Background: The goal of this study was to assess small bowel motion and explore the feasibility of using peritoneal space (PS) to replace bowel loops (BL) via the dose constraint method to spare the small bowel during intensity-modulated radiotherapy (IMRT) for rectal cancer. Methods: A total of 24 patients with rectal cancer who underwent adjuvant or neoadjuvant radiotherapy were selected. Weekly repeat CT scans from pre-treatment to the fourth week of treatment were acquired and defined as Plan, 1W, 2W, 3W, and 4W. The 4 weekly CT scans were co-registered to the Plan CT, BL and PS contours were delineated in all of the scans, an IMRT plan was designed on Plan CT using PS constraint method, and then copied to the 4 weekly CT scans. The dose-volume, normal tissue complication probability (NTCP) of the small bowel and their variations during treatment were evaluated. Results: Overall, 109 sets of CT scans from 24 patients were acquired, and 109 plans were designed and copied. The BL and PS volumes were 250.3 cc and 1339.3 cc. The V15 of BL and PS based plan of pre-treatment were 182.6 cc and 919.0 cc, the shift% of them were 28.9% and 11.3% during treatment (p=0.000), which was less in the prone position than in the supine position (25.2% vs 32.1%, p =0.000; 9.9% vs 14.9%, p=0.000). The NTCPC and NTCPA based plan of pre-treatment were 2.0% and 59.2%, the shift% during treatment were 46.1% and 14.0% respectively. Majority of BL’s Dmax and V15 were meet the safety standard during treatment using PS dose limit method except 3 times (3/109) of V15 and 5 times of Dmax (5/109).Conclusions: This study indicated that small bowel motion may lead to uncertainties in its dose volume and NTCP evaluation during IMRT for rectal cancer. The BL movements were significantly greater than PS, and the prone position was significantly less than the supine position. It is feasibility of using PS to replace BL to spare the small bowel, V15<830 cc is the dose constraint standard.