International Journal of Gynecology and Obstetrics 125 (2014) 266–269

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo

CLINICAL ARTICLE

Long-term follow-up and recurrence rate after mesorectum-sparing bowel resection among women with rectovaginal endometriosis Mandy Mangler ⁎, Julia Herbstleb, Sylvia Mechsner, Julia Bartley, Achim Schneider, Christhardt Köhler Department of Gynecology, Charité Campus Mitte, Berlin, Germany

a r t i c l e

i n f o

Article history: Received 6 August 2013 Received in revised form 3 December 2013 Accepted 3 March 2014 Keywords: Bowel endometriosis Histopathology in endometriosis Nerve-sparing operative therapy in endometriosis Rectovaginal endometriosis

a b s t r a c t Objective: To demonstrate the quality of a combined vaginal–abdominal surgical approach to rectovaginal endometriosis by analyzing long-term outcome and recurrence rates. Methods: In a prospective cohort study in Berlin, Germany, women with endometriosis of the rectovaginal septum were enrolled between September 2004 and December 2012. Bowel infiltration was verified intraoperatively and treated by a nerve-sparing, mesenterypreserving vaginal–abdominal operative approach. Operative results were evaluated by assessing short- and long-term complications and recurrence rates. Results: During the study period, 110 women underwent surgery. For 71 (64.5%) patients, bowel infiltration was confirmed intraoperatively. Overall, 15% of the patients had perior postoperative complications. No long-term complications occurred. After a median follow-up of 64 months, no recurrence in the rectovaginal septum was observed among the study patients. The recurrence of pelvic endometriosis was 15%. Conclusion: The surgical nerve-sparing approach to rectovaginal endometriosis was confirmed to facilitate precise diagnosis and treatment with minimal morbidity and a long-term complication rate of 0%. © 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Deep infiltrating endometriosis of the rectovaginal septum is difficult to assess. Evaluating the rectovaginal septum by inspection and palpation requires high expertise [1,2] (Fig. 1). Therefore, rectovaginal endometriosis is often misdiagnosed, and patients have painful symptoms for a considerable time until definitive treatment is undertaken. Due to its misdiagnosis, rectovaginal endometriosis can lead in some cases to life-threatening complications such as ileus [3]. The alternative—potential surgical overtreatment, especially among asymptomatic patients with endometriotic rectovaginal disease—is also a matter of debate [4]. There is neither a guideline nor a standardized recommendation concerning the surgical approach to rectovaginal endometriosis. Therapeutic strategies (e.g. shaving, disc resection, and segmental resection) are as inconsistent [4–9] as the definition of this specific form of endometriosis itself. The revised version of the Enzian score might help to quantify and qualify the disease further [10]. We previously described a nerve- and mesorectum-sparing standardized vaginal–abdominal operative approach to rectovaginal endometriosis [11]. This approach enables the surgeon to distinguish between

⁎ Corresponding author at: Charitéplatz 1, Charité Campus Mitte, Department of Gynecology, 10117 Berlin, Germany. Tel.: +49 177 2359289; fax: +49 30450564917. E-mail address: [email protected] (M. Mangler).

patients with infiltrating bowel disease and those with non-infiltrating rectovaginal endometriosis. The significance of microscopically affected resection margins among patients with rectovaginal endometriosis has not been sufficiently evaluated. It has been reported that recurrence might not be related to microscopic disease in resection margins [12]. Rectovaginal endometriosis is often multifocal and, if infiltrative, almost always involves the rectal wall as far as the level of the muscularis propria [13]. The aim of the present study was to analyze prospectively the results of the vaginal–abdominal approach to rectovaginal endometriosis with respect to the operative results, short- and long-term complications, and recurrence rate. 2. Materials and methods The present study enrolled consecutive patients with symptomatic rectovaginal endometriosis who underwent surgery at the Department of Gynecology of Charité University Berlin, Germany, between September 1, 2004, and December 31, 2012. All data were recorded prospectively in an observational study. Ethical approval for the study was obtained from the institutional review board and local ethical committee. Informed consent was obtained from all women. Patients were included in the study if rectovaginal endometriosis was either suspected by clinical examination or known through previous operations or diagnostic tools, such as vaginal sonography, magnetic resonance imaging (MRI), rectal endosonography, or colonoscopy.

http://dx.doi.org/10.1016/j.ijgo.2013.12.010 0020-7292/© 2014 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

M. Mangler et al. / International Journal of Gynecology and Obstetrics 125 (2014) 266–269

267

Fig. 1. Rectovaginal endometriosis in the rectovaginal septum on the ventral surface of the bowel. Fig. 2. The rectosigmoid can be mobilized and elevated while sparing both inferior hypogastric nerves. Arrows indicate direction of mobilization.

Before the operation, patients had a bimanual rectovaginal examination, vaginal ultrasound, MRI, endosonography of the rectum, and blood testing. During the rectovaginal septum examination, palpable nodules of the septum were evaluated including their consistence and extension. The elasticity of the septum and mucosal mobility of the rectal wall were recorded. The patients underwent a standardized surgical technique, consisting of a vaginal component, laparoscopic component, and only if needed a bowel anastomotic component. The technique has been previously described in detail [11]. In brief, the operation always starts vaginally. During the vaginal part of the operation, the surgeon can confirm or exclude endometriotic involvement of the bowel by examining the nodule directly in the rectovaginal septum. After the vaginal incision with simultaneous rectal palpation, it can be verified whether or not the endometriotic lesion infiltrates the rectal wall. If bowel-infiltrating disease is present, the endometriosis may be further mobilized and centralized; as a result, it will be necessary to conduct bowel resection. The extent of the disease guides the surgical preparation; once normal tissue is reached, the excision can be made close to the bowel wall. If the lesion lies superficially in the rectovaginal septum and is not infiltrating the intestine, the endometriotic structure can be completely removed transvaginally. Hence, the vaginal part of the operation serves to discriminate exactly between bowel infiltrating and non-infiltrating disease. Thereafter, all patients undergo laparoscopy to detect and extirpate other visible endometriotic lesions and/or to prepare for resection of the affected rectosigmoid colon if infiltrating disease was found vaginally. In the latter case, the affected bowel tube is isolated from its mesentery while leaving the mesorectum unaffected and thus preserving the inferior hypogastric nerves and blood supply (Fig. 2). Finally, bowel resection is performed as the third part of the operation by mini-laparotomy. For all patients, postoperative alimentation was carried out in accordance with the fast track principle. Resected bowel specimens were histologically examined by a standard method to exactly define the endometriotic involvement and its location. The bowel was examined in its whole circumference with special care to cephalic and caudal resection margins. All patients were followed up in the out-patient department or contacted via telephone. Clinical symptoms such as chronic abdominal

symptoms or pain associated with menstruation before and after the operation were documented. All data concerning the operation and alterations due to it were reported. Data were collected via SPSS Statistics version 20.0 (IBM, Armonk, NY, USA). All metric data were reported as the median (range). Fisher exact test was used to verify significance and the t test was used to compare the groups. A P value of less than 0.05 was considered statistically significant. 3. Results During the study period, 110 women underwent vaginal–abdominal operative surgery for rectovaginal endometriosis. The median age of the patients was 33 years (range, 19–51 years). The median body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) was 23 (range, 17–31), and 75.5% of the women were nulliparous. Bowel infiltrating disease was confirmed during the vaginal part of the operation for 71 (64.5%) women; only non-infiltrating endometriosis with no involvement of the bowel’s surface was detected among 39 (35.5%) women. Overall, 76.3% of the 110 patients had a history of prior surgery for endometriosis. A median of 2.7 (range, 0–8) operations for endometriosis per patient had previously been carried out. Patients with bowel infiltration were operated on significantly more often as compared with those without bowel infiltration (P = 0.013). In total, 65.5% of all patients had a history of hormonal therapy using various substances, such as gonadotropin-releasing hormone analogs, danazol, progesterone, intrauterine device, and combined oral contraceptives. Pain was reported by 108 patients. The remaining 2 patients presented with visible and palpable endometriosis in the rectovaginal septum and infertility. Dysmenorrhea was reported among 83.6% (92/110), and bowel-associated symptoms such as dyschezia, cyclic constipation, and/or diarrhea among 61.8% (68/110) of all women. The clinical symptoms of women in both groups are summarized in Table 1. Among the preoperative diagnostic tools, bimanual rectovaginal gynecologic examination demonstrated the highest accuracy (sensitivity,

268

M. Mangler et al. / International Journal of Gynecology and Obstetrics 125 (2014) 266–269

Table 1 Characteristics of patients with rectovaginal endometriosis.a Characteristics

Non-infiltrating bowel endometriosis (n = 39)

Infiltrating bowel endometriosis (n = 71)

P value

Age, y Parity 0 1 2 3 Undergone previous operations for endometriosis No. of previous operations 0 1 2 3 4–8 Hormonal therapy before surgery Clinical symptoms Dysmenorrhea Hypermenorrhea Dyspareunia Chronic pelvic pain Defecating symptoms Dyschezia Hematochezia Cyclic rectal bleeding Diarrhea and constipation Dysuria Back pain Infertility Primary Secondary Follow up, months

33.02 (19–51)

33.35 (24–39)

0.475

54 (76.0) 12 (16.9) 4 (5.6) 1 (1.4) 58 (81.7)

N0.99 0.78 0.45 N0.99 0.1008

13 (18) 22 (31) 13 (18) 13 (18) 10 (14) 49 (69) 71 (100) 61 (86) 9 (13) 46 (65) 56 (79) 50 (70) 39 (55) 5 (7) 5 (7) 32 (45) 16 (23) 17 (24) 32 (45) 30 (42) 2 (3) 63.9 (6–98)

0.1008 0.1456 0.4086 0.7961 0.013 0.3029 0.1236 0.4253 0.2848 0.6805 0.0046 0.0147 0.073 N0.99 0.1588 0.0017 0.3110 0.0161 0.0249 0.0014 0.1829 0.618

a

29 (74.3) 5 (12.8) 4 (10.25) 1 (2.6) 26 (66.6)

13 (33) 16 (40) 4 (11) 6 (15) 0 23 (59) 37 (95) 31 (84) 8 (21) 23 (59) 20 (54) 18 (46) 14 (36) 2 (5) 0 6 (15) 5 (13) 2 (5) 9 (23) 5 (13) 4 (10) 64.1 (6–98)

Values are given as median (range) or number (percentage).

0.92; specificity, 0.32). Rectal endosonography obtained a sensitivity of 0.48 and a specificity of 0.76. All other diagnostic procedures had an even lower sensitivity and specificity, including MRI (sensitivity, 0.42; specificity, 0.83), gynecologic sonography (sensitivity, 0.2; specificity, 0.80) and rectosigmoidoscopy (sensitivity, 0.03; specificity, 0.92). The stages of endometriosis, as defined by the revised American Society of Reproductive Medicine (rASRM), were distributed as follows: rASRM I (13%), rASRM II (10%), rASRM III (13%), and rASRM IV (62%) in the non-infiltrating group; and rASRM I (6%), rASRM II (11%), rASRM III (23%), and rASRM IV (56%) in the infiltrating group. The mean operating time was 230 minutes (range, 82–600 minutes) for all patients. However, it was significantly longer in the group with bowel infiltrating endometriosis than in the non-infiltrating

group: 268.3 minutes (124–600 minutes) and 162.6 minutes (82– 360 minutes), respectively (P b 0.001). In total, 71 (64.5%) women underwent a segmental resection of the rectosigmoid colon for bowel infiltrating disease. For 39 (35.5%) patients, only superficial removal for non-infiltrating endometriosis was required. In addition, a cecal pole resection was performed for 7 patients in the bowel-infiltrating group (9.9%), and appendectomies were carried out for 2 patients in each group. In the bowel-infiltrating group, the urinary bladder was affected in 23 patients (32.5%), 5 of whom needed partial bladder resection (7.0%); and the ureter showed endometriosis in 12 patients (16.9%), 4 of whom required resection and ureteric neoimplantation (5.6%). In the non-infiltrating group, the bladder was affected in 9 patients (23.1%), 2 of whom required partial bladder resection (5.1%); the ureter showed endometriosis in 2 patients (5.1%). Hysterectomy was performed for 3 women (7.7%) in the noninfiltrating group. Sacrouterine ligaments were affected in 13 and 20 patients in the non-infiltrating and infiltrating bowel groups, respectively. Overall, 6 patients had a perioperative drop in hemoglobin, and 2 of these 6 patients required transfusions of blood. A revision on the first day after surgery was carried out for 1 of the 6 patients owing to anemia and suspicion of bleeding. Only diffuse hemorrhage was found and coagulated. A temporary artificial anus was constructed in 1 patient because the bowel resection was large and very close (b4 cm) to the sphincter ani. The protective ileostomy was reversed 3 months after the operation without problems. For 1 patient in the non-infiltrating group, a dehiscence of the vaginal scar occurred and a Vicryl mesh was applied. Two patients in the group with bowel infiltration had residual urine at discharge. One patient had pre-existing self-catheterization. Another patient was only able to void the bladder completely by pressing after the operation. For the 1 patient in the non-infiltrating group who had voiding problems, residual urine disappeared after 3 weeks. One patient presenting with symptomatic ileus was treated conservatively. No postoperative pelvic abscesses, anastomotic insufficiencies, or rectovaginal fistulas were observed. All complications are listed in Table 2. Endometriosis was confirmed histologically for all study women. Among all patients who underwent bowel resection, histologic infiltration of the bowel wall was found. Endometriosis affected the serosa and muscularis of the bowel in all patients with bowel infiltration; in 27 women (38.0%), the submucosa was infiltrated; and in 6 women (8.5%), the mucosa was infiltrated. For 1 woman, an ovarian borderline tumor was diagnosed, and the patient was treated according to guidelines. Patients stayed in hospital for a median of 7.2 days (range, 3–14 days). The median length of bowel resected was 7.5 cm. The median size of the endometriotic nodule was 23 × 15 × 11 mm. By counting cephalic and/or caudal resection margins in 19 of 142 (13.3%) of macroscopically

Table 2 Intraoperative and postoperative complications among patients with rectovaginal endometriosis.

a

Complication

Non-infiltrating bowel endometriosis (n = 39)

Infiltrating bowel endometriosis (n = 71)

P value

Operation time, min Intraoperative blood loss N500 mL Temporary artificial anus Intraoperative transfusion of blood Residual urine N50 mL at discharge Dehiscence of vaginal scar Insufficiency of anastomosis Transfusion of blood after operation Reoperation due to diffuse hemorrhage Sepsis, fever, abscess, rectovaginal fistula, thrombosis Ileus Infections (e.g. urinary tract infection) Intra- and postoperative complication rate Long-term complication rate

268.3 (124–600) 1 (2.5) 0 (0.0) 0 (0.0) 1 (2.5) 1 (2.5) 0 (0) 0 (0) 1 (2.5) 0 (0.0) 0 (0.0) 2 (5.1) 6 (15.4) 0 (0)

162.6 (82–360) 3 (4.2) 1 (1.4) 1 (1.4) 2 (2.8) 0 (0) 0 (0) 1 (1.4) 0 (0) 0 (0) 1 (1.4) 2 (2.8) 11 (15.5) 0 (0)

b0.001 N0.99 N0.99 N0.99 N0.99 N0.99 N0.99 N0.99 N0.99 N0.99 N0.99 0.6139 N0.99 N0.99

Values are given as median (range) or number (percentage).

M. Mangler et al. / International Journal of Gynecology and Obstetrics 125 (2014) 266–269

sound margins, endometrial glands were detected histologically in 14 of all patients (12.7%). All endometriotic lesions were located in the ventral and therefore anti-mesenteric surface of the bowel tube. No endometriotic lesions in the dorsal wall of the bowel, and/or mesosigmoid or mesorectum were observed. Patients were followed up for a medium of 64 months (6– 98 months). Using a standardized follow-up protocol, patients were examined gynecologically either by consultants in the study center (79%) or by their own gynecologists in an outpatient setting. Recurrence of endometriosis occurred among 7 of the study patients (6.4%). All recurrences were diagnosed and treated by laparoscopy. There was no significant difference in the recurrence rate of pelvic endometriosis between the women without 4 (10.3%) and those with 3 (4.3%) infiltrative disease (P = 0.69). Symptoms or clinical evidence for recurrence in the rectovaginal septum were not seen in any patient. One patient in the bowel-infiltrating group underwent resection of the rectosigmoid colon owing to inflammatory bowel disease 2 years after her operation for endometriosis, and a small endometriotic lesion of a 5 mm was found in the newly resected bowel wall. One patient in each group showed symptoms related to endometriosis, but recurrence was excluded in both cases by laparoscopy. Recurrence was not influenced by sound or affected resection margins. 4. Discussion Operative therapy among patients with rectovaginal endometriosis may be associated with highly morbid conditions including rectovaginal fistulas, a temporary or definitive artificial anus, and anastomosis insufficiency [14], in addition to considerable blood loss, blood transfusion, or the need for laparoconversions [4–9,11,15–19]. Because autonomic nerve structures in the pelvis are located in the operative field, there is also a risk of impairment of the autonomic bladder and bowel function. As a result, a nerve-sparing operative approach is mandatory in this benign disease to minimize adverse effects on bladder and rectum function. Because resection of the bowel might also affect the inferior hypogastric nerve and plexus, the identification and preservation of these vegetative nerves seems reasonable [20–22]. The technique used in the present study involves a nerve-sparing approach to rectovaginal endometriosis by which long-term symptoms that occur after damage to the hypogastric nerve and plexus might be reduced. In the present study, the incidence of voiding problems was 9% directly after the operation and 2.7% at discharge. Voiding problems had disappeared by 3 months after the operation when patients were examined in the follow-up. No persisting voiding problems occurred. In the present series, neither rectovaginal fistula nor anastomosis insufficiency occurred, and the long-term complication rate was 0%. A temporary artificial anus was required for 1 patient in whom the bowel resection was large and very close to the sphincter; the protective ileostomy was reversed 3 months later. However, a routine protective ileostomy does not seem to be necessary [7,17]. Notably, the incision in and resection of a small part of the posterior vagina wall did not lead to any enteric complications or rectovaginal fistulas among the present study women. For 7 patients (6.4%), pelvic endometriosis but no local rectovaginal endometriosis reoccurred; these patients had to undergo another operation for endometriosis on other locations. Unlike in malignant disease, the clinical significance of the microscopic involvement of the resection margins of the bowel remains uncertain in endometriosis [12,13,23]. Fourteen patients with unclear resection margins in the present study were free from local recurrence of rectovaginal endometriosis in the rectovaginal septum after a median follow-up of 64 months. Even though these data are reassuring, overall the number of study patients might be too small and the length of follow-up might be too short to draw a definitive conclusion on the significance of the involvement of the resection margins. In summary, the operative approach described in the present study gives the surgeon the opportunity to provide a precise therapeutic

269

approach to endometriosis of the rectovaginal septum depending on the depth of the bowel infiltration and to make a complete resection of the rectovaginal endometriosis more likely. Many different methods to operate on patients with rectovaginal endometriosis have been described before [24]. Standardization of the operative approach is still lacking, which might result in a higher complication rate. A standardized operation would help to minimize negative short- and long-term results.

Conflict of interest The authors have no conflicts of interest.

References [1] Kruse C, Seyer-Hansen M, Forman A. Diagnosis and treatment of rectovaginal endometriosis: an overview. Acta Obstet Gynecol Scand 2012;91(6):648–57. [2] Mangler M, Medrano N, Bartley J, Mechsner S, Speiser D, Schneider A, et al. Value of diagnostic procedures in rectovaginal endometriosis. Aust N Z J Obstet Gynaecol 2013;53(4):389–94. [3] Takai N, Ueda T, Nishida M, Nasu K, Narahara H. Bowel obstruction due to endometriosis in the rectovaginal septum. Clin Exp Obstet Gynecol 2008;35(4):295–6. [4] Fedele L, Bianchi S, Zanconato G, Bettoni G, Gotsch F. Long-term follow-up after conservative surgery for rectovaginal endometriosis. Am J Obstet Gynecol 2004;190(4): 1020–4. [5] Golfier F, Sabra M. Surgical management of endometriosis. J Gynecol Obstet Biol Reprod (Paris) 2007;36(2):162–72. [6] Keckstein J, Wiesinger H. Deep endometriosis, including intestinal involvement–the interdisciplinary approach. Minim Invasive Ther Allied Technol 2005;14(3):160–6. [7] Darai E, Ackerman G, Bazot M, Rouzier R, Dubernard G. Laparoscopic segmental colorectal resection for endometriosis: limits and complications. Surg Endosc 2007;21(9): 1572–7. [8] Maytham GD, Dowson HM, Levy B, Kent A, Rockall TA. Laparoscopic excision of rectovaginal endometriosis: report of a prospective study and review of the literature. Colorectal Dis 2010;12(11):1105–12. [9] Kavallaris A, Mebes I, Evagyelinos D, Dafopoulos A, Beyer DA. Follow-up of dysfunctional bladder and rectum after surgery of a deep infiltrating rectovaginal endometriosis. Arch Gynecol Obstet 2011;283(5):1021–6. [10] Haas D, Shebl O, Shamiyeh A, Oppelt P. The rASRM score and the Enzian classification for endometriosis: their strengths and weaknesses. Acta Obstet Gynecol Scand 2013;92(1):3–7. [11] Mangler M, Loddenkemper C, Lanowska M, Bartley J, Schneider A, Köhler C. Histopathology-based combined surgical approach to rectovaginal endometriosis. Int J Gynecol Obstet 2008;103(1):59–64. [12] Remorgida V, Ragni N, Ferrero S, Anserini P, Torelli P, Fulcheri E. How complete is full thickness disc resection of bowel endometriotic lesions? A prospective surgical and histological study. Hum Reprod 2005;20(8):2317–20. [13] Kavallaris A, Köhler C, Kühne-Heid R, Schneider A. Histopathological extent of rectal invasion by rectovaginal endometriosis. Hum Reprod 2003;18(6):1323–7. [14] Jelenc F, Ribič-Pucelj M, Juvan R, Kobal B, Sinkovec J, Salamun V. Laparoscopic rectal resection of deep infiltrating endometriosis. J Laparoendosc Adv Surg Tech A 2012;22(1):66–9. [15] Ford J, English J, Miles WA, Giannopoulos T. Pain, quality of life and complications following the radical resection of rectovaginal endometriosis. BJOG 2004;111(4): 353–6. [16] Dubernard G, Rouzier R, David-Montefiore E, Bazot M, Daraï E. Urinary complications after surgery for posterior deep infiltrating endometriosis are related to the extent of dissection and to uterosacral ligaments resection. J Minim Invasive Gynecol 2008;15(2):235–40. [17] Dubernard G, Piketty M, Rouzier R, Houry S, Bazot M, Darai E. Quality of life after laparoscopic colorectal resection for endometriosis. Hum Reprod 2006;21(5):1243–7. [18] Slack A, Child T, Lindsey I, Kennedy S, Cunningham C, Mortensen N, et al. Urological and colorectal complications following surgery for rectovaginal endometriosis. BJOG 2007;114(10):1278–82. [19] Spencer C, Robarts P. Urological and colorectal complications following surgery for rectovaginal endometriosis. BJOG 2008;115(4):539–40. [20] Ceccaroni M, Clarizia R, Bruni F, D’Urso E, Gagliardi ML, Roviglione G, et al. Nervesparing laparoscopic eradication of deep endometriosis with segmental rectal and parametrial resection: the Negrar method. A single-center, prospective, clinical trial. Surg Endosc 2012;26(7):2029–45. [21] Volpi E, Ferrero A, Sismondi P. Laparoscopic identification of pelvic nerves in patients with deep infiltrating endometriosis. Surg Endosc 2004;18(7):1109–12. [22] Kovoor E, Nassif J, Miranda-Mendoza I, Lang-Avérous G, Wattiez A. Long-term urinary retention after laparoscopic surgery for deep endometriosis. Fertil Steril 2011;95(2): 803.e9–803.e12. [23] Koh CE, Juszczyk K, Cooper MJ, Solomon MJ. Management of deeply infiltrating endometriosis involving the rectum. Dis Colon Rectum 2012;55(9):925–31. [24] Possover M, Diebolder H, Plaul K, Schneider A. Laparascopically assisted vaginal resection of rectovaginal endometriosis. Obstet Gynecol 2000;96(2):304–7.