ENDOLUMINAL MAGNETIC RESONANCE IMAGING IN THE EVALUATION OF URETHRAL DIVERTICULA IN WOMEN DANIEL S. BLANDER, ERIC S. ROVNER, MITCHELL D. SCHNALL, PARVATI RAMCHANDANI, MARC P. BANNER, GREGORY A. BRODERICK, AND ALAN J. WEIN
ABSTRACT Objectives. Accurate determination of the size and extent of urethral diverticula can be important in planning operative reconstruction and repair. Voiding cystourethrography (VCUG) is currently the most commonly used study in the preoperative evaluation of urethral diverticula. We reviewed our experience with the use of endoluminal (endorectal or endovaginal) magnetic resonance imaging (eMRI) in these patients as an adjunctive study to VCUG to evaluate whether the MRI provided anatomically important information that was not apparent on VCUG. Methods. A retrospective analysis of all patients with a clinical diagnosis of urethral diverticula undergoing MRI at a single institution was performed. Patients were evaluated with history, physical examination, cystoscopy, VCUG, and eMRI. Endoluminal MRI was retrospectively compared to VCUG with respect to size, extent, and location found at operative exploration. Results. Twenty-seven consecutive patients underwent endorectal or endovaginal coil MRI in the evaluation of suspected urethral diverticula. Twenty patients subsequently had attempted transvaginal operative repair of the diverticulum. In 2 patients, eMRI demonstrated a urethral diverticulum, whereas VCUG did not. Operative exploration in these patients revealed a urethral diverticulum. In 14 of 27 patients, the VCUG underestimated the size and complexity of the urethral diverticulum as compared to eMRI and operative exploration. In 13 of 27 patients, the size, location, and extent of the urethral diverticulum on VCUG correlated well with the eMRI and/or operative findings. Conclusions. We have found endorectal and endovaginal coil MRI to be extremely accurate in determining the size and extent of urethral diverticula as compared to VCUG. This information can be critical when planning the approach, dissection, and reconstruction of these sometimes complex cases. UROLOGY 57: 660–665, 2001. © 2001, Elsevier Science Inc.
rethral diverticula are a relatively uncommon urologic entity; however, they have been reported to occur in up to 5% of the female population.1,2 They may present with pain, irritative voiding symptoms, urinary incontinence, chronic urinary tract infections, or simply as a painless anterior vaginal mass. Because of the variable nature of presenting symptoms, the urologist must have a high degree of suspicion to confirm the diagnosis in some cases. Many radiologic examinations have been used to From the Division of Urology and Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Reprint requests: Daniel S. Blander, M.D., Southern California Permanente Medical Group-Urology, 6 Willard, Irvine, CA 92604 Submitted: January 28, 2000, accepted (with revisions): November 20, 2000
© 2001, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
aid in the diagnosis of urethral diverticula, including intravenous urography, transvaginal ultrasound, and voiding cystourethrography (VCUG). It is generally felt that the standard of diagnosis for urethral diverticula is positive-pressure urethrography with a double-balloon catheter (PPU).3 Because many institutions lack expertise or equipment necessary to perform PPU, VCUG is often the study used for investigating patients suspected of harboring a urethral diverticulum. Potential problems with VCUG include the necessity for catheterization to perform the study, along with its attendant risk of urinary tract infection. A patient must also void during the study to demonstrate the urethral diverticulum. Given the circumstances of the study, voiding is often inhibited, and subsequently the urethral diverticulum may not be demonstrated, producing a false-negative study. The urethra may be incompletely dis0090-4295/01/$20.00 PII S0090-4295(00)01082-7
TABLE I. Presenting symptoms in 27 patients with confirmed urethral diverticulum Dysuria Recurrent urinary tract infection Urge incontinence Stress urinary incontinence Periurethral mass Hematuria Dyspareunia Urinary retention
imaging, including VCUG and eMRI. We compared eMRI with VCUG in the preoperative evaluation of urethral diverticula to determine whether the eMRI provided additional information that was used during excision and reconstruction of urethral diverticula. MATERIAL AND METHODS
TABLE II. Operative procedures performed in 23 patients with urethral diverticulum Simple excision Complex reconstruction with Martius flap Complex reconstruction without Martius flap Excision with PVS or AVWS with/without Martius flap Transvaginal marsupialization
6 5 2 9 1
KEY: PVS ⫽ pubovaginal sling; AVWS ⫽ anterior vaginal wall suspension.
tended during VCUG due to poor flow, and this may also lead to poor visualization of the urethral diverticulum. Finally, a urethral diverticulum may be incompletely visualized due to loculations or a stenotic ostium, both of which may prevent contrast from delineating the entire urethral diverticulum during voiding. The use of surface or body coil magnetic resonance imaging (MRI) has been described for diagnosis of urethral diverticula4 and other urethral pathology.5– 8 Endorectal and endovaginal coil MRI techniques differ from surface coil MRI in that the coil is placed within a body cavity adjacent to the tissues of interest, producing an improved signal-to-noise ratio and high resolution imaging of these areas. Kim et al.9 originally reported the use of such a coil in a small number of their patients with urethral diverticula. Recently, we have applied these endoluminal MRI (eMRI) techniques to the evaluation of urethral diverticula. We reviewed our experience with diagnosis and management of urethral diverticula with reference to preoperative
From July 1993 to October 1999, 32 patients were diagnosed at our institution with urethral diverticula. A total of 27 patients had both VCUG and eMRI, and, of these, 20 patients underwent attempted surgical repair. Five patients did not have eMRI for the following reasons: surgeon preference, claustrophobia, inability to tolerate the endorectal or endovaginal probe, and metallic foreign bodies. Patients were evaluated with a detailed history and physical examination with attention directed toward surgical and obstetric history. Presenting complaints are listed in Table I. Voiding symptomatology was also carefully evaluated, and those patients with urinary incontinence were further evaluated with pressure flow urodynamic studies. Cystourethroscopy was performed in all patients under either local or general anesthesia with attempts directed toward identifying the diverticular ostium. In all cases, VCUG was performed without double-balloon occlusion. VCUG was performed using sterile technique in all patients by an experienced uroradiologist (M.P.B. or P.R.). A Foley catheter was inserted into the bladder, and the bladder was filled with 300 to 500 cc of 30% Hypaque. Voiding was initiated, and anteroposterior, lateral, and right posterior oblique views were obtained during micturition. Endoluminal MRI was performed in all cases, using a 1.5Tesla magnet with either endovaginal or endorectal coil. MRI sequences included T1 axial and sagittal; T2 axial, sagittal, and coronal; and fat-suppressed sagittal fast spin echo. In general, diverticula appeared as low-signal intensity periurethral masses on T1-weighted images and as high-signal intensity periurethral masses on T2-weighted images. In 20 cases, surgical exploration with transvaginal excision of the urethral diverticulum was performed (Table II).10,11 All surgeries were performed by one of two surgeons (G.A.B. or E.S.R.). Operative and pathology reports were reviewed on all patients. Radiologic studies, including VCUG and MRI, were retrospectively reviewed. A discrepancy between the VCUG and eMRI was considered to exist if there was more than a total of a 1-cm difference in measurement of the three dimensions reported for the diverticulum, or if the diverticulum appeared simple on one study and multilocular or multilobed on the other study.
TABLE III. Comparison of staging procedures eMRI
Urethral diverticulum 27/27 (100%) 23/27 (85%) visualized Modality demonstrating 14/27 (52%) 0/27 (0%) greater complexity Neck of diverticulum 11/27 (41%) 0/27 (0%) visualized Loculations visualized 5/27 (19%) 2/27 (7%)
Cystoscopy N/A N/A 14/27 (52%) N/A
No Difference N/A 13/27 (48%) N/A N/A
KEY: eMRI ⫽ endoluminal magnetic resonance imaging; VCUG ⫽ voiding cystourethrography; N/A ⫽ not applicable.
UROLOGY 57 (4), 2001
FIGURE 2. VCUG (A) and axial T2-weighted eMRI (B) in a case in which the eMRI demonstrates loculations of a urethral diverticulum (white arrows) that are not seen on VCUG.
FIGURE 1. VCUG (A) and axial T2-weighted eMRI (B) in a case in which VCUG underestimated the size of the diverticulum (white arrows).
RESULTS In all 27 patients, eMRI identified the presence of a urethral diverticulum. The diverticular neck (ostium) was identified on 11 of 27 eMRI examinations. VCUG identified the presence of a diverticulum in 25 of 27 patients. The diverticular neck was not clearly defined on any VCUG. Cystoscopy demonstrated the neck of the diverticulum in 14 of 662
27 patients. Surgical exploration confirmed the presence of a urethral diverticulum in all 20 patients undergoing surgical exploration. The results of the staging studies are summarized in Table III. The VCUG underestimated the size and/or complexity of the diverticulum in 14 of 27 cases (Fig 1), whereas eMRI and surgical exploration were in near agreement in all cases. In contrast, the eMRI did not underestimate the size or complexity of the diverticulum in any case. The results were concordant in 13 of 27 patients. In all cases, the ostium of the diverticulum was identified at the time of surgical exploration. When the diverticulum was noted to extend in a saddlebag fashion or circumferentially on eMRI, this observation was confirmed by operative exploration in all cases. UROLOGY 57 (4), 2001
TABLE IV. Summary of previous pertinent MRI studies describing diverticula of the female urethra Author (yr)
Hricak (1991) Reuter (1991)6 Kim (1993)9
Retrospective review Case report Retrospective review
Surface coil Surface coil Body coil or pelvic coil
Klutke (1995)7 Debaere (1995)4
Case report Case report
Surface coil Body coil
Endorectal coil vs. body coil
Blander (1999)14 Daneshgari (1999)15
Case report Case report
Endorectal coil Endorectal coil
9 1 20
13 1 3
Findings Excellent image quality Excellent image quality MRI very accurate compared to urethrography and cystoscopy Tumor mass with diverticulum MRI superior to ultrasound in diagnosing diverticulum MRI demonstrated intraurethral wall diverticulum not visualized on VCUG Endovaginal coil provides superior imaging of female pelvis compared to body coil MRI more sensitive than double balloon urethrography Excellent image quality Excellent image quality
KEY: MRI ⫽ magnetic resonance imaging; VCUG ⫽ voiding cystourethrography.
COMMENT Urethral diverticula are an uncommon surgical entity. Their presentation is highly variable, and they may elude detection for many years. In most cases, a thorough history and physical examination should yield some clues as to the presence of a diverticulum, but radiologic assessment is often necessary to establish the diagnosis. Radiographic evaluation may also define the extent of a diverticulum or note the presence of loculations or coexistent masses, such as stones or tumors that may aid in the planning of appropriate surgery. Once diagnosed, repair and reconstruction of urethral diverticula can be quite challenging. The principles of urethral diverticulectomy include complete excision or destruction of the mucosal surface, watertight urethral closure, and multilayered closure with preservation and reconstruction of the periurethral fascia.10 We believe that an accurate preoperative radiographic evaluation and staging of urethral diverticula is critical to understand the complexity of the diverticulum as well as to plan an appropriate and successful operation. PPU is generally accepted as the standard in evaluation of suspected urethral diverticula. Simultaneous balloon occlusion of the bladder neck and external urethral meatus during PPU permits sufficient distention of the urethral lumen under pressure such that most urethral diverticula fill with contrast and are well delineated. Unfortunately, because they lack expertise or the specialized equipment needed for this technique, many centers do not routinely perform it. Additional limitaUROLOGY 57 (4), 2001
tions with PPU include the inherent risk of urinary tract infection and urethral injury due to urethral instrumentation as well as the inability to visualize loculated portions of the diverticulum or urethral diverticula with very narrow ostia (Fig 2). Thus, VCUG is currently performed in most centers in the evaluation of suspected urethral diverticula. This study carries with it the same limitations as PPU but in addition requires that the patient void to demonstrate the diverticulum. An inability to void or a poor urinary flow rate may preclude visualization of the diverticulum. VCUG may also not be predictive of the pathology found at operative exploration, as it may underestimate the size and complexity of urethral diverticula for the reasons noted previously. Endorectal and surface coil MRI have previously been used to evaluate female urethral diverticula (Table IV). With the access to eMRI technology at our institution and the advent of endoluminal techniques, we investigated whether noninvasive endoluminal MRI imaging of urethral diverticula would provide improved preoperative staging that would be helpful in the reconstruction of these cases. We found that eMRI is very accurate in defining the complexity, location, and extent of urethral diverticula as compared to VCUG. In the current series, VCUG underestimated the size or complexity of the diverticulum in 52% of the cases. Three cases had loculated collections missed on VCUG that were seen on MRI and thus excised. Similarly, Daneshgari et al.15 recently demonstrated noncommunicating intraurethral wall 663
FIGURE 3. VCUG (A) and axial T2-weighted eMRI (B), demonstrating a saddlebag urethral diverticulum (white arrows) in which the dorsal wall was used for urethral reconstruction.
diverticula, using eMRI. These investigators suggest that, in addition to its other advantages, eMRI is superior to VCUG in the diagnosis of urethral processes that do not communicate with the urethral lumen. In the complex group of patients presented herein, several of whom required complex urethral reconstruction after diverticular excision, eMRI was particularly helpful. Not uncommonly, urethral diverticula may extend partially or completely around the urethra (a saddlebag configuration). Indeed, 5 patients in this series had circumferential urethral diverticula. Not only is removal or destruction of the entire mucosal surface important to prevent recurrence in these cases, but also extensive urethral reconstruction using flaps or grafts may be required given the extent of ure664
thral wall involvement. The eMRI predicted the complexity of urethral involvement and, by completely defining the extent of the diverticulum, permitted complete excision and reconstruction. In 3 of these cases, a dorsal portion of the wall of the diverticulum was used as a flap for urethral reconstruction.16 Preoperative knowledge of the presence of this favorable anatomic configuration was critical in preserving this portion of the diverticulum intraoperatively for a flap (Fig 3). Although surface coil MRI has been previously reported in the evaluation of urethral diverticula,5,6 the use of endoluminal coils may provide greater anatomic detail and resolution of the area of interest.13–15 Although this study did not directly compare surface coil MRI to eMRI, we believe that the images provided by an endoluminal coil in the investigation of urethral pathology are outstanding, and we, therefore, no longer use surface coil MRI for these cases unless the patient cannot tolerate the endoluminal probe. There are some potential disadvantages to the use of eMRI imaging of the female urethra. Endoluminal technology is neither universally available nor routinely performed. Additionally, the endoluminal coil may be uncomfortable especially when performed in the setting of an actively infected diverticulum, although probably less so than a VCUG performed under similar circumstances. Finally, eMRI is costly as compared to standard VCUG, but it is possible that, with the expanding use of eMRI as an imaging modality, the expense will decrease over time. Two technical points regarding our study deserve mention. First, retrospective review of operative reports is subject to bias. In the future, a prospective study, including specific documentation of operative findings and how they compare to the eMRI, would be more accurate. In no instance in this series, however, was our planned procedure significantly altered on appreciation of the patient’s actual anatomic findings. Furthermore, this group of patients represents only a select subset of patients with urethral diverticula. This group of patients had significantly complicated diverticula, representative of those seen at a referral center. Thirty percent of these patients had undergone previous surgery, and 52% had multiloculated or multilobed type diverticula. It is possible that, outside a tertiary care center, a greater percentage of patients would have simple urethral diverticula, and, in the imaging of less complex, primary urethral diverticula, there might be less divergence between VCUG and eMRI findings. Transvaginal ultrasound (TVU) of the urethra may also prove to be a useful imaging modality to investigate urethral diverticula. TVU has been UROLOGY 57 (4), 2001
demonstrated to be helpful in diagnosing masses of the female urethra,17,18 but, to our knowledge, it has not been compared to endoluminal MRI. CONCLUSIONS We find that endoluminal MRI is very accurate in assessing the size, complexity, extent, and relative location of urethral diverticula. All of these factors are important in planning operative reconstruction of urethral diverticula. Unlike voiding cystourethrography, eMRI does not require invasive urethral catheterization or a specially trained technician. This preliminary study demonstrates the utility and accuracy of eMRI in the evaluation of urethral diverticula, especially in the evaluation of complex and recurrent cases. REFERENCES 1. Dmochowski RR, Ganabathi K, Zimmern PE, et al: Benign female periurethral masses. J Urol 152: 1943–1951, 1994. 2. Young GPH, Wahle GR, and Raz S: Female urethral diverticulum, in Raz S (Ed): Female Urology. Philadelphia, WB Saunders, 1996, pp 477– 489. 3. Davis HJ, and Cian LG: Positive pressure urethrography: a new diagnostic method. J Urol 75: 753, 1956. 4. Debaere C, Rigauts H, Steyaert L, et al: MR imaging of a diverticulum in a female urethra. J Belge Radiol 78: 345–346, 1995. 5. Hricak H, Secaf E, Buckley DW, et al: Female urethra: MR imaging. Radiology 178: 527–535, 1991. 6. Reuter KL, Young SB, Davidoff A, et al: Magnetic resonance imaging of an infected urethral diverticulum: a case report. Magn Reson Imaging 9: 955–957, 1991. 7. Klutke CG, Akdman EI, and Brown JJ: Nephrogenic
UROLOGY 57 (4), 2001
adenoma arising from a urethral diverticulum: magnetic resonance features. Urology 45: 323–325, 1995. 8. Nurenberg P, and Zimmern PE: Role of MR imaging with transrectal coil in the evaluation of complex urethral abnormalities. Am J Radiol 169: 1335–1338, 1997. 9. Kim B, Hricak H, and Tanagho EA: Diagnosis of urethral diverticula in women: value of MR imaging. AJR Am J Roentgenol 161: 809 – 815, 1993. 10. Leach GE, Schmidbauer CP, Hadley HR, et al: Surgical treatment of female urethral diverticulum. Semin Urol 4: 33– 42, 1986. 11. Ganabathi K, Leach GE, Zimmern PE, et al: Experience with the management of urethral diverticulum in 63 women. J Urol 152: 1445–1452, 1994. 12. Tan IL, Stoker J, and Lameris JS: Magnetic resonance imaging of the female pelvic floor and urethra: body coil versus endovaginal coil. Magma 5: 59 – 63, 1997. 13. Neitlich JD, Foster HE, Glickman MG, et al: Detection of urethral diverticula in women: comparison of a high resolution fast spin echo technique with double balloon urethrography. J Urol 149: 408 – 410, 1998. 14. Blander DS, Broderick GA, and Rovner ES: Magnetic resonance imaging of a “saddle bag” urethral diverticulum. Urology 53: 818 – 819, 1999. 15. Daneshgari F, Zimmern PE, and Jacomides L: Magnetic resonance imaging detection of symptomatic noncommunicating intraurethral wall diverticula in women. J Urol 161: 1259 –1262, 1999. 16. Rovner ES, Wein AJ, and Raz S: Reconstruction of the extended saddle bag urethral diverticulum in the female: the diverticular sac flap. Presented at the American Urological Association Mid-Atlantic Section Meeting, Palm Beach, Florida, October 1998. 17. Fontana D, Porpigli F, Morra I, et al: Transvaginal ultrasonography in the assessment of organic diseases of female urethra. J Ultrasound Med 18: 237–241, 1999. 18. Siegel CL, Middleton WD, Teefey SA, et al: Sonography of the female urethra. AJR Am J Roentgenol 170: 1269 –1274, 1998.