Jung-Hoon Park 1 Ho-Young Song 1 Jin Hyoung Kim 1 Deok Ho Nam 2 Jae-Ik Bae 3 Min-Hee Ryu 4 Hwoon-Yong Jung 4

Transcription

1 Vascular and Interventional Radiology Original Research Park et al. Stent Placement to Treat Malignant Jejunal Obstruction Vascular and Interventional Radiology Original Research Jung-Hoon Park 1 Ho-Young Song 1 Jin Hyoung Kim 1 Deok Ho Nam 2 Jae-Ik ae 3 Min-Hee Ryu 4 Hwoon-Yong Jung 4 Park JH, Song HY, Kim JH, et al. Keywords: expandable metallic stent, gastric cancer, malignant jejunal obstruction, total gastrectomy DOI:1.2214/JR Received June 16, 211; accepted after revision September 5, Department of Radiology and Research Institute of Radiology, san Medical Center, University of Ulsan College of Medicine, 388-1, Poongnap 2-dong, Songpa-gu, Seoul , Republic of Korea. ddress correspondence to H. Y. Song (hysong@amc.seoul.kr). 2 Department of Radiology, Kyung Hee University Hospital at Gangdong, School of Medicine, Kyung Hee University, Seoul, Republic of Korea. 3 Department of Radiology, jou University School of Medicine, Suwon, Republic of Korea. 4 Department of Internal Medicine, san Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea. JR 212; 198: X/12/ merican Roentgen Ray Society Placement of a Covered Expandable Metallic Stent to Treat Nonanastomotic Malignant Jejunal Obstructions fter Total Gastrectomy With Esophagojejunostomy OJECTIVE. The objective of our study was to assess the technical feasibility and clinical effectiveness of expandable metallic stent placement in patients with nonanastomotic malignant jejunal obstruction after total gastrectomy with esophagojejunostomy. MTERILS ND METHODS. We retrospectively analyzed data from 21 patients with malignant jejunal obstruction after total gastrectomy with esophagojejunostomy who received one of two types of expandable metallic stent. Clinical effectiveness was assessed using the following variables: technical and clinical outcomes, complications, dysphagia scores before and after stent placement, patient survival, and stent patency. Complications with related interventions were evaluated and compared between the two stent types. RESULTS. Stent placement was technically successful in 2 of the 21 patients (95%) with 19 of 2 patients (95%) showing symptomatic improvement. Type stents were used in 1 patients and type stents in the remaining 1 patients. Complications occurred with seven of 2 stents (35%) and involved stent migration (n = 3), tumor overgrowth (n = 3), or pain (n = 1). The dysphagia score before stent placement (mean ± SD, 3.2 ±.5) had improved by 3 days after stent placement (1.3 ±.9, p <.1) and was maintained compared with the initial score up to 1 month (1.7 ± 1.1, p <.1) and 3 months (2.1 ± 1.5, p =.21) after stent placement. The median patient survival and stent patency were 114 and 46 days, respectively. The type of stent was not significantly related to complications (p =.35). CONCLUSION. Placement of expandable metallic stents to treat nonanastomotic malignant jejunal obstruction in patients who have undergone total gastrectomy with esophagojejunostomy is feasible and clinically effective. T otal gastrectomy with esophagojejunostomy has become an increasingly common operation for patients with gastric carcinoma and other gastric malignancies [1, 2]. fter total gastrectomy, tumor recurrence leads to anastomotic narrowing in 19% of patients and to jejunal obstruction in 4% of patients [3]. These recurrent obstructions lead to severe nausea, vomiting, and cachexia and to progressive deterioration [1 3]. Most tumors that recur after gastric surgery are unsuitable for further resective or palliative surgery. Fluoroscopic or endoscopic placement of expandable metallic stents has been used in the palliative treatment of unresectable recurrent malignant obstruction after gastric surgery [4 9]. Placement of expandable metallic stents is safe, easy, and clinically effective in patients with anastomotic strictures caused by recurrent malignancy [5, 6, 8, 9]. Most previous studies have focused on stent placement for recurrent anastomotic stricture after palliative gastric surgery, and there have been few case reports focusing on endoscopic, fluoroscopic, or percutaneous placement of jejunal stents for the palliation of malignant jejunal obstruction after gastric surgery [6, 1, 11]. dditional studies on the use of expandable metallic stents to treat malignant jejunal obstruction in patients who have undergone total gastrectomy are required on a larger scale. Therefore, the purpose of this study was to assess the technical feasibility and clinical effectiveness of expandable metallic stent placement in 21 patients with jejunal obstruction due to recurrent malignancy after total gastrectomy with esophagojejunostomy. Materials and Methods Patient Population Informed consent for stent placement was obtained from each patient. This retrospective study was approved by our institutional review board. JR:198, May

2 Park et al. The patient cohort consisted of 21 patients (14 men, seven women) ranging in age from 25 to 72 years (mean, 51 years) who underwent expandable metallic stent placement between July 23 and March 211 for palliation of severe nausea, dysphagia, and vomiting secondary to a recurrent malignant stricture at the jejunum distal to the anastomotic site. ll patients had previously undergone total gastrectomy with esophagojejunostomy for gastric cancer. Malignant jejunal obstructions occurred days (mean, 732 days) after gastric surgery. The inclusion criteria for stent placement were documented unresectable recurrent malignancy and nausea or vomiting indicating the need for stent placement in the jejunum. The exclusion criteria were patients with mild symptoms in whom an adult-sized endoscope could be passed through the malignant jejunal obstruction, clinical evidence of perforation or peritonitis, jejunal obstruction extending to the anastomotic site or esophagus, and peritoneal seeding with multiple small-bowel obstructions. The diagnosis of Fig. 1 Two types of stents used in this study., Fully covered retrievable nitinol stent., Partially covered dual stent. Photograph shows, from top to bottom, outer partially covered stent, inner bare nitinol stent, and assembled dual-expandable nitinol stent. recurrent malignancy at the stricture site was established by endoscopy and abdominal CT. The causes of malignant jejunal obstructions were recurrent gastric cancer (n = 13), metastatic infiltrative seeding mass (n = 6), and extrinsic compression by metastatic lymph nodes (n = 2). Stents and Placement Technique Two types of expandable metallic stents were used in this study (Fig. 1): a fully covered retrievable stent (type ) (Niti-S Esophageal covered stent, Taewoong) and a partially covered, nonretrievable dual stent (type ) (Hercules SP Pyloric, S&G iotech). oth ends of the type stents were 24 mm in diameter, and the type stents were 2- mm long when fully expanded; the body of the type stent was 16 mm in diameter. The type stent consisted of two stents: an outer, partially covered stent and an inner, bare nitinol stent. oth the inner and outer stents were 18 mm in diameter, with the ends flared up to 28 mm. detailed description of the stents and their delivery systems has been reported in previous studies [12, 13]. We routinely used the type stent because it can be removed. In areas with a high risk of stent migration or in highly tortuous stricture sites, type stents were used because of their low migration rates and ability to conform to tortuous strictures [12, 14] (Fig. 2). Endoscopy and a study using barium- or water-soluble contrast medium were performed 1 9 days before stent placement to evaluate the site, severity, and length of the stricture. Topical anesthetic of the pharynx was routinely applied using an aerosol spray before the procedure; conscious sedation was not used. detailed description of stent placement techniques and stricture measurement techniques has been described previously [12, 13]. In brief, the stricture was cannulated with a gastric catheter (Song-Lim, S&G iotech) and a.35-inch exchange guidewire (Radiofocus M, Terumo) under fluoroscopic guidance. fter the full length of the Fig. 2 arium studies of patients with nonanastomotic malignant jejunal obstruction after total gastrectomy with esophagojejunostomy., 68-year-old man. arium study obtained 3 days after placement of fully covered retrievable stent shows good patency of stent (arrows)., 64-year-old man. arium study obtained 1 day after placement of partially covered dual stent (arrows) shows good flow of barium through stent at tortuous jejunal loop. Fully covered retrievable nitinol stent returns immediately to its original shape, whereas partially covered dual stent better conforms to bend. 124 JR:198, May 212

3 Stent Placement to Treat Malignant Jejunal Obstruction stricture was delineated with a water-soluble contrast medium, a stent was placed over a stiff wire using fluoroscopic guidance. Follow-Up Patients were advised to resume oral intake of liquids within 24 hours of stent placement. Patients underwent a barium study 1 3 days after stent placement to verify the position and patency of the stent. If the barium study showed incomplete stent expansion, patients were not allowed a soft or solid diet until full stent expansion was seen at follow-up radiography. Patients also underwent a barium study 1 month after stent placement to identify any delayed complications (e.g., stent migration or obstruction). dditional follow-up in each patient involved monthly radiography and clinical examinations in an outpatient clinic. follow-up barium study, endoscopy study, or CT study was performed only in patients with recurrent symptoms. In cases in which it was not possible to perform clinical examinations, the patients or their families were contacted by telephone every 6 months. Information was obtained about nausea, vomiting, pain, and food intake capacity. Definition and nalysis of Data Outcomes were assessed with reference to the following variables: technical and clinical success, dysphagia scores before and after stent placement, complications, patient survival, and stent patency. Technical success was defined by successful insertion of the stent in the proper position and good passage of contrast medium through the stent under fluoroscopic guidance. To assess clinical improvement, reviewers scored the extent of subjective dysphagia before and after stent placement on a 5-point scale as follows:, normal swallowing; 1, ability to swallow a semisolid diet; 2, ability to swallow a soft diet; 3, ability to swallow liquids only; or 4, complete dysphagia. Dysphagia scores were assessed before stent placement and 3 days, 1 month, and 3 months after stent placement and were analyzed using the Wilcoxon signed rank test. Clinical success was defined as an increase of at least one grade in the dysphagia score within 3 days of stent placement. Recurrence of obstructive symptoms was defined as a decrease of at least one grade in the dysphagia score. Complications associated with type and stents with a focus on management were evaluated. The Fisher exact test was used to evaluate the relationship between complications and the type of stent. The overall patient survival and stent patency rates were calculated according to the Kaplan- Meier method. The stent patency period was defined as the period between initial stent placement and recurrence of obstructive symptoms. In calculating the stent patency period, patients were removed from the calculations if the stent remained patent throughout the patient s life. ll statistical analyses were performed using SPSS software (version 18, SPSS). two-sided p value of <.5 was considered to indicate statistical significance. TLE 2: Complications ccording to Stent Types and Secondary Treatments Complication No. of Cases TLE 1: ssessment of Dysphagia Scores efore and fter Stent Placement Score Score efore Stent Placement (n = 2) Type Stent (n = 1) Type Stent (n = 1) Mean Range Score fter Stent Placement 3 Days (n = 2) 3 Days (n = 18) 9 Days (n = 13) Grade Grade Grade Grade Grade Mean ± SD 3.2 ± ± ± ± 1.5 p before vs 3 days after <.1 3 days after <.1 9 days after.21 Note Data are the number of patients within each dysphagia score group. Results Technical and Clinical Outcomes Fluoroscopic placement of a stent was technically successful in 2 of the 21 patients (95%). In one patient, the guidewire could not pass through the stricture owing to the tightness of the stricture and acute angulation of the bowel loop. Of the 2 patients in whom stent placement was technically successful, 1 patients initially received a type stent and 1 received a type stent. Food intake capacity scores before and after stent placement are summarized in Table 1. The dysphagia score before stent placement (mean ± SD, 3.2 ±.5) had improved by 3 days after stent placement (1.3 ±.9, p <.1), and this improvement was maintained compared with the initial score at 1 month (1.7 ± 1.1, p <.1) and 3 months (2.1 ± 1.5, p =.21) after stent placement. Overall, 19 of 2 patients (95%) had improvement in their level of dietary intake of at least one grade. The one patient who showed no improvement in dysphagia required two stents because the stricture was long. In that patient, a contrast-enhanced study performed 1 day after stent placement showed acute angulation of the placed stents, with the passage of contrast medium from the stent to the distal jejunum being severely delayed. Recurrence of obstructive symptoms in the absence of stent-related complications occurred in five patients: Multiple small-bowel obstructions due to disease progression were noted days after stent placement (mean, 72 days). total of five patients did not undergo additional interventions. Second Treatment, (No. of Cases) Migration Removal and restenting (3).211 Overgrowth dditional stenting (3) >.999 Severe pain 1 Immediate None b (1) >.999 Total a Comparison of type stent versus type stent. b Pain subsided with analgesics. p a JR:198, May

4 Park et al. Complications and Secondary Treatments total of seven complications (35%, 7/2) occurred days after stent placement (mean, 51 days). Complications and related secondary treatments are summarized in Table 2. Stent migration occurred in three of the 1 patients who received a type stent: Partial downward migration occurred in two patients and complete downward migration in one patient. In the patients with partial migration, the stent was removed using a stent removal set 24 and 3 days after stent placement, respectively. In the patient with complete migration, the stent passed through the rectum 91 days after stent placement. In the three patients with type stent migration, a type stent was placed after removal of the type stent and no further migration occurred. Tumor overgrowth occurred in three patients (type stent, n = 2; type stent, n = 1) days after stent placement (mean, 63 days); all were treated with the placement of an additional stent. One patient complained of severe pain just after stent placement; however, the severe pain subsided with analgesics. The development of complications was not significantly associated with the type of stent used (p =.35) (Table 2). Patient Survival and Stent Patency Periods During the mean follow-up period of 158 days after stent placement (range, days), 18 of the 2 patients died. The median and mean survival periods were 114 days (95% CI, ) and 158 days (95% CI, ) (Fig. 3), respectively. locking of stent patency occurred in 11 patients because of recurrence of obstructive symptoms (n = 5), stent obstruction (n = 3), or Survival Rate (%) stent migration (n = 3). The median and mean stent patency periods were 46 days (95% CI, 9 83) and 58 days (95% CI, 32 85) (Fig. 3). Discussion We found that placement of a covered expandable metallic stent is effective as palliative therapy in patients with jejunal obstruction caused by recurrent cancer after total gastrectomy with esophagojejunostomy. Most of the recurrent tumors were unsuitable for further resective or palliative surgery, leaving palliative therapy with stent placement as the only option for relieving dysphagia and improving nutritional deficiencies. Stent placement to treat malignant jejunal obstruction in patients who have undergone total gastrectomy with esophagojejunostomy has been reported in a few studies with a small number of cases [6, 1, 11]. The results of our present study showed that palliative treatment with placement of an expandable metallic stent under fluoroscopic guidance is associated with a high technical success rate (95%, 2 of 21 patients) and a high clinical success rate (95%, 19 of 2 patients) in patients with malignant jejunal obstruction after total gastrectomy with esophagojejunostomy. Dietary intake improved in 19 of 2 patients (95%) after stent placement, which is comparable to the clinical success rate of stent placement in the anastomotic stricture caused by recurrent gastric cancer, which has been reported to be 8 9% [5, 6, 8, 15]. Our clinical success rate was higher than that observed in patients with recurrent anastomotic obstructions at the sites of gastrojejunostomy and esophagojejunostomy. In our study, malignant jejunal obstructions with the progress of disease occurred Stent Patency Rate (%) days (mean, 732 days) after gastric surgery. The decreases in the overall patient survival rate and the stent patency rate, in comparison with gastrointestinal obstruction or anastomotic strictures, are caused by multiple metastases from disease progression (n = 7) and by peritoneal carcinomatosis with multiple small-bowel obstructions (n = 5). Peritoneal carcinomatosis with multilevel obstructions or autonomic infiltration with dysmotility was presumed to be responsible for symptom persistence. Several authors have reported that peritoneal carcinomatosis is a relative contraindication to stent placement for enteral obstructions because of the risk of multiple small-bowel obstructions [12, 16 2]. However, Mendelsohn et al. [21] reported that stent placement in selected patients with early signs of carcinomatosis and a single focal area of obstruction was not associated with adverse clinical outcomes. In our study, six patients with an infiltrative seeding mass had a single focal obstruction; however, peritoneal carcinomatosis with multiple small-bowel obstructions as a result of disease progression occurred in five of these patients (5/6, 83%) days after stent placement (mean, 72 days). These patients showed recurrence of obstructive symptoms directly as a result of multiple small-bowel obstructions and died days after the onset of peritoneal carcinomatosis (mean, 89 days). It is important that careful examination for the presence of multiple obstructions from the stomach to the colon is performed by abdominal CT and a contrast medium study of the small bowel before and after stent placement. Stent migration occurred in three patients with type stents (15%). lthough there was Fig. 3 Survival and stent patency rates. and, Graphs show cumulative patient survival () and stent patency () rates after stent placement (Kaplan-Meier analysis). 126 JR:198, May 212

5 Stent Placement to Treat Malignant Jejunal Obstruction no significant difference between type and stents (p =.211), the incidence of stent migration was greater in patients with type stents. Type stents have low migration rates (2%) not only because the bare parts of the stent are incorporated into the jejunal wall, but also because the stent is fixed above and below the obstruction by means of the large proximal and distal bare rims of the outer stent [6, 12]. With regard to stent migration in this study, we usually attempted to remove the migrated type stents under fluoroscopic guidance with a retrievable hook [14]. We successfully addressed stent migration by replacing the type stents with type stents. Placement of a type stent seems to be more effective in cases with a risk of stent migration or with highly tortuous stricture sites. In our retrospective study, the rate of complications did not differ significantly between type stents and type stents (5/1 vs 2/1, respectively; p =.35). large-scale prospective study is needed to determine with greater certainty which type of stent is preferable. In conclusion, placement of expandable metallic stents in patients with nonanastomotic malignant jejunal obstruction after total gastrectomy with esophagojejunostomy is technically feasible and clinically effective. References 1. Paolini, Tosato F, Cassese M, et al. Total gastrectomy in the treatment of adenocarcinoma of the cardia: review of the results in 73 resected patients. m J Surg 1986; 151: utler J, Dubrow TJ, Trezona T, Klassen M, Nejdl RJ. Total gastrectomy in the treatment of advanced gastric cancer. m J Surg 1989; 158: 62 64; discussion, Levine MS, Fisher R, Rubesin SE, Laufer I, Herlinger G, Rosato EF. Complications after total gastrectomy and esophagojejunostomy: radiologic evaluation. JR 1991; 157: Johnsson E, Thune, Liedman. Palliation of malignant gastroduodenal obstruction with open surgical bypass or endoscopic stenting: clinical outcome and health economic evaluation. World J Surg 24; 28: Song HY, Kim TH, Choi EK, et al. Metallic stent placement in patients with recurrent cancer after gastrojejunostomy. J Vasc Interv Radiol 27; 18: Kim JH, Song HY, Shin JH, et al. nastomotic recurrence of gastric cancer after total gastrectomy with esophagojejunostomy: palliation with covered expandable metallic stents. J Vasc Interv Radiol 27; 18: Cho YK, Kim SW, Nam KW, et al. Clinical outcomes of self-expandable metal stents in palliation of malignant anastomotic strictures caused by recurrent gastric cancer. World J Gastroenterol 29; 15: Yang ZQ, Song HY, Kim JH, et al. Covered stent placement in patients with recurrent cancer after a illroth I reconstruction. J Vasc Interv Radiol 27; 18: Tong DK-H, Law S, Wong KH. The use of selfexpanding metallic stents (SEMS) is effective in symptom palliation from recurrent tumor after esophagogastrectomy for cancer. Dis Esophagus 21; 23: Ota Y, Takagi Y, Osaka Y, et al. case of stenosis of the reconstructed jejunum due to recurrent cancer after total gastrectomy in which stent-in-stent placement was effective [in Japanese]. Gan To Kagaku Tyoho 28; 35: Takayama S, Ochi Y, Yasuda, et al. Palliative percutaneous jejunal stent for patients with short bowel syndrome. Case Rep Gastroenterol 29; 3: Song HY, Shin JH, Yoon CJ, et al. dual expandable nitinol stent: experience in 12 patients with malignant gastroduodenal strictures. J Vasc Interv Radiol 24; 15: Song HY, Lee DH, Seo TS, et al. Retrievable covered nitinol stents: experiences in 18 patients with malignant esophageal strictures. J Vasc Interv Radiol 22; 13: Ko HK, Song HY, Shin JH, Lee GH, Jung HY, Park SI. Fate of migrated esophageal and gastroduodenal stents: experience in 7 patients. J Vasc Interv Radiol 27; 18: Kim J, Choi IJ, Kim CG, et al. Self-expandable metallic stent placement for malignant obstruction in patients with locally recurrent gastric cancer. Surg Endosc 211; 25: aron TH, Harewood GC. Enteral self-expandable metal stents. Gastrointest Endosc 23; 58: Sabharwal T, Irani FG, dam ; Cardiovascular and Interventional Radiological Society of Europe. Quality assurance guidelines for placement of gastroduodenal stents. Cardiovasc Intervent Radiol 27; 3: Katsanos K, Sabharwal T, dam. Stenting of the upper gastrointestinal tract: current status. Cardiovasc Intervent Radiol 21; 33: Lopera JE, razzini, Gonzeles, Castaneda- Zuniga WR. Gastroduodenal stent placement: current status. RadioGraphics 24; 24: Morgan R, dam. Use of metallic stents and balloons in the esophagus and gastrointestinal tract. J Vasc Interv Radiol 21; 12: Mendelsohn R, Gerdes H, Markowitz J, Di- Maio CJ, Shattner M. Carcinomatosis is not a contraindication to enteral stenting in selected patients with malignant gastric outlet obstruction. Gastrointest Endosc 211; 73: JR:198, May