Most iatrogenic ureteral injuries are undetected during gynecologic surgery.1–5 Ureteral injury rates during gynecologic surgery range from 0.2% to 2.5% and higher rates (7.3%–11%) have been reported during procedures for incontinence and pelvic organ prolapse.1,4,6–8 Although universal cystoscopy has increased the detection of lower urinary tract injuries, it has not affected the overall injury rate.6 Failure to identify ureteral injury intraoperatively can lead to significant morbidity.4,7,9 Therefore, suspected ureteral injury warrants prompt intraoperative evaluation.
Knowledge of anatomy and careful intraoperative ureteral identification are hallmarks in injury prevention.3,5 The pelvic ureter is easily identified from the pelvic brim to uterine artery. After crossing under the uterine artery, it enters the parametrium and courses within a “ureteral tunnel,” where it roughly separates the anterior fibers of the cardinal ligament from the posterior fibers of the uterosacral ligament.10 In this region, the ureter lies along the anterolateral cervix and upper vagina before coursing inferomedially over the anterior vaginal wall to enter the bladder.3 Lastly, it courses within the bladder wall to terminate at the ureteric orifice.10
Data suggest that ureteral injury most commonly occurs while securing the uterine artery and with division of the cardinal ligament during hysterectomy.9,11,12 Understanding the ureter's anatomy within the parametrium and its relationship to other pelvic structures is critical to safely complete gynecologic surgery.3,5 The aim of this study is to evaluate the relationship of the pelvic ureter to clinically relevant structures and to characterize the histologic composition and nerve density of the parametrial tissue surrounding the distal ureter.
In this observational cadaveric study, detailed dissections of the pelvic ureter and microscopic examination with nerve density assessment of the distal ureter and its surrounding parametrial tissue were performed in adult female cadavers obtained from the Willed Body Program at the University of Texas Southwestern Medical Center in Dallas. This study was exempt from review by the University of Texas Southwestern Medical Center's institutional review board in accordance with the Code of Federal Regulations, Title 45. Age, race, height, weight and cause of death were obtained for all cadavers. The time from death to tissue harvest and immersion in fixative was documented for specimens used for histologic analysis.
In all specimens, a wide low transverse incision was used to enter the abdominal cavity and the incision was extended laterally and superiorly to a level just below the lower ribs. The flap of anterior abdominal wall was displaced superiorly to expose the abdominopelvic contents, including the kidneys. The bowel was packed into the upper abdomen to expose the entire pelvic cavity. Figure 1 illustrates ureteral lengths and distances measured relative to specific points along the ureter. For all measurements, care was taken to avoid excessive traction and manipulation, which could change or distort distances.
First, the ureter and ovarian vessels, within the proximal part of the infundibulopelvic ligament, were identified transperitoneally in the region of the pelvic brim. A window was made in the peritoneum over the pelvic brim with care not to disrupt the position of underlying structures. Distances between the most medial aspect of the ovarian vessels and the lateral border of the ureter were recorded. A peritoneal incision was then made just cephalad to the pelvic brim and lateral to the ureter to expose the common iliac and its bifurcation into external and internal iliac arteries. The point at which the ureter first passed over the bifurcation of the common iliac or external iliac arteries was noted and a metal pin was placed at the lateral aspect of this point. Position, medial or lateral, and distance from the bifurcation was recorded when the ureter did not pass exactly over the bifurcation of the common iliac artery. The peritoneal incision was then extended distally to the point at which the ureter passed under the uterine artery. In all specimens, the uterine artery was dissected to its origin from the internal iliac artery. Care was taken to preserve the ureter's attachments to the medial leaf of the broad ligament. A metal pin was then placed adjacent to the ureter at the superomedial point of the ureter-uterine artery crossover. This specific point was chosen as it represents the initial contact between the uterine artery and ureter and is closest to the cervix. Distances from the point at which the ureter entered the pelvic brim to the crossover with the uterine artery were measured. Given the indirect path of this segment of ureter along the pelvic sidewall, both direct and indirect measurements were obtained. Direct measurements were taken using a caliper positioned at the two previously marked points. Indirect measurements were obtained using an 0-vicryl suture that followed the exact path of the ureter between the aforementioned points. Both direct and indirect measurements were taken to determine whether significant differences were present between these values, because this would be important when selecting ureteral stent length and while performing ureterolysis. If the cadaver had a uterus, the closest distance between the ureter and uterine isthmus was recorded with the uterus in midline position. In addition, distances between the medial border of ureter and the most lateral portions of the anterior vaginal fornix were recorded in cadavers with uteri; in those without uteri, distances were measured to the most lateral portions of the vaginal apex. Dissection was then continued along the path of the anterior wall of the ureter and the cardinal ligament (anterior portion of parametrium) was sharply transected to expose this section of ureter and clearly identify its entrance into the bladder wall (Fig. 2). The distance between the ureter-uterine artery crossover point and the point at which the ureter entered the bladder wall was then recorded. Again, direct and indirect measurements were taken for this distal segment of the ureter in a similar fashion as described earlier. The point at which the ureter first passed over the anterolateral vaginal wall was then marked with a metal pin and the distance between this point and the most lateral border of the anterior vaginal fornix or the vaginal apex was measured. A large cystotomy was made at the bladder dome and the ureteric orifices were identified. Total ureter length was then measured by placing a centimeter graded 4 French Whistle Tip ureteral catheter into the ureters in a retrograde fashion, from ureteric orifices to the level of renal pelvis. Positioning of the catheter within the renal pelvis was confirmed by palpation. The catheter was then clamped with a hemostat at the ureteric orifice, removed and distances measured. The intramural ureteral length was recorded by placing a curved hemostat at the point where the ureter entered the bladder wall. The 4 French Whistle Tip ureteral catheter was then passed through the ureteric orifice to the level of the clamp. Another curved hemostat was used to clamp the catheter at the ureteric orifice interface; the clamped catheter was then removed and the length distal to the clamp was measured. Total vaginal length was recorded from the hymeneal remnants to the mid posterior fornix or vaginal apex.
Indirect ureter length measurements were used for data analyses for the two segments of pelvic ureter where both direct and indirect measurements were obtained. All measurements were taken twice by the same examiner (LAJ) using the same steel electrocardiogram caliper and plastic ruler. The mean of the two measurements was used for analyses. Measurements were tabulated and descriptive statistics were used for data analysis and reporting using Microsoft Excel 2011.
In five additional adult cadaver specimens, tissue was harvested within 12 hours from time of death for histologic analysis. The uterus, cervix, upper vagina and surrounding parametrial tissue were resected en-bloc to the level of the pelvic sidewall. Harvested tissue was immersed in 10% formalin solution for a minimum of 48 hours. Fixed tissue was transected first in the midsagittal plane into a right and left side. Each side was then cut into four tissue blocks in the axial plane perpendicular to the long axis of the ureter, from the level of the ureteric orifice to a level just above the crossover point between the ureter and uterine artery with the lateral border of tissue analyzed being the pelvic side wall and the medial border the uterus or cervix. This was done to analyze sections that spanned the entire length of the ureteral tunnel. Specimens were submitted in Super Mega cassettes (3″×2″×0.75″) for tissue processing.
Fixed tissues were processed, embedded and sectioned using standard methods. Adjacent series were stained with hematoxylin-eosin and Gomori trichrome for structural identification, and with antibodies against βIII tubulin (a specific marker of neuronal axons), and detected using standard colorimetric immunostaining protocols (Fig. 3). Hematoxylin-eosin and Gomori trichrome staining were performed with standard technique by the University of Texas Southwestern Medical Center Histo Pathology Core. For βIII tubulin immunohistochemistry, standard colorimetric methods were used. A Vectastain Elite mouse ABC kit (Vector Labs, #PK-6102) was used for staining according to the manufacturer's instructions. Incubation with primary antibody (mouse monoclonal anti-βIII tubulin [1:200; Abcam #ab78078]) was performed at 4°C overnight. A DAB (3,3′-diaminobenzidine) substrate kit (Vector Labs, #SK-4100) was used for colorimetric detection following the manufacturer's instructions. The sections were counterstained with hematoxylin (Leica SelecTech Hematoxylin 560, #3801570) and dehydrated according to standard procedures.
Microscopic slide analysis of all three specimens, using a Nikon Eclipse 80i microscope, was performed by a single gynecologic pathologist (K.S.C.), with all authors present for review and collaboration. Relevant microscopic features were noted, including the composition of the tissue surrounding the ureter in the area of interest. Brightfield images of entire tissue sections were collected for histologic examination using a Zeiss Axioscan.Z1 slide scanning microscope equipped with a 40X, 0.95NA objective. Nerve density was analyzed automatically in four quadrants relative to ureter for each whole slide image using the “Positive Pixel Count 2004-08-11” algorithm in Aperio ImageScope (126.96.36.19913) using appropriate settings to detect DAB-positive pixel areas. Total nerve area was quantified in individual anatomical regions of interest by summing the numbers of weakly positive, positive and strongly positive pixels to get a total number of DAB-positive pixels, converting this number to area in micrometers2 using the appropriate scaling factor and dividing by the total area covered by the region of interest. Relative nerve areas from anterolateral, anteromedial, posterolateral and posteromedial quadrants from a total of 20 tissue sections from five cadavers were measured. Average measurements of nerve density in the anterior and posterior quadrants were statistically compared using a two-tailed t test. A computer workstation (Lenovo ThinkStation D30) running Windows 7 64-bit equipped with a Xeon E5-1660 v2 @3.7 GHz processor, Nvidia Quadro K2000 (2-GB) graphics card, 96-GB RAM and 8-TB hard drive was used for analysis of whole slide images.
Thirty adult female cadavers were examined grossly, 28 were Caucasian, one was African American, and one was Native American. The age at time of death was 71±15 years (mean±SD). The body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) was 24±6. The most common cause of death was respiratory complications. Twenty specimens had uteri in situ, and 10 specimens had evidence of prior hysterectomy. Tissue from five separate adult female cadavers was examined microscopically, one nulliparous and four parous. Causes of death for these specimens included respiratory failure in three specimens, lymphoma and Alzheimer's disease. No specimen used for histologic analysis had undergone hysterectomy. Time from death to tissue harvest was less than 12 hours. Available medical histories and dissections of all specimens revealed no evidence of pelvic disease, pelvic cancers or prolapse beyond the hymen.
A single ureter on each side was identified in all specimens examined. At the pelvic brim, the ureter passed over the bifurcation of the common iliac artery in 55% of specimens (Fig. 4). In the remaining 45% of specimens, the ureter entered the pelvis by passing over the proximal part of external iliac artery; the distance from the bifurcation was 7.8±4.1 (range 2–19) mm.
In all specimens, the ureter was found attached to the posterior abdominal wall peritoneum and medial leaf of broad ligament until it passed under the uterine artery (Fig. 5). A single uterine artery anterior to the ureter was identified in all specimens, including those six specimens with evidence of hysterectomy. The ureter coursed under the uterine artery in all specimens and continued through the “ureteral tunnel” within the parametrium before entering the bladder wall. Transection of the parametrium anterior to the ureter (cardinal ligament) revealed the ureter positioned atop the posterior portion of the parametrium, which was continuous with the lateral and inferior extensions of the uterosacral ligament tissue (Figs. 2 and 4). Both the anterior and posterior portions of parametrium were continuous with the upper part of paravaginal tissue that attached to the pelvic sidewalls.
Pelvic ureter lengths between clinical points of interest are presented in Table 1. Table 2 summarizes the closest distances from the ureter to: ovarian vessels at the pelvic brim, uterine isthmus and lateral border of the anterior vaginal fornix in those specimens with uteri; and the lateral border of the vaginal apex in specimens without uteri. Distances from the point at which the ureter first passed over the anterolateral vaginal wall and the most lateral border of the anterior vaginal fornix or the vaginal apex are also presented in Table 2. Measurements in specimens with uteri and those without uteri were not statistically significant (P>.05), data not shown.
The total ureter length was 26.3±1.4 (range 24–29) cm on the right and 27.6±1.6 (25–30.5) cm on the left. The combined measurements of indirect pelvic ureter lengths plus the intramural ureter length was 13.0±2.0 (8.9–16.4) cm on the right and 13.2±1.5 (9.1–16.4) cm on the left. Based on these combined pelvic and intramural ureteral lengths, approximately 48% of the ureter lies within the pelvis. Average total vaginal length was 9.3±1.1 (7.3–11.5) cm in cadavers with uterus and 9.0±2.0 (6.7–11.3) cm in those without.
Microscopic examination of the distal pelvic ureter and surrounding connective tissue demonstrated the greatest density of fibrovascular tissue posteromedial to the ureter in all specimens examined (Fig. 6). The tissue of greatest density corresponded to the paracervical and parametrial tissues adjacent to the distal aspect of the lower uterine segment. This tissue, which gradually decreased in density from medial to lateral, was composed of prominent vessels with less conspicuous nerves and ganglia, disposed within a fibromuscular stroma. The connective tissue immediately surrounding the ureter, which corresponds to the clinical “ureteral tunnel,” was of approximately equal density in all quadrants and consisted predominantly of adipose tissue with inconspicuous delicate vasculature. Automated quantification of positive immunostaining using nerve-specific βIII tubulin antibodies revealed that the posterior quadrants surrounding the distal third of the pelvic ureter contained the greatest nerve density. The average nerve density measured as total tissue area covered by βIII tubulin-positive staining in posterolateral and posteromedial quadrants was 44% greater than in anterolateral and anteromedial quadrants (0.76% vs 0.43%, respectively), and this difference was statistically significant (P=.03).
Our main anatomic findings confirm the close proximity of the ureter to critical points in the pelvis where clamps are placed and tissue ligated during hysterectomy. We found the ureter to be an average of 0.8 cm from the ovarian vessels at the pelvic brim, 1.7 cm from the uterine isthmus, and 1.3 cm from the lateral vaginal apex; these findings support previous observations that the ureter is at greater risk of injury during the following steps of a hysterectomy: ligation of the ovarian vessels, transection of the uterine artery and cardinal ligament, and suturing the vaginal cuff.9,11,12 Our findings suggest a substantial injury risk is added when clamps are placed beyond 1.5 cm lateral to the uterine isthmus or 1 cm lateral to the vaginal cuff, which corroborates previous observations made by Hurd et al.13 Furthermore, we found that in approximately 50% of specimens the ureter entered the pelvis by passing directly over the bifurcation of the iliac vessels and in the remaining 50% it passed over the proximal part of the external iliac vessels; this finding may be helpful in intraoperative identification of the ureter in settings of bleeding or altered anatomy.
Another important observation was the short distance (approximately 3 cm) between the lateral border of the vaginal apex or anterior fornix to the point at which the ureter passes over the anterior vaginal wall. Dissection of the bladder off the anterior vagina beyond 3 cm might increase injury risk, especially when extended laterally, as done during radical hysterectomy or anterior mesh placement during sacrocolpopexy. This underlines the importance of maintaining a meticulous midline dissection plane while first developing the vesicovaginal space and the use of retractors to displace the bladder before operating on the anterolateral vaginal wall.
Intraoperative identification and management of ureteral injury can prevent long-term complications reported with delayed diagnosis.9 Retrograde passage of a ureteral catheter is often an initial step of injury evaluation. Knowledge of specific ureteral lengths at various pelvic regions susceptible to injury is important in diagnosis and management or ureteral injury. Our combined total pelvic and intramural ureter length of 13 cm suggests that, if catheter resistance is encountered approximately 13 cm from the ureteric orifice, the injury most likely occurred at the pelvic brim while securing the ovarian vessels. If catheter resistance is met 4–5 cm from the ureteric orifice, injury most likely occurred while securing the uterine vessels. Lastly, if resistance is met at 1–2 cm, injury most likely occurred during vaginal cuff closure.14
In this study, we provide a comprehensive and quantitative analysis of nerve density across the entire length of the ureteral tunnel. Our histologic analysis of the distal ureter and surrounding parametrial tissue revealed an increased density of nerve fibers in the posterior quadrants surrounding the ureter. This is similar to the findings of greater nerve content in the deep uterosacral ligament when compared with the superficial layers in a study evaluating tissue samples from women undergoing radical hysterectomy.15 Our findings of greater nerve density posterior to the ureter are supported by gross dissections, which consistently identified a vesical branch from the inferior hypogastric plexus posterior to the ureter.16,17 This suggests that dissection or coagulation posterior to the ureter, as compared with dissection of the cardinal ligament anterior to ureter, may pose a higher risk of injury to parasympathetic and sympathetic fibers contained within the inferior hypogastric plexus.17 Consequently, voiding dysfunction may result, as reported during radical hysterectomies.15,18,19
Limitations of this study are those inherent to cadaver studies. The cadavers in this study had small uteri without evidence of fibroids or pelvic pathology, which limits generalizability. Strengths of this study include the large sample size for gross examination with a diverse age group and BMI and the inclusion of specimens without uteri. Limitations specific to histologic analyses include a small sample size. Tissue quality may also have been compromised during tissue fixation. However, tissue harvesting within 12 hours of death should prevent tissue changes that can be seen with repetitive tissue manipulation, freezing and cellular lysing, thus leading to improved histologic evaluation.
In summary, the proximity of the ureter to the uterine isthmus (1.7 cm) and the lateral anterior vaginal fornix (1.8 cm) or vaginal cuff (1.3 cm) mandates careful surgical technique and adequate identification, especially in the setting of bleeding or altered anatomy. Increased nerve density and vascularity along the posterior aspect of the distal ureter emphasizes the importance of avoiding extensive ureterolysis in this region. If ureterolysis is necessary, avoiding the tissue posterior to the ureter might decrease the risk of nerve injury and therefore postoperative voiding dysfunction. Future research of the ureter as it courses through the ureteral tunnel should include a larger sample size for histologic evaluation to allow for a more detailed comparison of nerve density along the antero-posterior axis of the ureter proper. Furthermore, continuing the comprehensive analysis of nerve density in other regions of the female genitourinary tract may continue to aid in refining surgical technique.
1. Chan JK, Morrow J, Manetta A. Prevention of ureteral injuries in gynecologic surgery. Am J Obstet Gynecol 2003;188:1273–7.
2. Findley AD, Solnik MJ. Prevention and management of urologic injury during gynecologic laparoscopy. Curr Opin Obstet Gynecol 2016;28:323–8.
3. Clarke-Pearson DL, Geller EJ. Complications of hysterectomy. Obstet Gynecol 2013;121:654–73.
4. Vakili B, Chesson RR, Kyle BL, Shobeiri SA, Echols KT, Gist R, et al. The incidence of urinary tract injury during hysterectomy: a prospective analysis based on universal cystoscopy. Am J Obstet Gynecol 2005;192:1599–604.
5. Gilmour DT, Dwyer PL, Carey MP. Lower urinary tract injury during gynecologic surgery and its detection by intraoperative cystoscopy. Obstet Gynecol 1999;94:883–9.
6. Teeluckdharry B, Gilmour D, Flowerdew G. Urinary tract injury at benign gynecologic surgery and the role of cystoscopy: a systematic review and meta-analysis. Obstet Gynecol 2015;126:1161–9.
7. Ibeanu OA, Chesson RR, Echols KT, Nieves M, Busangu F, Nolan TE. Urinary tract injury during hysterectomy based on universal cystoscopy. Obstet Gynecol 2009;113:6–10.
8. Barber MD, Visco AG, Weidner AC, Amundsen CL, Bump RC. Bilateral uterosacral ligament vaginal vault suspension with site-specific endopelvic fascia defect repair for treatment of pelvic organ prolapse. Am J Obstet Gynecol 2000;183:1402–10.
9. Sharp HT, Adelman MR. Prevention, recognition, and management of urologic injuries during gynecologic surgery. Obstet Gynecol 2016;127:1085–96.
10. Jones H. Te Linde's operative gynecology. Vol. 11. Philadelphia (PA): Lippincott Williams & Wilkins; 2015.
11. Tamussino KF, Lang PF, Breinl E. Ureteral complications with operative gynecologic laparoscopy. Am J Obstet Gynecol 1998;178:967–70.
12. Utrie JW Jr. Bladder and ureteral injury: prevention and management. Clin Obstet Gynecol 1998;41:755–63.
13. Hurd WW, Chee SS, Gallagher KL, Ohl DA, Hurteau JA. Location of the ureters in relation to the uterine cervix by computed tomography. Am J Obstet Gynecol 2001;184:336–9.
14. Hutch JA. Theory of maturation of the intravesical ureter. J Urol 1961;86:534–8.
15. Butler-Manuel SA, Buttery LD, Polak JM, A'Hern R, Barton DP. Autonomic nerve trauma at radical hysterectomy: the nerve content and subtypes within the superficial and deep uterosacral ligaments. Reprod Sci 2008;15:91–6.
16. Mauroy B, Bizet B, Bonnal JL, Crombet T, Duburcq T, Hurt C. Systematization of the vesical and uterovaginal efferences of the female inferior hypogastric plexus (pelvic): applications to pelvic surgery on women patients. Surg Radiol Anat 2007;29:209–17.
17. Ripperda CM, Jackson LA, Phelan JN, Carrick KS, Corton MM. Anatomic relationships of the pelvic autonomic nervous system in female cadavers: clinical applications to pelvic surgery. Am J Obstet Gynecol 2017;216:388.e381–7.
18. Kraima AC, Derks M, Smit NN, van de Velde CJ, Kenter GG, DeRuiter MC. Careful dissection of the distal ureter is highly important in nerve-sparing radical pelvic surgery: a 3D reconstruction and immunohistochemical characterization of the vesical plexus. Int J Gynecol Cancer 2016;26:959–66.
19. Plotti F, Angioli R, Zullo MA, Sansone M, Altavilla T, Antonelli E, et al. Update on urodynamic bladder dysfunctions after radical hysterectomy for cervical cancer. Crit Rev Oncol Hematol 2011;80:323–9.