Malignant distal biliary obstruction (MDBO) is a common complication of peri-ampullary cancers associated with significant morbidity and mortality. Although, recent advances such as endosonography guided biliary drainage (EUS-BD) have emerged, methodological rigorous trials are needed to validate these promising developments. We aim to evaluate the quality of published randomized control trials (RCTs) on drainage of MDBO. Our findings are intended to a guide subsequent consensus on a methodological framework for clinical trials in MDBO. 

We conducted a systematic review of RCTs assessing biliary drainage in patients with MDBO (PRISMA 2020). We included RCTs published in full that compared at least two biliary drainage modalities or stents in MDBO. Studies involving hilar obstruction were excluded.  A systematic search was conducted in MEDLINE, Embase, CENTRAL, and Web of Science from database inception to Jan. 9, 2025. Study quality was assessed using the revised risk of bias (ROB 2.0).

Overall, 4,663 citations were identified with 71 trials ultimately selected. These RCTs compared ERCP stent types in 67.6% with the remaining comparing different modalities of drainage including EUS-BD, percutaneous drainage, and surgery. Most trials assessed patients with unresectable or locally advanced cancer (87.3%). A hypothesis statement was explicitly stated in only 22.5% of the studies. A sample size estimate was lacking in 26.8% of the RCTs. In terms of randomization, timing was missing in 43.7%, with 56.3% of the studies having no consort randomization diagram. Moreover, information on blinding was not described in 47.9% of the trials. In study results, key baseline characteristics such as performance status and comorbidity index were reported in only 29.6% and 11.3% of the studies, respectively. The definition of biliary obstruction was heterogeneous with 12 different definitions identified. In terms of outcomes, stent dysfunction was the most common; however, a precise definition was missing in 19.7% of the studies while 42 different definitions of stent dysfunction were used. Technical and clinical success were defined in only 40.8% and 47.9% of the studies, respectively. Although adverse events were ubiquitously reported, only 36.6% of the trials used a standardized lexicon for reporting. Hospital length of stay and quality of life data were reported in only 21.1% and 11.3% of the trials, respectively. According to the ROB2, 85% of the trials were at high-risk of bias in the measurement of outcomes. 

Our review highlights the methodological deficiencies and heterogeneity of current published trials in MDBO. A methodological framework and standardized definitions of outcomes are urgently needed to improve evidence-based medicine for the management of MDBO