Can transvaginal ultrasound be used to predict the need for ureterolysis in women undergoing laparoscopy for suspected endometriosis?

Over recent years, there has been major advancement in the use of transvaginal ultrasound (TVU) for the pre‐operative diagnosis and staging of deep endometriosis. The International Deep Endometriosis Analysis (IDEA) Consensus paper established guidelines for identifying, classifying and measuring endometriotic lesions and pelvic adhesions.1 In addition, the development of an ‘Ultrasound‐Based Endometriosis Staging System’ (UBESS) has provided us with a framework to predict the laparoscopic surgical skill level required to perform surgery for women with suspected endometriosis.2 Although the UBESS validation study performed by our centre demonstrated a high accuracy (98%) for the ultrasound prediction of corresponding RANZCOG/AGES surgical skill level required,3 there are some limitations with the UBESS classification. In particular, TVU is not able to predict which women will require ureterolysis (i.e. dissection of the ureter) at the time of laparoscopy, especially in the absence of any obvious ureteric disease at ultrasound.

Ureterolysis is considered an advanced laparoscopic skill and is not typically performed by general gynaecologists. This surgical procedure is performed to isolate and expose the ureter, in order to free it from external pressure or adhesions or to avoid injury to it during pelvic surgery. The anatomical course of the ureters is of surgical importance, as ureters travel close to other structures (i.e. ovaries, cervix and uterine artery) in the pelvis. Ureters must be identified during pelvic surgery to ensure that they are not accidentally damaged.

The course of the pelvic ureter begins with its descent into the posterior lateral pelvis, lateral to the sacrum and immediately ventral to internal iliac artery. The ureter then travels medially to the internal iliac artery and then passes underneath the uterine artery (i.e. referred to as ‘water under the bridge’) and into the paracervical tissue. This paracervical tissue is often referred to as ‘the tunnel’ of the cardinal ligament/anterior bladder pillar (or ‘the tunnel of Wertheim’). The ureter is located only 1–2 cm lateral to the cervix and is immediately lateral to the uterosacral ligament (USL). The risk of ureteric injury is particularly increased if endometriosis or fibrosis is present in this area, as the ureter can be drawn close to the adjacent uterine isthmus and USL by the disease. Once the ureter passes through the tunnel of Wertheim, it courses medially over the anterior vaginal fornix to enter the trigone of the bladder.

Whether a full or partial ureterolysis is performed is dependent upon the extent and location of disease, as well as and the skill of the surgeon. Full ureterolysis is described when dissection of the ureter is performed from the level of the pelvic brim to the level of the cardinal ligament/tunnel of Wertheim. This procedure can be performed by either the medial or lateral approach. During the medial approach, the ureter is first identified by visualisation through the peritoneum at the pelvic brim. The pelvic sidewall peritoneum is then grasped just above the ureter, tented and incised, allowing for entry into the pelvic sidewall. The ureter is then dissected off the peritoneum using blunt and/or sharp dissection along its entire course, to the level of the paracervical tissue. Care is taken during the ureterolysis procedure to ensure the ureter is not de‐vascularised. For the lateral approach, the pelvic sidewall triangle is entered and the ureter is dissected within the retroperitoneal space. The pelvic triangle peritoneum is bordered anteriorly by the round ligament, medially by the infundibulopelvic ligament and laterally by the external iliac artery. The peritoneum of the pelvic triangle is incised and extended parallel to the infundibulopelvic ligament. The infundibulopelvic ligament is then grasped and pulled medially, allowing for exposure of the retroperitoneal space and pelvic sidewall structures. The pelvic sidewall areolar tissue is dissected with sharp and blunt dissection until the ureter is visualised at the level of the pelvic brim. The ureter can then be dissected within the retroperitoneal space to the tunnel of Wertheim. Once the ureter is isolated and freed from the diseased tissue, endometriotic implants and fibrotic areas can be excised.

Full ureterolysis is generally required for the surgical excision of ureteric DE (intrinsic and extrinsic disease), as well as for the removal of endometriosis/fibrotic tissue involving structures in close proximity to the ureter (i.e. overlying pelvic sidewall/USL). When ovarian endometriomas are fixed to the underlying pelvic sidewall peritoneum, fibrosis of the sidewall can be extensive. As a result of the inflammatory reaction, the ureter can become densely adherent to the fibrotic sidewall peritoneum. Full ureterolysis is essential in these cases, in order to safely excise the diseased area. The same is true for USL endometriosis, as the ureter can be tethered towards the USL disease and is at risk for injury unless adequate ureterolysis is performed.

Partial ureterolysis (i.e. when only a portion of the ureter is dissected, rather than the entire course of the ureter) is usually adequate when the endometriosis is superficial and localised to a small area on the pelvic sidewall Figure 1a. is an example of superficial endometriosis located on the pelvic sidewall, and Figure 1b. demonstrates the close relationship of the superficial disease to the course of the underlying ureter. Partial ureterolysis (from the medial approach) was performed in this case, in order to excise the pelvic sidewall peritoneal disease safely (Figure 2).

Figure 1.

(a) A picture of superficial endometriosis lesions (indicated by white arrows) along the left pelvic sidewall peritoneum, overlying the left ureter, seen at laparoscopy. (LO = left ovary). (b) The same picture as displayed in (a), with the course of the ureter outlined in yellow. Note the close proximity of the superficial endometriosis lesions to the underlying left ureter.

Figure 2.

A picture following laparoscopic left ureterolysis, performed in order to safely excise the superficial left pelvic sidewall endometriosis lesions overlying the left ureter in Figure 1a. The white arrows point to the isolated left ureter. (white star = internal iliac artery).

Experienced TVU operators are able to visualise the course of the pelvic ureter from the iliac vessels to the vesicoureteric junction. The presence of hydroureter and/or ureteric DE at TVU are usually indicators for ureterolysis at surgery; however, there is no current method to predict when ureterolysis is required in the absence of these findings. In the case of severe pelvic sidewall fibrosis overlying the ureter and/or USL endometriosis immediately adjacent to the ureter, an advanced laparoscopic surgeon is required to excise the disease safely. Ureteric catheterisation may also be prudent for these women, as they are at increased risk of ureteric injury during a difficult ureteric dissection.

In a study by Lima et al, the researchers found that USL nodule size, left‐sided endometrioma, reduced ovarian mobility and ureteral changes on the ipsilateral side at pre‐operative TVU were all significantly associated with ureteral DE at laparoscopy.4 The best ultrasound predictor of ureteric involvement for this study was USL size; USL nodules measuring 1.75 cm and 1.95 cm on the right and left sides, respectively, significantly increased the risk of ureteral involvement. In another study by Ghezzi et al.,5 33 women underwent surgical excision of ureteric DE and the following associated intraoperative findings were noted: ovarian endometriomas in 61%, complete pouch of Douglas obliteration in 42%, and USL DE in 100%.

In cases of isolated superficial endometriosis (i.e. no endometrioma or pelvic DE) overlying the pelvic sidewall and underlying ureter, at least partial ureterolysis is required in order to safely excise the disease. This group of women with isolated superficial pelvic sidewall disease is essentially impossible to predict pre‐operatively; however, there may be some merit in the use of TVU soft markers such as site‐specific tenderness and ovarian immobility for the prediction of sidewall disease in these women. A recent study by our group found that left ovarian immobility was significantly associated with isolated superficial left USL endometriosis and left adnexal site‐specific tenderness corresponded to left pelvic sidewall superficial endometriosis. The accuracy, sensitivity, specificity, PPV and NPV for ovarian immobility at TVU and the presence of ipsilateral pelvic sidewall superficial endometriosis for the left ovary were as follows: 71%, 16%, 87%, 27% and 78%, respectively; and for the right ovary was as follows: 82%, 7.0%, 94%, 14% and 87%, respectively.6

In order to allow for the appropriate pre‐operative referral and surgical planning for women with pelvic sidewall disease (in the absence of obvious ureteric disease at TVU), future research should focus on identifying TVU hard markers to predict the need for ureterolysis at laparoscopy. The most promising ultrasound features at this point appear to be ipsilateral ovarian fixation, USL DE (especially the size of the nodule) and fixed endometrioma; these TVU markers need to be externally validated for the prediction of ureterolysis. The use of a more standardised TVU technique for the assessment of USL disease location may also improve our ability to predict the need for ureterolysis.7

For women with isolated superficial endometriosis and an essentially normal TVU scan, predicting the need for ureterolysis will likely remain a challenge. TVU soft markers for superficial disease (i.e. site‐specific tenderness and ovarian immobility) require further investigation to determine whether they are associated with the need for ureterolysis. Once the TVU markers associated with the need for ureterolysis are identified, the UBESS can then be modified to more accurately reflect the surgical skill level required, and in turn, improve the pre‐operative planning and surgical outcomes for these women.

Authorship statement

Authorship listing conforms with the journal's authorship policy, and all authors are in agreement with the content of the submitted manuscript.

Funding

No funding information is provided.

Conflict of Interest

No disclosures or conflicts of interest to declare.