Although surgical resection is considered the most effective therapy for locally recurrent rectal cancer, surgical stress and sacral nerve damage accompanied by high sacral resection often yield dismal results . The use of radiotherapy as an alternative modality has increased in recent years, especially for locally recurrent rectal cancer due to its high efficacy. In radiotherapy for locally recurrent rectal cancer, the small bowel is the major dose-limiting structure. Spacer placement to physically separate the small bowel from the recurrent tumor can reduce the bowel radiation dose. Spacer should be sutured securely to cover the entire area of tumor. Usually, artificial material, greater omentum or muscle is used as a spacer. In this case, a polytetrafluoroethylene sheet was placed between the tumor and small bowel. Complications related to the use of the spacer artificial material include organ injury, bleeding, foreign body reaction, infection, and fistula formation. Complete migration of the spacer into the lumen of the small intestine is extremely rare. To our knowledge, this report is the first case of complete spacer migration inside the small bowel.
Intraluminal migration of artificial materials has been described previously, such as the migration of mesh after hernioplasty [3–8]. Agrawal and Avill  reported finding intravesicular mesh following laparoscopic transabdominal preperitoneal inguinal hernia repair. Celik et al.  reported intraluminal mesh migration into the splenic flexure following transabdominal preperitoneal repair of inguinal hernia, which was subsequently removed by colonoscopy. Benedetti et al.  reported a patient who presented with rectal bleeding 2 years after inguinal hernia repair and was found to have sigmoid colon perforation with the plug partially within the colonic lumen. Lange and coworkers  reported a patient with sigmoid penetration following transabdominal preperitoneal repair of inguinal hernia. Goswamiet and colleagues  reported mesh migration into the cecum after transabdominal preperitoneal hernioplasty. Borchert et al.  reported a female patient who presented with small bowel obstruction 7 years after obturator hernia repair. The above reported cases involved small bowel obstruction after bowel adhesion to the mesh rather than transmural migration.
Two published case reports have described complete migration of artificial materials into the small bowel [9, 10]. Majeski  reported intraluminal mesh migration in a woman who had undergone incisional hernia repair with wire mesh 40 years prior to presentation with small bowel obstruction. A piece of wire mesh was found to have broken off and was within the lumen of the small bowel. Steinhagen et al.  reported complete transmural mesh migration from the abdominal wall into the small bowel one year after repair of incisional hernia with polytetrafluoroethylene mesh.
While migration of artificial material is quite uncommon, it is possible for mesh to translocate into the hollow abdominal viscus, although the exact mechanism involved in this process is not clear at present. Robinson and Levin  suggested that the retained foreign body, e.g., surgical sponges, becomes encapsulated by loops of small bowel and by peritoneal reaction. Riaz et al.  hypothesized the edges of the mesh might cause damage to the surface of nearby organs and evoke an inflammatory response leading to erosion and migration. Celik et al.  argued that it occurs when part of the bowel is carelessly included in the fixation of the mesh. Growswami et al.  proposed that adhesions at hernioplasty can predispose the patient to further adhesion formation leading to migration.
The above proposed mechanisms do not provide a reasonable explanation for the present case. Our patient was in good nutritional status and was not on long-term steroid use. How did the spacer pass from the outside of the small bowel into the lumen? A huge defect in the intestinal wall is required for the spacer to migrate into the lumen. However, this patient never developed sepsis-like symptoms that would be expected during the course. Based on the presence of dense adhesions in the pelvic cavity, we propose that the contact between the spacer and intestinal wall elicited local inflammatory reaction without systemic symptoms, during which the spacer eroded the bowel and ultimately penetrated the entire wall. However, the small intestine site containing the spacer was not near the pelvic cavity. This suggests that the spacer traveled within the lumen from the penetration site retrogradely to the site of small bowel obstruction. This case presents extremely uncommon complication of surgery using spacer. Such complication could be perhaps avoided by leaving non-viscus tissue between the spacer and intestinal wall, leaving no direct contact between the two.