Synopsis  

Gastroesophageal reflux disease (GERD) is increasingly recognized around the world and has been linked to the development of esophageal carcinoma. Endoscopy is widely performed not only for diagnosis but also for therapy (i.e. dilation). The majority of GERD cases do not have endoscopic abnormalities and a complete evaluation with biopsies, esophageal motility, and 24-hour ambulatory pH monitoring are frequently needed before therapy is initiated. This chapter reviews the endoscopic recognition of GERD, the various staging classifications of the disease and its complications, and the assessment and management of esophageal strictures. We further expand on the promising new technologies for endoscopic therapy of GERD and provide some concepts and trends for the future.

 Introduction   

Several studies on the value of symptoms and endoscopic findings in GERD have concluded that it is difficult to establish the diagnosis by patients' history alone and that reflux disease may be present even with normal endoscopy [1]. Therefore, for the correct diagnosis of GERD and the assessment of its severity, many authorities recommend a complete esophageal evaluation that includes endoscopy with biopsies, esophageal motility and 24-hour ambulatory pH monitoring (Fig. 1) [2].

 Diagnostic upper endoscopy   

Diagnostic upper endoscopy in GERD may reveal either a normal or an abnormal esophagus characterized by ulcers, erosions, exudates, and friability. There are four widely used systems of grading the severity of the esophageal damage, the Savary–Miller, the Hetzel–Dent, the MUSE, and the Los Angeles classification (Figs 2–4) [3–5]. The refinement of endoscopes, the advent of transnasal endoscopy, and the use of absorptive stains to detect and define abnormalities have facilitated imaging of the esophageal mucosa to a level of excellent detail and resolution. Although diagnostic upper endoscopy may reveal characteristic abnormalities of GERD, biopsy or brush cytology may be needed to obtain histological or cytological confirmation [6].

Normal esophagus   

Despite pathological gastroesophageal reflux, the esophagus may be normal from the cricopharyngeus to the esophagogastric junction. Since GERD may cause several extraesophageal manifestations (i.e. laryngitis, asthma, difficulty with throat clearing) endoscopy should always explore the posterior aspect of the pharynx, the glottis, the posterior cricoid cartilage, and both the pyriform sinuses [7–9].

Upon passing the endoscope just beyond that level, one appreciates the concentric, transient effect of the upper esophageal sphincter (UES) and then the pink appearance of the esophageal mucosa with its longitudinal folds. These folds usually measure 2–4 mm, and run parallel to each other, converging towards the centre of the lumen. At the level of the middle third of the esophageal body, a small amount of saliva may be normally appreciated. The presence of a larger amount of liquid or the existence of a fluid level in the midesophagus suggests significant impairment of esophageal clearance (i.e. scleroderma or achalasia) [10].

The length of the esophagus varies from 35 to 45 cm, depending on the height of the individual. As the endoscope is passed into the distal third of the esophageal body the esophageal mucosa may become slightly more erythematous.

Hiatus and Z-line   

At this level, two landmarks should be specifically assessed: the integrity of the gastroesophageal junction (Z-line) and the diaphragmatic hiatus. The Z-line marks the transition of the columnar gastric epithelium (reddish colour) to the squamous esophageal epithelium (whitish-pink colour). It is visualized both in an antegrade fashion as well as during retroflexion in the stomach (Fig. 5). The Z-line should be at the same level as the diaphragm or a bit below (Fig. 6). It is usually a distinct line, but it can be irregular or exhibit small ectopic islands of reddish mucosa above it. The endoscope traverses the esophagogastric junction easily upon air insufflation; resistance to passage or friability implies disease (i.e. stricture). The gastroesophageal junction should remain closed unless swallowing occurs or air is insufflated through the endoscope. Persistence of a patulous junction is abnormal and implies the presence of reflux.

In contrast, sometimes the gastroesophageal junction is difficult to evaluate because the mucosal folds converge excessively. This is not a feature of GERD but it can be seen in patients after fundoplication (rosette formation). Rarely, in older individuals with thoracic aortic dilation, an extrinsic compression of the distal esophagus is seen just above the diaphragmatic orifice and it should not be misinterpreted as a stricture. At this point, it is also important to measure the distance—if any—between the Z-line and the diaphragmatic hiatus (Fig. 7). Normally, this distance should be 0–2 cm; otherwise a sliding hiatal hernia is present.

Glycogenic acanthosis   

Glycogenic acanthosis of the esophagus consists of whitish elevations of the squamous mucosa of variable size (1–3 mm) scattered throughout the esophagus, but mostly in the distal two-thirds of the esophageal body. Histological examination of these lesions reveals the presence of excess glycogen that is easily demonstrated using the periodic acid–Schiff stain (PAS). Chromoendoscopy using Lugol's solution creates a brownish discoloration of these lesions that is more pronounced than the adjacent esophageal mucosa. The exact significance of glycogenic acanthosis is unknown, but an association with GERD has been reported [11].

Erythema   

The presence of erythema in the distal esophagus has created a lot of controversy since it is a poorly reproducible finding. Because the colour controls of video-endoscopes and image-processing units are variable, it is very difficult to attribute erythema to the underlying presence of esophagitis unless biopsies are taken. Therefore, the reporting of erythema should not necessarily be viewed as an abnormal finding, suggestive of GERD. Further, the occasional presence of gastric-like mucosa within 1 cm above the diaphragmatic hiatus does not imply the presence of Barrett's esophagus, and it may be seen in asymptomatic individuals as well as patients with GERD [12].

GEJ variants   

Several variants of the gastroesophageal junction should not be misinterpreted as contributing to GERD. For example, sometimes the normal esophagogastric angle is absent and the distal esophageal folds merge directly with the proximal gastric folds of the lesser curvature [13]. In contrast, the presence of a normal esophagogastric angle, that is, the point where the left esophageal wall merges with the greater gastric curvature, does not exclude pathological reflux. Occasionally, the Z-line is not easily perceptible as a result of gastric atrophy or anaemia, and this does not imply distal esophageal inflammation.

Abnormal esophagus   

Erythematous areas and erosions   

Careful distension of the distal esophagus by air insufflation may reveal small superficial erythematous areas or erosions. These changes are more frequently seen with magnification endoscopes or upon close-up distension view. Unfortunately, the smaller the surface alteration, the more there is interobserver variation. Sometimes, erythematous lesions are difficult to distinguish from gastric mucosa or islands of Barrett's esophagus. When there is doubt, biopsies should be done. Depending on their size, the absence of perceptible depth, and confluence, they are classified as either Savary–Miller stage 1 or Los Angeles grade A [3,4] (Fig. 8).

If biopsies of erythematous lesions reveal intestinal metaplasia, the diagnosis of Barrett's esophagus is made. Regardless of its length or surface involvement, this condition is associated with an increased risk for malignancy (adenocarcinoma) (14).

Ulcers   

Ulcers are characterized by a perceptible depth upon air insufflation, a white base (crater), coating exudates or pseudomembranes, and surrounding erythema. Close visualization reveals either elevated margins around the ulcer or a depressed appearance. Surrounding the ulcer(s) there is oedematous, congested mucosa. All these characteristics, namely depth, margins, and surrounding erythema, help with the assessment of GERD by either the Savary–Miller or the Los Angeles classification [3–5].

Specifically, Savary stage 1 or LA grade B: isolated longitudinal erosion or ulcer larger than 5 mm, on red background associated with an elevated mucosa along a longitudinal fold. Savary stage 2 or LA grade C: longitudinal erosions and ulcers that are confluent but affect only one part of the esophageal wall along the distal esophageal folds (Figs 9–11). Savary stage 3 or LA grade D: confluent and circumferential ulcers covered by exudates or pseudomembranes with perceptible depth upon insufflation and surrounding erythematous mucosa (Fig. 12). Savary stage 4: wall thickening and fibrosis with stricture formation with or without associated ulceration.

Sometimes, particularly after empirical therapy with antisecretory drugs, the stricture may be annular, distinct, short in length (< 1 cm), and without associated mucosal lesions (see strictures below) (Fig. 13).

At other times, the stricture is associated with an ulcer that extends from the Z-line to the level of the stricture in a longitudinal fashion. In such cases, a necrotic, haemorrhagic base is frequently encountered and is associated with bleeding and friability. Biopsies are important to rule out malignancy and Barrett's esophagus.

Strictures   

Typically a peptic stricture (Savary–Miller stage 4) is the result of a long-standing reparative process during the course of erosive esophagitis. It is noted most frequently in the distal esophagus as an annular constriction that impairs the easy passage of the endoscope [15]. However, it may not be easily appreciated with the newer, small caliber endoscopes, and air insufflation may be needed to outline its presence. A peptic stricture may or may not be associated with an esophageal ulcer, Barrett's metaplasia, or cancer, and biopsies should be obtained [16].

The presence of a peptic stricture deserves immediate attention in order to relieve dysphagia and prevent aspiration of stagnant esophageal contents. When performing endoscopy in patients with dysphagia suspected to have a peptic stricture, it is mandatory to do it under direct vision. If difficulty is encountered with passing the endoscope, guidewire-assisted dilation should be performed under fluoroscopic control either with a C-arm-type mobile unit in the endoscopy unit or a dedicated fluoroscopy table in the radiology suite. Although not always necessary, a barium study may be useful in delineating the severity, location, length, and complexity of the stricture [17].

Most peptic strictures are smooth on their surface, short, straight, and most frequently distal (Fig. 14). If such strictures can be traversed with the endoscope, use of a guidewire or balloon system is not needed and Maloney dilation can be safe and effective. Rarely, complex esophageal strictures are encountered. Such strictures are long, narrow, tortuous, or associated with esophageal diverticula and do not allow the tips of freely passed dilators to enter the stomach. These strictures require a guidewire-based system or a balloon dilator under fluoroscopic guidance. A barium swallow may assist with the localization, length, and diameter of strictures, their complexity, and the coexistence of esophageal diverticula or hiatal hernia. This in turn determines the preferred dilation technique, the number of sessions required, and the dilation risk [18].

Hiatal hernia   

A hiatal hernia is an intrathoracic ascent of the proximal portion of the greater curvature of the stomach through the esophageal hiatus of the diaphragm [19] (Figs 15 and 16). It can be either sliding when the cardia moves up into the thorax, or rolling (paraesophageal) when the cardia remains in place.

In sliding hernias, the upper border of the hernia is the squamocolumnar junction dislocated into the chest for a variable length, and the lower border is the diaphragmatic indentation that can be appreciated—if possible—by asking the patient to sniff, a maneuver that approximates the diaphragmatic crura. In paraesophageal hernias, a gastric sac of variable size is displaced into the mediastinum, but the squamocolumnar junction and the cardia remain in place. Sometimes, a mixed variety is seen.

A sliding hiatal hernia exists in 2–5% of the population, but with increasing age its prevalence reaches 10–15%. It may be found in up to 85% of patients with symptomatic reflux and it plays a pathogenetic role. The retention of acidic gastric contents within the hernia pouch allows for repetitive episodes of reflux during the transient lower esophageal relaxations (tLESRs) that happen either spontaneously or in response to swallows. Further, in the presence of sliding hiatal hernia, when the intra-abdominal pressure increases there is no compensatory, parallel increase in lower esophageal sphincter (LES) pressure.

These two factors contribute to pathological reflux in patients with sliding hiatal hernia [20]. In contrast, paraesophageal hernias are quite rare and may not be associated with GERD.

 Endoscopy in esophagitis   

Endoscopy is not necessary in every patient with heartburn but it should be performed if the symptoms become chronic and/or refractory to treatment, if the diagnosis is unclear, or if complications are suspected. However, the absence of endoscopic features of GERD does not exclude the diagnosis. Ambulatory 24-h esophageal pH monitoring should be considered in cases where endoscopy is negative [21].

Endoscopic staging of esophagitis   

Because of wide observer variability in assessing the severity of esophagitis, the four most common classifications are listed separately below.

Savary–Miller classification   

This is the most widely used grading of esophagitis and distinguishes four grades (1–4) ranging from erythema to erosions, ulcers, and stricture formation [3]Fig. 2). Because of the ambiguity of grade 4 esophagitis, there are modifications that either offer subdivisions of grade 4 or define Barrett's esophageal metaplasia as grade 5 (Figs 17 and 18).

Hetzel–Dent classification   

Because most patients with GERD either do not have endoscopic lesions or have mild esophagitis, the Hetzel–Dent system focuses on subtle mucosal abnormalities [4]Fig. 3). Of note, esophageal strictures and Barrett's metaplasia are not parts of this grading.

MUSE classification   

This system is analogous to the TNM classification system used in oncology and describes several independent features: metaplasia (M), ulceration (U), stricture formation (S), and mucosal erosion (E). Each of these elements is graded independently in four increments (0, absent; 1, mild; 2, moderate; 4, severe) [12]. The result of an endoscopic examination is described as a grade for each lesion type, i.e. M0U1S0E1.

Los Angeles classification   

This system classifies severity by the extent of mucosal abnormalities, and complications are recorded separately [5], Fig. 4). A mucosal break implies erosion in the squamous epithelium with or without overlying exudate. The system is very sensitive to early disease and is preferred in drug trials assessing drug efficacy.

Other lesions associated with GERD   

Schatzki's ring   

Schatzki's ring is a diaphragm-like constriction that results from mucosal hyperplasia of the gastroesophageal junction [22] (Fig. 19). It can be seen in the antegrade position, upon air insufflation, or during retroflexion at the level of the cardia. Although its relationship to GERD is not established, many studies have shown associated pathological reflux [23].

Mallory–Weiss tear   

A tear at the gastroesophageal junction (Mallory–Weiss tear) typically results from emetogenic injury to the distal esophagus and accounts for approximately 5% of upper gastrointestinal bleeding episodes [24]. It is usually not seen in conjunction with esophagitis, but it is associated with symptoms of reflux.

Scleroderma   

Esophageal involvement by scleroderma leads to severe mucosal alterations because of the lack of effective esophageal peristalsis in the distal two-thirds of the esophageal body and the absence or significant diminution of the lower esophageal sphincter (LES) pressure (25). This frequently results in severe involvement of the wall of the esophagus, ulcers and exudates in conjunction with a patulous gastroesophageal junction, and free reflux of gastric contents during endoscopy [26].

Postoperative changes after fundoplication   

After an antireflux operation, the anatomy of the gastroesophageal junction is permanently altered. The endoscopic appearance of the distal esophagus and cardia, both in the antegrade and retroflexed views, is characterized by confluence of the distal esophageal folds that create a rosette [27] (Fig. 20). Significant air insufflation is required to adequately visualize the mucosa and areas of short-segment Barrett's esophagus may become imperceptible.

In cases of failure of an antireflux procedure, several causes need to be entertained. If GERD symptoms persist or recur, a laxity of the fundoplication procedure leading to a patulous gastroesophageal junction is frequently found [28]. Rarely, an intrathoracic mobilization or a 'slipped' Nissen fundoplication is seen (Fig. 21). If new symptoms of poor esophageal emptying occur (dysphagia or regurgitation), a very tight fundoplication or scarring may be noted (Fig. 22).

Endoscopic biopsy   

Endoscopic esophageal mucosal biopsies should be obtained as part of the evaluation of patients with GERD. Changes of reflux esophagitis (papillary hyperplasia, eosinophilic or chronic inflammatory infiltration) may be present even in the absence of endoscopic abnormalities [29] (Fig. 23).

Chromoendoscopy in GERD   

Chromoendoscopy in patients with GERD is performed with either Lugol's solution or methylene blue [30,31]. Lugol's solution stains normal esophageal mucosa dark green-brown. Any lesion of the esophagus that does not contain glycogen (i.e. carcinoma, intestinal metaplasia, or inflammation) does not stain. Spraying 20–50 ml of Lugol's solution on the mucosa may therefore reveal dysplasia or early cancer that would not be visible endoscopically.

 Endoscopic management of esophageal strictures   

Esophageal dilation   

Although dysphagia is the main symptom of a peptic stricture, other conditions such as non-obstructive GERD, motility disorder, infection, malignancy, and esophageal rings should be considered [32]. The goals of therapy for peptic strictures are the relief of dysphagia and the prevention of stricture recurrence with adjunctive, continuous acid-suppressive therapy [33]. Conditions that may increase the risk of esophageal dilation are cervical deformity, bleeding disorder, recent esophageal surgery, esophageal perforation, large thoracic aneurysm, or foreign body impaction.

The use of fluoroscopy allows for accurate positioning of the endoscope tip, monitoring the guidewire placement through the stricture for subsequent Savary-type dilation, or for monitoring the performance of a tapered-tip dilation using either Maloney or Hurst dilators [34]. Further, fluoroscopy is quite useful in the performance of balloon dilation as it delineates the waist of the balloon at the level of the stricture and its obliteration during inflation. Upon completion of the dilation, if there is suspicion of a perforation, a contrast study should be performed using gastrograffin, in order to fluoroscopically look for contrast extravasation [35].

The use of fluoroscopy for esophageal dilation may not improve the outcome. In a randomized trial of benign esophageal stricture dilation with Maloney dilators with or without fluoroscopic guidance, those who underwent dilation with fluoroscopy more often tolerated passage of a 12.5 mm barium pill after dilation (62 vs. 42%), and had better dysphagia relief (93 vs. 69%) (36). In another study, however, it was the operator's experience rather than fluoroscopy that was associated with procedural success [37].

Principles of esophageal dilation   

Esophageal dilation circumferentially stretches the stricture and splits its fibrous composition. Typically, the mechanical dilators exert both a longitudinal and radial force, and dilate the stricture progressively from the proximal to its distal end. In contrast, balloon dilators exert their dilating force in a radial fashion and simultaneously over the entire length of the stricture [38].

Mechanical dilation   

There are two types of mechanical dilators: those that are passed freely, and those that are inserted over a guidewire. The tapered-tip Maloney (Medovations, Germantown, WI) are the most commonly used dilators; they are available in multiple sizes and do not require a guidewire for placement. The Hurst dilators have a rounded tip that is harder to pass, and are rarely used today. Both Maloney and Hurst dilators are filled with tungsten, which provides flexibility and weight. The Savary–Gilliard guidewire-assisted dilators (Wilson Cook, Winston-Salem, NC; American Dilatation System by Bard Interventional Products, Tewksbury, MA) are widely used [39]. They are made from plastic, have a tapered tip, and are available in multiple sizes. The Eder–Peustow dilator system (Eder Instruments Co., Chicago, IL), uses progressively larger elliptical (olive-shaped) metal dilators that are passed over a guidewire.

Pneumatic (balloon) dilation   

There are two types of balloon dilators: the TTS, a through-the-scope dilator (various manufacturers), which is passed directly through the endoscope, and the OTW, an over-the-guidewire balloon dilator (Boston Scientific Corp., Natick, MA) [40]. Balloons exert a greater radial force and retain their maximum predetermined diameter. The new generation of TTS balloons (Boston Scientific Corp., Natick, MA) expand to three different diameters at 1.5 mm increments without having to change balloons. Studies comparing mechanical to balloon dilation have shown mixed results and success depends mostly upon the experience of the endoscopist [41,42]. In general, complex strictures should be dilated with a guidewire-based mechanical or balloon dilating system.

Techniques of dilation   

Esophageal dilation is generally performed as an outpatient procedure under conscious sedation and meticulous local pharyngeal anaesthesia [42]. Some patients with simple peptic strictures who require frequent dilations may not need sedation or may self-dilate at home using Maloney dilators. Typically Maloney dilation is performed while the patient is seated. In order to improve visualization and to avoid aspiration, fasting before dilation is important. Drugs that affect platelet function should be stopped 1 week before the procedure and antibiotic prophylaxis should be given to patients at high-risk for endocarditis.

When using mechanical dilators, the diameter of the initial dilator should be approximately the same width as the stricture. For balloon dilators, the initial balloon size reflects the estimated stricture diameter: a 10 mm balloon for diameters between 2 and 4 mm; a 12 mm balloon for diameters between 5 and 9 mm; and a 15 mm balloon for a diameter greater than 9 mm. In order to reduce the likelihood of complications, no more than three dilators of progressively increasing size should be passed in a single session, and the luminal diameter should not be increased by more than 2 mm. The effectiveness of dilation is approximated by the degree of resistance during passage of the dilator.

The need for, and frequency of repeated dilations depends upon the patient's response. If the stricture has a narrow diameter and exhibits significant resistance, repeated sessions every 5–7 days may be required. As a general rule, the last dilator used in the previous session is usually passed first. However, since shrinkage of strictures following dilation may occur, it is useful to evaluate the stricture diameter prior to dilation, and if needed to revert back to a smaller-sized dilator.

TTS balloon dilators and mechanical dilators are equally safe for dilating esophageal strictures, but recurrences may be less during the second year following TTS compared to mechanical dilation [43]. TTS balloon dilators are passed through the channel of the endoscope into the stricture under direct vision or during fluoroscopy. Obliteration of the balloon 'waist' on fluoroscopy reflects effective stricture dilation. Two dilations of 30–60 s in each session are usually performed but the optimal number and duration of inflations is not known. The ability to move the inflated balloon freely through the treated stricture implies a successful dilation. If several sessions are required, they are usually performed every 3–5 days [44].

In general, dilation to 18 mm allows intake of a regular diet; a diameter of 15 mm allows for a modified regular diet and patients should be instructed to chew their food well. In contrast, patients with an esophageal lumen less than 13 mm will usually experience dysphagia with solids. The free passage of a 12 mm barium tablet into the stomach can be used as an objective endpoint for dilation [45,46](Figs 24–26).

Dilation of distal esophageal (Schatzki's) rings   

Given their dense, fibrous nature, Schatzki's rings often fail to respond to standard Savary-type dilation. In these instances, disruption using balloon dilators that are capable of applying radial force on the ring is preferable. If this approach fails and the patient remains symptomatic, consideration should be given to ring incision, using electrocautery [47]. Under direct visualization and minimization of movement, four-quadrant diathermy cuts are made using a regular or a needle-knife sphincterotome, resulting in restoration of the esophageal lumen to its original diameter. However, this procedure carries the risk of perforation and should only be performed by experienced endoscopists.

Concomitant medical therapy for strictures   

Proton pump inhibitors should be used in patients with peptic strictures since they reduce the need for subsequent dilation better than H2 antagonists. In one study, 336 patients with peptic strictures were randomized to omeprazole 20 mg QD or ranitidine 150 mg BID for 1 year after esophageal dilation to 12–18 mm and follow-up endoscopy with dilation was performed as indicated. The omeprazole-treated patients required significantly fewer repeated dilation sessions (30 vs. 46%) and had improved dysphagia scores compared to the ranitidine-treated group [48]. Patients with recurrent peptic strictures despite use of a proton pump inhibitor may benefit from ambulatory 24-h pH monitoring while on therapy to assure the adequacy of acid suppression.

Refractory strictures   

Some patients with benign esophageal strictures do not achieve acceptable symptom relief despite an intensive dilation schedule and aggressive proton pump inhibitor therapy [49]. These patients should be evaluated for, and offered surgery. Those who are poor surgical candidates may benefit from regular self-dilations with a Maloney dilator. However, home dilation has risks, and many patients are not motivated or suitable candidates.

Intralesional injection of corticosteroids   

The injection of corticosteroids into the stricture may reduce stricture recurrence following dilation [50]. Although the mechanism of action is not well understood, corticosteroids may reduce scar tissue formation by inhibiting collagen deposition. A standard sclerotherapy needle is used to inject an average of 0.2 ml of triamcinolone acetonide (Kenalog, Bristol-Myers Squibb, Princeton, NJ) into all four quadrants at the narrowest region of the stricture. Some use an ultrasound miniprobe passed through the endoscope to guide the injection into the thickest part of the stricture [51].

Other endoscopic methods for strictures   

Electrosurgical incision of peptic strictures has been reported but the safety of this technique has not been rigorously tested. A biodegradable esophageal stent made from a coil of poly L-lactide (Instent, Eden Praire, Minn) has been used for relief of dysphagia due to a benign refractory, radiation-induced stricture. Because of the possibility for migration or stent-induced fistula formation, expandable metal stents are not recommended for peptic strictures. A retrievable, expandable stent is under testing.

Complications of esophageal stricture dilation   

The major complication of esophageal dilation is esophageal perforation [52]. The incidence of perforation depends on the etiology of the stricture, the technique, and operator experience, and ranges between 0.1 and 0.3%. The clinical manifestations of esophageal perforation depend upon the location and extent of the perforation and the duration since the onset of injury. Cervical perforation may be associated with neck pain and tenderness, hoarseness, and subcutaneous emphysema. Perforation of the thoracic esophagus is most commonly associated with chest, back, or epigastric pain, which is worse with inspiration or swallowing. There may be associated dysphagia, odynophagia, bleeding, or dyspnea. Patients may be tachycardic, tachypneic, and febrile, and eventually develop septic shock.

Patients with suspected perforation should undergo chest and neck X-rays to look for subcutaneous emphysema or pneumomediastinum [53]. Contrast esophagography usually establishes the diagnosis. Water-soluble contrast agents (such as gastrograffin) are used initially, followed by barium studies if results are negative. Barium has a higher sensitivity for small perforations. If clinical suspicion remains high despite negative findings on esophageal contrast studies, computed tomography should be performed as it is highly sensitive for detecting soft tissue air. Patients whose perforation is recognized early have a good prognosis and, in many cases, can be managed non-surgically. A delay in diagnosis usually necessitates surgery and carries a mortality rate of 25–50%. Esophageal dilation may cause bacteremia in up to 45% of cases and antibiotic prophylaxis is needed for patients at high risk for endocarditis [54]. Significant bleeding requiring blood transfusions after esophageal dilation occurs in 0.2–2% of cases.

 Endoscopic therapies for GERD   

Several endoscopically assisted interventional techniques have emerged for the treatment of GERD. These therapies should be considered for patients who respond well to proton pump inhibitors (PPIs), but do not wish to take long-term medications, patients who respond partly to PPIs or have regurgitation not responsive to PPIs but do not want surgery, patients who are afraid of possible long-term sequelae of PPI therapy, patients intolerant to PPIs, and possibly patients who have failed fundoplication for GERD [55].

The best candidates for endoscopic therapies are those who have well-established GERD documented by endoscopy or pH monitoring and esophageal motility studies and have at least in part responded to PPIs. Other potential candidates are those with PPI-responsive GERD despite normal 24-h pH study (Non-Erosive Reflux Disease, NERD). Consideration should also be given to patients with PPI-responsive, extraesophageal manifestations of reflux. The endoscopic procedures currently available are not appropriate for the late, complicated disease states of esophageal shortening or stricture, poor esophageal function, and dysphagia [56].

Many technique modifications will certainly improve the safety and efficacy profiles of all endoscopic therapies. Efficacy comparisons and sham trials will be needed in order to show differences among these new endoscopic modalities in the management of GERD. Comparisons to surgery and drug therapy will probably ensue. Safety and tolerability will continuously be assessed in a postmarketing surveillance. Longer observations of patients already treated will address the durability of effect over periods longer than 12 months. The general applicability in all GERD groups and cost-effectiveness will be determined with wider application and future technological advances [57].

Many questions about endoscopic therapies for GERD remain unanswered: How do these procedures work? How well do they work? How safe are they? Who are the best candidates for them? How long does their effect last? Answers to all these questions will become available with the increased and widespread utilization of these procedures and the design and execution of well-controlled, randomized clinical trials.

Radiofrequency energy delivery (Stretta^®)   

Radiofrequency (RF) delivery for GERD, or the Stretta^® procedure, is performed using the Stretta^® system (Curon Medical Inc., Sunnyvale, CA), which incorporates a flexible catheter with a control module/generator. The catheter has a soft bougie tip and a 6 mm shaft, and contains a balloon/basket that inflates to a maximum 3 cm diameter and deploys four electrodes. The generator delivers RF in an automated fashion under temperature control using thermocouple monitoring while the power output is regulated by computer algorithm. Target tissue as well as mucosal temperature are achieved and maintained through a special suction and irrigation system [58].

Stretta^® technique   

Endoscopy is first performed, the distance from the incisors to the squamocolumnar junction (Z-line) is measured, a super-stiff guidewire is placed in the duodenum, and the endoscope is removed, leaving the guidewire in place.

The Stretta^® catheter, comprised of the bougie tip, balloon/basket assembly, and four electrode delivery sheaths positioned radially around the balloon, is then passed through the mouth block over the guidewire and positioned up to 1 cm above the Z-line. The wire is then removed. The balloon is inflated and four nickel–titanium needle electrodes (22 gauge, 5.5 mm length) are deployed into the muscle of the GEJ. RF energy is delivered to each electrode, while cooling the mucosa with irrigation, using the temperature controlled RF generator system described above.

Additional lesion sets are created in the region from 1 cm above to 2 cm below the Z-line by rotating the catheter 45° and varying its linear position. A total of 14 lesion sets are created, a number determined by previous animal model investigations and the individual patient anatomy. Patients then undergo endoscopy immediately after delivery of RF energy to assess the early post-RF appearance of the mucosa.

Typically, collagen contraction is seen immediately after RF delivery with tightening of the GEJ [59,60] (Fig. 27).

Mechanisms of action   

There are two potential mechanisms of action of Stretta^®:

• Mechanical: increased gastric yield pressure (GYP) by 75% and lower esophageal sphincter pressure (LESP) by 21% have been shown in a porcine model of reflux. After Stretta^®, increased muscular wall thickness has been shown histopathologically in a canine model and endosonographically in a porcine model. A modest increase of the lower esophageal sphincter pressure (LESP) 6 months after Stretta^® has also been found in a recently reported human trial.
  
• Neurological: decreases in the tLESR frequency by 54% in a canine model and by 25–44% in two human Stretta^® studies incorporating 3-h postprandial esophageal manometry and pH monitoring after a standard meal have been reported [61].
  

Complications   

The following complications have been encountered with Stretta^®: bleeding 0.17%, mucosal ulceration 0.08%, pleural effusion 0.08%, perforation 0.33%, aspiration/death 0.17%. The total number of complications with 1200 procedures is 0.83%. In order to prevent such complications patient selection is paramount. Technical considerations, such as catheter placement < 1 cm above the Z-line, balloon distension to ~2.5 psi, proper sedation, endoscopy confirmation after the first set or therapy, guidewire catheter placement, and adherence to patient management guidelines have played a role in curtailing the number of complications [62].

Endoscopic implantation of bulking agents   

The injection of bulking agents that are of low viscosity to traverse a needle and then change to a solid state at the local injection site has been studied for many years [63]. The site of injection is near the squamocolumnar junction in a circumferential fashion.

The injection needle diameter must match implant viscosity. Preferably, 23–27 per cent gauges are used since they are easier to manipulate and atraumatic enough to minimize back-leak of the implantable material from the puncture site.

The depth of implant (submucosal or intramuscular) also varies depending on the material, which tends to flow 1–2 cm up or down from the injection site. If a submucosal injection is made, the endoscopist observes a bulging effect.

If not, and a deeper implantation is desired, fluoroscopy is used. Initial trials with bulking agents were performed using Teflon paste and bovine dermal collagen in a refluxing dog model. These trials showed that there is a transient increase in gastric yield pressure and a decrease in esophagitis [64]

Inert polymer microspheres   

Inert polymer microspheres made of polymethyl-acrylate (PMMA, Plexiglass) dispersed in gelatin solution have been injected into the mucosa of the lower esophageal sphincter region in 10 patients, aiming at persistent bulking of the gastroesophageal junction barrier. The microspheres resist phagocytosis and do not migrate, thereby providing durability for the effect. The mean follow-up of patients in that study was 7.2 months and 70% of patients were off medications at last follow-up. The procedure has not been approved in the US, and wide-application and long-term data are lacking [65].

Enteryx™   

Deep implantation of ethinyl-vinyl alcohol (Enteryx™, Enteric Medical Technologies, Palo Alto, CA), an inert, biocompatible polymer, through a sclerotherapy needle under fluoroscopy has been recently studied. Seventy-five patients received the biopolymer injection under fluoroscopic guidance to confirm intramuscular injection. The procedure lasted about 1 h and there were no serious complications other than chest pain for up to 1 week following the procedure [66] (Fig. 28).

Expandable hydrogel prosthesis (Gatekeeper™)   

The procedure involves submucosal placement of an expandable miniature hydrogen prosthesis in the region of the gastroesophageal junction under direct endoscopic visualization. After placement, the prosthesis swells up and bulks the LES region (Medtronics, Minneapolis, MN) [67]. A pilot study with nine patients reported significant improvement of GERD health-related quality of life scores at 1 month. Proper positioning of the prosthesis was confirmed by endoscopic ultrasound in eight of the patients. Minimal complications were reported. Long-term data are awaited.

The placement of the prostheses in the submucosa of the esophagus can be performed under conscious sedation. After endoscopic inspection of the esophagus, stomach, and duodenum a guidewire is placed in the duodenum.

The Gatekeeper™ System uses a 16 mm over-tube as a conduit for the endoscope and for the 2.4 mm diameter hydrogel delivery system. Distally in the over-tube a shelf is present in a rigid segment. The endoscope (contained within the lumen of the over-tube) and over-tube are passed into the lower esophagus over a guidewire.

Once in position, suction is applied via the endoscope, which pulls the esophageal wall into a shelf at the end of the over-tube. A conventional injection needle is passed through a second channel in the over-tube and saline is injected into the submucosal tissue, creating a tissue bleb or submucosal pocket. The injection needle is removed and the prosthesis delivery system (1 mm diameter needle, dilator and 2.4 mm diameter trocar) is passed through the same channel and advanced into the tissue bleb.

The needle assembly and dilator are then removed, leaving the sheath in the submucosal plane. A dry 1.5 mm diameter by 18 mm long hydrogel rod is then loaded into the proximal end of the sheath and advanced with a push rod through the sheath into the submucosa. Once the hydrogel has been delivered the push rod and sheath are removed. Finally, the sheath can be pulled out of the submucosa and the fold is removed from the chamber by inflating some air, and slightly twisting the over-tube. Axial rotation or further insertion or removal of the over-tube repositions the shelf to place additional prostheses.

The complete procedure takes approximately 15 min for one hydrogel and 5 min for each additional hydrogel. Regardless of the number of deliveries only one pass of the over-tube and endoscope is required. Within 24 h the prostheses expand to their full state, creating submucosal swellings in the esophageal wall, thus creating a mechanical anti-reflux barrier (Fig. 29).

Endoscopic plication systems   

Endoluminal gastric plication (ELGP)   

This is an endoscopic suturing method introduced by CR Bard, Inc. (Murray Hill, NJ) under the name Endocinch. The procedure requires two endoscopes and an over-tube [68]. The first endoscope carries the metal sewing capsule on its tip. The second endoscope cinches the sutures through a catheter device that deploys a ceramic plug and ring through which the sutures are threaded. A short, 18 mm OD over-tube allows for repeated (~10) intubations.

Two to three plications are usually performed either longitudinally (one above the other), radially (next to each other), or spirally at the lesser curvature side of the cardia within 1 cm below the squamocolumnar junction. Each plication is formed by two sutures which are placed into the gastric submucosa, about 1.5 cm apart, and then pulled together.

After correct placement of the sewing capsule, suction is applied, the gastric wall is pulled into the hollow chamber of the capsule, and a straight needle loaded with non-absorbable prolene 3-0 suture is fired through the suctioned tissue. After placement of the second suture, the two stitches are pulled together and cinched by the ceramic plug and ring using the second endoscope. Depending on anatomy and technical expertise, the procedure takes 40–60 min [69] (Fig. 30).

The procedure intends to alter the anatomy of the gastric cardia and the angle of His by tightening the lax open end of the cardia within the lesser curve accentuating the angle of His. The majority of the stitches are placed submucosally, not transmurally, just below the squamocolumnar junction. Complications may include sore throat, minor bleeding, transient dysphagia, and mucosal perforation [70].

NDO   

Currently undergoing human clinical trials and not yet FDA-approved, this device (NDO Surgical Inc., Mansfield, MA) uses a small caliber endoscope to visually guide the plication device [71]. Both are then retroflexed to view the cardia region at the greater curvature side, just 1 cm below the squamocolumnar junction. After such positioning, a set of jaws is opened and a catheter with a corkscrew tip is advanced and screwed into the muscularis propria of that area bringing the gastric wall into the span of the jaws.

Upon closure of the jaws, one large plication is accomplished using a double-pronged rivet-like implant opposing two full thickness portions of the cardia wall (Fig. 31). Over time, the serosal sides of the gastric wall fuse and tighten the cardia, and accentuate the angle of His.

The procedure was introduced at DDW 2001 and its efficacy with six patients in India was reported. All plications were successful and there were no complications. These patients had 80% reduction of GERD scores at 3 months, five out of six were completely off medications, and pH scores were improved.

Boston Scientific device   

Using a special device (Boston Scientific/Microvasive, Natick, MA) the distal esophagus and proximal fundus are grasped, folded, and remodeled to create a valvular structure comparable to that of a Nissen fundoplication. A small caliber gastroscope is advanced through the lumen of the grasping device so that the procedure can be performed under endoscopic visualization. Bioabsorbable clips are used to fasten the esophageal and gastric tissue together. Clinical experience thus far is limited to adult baboons but the method appears promising [72].

 Outstanding issues and future trends   

GERD is a Western world pandemic and will remain a major health problem in the near future as the developing world modernizes. Although of lesser impact on morbidity and mortality, GERD will continue to chronically compromise quality of life and thus will contribute to large health-care costs. An increasingly sophisticated set of tools and devices will be available that will allow easier and more accurate diagnosis and definition of disease severity.

Endoscopic enhancement technologies (chromoendoscopy, magnification endoscopy, etc.) will allow easier, more specific recognition of inflammatory, metaplastic, or neoplastic changes of the esophageal mucosa. In vivo functional measurements (such as intracellular pH, cell permeability, cell integrity) and optical measurements of the depth and nature of injury will provide an additional dimension of the dysfunction associated with a given endoscopic appearance of disease.

Barrett's esophagus will be detected more and more frequently as transnasal endoscopy enters daily office practice as a screening tool, and surveillance practices will be better defined and provide maximum benefits with minimization of costs.

Peptic strictures will probably vanish because of early therapy. Pharmacological therapy will continue aiming at complete symptom control in GERD, but it will carry with it the need for chronic medication use.

Quick-fix approaches, such as endoscopic therapy, fundoplication, or laparoendoscopic approaches will continue to evolve and will play a major role for this chronic disease.

All this progress will require an increasingly frequent exchange between gastroenterologists and surgeons, who will work together in centres of excellence to care for such patients.

Disclosure   

Dr Triadafilopoulos is a consultant to almost all the major pharmaceutical companies that are active in the area of gastroesophageal reflux. He has received funding for studies, seminars, and travel from such companies and has an equity position in Curon Medical Inc.

 References  

 1 Johnsson, F, Joelsson, B, Gudmundsson, K & Greiff, L. Symptoms and endoscopic findings in the diagnosis of gastroesophageal reflux disease. Scand J Gastroenterol 1987; 22: 714–18. PubMed

 2 Patti, MG, Diener, U, Tamburini, A, Molena, D & Way, LW. Role of esophageal function tests in diagnosis of gastroesophageal reflux disease. Dig Dis Sci 2001; 46: 597–602. PubMed

 3 Savary, M & Miller, G. (1978) The esophagus: handbook and atlas of endoscopy. Gussman, Solothurn, Switzerland.

 4 Hetzel, DJ, Dent, J, Reed, WD, Narielvala, FM, McKinnon, M & McCarthy, JH et al. Healing and relapse of severe peptic esophagitis after treatment with omeprazole. Gastroenterology 1988; 95: 903–12. PubMed

 5 Armstrong, D, Bennett, JR, Blum, AL, Dent, J, DeDombal, FT & Galmiche, JP et al. The endoscopic assessment of esophagitis: a progress report on observer agreement. Gastroenterology 1996; 111: 85–92. PubMed

 6 Monnier, P & Savary, M. Contribution of endoscopy to gastroesophageal reflux disease. Scand J Gastroenterol 1984; 19 (Suppl. 106): 26–31.

 7 Wong, RK, Hanson, DG, Waring, PJ & Shaw, G. ENT manifestations of gastroesophageal reflux. Am J Gastroenterol 2000; 95 (8 Suppl.): S15–22. PubMed

 8 DeVault, KR. Overview of therapy for the extraesophageal manifestations of gastroesophageal reflux disease. Am J Gastroenterol 2000; 95 (8 Suppl.): S39–44. PubMed

 9 Fennerty, MB. Extraesophageal gastroesophageal reflux disease: presentations and approach to treatment. Gastroenterol Clin North Am 1999; 28 (4): 861–73. PubMed

10 Cameron, AJ, Malcolm, A, Prather, CM & Phillips, SF. Video-endoscopic diagnosis of esophageal motility disorders. Gastrointest Endosc 1999; 49: 62–9. PubMed

11 Vadva, M & Triadafilopoulos, G. Glycogenic acanthosis of the esophagus and gastroesophageal reflux. J Clin Gastroenterol 1993; 17: 79–83. PubMed

12 Armstrong, D, Blum, AL & Savary, M. Reflux disease and Barrett's esophagus. Endoscopy 1992; 24: 9–17. PubMed

13 Kahrilas, PJ. The role of hiatus hernia in GERD. Yale J Biol Med 1999; 72: 101–11. PubMed

14 Lagergren, J, Bergstrom, R, Lindgren, A & Nyren, O. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 1999; 340: 825–31. PubMed

15 Marks, RD & Richter, JE. Peptic strictures of the esophagus. Am J Gastroenterol 1993; 88: 1160–73. PubMed

16 Atkinson, M & Robertson, CS. Benign esophageal strictures in Barrett's columnar epithelialized oesophagus and its responsiveness to conservative management. Gut 1988; 29: 1721–4. PubMed

17 Dakkak, M, Hoare, RC, Maslin, SC & Bennett, JR. Oesophagitis is as important as oesophageal stricture diameter in determining dysphagia. Gut 1993; 34: 152–5. PubMed

18 Tulman, AB & Boyce, HW Jr Complications of esophageal dilation and guidelines for their prevention. Gastrointest Endosc 1981; 27: 229–34. PubMed

19 Mittal, RK. Hiatal hernia and gastroesophageal reflux: another attempt to resolve controversy. Gastroenterology 1993; 105: 941–2. PubMed

20 Sloan, S, Rademaker, AW & Kahrilas, PJ. Determinants of gastroesophageal junction incompetence: hiatal hernia, lower esophageal sphincter, or both? Ann Intern Med 1992; 117: 977–82. PubMed

21 Nasi, A, Moraes Filho, PP, Zilberstein, B, Cecconello, I, Gama-Rodrigues, JJ & Pinotti, HW. Gastroesophageal reflux disease: clinical, endoscopic, and intraluminal esophageal pH monitoring evaluation. Dis Esophagus 2001; 14: 41–9. PubMed

22 Ott, DJ, Gelfand, DW, Wu, WC & Castell, DO. Esophagogastric junction and its rings. Am. J. Roentgenol 1984; 142–281.

23 Marshall, JB, Kretschmer, JM & Diaz-Arias, AA. Gastroesophageal reflux as a pathogenic factor in the development of symptomatic lower esophageal rings. Arch Int Med 1990; 150: 1669–72.

24 Mallory, GK & Weiss, S. Hemorrhages from lacerations of the cardiac orifice of the stomach due to vomiting. Am J Med Sci 1929; 178: 506–15.

25 Cameron, AJ, Malcolm, A, Prather, CM & Phillips, SF. Videoendoscopic diagnosis of esophageal motility disorders. Gastrointest Endosc 1999; 49: 62–9. PubMed

26 Zamost, BJ, Hirschberg, J, Ippoliti, AF, Furst, DE, Clements, PJ & Weinstein, WM. Esophagitis in scleroderma: prevalence and risk factors. Gastroenterology 1987; 92: 421–8. PubMed

27 Wu, JM, Trus, TL, Richardson, WS, Hunter, JG, Branum, GD & Mauren, SJ et al. Evaluation and management of post-fundoplication dysphagia. Am J Gastroenterol 1996; 91: 2318–22. PubMed

28 Hinder, RA, Perdikis, G, Klinger, PJ & DeVault, KR. Management of the failed anti-reflux operation. Am J Med 1997; 103: S144–8.

29 Bowrey, DJ, Williams, GT & Clark, GW. Histological changes in the oesophageal squamous mucosa: correlation with ambulatory 24 hour pH monitoring. J Clin Pathol 2003; 56: 205–8. PubMed

30 Connor, MJ & Sharma, P. Chromoendoscopy and magnification endoscopy in Barrett's esophagus. Gastrointest Endosc Clin N Am 2003; 13: 269–77. PubMed

31 Inoue, H, Rey, JF & Lightdale, C. Lugol chromoendoscopy for esophageal squamous cell cancer. Endoscopy 2001; 33: 75–9. PubMed

32 Owen, W. ABC of the upper gastrointestinal tract: dysphagia. BMJ 2001; 323: 850–3. PubMed

33 Hands, LJ, Dennison, AR, Papavramidis, S, McIntyre, RL, Bishop, H & Kettlewell, MGW. The natural history of peptic esophageal strictures treated by dilatation and antireflux therapy alone. Ann R Coll Surg Eng 1989; 71: 36–09.

34 McClave, SA, Brady, PG, Wright, RA, Goldschmidt, S & Minocha, A. Does fluoroscopy guidance for Maloney esophageal dilation impact on the clinical endpoint of therapy: relief of dysphagia and achievement of luminal patency. Gastrointest Endosc 1996; 43: 93–7. PubMed

35 Bladergroen, MR, Lowe, JE & Postlethwait, RW. Diagnosis and recommended management of esophageal perforation and rupture. Ann Thorac Surg 1986; 42: 235–9. PubMed

36 Kozarek, RA, Patterson, DJ, Ball, TJ, Gelfand, MG, Jiranek, GE & Bredfeldt, JE et al. Esophageal dilation can be done safely using selective fluoroscopy and single dilating sessions. J Clin Gastroenterol 1995; 20: 184–8. PubMed

37 Ho, SB, Cass, O, Katsman, RJ, Lipschultz, EM, Metzger, RJ, Onstad, GR & Silvis, SE. Fluoroscopy is not necessary for Maloney dilation of chronic esophageal strictures. Gastrointest Endosc 1995; 41: 11–14. PubMed

38 Kozarek, RA. To stretch or to shear: a perspective on balloon dilators. Gastrointest Endosc 1987; 33: 459–61. PubMed

39 Shemesh, E & Czerniak, A. Comparison between Savary-Gilliard and balloon dilatation of benign esophageal strictures. World J Surg 1990; 14: 518–21. PubMed

40 Cox, JG, Winter, RK, Maslin, SC, Dakkak, M, Jones, R & Buckton, GK et al. Balloon or bougie for dilatation of benign esophageal stricture? Dig Dis Sci 1994; 39: 776–81. PubMed

41 Pereira-Lima, JC, Ramires, RP, Zamin, I Jr, Cassal, AP, Marroni, CA & Mattos, AA. Endoscopic dilation of benign esophageal strictures: report on 1043 procedures. Am J Gastroenterol 1999; 94: 1497–501. PubMed

42 Saeed, ZA. Balloon dilatation of benign esophageal stenosis. Hepatogastroenterology 1992; 39: 490–3. PubMed

43 Saeed, ZA, Winchester, CB, Ferro, PS, Michaletz, PA, Schwartz, JT & Graham, DY. Prospective randomized comparison of polyvinyl bougies and through-the-scope balloons for dilation of peptic strictures of the esophagus. Gastrointest Endosc 1995; 41: 189–95. PubMed

44 Scolapio, JS, Pasha, TM, Gostout, CJ, Mahoney, DW, Zinsmeister, AR, Ott, BJ & Lindor, KD. A randomized prospective study comparing rigid to balloon dilators for benign esophageal strictures and rings. Gastrointest Endosc 1999; 50: 13–17. PubMed

45 Saeed, ZA, Ramirez, FC, Hepps, KS, Cole, RA, Schneider, FE, Ferro, PS & Graham, DY. An objective end point for dilation improves outcome of peptic esophageal strictures: a prospective randomized trial. Gastrointest Endosc 1997; 45: 354–9. PubMed

46 Patterson, DJ, Graham, DY, Smith, JL, Schwartz, JT, Alpert, E, Lanza, FL & Cain, GD. Natural history of benign esophageal stricture treated by dilatation. Gastroenterology 1983; 85: 346–50. PubMed

47 Guelrud, M, Villasmil, L & Mendez, R. Late results in patients with Schatzki ring treated by endoscopic electrosurgical incision of the ring. Gastrointest Endosc 1987; 33: 96–8. PubMed

48 Smith, PM, Kerr, HD, Cockel, R, Ross, BA, Bate, CM & Brown, P et al. and Restore Investigator Group. A comparison of omeprazole and ranitidine in the prevention of recurrence of benign esophageal stricture. Gastroenterology 1994; 107: 1312–18. PubMed

49 Swarbrick, ET, Gough, AL, Foster, CS, Christian, J, Garrett, AD & Langworthy, CH. Prevention of recurrence of esophageal stricture: a comparison of lansoprazole and high dose ranitidine. Eur J Gastroenterol Hepatol 1996; 8: 431–8. PubMed

50 Zein, NN, Greseth, JM & Perrault, J. Endoscopic intralesional steroid injections in the management of refractory esophageal strictures. Gastrointest Endosc 1995; 41: 596–8. PubMed

51 Lee, M, Kubik, CM, Polhamus, CD, Brady, CE 3rd & Kadakia, SC. Preliminary experience with endoscopic intralesional steroid injection therapy for refractory upper gastrointestinal strictures. Gastrointest Endosc 1995; 41: 598–601. PubMed

52 Panzini, L, Burrell, MI & Traube, M. Instrumental esophageal perforation: chest film findings. Am J Gastroenterol 1994; 89: 367–70. PubMed

53 White, CS, Templeton, PA & Attar, S. Esophageal perforation: CT findings. AJR Am J Roentgenol 1993; 160: 767–70. PubMed

54 Botoman, VA & Surawicz, CA. Bacteremia with gastrointestinal procedures. Gastrointest Endosc 1986; 32: 342–4. PubMed

55 Fennerty, MB. Endoscopic therapy for gastroesophageal reflux disease: what have we learned and what needs to be done? Gastrointest Endosc Clin N Am 2003; 13: 201–9. PubMed

56 Hogan, WJ. Endoscopic treatment modalities for GERD: technologies score or scare? Gastrointest Endosc 2001; 53: 541–5. PubMed

57 Lehman, G. Endoscopic and endoluminal techniques for control of gastroesophageal reflux: are they ready for widespread clinical application? Gastrointest Endosc 2001; 52: 808–11.

58 Triadafilopoulos, G, DiBaise, JK & Nostrant, TT et al. Radiofrequency energy delivery to the gastroesophageal junction for the treatment of GERD. Gastrointest Endosc 2001; 53: 407–15. PubMed

59 Utley, DS, Kim, MS, Vierra, MA & Triadafilopoulos, G. Augmentation of lower esophageal sphincter pressure and gastric yield pressure after radiofrequency energy delivery to the gastroesophageal junction: a porcine model. Gastrointest Endosc 2000; 52 (1): 81–6. PubMed

60 Triadafilopoulos, G, Dibaise, JK & Nostrant, TT et al. The Stretta procedure for the treatment of GERD: 6 and 12 month follow-up of the U.S. open label trial. Gastrointest Endosc 2002; 55: 149–56. PubMed

61 Utley, D. The Stretta procedure: device, technique, and pre-clinical study data. Gastrointest Endosc Clin N Am 2003; 13: 135–45. PubMed

62 Triadafilopoulos, G. Clinical experience with the Stretta procedure. Gastrointest Endosc Clin N Am 2003; 13: 147–55. PubMed

63 Lehman, GA. The history and future of implantation therapy for gastroesophageal reflux disease. Gastrointest Endosc Clin N Am 2003; 13: 157–65. PubMed

64 O'Connor, KW & Lehman, GA. Endoscopic placement of collagen at the lower esophageal sphincter to inhibit gastroesophageal reflux: a pilot study of 10 medically intractable patients. Gastrointest Endosc 1988; 34: 106–12. PubMed

65 Feretis, C, Benakis, P & Dimopoulos, C et al. Plexiglass (polymethylmethacrylate) implantation: technique, pre-clinical and clinical experience. Gastrointest Endosc Clin N Am 2003; 13: 167–78. PubMed

66 Louis, H & Deviere, J. Endoscopic implantation of Enteryx for the treatment of gastroesophageal reflux disease: technique, pre-clinical and clinical experience. Gastrointest Endosc Clin N Am 2003; 13: 191–200. PubMed

67 Fockens, P. Gatekeeper reflux repair system: technique, pre-clinical and clinical experience. Gastrointest Endosc Clin N Am 2003; 13: 179–89. PubMed

68 Swain, P, Park, P-O & Mills, T. Bard EndoCinch: the device, the technique, and pre-clinical studies. Gastrointest Endosc Clin N Am 2003; 13: 75–88. PubMed

69 Filipi, CJ, Lehman, GA & Rothstein, RI et al. Transoral, flexible endoscopic suturing for treatment of GERD. A multicenter trial. Gastrointest Endosc 2001; 53: 416–22. PubMed

70 Rothstein, RI & Filipi, CJ. Endoscopic suturing for gastro-esophageal reflux disease: clinical outcome with the Bard EndoCinch. Gastrointest Endosc Clin N Am 2003; 13: 89–101. PubMed

71 Chuttami, R. Endoscopic full-thickness placation: the device, technique, pre-clinical and early clinical experience. Gastrointest Endosc Clin N Am 2003; 13: 109–16. PubMed

72 DeMeester, TR. Microvasive gastric stapler: the device, technique and pre-clinical results. Gastrointest Clin N Am 2003; 13: 117–33.

Copyright © Blackwell Publishing, 2004