Oesophagus, thoracic duct and lymphatics


Core Procedures

  • Transthoracic oesophagectomy

  • Minimally invasive oesophagectomy (MIO)

  • Thoracic lymphadenectomy

Embryology

The oesophagus develops from the primitive foregut, a ventral diverticulum caudal to the fifth pharyngeal pouch in the third week of gestation. It is initially a single common gastrointestinal (oesophagus) and respiratory (trachea) tube, which later develops a respiratory bud from which the tracheobronchial tree develops. Failure of division results in congenital problems, such as oesophageal atresia and tracheo-oesophageal fistula, which are beyond the scope of the present chapter. In adults, foregut duplication cysts include cystic or tubular duplications of the oesophagus or bronchus. They are usually located in the posterior mediastinum, and although they commonly share a common wall with the oesophagus, are rarely in continuity with it. Treatment is required only if the cyst is causing respiratory, gastrointestinal or cardiac effects due to its size.

Surgical surface anatomy

The anatomy of the chest wall has already been discussed in detail in Chapter 42 and the anatomical location of the intercostal neurovascular bundle should be reviewed.

The oesophagus is usually approached via a right-sided thoracotomy with the patient placed in a left lateral position (right side up). For MIO procedures, patients may be positioned prone or in a left lateral position with the table rotated so the patient lies almost prone. For open thoracotomy, the surface landmarks of the nipple and tip of the scapula are used to site the incision. The skin, subcutaneous fat and muscles of the chest walls are divided using diathermy. Superficial arteries running within the subcuticular adipose tissue and muscle should be ligated.

The scapula is mobilized by developing the space beneath the scapula using blunt dissection of areolar tissue to lift the scapula from the chest wall. The second rib will be palpable at the upper level of the thorax; the ribs can then be counted and a thoracotomy via the fourth intercostal space is recommended to allow easy access to both the infra- and supra-azygos oesophagus. Ligation and excision of the intercostal neurovascular bundle inferior to the thoracotomy wound may reduce the incidence of postoperative pain.

Clinical anatomy

The oesophagus enters the thoracic cavity from the neck posterior to the trachea and anterior to the vertebral column. It passes inferiorly behind the aortic arch and at the level of the T4/5 vertebral body is said to occupy the posterior mediastinum along the right side of the descending thoracic aorta. (The posterior mediastinum is defined as the area bounded anteriorly by the tracheal bifurcation, pericardium and pulmonary vessels, and posteriorly by the bodies of the fifth to the twelfth thoracic vertebrae. The arched posterior third of the central portion of the diaphragm constitutes the anteroinferior limit of the posterior mediastinum laterally as the mediastinal pleurae come close together). Below, as it inclines left, the oesophagus crosses anterior to the aorta and enters the abdomen through the diaphragm at the level of the tenth thoracic vertebra. From above downwards, the trachea, right pulmonary artery, left principal bronchus, pericardium (separating the oesophagus from the left atrium) and diaphragm are anterior. The vertebral column, longus colli, right posterior intercostal arteries, thoracic duct, azygos vein, terminal parts of the hemiazygos and access­ory hemiazygos veins, and, near the diaphragm, the aorta, are all posterior relations ( ).

The thoracic duct originates from the condensation of the right and left lumbar trunks with the intestinal trunk, forming the cisterna chyli at the level of L2, and enters the thorax through the aortic diaphragmatic hiatus. The duct traverses the hiatus alongside the aortic opening and ascends between the aorta and the azygos vein, running posterolateral to the oesophagus. At the T4/T5 vertebral level, it curves posterior to the left carotid artery and left internal jugular vein, and drains into the venous system at the angle of the left subclavian and internal jugular vein.

Surgical approaches and considerations

Oesophagectomy is the key surgical procedure for oesophagogastric surgeons with respect to the thoracic oesophagus. This section summarizes the key evidence underpinning the choice of approach to oesophagectomy. Discussion of the anatomical principles of treatment of paediatric oesophageal atresia and tracheo-oesophageal fistulae is beyond the scope of the present chapter.

The oesophagus may be removed using a transthoracic approach: i.e. using a midline laparotomy to mobilize the gastric conduit and resect the draining abdominal lymph node stations, followed by a right-sided posterolateral thoracotomy to resect the thoracic oesophagus and perform an intrathoracic anastomosis. Alternative approaches for oesophagectomy include the transhiatal approach, whereby the gastric conduit preparation and lymphadenectomy proceed as per the first phase of the two-phase transthoracic approach but the oesophagus is mobilized via the hiatus under vision as far as the inferior pulmonary ligament and then bluntly dissected blindly by developing the peri-oesophageal plane alongside the oesophagus. A left-sided cervical incision exposes the oesophagus and allows blunt dissection of the oesophagus into the superior mediastinum. The gastric conduit is anastomosed to the remnant oesophagus in the neck. The left anterolateral thoracoabdominal approach utilizes extension of the midline laparotomy incision across the left side of the chest to enter the chest at the level of the sixth costal cartilage, and the diaphragm is opened to expose the oesophagus, with formation of the intrathoracic anastomosis below the aortic arch. A three-phase oesophagectomy involves the same first two stages as a transthoracic approach with a third incision in the neck, generally used to remove high (usually squamous cell) oesophageal tumours. In Eastern locations, where squamous cell tumours are most prevalent, resection of the draining lymph nodes within the supra-azygos compartment, as well as in the neck, is standard practice during resection of tumours of the oesophagus, but this practice is infrequent in Western centres.

Choice of approach partly depends on the location and histology of the tumour, carefully assessed endoscopically, as well as using CT of the chest and abdomen. The principles of surgical resection involve removal of the primary tumour with negative margins (denoted an R0 resection), along with removal of the draining lymph nodes and creation of a healthy, well-perfused gastric conduit that can be anastomosed in a tension-free manner to the remnant oesophagus. Creation of the gastric conduit, as well as alternate conduits in cases where tumour involvement of the stomach precludes use of the gastric tube, is discussed in more detail in Chapter 58 .

Adenocarcinomas tend to develop in the lower oesophagus and at the oesophagogastric junction, meaning that three-phase surgery with three-field lymphadenectomy is not standard treatment, as most lymph node spread is in a caudal distribution. Junctional tumours are defined as those arising within 5 cm of the oesophagogastric junction (OGJ), defined anatomically as where the oesophagus meets the proximal gastric rugal folds. Originally classified by Siewert et al, tumours of the junction with their centre or more than two-thirds of the tumour more than 1 cm above the OGJ are denoted Siewert type 1. Siewert type 2 tumours lie within 1 cm proximally or within 2 cm distally of the junction. Siewert type 3 tumours have their centre or more than two-thirds of the visible tumour within 2–5 cm of the OGJ. Since 2010, tumours of the oesophagogastric junction have been staged in the American Joint Committee on Cancer–Union for International Cancer Control (AJCC–UICC) TNM staging system (7th edition) as oesophageal cancers rather than gastric cancers, since their biological behaviour more closely resembles that of oesophageal tumours.

Although ongoing application of Siewert grading is recommended, there is a lack of evidence regarding its suitability to guide treatment decisions, particularly with respect to selection of operative approach: i.e. whether type 3 tumours are adequately treated by a total gastrectomy with abdominal lymphadenectomy only rather than a two-phase, two-field lymphadenectomy.

Squamous cell tumours tend to develop higher in the oesophagus and are associated with a higher risk of developing nodal metastases in the draining nodes of the neck, as well as the possibility of developing ‘skip’ metastases, whereby involvement of lymph nodes is not contiguous and may occur at an earlier stage in the tumour development.

A proximal margin of 5 cm is generally considered standard and the location of the anastomosis is partly determined by location of the tumour. To allow adequate proximal margins, as well as choice of anastomotic technique, a hand-sewn anastomosis is generally placed at the level of the azygos arch, versus a stapled approach which allows placement of the anastomosis higher in the apex of the thoracic cavity. Barbour et al have reported that an ex vivo proximal margin of more than 3.8 cm of normal oesophagus (which equates to 5 cm in vivo ) is associated with a minimal risk of anastomotic recurrence and is an independent predictor of survival.

The extent of the thoracic lymphadenectomy is considered in more detail below. Other considerations for operative approach depend on the background surgical training and experience of the operator; presence of background Barrett's metaplasia; presence of suspicious lymph nodes within the operative field versus lymphadenectomy guided by primary tumour location; and patient comorbidities and preferences.

Open versus minimally invasive approach

Minimally invasive oesophagectomy (MIO) strategies include minimally invasive Ivor Lewis oesophagogastrectomy (laparoscopy and limited thoracotomy or thoracoscopy) and minimally invasive McKeown oesophagogastrectomy (thoracoscopy, limited laparotomy or laparoscopy, and cervical incision). MIO strategies may be associated with decreased morbidity and shorter recovery times. In a study of MIO (mainly using thoracoscopic mobilization) in over 1000 patients, mortality rate was only 1.7% and hospital stay was only 8 days, which is less than most open procedures; only 4.5% required conversion to an open procedure. However, it is important to note that the volume of cases reported in the literature exceeds that of other centres reporting MIO and may not be generalizable. The prospectively maintained Esophagectomy Complications Consensus Group (ECCG) database reports outcomes from 24 high-volume centres internationally and is the largest database comprehensively reporting complication outcomes. These data indicate no significant differences in terms of length of stay or complications between open and MIO approaches and may be more reflective of current practice than trial data.

No randomized trials have assessed whether MIO improves oncological outcomes when compared with open procedures. At meta-analysis of 16 case control studies, there were fewer lymph nodes resected in the MIO group, but it is not clear whether this has an impact on overall survival. A randomized controlled trial of MIO (n=59) versus open oesophagectomy (n=56), powered on pulmonary complications as a primary endpoint, reported fewer postoperative respiratory tract infections in the MIO group (12% versus 34%, p=0.005). A prospective cohort study using validated measures of health-related quality of life (HRQL) shows that following minimal access oesophagectomy, the impact on HRQL may be less severe than standard open surgery. However, the risk of observer bias in these trials is high because non-blinded observers, and not the patients themselves, performed the assessments.

A prospective phase II multicentre feasibility trial enrolled 104 patients over a 6-year period and reported broadly satisfactory outcomes for a totally minimally invasive approach with a 2.9% 30-day mortality rate, 8.6% anastomotic leak rate, a median of 19 (range 2–55) lymph nodes harvested and survival outcomes broadly similar to international norms. The caveat remains that 45 patients in this study came from one centre and the median number of cases was 4 cases per centre (IQR: 3–5) in the 17 centres participating. This indicates that minimally invasive approaches, while feasible, require extensive experience and developed expertise.

Results of the French MIRO trial report a reduced postoperative morbidity in the hybrid minimally invasive arm (laparoscopic gastric mobilization and open thoracotomy, n=103) versus open controls (n=104) at 35.9% versus 64.4% (p=0.0001) and similar mortality rate (4.9%).

Robotic approaches to minimally invasive oesophagectomy are in development at present, although early series report a relatively large proportion of anastomotic leaks, reoperations and airway injuries. Whether these issues are due to the learning curve associated with the development of newer approaches or are inherent to the decreased haptic feedback of the technique will need careful scrutiny within controlled environments. Larger single-centre series report fewer complications as experience increases and a randomized controlled trial comparing robotic to open surgery is under way.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here