Background: The surgical approaches and resection extent for rectal gastrointestinal stromal tumors (GISTs) are controversial due to the low incidence of this disease. A multicenter retrospective cohort study was conducted to compare the postoperative and oncologic outcomes of local excision (LE) and radical resection (RR) in patients with low rectal GIST. Patients and methods: The medical records of rectal GIST patients from 11 large-scale medical centers in China (January 2000-December 2019) were reviewed. All patients were divided into either the LE group or the RR group. Propensity score matching (PSM) was conducted to reduce confounders. Results: A total of 280 patients with low rectal GIST were enrolled. After PSM, 144 patients were included (72 in each group). The LE group showed a higher anal preservation rate (100.0% vs. 76.4%, P < 0.001), shorter operation time (77.1 +/- 68.4 min vs. 159.1 +/- 83.6 min, P < 0.001), fewer complications (8.3% vs. 22.2%, P = 0.021) and shorter postoperative hospital stay (4.9 +/- 4.1 d vs. 10.7 +/- 8.1 d, P < 0.001) than the RR group. There was no significant difference in recurrence-free survival (RFS) between the RR and LE groups among patients with tumors <= 2 cm(P = 0.220), and the RR group had a superior RFS than the LE group in patients with tumors >2 cm (P = 0.046). Conclusions: LE resulted in improved postoperative outcomes and comparable oncological safety with a low rectal GIST of <= 2 cm. However, for patients with a low rectal GIST of >2 cm, RR might be a more appropriate option with better RFS. (C) 2021 Elsevier Ltd, BASO similar to The Association for Cancer Surgery, and the European Society of Surgical Oncology. All rights reserved.