Non-Neoplastic and Inflammatory Disorders of the Small Bowel

Flattened Villi Pattern

The flattened villi pattern of injury has also been called the malabsorption pattern and is characterized by villous blunting, crypt hyperplasia with or without significant intraepithelial lymphocytosis, and expansion of the lamina propria by lymphocytes and plasma cells. The prototypical malabsorption disorder is celiac disease (CD), but a number of other disorders can produce similar histologic findings, including peptic injury, tropical sprue, collagenous sprue, nongluten protein sensitivity, small intestinal bacterial overgrowth (SIBO), autoimmune enteropathy, common variable immunodeficiency (CVID), and drug-related sprue-like enteropathy, among others.

Intact Villi Pattern

Several conditions result in small bowel mucosal injury without altering the villous architecture. These may or may not be accompanied by prominent inflammation and can be further categorized based on the inflammatory component into conditions with neutrophilic inflammation, eosinophilic inflammation, lymphoplasmacytic inflammation, and disorders with minimal or no inflammation.

Applying the Pattern-Based Approach

Identifying the two major patterns of small bowel injury (flat villi and intact villi) is helpful in generating a differential diagnosis and determining the cause of mucosal injury. However, the patterns are not etiologically specific. For instance, many conditions can manifest with flattened villi pattern of injury. Furthermore, one specific etiology may result in a variety of patterns of injury. For instance, NSAID-related injury can manifest with all major patterns of injury. It is therefore important to identify the pattern of injury and provide a comment enlisting all the potential etiologies that may contribute to injury. The sections in this chapter have been organized based on causes of small bowel injury.

Clinical Features

Once considered to be a rare childhood disorder, CD (also known as gluten-sensitive enteropathy or celiac sprue) is a chronic immune-mediated disorder induced by dietary exposure to gluten in genetically predisposed individuals. Confirmation of a diagnosis also requires prompt improvement of clinical symptoms and mucosal pathology after withdrawal of dietary gluten.

Celiac disease occurs both in adults and children with a female predominance (female-to-male ratio, 2–3 to 1). It is estimated to affect between 0.5% and 1% of individuals in both Europe and the United States. The peak age of incidence is the third to fourth decades of life. However, up to 20% of individuals may be diagnosed after 60 years of age.

The classification and diagnosis of CD are based on GI manifestations; however, patients with extraintestinal complications are increasingly being recognized ( Table 5.2 ). The clinical presentation of CD ranges from asymptomatic to severe malnutrition. CD may be divided into four clinical subtypes, two symptomatic types and two asymptomatic types. In the classic symptomatic subtype, patients present with chronic diarrhea, abdominal distension, and pain. In the atypical symptomatic subtype, GI symptoms may be less pronounced to absent. Instead, extraintestinal features such as anemia, short stature, neurologic symptoms, dermatitis herpetiformis, delayed puberty, osteoporosis, infertility, vitamin and protein deficiencies, and elevated liver enzyme levels are predominant. Some patients have asymptomatic or silent CD; they lack classic or atypical symptoms but have unequivocal villous atrophy found incidentally for other reasons or after serologic screening. Finally, latent CD is defined by a positive serology but lack of villous atrophy on initial biopsies. Individuals are asymptomatic but may develop symptoms, histologic changes, or both.

High-risk groups to be screened for CD are those with other autoimmune disorders such as type 1 diabetes mellitus and autoimmune thyroiditis, a family history of CD, selective immunoglobulin (Ig) A deficiency, or dermatitis herpetiformis. CD is also more common in certain pediatric populations, including those with Down syndrome, in whom the frequency of CD is as high as 10%. In this group, CD is usually not detected on the basis of GI symptoms.

Celiac disease is a clinicopathologic diagnosis that is most often based on positive serology and specific histologic findings. Antitissue transglutaminase (anti-tTG) antibodies are both highly sensitive (90%–98%) and specific (95%–97%) for CD. Antiendomysial (EMA) antibodies are specific (97%–100%) but have slightly inferior sensitivity (85%–98%). Antigliadin antibody testing should no longer be used because of its low positive predictive value. A diagnosis of CD may be made if both tissue biopsy and serologic testing are consistent with CD. CD may be excluded if results of both are negative.

Sometimes there is discordance between the histologic and serologic findings, which may occur for a number of reasons. First, IgA deficiency is more common in individuals with CD than in the general population. Because tTG and EMA serologic tests are based on IgA, they may be falsely negative in patients with IgA deficiency. Total IgA can be measured to identify these patients, and IgG-based assays, particularly deamidated gliadin peptide or IgG-tTG, may be used in these situations. Second, the flattened villi pattern of injury is not specific for CD. Consideration of an alternative cause (e.g., SIBO, tropical sprue, drug-associated enteropathy) should be explored in patients with negative celiac serologies. Sometimes HLA genotyping is useful from excluding the possibility of CD. HLA-DQ2 or HLA-DQ8 is present in almost all cases of CD. Negative testing for these alleles virtually eliminates the possibility of CD. Human leukocyte antigen (HLA) typing is particularly useful in patients who may have initiated a gluten-free diet (GFD) before proper diagnostic work-up. This complicates diagnosis because both serologic and histologic abnormalities may normalize with GFD. HLA genotyping can be used to rule out CD in a subset of these patients.

In patients who test positive for HLA-DQ2 or HLA-DQ8, a gluten challenge can be performed, in which biopsy and serologic analysis takes place after gluten has been reintroduced for a period time. In patients with CD, gluten challenge should elicit typical serologic and histologic features. However, some patients may experience a return of their symptoms (e.g., abdominal pain, bloating) but without serologic or histologic evidence of CD. An alternate diagnosis should be considered in these patients, such as IgE-mediated wheat allergy or nonceliac gluten sensitivity.

Genetic

The major genetic risk factors for CD are the HLA class II genes HLA-DQ2 and HLA-DQ8 . Approximately 95% of all patients with CD have a DQ2 heterodimer, and almost all remaining patients with CD have a DQ8 heterodimer. Gene dosage correlates with the risk for developing CD, with patients homozygous for DQ2 carrying the greatest risk. CD is concordant in 70% to 80% of monozygotic twins, 30% to 40% of HLA-identical siblings, and fewer than 20% of dizygotic twins, a rate similar to that for all first-degree relatives. However, recent linkage studies suggest that non-HLA disease-associated genes may also play a role.

Pathologic Features

Gross Findings

If the mucosal biopsy specimen is examined under a dissecting microscope (which is seldom done), delicate villi are found to be attenuated or even absent. Endoscopically, a reduction in folds, scalloping, a mosaic pattern, or nodular mucosa may be seen ( Fig. 5.1 ).

Microscopic Findings

Histologic abnormalities in the duodenum may be patchy. Current guidelines from the American College of Gastroenterology (ACG) recommend multiple biopsies of the duodenum, with one or two biopsies from the bulb and at least four biopsies from the distal duodenum. CD produces the flattened villi pattern of injury—intraepithelial lymphocytosis, villous blunting, and crypt hyperplasia ( Figs. 5.2 and 5.3 ). The earliest change in CD is the presence of increased intraepithelial lymphocytes (IELs), most conspicuously within the villous tips, with intact villous architecture ( Fig. 5.4 ). With persistent disease, epithelial damage ensues with reactive mucin loss. The lamina propria is expanded by lymphocytes and plasma cells, thus contributing to partial or complete villous blunting. Villous blunting is accompanied by crypt hyperplasia. Active inflammation is occasionally seen but is generally not prominent. Marked neutrophilic epithelial injury should suggest an additional or alternative cause, such as peptic duodenitis, NSAID injury, infection, or Crohn’s disease.

Duodenal biopsies exhibiting all the features of CD—increased IELs, villous blunting, crypt hyperplasia—are easy to recognize. However, changes of early CD can be quite subtle and include intact villous and crypt architecture with increased IELs. The increase in IELs is sometimes obvious, but in some cases, an objective measure is needed to resolve borderline cases. The normal range of IELs has been reported as 11 to 23 per 100 enterocytes. According to ACG guidelines, more than 25 IELs per 100 enterocytes is abnormal. One rapid method to determine this number is to evaluate 5 villous tips, counting the number of lymphocytes present within the distal-most 20 enterocytes. The final values can be expressed as IEL per 100 enterocytes. When possible, evaluate only well-oriented villi that are anchored by intact muscularis mucosae because tangentially sectioned lateral aspects of the villi may result in overestimation and inaccurate estimation of the lymphocyte count. A CD3 immunostain can also be used to highlight IELs. However, the utility of this approach has not been established or validated for clinical practice. Some studies have shown that CD3 immunohistochemistry (IHC) does not improve detection of CD in cases in which the hematoxylin and eosin–stained sections are normal.

Upon initiation of a GFD, patients often report a marked improvement in clinical symptoms. Microscopically, intestinal biopsies show diminished surface epithelial injury, a reduced number of IELs, and partial to complete resolution of villous architectural abnormalities. However, in some cases, despite resolution of clinical symptoms and normal serology, the inflammatory component may persist for up to a period of 1 to 2 years after GFD. If patients consume gluten-containing products, there is a rapid return of all lesions, and malabsorption ensues.

Celiac Disease—Pathologic Features

Gross Findings

• ■

  Attenuated to absent villi (as seen under a dissecting microscope)

• ■

  Scalloping or mosaic appearance of the mucosa

Microscopic Findings

• ■

  Variable degrees of villous blunting

• ■

  Intraepithelial lymphocytosis, most prominent on villous tips (>25 IELs per 100 enterocytes)

• ■

  Crypt hyperplasia

Differential Diagnosis

• ■

  Bacterial overgrowth

• ■

  Infection

• ■

  Tropical sprue

• ■

  Nongluten protein sensitivity

• ■

  Crohn’s disease

• ■

  Protein intolerance

• ■

  Autoimmune enteropathy

• ■

  Medications (nonsteroidal antiinflammatory drugs, olmesartan)

Differential Diagnosis

The flattened villi pattern of injury is not specific for CD, and therefore, a diagnosis of CD can only be confirmed with the knowledge of positive celiac serologies. Without this information, the pathologist must be descriptive in the report by stating the pattern of injury (e.g., duodenal mucosa with villous blunting and increased IELs) and providing a comment with the differential diagnosis. In biopsies showing well-developed features of the flattened villi pattern of injury, the differential diagnosis includes infection, protein intolerance, tropical sprue, nongluten protein sensitivity, bacterial overgrowth, Crohn’s disease, CVID, autoimmune enteropathy (AIE), and medication injury (particularly from NSAIDs and olmesartan). In biopsies with increased IELs and normal villous and crypt architecture, the differential diagnosis is slightly different and includes viral infections, H. pylori infection, NSAID injury, peptic injury, tropical sprue, nongluten protein sensitivity, bacterial overgrowth, Crohn’s disease, and AIE.

Clinical Features

Peptic duodenitis and peptic ulcer disease (PUD) represent a continuum of the same disease process, namely damage to the duodenal mucosa caused by exposure to excessive gastric acid. It is estimated to affect up to 10% of the Western population. Peptic disease is more common among men who are older than 40 years. Patients often present with burning epigastric pain relieved by eating. In severe cases, the pain may be constant and accompanied by nausea and vomiting.

Chronic infection with H. pylori is highly correlated with peptic disease of the duodenum and is associated with more than 80% of peptic ulcers. Mechanistically, this occurs with antrum-predominant H. pylori infection, which stimulates gastrin secretion, and in turn, increases acid production. Furthermore, peptic injury often results in gastric foveolar metaplasia, which can become secondarily colonized with H. pylori , further contributing to mucosal injury. Gastric foveolar metaplasia may represent an adaptation to chronic exposure to hyperacidity. Other factors, such as smoking, NSAIDs, renal insufficiency, or duodenal dysmotility, compromise mucosal defense mechanism and allow prolonged contact with gastric acid, resulting in peptic injury.

Patients with multiple duodenal ulcers represent the most severe form of the disease; multiple duodenal or jejunal ulcers should raise concern for ZES. In the absence of ZES, refractory ulcers may be seen in patients with chronic NSAID use, gastric outlet obstruction or duodenal stenosis, or those with a history of gastric bypass procedure.

Pathologic Features

Gross Findings

Peptic duodenitis is most commonly seen in the duodenal bulb. The endoscopic appearance varies from simple erythema to friability, erosions, and nodularity of the mucosa ( Fig. 5.7 ). Nodularity and polypoid lesions most often correspond to Brunner’s gland hyperplasia or prominent foveolar hyperplasia. In PUD, the ulcers are usually circular and rarely exceed 3 cm in diameter. They are typically located along the posterior wall of the duodenal bulb and are more likely to present with perforation and massive hemorrhage because of their proximity to pancreaticoduodenal and gastroduodenal arteries.

Microscopic Findings

In general, peptic duodenal disease shows the intact villi with a neutrophilic inflammation pattern of injury. There are expansion of the lamina propria by neutrophils and neutrophil-mediated epithelial injury ( Fig. 5.8 ). There are an accompanying increase in lamina propria plasma cells, reactive epithelial changes (mucin loss, nuclear hyperchromasia), gastric foveolar metaplasia ( Fig. 5.9 ) and Brunner’s gland hyperplasia. In some instances, the Brunner’s glands may extend from the submucosa into the basal aspect of the mucosa. In severe cases, villous blunting can be present as well—a feature more in keeping with the flattened villi pattern of injury. Severe peptic injury leads to ulceration (PUD).

Ancillary Studies

Ancillary stains are not generally used for diagnosis of peptic disease but may be used to highlight a few interesting features. Gastric foveolar metaplasia can be highlighted by a combination periodic acid–Schiff (PAS) and Alcian blue (AB) stain that highlights the presence of neutral mucin characteristic of the gastric foveolar metaplasia ( Fig. 5.10 ). H. pylori may colonize areas of metaplastic gastric epithelium because these cells express the same surface receptors as the native gastric epithelium. The organisms may be highlighted by special stains such as IHC, Giemsa, or Diff-Quik ( Fig. 5.11 ). However, evaluation of the gastric mucosa, when provided, is a preferred method to detect H. pylori infection.

Differential Diagnosis

Peptic duodenal disease typically shows a neutrophilic pattern of inflammation, and other causes that produce this pattern of injury include drug or medication injury (particularly from NSAIDs), infections (other than from H. pylori ), and IBD. Occasionally, there are specific clues to distinguish between these possibilities. For example, crypt distortion and granuloma formation in conjunction with involvement of the rest of the GI tract would be highly suspicious for Crohn’s disease. Some infectious causes, including CMV and adenovirus, can be detected microscopically.

Although gastric foveolar metaplasia is a common feature of peptic duodenal disease, “lipid hang-up” is a common mimic. It results from physiologically normal vacuolization of the enterocyte cytoplasm resulting from a recent fatty meal ( Fig. 5.12 ). Compared with gastric foveolar metaplasia, which is PAS positive, these vacuoles are PAS–AB negative and are only localized to the tips of the villi. A similar finding is seen in abetalipoproteinemia; however, the vacuolization affects the tips as well as the lateral aspect of the villi. Gastric heterotopia may also be included in the differential diagnosis with peptic duodenitis resulting from the prominent gastric foveolar epithelium. However, unlike peptic duodenitis, gastric heterotopia can be recognized by normal-appearing oxyntic glands in association with the surface gastric mucinous epithelium. Finally, chronic peptic injury can also cause villous blunting with a variable degree of intraepithelial lymphocytosis. Therefore, these findings should prompt consideration of other disorders that produce the flattened villi pattern of injury (e.g., CD).

Finally, although marked histologic abnormalities are highly correlated with H. pylori infection, many duodenal biopsies show mild changes, with increased lymphoplasmacytic inflammation in the lamina propria, gastric foveolar cell metaplasia, and reactive epithelial changes (hyperchromasia and apical mucin attenuation). Neutrophils may or may not be present. If a concurrent stomach biopsy shows no suggestion of H. pylori infection, it is unclear what diagnosis is most appropriate. Various terms are commonly used, including peptic duodenitis and peptic duodenopathy . However, in the absence of H. pylori , there is little evidence that these changes are caused by stomach acid or any other “peptic” etiology. Therefore, when these cases are encountered, it is probably best to be descriptive and simply list the specific histologic abnormalities that are present.

Prognosis and Therapy

The eradication of H. pylori and the use of acid-suppressive therapies improve symptoms in most patients with peptic duodenitis. Medical therapy is mainly intended to reduce gastric acid by small meals, antacids, and histamine-2 (H ₂ ) blockers. Patients should avoid substances that promote gastric acid secretion, such as alcohol. Aspirin and other NSAIDs should also be avoided because they inhibit mucosal defenses.

Bleeding ulcers are estimated to be responsible for up to 50% of all cases of acute hemorrhage from the upper GI tract. Up to 5% of patients may have bleeding that is brisk enough to cause hematochezia. Even if initially controlled, rebleeding can affect up to one-third of patients, most commonly patients in the seventh decade of life or beyond or those with a visible vessel in the ulcer base. Refractory ulcers heal slowly or follow a relapsing and remitting course. Scarring of duodenal ulcers may lead to duodenal stricturing and obstruction.

Free perforation of a duodenal ulcer into the peritoneal cavity can be a catastrophic, life-threatening event. Most patients with this complication are older adults, and the perforations are associated with NSAID use in up to half of cases. However, in patients younger than 75 years, smoking may be a stronger risk factor for perforation than NSAIDs.

Clinical Features

Nonsteroidal antiinflammatory drugs are a widely prescribed class of medication and can cause injury throughout the GI tract in those consuming the medication. NSAID injury occurs via both local and systemic mechanisms. Locally, NSAIDs may pass directly into epithelial cells and become ionized and trapped within. Inside the cell, these compounds can disrupt oxidative phosphorylation and deplete adenosine triphosphate stores. Systemically, NSAIDs nonselectively inhibit cyclooxygenase isoenzymes and suppresses prostaglandin synthesis. Certain prostaglandins downregulated by NSAIDs are important for maintaining several aspects of mucosal protection, including supporting mucin production, mucosal blood flow, and bicarbonate secretion. Thus, although ulceration is relatively common, chronic use can cause fibrosis, stricture, perforation, and formation of mucosal diaphragms—so-called diaphragm disease .

Pathologic Features

Microscopic Findings

Typically, NSAID damage shows the intact villi with neutrophilic inflammation pattern of injury, with expansion of the lamina propria by neutrophils. Epithelial injury is usually present and may be associated with erosion or ulcer surrounded by reactive epithelial changes. NSAID injury may also take the form of prominent mucosal eosinophilia, presumably via a hypersensitivity mechanism ( Fig. 5.14 ). In resection specimens, these changes are patchy in distribution with multifocal superficial ulcers.

In chronic NSAID injury, there are villous blunting and crypt hyperplasia—features of the flattened villi pattern of injury. In other cases, crypt distortion and pyloric gland metaplasia are often present. With ongoing injury, there are thickening of the muscularis mucosae and a disorganized proliferation of smooth muscle fibers, nerves, and vessels within the submucosa. These findings indicate reparative changes that are most prominent in areas with stricture formation. It has been suggested that ischemia, inflammation, and repeated mucosal injury contribute to these pathologic changes.

Rarely, chronic injury can lead to diaphragm disease . The pathogenesis of this phenomenon is not well understood but likely involves repeated cycles of injury and repair. There may be superficial annular or longitudinal ulcers. The submucosa is widened, creating annular constriction of the lumen, and it is replaced by disorganized fascicles of smooth muscles (i.e., muscularization of the submucosa), bundles of unmyelinated nerve fibers with scattered ganglion cells, and numerous blood vessels. These changes may extend to the serosa.

Nonsteroidal antiinflammatory drugs can produce any major patterns of small intestinal injury that includes neutrophilic inflammation, eosinophilic inflammation, the flattened villi pattern, and the architectural distortion pattern. Again, NSAID injury can also produce lesions with a paucity of chronic inflammation, which is a relatively distinct feature. Overall, the diagnosis of NSAID injury always requires correlation with clinical findings and exclusion of other causes.

Differential Diagnosis

Because NSAIDs injury can produce each of the major patterns of small intestinal injury, the differential diagnosis depends on which pattern is present. The neutrophilic inflammation pattern is most common, the differential diagnosis for which includes H. pylori infection, Crohn’s disease, and reactive duodenopathy. In patients with chronic NSAID use, chronic ischemia, Crohn’s disease, and Behçet’s disease are in the differential diagnosis. Clinical findings are key in distinguishing NSAID injury from Behçet’s disease. Behçet’s syndrome is characterized by lymphocytic vasculitis that affects veins and less commonly the arteries. Crohn’s disease is distinguished from NSAID injury by the presence of deep fissuring ulcers, transmural lymphoid aggregates, and granulomas. Chronic ischemia is a close mimic and requires assessment of mesenteric vessels. When villous blunting is a prominent feature, it should prompt consideration of other malabsorptive disorders. Ultimately, as with any other medication, confirmation of causal association requires correlation with history of drug intake and resolution of symptoms after discontinuation of the drug.

Prognosis and Therapy

The prognosis is excellent for patients with NSAID ulcers if the medication is discontinued. For patients requiring resections owing to strictures or diaphragm disease , the prognosis is good, and the risks are typically those associated with the surgical procedure itself.