Risk for anastomotic leakage in the thorax with possibly fatal consequences has resulted in the development of the three-stage approach with a cervical anastomosis by McKeown [30], and the transhiatal approach with a cervical anastomosis by Orringer and Sloan [31]. In case of anastomotic leakage in the neck, a subsequent cervical fistula is a manageable complication. However, some evidence shown that cervical anastomosis has some significant disadvantages, including excessive tension on the anastomosis, an ischemic tip of the gastric conduit resulting in a higher leak rate, a risk for recurrent laryngeal nerve injury, and development of postoperative oropharyneal dysfunction. Therefore, a simple and safe thoracoscopic intrathoracic anastomosis technique has always been the challenge for thoracic surgeons.

An Ivor Lewis procedure may reduce recurrent nerve lesion and other complications associated with a cervical dissection. Moreover, a shorter gastric conduit will permit a more extended gastric resection and a well-vascularized anastomosis lead to less anastomotic leakages and stenosis. This transthoracic procedure may be performed by a minimally invasive approach. MIE is increasingly implemented with potential benefits of having less pain, less respiratory infection, and reduced intensive care unit stay, preserving the completeness of the resection. The combination of Ivor Lewis esophagectomy with minimally invasive surgery has the potential to improve the postoperative outcome.

Anastomosis can be accomplished by a handsewn or stapler technique.

The first description of a totally endoscopic Ivor Lewis esophagectomy with an intrathoracic anastomosis was reported in 1999 by Watson et al. [32]. They described two patients in which the intrathoracic anastomosis was achieved with a handsewn single-layer technique. However, a higher incidence of dysphagia and a fourfold higher incidence of stricture were found after the handsewn technique. Therefore, stapler esophageal anastomosis is more often used in current clinical practice than handsewn techniques [33].

Stapled anastomosis in the thoracic cavity has been supported by Blackmon et al. [34] who analyzed three techniques of intrathoracic esophagogastric anastomosis: handsewn, circular stapled, and side-to-side stapled anastomosis. In this matched analysis, no significant differences were reported concerning anastomotic leakage.

Stapler technique include circular and side-to-side stapled anastomosis.

An important development is the introduction of the anvil transorally into the proximal esophagus, as described originally by Wittgrove et al. [39] for the gastrojejunostomy construction of the gastric bypass in bariatric surgery after the initial work of Sutton et al. in 2002 using a self-adopted circular anvil system [40].

Nguyen et al. [41] described the transoral technique in a series of ten patients to perform the intrathoracic anastomosis after esophageal resection. Campos et al. [42] confirmed the good results using the transoral anvil technique in 37 patients with a distal esophageal cancer.

Side-to-side stapled anastomosis is another significant intrathoracic anastomotic technique. Ben-David et al. [43] described in 2010 six patients with gastroesophageal junction cancers in whom after laparoscopic dissection and formation of the gastric conduit, the thorax was approached through a lateral right thoracoscopy. Gorenstein et al. [44] described a slight different side-to-side anastomosis technique in which the proximal esophagus was not stapled, and used the whole lumen for the construction of the side-to-side anastomosis by means of a linear stapler. Some evidence shown that there were no differences for both circular stapled and side-to-side stapled methods.

There are different techniques used to perform a safe intrathoracic anastomosis after an Ivor Lewis thoracoscopic procedure. None of the techniques are found superior to the others, but stapled anastomosis offered a safe outcome with a low percentage of anastomotic leakage and stenosis. Furthermore, no important differences were found between the two most used stapled anastomoses: the transoral introduction of the anvil, and the transthoracic introduction. Clinical trials are needed to compare different methods to improve the quality of the intrathoracic anastomosis after esophagectomy for cancer.

Esophagectomy is a mainstay of curative therapy for adenocarcinoma of the esophagus and gastroesophageal junction, and often plays an important role in the management of squamous cell esophageal cancers. However, whether by virtue of stage, age, or comorbidity, not all patients are appropriate candidates for esophagectomy. In addition to a staging evaluation, the physiologic evaluation of individuals who are potentially operable is key to appropriate patient selection. Assessing risk for esophagectomy determines the need for preoperative interventions to mitigate physiologic risk, focuses the need for perioperative resources, and assists in informed discussions with patients and their families.

More than almost any other common operation, esophagectomy is associated with high risks of postoperative morbidity and mortality. The most common postoperative complications are listed in Table 12.2 [45–48]. Pulmonary complications typically top the list, although minimally invasive approaches to esophagectomy have mitigated their risk somewhat. The risk continues after the traditional 30-day assessment period, with mortality rates increasing by 2-fold by 90 days postoperatively [49]. Complications are associated with increased costs of medical care and decreased quality of life [50–52]. There is also evidence that postoperative complications also adversely affect long-term survival [53].

There are reasonable alternatives to surgical therapy for esophageal cancer, including definitive chemoradiotherapy and palliative radiation therapy. Although the alternative therapies may not provide as good long-term outcomes as when resection is performed, their immediate risk is less than that associated with surgery. It is therefore appropriate to perform a careful physiologic evaluation of esophagectomy candidates in order to adequately assess their risk of postoperative morbidity and mortality and optimize patient selection for surgery.

Pulmonary complications are generally considered to be the most common after esophagectomy and contribute the most to the incidence of operative mortality. Complications after esophagectomy have recently been codified into a consensus-based list [54]. Pulmonary complications typically include pneumonia, acute aspiration, adult respiratory distress syndrome (ARDS), pleural effusion, pneumothorax, atelectasis due to mucus plugging, respiratory insufficiency, tracheobronchial injury, and persistent air leak. Factors associated with pulmonary complications in multivariable analyses include advanced age, poor spirometry, low diffusing capacity, poor performance status, renal dysfunction, diabetes, hypertension, alcohol consumption, prior cardiac surgery, underweight status or sarcopenia, and current cigarette use [55–59]. In addition to these typical clinical factors, it has also been shown that colonization of the airway with pathological bacteria is associated with an increase in the incidence of pneumonia [60]. Similar findings have been reported for pathologic colonization of dental plaque, and efforts to eradicate such plaque have been shown to reduce the incidence of postoperative pneumonia after esophagectomy [61, 62]. A number of studies, including a randomized trial, have demonstrated no increased risk of pulmonary complications after induction chemoradiotherapy [45, 46, 63].

Scoring systems have been developed that help identify patients at increased risk for postoperative pulmonary complications. One is based on a composite score comprised of age, spirometry (forced expiratory volume in the first second, or FEV1), diffusing capacity, and performance status [55]. The utility of this system has been validated in a large contemporary single institution dataset [64].

Most of the predictors of pulmonary complications can be assessed with a careful medical history (age, alcohol consumption, cardiac status, body mass index, blood pressure, smoking history; diabetic status), physical examination (performance status), and simple blood work (serum creatinine). The single evaluation that may not be performed routinely is pulmonary function testing, including spirometry and diffusing capacity. This is particularly important in patients undergoing induction therapy, in whom a substantial change in lung function as a result of the induction therapy, particularly higher doses of radiation therapy, with a mean reduction in DLCO% of 20% [65]. These patients should have baseline lung function testing prior to initiation of therapy, and retesting following completion of induction therapy. A decrease in diffusing capacity of more than 10 percentage points should lead to consideration of postponing surgery until the DLCO improves; typically a period of 1 month is sufficient for this to occur.

There are methods that may mitigate the adverse effects of lung function on postoperative outcomes. Inspiratory muscle training preserves muscle strength postoperatively and has been shown to decrease the incidence of overall pulmonary complications [66–68]. Preoperative nutritional support has been shown to reduce the incidence of pulmonary complications in some studies, but strong evidence to support this is lacking [69]. Obviously cessation of tobacco use and alcohol must be accomplished at least several weeks in advance of the planned operation. Most of the other risk factors are not remediable.

Cardiovascular complications after esophagectomy include arrhythmias (particularly atrial fibrillation), thromboembolic events, and, rarely, myocardial infarction. Risk factors for myocardial infarction include hypertension, diabetes, and coronary artery disease as manifested by prior myocardial infarction or the presence of angina. Most patients can easily be screened for this complication using the Revised Cardiac Risk Index (RCRI), which provides an algorithm for evaluation for major non-cardiac surgery. This and professional society guidelines indicate when additional testing is necessary [70, 71]. Intervention for patients who are at increased risk due to coronary artery disease should be considered preoperatively.

Atrial fibrillation presents an important challenge to the esophagectomy patient. New onset atrial fibrillation frequently is associated with other complications, and may be a harbinger of an undiagnosed problem such as anastomotic leak or pneumonia. The incidence of new postoperative atrial fibrillation is 15–20% [72, 73]. Preoperative risk factors for atrial fibrillation include male sex, advanced age, diabetes, a prior cardiac history, and induction therapy. At present there are no means for reducing preoperative risk factors for atrial fibrillation.

Venous thromboembolism (VTE) is an insidious complication that can affect esophagectomy patients from the period of induction therapy and extending a month after the operation. Risk factors include older age, female sex, black race, increased comorbidity index, prior VTE, and lower socioeconomic status [74]. An overall reported incidence of 6% [75] is likely underestimated because of lack of routine screening and absence of screening/reporting during the preoperative period. In my institution the overall incidence of symptomatic perioperative VTE is 11%, and among those undergoing induction therapy the incidence is 13% [unpublished data]. The development of VTE is associated with a 2-fold increase in operative mortality [74]. Use of preoperative screening for VTE in patients who are at increased risk, especially those with chronic venous stasis changes and a history of prior VTE, should be routine. In patients at substantially increased risk, prophylactic insertion of a temporary inferior vena cava filter should be considered. Whether routine use of prophylactic anticoagulation should be considered in patients undergoing induction therapy is a topic of current discussion.

Infectious complications are common after esophagectomy. They are categorized as, in order of decreasing frequency, anastomotic leak, empyema, wound infection, and intraabdominal abscess. The etiologies of infectious complications are quite varied, and include anastomotic leak, aspiration, respiratory compromise, and intraabdominal iatrogenic injury. Risk factors for complications include advanced age, male gender, black race, and multiple comorbidities [76]. There is no specific evaluation possible for assessing the risk of infectious complications; it is appropriate to use prophylactic intravenous antibiotics as the most likely preventive measure currently available.

Many patients with esophageal cancer suffer weight loss and associated nutritional deficiencies prior to beginning treatment. This may be a result of dysphagia resulting from the primary tumor, loss of appetite, and cancer cachexia. Cachexia is manifested by loss of fat and skeletal muscle as well as systemic inflammation; it is a primary cause of protein malnutrition, which in turn affects tolerance to aggressive therapies including surgery [77]. A combined assessment of BMI and percentage of normal body weight lost is a reliable predictor of mortality, particularly in patients with esophageal cancer [78]. Increased catabolism can be assessed by measurement of C-reactive protein (CRP), serum albumin, and the Glasgow Prognostic Score or its modification (GPS/MGPS) [79, 80]. Other elements of cachexia, including decreased caloric intake, reduced fuel stores, and impaired function, are readily measured through calorie counts, history and physical, and screening for frailty.

Unfortunately, anticancer therapies, specifically including induction therapy, are associated with weight loss and can contribute to the development of pre-cachexia or cachexia in patients for whom esophagectomy is intended. In one study two-thirds of patients experienced substantial weight loss after induction therapy, and 50% of the patients suffered loss of more than 10% of their normal body weight [81].

Interventions for severe nutritional deficiencies have not been proven successful in the short term, and long-term interventions do not take into consideration the need for timely and aggressive management. Parenteral nutrition has offered little benefit and some have demonstrated negative outcomes associated with this intervention. Laparoscopic placement of a feeding jejunostomy tube is appropriate in undernourished patients and those who have moderate to severe dysphagia, particularly patients for whom induction therapy is planned. Once a patient transitions from a pre-cachectic state to cachexia, interventions are unlikely to be successful. Progression to refractory cachexia is associated with patients being unresponsive to anticancer therapies, and portends a short expected survival.

Sarcopenia is a condition represented by abnormally low core muscle mass and density. Frailty is a state of increased vulnerability to physiologic stressors, which reduces resiliency and places affected patients at increased risk for postoperative complications. There is increasing recognition that sarcopenia and frailty are closely linked. Frailty or pre-frailty is present in about many patients undergoing esophagectomy for cancer. In my institution it pre-frailty or frailty is identified in about 70% of patients who are candidates for major thoracic surgery. Its presence is associated with a substantial increase in complications and mortality after esophagectomy. In particular, a modified frailty index is linearly associated with increasing incidences of life-threatening complications overall, pneumonia, respiratory failure, cardiac arrest and myocardial infarction, VTE, shock, and operative mortality [82]. With the exception of age, frailty was the only determinant of adverse outcomes after esophagectomy in a multivariable analysis in Hodari’s study.

Sarcopenia is present in 25–75% of patients undergoing esophagectomy for cancer [59, 83]. Its presence is associated with increased postoperative complications after esophagectomy. It has been shown to be an independent predictor of postoperative respiratory complications [58, 59]. Sarcopenia may not be present at the time of diagnosis but instead may develop during induction therapy for esophageal cancer, a change that is associated with an increased rate of postoperative complications [83, 84].

Frailty can easily be assessed using simple screening tools in the outpatient setting. A typical screening assessment includes gait speed, grip strength, weight loss, energy levels, and the level of recent physical activity. Patients may be classified using the Fried criteria as not frail, pre-frail, or frail [85]. Assessment of sarcopenia is not as simple. In general, underweight patients are often sarcopenic, especially older underweight patients. But a large segment of the population, the so-called sarcopenic obese patients, cannot be screened using this assumption. Methods are being developed for automated measurement of core muscle density and mass that may facilitate the diagnosis of sarcopenia using routine staging computed tomography (CT) scans in the near future.

It may be possible to mitigate the adverse effects of sarcopenia and frailty on postoperative outcomes through strength training. This is aimed at improving core muscle strength, balance, and endurance. Strength training for a period of only 4 weeks has been shown to substantially improve muscle strength and endurance in elderly frail patients [86]. Studies are currently underway to determine whether these improved metrics correlate with improved postoperative outcomes.

Hepatic insufficiency and esophageal cancer, particularly squamous cell cancer, may share a common etiologic pathway such as alcohol abuse. Fortunately there are few patients who are potential candidates for esophagectomy who also suffer from cirrhosis. Hepatic cirrhosis raises concerns for esophagectomy because of related blood coagulation disorders, the risk of esophageal and gastric vascular abnormalities may that preclude successful reconstruction, the development of difficult to manage ascites and pleural effusions, and the possibility of hepatic encephalopathy. Common etiologies for mortality in patients with cirrhosis include pneumonia, hepatorenal syndrome, and sepsis. It is suggested that patients in Child-Pugh ‘A’ cirrhosis may be a reasonable candidate for esophagectomy, but patients in Child-Pugh ‘B’ and ‘C’ are at substantially increased risk for mortality and likely should not be recommended for esophagectomy [87, 88].

Preexisting renal insufficiency poses perioperative management problems for esophagectomy patients. Fluid shifts during long operations and the occasional need for high volume fluid resuscitation over a period of several days can create challenges. There is no strong evidence to suggest that preoperative renal insufficiency increases the risk of postoperative complications or surgical mortality after esophagectomy.

The presence of diabetes offers management challenges in the perioperative period that must be considered, including glucose management during preoperative dietary changes, intraoperative management of glucose levels, and changing needs for insulin in the postoperative period as patients are transitioned to enteral feedings and then to oral intake. Diabetes is a risk factor for surgical complications, specifically anastomotic leak and dehiscence [89, 90].

Candidates for esophagectomy should be carefully evaluated prior to receiving recommendations regarding surgery (Table 12.3). A careful history will disclose information regarding general activity levels and any recent changes, performance status, weight loss, dysphagia, caloric intake, cardiovascular risk, and co-morbidities. Formal screening for frailty and nutritional deficiencies should be performed in appropriately selected patients. All patients should undergo pulmonary function testing and assessment of cardiac risk score (RCRI). Patients who are at increased cardiac risk should have additional evaluation by a cardiologist. Cardiopulmonary rehabilitation can be considered in selected patients who are de-conditioned but are able and motivated to improve their physical condition preoperatively.

Esophagectomy represents one of the highest risk routine operations for cancer. In addition to the extent of the operation, risk is increased because patients are often deconditioned and malnourished, and many have recently completed induction therapy. Careful assessment across a number of domains is essential in assessing risk and in making appropriate recommendations that balance oncologic outcomes with perioperative risks.

There is considerable morbidity and mortality associated with esophageal resections. These are related to perturbations in respiratory muscle function, substantial interstitial fluid shifts, contamination of the surgical spaces, occasional recurrent nerve or thoracic duct injury, and a high incidence of anastomotic leak, to mention just a few contributing factors. The frequent use of induction chemotherapy or chemoradiotherapy may compound these risks. Although perioperative complications associated with esophagectomy have decreased as a result of increasing regionalization of care and the introduction of minimally invasive techniques, the mortality rate is still about 5% and morbidity rates range from 15–50% [91, 92]. Given these risks, appropriate selection of patients for esophagectomy is crucial in optimizing operative outcomes. Selection is based on clinical cancer stage and specific patient characteristics, including anatomy, physiology, and comorbidities. Surgeon experience and judgment are critical in this endeavor.

The incidences of squamous cell carcinoma (SC) and adenocarcinoma (AC) of the esophagus have increased worldwide over the last decade [93, 94]. Historically, important variations among staging and treatment modalities around the world have made the comparison of treatment outcomes difficult. Fortunately, unified criteria for staging esophageal cancer derived from a collection of worldwide data were first established for the Seventh Edition of the American Join Committee on Cancer (AJCC)/International Union Against Cancer (UICC) manual [95]. Anatomic and histologic variables were incorporated, including histological type and grade, tumor location, and number of lymph nodes involved.

In Western countries routine staging studies include a combination of endoscopic ultrasound (EUS) with or without EUS-directed needle aspiration (EUS-FNA), computed tomography (CT), and positron emission tomography fused with CT (PET/CT). Some centers routinely perform abdominal or neck ultrasound to evaluate for liver nodules and enlarged cervical or supraclavicular nodes. In developing countries the routine use of many of these modalities is considerably constrained by access and cost. Clinical staging in many of these countries is limited to CT. Staging of the primary tumor is performed with CT and EUS, and in select circumstances bronchoscopy. Nodal staging is accomplished with CT, PET, and EUS. Evaluation of distant metastases is performed with CT, PET, and possibly EUS to evaluate for potential liver metastases or nodal metastases outside of the regional nodal stations.

Metastatic disease to the peritoneal cavity is difficult to identify with routine testing. This is particularly relevant in patients with distal esophageal and gastroesophageal junction tumors because of the higher rate of metastases to the peritoneal cavity. Laparoscopy and peritoneal fluid cytology have a sensitivity of 96% to detect peritoneal metastases particularly in patients with esophageal AC [96]. Some studies have demonstrated un upstaging in close to 12% of patients with esophageal tumors after staging laparoscopy who did not have evidence of peritoneal involvement in the traditional radiographic methods (CT, PET/CT, EUS) [97]. However, the risks and costs associated with an additional anesthetic and surgical procedure should be taken into consideration. In addition, patients undergoing minimally invasive esophagectomy can have a thorough assessment of the peritoneal cavity before proceeding with a formal resection, thus eliminating the need for a prior staging laparoscopy. Occasionally, mediastinoscopy, thoracoscopy, bronchoscopy (routine for tumors abutting the major airways) or image-guided percutaneous biopsies are utilized when suspicious abnormalities are identified during routine assessment.