The Palliation of Dysphagia in Oesophageal Malignant Obstructions using Endoprostheses:

A Review of the Literature.

Helen Tate, RGN.

Advanced Diploma Health Science Student.

University of Northumbria, Newcastle upon Tyne.


Malignant oesophageal obstruction due to intrinsic oesophageal tumours or from extrinsic compression caused by medistinal tumours is not a curative condition for the majority of patients. Dysphagia, or the inability to swallow, is one of the most distressing and debilitating symptoms connected with this disease. The rapid palliation of oesophageal carcinoma consists of the use of stenting to try and provide relief of the dysphagia. This article reviews some of the existing literature regarding the use of endoprostheses to effectively alleviate this problem.


Key words: oesophageal carcinoma; dysphagia; endoprosthesis; rigid plastic stents; self-expanding metal stents; palliation.


Carcinoma of the oesophagus is the seventh most common malignancy in the world, although it is still relatively uncommon in Britain with a mean incidence of 7.5 per 100,000 of the population (Muller, Erasmi, Stelzner 1990). The management of this condition however is difficult, as 50-60% of patients diagnosed with oesophageal malignancies are suitable only for palliative treatments. Palliation of oesophageal carcinoma consists mainly of the symptomatic treatment of dysphagia. Without treatment these patients have a poor quality of life, suffer from rapid weight loss and experience a relatively quick and unpleasant death from total dysphagia and aspiration of their own saliva. The aim is therefore the relief of dysphagia with minimal morbidity and maximum quality of life. The quality of life, not quantity is deemed more desirable (Ellul, Watkinson, Khan, Adam, Mason 1995). The use of oesophageal endoprostheses or stents are a reliable way to allay dysphagia, thus increasing life expectancy and improving the standard of living for the sufferer. This review is a comparison of the two varieties of stents available, rigid plastic stents (RPS) and self-expanding stents (SEMS).


The number of patients used in the studies examined varied considerably, ranging from a small sample of six patients (Kato, Saji, Kanematsu, Hoshi, Ishiguchi, Kunieda, Takao, Sugiyama 1996) to a larger selection of 70 patients (Cowling, Hale, Grundy 1998). 30.9 patients being the mean number of all the studies examined. The majority of the patients were men and only a small number were women. The possibility for this is that in western society it is a male predominated disease, although it is on the increase in the female population.

All the patients were suffering from inoperable carcinoma of the oesophagus causing various degrees of dysphagia. In the study by Cowling et al. (1998) however, 11 of 70 cases had metastatic disease producing extrinsic compression. Patients on Knyrim, Wagner, Bethge, Keymling, Valkil (1993) study were excluded if the tumour was within 2cms of the oesophageal sphincter or if they had a tracheal-oesophageal fistula. The other studies examined in this review did not conform to this criteria and patients with fistulas and high or low-sited tumours were included in the project. Throughout the surveys, all of the cases were assessed prior to stent insertion using the dysphagia scoring system first utilised by Mellow and Pinkas (Knyrim et al.1993).


0 = able to eat normal diet / no dysphagia.

1 = able to swallow some solid foods

2 = able to swallow only semi solid foods

3 = able to swallow liquids only

4 = unable to swallow anything / total dysphagia

Cowling et al. (1998) used a modified version of this score:

0 = no dysphagia

1 = semi solids only

2 = liquids only

3 = complete dysphagia.

The patients all started with dysphagia scores of 3 - 4. This score was measured 24 hours prior to stent insertion and 24 hours post insertion and then at regular intervals of 4-6 weeks until either, the end of the study or until death occurred in the subject.

Knyrim et al. (1993) however, also looked at the quality of life in his patients; this was measured using the Karnofsky index, which is a questionnaire about perceived well being.

0 = death.

100 = normal.

He measured this prior to stent insertion, but patients with a score of less than 30 were not included in the trial. The score was measured at the same timed intervals as the dysphagia score.

As well as the latter measurements, data regarding length of hospital stay, complications and the 30-day mortality percentage. (This was the percentage of the subjects that died before 30 days post insertion) was ascertained.

All information was collected and was correlated over approx. 6 months (the majority of studies ended after 6 months, this being the approximated length of life expectancy of someone with an inoperable oesophageal neoplasm). The pre and post- insertion dysphagia scores were compared and the levels of improvement and the percentage of patients with improvement were calculated. The data throughout the trails was found quantitative, although the Karnofsky index is about quality of life it is quantitative in substance. In Knyrims et al. (1993), he analysed his data using the Mann-Whitney test, Fisher's extract, and Kaplan Myer methods. Unlike the other articles examined whose results were shown with no information available about how they were statistically calculated.


Plastic Prostheses:

The rigid plastic stents used in the studies were of different manufacture, but were intrinsically the same in design. The largest lumen stents were used with the outer diameter being approx. 16mm and the inner diameter being approx. 12mm. In this group, patients were dilated up to 20mm endoscopically the day before stent placement by means of balloon catheters. The RPS was placed in position under general anaesthetic; this procedure was carried out endoscopically with fluoroscopic enhancement. Stent placement was checked post procedure with chest X-ray and radio-opaque swallows.

Self-Expanding Metal Stents.

The self-expanding metal stents used in the research were of two types, uncovered and covered metal mesh, again being of different manufacture. These endoprostheses were placed in position under fluoroscopic control only, thus eliminating the need for general anaesthetic. The patients had only sedation and no pre treatment of the stricture was necessary. In SEMS, "the basic design involves a cylindrical mesh of metal compressed into a sheath and mounted on a delivery system intended for sliding over a guide wire." Ellul et al. (1995). This enables a large lumen to be achieved on release with a minimal insertion diameter. (8 – 12mm introducers resulted in up to 24mm internal lumen after release). Due to the fact that only X-ray guidance was used, patients did not require any post placement tests to check validity. In some of the trials reviewed, these tests were carried out for data collection purposes only.

Insertion of the SEMS was successful in every instance compared with the RPS, where some oesophageal obstructions were unable to be negotiated despite a number of attempts.


Complications with the RPS were reported to be significantly, more frequent than with the SEMS. Ellul and Morgan (1996) reported that in 36% of cases, complications had taken place and Knyrim et al. reported that in 43% of his cases complications occurred with the insertion of the rigid stent.

Some of the complications seen include:

(a) The inability to negotiate the obstruction/stricture with the rigid stent even after pre dilation treatment. Multiple attempts were indicated in some trials with a 100% failure rate in 2.4% of cases (Knyrim et al.1993).

(b) Perforation of the oesophagus during the stent insertion procedure and post insertion due to migration of the stent was found to be as high as 8% by Ellul et al. (1996).

(c) Aspiration pneumonia was reported in this group, probably due to the use of general anaesthesia for the stent placement.

(d) Tension pneumothorax and pleural effusions were reported in, Knyrim et al. (1993), Ellul et al. (1996) and in Loizou, Grigg, Atkinson, Robertson, Brown (1991). These may have occurred because of trauma during the placement procedure, some of these tumours had to be manipulated with some force, again despite pre-treatment of the stricture with balloon dilatation. Heamatemesis occurred because of trauma especially on friable neoplasms.

(e) Migration of the stent and the reoccurrence of dysphagia were major problems in this group of patients. Migration occurred due to ill fitting stents, poor placement of the stent and also because of dislodgement of the endoprosthesis due to tumour growth. This is due to the rigid plastic stents not being flexible when the malignancy extends and grows, and it can dislodge the position of the stent. Recurrent dysphagia can be caused by migration of the stents and with overgrowth of the stent ends. More commonly as they have a smaller lumen than self-expanding metal stents and they can be blocked easily with food bolus’s. Loizou et al. (1991) indicated that half of the cases experiencing complications were caused by this problem. 52.4% of patients in Knyrims et al. (1993) study were effected just by migration and food impaction alone.

All of these complications were rectified with a variety of treatments ranging from antibiotics, endoscopic clearance of the stent, to retrieval and reinsertion or repositioning of the endoprosthesis. However, complications could effect the prognosis of the patient and lead to an earlier death. This happened in some of the studies.

Complications did arise with self-expanding metal stents, although the complication rates are indicated to be much lower. In Knyrims study, none of the latter problems occurs in the SEMS patients except for recurrent dysphagia. This was due to tumour regrowth and food bolus impaction, but Clements, Johnston, Mcllwrath, Spence and McGuigan (1996) reports one case of perforation of the oesophagus but no incidence of dysphagia due to tumour regrowth or food bolus. May, Mühdorfer, Hahn, Ell (1995) also stated that perforation on insertion of the stent occurred and also dysphagia due to regrowth was detected, however both he and Cowling et al. (1998) found that in some cases the incomplete expansion of the SEMS post insertion occurred.

Nobody else has reported this particular problem. It may have been the technique used in placement or due to the make of the stent, especially in May et al. (1995) study as 40% of the patients experienced this problem. It was rectified using post insertion balloon dilatation.

The most common complication seen in the self expanding metal stent group was ingrowth of the tumour through the uncovered mesh stents. The occurrence of this was high and was reported to occur in all of the studies reviewed that used uncovered SEMS.


In both groups, rigid plastic and self-expanding metal there was an improvement in the levels of dysphagia. All of the patients began with scores of 3 - 4 pre insertion of the endoprosthesis. Their levels of dysphagia were measured post insertion 24 hours later, patients with RPS; scores improved but were found in most cases to be 1-2. This was due to the lumen of the plastic stents being much smaller, and being unable to cope with solid food. Although dysphagia wasn't completely alleviated in this group, it was a marked improvement. In the SEMS group the dysphagia scores improved dramatically, most cases up to 0 - 1. This being that the lumens in SEMS were greater in size (up to 24mm).

Secondary dysphagia post insertion of the stent presented itself in both groups. In the rigid plastic stents it was due mainly to food impaction and migration, occurring in 15% of patients in Loizou et al. (1991) research and in a larger percentage in Knyrim et al. (1993) report, 33% of patients with plastic stents experienced reoccurring dysphagia.

In the SEMS group however migration of the stent was not reported often, and when it did occur it was found to be mainly in the covered stents. Raijman, Siddique, Ajani, Lynch (1997) reported 2.9% (3 out of 101 patients), Kato et al. (1996) reported 3.7% and Ellul et al. (1995) reported a migration of 9.0%. Although Ellul et al. (1995) migration levels where higher than some of the other trials, they are still considerably lower than the levels for the rigid plastic prosthesis.

The 30-day mortality percentages were found to be higher in the RPS. This may be due to a higher incidence of complications such as perforation / aspiration etc. In Knyrim et al. (1993) study, the rigid plastic prosthesis score was more than 50% greater than the self-expanding stents.

In Knyrim et al. (1993) study, the Karnofsky score was measured, although the results showed that both groups (RPS and SEMS) improved to the same level. The SEMS group had pre scores of 40 and the RPS group had pre scores of 50. Both groups increased up to 65, showing that the SEMS group actually improved more.

The length of stay in hospital varied across all the studies but was reported in most, to be that the patients with the self expanding metal stents stayed in hospital for less time and the patients with rigid plastic stents. The average length of stay in the RPS group was 9.1 days but in the SEMS group, it was 4.1 days (Knyrim et al.1993). This was due to the rigid plastic stent group having greater complication levels and also that they required a general anaesthetic for insertion.

It was found that the RPS patients had a higher incidence of reintervention due to complications and that hospital admissions during the following months were increased. Knyrim et al. (1993) showed the duration of hospital stay due to complications for RPS was 3.4 days but was only 1.9 for SEMS. Total hospital stays post placement for the RPS patients until the end of the study was more than 50 % higher than the SEMS patients. 5.4 days for the self-expanding metal stent patients and 12.5 for the rigid plastic stent patients.

Cost Effectiveness.

Rigid plastic stents are relatively inexpensive in comparison with the self-expanding metal stents. This would seem to make the use of SEMS questionable when alleviation of dysphagia with RPS is satisfactory, but as seen in a number of studies there are less complications than with rigid plastic stents and therefore the hospital stay is less .The insertion costs can be cheaper because only fluoroscopy is used for placement, only light sedation is used and no post insertion tests to verify placement are needed. Knyrim et al. (1993) explains that patients receiving SEMS could have the procedure carried out in an outpatient environment.

So, although the actual cost of the self-expanding metal stent is high, the overall cost, once factors such as hospital stay, cost of reintervention procedures and the fact that post insertion tests aren't needed, is comparable with the cost of the rigid plastic stent.


As the quality of life, and to some extent the quantity of life remaining to these patients depends to a large degree on their ability to swallow, the relief of dysphagia plays a vital role in the palliation of this disease. Treatment should ensure that the majority of these patients could avoid the consternation of total dysphagia, regardless to which stent is offered. It seems that there is an agreement that although rigid plastic stents relieved dysphagia sufficiently, the self-expanding metal stents have some outstanding merits in their favour. The insertion of the SEMS imposes less strain on the patient, can be carried out in an outpatient environment and has a lower complication rate than the RPS. The disadvantage of the self- expanding metal stent is that the costs are high. However, as discussed it is seen to be cost effective, and although that there is a higher incidence of tumour or overgrowth, it is still however has a lower reintervention rate than the rigid plastic stent. Relief of dysphagia can be achieved adequately with both types of stents; it emerges that there is a covenant that self-expanding metal stents are an analogous choice for the palliation of this excrescence.


Clements WDB, Johnston LR, Mcllwrath E, Spence RAJ, McGuigan J. (1996).

Self-expanding metal stents for malignant dysphagia. Journal of the Royal Society of Medicine; 89:454-456.

Cowling MG, Hale H, Grundy A. (1998). Management of malignant oesophageal obstruction with self-expanding metallic stents. British Journal of Surgery; 85:264-266.

Ellul JPM, Morgan R. (1996). Palliation of malignant dysphagia from oesophageal cancer. British Journal of Hospital Medicine; 55(5): 272-274.

Ellul JPM, Watkinson A, Khan RJK, Adam A, Mason RC. (1995). Self-expanding metal stents for the palliation of dysphagia due to inoperable oesophageal carcinoma. British Journal of Surgery; 82:1678-1681.

Kato M, Saji S, Kanematsu M, Hoshi H, Ishiguchi T, Kunieda K, Takao H, Sugiyama Y. (1996). Palliative therapy using polyurethane covered self-expandable metallic stents for malignant oesophageal strictures: experiences in six patients. Japanese Journal of Clinical Oncology; 26(6): 461-464.

Knyrim K, Wagner H, Bethge N, Keymling, Vakil N. (1993). A controlled trial of an expansile metal stent for palliation of esophageal obstruction due to inoperable cancer. The New England Journal of Medicine; 78:1302-1307.

Louzou LA, Grigg D, Atkinson M, Robertson C, Brown SG. (1991). A prospective comparison of laser therapy and intubation in endoscopic palliation for malignant dysphagia. Gasterenterology; 100:1303-1310.

May A, Selmaier M, Hochberger J, Gossner L, Mühdorfer S, Hahn EG, Ell C. (1995). Memory metal stents for palliation of malignant obstruction of the oesophagus and cardia. Gut; 37:309-313.

Muller JM, Erasmi H, Stelzner M. (1990). Surgical therapy of oesophageal carcinoma. British Journal of Surgery; 77:845-857.

Raijman JM, Siddique I, Ajani J, Lynch P. (1997). The overall complication rate after esophageal Wallstent is related to the esophageal location of the lesion. Gastrointestinal Endoscopy: Program and Extract issue; 45(4).



Home • Journals • Search • Rules for Authors • Submit a Paper • Sponsor us
Rules for Authors
Submit a Paper
Sponsor Us



Default text | Increase text size