Laparoscopic versus EUS-guided gastroenterostomy for gastric outlet obstruction: an international multicenter propensity score–matched comparison (with video)

Background and Aims In the management of gastric outlet obstruction (GOO), EUS-guided gastroenterostomy (EUS-GE) seems to be safe and more effective than enteral stent placement. However, comparisons with laparoscopic GE (L-GE) are scarce. Our aim was to perform a propensity score–matched comparison between EUS-GE and L-GE. Methods An international, multicenter, retrospective analysis was performed of consecutive EUS-GE and L-GE procedures in 3 academic centers (January 2015 to May 2020) using propensity score matching to minimize selection bias. A standard maximum propensity score difference of .1 was applied, also considering underlying disease and oncologic staging. Results Overall, 77 patients were treated with EUS-GE and 48 patients with L-GE. By means of propensity score matching, 37 patients were allocated to both groups, resulting in 74 (1:1) matched patients. Technical success was achieved in 35 of 37 EUS-GE–treated patients (94.6%) versus 100% in the L-GE group (P = .493). Clinical success, defined as eating without vomiting or GOO Scoring System ≥2, was achieved in 97.1% and 89.2%, respectively (P = .358). Median time to oral intake (1 [interquartile range {IQR}, .3-1.0] vs 3 [IQR, 1.0-5.0] days, P < .001) and median hospital stay (4 [IQR, 2-8] vs 8 [IQR, 5.5-20] days, P < .001) were significantly shorter in the EUS-GE group. Overall (2.7% vs 27.0%, P = .007) and severe (.0% vs 16.2%, P = .025) adverse events were identified more frequently in the L-GE group. Conclusions For patients with GOO, EUS-GE and L-GE showed almost identical technical and clinical success. However, reduced time to oral intake, shorter median hospital stay, and lower rate of adverse events suggest that the EUS-guided approach might be preferable. In the management of gastric outlet obstruction (GOO), EUS-guided gastroenterostomy (EUS-GE) seems to be safe and more effective than enteral stent placement. However, comparisons with laparoscopic GE (L-GE) are scarce. Our aim was to perform a propensity score–matched comparison between EUS-GE and L-GE. An international, multicenter, retrospective analysis was performed of consecutive EUS-GE and L-GE procedures in 3 academic centers (January 2015 to May 2020) using propensity score matching to minimize selection bias. A standard maximum propensity score difference of .1 was applied, also considering underlying disease and oncologic staging. Overall, 77 patients were treated with EUS-GE and 48 patients with L-GE. By means of propensity score matching, 37 patients were allocated to both groups, resulting in 74 (1:1) matched patients. Technical success was achieved in 35 of 37 EUS-GE–treated patients (94.6%) versus 100% in the L-GE group (P = .493). Clinical success, defined as eating without vomiting or GOO Scoring System ≥2, was achieved in 97.1% and 89.2%, respectively (P = .358). Median time to oral intake (1 [interquartile range {IQR}, .3-1.0] vs 3 [IQR, 1.0-5.0] days, P < .001) and median hospital stay (4 [IQR, 2-8] vs 8 [IQR, 5.5-20] days, P < .001) were significantly shorter in the EUS-GE group. Overall (2.7% vs 27.0%, P = .007) and severe (.0% vs 16.2%, P = .025) adverse events were identified more frequently in the L-GE group. For patients with GOO, EUS-GE and L-GE showed almost identical technical and clinical success. However, reduced time to oral intake, shorter median hospital stay, and lower rate of adverse events suggest that the EUS-guided approach might be preferable.