- Email: [email protected]
Ureteral Endometriosis: Preoperative Risk Factors Predicting Extensive Urologic Surgical Intervention
Accepted Manuscript Title: Ureteral Endometriosis: Preoperative Risk Factors Predicting Extensive Urologic Surgical Intervention Author: Kyle H. Gennaro, Jennifer Gordetsky, Soroush Rais-Bahrami, John P. Selph PII: DOI: Reference:
S0090-4295(16)30524-6 http://dx.doi.org/doi: 10.1016/j.urology.2016.08.016 URL 19972
To appear in:
Urology
Received date: Accepted date:
16-6-2016 10-8-2016
Please cite this article as: Kyle H. Gennaro, Jennifer Gordetsky, Soroush Rais-Bahrami, John P. Selph, Ureteral Endometriosis: Preoperative Risk Factors Predicting Extensive Urologic Surgical Intervention, Urology (2016), http://dx.doi.org/doi: 10.1016/j.urology.2016.08.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 Ureteral Endometriosis: Preoperative risk factors predicting extensive urologic surgical intervention Kyle H. Gennaro BS1, Jennifer Gordetsky MD2,3, Soroush Rais-Bahrami MD2,4, John P. Selph MD2 1. University of Alabama, School of Medicine, Birmingham, AL 2. University of Alabama at Birmingham, Department of Urology, Birmingham, AL 3. University of Alabama at Birmingham, Department of Pathology, Birmingham, AL 4. University of Alabama at Birmingham, Department of Radiology, Birmingham, AL
Conflicts of Interest and Source of Funding: None
* Correspondence: J. Patrick Selph, MD Department of Urology Faculty Office Tower 1120 1720 2nd Avenue South, Birmingham, AL 35294 E-mail: [email protected] Word Count: 2257
Page 1 of 22
2 Abstract Objective: To identify risk factors for urologic reconstruction during surgery for endometriosis
Patients and Methods: We retrospectively identified patients in a surgical pathology database undergoing surgery for endometriosis at our institution from 2010-2015 and subsequently identified those patients with ureteral involvement. Patients were categorized as requiring minimal urologic surgery (e.g. ureterolysis only) or more extensive urologic surgery (e.g. ureteral reimplant). All patients were undergoing surgery for endometriosis, and preoperative risk factors were then identified to predict the need for intraoperative extensive urologic surgery.
Results: Of 386 women undergoing surgery for endometriosis, 82 (21%) women required a surgical procedure on the ureter. 15 of these 82 patients (18.3%) with ureteral involvement required urologic surgical expertise in the form of either ureteral reimplantation with or without psoas hitch, or ureterolysis with ureteral stenting or omental wrap. The remaining 67 underwent ureterolysis alone or no intervention. The presence of flank pain, any urinary symptom, or hydronephrosis on preoperative imaging was a significant predictor of the need for major urologic intervention.
Conclusion: In patients with endometriosis undergoing surgery who complain of flank pain, any urinary symptom, or have hydronephrosis on preoperative imaging, one should have a high suspicion for needing to perform urologic reconstruction during
Page 2 of 22
3 surgery. Planning for this additional operation can afford the opportunity for appropriate urologic consultation and patient counseling. Key Words: endometriosis, ureteral obstruction, urogenital surgical procedures, ureteral stricture Introduction: Endometriosis is the presence of endometrial-type tissue in a location outside of the uterus, most commonly the pelvic peritoneum, ovaries, and rectovaginal septum [1]. The literature estimates the prevalence of endometriosis to be between 1 to 10% in reproductive age women [1-5]. Endometriosis involves the urinary tract in approximately 0.3 to 12% of patients, with the bladder, ureter, and kidneys most commonly affected [6,7]. Ureteral endometriosis (UE) is often classified as intrinsic or extrinsic based on histologic findings. Extrinsic UE, which includes periureteral or serosal involvement, is the most common type and exists in 80% of cases. Intrinsic UE, which is the presence of endometrial tissue in the mucosa or muscularis propria, is more rare [8]. Presentation of UE is variable depending on location and extent of the disease. Patients may present with abdominal pain or urinary obstruction; however, 50% of patients lack urinary symptoms at the time of their presentation, making the diagnosis of UE difficult [6,8,9]. Surgical management is commonly used to treat UE, with different levels of intervention required based on the extent of disease [6]. We investigated preoperative clinical factors in patients with UE to predict those who might need more extensive urologic surgical intervention.
Patients and Methods:
Page 3 of 22
4 We performed an Institutional Review Board-approved search of our surgical pathology database from 2010-2015 for cases of endometriosis. Clinical, pathologic, and surgical data were analyzed for patients who had pathology-proven endometriosis and macroscopic intraoperative surgical findings of endometriosis involving the ureter. The study population was divided into two groups. Group 1, the urologic intervention group, consisted of patients that required any urologic surgical expertise, including ureteral stenting (with or without ureterolysis), ureteroneocystostomy with or without psoas hitch, or ureterolysis with omental wrap. Group 2, the non-urologic surgical intervention group, included patients that required only ureterolysis or no urologic surgical intervention. The two groups were compared regarding preoperative clinical symptoms, imaging, and laboratory data. All statistical analysis was performed in RStudio Version 0.99.896, and the epiR package was used to calculate odds ratios and for chi-squared testing. Continuous data was expressed in terms of mean, standard deviation, and range. Categorical data was expressed as percentages. Continuous data were assessed for normality using graphical representation and compared using a t-test. Categorical data were compared using the chi-squared test and a p-value of <0.05 was considered statistically significant for all tests. Results: We identified 405 cases with pathologically proven endometriosis (figure 1). 19 cases were excluded: 4 without any clinical information, and 15 that were repeat biopsies on the same patient. A total of 386 patients with histologic evidence of endometriosis were included. Of these, 82 (21%) underwent a procedure involving the
Page 4 of 22
5 ureter and/or had macroscopic surgical findings of ureteral involvement. The average age of our patient population was 41.3 years (Table 1). 27 (32.9%) patients had a previous history of endometriosis, with an average age at diagnosis of 31.7 years (range 14-53). Eleven (13.4%) patients (7 surgically-induced, 2 natural, 2 unknown cause) were postmenopausal, with an average age at menopause of 42.3 years (range 36-45). 27/82 (32.9%) patients had >1 site of biopsy proven endometriosis. 19/82 (23.1%) patients had a prior history of hysterectomy, 10/82 (12.2%) unilateral oophorectomy, and 2/82 (2.4%) bilateral oophorectomy. 2/82 (2.4%) patients had previously received systemic hormone replacement therapy. The most common presenting symptom in both Group 1 and Group 2 was abdominal pain; 40.0% and 46.3% respectively (Table 2). The presence of any urinary symptom including urinary urgency, frequency, hematuria, dysuria (OR 7.88, 95% CI 1.80, 34.4), flank pain (OR 4.58; 95% CI 1.26,16.69), or hydronephrosis (OR 76.7; 95% CI 13.4, 439) was significantly more common in Group 1. The differences between physical exam findings, white blood cell count, and creatinine were not statistically significant between the two groups. 37/82 (45.1%) patients received preoperative imaging of the urinary tract. Radiologic evidence of hydronephrosis was identified in 18/37 (43.2%) patients. 13/18 (72.2%) patients with preoperative hydronephrosis required a more extensive urologic intervention. Of the 5 who did not undergo extensive urologic surgery, 4 had ureterolysis alone, and one underwent no intervention. In comparing patients who had hydronephrosis with those who did not, patients with hydronephrosis were more likely to
Page 5 of 22
6 have urinary symptoms (OR 9.67; 95% CI 2.12, 44.15) and/or flank pain (OR 4.58; 95% CI 1.26, 16.69). As highlighted in Table 3, 15/82 (18.3%) patients required more extensive urologic intervention (Group 1). Among these, 10/15 (60.0%) patients were treated with ureteroneocystostomy, of which 7/10 (70%) were left-sided, 2/10 (20%) were rightsided, and 1/10 (10%) was bilateral. One patient required partial cystectomy in addition to ureteroneocystostomy, and all patients received ureteral stents. Three of the reimplantations were done in conjunction with a psoas hitch. 3/15 (20.0%) patients were treated with ureterolysis and stenting of which 1 (33%) was left-sided, 1 (33%) was right-sided, and 1 (33%) was bilateral. Of the remaining two patients in Group 1, one was treated with ureterolysis with omental wrap, and the second had ureterolysis with failed ureteral stenting that ultimately required nephrostomy tube placement and was diagnosed with a non-functioning kidney. No patients were treated with a ureteroureterostomy. One patient who had undergone left ureterolysis with omental wrapping developed a recurrent pelvic mass requiring repeat ureterolysis. Follow-up imaging demonstrated no hydronephrosis over 4 years later. No other patients from this group required additional surgeries related to their initial pathology, though a second patient did develop endometrial carcinoma at a later date causing recurrence of hydronephrosis that was conservatively managed. 67/82 (81.7%) patients were managed conservatively with only ureterolysis or no intervention (Group 2). 58/67 (87.8%) patients were treated with ureterolysis, 1/67 (1.5%) patient had excision of endometriotic tissue abutting the ureter, and 8/67 (11.9%) had macroscopic findings suggesting ureteral involvement but received no intervention.
Page 6 of 22
7 Of those who received ureterolysis, 20/58 (34.5%) were bilateral, 18/58 (31.0%) were left-sided, and 20/58 (34.5%) were right-sided. One patient that received conservative management was found to have a right hydroureter 3 months after surgery due to a periureteral mass, suggesting recurrence of disease.
Page 7 of 22
8 Discussion: In patients with endometriosis, rates of urinary tract involvement range from 0.3 to 12%, with UE reported in 4-5% of cases [5,6,10,12]. Our study demonstrated ureteral involvement in 21% of cases of endometriosis, which is significantly higher than previous studies. In our experience, UE more commonly involved the left side, which has been similarly reported in other studies [5-14]. UE is difficult to diagnose due to the wide variation of presentations and lack of physical exam findings [15]. The majority of our cases were in premenopausal women near age 40 presenting with non-specific abdominal pain. In a study by Nezhat et al, UE was unsuspected in two-thirds of surgical cases [14]. Similarly, in our study endometriosis was listed as the indication for surgery in only 26.8% of cases. Furthermore, we found preoperative physical examination and serum creatinine to be unhelpful for predicting the necessity of more extensive urologic intervention. We found that the presence of preoperative urinary symptoms, flank pain, or hydronephrosis significantly predicted the necessity of more extensive intervention. These symptoms are similar to the findings in Gabriel and colleagues study of their endometriosis series, where the presence of dysuria, hematuria, or frequent daytime urination and recurrent urinary tract infections were predictive of ureteral involvement of endometriosis [16]. UE is classically described as being intrinsic or extrinsic [8]. In many cases, extrinsic involvement of the ureter can be cured with simple ureterolysis, while intrinsic involvement necessitates more extensive surgery, including ureteroureterostomy or ureteroneocystostomy. Some groups have suggested varying levels of surgical treatment based on intrinsic versus extrinsic disease with conservative
Page 8 of 22
9 management (including ureterolysis) recommended for extrinsic disease and more extensive resections (including ureteral resection) recommended for intrinsic disease [6]. Knabben and colleagues recently noted that 98% of patients in their series had ureteral involvement treated with ureterolysis alone, and others have shown success rates of 85-90% with ureterolysis only [13, 17-19]. Although extrinsic type UE is thought to be more common, some studies have shown that the true rates of intrinsic disease may be greatly underestimated [8,12]. A limitation of the extrinsic and intrinsic classification system, however, is that it is a postoperative diagnostic system - a definitive diagnosis of intrinsic UE requires a ureteral tissue sample, which is typically not performed preoperatively. Even if preoperative ureteral biopsy is performed, it may not be positive if endometrial tissue has not invaded into the lumen or mucosa of the ureter. In our study, three patients who underwent ureteral reimplantation had preoperative ureteroscopic biopsy to rule out malignancy, and none of the samples were positive for endometriosis. Only a final pathologic specimen was positive for endometriosis. For preoperative planning purposes, using intrinsic versus extrinsic disease as a prognostic indicator for the level of intervention is not ideal. We found that the presence of preoperative urinary symptoms, flank pain, or hydronephrosis significantly predicted the necessity of more extensive intervention. For this reason, we recommend the involvement of a urologic surgeon in the preoperative planning for patients with these symptoms and suspected endometriosis. Others have looked at the histologic appearance of the endometrioma to try and predict the need for urologic surgery. Serachiolli and colleagues found no difference in the risk of ureteral reconstruction for patients who had either endometrial
Page 9 of 22
10 glands or fibrotic endometrial tissue present in the pathologic specimen [20]. Again, however, this assessment can only be done with a biopsy specimen, which is usually obtained after radical surgery has been performed. In multiple cases in our series, women presented with primarily urologic complaints – flank pain, recurrent kidney infections, and obstruction on renal scintigraphy. In these cases, ureteroscopic biopsy was performed to rule out malignancy, but in all three cases in which this procedure was done, the biopsy specimen was negative for endometriosis. Others have found that ureteroscopic biopsy can be inconclusive in UE [21]. This poor sensitivity of ureteroscopic biopsy for endometriosis argues that a more definitive imaging modality might help better guide the surgeon in preoperative planning and patient counseling. Several studies have attempted to analyze the best preoperative imaging modality for detecting UE. In a study by Pateman et al, 2% of women with chronic pelvic pain had evidence of UE [5]. They showed that ultrasound with attention to the urinary tract has a high sensitivity (92%) and specificity (100%) for UE. For this reason they recommended routine assessment of the urinary tract in all women with endometriosis. Given the low rate of UE, however, this may not be necessary as a broad screening modality in all patients, but perhaps patients with other predictors of urinary tract involvement, e.g., urinary symptoms or flank pain. In our study, 37 patients underwent preoperative imaging of the urinary tract. Of these 37 women, 18 were found to have hydronephrosis. Predictive factors for hydronephrosis were identical to those predicting the need for more extensive surgery. Urinary symptoms and flank pain, with an OR of 9.67 and 4.58 respectively (Table 3), were the only statistically significant predictors of
Page 10 of 22
11 hydronephrosis. Physical exam findings, preoperative WBC, and preoperative Cr, were not significantly different between patients with and without hydronephrosis. Additionally, routine assessment of the urinary tract for endometriosis does not seem necessary in our patient population, as ultimately only 15/386 (3.9%) required significant urologic intervention. On the contrary, most patients receive pelvic ultrasound as a diagnostic tool in endometriosis, so the short additional time and cost it takes to evaluate the urinary tract simultaneously may be prudent. Sillou’s group recently published their experience with preoperative magnetic resonance (MR) imaging in endometriosis and found a sensitivity of 91% for diagnosing intrinsic UE, while also noting that in cases where the endometrioma surrounded less than 180° of the ureter, intrinsic UE was found in less than 10% of cases [7]. Certainly this information could aid the surgeon in preoperative planning, though the cost and relative rarity of intrinsic UE argues against routine MR before endometriosis surgery. The goals of surgical treatment for UE are to relieve the obstruction, preserve renal function, and prevent future recurrence of disease [6]. Due to the low prevalence of UE, there have been no prospective studies to date comparing the outcomes between conservative management and more extensive urologic management. Some studies suggest that conservative management with ureterolysis is adequate and should be offered initially to all patients with UE [10,13]. Ghezzi and colleagues reported that all patients were symptom free and showed no evidence of obstruction on intravenous pyelogram 3-months after ureterolysis; however, 12.1% of their patients required one or more operations for disease recurrence [13]. Knabben’s group found a 98% success rate with ureterolysis alone [17]. Antonelli’s group noted only one recurrence out of 30
Page 11 of 22
12 patients with ureteral or bladder endometriosis when radical removal of extensive and obstructive genitourinary endometriosis was performed [11]. Our data seem to support both notions, as only one patient that received management with ureterolysis alone was found to have a right hydroureter three months after surgery, and a second patient developed a recurrent pelvic mass requiring resection and repeat ureterolysis. Limitations of our study include the lack of a standardized protocol for patients with endometriosis. Because it is not routine practice at our institution to perform imaging of the urinary tract prior to surgery for endometriosis, it is possible that more patients with ureteral endometriosis had preoperative hydronephrosis but were not identified; however, review of the operative reports indicated that these patients did not have any hydroureter, and thus the likelihood of a positive imaging test preoperatively was low. Because this is a retrospective study, it was not designed to ascertain the success of urologic reconstruction in these patients, and thus our follow-up is short (mean seven months). It is possible some patients who underwent a more extensive reconstruction developed recurrent urinary obstruction. Lastly, we noted a higher rate (21%) of ureteral endometriosis than in many other published studies – this may be a reflection of the fact that we serve as a tertiary referral center for complicated endometriosis, or simply a reflection of how different surgeons characterize ureteral “involvement.”
Conclusion: The diagnosis of UE is difficult with few guidelines on surgical management. The presence of urinary symptoms, flank pain, and/or hydronephrosis increases the
Page 12 of 22
13 likelihood of requiring a more extensive surgical intervention. We recommend the involvement of a urologic surgeon in the preoperative planning for patients with these symptoms and suspected endometriosis.
Page 13 of 22
14 References 1. Giudice LC. Clinical practice. Endometriosis. N Engl J Med. 2010;362(25):2389-2398. 2. Barbieri RL. Etiology and epidemiology of endometriosis. Am J Obstet Gynecol. 1990;162(2):565-567. 3. Dunselman GA, Vermeulen N, Becker C, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29(3):400-412. 4. Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstet Gynecol Clin North Am. 1997;24(2):235-258. 5. Pateman K, Holland TK, Knez J, et al. Should a detailed ultrasound examination of the complete urinary tract be routinely performed in women with suspected pelvic endometriosis? Hum Reprod. 2015;30(12):2802-2807. 6. Maccagnano C, Pellucchi F, Rocchini L, et al. Ureteral endometriosis: proposal for a diagnostic and therapeutic algorithm with a review of the literature. Urol Int. 2013;91(1):1-9. 7. Sillou S, Poirée S, Millischer AE, et al. Urinary endometriosis: MR imaging appearance with surgical and histological correlations. Diagn Interv Imaging. 2015;96(4):373-381. 8. Yohannes P. Ureteral endometriosis. J Urol. 2003;170(1):20-25. 9. Stepniewska A, Grosso G, Molon A, et al. Ureteral endometriosis: clinical and radiological follow-up after laparoscopic ureterocystoneostomy. Hum Reprod. 2011;26(1):112-116. 10. Donnez J, Nisolle M, Squifflet J. Ureteral endometriosis: a complication of rectovaginal endometriotic (adenomyotic) nodules. Fertil Steril. 2002;77(1):32-37.
Page 14 of 22
15 11. Antonelli A, Simeone C, Zani D, et al. Clinical aspects and surgical treatment of urinary tract endometriosis: our experience with 31 cases. Eur Urol. 2006;49(6):10931097; discussion 1097-1098. 12. Chapron C, Chiodo I, Leconte M, et al. Severe ureteral endometriosis: the intrinsic type is not so rare after complete surgical exeresis of deep endometriotic lesions. Fertil Steril. 2010;93(7):2115-2120. 13. Ghezzi F, Cromi A, Bergamini V, et al. Outcome of laparoscopic ureterolysis for ureteral endometriosis. Fertil Steril. 2006;86(2):418-422. 14. Nezhat C, Nezhat F, Nezhat CH, et al. Urinary tract endometriosis treated by laparoscopy. Fertil Steril. 1996;66(6):920-924. 15. Ghezzi F, Cromi A, Bergamini V, et al. Management of ureteral endometriosis: areas of controversy. Curr Opin Obstet Gynecol. 2007;19(4):319-324. 16. Gabriel B, Nassif G, Trompoukis P, et al. Prevalence and Management of Urinary Tract Endometriosis: A Clinical Case Series. Urology 2011; Dec;78(6):1269-74. 17. Knabben L, Imboden S, Fellmann B, et al. Urinary tract endometriosis in patients with deep infiltrating endometriosis: prevalence, symptoms, management, and proposal for a new clinical classification. Fertil Steril. 2015;103(1):147-152. 18. Soriano D, Schonman R, Nadu A, et al. Multidisciplinary team approach to management of severe endometriosis affecting the ureter: long-term outcome data and treatment algorithm. J Minim Invasive Gynecol. 2011;18(4):483-488. 19. Uccella S, Cromi A, Casarin J, et al. Laparoscopy for ureteral endometriosis: surgical details, long-term follow-up, and fertility outcomes. Fertil Steril. 2014;102(1):160-166.e162.
Page 15 of 22
16 20. Seracchioli R, Raimondo D, Di Donato N, et al. Histological evaluation of ureteral involvement in women with deep infiltrating endometriosis: analysis of a large series. Hum Reprod. 2015; 30(4):833-839. 21. Seyam R, Mokhtar A, Al Taweel W, et al. Isolated ureteric endometriosis presenting as a ureteric tumor. Urol Ann. 2014;6(1):94-97.
Figure 1. Patient selection for surgical intervention cohorts Table 1. Patient demographics Average age (years; sd, range) Average follow-up (months; range) Race Caucasian African-American Hispanic Endometriosis Biopsy Location Ureter Bladder Wall Perivaginal Cuff Cervix Uterus Fallopian tube Paratubal Ovary Peritoneum Retroperitoneum Bowel Omentum
41.3 +/- 8.4 (22.4-68.5) 7.5 +/- 11.0 (0.0-56.2) n (%) 45 (54.9%) 33 (40.2%) 4 (4.9%) 5 (6.1%) 4 (4.9%) 1 (1.2%) 5 (6.1%) 16 (19.5%) 6 (7.3%) 3 (3.7%) 56 (68.3%) 12 (14.6%) 2 (2.4%) 6 (7.3%) 1 (1.2%)
Page 16 of 22
17 Table 2. Comparison of patients requiring urologic surgical intervention and no urologic surgical Intervention Extensive Surgical Intervention (n=15)
Minimal or No Surgical Intervention (n=67)
n (%)
n (%)
OR (p-value, 95% CI)
Urinary Symptoms
5 (33.3%)
4 (6.0%)
7.88 (p=0.002, 1.80-34.40)
Dyspareunia
1 (6.7%)
9 (13.4%)
0.46 (p=0.469, 0.05-3.94)
Dysmenorrhea
3 (20.0%)
17 (25.4%)
0.74 (p=0.661, 0.19-2.92)
Pelvic Pain
5 (33.3%)
26 (38.8%)
0.79 (p=0.693, 0.24-2.57)
Abdominal Pain
6 (40.0%)
31 (46.3%)
0.77 (p=0.659, 0.25-2.42)
Flank Pain
5 (33.3%)
7 (10.4%)
4.29 (p=0.023, 1.14-16.18)
Back Pain
0 (0.0%)
2 (3.0%)
0.00 (p=0.915, 0.00-NA)
Abnormal Uterine Bleeding 3 (20.0%)
21 (31.3%)
0.55 (p=0.383, 0.14-2.15)
Asymptomatic
6 (9.0%)
0.00(p=0.402, 0.00-NA)
Abdomen/Pelvis Tender to 3 (20.0%) Palpation
23 (34.3%)
0.48 (p=0.281, 0.12-1.87)
Vaginal Bleeding
3 (20.0%)
7 (10.4%)
2.14 (p=0.307, 0.48-9.49)
Abdominal/Pelvic Mass
2 (13.3%)
26 (38.8%)
0.24 (p=0.06, 0.05-1.16)
Laboratory Findings
Mean +/- st. dev (range)
Mean +/- st. dev (range)
Average Cr
0.82 +/- 0.19 (0.5-1.3)
0.76 +/- 0.13 (0.5-1.1))
p=0.20
Average WBC
9.07+/- 4.62 (2.9-19.0)
7.80 +/- 2.36 (3.6-16.5)
p=0.14
13 (86.7%)
5 (7.8%)*
76.70 (p<0.001, 13.38-439.76)
Presenting Symptoms
0 (0.0%)
Physical Exam Findings
Imaging Findings Hydronephrosis
*5/64, 3 patients with unknown pre-operative hydronephrosis status.
Page 17 of 22
18 Table 3. Characteristics of patients requiring urologic surgery for endometriosis
Cas e
1
2
3
4
Preoperativ e Urologic Interventio n Clot evacuation, gross hematuria CT showing mass Bladder requiring invading bladder, biopsy and continuous vagina, and left left ureteral bladder irrigation hydronephrosis stent placement Presenting Symptoms
flank pain and dysmenorrhea
Upper tract Imaging preop
CT scan with left hydronephrosis
None
Intraoperativ Surgery Performed Pathology e Findings (Team performing) Findings
Followup
endometriosi Direct lasix renal s with invasion of Partial cystectomy scan exuberant mass into left with left ureteral demonstrating histiocytic side of reimplant & psoas no obstruction reaction and bladder and hitch (urology) at 9 months decidua-like ureter after surgery changes Endometriosi Left ureterolysis Normal renal Dense fibrosis s in left (gyn) & ureteral ultrasound 3 encasing left pelvic stent placement months after distal ureter sidewall (urology) surgery biopsy
Cystoscopy, Dense fibrotic Robotic-assisted Lasix renal left ureteral left reaction laparoscopic left Endometriosi scan 9 months CT: severe left stent hydronephrosis surrounding ureteroneocystosto s invading after surgery hydronephrosis placement & with sepsis left distal my & psoas hitch left ureter with no ureteral ureter (urology) obstruction biopsy Renal ultrasound 6 Cystoscopy, Benign Dense fibrotic Robotic-assisted months CT: severe left two ureteral ureter with left reaction laparoscopic left postop: mild hydronephrosis, left stents & stromal hydronephrosis & surrounding ureteroneocystosto left ureteral stricture, no nephrostom endometriosi pyelonephritis left distal my & psoas hitch hydronephrosi contrast excretion y tube s and acute ureter (urology) s, normal Cr, placed ureteritis no pain; lasix scan pending
Page 18 of 22
19
5
6
7
8
9
Cystoscopy, CT: left retrograde left hydronephrosis & pyelogram, hydronephrosis delayed excretion of nephrostom and pelvic pain contrast y tube placement
Dense fibrosis Robotic-assisted surrounding laparoscopic left left ovary & ureteral reimplant & distal ureter, psoas hitch vagina, (urology) sigmoid colon Disseminated pelvic pain, endometriosis Open right menometrorrhagi with ureteroneocystosto None None a, urinary obliterated my with psoas hitch frequency posterior cul(gyn) de-sac 2cm mass CT scan with right Cystoscopy obstructing Ureterolysis (gyn) Right 5cm adnexal mass and ureteral the right ureter with ureteral stent Pyelonephritis and right stent and invading placement (urology) hydronephrosis placement retroperitoneu m Flank pain, nausea and vomiting
CT: right 4x6 cm pelvic mass and right moderate hydroureter
CT: bilateral Abdominal pain hydronephrosis & and dysuria invasive endometrial vs bladder cancer
None
None
Endometriosi negative lasix s involving renal scan 4 vaginal cuff, months after benign surgery urothelium Endometriosi s involving left ovary,
no imaging performed
Endometriosi unknown s involving patient right pelvic followed up sidewall with local mass and urologist right ovary Right ovary, fallopian Fibrotic mass Bilateral ureterolysis tube, & lost to followencasing right (gyn) with stent uterine up ureter placement (urology) serosa endometriosi s Mass invading Open bilateral trigone patient ureteroal reimplant Endometriosi involving followed up (gyn) with ureteral s invading ureteral with local stent placement bladder wall orifices physician (urology) bilaterally
Page 19 of 22
20
10
11
12
13
14
left flank pain
pelvic pain
cystoscopy CT with mass significant with ureteral surrounding mid left retroperitoneal stent ureter fibrosis placement
CT: 11x7 cm R adnexal mass & severe right hydro
none
Fibrosis encasing ureter, inability to place wire
had recurrent pain & large left ovarian Laparoscopic left mass, repeat ureterolysis with Endometriosi surgery with omental wrap s extensive (urology) ureterolysis, no hydro on CT 4.5 yrs later Endometriosi Right ureteral s involving reimplant & Lost to followb/l ovaries, ureterolysis up cervical (urology) stroma Robotic-assisted Negative renal laparoscopic left Endometriosi ultrasound 8 ureteroneocystosto s involving months after my with psoas hitch ureter surgery (urology)
Significant CT: left failed inflammation hydroureteronephro ureteral & 5cm section sis down to distal stent of obliterated left ureter placement ureter Solid mass in CT: 2.8cm pelvic cystoscopy Left no evidence of Flank pain, left pararectal Endometriosi mass, severe left with ureteral ureteroneocystosto recurrence 6 nausea and space s involving hydronephrosis & stent my with psoas hitch years after vomiting adherent to ureter delay nephrogram placement (gyn) surgery left ureter patient Endometriosi Abdominal pain, Left ureteral followed up none none unknown s involving vaginal bleeding reimplant (gyn) with referring apical vagina physician urinary incontinence & ureterovaginal fistula
Page 20 of 22
21
15
8
9
CT: large pelvic pelvic pain with mass invading deep venous bladder, small bowel thrombosis obstruction, bilateral hydro
CT: right 4x6 Flank pain, cm pelvic mass nausea and and right vomiting moderate hydroureter CT: bilateral Abdominal hydronephrosis pain and & invasive dysuria endometrial vs bladder cancer
none
Invasion through bladder wall with obliteration of left ureteral orifice
None
Fibrotic mass encasing right ureter
None
Mass invading trigone involving ureteral orifices bilaterally
Left ureterolysis & failed stent placement requiring nephrostomy tube placement (urology)
follow-up Endometriosi imaging: s involving nonfunctional bladder wall, left kidney so peritoneum, nephrostomy & left ovary tube was removed
Right ovary, Bilateral ureterolysis fallopian tube, (gyn) with stent & uterine lost to follow-up placement (urology) serosa endometriosis Open b/l ureteral reimplant (gyn) ureteral stents (urology)
Endometriosis patient followed invading up with local bladder wall physician
had recurrent pain & large left cystoscopy Laparoscopic left ovarian mass, CT with mass significant with ureteral ureterolysis with repeat surgery 10 left flank pain surrounding mid retroperitoneal Endometriosis stent omental wrap with extensive left ureter fibrosis placement (urology) ureterolysis, no hydro on CT 4.5 yrs later Endometriosis CT: 11x7 cm R Fibrosis Right ureteral involving b/l adnexal mass & encasing 11 pelvic pain none reimplant & ovaries, Lost to follow-up severe right ureter, inability ureterolysis (urology) cervical hydro to place wire stroma
Page 21 of 22
22 urinary CT: left Significant Robotic-assisted incontinence hydronephrosis failed inflammation & laparoscopic left Endometriosis 12 & and hydroureter ureteral stent 5cm section of ureteroneocystostomy involving ureterovaginal to distal left placement obliterated with psoas hitch ureter fistula ureter ureter (urology)
13
CT: 2.8cm pelvic mass, cystoscopy Solid mass in no evidence of Flank pain, Left Endometriosis severe left with ureteral left pararectal recurrence 6 nausea and ureteroneocystostomy involving hydronephrosis stent space adherent years after vomiting with psoas hitch (gyn) ureter & delay placement to left ureter surgery nephrogram
Abdominal 14 pain, vaginal bleeding
15
Negative renal ultrasound 8 months after surgery
pelvic pain with deep venous thrombosis
none
CT: large pelvic mass invading bladder, small bowel obstruction, bilateral hydro
none
none
unknown
Left ureteral reimplant (gyn)
Endometriosis patient followed involving up with referring apical vagina physician
follow-up Invasion Left ureterolysis & Endometriosis imaging: through bladder failed stent placement involving nonfunctional left wall with requiring bladder wall, kidney so obliteration of nephrostomy tube peritoneum, & nephrostomy left ureteral placement (urology) left ovary tube was orifice removed
Page 22 of 22
S0090-4295(16)30524-6 http://dx.doi.org/doi: 10.1016/j.urology.2016.08.016 URL 19972
To appear in:
Urology
Received date: Accepted date:
16-6-2016 10-8-2016
Please cite this article as: Kyle H. Gennaro, Jennifer Gordetsky, Soroush Rais-Bahrami, John P. Selph, Ureteral Endometriosis: Preoperative Risk Factors Predicting Extensive Urologic Surgical Intervention, Urology (2016), http://dx.doi.org/doi: 10.1016/j.urology.2016.08.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 Ureteral Endometriosis: Preoperative risk factors predicting extensive urologic surgical intervention Kyle H. Gennaro BS1, Jennifer Gordetsky MD2,3, Soroush Rais-Bahrami MD2,4, John P. Selph MD2 1. University of Alabama, School of Medicine, Birmingham, AL 2. University of Alabama at Birmingham, Department of Urology, Birmingham, AL 3. University of Alabama at Birmingham, Department of Pathology, Birmingham, AL 4. University of Alabama at Birmingham, Department of Radiology, Birmingham, AL
Conflicts of Interest and Source of Funding: None
* Correspondence: J. Patrick Selph, MD Department of Urology Faculty Office Tower 1120 1720 2nd Avenue South, Birmingham, AL 35294 E-mail: [email protected] Word Count: 2257
Page 1 of 22
2 Abstract Objective: To identify risk factors for urologic reconstruction during surgery for endometriosis
Patients and Methods: We retrospectively identified patients in a surgical pathology database undergoing surgery for endometriosis at our institution from 2010-2015 and subsequently identified those patients with ureteral involvement. Patients were categorized as requiring minimal urologic surgery (e.g. ureterolysis only) or more extensive urologic surgery (e.g. ureteral reimplant). All patients were undergoing surgery for endometriosis, and preoperative risk factors were then identified to predict the need for intraoperative extensive urologic surgery.
Results: Of 386 women undergoing surgery for endometriosis, 82 (21%) women required a surgical procedure on the ureter. 15 of these 82 patients (18.3%) with ureteral involvement required urologic surgical expertise in the form of either ureteral reimplantation with or without psoas hitch, or ureterolysis with ureteral stenting or omental wrap. The remaining 67 underwent ureterolysis alone or no intervention. The presence of flank pain, any urinary symptom, or hydronephrosis on preoperative imaging was a significant predictor of the need for major urologic intervention.
Conclusion: In patients with endometriosis undergoing surgery who complain of flank pain, any urinary symptom, or have hydronephrosis on preoperative imaging, one should have a high suspicion for needing to perform urologic reconstruction during
Page 2 of 22
3 surgery. Planning for this additional operation can afford the opportunity for appropriate urologic consultation and patient counseling. Key Words: endometriosis, ureteral obstruction, urogenital surgical procedures, ureteral stricture Introduction: Endometriosis is the presence of endometrial-type tissue in a location outside of the uterus, most commonly the pelvic peritoneum, ovaries, and rectovaginal septum [1]. The literature estimates the prevalence of endometriosis to be between 1 to 10% in reproductive age women [1-5]. Endometriosis involves the urinary tract in approximately 0.3 to 12% of patients, with the bladder, ureter, and kidneys most commonly affected [6,7]. Ureteral endometriosis (UE) is often classified as intrinsic or extrinsic based on histologic findings. Extrinsic UE, which includes periureteral or serosal involvement, is the most common type and exists in 80% of cases. Intrinsic UE, which is the presence of endometrial tissue in the mucosa or muscularis propria, is more rare [8]. Presentation of UE is variable depending on location and extent of the disease. Patients may present with abdominal pain or urinary obstruction; however, 50% of patients lack urinary symptoms at the time of their presentation, making the diagnosis of UE difficult [6,8,9]. Surgical management is commonly used to treat UE, with different levels of intervention required based on the extent of disease [6]. We investigated preoperative clinical factors in patients with UE to predict those who might need more extensive urologic surgical intervention.
Patients and Methods:
Page 3 of 22
4 We performed an Institutional Review Board-approved search of our surgical pathology database from 2010-2015 for cases of endometriosis. Clinical, pathologic, and surgical data were analyzed for patients who had pathology-proven endometriosis and macroscopic intraoperative surgical findings of endometriosis involving the ureter. The study population was divided into two groups. Group 1, the urologic intervention group, consisted of patients that required any urologic surgical expertise, including ureteral stenting (with or without ureterolysis), ureteroneocystostomy with or without psoas hitch, or ureterolysis with omental wrap. Group 2, the non-urologic surgical intervention group, included patients that required only ureterolysis or no urologic surgical intervention. The two groups were compared regarding preoperative clinical symptoms, imaging, and laboratory data. All statistical analysis was performed in RStudio Version 0.99.896, and the epiR package was used to calculate odds ratios and for chi-squared testing. Continuous data was expressed in terms of mean, standard deviation, and range. Categorical data was expressed as percentages. Continuous data were assessed for normality using graphical representation and compared using a t-test. Categorical data were compared using the chi-squared test and a p-value of <0.05 was considered statistically significant for all tests. Results: We identified 405 cases with pathologically proven endometriosis (figure 1). 19 cases were excluded: 4 without any clinical information, and 15 that were repeat biopsies on the same patient. A total of 386 patients with histologic evidence of endometriosis were included. Of these, 82 (21%) underwent a procedure involving the
Page 4 of 22
5 ureter and/or had macroscopic surgical findings of ureteral involvement. The average age of our patient population was 41.3 years (Table 1). 27 (32.9%) patients had a previous history of endometriosis, with an average age at diagnosis of 31.7 years (range 14-53). Eleven (13.4%) patients (7 surgically-induced, 2 natural, 2 unknown cause) were postmenopausal, with an average age at menopause of 42.3 years (range 36-45). 27/82 (32.9%) patients had >1 site of biopsy proven endometriosis. 19/82 (23.1%) patients had a prior history of hysterectomy, 10/82 (12.2%) unilateral oophorectomy, and 2/82 (2.4%) bilateral oophorectomy. 2/82 (2.4%) patients had previously received systemic hormone replacement therapy. The most common presenting symptom in both Group 1 and Group 2 was abdominal pain; 40.0% and 46.3% respectively (Table 2). The presence of any urinary symptom including urinary urgency, frequency, hematuria, dysuria (OR 7.88, 95% CI 1.80, 34.4), flank pain (OR 4.58; 95% CI 1.26,16.69), or hydronephrosis (OR 76.7; 95% CI 13.4, 439) was significantly more common in Group 1. The differences between physical exam findings, white blood cell count, and creatinine were not statistically significant between the two groups. 37/82 (45.1%) patients received preoperative imaging of the urinary tract. Radiologic evidence of hydronephrosis was identified in 18/37 (43.2%) patients. 13/18 (72.2%) patients with preoperative hydronephrosis required a more extensive urologic intervention. Of the 5 who did not undergo extensive urologic surgery, 4 had ureterolysis alone, and one underwent no intervention. In comparing patients who had hydronephrosis with those who did not, patients with hydronephrosis were more likely to
Page 5 of 22
6 have urinary symptoms (OR 9.67; 95% CI 2.12, 44.15) and/or flank pain (OR 4.58; 95% CI 1.26, 16.69). As highlighted in Table 3, 15/82 (18.3%) patients required more extensive urologic intervention (Group 1). Among these, 10/15 (60.0%) patients were treated with ureteroneocystostomy, of which 7/10 (70%) were left-sided, 2/10 (20%) were rightsided, and 1/10 (10%) was bilateral. One patient required partial cystectomy in addition to ureteroneocystostomy, and all patients received ureteral stents. Three of the reimplantations were done in conjunction with a psoas hitch. 3/15 (20.0%) patients were treated with ureterolysis and stenting of which 1 (33%) was left-sided, 1 (33%) was right-sided, and 1 (33%) was bilateral. Of the remaining two patients in Group 1, one was treated with ureterolysis with omental wrap, and the second had ureterolysis with failed ureteral stenting that ultimately required nephrostomy tube placement and was diagnosed with a non-functioning kidney. No patients were treated with a ureteroureterostomy. One patient who had undergone left ureterolysis with omental wrapping developed a recurrent pelvic mass requiring repeat ureterolysis. Follow-up imaging demonstrated no hydronephrosis over 4 years later. No other patients from this group required additional surgeries related to their initial pathology, though a second patient did develop endometrial carcinoma at a later date causing recurrence of hydronephrosis that was conservatively managed. 67/82 (81.7%) patients were managed conservatively with only ureterolysis or no intervention (Group 2). 58/67 (87.8%) patients were treated with ureterolysis, 1/67 (1.5%) patient had excision of endometriotic tissue abutting the ureter, and 8/67 (11.9%) had macroscopic findings suggesting ureteral involvement but received no intervention.
Page 6 of 22
7 Of those who received ureterolysis, 20/58 (34.5%) were bilateral, 18/58 (31.0%) were left-sided, and 20/58 (34.5%) were right-sided. One patient that received conservative management was found to have a right hydroureter 3 months after surgery due to a periureteral mass, suggesting recurrence of disease.
Page 7 of 22
8 Discussion: In patients with endometriosis, rates of urinary tract involvement range from 0.3 to 12%, with UE reported in 4-5% of cases [5,6,10,12]. Our study demonstrated ureteral involvement in 21% of cases of endometriosis, which is significantly higher than previous studies. In our experience, UE more commonly involved the left side, which has been similarly reported in other studies [5-14]. UE is difficult to diagnose due to the wide variation of presentations and lack of physical exam findings [15]. The majority of our cases were in premenopausal women near age 40 presenting with non-specific abdominal pain. In a study by Nezhat et al, UE was unsuspected in two-thirds of surgical cases [14]. Similarly, in our study endometriosis was listed as the indication for surgery in only 26.8% of cases. Furthermore, we found preoperative physical examination and serum creatinine to be unhelpful for predicting the necessity of more extensive urologic intervention. We found that the presence of preoperative urinary symptoms, flank pain, or hydronephrosis significantly predicted the necessity of more extensive intervention. These symptoms are similar to the findings in Gabriel and colleagues study of their endometriosis series, where the presence of dysuria, hematuria, or frequent daytime urination and recurrent urinary tract infections were predictive of ureteral involvement of endometriosis [16]. UE is classically described as being intrinsic or extrinsic [8]. In many cases, extrinsic involvement of the ureter can be cured with simple ureterolysis, while intrinsic involvement necessitates more extensive surgery, including ureteroureterostomy or ureteroneocystostomy. Some groups have suggested varying levels of surgical treatment based on intrinsic versus extrinsic disease with conservative
Page 8 of 22
9 management (including ureterolysis) recommended for extrinsic disease and more extensive resections (including ureteral resection) recommended for intrinsic disease [6]. Knabben and colleagues recently noted that 98% of patients in their series had ureteral involvement treated with ureterolysis alone, and others have shown success rates of 85-90% with ureterolysis only [13, 17-19]. Although extrinsic type UE is thought to be more common, some studies have shown that the true rates of intrinsic disease may be greatly underestimated [8,12]. A limitation of the extrinsic and intrinsic classification system, however, is that it is a postoperative diagnostic system - a definitive diagnosis of intrinsic UE requires a ureteral tissue sample, which is typically not performed preoperatively. Even if preoperative ureteral biopsy is performed, it may not be positive if endometrial tissue has not invaded into the lumen or mucosa of the ureter. In our study, three patients who underwent ureteral reimplantation had preoperative ureteroscopic biopsy to rule out malignancy, and none of the samples were positive for endometriosis. Only a final pathologic specimen was positive for endometriosis. For preoperative planning purposes, using intrinsic versus extrinsic disease as a prognostic indicator for the level of intervention is not ideal. We found that the presence of preoperative urinary symptoms, flank pain, or hydronephrosis significantly predicted the necessity of more extensive intervention. For this reason, we recommend the involvement of a urologic surgeon in the preoperative planning for patients with these symptoms and suspected endometriosis. Others have looked at the histologic appearance of the endometrioma to try and predict the need for urologic surgery. Serachiolli and colleagues found no difference in the risk of ureteral reconstruction for patients who had either endometrial
Page 9 of 22
10 glands or fibrotic endometrial tissue present in the pathologic specimen [20]. Again, however, this assessment can only be done with a biopsy specimen, which is usually obtained after radical surgery has been performed. In multiple cases in our series, women presented with primarily urologic complaints – flank pain, recurrent kidney infections, and obstruction on renal scintigraphy. In these cases, ureteroscopic biopsy was performed to rule out malignancy, but in all three cases in which this procedure was done, the biopsy specimen was negative for endometriosis. Others have found that ureteroscopic biopsy can be inconclusive in UE [21]. This poor sensitivity of ureteroscopic biopsy for endometriosis argues that a more definitive imaging modality might help better guide the surgeon in preoperative planning and patient counseling. Several studies have attempted to analyze the best preoperative imaging modality for detecting UE. In a study by Pateman et al, 2% of women with chronic pelvic pain had evidence of UE [5]. They showed that ultrasound with attention to the urinary tract has a high sensitivity (92%) and specificity (100%) for UE. For this reason they recommended routine assessment of the urinary tract in all women with endometriosis. Given the low rate of UE, however, this may not be necessary as a broad screening modality in all patients, but perhaps patients with other predictors of urinary tract involvement, e.g., urinary symptoms or flank pain. In our study, 37 patients underwent preoperative imaging of the urinary tract. Of these 37 women, 18 were found to have hydronephrosis. Predictive factors for hydronephrosis were identical to those predicting the need for more extensive surgery. Urinary symptoms and flank pain, with an OR of 9.67 and 4.58 respectively (Table 3), were the only statistically significant predictors of
Page 10 of 22
11 hydronephrosis. Physical exam findings, preoperative WBC, and preoperative Cr, were not significantly different between patients with and without hydronephrosis. Additionally, routine assessment of the urinary tract for endometriosis does not seem necessary in our patient population, as ultimately only 15/386 (3.9%) required significant urologic intervention. On the contrary, most patients receive pelvic ultrasound as a diagnostic tool in endometriosis, so the short additional time and cost it takes to evaluate the urinary tract simultaneously may be prudent. Sillou’s group recently published their experience with preoperative magnetic resonance (MR) imaging in endometriosis and found a sensitivity of 91% for diagnosing intrinsic UE, while also noting that in cases where the endometrioma surrounded less than 180° of the ureter, intrinsic UE was found in less than 10% of cases [7]. Certainly this information could aid the surgeon in preoperative planning, though the cost and relative rarity of intrinsic UE argues against routine MR before endometriosis surgery. The goals of surgical treatment for UE are to relieve the obstruction, preserve renal function, and prevent future recurrence of disease [6]. Due to the low prevalence of UE, there have been no prospective studies to date comparing the outcomes between conservative management and more extensive urologic management. Some studies suggest that conservative management with ureterolysis is adequate and should be offered initially to all patients with UE [10,13]. Ghezzi and colleagues reported that all patients were symptom free and showed no evidence of obstruction on intravenous pyelogram 3-months after ureterolysis; however, 12.1% of their patients required one or more operations for disease recurrence [13]. Knabben’s group found a 98% success rate with ureterolysis alone [17]. Antonelli’s group noted only one recurrence out of 30
Page 11 of 22
12 patients with ureteral or bladder endometriosis when radical removal of extensive and obstructive genitourinary endometriosis was performed [11]. Our data seem to support both notions, as only one patient that received management with ureterolysis alone was found to have a right hydroureter three months after surgery, and a second patient developed a recurrent pelvic mass requiring resection and repeat ureterolysis. Limitations of our study include the lack of a standardized protocol for patients with endometriosis. Because it is not routine practice at our institution to perform imaging of the urinary tract prior to surgery for endometriosis, it is possible that more patients with ureteral endometriosis had preoperative hydronephrosis but were not identified; however, review of the operative reports indicated that these patients did not have any hydroureter, and thus the likelihood of a positive imaging test preoperatively was low. Because this is a retrospective study, it was not designed to ascertain the success of urologic reconstruction in these patients, and thus our follow-up is short (mean seven months). It is possible some patients who underwent a more extensive reconstruction developed recurrent urinary obstruction. Lastly, we noted a higher rate (21%) of ureteral endometriosis than in many other published studies – this may be a reflection of the fact that we serve as a tertiary referral center for complicated endometriosis, or simply a reflection of how different surgeons characterize ureteral “involvement.”
Conclusion: The diagnosis of UE is difficult with few guidelines on surgical management. The presence of urinary symptoms, flank pain, and/or hydronephrosis increases the
Page 12 of 22
13 likelihood of requiring a more extensive surgical intervention. We recommend the involvement of a urologic surgeon in the preoperative planning for patients with these symptoms and suspected endometriosis.
Page 13 of 22
14 References 1. Giudice LC. Clinical practice. Endometriosis. N Engl J Med. 2010;362(25):2389-2398. 2. Barbieri RL. Etiology and epidemiology of endometriosis. Am J Obstet Gynecol. 1990;162(2):565-567. 3. Dunselman GA, Vermeulen N, Becker C, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29(3):400-412. 4. Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstet Gynecol Clin North Am. 1997;24(2):235-258. 5. Pateman K, Holland TK, Knez J, et al. Should a detailed ultrasound examination of the complete urinary tract be routinely performed in women with suspected pelvic endometriosis? Hum Reprod. 2015;30(12):2802-2807. 6. Maccagnano C, Pellucchi F, Rocchini L, et al. Ureteral endometriosis: proposal for a diagnostic and therapeutic algorithm with a review of the literature. Urol Int. 2013;91(1):1-9. 7. Sillou S, Poirée S, Millischer AE, et al. Urinary endometriosis: MR imaging appearance with surgical and histological correlations. Diagn Interv Imaging. 2015;96(4):373-381. 8. Yohannes P. Ureteral endometriosis. J Urol. 2003;170(1):20-25. 9. Stepniewska A, Grosso G, Molon A, et al. Ureteral endometriosis: clinical and radiological follow-up after laparoscopic ureterocystoneostomy. Hum Reprod. 2011;26(1):112-116. 10. Donnez J, Nisolle M, Squifflet J. Ureteral endometriosis: a complication of rectovaginal endometriotic (adenomyotic) nodules. Fertil Steril. 2002;77(1):32-37.
Page 14 of 22
15 11. Antonelli A, Simeone C, Zani D, et al. Clinical aspects and surgical treatment of urinary tract endometriosis: our experience with 31 cases. Eur Urol. 2006;49(6):10931097; discussion 1097-1098. 12. Chapron C, Chiodo I, Leconte M, et al. Severe ureteral endometriosis: the intrinsic type is not so rare after complete surgical exeresis of deep endometriotic lesions. Fertil Steril. 2010;93(7):2115-2120. 13. Ghezzi F, Cromi A, Bergamini V, et al. Outcome of laparoscopic ureterolysis for ureteral endometriosis. Fertil Steril. 2006;86(2):418-422. 14. Nezhat C, Nezhat F, Nezhat CH, et al. Urinary tract endometriosis treated by laparoscopy. Fertil Steril. 1996;66(6):920-924. 15. Ghezzi F, Cromi A, Bergamini V, et al. Management of ureteral endometriosis: areas of controversy. Curr Opin Obstet Gynecol. 2007;19(4):319-324. 16. Gabriel B, Nassif G, Trompoukis P, et al. Prevalence and Management of Urinary Tract Endometriosis: A Clinical Case Series. Urology 2011; Dec;78(6):1269-74. 17. Knabben L, Imboden S, Fellmann B, et al. Urinary tract endometriosis in patients with deep infiltrating endometriosis: prevalence, symptoms, management, and proposal for a new clinical classification. Fertil Steril. 2015;103(1):147-152. 18. Soriano D, Schonman R, Nadu A, et al. Multidisciplinary team approach to management of severe endometriosis affecting the ureter: long-term outcome data and treatment algorithm. J Minim Invasive Gynecol. 2011;18(4):483-488. 19. Uccella S, Cromi A, Casarin J, et al. Laparoscopy for ureteral endometriosis: surgical details, long-term follow-up, and fertility outcomes. Fertil Steril. 2014;102(1):160-166.e162.
Page 15 of 22
16 20. Seracchioli R, Raimondo D, Di Donato N, et al. Histological evaluation of ureteral involvement in women with deep infiltrating endometriosis: analysis of a large series. Hum Reprod. 2015; 30(4):833-839. 21. Seyam R, Mokhtar A, Al Taweel W, et al. Isolated ureteric endometriosis presenting as a ureteric tumor. Urol Ann. 2014;6(1):94-97.
Figure 1. Patient selection for surgical intervention cohorts Table 1. Patient demographics Average age (years; sd, range) Average follow-up (months; range) Race Caucasian African-American Hispanic Endometriosis Biopsy Location Ureter Bladder Wall Perivaginal Cuff Cervix Uterus Fallopian tube Paratubal Ovary Peritoneum Retroperitoneum Bowel Omentum
41.3 +/- 8.4 (22.4-68.5) 7.5 +/- 11.0 (0.0-56.2) n (%) 45 (54.9%) 33 (40.2%) 4 (4.9%) 5 (6.1%) 4 (4.9%) 1 (1.2%) 5 (6.1%) 16 (19.5%) 6 (7.3%) 3 (3.7%) 56 (68.3%) 12 (14.6%) 2 (2.4%) 6 (7.3%) 1 (1.2%)
Page 16 of 22
17 Table 2. Comparison of patients requiring urologic surgical intervention and no urologic surgical Intervention Extensive Surgical Intervention (n=15)
Minimal or No Surgical Intervention (n=67)
n (%)
n (%)
OR (p-value, 95% CI)
Urinary Symptoms
5 (33.3%)
4 (6.0%)
7.88 (p=0.002, 1.80-34.40)
Dyspareunia
1 (6.7%)
9 (13.4%)
0.46 (p=0.469, 0.05-3.94)
Dysmenorrhea
3 (20.0%)
17 (25.4%)
0.74 (p=0.661, 0.19-2.92)
Pelvic Pain
5 (33.3%)
26 (38.8%)
0.79 (p=0.693, 0.24-2.57)
Abdominal Pain
6 (40.0%)
31 (46.3%)
0.77 (p=0.659, 0.25-2.42)
Flank Pain
5 (33.3%)
7 (10.4%)
4.29 (p=0.023, 1.14-16.18)
Back Pain
0 (0.0%)
2 (3.0%)
0.00 (p=0.915, 0.00-NA)
Abnormal Uterine Bleeding 3 (20.0%)
21 (31.3%)
0.55 (p=0.383, 0.14-2.15)
Asymptomatic
6 (9.0%)
0.00(p=0.402, 0.00-NA)
Abdomen/Pelvis Tender to 3 (20.0%) Palpation
23 (34.3%)
0.48 (p=0.281, 0.12-1.87)
Vaginal Bleeding
3 (20.0%)
7 (10.4%)
2.14 (p=0.307, 0.48-9.49)
Abdominal/Pelvic Mass
2 (13.3%)
26 (38.8%)
0.24 (p=0.06, 0.05-1.16)
Laboratory Findings
Mean +/- st. dev (range)
Mean +/- st. dev (range)
Average Cr
0.82 +/- 0.19 (0.5-1.3)
0.76 +/- 0.13 (0.5-1.1))
p=0.20
Average WBC
9.07+/- 4.62 (2.9-19.0)
7.80 +/- 2.36 (3.6-16.5)
p=0.14
13 (86.7%)
5 (7.8%)*
76.70 (p<0.001, 13.38-439.76)
Presenting Symptoms
0 (0.0%)
Physical Exam Findings
Imaging Findings Hydronephrosis
*5/64, 3 patients with unknown pre-operative hydronephrosis status.
Page 17 of 22
18 Table 3. Characteristics of patients requiring urologic surgery for endometriosis
Cas e
1
2
3
4
Preoperativ e Urologic Interventio n Clot evacuation, gross hematuria CT showing mass Bladder requiring invading bladder, biopsy and continuous vagina, and left left ureteral bladder irrigation hydronephrosis stent placement Presenting Symptoms
flank pain and dysmenorrhea
Upper tract Imaging preop
CT scan with left hydronephrosis
None
Intraoperativ Surgery Performed Pathology e Findings (Team performing) Findings
Followup
endometriosi Direct lasix renal s with invasion of Partial cystectomy scan exuberant mass into left with left ureteral demonstrating histiocytic side of reimplant & psoas no obstruction reaction and bladder and hitch (urology) at 9 months decidua-like ureter after surgery changes Endometriosi Left ureterolysis Normal renal Dense fibrosis s in left (gyn) & ureteral ultrasound 3 encasing left pelvic stent placement months after distal ureter sidewall (urology) surgery biopsy
Cystoscopy, Dense fibrotic Robotic-assisted Lasix renal left ureteral left reaction laparoscopic left Endometriosi scan 9 months CT: severe left stent hydronephrosis surrounding ureteroneocystosto s invading after surgery hydronephrosis placement & with sepsis left distal my & psoas hitch left ureter with no ureteral ureter (urology) obstruction biopsy Renal ultrasound 6 Cystoscopy, Benign Dense fibrotic Robotic-assisted months CT: severe left two ureteral ureter with left reaction laparoscopic left postop: mild hydronephrosis, left stents & stromal hydronephrosis & surrounding ureteroneocystosto left ureteral stricture, no nephrostom endometriosi pyelonephritis left distal my & psoas hitch hydronephrosi contrast excretion y tube s and acute ureter (urology) s, normal Cr, placed ureteritis no pain; lasix scan pending
Page 18 of 22
19
5
6
7
8
9
Cystoscopy, CT: left retrograde left hydronephrosis & pyelogram, hydronephrosis delayed excretion of nephrostom and pelvic pain contrast y tube placement
Dense fibrosis Robotic-assisted surrounding laparoscopic left left ovary & ureteral reimplant & distal ureter, psoas hitch vagina, (urology) sigmoid colon Disseminated pelvic pain, endometriosis Open right menometrorrhagi with ureteroneocystosto None None a, urinary obliterated my with psoas hitch frequency posterior cul(gyn) de-sac 2cm mass CT scan with right Cystoscopy obstructing Ureterolysis (gyn) Right 5cm adnexal mass and ureteral the right ureter with ureteral stent Pyelonephritis and right stent and invading placement (urology) hydronephrosis placement retroperitoneu m Flank pain, nausea and vomiting
CT: right 4x6 cm pelvic mass and right moderate hydroureter
CT: bilateral Abdominal pain hydronephrosis & and dysuria invasive endometrial vs bladder cancer
None
None
Endometriosi negative lasix s involving renal scan 4 vaginal cuff, months after benign surgery urothelium Endometriosi s involving left ovary,
no imaging performed
Endometriosi unknown s involving patient right pelvic followed up sidewall with local mass and urologist right ovary Right ovary, fallopian Fibrotic mass Bilateral ureterolysis tube, & lost to followencasing right (gyn) with stent uterine up ureter placement (urology) serosa endometriosi s Mass invading Open bilateral trigone patient ureteroal reimplant Endometriosi involving followed up (gyn) with ureteral s invading ureteral with local stent placement bladder wall orifices physician (urology) bilaterally
Page 19 of 22
20
10
11
12
13
14
left flank pain
pelvic pain
cystoscopy CT with mass significant with ureteral surrounding mid left retroperitoneal stent ureter fibrosis placement
CT: 11x7 cm R adnexal mass & severe right hydro
none
Fibrosis encasing ureter, inability to place wire
had recurrent pain & large left ovarian Laparoscopic left mass, repeat ureterolysis with Endometriosi surgery with omental wrap s extensive (urology) ureterolysis, no hydro on CT 4.5 yrs later Endometriosi Right ureteral s involving reimplant & Lost to followb/l ovaries, ureterolysis up cervical (urology) stroma Robotic-assisted Negative renal laparoscopic left Endometriosi ultrasound 8 ureteroneocystosto s involving months after my with psoas hitch ureter surgery (urology)
Significant CT: left failed inflammation hydroureteronephro ureteral & 5cm section sis down to distal stent of obliterated left ureter placement ureter Solid mass in CT: 2.8cm pelvic cystoscopy Left no evidence of Flank pain, left pararectal Endometriosi mass, severe left with ureteral ureteroneocystosto recurrence 6 nausea and space s involving hydronephrosis & stent my with psoas hitch years after vomiting adherent to ureter delay nephrogram placement (gyn) surgery left ureter patient Endometriosi Abdominal pain, Left ureteral followed up none none unknown s involving vaginal bleeding reimplant (gyn) with referring apical vagina physician urinary incontinence & ureterovaginal fistula
Page 20 of 22
21
15
8
9
CT: large pelvic pelvic pain with mass invading deep venous bladder, small bowel thrombosis obstruction, bilateral hydro
CT: right 4x6 Flank pain, cm pelvic mass nausea and and right vomiting moderate hydroureter CT: bilateral Abdominal hydronephrosis pain and & invasive dysuria endometrial vs bladder cancer
none
Invasion through bladder wall with obliteration of left ureteral orifice
None
Fibrotic mass encasing right ureter
None
Mass invading trigone involving ureteral orifices bilaterally
Left ureterolysis & failed stent placement requiring nephrostomy tube placement (urology)
follow-up Endometriosi imaging: s involving nonfunctional bladder wall, left kidney so peritoneum, nephrostomy & left ovary tube was removed
Right ovary, Bilateral ureterolysis fallopian tube, (gyn) with stent & uterine lost to follow-up placement (urology) serosa endometriosis Open b/l ureteral reimplant (gyn) ureteral stents (urology)
Endometriosis patient followed invading up with local bladder wall physician
had recurrent pain & large left cystoscopy Laparoscopic left ovarian mass, CT with mass significant with ureteral ureterolysis with repeat surgery 10 left flank pain surrounding mid retroperitoneal Endometriosis stent omental wrap with extensive left ureter fibrosis placement (urology) ureterolysis, no hydro on CT 4.5 yrs later Endometriosis CT: 11x7 cm R Fibrosis Right ureteral involving b/l adnexal mass & encasing 11 pelvic pain none reimplant & ovaries, Lost to follow-up severe right ureter, inability ureterolysis (urology) cervical hydro to place wire stroma
Page 21 of 22
22 urinary CT: left Significant Robotic-assisted incontinence hydronephrosis failed inflammation & laparoscopic left Endometriosis 12 & and hydroureter ureteral stent 5cm section of ureteroneocystostomy involving ureterovaginal to distal left placement obliterated with psoas hitch ureter fistula ureter ureter (urology)
13
CT: 2.8cm pelvic mass, cystoscopy Solid mass in no evidence of Flank pain, Left Endometriosis severe left with ureteral left pararectal recurrence 6 nausea and ureteroneocystostomy involving hydronephrosis stent space adherent years after vomiting with psoas hitch (gyn) ureter & delay placement to left ureter surgery nephrogram
Abdominal 14 pain, vaginal bleeding
15
Negative renal ultrasound 8 months after surgery
pelvic pain with deep venous thrombosis
none
CT: large pelvic mass invading bladder, small bowel obstruction, bilateral hydro
none
none
unknown
Left ureteral reimplant (gyn)
Endometriosis patient followed involving up with referring apical vagina physician
follow-up Invasion Left ureterolysis & Endometriosis imaging: through bladder failed stent placement involving nonfunctional left wall with requiring bladder wall, kidney so obliteration of nephrostomy tube peritoneum, & nephrostomy left ureteral placement (urology) left ovary tube was orifice removed
Page 22 of 22