STUDY QUESTION Is there a difference in functional outcomes and recurrence rate between conservative versus radical rectal surgery in patients with large deep endometriosis infiltrating the rectum 5 years postoperatively? SUMMARY ANSWER No evidence was found that long-term outcomes differed when nodule excision was compared to rectal resection for deeply invasive endometriosis involving the bowel. WHAT IS KNOWN ALREADY Functional outcomes of nodule excision and rectal resection for deeply invasive endometriosis involving the bowel are comparable 2 years after surgery. Despite numerous previously reported case series enrolling patients managed for colorectal endometriosis, long-term data remain scarce in the literature. STUDY DESIGN, SIZE, DURATION From March 2011 to August 2013, we performed a two-arm randomized trial, enrolling 60 patients with deep endometriosis infiltrating the rectum up to 15 cm from the anus, measuring >20 mm in length, involving at least the muscular layer in depth, and up to 50% of rectal circumference. Among them, 55 women were enrolled at one tertial referral centre in endometriosis, using a randomization list drawn up separately for this centre. Institute review board approval was obtained to continue follow-up to 10 years postoperatively. One patient requested to stop the follow-up 2 years after surgery. PARTICIPANTS/MATERIALS, SETTING, METHODS Patients underwent either nodule excision by shaving or disc excision, or segmental resection. Randomization was performed preoperatively using sequentially numbered, opaque, sealed envelopes, and patients were informed of randomization results. The primary endpoint was the proportion of patients experiencing one of the following symptoms: constipation (1 stool/>5 consecutive days), frequent bowel movements (≥3 stools/day), anal incontinence, dysuria or bladder atony requiring self-catheterization 24 months postoperatively. Secondary endpoints were values taken from the Knowles–Eccersley–Scott-symptom questionnaire (KESS), the gastrointestinal quality of life index (GIQLI), the Wexner scale, the urinary symptom profile (USP) and the Short Form 36 Health Survey (SF36). MAIN RESULTS AND THE ROLE OF CHANCE Fifty-five patients were enrolled. Among the 27 patients in the excision arm, two were converted to segmental resection (7.4%). One patient managed by segmental resection withdrew from the study 2 years postoperatively, presuming that associated pain of other aetiologies may have jeopardized the outcomes. The 5 year-recurrence rate for excision and resection was 3.7% versus 0% (P = 1), respectively. For excision and resection, the primary endpoint was present in 44.4% versus 60.7% of patients (P = 0.29), respectively, while 55.6% versus 53.6% of patients subjectively reported normal bowel movements (P = 1). An intention-to-treat comparison of overall KESS, GIQLI, Wexner, USP and SF36 scores did not reveal significant differences between the two arms 5 years postoperatively. Statistically significant improvement was observed shortly after surgery with no further improvement or impairment recorded 1–5 years postoperatively. During the 5-year follow-up, additional surgical procedures were performed in 25.9% versus 28.6% of patients who had undergone excision or resection (P = 0.80), respectively. LIMITATIONS, REASONS FOR CAUTION The presumption of a 40% difference concerning postoperative functional outcomes in favour of nodule excision resulted in a lack of power for demonstration of the primary endpoint difference. WIDER IMPLICATIONS OF THE FINDINGS Five-year follow-up data do not show statistically significant differences between conservative and radical rectal surgery for long-term functional digestive and urinary outcomes in this specific population of women with large involvement of the rectum. STUDY FUNDING/COMPETING INTEREST(S) No specific funding was received. Patient enrolment and follow-up until 2 years postoperatively was supported by a grant from the clinical research programme for hospitals in France. The authors declare no competing interests related to this study. TRIAL REGISTRATION NUMBER This randomized study is registered with ClinicalTrials.gov, number NCT 01291576. TRIAL REGISTRATION DATE 31 January 2011. DATE OF FIRST PATIENT’S ENROLMENT 7 March 2011.

STUDY QUESTION What is the risk of progression of deep endometriotic nodules infiltrating the rectosigmoid? SUMMARY ANSWER There is a risk of progression of deep endometriotic nodules infiltrating the rectosigmoid, particularly in menstruating women. WHAT IS KNOWN ALREADY Currently, there is a lack of acceptance in the literature on the probability that deeply infiltrating rectosigmoid endometriotic nodules progress in size. STUDY DESIGN, SIZE, DURATION We conducted a monocentric case–control study between September 2016 and March 2018 at Rouen University Hospital. We enrolled 43 patients who were referred to our tertiary referral centre with deep endometriosis infiltrating the rectosigmoid, who had undergone two MRI examinations at least 12 months apart and had not undergone surgical treatment of rectosigmoid endometriosis during this interval. PARTICIPANTS/MATERIALS, SETTING, METHODS MRI images were reinterpreted by a senior radiologist with experience and expertise in endometriosis, who measured the length and thickness of deep infiltrating colorectal lesions. Intra- and inter-observer reliability were tested on 30 randomly selected cases. We defined ‘progression’ of a nodule as an increase of ≥20% in length or in thickness and ‘regression’ of a lesion as a decrease of ≥20% in length or in thickness between two MRIs. Any nodule for which the variation in length and thickness was <20% was considered as ‘stable’. Patients were divided into three groups based on evidence of progression, regression or stability of deep endometriotic nodules between their two MRI examinations. The total length of any period of amenorrhoea between the two MRI examinations, due to pregnancy, breastfeeding or hormonal treatment, was recorded. The total proportion of the time between MRIs where amenorrhoea occurred was compared between groups. MAIN RESULTS AND THE ROLE OF CHANCE Eighty-six patients underwent at least two MRIs for deep endometriosis infiltrating the sigmoid or rectum between September 2016 and March 2018. Of these, we excluded 10 patients with an interval of <12 months between MRIs, 10 patients who underwent surgery between MRIs, 17 patients for whom at least 1 MRI was considered to be of poor quality and 6 patients for whom no deep colorectal lesion was found on repeat review of either MRI. This resulted in a total of 43 patients eligible for enrolment in the final analysis. Mean time (SD) between MRIs was 38.3 (22.1) months. About 60.5% of patients demonstrated stability of their colorectal lesions between the two MRIs, 27.9% of patients met the criteria for ‘progression’ of lesions and 11.6% met the criteria for ‘regression’ of lesions. There was no significant difference in time interval between MRIs for the three groups (P = 0.76). Median duration of amenorrhoea was significantly lower in women with progression of lesions (7.5 months) when compared to those with stability of lesions (8.5 months) or regression of lesions (21 months) (P < 0.001). Median duration of amenorrhoea (expressed as percentage of total time between two MRIs) was also found to be significantly lower in the group demonstrating progression (15.1%) when compared to the group demonstrating stability (19.2%) and the group demonstrating regression (94.1%; P = 0.006). Progression of rectosigmoid nodules was observed in 34% of patients without continuous amenorrhoea, in 39% who had never had amenorrhoea and in no patients with continuous amenorrhoea. LIMITATIONS, REASONS FOR CAUTION Due to a lack of universally accepted criteria for defining the progression or regression of deep endometriotic nodules on MRI, the values used in our study may be disputed. Due to the retrospective design of the study, there may be heterogeneity of interval between MRIs, MRI techniques used, reason for amenorrhoea and duration of amenorrhoea. The mean inter-MRI interval was of short duration and varied between patients. Our findings are reported for only deep endometriosis infiltrating the rectosigmoid and cannot be extrapolated, without caution, to nodules of other locations. WIDER IMPLICATIONS OF THE FINDINGS Patients with deeply infiltrating rectosigmoid endometriotic nodules, for which surgical management has not been performed, should undergo surveillance to allow detection of growth of nodules, particularly when continuous amenorrhoea has not been achieved. This recommendation is of importance to young patients with rectosigmoid nodules who wish to conceive, in whom first line ART is planned. There is a very low risk of progression of deep endometriotic nodules infiltrating the rectosigmoid in women with amenorrhoea induced by medical therapy, lactation or pregnancy. STUDY FUNDING/COMPETING INTEREST(S) No funding was received for this study. The authors declare no competing interests related to this study.

NCT02294825.

Objective: To assess the relationship between history of surgery for endometriosis and adverse obstetrical outcomes. Design: Retrospective study using prospectively recorded data. Setting: Referral center. Patient(s): Total of 569 women with history of surgery for endometriosis, postoperative conception, and pregnancy evolution over 22 weeks of gestation. Interventions(s): Surgery for endometriosis. Main Outcome Measure(s): Small for gestational age (SGA) status of the newborn, spontaneous preterm birth (PT, before 37 weeks' gestation), and placenta previa. Results: Among 733 pregnancies included in the study, 566 deliveries were recorded (77.2%), of which 535 were singleton (72.9% of pregnancies) and 31 twins (4.2%). SGA was observed in 81 of 535 (15.1%) singleton pregnancies and in 9 of 31 (29%) twin pregnancies. PT occurred in 53 of 535 (9.9%) singleton pregnancies and in 19 of 31 (61.2%) twin pregnancies. The number of singleton and multiple pregnancies complicated by placenta previa were, respectively, 9 of 535 (1.7%) and 0 of 31. The independent factor found to relate to SGA was the absence of endometriomas; conception with the use of assisted reproductive technologies (ART) only tended toward statistical significance. Independent factors found to increase risk of PT were conception with the use of ART, body mass index >30 kg/m 2 , and surgery of deep endometriosis infiltrating the rectum and the bladder. Independent factors associated with placenta previa were conception with the use of ART and history of stage III or IV endometriosis. Conclusion(s):The risk of SGA and PT is increased in women with a history of surgery for endometriosis, and a high rate of conception with the use of ART may jeopardize outcomes. (Fertil Steril Ò 2020;113:996-1004. Ó2020 by American Society for Reproductive Medicine.) El resumen está disponible en Español al final del artículo.

No abstract