Endometriosis is a benign disease defined by the presence of ectopic endometrial glands and stroma, often associated with pelvic pain and infertility. The incidence of endometriosis in the reproductive age group is estimated at 3-10% [1].

Endometriosis involving the urinary tract includes the presence of endometrial tissue within or around the bladder, ureters, urethra, or kidney. Ureteric endometriosis is usually unilateral, most commonly involving a small segment of distal left ureter [2, 3]. It is often associated with retroperitoneal fibrosis and peri-ureteric cicatrization. It has an incidence of < 1% [3–5].

Ureteric obstruction resulting in hydronephrosis is a rare manifestation of ureteric endometriosis. It occurs as a consequence of intrinsic involvement within the ureter of endometriosis, or from extrinsic compression of the ureter by a pelvic endometrioma [6]. In cases of intrinsic ureteric endometriosis, ectopic endometrial tissue is present within the muscularis propria, lamina propria or ureteral lumen [6] In extrinsic cases endometriosis occurs within the ureteral adventitia and adjacent soft tissues only [6]. Extrinsic involvement is approximately 4 times more common than intrinsic disease [6].

It has been shown that pharmacological management with GnRH analogues or aromatase inhibitors alone, in treating ureteric endometriosis, will not prevent the development of hydronephrosis and renal deterioration [7, 8]. Despite this, the optimal surgical management of the obstructed ureteric segment has yet to be defined. The current debate, in cases of hydronephrosis associated with ureteric endometriosis, is whether a primary resection with re-implantation or re-anastomosis should be performed, versus a more conservative ureterolysis of the effected ureteric segment.

Given the rarity of the condition there is limited evidence in the literature. There have been several case series published with relatively small numbers and only short to intermediate follow up. Despite this, there is a growing body of evidence favouring ureterolysis as the treatment of choice [3, 4, 9, 10]. We have retrospectively reviewed the records of 126 patients who underwent ureterolysis for ureteric endometriosis by a single gynaecologist. Of these, 13 had macroscopic hydroureter at laparoscopy. This article discusses the surgical management and follow up of these 13 cases, highlighting the favourability of conservative ureterolysis.

Laparoscopy is widely accepted as the gold standard of surgical management of ureteric endometriosis due to the enhanced vision [3–5, 10–12]. High power density monopolar electrosurgical instruments, combined with excellent vision have meant that laparoscopic dissection around the ureter can occur with minimal complications [3, 4, 10].

In cases of ureteric obstruction, controversy exists as to whether the effected ureteric segment should be primarily resected, or managed with conservative ureterolysis. Few details have been given as to the actual operative findings, and many cases of obstruction have been managed with ureteric re-implantation [11]. We have retrospectively reviewed the records of 13 cases, and it is our feeling that the majority of patients have an obstruction at the level of the uterine artery, as it crosses the ureter. Endometriosis is a cicatrizing disease and we have been impressed that the disease process often seems to follow the course of adjacent blood vessels such as the uterine artery. The close proximity of the ureter to the uterine artery means that any cicatrizing process adjacent to or surrounding the artery will result in obstruction of the ureter. The fibrotic process forms a constricting band over the ureter and the distal ureter is often free of disease and therefore most of these situations should be resolved with simply excising the affected tissue without recourse to re-implantation.

The current literature relating to hydronephrosis secondary to ureteric endometriosis, consists of mostly case reports and several recent case series. Ghezzi et al (2007), have reported a prospective multi-centre cohort study involving 33 patients with a median follow up of 16 months. In this study only patients with moderate to severe hydronephrosis were included. These authors reported that 85% of the patients in their cohort were successfully managed with ureterolysis as the primary procedure [5]. Mereu, et al 2008, in a single-centre prospective case series with fifty-six patients, all with moderate to severe hydronephrosis, only performed ureterolysis in 62% of cases, while 38% underwent a ureteric resection [4]. Schneider et al (2006) have reported the findings of a prospective case series consisting of 22 women with urinary tract endometriosis. In seven of the 22, endometriosis involved the ureter, including six with distal ureteral endometriosis, and one with endometriosis involving a ureteral stump. Four of the women were suffering mild renal impairment at the time of diagnosis. In this series, six of the seven women underwent segmental ureteric resection with psoas hitching and re-implantation, and excision of the ureteral stump. These authors report no long term complication of relapse at 20 months follow up [13].

In our series ten of the patients underwent ureterolysis alone and three were managed with ureterolysis and a temporary JJ ureteric stent. Therefore 77% of the cohort were successfully treated without resection of the ureteric segment. Of the three patients that were managed with a ureteric resection, two underwent a psoas hitch with a submucosal tunnel of the ureter into the bladder, while one underwent a primary ureteric re-anastomosis. In these cases ureterolysis alone was not sufficient to free the ureter of disease therefore a segmental resection was performed. All of the patients in this series demonstrated a resolution of hydronephrosis at six months follow up.

The seemingly high open conversion rate of 42% reported in this series can be explained by the extensive degree of endometriosis seen in these patients and the need for concomitant procedures. There were two complications in this series of 126 patients (1.5%). These were, one inadvertent ureteric injury that was managed with a ureteric stent, and an injury to the iliac vein which required open conversion. No long term complications occurred.

Histological examination of the three resected ureteric specimens revealed extrinsic endometriosis. This finding is in keeping with the literature, whereby the vast majority of cases have demonstrated extrinsic ureteric endometriosis, and should therefore be suitable for ureterolysis as primary management [3, 4, 6, 14].

When dealing with uncommon disorders it is difficult to evaluate the efficacy of the evidence. The limitations stem from small cohort numbers, due to the rarity of the condition. The majority of evidence in the literature consists of case series, and therefore subject to bias. Moreover, the natural history of endometriosis is largely unknown. If it tends to become inactive then this would favour a ureteric conserving approach.

A major limitation of this study was the short term follow up and no conclusion can be made regarding the risk of recurrent disease. Despite this the data presented in this case series favours ureterolysis as a primary management of ureteric endometriosis, with minimal morbidity.

Dr Ian AR Smith is Associate Conjoint Lecturer in Faculty of Medicine at the University of New South Wales and University of Western Sydney. Dr Michael Cooper is Clinical Associate Professor in the Department of Obstetrics and Gynaecology at the University of Sydney.