Plastic versus self-expanding metallic stents for malignant hilar biliary obstruction: a prospective multicenter observational cohort study.

BACKGROUND: There are few comparative data as to whether plastic or self-expanding metallic stents are preferable for palliating malignant hilar biliary obstruction. METHODS: Thirty-day outcomes of consecutive endoscopic retrograde cholangiopancreatographies performed for malignant hilar obstruction at 6 private and 5 university centers were assessed prospectively. RESULTS: Patients receiving plastic (N=28) and metallic stents (N=34) were similar except that metallic stent recipients more often had: Bismuth III or IV tumors (16/34 vs. 5/28 P=0.043), higher Charlson comorbidity scores (P=0.003), metastatic disease (P=0.006), and management at academic centers (P=0.018). The groups had similar rates of bilateral stent placement (4/28 vs. 5/34), and similar frequency of opacified but undrained segmental ducts (7/28 vs. 5/34). Adverse outcomes including cholangitis, stent occlusion, migration, perforation, and/or the need for unplanned endoscopic retrograde cholangiopancreatography or percutaneous transhepatic cholangiography occurred in 11/28 (39.3%) patients with plastic versus 4/34 (11.8%) with metal stents (P=0.017). By logistic regression, factors associated with adverse outcomes included plastic stent placement (odds ratio 6.32; 95% confidence interval 1.23, 32.56) and serum bilirubin (1.11/mg/dL above normal: 1.01, 1.22) but not center type or Bismuth class. CONCLUSIONS: Metallic stent performance was superior to plastic for hilar tumor palliation with respect to short-term outcomes, independent of disease severity, Bismuth class, or drainage quality.

Predictors of outcomes after biliary and pancreatic sphincterotomy for sphincter of oddi dysfunction.

BACKGROUND: There are few data on combined pancreatic and biliary sphincterotomy for sphincter of Oddi dysfunction (SOD), especially regarding clinical features that might predict outcomes. We sought to examine the relative importance of various clinical features and the presence or absence of objective biliary abnormalities in determining responses to endoscopic therapy. METHODS: A cohort of consecutive patients with suspected SOD was treated with biliary sphincterotomy, with additional pancreatic sphincterotomy at initial or subsequent endoscopic retrograde cholangiopancreatography if there was abnormal pancreatic manometry in conjunction with pain refractory to biliary sphincterotomy, continuous pain, or a history of amylase elevation. Repeat intervention was offered until response was achieved or complete ablation of all treated sphincters was achieved. Response was assessed by patients using a 5-point Likert scale, and multivariate logistic regression analysis used to identify predictors of response. RESULTS: Of 121 patients, 112 (92%) were female, 105 (87%) postcholecystectomy, and by modified Milwaukee biliary classification 18 (15%) were type I, 53 (44%) type II, and 50 (41%) type III. All patients underwent biliary sphincterotomy and 49 (40%) pancreatic sphincterotomy. Good or excellent response at final follow-up was reported by 83 (69%) of 121 patients, including 37 (61%) of 61 patients requiring repeated intervention. Response was not significantly different between biliary types I, II, and III. Patient characteristics (with adjusted odds ratios) that were significant predictors of poor response were normal pancreatic manometry (4.6), delayed gastric emptying (6.0), daily opioid use (4.0), and age <40 (2.7). Abnormal liver function tests or dilated bile duct were not significant. CONCLUSIONS: For the treatment of SOD incorporating pancreatic and biliary sphincterotomy, patient characteristics and pancreatic sphincter manometry may be more important predictors of outcome than the traditional classification based on liver chemistries and bile duct dilation.

Pancreatic stent insertion: consequences of failure and results of a modified technique to maximize success.

BACKGROUND: Increasingly, pancreatic stents are being placed to prevent post-ERCP pancreatitis. However, guidewire and stent placement may fail if the duct is small or tortuous, potentially exacerbating the risk. This study assessed the impact of unsuccessful pancreatic stent placement on complications and the efficacy of a modified technique for stent insertion when pancreatic ductal anatomy makes stent insertion technically difficult. METHODS: Technical variables and 30-day complications of consecutive therapeutic ERCPs, including attempted major papilla pancreatic stent insertion were prospectively studied. Success rates for pancreatic stent placement were compared for a 1-year period during which conventional deep guidewire insertion was used and another 1-year period in which a modified technique was used as needed in patients with ductal anatomy that made stent placement technically difficult. In the modified technique, a short (2-3 cm) small diameter (3F-5F) stent was placed over a 0.018-in nitinol-tipped guidewire, passed as little as 1 to 2 cm beyond the pancreatic sphincter. RESULTS: In 225 high-risk therapeutic ERCPs, pancreatitis occurred after the procedure in two of 3 (66.7%) patients in whom pancreatic stent insertion failed vs. 32 of 222 (14.4%) patients with successful insertion (p=0.06). Severe pancreatitis occurred only after unsuccessful stent insertion. Significant multivariate risk factors for post-ERCP pancreatitis were unsuccessful pancreatic stent insertion (odds ratio 16.1: 95% CI[1.3, 200]), sphincter of Oddi dysfunction (odds ratio 3.2: 95% CI[1.4, 7.5]), and prior post-ERCP pancreatitis (odds ratio 3.2: 95% CI[1.4, 7.1]). The following were not risk factors: performance of pancreatic, biliary, or needle-knife pre-cut sphincterotomy; number of pancreatic contrast injections; and difficult cannulation. Stent placement was unsuccessful in 3 (3.2%) of 93 attempts during the 1-year period in which a conventional technique was used vs. none of 132 attempts in a subsequent year in which the modified technique was used. CONCLUSIONS: Failed attempts at pancreatic stent placement are associated with an extremely high risk of post-ERCP pancreatitis. Success can be consistently achieved by use of a modified technique.

Selective MRCP and CT-targeted drainage of malignant hilar biliary obstruction with self-expanding metallic stents.

BACKGROUND: Endoscopic management of malignant hilar biliary obstruction is controversial with respect to optimal types of stents and extent of drainage. This study evaluated outcomes of selective MRCP and CT-targeted drainage with self-expanding metallic stents. METHODS: Consecutive patients undergoing attempted palliative ERCP for malignant hilar biliary obstruction were prospectively followed. Whenever possible, management strategy included evaluation and staging for potential resectability before ERCP, with primary placement of metallic stents at the first ERCP in nonsurgical candidates, and early conversion to a metallic stent when a tumor proved to be unresectable. MRCP and/or CT were used to plan selective guidewire access, opacification, and drainage only of the largest intercommunicating segmental ducts. Unilateral stent placement was intended in all cases except for selected patients with Bismuth II cholangiocarcinoma. RESULTS: Thirty-five patients were included. Bismuth classification was I, 10; II, 6; III, 8; and IV, 11. Tumor origin was bile duct (17), gallbladder (5), and metastatic (13). Metallic stents were placed in 27 patients as the initial stent, and in 8 after plastic stent placement. Initial stents were placed endoscopically in 33 patients and percutaneously in 2 patients in whom lumenal tumor precluded ERCP. Stent placement was unilateral in 31 patients and bilateral in 4 patients. There were no episodes of cholangitis or other complications within 30 days after any procedures. Initial metallic stents were clinically effective in 27 (77%) of the 35 patients. Additional percutaneous drainage in 3 patients who did not respond to initial stent placement did not resolve jaundice. Median patency of first metallic stents was 8.9 months for patients with primary bile duct tumors and 5.4 months for all patients, and was not related to Bismuth classification. No further intervention was needed in 25 (71%) patients. CONCLUSIONS: Unilateral metallic stent placement by using MRCP and/or CT to selectively target drainage provides safe and effective palliation in most patients with malignant hilar biliary obstruction.