CHAPTER 26: Genitourinary Fistula and Urethral Diverticulum

A genitourinary fistula is defined as an abnormal communication between the urinary (ureters, bladder, urethra) and the genital (uterus, cervix, vagina) systems. The true incidence of genitourinary fistula is unknown and varies according to whether the etiology is obstetric or gynecologic. In Asia and Africa, up to 100,000 new cases of obstetric genitourinary fistula are added each year to the estimated pool of 2 million women with unrepaired fistulas (World Health Organization, 2014). For industrialized countries, most fistulas occur iatrogenically from pelvic surgery, and the generally accepted incidence derives from data on surgeries to correct these fistulas. For example, numbers from the National Hospital Discharge Survey of inpatient women show that approximately 4.8 per 100,000 women underwent lower reproductive tract fistula repair (Brown, 2012). This likely is underestimated as many cases are unreported, unrecognized, or treated conservatively. Of genitourinary fistulas, the vesicovaginal fistula is most common and develops significantly more frequently than ureterovaginal fistulas (Goodwin, 1980; Shaw, 2014).

The principles and phases of wound healing aid the understanding of genitourinary fistula pathogenesis. After injury, tissue damage and necrosis stimulate inflammation, and the process of cell regeneration begins (Kumar, 2015). Initially at the injury site, new blood vessels form, that is, angiogenesis. Three to 5 days after injury, fibroblasts proliferate and subsequently synthesize and deposit extracellular matrix, in particular collagen. This fibrosis phase determines the final strength of the healed wound. Collagen deposition peaks approximately 7 days after injury and continues for several weeks. Subsequent scar maturation and organization, termed remodeling, augments wound strength. These phases are interdependent and any disruption of this sequence eventually may create a fistula. Most defects tend to present 1 to 3 weeks after tissue injury. This is a time during which tissues are most vulnerable to alterations in the healing environment, such as hypoxia, ischemia, malnutrition, radiation, and chemotherapy. Edges of the wound eventually epithelialize, and a chronic fistulous tract is thus formed.

Although many classification systems exist for genitourinary fistula, no single system is considered the accepted standard nor is any one scheme superior in predicting surgical success. Fistulas can develop at any point between the genital and urinary systems, and one classification method reflects the anatomic communication (Table 26-1).

Vesicovaginal fistulas can also be characterized by their size and location in the vagina. They are termed high vaginal, when found proximally in the vagina; low vaginal, when noted distally; or midvaginal, when identified centrally. For instance, posthysterectomy vesicovaginal fistulas are often proximal, or “high” in the vagina, and located at the level of the vaginal cuff.

Others classify vesicovaginal fistula based on the complexity and extent of involvement (Table 26-2) (Elkins, 1999). In this scheme, complicated vesicovaginal fistulas are those that involve pelvic malignancy, prior radiation therapy, a shortened vaginal length, or bladder trigone; those that are distant from the vaginal cuff; or those that measure >3 cm in diameter.

In one obstetric classification system, high-risk vesicovaginal fistulas are described by their size (>4 to 5 cm in diameter); involvement of urethra, ureter(s), or rectum; juxtacervical location with an inability to visualize the superior edge; and reformation following a failed repair (Elkins, 1999).

A surgical classification to objectively evaluate obstetric urinary fistula repair has also been introduced (Waaldijk, 1995). In this system, type I fistulas are those that do not involve the urethral closure mechanism, type II fistulas do, and type III fistulas involve the ureter and include other exceptional fistulas. Type II fistulas are divided into: (A) without or (B) with subtotal or total urethra involvement. Type IIB fistulas are further subdivided as: (a) without or (b) with a circumferential configuration around the urethra.

To aid objective comparison of surgical outcomes, a more comprehensive and standardized classification system has been developed (Table 26-3). It integrates fistula distance from the external urethral meatus, fistula size, degree of surrounding tissue fibrosis, and extent of vaginal length reduction (Goh, 2004). This system has good inter- and intraobserver reproducibility and has demonstrated efficacy in predicting which patients are at risk of postfistula urinary incontinence and failure of closure (Goh, 2008, 2009). Despite the availability of these numerous classification systems, most clinicians will, from a practical standpoint, often use the anatomic communication and the relative position in the vagina (high, mid, low) in their initial description of the fistula.

Congenital genitourinary fistulas are rare, but if found, are commonly associated with other renal or urogenital abnormalities. Thus, most vesicovaginal fistulas are acquired and typically result from either obstetric trauma or pelvic surgery.

Obstetric Trauma

In developing countries, more than 70 percent of genitourinary fistulas arise from obstetric trauma, specifically from prolonged or obstructed labor or complicated cesarean delivery (Arrowsmith, 1996; Kumar, 2009; Raassen, 2014). Their development in this setting often reflects social practices or obstetric management common to a particular community or geographic region. For example, both childbearing at a young age, before the pelvis has completely developed or fully grown, and female circumcision, more correctly termed female genital mutilation, may significantly narrow the vaginal introitus and obstruct labor. Prolonged obstructed labor or anatomic malpresentation of the presenting fetal part can cause pressure and ischemic necrosis of the anterior vaginal wall and bladder, subsequently resulting in fistula formation. Alternatively, the vagina may be damaged by instruments used to deliver stillborn infants or perform abortion. Malnutrition and limited health care in many of these countries can further diminish wound healing.

In contrast, in most developed countries, fistulas uncommonly follow obstetric procedures or deliveries. Rarely, cesarean deliveries, usually those accompanied by obstetric complications, have led to complex urinary fistula (Billmeyer, 2001). Similarly, rare cases following cervical cerclage have been reported (Massengill, 2012).

Pelvic Surgery

In developed countries, iatrogenic injury during pelvic surgery is responsible for 90 percent of vesicovaginal fistulas, and the accepted incidence of fistula formation after pelvic surgery is 0.1 to 2 percent (Harris, 1995; Hilton, 2012a,b; Tancer, 1992). The remaining fistulas result from procedures performed by urologists and by colorectal, vascular, and general surgeons. In industrialized countries, hysterectomy is the most common surgical precursor to vesicovaginal fistula, accounting for approximately 75 percent of fistula cases (Symmonds, 1984). When all hysterectomy types are considered, vesicovaginal fistula is estimated to complicate 0.8 per 1000 procedures (Harkki-Siren, 1998). In their review of more than 62,000 hysterectomy cases, laparoscopic hysterectomies were associated with the greatest incidence (2 per 1000), followed by abdominal (1 per 1000), vaginal (0.2 per 1000), and supracervical (0 per 1000) hysterectomies. With hysterectomy for benign disease, Duong and colleagues (2009) noted that bladder wall laceration extending into the bladder neck or a ureteral orifice (trigone) significantly increased the risk of subsequent vesicovaginal fistula.

Because most genitourinary fistulas follow pelvic surgery, prevention and intraoperative recognition of lower urinary tract injury is imperative. As discussed extensively in Chapter 40, intraoperative cystoscopy has been shown to improve the detection rate of lower urinary tract injuries. This in turn may ultimately translate into a lower incidence of genitourinary fistula. Thus, intraoperative cystoscopy can be a useful adjunct, particularly in cases in which the ureters or bladder are suspected to have been at increased injury risk.

Other Causes

Other etiologies for urinary tract fistulas include radiation therapy, malignancy, trauma, foreign bodies, infections, pelvic inflammation, and inflammatory bowel disease. Of these, radiation therapy induces an endarteritis that can lead to tissue necrosis and subsequent potential fistula formation. This modality is a frequent cause, and some series have reported that up to 6 percent of genitourinary fistulas can result from radiation (Lee, 1988). Although most damage following this therapy develops within weeks and months, associated fistulas have been reported to present up to 20 years after the original insult (Graham, 1967; Zoubek, 1989).

Malignancy is commonly linked with tissue necrosis and may lead to urinary fistula formation. Emmert and Kohler (1996) found a 1.8-percent incidence of rectovaginal and vesicovaginal fistula in their analysis of nearly 2100 women with cervical cancer. Thus, tissue biopsy is routinely considered during diagnostic evaluation of women with a fistula and history of malignancy.

Trauma sustained during sexual activity or sexual assault can result in genitourinary fistula formation and has been estimated to precede 4 percent of these defects (Kallol, 2002; Lee, 1988). Foreign bodies such as a neglected pessary or vesical calculi are also documented causes (Arias, 2008; Dalela, 2003). Given that transurethral catheter placement has been linked to urethrovaginal fistula, this commonly used device should be placed, maintained, and removed with care (Dakhil, 2014). Foreign material introduced during surgery such as collagen injected transurethrally and complications resulting from synthetic mesh placement for urinary incontinence or pelvic organ prolapse are other inciting agents (Blaivas, 2014; Firoozi, 2012; Pruthi, 2000). Also, during sling surgeries, excess sling tension may increase tissue stress and necrosis. Thus, initial material selection and patient evaluation for poor wound healing risk factors are important prevention steps (Giles, 2005). Ideally, the material selected minimizes the normal foreign-body reaction, is nontoxic and nonantigenic, and is porous enough to admit immune and phagocytic cells and promote native tissue ingrowth (Birch, 2002). Mesh selection is further discussed in Chapter 24.

Other rare causes of fistula formation include infections such as lymphogranuloma venereum, urinary tuberculosis, pelvic inflammation, and syphilis; inflammatory bowel disease; and autoimmune disease (Ba-Thike, 1992; Monteiro, 1995). Additionally, conditions that interfere with healing such as poorly controlled diabetes mellitus, smoking, local infection, peripheral vascular disease, and chronic corticosteroid use are potential risks.

Vesicovaginal fistula classically presents with unexplained continuous urinary leakage from the vagina after a recent operation. Depending on the size and location of the fistula, the urine amount will vary. Occasionally small-volume, intermittent leakage is mistaken for postoperative stress incontinence. For this reason, patients with new-onset urinary leakage, particularly in the setting of recent pelvic surgery, are examined thoroughly to exclude fistula formation. Other less specific symptoms of genitourinary fistula include fever, pain, ileus, and bladder irritability.

Vesicovaginal fistula may present days to weeks after the initial inciting surgery, and those following hysterectomy typically present at 1 to 3 weeks. Some fistulas, however, have longer latency, and symptoms may develop several years later.

A thorough history and physical examination identifies most cases of vesicovaginal fistula. Accordingly, information is documented regarding obstetric deliveries, prior surgeries, previous fistula management, and malignancy treatment, especially pelvic surgery and radiation therapy.

Physical examination is equally informative, and vaginal inspection often will identify the defect. A meticulous assessment for other fistulous tracts is performed, and their location and size noted. Visual assistance with an endoscopic lens and translucent vaginal speculum can sometimes help identify a vaginal-apex fistula, which can be more difficult to detect.

It is essential to differentiate between “extraurethral” urinary leakage, as with a fistula, and “transurethral” leakage, that is, through the urethra, as with stress urinary incontinence. Occasionally, the vaginal fluid source is unclear, and a small amount of urine can easily be mistaken for vaginal discharge. Measurement of the vaginal fluid’s creatinine content can sometimes be used to confirm its origin. Although creatinine levels in urine vary, with mean levels reaching 113.5 mg/dL, a value >17 mg/dL is consistent with urine (Barr, 2005). In contrast, fluid with a concentration <5 mg/dL is highly unlikely to be human urine.

Although a genitourinary fistula ideally is visualized directly, inspection at times is unrevealing. In these circumstances, retrograde bladder instillation of visually distinct solutions such as sterile milk or dilute methylene blue or indigo carmine can often indicate a fistula and aid in its localization.

If the presence of a urinary fistula is uncertain or its vaginal location is not identified, a three-swab test, commonly known as the “tampon test,” is recommended (Moir, 1973). This test is commonly performed with a tampon. However, we recommend using two to four pieces of gauze sequentially packed into the vaginal canal. A diluted solution of methylene blue or indigo carmine is instilled into the bladder using a transurethral catheter. Notably, use of the former may increase given current indigo carmine shortages. After 15 to 30 minutes of routine activity, the gauze is removed serially from the vagina, and each is inspected for dye. The specific gauze colored with dye suggests the fistula location—a proximal or high location in the vagina for the innermost gauze and a low or distal fistula for the outermost. If the distally placed sponge is stained with dye, however, confirmation that it was not contaminated by urine leaking out through the urethra, as in the case of stress urinary incontinence, is essential.

Cystourethroscopy is another valuable diagnostic tool (Fig. 26-1). It permits fistula localization, determination of its proximity to the ureteral orifices, inspection for multiple fistula sites, and assessment of surrounding bladder mucosa viability. In addition, the use of cystourethroscopy and vaginoscopy concurrently to identify vesicovaginal fistula has been described (Andreoni, 2003).

Concomitant ureteral involvement is estimated to complicate 10 to 15 percent of vesicovaginal fistula cases and is sought during diagnostic evaluation (Goodwin, 1980). At our institution, intravenous contrast-enhanced computed tomography (CT) scanning in the excretory phase has become the preferred diagnostic test after initial cystourethroscopic survey is completed (Fig. 26-2). Selection of modalities other than CT for fistulous tract identification may be considered based on cost or availability. First, intravenous pyelography (IVP) can adequately confirm integrity of the upper collecting system and exclude ureteral involvement in a fistula. Second, retrograde pyelography may be used. Often carried out in conjunction with cystoscopy, it is performed by placing a small catheter into the distal ureter. Contrast material is injected through the catheter into one or both ureters. Fluoroscopic or conventional radiographs are then obtained. Retrograde pyelography generally has been reported to have the same diagnostic value as IVP.

In some instances resources are scarce, cost may be a limitation, and access to specialized diagnostic imaging is a challenge. With some advance planning, phenazopyridine hydrochloride (Pyridium) can be used in conjunction with the three-swab test to determine ureteral involvement, as a very rudimentary alternative to the aforementioned more sophisticated imaging. This tablet is administered orally, is excreted renally, acts as a topical bladder analgesic, and stains urine orange as a side effect. Women with suspected ureteral involvement are instructed to take a 200-mg dose a few hours before their clinic appointment. The steps of the three-swab test are then performed, as described earlier. In this case, if the most proximal (innermost) gauze is colored with orange dye, ureteral involvement is suspected. If both orange and blue dyes are seen, then involvement of both the bladder and ureter(s) is suspected.

Voiding cystourethrography (VCUG) can help confirm the presence, location, and number of fistulous tracts. In this, the bladder is filled via catheter with contrast dye, and fluoroscopic images of the lower urinary tract are obtained during patient micturition. However, CT has largely replaced this modality. Transabdominal sonography with applied color Doppler to identify flow through the fistula has been suggested as another diagnostic option (Volkmer, 2000). However, without color Doppler, sonography failed to identify 29 percent of vesicovaginal fistula cases in one study (Adetiloye, 2000).

Conservative Treatment

Occasionally, genitourinary fistulas may spontaneously close during continuous bladder drainage using an indwelling urinary catheter. Approximately 12 percent of women treated by sustained catheterization alone had fistulas that healed spontaneously (Oakley, 2014; Waaldijk, 1994). Romics and colleagues (2002) found that in 10 percent of cases, urinary fistulas close spontaneously after 2 to 8 weeks of transurethral catheterization, especially if the fistula is small (2- to 3-mm diameter). Another series reported fistulas up to 2 cm in diameter spontaneously healed in 50 to 60 percent of patients treated with an indwelling catheter (Waaldijk, 1989).

Despite these series, data that correlate fistula size and success of conservative management are limited. Many reports of successful spontaneous closure with catheter drainage have been limited to fistulas that were 1 cm in size or smaller (Lentz, 2005; Ou, 2004). Many studies are vague regarding how fistula size is measured, and each series has potential for considerable bias in its selection criteria. However, in general, the larger a fistula, the less likely it is to heal without surgery.

Evidence regarding the duration of catheter drainage also varies. Regardless, many agree that if a fistula has not closed within 4 weeks, it is unlikely to do so. This may be secondary to epithelialization of the fistulous tract (Davits, 1991; Tancer, 1992). Moreover, continued urinary drainage may lead to further bladder inflammation and irritation (Zimmern, 1991). Importantly, if attempting conservative treatment of a vesicovaginal fistula with catheter insertion and chronic drainage, urinary drainage ideally begins shortly after the inciting event.

Fibrin sealant (Tisseal, Evicel), also colloquially called fibrin glue, is formed from concentrated fibrinogen combined with thrombin to simulate the final clotting cascade stages. In gynecologic surgery, it is mainly used to control low-pressure bleeding. Although fibrin sealant has been described for the treatment of vesicovaginal fistula, it is often selected as a surgical adjunct rather than primary surgical treatment (Evans, 2003). Data regarding fibrin sealant effectiveness are sparse, and well-designed trials are lacking. Thus, fibrin sealant monotherapy may not be the initial recommended treatment in most vesicovaginal fistula cases due to potential lack of durability and thus a risk for recurrence. However, it may provide a viable alternative in patients with multiple comorbidities that contraindicate a prolonged fistula repair surgery.

In sum, a trial of conservative therapy is usually warranted and reasonable, especially if instituted shortly after the inciting event and if the fistula is small. However, gains from a conservative approach are balanced against a patient’s desire for an expedited repair to resolve the leak. Thus, the timing of intervention ideally achieves a compromise between reasonable conservative efforts and addressing the patient’s immediate distress and quality of life. As noted, most urinary fistulas ultimately require surgical intervention.

Surgical Treatment

Incorporating the fundamentals of genitourinary fistula repair is essential to successful resolution. These include accurate fistula delineation; adequate assessment of surrounding tissue vascularity; timely repair; multilayer, tension-free, and watertight defect closure; and postoperative bladder drainage.

Primary surgical repair of genitourinary fistula is associated with high cure rates (75 to 100 percent) (Rovner, 2012b). Factors that support this rate include adequate vascularity of the surrounding tissue, brief fistulous tract duration, no prior radiation therapy, meticulous surgical technique, and surgeon experience. The first attempt at surgical repair is usually associated with the best chance of successful healing. Surgical repair success rates specifically for obstetric fistulas are also high. Of these, 81 percent are corrected with the first attempt, and 65 percent with the second (Elkins, 1994; Hilton, 1998).

Timing of Repair

One principle of fistula repair dictates that a repair be performed in noninfected and noninflamed tissues. Early surgical intervention of uncomplicated fistulas within the first 24 to 48 hours following the inciting surgery is possible as it avoids the brisk postoperative inflammatory response. Such early closure does not affect success rates, yet it appears to reduce social and psychologic patient distress (Blaivas, 1995; Persky, 1979).

In instances of extensive and severe inflammation, we recommend delaying operative repair for 6 weeks until the inflammation subsides. During this time, a trial of catheter drainage, while the surrounding tissue has an opportunity to heal, is reasonable.

Route of Surgical Repair

Many different surgical repair options are available for vesicovaginal fistula. However, data that support an optimal route are limited, and the lack of consensus may reflect variances in surgeon experience and preference. Among important surgical considerations, ability to gain access to the fistula is essential and commonly dictates the surgical approach. Fortunately, success rates are high whether the route of repair is transvaginal or transabdominal.

Vaginal

The transvaginal approach to genitourinary fistula repair is straightforward and direct. Compared with abdominal approaches, it is associated with shorter operative times, decreased blood loss, less morbidity, and shorter hospital stays (Wang, 1990). The transvaginal route also allows easy access for the use of ancillary equipment, such as ureteral stents. This is particularly useful if the fistula is located near ureteral orifices.

One transvaginal approach used most commonly by gynecologists, the Latzko technique, is illustrated in Section 45-10. In this technique, likened to a partial colpocleisis, the most proximal portions of the anterior and posterior vaginal walls are surgically apposed to close the defect, without completely removing the fistulous tract. This partially obliterates the upper vagina, similar to that achieved by colpocleisis. Because of the potential for vaginal shortening, this technique may not be appropriate if vaginal depth has already been compromised or if there is preexisting sexual dysfunction. If use of the Latzko technique is anticipated, patient counseling specifically addresses these issues and potential sequelae. That said, recent studies evaluating sexual function show similar or higher functioning scores following vaginal repair routes compared with abdominal routes (Lee, 2014; Mohr, 2014).

The classical technique, in contrast to the Latzko method, involves total excision of the fistulous tract and mobilization of the surrounding anterior vaginal wall epithelium. After tract resection, the bladder mucosa is first closed, and a watertight repair is confirmed. This is followed by closure of one or two layers of fibromuscular tissue. Vaginal epithelium is then reapproximated.

Of the two approaches, some favor incomplete fistulous tract excision (Latzko repair) to avoid weakening the surrounding tissue, enlarging the defect, and thereby potentially compromising the repair. By preserving the presumptively stronger scar tissue surrounding the fistula, it theoretically permits a more secure reapproximation of surrounding tissue.

Abdominal (Transperitoneal)

With this route, the fistula is accessed and excised through an intentional cystotomy on the preperitoneal side of the bladder as shown in Section 45-10. This approach is used for situations in which the fistula: (1) is located proximally in a narrow vagina, (2) lies close to the ureteral orifices, (3) is complicated by a concomitant ureteric fistula, (4) persists after prior repair attempts, (5) is large or complex in configuration, or (6) requires an abdominal interposition graft, described in the next section.

Evidence-based support for laparoscopic genitourinary fistula repair has been limited to case reports and expert opinion (Miklos, 2015; Nezhat, 1994). The technique requires advanced laparoscopic surgical skills. Accordingly, success with this approach appears to be highly dependent on surgeon experience.

Interpositional Flaps

Surrounding tissue vascularity is essential for successful genitourinary fistula healing after repair. When intervening tissues for fistula closure are thin and poorly vascularized, various tissue flaps may be placed vaginally or abdominally between the bladder and the vagina in an attempt to enhance the repair and to lend support and blood supply (Eisen, 1974; Martius, 1928). Sections 45-10 and 45-11 illustrate the omental J-flap, which is an abdominal option, whereas the Martius bulbocavernosus fat pad flap is used during vaginal procedures. Although interposition flaps are useful in situations where tissue viability is in question, their utility in uncomplicated cases of vesicovaginal fistula is unclear.

Other Genitourinary Fistulas

Although vesicovaginal fistulas are the most common type of genitourinary fistula, other fistulas can develop and may be described based on their communication between anatomic structures. Urethrovaginal fistulas can result from surgery involving the anterior vaginal wall, in particular anterior colporrhaphy and urethral diverticulectomy (Blaivas, 1989; Ganabathi, 1994a). In developing countries, as with vesicovaginal fistula, obstetric trauma remains the most common cause of urethrovaginal fistulas. Frequently, patients present with continuous urinary drainage into the vagina or with stress urinary incontinence. The principles of repair are similar, namely, layered closure, tension-free repair, and postoperative bladder drainage. Other types of genitourinary fistula can also develop (see Table 26-1).

Paraurethral glands are found along the anterior vaginal wall and communicate directly with the urethra. A urethral diverticulum commonly forms from a cystic enlargement of one of these glands. This isolated outpouching is commonly asymptomatic and is frequently diagnosed incidentally during routine examination. However, many present with symptoms and often require surgical excision.

Urethral diverticulum is reported to develop in 1 to 6 percent of the general female population. With greater awareness and radiologic advances, rates of diagnosis are increasing (Rovner, 2012a). However, the true incidence may be underestimated because diverticula are frequently asymptomatic and thus underreported. In women with lower urinary tract symptoms, the incidence dramatically increases and may reach 40 percent (Stewart, 1981).

Urethral diverticulum is diagnosed most often in the third to sixth decades of life and more commonly in females than in males (Aldridge, 1978). A 6:1 predominance of urethral diverticula in African-Americans compared with whites has been reported, although others have found no racial predisposition (Davis, 1970; Leach, 1987).

The etiology of urethral diverticula is unclear. Although most are thought to be acquired, rare congenital diverticula have been reported. Congenital causes include persistence of embryologic remnants, defective closure of the ventral portion of the urethra, and congenital cystic dilatation of paraurethral glands (Ratner, 1949). Embryology of the female genital tract, described in Chapter 18, contributes to our understanding of congenital urethral diverticulum. During female development, the müllerian ducts form the upper vagina, whereas the urogenital sinus gives rise to the distal vagina, vestibule, and female urethra (Fig. 18-4). In the vagina, müllerian mucinous columnar epithelium is replaced by squamous epithelium of the urogenital sinus. When the process of epithelial replacement is arrested, small foci of müllerian epithelium may persist and form cysts or diverticula.

More commonly, diverticula are acquired and can result from infection, birth trauma, or traumatic instrumentation. The most widely held theory regarding urethral diverticular development dates back to Routh (1890) and involves the paraurethral glands and their ducts. The paraurethral glands surround and cluster most densely along the urethra’s inferolateral border (Fig. 26-3). Of these glands, the bilateral Skene glands are the most distal and typically the largest. Paraurethral glands connect to the urethral canal via a network of branching ducts. The arborizing pattern in portions of this network helps to explain the complexity of some urethral diverticula (Vakili, 2003).

Routh theorized that infection and inflammation obstruct individual ducts and lead to cystic dilatation. If these do not spontaneously resolve or if infection is not treated promptly, an abscess can form. Subsequent abscess expansion and continued inflammation may rupture the gland into the urethral lumen to create a fistulous communication between the two (Fig. 26-4). As infection clears, the dilated diverticular sac and new communicating ostium into the urethra persist. Of infectious agents, Neisseria gonorrhoeae and Chlamydia trachomatis are organisms that were once commonly associated with urethritis and severe inflammation of the paraurethral glands. However, positive identification of these organisms is not mandatory for diagnosis, and more often than not, urethral cultures from women who have diverticula yield a polymicrobial result, which typically does not require treatment.

In addition to infection, injured urethral tissue can swell and obstruct paraurethral ducts. Logically, urethral trauma sustained during childbirth or during urethral instrumentation are suggested etiologies (McNally, 1935). Moreover, as described earlier, different social customs and obstetric practices in the developing world can potentially contribute to urethral trauma. In addition, female genital mutilation or repeated urethral dilatations can traumatize the urethra.

Early classification systems organized urethral diverticula according to their radiographic complexity and described them as: (1) simple saccular, (2) multiple, or (3) compound or complex with branching sinuses (Lang, 1959). To help standardize surgical treatment, Ginsburg and Genadry (1983) created a preoperative classification system based on urethral location. This system organized diverticula based on their urethral location and described lesions as type 1 (proximal third), type 2 (middle third), and type 3 (distal third).

In an attempt to fully incorporate all characteristics necessary to adequately assign treatment, Leach and coworkers (1993) constructed the L/N/S/C3 classification system. In this system, the diverticulum characteristics are described according to its location (L), number (N), size (S), and configuration, communication, and continence status of the patient (C3). Of these, location is described in relation to the urethra and defined as distal, mid-, or proximal urethral and as with or without extension beneath the bladder neck. In Leach’s series of 61 patients, investigators found that most diverticula were in the midurethra. Logically, this distribution reflects the predominance of paraurethral glands along the middle third of the urethra. Ascertainment of the number of diverticula is important to prevent incomplete excision of multiple lesions and thus symptom persistence. Diverticulum size similarly can influence treatment. For example, some recommend concomitant interpositional tissue flaps for large diverticula. Moreover, urinary incontinence may develop de novo or persist if the diverticulum is extremely large and involves sphincter continence mechanisms.

Of the three “C”s, diverticula configuration may be described as solitary or multiloculated and as simple, saddle-shaped, or circumferential (Fig. 26-5). Preoperative delineation assists in complete surgical excision and aids preparation for an interpositional flap in those cases requiring extensive urethral resection (Rovner, 2003).

Obviously, successful repair of the urethral wall defect depends in great part on identifying the diverticular opening into the urethral canal. Thus, the communication site is sought preoperatively and classified as proximal, mid-, or distal urethral. In the previously cited study, midurethral communication sites were the most common (60 percent), followed by proximal (25 percent) and distal (15 percent) sites.

Finally, in this classification system, the continence status and urethral hypermobility of the patient are documented. Almost half of the patients in Leach’s series had stress urinary incontinence, and these authors suggest that urethral hypermobility is an indication for concomitant incontinence surgery. Although several studies have documented the safety of performing concurrent bladder-neck suspension, some still consider this approach controversial due to concerns of urethral erosion (Bass, 1991; Ganabathi, 1994b; Leng, 1998; Swierzewski, 1993). Although consensus is lacking on this issue, repairing the diverticulum first and then considering antiincontinence surgery if urinary incontinence persists is reasonable. This staged fashion is a particularly realistic option because of the current array of effective minimally invasive surgical procedures, such as midurethral slings. For postoperative comparison, we typically perform baseline preoperative urodynamic testing, although this step may not be adopted by all. Importantly, these data and postoperative expectations are discussed with the patient during preoperative counseling.

Urethral diverticula are frequently asymptomatic and discovered incidentally during gynecologic or urologic examination. However, symptoms may vary and reflect diverticulum characteristics, especially size, location, and extension. Although postvoid dribbling, dysuria, and egress of discharge through the urethra with finger compression of a suburethral mass are pathognomonic, not all women present so classically (Fig. 26-6). For most patients, symptoms are nonspecific and include pain, dyspareunia, and several urinary symptoms. Romanzi and colleagues (2000) found pain in 48 percent of affected women. Pain most likely stems from cystic dilatation and also possibly from concurrent inflammation. Those with dyspareunia may note either entry or deep dyspareunia, depending on whether diverticula are distal or proximal, respectively.

A large diverticulum can sometimes be mistaken for early-stage pelvic organ prolapse, especially when the presenting complaint is vaginal fullness, bulge, or pressure. In these cases, the palpable diverticular vaginal mass may mimic a cystocele or rectocele. In most, careful systematic palpation of the vaginal wall will distinguish prolapse from a discrete vaginal wall cyst or diverticulum.

Various lower urinary tract symptoms are frequently associated with urethral diverticulum. Specifically, urinary incontinence is noted by 35 to 60 percent of affected women (Gabanathi, 1994b; Romanzi, 2000). In addition, during micturition, urine may enter the diverticular sac, only to later spill from the sac and present as postvoid dribble or as urinary incontinence. In contrast to urine loss, urinary retention has also been reported (Nitti, 1999). Retention frequently accompanies periurethral or diverticular sac cancers, discussed later in this section. Urinary tract infection often complicates urethral diverticulum. In their review of 60 women with diverticula, Pathi and associates (2013) found that recurrent urinary tract infection was highly specific for urethral diverticula.

Less frequently, stones may form from urine stagnation and salt precipitation within the diverticular sac. As such, stones may be singular or multiple and are usually composed of calcium oxalate or calcium phosphate. The reported frequency approximates 10 percent (Perlmutter, 1993).

Malignant transformation within a urethral diverticulum is rare and accounts for only 5 percent of urethral cancers. Most of these tumors are adenocarcinomas, although transitional cell and squamous cell carcinomas have also been identified (Clayton, 1992). Tumors typically present in the sixth or seventh decade of life, and hematuria, irritative voiding complaints, and urinary retention are common. Thus, palpation of an indurated or fixed periurethral mass, coupled with urinary obstructive symptoms, typically prompts further diagnostic evaluation and tissue biopsy (von Pechmann, 2003). Given the rarity of cancers within these diverticula, codified treatment strategies are lacking. Currently, these malignancies are treated by anterior exenteration or by diverticulectomy, alone or with adjuvant radiation therapy (Shalev, 2002).

For many women, urethral diverticula may be diagnosed using simply a detailed history, physical examination, and high index of suspicion. Patient evaluation focuses on the common characteristics and symptoms noted earlier. However, despite available clinical tools, the diagnosis for many women is delayed as they may initially be treated for stress or urgency incontinence, chronic cystitis, trigonitis, urethral syndrome, vulvovestibulitis, pelvic organ prolapse, and idiopathic chronic pelvic pain. Moreover, the diverticulum itself may mimic a Gartner duct cyst, müllerian remnant vaginal cyst, vaginal epidermoid inclusion cyst, ectopic ureterocele, or endometrioma.

Physical Examination

The most frequent physical finding, an anterior vaginal mass beneath the urethra, is detected in 50 to 90 percent of symptomatic patients (Ganabathi, 1994b; Gerrard, 2003; Romanzi, 2000). Although urethral expression of purulent material during compression of the mass with a finger is common, failure to demonstrate transurethral expression of discharge does not exclude the diagnosis. Stenosis of the diverticular duct may obstruct sac emptying in these cases. Thus, meticulous examination and palpation is performed along the entire length of the urethra. Once diverticula are identified, their number, size, consistency, and configuration are determined.

However, physical examination alone is typically insufficient to completely characterize a mass. Of available ancillary tests, each has its advantages and disadvantages, and investigators may differ as to which is chosen primarily. Accordingly, familiarity with each modality’s strengths aids selection of the most appropriate test for a given clinical setting.

Cystourethroscopy

Of the diagnostic procedures used to detect urethral diverticula, cystourethroscopy is the only tool that allows direct inspection of the urethra and bladder. During endoscopy, fingers pressed upward against the proximal anterior vaginal wall help occlude the bladder neck and allow the distending medium to create positive pressure and open diverticular ostia (Fig. 26-7). A zero-degree cystoscope lens allows complete radial assessment of the urethra to aid identification of diverticular ostia and at times, purulent discharge extruding from them.

A primary advantage to cystourethroscopy includes its accuracy for diverticulum detection (Summitt, 1992). Moreover, in those with nonspecific lower urinary tract symptoms, other causes such as urethritis, cystitis, stones, or stenosis can be excluded. Despite these advantages and its common use by urogynecologists, gynecologic generalists employ cystourethroscopy less frequently due to obstacles such as inexperience evaluating bladder and urethral mucosal anatomy, cystoscopic expertise, instrumentation costs, and credentialing challenges. Even for clinicians who are experienced with this tool, it may still fail to display all diverticula. For example, a poor seal between the cystoscope and distal urethral mucosa may lead to inadequate sac distention and failure to identify distally located diverticula. Also, diverticula whose ostia are stenotic, and thus do not communicate with the urethral lumen, can be missed. Although this endoscopy is minimally invasive, patient pain and risk of postprocedural infection are additional legitimate concerns. Last, important information regarding diverticular size and configuration may not be obtained with this tool.

Magnetic Resonance Imaging

This imaging modality has slowly become the diagnostic standard to characterize periurethral masses because of its superior resolution at soft tissue interfaces. Specifically, for detecting urethral diverticula, magnetic resonance (MR) imaging has comparable or superior sensitivity compared with other radiologic techniques (Lorenzo, 2003; Neitlich, 1998). To improve image resolution with MR imaging, an imaging coil may be placed into the rectum or vagina. The coil, which is housed inside a probe, improves the image quality of structures surrounding the rectum or vagina. Alternatively, an external plate or coil for image resolution enhancement can be used to minimize patient discomfort. Many institutions have opted for this method largely due to patient comfort, without significant loss of diagnostic accuracy. Despite MR imaging’s advantages, procedure costs are balanced against the need for additional anatomic information. For a solitary diverticulum with clearly demarcated boundaries and no evidence of extension, costly and extensive imaging may not be necessary.

Other Imaging Tools

Although we use the two previously mentioned techniques most commonly, other techniques may be performed at other institutions. Of these, voiding cystourethrogram (VCUG) is used by some as an initial tool in urethral diverticulum evaluation. Radiographic contrast instilled into the bladder fills a diverticular sac during voiding, and postvoid radiographs then highlight sac volume. This painless, simple test has an overall reported accuracy that approximates 85 percent. However, some prefer positive-pressure urethrography as a primary diagnostic tool, since VCUG requires a diverticular communication to be patent during testing (Blander, 2001). Additionally, VCUG involves ionizing radiation exposure for the patient, although this is minimal.

Positive pressure urethrography (PPUG) also uses radiographic contrast and x-ray. With PPUG, a double-balloon, triple-lumen catheter (Trattner catheter) is inserted through the urethra, and its tip enters the bladder (Fig. 26-8). Inflating the proximal balloon allows it to be pulled snug against and occlude the urethra at the urethrovesical junction. A distal balloon obstructs the distal urethra. A single catheter port between the two balloons allows instillation of radiopaque contrast, subsequent urethral distention, and expansion of the diverticulum under positive pressure. For accurately identifying diverticula, PPUG has a sensitivity surpassing that of VCUG (Golomb, 2003). Disadvantageously, PPUG can be time-consuming, technically difficult, and associated with patient discomfort and postprocedural infection risk. Additionally, similar to VCUG, diverticula may be missed if thick pus or debris prevents adequate filling with contrast medium or if the diverticular ostium is stenotic. Accordingly, although PPUG has been a primary tool for many for diagnosing urethral diverticula, fewer radiologists are proficient with the technique, and it has gradually been replaced by other modalities.

Sonography is a relatively new tool for urethral diverticulum assessment and appears to have some efficacy (Gerrard, 2003). Suggested technical advantages include visualization of diverticula that do not fill during radiographic contrast studies and characterization of diverticular wall thickness, size, and internal architecture (Yang, 2005). Transabdominal, -vaginal, -rectal, -perineal and -urethral sonographic techniques have all been described (Keefe, 1991; Vargas-Serrano, 1997). Although the endourethral technique has been reported to have excellent specificity, it is expensive and may be more invasive that the other sonographic routes (Chancellor, 1995). Advantages of sonography include patient comfort, avoidance of ionizing radiation and contrast exposure, relative low cost, and reduced invasiveness. However, in addition to its learning curve, sonography’s role in urethral diverticulum diagnosis has not been clearly established. Currently, it remains an academic or adjunctive technique.

Because there is still no consensus on which modality is best used primarily, it is reasonable to begin with cystourethroscopic evaluation followed by VCUG. At our institution, if the initial evaluation is unrevealing but the diagnostic suspicion remains high or if the lesion appears more complex, then MR imaging combined with an endorectal coil or external plate may improve the resolution of images and add to the information gained.

Acute Treatment

Women with a urethral diverticulum may present acutely with pain, urinary symptoms, or focal tenderness during examination. Conservative management is recommended in the acute phase and includes sitz baths, administration of a broad-spectrum oral antibiotic such as a cephalosporin or fluoroquinolone, and oral analgesics.

Chronic Diverticula

For a chronic diverticulum, a conservative approach may be elected by women who have few or no symptoms and decline surgery due to its associated risks of urethrovaginal fistula and sphincter defect incontinence. However, in women electing observation, long-term data are lacking regarding rates of subsequent symptom development, diverticulum enlargement, and eventual need for surgical excision.

Many practitioners may deliberate as to whether an enlarged inflamed cystic connection with the urethra is termed a “Skene gland cyst” or a “urethral diverticulum.” Pragmatically, for those with persistent bothersome symptoms the treatment is the same with surgical intervention often indicated. Procedures include diverticulectomy, transvaginal partial ablation, and marsupialization, which are all described in Section 45-9 of the atlas.

Of these, diverticulectomy is the most frequently chosen to treat diverticula at any site along the urethra. Excision of the entire diverticulum provides long-term correction of the urethral defect, normal urine stream, and high rates of postoperative continence. However, disadvantages include risks for postsurgical urethral stenosis, urethrovaginal fistula, potential injury to the urinary sphincter continence mechanism with subsequent incontinence, and the possibility of recurrence. For patients who also have stress urinary incontinence, some practitioners recommend concomitant placement of either a midurethral or pubovaginal sling. As noted earlier, although this practice is supported by some studies, our preference is to approach it as a staged procedure. We perform the defect repair first and later reassess the need for an antiincontinence procedure.

Another surgery, partial diverticular sac ablation, may be preferred for proximal diverticula to avoid bladder entry or bladder neck injury risks. Tancer and associates (1983) described this procedure, during which excess diverticular wall is excised, the diverticular neck is not removed, but remaining diverticular wall tissue is reapproximated to close the defect.

Last and less frequently, diverticulum marsupialization, also known as the Spence procedure, has been used for distal diverticula (Spence, 1970). The procedure is a meatotomy that when healed forms a new wider urethral meatus. Although simple to perform, this procedure alters the shape and function of the urethral meatus, with patients often noting spraying of their urine stream.

Other procedures described in case reports include urethroscopic transurethral electrosurgical fulguration of the diverticular sac and transurethral incision to widen the diverticular ostia (Miskowiak, 1989; Saito, 2000; Vergunst, 1996). Data are lacking, however, regarding long-term efficacy and complication rates with these techniques.