Deeply infiltrating endometriosis of the rectum: shave, disc or segmental excision?

Abstract

Dear Editor, The article by Ruffo et al. [1] describes a single-center series of 750 segmental resections of the mid or low rectum for deeply infiltrating endometriosis. Although the authors’ relatively low complication rate is impressive, we are concerned about the large number of low or very low anastomoses being performed in otherwise well and young (median age, 33 years) women without accompanying report of bowel function. Anterior resection syndrome is well documented, particularly after low rectal anastomoses, the management of which can be difficult [2–4]. Considering that endometriosis surgery is in many ways a quality-of-life operation to alleviate symptoms, we are interested in the authors’ comments. In our experience as well as that of others, most rectal endometriosis can be managed with shave or disc excisions without recourse to segmental resection [5, 6]. Although Ruffo et al. describe pathological confirmation of endometriosis, it will be important to determine the extent of disease on pathology. In our opinion, segmental resection should be reserved for patients with multicentric involvement or where a less invasive procedure is deemed inadequate. The authors did not make any comments on discs or shave excisions; if all patients were managed with segmental resection, then we consider this approach overly aggressive. We recently published our experience with rectal endometriosis and reported that the majority of patients (71 %) can be spared the morbidity of segmental resection by undergoing a disc excision of the anterior rectal wall because most will have focal rectal wall involvement [5]. This avoids the morbidity of full rectal mobilization, which risks rectal denervation and preserves the native rectal reservoir, which in turn reduces the likelihood of bowel dysfunction. Routine mobilization of the splenic flexure, as highlighted by Ruffo et al., necessarily implies a generous segmental resection. We are concerned about the appropriateness of this aggressive approach. We do not contest the safety and feasibility of laparoscopic management of rectal endometriosis, and recent publications testify that rectal resection can indeed be performed with low complication rates. Other than cyclical bleeding in 20 % and a preoperative median dyschezia score of 6.7, it was difficult to grasp the severity of patients’ symptoms in the series of Ruffo et al. to justify such aggressive surgical management. One could argue that even a 3 % anastomotic leak rate, 2 % rectovaginal fistula rate, 0.7 % risk of ureteric injury, and 14.5 % temporary ileostomy rate is excessive if the rectal resection was unwarranted in the first place. As we discuss in our recent article, the literature seems divergent in that there seems to be a lack of consensus as to whether bowel resection is warranted [6]. In the absence of definitive evidence to support one approach or the other, and in an era of patient-centered care, we urge all authors to move away from only reporting on short-term surgical outcomes and instead focus on functional outcomes with regard to symptom control, fertility, and bowel function outcomes. C. E. Koh (&) M. J. Solomon Surgical Outcomes Research Center, Royal Prince Alfred Hospital, Camperdown, NSW, Australia e-mail: cherry_koh@hotmail.com