`GASTROINTESTINAL ENDOSCOPY
`Copyright © 1992 by the American Society for Gastrointestinal Endoscopy
`
`Treatment of malignant esophageal
`obstruction with silicone-coated metallic self-
`
`expanding stents
`
`John Schaer, MD, Ronald M. Katon, MD
`Krassi Ivancev, MD, Barry Uchida, BS, FIT
`Josef Bosch, MD, Kenneth Binmoeller, MD
`Portland. Oregon
`
`Six patients with high-grade malignant esophageal obstruction were treated with
`silicone-coated metallic self-expanding esophageal stents (Z stents). Endoscopic
`placement of stents was well tolerated. All patients achieved excellent palliation,
`defined by a decrease of at least two dysphagia grades, which was sustained.
`Complications occurred during follow-up in four patients and included stent
`migration, silicone disruption with tumor ingrowth, food impaction, and perforation
`(discovered at autopsy) at the distal stent site. Three of the four complications
`were promptly treated by endoscopic or radiologic intervention. Recent
`modification in stent design and placement technique will hopefully reduce
`complications. The self-expanding stent has several theoreticat advantages over
`the rigid plastic stem and Nd:YAG laser for palliation of obstructing esophageal
`malignancy. (Gastrointest Endosc 1992;38:7-11)
`
`A variety of therapies are employed in the palliative
`treatment of malignant esophageal obstruction. Un—
`fortunately, no single modality provides effective, safe,
`inexpensive relief of dysphagia in all cases. Two of the
`more commonly used modalities are photocoagulation
`with the Nd:YAG laser and the placement of rigid
`plastic endoprostheses. Nd:YAG laser treatment is
`limited by its high cost, the requirement for frequent
`treatment sessions, and tumor recurrence."3 The use
`of rigid plastic endoprostheses is plagued by high
`complication rates.H Recently, self-expanding metal-
`lic stents have shown promise in the treatment of
`biliary obstruction.9 It seemed possible that a modified
`version of a self-expanding metallic stent might pro-
`vide benefit in refractory esophageal obstruction. This
`report describes our experience in six consecutive
`patients with malignant esophageal obstruction pal—
`liated with silicone-coated self—expanding metallic Z
`stents.
`
`Received July 10, 19.91. For revision August 1, 1991. Accepted August
`26, 199:.
`
`Mm the Department of Medicine, Division of Gastroenteroiogy, and
`The Charles Dotter Institute for Interventionat Therapy, Oregon
`Health Sciences University, Porttand, Oregon. Reprint requests: Ron—
`std M. Katon, MD. Division of Gostroenteroiogy, L461. Oregon
`Hearts Sciences University, 3181 aw. Sam Jackson Park Rd,
`Portland, Oregon 97201-3098.
`
`VOLUME 38, No. 1,1992
`
`PATIENTS AND METHODS
`
`Patients
`
`Six patients with malignant esophageal obstruction were
`studied. All six had severe grade 3 or 4 dysphagia, and none
`were Operative candidates. Mean age was 71 years (range,
`57 to 87 years). There were four men and two women. Three
`patients had esophageal squamous cell carcinoma, two pa-
`tients had esophageal adenocarcinoma, and one patient had
`an adenocarcinoma of the lung which was metastatic to the
`mediastinum. Five of the six patients had been refractory to
`prior therapies. All patients had extensive obstructive le-
`sions in the middle or distal esophagus. In three patients,
`the lesion extended to the gastroesophageal junction (Table
`1).
`A total of nine stents were placed in six patients. Informed
`consent was obtained under a compassionate use basis. Five
`of six patients were treated in an outpatient setting, while
`the sixth was hoepitalized for 24 hours. Patients were first
`evaluated with a barium esophagram and endoscopy with
`the Olympus GIFXVIO endoscope to determine the location
`and length of esophageal obstruction. Five of the six patients
`underwent initial dilation with American Endoscopy Dila—
`tors (Mentor, Ohio) to 42 to 45 French diameter.
`
`Stents
`
`The prostheses used were modified self—expanding Z
`stents coated with a silicone membrane and were homemade
`
`7
`
`W.L. Gore & Associates, Inc.
`W.L. Gore & Associates, Inc.
`Exhibit 1003-1
`Exhibit 1003-1
`
`
`
`Table 1.
`
`Patient demographics
`
`Prior treatment
`Tumor type L123?“
`Patient Age Sex
`1
`67 M ESOPH-SCCA" (27—32} Laserx 8,chemo—
`therapy, radia-
`tion
`
`2
`3
`4
`5
`6
`
`(30—40) Pericardial window
`Lung-ADCA
`F
`70
`5'? M ESOPH-SCCA (37-47? Radiation
`S? M ESOPH-SCCA (37—47)” Radiation
`62 M ESOPH ADCA (28—36) Chemotherapy
`83 F ESOPH ADCA (28—34? Laser X 2, dilation,
`alcohol inj. X 3
`
`“ SCCA, squamous cell carcinoma; ADCA. adenocarcinoma.
`” Tumor extended beyond gastroesophageal junction.
`
`in our institution. Stents were handmade from 0.018-inch
`stainless steel wire. Individual stent bodies had an internal
`diameter of 15 mm and a length of 2 cm. Multiple stent
`bodies were interconnected for a total stent length of 8 to
`12 cm. The last four patients treated had a flanged proximal
`end which had a 20-min internal diameter (Fig. 1). All stents
`had 3-mm long wire hooks on the exterior of the stent that
`“anchored” the stent within the tumor.
`
`Stent placement technique
`
`Stents were placed under fluoroscopic control using light
`conscious sedation. A 14 French orogastric tube was first
`passed distal to the esophageal obstruction. Two guidewires
`were placed through the orogastric tube, which was then
`removed. Over one guidewire a 5.5 French angiography
`catheter was placed for contrast injection. Over the second
`guidewire, an 18 French Teflon stent delivery catheter was
`placed. The stent was then pushed out of the delivery
`catheter under fluoroscopic guidance and delivered into the
`esophagus.
`Patients underwent a follow-up esophagram and endo—
`scopic examination immediately after stent placement and
`also at 1 week, 4 weeks, 8 weeks, and as needed thereafter.
`Clinical assessment was performed at monthly intervals
`until patient death or the completion of the study period. In
`the three patients whose stent extended beyond the gastro—
`esophageal junction, H2 blockers and anti-reflux measures
`were prescribed prophylactically to minimize gastroesopha-
`geal reflux. Mean duration of follow-up was 4 months. The
`severity of dysphagia was quantified with a 0 to 4 dysphagia
`scale as follows: 0 = no dysphagia; 1 = dysphagia to normal
`solids; 2 = dysphagia to soft solids; 3 = dysphagia to solids
`and liquids; and 4 = inability to swallow saliva.
`
`RESULTS
`
`Stent placement was successful and well tolerated
`in all six patients. There were no complications noted
`at the time of stent placement. Mean dysphagia grade
`prior to stenting was 3.3. All patients had improve-
`ment of at least two dysphagia grades after stent
`placement, and mean dysphagia grade post-stenting
`was 0.67 (Table 2). Endoscopic and radiologic exami-
`nation confirmed the improvement in esophageal lu—
`
`8
`
`minal diameter in each patient. In four of six patients,
`the stents expanded to maximal diameter within 24
`hours, while in two patients the stents expanded to
`approximately 75% of their maximal diameter. Relief
`of dysphagia was sustained in all patients with a mean
`follow-up of 4 months. Four patients died during fol-
`low-up with a mean survival of 4 months (Table 2).
`Figures 2 and 3 depict an 83—year—old woman with
`adenocarcinoma of the esophagus (patient 6, Tables 1
`and 2). She had undergone five esophageal dilations,
`four
`treatments with ethanol
`injection, and two
`Nd:YAG laser treatments prior to stenting. She had
`complete dysphagia and was receiving enteral alimen—
`tation through a gastrostomy tube. After stenting she
`was able to ingest a semi-solid diet. Improvement has
`been sustained for 3 months and she no longer requires
`gastrostomy feedings. Figures 4 and 5 represent a 70-
`year-old woman with adenocarcinoma of the lung and
`mediastinal metastases (patient 2, Tables 1 and 2).
`She had an extra-esophageal tumor with esophageal
`compression and complete dysphagia. With placement
`
`
`
`Figure 1. Stainless steel self-expanding Z stent with silicone
`membrane.
`
`GASTROINTESTINAL ENDOSCOPY
`
`W.L. Gore & Associates, Inc.
`W.L. Gore & Associates, Inc.
`Exhibit 1003-2
`Exhibit 1003-2
`
`
`
`Table 2.
`
`Results and complications of sell-expandable slant
`Dysphagia grade
`Patient
`
`Pre«stcnt
`Post—stout
`
`Complications
`
`Follow—up
`
`1
`
`2
`3
`4
`5
`
`3
`
`4
`3
`3
`3
`
`Foreign body sensation,
`torn membrane with
`tumor ingrowth
`None
`Shoot migration
`Meat impaction
`Distal esophageal pcr—
`foration
`
`Alive {3 mo)
`None
`4
`1
`6
`
`X=0.6T2:33 X=4mo
`
`
`1
`
`0
`0
`1
`1
`
`Died {8 mo)
`
`Died {2 mo)
`Died {3.5 mo)
`Alive (5 mo)
`Died {2 mo}
`
`
`
`1.-
`
`
`
`Figure 2. Barium swallow. A, Approximately lo-cm long
`compression of mid-esophagus with near total obstruction.
`8. Full stent expansion and excellent flow of barium.
`
`of two overlapping stents (total length 14 cm), she was
`able to take a regular diet until her death from wet
`astatic disease 2 months later.
`
`Complications
`
`Two patients experienced no complications. In the
`remaining four patients, five complications were en-
`countered as follows (Table 2). In patient 1, the use
`of an inadequate 14 French diameter delivery catheter
`resulted in tears in the silicone coating of the stent
`Which allowed tumor ingrowth. In subsequent pa-
`tients, we employed an 18 French delivery catheter,
`
`VOLUME 38, NO. 1.19.92
`
`and no further injury to the silicone membrane oc—
`curred. This problem was treated by placing a second
`stent within the first stent, which provided relief of
`dysphagia and blocked tumor ingrowth. This patient
`was also the only patient with a foreign body sensation
`in the chest after stenting. This sensation persisted,
`but was easily controlled with Tylenol. Patient 3 suf—
`fered migration of his stent into the stomach. The
`stent was retrieved endoscopically by grasping it with
`a polypectomy snare and pulling it into an overtube.
`A second stent with a flanged proximal border was
`then placed. All four stents placed subsequently had
`this flanged stent design. No other problems with
`stent migration were encountered. Patient 4 had one
`episode of meat impaction, which was easily cleared
`endoscopically. Patient 6 died 2 months following
`stent placement, and was found at autopsy to have an
`esophageal perforation near the distal margin of the
`stent. The perforation occurred within tissue invaded
`by malignancy and it was not clear whether the per-
`foration occurred from stent trauma or from degen-
`eration of the esophageal tumor. Three other patients
`died at 2 months, 3.5 months, and 8 months following
`stent insertion. No autopsies were obtained in these
`patients.
`
`DISCUSSION
`
`Self-expanding metallic Z stents were found to be
`efficacious in the palliative therapy of malignant
`esophageal obstruction. Despite the selection of pa-
`tients with challenging strictures, an esophageal lu-
`men adequate to maintain oral alimentation was
`achieved in all six patients. In each patient, dysphagia
`improved by at least two grades, and improvement
`was sustained in all patients.
`The use of self-expanding Z stents has several po-
`tential advantages over thYAG laser or rigid plastic
`stents in the palliation of malignant esophageal
`stenosis. The Nd:YAG laser is expensive, requires
`frequent treatment sessions, and recurrence of ob-
`struction after the initial series of treatments is a
`
`9
`
`W.L. Gore & Associates, Inc.
`W.L. Gore & Associates, Inc.
`Exhibit 1003-3
`Exhibit 1003-3
`
`
`
`problem.H In addition, submucosal lesions are inac—
`cessible to treatment with the Nd:YAG laser. Al—
`
`though rigid endoprostheses are relatively inexpensive
`and readily available, complication rates are high.
`Perforation rates range between 7 and 18%, and stent
`migration, tumor overgrowth, and bleeding are trou—
`blesome.H
`
`Endoscopic placement of the Z stent is gentle and
`does not generate the shearing forces necessary for
`right stent insertion. Stent placement is a one-stage
`procedure, obviating the need for multiple sessions as
`in Nd:YAG laser treatment.
`
`In this investigation there was no morbidity en-
`
`7
`
`1
`
`a
`
`_JEL .4
`
`Figure 3. Endoscopic photographs. A, Pin hole size lumen
`prior to stenting. B, Marked improvement in luminal diameter.
`Star“ is partially obscured by retained barium.
`
`countered during placement of the stents in any pa-
`tient. Stent placement was performed with light can-
`scious sedation and in five of six patients was accom-
`plished in an outpatient setting. Pre-stenting dilation
`was performed to a modest 42 to 45 French diameter
`in comparison to the 54 to 60 French diameter dila-
`tions usually required for placement of rigid endo-
`prosthesesf”8 Another potential advantage of Z stents
`over rigid stents is in the management of tumor over-
`growth. With Z stents an overlapping stent can be
`placed at either end of the original stent to, in effect,
`extend the stented region. In patient 1, an overlapping
`stent was successfully used to cover a defect in the
`silicone membrane, which had allowed tumor en-
`croachment. This capability is not possible with rigid
`stents. Whether stent migration will be less common
`with the flanged Z stents then with rigid stents re-
`mains to be determined in larger studies.
`The complications encountered with Z stents during
`the follow-up period were easily managed with endo-
`scopic or radiologic interventions in three of four
`patients. Modifications in stent design, such as the
`flanged proximal stent body used to prevent migra-
`tion, should decrease future similar problems.
`In the one case of esophageal perforation found at
`autopsy, a question remains as to whether the pointed
`distal ends of the stent may have contributed to the
`perforation. Accordingly, future stents will have yet
`another design modification, with rounded wires at
`the distal end of the stent and a distal silicone bumper
`
`
`
`Figure 4. Barium swallow. A, Long. irregular, near-total obstruction in distal esophagus. B, Expansion of stent is nearly complete.
`except in its mid-portion, which is expanded to 70 to 75%. 0, Excellent flow at barium through the stent.
`
`10
`
`GASTROINTESTINAL ENDOSCOPY
`
`W.L. Gore & Associates, Inc.
`W.L. Gore & Associates, Inc.
`Exhibit 1003-4
`Exhibit 1003-4
`
`
`
`
`
`Figure 5. Endoscopic photographs. A, Nodular tumor with near-total esophageal obstruction. B. Proximal view of stem with full
`expansion. 0, View of partially expanded stent in mid-esophagus. Note visualization of esophageal wall through silicone
`membrane.
`
`added to prevent trauma to the esophageal mucosa.
`Three other patients died during the course of this
`study, at 2 months, 3.5 months, and 8 months after
`stent insertion. Unfortunately, autopsies could not be
`obtained in these patients, so no further information
`exists regarding possible inflammation and/or injury
`to the esophageal tissue from these metallic stents.
`Long—term placement of metallic stents in the biliary
`tract has not resulted in perforation or necrosis.9 As
`with rigid stents, the only patients unsuitable for the
`Z stent are those with neoplasms completely obstruct-
`ing the esophagus or in close proximity to the crico-
`pharyngens.
`The mesh stent (Wallstent-Medinvent SA) is an-
`other seif—expanding stent design which has been em—
`ployed in a small number of patients for palliation of
`esophageal I.'I:|alignancy.m'11 Due to their inherent de-
`sign, mesh stents have openings between the wire
`filaments which cannot be coated, therefore allowing
`tumor ingrowth. This complication occurred in two of
`six patients reported by Knyrim et a1.” The silicone
`coating of the Z stent provides a barrier to tumor
`ingrowth. Also, due to this intact barrier, Z stents may
`have an application in the treatment of tracheo-esOph-
`ageal fistula, whereas mesh stents appear unsuitable
`for this purpose.
`Finally,
`the central question is how will self-
`expanding stents fit into the armamentarium available
`for palliation of esophageal malignancy? Future com-
`parative trials between Nd:YAG laser and stents of
`
`various designs, including both rigid and expandable
`stents, are needed to determine the optimal applica-
`
`tion of these various modalities in the palliative treat—
`ment of malignant esophageal stenosis.
`
`REFERENCES
`
`1. Fleischer D, Kessler F. Endoacopic Nd:YAG laser therapy for
`carcinoma of the esophagus: a new form of palliative treatment.
`Gastroenterology 1983;85:600—6.
`2. Mellow MH, Pinkas H. Endoscopic therapy for esophageal
`carcinoma with Nd:YAG laser: prospective evaluation of effi-
`cacy,
`complications,
`and survival. Gastrointest Endosc
`1984;30:334—9.
`3. Fleischer D, Sivak M. Endoscopic Nd:YAG laser therapy as
`palliation for oesophagogastric carcinoma. Gastroenterology
`1935;89:827—31.
`4. Ogilivie AL, Droniield MW, Ferguson R, Atkinson M. Palliative
`intubation of oesophagogastric neoplasms at fibreoptic endos—
`copy. Gut 1984;23:1060—7.
`5. Tytgat GNJ, Huibregste K, Bartlesman JFWM, Den Hartog
`Jaeger FCA. Endoscopic palliative therapy of gastrointestinal
`and biliary tumors with prostheses. Clin Gsstroenterol
`1986;15:249-71.
`6. Loizou LA, Grigg D, Atkinson M, Robertson C, Brown SG. A
`prospective comparison of laser therapy and intubation in en—
`doscopic palliation for malignant dysphagia. Gsstroenterology
`1991;100:130340.
`7. Graham DY, Dobbs SM, Zubler M. What is the role of prosthe-
`ses insertion in esophageal carcinoma? Gastrointest Endosc
`1983;29:1—5.
`8. Buset M, Des Marez B, Cremer M. Endoscopic palliative intu-
`bation of the esophagus invaded by lung cancer. Gastrointest
`Enclose 1990;36:357—9.
`9. Huibregtse K, Cheng J, Coene PPLO, Fockens P, 'Tytgat GNJ.
`Endoscopic placement of expandable metal stents for biliary
`strictures—a preliminary report on experience with 33 patients.
`Endoscopy 1989;21:280—2.
`10. Knyrim K, Wagner HJ, Pausch J, Vakil N, Starch E. Expand-
`able metal stents for the palliative treatment of es0phageal
`obstruction [Abstract]. Gastrointest Endosc 1990;36:236.
`11. Domschke W, Foerster ECH, Matek W, Rodl W. Self-expand-
`ing mesh stent for esophageal cancer stenosis. Endoscopy
`1990;22:134—6.
`
`VOLUME 38, NO. I, 1992
`
`ll
`
`W.L. Gore & Associates, Inc.
`W.L. Gore & Associates, Inc.
`Exhibit 1003-5
`Exhibit 1003-5
`
`