Surgical treatment of deep endometriosis and risk of recurrence

Journal of Minimally Invasive Gynecology (2005) 12, 508-513 Surgical treatment of deep endometriosis and risk of recurrence Michele Vignali, MD, Stefano Bianchi, MD, Massimo Candiani, MD, Giovanna Spadaccini, MD, Giulia Oggioni, MD, and Mauro Busacca, MD From the Department of Obstetrics and Gynaecology, University of Milano, Macedonio Melloni Hospital (Drs. Vignali, Spadaccini, Oggioni, and Busacca); and Department of Obstetrics and Gynaecology, University of Milano, Clinica Mangiagalli (Drs. Bianchi and Candiani), Milano, Italy. KEYWORDS: Endometriosis; Laparoscopy; Deep endometriosis Abstract STUDY OBJECTIVE: To evaluate the risk of recurrence of deep endometriosis after conservative surgery. DESIGN: Retrospective analysis (Canadian Task Force classification II-3). SETTING: Tertiary care university hospital. PATIENTS: One hundred fifteen symptomatic patients operated on in our department from 1996 through 2002 with postoperative follow-up of at least 12 months. INTERVENTION: All patients underwent conservative surgery for deep infiltrating endometriosis. MEASUREMENT AND MAIN RESULTS: Risk factors for recurrence of symptoms and clinical findings and for repeated surgery were evaluated by univariate and multivariate analysis. During follow-up, we observed 28 patients with pain recurrence and 15 patients with recurrent clinical findings, and 12 patients required reoperation for deep endometriosis. Recurrence rates of pain and clinical findings during 36 months were 20.5% and 9%, respectively. Multivariate analysis showed that only age was a significant predictor of pain recurrence (OR 0.9, 95% CI 0.81-0.99, p.05), enhancing the risk in younger patients. Recurrence of clinical signs of deep endometriosis was predicted by obliteration of the pouch of Douglas (OR 1.46, 95% CI 1.16-16.2, p.05). Reoperation for deep endometriosis was predicted only by the incompleteness of first operation (OR 21.9, 95% CI 3.2 146.5, p.001). CONCLUSION: Our study indicates that age, obliteration of the pouch of Douglas, and surgical completeness may have a significant influence on the recurrence of the disease. 2005 AAGL. All rights reserved. Corresponding author: Michele Vignali, MD, Department of Obstetrics and Gynaecology, University of Milano, Via Macedonio Melloni, 52, 20122 Milano, Italy. E-mail: michele.vignali@unimi.it Submitted November 23, 2004. Accepted for publication June 15, 2005. The definition of deep endometriosis is still debated. Several authors agree that deep endometriosis is a separate entity of the disease that penetrates below the surface of the peritoneum more than 5 mm. 1 These lesions most commonly are located at the level of the uterosacral ligaments and the pouch of Douglas, 2 but often may extend along the rectovaginal septum and may infiltrate and extend backwards into the rectum and sigmoid, leading to cul-de-sac obliteration. 3 5 Deep infiltrating endometriosis differs from other peritoneal and ovarian lesions in its particular histopathologic features and its strong correlation with pelvic pain and severe dyspareunia. 6 When rectal infiltration occurs, dyschezia and rectal bleeding can be observed. 7 Because of 1553-4650/$ -see front matter 2005 AAGL. All rights reserved. doi:10.1016/j.jmig.2005.06.016

Vignali et al Recurrence of deep endometriosis 509 such debilitating symptoms, adequate management is advocated. While hormone-based medical treatment allows temporary quiescence of active lesions 8,9 with subsequent short-term pain relief, several authors consider surgery, both laparoscopy and laparotomy, the optimal treatment for deep endometriosis, since it allows for the complete excision of the lesions and final diagnosis with successful long-term results. 10 15 Furthermore, it has been demonstrated that surgical treatment of deep endometriosis may improve the recovery of fertility. 16,17 In the present study, we evaluated the results of conservative pelvic surgery, either by laparoscopy or laparotomy, in the treatment of deep infiltrating endometriosis, considering the rates of pain and disease recurrence as the main indicators of effectiveness of the surgery. Materials and methods We retrieved the clinical records of all women who were referred to the Second Department of Obstetrics and Gynecology from January 1996 through January 2002 who underwent conservative surgical treatment for deep peritoneal pelvic endometriosis. Institutional review board approval was not requested because this was a retrospective study. Deep endometriosis was defined as endometriosis infiltrating deeper than 5 mm under the peritoneal surface. Deep endometriosis was confirmed by histologic analysis in all patients. The depth of infiltration of endometriosis was assessed during both laparoscopy and laparotomy by means of a graded palpator held adjacent to the lesion. To be enrolled in the study, the following criteria had to be satisfied: age younger than 45 years at the time of surgery, presence of at least one painful symptom that was moderate to severe, conservative treatment with retention of the uterus and at least one ovary, and postoperative follow-up of at least 12 months after surgery or after the suspension of postoperative medical therapy. According to these criteria, we excluded patients for whom treatment was not conservative and those (six patients) who underwent a segmental bowel resection with anastomosis for full-thickness involvement by endometriosis. All surgical procedures were performed by one of four different skilled surgeons (MB, MV, MC, and SB) who were actively engaged in endometriosis clinics and research. Information regarding clinical and surgical data was obtained from detailed patients charts. The records of 115 patients were scrutinized. In fortynine women, both pelvic pain and infertility were associated with endometriosis. Infertility was defined as the unsuccessful attempt to conceive during at least 12 months. Of the women, 17.5% had previously delivered. Bowel preparation was done in all patients with suspected rectovaginal lesions. Three days before surgery, the patients followed a semi-liquid diet and the evening before surgery had a 1 L enema. They took metronidazole 1 g orally per day starting two days before the operation, and 1 g of intravenous amoxicillin was administered 30 minutes before the induction of anesthesia. All patients who had operative laparoscopy were operated on according to the following technique. After the induction of pneumoperitoneum and introduction of the laparoscopic fiberoptic and two or three suprapubic ancillary ports, the pelvis was inspected and cleared of adhesions. Adhesiolysis and dissection were performed with laparoscopic scissors, and hemostasis was achieved always with bipolar coagulation. An assistant stood between the patient s legs ready to perform a rectovaginal examination if required. After adhesiolysis, the deep endometriotic implants were dissected and isolated, then removed. Dissection was carried out until the loose areolar tissue of the retroperitoneal space was reached. In patients in whom the rectal wall was involved, a careful dissection was done until the normal muscularis layers of the rectum were reached. When necessary, ureterolysis and dissection of the lateral rectal space were carried out. Ureters were always identified before proceeding to dissection of deep implants infiltrating uterosacral ligaments. Endometriosis infiltrating the whole thickness of the vaginal wall through the mucosa required partial colpectomy combined with rectovaginal septum dissection. Associated endometriotic lesions (i.e., ovarian cysts, superficial peritoneal implants, adnexal adhesions) were treated during the same operative procedures, following the technique previously described. 18 Dissection and removal of deep infiltrating lesions by laparotomy was performed using the same procedures used with laparoscopy and following the technical principles of pelvic microsurgery. Ovarian and peritoneal lesions were evaluated at the time of laparoscopy or laparotomy and staged according to the American Fertility Society (AFS) classification. 19 At the end of the operation, the surgeons were requested to express their subjective opinion on the completeness of deep implants dissection and removal, and to judge whether residual grossly visible implants were left in the pelvis. On the basis of this opinion, the operation was defined as complete or not complete. Clinical examinations and transvaginal ultrasonographies were scheduled at 3-, 6-, and 12 months after surgery and subsequently at least once a year. The severity of pain symptoms (i.e., dysmenorrhea, dyspareunia, pelvic pain) was assessed before intervention and at follow-up visits using a questionnaire previously described. 20 In this questionnaire, two different scales to evaluate pain severity were used: a multidimensional one that emphasized the social impact of pain symptoms, and an analogic linear scale that expressed the subjective aspects of pain. Only moderate and severe pain were considered in the analysis. Recurrence of pain symptoms was defined as a recurrence or persistence of at least one symptom of pain that was moderate to severe, while recurrence of clinical

510 Journal of Minimally Invasive Gynecology, Vol 12, No 6, November/December 2005 findings was defined as the presence of fibrotic areas or tender nodularities affecting the pouch of Douglas or the uterosacral ligaments. Ultrasonographic recurrence of ovarian endometrioma, defined as previously reported 17 also was recorded. Statistical analysis Probabilities of recurrence of pain symptoms and clinical findings and of repeated surgery were calculated using the Kaplan-Meier method. The curves we obtained were stratified according to age, stage of disease, previous surgery, depth of infiltration of 1 cm or more or less than 1 cm, obliteration of the pouch of Douglas, surgical completeness, and pregnancy after surgery. The comparison between the curves was carried out with the log-rank test. Univariate analysis of the possible risk factors for recurrence of pain symptoms and clinical findings and of repeat surgery was done by 2 analysis and t test. The variables found by univariate analysis that were significantly associated with recurrence of pain symptoms and clinical findings and of repeated surgery were included in a multivariate analysis that was carried out using Cox regression. The results of the Cox regression were given as odds ratios and their 95% CIs. Results Of the 115 patients, ninety-six (83.5%) were treated with laparoscopy, reflecting the fact that in our department operative laparoscopy is considered the elective surgical procedure in the treatment of endometriosis, but a nonmarginal proportion of our patients required a laparotomy for the technical difficulties associated with the excision of deep endometriosis. The clinical characteristics of patients are given in Table 1. The mean duration of follow-up was 33.9 months. Deep endometriotic lesions involved the rectal wall in 15 women, the uterosacral ligaments in 57, the vesical fold in 17, and the pouch of Douglas in 36. The sum of these localizations exceeds the total number of patients because of multiple locations in some patients. No preoperative medical therapy was administrated. Fortyeight patients (42%) received postoperative medical therapy according to the preference of the referring gynecologist. Thirty of them were treated with an intramuscular injection of gonadotropin-releasing hormone analog every 4 weeks for 12 weeks, while 18 were given 600 mg of danazol daily for 3 months. Unilateral ureteral lesions were present at the time of intervention in one woman who required suturing but not anastomosis, and six patients had postoperative complications (two with vaginoperitoneal fistula and four with postoperative fever for more than 2 days). Table 1 Characteristics of patients who underwent conservative surgical treatment for deep infiltrating endometriosis Characteristic N 115 Age yrs* 30.1 5.4 Infertile, No. 49 (42.6) Previous surgery for endometriosis, No. 23 (20) AFS stage of disease, No. I 22 (19) II 27 (23.5) III 50 (43.5) IV 16 (14) Dysmenorrhea, No. 97 (84.3) Pelvic pain, No. 55 (47.8) Dyspareunia, No. 55 (47.8) Deep infiltrating endometriosis 1 cm, 49 (42.6) No. Type of operation, No. Laparoscopy 96 (83.5) Laparotomy 19 (16.5) Follow up mos* 33.9 23.2 AFS American Fertility Society. *Values are given as mean SD. According to the criteria previously described, 22 operations were not considered complete by the surgeons. During follow-up, we observed 28 patients with pain recurrence, and 15 patients with recurrent clinical findings. Reoperation for deep endometriosis was performed in 12 patients. The actuarial rates of pain and recurrence of clinical findings during 36 months were 20.5% and 9%, respectively. However, the actuarial rates of pain and recurrence of clinical findings during 60 months were 43.5% and 28%, respectively. In those women who underwent a second operation, the recurrence of deep endometriosis was observed in the same area of the pelvis involved in the first operation. Table 2 illustrates the differences between patients with pain recurrence and those without, in terms of age, stage, postoperative pregnancy, depth of endometriosis infiltration, surgical completeness, and obliteration of pouch of Douglas. The two groups differed significantly only by age and completeness of the operation. Further analysis evaluating the total, implants, and adhesions AFS score did not reveal other significant differences between patients with and without pain recurrence. Table 3 shows the differences between patients with and without clinical findings suggestive of recurrent deep endometriosis, in terms of age, stage, postoperative pregnancy, depth of endometriosis infiltration, surgical completeness, and obliteration of the pouch of Douglas. Stage, obliteration of pouch of Douglas, and surgical completeness were all significantly correlated with the risk of recurrence. In particular, patients with more severe endometriosis and those with pouch of Douglas obliteration had a higher probability of recurrence of clinical signs, whereas the patients who conceived after the operation had a trend towards low probability of recurrence, (p.07). Furthermore, further anal-

Vignali et al Recurrence of deep endometriosis 511 Table 2 Univariate analysis of predictive factors that are associated with recurrence of moderate or severe pain after surgery for deep infiltrating endometriosis Factor No pain recurrence n 87 Pain recurrence n 28 Age (yrs)* 31.5 5.5 29.1 4.6.04 AFS stage of disease, No. (%) I 17 (19.5) 5 (17.8) NS II 21 (24.1) 6 (21.4) III 39 (44.8) 11 (39.3) IV 10 (11.5) 6 (21.4) Pregnancy, No. (%) No 63 (72.4) 24 (85.7) NS Yes 24 (27.6) 4 (14.3) Endometriosis infiltrating 1 cm, No. (%) No 50 (57.5) 16 (57.1) NS Yes 37 (42.5) 12 (42.9) Surgical completeness, No. (%) No 13 (14.9) 9 (32.1).03 Yes 74 (85.1) 19 (67.9) Pouch of Douglas obliteration, No. (%) No 66 (75.9) 17 (60.7) NS Yes 21 (24.1) 11 (39.3) AFS American Fertility Society. *Values are given as mean SD. p ysis evaluating the total, implants, and adhesions AFS score of the total implants and adhesions did not reveal other significant differences between patients with and without pain recurrence. Patients who underwent a second operation for recurrent deep endometriosis were found to have undergone surgery that was not complete more frequently than those who were not reoperated on (7 of 22 vs 5 of 93, p.01). Other factors considered in the univariate analysis were not significantly associated with reoperation. Multivariate analysis showed that only age was a significant predictor of pain recurrence (OR 0.9, 95% CI 0.81-0.99, p.05), enhancing the risk in younger patients. Recurrence of clinical signs of deep endometriosis was predicted by obliteration of the pouch of Douglas (OR 1.46, 95% CI 1.16-16.2, p.05). The OR of the surgical com- Table 3 Univariate analysis of predictive factors that are associated with clinical recurrence of deep endometriosis after surgery Factor No recurrence n 100 Recurrence n 15 Age yrs* 31.1 5.6 30.6 4.3 NS AFS stage of disease, No. I 20 (20) 2 (13.3).02 II 26 (26) 1 (6.7) III 43 (43) 7 (46.7) IV 11 (11) 5 (33.3) Pregnancy, No. No 73 (73) 14 (93.3).07 Yes 27 (27) 1 (6.7) Endometriosis infiltrating 1 cm, No. No 58 (58) 8 (53.3) NS Yes 42 (42) 7 (46.7) Surgical completeness, No. No 17 (17) 5 (33.3).03 Yes 83 (83) 10 (66.7) Pouch of Douglas obliteration, No. No 76 (76) 7 (46.7).01 Yes 24 (24) 8 (53.3) AFS American Fertility Society. *Values are given as mean SD. P

512 Journal of Minimally Invasive Gynecology, Vol 12, No 6, November/December 2005 A ctuar ai l pain recurrencerate (% ) 35 30 25 20 15 10 5 0 pleteness factor was 0.24 (95% CI 0.05 1.24), but it did not reach statistical significance (p.07). Reoperation for deep endometriosis was predicted only by the incompleteness of the first operation (OR 21.9, 95% CI 3.2-146.5, p.001). Discussion 0 6 12 18 24 30 36 Time (months) Figure 1 The graph shows the actuarial rate of pain recurrence over 36 months. The results of our study indicate that conservative surgery for deep infiltrating endometriosis allows persistent relief of pain symptoms 3 years after the operation in 80% of the symptomatic patients (Figure 1) and an apparent disease-free interval of 36 months in more than 90% (Figure 2). However, as suggested by survival analysis, these recurrence rates may double in the 2 years subsequent to followup. These estimates of the survival analysis must be interpreted with caution, because most of our patients dropped out after the third year of follow-up. The evaluation of risk factors for deep endometriosis recurrence by means of multivariate analysis showed that obliteration of the pouch of Douglas was the main predictor of recurrence. The incompleteness of the first operation, although it did not reach statistical significance in predicting recurrence (p.07), was associated with a statistically significant higher risk of reoperation (p.001). Furthermore, younger patients had a greater probability of pain symptom recurrence independent from other anatomic or surgical factors. Our study undoubtedly has some limitations that do not allow for generalization of our results. In fact, the method used to measure the depth of subperitoneal endometriotic infiltration is rather empirical; furthermore, the evaluation of endometriotic infiltration depth depends mainly on the accuracy of dissection and delimitation of the borders of the lesions. Our series is too small to take into account all the anatomic and surgical variables potentially involved in determining the resolution of disease and its symptoms. Moreover, the anatomic variables we have considered in the analysis may not represent those most adequate to reflect disease extension and the surgical difficulties imposed by some locations. From another viewpoint, our series is relatively large, taking into account that deeply infiltrating endometriosis is the less frequent form of the disease. Statistical analysis of the data has been done with the aim to better define the risk of recurrence and the main predictors of this risk considering the complexity of the interaction of multiple variables in a long follow-up. One of the most relevant results of our study is the lack of correlation between the risk of pain and disease recurrence and depth of endometriosis infiltration. In fact, women who underwent excision of lesions infiltrating 1 cm or more did not have a greater risk of pain or disease recurrence than those with lesser infiltrating implants. This result was not predictable because it has been demonstrated that the depth of infiltration, in particular when greater than 1 cm, is markedly correlated with the severity of pain. 21 In addition, the deeper the lesion, the more difficult the dissection from the surrounding tissue and its complete excision. The recurrence of clinical disease signs and reoperation were significantly more frequent in women who had an incomplete operation in the surgeon s opinion. The relevance of this subjective parameter is demonstrated through univariate analysis. However, the persistence of deep endometriotic implants after surgery does not mean inevitable persistence or an early recurrence of pain symptoms. In fact, women who underwent an incomplete operation did not report more pain than patients who underwent a complete operation. Pain symptoms in our series recurred similarly in patients with and without recurrent or persistent deep endometriosis. This finding can raise further doubt on the muchdebated association between endometriosis and pelvic pain, but we also have to recognize that nondeep endometriotic lesion recurrence or persistence may have contributed to pain symptoms. The lack of a significant association between pain and deep lesion recurrence confirms the nonspecificity of pain as a marker of a particular form or location of the disease. All the papers published until now on the surgical treatment of deep endometriosis have reported a high cure rate and an almost complete resolution of symptoms after surgery. 10 13,21 24 Several papers have focused on specific Figure 2 The graph shows the actuarial rate of clinical findings recurrence over 36 months.

Vignali et al Recurrence of deep endometriosis 513 locations of deep endometriosis, and the authors have stressed the technical details of the surgical procedures 7,10,11 rather than the prognostic value of potential anatomic and nonanatomic predictive factors. From this aspect, our work may be considered original. Our results seem to suggest a worse prognosis compared with those of previous series published by different authorities in the field of advanced laparoscopic surgery. However, the postsurgical outcome of a selected series of patients operated on by a few skilled surgeons may not apply to a nonselected population. Our study indicates also that the ability to perform a complete operation may have a significant influence on the risk of disease recurrence. References 1. Koninckx PR, Martin D. Treatment of deeply infiltrating endometriosis. Curr Opin Obstet Gynecol. 1994;6:231-241. 2. Cornillie FJ, Oosterlynck D, Lauweryns JM, et al. Deeply infiltrating pelvic endometriosis: Histology and clinical significance. Fertil Steril. 1990;53:978-983. 3. Chapron C, Fauconnier A, Vieira M, et al: Anatomic distribution of deeply infiltrating endometriosis: surgical implications and proposition for a classification. Hum Reprod. 2003;18:157-162. 4. Martin DC, Batt RE. Retrocervical, retrovaginal pouch, and rectovaginal septum endometriosis. J Am Assoc Gynecol Laparosc. 2001;8:12-17. 5. Vercellini P, Frontino G, Pietropaolo G, et al. Deep endometriosis: definition, pathogenesis, and clinical management. J Am Assoc Gynecol Laparosc. 2004;11:153-161. 6. Koninckx PR, Meuleman C, Demeyere S, et al. Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertil Steril. 1991;55:759-765. 7. Redwine DB. Laparoscopic en bloc resection for treatment of the obliterated cul-de-sac in endometriosis. J Reprod Med. 1992;37:695-698. 8. Brosens IA, Verleyen A, Cornillie F. The morphologic effect of short-term medical therapy of endometriosis. Am J Obstet Gynecol. 1987;157:1215-1221. 9. Nisolle-Pochet M, Casanas-Roux F, Donnez J. Histologic study of ovarian endometriosis after hormonal therapy. Fertil Steril. 1988;49: 423-426. 10. Nezhat C, Nezhat F, Pennington E. Laparoscopic treatment of infiltrative rectosigmoid colon and rectovaginal septum endometriosis by the technique of videolaparoscopy and the CO 2 laser. Br J Obstet Gynaecol. 1992;99:664-667. 11. Reich H, McGlynn F, Salvat J. Laparoscopic treatment of cul-de-sac obliteration secondary to retrocervical deep fibrotic endometriosis. J Reprod Med. 1991;36:516-522. 12. Redwine DB. Conservative laparoscopic excision of endometriosis by sharp dissection: life-table analysis of reoperation and persistent or recurrent disease. Fertil Steril. 1991;56:628-634. 13. Chapron C, Dubuisson DB, Fritel X, et al. Operative management of deep endometriosis infiltrating the uterosacral ligaments. J Am Assoc Gynecol Laparosc. 1999;6:31-37. 14. Crosignani PG, Vercellini P. Conservative surgery for severe endometriosis: should laparotomy be abandoned definitively? Hum Reprod. 1995;10:2412-2418. 15. Crosignani PG, Vercellini P, Biffignandi F, et al. Laparoscopy versus laparotomy in conservative surgical treatment for severe endometriosis. Fertil Steril. 1996;66:706-711. 16. Chapron C, Fritel X, Dubuisson DB. Fertility after laparoscopic management of deep endometriosis infiltrating uterosacral ligaments. Hum Reprod. 1999;14:329-332. 17. Redwine DB, Wright JT. Laparoscopic treatment of complete obliteration of the cul-de-sac associated with endometriosis: long-term follow-up of en bloc resection. Fertil Steril. 2001;76:358-365. 18. Busacca M, Marana R, Caruana P, et al. Recurrence of ovarian endometrioma after laparoscopic excision. Am J Obstet Gynecol. 1999;180: 519-523. 19. American Society for Reproductive Medicine. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril. 1997;67:817-821. 20. Fedele L, Bianchi S, Bocciolone L, et al. Buserelin acetate in the treatment of pelvic pain associated with minimal and mild endometriosis: a controlled study. Fertil Steril. 1993;59:516-521. 21. Koninckx PR, Oosterlynck D, D Hooghe T, et al. Deeply infiltrating endometriosis is a disease whereas mild endometriosis can be considered a non-disease. Ann NY Acad Sci. 1994;734:333-341. 22. Donnez J, Nisolle M, Gillerot S, et al. Rectovaginal septum adenomyotic nodules: a series of 500 cases. Br J Obstet Gynaecol. 1997;104: 1014-1018. 23. Donnez J, Nisolle M. Advanced laparoscopic surgery for the removal of rectovaginal septum endometriotic or adenomyotic nodules. Bailleres Clin Obstet Gynaecol. 1995:9:769-774. 24. Maher P, Wood C, Hill D. Excision of endometriosis in the pouch of Douglas by combined laparovaginal surgery using the Maher abdominal elevator. J Am Assoc Gynecol Laparosc. 1995;2:199-202.