37: Overview: Anatomy and Pathophysiology of Esophageal Reflux Disease

Gastroesophageal reflux disease (GERD) is one of the most common disorders seen in the general population. Approximately 10% of all people experience heartburn daily, 15% in any given week, and about half the population in any given year.¹ The spectrum of disease ranges from occasional postprandial substernal discomfort to the development of peptic stricture or even carcinoma in the setting of Barrett esophagus. It is important for physicians to identify and select appropriate treatments for patients at risk of developing complications. This chapter reviews the etiology, diagnosis, complications, and treatment of GERD.

The esophagus is lined with a stratified nonkeratinizing squamous epithelium consisting of three layers: superficial, intermediate, and basal. The basal cell layer comprises 15% of the total epithelial thickness and is the only layer that normally contains mitotic figures. The lamina propria, which lies deep to the squamous epithelium, contains glandular structures similar to those found in the gastric cardia. The epithelium is covered by a protective layer of mucin and surface bicarbonates that are produced by salivary glands located in the proximal esophagus and in the region adjacent to the gastroesophageal junction (GEJ). The mucus produced in these glands reaches the intervening sections of the esophagus through peristalsis. Unlike the stomach, the mucus layer that lines the esophagus is rudimentary and provides little protection against prolonged acid exposure.

The sensations of heartburn or discomfort from reflux are transmitted from the esophagus by the spinal splanchnic afferent nerves. These sensations may be modulated by vagal afferent nerves. A series of high-pressure zones, or sphincters, and zones of complex neural interaction propel the forward movement of food and liquid into the stomach and retard the return or reflux of gastric contents back into the esophagus. The intrinsic lower esophageal sphincter (LES), along with its extrinsic components, is the mechanism chiefly responsible for preventing gastric reflux back into the esophagus (Fig. 37-1). The LES is identified on intraluminal manometry as a 2- to 4-cm-long zone of high pressure at the GEJ. A ringed circular muscle is also present at this junction. The vessel density and amount of connective tissue are greater in LES muscle than in the remainder of the esophagus. This region is also rich in mitochondria and smooth endoplasmic reticulum.

The LES exerts a resting basal pressure in normal individuals. The tonic muscular contraction that is the hallmark of the LES contributes to the resting tone, distinguishing this region from the esophageal body. Inhibition of LES contraction is mediated largely through nitric oxide, whereas excitation is mediated through acetylcholine. Even with ablation of all neural input, the LES maintains its intrinsic basal smooth muscle tone. This muscular tone is distributed unevenly throughout the sphincter. The pressure is greatest in the lowest 2 cm of the LES and on the left side, where the left crus and sling fibers of the stomach exert force on the LES. The basal pressure also varies according to the respiratory cycle. During inspiration, the pressure in the upper half of the LES decreases, whereas the pressure in the lower half increases. This finding may be related to the respiratory contractions of the diaphragmatic crural fibers or relative intrathoracic and intra-abdominal pressures. A small shift in the position of the LES can have dramatic effects on these pressure gradients.

Relaxation of the LES occurs naturally during deglutition, about 2 seconds after the initiation of swallowing. At this time, the pressure in the LES decreases to a level approximating the intragastric pressure. This period of decreased pressure may last as long as 8 to 10 seconds. Distention of the esophagus produces a reflex relaxation of the LES that is distinct from the relaxation elicited during initiation of swallowing, such as that produced from tactile pharyngeal stimulation. A number of hormones and neurotransmitters also can produce LES relaxation. These include gastric inhibitory peptide, glucagon, cholecystokinin, nitric oxide, progesterone, vasoactive intestinal peptide, and prostaglandin E. Conversely, hormones that stimulate contraction include gastrin, bombesin, motilin, serotonin, and somatostatin (Table 37-1).

The two components of the extrinsic portion of the LES are the transmitted intra-abdominal pressure and the crural fibers. As long as the LES is situated below the diaphragm (the normal anatomic position), increases in intra-abdominal pressure are exerted equally on the LES and stomach. The crural fibers serve as a redundant means of maintaining LES pressure. A normal LES should be able to resist most increases in intragastric pressure. For this reason, many individuals with a hiatal hernia have no reflux. The crural fibers exert increased pressure on the LES when the tendency toward reflux is greatest (i.e., during inspiration and with increases in intra-abdominal pressure). During inspiration, the intrathoracic pressure decreases, and the intra-abdominal pressure increases. Crural fibers contract along with the diaphragm, causing an increase in LES pressure. Pressure from the crural fibers also increases during independent increases in intra-abdominal pressure.

The anatomy of the proximal stomach and its angle of attachment to the LES (angle of His) counteract the tendency for reflux to occur. The esophagus joins the stomach not at its apex but farther down along the lesser curvature. It is believed that the lateral fold of the gastric mucosa acts as a flap valve, helping to close the entrance to the LES during episodes of raised intra-abdominal and intragastric pressure. The fundus and cardia of the stomach act as a reservoir to minimize the increase in intragastric pressure associated with meals. The anatomic significance of the angle of His and its relationship to the fundic reservoir is apparent in studies showing a larger degree of reflux when individuals lie on the right side as opposed to the left.

Normal esophageal motility and clearance are important to minimizing the effects of gastric refluxate. Reflux into the esophagus is normally followed by waves of secondary peristalsis that move the contents back into the stomach. This secondary wave of peristalsis also brings bicarbonate-rich saliva into the esophagus to neutralize the remaining gastric acid. Although impaired esophageal motility occasionally may result from severe reflux esophagitis, it also may contribute to it, such as occurs in patients with scleroderma.

Factors that affect the degree of gastroesophageal reflux include esophageal motility, the amount and composition of the saliva and other protective esophageal factors, the intrinsic LES, alterations in the transthoracic—transabdominal pressure gradient, the extrinsic LES (including the crural muscles), the integrity of the angle of His, gastric emptying, and the nature of the refluxate.

The LES is the most important component of the antireflux apparatus. Decreased resting LES tone and transiently decreased LES tone both may contribute to reflux. Transient relaxation is induced by gastric distention and pharyngeal stimulation. The former condition allows eructation with relief of gastric pressure, whereas the latter is important in swallowing. Transient relaxation is inhibited in the supine position and during sleep. The consumption of inappropriately large meals not only provides a pressure gradient favoring reflux but also induces transient relaxation of LES tone. Inappropriately low transient LES pressure is believed to be the most common cause of reflux and may account for 60% to 70% of episodes of heartburn. Pathologic transient decreases in LES tone occur when the LES pressure decreases by more than 5 mm Hg in the absence of a swallow. Reflux in patients with mild to moderate disease and no endoscopic evidence of esophagitis most often occurs as a consequence of transient decreases in LES pressure, whereas patients with severe reflux typically also have deficits in resting LES tone. Normal LES tone is 10 to 30 mm Hg above gastric pressure. Less than 2.5% of the population has a resting LES pressure of less than 6 mm Hg. Most patients with mild reflux have normal LES tone. Patients with erosive esophagitis usually have a low resting LES tone, and the degree of esophagitis typically correlates with the lack of LES tone. There is evidence in animals that repeated exposure of the LES to acid via transient relaxation may damage the sphincter, leading to a lower resting LES tone.

A hiatal hernia is defined as the separation and cranial displacement of the GEJ from the diaphragmatic hiatus. In normal individuals, the GEJ is anchored to the crus by the phrenoesophageal ligament, which is composed of a reflection of peritoneum fused with fibrous tissue (Fig. 37-2). When there is a hiatal hernia, the crura are often stretched and thin, and the phrenoesophageal ligament is lax. The redundant tissue is termed the hernia sac. In normal individuals, increased intra-abdominal pressure does not lead to reflux. In patients with hiatal hernia, however, the contribution of the diaphragmatic crura and effects of increased intra-abdominal pressure on the intra-abdominal esophagus are lost, leading to increased susceptibility to strain-induced reflux. Whether prolonged reflux leads to fibrosis and shortening of the esophagus with development of a hiatal hernia, or if the primary laxity in the crura and phrenoesophageal membrane leads to the development of reflux is unknown. As the hiatal hernia enlarges, the resting LES tone diminishes.

Deficits in esophageal peristalsis are seen commonly in individuals with moderate-to-severe GERD. An esophageal contraction of 30 mm Hg or greater is usually adequate to clear an acid bolus in the supine patient. Poor contractility, with a decreased number or amplitude of contractions, leads to acid retention in the esophagus. Repeated exposure of the esophagus to acid is believed to injure the intrinsic muscles of the esophagus and may lead to further deficits in esophageal motility and clearance. Twenty-five percent of patients with mild and 50% of patients with severe esophagitis have evidence of severe dysfunction.² Individuals with decreased salivary volume (e.g., patients with Sjögren's syndrome and smokers) have increased acid exposure in the esophagus owing to decreased clearance.

Most individuals with symptomatic GERD have normal gastric emptying. It is evident, however, that a delay in gastric emptying will increase gastric volume and pressure and exacerbate GERD in patients predisposed to the condition. This can be seen after any surgery on the GEJ if the vagus nerves are injured, and it must be considered in any patient who has reflux despite fundoplication. Patients with gastroduodenoesophageal reflux and GERD may be more susceptible to erosive esophagitis than those with GERD alone. Pepsin and bile salts from the duodenum have proteolytic and detergent effects that exacerbate the damaging effects of gastric acid.

Most individuals experience the typical symptoms of heartburn, that is, regurgitation or midline chest discomfort, at some point during their lifetime (Table 37-2). The burning pain or pressure associated with heartburn is caused by reflux of acid into the esophagus. Frequently, symptoms are more prevalent after meals or in the supine position. Certain foods associated with decreased LES tone, such as high-fat meals, caffeine, chocolate, peppermint, and alcohol, may exacerbate these symptoms. In general, individuals who experience heartburn only after certain foods tend to have higher resting LES pressures than those who experience heartburn with almost all types of foods. Patients with daily symptoms are more likely to have resting LES pressures of less than 10 mm Hg.¹ Relief of symptoms immediately after ingestion of water or antacids supports the diagnosis of GERD. Substernal pressure may be related to esophageal spasm caused by esophagitis or exposure to acid. If the pain is related to exertion and stops on cessation of activity, cardiac ischemia should be considered, although a significant number of these patients also may have reflux. The severity and frequency of symptoms do not necessarily reflect the severity of the underlying esophagitis. One-quarter of all patients with severe reflux and Barrett esophagus have no symptoms at all.³ Likewise, a significant number of patients with frequent and severe complaints are not found to have esophagitis by endoscopy.

Respiratory symptoms such as hoarseness, cough, asthma, and recurrent pneumonia are referred to as atypical symptoms (Table 37-2). Although a significant number of patients with chronic asthma or cough also may have GERD, these symptoms are less specific than typical symptoms, and surgical fundoplication is, in general, less effective for symptom control. A randomized controlled trial by Larrain et al.⁴ treated 94 intrinsic asthma patients with reflux documented by either barium swallow or pH probe analysis. The patients were randomized to placebo, cimetidine, or surgery. The surgery and cimetidine groups had roughly 75% improvement in wheezing and respiratory function versus only 34% in the control group. Long-term follow-up showed that 50% of the surgical group was asthma-free compared with only 5% of the control group. Individuals with respiratory symptoms who may be more likely to respond to GERD treatment are those with nonallergic asthma, nocturnal symptoms, abnormal proximal esophageal acid exposure, and those who had reflux symptoms before developing asthma.

Patients with odynophagia (i.e., pain on swallowing) or dysphagia (i.e., difficulty swallowing) are more likely to have severe esophagitis, abnormal motility, or structural abnormalities of the esophagus. Odynophagia may be associated with severe esophagitis or esophageal spasm. Dysphagia may be related to motor diseases (typically, dysphagia to both solids and liquids) or to a narrowing caused by a peptic stricture, ring, or tumor (dysphagia more with solids than with liquids).

Diagnostic techniques focus on three aspects of GERD: objective documentation and quantification of reflux (usually by pH probe), definition of the pathophysiology (usually by manometry), and assessment of mucosal complications of reflux such as stricture, metaplasia, ulcers, or neoplasia (by barium swallow or endoscopy) (Table 37-3).

Barium Swallow

Barium swallow is a noninvasive means of obtaining valuable information about the esophagus. The fundamental techniques employed with barium swallow include (1) full-column examination, in which the esophagus is filled with barium and inspected for hiatal hernia, contour deformity, rings, and strictures; (2) analysis of mucosal patterns, in which the empty esophagus is coated with barium and inspected for irregular, thickened folds or varices; (3) fluoroscopic observation and motion recordings employing real-time observation of contractions and reflux; and (4) double-contrast examination, in which the esophagus is coated with barium and then distended with air to reveal fine mucosal abnormalities. The barium swallow is overall a relatively insensitive tool for diagnosing GERD. Mild esophagitis may not be visible on barium swallow, although air-contrast techniques increase the sensitivity. Moreover, reflux of barium from the stomach into the esophagus can be seen in a significant number (20%–25%) of patients with no reflux symptoms and infrequently in patients with severe symptoms. Barium swallow is most effective for detecting the following conditions: moderate-to-severe esophagitis, presence of a hiatal hernia, and any associated stricture or mass lesion. The presence of a hiatal hernia containing gastric mucosal folds and the presence of a stricture above these folds are fairly specific for peptic stricture in the setting of Barrett esophagus.

Radionuclide Scintigraphy

Radionuclide scintigraphy was proposed initially as a means of documenting reflux and assessing reflux severity. A radioisotope (technetium-99 m sulfur colloid) is swallowed and followed for reflux. Manual compression of the abdomen or straining maneuvers can be used to elicit reflux. The sensitivity and specificity of this test have been questioned, and it is not employed commonly in the workup of GERD. Radionuclide scintigraphy is sensitive and specific for detecting delayed gastric emptying and can be employed if this disorder is suspected. It should be part of the routine workup of recurrent or persistent reflux after fundoplication or for patients in whom gastric emptying dysfunction is suspected based on symptoms, associated disorders (diabetes mellitus), or in patients taking medications that may slow gastric emptying.

Endoscopy

Endoscopy is the first and preferred diagnostic procedure when reflux fails medical treatment, if symptoms recur after a course of medical treatment, or when patients have significant odynophagia or dysphagia. Endoscopy permits direct visualization and biopsy. Metaplasia, ulcers, strictures, and erosions seen during endoscopy can be graded to standardize reflux severity. Patients with reflux and mild esophagitis may have an endoscopically normal-appearing esophagus. Random mucosal biopsy should be taken 5 cm above the LES because even normal individuals may exhibit some metaplastic change in the most distal esophagus. Microscopic changes that are apparent with esophagitis include a basal epithelial layer exceeding 15% of the total epithelial thickness and the presence of lamina propria papillae that extend more than two-thirds deep into the epithelial layer. Neutrophils and eosinophils in the lamina propria are considered to be better indicators of reflux esophagitis, as are erosion, ulceration, stricture, and a columnar-lined mucosa. The presence of retained food in the stomach during endoscopy should also alert the physician to the possibility of gastric emptying dysfunction.

pH Probe Analysis

pH probe analysis is the most sensitive and specific means for quantifying acid reflux in the esophagus. In patients with typical symptoms of reflux and esophagitis documented by endoscopy, pH probe analysis adds little to the overall management. This technique is applicable to (1) patients with typical reflux symptoms in the absence of esophagitis, (2) patients with atypical symptoms with or without esophagitis, (3) before consideration of antireflux surgery, or (4) for the evaluation of unsuccessful antireflux treatment. Proton pump inhibitors (PPIs) must be stopped 7 days before study, and H₂-receptor antagonists should be stopped 48 hours before study.

The electrode, made of either glass or antimony, is positioned 5 cm above the LES (as defined by manometry). Dual-probe catheters are available that simultaneously measure acid in the upper esophagus and lower pharynx. The catheter is left in place for 24 hours, and the patient is asked to record the duration of meals, time spent in the supine position, and the occurrence of any reflux symptoms. Studies have shown that the variable percentage of time with pH <4 is the most reproducible measurement (85% reproducibility). This measurement is termed the reflux time or acid exposure time and correlates better with esophagitis grade than the variable number of episodes with pH <4. Normal patients typically have acid exposure times of less than 7%. Patients with values in the 7% to 12% range may have some degree of mild inflammation, whereas acid exposure times of greater than 12% are usually found with more severe forms of esophagitis. For example, patients with columnar-lined esophagitis have an average exposure time of 26%.

DeMeester and Johnson developed a composite score based on six variables of acid exposure: total time with pH <4, upright time with pH <4, supine time with pH <4, total number of episodes with pH <4, number of episodes >5 minutes with pH <4, and duration of longest episode of pH <4. It is not clear that this composite score characterizes the severity of reflux any better than the simple variable number of episodes of pH <4.⁵

Some patients are bothered by placement of a 24-hour transnasal pH probe catheter that is worn externally. A Bravo capsule, consisting of a wireless transmitter and pH electrodes in a 25 mm by 6 mm capsule, can monitor acid reflux in the esophagus and transmit information wirelessly. This capsule sends data to an external receiver and recorder, which is worn by the patient. The capsule is typically placed at the time of endoscopy and is secured on the wall of the esophagus via small pins into the mucosa. Twenty-four-hour collection is typically performed, and the capsule usually sloughs off and is passed within 5 to 7 days.

Approximately 10% to 20% of patients who have endoscopy and biopsy-proved esophagitis demonstrate negative 24-hour pH studies. A percentage of these patients may have significant reflux of duodenal contents into the esophagus. In theory, patients with bilious reflux should have an esophageal pH greater than 7, but duodenogastroesophageal reflux typically occurs with gastroesophageal reflux, and the esophageal pH is rarely, if ever, greater than 7. Reflux of bile and duodenal enzymes has been proposed as factors contributing to the development of esophagitis and Barrett esophagus. Bilirubin-sensitive probes are available for measuring bile reflux. These probes rely on detecting significant light absorption near 450 nm, the characteristic absorption peak of bilirubin. A fiberoptic probe is used to calculate the difference in light absorption at 470 nm (bilirubin peak) and 565 nm (reference) at the tip of the probe. There is a reasonably good correlation between these readings and direct aspiration and measurement of bilirubin levels. Manometry, often used to evaluate esophageal function before fundoplication, is discussed in Chapter 33.

Impedance

Impedance is the ratio of voltage to current and is inversely proportional to conductivity. Using a catheter placed in the esophagus one can make sum measurements of the impedance of the esophagus and its contents. Impedance electrodes are typically placed at 2-cm intervals along a transnasal catheter that traverses the length of the esophagus. By measuring bolus transit, conclusions regarding functional peristalsis can be made, although it is not clear yet whether impedance transit measurements offer anything additional to high-resolution manometry. Impedance measurements do, however, add to our knowledge of the chemical composition of refluxate in the esophagus. Multichannel intraluminal catheters which combine impedance and pH measurements are more sensitive to the detection of combined acidic and nonacidic reflux events. Typically, the catheter is placed for 24 hours, and patients are asked to keep a log of symptoms. Using impedance, for example, it is now known that many patients on PPIs do not have less reflux, but only have fewer episodes of low pH reflux. Patients with continued symptoms on PPI may be more appropriately offered antireflux surgery if it can be documented that continued nonacidic reflux events persist. In theory, a single multichannel impedance pH catheter could give complete information regarding esophageal peristalsis, reflux episodes, and reflux composition.

Lifestyle Modifications

The first phase of therapy for symptomatic GERD involves lifestyle modifications aimed at factors that have been shown to increase symptoms and acid exposure in the esophagus (Table 37-3). Elevating the head by 6 inches in the supine position has been shown to decrease esophageal clearance time and esophageal acid exposure in reflux patients. Certain foods have been shown to increase acid exposure in the esophagus by causing increased acid production, decreased LES tone, decreased gastric emptying, decreased esophageal clearance, or a variety of these factors. Chocolate, peppermint, alcohol, high-fat meals, and smoking all have been shown to decrease LES pressure. Fatty foods also significantly delay gastric emptying. The effect of caffeine on LES pressure is less clear, but it is believed to increase acid production and acid exposure in the esophagus. There is a correlation between obesity, low LES pressure, and GERD. Alterations in diet and lifestyle with antacid therapy may provide significant relief to individuals with occasional heartburn or mild GERD, but studies in patients with reflux esophagitis show a good or excellent response in only approximately 20% (vs. a 75% response rate in surgical patients).⁶

Motility Agents

Motility agents should, in theory, increase acid clearance, and if gastric emptying and LES pressure are increased, they should decrease acid exposure of the esophagus. Metoclopramide increases LES tone, improves esophageal contraction, and increases gastric emptying. When used for prolonged periods, however, there is a significant incidence of side effects such as restlessness and agitation. Even in combination with H₂-receptor antagonists, it does not seem to be as effective as the better-tolerated PPIs. Cisapride, another promotility agent, is similar in efficacy to metoclopramide but is no longer available for use in the US owing to its proarrhythmic effects.

H₂-Receptor Antagonists

Although antacids are effective in raising the esophageal pH above 4 (the threshold for esophageal healing), they are short-acting and give only transient relief. H₂-receptor antagonists such as cimetidine (300 mg qid), ranitidine (150 mg bid), famotidine (20 mg bid), and nizatidine (150 mg bid) were used initially for 6- to 12-week durations to treat reflux that did not respond to conservative measures. These dosages are effective in 50% to 70% of patients, with higher doses needed to reliably heal moderate-to-severe cases of esophagitis.

Proton Pump Inhibitors

PPIs provide more rapid relief of symptoms and promote healing of erosive esophagitis far more effectively. They are used today in preference to H₂-receptor antagonists. PPIs omeprazole (40 mg per day), lansoprazole (30 mg per day), pantoprazole (40 mg per day), and rabeprazole (20 mg per day) are approximately 90% effective in healing erosive esophagitis after 8 weeks. Patients who do not heal with once-daily dosing should be put on a twice-daily regimen for better 24-hour control of gastric acid production. Most patients with erosive esophagitis treated with PPIs heal, but they have recurrent symptoms within 6 to 9 months. An additional 10% have recurrent esophagitis without symptoms. PPIs have few known side effects. However, the complete suppression of acid secretions is known to stimulate goblet cells, resulting in hypergastrinemia, although the development of carcinoid tumors or gastrinomas has not been seen in humans. The appearance of significant numbers of metaplastic cells in the esophageal mucosa with PPI treatment is also theoretically of concern in relation to Barrett esophagus. Although PPIs stop acid production, limit acid exposure in the esophagus, and decrease the quantity of refluxate, they do not stop the act of reflux. The possibility of bile salts and digestive enzymes contributing to the progression of Barrett esophagus and the development of dysplasia, cancer, or both also has been raised. Despite these concerns, there is no clear-cut relationship between the use of PPIs and an increased risk of Barrett metaplasia.

Role of Surgery

For the uncomplicated patient with GERD, surgical therapy cannot be recommended over PPI therapy. There appears to be no difference in response of esophagitis, symptoms, patient satisfaction, or death from cancer between the two modes of treatment. After fundoplication, the need for PPIs decreases, but a significant percentage of patients continue to take these medications (in one study 37%).⁷ In a patient with complicated GERD (Barrett and/or stricture), surgery may very well be more effective than PPI therapy (see Chapter 41).

Laparoscopic fundoplication is indicated for patients who do not respond to or cannot tolerate high-dose PPI therapy, as well as for those who require long-term therapy but do not wish to take PPIs long term (see Chapters 38-40). A thorough evaluation before fundoplication includes a 24-hour pH probe analysis, esophageal manometry, and an esophagogastroduodenoscopy (EGD). Patients must meet two criteria before being offered the option of fundoplication. They must have symptoms of acid reflux into the esophagus (i.e., esophagitis or sequelae of esophagitis, pain, or atypical symptoms) and documentation of abnormal acid exposure. Alternatively, patients should manifest complications clearly caused by GERD, such as peptic strictures, Barrett esophagitis, or ulcers. Patients are stratified as those with typical or atypical symptoms and those without esophagitis, with esophagitis, or with complicated esophagitis. Patients with typical symptoms and evidence of esophagitis are good candidates for fundoplication. (Some surgeons would argue that pH and manometry in these patients adds little to overall management.⁸) Those with typical symptoms but without esophagitis also may be candidates for surgical therapy if PPIs are not effective or not tolerated, provided that the acid reflux has been documented by pH/manometry testing. Patients with atypical symptoms, such as frequent pneumonia, chronic cough, adult-onset asthma, and esophagitis, also should undergo pH/manometry testing. Overall, laparoscopic Nissen fundoplication is more successful in relieving typical symptoms (93% successful) than atypical symptoms⁹ (56% successful) (see Chapter 39). Patients with atypical symptoms without evidence of esophagitis must be studied carefully before being considered for fundoplication. The absence of mucosal damage does not mean that reflux is absent. These patients still may be suffering from reflux and aspiration but have the ability to quickly clear residual fluid from the esophagus. PPIs stop acid production, decrease the amount of gastric secretions, and decrease the amount of gastric contents refluxed, but the upper airway still may be exposed to damage from bile acids and enzymes. Nonetheless, patients with atypical symptoms without esophagitis who respond to PPIs have a better response to fundoplication than those who do not.¹⁰ This is attributable in large part to the diagnostic capability of PPI therapy to select atypical patients with symptoms secondary to reflux.

Patients who have recurrent symptoms after surgical fundoplication should undergo EGD to document the presence of esophagitis and recurrent hernia. Repeat pH and manometry testing provides objective evidence of recurrent reflux or motor disorders. Gastric emptying also must be assessed. If vagal nerves were injured at the time of the original fundoplication, delayed gastric emptying may be contributing to reflux despite the presence of an adequate fundoplication. Treatment for this condition would be a promotility agent, such as metoclopramide, followed by either pyloromyotomy or pyloroplasty if symptoms continue.

A hiatal hernia is a type of paraesophageal hernia defined as a loosening of the phrenoesophageal membrane resulting in displacement of the LES away from its attachments to the diaphragm. There are four types of paraesophageal hernias (Fig. 37-3). All four types are associated with symptomatic and asymptomatic GERD. Type I, also termed a sliding hiatal hernia, is a simple herniation of the LES into the chest, often associated with esophageal shortening. The hernia itself is not harmful and requires no surgical correction. A large number of people have an asymptomatic hiatal hernia. Any surgical treatment offered for a type I hernia is dictated by the need to treat the associated GERD. Adequate surgical correction of a significant type I hernia often requires esophageal lengthening (see Chapter 42).

The LES in type II hiatal hernia, also termed paraesophageal, remains fixed in position at the level of the hiatus, but a portion of the stomach herniates along with the LES into the chest through a small defect in the phrenoesophageal membrane. Type II hiatal hernias are quite rare, but because they involve only a partial defect of the phrenoesophageal membrane, the incidence of incarceration and strangulation is high. Surgical repair is indicated to prevent strangulation or further herniation of the involved portion of the stomach.

Type III hernias have both sliding and paraesophageal components. As a sliding hernia progresses, the mobile greater curvature of the stomach is drawn into the chest and it begins to rotate until it eventually resides in the right chest. This condition is termed organoaxial rotation. Several complications can occur from large type III hernias. Vascular insufficiency may arise from arterial obstruction or venous congestion. There also may be ulceration, bleeding, or frank necrosis and perforation. Chronic iron-deficiency anemia is common in patients with type III hernias and is thought to be due to the slow, chronic loss of blood from the stomach. Partial or complete obstruction may occur either from organoaxial rotation or from compression of the conduit which often occur as a result of a meal.

Type IV hernias are simply type III hernias that involve displacement of additional structures and organs into the chest, such as the omentum, spleen, or colon.

Peptic Stricture

Approximately 10% of patients seeking treatment for GERD suffer from peptic stricture. Like other patients suffering from complicated GERD, these patients tend to have a lower ­resting LES pressure. One study noted that 64% of patients with stricture suffered from motility disorders versus 32% of patients without stricture, suggesting that impaired clearance of acid may be a causative factor in GERD.¹¹

Strictures usually form in the lower esophagus as a result of severe esophagitis. When esophagitis extends to the full thickness of the esophageal wall, healing may result in stricture. Schatzki rings are short, web-like stenoses associated with GERD. They may be the result of mucosal and submucosal inflammation and subsequent fibrosis. The ring does not involve the muscular portion of the esophagus. Peptic strictures typically present at the squamocolumnar junction. (In patients with Barrett esophagitis, the squamocolumnar junction lies above the GEJ.) These strictures usually measure 1 cm in length or less and are rarely longer than 3 cm. Other factors predisposing to stricture formation include Zollinger—Ellison syndrome, prolonged nasogastric tube placement, and scleroderma.

Dysphagia is the most common presenting symptom of peptic stricture. Many strictures form gradually. Patients tend to adapt their diet to prevent the physical discomfort, and therefore, complaints regarding the severity of dysphagia correlate very loosely with severity of the stricture. Because the stricture is a mechanical obstruction, the dysphagia typically is more pronounced with solid foods than with liquids. Most patients have a prior history of GERD, along with an associated hiatal hernia. Weight loss is uncommon with peptic stricture in contrast to malignant strictures, which typically occur in patients over 50 years of age, develop more suddenly, may not be accompanied by a hiatal hernia, and are almost always associated with weight loss.

Strictures typically are diagnosed by barium swallow. As noted earlier, benign strictures are located at the squamocolumnar junction; associated with a hiatal hernia; and typically have smooth, tapered proximal and distal segments. Associated esophagitis may be present. In contrast, malignant strictures may not be associated with a hiatal hernia and have irregular narrowing with abrupt or “shouldered” proximal or distal edges. EGD should be performed. Associated esophagitis is seen often. Biopsy should be performed proximally throughout the stricture and beyond, if possible. Use of a thinner pediatric endoscope may facilitate passage. Coexisting Barrett metaplasia may be found. Dilation of the stricture with serial Maloney dilators may permit passage of the endoscope and yield a more detailed examination.

Initial management of peptic stricture involves dilation (see Chapter 44) and PPI therapy. Use of PPIs rather than H₂-receptor antagonists results in a lower incidence of stricture recurrence. Three different types of esophageal dilators may be used. The soft, flexible, mercury-filled Maloney dilator is used more commonly for benign strictures. These tapered dilators come in various sizes up to 60F (20 mm). They are designed with flexible tips to follow the course of the esophageal lumen. These dilators are preferred for their ease of use, provided that the stricture is not extremely tight or tortuous. Savary dilators are similar in shape to the Maloney dilators, but they are stiffer, have a central channel, and are designed to be passed over a flexible guidewire. These dilators are used for more tortuous or tight strictures, where there is a greater concern for perforation, or if there may be difficulty engaging the stricture. The wire is passed through the stricture under direct vision and is verified with fluoroscopy to end in the stomach. A series of dilators then may be passed using Seldinger technique. These dilators range in size from 15 to 60 F (5–20 mm). Through-the-scope balloon dilators are also available, although they are disposable and more expensive.

The typical peptic stricture may be dilated safely with Maloney dilators without the use of fluoroscopy. The risk of perforation should be <1%. One-third to one-half of patients require long-term repeat dilations. After two or more dilations, additional dilations are required in nearly all patients. Surgery is indicated in patients who cannot be dilated, as well as in patients who have recurrent strictures while on PPI therapy. The preferred treatment in patients who can be dilated is a laparoscopic fundoplication. Any patient who has concomitant esophageal shortening also should undergo an esophageal lengthening procedure. If the patient cannot be dilated, then esophagectomy is usually needed; however, this is extremely rare, and a thorough effort should be made to exclude unusual malignancies. In a patient with an otherwise healthy esophagus, distal esophagectomy is an acceptable procedure. Colon or jejunum interposition typically is used because a high incidence of recurrent reflux and stricture has been described with the gastric pull-up procedure after distal esophagectomy. Distal esophagectomy is also reasonable in patients suffering from perforation after dilation, especially if they have required several dilations. In these patients, repair of the perforation is associated with a very high incidence of recurrent, difficult-to-dilate strictures. Repair of the stricture and perforation using a Thal patch of stomach to widen the esophagus has been described. This technique is prone to leakage and often leaves the patient with wide-open reflux. For this reason, distal esophagectomy usually is preferred.

Barrett Metaplasia

Barrett metaplasia is defined as the presence of a “specialized” columnar epithelium in the distal esophagus. The term specialized is used because the epithelium is different from that found in the gastric cardia, with features of gastric, small intestinal, and colonic mucosa, including goblet cells. It is most similar to intestinal epithelial cells. Before Barrett metaplasia can be diagnosed, first it must be identified by the endoscopist. The GEJ is defined endoscopically as the point where the gastric folds are first apparent in the minimally distended esophagus. The squamocolumnar junction, or Z-line, typically coincides with the GEJ. Barrett metaplasia exists when the usual pale pink stratified squamous epithelium of the distal esophagus is replaced by a beefy red columnar intestinal mucosa. The exact location of the GEJ may be difficult to discern, especially in patients with a hiatal hernia. Varying distances have been proposed for the length of columnar mucosa that must be identified before the diagnosis of Barrett esophagus can be made. Patients with less than 3 cm of Barrett metaplasia have short-segment Barrett. Those with more than 3 cm of metaplasia have long-segment Barrett. This distinction is important because patients with short-segment Barrett may not have the same risk of dysplasia.

Barrett metaplasia is caused by frequent, repeated, and prolonged exposure of the esophagus to the gastroduodenal contents. Patients with Barrett metaplasia have been shown to have more severe reflux than patients with esophagitis. Average LES pressure in Barrett patients is about half that of patients with uncomplicated reflux esophagitis (5 vs. 9 mm).¹² Most Barrett patients have a hiatal hernia of significant size (76% vs. 50% with uncomplicated esophagitis).¹³ Exposure to acid and bile is higher in Barrett patients than in patients with non-Barrett esophagitis. The origin of the specialized metaplastic cells is unclear. A number of hypotheses have been proposed. For example, these specialized cells may result from migration of gastric cardia cells to the distal esophagus, they may represent metaplasia of esophageal squamous cells, or they may develop from esophageal glandular cells. Experimental and clinical evidence exists for each of these hypotheses without clear consensus.

Approximately 10% of patients with esophagitis eventually will develop Barrett metaplasia. The disease is twice as prevalent in men as in women. Barrett patients have a 30- to 125-fold chance of developing esophageal adenocarcinoma compared with the general population. The rate of cancer development is 1 in 200 patient-years for those with metaplasia. Not all Barrett patients are at equal risk of developing adenocarcinoma. Risk factors for the development of adenocarcinoma in these patients include male gender, smoking, greater length of Barrett esophagus, Barrett ulcer, peptic stricture, white race, and older age. Most Barrett patients have symptoms that are indistinguishable from those of uncomplicated esophagitis. Barrett patients tend to have more complications from esophagitis (e.g., bleeding and stricture). The development of a stricture at the squamocolumnar junction well above the GEJ in a patient with a hiatal hernia is nearly pathognomonic for Barrett esophagus.

Barrett carcinoma is believed to progress in the following sequence: metaplasia < low-grade dysplasia < high-grade dysplasia < adenocarcinoma. Progression from low-to-high—grade dysplasia may take many years, whereas progression from high-grade dysplasia to adenocarcinoma occurs on average in 14 months.¹⁴ The annual incidence of the development of cancer in a patient with Barrett esophagus has been estimated to be from 0.2% to 1.9%.¹⁵ Dysplasia indicates the development of neoplastic changes that consist of alterations in the cellular and glandular architecture. Cell nuclei may be hyperchromatic and enlarged, with increased mitotic figures. The nuclei may show loss of polarity and, instead of appearing uniformly at the base of cells, may migrate toward the luminal surface. In high-grade dysplasia, nuclei extend toward the luminal one-third of the cell. The differentiation between low- and high-grade dysplasia is often difficult.

The treatment of Barrett esophagus with low-grade and high-grade dysplasia is controversial and continues to evolve. The natural history and progression of low-grade dysplasia to high-grade dysplasia to intramucosal carcinoma and eventually to cancer with metastatic potential has yet to be well defined. The surveillance and treatment of Barrett Esophagus is discussed in detail in Chapter 41.

GERD is both common and heterogeneous. It is related but not limited to obesity and modern Western lifestyle. Careful elucidation of all symptoms, as well as thorough clinical evaluation including endoscopy, pH study, and a manometric evaluation, are recommended prior to surgical therapy or any other interventions. Whenever suspected because of medical history or symptoms, gastric emptying should be assessed to avoid operating on patients with poor gastric motility. This disorder is often treated by multiple specialists including gastroenterologists, general surgeons, thoracic surgeons, otolaryngologists, and pulmonologists. Behavior modification can improve the symptoms of many but not all patients. It is important to remember to design individualized treatments that are best for each patient rather than fit the patient to a specialized treatment strategy.