Bowel anastomosis leakage following endometriosis surgery: an evidence based analysis of risk factors and prevention techniques.

A. Vigueras Smith, Andres Vigueras Smith, Vigueras Smith A, Rok Šumak, Sumak R, R Sumak, R. Cabrera, Cabrera R, Ramiro Cabrera, Kondo W, William Kondo, W Kondo, H. Ferreira, Hélder Ferreira
Facts, views & vision in ObGyn · 2020 · vol. 12(3) , pp. 207–225 · PMID:33123696 · PMC7580259 · W3096912084
article OA: gold CC0 ⤵ 19 in-corpus citations
📄 Open PDF Full text JSON View on OpenAlex View on PubMed
⚙ AI-generated summary by gemini-2.5-flash-lite, 2026-06-08 ⓘ

This review analyzed risk factors and prevention techniques for anastomotic leakage after endometriosis bowel surgery, identifying modifiable factors and the role of protective stomas.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

⚙ AI-generated deep summary by claude@2026-06, 2026-06-12 · read from full text ⓘ

This paper is an evidence-based review analyzing risk factors and prevention techniques for bowel anastomotic leakage after surgical treatment of bowel endometriosis, using a structured literature search of PubMed and Google Scholar for English publications. It describes how anastomotic leakage is defined and clinically classified, summarizes patient, intra-operative, and surgical-technique risk factors cited across studies, and discusses definitions of major complications as well as the role of protective stomas in prevention strategies. The review notes that segmental resection is associated with higher leakage and other postoperative morbidity than shaving or discoidal excision, while also emphasizing that leakage rates vary substantially depending on definitions, anastomosis level, and follow-up length; it also highlights limitations such as lack of universal operative reporting standards for deep endometriosis classification. This paper is centrally about endometriosis—specifically bowel endometriosis surgery and anastomotic leakage risk and prevention.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

BACKGROUND: Deep endometriosis most commonly involves the rectosigmoid junction and its management often requires a colorectal resection. Anastomotic leakage is a severe complication after resection and affects 1-6% of the cases. OBJECTIVE: To evaluate the risk factors related to anastomotic leakage following endometriosis sur-gery, its prevention techniques and the role of protective stomas. METHODS: A comprehensive literature review was carried out for English-language publications in Pubmed and Google Scholar. We included all studies including the following MeSH terms and key words: Anastomotic leakage AND bowel surgery OR Endometriosis OR Colorectal surgery OR Bowel endometriosis. Two authors independently made a selection and analysed relevant abstracts according to the aim of this review. RESULTS: Risk factors and preventive measures were categorised considering the patient condition, the intra- operative setting and the surgical procedure itself. Level I and II recommendations include modifiable risk factors such as the use of stapled or handsewn anastomosis; intra-operative air leak test to check the integrity of the anastomosis; systematic use of pelvic and trans-anal drainage; application of protective or ghost ileostomy in low rectal resections; vaginal closure before the bowel resection; use of oral antibiotics the day before surgery and performing partial mesorectal resection near the bowel wall. Diverting stomas may decrease the morbidity and the clinical consequences of leakage over 65% of low rectal resections but may cause significant adverse effects. CONCLUSION: Evidence-based protective actions are crucial to reduce clinical consequences of anastomotic leakage and to minimise the use of protective stomas in endometriosis surgery.
Full text 51,927 characters · extracted from pmc-nxml · 8 sections · click to expand

A

Appropriate fluid administration should be part of intra-operative and post-operative care since there is evi-dence that both over-hydration and restriction are directly associated with a high risk of AL after colecto-my/gastrointestinal surgery ( Schnuriger et al., 2011 ). The National Institute for Health and Care Excellence in 2011 recommended the use of goal-directed therapy to reduce post-operative complications. Nevertheless, there are no studies that prove it reduces AL rates ( Futier et al., 2010 ). Patients with prolonged diastolic blood pressure drops have a 3-fold higher risk of AL ( Post et al. 2012 ; Choudhuri et al., 2013 ). Similarly, patients who have post-operative treatment with vasopressors have a three to four-fold increase in AL rate, directly related to the time exposure to these drugs ( Zakrison et al., 2007 ). Higher intra-operative blood loss is associated with increased risk of AL ( Defazio et al., 2014 ) by reducing colonic blood flow at the anastomotic level, leading to impaired wound healing and tissue necrosis ( Irwin et al. 1990 ). Furthermore, blood transfusion in the perioperative period also increases the risk between 2 and 10-fold ( Krarup et al., 2012 ). Moreover, the risk of blood-borne infections associated with transfusions is another reason for the use of a restrictive rather than liberal transfusion protocol ( Burden et al., 2012 ). Inadequate perfusion and partial O 2 pressure are responsible for impaired wound healing. Normal pre- operative levels of haemoglobin must be checked and corrected when needed. A retrospective analysis of over 1200 major abdominal surgeries found that haemoglobin levels under 8g/dl independently increase the rate of AL 1.91 times ( Choudhuri et al., 2013 ). Duration of surgery is positively correlated with postoperative morbidity in both major and minor proce-dures ( Scott 1982 ). The retrospective analysis of Silva-Velazco et al. ( 2016 ) found an AL increase of 3% for every 30 minutes of surgical time. The threshold for an increased chance of leakage was between 220 to 300 minutes ( Huh et al., 2010 ). It has been suggested that poor perfusion of anastomotic site as demonstrated by indocyanine green (ICG) may increase the risk of AL. Apart from the use of fluorescence-guided surgery for the detection of superficial endometriosis, intra- operative ICG assessment of the bowel wall (after shaving procedure) or the anastomotic line (after bowel DIE resection) vascularisation is a potential tool that might be helpful in confirming com- plete macroscopic resection of the disease and reducing the rate of bowel perforations. Therefore, the two theoretical benefits are choosing the adequate transecting line and evaluating the rectal vascularisation after mechanical anastomosis ( Seracchioli et al., 2018 ). ICG identifies the vascularisation of a specific anatomical structure or tissue, showing vascular anatomy and local perfusion ( Alander et al., 2012 ). After direct intrave-nous administration (0.25mg/Kg), a fast and objective evaluation of neoanastomosis vascularisation could be performed. When vascularisation is normal, the ICG turns fluorescent (dark green) once excited with a light in the NIR spectrum ( De Neef et al., 2018 ). Although some prospective studies show less leakage incidence when this technique is applied (compared to overall rate) ( Jafari et al., 2015 ; Kawada et al., 2014 ; Blanco-Colino and Espin- Basany, 2018 ), evidence quality is still poor and scant to recommend their use routinely in bowel DIE surgical cases and must remain as a part of analytical, experimental protocols to demonstrate their real benefit. Currently, there is an ongoing clinical trial conducted by the group of Clermont Ferrand assessing the potential role of ICG in reducing fistula rates after rectal shaving surgery.

B

Whether to perform a bowel stoma or not depends on the surgical team preferences, but must be guided by some specific factors: The site and height of anastomosis are crucial. It is well accepted that the serosal layer has a critical role in anastomotic healing. Since the lower rectum is lacking in this layer, a higher risk of leakage is expected at this level ( Moran et al., 2017 ). Furthermore, lower lesions usually require extensive para-rectal dissection which can harm vascular structures, compromising the final vascularisation of the bowel at the anastomotic line, increasing the risk of leakage and temporary defunctioning stoma ( Nezhat et al., 2018 ). The level of the anastomosis can be classified into three types according to their distance from the anal verge (AV) ( Figure 1 ): Bowel anastomotic levels.Three consecutive images representing the three levels of anastomotic lines according to their distance from the anal verge. A: Medium or High anastomosis (> 8 cm from AV) ; B: Low anastomosis (5 to 8 cm from AV) ; C: MUltra-Low anastomosis (< 5 cm from AV). White line: Anastomotic line. Yellow line: 8 cm form AV. Blue line: 12 cm from AV. High/Medium: Equal or more than 8 cm. Low: Less than 8 cm but more than 5 cm. Ultra-low: 5 cm or lower. Even though initial studies did not show a significant relationship between the rate of fistulas and level of anastomosis, probably due to the bias of using protective ileostomy in low and ultra-low anastomosis ( Mereu et al., 2007 ), recent evidence has consistently shown that the rate is significantly higher in the left side of the colon, and specifically in those performed within 10 cm from AV ( Bakker et al., 2014 ; Trencheva et al., 2013 ; Abrao et al., 2015 ). Furthermore, several prospective studies have shown that the lower the anastomosis; the higher the risk of leakage ( Park et al., 2013 ). The leakage rate is up to 3.4 times higher for tumours located less than 7 cm from the AV ( Hamabe et al., 2018 ) and ten times higher for those located under 5 cm of the AV ( Choi et al., 2010 ). Total meso-rectal excision (TME) is the standard treatment for locally advanced rectal cancer, reducing the risk of recurrence and improving global prognosis ( Bianchi et al., 2014 ), while benign conditions need just partial meso-rectal excisions near the bowel. When rates of leakage in bowel endometriosis surgery are com-pared to rectal cancer, percentages are consistently higher in the latter, reaching up to 17% ( Sartori et al., 2011 ). A meta-analysis reports a leakage rate of post-TME ranging from 5.4% to 5.8% ( Hua et al., 2014 ). This is explained in part by the fact that endometriosis is a benign disease affecting healthy young women without major comorbidities. Additionally, and although surgical techniques for segmental resection vary widely among different teams, DIE bowel resection could encompass a “nerve-vessel sparing segmental resection”, where mesorectum resection is limited to the macroscopic DIE infiltration area and cutting of the inferior mesenteric vessels is avoided. This strategy results in a tubular fashion dissection which spares all the fatty tissue, hypogastric nerve plexus and vessels lateral to the bowel segment resected. Despite the absence of solid evidence about the benefits of this type of segmental resection compared to others, the theoretical im-provement of the anastomosis vascularisation could enhance the bowel healing process and reduce the risk of leak and micro-leaks ( Hudelist et al., 2018 ) ( Figure 2 ). Meso-rectal resection for bowel DIE resection. 9 consecutive images showing the technique for the meso-rectal dissection in the endometriotic scenario. Since this is a benign disease, dissection must be performed as close as possible to the bowel in order to obtain maximum preservation of irrigation and innervation in between the white and the orange lines. Dissection must be performed up to 2 cm away from the DIE nodule edges. Currently, an air leak test is one of the most frequent intra-operative tests for evaluation of mechanical anastomosis competence, as well as diagnosis and treatment of occult disruptions ( Umanskiy and Hyman, 2019 ). A positive test leads to the performing of further procedures, such as reinforcement stitches or protective ileostomy to avoid anastomotic complications ( Nachiappan et al., 2014 ) . Even though it is widely used, a unique and clear definition and standardisation of the test does not exist. Many different techniques in terms of in- sufflation methods (syringe, catheters, endoscope, sigmoidoscope, rectoscope, proctoscope), solutions (air, saline solution) and volumes (60 mL to 400 mL) have been described ( Wu et al., 2016 ). However, clinicians must consider that colorectal anastomosis support can stand pressures around 70 to 184 mmHg; therefore, a volume near 400ml must be injected carefully under barometric intra-luminal measurement to avoid damage of anastomosis ( Schwab et al., 2002 ). A recent meta-analysis of 20 studies with 5283 patients evaluate their role in the prevention of AL. Although the studies are biased, the rate of leakage after performing the air- leak test is consistently lower than those without testing (OR: 0.61), but not statistically significant (p=0.15). Nevertheless, among all the patients with the air-leak test performed, those with a positive test have significantly higher chances of presenting leakage than those with a negative one (4.2% vs 11.4%) ( Wu et al. 2016 ). Surgeons must be aware that this test detects only mechanical disruption of the anastomosis, leaving out other pathologic mechanisms of AL such as healing disturbances or infection. In conclusion, the systematic use of this test does not significantly reduce the rate of leakage. However, its use is highly recommended since, in case of a positive result, the risk of leakage rises dramatically, and further protective procedures must be considered (thorough revision of anastomosis, suture reinforcement, redoing anastomosis, oversewing and re-testing) ( Figure 3 ). Positive air-leak test. 4 consecutive images demonstrating a positive air-leak test. After bowel occlusion with laparoscopic atraumatic grasper distal to the bowel anastomosis, 60-400 cc of air is directly inserted trans-anally to distend the rectum. The blue arrows show the bubbles coming from the mechanical anastomosis dysfunction. Studies in colorectal cancer suggest that the nodule diameter may be a predicting factor for AL. With the in-crease in lesion size, intra-pelvic manipulation becomes restricted and rectal transection is more challenging, starting from 3 to 4 cm diameter ( Eberl et al., 2008 ; Zhu 2010 ). Thus, when tumours are bigger than 5 cm, a 4-fold higher risk of leakage is seen ( Kawada et al., 2014 ). In endometriosis, the nodule size is closely related to the type of surgical bowel treatment, and therefore, the expected risk of leakage. According to Abo et al., ( 2018 ) shaving techniques are usually performed in nodules up to 3 cm. In comparison, discoidal or segmen-tal resections are conducted on nodules more prominent than 3 cm ( Ferrero et al., 2009 ; Ghezzi et al., 2008 ). Since segmental resection appears to have higher rates of AL than nodulectomies, the bowel DIE nodule size seems to be an indirect risk factor of AL when over 3 cm. As previously noted, AL events can occur in any of these surgical modalities, but predominantly in the low/ultra-low segmental bowel resection (See anastomotic leak in bowel endometriosis treatment). Concerning RVF, concomitant vaginal resection appears to be a predisposing factor. According to Abo et al ( 2018 ) RVF was present in 3.8% of 364 bowel resections (without any differences between the type of bowel resection) with 50% of the cases having concomitant vaginal resection. It is always recommended to perform vaginal closure before bowel resection, and in cases of low bowel anastomosis, closing the vaginal opening and making interposition of the omental flap should usually be considered ( Minelli et al., 2009 ; Ruffo et al., 2010 ) ( Figure 4 ). However, for isolated anastomotic leakage, vaginal opening does not appear to be a primary factor. Concomitant bowel and vaginal resection . Images showing the anatomical relationship between vaginal closure and bowel anastomotic line. Upper set without annotation, lower set with annotation. Closer location increases the risk of rectovaginal fistula. A:Vaginal cuff closure after total laparoscopic hysterectomy and bowel anastomosis ; B: Vaginal closure after resection of vaginal DIE nodule plus bowel anastomosis. Surgeon’s experience is still one of the significant factors in deciding whether to perform a stoma or not. Colo-rectal surgery is an advanced and difficult procedure which expose patients to a relatively high risk of severe complications. Thus, it should be performed in expert centres to reduce the instances of such complications . The French group of Bendifallah et al. ( 2017 ) analysed the relationship between case volume (rectum and sigmoid colon DIE) and incidence of complications, establishing an optimal cut-off value of 20 cases a year per centre and 7-13 procedures a year per surgeon for significant reduction of grade III and IV complication rates. It is clear that this type of colorectal surgery is certainly not an innocuous procedure and an evidence-based approach in the decision making should be adapted.

C

General surgical principles and technical points for bowel closure must be followed and maintained to avoid anastomotic leakage ( Nezhat et al., 2018 ; De Cicco et al., 2011 ) ( Table V ). Surgical principles and technical points for bowel anastomosis construction. Excluding ileocolic and oesophago-gastric anastomosis evaluations, three significant meta- analyses have been published in recent years. Naumann et al. ( 2015 ) evaluated the risk of leak, abscess, and fistula after bowel anastomosis in the emergency setting. In seven studies (5 retrospectives, one RCT, and one prospective non-randomised) and more than 1,205 anastomoses, there were no significant differences between the handsewn and the stapler technique in the risk of AL (OR:1.00). Furthermore, Neutzling et al. ( 2012 ) presented their results specifically for colorectal anastomosis, including 9 RCTs and 1233 patients. No significant differences were seen in both clinical (6.3% vs. 71%) and radiological (7.8% vs. 7.2%) anastomotic dehiscence between the arms, regardless of the level of the anastomosis. Finally, a systematic review of eleven systematic reviews concerning handsewn versus stapled anastomosis reported no evidence of the superiority of any specific technique ( Hemming et al., 2013 ). The conclusion is that the decision of the type of anastomosis is likely a matter of surgeon’s preference and experience, as techniques appear to be substantially equivalent concerning leak rate. However, in the case of stapler use, recent retrospective evidence supports the fact that the number of cartridges fired is a relevant factor for AL occurrence, significantly increasing when three or more cartridges are used ( Braunschmid et al., 2017 ). The use of bio-absorbable staple-line reinforcing material is appealing to some. Although studies have shown that these reinforcements are safe and there have been several RCTs on the subject, to date, there have been no compelling studies which have demonstrated a decrease in leakage rates when they are used ( Placer et al., 2014 ). Reinforcement sutures are typically placed around the anastomosis, but intra-luminal reinforcement has also been carried out ( Kim et al., 2015 ). To date, there is no compelling evidence indicating that suture rein-forcement reduces leakage, yet these techniques may improve a surgeon’s confidence regarding the strength of one’s anastomosis ( Figure 5 ). Suture reinforcement of anastomosis. 5 consecutive images showing the manual suture reinforcement after stapler anastomosis. Using delayed absorbable or nonabsorbable mono-filament sutures, 1 to 5 intra-corporeal stitches are performed to secure the anastomosis strength against AL during the first days PO. A,B,C and D: Reinforcement stitch with a triple-double blocking sequence; E: Final view after 5 stitches. To date, one RCT has failed to show a decrease in leakage rate with the use of fibrin glue ( Silecchia et al., 2008 ). At the same time, there have been several case series showing low meagre anastomotic leakage rates with the use of these glues ( Lee et al., 2004 ). Fibrin glue application over the stapled anastomosis was found not to be significantly associated with leakage following laparoscopic rectal cancer surgery without stool diversion ( Huh et al., 2010 ). Decades ago, several materials, such as silk, linen, catgut, polyglactin 910, and nylon, were commonly used for colorectal anastomosis. Nowadays, it is evident that absorbable sutures are safe, leaving no channel for luminal microbial migration once absorbed. Most gastrointestinal anastomoses, including colorectal, are constructed with polydioxanone sutures. Absorbable sutures compared with non- absorbable or slowly absorbable sutures cause more tissue reaction and dissolve too rapidly, reducing the anastomotic strength ( Van Winkle et al., 1975 ). Multifilament, compared to monofilament sutures, cause more tissue damage and easier adherence of material within the interstices of the multifilament ( Deveney and Way 1977 ), creating a risk of infection ( Durdey and Bucknall 1984 ). Polydioxanone thread possesses all characteristics considered important; monofilament, little tissue reaction, slowly absorbable with long preservation of strength and low bacterial adherence risk. Based on experimental studies, non-absorbable, or slowly absorbable monofilament sutures seem to be the first choice for colorectal anastomosis ( Slieker et al., 2013 ). Continuous suture provides a tighter seal than an interrupted one. The main fear is that if this suture breaks, the entire suture line could open. RCTs investigating interrupted and continuous sutures for colorectal anastomosis are lacking. Only one small, non-randomised, comparative clinical study found no significant differences ( Houdart, 1994 ). Two clinical and experimental studies have not concluded that one technique is superior to the other and a high level of evidence is lacking; however, from a technical and time-consuming point of view, a continuous suture is preferable over interrupted sutures ( Slieker et al., 2013 ). Lembert described the construction of intestinal anastomoses in dogs using suture bites with 5-mm distance to the cut edge nearly two centuries ago ( Breschet, 1828 ). An RCT allocated patients to have bowel sutures placed either 5 or 10 mm from the cut edges, with no significant differences in leakage rates ( Greenall et al., 1979 ). Lembert reported a distance of approximately 1 cm between sutures ( Breschet, 1828 ). Neither comparative clinical studies nor cohort studies were found. Animal experiments indicate that small distance between sutures (1.5 mm) improves apposition compared with a more considerable distance (2.5 mm) ( Waninger et al., 1992 ). However, due to lack of clinical studies on this topic, there is no precise conclusion in the literature regarding this issue. In routine clinical practice, two undefined schools of thought seem to exist. The first believes that sutures should be tightened to prevent dehiscence of the anastomosis, and the second considers that sutures should be applied more loosely, allowing maximal perfusion of the cut edges. The bowel is highly supplied with blood and may become oedematous and hardened when tight stitches are used. Only one rat study investigated this, with moderate tension giving the best histological and micro-angiographic results ( Waninger et al., 1992 ). Whether pressure on knots could influence the incidence of AL in a clinical setting has not been investigated, and therefore, the optimal tension on the thread or the knot is unknown. The main strength of the gastrointestinal tract is in the muscularis and submucosa. Thus, effective closure involving at least these two layers is needed. Additionally, avoiding entering into the mucosa could help to prevent leakage. Cohort studies report low rates of AL for both serosa-submucosal and full-thickness suture types ( Leslie and Steele 2003 ). We can conclude that both serosa-submucosal and full-thickness sutures seem to provide low rates of leakage. Both everting and inverting (as well as end-to-end) techniques have been performed, but both have drawbacks. An RCT showed a 5-fold increased incidence of AL in patients receiving an everting compared with those receiving an inverting suture. Hence, there seems to be an advantage of inverting over everting colorectal anastomosis ( Goligher et al., 1970 ). The classic technique is based on a double-layer inverting anastomotic method. One RCT ( Everett, 1975 ) matched the inclusion criteria, showing no significant differences in AL between single- and double-layer colorectal anastomosis in 92 patients. However, in the subgroup analysis of low colorectal anastomosis, the incidence of AL in those with the double-layer technique was significantly higher. Single-layer anastomosis has the additional advantage of being less time consuming to perform and is less costly ( Burch et al., 2000 ); hence the published literature favourings single layer anastomosis. A pelvic drain may prevent haematomas or seromas that constitute a medium for bacterial infection which can involve the anastomosis, thereby causing dehiscence. Moreover, a pelvic drain may help control leaks if they do occur, leading to a less severe clinical course ( Qu et al., 2015 ). However, routine prophylactic use is debatable ( Emile and Abd El-Hamed, 2017 ). Two retrospective studies found pelvic drainage associated with low-er rates of AL, though without reaching statistical significance ( Akiyoshi et al., 2011 ). However, the lack of pelvic drain was found to be independently predictive of leakage at multivariate analysis. Two main systematic reviews of RCTs have been published. A Cochrane meta-analysis included 6 RCTs and over 1100 patients, and reported no reduction of leakage after prophylactic use of drainage ( Jesus et al., 2004 ). A recent meta-analysis including RCTs and retrospective studies indicate a significant decrease (49%) in AL when drains are used in rectal infra- peritoneal bowel resections. However, when RCTs were analysed alone, this reduction did not persist ( Rondelli et al., 2014 ). The evidence is contradictory. One RCT showed no reduction in AL when trans-anal stents were used in 194 patients subjected to anterior rectal resection ( Bulow et al., 2006 ). Nevertheless, recent prospective and retrospective studies show that the use of a trans- anal drainage tube significantly reduces AL and other unfavourable effects of post-operative diarrhoea ( Tanaka et al., 2017 ; Nishigori et al., 2014 ; Zhao et al., 2013 ) by lowering endo-luminal pressure (gas and fluids) at the anastomotic line in the early period. Despite this, their systematic use remains questionable and is a matter of preference of the surgeon. Theoretically, the interposition of omental graft on a vascular pedicle, covering the area of the anastomosis, offers two main benefits ( Wiggins et al., 2015 ; Hayari et al., 2004 ): Re-enforcement of the anastomotic line during the first post-operative days (when there is a higher risk of leakage,) acting as a biologically viable plug which can seal microscopic leaks. Increased angiogenesis and neo-vascularisation at the anastomotic site by providing vascular endo- thelial growth factor, promoting microvascular anastomosis between the omentum and the bowel wall ( Adams et al., 1992 ). Limitation of their use is mainly due to the fear of omental necrosis and increased risk of recurrence in the cancer setting ( van Garderen et al., 1991 ) ( Figure 6 ). Omentoplasty. Sequence of 6 images showing the dissection of the major oomentum in order to create the oomental flap. A and B: Dissection line in order to create the flap ; C: Omental flap is done and ready to interpose. D and E: Fixation of the flap into the vaginal wall. F: Final position of the flap between the bowel and vaginal suture lines. In the prospective series of Ozben et al. ( 2016 ), no reduction in AL or surgical site infection rates was seen when the omental flap was used after rectal cancer surgery. Moreover, the systematic review published by Wiggins et al. ( 2015 ) including three RCTs and 943 colorectal anastomoses showed no significant differences in the rate of leakage (5% vs 8.4%), in-hospital mortality (4.2% vs 4.1%) and anastomotic stricture (1.9% vs 5%) between patients with and without omentoplasty.

Diverting

For low and ultra-low colorectal anastomosis at high risk of developing fistula and leakage, the use of temporary protective ileostomy is usually recommended in order to prevent these complications. However, it is asso-ciated with stoma-related risks, such as hernia, retraction, stenosis, sinus formation, dehydration, prolapse and necrosis ( Bertocchi et al., 2019 ). Randomised controlled trials (RCTs) and meta-analyses have demonstrated that the use of defunctioning stoma in low colorectal anastomosis may reduce the morbidity and clinical consequences of leakage (up to 68% reduction of clinically symptomatic AL and 73% fewer re-operations). However, they do not seem to reduce the likelihood of occurrence itself or the mortality rates ( Huser et al., 2008 ; Matthiessen et al., 2007 ; Boyce et al., 2017 ) Considering that endometriosis is a benign disease in young and healthy people, adequate case selection and patient counselling is crucial ( Minelli et al., 2009 ). Treatment must always be tailored according to the patient’s disease, desires and expectations.

Prevention

Up to now many techniques have been proposed to prevent or reduce the rate of leakage, but many of them do not have a corresponding evidence-based background ( Table IV ). Classical preventive techniques for anastomotic leakage. After evaluating the literature, we were able to classify the general and specific techniques for the prevention of leakage. Following the Canadian Task Force Levels of Evidence, we present the classification of these techniques based on the level of evidence ( Burns et al., 2011 ).

Conclusions

Anastomosis leakage is the leak of intra-luminal content from the suture line between 2 hollow viscera. It is a heterogeneous pathology by definition, but severe in its nature, causing severe morbidity, re-admissions, re-operations, a more extended hospital stay, lower quality of life and up to 15% mortality. All types of bowel endometriosis surgical treatment carry a risk of leakage and RVF, and even when these complications are predominantly seen in the segmental resection group, with an overall risk of 1.7% (0-6%) and 3.9% (0-10.3%), evidence is predominantly retrospective and AL definitions are heterogeneous among the different studies. Careful monitoring is essential after any of these procedures. Control of modifiable risk factors, together with strictly following surgical principles such as avoidance of anastomotic tension, tissue ischemia and overlapping remain paramount in general prevention. Results of this evidence-based analysis lead us to recommend the following peri-operative modifiable measures; the use of either stapler or handsewn (single layer closure) anastomosis construction; intra-operative use of air leak test to check the mechanical integrity of anastomotic line; systematic use of pelvic (in infra-peritoneal anastomosis) and trans-anal drainage; application of further preventive interventions (protective or ghost ileostomy) when the nodule is located under 8 cm from the anal verge and in high-risk patients; closure of the vagina before performing bowel resection (when colpotomy is required); systematic use of non-absorbable oral antibiotics one day before surgery and performing partial mesorectal resection near the bowel wall ( Table VI ). Summary of recommendations for main risk factors and preventive techniques of anastomotic leakage. Temporary defunctioning stomas may decrease the morbidity and clinical consequences of the leakage in over 65% of low colorectal anastomosis, but are associated with significant side effects that must be balanced against the risk of leakage. The treatment, considering the benign nature of endometriosis, must always be tailored according to the patient’s disease, desires and expectations, with comprehensive case-by-case selection and patient counselling. Finally, readers must be aware that the majority of the studies on this topic come from colorectal surgeons’ experience. This is relevant since colorectal oncology patients usually have a different demographic to the young, healthy patients in the endometriosis setting. However, the large endometriosis series, including more than 5500 segmental resections, support the conclusions presented here.

Introduction

Bowel endometriosis is defined as the presence of endometrial-like glands and stroma infiltrating the bowel wall and affects 5% to 12% of patients with deep infiltrating endometriosis (DIE) ( Abo et al., 2018 ). When surgery is indicated, amongst other challenges, anastomotic leakage (AL) appears as a major life-threatening complication affecting around 1-2% of segmental resections, significantly increasing morbidity, mortality and reoperation rate ( Richards et al., 2012 ). Due to the high number of bowel endometriosis cases requiring surgery in current practice, it is necessary to have a thorough knowledge of AL presentation and its preventive methods in order to reduce their numbers to a minimum. The primary objective of this review is to analyse the currently available information relating to AL risk factors and preventive techniques following surgical treatment of bowel endometriosis, and the role of systematic use of protective stomas.

Materials|Methods

A comprehensive review of the literature was carried out for English publications in Pubmed and Google Scholar relating to bowel anastomotic leakage following endometriosis surgery. We included all studies found under the search of following MeSH and the keywords terms: Anastomotic leakage AND Bowel surgery OR Endometriosis OR Colorectal surgery OR Bowel endometriosis. Initially, a structured investigation question was created using the PICO strategy as shown in Table I . Structured investigation strategy used in this review. AL is defined as the leakage of luminal contents from the surgical join between two hollow viscera ( Peel and Taylor 1991 ). The luminal contents may emerge either through the wound or at the drain site, or they may collect near the anastomosis causing fever, abscess, septicaemia, metabolic disturbance and/or multiple organ failure. The asymptomatic leak of luminal content from the anastomotic site into an adjacent localised area detected by imaging exams should be recorded as a sub-clinical leak ( Bruce et al., 2001 ). Diagnosis frequently encompasses clinical, biochemical and imaging exams. Leakage may present itself as pain, fever or feculent discharge from a drain. Clinical signs may include tachycardia, abdominal tenderness and signs of peritonitis ( Table II ). Laboratory findings usually show an elevated white cell count and an increase in acute-phase biochemical markers (pro-calcitonin, C-reactive protein) during the first 3 to 5 days postoperatively. Radiological investigations (CT scan, water-soluble contrast media enema, contrast media MRI) show signs of a fluid collection which may contain gas or contrast in proximity to the anastomosis ( Kanellos et al., 2004 ). Clinical symptoms and radiologic signs of anastomotic leak. Bruce et al. ( 2001 ) proposed three types of lower gastrointestinal AL based on signs, symptoms and severity, independent of the level of the colorectal anastomosis. Radiological: No clinical signs. Clinical-Minor: Leakage of luminal contents through the drain/wound - local inflammation - fever (>38°C) - tachycardia - leukocytosis (over 10.000/ litre) - faecal purulent discharge from drain/wound (abscess). Clinical-Major: Same as minor plus severe disruption of the anastomosis. There are known risk factors for leakage inherent to the patient condition, intra-operative setting and surgical technique ( Phillips 2016 ). A summary of these are presented in Table III ( Law et al., 2000 ; Lipska et al., 2006 ). Risk factors for bowel anastomotic leakage. A thorough analysis of the principal risk factors will be presented later (See Prevention of Risk Factors).

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy is the canonical version.

My notes (saved in your browser only)

⚙ Ask this paper AI returns verbatim quotes from the full text · source: pmc-nxml ⓘ

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Condition tags

endometriosisbowel_endometriosis

Citation neighborhood

Papers in the corpus that this work cites (lower rings, blue) and that cite this one (upper rings, green). Dot size scales with the paper's in-corpus citation count — bigger dot = more influential within the endo/adeno field. Click a dot to open that paper. [ expand to 2 hops ] — adds papers reached through this work's immediate citers/citees. Heavier; up to 60 extra dots.

References (100)

Cited by (19)

Source provenance

europepmc
last seen: 2026-10-05T06:18:27.067365+00:00
openalex
last seen: 2026-06-10T17:14:06.276822+00:00
pubmed
last seen: 2026-05-13T22:21:36.268089+00:00
License: CC0 · commercial use OK