Begin tumors of the esophagus

Begin tumors of the esophagus

In ordinary surgical practice, benign tumors of the esophagus are rarely encountered. For instance, at the Mayo Clinic, Moersch discovered 59 (0.3 percent) benign tumors among 18,459 patients who complained of dysphagia. Most of the benign mucosal tumors, such as polyps and papillomas, are seen by the endoscopist rather than by the surgeon and are usually removed locally through the esophago – scope. The importance of careful histological study of these tumors hardly needs emphasis.

By far the most common benign tumor is the leiomyoma, and this is the one which is referred to the surgeon when it attains sufficient dimensions to cause some degree of dysphagia. These myomas, found most often in the middle and lower thirds of the esophagus, occur singly or in multiple arrangement. They arise from the circular muscle layer and vary greatly in size. In my experience, they are always encapsulated and never involve the esophageal mucosa.

The degree of dysphagia produced by these tumors will depend largely on their size and extent of impingement on the esophageal lumen. I have seen fairly large lobulated tumors practically encircling the organ with little dysphagia. The diagnosis is made by roentgen examination. This discloses the characteristic appearance of a smoothly outlined tumor with sharply defined borders (Fig. 22). Esophagoscopy confirms the bulging of the tumor into the lumen, and always discloses an intact overlying mucosa. With the x-ray films and these findings on endoscopy, biopsy is not necessary. As a matter of fact, I would urge that biopsy be avoided. Biting into the normal mucosa could complicate matters for the surgeon should operation for the removal of the myoma be undertaken soon after.

The surgical treatment of myoma or fibroma of the esophagus is concerned mainly with the enucleation of an encapsulated tumor without entering the lumen of the organ. If the tumor is located in the middle third of the organ, a right transthoracic approach is preferable. A tumor located in the lower third is best treated by a left transthoracic approach. In either case, the chest is opened by an inter-

Begin tumors of the esophagus

Fig. 22. Roentgenogram indicating the smooth appearance of a myoma of the lower esophagus.

costal incision as described in the section on epiphrenic diverticulum. Gradual opening of the rib-spreader over a 10-minute period affords a wide exposure without rib fracture. The mediastinal pleura overlying the esophagus is incised, and the esophagus is freed bluntly from the mediastinum. Holding the tumor-bearing portion of the esophagus between the index finger and thumb of the left hand, the surgeon incises the thinned-out longitudinal musculature overlying the tumor (Fig.

Begin tumors of the esophagus

Fig. 23. Details of the operation for myoma of the esophagus showing the mobilization of the esophagus, the incision in the muscularis, enucleation of the myoma, and repair of the muscularis.

23). When the proper plane is reached, the tumor may then be easily enucleated without injuring the mucosa. The divided musculature is now repaired with interrupted sutures of fine silk. The esophagus is replaced in the mediastinum, and the incised pleura is closed with interrupted sutures. The chest may be closed without underwater drainage, provided that full expansion of the lung is maintained by the anesthetist until the wound dressing is applied.

The postoperative care of these patients should present no difficulties. Swallowing of fluids may be started the following day with rapid increase in the food intake thereafter. A postoperative hospital stay of seven or eight days should suffice.

The literature contains instances of diffuse myomas of portions of the esophagus where it was impossible to accomplish enucleation of the tumors. This necessitated resection of the organ with esophagogastric anastomosis. This must be an extremely rare occurrence nowadays. In our group of 16 patients with esophageal myomas, enucleation was possible in each instance. None of the patients died.


Despite recurring reports in the literature during the past decade which depict a rather bleak outlook for surgical therapy, the problem of malignancy of the esophagus remains a most important one in the overall consideration of diseases of the gastrointestinal tract. In this section, I will try to indicate how certain misconceptions have crept into the present viewpoint of a large number of American surgeons and make an effort to clarify many controversial features with a view to reaching a more rational approach to this difficult problem.

Although of different origin and different histological morphology, it has become customary to consider cancer of the esophagus and cancer of the cardiac end of the stomach under the same heading. I would like to emphasize that these two forms of malignancy have different biological characteristics, different anatomical extensions, and, very often, differ in their symptomatology and clinical course. For these reasons, they should be treated as separate entities. For purposes of clarification, however, both will be considered in this chapter, but under separate headings.


Aside from rare exceptions, the overwhelming majority of malignant tumors are squamous-cell carcinomas. Occasionally, one may encounter primary adenocarcinoma of the organ arising, apparently, from either ectopic gastric mucosa or from the deeper cell layers of the esophageal mucosa. In our overall series of patients, we have had seven such examples. In one the tumor was located above the aortic arch. An almost total esophagectomy was performed with transposition of the stomach as far as the upper cervical espohagus. For the record, this patient is alive and quite well nine years later. There is no evidence whatsoever of esophageal inflammation in the remaining small segment. Swallowing is normal, but the patient must sleep on three pillows to prevent regurgitation of gastric contents (Fig. 24).

Other rare malignant tumors of the organ are nonepithelial in origin. These comprise the leiomyosarcomas and the still rare melanocarcinomas and carcinosarcomas. I have had one of each in our group of patients. The leiomyosarcoma arose from the middle third of the organ and projected downward like a thick sausage toward the cardia (Fig. 25). Because the pedicle was infiltrated with tumor cells, a radical esophagectomy with supra-aortic gastroesophageal anastomosis was performed in 1946. This woman is alive and well 18 years later without any evidence of regurgitant esophagitis. The patient with the melanocarcinoma was reported in the Annals of Surgery for August 1955. She is alive and well 10 years after operation. It was suggested at the time of publication that these organ melanomas arise only from the melanoblast (the dendritic cell) and that primary melanocarcinoma of the esophagus arises from melanoblasts ectopically located in the esophagus.

Begin tumors of the esophagus

Fig. 24. Roentgenogram of patient referred to in the text, nine years after resection of the esophagus for primary adenocarcinoma of the esophagus behind the aortic arch. The esophagogastric anastomosis is in the cervical region.

Begin tumors of the esophagus

Fig. 25. Roentgenogram of a large leiomyosarcoma arising in the middle third ot the esophagus. Patient is alive and well 18 years after operation.

Begin tumors of the esophagus

Fig. 26. Roentgenogram of a carcinoma at the cardiac end of the stomach in a long-standing hiatus hernia.

Although one may find mention in surgical textbooks of many factors which could be considered as predisposing to the development of cancer of the esophagus, I believe that most of them can be discarded because of lack of any confirmatory evidence. A few would seem to possess some significance. For instance, I have noted a tremendous preponderance of esophageal cancer among the Chinese and Japanese, who ingest their rice piping hot. Again, I have had 11 instances of cancer of the cardia in long-standing hiatus hernia, approximately 8 percent of the total series of this variety of hernia (Figs. 26 and 27). On the other hand, I have seen only one example of squamous-cell cancer of the lower esophagus occurring seven years after a Heller operation for achalasia, the patient being symptom free during the intervening years.

Begin tumors of the esophagus

Fig. 27. Roentgenogram of a small carcinoma at the cardiac end of the stomach producing symptoms of obstruction for six weeks prior to operation. The hiatus hernia had been present for some years.

Begin tumors of the esophagus

Fig. 28. Gross specimen of resected esophagus showing the scirrhous or infiltrating variety of squamous cell carcinoma.

From the standpoint of therapeutic approach, the esophagus may be divided into four segments: 1, the cervical portion, extending from the pharyngoesophageal junction to the superior thoracic aperture; 2, the retro-aortic and superior mediastinal segment, stretching from the superior aperture to the under edge of the aortic arch; 3, the infra-aortic portion, extending from the inferior border of the aortic arch halfway to the diaphragm; and 4, the distal segment which ends at the hiatus of the diaphragm. In order of frequency, cancers of the organ occur in segments 3, 4, 1, and 2. The gross appearance of these tumors conforms to a loosely defined classification of three main types, in order of frequency: 1, scirrhous or infiltrating variety (Fig. 28); 2, the ulcerating form (Fig. 29); and 3, the fungating polypoid type (Fig. 30). A few years ago Wu-Ying-Kai of Peking presented a tentative gross pathological classification at a meeting of the International Society of Surgery. It has not been generally adopted.

Begin tumors of the esophagus

Fig. 29. Typical appearance of the ulcerating form of carcinoma of the esophagus with sharply defined overhanging edges and excavated base.

The scirrhous infiltrating type of esophageal cancer possesses a marked tendency to grow peripherally, involving at an early stage the hilus of the left lung or the wall of the descending thoracic aorta. Attempts to separate the tumor from these sites have often resulted disastrously. I remember, in the early days of our experience, attempting to separate a tumor from the undersurface of the aortic arch. In my youthful enthusiasm, I did not realize the significance of this tumor invasion and was soon left with a hole in the aorta plugged by my left index finger in the manner of the famed Dutch boy and the dike. When aortic grafts were unknown, repair of the hole was impossible because of carcinomatous infiltration. The ulcerating type may also extend peripherally, but not nearly as frequently. This variety resembles the gross appearance of a skin epithelioma with an ulcerated base and raised hard overhanging edges. There is usually considerable pulling inward toward the ulcerated tumor of the remaining circumference of the esophagus at this site resulting in a marked narrowing of the esophageal lumen. The least common variety, the polypoid tumor, projects into the lumen and may reach large proportions. Frequently the external musculature is completely free of tumor invasion, thus greatly simplifying mobilization of the organ from the mediastinum. In our one example of carcinoma associated with achalasia, the tumor reached the surprising size of 3 by 5 inches.

Begin tumors of the esophagus

Fig. 30. Demonstrating the gross appearance of a large fungating carcinoma of the esophagus, rather sharply demarcated and projecting into the lumen.

During the life history of cancer of the esophagus, especially in the middle third of the thoracic portion, complications may result from penetration of the tumor into neighboring structures. Slow perforation is probably the most common pathological complication and frequently results in the establishment of a fistula between the esophagus and the respiratory apparatus (left main bronchus or trachea). In middle third cancers, it is desirable to demonstrate the presence or absence of infiltration of the left bronchus or trachea by preoperative bronchoscopic examination. A positive biopsy precludes radical surgery. Neoplastic invasion of the aorta is a not uncommon complication and may result in sudden death from hemorrhage. Dilatation of the esophagus proximal to an obstructing cancer may occur but never reaches the proportions usually associated with achalasia.

The most important avenue of spread in esophageal cancers is via the lymphatics. Experience has demonstrated that there is a propensity for spread along the intramural lymphatic channels proximal to the tumor area. Thus, tumor cells may be found in these vessels at some distance from the primary growth. The frequency of this finding some years ago was the evidence necessary to explain the frequent suture line recurrences which were then encountered. This led to the need for more radical proximal excision of the organ in the performance of the operation. Between 70 percent and 80 percent of patients operated upon for esophageal cancer have metastatic spread to the regional lymph nodes. Cancers of the cervical segment spread to the superficial and deep nodes on both sides of the neck and, also, not infrequently to the paratracheal nodes in the superior mediastinum. In the thorax, lymphatic extension occurs to the paratracheal, hilar, and periesophageal groups of nodes. Operative experience has demonstrated the important fact that cancers of the thoracic esophagus, even as high as the level of the aortic arch, metastasize to the nodes below the diaphragm in the paracardial and left gastric artery groups. From my experience, I would say that between 50 percent and 60 percent of patients with cancer of the thoracic esophagus have metastatic node involvement below the diaphragm. It is presumed that this retrograde spread is caused by blockage at the site of the growth which prevents proximal dissemination. This single feature of the biological behavior of these tumors, excluding all other local factors, singles out this disease as a particularly malignant one.

Another curious facet of the story of esophageal cancer is the variation in the grade of malignancy and the rate of growth and spread. I have seen patients whose history extended over the short period of six weeks with hopeless extensive local and distant dissemination. On the other hand, there have been instances of prolonged history with confirmatory fluoroscopic evidence, and at operation easily resectable tumors were encountered. Barring obvious local factors of inop – erability, such as left recurrent nerve paralysis, invasion of the trachea or bronchus, fistula formation, cervical node involvement, positive azygogram (to be discussed shortly), and extensive spread laterally as demonstrated in the esophagus films, this biological variation calls for exploration to determine operability.

I have very little to say about the symptomatology of this disease except to stress the importance of investigating the reason for repeated episodes of dysphagia when swallowing solid food or substernal burning when swallowing hot liquids. By the time dysphagia becomes an everyday occurrence, more than three quarters of the esophageal lumen has been compromised and the tumor may very well be inoperable. As the lumen becomes narrowed progressively, the patient expectorates increasing amounts of thickened saliva which accumulates in the proximal segment of the organ. Loss of weight is a regular occurrence, but the extremes of yesteryear are rarely encountered nowadays. Boring back pain has serious prognostic significance, usually indicating tumor invasion of the thoracic aorta or spine.

Roentgenographic study of the esophagus will usually demonstrate the reason for the dysphagia (Figs. 31, 32, 33). Modern methods of x-ray examination will disclose even the smallest mucosal aberration. In recent years, there has been developed an additional form of roentgen study to demonstrate operability for middle third cancers, namely, the azygogram. By injecting Diodrast or a similar substance into the bone marrow of one of the ribs, the azygos venous system becomes visible in the roentgenogram. Nonvisualization of the azygos major vein indicates blockage of this venous system in the mediastinum, usually pointing toward extensive mediastinal growth of the tumor. This method of preoperative investigation is becoming a valuable aid in determining operability of middle third cancers (Figs. 34, 35, 36, 37).

Begin tumors of the esophagus

Fig. 31. Esophagram indicating the presence of a constricting scirrhous carcinoma of the middle third of the esophagus.

Begin tumors of the esophagus

Fig. 32. Roentgenogram of the esophagus showing the carcinoma of the middle third of the ulcerating variety.

Begin tumors of the esophagus

Fig. 33. X-ray of the esophagus depicting a small projecting tumor in the lower third of the organ.

Begin tumors of the esophagus

Fig. 34. Normal azygogram following injection into the left eighth rib demonstrating the hemiazygos system and the major azygos vein.

Begin tumors of the esophagus

Fig. 35. Azygogram following injection of the contrast medium in the right eighth rib showing an intact azygos system.

Begin tumors of the esophagus

Fig. 36. Esophagram indicating the presence of a large carcinoma in the upper portion of the middle third.

Chapter 2: The Esophagus

Begin tumors of the esophagus

Fig. 37. Azygogram of the same patient as in Figure 29 showing complete blockage of the azygos system at the lower edge of the tumor. This form of examination has special applicability in preoperative determination of operability.

Although recognizing the inherent dangers of esophagoscopy, especially in the hands of the inexperienced, I have always been a strong advocate of the importance of this form of examination in the complete work-up of the patient with esophageal disease. In suspected cancer of the organ, the esophagoscopist not only demonstrates the presence of the tumor but also locates it for the surgeon with respect to the distance from the upper incisor teeth, thus aiding in the planning of the operative approach. A positive biopsy, of course, clinches the diagnosis. Occasionally, when insufficient biopsy material is obtained for positive diagnosis, cytological study of esophageal washings may often demonstrate cancer cells.

The primary objective in the surgical treatment of esophageal cancer is twofold, the radical removal of the growth with a view to subsequent cure and the reestablishment of esophagogastric continuity to relieve the dysphagia. Unfortunately, this is not always possible. But this fact alone should not influence the surgeon to deny the patient the opportunity of an exploration, in the absence, of course, of obvious evidence of inoperability. The disclosure of certain pathological findings at the time of exploration will indicate to the surgeon which patient might be considered a candidate for possible cure and which patient will be afforded only palliative relief. But it must be emphasized that this differentiation between operations for cure and those for palliation be kept clearly defined when the surgeon’s experience is made the subject of a statistical survey. A definite policy with regard to the value of surgical treatment in this disease cannot be enunciated on the basis of limited experience. Yet, in recent years such attempts have been made in the surgical literature. It should be remembered that the operation of esophagectomy for cancer may be an extremely difficult undertaking. There are many technical pitfalls and unexpected complicating factors which could develop during the course of the procedure. In my opinion, many repetitive performances are needed before the surgeon can learn the myriad of detailed maneuvers of the operation. Increasing experience alone will enable him to handle the complicating factors which may arise during the conduct of the operation. This repeated performance will eventually be followed by a declining operative mortality, the performance of a better cancer operation, a more quickly performed operation which is most desirable in this group of poor-risk subjects and an increasing number of long-term survivors. The surgeon who undertakes the treatment of cancer of the esophagus should realize that he must pass through a discouraging period of trial and error before he has mastered the numerous technical details, during which time he will have developed a discriminatory approach to the many problems that confront him in each individual patient.

Preoperative Preparation. Once the decision has been made to submit the patient to operation, adequate preoperative preparation should be carried out. The time required for this will depend, in large part, on the patient’s general status. Among the elderly depleted group, a week or two may be needed before satisfactory improvement is noted. This is time well spent. The diet consists of high – caloric, high-protein liquids and soft foods with necessary vitamins if they can be swallowed. One or two preoperative transfusions are frequently necessary. These may be supplemented by intravenous fluids if dehydration is a prominent factor. Many of these patients enter the hospital so dried out that the daily intravenous injection of 2,000 ml of glucose and water and physiological salt solution will produce a marked improvement in a short period of time. For two days prior to operation, penicillin is administered by the aerosol inhalator in the dosage of 50,000 units in 1 ml of saline four times daily. I am convinced that this therapy has been a factor in practically eliminating postoperative pulmonary complications which were almost routine aftermaths in the early years of our experience with this operation. Cleansing esophageal lavages with warm saline solution remove retained food, mucus, and other detritus proximal to the obstructing tumor and also cleanse the surface of an ulcerating or polypoid tumor.

The question of preoperative digitalization always arises in the care of these patients, frequently giving cause for acrimonious discussions between internists and surgeons. While I am completely aware of all the academic arguments advanced against digitalization, I am convinced that this form of preoperative preparation of the elderly patient has much to commend it and make use of it almost routinely. Improvement of cardiac tone for the stress that accompanies this extensive procedure is something to be desired. Certainly a digitalized heart is easier to treat should cardiac difficulties arise in the postoperative period. Finally, a Levin tube is passed prior to transfer of the patient to the operating room.

Esophagectomy for Carcinoma of the Thoracic Portion. It should be constantly remembered that cancers of the esophagus frequently spread proximally in the submucosal and muscle lymphatics. This involvement is microscopic, thus eluding the palpatory and visual examination by the surgeon after the organ is mobilized. Therefore, in order to minimize suture line recurrences, the site of proximal transection of the organ should be as far removed from the tumor site as is technically feasible. Accepting this as a primary objective, the thoracic esophagus may be divided into two segments, each requiring a different surgical approach. Tumors located below the arch of the aorta are approachcd by the left transthoracic route, with esophagogastric anastomosis above the aortic arch. It seems to me that this approach greatly simplifies the continuity of the technical details of the operation, and for this reason 1 have always preferred it. Occasionally, when a tumor is located at the very lower end of the organ, it may be justifiable to avoid the difficulties of a supra-aortic anastomosis and content oneself with an infra-aortic anastomosis. To exclude cutting across areas with lymphatic spread, we have resorted to multiple frozen sections of the cut edge of the proximal esophagus. This is not a completely reliable method because, in spite of a negative report, a nest of cancer cells may be located a few millimeters proximal to the site of transection. I, therefore, am of the opinion that as a general rule the higher transection and anastomosis is preferable. For tumors located behind and above the arch (fortunately much less frequent), the right transthoracic approach is preferable because of the greater ease of mobilizing the esophagus in this region. This is the procedure of Ivor Lewis, which may be modified somewhat by circumstances arising during the progress of the operation.




    Heya, excellent page, especially many thanks

  2. I must say you have high quality articles here.

    Your website can go viral. You need initial traffic only.
    How to get it? Search for: Miftolo’s tools go viral

Leave a Reply