Background: Although self-expandable metal stents (SEMS) are widely used for distal malignant biliary obstruction, one-step SEMS (direct placement without a prior plastic stent) and two-step SEMS (placement at second endoscopic retrograde cholangiopancreatography [ERCP] following plastic stent placement) have not been fully compared.
Methods: In this multicenter retrospective study, patients were included who underwent first-time endoscopic SEMS placement between September 1994 and December 2010. We compared the one-step and two-step strategies using a propensity analysis.
Results: In total, 370 patients were identified and one-step SEMS was performed in 59 patients. After adjustment using propensity scores, the median times to dysfunction were 116 and 219 days, respectively, for one-step and two-step SEMS (P = 0.058). Stent migration was more frequently observed in one-step SEMS as compared with two-step SEMS (25 vs. 11 %, P = 0.031). In one-step SEMS, the number of days of hospitalization associated with first-time SEMS placement was shorter compared with that in two-step SEMS (21 vs. 30 days, P = 0.001), and the total costs of SEMS-related interventions within 6 months were lower (6510 and 8100 USD, P = 0.004). The pathological diagnosis rates for pancreatic and biliary tract cancer at initial ERCP were 52 and 61 %. After failed diagnosis at initial ERCP, pathological diagnosis rates for pancreatic cancer were 32 versus 76 % (P = 0.005) by repeated ERCP versus endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA).
Conclusions: One-step SEMS was associated with increased stent migration, despite having potential cost-effectiveness. The additional yield of pathological diagnosis at repeated ERCP was low compared with that yielded by EUS-guided FNA.