Fluoroscopically Guided Placement of Self-Expandable Metallic Stents and Stent-Grafts in the Treatment of Acute Malignant Colorectal Obstruction

Transcription

1 Fluoroscopically Guided Placement of Self-Expandable Metallic Stents and Stent-Grafts in the Treatment of Acute Malignant Colorectal Obstruction Honsoul Kim, MD, Seung Hyoung Kim, MD, Sun Young Choi, MD, Kwang-Hun Lee, MD, PhD, Jong Yoon Won, MD, Do Yun Lee, MD, and Jong Tae Lee, MD PURPOSE: To evaluate the technical feasibility and clinical effectiveness of fluoroscopically guided placement of self-expandable metallic stents and stent-grafts for acute malignant colorectal obstruction. MATERIALS AND METHODS: Radiologic images and clinical reports of 42 patients (22 men, 20 women; age range, years; median age, 65.5 years) who underwent fluoroscopically guided colorectal stent insertion without endoscopic assistance for acute malignant obstruction were reviewed retrospectively. Eighteen patients received bare stents as a bridge to surgery. Twenty-four patients received 27 insertions of either a bare stent (n 15) or a stent-graft (n 12) for palliation. The obstruction was located in the rectum (n 8), sigmoid (n 17), descending (n 8), splenic flexure (n 3), and transverse (n 6). RESULTS: Clinical success, defined as more than 50% dilatation of the stent with subsequent symptomatic improvement, was achieved in 41 of the 42 patients (98%). No major procedure-related complications occurred. Minor complications occurred in eight of the 45 procedures (18%). No perioperative mortalities occurred within 1 month after surgery. In the palliative group, the median stent patency was 62 days (range, 0 1,014 days). There was no statistically significant difference in stent patency between the bare stents (range, days; median, 68 days) and stent-grafts (range, 1 1,014 days; median, 81 days). CONCLUSIONS: Fluoroscopically guided placement of self-expandable metallic stents and stent-grafts for the relief of acute malignant colorectal obstruction was technically feasible without endoscopic assistance even in lesions proximal to the splenic flexure and transverse and clinically effective in both bridge to surgery and palliative management. J Vasc Interv Radiol 2008; 19: From the Department of Radiology and Research Institute of Radiological Science, Yonsei University College of Medicine, 134 Shinchon-dong, Seodaemun-gu, Seoul , Republic of Korea (H.K., S.Y.C., K.H.L., J.Y.W., D.Y.L., J.T.L.) and the Department of Radiology, Cheju University College of Medicine, Cheju, Republic of Korea (S.H.K.). Received December 8, 2007; final revision received August 15, 2008; accepted August 19, Address correspondence to K.H.L.; doctorlkh@ yuhs.ac None of the authors have identified a conflict of interest. SIR, 2008 DOI: /j.jvir ACUTE colorectal malignant obstruction has been traditionally managed with surgical approaches. In patients with unresectable cancer, surgical colostomy for palliative purposes has been the classic method (1,2). In patients with resectable cancer, the Hartman procedure with or without subsequent repair of colostomy or tumor resection combined with primary in an emergency setting are usually performed (1 4). The results of a controlled study of emergency surgery for cancer, however, were unfavorable. Smothers et al (2) reported a 34% and 64% surgical mortality and morbidity rate, respectively, in emergency settings, compared to 6.9% and 24% mortality and morbidity in elective settings, reflecting a considerable amount of additional risk. Recently, self-expandable metallic stents have emerged as a promising alternative to surgical methods for reducing perioperative mortality and morbidity in cases of acute malignant colorectal obstruction (1,3 20); however, other studies usually describe 1709

2 1710 Metallic Stents and Stent-Grafts in Malignant Colorectal Obstruction December 2008 JVIR the metallic stent to be delivered with endoscopy alone or with fluoroscopic guidance with endoscopic assistance. Moreover, most of the obstructing lesions were confined to the rectosigmoid area, which is technically not difficult. Herein, we present our experience with fluoroscopically guided placement of self-expandable metallic stents and stent-grafts without endoscopic assistance in acute malignant colorectal obstructions, which included technically challenging lesions proximal to the splenic flexure of the and transverse, and compared the patency of the metallic stents and stent-grafts. MATERIALS AND METHODS This retrospective study was approved by the institutional review board. Patients From February 2003 to June 2005, the images and clinical reports from 42 consecutive patients (22 men, 20 women; age range, years; median age, 65.5 years) with acute malignant colorectal obstruction who underwent fluoroscopically guided placement of self-expandable metallic stents and stent-grafts were reviewed retrospectively. All patients had clinical features of acute bowel obstruction (eg, abdominal distention and pain combined with bowel habit changes). Supplemental radiologic features of acute bowel obstruction were present in all patients. The cause of obstruction was primary colorectal cancer in 31 patients and secondary metastatic malignancy (advanced gastric cancer in eight patients, cervical cancer in two patients, and pancreatic cancer in one patient) in the remaining 11. The obstruction sites were confirmed with computed tomography (CT) to be the rectum (n 8), sigmoid (n 17), descending (n 8), splenic flexure (n 3), and transverse (n 6). Stent Placement The inclusion criterion for fluoroscopically guided self-expandable metallic stent placement was acute malignant colorectal obstruction. Patients Figure 1. Radiographs in a 76-year old man (patient no. 18) with acute high-level obstruction of the large bowel. Acute obstruction symptoms were relieved with fluoroscopically guided placement of a self-expandable metallic stent. (a) Upright radiograph of the abdomen shows the step-ladder sign of mechanical obstruction. A CT scan showed obstruction at the transverse by cancer (not shown). (b) With the patient in a prone position, the catheter (catheter tip, arrow) was advanced just distal to the obstruction site at the mid-transverse. Note the curvature of the splenic flexure of the, which causes difficulties for catheter and guide wire manipulation and stent placement. This acute curvature was overcome with manipulation of the stiff guide wire, which resulted in the successful placement of the self-expandable metallic stent. (c) The acute curvature of the splenic flexure was flattened by a stiff guide wire, and the self-expandable metallic stent (arrow) was deployed. (d) Upright radiograph of the abdomen the day after the procedure shows complete resolution of the acute obstruction. whose obstruction site was located proximal to the hepatic flexure were excluded. Informed consent for the procedure was obtained from each patient and/or his or her legal guardian. Stent placement was performed under fluoroscopic guidance without endoscopic assistance. Patients were placed in the prone or lateral decubitus position, and a 26-F rectal catheter (Sewoon Medical, Seoul, Korea) was inserted into the rectum. This was followed by balloon inflation with 40 ml of room air. A double-contrast study using a water-soluble contrast medium (Telebrix; Lothalamate Sodium, Guerbet, Aulnay-sous-Bois, France) and room air was performed to verify the colorectal course and distal portion of the obstructive lesion. A inch Radiofocus hydrophilic nitinol guide wire (Terumo, Tokyo, Japan) and a 5-F angiographic catheter (Cobra, C1 or Headhunter, HN1; Cook, Bloomington, Indiana) were then inserted coaxially into the rectal catheter

3 Volume 19 Number 12 Kim et al 1711 Table 1 Summary of Treatment Outcomes in the Bridge to Surgery Group Patient No./Sex/ Age (y) 1/M/80 2/M/79 3/M/56 4/F/28 5/M/66 6/F/82 7/M/39 8/M/75 9/F/65 10/F/76 11/F/60 12/M/50 13/M/70 14/M/50 15/F/78 16/M/70 17/F/73 18/M/76 Cancer Diagnosis Location of Obstruction Previous Treatment Technical Success Stent Type Subsequent Treatment Postoperative Complication Transverse No Bare Resection and Colon twisting (Re-Op) Rectum No Bare Resection and Anastomosis site disruption Rectum No Bare Resection and None Transverse No Bare Resection and None Transverse No Bare Resection and None Note. Procedural-related transient hematochezia occurred in three patients. Survival (d) 459 (hospice care) 399 (hospice care) 1,187 (lost to follow-up) 616 (lost to follow-up) 51 (died) 818 (died) 890 (undergoing follow-up) 161 (lost to follow-up) 1,117 (undergoing follow-up) 943 (undergoing follow-up) 127 (lost to follow-up) 964 (undergoing follow-up) 1,128 (lost to follow-up) 1,386 (undergoing follow-up) 1,535 (undergoing follow-up) 838 (lost to follow-up) 1,474 (undergoing follow-up) 990 (undergoing follow-up) and passed over the obstructive lesion under fluoroscopic guidance. When the 5-F angiographic catheter could no longer pass through the tight stricture even though the guide wire had been passed a 4-F gliding catheter (Terumo) was substituted for the 5-F catheter. After the catheter tip was confirmed to hold position at the proximal side of the obstruction by means of contrast medium injection, the wire was exchanged with a inch Amplatz super stiff guide wire (Boston Scientific/Medi-tech, Miami, Florida). In our experience, one of the more technically difficult portions of the procedure is negotiating the tortuous sigmoid curvature in cases with obstruction proximal to the sigmoid. We have found that straightening of the sigmoid can be achieved by manipulating a inch Amplatz super-stiff guide wire back and forth. In cases of obstruction proximal to the splenic flexure, an 8-F guiding sheath (Balkin sheath, Cook) was placed in advance into the descending for easier selection of the obstructing core. This guiding sheath kept the sigmoid straight and helped select the obstructing core by enhancing the torque manipulation of the 5-F catheter and inch hydrophilic nitinol guide wire. Finally, a self-expandable metallic stent of appropriate size was inserted over the inch Amplatz super-stiff guide wire. The metallic stent measured either 22 or 24 mm in diameter. The length of the metallic stent was determined so that at least an additional 4 cm (2 cm extra-long on each side of the proximal and the distal portions of the actual stricture) would be provided in addition to the actual stricture length to cover the entire lesion. If the self-expandable metallic stent insertion was intended for bowel decompression as a bridge to surgery, a bare stent (Hanarostent, M.I.Tech, Seoul, Korea; 22 mm in diameter, 8 14 cm in length) loaded in an 8-F introducer was used. If stent placement was performed for palliative treatment in an unresectable/in-

4 1712 Metallic Stents and Stent-Grafts in Malignant Colorectal Obstruction December 2008 JVIR Table 2 Summary of Treatment Outcomes in the Palliative Group Patient No./Sex/ Age (y) Cancer Diagnosis Location of Obstruction Previous Treatment Technical Success 1/M/56 Pancreas Splenic flexure Pancreatectomy 2/M/66 Stomach Rectum Palliative RTG 3/F/62 Stomach Rectum RSTG 4/M/63 Rectum Lower anterior resection 5/M/74 Rectum Lower anterior resection 6/F/88 (1st stent) Sigmoid None 6/F/89 (2nd stent) Sigmoid Bare stent 6/F/89 (3rd stent) Sigmoid Stent-graft insertion 7/M/62 Descending Anterior resection 8/F/33 Cervix Sigmoid RAH and BSO 9/M/77 Descending None (refused) 10/F/51 Transverse None 11/F/63 Cervix Sigmoid None 12/M/64 Splenic flexure None 13/F/52 Sigmoid None 14/M/35 Sigmoid Anterior resection 15/F/80 Rectum Lower anterior resection 16/F/85 Stomach Sigmoid RSTG 17/M/79 Sigmoid None 18/F/65 Stomach Transverse RSTG 19/M/36 Stomach Splenic flexure None 20/F/60 Stomach Transverse RSTG 21/F/93 and esophagus Rectum Esophageal metallic (double primary) stent insertion 22/M/83 Stomach Sigmoid Pyloric stent insertion 23/F/34 (1st stent) Stomach Sigmoid RSTG 23/F/34 (2nd stent) Stomach Sigmoid RSTG, metallic stent insertion 24/F/88 Sigmoid None Note. Three patients had transient hemathochezia. RSTG radical subtotal gastrectomy, RTG radical total gastrectomy, RAH radical abdominal hysterectomy, BSO bilateral salpingo-oophorectomy. operable case, either a metallic stent (Hanarostent) or an outer polyester stent-graft with inner bare metallic stent (S&G Biotech, Seoul, Korea; 24 mm in diameter, 6 10 cm length) designed to prevent stent-graft migration and to reduce the diameter of the stent introducer was used. The introducer size of the stent-graft was 10 F. Definitions Technical success was defined as self-expandable metallic stent placement that covered the obstructing lesion as well as at least 2 cm of normal bowel proximal and distal to the lesion. Clinical success was declared when expansion of the metallic stent was confirmed to be at least 50% in diameter, evidence suggesting improvement of the obstruction appeared at subsequent radiologic evaluations, gas and stool passage were restored, and symptoms improved. Minor complications were defined as events that caused no significant clinical sequelae necessitating further therapy other than overnight observation (eg, transient cancer bleeding, most likely the result of minor trauma caused by the guide wire during lumen selection of the tumor core). Major complications were defined as those necessitating therapy, an unplanned increase in the level of care, or prolonged hospitalization or those that resulted in permanent adverse sequelae (eg, cancer bleeding necessitating transfusion or perforation) (21). Stent patency was declared when no evidence of obstructive symptoms and signs could be found at subsequent clinical and radiologic evaluations. Statistical Analysis In the palliative group, the period of patency of the obstructive lesion after stent insertion was analyzed to see if a difference existed between bare metallic stents and stent-grafts. The Kaplan-Meier method was used in the analysis, and SPSS version 14.0 for Windows software (SPSS, Chicago, Illinois) was used for the actual calculation.

5 Volume 19 Number 12 Kim et al 1713 Clinical Success Stent Type Subsequent Treatment Morbidity Patency Survival (d) Graft None Hematochezia (died) Graft None None (died) Graft None None (lost to follow-up) Bare None None (died) Graft None None (hospice care) Bare None Tumor ingrowth, (died) stent fracture Graft Graft stent reinsertion, Migration 219 chemotherapy Graft None Hematochezia 26 Bare None None (died) No Bare Palliative colostomy Persistent 0 51 (hospice care) obstruction Bare None None (died) Bare None None (died) Graft None Tenesmus (lost to follow-up) Bare None None (hospice care) Bare Abscess drainage None (hospice care) Bare None None 2 2 (died) Graft Chemotherapy, radiation Anal pain 1,014 1,014 (undergoing follow-up) therapy Graft None None (lost to follow-up) Graft None None (died) Bare None None (died) Bare None None (hospice care) Bare None None (lost to follow-up) Bare None Hematochezia 1 1 (hospice care) Bare None None (died) Graft Metallic stent reinsertion Tumor ingrowth (died) Graft None None 1 Bare None None (hospice care) The patency period of the sef-expandable metallic stent and stent-graft was defined as the time between stent insertion and the occurrence of one of the following: reobstruction of the inserted stent, patient death, or refusal of additional treatment by the patient and/or guardian accompanied by hospice care. If the specific time could not be documented (eg, in a patient who did not revisit the outpatient clinic and was classified as a follow-up loss or in a patient who was still regularly visiting the outpatient clinic without event at study termination), then the data from the specific patient were classified as censored and processed accordingly. RESULTS Overall Initial Technical and Clinical Success Technical success of fluoroscopically guided placement of self-expandable metallic stents was achieved in all patients, including those with lesions of high-level obstruction (Fig 1). Clinical success, defined as more than 50% expansion of the stent with subsequent symptomatic improvement, was achieved in 41 of the 42 patients (98%). The patient who failed to achieve clinical success had underlying cervical cancer with carcinomatosis. Although precise positioning and full expansion of the self-expandable metallic stent was documented, the obstruction was not relieved because hard fecal lumps stuck to the stent lumen and caused obstruction. Five days after insertion of the metallic stent, the patient had to receive a colostomy. Complications There were no procedural-related major complications, and minor complications occurred in eight of the 45 procedures (18%). After stent placement, six patients reported newly developed hematochezia; however, the bleeding lasted no longer than 7 days and was self-limited in five of the patients. One patient refused further

6 1714 Metallic Stents and Stent-Grafts in Malignant Colorectal Obstruction December 2008 JVIR Table 3 Patency Analysis of Bare Stents in Patients Who Received Stents for Palliative Purpose Patency (d) Status Cumulative Patency Survival Standard Error Cumulative Events No. Remaining 0 Event Event Event Event Event Event Censored Event Event Event Event Event Event Event Event Table 4 Patency Analysis of Stent-Grafts in Patients Who Received Stents for Palliative Purpose Patency (d) Status Cumulative Patency Survival Standard Error Cumulative Events No. Remaining 1 Event Event Event Censored Event Event Event Censored Event Event Censored 8 1 1,014 Censored 8 0 treatment and was discharged for hospice care the day after stent insertion. Thus, the course of hematochezia in this patient is unknown. Tenesmus was reported in one patient who had underlying cervical cancer. She had two separate obstructive lesions, one in the proximal rectum and another in the proximal sigmoid, and received two stent-graft insertions. After adjusting to her stents, her tenesmus resolved. Intermittent anal pain was reported in one patient who had recurrent rectal cancer. A double-contrast study revealed a lesion extending to the distal rectum, so irritation by the inserted stent was expected. This patient also adjusted to her stent, and at present she is under observation and has not presented additional symptoms for 1,014 days after the procedure. Because none of our stent placements involved the anorectal junction, fecal incontinence was not reported. Bridge to Surgery Group The data for the bridge to surgery group are included in Table 1. Among the group of 31 patients who had primary colorectal cancer, 18 underwent subsequent surgical resection after stent insertion, and the stent acted as a temporary measure to decompress obstruction. The use of self-expandable metallic stents was successful in all cases. Stool and gas passage were restored within 48 hours. Each patient underwent a subsequent elective tumor resection within 6 41 days after stent placement. Preoperative mechanical bowel preparation with polyethylene glycol solution was successfully achieved without complications. No immediate postoperative mortality has occurred, but perioperative morbidity occurred in two of the 18 patients (11%). In one patient, site disruption was confirmed with oscopy 21 days after surgery. Fortunately, the lesion was limited focally by extensive surrounding adhesions. After conservative treatment, the patient was discharged and additional surgical interventions were not necessary. The other patient developed a twisted bowel loop proximal to the resection site, and manual reduction was conducted. Among the 18 patients who underwent a subsequent surgery after stent insertion, eight are currently in a followup program without adverse events (two died, six were lost to follow-up, two were discharged to hospice care). The survival period for these patients,

7 Volume 19 Number 12 Kim et al 1715 Figure 2. Graph shows cumulative patency (Kaplan-Meier method) of bare stents and stent-grafts in patients who received stent insertion for palliative purposes (P.39, not statistically significant by log-rank analysis). which was last checked when the study ended, ranged from 51 to 1,535 days (median, days) (Table 1). Palliative Group The data for the palliative group are included in Table 2. Self-expandable metallic stents and stent-grafts were used in 24 patients to relieve the large bowel obstruction without surgery. The overall patency of the 27 stent insertion procedures in the palliative group ranged from 0 to 1,014 days (median, 62 days) at study termination. The patency of bare stents in 15 procedures ranged from 0 to 855 days (median, 68 days). The patency of stent-grafts in 12 procedures ranged from 1 to 1,014 days (median, 81 days). With the bare stent, one of 15 procedures (6.7%) was classified as censored data. With the stent-grafts, four of 12 procedures (33%) were classified as censored data. The P value calculated by the Log-rank test comparing the patency period between the bare stent and stent-graft insertion groups was calculated to be 0.39, suggesting no statistically significant difference in the patency period between the two groups (Tables 3, 4; Fig 2). During the follow-up period, 10 patients died from advanced cancer without episodes of reobstruction. Two patients had to undergo additional stent-graft placements due to tumor ingrowth and migration. The first patient underwent three stent insertions. She initially received a bare stent for sigmoid cancer, which resisted obstruction for 195 days. Obstructive symptoms reappeared, and a stent fracture with tumor ingrowth was identified. She received a stentgraft, which functioned for another 219 days, but migration of her second stent occurred shortly after chemotherapy. She died 26 days after undergoing a third stent-graft insertion. Another patient underwent a stentgraft insertion for sigmoid obstruction secondary to stomach cancer metastasis, but obstruction recurred 62 days after the procedure. Tumor obstruction proximal to the margin of the first stent was noted, and the patient received a second stent-graft. However, she died the day after her second stent insertion. DISCUSSION The insertion of self-expandable metallic stents can achieve not only effective decompression of obstructive symptoms, but also benefits in morbidity and mortality during the subsequent elective surgery (4,20). The chance of successful primary instead of stoma formation after resection of the cancer segment is also thought to be markedly increased (8). This is most likely due to adequate bowel preparation and sufficient decompression, which makes more technically feasible when compared with the markedly dilated bowels encountered in emergency settings (9,10,18). Furthermore, the degree of venous engorgement and potential ischemia caused by acute bowel obstruction is expected to be relieved. As a result, the promoted healing process could provide a better environment for surgery. The results of this study seem to support these theories, as no immediate postoperative mortalities occurred. Although the placement of a self-expandable metallic stent failed to prevent surgical morbidities in two cases (morbidity rate, 11%), eventually all 18 patients in the bridge to surgery group succeeded in receiving primary instead of colostomy, which would have been a very unusual result without the placement of a self-expandable metallic stent before surgery. For patients in the palliative group, self-expandable metallic stents and stentgrafts were effective in restoring and maintaining bowel luminal patency, even in the absence of stoma formation (5,11,12,17 19). Lesions located at or more proximal to the splenic flexure are technically more challenging due to the remote distance from the anus and the tortuous course of the. Our study showed a technical success rate of 100% even when including three patients with a lesion at the splenic flexure and six patients with a lesion at the transverse, which provides encouraging data for fluoroscopically guided self-expandable metallic stent placement. The median period of patency in the palliative group with bare stents and stent-grafts was 68 and 81 days, respectively. The group that underwent stent-graft insertion showed a slightly longer period of patency than did the group that received bare stent insertion, but the difference was statistically insignificant. The statistically insignificant findings observed in this study seem to be from the small number of patients and the relatively high ratio of censored data. Some complications occurred, including newly developed hematochezia, tenesmus, and anal pain, which

8 1716 Metallic Stents and Stent-Grafts in Malignant Colorectal Obstruction December 2008 JVIR seemed to be related to the procedure or the self-expandable metallic stent itself. Hematochezia was most likely the result of mechanical trauma caused by the guide wire during the selection procedure. However, as previously mentioned, with the exception of one case in which the ultimate outcome is still unknown, the remaining cases were self-limiting in nature. Anal pain and tenesmus were probably the result of irritation caused by insertion of the metallic stent near anorectal junction (10,11). Even if some discomfort is felt at metallic stent insertion, the patient most likely will adjust, as was demonstrated in our study. Even though there were some complications, in most cases, the clinical significance is minor (10). In summary, fluoroscopically guided placement of self-expandable metallic stents and stent-grafts in patients with acute obstruction caused by malignant lesions is highly effective and relatively safe in both the bridge to surgery group and the palliative treatment group. The stent insertion procedure is technically feasible, with a reliable clinical success rate with fluoroscopic guidance and without endoscopic assistance even in lesions at the splenic flexure and transverse. There was no significant difference in the patency period between bare stents and stentgrafts in patients who received the stent insertion for a palliative purpose. This result would be one of the limitations of this study due to the small number of patients and relatively large number of censored data. In addition, the results were not evaluated according to the pathologic type of the lesion and the concurrent chemotherapy. Further study with regard to these points would be expected. In conclusion, fluoroscopic placement of self-expandable metallic stents and stent-grafts for the relief of acute malignant colorectal obstruction was technically feasible without endoscopic assistance even in the lesions proximal to splenic flexure and transverse and clinically effective in both bridge to surgery and palliative management. References 1. Maetani I, Tada T, Ukita T, et al. Self-expandable metallic stent placement as palliative treatment of obstructed colorectal carcinoma. J Gastroenterol 2004; 39: Smothers L, Hynan L, Fleming J, Turnage R, Simmang C, Anthony T. Emergency surgery for carcinoma. Dis Colon Rectum 2003; 46: Camunez F, Echenagusia A, Simo G, Turegano F, Vazquez J, Barreiro-Meiro I. Malignant colorectal obstruction treated by means of self-expanding metallic stents: effectiveness before surgery and in palliation. Radiology 2000; 216: Martinez-Santos C, Lobato RF, Fradejas JM, Pinto I, Ortega-Deballon P, Moreno-Azcoita M. Self-expandable stent before elective surgery vs. emergency surgery for the treatment of malignant colorectal obstructions: comparison of primary and morbidity rates. Dis Colon Rectum 2002; 45: Mucci-Hennekinne S, Kervegant AG, Regenet N, et al. Management of acute malignant large-bowel obstruction with self-expanding metal stent. Surg Endosc 2007; 21: Mergener K, Kozarek RA. Stenting of the gastrointestinal tract. Dig Dis 2002; 20: Saida Y, Sumiyama Y, Nagao J, Uramatsu M. Long-term prognosis of preoperative bridge to surgery expandable metallic stent insertion for obstructive colorectal cancer: comparison with emergency operation. Dis Colon Rectum 2003; 46:S44 S Law WL, Chu KW, Ho JW, Tung HM, Law SY, Chu KM. Self-expanding metallic stent in the treatment of ic obstruction caused by advanced malignancies. Dis Colon Rectum 2000; 43: Dauphine CE, Tan P, Beart RW Jr, Vukasin P, Cohen H, Corman ML. Placement of self-expanding metal stents for acute malignant large-bowel obstruction: a collective review. Ann Surg Oncol 2002; 9: Balague C, Targarona EM, Sainz S, et al. Minimally invasive treatment for obstructive tumors of the left : endoluminal self-expanding metal stent and laparoscopic colectomy: preliminary results. Dig Surg 2004; 21: Johnson R, Marsh R, Corson J, Seymour K. A comparison of two methods of palliation of large bowel obstruction due to irremovable cancer. Ann R Coll Surg Engl 2004; 86: Liberman H, Adams DR, Blatchford GJ, Ternent CA, Christensen MA, Thorson AG. Clinical use of the selfexpanding metallic stent in the management of colorectal cancer. Am J Surg 2000; 180: , discussion Aviv RI, Shyamalan G, Watkinson A, Tibballs J, Ogunbaye G. Radiological palliation of malignant ic obstruction. Clin Radiol 2002; 57: Park KB, Do YS, Kang WK, et al. Malignant obstruction of gastric outlet and duodenum: palliation with flexible covered metallic stents. Radiology 2001; 219: Mauro MA, Koehler RE, Baron TH. Advances in gastrointestinal intervention: the treatment of gastroduodenal and colorectal obstructions with metallic stents. Radiology 2000; 215: Bhardwaj R, Parker MC. Palliative therapy of colorectal carcinoma: stent or surgery? Colorectal Dis 2003; 5: Siddiqui A, Khandelwal N, Anthony T, Huerta S. Colonic stent versus surgery for the management of acute malignant ic obstruction: a decision analysis. Aliment Pharmacol Ther 2007; 26: Jost RS, Jost R, Schoch E, Brunner B, Decurtins M, Zollikofer CL. Colorectal stenting: an effective therapy for preoperative and palliative treatment. Cardiovasc Intervent Radiol 2007; 30: Ptok H, Meyer F, Marusch F, et al. Palliative stent implantation in the treatment of malignant colorectal obstruction. Surg Endosc 2006; 20: Ng KC, Law WL, Lee YM, Choi HK, Seto CL, Ho JW. Self-expanding metallic stent as a bridge to surgery versus emergency resection for obstructing left-sided colorectal cancer: a casematched study. J Gastrointest Surg 2006; 10: Sacks D, McClenny TE, Cardella JF, Lewis CA. Society of Interventional Radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14: S199 S202.