Background: Endoscopic diagnosis and treatment of gastrointestinal submucosal lesions is still not established. We evaluated the clinical usefulness of two resection methods for submucosal lesions, using a "nonrandomized surgeon" design.
Methods: The strip biopsy method was evaluated at Tenri Hospital and the aspiration lumpectomy method was used at Kyoto University Hospital. The inclusion criteria for selecting patients were endosonographic findings indicating a tumor location within the submucosa.
Results: Seventy-seven patients were treated. The size of the specimens (mean +/- SEM) was 20.7 +/- 0.9 mm for the aspiration lumpectomy group and 14.0 +/- 0.8 mm for the strip biopsy group (p < 0.01). Aspiration lumpectomy was adequate for a definitive histologic diagnosis in 95% of the cases (36 of 38) as compared with 77% (30 to 39) of the cases treated by strip biopsy (p = 0.047). The eradication rate was 87% for the aspiration lumpectomy group, and 74% for the strip biopsy group (p = 0.274). The complication rate resulting from these procedures (hemorrhage, mean 5%) was acceptable. Repeated endoscopic examinations revealed no recurrence except for one lymphangioma.
Conclusions: Endoscopic resection, especially aspiration lumpectomy, provides a significant benefit for accurate final diagnosis and eradication of submucosal lesions.