123 - Anatomy of the Esophagus

Editors: Shields, Thomas W.; LoCicero, Joseph; Ponn, Ronald B.; Rusch, Valerie W.

Title: General Thoracic Surgery, 6th Edition

Copyright 2005 Lippincott Williams & Wilkins

> Table of Contents > Volume II > The Esophagus > Section XXIII - Benign Esophageal Disease > Chapter 143 - Surgical Therapy for Gastroesophageal Reflux Disease

Chapter 143

Surgical Therapy for Gastroesophageal Reflux Disease

Mark J. Krasna

Gastroesophageal reflux (GER) describes the retrograde flow of gastric material back into the esophageal lumen from a site distal to the lower esophageal sphincter (LES). It is a physiologically normal occurrence in humans. The refluxate can contain bile salts and pancreatic enzymes, but its more noxious and typical constituents are hydrochloric acid and pepsin produced by the stomach. Depending on the volume of refluxed contents, the time interval between reflux episodes, and the duration that refluxed agents remain in contact with the esophageal mucosa, undesirable changes can ensue. These phenomena make up the principal pathophysiologic conditions that lead to gastroesophageal reflux disease (GERD). It is recognized, however, that the diagnosis of GERD connotes more than aberrant physiology. Here, the word disease refers to a much larger entity, including the various mechanisms for its development, associated anatomic abnormalities, a full array of symptoms, and a spectrum of potential tissue injuries.

Realizing the scope of GERD provides the necessary context in which to place the seeming exponential rise in diagnosis. The prevalence of GERD and its salient manifestations has achieved near-epidemic proportions. A 1998 Gallup survey disclosed that 44% of adult Americans experience heartburn at least once every month. According to Nebel and colleagues in 1976, 14% of American adults experienced heartburn weekly, and a population-based study by Locke and associates in 1997 revealed a weekly heartburn incidence of about 20%. Estimates of daily heartburn approach the one-tenth mark (U.S. population). The lay press has reported that Americans spent $5 billion dollars on antireflux medications in 1995; more current data are unavailable, but expenditures in excess of this figure are likely given the present state of the pharmaceutical industry and overall health care costs. Equally impressive is the 10% to 20% incidence of serious complications, including esophageal stricture formation and Barrett's esophagus, in patients seeking medical attention for GERD-related symptoms. Complementing these statistics is the documented fourfold increase in esophageal adenocarcinoma over the past 20 years. As described by Lagergren and associates (1999) and others, the causal link between symptomatic GER and this malignancy is certainly evident. But heartburn is not synonymous with GERD. A documented hiatal hernia is not either. And although clinical improvement, even complete symptomatic relief with proton pump inhibitor (PPI) therapy, has a strong positive predictive value relative to the GERD diagnosis, response to treatment cannot be used as the sole criterion on which to make this diagnosis.

PHYSIOLOGY AND PATHOLOGY

Three basic defense systems exist to prevent pathologic GER and minimize the effects of physiologic reflux. They are anatomic reflux barriers, esophageal luminal clearance mechanisms, and intrinsic tissue resistance to esophageal mucosal damage. Orlando (2002) has suggested that these defenses are hierarchically arranged according to their significance and potential impact. Anatomic barriers are the first line of defense and physically limit both reflux episode frequency and material amount. The second tier of defenses acts to reduce the duration of contact between the deleterious components of the refluxate and the esophageal mucosa. The third level of defense is more theoretical and pertains to the esophageal epithelium itself. It has been proposed that histochemical and cytoprotective elements exist that can curtail acid or other toxin-mediated tissue destruction.

The most critical anatomic reflux barrier is the LES. Esophageal contraction waves are noted to slow immediately above the LES segment. In primates, the LES is not a true sphincter, but a region of relative high pressure in the distal esophagus at or just above the esophagogastric junction (EGJ). In adults, it measures 3 to 5 cm in length. A circumscribed

P.2174

area of thickened smooth muscle in this location can be found, but a distinct valvelike entity, such as at the pylorus or anus, is not identifiable. In normal individuals, manometry demonstrates pressures between 15 and 30 mm Hg within the LES segment, separating lower-pressure zones above and below. Above the LES, the esophageal pressure is normally 5 mm Hg less than atmospheric pressure; below it (gastric), 5 mm Hg greater than atmospheric pressure. Fundamentally, LES integrity depends on: (a) its overall length (dP/dT), (b) its external pressure environment relative to body compartment (intraabdominal positive pressure versus intrathoracic negative pressure), and (c) its abdominal length, a synthesis of these characteristics.

The LES remains contracted at rest owing to intrinsic myogenic activity, and it normally relaxes with the approach of a primary peristaltic wave. Once food passes through the LES, the LES returns to its contracted state, achieving a closing pressure that is about twice its resting value. Lower esophageal pressure relaxation also occurs with secondary peristalsis. Holloway and co-workers (1985) have shown that postprandial gastric distention leads to transient LES relaxation, whereby gas is vented from the stomach. According to Mittal and coinvestigators (1995), physiologic and pathologic GER occurs principally during these vagally mediated and non swallow-induced relaxations. Longer reflux events and increased episode frequency dominate in the development of GERD. Ultimately, Dodds and colleagues (1982) have ascribed these GERD-inherent alterations to: (a) attenuated LES tone, (b) intraabdominal pressure spikes that overcome the LES (stress reflux), or (c) increased transient LES relaxations. This mechanistic account, however, considers little beyond sphincter impairment in the multifactorial etiology of GERD.

Equally integral to the barrier defenses are the other components of the diaphragmatic apparatus. Encircled by the right crus of the diaphragm, the LES is mechanically buttressed. On inspiration and other states of increased intraabdominal pressure, the hiatal crural complex contracts, amplifying LES closure strength. Also, the pincerlike arrangement of the crural fibers creates an oblique entry for the esophagus as it meets the stomach. This configuration, termed the angle of His, augments LES pressure more indirectly. Because of this acute angle, a distended stomach's walls compress the esophagus, reciprocally applying force to the LES from outside the esophageal wall. The phrenoesophageal ligament provides further scaffoldlike support to the LES. This membranous continuation of the endothoracic fascia and peritoneum circumferentially envelops the distal esophagus within its diaphragmatic canal. It anchors the distal esophagus in the abdomen, maximizing LES abdominal length. Finally, the redundant nature of the mucosal folds within the gastric cardia acts like a sponge and soaks up the refluxate, limiting its total overall volume.

In addition to an anatomic barrier defense, the esophagus defends against caustic refluxed agents by clearing them from its lumen. The bulk of refluxate volume is cleared by a combination of gravity and esophageal peristalsis. In terms of gravity, a patient's posture, whether upright or supine, warrants consideration. In nocturnal reflux, the effects of gravity's force are difficult to realize; while recumbent, little benefit is derived. Similarly, the patient who remains upright following meals may minimize the consequences of reflex-induced LES relaxations. More essential to effective luminal clearance is normal esophageal motility and preserved motor function. An intact peristaltic pump clears the esophagus of refluxed irritants much like a coordinated peristaltic wave propels a food bolus from the cervical esophagus through to the stomach. Although the propulsive contractions following deglutition are initiated quite differently, reactive peristalsis that empties a reflux-filled esophagus ultimately relies on the same end-network of striated and smooth muscle. Proper esophageal body motor activity is critical, therefore, for clearance by peristalsis to succeed. Failed peristalsis and diminished muscular strength, moreover, can arise secondary to reflux-mediated injury. A vicious, perpetual cycle includes mucosal inflammation leading to submucosal and mural thickening, leading to loss of myogenic tone, leading to compromised luminal clearance, leading to stasis of refluxed material, leading to a reinvigoration of this cascade through further tissue damage. Furthermore, Kahrilas and Shi (2000) and others have demonstrated GERD-associated motor disturbances. Patients with GERD have lower-amplitude peristaltic contractions, more frequent aperistaltic contractions, and more overall failed contractions with swallowing than expected in nonpathologic GER.

Salivary and other esophageal gland secretions provide further means of clearance through chemical neutralization. Although produced in quantities nearing 1 L per day, these glandular fluids do not mechanically wash out or dilute refluxed matter to any significant degree. Rather, their bicarbonate-rich composition enables neutralization of refluxed stomach acid, according to Helm and co-workers (1984). While contributing minimally to bolus clearance, these secretions are factors in the effective clearance of luminal acid.

The esophageal mucosal inherent resistance to tissue destruction constitutes the final line of defense against GER. Although defects in physical barrier defenses have been emphasized historically, it is now increasingly evident that the regressive slide from GER to GERD can occur without failures in anatomic structure. Monitoring of esophageal pH by Schlesinger and colleagues (1985) has revealed that normal acid contact times can be found in a significant proportion of patients with histologically proven erosive and nonerosive esophagitis. In healthy subjects, Schindlbeck and associates (1987) have shown daily cumulative acid contact times of 1 to 2 hours. Explanations for esophagitis in the setting of normal acid contact times are limited. Theorized increased rates of gastric acid and pepsin production in GERD patients have been disproved. The alternative explanation offered by Orlando (2002) and other investigators

P.2175

is that impaired tissue resistance accounts for reflux-associated disease and normal acid contact times. A multidimensional schema has been established, detailing cellular and biochemical protection at the level of the esophageal endothelium. Cell membrane and intercellular junctions limiting the diffusion of HCl are being characterized, proton-blocking glycoconjugates have been isolated, and tissue-specific buffering machinery with upregulation of epithelial carbonic anhydrase and transmembrane Na/H and Cl/HCO₃ exchangers has been found. Susceptibility to breaks in tissue resistance has been shown to be greater in cigarette smokers, alcohol users, and consumers of hot or hypertonic solutions, and with advancing age.

Although independent reference is frequently made to the causal relationship between hiatal hernia and GERD, such separate discussion is unwarranted. As Hiebert (1991) has observed, at least half of patients with hiatal hernia do not have symptomatic GER, and not every patient with severe GERD has a concomitant hiatal hernia. The altered anatomy that defines hiatal hernia also defines critical lapses in antireflux barrier and luminal clearance defenses. The functional hiatal crural unit disassembles. Lost is the pincer grasp by the right crus on the region of the LES. The angle of His becomes flattened and obtuse, minimizing the counterpressure phenomenon that occurs with gastric distention. In the most common forms of hiatal hernia, the gastroesophageal junction is displaced into the chest, whereby the antireflux positive pressure in the abdomen is replaced with the proreflux negative pressure characteristics in the thorax. The herniated gastric portion becomes functionally obstructed at its distal extent because of diaphragmatic impingement against the stomach's body. This retained pouch becomes a reservoir for secretions and swallowed air, effectively maintaining a source of refluxate volumes from which individual GER episodes are essentially guaranteed. Luminal clearance is also affected because of esophageal foreshortening in a long-standing hiatal hernia. Esophageal stripping waves become weakened and attenuated in hiatal hernia with degradation of coordinated peristalsis. The herniated portion of the stomach fails to empty completely or clear thoroughly. Interestingly, Sloan and Kahrilas (1991) have shown that fundoplication, even outside the context of hiatal hernia, improves gastric emptying through sacrifice of the fundic reservoir. Finally, the laxity of the mucosal folds of the cardia is diminished in hiatal hernia, relinquishing trapped gastric juice for its potential increased backflow across the LES.

PATIENT SELECTION AND PREOPERATIVE EVALUATION

Clinical manifestations of GERD are divided into two subsets. Patients are grouped according to their presenting symptoms: typical or esophageal GERD, as opposed to atypical or extraesophageal GERD. Typical GERD symptoms include heartburn, regurgitation, and dysphagia. Heartburn is classically characterized as a burning sensation in the epigastrium or substernal area. Patients with regurgitation complain of a sudden bitter or acid taste that develops following eating, commonly at night once supine in bed. Dysphagia is best recognized as a complication of GERD in the setting of an inflammation-induced esophageal stricture or problems with esophageal peristalsis. Atypical symptoms include sore throat, cough, nonallergic asthma with wheezing, dysphonia, and chest pain. These extraesophageal difficulties likely relate to reflux that damages the laryngopharyngeal apparatus and, as suggested by Wetscher and colleagues (1997) and other investigators, affects the respiratory tract in a complex and only preliminarily understood manner.

Distinguishing between these subpopulations is more essential than perhaps first realized. The correlation between typical symptoms and increased esophageal acid exposure is certainly stronger than the pathophysiologic link between atypical symptoms and acid reflux. More importantly, atypical symptoms are much less reliably relieved through antireflux surgery than their typical counterparts. The explanation for this difference is likely multifactorial. Overall, the chronicity of the injury pattern may be the most important factor. Extraesophageal manifestations of GERD likely develop gradually and consequently may prove less reversible. Acutely reducing or even removing the insult completely with fundoplication may have little impact on the more permanent changes that have occurred in reflux-related respiratory disease.

Determining that GER is the underlying cause of a patient's symptoms is not only the most critical element in proper patient selection, it also is central to the diagnostic testing and evaluation of the patient considered for antireflux surgery. Every patient being considered for antireflux surgery undergoes preoperative testing. Although response to PPI therapy has definite clinical implications, amelioration of symptoms with PPI does not confirm the diagnosis of GERD. All of these preoperative studies are designed to prove a correlation between symptoms and reflux-mediated disease. Again, establishing this causal relationship, not merely its association, maximizes the chances before the operating room that antireflux surgery will help the patient. This concept is even more important in the laparoscopic era, in which the threshold for surgical referral is markedly lower. Complementing this increased tendency for referral is the knowledge that the GERD patient now presents sooner with an earlier form of the disease, and rarely with any significant degree of disease progression. Of course, meticulous intraoperative technique and judgment, followed by detailed postoperative care and instruction, are integral to a successful surgical outcome. It is this preoperative attitude and assessment, however, that deserves the most emphasis and demands the most skill. At minimum, three preoperative tests are routinely undertaken on all patients: (a) 24-hour continuous esophageal pH monitoring,

P.2176

(b) esophageal manometry, and (c) anatomic imaging by barium swallow, esophagogastroscopy, or both.

Monitoring of pH objectively qualifies the existence of pathologic esophageal acid exposure. A summary 24-hour pH score is computed based on the number of reflux episodes per day, those episodes lasting longer than 5 minutes, and the total time that esophageal pH drops below 4.0. The patient-activated event marker, or a record tracking patient complaints alongside continuous pH measurements, also supports reflux symptom correlation. Realizing that most patients with typical symptoms, especially in the setting of confirmed erosive esophagitis or clear improvement with PPI therapy, will have an abnormal 24-hour pH score, pH testing also quantifies the degree and pattern of reflux disease. As characterized by Fein (1996) and Fein and coinvestigators (1997), failure of medical therapy is more common, and GERD-related strictures, ulcers, and other complications are higher with nocturnal or bipositional reflux. Furthermore, it has been established that a structurally normal LES may function abnormally. A characteristic pattern of reflux is seen in this situation, namely multiple short reflux episodes following meals and while upright. Interpretation of these patterns may provide added insight into the GERD syndrome beyond an abnormal 24-hour score.

The LES and intrinsic esophageal tone (effective luminal clearance) constitute the chief anatomic defenses limiting the duration and potential impact of refluxed gastric juice. Esophageal manometry provides information relevant to the status of these barriers. Pressure transducers positioned at standard intervals along the length of the esophagus record in real time the propagation of a peristaltic wave with swallowing. A significant correlation between esophageal motor dysfunction and GERD has been established by Rakic and associates (1997), among others. Absent or disordered peristalsis can significantly complicate the surgical treatment of GERD. Undetected poor peristaltic function not only can undermine the relief potentially garnered through fundoplication but also can frequently exacerbate a patient's overall difficulties. LES pressure less than 10 mm Hg anticipates the theoretical salutary effect that reinforcement of the LES should have. LES incompetence, however, may have little influence on GERD or antireflux surgery. Many believe that, in GERD, normal transient receptive relaxation of the LES occurs too frequently and independent of an initiated peristaltic wave. Others contend that GERD arises because of stress reflux in which sudden spikes in intraabdominal pressure force acid back across the sphincter irrespective of its integrity. Regardless, as defined by Costantini and colleagues (1996), the association between low LES tone and medical nonresponders has important implications for GERD therapy in general and surgical therapy in particular. Manometry also assists in localizing the LES, a key factor in determining overall esophageal length. Esophageal manometry is not essential, therefore, to document a lax LES; manometry principally identifies insufficient esophageal propulsive function that can be caused by long-standing reflux. Preserved peristaltic pressure is needed to propel a bolus of food through the fixed and static obstruction caused by the fundoplication itself; a partial fundoplication may be more appropriate if normal esophageal motility has been lost to any degree.

Barium esophagram and upper endoscopy provide nonfunctional information that also should be integrated in the overall surgical evaluation of the patient with GERD. The knowledge gained through these investigations extends beyond anatomy or road mapping. The possibility of a foreshortened esophagus, a significant cause of reherniation and fundoplication failure, is best suggested by findings evident through imaging. A large hiatal hernia (>5 cm) and esophageal stricture are two findings that can herald the existence of a short esophagus. Although a shortened esophagus can be inferred through the manometric localization of the LES, a large hiatal hernia that fails to reduce in the upright position on video esophagram is considerably more indicative of this phenomenon. The patient with long-standing GERD, whether refractory to medical therapy or not, and certainly patients with GERD-associated dysphagia, should undergo endoscopy with biopsy. Mucosal injury is best detected through direct visualization. Endoscopic biopsy of any mucosal irregularity is indicated and can reveal Barrett's metaplasia that needs to be monitored even after fundoplication. This is particularly noteworthy given the symptomatic improvement that should follow fundoplication; the impetus for routine surveillance may be lost, from both the patient's and physician's standpoint. The discovery of Barrett's dysplasia, even low-grade, warrants careful consideration. It may be imprudent to manipulate the EGJ surgically, even minimally, with the prospect of a definitive extirpative procedure in the not-so-distant future.

INDICATIONS AND CONTRAINDICATIONS

Symptomatic and documented GER that persists despite maximal medical therapy (specifically high-dose, usually twice daily, PPI therapy) is recognized as the most common indication for surgical intervention. As DeVault and Castell (1999) have pointed out, however, medical intractability is quite a rare phenomenon in current practice. Before the introduction of PPI, the medical treatment of GERD was suboptimal. Treatment failures with diet modification, weight loss, alkali-based agents, neutralizing antacids, and histamine receptor antagonists (H₂ blockers) were common. Historically, nonresponders represented most of those referred for antireflux procedures, irrespective of approach. GERD-related symptomatology refractory to high-dose PPI, as Spechler (2000) argues, should raise essential doubts concerning the very etiology of these symptoms. In fact, a well-established correlation has been drawn by Campos and co-workers (1999a, 1999b)

P.2177

regarding the effectiveness of PPI therapy and the effectiveness of antireflux surgery: there is a strong positive predictive value between PPI response and fundoplication success. This parallel pattern of therapeutic success should and will change practice patterns eventually. The present candidate pool for antireflux surgery is composed less of PPI failures, but primarily of those patients who fail to comply with PPI treatment. Medication expense and the need to follow closely a frequently inflexible dosing schedule are often cited as the chief causes of noncompliance. Many GERD patients are young and unwilling or unable to adhere to long-term, if not life-long, medical therapy programs.

The medically optimized patient who develops GERD-related esophageal injury or complications is also appropriately referred for surgery. Endoscopically detected severe esophagitis, mucosal ulceration, or stricture formation should all be recognized as disease progression. Recurrent aspiration with chemical pneumonitis and chronic laryngopharyngeal irritation due to reflux are also indications to proceed with antireflux surgery.

Although GERD-induced epithelial injury leading to metaplasia is a reproducible sequence of events, the ability of surgery to interrupt or delay this pathologic cascade remains dubious. Evidence would suggest that once triggered, this degradation continues without the possibility for interruption. Repeatedly, it has been shown that antireflux surgery cannot prevent dysplastic change. Accepting that GERD surgery has little impact on the natural history of Barrett's esophagus, it seems prudent to view Barrett's not only as a controversial indication for antireflux surgery but also in the absence of symptoms or other complications, not an indication for surgical intervention.

SURGICAL STRATEGY

Antireflux surgery has had two main challenges it has needed to overcome in order to be accepted according to Hagen and Peters (2000). As initially performed, the standard fundoplication was a lengthy, tight wrap that overly compensated for a hypotensive LES. The result was dysphagia and gas bloating. Rossetti and Allgower (1973), as well as Negre (1983) and Donahue (1985) and their colleagues, and others all recognized that reinforcement of the LES could be accomplished with a shorter and looser fundoplication, preserving the capacity to swallow normally the majority of food types and to belch. The second hurdle that faced antireflux surgeons was the perioperative morbidity (and mortality) associated with historical antireflux surgery. Hospital stays were typically longer than 1 week with 2-month disabilities from work. Minimal access fundoplication has radically, if not completely, reversed these negative events. Ultimately, although the medical management of GERD is less invasive in design, it has potentially much more complexity in practice. GERD treated medically requires long-term (chronic) therapy, may have unforeseen side effects, and ultimately, may be too one-dimensional in its concept.

Operative therapy for GERD incorporates a multitude of surgical decision-making skills. There is no standard antireflux procedure against which all other procedures can be compared; no single operation can repair every GERD-related abnormality or treat every patient with GERD. Important factors that need to be integrated in a comprehensive surgical plan include: (a) operative approach (transabdominal, transthoracic, laparoscopic, even thoracoscopic); (b) degree of fundic wrapping complete (total) or partial; and (c) adjunctive procedures esophageal lengthening techniques (by gastroplasty), pyloroplasty, and the occasional gastropexy (by gastrostomy tube insertion.).

Every antireflux procedure initiates with the same three basic goals. The principal steps of the operative procedure are ordered as to reflect these essential endpoints. First, thorough esophageal mobilization is necessary to restore at least 2.5 cm (optimally, 3 to 5 cm) of intraabdominal esophageal length. Second, the invariably widened esophageal hiatus must be closed. Third, the lower esophageal high-pressure zone must be reestablished. The fundic wrap recreates an acute angle of His and in turn refashions normal orientation for the gastric sling and clasp fibers, foundations for LES tone and cardial competence. The wrap secondarily calibrates the esophagogastric orifice. Omission of or compromise in any of these steps will negatively affect the final result.

Finally, every attempt should be made to minimize potential complicating forces before surgical intervention. Weight reduction, both as an inherent counter-reflux measure and postoperative decrease in tissue strain, must be encouraged. Optimization of pulmonary function, whether through smoking cessation, treatment of bronchospasm, or reduction in airway obstruction, is necessary. Liberal use of antiemetics, restricted use of anticholinergics, and avoidance of narcotic-induced constipation are important concepts in the immediate postoperative period.

OPERATIVE TECHNIQUES

In 1951, Allison described the first procedure designed specifically to treat GERD. Although unsuccessful, the essential concepts governing antireflux surgery were defined. In 1955, Rudolf Nissen folded the gastric fundus on itself to envelop the distal esophagus in a 49-year-old woman with chronic GERD. This technique, developed by Nissen serendipitously almost 20 years before, proved successful and was published in 1956. Presently referred to as a Nissen fundoplication, this procedure has undergone considerable modification and refinement. Total fundoplication, meaning a 360-degree wrap, can be accomplished through the abdomen, the left chest, and using the laparoscope.

P.2178

Open Transabdominal Nissen Fundoplication

The peritoneal cavity is opened in the upper midline. The liver's left lateral lobe is displaced off the hiatus and EGJ. The distal esophagus is circumferentially dissected. Care is taken to preserve both vagal trunks, anterior and posterior. Adequate esophageal length is confirmed; proper placement of a wrap requires the tension-free reduction of the EGJ below the diaphragm and a 2.5 cm or longer intraabdominal esophagus. A Penrose drain passed posterior to the esophagus facilitates its retraction. The hepatogastric omentum is left intact. According to Ellis (1996), this measure best ensures correct placement of the wrap around the distal esophagus instead of the proximal stomach. The esophagus is displaced leftward, exposing the right crus and its decussation of fibers (left crus). These crural pillars are approximated posterior to the esophagus, closing the hiatus until only a tip of a finger can enter the neohiatus. The short gastric vessels are divided. The fundus is freed from any remaining retrogastric adhesions. The characteristic esophagogastric fat pad is removed.

The mobilized fundus is passed posterior to the esophagus using the surgeon's right hand and received in the surgeon's left hand to the right of the esophagus' native position. In general, a fully and properly mobilized fundus can remain in this position without constant traction. Interrupted nonabsorbable sutures are placed to approximate the seromuscular walls of the adjacent fundic folds anterior to the esophagus, incorporating a small bite of esophageal wall in between. Usually, three sutures are placed 1 cm apart to create a 2-cm wrap. The sutures are tied with an intraluminal 50 to 60F bougie in place, simultaneously gauging the wrap's appropriate tightness and floppiness.

Open Transthoracic Nissen Fundoplication

Two circumstances may encourage the use of a transthoracic approach to fundoplication as opposed to the more direct and less morbid transabdominal route. The first and more important consideration concerns the foreshortened esophagus, anticipated or documented. Although a gastroplasty to lengthen the esophagus can be performed from the abdomen, it is more straightforward and potentially safer if completed through the chest. In patients having already undergone left thoracotomy (especially in patients having had an initial antireflux procedure performed through the left chest), the fundoplication may be performed through the same left (lower) thoracotomy, or redo-thoracotomy. It is recognized that thoracotomy is more painful and carries more attendant cardiopulmonary risks than laparotomy.

The patient is placed in the right lateral decubitus position. Provisions are made for single-lung (right) ventilation with either a double lumen tube (preferred) or bronchial blocker. A lateral thoracotomy is made in the seventh intercostal space. The inferior pulmonary ligament is divided and the mediastinal pleura opened to expose the esophageal bed. Care is taken not to enter the right pleural space. The esophagus is dissected from the posterior mediastinum. Again, for convenience, the distal thoracic esophagus is encircled with a rubber drain for atraumatic retraction. The opened mediastinum is extended to the diaphragm, and both the anterior aspect and posterior decussating tail or tails of the right crus are identified. The phrenoesophageal membrane is incised at the hiatus, enabling entry into the peritoneal cavity. An assessment of esophageal length is made; the future fundoplication and wrapped distal esophagus must reduce below the diaphragm without tension. The EGJ is defined; posteriorly, the retroperitoneal tissues are divided, avoiding inadvertent disruption of Belsey's artery communicating between the left gastric and inferior phrenic arteries.

The stomach is drawn into the left chest while the fundus is mobilized by dividing the short gastric vessels. The esophagogastric fat pad is excised. The posterior crural pillars are sutured together; the sutures are not tied. The fundoplication is made over a 50 to 60F bougie in the same manner as in the transabdominal approach outlined previously. The bougie is slowly removed and the fundoplication positioned below the diaphragm. The reapproximating crural sutures are tied, closing the hiatus without compressing the distal thoracic esophagus.

Toupet Modification of Nissen Fundoplication

Toupet (1963) reported the technique of a posterior partial fundoplication. The partial wrap may include 180 to 270 degrees of the circumference of the esophagus. After division of the short gastric vessels, the distal 2 to 3 cm of the esophagus is imbricated by the posterior gastric wrap. The hiatus is closed, leaving adequate space for the esophagus superiorly. The medial aspects of the left and right portions of the posterior gastric wrap are sutured directly to the adjacent wall of esophagus to accomplish the desired degree of the wrap. The lateral portions of the wrap are sutured to the respective limbs of the crus to maintain the fundoplication within the abdomen. Three sutures as a rule are used for each suture line. Of interest is that in comparative studies between this procedure and the complete Nissen fundoplication by Lundell and associates (1991), there was no significant clinical difference between the two procedures, except for a higher incidence of dysphagia at 3 months after the complete fundoplication that disappeared with time. In a long-term study by Lundell and colleagues (1996), the cumulative relapse rates at 3 years were essentially the same: 5% for the complete wrap and 6% for the partial wrap. Similar findings were reported by Thor and Silander (1989).

Transthoracic Partial Fundoplication (Belsey Mark IV Repair)

The antireflux procedure devised by Belsey is also performed through the left chest, but otherwise is uniquely dissimilar

P.2179

to a transthoracic Nissen fundoplication. In concept, Belsey's operation addresses aberrant physiology more than distorted anatomy. Belsey's exacting use of rigid esophagoscopy allowed him to correlate LES competence with LES position relative to the diaphragm. Examining patients in a seated position, Belsey (1977) observed in real time the loss of LES integrity with migration of the EGJ above the hiatal arch (a gapped opening was noted with inspiration). This induced patulous cardia could be reversed, he believed, by returning the EGJ and LES to their proper infradiaphragmatic position. Categorized as a partial fundoplication, the 240-degree Belsey wrap incites only a modicum of dysphagia and preserves the capacity for belching and vomiting in more than three fourths of patients. Although characteristically demanding of the surgeon, a Belsey partial fundoplication obtains equivalent GERD control (>85%) compared with the less formidable total fundoplication ascribed to Nissen. And although overall, reflux recurrence rates are slightly higher with partial fundoplications, the difference is particularly minimal in the Belsey type of construction. This favorable balance of outcomes with the Belsey Mark IV may be due to the distinctive architecture of its wrap. In contrast to standard fundoplication design, the wrap is formed by concentric folding of the fundus upon itself, theoretically providing a more moderated force than that which develops with direct opposition of the wrap against the esophagus. Admittedly counterintuitive, Lerut and colleagues (1990) suggest that this imbricated fundic overlay also discourages dilation of the intraabdominal esophagus. Finally, the Belsey approach preserves the opportunity for gastroplasty, realizing that the Pearson gastroplasty described by Pearson and associates (1971,1987) is significantly more complex than its cousin, the Collis-Nissen.

The Collis gastroplasty was originally reported by Collis in 1957. After lateral thoracotomy in the sixth or seventh intercostal space, the esophagus is freed from its mediastinal connections. Mobilization of the esophagus up to the level of the aortic arch has been emphasized. The phrenoesophageal membrane is incised, entering the peritoneal cavity first along the anteromedial aspect of the hiatal ring. In contrast to Nissen fundoplication, it is usually not necessary to divide the short gastric vessels because complete fundic mobilization is not required. The EGJ is defined, and the overlying fat pad is removed. Crural sutures are placed but left untied.

The Belsey fundoplication is formed using two overlapping rows of 2 0 silk, double-armed, mattress sutures. The first row consists of three sutures placed between the fundus and adjacent esophagus to initiate a crescentic fold that brings these structures into opposition. The vagi are used as guides for localizing these sutures: the most anterior suture is placed alongside the anterior trunk; the most posterior, adjacent to the posterior trunk; and the remaining one divides the distance in between. Constant tension is maintained when tying these sutures to avoid avulsion from the submucosal plane in which they are seated. The second fold develops from three more mattress sutures placed at a distance of 1.5 to 2.0 cm on either side of the knotted sutures of the first row. The suture needles from the second row are kept attached and, after creating their fold, are passed through the diaphragm from below upward. The classic Belsey teaspoon facilitates passage of these anchoring sutures through the diaphragm. These second-row sutures are then gradually tightened, inverting the entire fundoplication complex below the diaphragm in the process. These sutures are tied, and their knots come to lie against the left crus in the chest. Because a 360-degree wrap is not made, a calibrating bougie is not required. The crural sutures are then secured with the intention of forming a strong posterior buttress rather than a constricted hiatus.

Open Gastroplasty (Transthoracic Collis)

Gastroplasty refers to the surgical reshaping of a stomach's contour, typically through the process of gastric tubularization. In the context of antireflux surgery, a gastroplasty is an esophageal lengthening technique used in patients with esophageal shortening. Longitudinal contracture of the esophagus is an acquired, irreversible process caused by chronic esophagitis and subsequent transmural fibrosis. To regain esophageal length, a new distal segment is created in continuity with the native, diseased esophagus. This neoesophagus consists of a tube fashioned out of the gastric cardia. The new distal end effectively renders the intraabdominal portion of the esophagus longer. This fundamental change is both physiologically and technically advantageous. After gastroplasty, the distal esophagus and EGJ are transposed into the favorable pressure environment of the abdomen, and the lengthened esophagus provides additional terrain around which to construct a tension-free intraabdominal fundoplication.

Following the initial steps for a transthoracic Nissen fundoplication, the esophagus is exposed through the left chest. Mobilization is extended to the level of the aortic arch, as in a Belsey repair. The phrenoesophageal membrane is divided, the peritoneum is entered, and the stomach is brought into the chest. The uppermost short gastric vessels are usually ligated, and the esophagogastric fat pad is excised. A 50 to 60F bougie is advanced through the EGJ carefully and guided along the stomach's lesser curve. A linear stapler is placed parallel to the bougie on the left side and closed. In a cut Collis gastroplasty, the cutting stapler is fired, creating a staple line extending 5 cm from the EGJ inferiorly. Complete fundic area and volume are preserved. The staple line is then oversewn along is esophageal and gastric margins. In an uncut Collis, a noncutting linear stapler is used, whereby a functional but not anatomic gastric tube is made; all the bowel walls remain intact, minimizing leak risk. The semiseparated fundus is then plicated according to either the Belsey or Nissen technique. The

P.2180

fundoplication and neoesophagus are reduced below the diaphragm, and the crura are reapproximated in the manner already described.

Minimal Access Fundoplication

With the first published report on laparoscopic cholecystectomy in 1989, a relative revolution in surgical technique began. The integration of a minimal access approach to the surgical treatment of GERD has been equally spectacular. An unpublished report in 1999 estimated that nearly 35,000 laparoscopic Nissen fundoplications were performed in the United States alone. This reflects both the prevalence of GERD and the recognized patient preference for small incisions. Nonetheless, Spechler and co-workers' (2001) long-term outcome analysis comparing medical and surgical therapy for GERD has raised new questions regarding the overall efficacy of GERD managed by surgery. Realizing both the historical role played by the thoracic surgeon in antireflux surgery and current concerns regarding the surgical treatment of GERD, a comprehensive discussion on minimal access fundoplication attains heightened relevance and purpose.

As in all minimal access surgery, the indications for a particular procedure should correspond exactly with those defined for the same procedure performed with standard open techniques. Irrespective of approach, a surgical procedure is valued only in terms of its therapeutic intent; its success only in terms of effective disease treatment. Unfortunately, this principle has been sometimes overlooked in the age of laparoscopic surgery. This oversight has likely also influenced the application of minimal access techniques to antireflux surgery. It should not. Ultimately, standard and proven surgical decision making must remain at the center of the process and administration of appropriate care.

Particular contraindications to minimal access fundoplication relate more to reflux disease state and extent than to the inherent limitations of laparoscopy. Besides an inability to tolerate general anesthesia, or exclusion based on significant medical comorbidities, laparoscopic fundoplication can be technically challenging and occasionally not feasible in certain clinical situations. Obesity can prevent an endoscopic instrument from reaching the hiatus, and obesity-related organomegaly can thwart repositioning of organs and tissues using fine graspers or other tools measuring 10 mm or less. A large lateral segment of the liver's left lobe can be difficult to retract, obscuring the needed view of the gastroesophageal junction or, more commonly, the left crural region. Previous upper abdominal surgery, particularly in the left upper quadrant, can lead to dense adhesion formation. Adhesiolysis is usually possible, but it may be a complicated and time-consuming process. Port placement and standard operative routine can remain compromised even after the successful division of adhesions. Adhesions and obesity can also limit the development of pneumoperitoneum or lead to its inhomogeneous distribution. Rarely, carbon dioxide insufflation can aggravate an acidemia in patients with obstructive lung disease. And finally, as Dahlberg and colleagues' (2001), experience has shown, laparoscopic fundoplication is contraindicated in the setting of giant hiatal hernia (>5 cm).

One technique for minimal access fundoplication places the patient in a modified lithotomy position. Legs are bent in a neutral position with the calves and feet supported in padded stirrups. This position allows the surgeon to stand between the legs and best triangulate the gastroesophageal junction. The two-handed technique is facilitated with video monitor(s) placed at the head of the bed. Care is taken to secure the buttocks to the end of the bed in order to minimize slipping of the patient placed steep in reverse Trendelenburg; a roll turned under the draw sheet can be useful. An orogastric tube is placed to decompress the stomach, and a urinary catheter is used similarly. These are removed at the procedure's end.

In the abdomen having no upper midline scar, a point is selected two thirds of the distance from the xiphoid to the umbilicus. The proper selection of this point for insertion of the laparoscope is critical. A point too far from the epigastrium will prevent close viewing of the hiatus, whereas a point too close to the xiphoid likely leads to sword fighting between endoscopic instruments. A Veress needle may be used to insufflate CO₂, creating a pneumoperitoneum slowly to a maximal pressure of 15 mm Hg. The Veress needle is exchanged for a 10- to 12-mm trocar protected with a governing shield. A 30-degree laparoscope is inserted and the peritoneal cavity explored.

Four additional ports are placed under vision. A 5-mm port is placed in the right midabdomen along the anterior axillary line. A foldable liver retractor is inserted through this port, and the liver's left lateral segment is retracted cephalad and against the anterior abdominal wall. The triangular ligament, particularly its left aspect, is left intact. A second 5-mm port is inserted at the epigastrium and angled slightly leftward and superiorly. A third 10- to 12-mm port is positioned beneath the middle left subcostal margin. With the pneumoperitoneum evacuated at the end of the procedure, this incision is about 1 cm below the costal fold. By way of these upper ports, most of the operative steps are performed: dissection, suturing, and manipulation. A fourth 5-mm port is placed in the left midabdomen in line with the right abdominal port. Instrumentation through this port provides the bulk of gastric retraction and facilitates division of the short gastric vessels to free the stomach's greater curve.

The gastrohepatic ligament is incised, exposing the liver's caudate lobe and entering the superior aspect of the lesser sac. In most patients, this is an avascular plane. Prominent vessels in the area should raise the suspicion that an aberrant left hepatic artery exists. Every attempt should be made to preserve this vessel, although subsequent dissection certainly

P.2181

is made more difficult if it is saved. The right crus' peritoneal and fascial investments are left intact. The junction, or V, with the left crus is defined caudally and posterior. Blunt dissection is safer and preferred. Dissection proceeds toward the left lateral edge of the left crural pillar, creating a retroesophageal window to the left upper quadrant. The proximity of the superior splenic pole and hilum is anticipated. The posterior vagal trunk is protected. The supreme short gastric artery sometimes can be seen through this window, and a harmonic scalpel can be used to divide it from this perspective. Focus is now shifted anteriorly. Dissection continues to mobilize the anterior 150 degrees of the esophagus, incising the phrenoesophageal membrane from crus to crus at the level of the hiatal opening. The anterior vagus is protected. At this point, the entire intraabdominal esophagus has been freed. It is measured and assessed to be at least 2.5 cm in length; a standard grasper with its jaws fully opened approximates this distance and can be used as an in situ ruler; other authors employ a Penrose drain for retraction inserted through the retroesophageal window with its ends held together anteriorly as a sling. The same degree of exposure can be obtained using the front and back aspect of each endoscopic tool, two-handed technique, and a continual shift between the retraction and distraction of the pertinent tissues. It is the practice of many surgeons to close the hiatus at this juncture.

Complete fundic mobilization and division of the short gastric vessels is the next step. Physiologically, the fundus works in concert with the LES. Swallow-mediated receptive relaxation of the LES continues aborally into the fundus, which also relaxes in series with the LES. The gastric body, however, does not relax with swallowing. Failure to mobilize the fundus predisposes to using the gastric body to create a wrap. The loss of coordinated receptive relaxation aggravates improper LES compliance and prolongs gastric emptying. The fundoplasty can only be fashioned without tension if the fundus is not tethered in any way. The main tethering structures are the short gastric vessels contained within the splenogastric ligament. These are divided using the harmonic scalpel. The gastrosplenic ligament oriented in an anteroposterior fashion, not mediolaterally, exposes the short gastric arcades best. In dividing the lower arcades (more distal along the greater curvature), particular care is taken to avoid injury to the colon's splenic flexure. A trial fundoplication is performed. The mobilized fundus is brought through the retroesophageal space. A reticulating grasper is used to take hold of the posterior wall of the fundus, but not to pull it. Instead, the grasper guides the fundus as it is being fed from the left, behind the esophagus and through the retroesophageal window, to its final position on the right side of the esophagus. This path must be tension free. Testimony to the wrap's friction-free passage is obtained with the shoeshine maneuver, so-termed by John Hunter at Emory. Held by its posterior and anterior folds, the unfettered fundus should move back and forth through the retroesophageal window without resistance.

Regardless of the size of the hiatal defect or degree of herniation, the crura are closed. The orogastric tube is removed (with any esophageal stethoscope or thermometer), and a 50 to 60F (usually, 52 to 56F) bougie is placed transorally. The right and left crural pillars are brought together using two to four interrupted nonabsorbable sutures. Full-thickness bites of the crura are taken, and the self-contained needle-driver unit limits pleural penetration and esophageal perforation and reduces the risk for venocaval puncture. Only the posterior diaphragmatic opening behind the esophagus is closed. Excessive tightening while tying these sutures can prove problematic. It can lead to strangulation of the crural muscles, distal esophageal obstruction unrelated to the wrap, or both. The calibrating bougie is removed.

Correct orientation and geometry of the wrap are considered paramount. Distortion of the wrap can have functional and other physiologic consequences. The easiest and most common technical oversight involves wrapping the anterior stomach rather than its posterior fundic wall around the esophagus. This wrap architecture forms the basis for the Rossetti-Hell (1977) modification of a Nissen fundoplication. Experience has shown a significant increase in postoperative dysphagia associated with this modification, suggesting that the Nissen posterior fundoplication optimally preserves gastric and fundic motility while properly supporting the incompetent LES. Again, the posterior fundus is passed posterior to the esophagus through the opened retroesophageal window. The bougie (>56F) is reinserted, and a truly well-mobilized fundus will remain in a wrapped stance. The posterior lip of the fundus (patient's right) and the anterior lip (patient's left) are held together in opposition, meeting along the right anterolateral aspect of the esophagus. The looseness of the fundoplication is assured. The gastroesophageal fat pad is removed. Our rationale for removing this tissue is twofold: (a) definition of the EGJ the wrap should envelop the esophagus above this landmark; and (b) promotion of adhesion between the folds and lips of the wrap and the overlapped margin of anterior esophagus. Typically, three sutures are placed to create a 2- to 2.5-cm fundoplication. Nonabsorbable suture material is used, and the stitch is passed first through the anterior side, then through the anterior wall of the esophagus, and then through the posterior fundus. An Endo-stitch device is convenient here, too, although the fundoplication can be secured using free sutures. Nasogastric drainage postoperatively is optional.

Minimal Access Partial (Toupet) Fundoplication

The laparoscopic approach to partial fundoplication, and in particular the eponymous Toupet, has been well characterized by Swanstrom (1996). Applying the same consideration as in open antireflux surgery, a less than 360-degree wrap is justified in GERD patients who also have documented impaired esophageal motility. A complete fundoplication

P.2182

may precipitate iatrogenic achalasia if esophageal body contracture amplitude is low or the peristaltic stripping wave is abnormally coordinated. From an analogous perspective, the adjunctive use of only partial fundoplications to minimize reflux that may accompany therapeutic esophagomyotomy for achalasia is appropriate. In patients having diminished esophageal tone, a balance between optimal antireflux barrier effect achieved through fundoplication and possible esophageal obstruction consequent to fundoplication must be struck. As demonstrated by Lund and co-workers (1997), a 270- versus 360-degree wrap in patients with compromised esophageal motility causes less gas bloat and dysphagia following fundoplication. The unconditional or unqualified use of partial fundoplication also has its limitations. Jobe and colleagues (1997) have shown that, if uniformly performed regardless of esophageal motor status, partial fundoplication is associated with reflux failure rates in excess of 20% at 22 months and positive pH scores in 51% of patients restudied (also at 22 months).

From a technical standpoint, minimal access partial fundoplication duplicates the principal steps used in laparoscopic Nissen fundoplication. Patient positioning, port placement, hiatal dissection, fundic mobilization, and posterior crural closure are the same for the two procedures. In fashioning the partial wrap, the posterior fundus is passed through the retroesophageal window, as already described for laparoscopic complete fundoplication. Leaving 90 degrees of the anterior esophageal wall unwrapped, the right side of the fundus is endoscopically sutured to the right anterolateral aspect of the esophageal wall; four sutures are typically sufficient. Next, the right posterior aspect of the fundus is secured to the right diaphragmatic pillar using three or four interrupted sutures. To complete the wrap, the left posterior aspect of the fundus forming the anterior fundic lip is sutured to the left crural pillar. Three sutures are usually enough to approximate this aspect of the fundus to the left anterolateral wall of the esophagus. Some authors incorporate the posteriorly wrapped fundus with the closure of the crura. Others use an intraesophageal bougie to calibrate the tightness of the wrap; I have found this measure to be unnecessary.

Minimal Access Gastroplasty

In the early laparoscopic era, one contraindication to minimal access fundoplication was the suspected shortened esophagus. Similarly, before 1996, the short esophagus recognized at laparoscopy was an indication for conversion to an open procedure. Advanced laparoscopic techniques have continued to evolve, however, and currently, minimal access gastroplasty has transcended theory into clinical practice. Importantly, the incidence of GERD-associated esophageal shortening has remained constant at about 10%. This rate has been documented in the open, transitional, and laparoscopic eras by Polk (1976) as well as by Kauer (1995) and Johnson (1998) and their associates. More importantly, this pathologic development has remained supremely relevant for the antireflux surgeon. The unrecognized shortened esophagus around which fundoplication is created can lead to crural disruption, intrathoracic herniation, or distal misplacement of the wrap (a slipped fundoplication). In the age of open antireflux surgery, Siewert (1989) and Ellis (1996) and their co-workers linked an untreated shortened esophagus to a 20% to 33% fundoplication failure rate. Failure rates and morbidity following laparoscopic fundoplication in cases of missed short esophagus are comparable according to Swanstrom and colleagues (1996), followed subsequently by Johnson and co-workers (1998).

The minimal access approach to the short esophagus starts with complete distal esophageal mobilization. Circumferential dissection typically requires division of the phrenoesophageal ligament and retrogastric attachments; complete fundic mobilization may also be required. According to Horvath and associates (2000), this type I dissection enables the distal 3 to 4 cm of the esophagus to be freed in 90% of cases. An assessment of intraabdominal esophageal length is then made. A length of at least 2.5 cm is ensured. Excessive inferior retraction of the stomach is avoided because any hiatal tethering can lead to elongation of the proximal stomach; a misperception of adequate esophageal length results, given the similar appearance between a stretched cardia and the distal esophagus. After type I dissection, 5% to 7% of patients persist in having less than 2.5 cm of true intraabdominal length. A type II dissection is thereby indicated, in which transhiatal dissection is employed to mobilize the lower mediastinal esophagus. A second assessment of intraabdominal length is made, and in 3% to 5% of cases, the esophagus will still be short. A minimal access lengthening procedure is then undertaken.

As of yet, a classic open Collis gastroplasty has not been replicated using minimal access techniques. This would require a total thoracoscopic approach to the short esophagus, a somewhat impractical construct given the limited space in the chest, the inability to mobilize adequately the gastric fundus from the chest, and the overall inaccessibility to the nonherniated stomach from the chest. Nonetheless, equivalent esophageal lengthening using gastric tubularization has been achieved with other advanced endoscopic methods.

Swanstrom and co-workers (1996) were the first to describe minimal access gastroplasty using a combined laparoscopic and right thoracoscopic approach. In this procedural tour-de-force, the gastric tube is created with the use of an articulating endoscopic 3-cm linear stapler guided thoracoscopically through the right chest, across the mediastinum, and into the left upper abdomen through the esophageal hiatus. The already laparoscopically mobilized fundus is oriented in an anteroposterior direction. A bougie is advanced across the EGJ and into the stomach along its lesser curve. The gastric fundus is fed into the jaws of the opened stapler,

P.2183

and the stapler is closed flush with the free side of the bougie. Firing the stapler creates a 3-cm gastric tube in continuity with the distal esophagus. The fundus remaining is then used to create the wrap around the lengthened esophagus. Awad and co-workers (2000) published their experience using left-sided thoracoscopically assisted minimal access gastroplasty to treat a foreshortened esophagus in eight patients.

Limiting the procedure to just one body cavity, Johnson and colleagues (1998) laparoscopically reproduced Steichen's (1986) total abdominal approach to Collis gastroplasty. In this elegant procedure, a circular 25-mm (EEA) stapler is used to create a full-thickness hole in the stomach 3 cm below the angle of His. The EEA stapler components are introduced through an upper epigastric subxiphoid incision or enlargement of the laparoscopic port closest to that location. With a bougie positioned along the right perimeter of the EEA hole, a separate linear stapler is inserted through the hole and fired in parallel with bougie. The gastric tube is thus formed from the bottom up, rather than from the angle of His downward. The fundic remnant is ultimately plicated, as in the Swanstrom endoscopic Collis-Nissen.

MORTALITY AND MORBIDITY

Mortality

The mortality associated with open antireflux operations, abdominal or transthoracic, is about 1% or less in most series, as noted in the review of Beauchamp and associates (2001). After laparoscopic antireflux procedures (almost exclusively Nissen fundoplications), the mortality rate has been 0% to 1%, as noted in the report of Tirnaksiz and Deschamps (2001).

Morbidity

Intraoperative Complications

The three major complications of a laparoscopic antireflux procedure are: (a) bleeding from an injured spleen or, rarely, a torn short gastric vessel [0.26% as noted by Collet and Cadiere (1995)]; (b) perforation of the stomach or the esophagus [an incidence of 4.6% in the study by Schauer and coinvestigators (1996)]; and (c) the necessity of conversion to an open procedure due to technical difficulties [0.3% to 12% in the series collected by Tirnaksiz and Deschamps (2001)]. The same complications, except for the necessity for conversion, may occur in the open procedures.

Early Postoperative Complications

Similar complications occur after either an open or a laparoscopic antireflux operation, although the incidence may differ with the various procedures. However, wound infection and incisional hernia are not seen with a thoracic approach, nor is the postoperative thoracotomy pain syndrome (more often seen after these operations than those for other thoracic pathology; the reason remains unknown) seen after an abdominal operation.

The major early problems are the gas-bloat syndrome (the patient can neither belch nor vomit), dysphagia, and increased flatulence. The incidence of the gas-bloat syndrome is less with the floppy Nissen than after the original Nissen procedure. The incidences of the aforementioned problems vary greatly, but as a rule, the magnitude of the symptoms decreases with time.

Late Complications

The antireflux repair may break down with the subsequent return of the symptoms of gastric reflux. This occurs usually in less than 5% of the patients. The Nissen wrap may infrequently herniate into the chest due to lack of fixation or a failure to recognize a shortened esophagus, but more often the fundoplication may be created on the proximal portion of the stomach (the so-called slipped Nissen, as noted previously). A paraesophageal hernia may occur in 2% to 8% of cases, as noted by Soper and Dunnegan (1999). Beauchamp and co-workers (2001) noted a recurrence or persistence of a hiatal or parahiatal hernia in 10% of 179 patients who had undergone a short, floppy, transabdominal Nissen fundoplication. This complication is often the reason for a revision procedure.

When the entire or a portion of the Nissen wrap herniates into the thorax, either early postoperatively or late (from 1 to 10 or more years postoperatively), infrequently the wrap may become incarcerated, as noted by Idani and associates (2000). Subsequently, the incarceration may result in obstruction of the esophagus, or an area of the gastric wrap may become necrotic, and a perforation into the left pleural space may occur. A pyothorax with its attendant signs and symptoms develops and may be fatal if not recognized. This complication was first recorded by Stilwell (1978) and subsequently by Mansour and colleagues (1981) and Mansour and Sharma (2003). Surgical treatment consists of decortication of the lung, drainage of the pleural space, reduction of the wrap to below the diaphragm, and drainage of the gastric wrap by a large drainage tube usually inserted through the site of perforation, with the tube carried beneath the left hemidiaphragm and established as a gastrostomy on the anterior abdominal wall. Closure of any diaphragmatic defects is accomplished before closure of the thoracotomy incision.

In a small number of patients, dysphagia or the gas-bloat syndrome may persist. Esophageal dilations or redo surgery may be necessary to correct these persistent late complications.

P.2184

RESULTS

Satisfactory to excellent results may be expected in 85% to 95% of the patients who undergo either open or laparoscopic antireflux surgery. Many surgeons believe that overall the results are better in the laparoscopic group, but this is yet to be proved conclusively. Nonetheless, because of many of the other advantages of the laparoscopic approach (e.g., length of hospital stay, less postoperative pain), it was the most commonly performed antireflux operation in the last decade of the 20th and in the first years of the 21st century.

REFERENCES

Allison PR: Reflux esophagitis, sliding hiatal hernia, and the anatomy of repair. Surg Gynecol Obstet 92:419, 1951.

Awad ZT, et al: Left side thoracoscopically assisted gastroplasty: a new technique for managing the shortened esophagus. Surg Endosc 14:508, 2000.

Beauchamp G, Ouellette D, Dup r S: Open antireflux repairs for simple reflux esophagitis; short- and long-term results. Chest Surg Clin North Am 11:555, 2001.

Belsey RH: Mark IV repair of hiatal hernia by the transthoracic approach. World J Surg 1:475, 1977.

Campos GM, et al: Multivariate analysis of factors predicting outcome after laparoscopic Nissen fundoplication. J Gastrointest Surg 3:292, 1999a.

Campos GM, et al: The pattern of esophageal acid exposure in gastroesophageal reflux disease influences the severity of the disease. Arch Surg 134:882, 1999b.

Collet D, Cadiere GB: Conversions and complications of laparoscopic treatment of gastroesophageal reflux disease. Formation for the Development of Laparoscopic Surgery for Gastroesophageal Reflux Disease Group. Am J Surg 169:622, 1995.

Collis JL: An operation for hiatus hernia with short esophagus. J Thorac Cardiovasc Surg 34:768, 1957.

Costantini M, et al: The role of a defective lower esophageal sphincter in the clinical outcome of treatment for gastroesophageal reflux disease. Arch Surg 131:655, 1996.

Dahlberg PS, et al: Laparoscopic repair of large paraesophageal hiatal hernia. Ann Thorac Surg 72:1125, 2001.

DeVault KR, Castell DO: Update guidelines for the diagnosis and treatment of gastroesophageal reflux disease. The Practice Parameters Committee of the American College of Gastroenterology. Am J Gastroenterol 94:1434, 1999.

Dodds WJ, et al: Mechanisms of gastroesophageal reflux in patients with reflux esophagitis. N Engl J Med 307:1547, 1982.

Donahue PE, et al: The floppy Nissen fundoplications. Effective long-term control of pathologic reflux. Arch Surg 120:663, 1985.

Ellis FH Jr: Nissen fundoplication. In Braasch JW, et al (eds): Atlas of Abdominal Surgery. Philadelphia: WB Saunders, 1990, p. 11.

Ellis FH Jr., Gibb SP, Sheatly GJ: Reoperation after failed antireflux surgery. Review of 101 cases. Eur J Cardiothorac Surg 10:225, 1996.

Fein M: Duodenogastroesophageal reflux parallels acid and not alkaline exposure in the esophagus and contributes to complications of reflux disease. Am J Gastroenterol 91:1663, 1996.

Fein M, et al: Isolated upright gastroesophageal reflux is not a contraindication for antireflux surgery. Surgery 122:859, 1997.

Hagen JA, Peters JH: Minimally invasive approaches to antireflux surgery. Semin Thorac Cardiovasc Surg 12:157, 2000.

Heartburn Across America: A Gallup Organization National Survey. Princeton, NJ: Gallup Organization, 1988.

Helm JF, et al: Effect of esophageal emptying and saliva on clearance of acid from the esophagus. N Engl J Med 310:284, 1984.

Hiebert CA: Surgical management of esophageal reflux and hiatal hernia: 1951. Ann Thorac Surg 52:159, 1991.

Holloway RH, et al: Gastric distention: a mechanism for postprandial gastroesophageal reflux. Gastroenterology 89:779, 1985.

Horvath KD, Swanstrom LL, Jobe BA: The short esophagus: pathophysiology, incidence, presentation, and treatment in the era of laparoscopic antireflux surgery. Ann Surg 232: 630, 2000.

Hunter JG, et al: A physiologic approach to laparoscopic fundoplication for gastroesophageal reflux disease. Ann Surg 223:673, 1996.

Idani H, et al: Acute intrathoracic incarceration of the stomach after laparoscopic Nissen fundoplication. Surg Laparosc Endosc Percutan Tech 10:99, 2000.

Jobe BA, et al: Evaluation of laparoscopic Toupet fundoplication as a primary repair for all patients with medically resistant gastroesophageal reflux. Surg Endosc 11:1080, 1997.

Johnson AB, Oddsdottir M, Hunter JG: Laparoscopic Collis gastroplasty and Nissen fundoplication. A new technique for the management of esophageal foreshortening. Surg Endosc 12:1055, 1998.

Kahrilas PI, Shi G: Pathophysiology of gastroesophageal reflux disease: the antireflux barrier and luminal clearance mechanisms. In Orlando R (ed): Gastroesophageal Reflux Disease. New York: Marcel Dekker, 2000, p. 137.

Kauer WK, et al: A tailored approach to antireflux surgery. J Thorac Cardiovasc Surg 110:141, 1995.

Lagergren J, et al: Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 340:825, 1999.

Lerut T, et al: The Belsey Mark IV antireflux procedure: reflections on indications and long-term results. Acta Gastroenterol Belg 13:585, 1990.

Locke GR III, et al: Prevalence and clinical spectrum of gastroesophageal reflux: a population-based study in Olmsted County, Minnesota. Gastroenterology 112:1448, 1997.

Lund RF, et al: Laparoscopic Toupet fundoplication for gastroesophageal reflux disease with poor esophageal body motility. J Gastrointest Surg 1:301, 1997.

Lundell L, et al: Lower esophageal sphincter characteristics and esophageal acid exposure following partial or 360 degree fundoplication: results of a prospective randomized clinical study. World J Surg 15:115, 1991.

Lundell L, et al: Long-term results of a prospective randomized comparison of total fundic wrap (Nissen-Rossetti) or semifundoplication (Toupet) for gastro-oesophageal reflux. Br J Surg 83:830, 1996.

Mansour KA, Sharma J: Delayed intrathoracic rupture of herniated Nissen fundoplication: report of two cases. Ann Thorac Surg 75:1957, 2003.

Mansour KA, et al: Complications of intrathoracic Nissen fundoplication. Ann Thorac Surg 32:173, 1981.

Mittal RK, et al: Transient lower esophageal sphincter relaxation. Gastroenterology 109:601, 1995.

Nebel OT, Fornes MF, Castell DO: Symptomatic gastroesophageal reflux: incidence and precipitating factors. Am J Dig Dis 21:953, 1976.

Negre JB, et al: Nissen fundoplication. Results at 10 year follow-up. Am J Surg 146:635, 1983.

Orlando RC: Pathogenesis of gastroesophageal reflux disease. Gastroenterol Clin North Am 31:S35, 2002.

Pearson FG, Langer B, Henderson RD: Gastroplasty and Belsey hiatus hernia repair. An operation for the management of peptic stricture with acquired short esophagus. J Thorac Cardiovasc Surg 61:50, 1971.

Pearson FG, et al: Gastroplasty and fundoplication for complex reflux problems. Long-term results. Ann Surg 206:473, 1987.

Polk HC: Fundoplication for reflux esophagitis: misadventures with the operation of choice. Ann Surg 183:645, 1976.

Rakic S, et al: Role of esophageal body function in gastroesophageal reflux disease: implications for surgical management. J Am Coll Surg 185:380, 1997.

Rossetti M, Allgower M: Fundoplication for treatment of hiatal hernia. Prog Surg 12:1, 1973.

Rossetti M, Hell K: Fundoplication for the treatment of gastroesophageal reflux in hiatal hernia. World J Surg 1:439, 1977.

Schauer PA, et al: Mechanisms of gastric and esophageal perforations during laparoscopic Nissen fundoplication. Ann Surg 223:43, 1996.

Schindlbeck NE, et al: Optimal thresholds, sensitivity, and specificity of long-term pH-metry for the detection of gastroesophageal reflux disease. Gastroenterology 93:85, 1987.

Schlesinger PK. et al: Limitations of 24 hour intraesophageal pH monitoring in the hospital setting. Gastroenterology 89:797, 1985.

Siewert JR, Isolari J, Reussner H: Reoperation following failed fundoplication. World J Surg 13:791, 1989.

Sloan S, Kahrilas PJ: Impairment of esophageal emptying with hiatal hernia. Gastroenterology 100:596, 1991.

P.2185

Soper NJ, Dunnegan D: Anatomic fundoplication failure after laparoscopic antireflux surgery. Ann Surg 229:669, 1999.

Spechler SJ: GERD and its complications. Mt Sinai J Med 67:106, 2000.

Spechler SJ, et al: Long-term outcome of medical and surgical therapies for gastroesophageal reflux disease: follow-up of a randomized controlled trial. JAMA 285:2331, 2001.

Steichen FM: Abdominal approach to the Collis gastroplasty and Nissen fundoplication. Surg Gynecol Obstet 162:372, 1986.

Stilwell JH: Intrathoracic perforation of hiatus hernia. Br Med J 2:605, 1978.

Swanstrom LL: Laparoscopic partial fundoplication. Probl Gen Surg 13: 75, 1996.

Swanstrom LL, Marcus DR, Galloway GQ: Laparoscopic Collis gastroplasty is the treatment of choice for the shortened esophagus. Am J Surg 171: 477, 1996.

Thor KBA, Silander TA: A long-term randomized prospective trial of the Nissen procedure versus a modified Toupet technique. Ann Surg 210: 719, 1989.

Tirnaksiz MB, Deschamps C: Laparoscopic repair for simple ulcerative esophagitis: current results. Chest Surg Clin North Am 11:547, 2001.

Toupet A: Technique d'oesophago-gastroplastie avec phr no-gastropexie appliqu e dans la cure radicale des hernies hiatales et comme complement de l'op ration de Heller dans les cardiospasmes: M moire de l'Acad mie de Chirurgie 11:394, 1963.

Wetscher GJ, et al: Respiratory symptoms in patients with gastroesophageal reflux disease following medical therapy and following antireflux surgery. Am J Surg 174:639, 1997.

Reading References

Bowrey DJ, Peters JH: Laparoscopic esophageal surgery. Surg Clin North Am 80:1213, 2000.

Champault G, et al: Gastroesophageal reflux: conventional surgical treatment versus laparoscopy; a prospective study of 61 cases. Surg Laparosc Endosc 6:434, 1996.

Chrysos E, et al: Laparoscopic surgery for gastroesophageal reflux disease patients with impaired esophageal peristalsis: total or partial fundoplication. J Am Coll Surg. 197:8, 2003.

Coelho JC, et al: Complications of laparoscopic fundoplication in the elderly. Surg Laparosc Endosc Percutan Tech 13:6, 2003.

Coelho JC, et al: Conversions and complications of laparoscopic treatment of gastroesophageal reflux disease. J Am Coll Surg 189:356, 1999.

Dallemagne B, et al: Laparoscopic Nissen fundoplication: preliminary report. Surg Laparosc Endosc 1:138, 1991.

Dent J, et al: Mechanism of gastroesophageal reflux in recumbent asymptomatic human subjects. J Clin Invest 65:256, 1980.

DeVault KR, Castell DO: Current diagnosis and treatment of gastroesophageal reflux disease. Mayo Clin Proc 69:867, 1994.

Fass R: Gastroesophageal reflux disease revisited. Gastroenterol Clin North Am 31:S1, 2002.

Flum DR, et al: The nationwide frequency of major adverse outcomes in antireflux surgery and the role of surgeon experience, 1992 1997. J Am Coll Surg 195:611, 2002.

Heine KJ, Mittal RK: Crural diaphragm and lower esophageal sphincter as antireflux barriers. Viewpoints Dig Dis 23:1, 1991.

Hirschowitz BI: A critical analysis with appropriate controls of gastric acid and pepsin secretion in clinical esophagitis. Gastroenterology 101:1149, 1991.

Kahrilas PJ: Gastroesophageal reflux disease. JAMA 276:983, 1996.

Kahrilas PJ: Surgical therapy for reflux disease. JAMA 285:2376, 2001.

Lafullarde T, et al: Laparoscopic Nissen fundoplication: five-year results and beyond. Arch Surg 136:180, 2001.

Laws HL, Clements RH, Swillie CM: A randomized prospective comparison of the Nissen fundoplication versus the Toupet fundoplication for gastroesophageal reflux disease. Ann Surg 225:647, 1997.

Mabrut JY, et al: Impact of antireflux operation on columnar-lined esophagus. J Am Coll Surg 196:60, 2003.

Naunheim KS, et al: Laparoscopic fundoplication: a natural extension for the thoracic surgeon. Ann Thorac Surg 61:1062, 1996.

Perdikis G, et al: Esophageal manometry and 24 hour pH testing in the management of gastroesophageal reflux patients. Am J Surg 174:634, 1997.

Rice TW: Why antireflux surgery fails. Dig Dis 18:43, 2000.

Rice TW, Gagner M: Laparoscopic antireflux surgery. Semin Thorac Cardiovasc Surg 9:173, 1997.

Ritter MP, et al: Outcome after laparoscopic fundoplication is not dependent on a structurally defective lower esophageal sphincter. J Gastrointest Surg 6:567, 1998.

Skinner DB, Belsey RH: Surgical management of esophageal reflux and hiatus hernia: long-term results with 1,030 patients. J Thorac Cardiovasc Surg 53:33, 1967.

Spechler SJ: Epidemiology and natural history of gastro-oesophageal reflux disease. Digestion 51[Suppl 1]:24, 1992.

Spechler SJ, et al: Comparison of medical and surgical therapy for complicated gastroesophageal reflux disease in veterans. The Department of Veterans Affairs Gastroesophageal Reflux Disease Study Group. N Engl J Med 326:786, 1992.

Tefera L, et al: Can the combination of symptoms and endoscopy confirm the presence of gastroesophageal reflux disease? Am Surg 63:933, 1997.

Terry M, et al: Outcomes of laparoscopic fundoplication for gastroesophageal reflux disease and paraesophageal hernia. Surg Endosc 15:691, 2001.

Viljakka MT, Luostarien ME, Isolauri JO: Complications of open and laparoscopic antireflux surgery: 32-year audit at a teaching hospital. J Am Coll Surg 185:446, 1997.

Waring JP: Surgical and endoscopic treatment of gastroesophageal reflux disease. Gastroenterol Clin North Am 31:S89, 2002.