Single-balloon enteroscopy with snare-assisted mechanical fragmentation for small bowel phytobezoar obstruction: a retrospective case series

Small bowel phytobezoar obstruction is uncommon but may require urgent intervention when conservative treatment fails or complete obstruction persists. Balloon-assisted enteroscopy enables direct endoscopic access to the small bowel, but evidence for single-balloon enteroscopy (SBE)-guided treatment remains limited. This retrospective case series evaluated the feasibility, clinical outcomes, and safety considerations of SBE with snare-assisted mechanical fragmentation for small bowel phytobezoar obstruction. We retrospectively reviewed 12 consecutive patients with CT-suspected and endoscopically confirmed or highly suspected small bowel phytobezoar obstruction who underwent attempted SBE-guided snare-assisted mechanical fragmentation at Weifang People’s Hospital from May 2024 to April 2025. Outcomes included technical access and fragmentation, clinical success without surgery, procedure time, adverse events, need for surgery, and short-term recurrence. The cohort included 4 men and 8 women, with a mean age of 72.83 ± 8.38 years. All patients reported recent hawthorn ingestion. The obstruction site was the jejunum in 10 patients and the duodenum in 2 patients. SBE reached and fragmented the bezoar in 11 of 12 patients. Clinical success without surgery was achieved in 10 of 12 patients (83.3%; exact 95% confidence interval [CI], 51.6–97.9%). Two patients required surgery: one because of postoperative gastric perforation and one because the distal jejunal bezoar could not be reached or visualized. No bleeding, aspiration pneumonia, or short-term recurrence was recorded. SBE with snare-assisted mechanical fragmentation may be a minimally invasive option for selected patients with accessible small bowel phytobezoar obstruction, particularly when bezoars are relatively soft and there are no signs of ischemia, perforation, or hemodynamic instability. Given the small sample size, retrospective design, potential selection bias, and relatively short follow-up, additional data from larger patient cohorts are required to further validate our findings.