Characteristics and Surgical Outcomes of Patients with Intestinal Endometriosis Undergoing Bowel Resection

Abstract

Objective:

To retrospectively evaluate the clinical features, perioperative findings, and long-term outcomes of patients who underwent bowel resection for intestinal endometriosis, and to clarify diagnostic and therapeutic challenges with the goal of supporting optimal surgical strategies that minimize recurrence while preserving function.

Methods:

Twenty-one patients who underwent bowel resection for intestinal endometriosis from 2002 to 2022 were evaluated regarding clinicopathological findings and short and long-term outcomes.

Results:

The median age was 38 years. All patients had gastrointestinal symptoms. Colonoscopy confirmed endometriosis histologically in only 1 patient (7.1%). Lesions were identified in the sigmoid colon to the upper rectum in 17 patients, some of which were accompanied by obliteration of the pouch of Douglas. Low anterior resection was performed in 15 of these patients, and temporary stomas were created in 4 patients. Gynecologic procedures were performed in 9 patients. Histopathology showed invasion into the submucosal and muscular layers. Early complications (Clavien-Dindo ≥2) occurred in 2 patients (anastomotic leak and ileus), while late complications included anastomotic stenosis, bowel stricture, and rectovaginal fistula (1 patient each). Only one recurrence (4.8%) was observed during a median follow-up of 2,088 days. Spontaneous pregnancy was achieved in 2 patients.

Conclusions:

This study highlights the clinical significance of bowel resection for intestinal endometriosis, suggesting the importance of comprehensive evaluation to improve diagnostic accuracy, the utility of collaboration with gynecology, and the potential for recurrence reduction and functional preservation through appropriate resection margins. These findings may contribute to the development of future treatment strategies for this condition.

Introduction

Endometriotic disease has an annual incidence of 0.1% among women with age between 15 and 49 years, and a prevalence of about 5%, with a peak between 25 and 35 years of age[1]. In rare cases, endometrial tissue implants can occur outside the genital tract, with the intestinal tract being one of the most common extragenital sites. Among these, the rectosigmoid region is most frequently affected, and patients may present with nonspecific gastrointestinal symptoms such as bloating, constipation, diarrhea, and hematochezia.

Intestinal endometriosis often presents with non-specific clinical symptoms, and because endometrial tissue rarely reaches the mucosal surface, there are challenges in making a definitive preoperative diagnosis and accurately assessing the extent of the lesion through imaging or endoscopic biopsy. Moreover, the extent of disease and the depth of invasion are often underestimated, complicating surgical planning.

Surgical resection is often required for symptomatic patient refractory to medical therapy. However, in women of reproductive age, treatment strategies must balance symptom control, recurrence prevention, and preservation of fertility. There is currently no standardized approach regarding the extent of resection or perioperative management, and literature on surgical outcomes - particularly long-term bowel function and reproductive impact - remains limited.

The aim of this study is to retrospectively evaluate the comprehensive preoperative assessment, perioperative outcomes, and long-term postoperative outcomes of patients who underwent bowel resection for intestinal endometriosis at our institution. By identifying the challenges associated with diagnosis and treatment, this study seeks to provide insights that may support the development of optimal surgical strategies to minimize the risk of recurrence while preserving postoperative function.

Methods

The subjects were patients who underwent surgery (bowel resection) for intestinal endometriosis at Juntendo university hospital from January 2002 to December 2022. Preoperative evaluation included clinical symptom assessment, gynecologic history, and imaging studies. Colonoscopy and MRI were performed when clinically indicated, and jelly-enhanced MRI was selectively used in patients with suspected deep pelvic lesions to better visualize the rectovaginal septum and pouch of Douglas. We retrospectively searched and collected the medical records of patients with histopathologically diagnosed intestinal endometriosis from surgical specimens. Twenty-one patients for intestinal endometriosis were performed at our hospital, 19 were laparoscopic bowel resections, and two were laparotomies.

The clinical characteristics of the patients were examined, including age at the time of surgery, abdominal surgery, and medical history of endometriosis. Clinical and imaging findings related to the diagnosis were obtained from the medical records. Preoperative treatment, preparation, location of lesions, operative time, blood loss, procedure of intestinal surgery, concomitant surgery, histopathological examination results, and postoperative course were analyzed. In addition, data on postoperative complications were collected and surgical outcomes were evaluated. Early postoperative complications were defined as those that occurred during hospitalization, and late postoperative complications were defined as those that occurred after discharge. Postoperative complications were defined as Grade-2 or higher by Clavien-Dindo classification (CD classification)[2]. This study was approved by the Juntendo University Ethics Committee (IRB approval number: E24-0439). This study has all patients and relevant persons (such as the parent or legal guardian) to publish the information, including photographs.

We performed an enema test, colonoscopy, and pelvic Magnetic Resonance Imaging (MRI) as preoperative examinations. The enema test findings showed various findings from only wall irregularity to the circumferential stenosis (Figure 1a, b). Similarly, colonoscopy showed various findings. In most patients, reddening, wall thickening, edema and even stenosis suspicious of intestinal endometriosis were observed (Figure 2a, b, c, d). Endoscopic findings were collected, and biopsy specimens were taken from lesions with abnormal changes using standard endoscopic forceps. In our hospital, the accuracy of image diagnosis is further improved by performing MRI by the jelly method in which jelly is injected transvaginal and transanal for imaging (Figure 3). This method clearly outlines the intestinal and vaginal tracts on MRI.

Figure 1.

Enema findings.

Enema examination revealed atypical radiographic findings, including circumferential luminal stenosis (white arrow) and a serrated appearance of the mucosal surface (black arrow).

Figure 2.

Colonoscopy findings.

Colonoscopy demonstrated heterogeneous and atypical findings, such as reddish mucosa (a), irregular and rough mucosal changes (b), a submucosal tumor–like elevation (c), and circumferential luminal stenosis (d).

Figure 3.

Jelly-enhanced MRI findings.

MRI with transvaginal or transanal jelly instillation revealed a circumferential stricture of the rectum (white arrow), enhancing the contrast between the lesion and surrounding structures.

Preoperative preparation & surgical procedure

Preoperative preparation

In patients with intestinal obstruction, intestinal decompression procedures such as ileus tubes, intestinal stents, and stomas are performed depending on symptoms and location of the lesion.

In patients scheduled for wait-and-see surgery, preoperative hormone therapy (about 6 months) should be performed.

Preoperative colon cleaning is not performed in our hospital if a stoma has been placed in advance. If there is no obstruction, mechanical preparations were performed on the day before surgery, and oral intake was continued until the night before. If there is an obstruction, the patient is fasted and given laxatives 2 days prior to surgery.

Surgical procedure

Surgery was performed under general anesthesia in the lithotripsy position.

Ureteral stents were placed when ureteral stenosis was suspected on preoperative imaging or when the risk of ureteral injury was considered high due to extensive endometriosis lesions in the pelvis.

The following is a description of laparoscopic procedures for left-sided colorectal lesions, which were performed in many patients.

For laparoscopic surgery, pneumoperitoneum pressure is controlled at 10 mmHg, and 5-port surgery is performed. For pelvic lesions, the gynecologists dissect severe adhesions in the Douglas' pouch caused by deep endmetriosis after inserting a uterine manipulator (Figure 4a). In particular, posterior colpotomy and dissection of rectovaginal adhesion will be performed (Figure 4b). If necessary, ovarian cystectomy will be performed for endometrioma at the same time. Careful manipulation is required during adhesion detachment because the layers are difficult to recognize, and bleeding is likely to occur. Next, rectal passive surgery is performed by colorectal surgeons, but it is necessary to perform sufficient detachment passive because it is difficult to diagnose the area of infiltration of endometriotic tissue. If the extension of the intestinal tract and mesentery was poor due to endometriotic tissue, resulting in tension at the anastomotic site, mobilization of the splenic flexure was performed. The extent of the serosal lesions caused by endometriosis is difficult to diagnose preoperatively; thus, the extent of the lesion should be carefully diagnosed according to the intraoperative findings. The extent of serosal involvement by endometriosis was determined intraoperatively based on macroscopic findings such as fibrotic thickening, surface nodularity, puckering, and visible invasion through the serosa. Dense adhesions and reduced bowel mobility were also considered indicators of deeper infiltration. The distal side of the rectum freed from endometriosis was resected using a laparoscopic linear stapler (Figure 4c). After a wound-protecting device was attached to a 4 cm median skin incision made just below the umbilicus, the resected tissue was removed. Anastomosis was performed intracorporally using the double stapling technique with a 25-31 mm circular device (Figure 4d). If the anastomotic site is in a low position or if there is a high risk of anastomotic leakage, construct a covering stoma.

Figure 4.

Intraoperative findings.

(a) Dense adhesions between the vagina and rectum formed a frozen pelvis (white arrow).

(b) A gynecologist performed adhesiolysis of the posterior vaginal fornix (white arrow).

(c) After rectal mobilization, laparoscopic dissection was continued from the peritoneal reflection toward the anal side, avoiding areas affected by endometriotic lesions.

(d) Following transection at a site with normal rectal tissue, anastomosis was completed using a circular stapling device.

Results

The age ranged from 28 to 48 years old with a median of 38 years old. Pregnancy history was found in 1 patient. Although the chief complaint was a total number of patients, bowel obstruction was the most common in 8 patients. Next were defecation pain and melena in each 6 patients, with melena observed coinciding with menstruation in 4 of those patients. In addition, difficulty in defecation was observed in 3 patients, abdominal pain in 3 patients, constipation and low back pain in each 2 patients, and symptoms caused by intestinal stricture (Table 1). The median preoperative CA125 was 74 U/ml (Range 7-201). These were higher than the standard range in 17 patients (standard 0-35 U/ml), and postoperative decrease was observed in all patients (Table 1 and 4).

Table 1.

Patient and Endometriosis Related Characteristics.

Variable	N=21
Age (year), median (range)	38 (28-48)
BMI (kg/m2), median (range)	20.2 (14.4-29.4)
Location*
Ileum		5
Sigmoid colon (S/C)		2
Rectosigmoid (RS)		10
Upper Rectum (Ra)		5
Symptoms**
Bowel obstruction		8
Painful defecation		6
Melena		6
Difficulty in defecation		3
Abdminal pain		3
Constipation		2
Low back pain		2
Narrowing of the stool column		2
Painful urination		1
Preoperative pregnancy	1 (4.8%)
Preoperative CA125 (U/ml), median (range) ***	74 (7-201)
Patients receiving preoperative pharmacotherapy	15
Gonadotropin releasing hormone (GnRH)		12
Dienogest		3
Colonoscopy**	17/21 (81.0%)
Redness		9
Stenosis		6
Wall thickening		2
Biopsy	14/17 (82.4%)
Inflammation		13
Endometriosis		1
Enema	16/21 (76.2%)
Stenosis		12
Failure of wall extension		2
Unclear		2
Magnetic resonance imaging**	19/21 (90%)
Douglas fossa adhesion		9
Chocolate cyst		5
Mass		1
No findings		4
Prior history of pelvic surgery	1 (4.8%)
Chocolate cystectomy		1
Preoperative procedures	7
Ileus tubes		3
Stoma		2
Colonic stent		1
Nephrostomy		1

*one case had two lesions (Ileum and RS) **duplicate *** standard: 0-35 U/ml

Pharmacotherapy performed before surgery was Gonadotropin releasing hormone analogue (GnRHa) in 12 patients, Dienogest in 3 patients, and 2 patients did not receive any medication because they wished to have a baby (Table 1).

The enema test was performed in 16 patients (76.2%). The findings were as follows: stenosis in 12 patients, failure of wall extension in 2 patients, and unclear in 2 patients. Similarly, colonoscopy was performed in 17 patients (81.0%). The main findings of colonoscopy were erythema in 9 patients, stenosis in 6 patients, and wall thickening in 2 patients (Figure 4a, b, c, d). However, intestinal endometriosis was detected on preoperative biopsy in only 1 patient (7.1%) (Table 1). The main site of intestinal endometriosis was sigmoid colon in 2 patients, rectosigmoid in 10 patients, upper rectum (Ra: the lesion is located from the height of the inferior border of the second sacral vertebra to the peritoneal inversion) 5 patients, ileum in 5 patients and both the rectum and ileum in 1 patient. Among these, 100% of the lesions from the sigmoid colon to the rectum were associated with obliteration of the pouch of Douglas due to deep endometriosis (Table 1). Seven patients underwent preoperative procedures related to intestinal endometriosis: ileus tube placement in 3 patients, preoperative stoma creation in 2 patients, and colonic stent in 1 patient, and nephrostomy in 1 patient.

As operations, in patients of left-sided lesions, low anterior resection (LAR) was performed in 15 patients (88.2%), and ileostomy was added in 4 patients with a high risk of anastomotic leakage. 3 patients were ileocecal resection. In one patient, an ileal lesion was found intraoperatively and LAR and ileocecal were performed. One patient had a thickened bowel wall that could not be cut laparoscopically, so the patient was transferred to laparotomy. In this study, concurrent procedures performed by gynecologists included ovarian cystectomy in 5 patients, myomectomy in 5 patients, unilateral oophorectomy in 2 patients, and salpingectomy in 2 patients. The operation time was 154 to 592 minutes, a median of 349 minutes. The blood loss was 5 ml to 600 ml with a median of 186 ml. The median distance from the anal verge to the anastomosis was 7.6 cm for the left-sided colorectal lesion (Table 2).

Table 2.

Surgical Issues and Short-Term Outcomes.

Variable	N=21
Approach
Laparoscopic	19
Open	2
Operation time (min), median (range)	345 (155-592)
Blood volume (ml), median (range)	120 (5-1,115)
Surgical procedure*	21
Low anterior resection		15
high anterior resection		2
Ileocecal		3
Partial resection of the small intestine		1
Gynecologic surgery*	14
Unilateral oophorectomy		2
Salpingectomy		2
Cystectomy		5
Myomectomy		5
Additional procedure	6
Ureteral stent		2
Diverting stoma		4
Conversion to open surgery	1/19 (5.3%)
		1
Early postoperative complication (CD≧2)	2
Anastomotic leakage		1
Ileus		1
Postoperative hospitalization period (day), median (range)	15 (6-34)

*duplicate

The median length of the resected intestine was 13 cm(Range 7-28). Histopathological examination revealed endometriosis tissue in the submucosa in 15 patients (68.2%) and in the intrinsic muscle layer in 7 patients (31.8%) (Table 3).

Table 3.

Pathological Findings.

Variable	N=21
Length of resected intestine (cm), median (range)	13 (7-28)
Depth of endometrial tissue invasion from the serosal side*
MP	7 (31.8%)
SM	15 (68.2%)

*duplicate

The postoperative course was favorable in 18 patients, regarding early postoperative complications (CD≧2), two patients were observed: anastomotic leakage and ileus. Anastomotic leakage was treated by conservative treatment while ileus was treated by conservative treatment of ileus tube placement. On the other hand, for late postoperative complications (CD≧2), One patient had anastomotic stenosis requiring endoscopic bougie. One patient each of intestinal stenosis and vaginal fistula was observed. The intestinal stenosis improved with conservative treatment. In a patient with a vaginal fistula, a LAR and ileal stoma were constructed at the first surgery. Since there were no postoperative anastomotic problems, the stoma was closed 5 months after surgery. After the closure of a diverting stoma, a vaginal fistula appeared, necessitating the re-construction stoma at transverse colon (Table 2 and 4). Postoperative drug therapy was given in 13 patients (Dienogest: 10, Oral contraceptive: 3). Two patients gave birth postoperatively, one of which was delivered by Cesarean section (Table 4). In our study, the median duration of postoperative follow-up was 2,088 days (range: 302-4,078 days).

Table 4.

Long-Term Postoperative and Reproductive Outcomes.

Variable		N=21
Postoperative pharmacotherapy	13
Dienogest		10
Oral contraceptive		3
Late postoperative complication (CD≧2)	3
Anastomotic stenosis		1
Intestinal stenosis		1
Vaginal fistula		1
Distance between DST anastomosis	17/17 (100%)
and anal verge (cm), median (range)*		7.64 (3-12)
Postoperative CA125, median (Range)		24 (5-48)
Postoperative pregnancy		2 (9.5%)
The duration of postoperative follow-up (day), median (range)	2,088 (302-4,078)

*cases of S, RS, and Ra lesions were mesured

Discussion

Extragenital endometriosis, also referred to as ectopic endometriosis, is a relatively rare condition, accounting for approximately 5-15% of all endometriosis patients. These ectopic lesions are most frequently located in the pelvic cavity, particularly involving the rectal wall and the rectosigmoid junction, comprising up to 93% of intestinal endometriosis patients; however, they may also involve other organs such as the bladder and ureters[3-5]. The average age of onset ranges from 33 to 43 years, indicating a higher prevalence among middle-aged women of reproductive age[6-8]. In addition to dysmenorrhea, the most common symptoms are related to bowel obstruction, such as bloating, constipation, and diarrhea, while hematochezia is reported in 16.7% of patients[7,8]. In the present study, specific symptoms of intestinal endometriosis were not observed, and most patients presented with nonspecific clinical features.

The diagnostic yield of endoscopic biopsy is reported to be as low as 13.3%[8], reflecting the rarity with which endometriotic tissue infiltrates the mucosa-less than 4% of patients reach the mucosal surface[9,10]. In our study, colonoscopy was performed in 17 patients, and biopsy was conducted in 14 patients, but endometriotic tissue was confirmed in only one patient (7.1%). Histopathological studies show that lesions infiltrate from the serosa in 63-72.7% of patients, involve the muscularis propria in 24.2-33.3%, and reach the mucosa or submucosa in only 3-3.7%[10]. Consistent with this, submucosal invasion was the most common finding in our study (15 patients), and the rarity of lesions extending to the mucosa likely contributes to the difficulty of preoperative diagnosis. As previously reported, endometriotic lesions tend to invade the bowel wall from the serosal side, and our pathological findings also demonstrated extension from the serosa to the submucosa. In the remaining 20 patients, a comprehensive clinical assessment based on symptoms, gynecologic history, endoscopic, and radiologic findings led to a strong clinical suspicion of intestinal endometriosis.

Given the limitations of preoperative diagnosis, surgical indication at our institution is based on a comprehensive evaluation of symptom progression, response to medical therapy, and imaging findings. Surgical intervention was considered for patients with worsening obstructive symptoms or gastrointestinal symptoms unresponsive to conservative treatment. Almost all patients underwent preoperative hormonal therapy; preoperative interventions such as ileus tube insertion or stoma formation were required in 7 patients. At our institution, preoperative hormonal therapy is primarily considered in elective surgical patients to reduce intraoperative bleeding and facilitate tissue dissection. In contrast, for patients requiring emergency surgery, there is insufficient time to administer preoperative medical treatment; therefore, it is generally not applied. The decision to initiate preoperative hormonal therapy is individualized and determined based on a comprehensive assessment of clinical factors, including symptom severity, extent of disease, and patient preferences.

A key therapeutic strategy for this disease is multidisciplinary surgery involving gynecologists. We actively utilized jelly-enhanced MRI, in which jelly is introduced transvaginally and transanally during imaging. This technique allowed three-dimensional evaluation of the extent of lesions and adhesions from the uterus and vaginal wall to the bowel[11]. Jelly-enhanced MRI clearly visualized the layered structure of the rectum and vagina and identified lesions suspected of involving the rectovaginal septum or obliteration of the pouch of Douglas. In this study, jelly-enhanced MRI was performed in 19 of 21 patients (90.5%) to improve visualization of rectovaginal adhesions and obliteration of the pouch of Douglas. Although our study did not include a comparative analysis of diagnostic accuracy with and without jelly, previous prospective studies have reported high diagnostic accuracy for jelly-enhanced MRI, with sensitivity and specificity of 90.9% and 77.8% for cul-de-sac obliteration, and 94.1% and 100% for deep endometriotic lesions, respectively[12]. These findings suggest that jelly-enhanced MRI is useful for preoperative assessment and surgical planning in patients with suspected deep infiltrating endometriosis. We actively utilized jelly-enhanced MRI in 19 of 21 patients (90.5%), which allowed detailed preoperative visualization of the rectovaginal septum and posterior pelvic adhesions. Rectovaginal adhesions or obliteration of the pouch of Douglas were identified preoperatively in 9 patients (47.4%), findings that matched intraoperative observations. Histopathological examination demonstrated that endometriotic tissue infiltrated the submucosa in 68.2% of patients and the muscularis propria in 31.8%. However, in this study, we did not specifically evaluate the correlation between jelly-enhanced MRI findings and histological depth of bowel wall infiltration. This remains an important subject for future investigation. Although its ability to predict transmural invasion was not analyzed, jelly-enhanced MRI contributed meaningfully to assessing pelvic adhesions and lesion extent, thereby facilitating surgical planning. We believe this modality significantly contributed to enhancing preoperative diagnostic accuracy and determining surgical planning, including the need for bowel resection or combined procedures[12]. Collaborative surgery with gynecologists enabled simultaneous resection of gynecological lesions such as ovarian cystectomy and oophorectomy, facilitating adhesiolysis and comprehensive pelvic disease management. This approach enabled both complete resection of intestinal endometriotic lesions and preservation of ovarian function and fertility.

Laparoscopic or open surgery are the primary approaches, while robot-assisted surgery is also an emerging option, accounting for 90.3%, 7.9%, and 1.7% of patients, respectively[6]. Surgical techniques include shaving (conservative excision without opening the bowel), full-thickness disc excision, and segmental resection, with reported proportions of 39.8%, 22%, and 38.2%, respectively[7].

For left-sided pelvic lesions, identifying the full extent of disease is particularly difficult, necessitating wide excision to prevent recurrence[10]. Since the serosal extent is not always clear preoperatively, the actual resection margin often becomes more distal than anticipated[13]. Jinushi et al. reported that intestinal endometriosis spreads concentrically along the long axis and from the superficial to deep layers along the short axis, with occasional skip lesions, based on pathological analyses. To reduce recurrence risk, a resection margin of 6-7 cm is considered necessary[14].

In our study, LAR was frequently selected to ensure an adequate margin in patients with rectal involvement and frozen pelvis, with 83.3% of patients with rectosigmoid lesions undergoing LAR. Complete resection of endometriotic tissue during initial surgery is crucial to prevent recurrence[15], as incomplete resection may lead to local relapse[16]. Although we did not systematically assess postoperative bowel and urinary function using tools such as the LARS score, notable late complications (Clavien-Dindo grade ≥ 2) included anastomotic stenosis (n=1), bowel stricture (n=1), and rectovaginal fistula (n=1). Although these complications are not directly reflective of LARS or neurogenic bladder, pelvic autonomic nerve injury associated with LAR cannot be excluded. However, the mean anastomotic level in our cohort was relatively high (7.6 cm from the anal verge), and visualization afforded by laparoscopic magnification in young women likely contributed to autonomic nerve preservation and reduced complications.

There are reports showing a low closure rate of temporary stomas created during the initial surgery[6]. In our study, a temporary ileostomy was created in 4 patients, and a colostomy prior to bowel resection in 2 patients. Among the 6 patients, 5 (83.3%) achieved stoma closure. One patient developed a rectovaginal fistula following stoma closure and required re-diversion. This patient had previously experienced an anastomotic leak, and the fistula may have resulted from a recurrent leak after closure. Although posterior colpotomy was performed in this patient, its direct contribution to the fistula remains unclear. Pathological examination confirmed that both the proximal and distal resection margins were negative for endometriotic tissue in this patient. The affected bowel segment was resected and anastomosed at a site without serosal changes, and no endometriotic involvement was identified at either margin. However, microscopic residual disease cannot be entirely excluded. Moreover, intestinal endometriosis has been reported to exhibit skip lesions, defined as discontinuous microscopic foci of endometriosis, which may potentially lead to recurrence despite negative resection margins. These factors highlight the difficulty of ensuring adequate resection margins and preventing recurrence in this condition.

Two patients underwent open surgery from the beginning. One had extensive pelvic endometriosis with ureteral involvement requiring preoperative nephrostomy and concurrent bowel resection, adhesiolysis, and ureteral reconstruction. The other had perforation and abscess formation at a colonic stent site, necessitating laparotomy after initial management at a referring hospital. Only one patient required conversion from laparoscopy to open surgery due to difficulty transecting a thickened bowel wall with a laparoscopic stapler.

Reported average operative time and blood loss are 349 minutes and 186 mL, respectively, which are longer and greater than in typical colorectal cancer surgery. These differences are attributed to the technical difficulty of adhesiolysis and layer identification in frozen pelvis.

Segmental resection is associated with higher risks of fistula and leakage[17], yet disc excision and shaving are also associated with high rates of rectovaginal fistula[7]. Although recurrence rates of intestinal endometriosis have been reported as high as 67% in some studies, our cohort had only one recurrence (4.8%) during long-term follow-up[18]. This favorable outcome may reflect our surgical policy of complete resection for deeply infiltrating lesions, especially when accompanied by stricture or full-thickness invasion. In this study, segmental resection included anterior resection and sigmoidectomy, depending on the location and extent of the lesion. Macroscopic and microscopic evaluations showed that several lesions extended beyond the stricture and serosal involvement to the muscularis and submucosa. However, no residual endometriotic tissue was identified at the resection margins, indicating that complete excision was achieved. We did not systematically assess longitudinal spread or satellite lesions pathologically, which remains a limitation. Furthermore, our long follow-up period (median: 2,088 days), appropriate resection margins, and collaborative surgery with gynecologists may have contributed to the low recurrence rate, though further validation in larger studies is warranted.

In addition to surgical and histopathological evaluations, tumor markers can support both the diagnosis and postoperative monitoring of intestinal endometriosis. Among these, CA125 is the most widely used marker and is commonly elevated in patients with endometriosis, with a reported sensitivity of 36% and specificity of 87% [19]. In our study, 18 of 21 patients exhibited elevated preoperative CA125 levels, all of which declined following surgery. These findings suggest that CA125 reflects disease activity and serves as a useful marker for monitoring therapeutic response.

Factors affecting fertility differ according to endometriosis phenotype (peritoneal, ovarian, deep infiltrating), with deep infiltrating endometriosis affecting 35-50% of women[20,21]. Although the impact of surgery for intestinal endometriosis on fertility is not consistent, it may improve spontaneous pregnancy rates while also posing risks for complications[22]. Two patients underwent unilateral oophorectomy and two underwent salpingectomy concurrently with bowel resection. In both patients of oophorectomy, the procedure was performed unilaterally, with the contralateral ovary preserved. Therefore, the impact on hormonal environment and reproductive function is considered to be minimal. Only one patient had a history of childbirth prior to surgery, and two patients achieved pregnancy postoperatively. As not all patients desired pregnancy, fertility outcomes must be interpreted cautiously. Moreover, many patients required continued postoperative hormonal therapy to prevent recurrence and control symptoms, potentially affecting pregnancy rates.

Despite the long operative time and significant blood loss, laparoscopic rectal resection for intestinal endometriosis is a safe and effective procedure for young women, offering fertility preservation, reduced complications, and improved cosmetic outcomes. While the malignant transformation rate of endometriosis is less than 1%, strict postoperative follow-up remains essential[23].

This study has several limitations. First, the sample size is relatively small, limiting the generalizability of our findings. Second, some data were self-reported, introducing potential bias. Further studies with larger sample sizes and objective evaluations are needed.

In conclusion, this study demonstrates several important clinical implications of bowel resection for intestinal endometriosis. These include the necessity of comprehensive evaluation to address diagnostic challenges, the utility of multidisciplinary surgical approaches, and the potential for reducing recurrence and preserving function through appropriate resection margins. Despite limitations such as the small sample size and lack of detailed functional assessment, our findings may contribute to the development of future treatment strategies for this condition.

Conflicts of Interest

There are no conflicts of interest.

Author Contributions

YK and MK were mainly involved in the patient treatment/surgery. MK, KS, SI, MK, KH, ST, YK, KM and KS were involved in patient treatment/follow-up. All authors read and approved the final manuscript.

Approval by Institutional Review Board (IRB)

This study was approved by the Juntendo University Ethics Committee (IRB approval number: E24-0439)