults does represent a skip lesion of UC and in isolation, therefore, should not be interpreted as evidence of CD.

The conundrum of backwash ileitis in UC
A perennial problem in IBD at the imaging, endoscopic and pathologic levels is the separation of ileitis due to CD from non-specific terminal ileal inflammation in patients with UC ("backwash ileitis"). Part of the problem relates to the lack of a modern consensus-derived gross and microscopic definition of backwash ileitis [39]. Also, the pathogenesis of the changes is unclear: originally though to represent a result of reflux of colonic contents into the terminal ileum, backwash ileitis is now felt to be a primary ileal process [64]. A contemporary provisional diagnosis of backwash ileitis in UC has been provided by Heuschen et al: a 5.0 cm or longer segment of terminal ileal mucosal inflammation with no granulomas. In resection specimens, no mural inflammation should be identified [64]. Using this definition, backwash ileitis was identified in colectomy specimens in 22% of predominantly adult patients with pancolitis and in none of those with localized left-sided disease; in this cohort, it predicted an increased risk of colorectal carcinoma [64]. In a recent pediatric series, backwash ileitis (not precisely defined) was identified in 22% of colectomy specimens with pancolitis. Its presence did not predict any excess of problems with ileoanal pouches (such as refractory pouchitis and pouch failure) (29). A combined imaging, endoscopic and histologic study of backwash ileitis (not precisely defined) published in 2003 gave a prevalence of 39% in 18 pediatric patients with UC [65]. In the absence of a standardized definition, such studies do not contribute to our understanding of this condition and its precise relation to IBD.

Upper gastrointestinal tract involvement in inflammatory bowel disease
Upper gastrointestinal inflammatory lesions of the esophagus, stomach, and/or duodenum in patients with CD have been well described in the literature. Among the possible findings, granulomas, focal duodenitis and focal gastritis in mucosal biopsy specimens are most closely associated with or are the best predictors of colonic CD [66, 67]. Focal (or "focally enhanced") gastritis is defined as a localized and sharply defined area of gastritis, with active and chronic inflammation and gland injury, that is surrounded by normal mucosa and is easily identified at low-power inspection [66, 67, 68, 69]. In contemporary medical practice, upper gastrointestinal endoscopy with biopsy in often performed before the institution of therapy in pediatric patients with newly diagnosed colonic IBD (both UC and CD). This procedure has lead to some interesting findings. The protocols and the extent of histologic sampling have not been uniform in these studies, but most are prospective in nature. Although the gross and microscopic abnormalities are not always described in detail, an intriguing finding in these studies is that the overall prevalence of endoscopic and histologic inflammatory lesions in the esophagus, stomach and duodenum is roughly equal in patients with newly diagnosed and typical colonic UC and CD. When known causes of such inflammation (such as reflux esophagitis and Helicobacter pylori-associated gastritis) are excluded, there still remains a high prevalence of non-specific lesions, particularly H. pylori-negative diffuse gastritis, in both conditions [69, 70, 71, 72]. Whether this gastritis is incidental or related to IBD is unknown, but the important point is that the mere presence of upper gastrointestinal inflammatory lesions can no longer be used to automatically categorize a patient as having CD. The presence of granulomas in upper intestinal mucosal biopsies is highly specific for a diagnosis of CD. Their prevalence has varied from 25% to 60% in recent pediatric series, and they are most often found in gastric mucosa [69, 70, 71, 72]. In contrast, they are much less common in the stomach of adults with CD (a prevalence of only 5% in the study of Parente and colleagues) [68]. In children, they are often detected in gastric and duodenal specimens even when synchronous colonic mucosal biopsies are negative for this finding [69, 71, 72]. Although focal gastritis can be seen in a minority of UC patients (8% to 12%) and non-IBD controls (19%), it is more common in patients with CD (43% to 52%) with a calculated positive predictive value for colonic CD of 70% to 80% [68, 69]. All examples of gastritis, whether focal or diffuse, should be carefully evaluated to exclude Helicobacter infections as well as allergic and chemical-type injuries. As in the colon, "IBD-associated gastritis" is a diagnosis of exclusion. In summary, upper endoscopy can be helpful in classifying colonic IBD. Biopsy specimens should be obtained from grossly normal as well as abnormal mucosa to detect treatable localized conditions, such as H. pylori-associated gastritis, and to detect granulomas and focal gastritis [68, 69, 70, 72]. In one recent pediatric study of the stomach in IBD, 87% of patients had various inflammatory changes in antral biopsies despite the fact that only 38% had endoscopically detectable mucosal changes (chiefly erythema and small ulcers) [69]. The recent recognition that diffuse, non-Helicobacter chronic gastritis is common in patients with classic UC raises an interesting question: Is it a manifestation of UC? Perhaps adding credence to this possibility is a small number of pediatric and adult cases of "ulcerative enteritis" developing in patients with classic UC, typically after colectomy for treatment of the colitis. In these patients, as recently reviewed by Rubenstein and colleagues, the enteritis is diffuse, confined to the mucosa, lacks granulomas, and responds to the usual medical therapies for UC [73]. Clarification of the exact nature of this enteritis (as well as the non-specific gastritis) will require additional prospective studies.

Indeterminate colitis
Approximately 5% to 10% of cases of IBD at some point in their course cannot be accurately classified as UC on CD based on evaluation of mucosal biopsy specimens or even after examination of colectomies. Such cases have been categorized as "indeterminate colitis", with the implication that many ( but not all) of these cases will declare their true identity over time [39, 74, 75]. Imprecision in diagnosis may relate to a lack of or insufficient clinical, imaging or endoscopic data, or, in biopsy specimens, to issues previously discussed. Among patients with indeterminate colitis, approximately 60% are ultimately reclassified as UC or CD; however, the others remain diagnostically obscure even after prolonged follow-up [75]. Reclassified cases turn out to be UC more often than CD [76]. In resection specimens, the major causes of diagnostic uncertainty are those cases clinically felt to be UC that have atypical gross features, such as deep-fissuring ulcers, suggestive of CD. This scenario develops in patients with fulminant colitis in whom bowel dilatation and subsequent ischemic injury are superimposed on the underlying IBD. Microscopically, deep mural inflammation is present, often with ulcers and fissures extending into the muscularis propria. Acute ischemic damage features marked vascular congestion, myocytolysis, and active inflammation accompanying the deep ulcers, all findings reminiscent of CD. In fulminant colitis, the best features to support a diagnosis of CD specifically are granulomas and/or transmural lymphoid nodules away from the areas of deep ulceration [39, 74, 75, 77]. In the absence of these features and mural ileal disease, such cases can be provisionally classified as indeterminate colitis. As noted earlier, subsequent evaluation will clarify the true nature of the disease in most cases [39, 75]. Serologic testing (p-ANCA; ASCA} has not yet proven to be especially helpful in sorting out cases of indeterminate colitis [78, 79].

Table 1
Histologic Features of Chronic Colitis

  • Increased mononuclear inflammation in the lamina propria

  • Crypt distortion and atrophy

  • Surface villiform change

  • Basal plasmacytosis and lymphoid aggregates

  • Paneth cell metaplasia (distal colon)



Table 2
Classic Distinguishing Features of Ulcerative Colitis (UC) and Crohn's Colitis (CC)

UC CC
Gross features
Isolated right-sided colitis No Yes
Rectal involvement Yes Variable
Distribution Diffuse Diffuse or focal
Method of extension Continuous(a) Often discontinuous ("skip areas" of normal colon)
Involvement of gut proximal to colon No(b) Common
Fistulas No Occasional
"Creeping" serosal fat No Common
Thickened bowel wall No Yes
Strictures Rare Occasional
Microscopic features
Inflammation confined to mucosa and submucosa Yes Uncommon
Transmural inflammation No(c) Common
Fissuring ulcers No(c) Yes
Fistulas No Yes
Sarcoid-like granulomas No Yes
Distribution of inflammatory changes in mucosal specimens Diffuse Focal or diffuse
Vasculitis No Yes

(a)Recent studies document appendiceal or cecal involvement in some cases of localized left-sided UC

(b)Possible upper gastrointestinal involvement by otherwise classic UC is a subject of current debate (see text for details)

(c)May be present in fulminant colitis/toxic megacolon



Table 3
Atypical Features and Controversial Issues in Pediatric
and Adult Inflammatory Bowel Disease

  • Atypical findings in rectosigmoid mucosal biopsies at the onset of ulcerative colitis

  • Effects of medical therapy on the histology of ulcerative colitis

  • Appendiceal involvement as a "skip lesion" in ulcerative colitis

  • Periappendiceal cecal inflammation as a "skip lesion" in ulcerative colitis

  • Backwash ileitis

  • Upper gastrointestinal involvement in Crohn's disease and ulcerative colitis

  • Indeterminate colitis
References
  1. Antonioli DA. Pediatric inflammatory bowel disease. Pediatr Develop Pathol 2005; 8:2-19.

  2. Goldman H. Interpretation of large intestinal mucosal biopsy specimens. Hum Pathol 1994; 25:1150-1159.

  3. Levine DS, Haggitt RC. Colon. In: Sternberg SS, ed. Histology for Pathologists, Second Edition. Philadelphia , Lippincott-Raven, 1997, pp 519-537.

  4. Symonds DA. Paneth cell metaplasia in diseases of the colon and rectum. Arch Pathol 1974; 97:343-347.

  5. Tsang P, Rotterdam H. Biopsy diagnosis of colitis. Am J Surg Pathol 1999; 23:423-430.

  6. Nostrant TT, Kumar NB, Appelman HD. Histopathology differentiates acute self-limited colitis from ulcerative colitis. Gastroenterology 1987; 92:318-328.

  7. Jenkins D, Balsitis M, Gallivan MF, et al. Guidelines for the initial biopsy diagnosis of suspected chronic idiopathic inflammatory bowel disease: the British Society of Gastroenterology initiative. J Clin Pathol 1997; 50:93-105.

  8. Tanaka M, Riddell R, Saito H, et al. Morphologic criteria applicable to biopsy specimens for effective distinction of inflammatory bowel disease from other forms of colitis and of Crohn's disease from ulcerative colitis. Scand J Gastroenterol 1999; 34:55-67.

  9. Lowichik A, Weinberg AG. A quantitative evaluation of mucosal eosinophils in the pediatric gastrointestinal tract. Mod Pathol 1996; 9:110-114.

  10. Pascal RR, Gramlich TL, Parker KM, Gansler TS. Geographic variation in eosinophil concentration in normal colonic mucosa. Mod Pathol 1997; 10:363-365.

  11. Odze RD , Wershil BK, Leichtner AM, Antonioli DA. Allergic colitis in infants. J Pediatr 1995; 126:163-170.

  12. O'Leary AD, Sweeney EC. Lymphoglandular complexes of the colon: structure and distribution. Histopathology 1986; 10:267-283.

  13. Snover DC, Sandstad J, Hutton S. Mucosal pseudolipomatosis of the colon. Am J Clin Pathol 1985; 84:575-580.

  14. Watts DA, Lessells AM, Penman ID, Ghosh S. Endoscopic and histologic features of sodium phosphate bowel preparation-induced colonic ulceration: case report and review. Gastrointest Endosc 2002; 55:584-587.

  15. Driman DK, Preiksaitis HG. Colorectal inflammation and increased cell proliferation associated with oral sodium phosphate bowel preparation solutions. Hum Pathol 1998; 29:972-978.

  16. Zwas FR, Cirillo NW, El-Serag HB, et al. Colonic mucosal abnormalities associated with oral sodium phosphate for colonoscopy preparation. Gastrointest Endosc 1998; 48:276-278.

  17. Berkelhammer C, Ekambaram A, Silva RG. Low-volume oral colonoscopy bowel preparation: sodium phosphate and magnesium citrate. Gastrointest Endosc 2002; 56:89-94.

  18. Wang NACS, Penman ID, Campbell S, et al. Microscopic focal cryptitis associated with oral sodium phosphate preparation. Histopathology 2000; 36:476-478.

  19. Xin W, Brown PI, Greenson JK. The clinical significance of focal active colitis in pediatric patients. Am J Surg Pathol 2003; 27:1134-1138.

  20. Jonas G, Mahoney A, Murray J, Gertler S. Chemical colitis due to endoscope cleaning solutions: a mimic of pseudomembranous colitis. Gastroenterology 1988; 95:1403-1408.

  21. Meisel JL, Bergmen D, Graney D, et al. Human rectal mucosa: prospective and morphological changes caused by laxatives. Gastroenterology 1977; 72:1274-1279.

  22. West AB, Kuan S-F, Bennick M, Lagarde S. Gluteraldehyde colitis following endoscopy: clinical and pathologic features and investigation of an outbreak. Gastroenterology 1995; 108:1250-1255.

  23. LeBerre N, Heresbach D, Kerbaol M, et al. Histological discrimination of idiopathic inflammatory bowel disease from the other types of colitis. J Clin Pathol 1995; 48:749-753.

  24. Washington K, Greenson JK, Montgomery E, et al. Histopathology of ulcerative colitis in initial rectal biopsy in children. Am J Surg Pathol 2002; 26:1441-1449.

  25. Geboes K, Dalle I. Influence of treatment on morphologic features of mucosal inflammation. Gut 2002; 50:S37-S42.

  26. Robert ME, Tang L, Hao LM, Reyes-Mugica M. Patterns of inflammation in mucosal biopsies of ulcerative colitis: perceived differences in pediatric populations are limited to children younger than 10 years. Am J Surg Pathol 2004; 28:183-189.

  27. Farmer M, Petras RE, Hunt LE, et al. The importance of diagnostic accuracy in colonic inflammatory bowel disease. Am J Gastroenterology 2000; 95:3184-3188.

  28. Sandborn WJ, Hanauer SB. Imfliximab in the treatment of Crohn's disease: a user's guide for clinicians. Am J Gastroenterol 2002; 97:2962-2972.

  29. Alexander F, Sarigol S, DiFiore J, et al. Fate of the pouch in 151 pediatric patients after ileal pouch anal anastomosis. J Pediatr Surg 2003; 38:78-82.

  30. Langholz E, Munkholm P, Krasnilikoff PA, et al. Inflammatory bowel diseases with onset in childhood. Clinical features, morbidity, and mortality in a regional cohort. Scand J Gastroenterol 1997; 32:139-147.

  31. Israel EJ, Kleinman RE. Inflammatory bowel disease: diagnosis and treatment. Semin Gastrointest Dis 1994; 5:95-105.

  32. Glickman JN, Bousvaros A, Farraye FA, et al. Pediatric patients with untreated ulcerative colitis may present initially with unusual morphologic findings. Am J Surg Pathol 2004; 28:190-197.

  33. Escher J, ten Kate F, Lichtenbelt K, et al. Value of rectosigmoidoscopy with biopsies for diagnosis of inflammatory bowel disease in children. Inflamm Bowel Dis 2002; 8:16-22.

  34. Mamula P, Telega GW, Markowitz JE, et al. Inflammatory bowel disease in children 5 years of age and younger. Am J Gastroenterol 2002; 97:2005-2010.

  35. Powell-Tuck J, Richie JK, Lennard-Jones JE. The prognosis of idiopathic proctitis. Scand J Gastroenterol 1977; 12:727-732.

  36. Farmer RG. Long-term prognosis for patients with ulcerative proctosigmoiditis (ulcerative colitis confined to the rectum and sigmoid colon). J Clin Gastroenterol 1979; 1:47 -50.

  37. Mir-Madjlessi SH, Michener WM, Farmer RG. Course and prognosis of idiopathic ulcerative proctosigmoiditis in young patients. J Pediatr Gastroenterol Nutr 1986; 5:570-575.

  38. Hyams J, Davis P, Lerer T, et al. Clinical outcome of ulcerative proctitis in children. J Pediatr Gastroenterol Nutr 1997; 25:149-152.

  39. Odze R. Diagnostic problems and advances in inflammatory bowel disease. Mod Pathol 2003; 16:347-358.

  40. Wakefield AJ, Sankey EA, Dhillon AP, et al. Granulomatous vasculitis in Crohn's disease. Gastroenterology 1991; 100:1279-1287.

  41. Surawicz CM, Meisel JL, Ylvisaker T, et al. Rectal biopsy in the diagnosis of Crohn's disease: value of multiple biopsies and serial sectioning. Gastroenterology 1981; 81:66-71.

  42. Schmitz-Moorman P, Schag M. Histology of the lower intestinal tract in Crohn's disease of children and adolescents: multicentric paediatric Crohn's disease study. Pathol Res Pract 1990; 186:479-484.

  43. Yao K, Yao T, Iwashita A, et al. Microaggregate of immunostained macrophages in non-inflamed gastroduodenal mucosa: a new useful histological marker for differentiating Crohn's colitis from ulcerative colitis. Am J Gastroenterol 2000; 95:1967-1973.

  44. Markowitz J, Kahn E, Grandcher K, et al. Atypical rectosigmoid histology in children with newly diagnosed ulcerative colitis. Am J Gastroenterol 1993; 88:2034-2037.

  45. Mantzaris GJ, Hatzis A, Archavlis E, et al. The role of colonoscopy in the differential diagnosis of acute, severe hemorrhagic colitis. Endoscopy 1995; 27:645-653.

  46. Greenson JK, Robert SA, Carpenter SL, et al. The clinical significance of focal active colitis. Hum Pathol 1997; 28:729-733.

  47. Volk EE, Shapiro BD, Easley KA, et al. The clinical significance of a biopsy-based diagnosis of focal active colitis: a clinicopathologic study of 31 cases. Mod Pathol 1998; 11:789-794.

  48. Odze R, Antonioli D, Peppercorn M, Goldman H. Effect of topical 5-aminosalicylic acid (5-ASA) therapy on rectal mucosal morphology in chronic ulcerative colitis. Am J Surg Pathol 1993; 17:869-875.

  49. Bernstein CN, Shanahan F, Anton PA, Weinstein WM. Patchiness of mucosal inflammation in treated ulcerative colitis: a prospective study. Gastrointest Endosc 1995; 42:232-237.

  50. Levine TS, Tzardi M, Mitchell S, et al. Diagnostic difficulty arising from rectal recovery in ulcerative colitis. J Clin Pathol 1996; 49:319-323.

  51. Kleer CG, Appelman H. Ulcerative colitis: patterns of involvement in colorectal biopsies and changes with time. Am J Surg Pathol 1998; 22:983-989.

  52. Kim B, Barnett JL, Kleer CG, et al. Endoscopic and histological patchiness in treated ulcerative colitis. Am J Gastroenterol 1999; 94:3258-3262.

  53. Scott IS, Sheaff M, Coumbe A, et al. Appendiceal involvement in ulcerative colitis. Histopathology 1998; 33:168-173.

  54. Goldblum JR, Appelman HD. Appendiceal involvement in ulcerative colitis. Mod Pathol 1992; 5:607-610.

  55. Davison AM, Dixon MF. The appendix is a "skip lesion" in ulcerative colitis. Histopathology 1990; 16:93-95.

  56. Kahn E, Markowitz J, Daum F. The appendix in inflammatory bowel disease in children. Mod Pathol 1992; 5:380-383.

  57. Groisman GM, George J, Harpaz N. Ulcerative appendicitis in universal and nonuniversal ulcerative colitis. Mod Pathol 1994; 7:322-325.

  58. Kroft SH, Stryker SJ, Rao MS. Appendiceal involvement as a skip lesion in ulcerative colitis. Mod Pathol 1994; 7:912-914.

  59. Perry WB, Opelka FG, Smith D, et al. Discontinuous appendiceal involvement in ulcerative colitis: pathology and clinical correlation. Gastrointest Surg 1999; 3:141-144.

  60. Yang SK, et al. Appendiceal orifice inflammation as a skip lesion in ulcerative colitis: an analysis in relation to medical therapy and disease extent. Gastrointest Endosc 1999; 49:743-747.

  61. d'Haens G, Geboes K, Ponette E, et al. Patchy cecal inflammation associated with distal ulcerative colitis: a prospective endoscopic study. Am J Gastroenterol 1997; 92:1275-1279.

  62. Okawa K, Aoki T, Sano K, et al. Ulcerative colitis with skip lesions at the mouth of the appendix: a clinical study. Am J Gastroenterol 1998; 93:2405-2410.

  63. Matsumoto T, Nakamura S, Shimizu M, Iida M. Significance of appendiceal involvement in patients with ulcerative colitis. Gastrointest Endosc 2002; 55:180-185.

  64. Heuschen UA, Hinz U, Allemeyer EH, et al. Backwas ileitis is strongly associated with colorectal carcinoma in ulcerative colitis. Gastroenterology 2001; 120:841-847.

  65. Laghi A, Borrelli P, Paolantonio P, et al. Contrast enhanced magnetic resonance imaging of the terminal ileum in children with Crohn's disease. Gut 2003; 52:393-397.

  66. Oberhuber G, Puspok A, Oesterreicher C, et al. Focally enhanced gastritis: a frequent type of gastritis in patients with Crohn's disease. Gastroenterology 1997; 112:698-706.

  67. Wright CL, Ridell RH. Histology of the stomach and duodenum in Crohn's disease. Am J Surg Pathol 1998; 22:383-390.

  68. Parente F, Cucino C, Bollani S, et al. Focal gastric inflammatory infiltrates in inflammatory bowel diseases: prevalence, immunohistochemical characteristics, and diagnostic role. Am J Gastroenterol 2000; 95:705-711.

  69. Kundhal PS, Stormon MQ, Zacho M, et al. Gastric antral biopsy in the differentiation of pediatric colitides. Am J Gastroenterol 2003; 98:557-561.

  70. Ruuska T, Vaajalahti P, Arajärvi P, Mäki M. Prospective evaluation of upper gastrointestinal mucosal lesions in children with ulcerative colitis and Crohn's disease. J Pediatr Gastroenterol Nutr 1994; 19:181-186.

  71. Tobin JM, Sinha B, Ramani P, et al. Upper gastrointestinal mucosal disease in pediatric Crohn's disease and ulcerative colitis: a blinded, controlled study. J Pediatr Gastroenterol Nutr 2001; 32:443-448.

  72. Abdullah BA, Gupta SK, Croffie JM, et al. The role of esophagogastroduodenoscopy in the initial evaluation of childhood inflammatory bowel disease: a 7-year study. J Pediatr Gastroenterol Nutr 2002; 35:636-640.

  73. Rubenstein J, Sherif A, Appelman H, Chey WD. Ulcerative colitis-associated enteritis: is ulcerative colitis always confined to the colon? J Clin Gastroenterol 2004; 38:46-51.

  74. Price AP. Overlap in the spectrum of non-specific inflammatory bowel disease-"colitis indeterminate". J Clin Pathol 1978; 31:567-577.

  75. Geboes K, DeHertogh G. Indeterminate colitis. Inflammatory Bowel Disease 2003; 9:324-331.

  76. Wells AD, McMillan I, Price AB, et al. Natural history of indeterminate colitis. Br J Surg 1991; 79: 179-181.

  77. Swan NC, Goeghan JG, O'Donohue DP, et al. Fulminant colitis in inflammatory bowel disease. Dis Colon Rectum 1998; 41: 1511-1515.

  78. Rutgeerts P, Vermeire S. Serological diagnosis of inflammatory bowel disease. Lancet 2000; 356:2117-2118.

  79. Joossens S, Reinisch W, Vermeire S, et al. The value of serologic markers in indeterminate colitis: a prospective follow-up study. Gastroenterology 2002; 122:1242-1247.
Additional recent references
  1. Geboes K, Van Assche G, D'hoore A. Handling the hot colon: when not to do a pouch. Pathol Case Reviews 2004; 9:97-105. (Indeterminate/fulminant colitis).

  2. Heresbach D, Alexandre JL, Branger B, et al. Frequency and significance of granulomas in a cohort of incident cases of Crohn's disease. Gut 2005; 54:215-222.

  3. Hofstaeder F. Granuloma formation in the different phenotypes of Crohn's disease. Gut 2005; 54:180-181.

  4. Pierik M, DeHertogh G, Vermiere S, et al. Epithelioid granulomas, pattern recognition receptors, and phenotypes of Crohn's disease. Gut 2005; 54:223-227.