Pelvic endometriosis and hydroureteronephrosis

ENDOMETRIOSIS Pelvic endometriosis and hydroureteronephrosis Luca Carmignani, M.D.,a Paolo Vercellini, M.D.,b Matteo Spinelli, M.D.,c Eleonora Fontana, M.D.,b Giada Frontino, M.D.,b and Luigi Fedele, M.D.b a

Chief of Urology Unit, I.R.C.C.S. Policlinico San Donato, University of Milan; b Department of Obstetrics and Gynecology, Fondazione ‘‘Policlinico-Mangiagalli-Regina Elena’’, University of Milan; and c Urology Unit, Fondazione ‘‘Policlinico-Mangiagalli-Regina Elena’’, University of Milan, Milan, Italy

Objective: To assess whether routine renal ultrasonography may be recommended in all patients with pelvic endometriosis, in order to avoid silent ureteral involvement of the disease. Design: Retrospective descriptive study. Settings: Tertiary center for the treatment of endometriosis at the Department of Obstetrics and Gynecology of the State University of Milan, Milan, Italy. Patient(s): Seven-hundred-fifty patients with a primary diagnosis of endometriosis, between January 2005 and July 2007. Intervention(s): Routine urinary ultrasound; recording of patient history, signs, and symptoms; gynecologic examination; blood and urinary analyses; magnetic resonance imaging; spiral multislice computerized tomography. Main Outcome Measure(s): Symptoms and signs of ureterohydronephrosis; diagnosis of ureterohydronephrosis. Result(s): Twenty-three patients (3%) of all 750 patients with endometriosis had associated ureterohydronephrosis diagnosed at renal ultrasound. Symptoms secondary to ureteral and renal involvement were present in 10 patients (43.5%); 6 reported lumbar pain (26.1%) and 4 patients (17.4%) had renal colic. Conclusion(s): In our study, the high number (56.5%) of asymptomatic ureteral involvement in patients with known pelvic endometriosis seems to warrant the need for further investigations regarding the possibility to avoid the high percentage of silent renal losses. Unfortunately there appears to be no specific risk factor to allow for early suspicion nor a validated preventive diagnostic and therapeutic program. It remains to be evaluated whether urinary ultrasound ensures a beneficial cost-benefit ratio if employed on a routine basis. (Fertil Steril 2010;93:1741–4. 2010 by American Society for Reproductive Medicine.) Key Words: Pelvic endometriosis, hydronephrosis, ultrasound, ureteral endometriosis, renal colic, lumbar pain

Urinary tract involvement in patients with endometriosis occurs in approximately 1%–2% of cases; the bladder is the most frequently involved organ, followed by the ureters and the kidneys with a proportion of 40:5:1 (1, 2). Patients with ureteral endometriosis often refer to nonspecific symptoms at clinical presentation, therefore posing differential diagnostic problems and a relatively high risk for subsequent loss of renal function (1). Ureteral endometriosis can ultimately cause the loss of renal function owing to ureteral urine flow obstruction caused by the endometriotic tissue enclosing the distal portion of the ureter.

Received May 21, 2008; revised December 5, 2008; accepted December 11, 2008; published online February 6, 2009. L.C. has nothing to disclose. P.V. has nothing to disclose. M.S. has nothing to disclose. E.F. has nothing to disclose. G.F. has nothing to disclose. L.F. has nothing to disclose. Reprint requests: Luca Carmignani, M.D., Department of Medicine and surgery, Urology Unit, University of Milan, IRCCS Policlinico San Donato Via Morandi 30, 20097, San Donato Milanese, Italy (TEL: þ39.02.527741; FAX: þ39.02.52774389; E-mail: luca.carmignani@ unimi.it).

0015-0282/10/$36.00 doi:10.1016/j.fertnstert.2008.12.038

The loss of renal units in patients with ureteral endometriosis causing hydronephrosis has been estimated to be 25%– 50% at the time of diagnosis, although these data derive from limited case series (2). The aim of this descriptive study is to assess whether routine renal ultrasonography may be recommended in all patients with pelvic endometriosis, in order to silence ureteral involvement of the disease.

MATERIALS AND METHODS Between January 2005 and July 2007, all new patients referred to the tertiary referral center for the treatment of endometriosis at the Department of Obstetrics and Gynecology of the State University of Milan, Italy, underwent additional urinary ultrasound during routine assessment, to exclude hydronephrosis. All patients were informed of the aims of the study and signed an informed consent before participating in the study. Renal ultrasound was repeated at the preoperative workup to confirm the diagnosis of ureterohydronephrosis. All patients who exhibited hydronephrosis at the preoperative

Fertility and Sterility Vol. 93, No. 6, April 2010 Copyright ª2010 American Society for Reproductive Medicine, Published by Elsevier Inc.

1741

routine urinary ultrasound were enrolled in this study and underwent further investigation by means of contrast-enhanced abdominal pelvic nuclear magnetic resonance imaging (MRI) or spiral multislice computerised tomography (spiral-CT) in patients with a metallic prosthesis and/or referred claustrophobia (3). In accordance with the literature, hydronephrosis was intended as renal pelvic dilation secondary to urinary obstruction, which will induce a progressive loss in renal function if left untreated (4). In cases of uncertain findings of ureteral endometriotic disease or if a double-J ureteral stent positioning was required for manifest hydronephrosis, the patients subsequently underwent cystoscopy and retrograde pyelography. Radionucleotide renal perfusion scans have not been routinely performed. Symptoms that we considered to be associated with hydronephrosis were lumbar pain (pain affecting the lumbar region) and renal colic (acute onset of intermittent intense flank pain irradiating downwards to the homolateral groin), which was then associated with nausea, fever, vomit, dysuria, pyelonephritis, and hematuria (4). The mean time from presentation of lumbar pain was determined at the assessment of the patients’ history. Suprapubic and pelvic pain, dyspareunia, and dysmenorrhea were not considered as specific symptoms of hydronephrosis. A routine vaginal examination was performed in all patients and data were collected on the presence of rectovaginal endometriotic nodules.

RESULTS Twenty-three patients (3%) of all 750 patients with endometriosis had associated ureterohydronephrosis diagnosed at renal ultrasound. The ureterohydronephrosis was indeed severe and did not show variability during the menstrual cycle. Eleven of these patients (47.8%) were taking hormonal treatment at the diagnosis of ureterohydronephrosis (10 with combined oral contraceptive and one with a progestin). Patient characteristics are shown in Table 1. Only 6 patients of 23 with hydronephrosis had normal results after a gynecological examination. At clinical examination, 12 patients exhibited posterior fornix nodularity, and two had palpable anterior isthmic nodule. Ovarian cysts were palpable in 17 patients. In 17 patients, transvaginal ultrasound identified ovarian cysts, of which in 11 cases were % 3 cm (Table 1). None of the patients presented evidence of altered renal function at blood (creatinine, blood urea nitrogen, Na, Ca, K, Cl) and urinary analysis, even in cases of an excluded kidney. In the 23 patients with ureterohydronephrosis, symptoms secondary to ureteral and renal involvement were present in 10 patients (43.5%). Of these, six reported lumbar pain (26.1%). In four patients (17.4%), the presenting symptom was a renal colic associated with fever, chills, and vomiting—three independently of the menstrual cycle and one during menstruation. In one case, hydronephrosis was diagnosed at ultrasound examination during assessments for an arterial 1742

Carmignani et al.

TABLE 1 Characteristics of 23 patients with hydronephrosis in pelvic endometriosis. Patient characteristic Mean age, y (range) Previous surgery for endometriosis Parity Ureteral localization Left Right Bilateral Other endometriotic localizations Adenomyosis Ovarian Cyst Left Right Bilateral Rectovaginal endometriosis staging (6) (N ¼ 16 and % of hydronephrosis) I II III IV Pelvic endometriosis staging (5) I II III IV Hydronephrosis Left Right Hormonal treatment Combined oral contraceptive Progestin

N (%) 35.6 (25–48) 6 (26.08) 5 (21.7) 16 (69.5) 6 (26.08) 1 (4.3) 23 (100)

3 (13.04) 6 (26.08)* 3 (13.04)* 2 (8.69)*

4 (17.39)y 2 (8.69)y 0y

4 (17.39) 2 (8.69) 7 (30.43) 3 (13.04)

0 (0) 4 (17.39) 5 (21.7) 14 (60.8) 17 (73.9) 6 (26.08) 11 (47.8) 10 (43.7) 1 (4.3)

* Diameter % 3 cm. y Diameter > 3 cm. Carmignani. Endometriosis and ureterohydronephrosis. Fertil Steril 2010.

hypertension. All patients presented symptoms caused by pelvic and/or bladder endometriosis (frequency, urgency, suprapubic pain, dysmenorrhea, deep dyspareunia) (Table 2).

Endometriosis and ureterohydronephrosis

Vol. 93, No. 6, April 2010

TABLE 2

FIGURE 1

Urologic symptoms at diagnosis. Asymptomatic patients Patients with urinary symptoms Lumbar pain Left Right Renal colic Left Right Hematuria Arterial hypertension

Right ureterohydronephrosis (urography I.V.). n (%) 10 (43.47) 6 (26.08) 4 (17.39) 2 (8.69) 4 (17.39) 3 (13.04) 1 (4.34) 0 (0) 1 (4.34)

Carmignani. Endometriosis and ureterohydronephrosis. Fertil Steril 2010.

In those patients with a renal colic, the diagnostic workup was more prompt and implied that second- and third-level examinations be undertaken immediately. When considering the American Fertility Society’s Endometriosis classification (AFS) stage (5), four (17.4%) of the 23 patients with ureterohydronephrosis had mild endometriosis. Sixteen (70%) patients with ureterohydronephrosis had deep endometriotic lesions. In these patients, the classification by Adamyan (6) was used and they were distributed as shown in Table 1. In 10 (43.5%) of the 23 patients, the uterosacral ligaments were involved or rectovaginal lesions were present. The mean time from presentation of lumbar pain to correct diagnosis of hydronephrosis was 11.3 months (2–24 months). Fifteen patients undergone ureteroneocystostomy with an associated psoas hitch in 15 patients, ureterolysis in six patients, and left ureteroneocystostomy and right ureterolysis in the patient with bilateral involvement. In one patient, a simple nephrectomy was performed as a nonfunctioning kidney was diagnosed. All 23 patients are currently healthy and have been free from relapse. DISCUSSION Our study highlights a relatively high incidence of hydronephrosis in regard to endometriosis of the urinary tract (estimated at 0.5%–1% of all endometriotic patients) (1) and the specific incidence of ureteral localization (a risk factor for the development of obstructive hydronephrosis) (7–9), as well as a high number of silent asymptomatic presentations of hydronephrosis (2). The high incidence of hydronephrosis in our series may be partially due to the patient population presenting at a tertiary referral centre for endometriosis, and as such the incidence rate here reported is most probably overestimated. Furthermore, it is also possible that a considerably large number of patients will not have hydronephrosis diagnosed at their first gynecological visit. Although considered a pathognomonic sign of urinary endometriosis, and as such described in 13%–18% of ureteral Fertility and Sterility

Carmignani. Endometriosis and ureterohydronephrosis. Fertil Steril 2010.

endometriosis by Comiter et al. (2), hematuria was never observed in our series. Undoubtedly, patients with pelvic endometriosis presenting nonspecific symptoms pose a diagnostic challenge, and perhaps a portion of the pelvic pain described by these patients is actually of lumbar or renal origin despite attempts to accurately localize the referred pain. The percentage of endometriotic-induced hydronephrosis in women with irritative urinary symptoms is in accordance with the most recently published data confirming the absence of specific urinary symptoms in nearly half of all patients (1, 3). Furthermore, if the symptoms are present in the majority of patients, they will not be recognized; we believe it is necessary that more attention be given in order to avoid irreversible renal damage (9, 10). Three-hundred-twenty of the 750 patients (42.6%) had deep infiltrating endometriosis (DIE), whereas the incidence of hydronephrosis in patients with DIE is 5%; DIE seems to represent a risk factor for development of hydronephrosis. When considering the AFS stage, 17.4% of the patients with ureterohydronephrosis had only a mild endometriosis; when considering Adamyan grading, 25% had grade 1 rectovaginal lesions. These descriptive data seem to show that 1743

there are a relatively limited number of patients who are currently not considered to be at true risk of urinary involvement and possible late diagnosis of renal failure. This study does not allow the definition of the functional renal loss by means of renal perfusion radionucleotide examinations. In fact, radionucleotide renal perfusion scans were not routinely performed because of medical costs and time issues. Renal functional and morphologic data were obtained by means of contrast perfusion phases during MRI or CT scans despite these are admittedly less precise. Renal function parameters available from the data collected are based on abdominal–pelvic contrast enhanced CT or MRI scans with three-dimensional reconstruction of the urinary excretory system performed for diagnostic or preoperative anatomical studies (Fig. 1). The majority of previous studies have principally focused their attention on the differences between intrinsic and extrinsic endometriosis (2, 7, 11)—thus the therapy to be proposed in patients with associated hydronephrosis. In regard to surgical therapy, the options are endoscopic or open procedures and the necessity of performing a neo-ureterocystostomy or ureterolysis, the latter of which is warranted by recent studies (11, 12). We decided to perform a neo-ureterocystostomy, as we believe that the endometrial ureteral tract is fibrotic and thus atonic, as is demonstrated by the pathologic examination of the resected ureteral segments. The present study seems to confirm the presence of a left lateral predisposition of ureteral endometriosis in 16 of the 23 patients (69.5%) (13). Early diagnosis of hydronephrosis is critical to safeguarding renal function in these young patients, rather than considering eventual therapeutic strategies. As such we believe that patients with suspected or proven pelvic endometriosis should systematically undergo a sonographic kidney evaluation during routine gynecological sonographic examination, because of the high frequency of silent presentations. Once diagnosed, periodic evaluation of hydronephrosis is mandatory to prevent the high risk of secondary loss of renal function. In our study, in which renal ultrasound was performed in all patients with pelvic endometriosis, the exclusion from this evaluation of those without deep endometriosis would have missed 30% cases of ureterohydronephrosis possibly followed by a subsequent silent renal loss. Of the 23 patients with ureterohydronephrosis, specific symptoms secondary to ureteral and renal involvement were present in 10 patients (43.5%), representing the only justification for requesting a nephrologic workup. Those patients with a primary urinary symptom at onset were first evaluated by the urologist to allow early recognition

1744

Carmignani et al.

of associated hydronephrosis independently of the causative etiology, but guarantee a better long-term outcome. Hydronephrosis secondary to endometriosis is a misrecognized disease that can lead to serious complications, such as silent loss of renal function. In only a small number of these patients, the presence of a renal colic will lead to prompt diagnosis and thus an earlier treatment. In our study, the high rate (56.5%) of asymptomatic ureteral involvement in patients with known pelvic endometriosis seems to warrant the need for further investigations regarding the possibility of avoiding the high percentage of silent renal losses. Unfortunately there appears to be no specific risk factor to allow for an early suspect or a validated preventive diagnostic and therapeutic program. Such difficult diagnostic conditions, along with the high risk of silent renal loss in about 25%–50% cases at the time of diagnosis, warrants perhaps an earlier and periodic diagnostic assessment in all patients with pelvic endometriosis. It remains to be evaluated whether urinary ultrasound ensures a beneficial cost/benefit ratio if used on a routine basis, especially concerning the relatively large patient number.

REFERENCES 1. Frego E, Antonelli A, Tralce L, Micheli E, Cunico SC. Ureteral endometriosis: urologic features. Arch Ital Urol Androl 2002;74:3–5. 2. Comitern C. Endometriosis of the urinary tract. Urol Clin N Am 2002;29: 625–35. 3. Kinkel K, Chapron C, Balleyguier C, Fritel X, Dubuisson JB, Moreau JF. Magnetic resonance imaging characteristics of deep endometriosis. Hum Reprod 1999;14–4:1080–6. 4. Wein A, Kavoussi L, Novick A, Partin A, Peters C. Campbell-Walsh Urology Review Manual, 3rd ed. 5. Revised American Fertility Society classification of endometriosis: 1996. Fertil Steril 1997;67:817–21. 6. Adamyan L. Additional international perspectives. In: Nichols DH, ed. Gynecologic and obstetric surgery. St. Louis: Mosby Year Book, 1993: 1167–82. 7. Yohannes P. Ureteral endometriosis. J Urol 2003;170:20–5. 8. Vercellini P, Pisacreta A, Pesole A, Vicentini S, Stellato G, Crosignani PG. Is ureteral endometriosis an asymmetric disease? BJOG 2000;107: 559–61. 9. Ghezzi F, Cromi A, Bergamini V, Serati M, Sacco A, Mueller M. Outcome of laparoscopic ureterolysis for ureteral endometriosis. Fertil Steril 2006;86:418–22. 10. Antonelli A, Simeone C, Zani D, Sacconi T, Minini G, Canossi E, et al. Clinical aspects and surgical treatment of urinary tract endometriosis: our experience with 31 cases. Eur Urol 2006;49:1093–8. 11. Donnez J, Nisolle M, Squifflet J. Ureteral endometriosis: a complication of rectovaginal endometriotic (adenomyotic) nodules. Fertil Steril 2002;77:32–7. 12. Generao S, Keene K, Das K. Endoscopic diagnosis and management of ureteral endometriosis. J Endourol 2005;19:1177–9. 13. Vercellini P, Pisacreta A, Pesole A, Vicentini S, Stellato G, Crosignani PG. Is ureteral endometriosis an asymmetric disease? BJOG 2000;107: 559–61.

Endometriosis and ureterohydronephrosis

Vol. 93, No. 6, April 2010