Adult Chest Surgery

Chapter 22. Palliative Options and Procedures 


Esophageal cancer has an annual incidence of approximately 14,000 in the United States. Unfortunately, most patients are not candidates for curative treatment with esophagectomy because symptoms of dysphagia usually do not present until at least 50% of the circumference of the esophagus is involved with tumor. By this time, the disease usually has spread to distant sites. The symptoms of dysphagia and weight loss that are commonly experienced have a significant impact on quality of life for these unfortunate patients. A number of palliative therapies are available to the surgeon that are not as invasive or as high risk as esophagectomy but nevertheless may provide significant symptomatic relief for the patient.

The options for treating dysphagia that is caused by advanced, unresectable esophageal cancer include stenting, photodynamic therapy (PDT), thermal laser ablation, and brachytherapy. Of these options, the most widely used therapies in the United States are esophageal stenting and PDT. These therapies are the focus of this discussion.



Endoscopic stent placement plays an important role in the palliation of dysphagia secondary to esophageal cancer. Over the past decade, advances in expandable stent technology have led to smaller, more flexible delivery systems that are easier to manipulate than the original plastic stents. These attributes permit successful deployment without exposing patients to the risks of aggressive mechanical dilation.

Most plastic stents have an internal diameter of 10 mm and include a proximal funnel to collect food and liquids and a flange at the distal end to prevent migration. The two methods of insertion are traction and pulsion. Traction stenting requires a general anesthetic and a laparotomy incision. For traction stenting, a pilot bougie is inserted orally and retrieved through a gastrostomy. The stent is sutured to the bougie and pulled into place at the site of obstruction. Then the tube is trimmed to length and the gastroscopy closed. Pulsion stenting can be performed with sedation; however, general anesthesia is also often used. A guidewire is placed, followed by dilation at the obstruction site.

All plastic stents necessitate aggressive dilation to a diameter of 45F before placement. The need for aggressive dilation imposes a significant risk of perforation. Additional problems include tube displacement, food impaction, and intractable reflux for stents placed across the gastroesophageal junction. In a study of these older methods of stenting, significantly lower mortality was seen in the pulsion group (14%) than in the traction group (23%).Length of hospital stay was 8.4 days, compared with 18.6 days in the pulsion and traction groups, respectively. Clearly, these results are not acceptable by current standards of care, particularly when the intent is palliation rather than cure.

A new generation of stents has evolved. Chief among them is the expandable metal stent (EMS). The EMS eliminates the need for aggressive dilation before insertion and can be placed with sedation alone. Because the stent itself is embedded in the tumor, the chance for migration is low. Morbidity, mortality, and length of hospital stay have been equally reduced in comparison with the plastic predecessors.

The EMS is constructed of various metals, including cobalt alloys (Wallstent, Schneider, Minneapolis, MN), stainless steel (Gianturco, Cook, Bloomington, IN), and a nickel-titanium alloy referred to as nitinol (Esophacoil, Medtronic, Minneapolis, MN; Ultraflex, Microvasive, Natick, MA). These materials are resistant to corrosion and biologically inert. The wire stents can be woven (Wallstent), knitted (Ultraflex), or bent into a zigzag (Gianturco) or coil (Esophacoil) configuration. The stent design influences its retraction (i.e., shortening) properties. Retraction percentage is higher with the coil than with the zigzag configuration. The shape-memory characteristics of these metals and metal alloys permit the stent to reexpand to its original tubular shape even after it has been compressed into a delivery system. The available systems include stents that can be deployed from the proximal end, the center, or the distal end. Proximal delivery is better suited for proximal strictures, whereas distal delivery is more suitable for obstructions located close to the gastroesophageal junction.

The EMS also can be partially covered with polyurethane or silicone. Covered designs help to reduce tumor ingrowth but also increase the risk of migration. Of note, covered stents have properties especially suited for the management of tracheoesophageal fistulas.The wide range of available diameters and lengths permits use of the EMS for most esophageal lesions.

A recent innovation in stent technology is the self-expanding plastic stent (Polyflex Esophageal Stent, Boston Scientific, Natick, MA). As with the EMS, these stents can be placed without need for aggressive predilation. However, the self-expanding plastic stent also can be removed because there is no tissue ingrowth, and it is therefore preferred for benign strictures. The self-expanding plastic stent also may be considered for patients receiving chemotherapy and radiation, where it may be possible to remove the stent at the end of therapy. The self-expanding plastic stent does, however, suffer from an increased migration rate (up to 25%).3

Photodynamic Therapy

PDT operates on the following basic principle: A photosensitive substance is injected that accumulates in target cells of the host tumor. This photosensitizing substance, when activated by light of a specific spectrum, causes selective tissue destruction (apoptosis). Photosensitizers used in surgical practice include purified hematoporphyrin derivatives [porfimer sodium (Photofrin), Axcan Scandipharm, Birmingham, AL], chlorines [Temoporfin, orm-tetrahydroxyphenyl chlorine (m-THPC)], and 5-aminolaevulinic acid (5-ALA). Optimal wavelengths of light absorption are 630 nm for porfimer sodium and 5-ALA and 652 nm for m-THPC. Currently, Photofrin is the only photosensitizer approved by the Food and Drug Administration for esophageal cancer and high-grade dysplasia of the esophagus.

Photosensitizers accumulate in all cells of the body. However, after 1–4 days, higher concentrations are found within tumor cells and in the interstitium of the tumor mass. Altered lymphatic drainage, neovascularization, and increased cellular proliferation are some of the mechanisms hypothesized to be responsible for this phenomenon. Laser light delivered directly to cancer cells that harbor the photosensitizer triggers a series of events culminating in cell destruction. When cells are bombarded with photons of a wavelength specific to the photosensitizer, the absorbed energy acts as a catalyst to the formation of highly reactive oxygen species (e.g., superoxide anions, peroxide anions, and singlet oxygen). The primary targets of photodamage are cellular membranes, amino acids, and nucleosides. Unlike the more commonly used thermal lasers, PDT uses a nonthermal laser consisting essentially of a wavelength of light that is specific to the photosensitizer (e.g., 630 nm for Photofrin). Since thermal energy is not used with this therapy, the risk of perforation is lower than with traditional thermal laser technique.


Although overall survival in esophageal cancer patients under consideration for palliative therapy is poor, both the EMS and PTD can improve the quality of life dramatically. In a report of 127 stent placements in 100 patients, immediate relief of dysphagia was observed with the EMS in 85% of patients.4

A more recent study of 78 patients treated with the EMS measured the effect of this treatment on quality of life at 1- and 2-month intervals using the QLQ-C30 questionnaire.Quality-of-life scores in the EMS group were compared with those of a group of patients treated with thermal laser ablation. At 1 and 2 months, scores were significantly improved in 96% and 75% of the EMS patients compared with 71% and 57% of the thermal laser ablation patients. Similar to other studies, the mean survival time for these inoperable patients was poor at 18 weeks.

Litle and colleagues assessed palliation of dysphagia using PDT in 215 patients with obstructing esophageal cancer.Dysphagia was seen to improve in 85% and bleeding in 90% of these patients. As expected, median survival was poor at 4.9 months; however, significant palliation was provided, as demonstrated by the improvement in dysphagia scores, which were measured before and after therapy.


All esophageal cancer patients with complaints of dysphagia for whom there are no plans for surgical resection should be considered for either stent or PDT therapy (Table 22-1). PDT is also an excellent option for controlling bleeding associated with esophageal cancer.PDT is most effective for tumors that involve the mucosa. As such, its use should be reserved for tumors with primarily an endoluminal component. Stents are more effective for tumors that are largely extra- or endoluminal.

Table 22-1. Comparison of PDT versus Stent Therapy for the Esophagus




Intraluminal tumors



Tumors causing extrinsic compression

Not effective


Palliation of cervical cancers


Limited by side effects

Palliation of gastroesophageal junction cancers


Limited by reflux

Effective for bleeding control



Onset of relief of dysphagia






Late stricture/dysphagia

Yes (stricture of normal esophagus)

Yes (stent migration or tumor ingrowth/overgrowth)


Tumor location can be an important consideration in patient selection. For instance, using palliative procedures for cervical cancers can be quite challenging. Stents placed close to the cricopharyngeus may be poorly tolerated because they can impart a persistent foreign-body sensation. Particularly in the case of bulky cervical tumors, stent deployment in the esophagus may lead to compression of the airway, requiring the need for a second stent in the airway. For these reasons, we prefer to use PDT in the cervical region.

Distal esophageal cancers are also better treated with PDT. Stents placed in close proximity to the gastroesophageal junction can cause significant reflux. Some centers recommend valved stents for distal tumors to minimize reflux. In a study comparing the effectiveness of valved stents versus the standard EMS in 50 patients, the incidence of reflux was significantly (p < 0.001) reduced with the valved stents from 96% to 12%.7

PDT and stenting are considered equally effective for midesophageal lesions. An advantage of stenting, however, is that patients do not have to endure a period of sun sensitivity requiring avoidance of direct sunlight. On the other hand, if definitive chemotherapy and radiation are part of the treatment regimen, PDT avoids the complications of stent erosion and esophageal perforation associated with chemotherapy.4

Mention has already been made of the use of covered stents for tracheoesophageal fistulas. Another group of patients difficult to treat are those who present with a perforated esophageal cancer. If resection is not a feasible option, then a covered stent may be effective.8

Contraindications to PDT include porphyria or allergies to porphyrins, the most commonly used photosensitizing agent. PDT is also contraindicated in the presence of an esophagorespiratory fistula.


Generally, all esophageal cancer patients are first evaluated for surgical resection. At minimum, therefore, all patients have a CT scan of the chest and abdomen. Barium swallow and endoscopy are helpful in determining the location and length of the mass, the degree of obstruction, and the degree of endoluminal and extrinsic obstruction.

For PDT, preoperative patient education is essential to minimize complications related to photosensitivity. Patients will remain photosensitive for a period of approximately 4 weeks after treatment. Direct sunlight exposure must be avoided during this period. After 4 weeks, patients are gradually reexposed to sunlight. When outdoors during the photosensitive period, patients must cover the skin surface by wearing hats, sunglasses, long-sleeve shirts, gloves, and long pants. Even indoors, it is necessary to avoid sitting next to windows and to keep blinds or curtains closed. Education of hospital staff is also important, and the use of labels to identify a photosensitive patient may minimize untoward light exposure during transfer for imaging studies or other procedures.


Expandable Metal Stent

EMS placement can be performed under conscious sedation or general anesthesia. Esophageal stents are available in a wide range of lengths (60–150 mm), but in contrast to airway stents, most have a similar maximal internal diameter (17–23 mm). Larger diameters are available from some manufacturers. If the opening of the obstruction is too small to accept the endoscope, the lumen may require dilation or laser ablation before length can be assessed accurately. The obstruction is identified endoscopically and measured. Fluoroscopy is used to mark the location of the obstruction. Two radiopaque markers (e.g., small stylets or paper clips) are taped to the skin at points corresponding to the proximal and distal edges of the obstruction. Alternatively, some clinicians prefer to inject the submucosa at the proximal and distal ends of the tumor with a radiopaque liquid (Conray).

Once the tumor length is established, the proper stent is selected. In general, we use an 18- to 23-mm-diameter stent that is 1–2 cm longer than the stricture to avoid crimping and infolding of the proximal and distal ends. A guidewire is passed through the obstruction, and the endoscope is withdrawn. Under fluoroscopic control, the delivery system is inserted over the guidewire through the obstruction and aligned with the skin or mucosal markers. The stent is deployed and expands within the lumen. Proper positioning and deployment are confirmed fluoroscopically and endoscopically. Minor adjustments are possible immediately after deployment by grasping the proximal end of the stent with endoscopic grasping forceps.

If a stent is planned for a bulky cervical esophageal tumor, bronchoscopy also should be performed. Most stents are 8 mm in diameter. Therefore, we recommend placing either an 18-mm balloon dilator or bougie before deploying the stent and assessing the airway with bronchoscopy during this maneuver. This will help to determine the need for a concomitant tracheal stent.

Photodynamic Therapy

The first step in PDT is administration of the photosensitizer. Porfimer sodium (1.5–2.0 mg/kg) and m-THPC (0.15 mg/kg) are injected intravenously, whereas 5-ALA (60 mg/kg) is given orally. Despite an improved side-effect profile and increased tumor specificity, 5-ALA has limited applications in thoracic surgery, because it produces necrosis only to a depth of 1 mm using current dosing regimens. Higher doses of 5-ALA cannot be used secondary to severe side effects, including nausea, vomiting, and transient elevation of liver enzymes. The chemical agent m-THPC is more photoactive than other agents and has a shorter elimination half-life than porfimer sodium, but it is not yet approved for clinical use in the United States.

Injection is usually performed in the outpatient setting, with the patient returning 24–48 hours later for endoscopy and delivery of the light therapy. The normal dose of porfimer sodium is 2 mg/kg injected over 3–5 minutes. Our preference is to use conscious sedation and flexible endoscopy in most patients. A cylindrical diffuser fiber is used to deliver light therapy to the tumor. The fibers are available in three sizes (1 cm, 2.5 cm, and 5 cm). The 5- and 2.5-cm sizes typically are used for treating esophageal cancers (Fig. 22-1). The diffuser fiber is placed endoluminally alongside the tumor. Multiple light illumination cycles may be needed depending on the length of the tumor relative to the length of the probe. We generally administer from 300 to 400 J/cm to the tumor. Care must be taken to minimize light therapy to the more normal areas of the esophagus, such as regions of high-grade dysplasia, because this may result in fibrous scarring and strictures. The propensity for PDT to damage relatively more normal areas of esophagushas led to the use of lower light dosimetry in patients with high-grade dysplasia.

Figure 22-1.


Photodynamic therapy cylindrical diffuser fibers.

Repeat endoscopy is performed at 48 hours and sometimes for a third time another 48 hours later. During the repeat endoscopic procedures, necrotic debris is removed using a combination of irrigation and suction forceps. Additionally, we often use a gently dilated balloon within the esophagus, moving this proximally and distally to debride the necrotic tumor and thus expose more underlying viable tumor. Additional light treatments are then delivered. With dysphagia, the palliative effect is usually evident by 5–7 days.


All patients undergoing the EMS or PDT should have a portable chest film performed in the post anesthesia care unit to note stent location as well as to rule out pneumomediastinum or pneumothorax. We routinely obtain an early barium esophagogram to assess patency and rule out perforation secondary to stent placement. After PDT, we usually defer the esophagogram for a few days to permit tissue edema to decrease. If there is any concern for the possibility of perforation, the esophagram is performed earlier. As mentioned earlier, all PDT recipients must strictly avoid exposure to sunlight. Medical therapy with proton pump inhibitors is administered to minimize problems with esophagitis after PDT or bleeding or reflux after stent placement.


Palliative procedures are often performed in high-risk patients who cannot tolerate esophagectomy. Despite this increased risk, mortality is low with both PDT and stenting. In Litle's report of 215 PDT patients, mortality was 1.8%.Christie's report of 127 stent placements in 100 patients yielded 0% mortality.These results are far more acceptable than the mortality associated with the older-generation plastic stents.For the most part, the complications observed were relatively minor. The most common complication after PDT in the Litle study was sunburn, which occurred in 6% of patients.Other complications included perforation (2%), stricture (2%),Candida esophagitis (2%), and pleural effusion (4%).

In the stent study described earlier, early complications included inadequate deployment (3.1%), pain requiring stent removal (1.6%), and perforation (0.8%).During follow-up, the most severe complication was erosion of the stent through the esophagus leading to sepsis. This occurred in three patients. Two of these patients also received chemotherapy. Sepsis led to death in one of these three patients. Reflux was reported in 11%, and stent migration occurred in 8.7%.

A number of patients may require reintervention during follow-up. In the stent series described earlier, tumor ingrowth or overgrowth occurred in 33% of patients.This can be treated with PDT or neodymium-yttrium-aluminium-garnet (Nd:YAG) laser. It is worth noting, however, that the Nd:YAG laser should not be used with a covered stent because of the risk of fire. Similarly, reintervention after PDT may include another course of PDT or stent placement depending on the reasons underlying the recurrent dysphagia. In Litle's study of PDT, 35 patients (16.3%) required stent placement, and 40 (18.7%) were treated with a second course of PDT at a mean dysphagia-free interval of 66 days.6


The thermal laser can be useful in limited circumstances. The device used most commonly in thoracic surgery practice is the Nd:YAG laser. In general, the thermal laser is more useful for tumors of the airway, which tend to be only a few centimeters in length. We do occasionally use the Nd:YAG laser to treat small endoluminal tumors of the esophagus in patients with tumor overgrowth close to a noncovered stent or at the site of a previous esophagogastric anastomosis. The Nd:YAG laser would not be indicated for treating a friable endoluminal tumor because identifying the lumen of a friable tumor may be difficult, resulting in prolonged treatment times and a higher risk of perforation with the thermal laser. This finding was reported in a randomized study comparing PDT and the Nd:YAG laser. A significantly higher perforation rate (7% versus 1%) was observed after Nd:YAG laser ablation than after PDT.10 For these reasons, we prefer stenting or PDT for the palliative treatment of esophageal cancers.

Brachytherapy can be defined as the placement of interstitial or intracavitary radioactive sources to facilitate the delivery of radiation doses to tumors with relative sparing of surrounding tissues. High-dose-rate brachytherapy involves the placement of radiation seeds via blind-ended afterloading catheters for short periods of time. Although high-dose-rate brachytherapy is used in many centers for airway tumors, this therapy has not gained the same popularity for esophageal cancer. We have found the afterload catheters currently approved by the Food and Drug Administration to be cumbersome and more difficult to place than those approved for the airway. Treatment response time is also slower than that seen with PDT. A recent randomized study from Sweden comparing stent insertion and brachytherapy demonstrated a more rapid relief of dysphagia in stent patients, with improved dysphagia scores at 1 month.11 Median survival was similar for both groups (132 versus 109 days, stent versus brachytherapy).


Although patients with advanced esophageal cancer have a poor survival, palliative therapies such as PDT or stenting can improve their quality of life significantly. These therapies fulfill the principal goals of palliative treatment of esophageal cancer. They are safe and yield relatively low morbidity and mortality yet are effective in improving dysphagia. The surgeon should be familiar with each of the above-described methods of intervention and should individualize therapy according to tumor characteristics and patient wishes. In many cases these therapies are complementary, and selection of one is not mutually exclusive of the other.



A 42-year-old man with a history of hypertension and heavy alcohol use presented with 3–4 days of hematemesis and dysphagia to both liquids and solids. A nasogastric tube was placed in the emergency room and immediately drained 200 mL of fresh blood. He was started on IV Protonix, transfused, and sent for endoscopy by the staff gastroenterologist. Endoscopy revealed an obstructing, fungating mass in the cervical esophagus. CT scan was performed and demonstrated a bulky esophageal cancer measuring 5.4 x 2.6 cm in its greatest dimension (Fig. 22-2).

Figure 22-2.


CT scan of cervical esophageal cancer.


Review of the CT scan revealed a bulky cervical cancer with some compression of the airway. The patient was not complaining of dyspnea. Cancers in this location are challenging to treat, and results of resection and radiation yield equivalent results.12 Therefore, this patient was referred for definitive radiation and chemotherapy. Since his dysphagia was so severe, endoscopic palliation also was thought to be necessary. Although our preference normally would be to use PDT, before subjecting the patient to unnecessary photosensitivity, we performed our own endoscopy in the operating room. First, an awake bronchoscopy was performed to view the degree of extrinsic compression by the mass. No gross tumor involvement of the airway was noted. An endotracheal tube was passed over the bronchoscope, and the patient was placed under general anesthesia. On esophagoscopy, the tumor appeared just 2–3 cm below the cricopharyngeus muscle, extending 5 cm distally. Biopsy demonstrated poorly differentiated squamous cell carcinoma. Although the gastroenterologist reported a fungating tumor, the mucosal component was actually quite small, and most of the obstruction was created by extrinsic compression. We concluded, therefore, that PDT would not be successful and elected to pursue stenting either of the esophagus alone or of the airway and esophagus combined. A guidewire was passed through the endoscope into the stomach. Savary (Cook Medical Inc., Bloomington, IN) dilators were passed over the guidewire. With a large dilator in the esophagus, bronchoscopy was again performed. There was no significant compression of the airway with the dilator in place. Using fluoroscopic guidance, a covered esophageal stent was placed.

The patient was extubated successfully. Chest x-ray performed postoperatively showed no pneumomediastinum or pneumothorax. A Gastrografin swallow was performed on postoperative day 2 and showed a widely patent esophagus without any extravasation of contrast material (Fig. 22-3). The patient was able to resume a normal diet and was discharged home with plans to start definitive radiation and chemotherapy.

Figure 22-3.


Gastrografin swallow after EMS placement.


This excellent review of palliative options stresses two main themes. First, surgical resection is now rarely if ever an appropriate palliative tool. Second, the concept of palliation demands a specific symptom management approach. Therefore, palliative options should be tailored to the patient and to his/her specific complaint at the same time. In general, we go from least to most invasive in our approach to palliation using laser debridement or PDT followed by stenting.



1. Unhurt HW, Pagliero KM: Pulsion intubation versus traction intubation for obstructing carcinomas of the esophagus. Ann Thorac Surg 40:337, 1985. 

2. van den Bongard HJ, Boot H, Baas P, et al: The role of parallel stent insertion in patients with esophagorespiratory fistulas. Gastrointest Endosc 55:110, 2002. 

3. Costamagna G, Shah SK, Tringali A, et al: Prospective evaluation of a new self-expanding plastic stent for inoperable esophageal strictures. Surg Endosc 17:891, 2003. [PubMed: 12618935]

4. Christie NA, Buenaventura PO, Fernando HC, et al: Results of expandable metal stents for malignant esophageal obstruction in 100 patients: Short-term and long-term follow-up. Ann Thorac Surg 71:1797, 2001. [PubMed: 11426750]

5. Xinopoulos D, Dimitroulopoulos D, Moschandrea I, et al: Natural course of inoperable esophageal cancer treated with metallic expandable stents: Quality of life and cost-effectiveness analysis. J Gastroenterol Hepatol19:1397, 2004. [PubMed: 15610314]

6. Litle VR, Luketich JD, Christie NA, et al: Photodynamic therapy as palliation for esophageal cancer: Experience in 215 patients. Ann Thorac Surg 76:1687, 2003. [PubMed: 14602313]

7. Laasch HU, Marriott A, Wilbraham L, et al: Effectiveness of open versus antireflux stents for palliation of distal esophageal carcinoma and prevention of symptomatic gastroesophageal reflux. Radiology 225:359, 2002. [PubMed: 12409567]

8. White RE, Mungatana C, Topazian M: Expandable stents for iatrogenic perforation of esophageal malignancies. J Gastrointest Surg 7:715, 2003. [PubMed: 13129545]

9. Panjehpour M, Overholt BF, Phan MN, et al: Optimization of light dosimetry for photodynamic therapy of Barrett's esophagus: Efficacy vs incidence of stricture after treatment. Gastrointest Endosc 61:13, 2005. [PubMed: 15672050]

10. Lightdale CJ, Heier SK, Marcon NE, et al: Photodynamic therapy with porfimer sodium versus thermal ablation therapy with Nd:YAG laser for palliation of esophageal cancer: A multicenter randomized trial. Gastrointest Endosc 42:507, 1995. [PubMed: 8674919]

11. Bergquist H, Wenger U, Johnsson E, et al: Stent insertion or endoluminal brachytherapy as palliation of patients with advanced cancer of the esophagus and gastroesophageal junction: Results of a randomized, controlled clinical trial. Dis Esophagus 18:131, 2005. [PubMed: 16045572]

12. Mendenhall WM, Sombeck MD, Parsons JT, et al: Management of cervical esophageal carcinoma. Semin Radiat Oncol 4:179, 1994. [PubMed: 10717106]

If you find an error or have any questions, please email us at Thank you!