BACKGROUND AND STUDY AIM: Despite high technical and functional success rates with endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS), rates of adverse events have also been high. No reports have focused on EUS-HGS alone with a large sample size about predictors of stent patency. The present study examined predictors of stent patency in patients who underwent EUS-HGS. PATIENTS AND METHODS: The consecutive 51 patients who underwent EUS-HGS using one metallic stent were retrospectively enrolled in this study. Baseline characteristics, stent length from the hepatic portion and in the luminal portion, kinds of stent dysfunction, and stent patency were reviewed. RESULTS: Median duration of stent patency was significantly shorter with stent length in the luminal portion <3 cm (52 days) than with ≥3 cm (195 days; P < 0.01). On the other hand, median duration of stent patency did not differ significantly between ≥4 cm (194 days) and <4 cm (127 days; P = 0.1726). Length of stent in the luminal portion ≥3 cm (Hazard ration [HR], 9.242; 95% confidence interval [CI], 3.255-26.244, P < 0.05) and performance of chemotherapy (HR, 3.022; 95% CI, 1.448-6.304, P < 0.05) were also associated with long stent patency on the Cox proportional hazards model. CONCLUSION: In conclusion, to obtain long-term stent patency, our data suggest that a stent length ≥3 cm in the luminal portion may be suitable for EUS-HGS.
BACKGROUND AND STUDY AIM: Despite high technical and functional success rates with endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS), rates of adverse events have also been high. No reports have focused on EUS-HGS alone with a large sample size about predictors of stent patency. The present study examined predictors of stent patency in patients who underwent EUS-HGS. PATIENTS AND METHODS: The consecutive 51 patients who underwent EUS-HGS using one metallic stent were retrospectively enrolled in this study. Baseline characteristics, stent length from the hepatic portion and in the luminal portion, kinds of stent dysfunction, and stent patency were reviewed. RESULTS: Median duration of stent patency was significantly shorter with stent length in the luminal portion <3 cm (52 days) than with ≥3 cm (195 days; P < 0.01). On the other hand, median duration of stent patency did not differ significantly between ≥4 cm (194 days) and <4 cm (127 days; P = 0.1726). Length of stent in the luminal portion ≥3 cm (Hazard ration [HR], 9.242; 95% confidence interval [CI], 3.255-26.244, P < 0.05) and performance of chemotherapy (HR, 3.022; 95% CI, 1.448-6.304, P < 0.05) were also associated with long stent patency on the Cox proportional hazards model. CONCLUSION: In conclusion, to obtain long-term stent patency, our data suggest that a stent length ≥3 cm in the luminal portion may be suitable for EUS-HGS.