| 063 |
|
CLINICAL RISK FACTORS OF PLASTIC BILIARY STENT OBSTRUCTION: A PROSPECTIVE TRIAL
NS Abraham, AN Barkun, L Joseph1, ER Valois, JS Barkun2
The Divisions of Gastroenterology, 2General Surgery and 1Epidemiology & Biostatistics, McGill University Health Centre, Montreal, Quebec
AIM: To determine early clinical risk factors associated with plastic stent obstruction in patients with malignant biliary obstruction.
METHODS: In the exploratory analysis of data generated from a prospective trial examining stent failure, possible risk factors for plastic stent blockage were examined. Assessment included: demographic data, presenting clinical symptoms/signs of malignant biliary obstruction, biochemical profile, ASA score, stent size/length, performance of a sphincterotomy (ES) and site of malignancy. All patients were followed 7 days after ERCP stent insertion, then monthly until final outcome (stent blockage, death or drop out) was reached. Data from the initial stent insertion and one week thereafter were analyzed to determine the presence of early clinical risk factors of stent blockage. Univariate and multivariate survival analysis was performed using Cox Proportional Hazards Estimation.
RESULTS: There were 64 patients (29 male, 35 female), 98% were Caucasian, and mean age was 71.7 yrs (SD:10.7). Upon initial presentation the most common clinical symptoms and signs included: jaundice (98%), fatigue (86%), weight loss (81%) and pruritus (69%). Three-quarters of the patients had an American Society of Anaesthesia score of 3 or greater and 64% had a distal bile duct obstruction. Baseline biochemistry revealed: AST 130 mmol/l (SD: 89), ALT 173 mmol/l (SD: 173), ALP 699 mmol/l (SD: 466) and T. Bili 275 mmol/l (SD: 141). ERCP and stent insertions were successfully performed on all patients. Sphincterotomies (ES) were performed in 25%. In total, 50% of stents blocked. Univariate analysis revealed that failure of improvement in ALT (RR=1.7; 95% CI: 1.2-2.4) and ES (RR=0.47; 95% CI:0.23-0.98) were associated with stent blockage. There was no significant association with stent size/length, baseline biochemistry profile, site of biliary obstruction or clinical symptoms. On multiple Cox regression analysis stent blockage was best predicted by the use of a sphincterotomy at ERCP and failure of improvement in ALT.
CONCLUSION: The use of sphincterotomy to insert a plastic stent for malignant biliary obstruction is an early clinical risk factor associated with subsequent blockage.