Advanced therapy in thoracic surgery - part 9 potx

440 / Advanced Therapy in Thoracic Surgery 54 Koul PA, Koul AN, Wahid A, et al CT in pulmonary hydatid disease — unusual appearances Chest 2000;118:1645–7 62 Fontenelle LT, Armstrong RG, Stanford W, et al The asymptomatic mediastinal mass Arch Surg 1971;102:98–102 55 Chevret R, Jouhari-Ouaraini A, Rahali R Kystes hydatiques du poumon: le probleme des recidives Chirurgie 1980;106:673 63 Benjamin SP, McCormack LJ, Effler DB, Groves LK Primary tumors of the mediastinum Chest 1972;62:297–303 56 Shaffer K, Rosadodechristenson ML, Patz FT, et al Thoracic lymphangioma in adults — CT and MR imaging features Am J Roentgenol 1994;162:283–89 57 Riquet M, Briere J, Pimpec-Barthes F, et al Cystic lymphangiomas of the neck and mediastinum: are there acquired forms? Revue des Maladies Respiratoires 1999;16:71–9 58 Charruau L, Parrens M, Jougon J, et al Mediastinal lymphangioma in adults: CT and MR imaging features Eur Radiol 2000;10:1310–4 59 Shields TW Primary lesions of the mediastinum and their investigation and treatment In: Shields TW, editor General thoracic surgery Baltimore (MD): Williams and Wilkins; 1994 p 1724–69 60 Sabiston DC Jr, Scott HW Jr Primary neoplasms and cysts of the mediastinum Ann Surg 1952;136:777–97 61 Burkell CC, Cross JM, Kent HP, et al Mass lesions of the mediastinum Curr Probl Surg 1969;2–57 64 Rubush JL, Gardner IR, Boyd WC, et al Mediastinal tumors Review of 186 cases J Thorac Cardiovasc Surg 1973;65:216–22 65 Vidne B, Levy MJ Mediastinal tumours Surgical treatment in forty-five consecutive cases Scand J Thorac Cardiovasc Surg 1973;7:59–65 66 Nandi P, Wong KC, Mok CK, et al Primary mediastinal tumours: review of 74 cases J R Coll Surg Edinb 1980;25:460–6 67 Davis RD Jr, Oldham HN Jr, Sabiston DC Jr Primary cysts and neoplasms of the mediastinum: recent changes in clinical presentation, methods of diagnosis, management, and results Ann Thorac Surg 1987;44:229–37 68 Azarow KS, Pearl RH, Zurcher R, et al Primary mediastinal masses A comparison of adult and pediatric populations J Thorac Cardiovasc Surg 1993;106:67–72 CHAPTER 36 DESCENDING NECROTIZING MEDIASTINITIS MARK D IANNETTONI, MD WILLIAM R LYNCH, MD, MS infection at operation or postmortem examination Establishment of the relationship of oropharyngeal infection with the development of the necrotizing mediastinal process Although established for the purposes of a retrospective review, these criteria are a useful framework for the clinical diagnosis of DNM Appreciating the pathophysiology of DNM prepares one to suspect this life-threatening infection The primary source is often an oropharyngeal abscess originating from an odontogenic infection of a second or third molar.3–5 The infections spread along the deep cervical fascial planes on the way to the mediastinum and may cause cellulitis, abscess formation, pericardial and pleural effusions, empyemas, tissue necrosis, mediastinitis, and sepsis Other sources are peritonsillar or retropharyngeal abscesses and Ludwig’s angina.6–8 Traumatic endotracheal intubation or endoscopic procedures can result in pharyngeal perforations, which become a portal for infection to reach the mediastinum.9–11 The majority of acute mediastinal infections result from complications of cardiac surgery or perforations of the esophagus Less commonly, acute mediastinitis can result from spread of cervical or odontogenic infections Suppurative infections of the oropharynx extending to the mediastinum have become rare since the advent of antibiotics When this does occur, the most severe infection is descending necrotizing mediastinitis Descending necrotizing mediastinitis (DNM) is a fulminant infection leading to uncontrolled sepsis and death if not promptly recognized and appropriately treated It is frequently the result of odontogenic infections or unrecognized injuries to the aerodigestive tract These virulent infections dissect through and travel along the fascial planes of the neck to reach the mediastinum The diagnosis of DNM is almost always delayed and is often not made until sepsis persists or develops following inadequate surgical drainage of the cervical infection Early diagnosis and aggressive treatment is paramount in salvaging these patients as the reported mortality can reach 40% 1–3 We review the pertinent anatomy and pathophysisolgy, clinical manifestations, diagnostic work-up, and recommended treatment of DNM Anatomy Understanding the anatomy of the neck and the various pathways from the oropharynx and aerodigestive tract into the mediastinum is necessary to make the association between the source of infection and resulting mediastinitis The neck contains viscera, muscles, nerves, blood vessels, and a bony skeleton As in the limbs and torso, these structures are contained and supported by fascial coverings of connective tissue organized as sheets and membranes These fasciae wrap around the structures of the neck, forming cylindrical connective tissue planes that span from the face and base of the skull to the mediastinum and thoracic inlet These cylindrical sheets Anatomy and Pathophysiology DNM is an aggressive, suppurative mediastinitis associated with odontogenic or cervicofacial infections A review by Estrera and colleagues in 1983 defined criteria for diagnosis of DNM The criteria included: Clinical manifestations of severe infection Demonstration of characteristic roentgenographic features Documentation of the necrotizing mediastinal 441 442 / Advanced Therapy in Thoracic Surgery of fascia confine and compartmentalize the deep structures of the neck, defining the cervicomediastinal region The cervicomediastinal region is traversed in its length by three major deep fascial layers: the superficial layer, the visceral layer, and the prevertebral layer The superficial layer invests the sternocleidomastoid muscle, trapezius muscle, strap muscles, and parotid and submandibular glands The visceral layer surrounds the thyroid gland, esophagus, and trachea Its upper limit attaches to the hyoid bone and extends inferiorly to the mediastinum The prevertebral layer is immediately adjacent to the vertebral column and runs from the base of the skull to the coccyx The alar layer, just anterior to the prevertebral layer, is intimate with the prevertebral layer but extends only to the second thoracic vertebra.12 These three deep fascial layers (Figure 36-1), individually and as a group, surround and define five potential spaces of the neck: the pretracheal space, the perivascular space, the retrovisceral space, the submandibular space, and the lateral pharyngeal space Of these five spaces, the first three (pretracheal, perivascular, and retrovisceral) provide avenues for infections that originate in the head and neck to descend into the mediastinum.3,4,13,14 The latter two spaces (submandibular and lateral pharyngeal) communicate with the first three, allowing infections of the oropharynx a pathway to the mediastinum Each of these five spaces will be described below The pretracheal space is anterior to the trachea and posterior to the strap muscles and pretracheal fascia The pretracheal fascia, together with the buccopharyngeal fascia, encloses the pharynx, esophagus, larynx, trachea, Omohyoid m Sternothyroid m Pretracheal space Sternohyoid m Angle of Dissection Carotid sheath Sternocleidomastoid m Retrovisceral space Buccopharyngeal fascia Prevertebral fascia FIGURE 36-1 The different fascial planes of the neck Infection may descend through these planes into the mediastinum The arrow demonstrates the standard surgical approach to the prevertebral fascia medial to the sternocleidomastoid muscle and thyroid gland Reproduced with permission from Wheatley et al.4 thyroid, and parathyroid The buccopharyngeal fascia externally invests the upper part of the alimentary tract It covers the pharynx, the buccinator muscle, and the posterior esophagus The buccopharyngeal fascia, together with the pretracheal fascia blends along the pharyngeal constrictors, the hyoid, the thyroid cartilage, and the thyroid The pretracheal fascia descends from the thyroid, covering the trachea anteriorly and the sternum posteriorly The inferior extent of the pretracheal fascia and pretracheal space is the aorta, pericardium, and parietal pleura at the level of the carina Perforations of the trachea and lateral pharyngeal walls allow infections to enter this space The infections can descend into the anterior mediastinum, causing purulent pericarditis, empyema, and mediastinitis The perivascular space is surrounded by the carotid sheath This sheath invests the internal and common carotid arteries, the internal jugular vein, and the vagus nerve The cervical sympathetic trunk lies behind, but not within, the sheath The areolar tissue of the sheath separates and invests the nerves and vessels mentioned The sheath is adherent to the thyroid, the sternocleidomastoid muscle, and prevertebral fascia The sheath blends with the fascia of the stylohyoid and digastric muscles in the upper neck and is attached to the base of the skull at the jugular foramen and carotid canal Infections can track along the vascular structures through this space to reach the mediastinum and pleural spaces The retrovisceral (or retropharyngeal) space is the largest and most important interfascial interval in the neck, when considering a pathway of infection for DNM This potential space is areolar in nature and is bordered by the buccopharyngeal fascia anteriorly and the prevertebral fascia posteriorly The lateral borders are the carotid sheaths The potential space extends from the base of the skull superiorly to the posterior mediastinum inferiorly This loose areolar connective tissue supports the movements of the pharynx during swallowing The retopharyngeal lymph nodes are in the lateral aspects of the retrovisceral space near the base of the skull These nodes are part of the superior deep cervical chain The retrovisceral space is further divided by the alar fascia The alar fascia is usually delicate; however, it sometimes is a definitive fascial plane that, along with the prevertebral fascia, descends to the seventh cervical vertebra The retrovisceral space communicates with the lateral pharyngeal spaces above via the styloglossus muscles This muscular avenue allows infections from the base of the tongue, the teeth, the tonsils, and the pharynx, which erupt into the lateral pharyngeal space, to descend into the retrovisceral space and beyond This is the most common route for descending mediastinitis Descending Necrotizing Mediastinitis / 443 The submandibular space spans from the floor of the mouth to the hyoid bone The mandible provides the anterior and lateral borders of this space with the superficial layer of the deep cervical fascia bordering the space inferiorly The mylohyoid muscle crosses the mandible and is responsible for directing the spread of dental infections, most importantly those from abscesses of the second and third molars The buccopharyngeal gap is a connection between the submandibular and lateral pharyngeal spaces that results when the styloglossus muscle passes between the middle and superior pharyngeal constrictors Infections from the submandibular space may travel the styloglossus into the lateral pharyngeal space, which in turn communicates with all the major spaces of the neck In 1836, Wilhelm von Ludwig described a gangrenous infection of the neck characterized by cellulitis, chest pain, and asphyxiation as a result of a submandibular abscess Today, Ludwig’s angina implies bilateral infections involving this space The lateral pharyngeal (or parapharyngeal) space, as mentioned, communicates with all major spaces of the neck The space is defined by the skull above, the hyoid below, the prevertebral fascia posteriorly, and the buccinator and superior pharyngeal constrictor muscles anteriorly The lateral borders are the parotid glands and the mandible The anterior aspects of the lateral pharyngeal space contain lymph nodes and fat The posterior aspects of this space include cranial nerves IX, X, XI, and XII as well as the carotid artery and jugular vein When infection invades this space, symptoms of trismus or cranial nerve palsies suggest its involvement Pathophysiology The potential spaces defined by the major fascial planes of the cervicomediastinal region provide a pathway for the spread of infection into the mediastinum Once a cervical infection has been established, invasion of any or all of these spaces can lead to life-threatening DNM Approximately 70% of the reported cases of DNM are thought to have spread via the retrovisceral space, 20% through the perivascular space, and the rest by way of the pretracheal space , , , , – Gravity and the negative intrathoracic pressure favor the descent of infection into the mediastinum Odontogenic infection is the most common source for DNM.1,3,4,6 Other reported sources include peritonsillar abscess, retropharyngeal abscess, epiglottitis, parotitis, lymphadenitis, and trauma.19–26 Some reports have implicated endotracheal intubation resulting in tracheal or esophageal injury as a source for infection leading to DNM.9–11 Commonly, odontogenic abscesses from second and third molars rupture into the submandibular or lateral pharyngeal space The infection progresses through the fascial lined spaces, typically tracking into the retrovisceral space on the way to the posterior mediastinum Iatrogenic pharyngeal injuries can introduce infection directly into the retrovisceral space Most of these infections are polymicrobial, representing the normal bacterial flora of the mouth and pharynx These organisms have the potential to become virulent and can be invasive when normal barriers have been broken Most odontogenic infections are polymicrobial, comprised of organisms that reflect the indigenous microflora of the oropharyngeal cavity Anaerobes generally outnumber the aerobes by a factor of 10:1 15,27,28 Synergistic effects in this mixed population can promote invasiveness and contribute to the virulence of these infections Some gram-positive cocci and gram-negative rods can cause tissue damage by gas production, resulting in a nonclostridial gas gangrene The organisms isolated in the various case reports on DNM reflect this polymicrobial oropharyngeal population Common isolates recovered from the mediastinum were Bacteroides fragilis, Enterobacter cloacae, Escherichia coli, Serratia marcescens, Staphylococcus aureus, and ␤-hemolytic streptococcus species.3,4,27,28 Other series identified Pseudomonas aeruginosa, Klebsiella pneumoniae, Peptostreptococcus, Actinomyces, and Clostridia.14,29 The necrotizing infections in the cervicomediastinal region share similarities with other necrotizing soft tissue infections in the body The infections are rapid, progressive, and lethal The infections are polymicrobial, synergistic, and virulent Some microorganisms possess virulence factors that enhance the fulminant nature of the process Fascial planes provide means of travel from space to space and cavity to cavity Patients with impaired immune systems (diabetics, alcoholics, patients with acquired immunodeficiency syndrome, cancer patients) are at increased risk for this necrotizing infection The infections spread rapidly, resulting in septic shock, multisystem organ failure, and death The aggressive nature of this type of infection requires an astute clinician to rapidly diagnose the condition and respond with an aggressive surgical approach Clinical Manifestations The clinical presentation of DNM depends on the origins of the infection and the time course of the disease process An odontogenic infection is the most common cause of DNM, followed by peritonsillar and retropharyngeal abscesses Abscess from a second or third molar is the odontogenic infection most typically reported in association with DNM The majority of these odontogenic infections are successfully treated by root canal, 444 / Advanced Therapy in Thoracic Surgery tooth extraction, or other periodontal procedures before they become serious However, a rare few become a rapidly progressing life-threatening infection Patients initially present to a local dentist or emergency room Early symptoms include fever and pain If the infection persists and abscess develops, signs and symptoms might suggest the fascial space that is involved The submandibular space may be involved when an abscessed molar erupts The patient might experience mouth pain, dysphagia or drooling The tongue and floor of the mouth can swell with infection and associated edema The tongue may be displaced anteriorly, and as the infection migrates into the neck, patients will complain of neck pain and stiffness Involvement of both submandibular spaces is referred to as Ludwig’s angina Airway compromise is the most common cause of early death in this rare infection Infection involving the lateral pharyngeal space may cause pain, neck swelling, and trismus The sternocleidomastoid muscle can become involved, making rotation of the neck painful and difficult If the perivascular space becomes involved, palsies of cranial nerves IX to XII may be apparent, and in severe cases, Horner’s syndrome may develop Thrombophlebitis of the jugular vessels is possible, as is carotid artery erosion or thrombosis.30 The infection progresses if untreated or if the patient does not respond to initial therapy The infection can migrate from the portal of infection into one of the three critical fascial spaces: the pretracheal space, the perivascular space, or the retrovisceral space Having reached one of these spaces, the infection can continue to descend, aided by both gravity and the negative pressure generated by the thorax, into the mediastinum As mediastinitis begins, the inflammatory process causes swelling of the mediastinal structures The pericardium may be involved, resulting in pericarditis or pericardial effusion The pleural spaces may develop sympathetic effusions or become directly infected Empyemas may develop Pneumomediastinum or pneumothorax may also be part of the evolving inflammatory and infective process leading to hemodynamic derangement The localized infection may evolve into a systemic process Tachycardia, tachypnea, pyrexia, and leukocytosis suggest systemic inflammatory response syndrome (SIRS) This syndrome evolves from a complex sequence of events initiated by proinflammatory cytokines such as tumor necrosis factor, interleukin (IL)-1, IL-2, IL-6, and interferon-␥ The body attempts to regulate this initial response with a series of anti-inflammatory cytokines and other soluble factors Secondary mediators such as nitric oxide, thromboxanes, leukotrienes, plateletactivating factor, prostaglandins, and the complement system are also triggered as SIRS progresses This milieu of primary and secondary mediators can cause endothelial cell damage leading to tissue and organ damage As the inflammatory process tumbles out of control, endorgan dysfunction, sepsis, septic shock, and hemodynamic collapse are the result Multiorgan system failure and death is the final course if the inflammatory process cannot be arrested and reversed.31–33 Diagnosis When called to evaluate a patient with suspected DNM, the patient is usually late in the course of the disease process Mediastinal sepsis is the reason these patients are so ill; however, the diagnosis is delayed because the clinical presentation suggests an infection isolated to the head and neck Mediastinitis can present as quickly as 24 to 48 hours after the odontogenic infection or procedure, or it may take to weeks before mediastinitis evolves 3–5 Many of these patients are elderly or immunocompromised Some have undergone surgical drainage of the neck but have not improved or have worsened A thorough and detailed history usually reveals a recent dental procedure or cervical trauma or procedure involving the hypopharynx or aerodigestive tract On initial inspection, the patient is generally lethargic with uncontrolled fevers despite the use of broadspectrum antibiotics On physical exam there may be severe pain and induration over the neck, trismus, and deviation of the tongue In the most severe cases, airway compromise secondary to cervical swelling and venous congestion may necessitate intubating the patient In patients who have undergone prior surgical intervention, there may be pus emanating from the wounds even though drains are still in place Further evaluation may demonstrate chest wall crepitus and chest pain in those with a delayed diagnosis Decreased breath sound may be present at the bases secondary to pleural effusion Blood work should include complete blood count, electrolytes, and blood cultures If SIRS or sepsis is evident, liver function tests, arterial blood gas, and lactic acid level will be helpful Invasive hemodynamic monitoring including arterial access, central access, and pulmonary artery catheter may be indicated Cultures usually reveal a mixed flora of anaerobic and aerobic bacteria and, in many cases, fungal elements Broad-spectrum antibiotics, resuscitation, and supportive measures are necessary as the diagnostic process continues A chest radiograph is always part of the work-up Findings may include widening of the superior mediastinal shadow, widening of the retrocervical space, anterior displacement of the trachea, mediastinal emphysema, and air–fluid levels However, these findings are difficult to recognize when the process presents late The single Descending Necrotizing Mediastinitis / 445 most important and influential diagnostic evaluation for a patient with suspected DNM is the computed tomography (CT) scan.3–5,15–18 CT scanning allows for evaluation of the mediastinum and surrounding structures (Figure 36-2) The CT will help define the extent of the infection, which is necessary when planning the surgical approach to the mediastinum Frequently, if the disease is diagnosed early, only the upper mediastinum is involved and limited drainage may be appropriate CT scanning is also necessar y to determine if the infection has been adequately drained after initial intervention since fevers and signs of sepsis may lag a progressing infection for days One retrospective review averaged six CT scans per patient after the initial surgical intervention.18 Although laboratory testing with a complete blood count and cultures are important, interval resolution of fevers and sepsis may not completely correlate with adequate drainage Treatment DNM is an aggressive infection with reported mortalities of 40 to 50% in the postantibiotic era.1,3,4 If left untreated, it is universally fatal With aggressive surgical intervention, the mortality can be reduced to less than 20%.5,15–18 This improvement in survival is the result of earlier diagnosis, broad-spectrum antibiotic therapy, and thorough surgical débridement and drainage Early in the course of DNM, the clinical picture is similar to that of a localized acute cervical infection The clinical progression is dramatically different with DNM, resulting in death if left untreated Clinical suspicion is necessary to recognize this disease early in its evolution, thereby improving the chances of arresting and reversing the inflammatory process The CT scan is the critical diagnostic study that differentiates DNM from a localized cervical infection The CT defines the extent of the disease process and the amount of destruction of medi- A C B D FIGURE 36-2 Computed tomography scans of the neck of a 40-year-old man days after self-extracting a lower second molar with a pair of pliers The patient had fevers, shaking chills, and woody edema of the neck A, Air in the deep neck spaces B, Mediastinal air–fluid levels C, Mediastinal inflammation and edema D, Bilateral effusions The patient underwent tracheostomy, cervical drainage, and mediastinal drainage through a transhiatal approach The patient survived to be discharged 42 days after original presentation 446 / Advanced Therapy in Thoracic Surgery astinal tissue Determining the extent of the disease is important because this dictates the magnitude of surgical débridement.15–18 The single most important aspect of successfully treating a patient with DNM is that of adequate surgical drainage and débridement Patients typically present well into the course of DNM They are often septic or in septic shock Some will have been inadequately drained Initial treatment involves resuscitation and stabilization The inflammation, venous congestion, and edema cause dramatic neck swelling that threatens the patient’s airway This swelling and edema is unpredictable, and should the airway be lost, regaining control would be virtually impossible For this reason, early tracheostomy is felt to be essential and is recommended in many reports.3,4,34 Others feel a surgical airway should be used selectively.35 Broad-spectrum antibiotics against aerobic and anaerobic organisms should be administered until cultures are returned and appropriate directed therapy can begin Once the patient has been stabilized, definitive surgical débridement and drainage can be performed The CT scan defines the extent of the infection and directs the surgical approach For deep neck infections, the surgical approach to the various spaces is well established The surgical approach to DNM, however, remains controversial When the infection is above the tracheal bifurcation anteriorly or the fourth thoracic vertebra posteriorly, most agree that cervical drainage is the appropriate first step as long as all fascial planes are opened and drained adequately A unilateral or bilateral approach may be used, and soft drains are recommended to minimize the potential of erosion into vascular structures.36 When the infection descends below these margins, a more aggressive surgical approach is required The CT scan will identify those patients with more extensive mediastinitis For these patients, in addition to the cervical drainage, drainage and débridement of the mediastinum is required When the DNM is limited to the anterior mediastinum, a subxyphoid approach may be used A subxyphoid incision is used to access the anterior mediastinum.3 Blunt manual retrosternal dissection is followed by placement of a retrosternal drain The anterior mediastinum may also be reached via a parasternal incision, standard thoracotomy,37–39 or even mediansternotomy Transpleural drainage violates and contaminates the pleural space The posterolateral thoracotomy allows access to the ipsilateral mediastinum, the pleural spaces, and the prevertebral and paraesophageal spaces.15,17,18,27 Controversy exists as to whether it is necessary to access and drain the pleural spaces The pleural effusions are often sympathetic, and entering theses spaces risks contaminating them, thereby enhancing the risk of empyema formation.3,27 One approach to avoid contaminating these spaces is to sample the fluid by thoracentesis A sterile collection can be safely observed Other recent reports have suggested thoracoscopic or CT-guided percutaneous drainage as an alternative to the aggressive surgical approach.41,42 The necrotizing nature of this infection mandates débridement of involved tissue, and this cannot be adequately achieved with a minimalist approach Endo and colleagues have suggested a classification of DNM to help direct the treatment of this lethal infection For infections localized to the space above the carina, cervical drainage could be used to treat the infection For infections that have involved the anterior mediastinum alone, a subxyphoid approach could be added For diffuse mediastinitis that involves the anterior and posterior mediastinum, a thoracotomy needs to be part of the surgical approach.43 Results DNM is perhaps the most aggressive and lethal form of mediastinitis Early case reports and reviews suggested a mortality rate approaching 50%.1 Even into the postantibiotic era, the mortality rate remained near 40% 3–5 With the widespread use of CT scanning, the diagnosis of DNM became better defined and more easily recognized Early use of this modality has been stressed by various reviews as a crucial and necessary step in improving the outcome of this fulminant infection Along with early diagnosis, it has also become clear that aggressive and repeated surgical drainage and débridement is required to improve the survival of this group of patients Employing this approach, reviews over the past decade have demonstrated mortality rates of to 23%.5,15–18 DNM remains a rare but aggressive form of mediastinitis With early recognition, adequate surgical drainage, and appropriate antibiotics, the disease may be successfully treated References Pearse HE Jr Mediastinitis following cervical suppuration Ann Surg 1938;107:588–611 Liptay MJ, Fry WA, Shields TW Acute and chronic mediastinal infections In: Sheilds TW, LoCicero III J, Ponn RB, editors General thoracic surgery 5th ed Philadelphia (PA): Lippincott Williams and Wilkins; 2000 p 2093–104 Estrera AS, Landay MJ, Grisham JM, et al Descending necrotizing mediastinitis Surg Gynecol Obstet 1983;157:545–52 Wheatley MJ, Stirling MC, Kirsh MM, et al Descending necrotizing mediastinitis: transcervical drainage is not enough Ann Thorac Surg 1990;49:780–4 Descending Necrotizing Mediastinitis / 447 Corsten MJ, Shamji FM, Odell PF, et al Optimal treatment of descending necrotizing mediastinitis Thorax 1997;52:702–8 Moreland LW, Corey J, McKenzie R Ludwig’s angina: a case report and review of the literature Arch Intern Med 1988;148:461–6 Snow N, Lucas AE, Grau M, Steiner M Purulent mediastinal abscess secondary to Ludwig’s angina Arch Otolaryngol 1983;109:53–5 Bounds GA Subphrenic and mediastinal abscess formation: a complication of Ludwig’s angina Br J Oral Maxillofac Surg 1985;23:313–21 Uram J, Hauser MS Deep neck and mediastinal necrotizing infection secondary to a traumatic intubation J Oral Maxillofac Surg 1988;46:788–91 10 Wolff AP, Kuhn FA, Ogura JH Pharyngeal-esophageal perforations associated with rapid oral endotracheal intubation Ann Otol Rhinol Laryngol 1972;81:258–61 11 Seaman M, Ballinger P, Sturgill TD, Maertins M Mediastinitis following nasal intubation in the emergency department Am J Emerg Med 1991;9:37–9 22 Zachariades N, Mezitis M, Stavrinidis P, KonsolakaAgouridaki E Mediastinitis, thoracic empyema and pericarditis as complications of a dental abscess: report of a case J Oral Maxillofac Surg 1988;46:493–5 23 Guardia SN, Cameron R, Phillips A Fatal necrotizing mediastinitis secondary to acute suppurative parotitis J Otolaryngol 1991;20:54–6 24 Chong WH, Woodhead MA, Millard FJC Mediastinitis and bilateral thoracic empyema complicating adult epiglottitis Thorax 1990;45:491–2 25 Watanabe M, Ohshika Y, Aoki T, et al Empyema and mediastinitis complicating retropharyngeal abscess Thorax 1994;49:1179–80 26 Zeitoun IM, Dhanarajani PJ Cervical cellulitis and mediastinitis caused by odontogenic infections: report of two cases and a review of literature J Oral Maxillofac Surg 1995;24:60–3 27 Marty-Ane C-H, Berthet J-P, Alric P, et al Management of descending necrotizing mediastinitis: an aggressive treatment for an aggressive disease Ann Thorac Surg 1999; 68:212–7 12 Woodburne RT, Burkel WE Essentials of human anatomy New York: Oxford University Press; 1988 28 Sancho LMM, Minamoto H, Fernandez A, et al Descending necrotizing mediastinitis: a retrospective surgical experience Eur J Cardiovasc Surg 1999;16:200–5 13 Moncada R, Warpeha R, Pickleman J, et al Mediastinitis from odontogenic and deep cervical infection: anatomic pathways of propogation Chest 1978;73:497–500 29 Brook I, Frazier E Microbiology of mediastinitis Arch Intern Med 1996;156:333–6 14 Sakamoto H, Takayuki A, Yoshifumi K, et al Descending necrotizing mediastinitis due to odontogenic infections Oral Surg Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;89:412–9 15 Papalia E, Rena O, Oliaro A, et al Descending necrotizing mediastinitis: surgical management Eur J Cardiothorac Surg 2001;20:739–42 16 Kiernan PD, Hernandez A, Byrn WD, et al descending cervical mediastinitis Ann Thorac Surg 1998;65:1483–8 17 Mitjans MS, Sanchis JB, Padro XB, et al Descending necrotizing mediastinitis Int Surg 2000;85:331–5 18 Freeman RK, Vallieres E, Verrier ED, et al Descending necrotizing mediastinitis: an analysis of the effects of serial surgical debridement on patient mortality J Thorac Cardiovasc Surg 2000;119:260–7 30 Mathieu D, Neviere R, Teillon J, et al Cervical necrotizing fasciitis: clinical manifestations and management Clin Infect Dis 1995;21:51–6 31 Oberholzer A, Oberholzer C, Moldawer LL Sepsis syndromes: understanding the role of innate and acquired immunity Shock 2000;14:59023 32 Pinsky MR Organ-specific therapy in critical illness: interfacing molecular mechanisms with physiological interventions J Crit Care 1996;11:95–107 33 Livingston DH Management of the surgical patient with multiple system organ failure Am J Surg 1993;165(2A Suppl):8S–13S 34 Alexander DW, Leonard JR, Trail ML Vascular complications of deep neck abscesses A report of four cases Laryngoscope 196;78:361–70 19 Levine TM, Wurster CF, Krepsi YP Mediastinitis occurring as a complication of odontogenic infections Laryngoscope 1986;96:747–50 35 Allen D, Loughnan TE, Ord RA A reevaluation of the role of tracheotomy in Ludwig’s angina J Oral Maxillofac Surg 1985;43:436–9 20 Steiner M, Grau MJ, Wilson DL, Snow NJ Odontogenic infection leading to cervical emphysema and fatal mediastintis J Oral Maxillofac Surg 1992;40:600–4 36 Ichimura H, Ishikawa S, Hiramatsu Y, et al Innominate artery rupture after transcervical draining for descending necrotizing mediastinitis Ann Thorac Surg 2001;71:1028–30 21 Rubin MM, Cozzi GM Fatal necrotizing mediastinitis as a complication of an odontogenic infection J Oral Maxillofac Surg 1987;45:529–33 37 Brunelli A, Sabbatini A, Catalini G, Fianchini A Descending necrotizing mediastinitis: cervicotomy or thoracotomy? J Thorac Cardiovasc Surg 1996;111:485–6 448 / Advanced Therapy in Thoracic Surgery 38 Kim JT, Kim KH, Lee SW, Sun K Descending necrotizing mediastinitis: mediastinal drainage with or without thoracotomy Thorac Cardiovasc Surg 1999;47:333–5 41 Roberts JR, Smythe WR, Weber RW, et al Thorascopic management of descending necrotizing mediastinitis Chest 1997;112:850–4 39 Ris HB, Banik A, Furrer M, et al Descending necrotizing mediastinitis: surgical treatment via a clamshell approach Ann Thorac Surg 1996;62:1650–4 42 Gobien RP, Stanley JH, Gobien BS, et al Percutaneous catheter aspiration and drainage of suspected mediastinal abscess Radiology 1984;151:69–71 40 Izumoto H, Komoda K, Okada O, et al Successful utilization of the mediansternotomy approach in the management of descending necrotizing mediastinitis: report of a case Surg Today 1996;26:286–8 43 Endo S, Murayama F, Hasegawa T, et al Guideline of surgical management based on diffusion of descending necrotizing mediastinitis Jpn J Thorac Cardiovasc Surg 1999;47:14–9 CHAPTER 37 DIAPHRAGMATIC PACING RICHARD B THOMPSON, MD THOMAS A D’AMICO, MD In his 1777 treatise on the uses of electricity for human ailments, Cavallo noted that electricity may be of use to assist respiration However, it was not until after Galvani discovered animal electricity in 1787, that Ure in 1818 applied Galvanic electricity to the phrenic nerve of a recently hanged criminal and first observed the powerful contractions that resulted “The chest heaved and fell; the belly was protruded and again collapsed, with the relaxing and retiring diaphragm.”1 Sarnoff and colleagues coined the term “electrophrenic respiration” as he began to investigate what is now recognized as diaphragmatic pacing as a possible treatment to aid respiration in victims of bulbar poliomyelitis.2,3 Sarnoff ’s early work remains critical to the understanding of diaphragmatic pacing; however, Glenn and colleagues and Farmer and colleagues are credited for developing the modern systems, employing chronic diaphragmatic stimulation using radiofrequency signals to stimulate phrenic nerves through intact skin.4,5 Moreover, this group initiated investigations regarding muscle fatigue and conditioning, and they are responsible for the implementation of safety guidelines for effective use of diaphragmatic pacing 55% slow-twitch fatigue-resistant, 21% fast-twitch fatigueresistant, and 24% fast-twitch fatigable.7 Although diaphragmatic dysfunction may be the result of a variety of disease processes, only a select few are eligible for treatment with diaphragmatic pacing An intact peripheral (phrenic) nerve and muscle (diaphragm) are prerequisites for potential pacing candidates Some patients with intrinsic paralysis of the accessory respiratory muscles (high cervical spine injuries) and those with central alveolar hypoventilation (Ondine’s curse) may be candidates for diaphragmatic pacing Patients with respiratory insufficiency secondary to lower motor neuron (phrenic nerve) dysfunction, amyotrophic lateral sclerosis, muscular dystrophy, or extensive pulmonary parenchymal disease have traditionally not been candidates Although the potential patient pool is limited, diaphragmatic pacing offers a clear advantage in quality of life, compared with traditional ventilatory systems, for appropriately selected patients.8 Apparatus There are currently three commercially available diaphragmatic pacing devices Though each has unique characteristics, there are four common basic components: receiver, electrode, antennae, and transmitter The receiver and electrode assembly require permanent surgical implantation, whereas the transmitter and antennae are external devices The basic components of a typical device are detailed in Figure 37-1 (Avery Laboratories of Farmingdale, New York) The subcutaneously implanted receiver transforms radiofrequency signals from the transmitter into electrical impulses carried to electrodes placed in the proximity Diaphragmatic Pacing Long-term electrical stimulation of motor nerves has been successful only in pacing the diaphragm The durability and fatigue resistance of the diaphragm is owing to its high oxidative capacity and blood flow The critical conditioning phase of diaphragmatic pacing further molds the muscle into an even more fatigue-resistant state through long-term application of slow stimulation frequencies.6 Normal diaphragmatic muscle fiber has been shown to be 449 CHAPTER 40 OPTIMAL MANAGEMENT OF BARRETT’S ESOPHAGUS ALAN G CASSON, MB, CHB, MSC, FRCSC, FACS convincing biological mechanism explaining these associations has yet been established Despite recent advances in multimodality therapy, the prognosis for invasive esophageal adenocarcinoma remains poor, with 5-year survival consistently reported below 20%.7 This generally reflects advanced tumor stage at diagnosis and, possibly, an aggressive tumor biology It therefore seems likely that progress with esophageal adenocarcinoma will only be made with the development of strategies for early detection, prevention, and an improved understanding of its etiology and tumor biology Indeed, Barrett’s esophagus is an excellent model to investigate the development and progression of a clinically relevant human malignancy, as the esophagus is relatively accessible by esophagogastroscopy and biopsy Endoscopic surveillance of patients with Barrett’s esophagus is currently used for the early diagnosis of malignancy, and for patients with biopsy-proven high-grade dysplasia, early intervention (usually surgical resection) is reported to improve long-term survival Barrett’s esophagus, an acquired condition in which normal esophageal squamous epithelium is replaced with metaplastic columnar mucosa, has remained a controversial diagnosis for several decades This chapter reviews current approaches to the diagnosis and management of patients with a columnar epithelium-lined (Barrett’s) esophagus New strategies and management of this entity will also be discussed Background Over the past several decades, there has been a dramatic change in the epidemiology of esophageal malignancy in North America, with the incidence of adenocarcinomas of the lower esophagus and esophagogastric junction (EGJ) increasing at a rate in excess of that for any other solid tumor.1,2 Although the reasons for this change are largely unknown and remain controversial, several lifestyle risk factors, including alcohol consumption, tobacco exposure, obesity, and dietary factors, have been proposed , Recent studies have implicated gastroesophageal reflux disease (GERD) as a significant risk factor for esophageal adenocarcinoma.5,6 It is hypothesized that GERD results in acute mucosal injur y (esophagitis), promotes cellular proliferation, and induces columnar metaplasia of the normal squamous epithelium lining the esophagus The resulting columnar epithelium-lined (Barrett’s) esophagus appears predisposed to develop malignancy Dysplasia is widely regarded as the precursor of invasive cancer, and highgrade dysplasia in Barrett’s epithelium is frequently associated with esophageal adenocarcinoma However, despite the plausibility of a link between GERD, Barrett’s epithelium, and esophageal adenocarcinoma, no Historical Perspective An appreciation of current controversies surrounding Barrett’s esophagus may be derived from briefly reviewing its history Although peptic ulceration of the columnar epithelium-lined esophagus was reported as early as 1906,8 recent interest in this condition dates from the early 1950s Barrett initially reported peptic ulceration in a tubular segment of foregut lined by columnar epithelium.9 As his definition of the esophagus required the lining to have squamous epithelium, he initially proposed (incorrectly) that this tubular segment was stomach In 1951, Bosher and Taylor described goblet cells in colum479 480 / Advanced Therapy in Thoracic Surgery nar epithelium lining the lower esophagus,10 and this observation was confirmed by Morson and Belcher in the following year, who also reported an association with esophageal adenocarcinoma.11 Although it was Allison who, in 1953, proposed the existence of a columnar epithelium-lined esophagus,12 the term Barrett’s esophagus persisted for what Barrett later (in 1957) described as the lower esophagus lined by columnar epithelium.13 This entity has remained controversial ever since Barrett’s esophagus was initially presumed to be of congenital origin, but subsequent reports suggested the columnar epithelium is acquired as a result of reflux esophagitis The association with gastroesophageal reflux has now clearly been established by several careful human and experimental studies since the 1970s Further insight into the histology of the columnar-lined esophagus followed the report of Paull and colleagues in 1976.14 By manometric-guided biopsy, three distinct epithelial types (junctional or cardia-type, fundic-type, and specialized columnar epithelium) were found in different zones of the lower esophagus Careful pathologic studies subsequently demonstrated that adenocarcinomas arose from specialized metaplastic columnar epithelium (progressing through dysplasia), to establish the now recognized association between Barrett’s esophagus and malignancy Over the last two decades, prompted by epidemiologic studies reporting increasing rates for esophageal adenocarcinomas, there has been renewed interest in Barrett’s esophagus With improved results, particularly reduced mortality rates, following esophageal resection, endoscopic surveillance to detect early stage disease has been critically evaluated and generally accepted into current clinical practice for detection of early stage disease In parallel, novel, less-invasive, organ-preserving (ie, mucosal ablation) approaches have been explored by several groups Recent advances in molecular technology have stimulated considerable interest in the fundamental biology of this disease Diagnosis The diagnosis of Barrett’s esophagus currently requires both careful endoscopic and histologic evaluation Critical to endoscopic evaluation of the upper gastrointestinal tract is the recognition of the squamocolumnar junction and the anatomic EGJ However, vague and conflicting terminology related to the anatomy of this region of the foregut has caused much confusion among clinicians and pathologists as to what precisely constitutes a columnar epitheliumlined (Barrett’s) esophagus With increasing interest in this entity, further attempts to refine the diagnosis of Barrett’s esophagus have incorporated key histologic characteristics (intestinal metaplasia) into recent definitions.15 Endoscopic Evaluation Widespread clinical application of endoscopy (esophagogastroscopy) generally led investigators to define Barrett’s esophagus on the basis of an arbitrary minimum length of columnar mucosa (generally cm, but ranging from to cm) above the anatomic EGJ Whereas the endoscopic identification of the squamocolumnar junction was rarely difficult, definition of the anatomic EGJ has caused considerable confusion Few studies have critically addressed clinically relevant techniques to identify the EGJ the esophagogastric junction Classic anatomical studies of the EGJ alluded to various structures such as the phrenoesophageal ligament and the peritoneal reflection and to changes in the character of muscular layers as the tubular esophagus enters the stomach Although these structures may be helpful to define the EGJ at surgery, they have limited clinical application for the majority of patients Further attempts to define the EGJ physiologically using esophageal manometry defined a high-pressure zone, but this physiologic lower esophageal sphincter (LES) did not generally correspond to any consistent anatomic structure Considerable individual variation in “normal” anatomy and the dynamic nature of the EGJ on swallowing have limited precise radiographic definitions In current clinical practice, endoscopic evaluation of the upper gastrointestinal tract is widely utilized, and the EGJ may be considered as the point at which the tubular esophagus dilates to become the stomach Practically, it may be difficult to precisely identify the point at which the esophagus ends and the stomach begins, especially with peristalsis and in the presence of a hiatus hernia In this situation, a useful endoscopic landmark for the EGJ corresponds to the proximal margin of the gastric folds the cardia (and related terminology) The use of the term “cardia” is also inconsistent Anatomists have defined a cardiac region of the proximal stomach, below or distal to the EGJ, lined by oxyntic mucosa (comprising acid- and pepsin-secreting cells) This is quite distinct from current clinical terminology related to the “cardia,” which refers, somewhat vaguely, to the general region of, or directly above, the EGJ, which is lined by cardia-type columnar mucosa (characterized by mucus-secreting cells) There is considerable debate as to whether cardia-type mucosa is a normal finding, particularly in children, or whether it represents the earliest reflux-induced metaplastic tissue, with potential to develop intestinal metaplasia Optimal Management of Barrett’s Esophagus / 481 Histology Recognizing the limitations of endoscopy and the importance of identifying specialized intestinal metaplasia (because of its association with malignancy) in the columnar-lined esophagus, recent attempts to define Barrett’s esophagus have incorporated tissue histology in the diagnosis The American College of Gastroenterology currently defines Barrett’s esophagus as “a change in the esophageal epithelium of any length that can be recognized at endoscopy and is confirmed to have intestinal metaplasia by biopsy.”16 This is not without difficulty, however, as foci of intestinal metaplasia are found relatively frequently in short segments (< cm) of columnar mucosa lining the lower esophagus and even in biopsies from the “normal” EGJ (or cardia) intestinal metaplasia Goblet cells are the hallmark of intestinal metaplasia and must be diagnosed histologically using hematoxylin and eosin staining of tissue sections Alcian blue staining may be used to complement the histologic diagnosis, as goblet cells contain acidic mucin, resulting in an intense blue color when stained with this agent at pH 2.5 Light blue staining may be seen with some reactive conditions, but in the absence of distinctive goblet cell morphology, should not be confused with specialized intestinal metaplasia The pathogenesis and natural history of intestinal metaplasia is not clearly understood Studies have reported that the frequency of intestinal metaplasia varies with the length of columnar epithelium lining the lower esophagus, with a prevalence of over 90% in patients with > cm (ie, classic long-segment) of columnar mucosa.17 However, it is increasingly recognized that endoscopic biopsies taken from the distal (< cm) tubular esophagus will have histologic evidence of intestinal metaplasia in 10 to 35% of cases.18 These observations have given rise to the entity of “short segment” Barrett’s esophagus, which is also reported to be associated with GERD and to have malignant potential.19 Furthermore, microscopic foci of intestinal metaplasia, within cardia-type mucosa, are also reported in up to 35% of biopsies taken from patients with a normalappearing EGJ.20 This entity has been referred to (variably) as “cardia intestinal metaplasia,” and the significance of this finding is extremely controversial.21 While a number of reports suggest that cardia intestinal metaplasia may represent an early manifestation of GERD (consequently associated with Barrett’s esophagus), others have reported stronger associations with chronic gastritis, Helicobacter pylori infection, gastric intestinal metaplasia, and gastric malignancy.22 The asso- ciation with H pylori is particularly interesting, as gastric infection with cagA-positive strains has been identified to have an inverse association with esophageal adenocarcinoma development.23 For practical purposes, intestinal metaplasia of the esophagus and cardia cannot be distinguished by routine histology Recent immunohistochemical studies, using monoclonal antibodies to cytokeratin and 20, have reported that it is possible to accurately differentiate between esophageal and cardia intestinal metaplasia.24 In esophageal intestinal metaplasia, cytokeratin positivity is found in superficial and deep glands, whereas cytokeratin 20 positivity is limited to superficial glands only (Barrett’s cytokeratin 7/20 pattern) In cardia intestinal metaplasia, cytokeratin immunoreactivity is absent (or weak or patchy), but cytokeratin 20 positivity is seen in both superficial and deep glands Although early reports suggest cytokeratin or 20 immunostaining patterns to be highly sensitive and specific, routine application of these immunohistochemical techniques require further critical evaluation, and interpretation of cytokeratin immunoreactivity should be in conjunction with current clinical, endoscopic, and histologic findings dysplasia Dysplasia may be identified histologically on the basis of phenotypic nuclear alterations resulting from deoxyribonucleic acid (DNA) abnormalities, and is graded as low-grade, high-grade, or indefinite Low-grade dysplasia is generally distinguished from high-grade dysplasia based on nuclear localization in relation to the luminal surface of the cell The histologic diagnosis of dysplasia is largely subjective and considerable intra- and interobserver variation is reported Furthermore, dysplastic change should be interpreted with caution when atypical epithelial cells (arising from a background of active inflammation) are present Dysplastic change, particularly high-grade dysplasia, in the columnar epithelium-lined esophagus is currently regarded as the most reliable predictor for invasive cancer and is frequently associated with invasive adenocarcinoma.25,26 However, as the natural history of dysplasia is not known with certainty, this has a number of implications for endoscopic surveillance adenocarcinomas of the esophagus and cardia The association between Barrett’s esophagus and malignancy is well established While adenocarcinoma is the principal histologic subtype, it is important to note that a limited number of studies have reported the development of squamous cell carcinomas in association with Barrett’s epithelium 482 / Advanced Therapy in Thoracic Surgery The classification of adenocarcinomas of the lower esophagus and EGJ (or cardia) is controversial, as is the staging of these tumors The classification proposed by Siewert has seen recent increased clinical application, particularly in Europe.27 It is based on an estimate of the tumor centre in relation to the EGJ and is therefore applicable to adenocarcinomas arising within cm of the EGJ, as follows Adenocarcinomas of the lower esophagus (Type I) arise from to cm above the EGJ; cardia adenocarcinomas (Type II) arise from the region cm above to cm below the EGJ; and subcardia gastric adenocarcinomas (Type III) arise from to below the EGJ Such measurements may be difficult to estimate for large, advanced stage tumors Over the past decade, we have used strict clinicopathologic criteria to stratify primary esophageal adenocarcinomas (Type I) from adenocarcinomas arising at the EGJ or cardia (Type II).28 Based on clinical, endoscopic, radiologic, operative, and pathologic findings, primary esophageal adenocarcinomas are defined as follows: (1) by the presence of Barrett’s epithelium, (2) when more than 75% of the tumor mass involves the tubular esophagus, (3) by direct invasion of periesophageal tissues histologically, (4) with minimal gastric involvement, and (5) by clinical symptoms of esophageal obstruction (ie, dysphagia) The most important criterion to establish the diagnosis of a primary esophageal adenocarcinoma is the presence of Barrett’s epithelium However, this may not be found in up to 50% of esophageal adenocarcinomas, likely as a consequence of tumor progression Therefore, criteria to should be applied to such “non-Barrett’s” adenocarcinomas, in order to establish their primary esophageal (vs cardia, subcardia, or gastric) origin Recently, molecular studies of adenocarcinomas of the esophagus (Type I) and cardia (Type II) have reported different frequencies of p53 mutations, MDM2 gene amplification, and cytokeratin expression, suggesting these two tumor types may be distinct pathologic entities.29 It is anticipated that such approaches may be useful to establish a precise diagnosis of adenocarcinomas arising in this region of the upper gastrointestinal tract and to direct therapy Current Definitions Current definitions of Barrett’s esophagus, incorporating both endoscopic and histologic findings, are generally applicable in the context of modern clinical practice Classic long-segment Barrett’s epithelium would therefore remain predominantly an endoscopic diagnosis (> cm columnar mucosa above the EGJ) It would be expected that intestinal metaplasia would be confirmed histologically in the majority (> 95%) of biopsies By contrast, short-segment Barrett’s esophagus is primarily a histologic diagnosis following biopsy of columnar epithelium within cm of the EGJ Although the term Barrett’s esophagus (or epithelium or mucosa) clearly is somewhat ambiguous, it is still widely used Descriptive terms such as “columnar epithelium-lined esophagus” (or columnar-lined esophagus) are much more accurate, especially when qualified by a measurement and a histologic description (eg, a cm columnar epithelium-lined esophagus with specialized intestinal mucosa) However, such terminology is linguistically more cumbersome, and it is likely that “Barrett’s esophagus” (qualified as long- or short-segment) will continue to be used in general medical terminology Cardia intestinal metaplasia, or specialized intestinal metaplasia at the EGJ, is difficult to classify as Barrett’s epithelium (even as “ultra-short” Barrett’s), as it is primarily a histologic diagnosis, with normal endoscopic findings It is reasonable to consider cardia intestinal metaplasia as a distinct entity until its etiology and pathogenesis is clarified and the association with malignancy has been precisely defined Epidemiology Estimates of the prevalence and incidence of Barrett’s esophagus have varied between populations and over time, depending on how Barrett’s esophagus is defined In North America, current estimates of the prevalence of long-segment Barrett’s esophagus are approximately to 2%, for all patients undergoing endoscopy This increases to over 10% for patients who undergo endoscopy for upper gastrointestinal symptoms (predominantly reflux-related).30 Autopsy series indicate the prevalence of Barrett’s esophagus to be much higher (approximately 20 times), suggesting that a large percentage of the general population have undiagnosed Barrett’s esophagus.31 This may not be unreasonable considering that up to 40% of the general population in North America experience reflux-related symptoms monthly The mean age at which Barrett’s esophagus is currently diagnosed is around 60 years, with a male-to-female ratio of 3:1 to 2:1 It is currently unclear whether the prevalence of Barrett’s esophagus is increasing or whether this diagnosis is being made more frequently because of widespread use of endoscopy.32 Preliminary data from the United Kingdom, adjusting for increasing numbers of endoscopic procedures, suggest a real increase in prevalence of Barrett’s esophagus.33 Other studies have reported that while the prevalence of long-segment Barrett’s esophagus remains unchanged, it is short-segment Barrett’s esophagus that is increasing 34 Currently, the prevalence of cardia intestinal metaplasia in the general population is not known with any degree of certainty.35,36 Optimal Management of Barrett’s Esophagus / 483 Although esophageal malignancy is relatively uncommon in North America, there has been a marked change in the epidemiology of this disease over the past three decades While the incidence of squamous cell carcinoma has remained steady, incidence rates for adenocarcinomas of the esophagus and EGJ have increased rapidly, particularly for white males.1,2 It is thought to be unlikely that changes in diagnostic practice and reporting have contributed to these trends.36 Although Barrett’s esophagus is likely the precursor lesion for esophageal adenocarcinoma, the risk of cancer development is unclear To date, most studies have reported relative risks of at least 40 times higher than that for the general population.37 However, the absolute risk of an individual patient with Barrett’s esophagus developing invasive adenocarcinoma is low and has recently been estimated at 0.5% per patient year.38 Etiology and Pathogenesis The precise etiology and natural history of Barrett’s esophagus is still unknown Although there is a plausible link between GERD, Barrett’s esophagus, and esophageal adenocarcinoma, no convincing biological mechanism explaining these associations has yet been established The challenge is to correlate histologic progression, reflected by the metaplasia–dysplasia–carcinoma sequence, with molecular alterations, and to incorporate these into current clinical practice.39 Association with GERD It is generally accepted that Barrett’s esophagus is an acquired condition resulting from gastroesophageal reflux Clinical and experimental studies of patients with Barrett’s esophagus (with and without reflux-related symptoms) have demonstrated several physiological abnormalities, including increased acid exposure, a defective LES (length, pressure, or both), and impaired motility and clearance from the esophageal body, compared with normal subjects.18 Reflux of duodenal contents has recently been suggested as an important contributing factor in the pathogenesis of esophagitis and Barrett’s esophagus.40 Pure alkaline reflux esophagitis is thought to be rare, and in most situations, a mixed refluxate (combined acid and alkaline secretions) appears to be important in causing esophageal mucosal damage This is supported experimentally, where mixed reflux was shown to cause esophagitis more frequently than acidic gastric juice, bile, lysolecithin, or pancreatic enzymes (trypsin, lipase, and carbopeptidase) alone As bile is a major component of duodenal juice, the role of bile acids and salts in alkaline reflux esophagitis has been studied extensively Bile acids alter the ionic permeability of mucous membranes, with back diffusion of hydrogen (H+) ions and intracellular acidification presumably resulting in mucosal damage Deconjugated and secondary acids appear to be more harmful than conjugated or primary bile acids, and activity of bile acids appears to be pH dependent pH also affects the activity of associated gastric and duodenal enzymes, with trypsin more proteolytically active in a more alkaline environment (pH to 8) and pepsin in an acidic environment (pH to 5) These may further interact to cause increased mucosal damage In a number of recent clinical studies, patients with Barrett’s esophagus were reported to have increased bile reflux when compared with patients with GERD (or normal subjects), although such patients also had significantly abnormal acid reflux profiles.41 These clinical observations are further supported by experimental animal models of duodenoesophageal reflux.42 To date, three large case-control studies have evaluated the association between GERD and risk of developing esophageal adenocarcinoma 5,6,43 Although individual cancer risk for individuals with GERD is low, patients with long-standing and severe reflux symptoms were reported to be at relatively increased risk (odds ratio of 43.5 in one study).6 Of interest, this association was not seen for esophageal squamous cell carcinoma or for adenocarcinomas of the EGJ (cardia) Association with Intrinsic Esophageal Disorders Barrett’s esophagus is also reported in association with other intrinsic esophageal diseases such as with scleroderma,44 following lye ingestion,45 postgastrectomy,46 and following myotomy for achalasia.47 These disorders may have associated abnormalities at the EGJ predisposing to reflux or, alternatively, reduced intrinsic peristalsis promoting mucosal injury through stasis Barrett’s esophagus has also been documented in association with anticancer chemotherapy.48 Hereditary Factors Although families with Barrett’s esophagus have been reported, this is quite uncommon and a genetic locus for a familial syndrome of Barrett’s esophagus has not been confirmed to date.49 However, the observation of an association between esophageal cancer and the rare autosomal dominant disease tylosis suggested the possibility of hereditary factors underlying esophageal tumorigenesis The tylosis esophageal cancer gene was recently mapped to a small region on chromosome 17q25, and recent loss of heterozygosity (LOH) studies have further implicated this gene in sporadic esophageal tumors.50,51 484 / Advanced Therapy in Thoracic Surgery Molecular Alterations Human tumors are thought to arise as a multistep process, modulated by both genetic and environmental factors The accumulation of genetic alterations leads to genomic instability and, through complex interactions between stimulatory oncogenes and regulatory tumor suppressor genes, results in widespread clonal outgrowth of cells exhibiting aberrant cell cycle regulation, with capacity for invasion In general, genomic instability precedes the appearance of histologic changes Barrett’s adenocarcinomas similarly appear to develop by a multistep process, recognized histologically as the metaplasia–dysplasia–adenocarcinoma sequence.39 Over the past decade, the widespread application of molecular technology has led to several molecular studies of esophageal carcinoma, many with pathological and clinical correlation This has been the subject of several recent reviews 52–54 The following section discusses selected biomarkers that may have potential clinical application in the future management of patients with Barrett’s esophagus proliferation markers Proliferating cell nuclear antigen (PCNA) and Ki67, a cell nuclear proliferation-associated antigen for G 1/S and G2/M of the cell cycle, have been studied in esophageal tissues by immunohistochemistry and flow cytometry PCNA immunostaining is normally seen in the basal layer of metaplastic Barrett’s epithelium but in highgrade dysplasia is seen to extend to more superficial layers.55,56 Immunohistochemical studies using the monoclonal antibody MIB-1 (against Ki67) demonstrated a higher percentage of proliferating cells in metaplastic Barrett’s mucosa compared with normal gastric epithelium Staining patterns for low-grade and high-grade dysplasia were similar to PCNA, suggesting a greater turnover of differentiated cells in the surface epithelium, with immature proliferating cells arising from basal layers Increased proliferative activity and altered cell cycle kinetics were also demonstrated in Barrett’s epithelium using flow cytometry.57 An increased G1 fraction appears to be the earliest finding, progressing to increased S phase fractions with aneuploidy, high-grade dysplasia, and carcinoma These findings suggest a functional instability of Barrett’s mucosa, predisposing to increasing dysplasia and malignancy aneuploidy Ploidy, or DNA content, may be studied in cell populations using flow cytometry Progression of normal esophageal epithelium to Barrett’s metaplasia is clearly associated with abnormal DNA content (aneuploidy).57–60 The prevalence of aneuploidy appears to increase with the degree of dysplasia determined histologically 58 In an ongoing prospective study of over 300 patients, patients whose baseline biopsy demonstrated no, indefinite, or low-grade dysphasia with a diploid cell population (without aneuploidy or increased 4N fraction) appeared to be at low risk for malignant progression For this group of patients, endoscopic surveillance up to years was proposed However, more frequent surveillance was proposed for patients at increased risk for cancer progression, whose tissue biopsies were found to have high-grade dysplasia, aneuploidy, increased 4N fraction and 17p (p53) loss of heterozygosity.59 Furthermore, specific flow cytometric variables (aneuploid DNA content > 4.7N; 4N fraction > 6%) were even more highly predictive of cancer progression.60 The exact significance of the 4N fraction is unclear but is thought to represent an unstable intermediate stage in progression to aneuploidy Mapping studies have shown that most aneuploid populations are localized to a single region of esophageal mucosa, suggesting clonal expansion of single progenitor cells to involve large regions of esophageal mucosa Although multiple aneuploid cell populations are occasionally encountered in Barrett’s epithelium, only one aneuploid cell population typically is found in the primary tumor chromosomal abnormalities The following nonrandom chromosomal abnormalities and allelic losses have consistently been reported in Barrett’s esophagus and primary esophageal adenocarcinomas: 2q, 3p, 5q, 9p, 11p, 12q, 13q, 17q, 17p, 18q, and Xq and loss of the Y-chromosome, which increased with high-grade dysplasia.61–63 loss of heterozygosity LOH, reflecting loss of chromosomal regions or loci from one or more alleles, is reported frequently in Barrett’s esophagus and associated esophageal adenocarcinoma The following loci (and genes) have been implicated: 5q (adenomatous polyposis coli; APC),64 17p (p53),61,65 18q (deleted in colorectal cancer; DCC),66 9p (p16, p15),67 3p (fragile histidine triad; FHIT),68 and 13q (retinoblastoma; RB).69 LOH is reported in several additional loci (1p, 3q, 4q, 5p, 6q, 9q, 11p, 12p, 12q, 17q, and 18q), but putative candidate tumor suppressor genes have yet to be characterized 61,62 Particularly interesting is a recent report suggesting that deletion of a locus (31 to 32.1) on chromosome 14q can be used to differentiate adenocarcinomas of esophageal versus gastric cardia origin.70 Studies of 5q and 17p allelic losses in aneuploid cell populations derived from patients with Barrett’s esophagus and adenocarcinomas suggest that 17p loss precedes 5q loss in esophageal tumorigenesis This is in contrast to Optimal Management of Barrett’s Esophagus / 485 colorectal malignancy, where 5q losses are found earlier in neoplastic progression This finding further supports the hypothesis that alterations of the p53 tumor suppressor gene (localized to 17p) occur as an early molecular event in esophageal tumorigenesis p53 The p53 tumor suppressor gene encodes a 53 kd polypeptide that regulates cell cycle progression, DNA repair, apoptosis, and neovascularization in normal and malignant cells via highly complex DNA and protein interactions.71 p53 mediates cell cycle arrest in part by inducing the expression of p21 (WAF-1), which sequesters a variety of cyclin-dependent kinases facilitating G1 as well as G2/M arrest Over 90% of p53 mutations locate in the conserved DNA binding domain (exons to 8), with “hot spots” at codons 175, 176, 245, 248, 249, 273 and 282 in many human tumors This gene appears to have a central role in human malignancy, and has been characterized extensively over the past decade Furthermore, it would appear that p53 may have potential clinical application for novel therapeutic strategies.72,73 p53 gene mutations were initially reported in esophageal squamous cell carcinomas74 and subsequently in primary esophageal adenocarcinomas and associated Barrett’s epithelium in 1991.28 These findings have now been confirmed by several other investigators, and the spectrum of p53 alterations in Barrett’s esophagus and esophageal adenocarcinomas has been studied extensively The finding of p53 mutations in nondysplastic Barrett’s epithelium further suggests that p53 is altered early in the metaplasia–dysplasia–carcinoma sequence and may therefore be a useful molecular marker in endoscopic surveillance programs In a comprehensive study of primary esophageal adenocarcinomas defined according to strict clinicopathologic criteria, p53 mutations were associated with poor tumor differentiation and with reduced disease-free and overall survival following surgical resection.75 Of particular biological interest was the observation that patterns of p53 mutations varied between adenocarcinomas and squamous cell carcinomas of the esophagus Patterns of p53 mutations in esophageal squamous cell carcinoma are predominantly transitions or transversions occurring at A:T base pairs (suggesting a relationship to metabolites of ethanol, a well-defined risk factor for squamous cell carcinomas), or G-to-T transversions (a characteristic mutation attributed to benzo[a]pyrene, suggesting an association with tobacco) However, for esophageal adenocarcinomas, predominant mutations are G:C-to-A:T transitions at CpG dinucleotides, suggesting that these mutations result from endogenous mechanisms, such as spontaneous deamination of 5Ј-methylated cytosine into thymine retinoblastoma The Rb gene, located on 13q14, encodes a 105 kd nuclear phosphoprotein that is intimately involved in regulation of the G restriction point Rb mutations have been observed in Barrett’s epithelium, and in approximately 20 to 40% of esophageal adenocarcinomas,76 and are more frequent in tumors with p53 mutations.77 Recent studies suggest that loss of Rb expression correlates with advanced stage of disease, nodal metastases, and reduced survival.78 p16 Inactivation of the p16 tumor suppressor gene by allelic deletion or point mutation has been detected in approximately 20% of esophageal carcinomas.79 LOH involving 9p21 was reported in 75% of aneuploid cell populations derived from esophageal adenocarcinomas, as well as in all Barrett’s epithelia associated with tumors having p16 mutations.67 Allelic loss involving p16 preceded the onset of aneuploidy in most specimens Somatic mutation silenced the remaining p16 allele in 23% of the aneuploid cell samples Recent reports suggest that p16 inactivation correlates with cyclin D1 overexpression and poor prognosis.80 Experimental studies have also demonstrated that restoration of p16 expression, using gene therapy techniques, profoundly inhibits the proliferation and tumorigenicity of esophageal cancer cells, implicating p16 as a key molecular event during esophageal carcinogenesis.81 fragile histidine triad gene Deletions (LOH) involving 3p have been detected in 60 to 100% of esophageal adenocarcinomas and Barrett’s epithelium and has been proposed as an early molecular event.68 Although the tumor suppressor gene associated with 3p has not been identified conclusively, one major target appears to be the FHIT gene, which acts by hydrolyzing dinucleotide triphosphates to modulate cell cycle progression and apoptosis.82 Inactivation of FHIT by point mutation or promoter methylation result in the loss of FHIT protein expression, which has been observed in 50 to 90% of esophageal tumors and 85% of specimens derived from Barrett’s esophagus examined epidermal growth factor and receptor Epidermal growth factor (EGF) has a stimulator effect on epithelial cell proliferation, and has been shown to be overexpressed in approximately 30% of esophageal adenocarcinomas.83 Epidermal growth factor receptor (EGFR) overexpression appears to correlate with the degree of dysplasia Similarly, transforming growth factor alpha (TGF-␣), structurally and functionally related to EGF, which binds to the EGFR to stimulate growth via 486 / Advanced Therapy in Thoracic Surgery autocrine mechanisms, is also overexpressed in dysplastic esophageal epithelium.84 ERBB2 gene The ERBB2 gene encodes a 185 kd tyrosine kinase receptor molecule that is structurally related to EGFR Using immunohistochemistry, ERBB2 protein was reported to be overexpressed in up to 70% of Barrett’s epithelia and esophageal adenocarcinomas.85 In one recent study, overexpression of ERBB2 was seen in 24% of adenocarcinomas of the distal esophagus and EGJ and was associated with an advanced tumor stage (III, IV).86 cyclin D1 Cyclin D1 (CCND1), a cell-cycle regulatory gene, was recently reported to be altered in Barrett’s esophagus In a large case-control study of patients with Barrett’s esophagus, cyclin D1 immunopositivity of esophageal biopsies was associated with a statistically significant increased risk of progression to adenocarcinoma, suggesting this molecular marker may be clinically useful in future endoscopic surveillance studies.87 Cyclin D1 overexpression, reported in 40 to 60% of esophageal adenocarcinomas, may have prognostic significance following esophageal resection.80 ras oncogene In contrast to other gastrointestinal tumors, the ras oncogene is rarely mutated in human esophageal adenocarcinoma However, overexpression of ras-regulated genes (osteopontin and cathepsin L) has been reported in 58% of esophageal adenocarcinomas.88 Management The preceding sections review current controversies in the diagnosis of Barrett’s esophagus and recent advances in our understanding of the biology of this disease These observations form the basis for current approaches to management, which are summarized in the following sections There are several distinct goals of management (1) For patients diagnosed with invasive adenocarcinoma (arising from Barrett’s epithelium), conventional oncologic treatment is directed towards cure or palliation, depending on tumor stage (2) On the basis that earlier detection of malignancy will improve survival, endoscopic surveillance has been recommended for patients newly diagnosed with Barrett’s epithelium Several aspects of endoscopic surveillance remain controversial, particularly the timing and frequency of endoscopy, accuracy of biopsy sampling, tissue histology, implied treatment, and the cost-efficacy of this approach (3) For patients with GERD who are found to have Barrett’s metaplasia, control of reflux-related symptoms remains the primary goal of management However, it is not clear to what extent current medical or surgical antireflux therapies alter the natural history of Barrett’s metaplasia (4) Finally, the role of recently described mucosal ablation techniques in current clinical practice requires further critical appraisal Invasive Esophageal Adenocarcinoma The prognosis for invasive esophageal adenocarcinoma of the esophagus is related to the stage of disease at diagnosis For early stage tumors, esophageal resection is potentially curative.89–91 Recent studies have reported significant reductions in operative mortality following esophagectomy over the past two decades, with mortality rates below 5% consistently achieved in high-volume units.92,93 However, postoperative morbidity remains relatively high, and therefore careful attention must be given to assessment of the physiological status of patients under consideration for surgery Preoperative staging of esophageal tumors is generally considered to be relatively inaccurate, although newer techniques (endoscopic ultrasonography, positron emission tomography, thoracoscopy and laparoscopy) appear promising Several surgical approaches are widely employed for esophageal resection, with generally comparable outcomes In current practice, reconstruction of the upper gastrointestinal tract is generally achieved using stomach Despite increasing use in current clinical practice, multimodality therapy does not appear to offer a survival benefit for esophageal adenocarcinoma.94 The reader is referred to a detailed recent review for a further discussion of the management of invasive esophageal cancer.95 Surveillance of Barrett’s Esophagus The rationale for surveillance of patients with Barrett’s esophagus is that the early detection of cancer will lead to intervention and improved survival Several reports have confirmed the efficacy of endoscopic surveillance,96,97 and this approach has been widely adopted in current clinical practice However, practice patterns are quite variable and inconsistent Dysplasia is widely regarded as the precursor lesion of invasive cancer and is diagnosed by histologic examination of esophageal biopsies obtained during upper gastrointestinal endoscopy As dysplastic epithelium may appear normal endoscopically, to minimize sampling error, multiple (four-quadrant), systematic biopsies of the columnar-lined esophagus (every cm) are recommended Adherence to strict endoscopic protocols is reported by some groups to be highly accurate in differentiating high-grade dysplasia from invasive carcinoma Optimal Management of Barrett’s Esophagus / 487 Improved sampling accuracy is reported by use of large (jumbo) biopsy forceps, in addition to extensive biopsy of any suspected mucosal abnormality The following techniques have recently been described to improve the diagnostic yield: endoscopic ultrasonography; chemoendoscopy, using vital stains such as Lugol’s iodine; laser-induced fluorescence endoscopy; optical coherence tomography; and concurrent use of brush or balloon cytology Although promising, further evaluation of such techniques is warranted before they are applied to routine clinical practice.98 The limitations of histologic examination of esophageal biopsy specimens are widely known, particularly to establish the diagnosis of high-grade dysplasia and intramucosal carcinoma As active inflammatory change (esophagitis) may lead to extensive cellular atypia, it is important that endoscopic biopsies be taken after maximal antireflux therapy Even for expert gastrointestinal histopathologists, the grading of dysplastic change is often subjective, accounting for interobserver disagreement in over 10% of cases Current recommendations regarding the frequency of endoscopic surveillance are somewhat empirical Practice guidelines recently proposed by the American College of Gastroenterology recommend follow-up endoscopy every to years for patients with Barrett’s metaplasia, without evidence of dysplastic change For patients with low-grade dysplasia, 6-month follow-up is recommended, then yearly if stable The management of high-grade dysplasia is discussed below Endoscopic surveillance and management of patients with short segment Barrett’s esophagus and cardia intestinal metaplasia is extremely controversial and was proposed as an area for future research High-Grade Dysplasia in Barrett’s Esophagus The optimal management of patients with high-grade dysplasia in Barrett’s esophagus is confused by current histologic definitions and a lack of understanding of the natural history of this entity In the absence of inflammatory change, which may confound the histologic diagnosis, high-grade dysplasia is currently the most reliable predictor of progression to invasive adenocarcinoma In current practice, endoscopic biopsies are repeated from to 12 weeks after the initial diagnosis of high-grade dysplasia, usually following intensive medical therapy If no high-grade dysplasia is seen, follow-up is increased to every months to ensure stability, then according to proposed guidelines (as above) In this situation, it is presumed that inflammatory changes confounded the initial diagnosis of high-grade dysplasia However, in selected patients with Barrett’s epithelium with an unequivocal diagnosis of high-grade dysplasia on two consecutive biopsies, esophageal resection would currently appear to be the most effective treatment.99 This approach requires that esophagectomy be performed with low mortality (< 5%) and morbidity and therefore selects for patients who have few comorbidities (eg, ischemic heart disease or chronic obstructive lung disease) and who are physiologically able to tolerate this procedure The extent of resection is controversial but ideally should include all metaplastic columnar epithelium and as much of the esophageal squamous mucosa as possible, as this may be at risk for second primary tumors and to minimize the possibility of submucosal tumor at the proximal resection margin In current clinical practice, a subtotal (total thoracic and abdominal) esophagectomy would be reasonable, leaving a short cervical esophagus for a neck anastomosis It would be anticipated that unsuspected invasive adenocarcinoma would be diagnosed by careful histologic examination of the resected specimen in up to 50% of cases While it is expected that most tumors would be early stage (with favorable prognosis), a proportion of tumors will be of more advanced stage, with a higher frequency or regional lymph node metastases The extent of lymphadenectomy is unclear, in terms of staging accuracy and therapeutic intent There has been increasing discussion about quality of life and swallowing following esophageal resection and reconstruction, although objective data are limited Alternative management of high-grade dysplasia requires strict endoscopic surveillance, with surgical intervention only with a definite histologic diagnosis of invasive adenocarcinoma.100 As the natural history of high-grade dysplasia is not known, it is proposed that progression to invasive adenocarcinoma may take many years and is not inevitable This approach may therefore be entirely suitable for patients who refuse surgical intervention or who are at high surgical risk because of associated comorbidity In this situation, the frequency of endoscopic surveillance is quite unclear and may actually be unnecessary if intervention is not planned Treatment of Reflux-Related Symptoms Symptoms secondary to gastroesophageal reflux may be controlled effectively by lifestyle modification or pharmacologic or surgical therapy Although lifestyle changes (eg, diet, weight loss, reduction of alcohol and tobacco consumption, or elevation of the head of the bed) are extremely effective, patient compliance limits the longterm success of this approach Pharmacologic therapy using powerful antacids (H2-receptor antagonists, proton pump blockers) or prokinetics is currently the mainstay of modern medical therapy In addition to effective control of symptoms, medical therapy may also result in 488 / Advanced Therapy in Thoracic Surgery healing ulcerative esophagitis However, there is currently no evidence that medical antireflux therapy alters the natural history of Barrett’s epithelium and reduces the risk of developing adenocarcinoma In selected patients, antireflux surgery may well be superior to medical therapy for GERD,101 with consequent avoidance of long-term antacid medication Several surgical procedures are described, but the most widely used, and durable, is the Nissen (360°) fundoplication There has been considerable interest in minimally invasive antireflux surgery (eg, laparoscopic fundoplication) over the past decade, consequently altering the threshold for operative intervention.102 In patients with complex hiatus hernia, transthoracic approaches may be required to lengthen the esophagus (eg, Collis gastroplasty) to permit the creation of a tension-free intraabdominal fundoplication.103 As with medical therapy, there is currently little evidence to support regression of Barrett’s epithelium, dysplasia change, or reduction of cancer risk after antireflux surgery Patients with Barrett’s epithelium are therefore advised to continue endoscopic surveillance postoperatively Current indications for antireflux surgery in patients with Barrett’s esophagus are essentially similar to those for patients with GERD: to control intractable symptoms and to treat reflux-related complications (ie, peptic stricture, ulcerative esophagitis, and pulmonary aspiration) Endoscopic Ablation Over the past decade, several techniques of endoscopic ablation of Barrett’s epithelium have been evaluated in an attempt to reduce cancer risk.104 Esophageal mucosa may be ablated by laser, electrocoagulation, argon beam coagulation, cryotherapy, or photodynamic therapy or mechanically (endoscopic mucosal resection) The depth of mucosal ablation is critical to the efficacy of each technique and is directly related to the development of complications such as stricture formation or esophageal perforation Concurrent antireflux therapy (medical or surgical) is essential to allow healing of normal squamous epithelium Results of studies are currently limited by relatively small patient numbers, heterogenous patient populations with variable and incomplete diagnosis and staging, lack of control subjects, and only short to intermediate-term follow up A few early reports have shown incomplete esophageal reepithelialization, often overlying Barrett’s epithelium, in which invasive carcinoma has developed.105 As the long-term risk of cancer development is unclear, such novel approaches should currently be considered experimental but warrant further careful evaluation Summary Over the past three decades, there has been a dramatic change in the epidemiology of esophageal cancer There is increasing evidence linking GERD to Barrett’s esophagus, which is considered premalignant for esophageal adenocarcinoma As Barrett’s esophagus appears central to esophageal tumorigenesis, this chapter has addressed several recent controversies related to its diagnosis, etiology, molecular pathogenesis, and management As it is likely that significant progress with this disease will only be made with an improved understanding its fundamental biology, recent advances in the identification of clinically relevant biomarkers has been discussed in detail These may provide a rational basis for future strategies for early detection in endoscopic surveillance programs, as intermediate prognostic markers in chemoprevention studies, and as molecular targets for novel gene therapies References Blot WJ, Devesa SS, Kneller RW, et al Rising incidence of adenocarcinoma of the esophagus and gastric cardia JAMA 1991;265:1287–9 Devesa SS, Blot WJ, Fraumeni JF Changing patterns in the incidence of esophageal and gastric carcinoma in the United States Cancer 1998;83:2049–53 Zhang Z-F, Kurz RC, Sun M, et al Adenocarcinomas of the esophagus and gastric cardia: medical conditions, tobacco, alcohol, and socioeconomic factors Cancer Epidemiol Biomarkers Prev 1996;5:761–8 Gammon MD, Schoenberg JB, Ahsan H, et al Tobacco, alcohol, and socioeconomic status and adenocarcinomas of the esophagus and gastric cardia J Natl Cancer Inst 1997;98:1277–84 Chow W-H, Finkle WD, McLaughlin JK, et al The relation of gastroesophageal reflux disease and its treatment to adenocarcinomas of the esophagus and gastric cardia JAMA 1995;274:474–7 Lagergren J, Bergstrom R, Lindgren A, et al Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma N Engl J Med 1999;340:825–31 Farrow DC, Vaughan TL Determinants of survival following the diagnosis of esophageal adenocarcinoma (United States) Cancer Causes Control 1996;7:322–7 Tileston W Peptic ulcer of the esophagus Am J Med Science 1906;132;240–65 Barrett NR Chronic peptic ulcer of the oesophagus and “oesophagitis.” Br J Surg 1950;38:175–82 10 Bosher LH, Taylor FH Heterotrophic gastric mucosa in the esophagus with ulceration and stricture formation J Thorac Surg 1951;21:306–12 Optimal Management of Barrett’s Esophagus / 489 11 Morson BC, Belcher JR Adenocarcinoma of the esophagus and ectopic gastric mucosa Br J Cancer 1952;6:127–32 12 Allison PR, Johnstone AS The oesophagus lined with gastric mucous membrane Thorax 1953;8:87–101 13 Barrett NR The lower esophagus lined by columnar epithelium Surgery 1957;38:175–82 14 Paull A, Trier JS, Dalton MD, et al The histologic spectrum of Barrett’s esophagus N Engl J Med 1976;295:476–80 15 Spechler SJ Columnar-lined esophagus Definitions Chest Surg Clin N Am 2002;12:1–13 16 Sampliner RE Practice guidelines on the diagnosis, surveillance, and therapy of Barrett’s esophagus Am J Gastroenterol 1998;93:1028–32 17 Oberg S, DeMeester TR, Peters JH, et al The extent of Barrett’s esophagus depends on the status of the lower esophageal sphincter and the degree of acid exposure J Thorac Cardiovasc Surg 1999;117:572–80 30 Cameron AJ Epidemiology of columnar-lined esophagus and adenocarcinoma Gastroenterol Clin North Am 1997;26:487–94 31 Cameron AJ, Zinsmeister AR, Ballard DJ, et al Prevalence of columnar-lined (Barrett’s) esophagus Comparison of population-based clinical and autopsy findings Gastroenterology 1990;99:918–22 32 Prach AJ, MacDonald TA, Hopwood DA, et al Increasing incidence of Barrett’s oesophagus: education, enthusiasm or epidemiology? Lancet 1997;350:933 33 Caygill CPJ, Reed PI, Johnston BJ, et al A single centre’s 20 year experience of columnar-lined (Barrett’s) oesophagus diagnosis Eur J Gastroenterol Hepatol 1999;11:1355–8 34 Sharma P, Morales TG, Sampliner RE Short segment Barrett’s esophagus: the need for standardization of the definition and of endoscopic criteria Am J Gastroenterol 1998;93:1033–6 18 DeMeester SR, DeMeester TR Columnar mucosa and intestinal metaplasia of the esophagus Fifty years of controversy Ann Surg 2000;3:303–21 35 Ekstrom AM, Signorello LB, Hansson L-E, et al Evaluating gastric cancer misclassification: a potential explanation for the rise in cardia cancer incidence J Natl Cancer Inst 1999;91:786–90 19 Spechler SJ Short and ultrashort Barrett’s esophagus: what does it mean? Semin Gastrointest Dis 1997;8:59–67 36 Devesa SS, Fraumeni JF The rising incidence of gastric cardia cancer J Natl Cancer Inst 1999;91:747–9 20 Spechler SJ, Zeroogian JM, Antonioli DA, et al Prevalence of metaplasia at the gastro-oesophageal junction Lancet 1994;344:1533–6 37 Shaheen N, Ransohoff DF Gastroesophageal reflux, Barrett esophagus and esophageal cancer Scientific review JAMA 2002;287:1972–81 21 Clark GWB, Smyrk TC, Burdiles P, et al Is Barrett’s metaplasia the source of adenocarcinomas of the cardia? Arch Surg 1994;129:609–14 38 Shaheen NJ, Crosby MA, Bozymski EM, et al Is there a publication bias in reporting cancer risk in Barrett’s esophagus? Gastroenterology 2000;119:333–8 22 Chen YY, Antonioli DA, Spechler SJ, et al Gastroesophageal reflux disease versus Helicobacter pylori infection as the cause of gastric carditis Mod Pathol 1998;11:950–6 39 Jankowski J, Wright NA, Meltzer S, et al Molecular evolution of the metaplasia-dysplasia-adenocarcinoma sequence in the esophagus Am J Pathol 1999;154:965–73 23 Vaezi MF, Falk GW, Peek RM, et al CagA-positive strains of Helicobacter pylori may protect against Barrett’s esophagus Am J Gastroenterol 2000;95:2206–11 40 Vaezi MF, Richter JE Bile reflux in columnar-lined esophagus Gastroenterol Clin North Am 1997;26:565–82 24 Ormsby AH, Vaezi MF, Richter JE, et al Cytokeratin immunoreactivity patterns in the diagnosis of shortsegment Barrett’s esophagus Gastroenterology 2000;119:683–90 25 McArdle JE, Lewin KJ, Randall G, et al Distribution of dysplasias and early invasive carcinoma in Barrett’s esophagus Hum Pathol 1992;23:479–82 26 Haggitt RC Barrett’s esophagus, dysplasia, and adenocarcinoma Hum Pathol 1994;25:983–93 41 Champion G, Richter JE, Vaezi MF, et al Duodenogastroesophageal reflux: relationship to pH and importance in Barrett’s esophagus Gastroenterology 1994;107:747–54 42 Goldstein SR, Yang G, Curtis SK, et al Development of esophageal metaplasia and adenocarcinoma in a rat surgical model without the use of a carcinogen Carcinogenesis 1997;18:2265–70 27 Siewert JR, Stein HJ Classification of adenocarcinoma of the oesophagogastric junction Br J Surg 1998;85:1457–9 43 Farrow DC, Vaughan TL, Sweeny C, et al Gastroesophageal reflux disease, use of H2 receptor antagonists, and risk of esophageal and gastric cancer Cancer Causes Control 2000;11:231–8 28 Casson AG, Mukhopadhyay T, Cleary KR, et al p53 gene mutations in Barrett’s epithelium and esophageal cancer Cancer Res 1991;51:4495–9 44 McKinley M, Sherlock P Barrett’s esophagus with adenocarcinoma in scleroderma Am J Gastroenterol 1984;79:438–9 29 Taniere P, Martel-Planche G, Maurici D, et al Molecular and clinical differences between adenocarcinomas of the esophagus and of the gastric cardia Am J Pathol 2001;158:33–9 45 Spechler SJ, Schimmell EM, Dalton JW Barrett’s esophagus complicating lye ingestion with sparing of the distal esophagus Gastroenterology 1981;81:580–3 490 / Advanced Therapy in Thoracic Surgery 46 Meyer W, Vollmar F, Bar W Barrett’s esophagus following total gastrectomy Endoscopy 1979;2:121–6 47 Kortan P, Warren RE, Gardner J, et al Barrett’s esophagus in a patient with surgically resected achalasia J Clin Gastroenterol 1981;3:357–60 48 Sartori S, Nielsen I, Indelli M, et al Barrett esophagus after chemotherapy with cyclophosphamide, methotrexate, and 5-fluorouracil (CMF): an iatrogenic injury? Ann Int Med 1991;114:210–1 49 Trudgill N Familial factors in the etiology of gastroesophageal reflux disease, Barrett’s esophagus, and esophageal adenocarcinoma Chest Surg Clin N Am 2002;12:15–24 50 Kelsell DP, Risk JM, Leigh IM, et al Close mapping of the focal non-epidermolytic palmoplantar keratoderma (PPK) locus associated with esophageal cancer (TOC) Hum Mol Genet 1996;6:857–60 61 Aoki T, Mori T, Xiqun D, et al Allelotype study of esophageal carcinoma Genes Chromosom Cancer 1994;10:177–82 62 Dolan K, Garde J, Gosney J, et al Allelotype analysis of esophageal adenocarcinoma: loss of heterozygosity occurs at multiple sites Br J Cancer 1998;78:950–7 63 Menke-Pluymers MB, van Drunen E, Vissers KJ, et al Cytogenetic analysis of Barrett’s mucosa and adenocarcinoma of the distal esophagus and cardia Cancer Genet Cytogenet 1996;90:109–17 64 Zhuang Z, Vortmeyer AO, Mark EJ, et al Barrett’s esophagus: metaplastic cells with loss of heterozygosity at the APC gene locus are clonal precursors to invasive adenocarcinoma Cancer Res 1996;56:1961–4 65 Blount PL, Galipeau PC, Sanchez CA, et al 17p allelic losses in diploid cells of patients with Barrett’s esophagus who develop aneuploidy Cancer Res 1994;54:2292–5 51 Risk JM, Mills HS, Garde J, et al The tylosis esophageal cancer (TOC) locus: more than just a familial cancer gene Dis Esophagus 1999;12:173–6 66 Barrett MT, Schutte M, Kern SE, et al Allelic loss and mutational analysis of the DPC4 gene in esophageal adenocarcinoma Cancer Res 1996;56:4351–3 52 Casson AG Role of molecular biology in the follow-up of patients who have Barrett’s esophagus Chest Surg Clin N Am 2002;12:93–111 67 Barrett MT, Sanchez CA, Galipeau PC, et al Allelic loss of 9p21 and mutation of the CDKN2/p16 gene develop as early lesions during neoplastic progression in Barrett’s esophagus Oncogene 1996;13:1867–73 53 Wijnhoven BPL, Tilanus HW, Dinjens WNM Molecular biology of Barrett’s adenocarcinoma Ann Surg 2001;233:322–7 54 Souza RF, Morales CP, Spechler SJ Review article: a conceptual approach to understanding the molecular mechanisms of cancer development in Barrett’s oesophagus Aliment Pharmacol Ther 2001;15:1087–100 55 Gillen P, McDermott M, Grehan D, et al Proliferating cell nuclear antigen in the assessment of Barrett’s mucosa Br J Surg 1994;81:1766–8 56 Hong MK, Laskin WB, Herman BE, et al Expansion of the Ki-67 proliferative compartment correlates with degree of dysplasia in Barrett’s esophagus Cancer 1995;75:423–9 57 Reid BJ, Blount PL, Rubin CE, et al Flow-cytometric and histological progression to malignancy in Barrett’s esophagus: prospective endoscopic surveillance of a cohort Gastroenterology 1992;102:1212–9 58 Reid BJ, Levine DS, Longton G, et al Predictors of progression to cancer in Barrett’s esophagus: baseline histology and flow cytometry identify low- and high-risk patient subsets Am J Gastroenterol 2000;95:1669–76 59 Reid BJ, Prevo LJ, Galipeay PC, et al Predictors of progression in Barrett’s esophagus II: baseline 17p (p53) loss of heterozygosity identifies a patient subset at increased risk for neoplastic progression Am J Gastroenterol 2001;96:2839–48 60 Rabinovitch PS, Longton G, Blount PL, et al Predictors of progression in Barrett’s esophagus III: baseline flow cytometric variables Am J Gastroenterol 2001;96:3071–83 68 Michael D, Beer DG, Wilke CW, et al Frequent deletions of FHIT and FRA3B in Barrett’s metaplasia and esophageal adenocarcinomas Oncogene 1997;15:1653–9 69 Boynton RF, Huang Y, Blount PL, et al Frequent loss of heterozygosity at the retinoblastoma locus in human esophageal cancers Cancer Res 1991;51:5766–9 70 van Dekken H, Geelen E, Dinjens WN, et al Comparative genomic hybridization of cancer of the gastroesophageal junction: deletion of 14q31–32.1 discriminates between esophageal (Barrett’s) and gastric cardia adenocarcinomas Cancer Res 1999;59:748–52 71 Prives C, Hall PA The p53 pathway J Pathol 1999;187:112–6 72 Harris CC Structure and function of the p53 tumor suppressor gene: clues for rational cancer therapeutic strategies J Natl Cancer Inst 1996;88:1442–5 73 Wallace-Brodeur RR, Lowe SW Clinical implications of p53 mutations Cell Mol Life Sci 1999;54:64–75 74 Hollstein MC, Metcalf RA, Welsh JA, et al Frequent mutation of the p53 gene in human esophageal cancer Proc Natl Acad Sci U S A 1990;87:9958–61 75 Casson AG, Tammemagi M, Eskandarian S, et al p53 alterations in esophageal cancer: association with clinicopathologic features and survival Mol Pathol 1998;51:71–9 76 Montesano R, Hollstein M, Hainaut P Genetic alterations in esophageal cancer and their relevance to etiology and pathogenesis: a review Int J Cancer (Pred Oncol) 1996;69:225–35 Optimal Management of Barrett’s Esophagus / 491 77 Xing EP, Yang GY, Wang LD, et al Loss of heterozygosity of the Rb gene correlates with pRb protein expression and associates with p53 alteration in human esophageal cancer Clin Cancer Res 1999;5:1231–40 91 van Sandick JW, van Lanschot JJB, ten Kate FJW, et al Indicators of prognosis after transhiatal esophageal resection without thoracotomy for cancer J Am Coll Surg 2002;194;28–36 78 Roncalli M, Bosari S, Marchetti A, et al Cell cycle-related gene abnormalities and product expression in esophageal carcinoma Lab Invest 1998;78:1049–57 92 Swisher SG, DeFord L, Merriman KW, et al Effect of operative volume on morbidity, mortality, and hospital use after esophagectomy for cancer J Thorac Cardiovasc Surg 2000;119:1126–34 79 Esteve A, Martel-Planche G, Sylla BS, et al Low frequency of p16/CDKN2 gene mutations in esophageal carcinomas Int J Cancer 1996;66:301–4 80 Takeuchi H, Ozawa S, Ando N, et al Altered p16/MTS/CDKN2 and cyclin D1/PRAD-1 gene expression is associated with the prognosis of squamous cell carcinoma of the esophagus Clin Cancer Res 1997;3:2229–36 81 Schrump DS, Chen A, Consuli U, et al Inhibition of esophageal cancer proliferation by adenoviral-mediated delivery of p16INK4 Cancer Gene Ther 1996;3:357–64 82 Chen YJ, Chen PH, Lee MD, et al Aberrant FHIT transcripts in cancerous and corresponding non-cancerous lesions of the digestive tract Int J Cancer 1997;72:955–8 83 Al-Kasspooles M, Moore JH, Orringer MB, et al Amplification and over-expression of the EGFR and erbB-2 genes in human esophageal adenocarcinoma Int J Cancer 1993;54:213–9 84 Iihara K, Shiozaki H, Tahara H, et al Prognostic significance of transforming growth factor alpha in human esophageal carcinoma: implication for the autocrine proliferation Cancer 1993;71:2902–9 85 Flejou JF, Paraf F, Muzeau F, et al Expression of c-erbB-2 oncogene product in Barrett’s adenocarcinoma: pathological and prognostic correlations J Clin Pathol 1994;47:23–6 86 Polkowski W, van Sandick JW, Offerhaus GJ, et al Prognostic value of Lauren classification and c-erbB-2 oncogene overexpression in adenocarcinoma of the esophagus and gastroesophageal junction Ann Surg Oncol 1999;6:290–7 87 Bani-Hani K, Martin IG, Hardie LJ, et al Prospective study of cyclin D1 overexpression in Barrett’s esophagus: association with increased risk of adenocarcinoma J Natl Cancer Inst 2000;92:1316–21 88 Casson AG, Wilson SM, McCart JA, et al ras mutation, and expression of the ras regulated genes osteopontin and cathepsin L, in human esophageal cancer Int J Cancer 1997;72:739–45 89 Steup WH, De Leyn P, Deneffe G, et al Tumors of the esophagogastric junction Long-term survival in relation to the pattern of lymph node metastasis and a critical analysis of the accuracy or inaccuracy of pTNM classification J Thorac Cardiovasc Surg 1996;111:85–9 90 Swanson SJ, Batirel HF, Bueno R, et al Transthoracic esophagectomy with radical mediastinal and abdominal lymph node dissection and cervical esophagogastrostomy for esophageal carcinoma Ann Thorac Surg 2001;72;1918–25 93 Dimick JB, Catteneo SM, Lipsett PA, et al Hospital volume is related to clinical and economic outcomes of esophageal resection in Maryland Ann Thorac Surg 2001;73:334–41 94 Entwistle JWC, Goldberg M Multimodality therapy for resectable cancer of the thoracic esophagus Ann Thorac Surg 2002;73:1009–15 95 Krasna MJ, editor Esophageal cancer Chest Surg Clin N Am 2000;10:441–651 96 Streitz JM, Andrews CW, Ellis FH Endoscopic surveillance of Barrett’s esophagus Does it help? J Thorac Cardiovasc Surg 1993;105:383–8 97 van Sandick JW, van Lanschot JJB, Kuiken BW, et al Impact of endoscopic biopsy surveillance of Barrett’s oesophagus on pathological stage and clinical outcome of Barrett’s carcinoma Gut 1998;43:216–22 98 El Khoury J, Sahai AV Endoscopy in Barrett’s esophagus Surveillance during reflux management and new advances in the diagnosis and early detection of dysplasia Chest Surg Clin N Am 2002;12:47–58 99 Collard J-M High-grade dysplasia in Barrett’s esophagus The case for esophagectomy Chest Surg Clin N Am 2002;12:77–92 100 Schnell TG, Sontag SJ, Chejfec G, et al Long-term nonsurgical management of Barrett’s esophagus with high-grade dysplasia Gastroenterology 2001;120:1607–19 101 Parilla P, de Haro LFM, Ortiz A, et al Standard antireflux operations in patients who have Barrett’s esophagus: current results Chest Surg Clin N Am 2002;12:113–26 102 Jamieson GG, France M, Watson DI Results of laparoscopic antireflux operations in patients who have Barrett’s esophagus Chest Surg Clin N Am 2002;12:149–56 103 Chen L-Q, Ferraro P, Duranceau A Results of CollisNissen gastroplasty to control reflux disease in patients who have Barrett’s esophagus Chest Surg Clin N Am 2002;12:127–48 104 Pacifico RJ, Wang KK Role of mucosal ablative therapy in the treatment of the columnar-lined esophagus Chest Surg Clin N Am 2002;12:185–204 105 van Laethem JL, Peny MO, Salmon I, et al Intramucosal adenocarcinoma arising under squamous re-epithelialization of Barrett’s esophagus Gut 2000;46:574–7 492 / Advanced Therapy in Thoracic Surgery CHAPTER 41 STRATEGIES FOR ESOPHAGEAL REPLACEMENT AND RECONSTRUCTION SCOTT SWANSON, MD Esophageal resection remains a formidable operation with one of the highest potential for mortality and morbidity of any commonly performed operation While, in theory, resection of the esophagus and replacement with conduit is a relatively straightforward procedure, complications such as recurrent laryngeal nerve injury, thoracic duct injury, and mediastinal enteric leak in a potentially malnourished patient have contributed to described mortality rates of 15 to 40%.2–4 More recent reports describe a mortality rate of approximately 15%,5 while a few describe mortality rates as low as 5%.6–9 It is generally agreed that healthy stomach is the preferred conduit for esophageal replacement The stomach is well vascularized, easily reaches to the neck, and requires only a single anastomosis for reestablishing intestinal continuity When the stomach is not available (usually because of prior surgery or disease) the choice of conduit includes right or left colon and jejunum Key considerations for esophageal resection include optimizing the patients’ chances at cure, minimizing the risk of mediastinal enteric leak (which carries a mortality rate as high as 50% ), and minimizing associated pulmonary insufficiency and infection The four commonly used routes for esophageal resection include transhiatal, right chest and abdomen, right chest and abdomen and left neck, and left chest Each of these techniques is described as is the use of colon and jejunum for esophageal replacement Esophageal Resection (Four Methods) The Brigham Tri-Incisional Esophagectomy McKeown, first described the use of three incisions (neck, right thoracotomy, and abdomen) for esophagec- tomy.11 Advantages of the tri-incisional technique include direct visualization of mediastinal dissection, ability to perform a complete thoracic lymph node dissection, total thoracic esophageal resection, and avoidance of an intrathoracic anastomosis A bulky or potentially invasive tumor in the mid or upper esophagus should be dissected under direct visualization via thoracotomy The use of preoperative chemotherapy and radiation for the treatment of T3 or N1 tumors results in increased intrathoracic scarring, and dissection under direct visualization is usually safer The anastomosis can be performed in either the right or left neck Generally the left neck is preferred, because there is less chance of injuring the right recurrent laryngeal nerve, which is farther from the lower cervical esophagus than is the left recurrent nerve Described is the Brigham and Women’s Hospital modification of the McKeown technique.12 Esophagogastroduodenoscopy is performed to pinpoint the location of the tumor and evaluate for any coexisting gastric or duodenal disease Bronchoscopy is performed to exclude any tracheal or main bronchial involvement A double-lumen endotracheal tube is placed, and the patient is placed in the left lateral decubitus position A right posterolateral thoracotomy incision approximately 10 cm in length is made, and the chest is entered in either the fifth or sixth interspace depending on the location of the tumor (Figure 41-1) The lung is retracted anteriorly, and the inferior pulmonary ligament is divided At a point of minimal scarring away from the tumor, the pleura overlying the esophagus is opened anteriorly and posteriorly, and the esophagus is encircled with a wide Penrose drain All lymph node tissue between the pericardium, aorta, and spine are swept onto the speci- ... columnar-lined (Barrett’s) esophagus Comparison of population-based clinical and autopsy findings Gastroenterology 199 0 ;99 :91 8–22 32 Prach AJ, MacDonald TA, Hopwood DA, et al Increasing incidence... for the rise in cardia cancer incidence J Natl Cancer Inst 199 9 ;91 :786? ?90 19 Spechler SJ Short and ultrashort Barrett’s esophagus: what does it mean? Semin Gastrointest Dis 199 7;8: 59? ??67 36 Devesa... intubation Ann Otol Rhinol Laryngol 197 2;81:258–61 11 Seaman M, Ballinger P, Sturgill TD, Maertins M Mediastinitis following nasal intubation in the emergency department Am J Emerg Med 199 1 ;9: 37–9