Comparative pregnancy rate after colorectal resection versus other surgical procedures for deep infiltrating rectal endometriosis: a systematic review and meta-analysis

We performed a systematic review and meta-analysis by following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 14 , and in line with the suggestions provided by the Cochrane Collaboration 15 . PROSPERO registration number: CRD42024512328. Included studies were cross-sectional, nonrandomized and randomized clinical trials, case-controls, while excluded studies were case-reports, general reviews, systematic reviews and studies with no control group. We searched PubMed, Web of Science, Cochrane Library and Clinical Trials for relevant studies published from inception to December 2024. The search strategy consisted of specific vocabulary and the National Library of Medicine’s Medical Subject Headings (MeSH) terms. We used the following search string: ((“infertility” OR “fertility” OR “reproductive”) OR (“in vitro” AND (fertilization OR “IVF” OR “assisted reproduction”))) AND ((colorectal AND (“endometriosis” OR endometriosis)) OR ((deep AND infiltrating AND (“endometriosis” OR endometriosis))) OR (“deep infiltrative endometriosis” OR (“intestinal” OR “bowel” OR “rectovaginal” OR “vaginal”)) AND (“endometriosis”)). We only considered English-language studies and we performed a thorough assessment of the references from the pertinent articles. Four authors (PFC, SH, RM, and MC) independently conducted the initial search to evaluate studies based on the eligibility criteria. Data extraction was performed by the same four authors (PFC, SH, RM, and MC), while the remaining author (JMA) reviewed and verified the extracted data. Studies were included if they met the following criteria: (i) they provided data on surgical management techniques (shaving, disc excision, and/or colorectal resection), and (ii) they detailed pregnancy outcomes in each subgroup, including spontaneous pregnancy and/or assisted reproductive technology (ART). The quality of each included study was independently assessed by four authors (PFC, SH, RM, and MC). Discrepancies in assessments were discussed among these authors, and if consensus could not be reached, a fifth reviewer (JMA) was consulted. We extracted and summarized the following data: author, year of publication, country, number of patients, type of study, surgical approach, surgical technique, period of inclusion, main outcome (pregnancy rate), modality of diagnosis, preoperative medical therapy, major complications, duration of infertility and follow-up duration. We assessed the pregnancy rate (number and percentage) in each group of patients. When data were available allowing the comparison for pregnancy rate, we performed a fixed-effects meta-analysis. We assessed the number, percentage, mean, standard deviation, median, interquartile range, and confidence interval (CI) for all measures of each study 16 . Odds ratios (ORs) and 95% confidence intervals (CIs) for dichotomous outcomes were calculated, with statistical significance defined as p  < 0.05. Statistical heterogeneity was evaluated using Cochran’s Q test and the I² statistic, where an I² value below 50% was indicative of low heterogeneity. In the absence of significant heterogeneity, a fixed-effects model was employed; otherwise, a random-effects model was utilized for pooled estimates. All statistical analyses were conducted using SAS software. During the data extraction process, researchers independently assessed the risk of bias for each study using the ROBINS-I tool (Risk Of Bias In Non-randomized Studies - of Interventions) 17 . The evaluation criteria included: Bias due to confounding, Bias in selection of participants into the study, Bias in classification of interventions, Bias due to deviations from intended interventions, Bias due to missing data, Bias in measurement of outcomes, Bias in selection of the reported results. We generated funnel plots to evaluate publication bias. We rated risk of bias for each criterion as low, moderate or high.

We included 13 studies in the systematic review and meta-analysis (Fig.  1 ). Six studies were prospective 7 , 18 – 22 , two were randomized trials 23 , 24 , and five were retrospective 25 – 29 . The 13 included studies enrolled a total of 3,248 patients. There was pregnancy information for 2,131 patients: 1073 who underwent colorectal resection, 502 who underwent shaving, 172 who underwent disc excision, and 384 who underwent other practices (expectant management). Tables  1 , 2 and 3 summarize the characteristics, outcomes, and postoperative outcomes of the 13 studies included in the review 7 , 18 – 29 . Eleven studies were conducted in Europe, in France 21 , 23 , 25 , 27 , 28 , Italy 19 , 24 , 26 , Finland 29 , Austria 18 , and Belgium 20 . One study 22 was conducted in the USA and one study 19 was conducted in Iran. Fig. 1 Flowchart. Flowchart. Table 1 Study characteristics. Study Year Country Study design Period Main outcome Modality of diagnosis for colopathology Modality of diagnosis for endometriosis Patients Preoperative medication Surgical approach Surgical techniques Complications Duration of infertility Follow up Hudelist, 2018 Austria Prospective study 2011 to 2016 Yes, short- and long-term outcomes regarding perioperative morbidity, long-term pain and fertility outcomes Not reported TVS 134 Not reported Laparoscopy (except one) Segmental resection, disk excision The rate of minor complications (Clavien-Dindo grade I and II) was 8.8% for the segmental resection (SR) group and 12.5% for the disk resection (DR) group. Major complications (grade III) occurred in seven of 102 women (6.9%) in the SR group and in one of 32 women (3.1%) in the DR group. Not reported 34 to 36 months Roman, 2022 France Randomized trial 2011 to 2013 Not, one of the following symptoms: constipation, frequent bowel movements, anal incontinence, or bladder dysfunction 24 months after surgery Not reported Pelvic magnetic resonance imaging (MRI) 55 Not reported Laparoscopy, N  = 51; Conversion to open route, N  = 4. Segmental resection, shaving or disk excision Among colorectal resection, N  = 28: 2 pelvic abscess or hematoma, 2 s laparoscopy, 1 transitory self-catheterization of the bladder; among excision group N  = 27, 1 rectovaginal fistula, 1 bladder fistula secondary bladder resection and 6 transitory self-catheterization of the bladder Not reported 84 months Lapointe, 2022 France Retrospective study 2009 to 2016 Yes, pregnancy rate after surgery Clinical assessment before and during surgery. Clinical and symptoms 94 none Laparoscopy Colorectal resection, shaving In the shaving group ( n  = 55), 6 (10.9%) patients presented with complications (Dindo Clavien grade II or higher)0.4 patients presented grade II complications: 3 had difficulty weaning from the urinary catheter and 1 patient required two red blood cell pellets transfusion. 2 patients presented grade IIIb complications: one ureteral wound with fistula and one hemoperitoneum. In the resection group ( n  = 39), 5 (12.8%) patients presented com- plications. 3 patients presented grade II complications: 3 had difficulties in weaning from the urinary catheter and one of them also required transfusion of two red blood cell pellets, 1 patient presented a large hemoperitoneum classified IIIb and 1 had a per-operative car- diac arrest classified IVb. 12.5 months 24 months Maggiore, 2017 Italy Retrospective study 2009 to 2015 Yes, Crude and cumulative SPR (spontaneous pregnancy rate) Not reported Transvaginal trasonography (TVS) 505 None Laparoscopy Segmental colorectal resection, shaving, disk excision Not reported 12 months 12 months Abo, 2018 France Retrospective study 2009 to 2015 Not, complications related to surgeries Not reported Not reported 364 None Laparoscopy Segmental colorectal resection, shaving, disk excision The mean rate of Clavien 3b postoperative complications was 11.8%, representing 43 patients, among whom more than half (29 patients) were in the segmental colorectal resection arm (P1 < 0.001). Eight cases of stenosis of colorectal anastomosis (5.8%) were recorded in the segmental colorectal resection arm (P1 = 0.003), six of which were managed by endoscopic dilation and two by secondary colorectal resection. Not reported 36 months Alborzi, 2022 Iran Prospective study 2015 to 2021 Yes, short and long term outcomes (including fertility) Endometriosis-related pain symptoms Colorectal endometriosis using imaging techniques and clinical symptoms 837 Not reported Laparoscopy (28–37% of the surgeries) and laparotomy Segmental colorectal resection, shaving, disk excision Fever and the need for blood transfusion, peritonitis and rectovaginal fistula were higher in segmental group than in shaving group. Abscess, external iliac artery injury, early and late bladder atony, and DVT, were seen in 2.7% of the patients and only in the segmental group. 11 to 14 months 12 months Bourdel, 2018 France Retrospective study 2000 to 2013 Yes, pain and fertility Invasion of rectum layers was further assessed using magnetic resonance imaging (MRI) and/or vaginal or rectal ultrasound Not reported 195 Not reported Laparoscopy (98.6%) and conversion Segmental colorectal resection, shaving The rate of minor complications (grade I and II) was 13% for the SG and 30% for the RG (RR = 2.83; 95% CI 1.01–7.63; p  = 0.05). Major complications (grade III) occurred in seven patients (4%) in the SG and in six patients (26%) in the RG (RR = 6.55; 95% CI 1.88–22.8; p  = 0.001). Not reported 60 to 67 months Meuleman, 2014 Belgium Prospective study 2006 to 2008 Not, clinical outcomes Not reported Vaginal examination, TVUS, bowel barium enema 203 GnRH analog therapy 3 months before surgery Laparotomy Colorectal resection, other surgery The prevalence of Clavien-Dindo grade III or higher (“major”) postoperative complications was 2% overall ( n  = 4/203) and was similar for the study (3%; n  = 2/76) and control (2%; n  = 2/127) groups. The prevalence of Clavien-Dindo grade I to II (“minor”) complications was 4% overall ( n  = 9/203) and was higher in the study group (11%; n  = 8/76) than in the control (1%; n  = 1/127) group ( P  = 0.002). Not reported 24 months Stepniewska, 2009 Italy Randomized trial 2000 to 2005 Yes, infertility questionnaire Double contrast barium enema Not reported 155 Not reported Laparoscopy Segmental colorectal resection, and no segmental resection The main surgical complications which occurred in group colorectal resection were two anastomotic fistulas (3.2%), both after a very low bowel resection. The other early (before a month after surgery) post-operative complications in group colorectal resection were: ureteral lesion (1.6%), bladder lesion (1.6%), bowel occlusion (1.6%) or severe blood loss (12.8%) treated by hemotransfusion (6.4%) or autotransfusion (6.4%). 30 months 27 months Tuominen, 2021 Finland Retrospective study 2004 to 2013 Yes, long-term pregnancy and delivery outcomes Clinical examination with typical findings, transvaginal ultrasound, or magnetic resonance Clinical examination, transvaginal ultrasound, or magnetic resonance 543 In conservative group: combined oral contraceptive pills, progestin-only pills, levonorgestrel-releasing intrauterine system, and gonadotropin-releasing hormone analogues with or without add-back as oral contraceptives or estrogen are used. Laparoscopy and laparotomy Operative and conservative Grade III–IV surgical complications were experienced by 22 (6.1%) women in the operative group, and four of the 35 women (11.4%) who underwent surgery later in the conservative group. Not reported 60 months Vercellini, 2006 Italy Prospective study Not reported Yes, incidence of pregnancy Vaginal and rectal examinations performed, transvaginal and transrectal ultrasonography, and histologic demonstration of endometriosis in a biopsy of the posterior fornix TVUS, TRUS, clinical examinations 105 None Laparotomy Colorectal resection, expectant management One major and 9 minor postoperative complications occurred. 23 to 25 months 27 months Mohr 2005 USA Retrospective study 1990 to 1997 Yes, complications and fertility rates Histologically or clinically, or both histologically and clinically modalities Not reported 178 Not reported Laparoscopy Segmental colorectal resection, shaving, disk excision Complications included 24 minor and 9 major ones for combined complication rates across all procedures of 13% and 5%, respectively. Four patients (3%) required an ad- ditional surgical procedure to address a complication. The least invasive procedure, shaving, had a significantly lower overall complication rate (6%) compared with 23% for disc excision ( P  < 0.007) and 38% for segmental resection ( P  < 0.001). Not reported 24 months Breteau 2019 France Retrospective study 2009 to 2016 Yes, overall pregnancy rate MRI examinations and confirmed by positive histological results + transvaginal and rectal ultrasound, or computed tomography-based virtual colonoscopy. Not reported 71 Hormonal treatment to relieve pain symptoms (e.g., GnRH analogues, progestins, or contraceptive pill) Not reported Segmental colorectal resection, shaving, disk excision Not reported Not reported 12 months DR: Disk resection is used to remove localized nodules or lesions of endometriosis that have infiltrated the bowel wall but do not require the removal of an entire segment of the bowel. SR: Segmental resection is employed to remove larger and deeper lesions of endometriosis that infiltrate multiple layers of the bowel wall. Shaving: The shaving procedure aims to remove superficial endometriotic lesions from the surface of organs, such as the bowel, bladder, or other pelvic structures, while preserving the underlying tissue. NR: Not Reported Study characteristics. In conservative group: combined oral contraceptive pills, progestin-only pills, levonorgestrel-releasing intrauterine system, and gonadotropin-releasing hormone analogues with or without add-back as oral contraceptives or estrogen are used. Vaginal and rectal examinations performed, transvaginal and transrectal ultrasonography, and histologic demonstration of endometriosis in a biopsy of the posterior fornix Complications included 24 minor and 9 major ones for combined complication rates across all procedures of 13% and 5%, respectively. Four patients (3%) required an ad- ditional surgical procedure to address a complication. The least invasive procedure, shaving, had a significantly lower overall complication rate (6%) compared with 23% for disc excision ( P  < 0.007) and 38% for segmental resection ( P  < 0.001). MRI examinations and confirmed by positive histological results + transvaginal and rectal ultrasound, or computed tomography-based virtual colonoscopy. DR: Disk resection is used to remove localized nodules or lesions of endometriosis that have infiltrated the bowel wall but do not require the removal of an entire segment of the bowel. SR: Segmental resection is employed to remove larger and deeper lesions of endometriosis that infiltrate multiple layers of the bowel wall. Shaving: The shaving procedure aims to remove superficial endometriotic lesions from the surface of organs, such as the bowel, bladder, or other pelvic structures, while preserving the underlying tissue. NR: Not Reported Table 2 Study characteristics with clinical information. Reference Characteristics of patients included Symptoms Dyspareunia Dysmenorrhea Rectorragy ? Dyschesia ? Urinary symptoms Adenomyosis Extensions (pelvic organs …) Hudelist, 2018 102 patients (SR), 32 patients (DR) Mean age: 34.5 years (SR), 34.0 years (DR), Infertility: 58.9% (SR), 37.5% (DR) Chronic pelvic pain 68.6% (SR), 84.4% (DR) 100% (both groups) Dyschezia: 65.7% (SR), 50% (DR) Dysuria: 12.7% (SR), 12.5% (DR) NR Vagina: 83.3% (SR), 87.5% (DR), Parametrium: 82.4% (SR), 96.9% (DR), Rectum/Sigmoid: 100% (both groups) Roman, 2022 55 patients with deep endometriosis infiltrating the rectum Mean age: 30 years (shaving/DR), 28 years (SR), Nullipara: 74.1% (shaving/DR), 71.4% (SR) Constipation, frequent bowel movements, anal incontinence, bladder dysfunction 74.1% (shaving/DR), 85.7% (SR) 100% (both groups) Pain during defecation: 77.8% (shaving/DR), 78.6% (SR) 29.6% (shaving/DR), 25.0% (SR) 77.8% (shaving/DR), 60.7% (SR) Endometrioma: 29.6% (shaving/DR), 46.4% (SR), Ureteral nodules 14.8% (shaving/DR), 10.7% (SR), Bladder nodule: 11.1% (shaving/DR), 3.6% (SR), Posterior vaginal fornix: 88.9% (shaving/DR), 71.4% (SR), Omentoplasty : 66.7% (shaving), 75% (SR) Lapointe, 2022 55 patients (shaving), 39 patients (SR), Mean age: 30.7 years, Nulliparous: 92.5%, Infertility: 39.4% NR NR NR Dyschesia : 43.6% (shaving), 53.9% (SR) Rectoragy: 14.6% (shaving), 30.8% (SR) Diarrhea : 7.3% (shaving), 5.1% (SR) NR 3.6% (shaving) 5.1% (SR) 100% stade 4 endometriosis. Not specifically detailed by organs. Maggiore, 2017 Mean age : 33.1 y (No SR), 33.3 y (SR) Mean age: 33 years (both groups), Nulliparous: 100% (both groups) Chronic pelvic pain, Infertility 6.3% (No SR), 9.5% (SR) 45.1% (No SR), 49.8% (SRl) Dyschezia: 41.5% (No SR), 46.2% (SR) NR Focal: 6.7% (No SR), 7.7% (SR), Diffuse: 16.9% (No SR), 19.0% (SR) Rectovaginal nodule with/without ovarian endometrioma Abo, 2018 145 patients (shaving), 80 patients (DR), 139 patients (SR) Mean age: 33 years, Nulliparous: 100% Chronic pelvic pain 80% (shaving), 82.5% (DR), 69.8% (SR) 96.5% (shaving), 96.2% (DR), 96.4% (SR) Dyschezia: 70.3% (shaving), 80% (DR), 70.5% (SR) Rectorrhage : 8.3% (shaving), 21.2% (DR), 21.6% (SR) 29.7% (shaving), 27.5% (DR), 23% (SR) NR Rectum: 100% (all groups) Rectosigmoid junction: 29.2% (shaving), 68.7% (DR), 66.9% (SR) Vagina: 6.9% (shaving), 42.5% (DR), 34.5% (SR) Diaphragm: 15.2% (shaving), 10% (DR), 24.5% (SR) Alborzi, 2022 263 patients (shaving), 248 patients (DR), 326 patients (SR) Mean age: 34.1 y (shaving), 34.8 y (DR), 35.4 years (SR) Nulliparous: 100% Chronic pelvic pain 77,08% (shaving), 92,59% (DR), 86,41% (SR) 77,36% (shaving) 84,48% (DR), 70,56% (SR) Constipation : 10,7% (shaving), 10.7% (DR), 14,8% (SR) Dyschezia 84,62% (shaving), 100% (DR), 81,61 (SR) NR NR Douglas pouch: 97%, bladder: 1.7%, Urters: 17.3%, External iliac artery: 0,11%, Nerve root: 0.45%. Bourdel, 2018 172 patients (shaving), 23 patients (SR), Mean age: 31.5 y (shaving), 32.7 y (SR), Nulliparous: 100% Chronic pelvic pain, infertility 60% (shaving), 65% (SR) 96.4% (shaving), 100% (SR) Rectorragy: 94% (both groups), Dyschezia: 70% (shaving), 70% (SR) NR NR Rectum: 100% (both groups) Vagina: 30% (SR) Uterosacral ligament: 68% (shaving), 67% (SR) Meuleman, 2014 76 patients (SR), 127 patients (No SR) Mean age: 32.9 y (SR), 32.1 y (No SR) Chronic pelvic pain: 56% (overall) 60% (SR), 51% (No SR) 90% (SR), 83% (No SR) Dyschezia: 70% (SR), 46% (No SR), Cyclical rectal bleeding: 19% (SR), 6% (No SR) Dysuria: 27% (SR), 15% (No SR) Hematuria 4% (SR), 3% (No SR) NR Nodules in Douglas pouch: 33% (SR), 14% (No SR), Nodules in rectovaginal septum: 18% (SR), 7% (No SR) Stepniewska, 2009 − 60 patients (SR, group A), Mean age: 31.7 y, Nulliparous: 100% − 40 patients (No SR, residual bowel endometriosis, group B), Mean age: 33.5 years, Nulliparous: 100% − 55 patients (No SR, no bowel involvement, group C), Mean age: 32.4 y, Nulliparous: 100% Chronic pelvic pain :52% (group A), 53% (group B), 35% (group C) 72% (group A) 65% (group B) 64% (group C) 98% (group A), 98% (group B), 65% (group C) Dyschezia: 72% (group A), 33% (group B), 42% (group C) Rectorrhage: 8% (group A), 5% (group B), 0% (group C) Dysuria: 20% (group A), 5% (group B), 9% (group C) NR Rectum: 100% (groups A, B), 0% (group C), Uterosacral ligament: 68% (group A), 67% (group B) Vagina: 15% (group A), 5% (group B), Parametrium: 5% (group A), 5% (group B) Endometrioma: 100% (group C) Tuominen, 2021 Mean age SR: 31.9 years No SR: 31.5 years Chronic pelvic pain: 64% (No SR), 95.8% (SR) Infertility: 47% (No SR), 34.4% (SR) NR NR NR NR NR NR Vercellini, 2006 SR: 44 patients, Mean age: 32.7 years, BMI: 22.8, Primary infertility: 93.2%, Secondary infertility: 6.8% No SR: 61 patients, Mean age: 33.7 years, BMI: 22.4, Primary infertility: 91.8%, Secondary infertility: 8.2% Chronic pelvic pain 52.3% (SR), 50.8% (No SR) 72.3% (SR), 75.4% (No SR) Dyschezia: 38.6% (SR), 36.1% (No SR) NR NR Pelvic extension or endometriomas excluded from the study. Mohr 2005 100 patients (shaving), 39 patients (DR), 48 patients (SR), Mean age: 35 years, Nulliparous : 100% Chronic pelvic pain NR 98.9% (all groups) Bowel symptoms: 74%, Constipation: 55%, Diarrhea: 41%, Rectorragy: 16% NR NR Rectovaginal septum (54.3%), bladder nodule (1.8%), ureterolysis (45.2%), diaphragm (2.7%), presacral neurectomy (5.9%), Breteau 2019 Pregnant group: 32 patients, Non-pregnant group: 41 patients, Mean age: 31.9 years, Mean duration of infertility: 48.4 months NR NR NR NR NR NR Rectum (57.5%), sigmoid: 34.2%, Ileal (9.6%), appendix (9.6%), diaphragm (15.1%), vagina (37%), bladder (8.2%) Study characteristics with clinical information. 102 patients (SR), 32 patients (DR) Mean age: 34.5 years (SR), 34.0 years (DR), Infertility: 58.9% (SR), 37.5% (DR) 55 patients with deep endometriosis infiltrating the rectum Mean age: 30 years (shaving/DR), 28 years (SR), Nullipara: 74.1% (shaving/DR), 71.4% (SR) Endometrioma: 29.6% (shaving/DR), 46.4% (SR), Ureteral nodules 14.8% (shaving/DR), 10.7% (SR), Bladder nodule: 11.1% (shaving/DR), 3.6% (SR), Posterior vaginal fornix: 88.9% (shaving/DR), 71.4% (SR), Omentoplasty : 66.7% (shaving), 75% (SR) 55 patients (shaving), 39 patients (SR), Mean age: 30.7 years, Nulliparous: 92.5%, Infertility: 39.4% Dyschesia : 43.6% (shaving), 53.9% (SR) Rectoragy: 14.6% (shaving), 30.8% (SR) Diarrhea : 7.3% (shaving), 5.1% (SR) 3.6% (shaving) 5.1% (SR) Mean age : 33.1 y (No SR), 33.3 y (SR) Mean age: 33 years (both groups), Nulliparous: 100% (both groups) Focal: 6.7% (No SR), 7.7% (SR), Diffuse: 16.9% (No SR), 19.0% (SR) 145 patients (shaving), 80 patients (DR), 139 patients (SR) Mean age: 33 years, Nulliparous: 100% 80% (shaving), 82.5% (DR), 69.8% (SR) 96.5% (shaving), 96.2% (DR), 96.4% (SR) Dyschezia: 70.3% (shaving), 80% (DR), 70.5% (SR) Rectorrhage : 8.3% (shaving), 21.2% (DR), 21.6% (SR) 29.7% (shaving), 27.5% (DR), 23% (SR) Rectum: 100% (all groups) Rectosigmoid junction: 29.2% (shaving), 68.7% (DR), 66.9% (SR) Vagina: 6.9% (shaving), 42.5% (DR), 34.5% (SR) Diaphragm: 15.2% (shaving), 10% (DR), 24.5% (SR) 263 patients (shaving), 248 patients (DR), 326 patients (SR) Mean age: 34.1 y (shaving), 34.8 y (DR), 35.4 years (SR) Nulliparous: 100% Constipation : 10,7% (shaving), 10.7% (DR), 14,8% (SR) Dyschezia 84,62% (shaving), 100% (DR), 81,61 (SR) Douglas pouch: 97%, bladder: 1.7%, Urters: 17.3%, External iliac artery: 0,11%, Nerve root: 0.45%. 172 patients (shaving), 23 patients (SR), Mean age: 31.5 y (shaving), 32.7 y (SR), Nulliparous: 100% Rectorragy: 94% (both groups), Dyschezia: 70% (shaving), 70% (SR) Rectum: 100% (both groups) Vagina: 30% (SR) Uterosacral ligament: 68% (shaving), 67% (SR) 76 patients (SR), 127 patients (No SR) Mean age: 32.9 y (SR), 32.1 y (No SR) Dyschezia: 70% (SR), 46% (No SR), Cyclical rectal bleeding: 19% (SR), 6% (No SR) Dysuria: 27% (SR), 15% (No SR) Hematuria 4% (SR), 3% (No SR) − 60 patients (SR, group A), Mean age: 31.7 y, Nulliparous: 100% − 40 patients (No SR, residual bowel endometriosis, group B), Mean age: 33.5 years, Nulliparous: 100% − 55 patients (No SR, no bowel involvement, group C), Mean age: 32.4 y, Nulliparous: 100% 72% (group A) 65% (group B) 64% (group C) Dyschezia: 72% (group A), 33% (group B), 42% (group C) Rectorrhage: 8% (group A), 5% (group B), 0% (group C) Rectum: 100% (groups A, B), 0% (group C), Uterosacral ligament: 68% (group A), 67% (group B) Vagina: 15% (group A), 5% (group B), Parametrium: 5% (group A), 5% (group B) Endometrioma: 100% (group C) Mean age SR: 31.9 years No SR: 31.5 years Chronic pelvic pain: 64% (No SR), 95.8% (SR) Infertility: 47% (No SR), 34.4% (SR) SR: 44 patients, Mean age: 32.7 years, BMI: 22.8, Primary infertility: 93.2%, Secondary infertility: 6.8% No SR: 61 patients, Mean age: 33.7 years, BMI: 22.4, Primary infertility: 91.8%, Secondary infertility: 8.2% 72.3% (SR), 75.4% (No SR) Dyschezia: 38.6% (SR), 36.1% (No SR) 100 patients (shaving), 39 patients (DR), 48 patients (SR), Mean age: 35 years, Nulliparous : 100% Pregnant group: 32 patients, Non-pregnant group: 41 patients, Mean age: 31.9 years, Mean duration of infertility: 48.4 months Table 3 Study characteristics with surgical information. Reference Surgical techniques endometrioma Preoperative characteristics of the digestive infiltration Previous ovarian endometriosis surgery Preoperative Infertile patients Objectives of surgery (sub occlusive symptoms, worst cases of DIE or worst reproductive prognosis at baseline) Additional surgery Modality of conception (spontaneous vs. ART) IVF before surgery outcome pregnancy (spontaneous abortion, normal deliv. or c-sections) Bias of the study Hudelist, 2018 SR N  = 102 DR N  = 32 Not reported Enzian C Classification (rectum/sigmoid) : C1 SR 1.9%, DR 75% C2 SR 18.7%, DR 25% C3 SR 79.4%, DR 0% Not reported SR 58.9% DR 37.5% Pain symptoms SR 41.3%, DR 62.5% Infertility SR 0.9%, DR 0% Pain symptoms and infertility SR 57.8% and DR 37.5% ureterolysis SR 40.2%, DR 43.8% ureteral reimplantation SR 2%, DR 3% Endometrioma SR 37.3%, DR 27.5% Spontaneous SR 40%, DR 54.5% IVF SR 24% , DR 9% Not reported Miscarriage: SR 15.6%, DR 14% Preterm birth: SR 12.5%, DR 14% Term delivery: SR 65.6%,DR 57.1% Roman, 2022 SR N  = 28 Shaving N  = 27 SR 29.6% Shaving 46.4% Not reported Not reported SR 28.6% Shaving 37% Not reported endometrioma SR 46.4% Shaving 29% bladder nodule SR 3.5% Shaving 11.1% ureterolysis SR 10.7% Shaving 11.1% vaginal fornix SR 71.4% Shaving 88,8% appendicectomy SR 7.1% Shaving 14.8% omentoplasty SR 75% Shaving 66.7% Spontaneous 57.4% IVF 42.5% Not reported Not reported Monocentric study Lapointe, 2022 SR N  = 39 shaving  N  = 55 SR 38% shaving 27% bowel infiltration SR 61% Shaving 42% Not reported Not reported All patients with a pregnancy desire who underwent surgery for DIE involving the bowel Not reported Spontaneous 25,5% SR 35% Shaving 18% IVF 26% shaving 23.6% SR 28% Not reported Monocentric retrospective study Maggiore, 2017 SR N  = 284 (group E), No surgery N  = 221 (group S) Unilateral SR 46.8% Np surgery 45.7% ; Bilateral SR 10.5% No surgery 10.8% Depth of infiltration of the bowel wall: Muscularis propria SR 10.2% No surgery 12.2% Submucosa SR 2.1% No surgery 2.7% Mucosa SR 1.4% No Surgery 0.9% Not reported Not reported live birth Remove all the endometriosis lesions and to reduce the pain Not reported Cumulative Spontaneous Pregnancy Rates) SR 23.8% No surrgery 39.5%; No Not reported Retrospective study Small sample size Abo, 2018 SR N  = 139 DR N  = 80 Shaving N  = 145 SR 61.9% DR 43.7% Shaving 16.5% Not reported Not reported SR 43.9% DR 22.5% Shaving 33.8% To reduce pain and infertility Hysterectomy + colpectomy SR 3.6%, DR 5%, Shaving 29% Bladder resection SR 10.1%, DR 3.7%, Shaving 5.5% . One year after the procedure 37.5% of infertile women became pregnant, and 33.3% of pregnancies were spontaneous. Three years after the procedure, the PR reached 66.7% and 50% of pregnancies were spontaneous. Not reported Not reported Small sample size Alborzi, 2022 SR N  = 326 DR N  = 248 Shaving N  = 263 Left endometrioma 58.3%, right endometrioma 61.2% Not reported Not reported failed IVF before surgery SR 51.1% DR 17.0% Shaving 29.6% To reduce pain and/or infertility right salpingectomy SR 47,7%, DR 50,6%, Shaving 41.3% left salpingectomy SR 53.5%, DR 56.5%, Shaving 51.7%, right ureterolysis SR 14.1%, DR 9.8%, Shaving 11.3% left ureterolysis SR 15.6%, DR 9.8%, Shaving 10% ureteral reanastomosis SR 2%, DR 0.7%, Shaving 0% right oophorectomy SR 3.9%, DR 16.9%, Shaving 14.8% left oophorectomy SR 2.7%, DR 18.8%, Shaving 17.8% spontaneous in shaving 27.6% Not reported in other group. failed IVF before surgery SR 51.1% DR 17.0% Shaving 29.6% abortion: Shaving 1.9% DR 4.4% SR 2.6% Lack of datas on modality of conception after surgery. Bourdel, 2018 SR N  = 23 Shaving N  = 172 SR 90.9% Shaving 52.3% Not reported Not reported SR 0% Shaving 6.9% Pain SR 78.2%, Shaving 51.2%, infertility SR 0%, Shaving 7.5% infertility and pain SR 21.7%, Shaving 41.5% Adhésiolysis SR 78.2%, Shaving 75% ureterolysis SR 69.6%, Shaving 68% colpectomy SR 30.4%, Shaving 75% Bladder resection SR 4.4%, Shaving 11.6% cystectomy SR 65.2%, Shaving 57.8% ureteral nodule SR 0%, Shaving 2.4% oophorectomy SR 21.7%, Shaving 4% SR Spontaneous pregnancies: 26%,IVF: 13% Shaving spontaneous pregnancies 36.2% IVF: 26.7% Not reported SR: 44.4% vaginal delivery, 44.4% CS, 11.1% miscarriage Shaving : 51% vaginal delivery, 25% CS, 13.8% miscarriages, 2.2% ectopic pregnancies, 2.2% therapeutic abortion, 1 abortion Confusion: bias additional surgeries Meuleman, 2014 SR N  = 76 No SR N  = 127 Not reported Transmural bowel invasion SR 90% no SR 40% Not reported, Not reported Infertility SR 5%, not SR 18% infertility + pain SR 90%, nit SR 74% Bladder resection SR 20%, no SR 8% Ureter SR 78%, no SR 66% Salpingectomy SR 29%, no SR 9% Cystectomy SR 11%, no SR 23% Ovariectomy SR 8%, no SR 2% SR spontaneous 23%, IVF 39% No SR spontaneous, 18.7% IVF Not reported Not reported Monocentric study, response rate lower in no SR group Stepniewska, 2009 SR N  = 60 No SR N  = 40 Not reported Stenosis bowel resection SR 49%, no SR 23%, Not reported All Sub occlusive symptoms all + infertility since > 1 year caecum resection SR 6.6%, no SR 0% ileal resection SR 8.3%, no SR 0% vaginal resection SR 8.3%, no SR 0% bladder resection ST 8.3%, no SR 0%, kystectomy SR 55%, no SR 60% ,appendiccectomy SR 1.6%, no SR 0% Salpingectomy SR 3.3%, no SR 5% SR spontaneous 20%, IVF 8.3% No SR spontaneous 2.5%, IVF 2.5% Not reported Not reported Tuominen, 2021 SR N  = 132 No SR = 192 39.6% all Not reported Not reported 50.5% all Pain 62.5% Infertility 1.1% Pain + infertility 33.3% Other 3.1% Additional surgery on ovaries SR 68%; no SR 48% Cumulative pregnancy rate 76.2% SR 76% no SR 74.5% Spontaneous 44.2%. SR 32.1% no SR 48.8% Not reported LBR SR 62%, no SR 60% Vaginal delivery SR 55.1%, no SR 53.7% CS SR 44.9%, no SR 46.3% Retrospective study, memory biais, selection biais and missing data Vercellini, 2006 Rectal Surgery N  = 44 (Shaving N  = 38 DR N  = 6) No surgery N  = 61 0% Not reported 0% All with rectal endometriosis as unique factor of infertility 100% infertility In rectal surgery group ureterolysis 13.6% segmental bladder resection 2.2% Pregnancy rate 34% in rectal surgery group, 36% no surgery group IVF 29.5% in rectal surgery group 31% in no surgery group Not reported Not reported Retrospective study, selection biais and missing data Mohr 2005 SR N  = 48 DR N  = 39 Shaving N  = 100 Not reported 100% infiltrative bowel disease Not reported 32% all Pain 99% Infertility 32% Bowel symptoms 74% ALL: ureterolysis 45%, Adhesiolysis 13%, myomectomy 12%, colpectomy 11%, ureter 14%, bladder 9%, diaphragm 3%, cystectomy 4%, neurectomy 6%, ovariectomy 8%, hysterectomy 17% Not reported Not reported At term delivery 78.5% ( among all pregnancies) 17.8% miscarriages, 3.6% termination. retrospective monocentric study Breteau 2019 Surgery N  = 104 SR N  = 25 DR N  = 4 Shaving N  = 22 Not reported Not reported Not reported All : 2 previous IVF or ICSI failures 100% infertility urinary tract all 10.5% Right ovarian endometrioma : Cyst drained and washed all 7.7% ; plasma energy ablation all 25.9% Left ovarian endometriosis : Cyst drained and washed all 5.7% plasma energy ablation all26.9% Cystectomy 2.9% Ovariectomy 1% Pregnancy rate 43.8% 21.8% spontaneous, 10 78.2% IVF All: 2 previous IVF or ICSI failures 53% delivery at term, 18.7% premature birth, 18.7% early miscarriage, 6.3% ectopic pregnancy. 1% molar pregnancy. DR: disk excision, SR: segmental resection, CS: cesarean section CPR: cumulative pregnancy rate, LBR: life birth rate. Study characteristics with surgical information. SR N  = 102 DR N  = 32 Enzian C Classification (rectum/sigmoid) : C1 SR 1.9%, DR 75% C2 SR 18.7%, DR 25% C3 SR 79.4%, DR 0% SR 58.9% DR 37.5% Pain symptoms SR 41.3%, DR 62.5% Infertility SR 0.9%, DR 0% Pain symptoms and infertility SR 57.8% and DR 37.5% ureterolysis SR 40.2%, DR 43.8% ureteral reimplantation SR 2%, DR 3% Endometrioma SR 37.3%, DR 27.5% Spontaneous SR 40%, DR 54.5% IVF SR 24% , DR 9% Miscarriage: SR 15.6%, DR 14% Preterm birth: SR 12.5%, DR 14% Term delivery: SR 65.6%,DR 57.1% SR N  = 28 Shaving N  = 27 endometrioma SR 46.4% Shaving 29% bladder nodule SR 3.5% Shaving 11.1% ureterolysis SR 10.7% Shaving 11.1% vaginal fornix SR 71.4% Shaving 88,8% appendicectomy SR 7.1% Shaving 14.8% omentoplasty SR 75% Shaving 66.7% Spontaneous 57.4% IVF 42.5% SR N  = 39 shaving  N  = 55 SR 38% shaving 27% bowel infiltration SR 61% Shaving 42% All patients with a pregnancy desire who underwent surgery for DIE involving the bowel Spontaneous 25,5% SR 35% Shaving 18% IVF 26% SR N  = 284 (group E), No surgery N  = 221 (group S) Unilateral SR 46.8% Np surgery 45.7% ; Bilateral SR 10.5% No surgery 10.8% Depth of infiltration of the bowel wall: Muscularis propria SR 10.2% No surgery 12.2% Submucosa SR 2.1% No surgery 2.7% Mucosa SR 1.4% No Surgery 0.9% Cumulative Spontaneous Pregnancy Rates) SR 23.8% No surrgery 39.5%; Retrospective study Small sample size SR N  = 139 DR N  = 80 Shaving N  = 145 SR 61.9% DR 43.7% Shaving 16.5% SR 43.9% DR 22.5% Shaving 33.8% Hysterectomy + colpectomy SR 3.6%, DR 5%, Shaving 29% Bladder resection SR 10.1%, DR 3.7%, Shaving 5.5% . One year after the procedure 37.5% of infertile women became pregnant, and 33.3% of pregnancies were spontaneous. Three years after the procedure, the PR reached 66.7% and 50% of pregnancies were spontaneous. SR N  = 326 DR N  = 248 Shaving N  = 263 failed IVF before surgery SR 51.1% DR 17.0% Shaving 29.6% right salpingectomy SR 47,7%, DR 50,6%, Shaving 41.3% left salpingectomy SR 53.5%, DR 56.5%, Shaving 51.7%, right ureterolysis SR 14.1%, DR 9.8%, Shaving 11.3% left ureterolysis SR 15.6%, DR 9.8%, Shaving 10% ureteral reanastomosis SR 2%, DR 0.7%, Shaving 0% right oophorectomy SR 3.9%, DR 16.9%, Shaving 14.8% left oophorectomy SR 2.7%, DR 18.8%, Shaving 17.8% spontaneous in shaving 27.6% Not reported in other group. failed IVF before surgery SR 51.1% DR 17.0% Shaving 29.6% SR N  = 23 Shaving N  = 172 SR 90.9% Shaving 52.3% SR 0% Shaving 6.9% Pain SR 78.2%, Shaving 51.2%, infertility SR 0%, Shaving 7.5% infertility and pain SR 21.7%, Shaving 41.5% Adhésiolysis SR 78.2%, Shaving 75% ureterolysis SR 69.6%, Shaving 68% colpectomy SR 30.4%, Shaving 75% Bladder resection SR 4.4%, Shaving 11.6% cystectomy SR 65.2%, Shaving 57.8% ureteral nodule SR 0%, Shaving 2.4% oophorectomy SR 21.7%, Shaving 4% SR Spontaneous pregnancies: 26%,IVF: 13% Shaving spontaneous pregnancies 36.2% IVF: 26.7% Confusion: bias additional surgeries SR N  = 76 No SR N  = 127 Transmural bowel invasion SR 90% no SR 40% Infertility SR 5%, not SR 18% infertility + pain SR 90%, nit SR 74% Bladder resection SR 20%, no SR 8% Ureter SR 78%, no SR 66% Salpingectomy SR 29%, no SR 9% Cystectomy SR 11%, no SR 23% Ovariectomy SR 8%, no SR 2% SR spontaneous 23%, IVF 39% No SR spontaneous, 18.7% IVF SR N  = 60 No SR N  = 40 caecum resection SR 6.6%, no SR 0% ileal resection SR 8.3%, no SR 0% vaginal resection SR 8.3%, no SR 0% bladder resection ST 8.3%, no SR 0%, kystectomy SR 55%, no SR 60% ,appendiccectomy SR 1.6%, no SR 0% Salpingectomy SR 3.3%, no SR 5% SR spontaneous 20%, IVF 8.3% No SR spontaneous 2.5%, IVF 2.5% SR N  = 132 No SR = 192 Pain 62.5% Infertility 1.1% Pain + infertility 33.3% Other 3.1% Cumulative pregnancy rate 76.2% SR 76% no SR 74.5% Spontaneous 44.2%. SR 32.1% no SR 48.8% LBR SR 62%, no SR 60% Vaginal delivery SR 55.1%, no SR 53.7% CS SR 44.9%, no SR 46.3% Rectal Surgery N  = 44 (Shaving N  = 38 DR N  = 6) No surgery N  = 61 In rectal surgery group ureterolysis 13.6% segmental bladder resection 2.2% Pregnancy rate 34% in rectal surgery group, 36% no surgery group IVF 29.5% in rectal surgery group 31% in no surgery group SR N  = 48 DR N  = 39 Shaving N  = 100 Pain 99% Infertility 32% Bowel symptoms 74% ALL: ureterolysis 45%, Adhesiolysis 13%, myomectomy 12%, colpectomy 11%, ureter 14%, bladder 9%, diaphragm 3%, cystectomy 4%, neurectomy 6%, ovariectomy 8%, hysterectomy 17% Surgery N  = 104 SR N  = 25 DR N  = 4 Shaving N  = 22 urinary tract all 10.5% Right ovarian endometrioma : Cyst drained and washed all 7.7% ; plasma energy ablation all 25.9% Left ovarian endometriosis : Cyst drained and washed all 5.7% plasma energy ablation all26.9% Cystectomy 2.9% Ovariectomy 1% 53% delivery at term, 18.7% premature birth, 18.7% early miscarriage, 6.3% ectopic pregnancy. 1% molar pregnancy. DR: disk excision, SR: segmental resection, CS: cesarean section CPR: cumulative pregnancy rate, LBR: life birth rate. Shaving is the least invasive procedure, consisting of a layer-by-layer excision of the rectal lesion, without full-thickness resection. Shaving is used in cases where the lesion has partially invaded the muscular layers, but not the full thickness of the bowel wall 9 . Disc excision consists of full-thickness excision of the rectal lesion on the bowel wall with opening of the mucosa and a transverse repair 9 . Anterior rectosigmoid segmental resection with end-to- end anastomosis is performed in cases of full-thickness large or multiple rectal lesions. It may or may not be protected by an ileostomy 9 . In five studies 22 , 24 – 27 , the surgical procedures were laparoscopic, and in two studies 7 , 20 the surgery was laparotomic. One study did not provide this information, and the other studies 18 , 19 , 23 , 28 , 29 presented different rates of both types of surgery, including conversion to laparotomy. There are different definitions for these surgeries 13 . Only three studies provided a clear definition of these types of surgery: Abo et al. 27 defined shaving as nodule excision without opening the rectum and disc excision as resection of the nodule with excision of the anterior rectal wall. Bourdel et al. 28 defined shaving as a “reverse technique” starting from the lateral and the vaginal parts of the nodule and ending with rectal invasion. Mohr et al. 22 defined shaving as the least invasive procedure, consisting of layer-by-layer combined carbon dioxide (CO 2 ) laser vaporisation and excision of the lesion, without full-thickness resection; they defined disc excision as full-thickness excision of the lesion on the bowel wall with transverse repair. Colorectal resection was associated with a lower pregnancy rate compared with other techniques ( N  = 2,131, 35.5% vs. 42.6%, OR = 0.64 [95% CI 0.52–0.79], p  < 0.001, I 2 = 35%) 7 , 18 – 29 (Fig.  2 ). There was a similar result when comparing colorectal resection with shaving ( N  = 952, 17.3% vs. 38.8%, OR = 0.51 [95% CI 0.36–0.73], p  < 0.001, I 2 = 0%) 19 , 21 , 22 , 25 – 28 (Fig.  3 ), but not when comparing colorectal resection with disc excision ( N  = 432, 29.2% vs. 35.8%, OR = 0.65 [95% CI 0.37–1.13], p = 0.13) 18 , 19 , 21 – 23 , 27 (Fig.  4 ). The test of asymmetry funnel plots (Fig.  5 ) indicated publication bias among the included studies. Fig. 2 Forest plot of the pregnancy rate comparing colorectal resection with other techniques. Forest plot of the pregnancy rate comparing colorectal resection with other techniques. Fig. 3 Forest plot of the pregnancy rate comparing colorectal resection with shaving. Forest plot of the pregnancy rate comparing colorectal resection with shaving. Fig. 4 Forest plot of the pregnancy rate comparing colorectal resection with disc excision. Forest plot of the pregnancy rate comparing colorectal resection with disc excision. Fig. 5 Funnel plots for (A) colorectal resection compared with all options, (B) colorectal resection compared with shaving, and (C) colorectal resection compared with disc excision. Funnel plots for (A) colorectal resection compared with all options, (B) colorectal resection compared with shaving, and (C) colorectal resection compared with disc excision. Six studies reported spontaneous pregnancy rate, and we observed no significant difference between colorectal resection and other techniques, N  = 912, 34.9% vs. 34.1%, (OR = 1.33 [0.97–1.83], p = 0.08) 18 , 20 , 25 , 26 , 28 , 29 (Fig.  6 ). Fig. 6 Forest plot of the occurrence of spontaneous pregnancy rate comparing colorectal resection to other techniques. Forest plot of the occurrence of spontaneous pregnancy rate comparing colorectal resection to other techniques. Included studies presented moderate risk of bias for non-randomized studies according to the ROBINS-I tool risk of bias. Maggiore et al. presented unclear definition of disc excision, thus this study has been removed from the analysis of colorectal resection vs. disc excision. Risk of bias is shown in Table 4 . Table 4 Risk of bias. Red: high risk, yellow: moderate risk, green: low risk of bias. Risk of bias. Red: high risk, yellow: moderate risk, green: low risk of bias.

This systematic review and meta-analysis highlights that colorectal resection for deep infiltrating endometriosis is associated with a lower pregnancy rate compared to less invasive surgical approaches, such as rectal shaving. While colorectal resection may provide thorough removal of endometriotic tissue, its invasiveness is linked to potential disruptions in pelvic anatomy, increased adhesions, and other factors that could negatively impact fertility outcomes. However, the findings must be interpreted with caution due to significant limitations, including the heterogeneity of the included studies, variability in surgical indications, and the lack of consistent reporting on associated procedures, such as treatment for ovarian endometriosis or prior IVF. Furthermore, the inclusion of only two randomized clinical trials underscores the need for more robust research. Future studies, particularly randomized controlled trials, are essential to clarify the impact of different surgical techniques on fertility outcomes. Such research should aim to standardize reporting criteria and evaluate long-term reproductive health to provide clearer guidance for clinical decision-making. Until then, surgical approaches should be carefully tailored to individual patient profiles, considering both reproductive goals and the risks associated with more invasive procedures.

We conducted a meta-analysis focusing on pregnancy rates following surgery for deep infiltrating endometriosis (DIE) affecting the rectum. Our findings indicate that colorectal resection is associated with a lower pregnancy rate compared to other surgical techniques, including rectal shaving and disc excision. The debate surrounding colorectal surgery for endometriosis persists, primarily due to concerns about the associated risks. Recent studies, including meta-analyses, have highlighted that this type of surgery can result in severe complications such as rectovaginal fistula, anastomotic leakage, anastomotic stenosis, and voiding dysfunction. In contrast, rectal shaving has been shown to have a lower risk of postoperative complications compared to other surgical techniques like disc excision and segmental colorectal resection 9 , 13 , 30 . Despite these findings, no prior study has reached a definitive conclusion regarding the differences in pregnancy rates between these surgical methods. Endometriosis can negatively affect both natural and assisted conception. Mechanisms proposed to explain its impact on reproductive dysfunction include impaired folliculogenesis leading to follicular dysfunction and reduced egg quality, luteal phase abnormalities, decreased fertilization rates, and abnormal embryogenesis 31 , 32 . In addition to its effects on conception, endometriosis can also influence obstetric and neonatal outcomes of pregnancy. These include an increased risk of small-for-gestational-age infants, caesarean section, miscarriage, hemorrhage, low placental adhesion, and preterm delivery 33 , 34 . During pregnancy, hormonal changes may alter endometriotic lesions, leading to atypical appearances that pose a clinical dilemma. Managing these events is challenging due to the limited number of reported cases. A review of the available literature highlights the lack of formal estimates of their incidence. While complications of endometriosis during pregnancy are rare, there is no evidence to suggest that the disease has a significant detrimental effect on pregnancy outcomes. Endometriosis can negatively impact both natural and assisted conception processes 35 . Proposed mechanisms, including impaired folliculogenesis resulting in follicular dysfunction and poor egg quality, as well as luteal phase abnormalities, reduced fertilization rates, and abnormal embryogenesis, help to explain the reproductive challenges associated with the condition 32 . Additionally, endometriosis may affect obstetric and neonatal outcomes, contributing to complications such as small-for-gestational-age infants, an increased rate of caesarean sections, miscarriages, hemorrhages, low placental adhesion, and preterm deliveries 36 . During pregnancy, hormonal changes can alter endometriotic lesions, which may appear atypical and present a clinical challenge. Managing these conditions is difficult due to the limited number of documented cases, and a review of the literature reveals a lack of formal incidence estimates. Despite these challenges, complications arising from endometriosis during pregnancy are uncommon, and there is no significant evidence indicating that the condition substantially worsens pregnancy outcomes 37 , 38 . The observed lower pregnancy rate associated with colorectal resection compared to rectal shaving and disc excision in cases of endometriosis may be attributed to several factors. A study comparing the laparoscopic surgical management of bowel endometriosis using three techniques—shaving, discoid excision, and colorectal resection—revealed that all methods were effective for immediate symptom relief. However, the shaving group experienced significantly higher rates of symptom recurrence and the need for reintervention. This suggests that while the less invasive shaving technique provides initial relief, it may not offer as durable a resolution as the more invasive segmental resection. Notably, that study did not directly compare pregnancy rates among the techniques 39 . The choice of surgical method is influenced by factors such as the size and location of endometriotic nodules, the patient’s fertility desires, and the risk of complications. Conservative techniques like shaving, which preserve more of the bowel, may reduce complication rates but might not completely remove all endometriotic tissue, potentially leading to higher recurrence rates. In contrast, more invasive procedures like segmental resection may achieve more thorough removal of endometriotic tissue but carry a higher risk of complications due to their invasiveness. This meta-analysis provides valuable insights into the impact of rectal resection on fertility. The decision regarding surgical technique should be tailored to each patient’s unique case and fertility goals. Lapointe et al. (2022) evaluated postoperative fertility and pregnancy outcomes in patients with colorectal DIE, comparing outcomes between those undergoing shaving and those undergoing digestive resection (discoid or segmental). The study included 94 patients, with 39 undergoing bowel resection and 55 undergoing shaving. While no significant difference was found in the overall pregnancy rate between the two groups, there was a higher rate of spontaneous pregnancies in the resection group 25 . However, our meta-analysis found no significant association between resection surgery and spontaneous pregnancy. This suggests that resection surgery may not confer a greater advantage for spontaneous pregnancy occurrence compared to overall pregnancy outcomes. Several physiological pathways may explain the lower pregnancy rate observed with colorectal resection compared to shaving and disc resection in women with endometriosis. Colorectal resection is a more extensive and invasive procedure, often involving the removal of a portion of the bowel. This can lead to significant alterations in pelvic anatomy, potentially affecting adjacent organs such as the uterus and fallopian tubes, thereby impacting fertility 40 . More invasive surgeries, like colorectal resection, are associated with increased post-surgical adhesions (scar tissue formation). These adhesions can impair the normal functioning of reproductive organs by obstructing the fallopian tubes or altering the uterine environment, which may reduce fertility 41 . Additionally, invasive procedures can indirectly affect ovarian reserve. Any surgery performed in the pelvic area carries the risk of disrupting the blood supply to the ovaries or causing inadvertent damage, potentially reducing the number of viable eggs available for fertilization 42 . Major surgeries also trigger a heightened immune and inflammatory response. This increased inflammatory environment in the pelvis following surgery could adversely impact implantation and early embryonic development 5 . Furthermore, colorectal resection carries a higher risk of nerve damage or alterations in pelvic innervation 40 . Such changes can affect uterine and tubal function, as well as overall pelvic health, which are critical for successful conception and pregnancy. The severity of pain symptoms plays a pivotal role in the decision-making process for the surgical treatment of endometriosis, significantly impacting both pain relief and fertility outcomes. Studies indicate that patients experiencing severe pain symptoms, such as dysmenorrhea, dyspareunia, and dyschezia, are more likely to pursue surgical interventions, which can result in substantial pain relief and an improved quality of life 7 , 24 . A shared decision-making approach, involving thorough discussions about the risks and benefits of surgery versus expectant management, is essential to ensure that treatment decisions align with the patient’s preferences and clinical indications. Consequently, a comprehensive evaluation of pain symptoms is critical for developing individualized treatment plans that effectively address pain management while optimizing reproductive health. The strength of this meta-analysis lies in its novelty, as it appears to be the first to compare pregnancy rates across different surgical approaches for endometriosis. However, several limitations warrant attention. According to the Cochrane guidelines, assessing the risk of bias is typically required. However, given that our systematic review included only two randomized controlled trials, we deemed it inappropriate to conduct a risk of bias evaluation. Additionally, the heterogeneity observed among some studies may impact the precision and reliability of the findings. Variability in how results were reported across studies further complicated the meta-analysis of other criteria, limiting its overall feasibility. The heterogeneity in the stages of endometriosis presented a significant limitation, particularly when comparing surgical types and pregnancy outcomes for different levels of deep infiltrative endometriosis. Furthermore, the inclusion criteria do not allow for a definitive conclusion that bowel resection results are inferior. This is because the indications for bowel resection were not consistently considered, and large lesions were less likely to be treated by bowel resection, especially since only two randomized clinical studies were included. Another key limitation is the lack of reporting on associated procedures, particularly those involving the treatment of ovarian endometriosis, as well as the number of patients who underwent IVF prior to surgery. These factors were not documented in the included studies, representing a significant limitation of this meta-analysis.

Endometriosis, a common condition affecting the female reproductive system, affects approximately 5–10% of women during their reproductive years 1 . This condition can manifest in the pelvic region in three distinct forms: superficial endometriosis affecting the peritoneum, ovarian endometriosis (endometriomas), and deep infiltrating endometriosis (DIE), which may involve various locations, such as the rectovaginal area 2 . DIE lesions in the rectovaginal area typically originate at the posterior cervix and progressively involve the anterior portion of the rectum 3 . In cases of DIE, disrupted anatomy and severe intra-abdominal adhesions interfere with gamete transport. Even women with minimal or mild endometriosis, who do not have adhesions, experience reduced fertility. This decline in fertility appears to be linked to disruptions in ovulation, fertilization, implantation, and an increased risk of miscarriage 4 . Patients with endometriosis undergoing in vitro fertilization (IVF) treatments face lower rates of fertilization, implantation, and pregnancy compared to women without the condition. The outcomes are particularly poor for those with advanced disease, with DIE specifically negatively impacting IVF and intracytoplasmic sperm injection (ICSI) results 5 . For patients with mild to minimal endometriosis, surgical treatment offers benefits with a low complication rate. However, the situation is more complex for those with advanced disease due to limited evidence and the potential for severe complications. Reduced fertility remains a significant issue for women with endometriosis. The effect of surgeries for DIE, particularly those involving the bowel wall, on fertility remains a controversial topic, with diverse opinions reflected in the literature 6 . Medical interventions can alleviate the symptoms of rectovaginal DIE, but they do not eliminate the disease and are frequently accompanied by side effects such as irregular bleeding, weight changes, reduced sexual desire, and headaches 7 . Additionally, pregnancy does not halt the progression of the disease 8 . Surgical excision of rectovaginal DIE, however, has been shown to improve fertility prospects 9 . Rectovaginal DIE in pregnant women has been associated with increased risks of preterm birth, hospital admissions, and low birth weight in newborns 10 . When rectovaginal DIE lesions involving the rectum cause symptoms that adversely affect pain, bowel, sexual, or reproductive functions, despite optimized medical treatment, surgical removal may be considered. Various surgical approaches—including laparoscopic bowel resection, disc excision, and rectal shaving—are utilized, but there is no consensus on the optimal surgical technique or timing for these interventions. Criteria for bowel resection, such as infiltration of the rectal mucosa or invasion involving more than half of the bowel’s circumference, have been proposed 9 . However, these criteria and the decision-making process for surgery in cases of rectovaginal DIE remain subjects of ongoing debate 9 , 11 . Some studies have suggested that colorectal surgery aimed solely at improving reproductive outcomes in infertile patients with intestinal endometriosis should be performed only in research settings and by highly experienced surgeons 12 . Women should be informed about the uncertainties regarding the risks and benefits of bowel surgery in various clinical scenarios. Preoperative counselling must be conducted impartially, with the goal of fostering a shared medical decision-making process. The lack of consensus regarding the optimal surgical management of DIE contributes to these discordant findings 9 , 13 . Furthermore, no meta-analysis has yet investigated the impact of different surgical techniques on pregnancy rates. This meta-analysis study aims to evaluate how colorectal resection surgery compares to other surgical techniques (e.g., rectal shaving, disc excision) in terms of pregnancy rates in women with deep infiltrating rectal endometriosis.