• imaging;
  • pelvis;
  • transvaginal;
  • ultrasound;
  • ureters


  1. Top of page


To investigate the feasibility of identifying pelvic segments of normal ureters and measuring their size on standard transvaginal ultrasound examination.


This was a prospective observational study from June to July 2012. All women in the study underwent a transvaginal ultrasound examination performed for various indications either by an expert or by an intermediate-level operator. A standardized assessment of the pelvic organs was performed, recording any congenital or acquired uterine pathology and ovarian abnormalities. Visualization of pelvic segments of both ureters was attempted in all cases. The success in finding the ureters, the time required to identify them and their dimensions at rest and while exhibiting peristalsis were all recorded.


A total of 245 consecutive women were included in the study. In all women at least one ureter was successfully identified. Both ureters were seen in 227 women (92.7% (95% CI, 89.4–96.0%)). In 17 (6.9% (95% CI, 3.7–10.1%)) the left ureter was not seen and in one woman (0.4% (95% CI, 0.0–1.2%)) the right ureter could not be visualized (P < 0.001). There were no significant differences in the median time required to visualize the right and left ureters (9.0 (interquartile range (IQR), 6.0–14.0) s vs 8.0 (IQR, 6.0 –14.0) s, respectively; P = 0.9). The median diameter of the right ureter was 1.7 (IQR, 1.4–2.2) mm at rest and 2.9 (IQR, 2.4–3.6) mm during peristalsis. The median diameter of the left ureter was 1.9 (IQR, 1.6–2.3) mm at rest and 2.9 (IQR, 2.4–3.6) mm during peristalsis.


Pelvic segments of normal ureters can be identified in most women on transvaginal ultrasound examination. Visualization of the ureters could be integrated into the routine pelvic ultrasound examination, particularly in women presenting with pelvic pain or in those with suspected pelvic endometriosis. Copyright © 2013 ISUOG. Published by John Wiley & Sons Ltd.


  1. Top of page

Transvaginal ultrasound provides clear images of pelvic organs and is widely used for the routine assessment of women with a wide range of gynecological complaints. Pelvic ultrasound examination is performed in a systematic manner, and involves examination of the uterine morphology, endometrial assessment and ovarian appearance and size. The urinary bladder and urethra can also be examined by transvaginal ultrasound[1]. Measurement of residual bladder volume is routinely performed in women presenting with urogynecological complaints[2]. Examination of the ureters is not included in the routine pelvic examination, as it has been assumed in the past that normal ureters cannot be visualized on the scan[3].

Anatomically, the ureters are tubular structures that measure approximately 25 cm in length, divided into abdominal and pelvic segments of similar length[4]. The ureteric lumen is narrow and is surrounded by transitional epithelium, subepithelial connective tissue, inner longitudinal and outer circular muscle layers and a layer of fibrous tissue. The ureters begin their course from the kidney, posterior to the renal artery, and continue on the anterior edge of the psoas major muscle. At the entrance to the pelvis, the ureters cross the iliac vessels anteriorly. They lie immediately under the peritoneum of the pelvic side wall, behind the lateral attachment of the broad ligaments. Curving medially and forwards, the ureters pass through the base of the broad ligaments below the uterine arteries, approximately 2 cm lateral to the supravaginal part of the cervix a short distance above the lateral fornices of the vagina. Approaching the bladder, the ureters pass medially in front of the upper vagina and enter the bladder base obliquely at the upper angles of the trigone[5].

On ultrasonography ureteric function can be indirectly assessed by identifying the ‘ureteric jet effect’. This was first described by Dubbins et al.[6] in a case series of six patients. This effect is caused by the presence of acoustic interfaces between urine of different specific gravities. Ureteric jet can also be assessed by color Doppler ultrasound and can be used to confirm ureteral patency[7].

The aim of this study was to investigate the possibility of visualizing the pelvic segment of normal ureters directly and measuring their size on a standard gynecological transvaginal ultrasound examination.


  1. Top of page

This was a prospective study of women referred by their general practitioner or consultant gynecologist for transvaginal ultrasound examination. The study was conducted between June 1 2012 and July 31 2012. Consecutive women were examined by one of two operators; Operator A was an ultrasound expert with more than 10 years' experience in gynecological ultrasound while Operator B was an intermediate-level operator with more than 12 months' experience working full-time in a tertiary-level gynecological ultrasound unit. Women who had never been sexually active and those who were unable to tolerate a transvaginal scan were excluded from the study. Indications for the examination are listed in Table 1.

Table 1. Main indications for transvaginal ultrasound examination in 245 women
Indicationn (%)
Pelvic pain64 (26.1)
Abnormal uterine bleeding58 (23.7)
Suspected ovarian cysts30 (12.2)
Ovarian cancer screening23 (9.4)
Suspected fibroids17 (6.9)
Other53 (21.6)

Before starting the study we consulted anatomical textbooks and radiological literature to assess different possible approaches for visualizing the ureters. These approaches were tested during routine pelvic ultrasound examinations over several weeks until agreement was reached on the approach that provided the most consistent way of visualizing the ureters on transvaginal scan. The examination was conducted as follows. All women were asked to empty their bladder prior to examination. Ultrasound scans were performed in a standardized fashion using a 5–7-MHz transvaginal probe with three-dimensional (3D) facility (Voluson E8, GE Healthcare Ultrasound, Milwaukee, WI, USA). First the uterus was examined in the transverse plane to identify the uterine cavity and cervix. Congenital uterine anomalies were suspected in women with complete or partial duplication of the uterine cavity and in those with a single interstitial tube visible on two-dimensional B-mode scan. In all cases of suspected congenital anomalies a 3D ultrasound scan was carried out to clarify the diagnosis[8]. Acquired uterine anomalies, such as fibroids or adenomyosis, were diagnosed based on direct visualization using previously described diagnostic criteria[9, 10]. The probe was then rotated 90° anticlockwise, the uterus and endometrium were visualized in the longitudinal plane and endometrial thickness was measured. The examination was concluded by examining the adnexa in the transverse plane. The ovaries were then identified and were measured in three orthogonal planes. The presence of any ovarian abnormalities was recorded. The diagnosis of endometriosis and adhesions was made in accordance with previously described criteria[11].

Once the examination of the abovementioned organs was complete the urinary bladder was examined in the transverse plane in order to exclude any bladder abnormalities. A longitudinal section through the urethra was performed next and the image was frozen with time displayed on the screen, which was marked as Time 1. The image was unfrozen and the probe was immediately moved towards the lateral pelvic wall in order to identify the distal part of the ureter adjacent to the bladder trigone. Ureters typically appeared as long tubular hypoechoic structures with thick hyperechoic mantle extending from the lateral aspect of the bladder base towards the common iliac vessels (Figure 1). Once the ureter had been clearly visualized the image was frozen again and Time 2 was marked. The total time required to complete the ureteric visualization was calculated as Time 2 minus Time 1. The ureter was then followed to the pelvic brim to the level at which it crosses the common iliac vessels. In women in whom visualization of the ureter was difficult, color and pulsed-wave Doppler were used to differentiate between blood vessels at the lateral pelvic wall and the ureter. The presence of a color Doppler signal and pulsed-wave Doppler waveforms consistent with arterial or venous flow were typical of blood vessels, while their absence was considered suggestive of a ureter. Once the ureter had been identified it was observed for up to 180 s in order to identify peristalsis. The ureteric diameter at rest was measured in all patients in a longitudinal section at the notional intersection of the ureter with a vertical line extending from the lateral edge of the bladder (Figure 2a). The measurement was taken by placing the caliper on the outer edge of the muscularis layer at the junction with the hyperechoic fibrous layer (Figure 2b). In women in whom ureteric peristalsis was observed the measurements were also performed at the peak of ureteric dilatation (Figure 2c). The same procedure was then repeated on the contralateral side.


Figure 1. Longitudinal section of pelvis showing urinary bladder (B) and paravesical position of ureter (U).

Download figure to PowerPoint


Figure 2. Transvaginal ultrasound images illustrating standardized approach to measurement of ureteric diameter. (a) Calipers mark notional intersection between ureter (U) and lateral edge of bladder (B) at which all measurements were taken. (b) Measurements of ureteric diameter were taken by placing calipers on outer edge of muscularis layer at junction with hyperechoic fibrous layer. (c) Measurement of ureteric diameter (calipers) at point of peak dilatation during peristalsis. NL, notional line.

Download figure to PowerPoint

Statistical analysis

The Kolmogorov–Smirnov test was used to test for normal distribution. Age was normally distributed and was expressed as mean ± SD. The time interval to identify the ureter was not normally distributed and was expressed as median (interquartile range (IQR)). We used the t-test to compare the means of normally distributed variables and the Mann–Whitney U-test to compare the medians of non-normally distributed variables. Proportions were expressed as percentage (95% CI). We used the chi-square test to compare proportions.

The separate effect of each variable upon the time taken to observe the ureter was examined separately in a series of univariable analyses. Subsequently the joint effect of the variables was examined together in a multivariable analysis. To limit the number of variables in the multivariable analysis, only those with P < 0.2 from the univariable analyses were included. A backward selection procedure was used to retain only statistically significant variables.

Ethical approval was sought from the local research ethics committee, who deemed that, as examination of the genitourinary system constitutes a standard part of pelvic examination, full ethical review was not required.


  1. Top of page

Two hundred and sixty-one consecutive women were examined by the two observers during the study period. Sixteen women were excluded, 14 because they were unable to tolerate transvaginal ultrasound, and two because they had ureteric abnormalities detected during the examination, thus 245 women were included in the study. The main indications for ultrasound scan are listed in Table 1. The mean age of the women participating in the study was 41.7 (range, 18–84) years. One hundred and ninety-five (79.6%) were premenopausal. Eighty-three (33.9%) women had normal pelvic ultrasound findings while the remaining 162 (66.1%) had various uterine, ovarian or other pelvic abnormalities detected on examination (Table 2).

Table 2. Transvaginal ultrasound findings in 245 women
Diagnosisn (%)
Normal pelvis83 (33.9)
Uterine fibroids36 (14.7)
Ovarian cysts20 (8.2)
Adenomyosis19 (7.8)
Pelvic endometriosis7 (2.9)
Bladder adhesions/abnormalities4 (1.6)
Ovarian adhesions2 (0.8)
Two abnormalities36 (14.7)
Three or more abnormalities38 (15.5)

In all women we successfully identified at least one ureter. Both ureters were seen in 227 women (92.7% (95% CI, 89.4–96.0%)). Only the right ureter was seen in 17 (6.9% (95% CI, 3.7–10.1%)), and only the left in one woman (0.4% (95% CI, 0.0–1.2%)) (P < 0.001). There were no significant differences in the median time required to visualize the right and the left ureters: 9.0 (IQR, 6.0–14.0) s and 8.0 (IQR, 6.0–14.0) s, respectively (P = 0.9). The median diameter of the right ureter was 1.7 (IQR, 1.4–2.2) mm at rest and 2.9 (IQR, 2.4–3.6) mm during peristalsis. The median diameter of the left ureter was 1.9 (IQR, 1.6–2.3) mm at rest and 2.9 (IQR, 2.4–3.6) mm during peristalsis.

Factors affecting visualization of right ureter

Initially the separate effect of each variable upon the time needed to observe the right ureter was examined using linear regression (Table 3). As the outcome was log-transformed, the size of effect of each variable is reported in the form of ratio (CI). For the categorical variables, these indicate the ratio of the time taken in each category relative to the time for a baseline category. For age and body mass index (BMI), which were measured on a continuous scale, the data represent ratios for change in time taken for a given increase in that variable. In all instances, a ratio > 1 would suggest an increased time taken, while a ratio < 1 would imply a decreased time. P-values indicating the significance of each variable are also reported.

Table 3. Univariable analysis of factors affecting length of time needed to visualize right and left ureters by transvaginal ultrasound in 245 women
VariableCategoryRight ureterLeft ureter
Ratio (95% CI)PRatio (95% CI)P
  1. a

    Ratios reported for a 10-unit change in this variable.

  2. b

    Ratios reported for a 5-unit change in this variable.

Agea0.98 (0.92–1.04)0.500.97 (0.90–1.05)0.44
Menopausal statusPre10.3710.21
Post0.92 (0.76–1.11) 0.86 (0.69–1.09)
Body mass indexb1.09 (0.99–1.20)0.091.13 (0.99–1.28)0.06
Pelvic painNo10.3210.33
Yes1.08 (0.93–1.25) 1.10 (0.91–1.32)
Yes1.38 (0.91–2.10) 1.33 (0.75–2.37)
1–20.91 (0.76–1.08) 1.08 (0.87–1.34)
≥30.92 (0.76–1.11) 1.07 (0.85–1.34)
1–20.94 (0.79–1.11) 1.13 (0.92–1.39)
≥30.86 (0.69–1.06) 1.06 (0.82–1.37)
Yes1.06 (0.88–1.28) 1.16 (0.92–1.47)
Pregnancy terminationNo10.7910.11
Yes0.97 (0.77–1.22) 0.80 (0.60–1.05)
Ectopic pregnancyNo10.0810.15
Yes1.45 (0.95–2.19) 1.45 (0.88–2.39) 
Past medical historyNone10.1910.47
Chronic illness1.12 (0.94–1.34) 1.08 (0.87–1.34)
Past surgical historyNone10.9110.66
≥1 operations0.99 (0.80–1.22) 1.06 (0.81–1.38)
Retroverted1.16 (0.90–1.49) 1.14 (0.84–1.55)
Axial0.68 (0.35–1.33) 1.74 (0.43–7.07)
Absent1.57 (1.01–2.45) 1.17 (0.68–2.00)
 Yes1.24 (1.05–1.46) 1.23 (1.00–1.51)
Yes1.06 (0.90–1.24) 0.96 (0.78–1.17)
Deep infiltrating endometriosisNo10.7210.19
Yes1.04 (0.84–1.27) 1.19 (0.92–1.55)
Ovarian cysts (right side)No10.7310.87
Yes1.04 (0.83–1.30) 1.02 (0.79–1.33)
Ovarian adhesionsNo10.1110.002
Yes1.18 (0.97–1.44) 1.46 (1.14–1.86)
20.88 (0.76–1.02) 0.88 (0.73–1.05)
OperatorB1< 0.00110.004
A0.72 (0.62–0.83) 0.76 (0.63–0.92)

The joint effect of the variables upon the time taken to observe the right ureter was examined by multivariable analysis. A backwards selection procedure was used to retain only the statistically significant variables, and the final model is summarized in Table 4.

Table 4. Multivariable backward logistic regression analysis of factors affecting length of time needed to visualize right and left ureters by transvaginal ultrasound in 245 women
VariableCategoryRatio (95% CI)P
  1. a

    Ratios reported for a 5-unit change in this variable. BMI, body mass index.

Right ureter   
 Retroverted1.22 (0.97–1.56)
 Axial0.55 (0.29–1.06)
 Absent1.49 (0.97–2.28)
OperatorB1< 0.001
 A0.70 (0.61–0.82)
Left ureter   
BMIa 1.14 (1.00–1.29)0.04
Ovarian adhesionsNo10.02
 Yes1.62 (1.10–2.39)

The results of the multivariable analysis indicate that uterine position/absence and operator experience were independent predictors of time needed to find the right ureter. After adjusting for these factors, there was no longer any significant effect of adenomyosis upon the outcome.

There was only slight evidence of a difference in effect of the uterine categories anteverted, retroverted, axial and absent on ureteric visualization in the univariable analyses. However, after adjusting for the two other significant factors, the differences were statistically significant. In women with an absent uterus the time needed to visualize the right ureter was on average 49% longer than in women with an anteverted uterus. In women with an axial uterus the examination time to visualize the right ureter was shorter.

As in the univariable analyses, it was found that the more experienced operator was able to visualize the ureter more quickly.

Factors affecting visualization of left ureter

The results showing the effects of the recorded variables on the time taken to visualize the left ureter are summarized in Table 3. Subsequently the joint effect of the variables upon the time taken to observe the left ureter was examined by multivariable analysis. A backwards selection procedure was used to retain only the statistically significant variables, and the final model is summarized in Table 4.

The multivariable analysis indicated that BMI and ovarian adhesions were both significantly associated with time taken to find the left ureter. After accounting for the effects of these variables, there was no longer found to be a significant effect of operator or adenomyosis upon this outcome.

The time needed to find the left ureter was longer for women with a greater BMI; a 5-unit increase in BMI was associated with a 14% increase in time taken. Furthermore, the presence of ovarian adhesions was also associated with a longer time. In women with ovarian adhesions it took, on average, 62% longer to find the left ureter than in women with no adhesions.

Comparison between operators

One hundred and forty-three (58.4%) women were examined by Operator A and 102 (41.6%) by Operator B. There were no significant differences between the operators in total number of ureters visualized on ultrasound scan; 273/286 (95.5% (95% CI, 93.1–97.9%)) for Operator A vs 195/204 (95.6% (95% CI, 92.8–98.4%)) for Operator B (P = 0.94). The median time taken for Operator A to find the right ureter and left ureter was 8.0 (IQR, 6–10) s and 8.0 (IQR, 6–11) s, respectively. For Operator B the median time taken to find the right ureter and left ureter was 13.0 (IQR, 7–18) s and 12.0 (IQR, 6–25) s, respectively.

When comparing the times required to visualize the ureter in the first half (Epoch 1) and second half (Epoch 2) of the study we found that Operator B was significantly faster in Epoch 2 than in Epoch 1 (P = 0.008 for left ureter and P < 0.001 for right ureter). There was no significant difference in time taken between Epoch 1 and 2 for Operator A.


  1. Top of page

The results of our study show that pelvic segments of normal ureters can be visualized on routine transvaginal ultrasound examination. The overall visualization rate was 96% and it was not significantly affected by the experience of the operator. The median time required to visualize each ureter was between 8 and 9 s. This indicates that visualization of the ureters could be incorporated into standard gynecological pelvic examination without significant increase in the examination time. As expected, the speed of identifying the ureters was influenced by the experience of the operator. It is reassuring, however, that the less experienced operator required significantly less time to complete the ureteric examination in the second half of the study, which suggests that the learning curve is steep. Since completion of the study we have taught all ultrasonographers in our unit the technique of visualizing the ureters. All of them have acquired the skill very quickly and they now routinely examine the ureters while conducting gynecological ultrasound examinations.

It was significantly harder to identify the right ureter in women with absent uteri, while the factors affecting visualization of the left ureter were higher BMI and the presence of ovarian adhesions. In the absence of a uterus, useful anatomical landmarks such as uterine arteries are missing, which makes visualization of the ureters more difficult. Larger amounts of intra-abdominal adipose tissue in women with a high BMI is a recognized factor limiting the quality of ultrasound images and therefore it is not surprising that it had a significant effect on the ability to visualize the ureters as well. The ovaries are located in close proximity to the ureters, and severe periovarian adhesions could alter the position of the ureters and make their identification on ultrasound harder.

The ability to visualize the ureters on routine pelvic examination offers some important diagnostic benefits. Yaqoob et al.[12] found that 85% of ureteric stones are located in the distal ureters. This can give rise to pelvic pain that may mimic pelvic inflammatory disease or ovarian cyst accidents. Previous studies have shown that transvaginal ultrasonography can be used to diagnose distal ureteral calculi, but the authors stated that normal ureters cannot be seen in the absence of pathology[13, 14].

In women with pelvic endometriosis, the ureters may be directly affected by the disease or they could be compressed ovarian endometriomas or ovarian adhesions[15]. Detection of ureteric involvement is of critical importance in conducting safe surgical treatment of severe pelvic endometriosis. Preoperative assessment of women with endometriosis usually involves renal ultrasound in order to identify signs of ureteric obstruction. The ability to visualize the ureters directly on transvaginal scan would be of great benefit in the assessment of the severity of pelvic endometriosis, and this would facilitate preoperative counseling of women and help to improve surgical planning. Preoperative detection of ureteric compression in women diagnosed with benign or malignant pelvic tumors would also contribute to the safety of surgery by reducing the risk of ureteric damage.

The ability to visualize the ureters may also facilitate detection of congenital abnormalities such as duplication of the ureter. Detection of a double ureter is particularly useful prior to pelvic surgery and may help to minimize the risks of accidental ureteric injury. Diagnosis of other ureteric abnormalities such as ureterocele could also be made with more confidence and facilitate the differential diagnosis between urinary tract and gynecological pathology.

In conclusion, our study has shown that pelvic ureteric segments can be consistently visualized on transvaginal ultrasonography and that their examination can usually be completed very quickly. We believe that ureteric visualization will be incorporated into standard pelvic ultrasound examination in the future, particularly in women with a history of pelvic pain and in those with clinical and ultrasound diagnosis of endometriosis.


  1. Top of page
  • 1
    Khullar V, Cardozo LD, Salvatore S, Hill S. Ultrasound: a non-invasive screening test for detrusor instability. Br J Obstet Gynaecol 1996; 103: 904908.
  • 2
    Kelly CE. Evaluation of voiding dysfunction and measurement of bladder volume. Rev Urol 2004; 6 (Suppl): S32S37.
  • 3
    Bates J. Anatomy, Physiology and Ultrasound Appearance. In Practical Gynaecological Ultrasound (2nd edn), Bates J (ed). Cambridge University Press: Cambridge, UK, 2006; 3253.
  • 4
    Delancey JOL. Surgical anatomy of the female pelvis. In Te Linde's Operative Gynecology (10th edn), Rock JA, Jones HW (eds). Lippincott Williams and Wilkins: Philadelphia, PA, 2008; 82111.
  • 5
    Drife J, Magowan B. Clinical Obstetrics and Gynaecology. Saunders: London, 2007; 7185.
  • 6
    Dubbins PA, Kurtz AB, Darby J, Goldberg BB. Ureteric jet effect: the echographic appearance of urine entering the bladder. A means of identifying the bladder trigone and assessing ureteral function. Radiology 1981; 140: 513515.
  • 7
    Timor-Tritsch IE, Haratz-Rubinstein N, Monteagudo A, Lerner JP, Murphy KE. Transvaginal color Doppler sonography of the ureteral jets: a method to detect ureteral patency. Obstet Gynecol 1997; 89: 113117.
  • 8
    Salim R, Woelfer B, Backos M, Regan L, Jurkovic D. Reproducibility of three-dimensional ultrasound diagnosis of congenital uterine anomalies. Ultrasound Obstet Gynecol 2003; 21: 578582.
  • 9
    Anderson J. The myometrium. In Gynecologic Imaging, Anderson J (ed). Churchill Livingstone: Philadelphia, PA, 1999; 243.
  • 10
    Naftalin J, Hoo W, Pateman K, Mavrelos D, Holland T, Jurkovic D. How common is adenomyosis? A prospective study of prevalence using transvaginal ultrasound in a gynaecology clinic. Hum Reprod 2012; 27: 34323439.
  • 11
    Holland TK, Hoo WL, Mavrelos D, Saridogan E, Cutner A, Jurkovic D. Reproducibility of assessment of severity of pelvic endometriosis using transvaginal ultrasound. Ultrasound Obstet Gynecol 2013; 41: 210215.
  • 12
    Yaqoob J, Usman MU, Bari V, Munir K, Mosharaf F. Unenchanced helical CT of ureterolithiasis: incidence of secondary urinary tract findings. J Pak Med Assoc 2004; 54: 25.
  • 13
    Yang JM, Yang SH, Huang WC. Transvaginal sonography in the assessment of distal ureteral calculi. Ultrasound Obstet Gynecol 2005; 26: 658662.
  • 14
    Yang JM, Yang SH, Hsu HC, Huang WC. Transvaginal sonography in the morphological and functional assessment of segmental dilation of the distal ureter. Ultrasound Obstet Gynecol 2006; 27: 449451.
  • 15
    Gambadauro P, Olovsson M, Persson P. Unusually rapid growth of bilateral endometriomas and acute bilateral hydronephrosis. Gynecol Endocrinol 2011; 27: 948950.