THE ENDOSCOPICALLY ABNORMAL ESOPHAGUS …

THE ENDOSCOPICALLY ABNORMAL ESOPHAGUS

The endoscopic abnormalities that may be seen in association with esophageal disorders are numerous, varying from subtle alterations in the esophageal mucosal surface to large ulcers or masses. This discussion will be focused on a few commonly encountered entities in which clinical, endoscopic and histologic correlation are critical for establishing a correct diagnosis. These include gastroesophageal reflux disease (GERD), eosinophilic esophagitis and Barrett's esophagus.

Gastroesophageal Reflux Disease

The most recent practice guidelines for diagnosis and management of gastroesophageal reflux disease support establishing a presumptive diagnosis of GERD without endoscopic evaluation in the setting of typical symptoms such as heartburn and regurgitation (1). Empiric medical therapy with a proton pump inhibitor is recommended; therefore, most patients that are being evaluated for GERD endoscopically are already on PPI therapy at the time of biopsy. Endoscopic examination is recommended in any patient with "alarm" symptoms, or in those at high risk for development of complications. Alarm symptoms include dysphagia, non-cardiac chest pain, or a lack of a response to PPI therapy. Although the presence of GERD-related symptoms does not provide a high level of sensitivity or specificity for predicting esophageal abnormalities on upper endoscopy, it may be reasonable to screen select individuals for Barrett's esophagus by endoscopy. Epidemiologic risk factors for BE have been well established and include age over 50, symptoms for > 5-10 years, obesity, and male sex. Recent data suggest that selectively screening patients that fall into this high risk group may be reasonable (2).

The endoscopic appearance of the esophagus varies with disease severity. Approximately one third of patients with chronic GERD symptoms are endoscopically normal (3). Areas of patchy erythema and red streaks are the first endoscopic abnormalities. Later erosions and ulcers develop; these predominate distally and taper off proximally. As the disease progresses, the ulcers become confluent, even circumferential. Strictures or BE characterize severe chronic disease. Several endoscopic classifications have been developed to evaluate the esophageal mucosa. The two most common are the Savary-Miller MUSE system and "Los Angeles" classifications (4,5). The use of these grading systems among gastroenterologists varies, likely depending on the practice setting. Among esophageal biopsies seen in our practice in 2010, 27.9% included an LA grade, and 12.2% a Savary-Miller grade (6).

Savary-Miller Classification

Grade I II

III

IV

Features One or more supravestibular reddish spots with or without exudates Erosive and exudative lesions in the distal esophagus that may be confluent, but not circumferential Circumferential erosions in the distal esophagus covered by hemorrhagic and pseudomembranous exudates Presence of chronic complications such as deep ulcers, stenosis, or scarring with Barrett's metaplasia

Los Angeles Classification

Grade A B

C

D

Features One or more mucosal breaks 5 mm in length At least one mucosal break > 5 mm long, but not continuous between the tops of adjacent mucosal folds At least one mucosal break which is continuous between adjacent mucosal folds, but not circumferential (< 75% of periphery) Mucosal breaks that involve at least three-quarters of the luminal circumference

Note: Ulcers, strictures, Barrett's metaplasia, and other findings are reported as an adjunct to each grade.

Reflux Esophagitis

GERD affects patients of all ages, even children and small infants, but is most common among patients over age 65 (7). GERD affects from 3% to 4% of the population, and its incidence has increased in the past few decades. The annual cost of managing the disease in the United States is estimated at 9 billions dollars (8). GERD is equally present among men and women, but there is a male predominance of esophagitis and Barrett esophagus. GERD affects whites more frequently than members of other races.

Conditions predisposing to GERD include smoking, decreased physical activity, increased intra-abdominal or intragastric pressure, including pregnancy, ascites, body mass index and obesity; and delayed gastric emptying. Postmenopausal estrogen therapy has been associated with GERD symptoms (9, 10). Motility disorders including diabetes, alcoholic neuropathies, and scleroderma also predispose to GERD. Patients with hiatal hernias and strictures are especially prone to develop GERD. It also follows surgical procedures.

Clinical Features Transient mild reflux affects most individuals including children and adults. The degree

of reflux must be severe for individuals to become symptomatic. Adults present with diverse symptoms including heartburn, regurgitation, bitter-tasting fluid in the mouth, dysphagia, odynophagia, nausea, vomiting, hiccups, anginalike chest, and hoarseness. The regurgitation can cause a spectrum of conditions, including asthma, chronic cough, chronic laryngitis or pharyngitis, subglottic stenosis, and dental disease. Rare patients present with bleeding from esophageal ulcers. Complications peak between ages 50 and 70 years (11).

As discussed above, the majority of patients with GERD symptoms are treated with proton pump inhibitor (PPI) therapy, often prior to being seen by a gastroenterologist (12). A recent survey found, however, that as many as 40% of patients receiving PPI treatment have residual symptoms (13). As a result, many of the patients undergoing endoscopy and biopsy for GERD symptoms are those that are refractory to PPI therapy. Such patients may have additional disorders superimposed on GERD. The differential diagnosis of esophageal disorders that may be associated with refractory symptoms in patients treated with PPIs is summarized below.

Disorders Associated with Residual Symptoms in Patients Treated with PPIs

Esophageal Disorders Reflux Related Incorrect medication dose timing Medication non-compliance Residual pathologic acid secretion Rapid PPI metabolism Hypersecretory state Anatomic abnormalities Defective LES function Hypersensitivity of esophagus to small amounts of refluxed material Non-reflux related Achalasia Esophageal spasm Scleroderma Eosinophilic esophagitis Pill esophagitis Infectious esophagitis

Non-Esophageal Disorders Gallbladder disease Cardiovascular disease Musculoskeletal disorders Malignancy (GI and non-GI)

Histologic Features Biopsies are performed to confirm the diagnosis of GERD; to document complications,

including esophagitis, BE, or tumor development; and to rule out the presence of coexisting infections. Since esophagitis tends to be a patchy process, it is easy to miss diagnostic changes on a single biopsy. The current wisdom is that biopsies should be taken in the area just distal to the Z line to detect carditis (see below), just proximal to the Z line to detect esophagitis, and 3 cm proximal to the Z line to detect the hyperplastic changes that are more predictive of the presence of GERD than more distally derived biopsies.

Various histologic features should be assessed when examining the biopsy for GERD (see below). No single feature represents an absolute criterion for the presence of GERD, but each is helpful in establishing the diagnosis. In the absence of a known drug history or the presence of specific microorganisms, biopsies, particularly distal biopsies showing esophagitis, are most likely to be due to GERD.

Epithelial Hyperplasia The normal basal cell layer is only one to four cells high; it should not constitute more

than 15% of the epithelial thickness. In the setting of GERD, the basal zone increases from 10% to more than 50%; papillary height can increase to more than 50% to 75% of the total epithelial thickness (14). This change affects patients with an endoscopically normal mucosa as well as those with endoscopic evidence of esophagitis. Regenerative changes are characterized by

nuclear enlargement, hyperchromasia, and mitoses that remain limited to the basal layer. Prominent nucleoli may be present.

Although recognition of the basal layer of the squamous epithelium is not difficult, determination of its uppermost limit often is. One definition that may be helpful (15) is that the upper limit of the basal zone is that point where the majority of epithelial cell nuclei are separated by a distance less than the diameter of one nucleus. In addition, accurate assessment of the thickness of the basal layer requires evaluating well-oriented specimens. The basal height should not be assessed near vascular papillae. It may be helpful to divide the epithelial thickness into thirds. When the lower third is divided in half, the basal cells should be confined to its lower half. In less optimally oriented specimens, basal zone thickness can be evaluated if one sees at least three to four papillae arranged in parallel to one another and not cut tangentially. In tangentially cut sections, a helpful feature is an increase in the number of papillae, which can be evaluated in an en face section. In this setting one may see overlapping capillaries. Since the biopsies may be small or have minimal or no lamina propria or they may be inappropriately oriented and therefore difficult to evaluate for basal hyperplasia and papillary elongation, we recommend that the biopsies be examined at three levels to increase their diagnostic accuracy.

The sensitivity of hyperplasia as a diagnostic feature of GERD is only 60% to 70% (16). Basal cell hyperplasia is a reversible change that disappears with treatment. Hyperplasia also complicates other forms of esophagitis so that it is not specific for GERD.

Papillary Height Papillary elongation is most often defined as papillae extending more than two-thirds of

the distance to the epithelial surface. The degree of papillary elongation correlates with severity of reflux (17), but is not a specific feature of GERD. Evaluation of papillary height, like basal layer thickness, requires a well-oriented biopsy in which the entire thickness of the epithelium is visible.

Dilated Intercellular Spaces Dilated intercellular spaces occur in patients with both erosive and non-erosive GERD

(18-20). In light microscopic studies, dilated intercellular spaces are defined as an increase in the distance between squamous epithelial cells. This distance varies from greater than 0.47 to 2.4 um depending on the study (18-22). Dilated intercellular spaces are seen predominantly in the basal layer, and are thought to occur as a result of stretching and detachment of desmosomes (20). The prevalence of dilated intercellular spaces in biopsies from patients with GERD ranges from 67-94% depending on whether or not clinical symptoms, endoscopically identifiable lesions or pH monitoring abnormalities are present (23).

Inflammation Intraepithelial lymphocytes. Small numbers of lymphocytes populate both the normal mucosa and the lamina propria so that their presence does not aid in making a diagnosis of esophagitis. However, they are very conspicuous in patients with GERD (24). Biopsies with esophagitis average greater than six lymphocytes per hpf (25). They are part of the inflammatory response in GERD but are not an independent marker of reflux esophagitis.

Neutrophils. The presence of isolated neutrophils, either in the squamous epithelium or in the lamina propria, serves as evidence for acute esophagitis of many etiologies, and are most

commonly observed in erosive GERD. Neutrophils are present in the epithelium of from 1040% of patients with reflux esophagitis, making them a relatively insensitive marker (26-28). They tend not to appear until the inflammation becomes severe and the epithelium ulcerated. They generally decrease in number the further one goes away from the erosion or ulcer. When numerous neutrophils are identified in an esophageal biopsy, the possibility of other ulcerating conditions including infection and pill-mediated injury should be considered.

Eosinophils. The normal number of eosinophils present in the esophagus is still somewhat controversial. However, many feel that eosinophils are normally absent (26, 29). Some, however, have found a modest number (usually less than 5 per high power field) may be seen (27). It is likely that these differences arise from differing definitions of normal controls versus GERD. Influx of eosinophils into the epithelium occurs early in the course of reflux esophagitis and may be seen in the absence of basal cell hyperplasia. Other causes for mucosal eosinophilia include the entities listed in the Table below.

Eosinophil-Associated Esophageal Disorders Primary eosinophilic disorders Eosinophilic esophagitis Secondary eosinophilic disorders Eosinophilic gastroenteritis Hypereosinophilic syndrome Secondary noneosinophilic disorders

Infection Pill esophagitis Gastroesophageal reflux disease Tumors Vasculitis Connective tissue disorders

One can easily appreciate eosinophils in small endoscopic biopsies, even when the biopsies are not well oriented. They affect up to 60% of adults with severe disease but the intraepithelial eosinophils may be focal in nature, necessitating a search for them on serial sections; their presence does not correlate with disease severity (29, 30). Eosinophils are not a sensitive marker for GERD, since they are only found in 40% to 50% of individuals with GERD (23, 28, 30). Significant esophageal eosinophilia (>15 intraepithelial eosinophils per hpf) may sometimes be seen in patients with GERD, but should prompt the pathologist to consider the diagnosis of eosinophilic esophagitis as discussed later.

Erosions and Ulcers in Gastroesophageal Reflux Disease The mucosal changes of reflux esophagitis range from the changes already described to

acute esophagitis, erosions and superficial ulcers. The epithelium close to erosions or ulcers often contains neutrophils, eosinophils, and many lymphocytes. The erosions or ulcers often contain granulation tissue, an inflammatory exudate, and fibrinoid necrosis in the ulcer base. Lymphoplasmacytic infiltrates, often forming lymphoid aggregates, tend to cluster around erosions and ulcers. Epithelium at the ulcer margin is usually attenuated. Marked basal cell

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download