Expandable Metal Stents in Achalasia—Is There a Role?Sandeep Mukherjee, M.D., David S. Kaplan, M.D., Gulshan Parasher, M.D., and Michael S. Sipple, M.D.Division of Gastroenterology, Department of Medicine, State University of New York Health Science Centerat Syracuse, Syracuse, New York; and Division of Gastroenterology, Department of Medicine, CrouseHospital, Syracuse, New York

OBJECTIVE: Achalasia is treated with pneumatic dilation ormyotomy, and botulinum toxin injections are occasionallyused. We review our community’s experience with expand-able metal stents in six patients who failed medical treat-ment or were poor surgical candidates.

METHODS: Eight stents were placed in six patients betweenJuly 1995 and November 1997. Four patients had achalasiaand two pseudoachalasia. Four patients underwent succes-sive botulinum toxin injections. One patient only agreed toperiodic Maloney dilatations or a stent. Pneumatic dilationwas performed in one patient and considered high risk in therest. All were poor surgical candidates. Three differentstents were used: Gianturco Rosch Z stent, Wallstent I, andWallstent II.

RESULTS: One-month mortality and morbidity were 33%and 50%, respectively. Two patients were asymptomatic ona liquid diet for$6 months but required repeat endoscopyfor recurrent dysphagia because of food bolus impaction andproximal stent migration in each.

CONCLUSIONS: Expandable metal stents in achalasia orpseudoachalasia do not provide sustained symptom relief,and their use is associated with unacceptably high morbidityand mortality. We do not recommend the use of thesedevices in patients who have failed medical therapy or whoare poor surgical candidates. (Am J Gastroenterol 2000;95:2185–2188. © 2000 by Am. Coll. of Gastroenterology)

INTRODUCTION

Achalasia is a disorder of esophageal motility characterizedby aperistalsis, elevated lower esophageal sphincter (LES)pressure, and failure of LES relaxation upon swallowing (1,2). It is treated with pneumatic dilation or myotomy and,recently, botulinum toxin injections (botox) have been de-scribed (3, 4). An additional treatment option that has cometo our attention was the use of expandable metal stents(EMS) for achalasia. This technique had also been described

by De Palmaet al. with encouraging results (5). In ourcommunity, these stents had been used as a last resort inpatients who had failed medical therapy or who were poorsurgical candidates, with reportedly excellent results. Wereviewed the clinical course of these patients to determinewhether this was a valid intervention.

MATERIALS AND METHODS

Eligibility CriteriaBetween July 1995 and November 1997, EMS were placedin six adults (patients A–F) with achalasia or pseudoacha-lasia who had failed medical therapy or were poor surgicalcandidates. The leading symptom for all patients was dys-phagia for solids (100%) and liquids (50%). Diagnosis wasconfirmed by barium swallow and/or esophageal manome-try and all patients underwent esophagogastroduodenoscopy(EGD) (6). All patients had undergone at least one dilationand successive botox injections were performed in fourpatients (66%) (7–9). Informed consent was obtained fromall patients.

Stent MaterialsThree different EMS were used: Gianturco Rosch Z stent(Wilson Cook Medical, Winston Salem, NC) and WallstentI and II (Schneider USA, Plymouth, MN). The GianturcoRosch stent consists of a wide Z stainless steel wire meshcoated by a polyethylene film with small barbs in the centerto prevent migration. Unconstrained stent length is 8 cm,and this stent releases without shortening to a diameter of 18mm.

The Wallstent consists of a self-expanding, cross-linkedstainless steel mesh coated with polyurethane. Uncon-strained length is 10 cm and unconstrained diameter 20 mm.The Wallstent position can be altered before 75% deploy-ment but shrinks by approximately 30% upon expansion.Once implanted, Wallstent removal is usually impossible.Each stent is compressed into a 6-mm (18-Fr) deliverysystem on a support catheter and sheathed by a lubricateddouble membrane. The membrane is manually unrolledunder a pressure of 3–4 atm for stent deployment. Two

Presented at the Annual Meeting of the American Gastroenterological Association,Digestive Diseases Week, May 17, 1998, New Orleans, LA, and published in abstractform (Gastroenterology 1998;114:A732).

THE AMERICAN JOURNAL OF GASTROENTEROLOGY Vol. 95, No. 9, 2000© 2000 by Am. Coll. of Gastroenterology ISSN 0002-9270/00/$20.00Published by Elsevier Science Inc. PII S0002-9270(00)01093-5

radiopaque markings on the support catheter allow radio-logical supervision for stent placement (10–13).

Implantation TechniqueThe procedures were performed under standard endoscopicconditions with fluoroscopy. All patients received topicalanesthesia, sedation (midazolam 1–5 mgi.v.) and whennecessary opiate analgesics (meperidine 50–150 mgi.v.).General anesthesia was not required.

During EGD (with the patient in the left lateral position),a radiopaque marker was attached to the patient’s torso tomark the level of the gastroesophageal junction. The endo-scope was removed after placement of a 0.038 inch hydro-philic guidewire into the stomach. The EMS was advancedalong the guide wire under continuous fluoroscopy anddeployed in the technique discussed above (14, 15).

Follow-UpPatients D and F had routine chest radiographs 48 h afterstent deployment to check stent position before feeding.However, follow-up examinations in the rest of the patientswere scheduled according to their symptoms and medicalstatus. All surviving patients continue to be followed.

RESULTS

Eight EMS were placed in six patients between July 1995and November 1997 (Table 1). There were four men andtwo women. Patient A was 31 yr old but the average age ofthe other five patients was 81 yr. Four patients had achalasiaand two pseudoachalasia (we defined pseudoachalasia asachalasia in the presence of metastatic adenocarcinoma inthe presence of normal gastroesophageal junction biopsies).Pseudoachalsia is usually diagnosed by endoscopic biopsieswhich have a false-negative rate of up to 25% (16). Allpatients underwent barium swallows and EGD, and manom-etry was performed in four patients (A, B, C, and E). Threedifferent EMS were used: the Gianturco Rosch Z stent

(patients A and F), Wallstent I (patients D, E, and F) andWallstent II (patients B and C).

The 1-month mortality and morbidity were 33% (patientsB and C died of myocardial infarctions) and 50%, respec-tively. The Gianturco Rosch Z stent in patient A slippedupon deployment and was removed immediately. This stentwas also removed from patient F 2 days after placementbecause of proximal migration above the gastroesophagealjunction. The range of symptom relief among all patientsvaried between 2 wk (patients B and C) and 10 months(patient F) before further intervention.

Recurrent dysphagia in patient B occurred because ofproximal stent occlusion from a highly dilated and tortuousesophagus, which eventually obstructed an otherwise patentstent. Removal of a food bolus in patient D 6 months afterEMS placement and strict maintenance on a liquid diet haskept her symptom-free. However, recurrent dysphagia inpatient E developed secondary to an inflammatory strictureand food bolus, which required repeated Savary dilationsand high-dose proton pump inhibitors. She eventually diedfrom an exsanguinating gastrointestinal bleed 8 months afterEMS placement. Autopsy revealed a gastroesophageal ulceradjacent to her stent, which had penetrated the aorta. Aftera symptom-free interval of 10 months, patient F required asecond EMS because of proximal stent migration.

As of September 1998, four patients (66%) have died(patient A, of metastatic esophageal cancer; patients B andC, of myocardial infarctions; and patient E, of GI bleed).Patients D and F have no dysphagia on a liquid diet andproton pump inhibitors after a food bolus removal andsecond EMS placement, respectively, averaging 14 monthssince their original treatment.

DISCUSSION

In this small series of six patients, only two patients with theWallstent I had a symptom-free interval an average of 8

Table 1. Clinical Course of Patients With Expandable Metal Stents for Achalasia or Pseudoachalasia

Case* Diagnosis Manometry Dilation Botox Stent Duration Comments

(A) 31F Pseudoachalasia Inc. LES 32 32 GianturcoRosch

Slippedimmediately

Death from metastatic esophageal cancer

(B) 81 F Achalasia Aperistalsis 32 32 Wallstent II 2 wks Proximal stent occlusion by tortuousesophagus; death from myocardialinfarction

(C) 81 M Pseudoachalasia Tertiarycontractions

33 32 Wallstent II 2 wks Death from myocardial infarction

(D) 81 F Achalasia Not done 36 Not done Wallstent I 6 months Stent occlusion by food bolus(E) 80 M Achalasia Aperistalsis 34 32 Wallstent I 3 months Inflammatory stricture required dilation;

death from gastrointestinal bleed(F) 86 M Achalasia Not done 36 Not done Gianturco

Rosch2 days Proximal stent migration

Wallstent I 10 months Proximal stent migrationGianturco

Rosch.7 months Occlusion by food bolus

* Case data include case/patient letter (A–F), age (yr), and gender (F5 female, M5 male).

2186 Mukherjee et al. AJG – Vol. 95, No. 9, 2000

months before further intervention. This interval may havebeen increased if there had been better compliance with aliquid diet. Complications such as inflammatory strictureand stent migration were more common with these stents butcould also be a reflection of these patients’ longer survival.The stricture in patient D was secondary to iatrogenic gas-troesophageal reflux disease induced by EMS deploymentacross the gastroesophageal junction, and subsequent pa-tients (E, F) were all started on proton pump inhibitors.

A very disturbing complication was the aorta-enteric fis-tula in patient E from a penetrating gastroesophageal junc-tion ulcer adjacent to the EMS. This may have occurred fortwo reasons. First, Wallstents have higher expansile forcescompared to other EMS. This has the advantage of treatingvery tight esophageal strictures but may be associated withsevere chest pain, esophageal perforation, or even esopha-geal hemorrhage (17). In their study on self-expandingmetal stents in malignant esophageal obstruction, Schmass-manet al. reported esophageal perforations in 23% of theirpatients but none in those who received the knitted nitinolstent (18). Secondly, achalasia itself may be a risk factor forEMS-related perforation. Unlike the desmoplasia and fibro-sis commonly associated with esophageal cancers, achalasiais characterized by a hypertensive LES and often hugeproximal dilation. These characteristics, together with thepressure of an expansile metal foreign body, may signifi-cantly increase the risk for perforation in contrast to that inpatients with malignant strictures.

De Palmaet al. recently described their experience withEMS in eight patients with achalasia. The average age was67.5 yr and all patients had failed myotomy and/or multipleballoon dilations or botox injections (two cases) (5). Fournitinol coil stents (Instent, Eden Praire, MN) and four cov-ered ultraflex stents (Microvasive, Boston Scientific,Quincy, MA) were inserted in eight patients. All patientswere able to eat a normal diet during an average follow-upperiod of 9 months. Complications occurred in two patients:in one, the nitinol stent migrated per rectum postdeploymentand was replaced by an ultraflex stent, and a second patientdeveloped mild reflux esophagitis that was controlled withmedical therapy.

Stent migration occurred in our patients as well, with twoGianturco Rosch Z stents removed endoscopically, one eachfrom patients A and F. As the stent in patient F remained inthe esophagus, it was easily removed with a medium-sizedpolypectomy snare and diagnostic gastroscope. An overtubewas not required, as the technique was straight-forward andthe polyethlene-coated stent protected the patient from anystent trauma. However, stent removal from patient A wasmuch more difficult after it had slipped into the stomach.The stent was eventually removed by a novel techniqueusing a modified snare in which the loop emerged perpen-dicular to the long axis of the snare (20).The open loop wasinitially placed proximal to the distal tip of the gastroscope

and closed before performing EGD. The stent was intubatedby the gastroscope, the snare opened, and with the aid offluoroscopy, carefully advanced over the stent. The snarewas tightly closed around the stent and the entire apparatusremoved uneventfully from the patient. Before using thistechnique we were repeatedly unsuccessful using a standardpolypectomy snare, with our multiple attempts lasting ap-proximately 2 h. However, we were successful in our firstattempt with the modified snare.

Institutional Review Board approval was not obtained, asthere was initially no intention of performing a study. Theattending gastroenterologist who was responsibe believedthat a signed consent form would suffice for these patients,who were treated on a highly individualized basis as theywere extremely debilitated from their underlying illness andcomorbidities, had failed to respond to previous attempts atachalasia treatment and were too sick to undergo surgery.Before providing informed consent, the patients and theirrelatives were made aware that this procedure was risky,that success could not be guaranteed, and that complicationswere possible. However, as we became aware of the ex-tremely disappointing results in our community, we werevery concerned that this procedure was not indicated, as ourresults were in contrast to the encouraging results reportedby De Palmaet al. (5). In addition, these procedures wereperformed by two different gastroenterologists who wereinitially unaware of the other physician’s results. Conse-quently, we believed that it was important to make themembers of our medical community aware that, althoughtechnically feasible, the results were unacceptable and theprocedure could not be justified. This was the impetus forour study.

In conclusion, we do not recommend using EMS inachalasia or pseudoachalasia. Unlike the experience of DePalmaet al. (5), the overall results in our community werevery poor, with complications or death occurring in all ofour patients. Although the best results were with the Wall-stent I, both patients required repeat endoscopic therapy.Both Gianturco Rosch stents were removed within 2 daysand the Wallstent II was effective for only 2 wk. Theextremely high morbidity and mortality of this interventionare of grave concern, given that untreated achalasia is usu-ally a nonmorbid and nonfatal illness. We wish to makephysicians aware that, although managing such patients canbe difficult and at times frustrating, we do not advocateEMS as a “last resort.” Laparoscopic surgery, anticholin-ergics to reduce salivation, or alternative nutritional routesmay be preferred.

Request reprints and correspondence:Sandeep Mukherjee,M.D., Section of Gastroenterology, 983285 Nebraska MedicalCenter, Omaha, NE 68198-3285.

Received Jan. 3, 2000; accepted May 15, 2000.

2187AJG – September, 2000 Expandable Metal Stents in Achalasia

REFERENCES

1. Gadenstatter M, Mehra D. Achalasia: Dilatation or myotomy.A continuing enigma. Am J Gastroenterol 1997;92:1572–3.

2. Hinder RA, Napoliello DA, Klinger PJ. Cutting edge manage-ment of achalasia. Am J Gastroenterol 1997;92:1078–9.

3. Pasricha PJ, Kalloo AN. Recent advances in the treatment ofachalasia. Gastrointest Endosc Clin North Am 1997;2:191–206.

4. Birgisson S, Richter JE. Achalasia. What’s new in treatment.Dig Dis 1997;15:1–27.

5. De Palma GD, Sivero L, Galloro G, et al. Self expanding metalstents for endoscopic treatment of the esophageal achalasia.Gastrointest Endosc 1998;47:A15.

6. Richter JE. Motility disorders of the esophagus. In: Yamada T,Alpers DH, Owyang C, et al., eds. Textbook of gastroenter-ology, 2nd ed. Philadelphia: JB Lippincott, 1995:1182–94.

7. Vallera RA, Brazer SR. Botulinum toxin for suspected acha-lasia. Am J Gastroenterol 1995;90:1319–21.

8. Bernstein D, Barkin JS. Botulinum toxin for suspected achal-sia. Am J Gastroenterol 1995;90:1319–21.

9. Pasricha PJ, Ravich WJ, Hendrix TR, et al. Intrasphinctericbotulinum toxin for the treatment of achalasia. N Engl J Med1995;332:774–8.

10. Axelrad AM, Fleischer DE, Gomes M. Nitinol coil esophagealprosthesis: Advantages of removable self expanding metalstents. Gastrointest Endosc 1996;43:155–60.

11. Sturgess RP, Morriss AI. Metal stents in the esophagus. Gut1995;37:593–4.

12. Song HY, Do YS, Han YM, et al. Covered expandable esoph-ageal metallic stent tubes. Experience in 119 patients. Radi-ology 1994;193:689–95.

13. Domschke W, Foerster ECL, Matek W, et al. Self expandingmesh stent for esophageal cancer stenosis. Endoscopy 1990;20:134–6.

14. Ell C, Hochberger J, May A, et al. Coated and uncoated selfexpanding metal stents for malignant stenosis in the upper GItract: Preliminary clinical experience with Wallstents. Am JGastroenterol 1994;89:1496–1500.

15. Tytgat GNJ, Bartelsman JFWM, Den Hartog Jager FCA, et al.Upper intestinal, and biliary tract endoprosthesis. Dig Dis Sci1986;31:57S–76S.

16. Kahrilas P, Kishk SM, Helm JF, et al. Comparison ofpseudoachalasia and pseudoachalasia. Am J Med 1987;82:439–46.

17. Angueira CE, Kadakia SC. Esophageal stents for inoperableesophageal cancer: Which to use? Am J Gastroenterol 1997;92:373–6 (editorial).

18. Schmassman A, Meyenberger C, Knuchel J, et al. Self ex-panding metal stent in malignant esophageal obstruction: Acomparison between two stent types. Am J Gastroenterol1997;92:400–6.

19. Meshkinpour H, Haghighat P, Meshkinpour A. Quality of lifeamong patients treated for achalasia. Dig Dis Sci 1996;41:352–6.

20. Mukherjee S, Kaplan DS. A new technique for removingdisplaced esophageal metal stents. Am J Gastroenterol 1999;94:1109–10 (letter).

2188 Mukherjee et al. AJG – Vol. 95, No. 9, 2000