4.7.

5 Summary of evidence and recommendations for surgery in women with both pelvic organ
prolapse and stress urinary incontinence

Summary of evidence LE
Women with POP and UI
Surgery for POP and SUI shows a higher rate of cure of UI in the short-term than POP surgery alone. 1a
There is conflicting evidence on the relative long-term benefit of surgery for POP and SUI vs. POP 1a
surgery alone.
Combined surgery for POP + SUI carries a higher risk of adverse events than POP surgery alone. 1a
Continent women with POP
Continent women with POP are at risk of developing SUI postoperatively. 1a
The addition of a prophylactic anti-UI procedure reduces the risk of postoperative UI but increases the 1a
risk of adverse events.
Women with POP and OAB
There is some low-level inconsistent evidence to suggest that surgical repair of POP can improve 2b
symptoms of OAB.

Recommendations for women requiring surgery for bothersome pelvic organ prolapse Strength rating
(POP) who have symptomatic or occult stress urinary incontinence (SUI)
Offer simultaneous surgery for POP and SUI only after a full discussion of the potential risks Strong
and benefits of combined surgery vs. POP surgery alone.
Inform women of the increased risk of adverse events with combined prolapse and anti- Strong
urinary incontinence surgery compared to prolapse surgery alone.
Recommendations for women requiring surgery for bothersome POP who do not have
symptomatic or occult SUI
Inform women that there is a risk of developing de novo SUI after prolapse surgery. Strong
Warn women that the benefit of combined surgery for POP and SUI may be outweighed by Strong
the increased risk of adverse events compared to prolapse surgery alone.

4.8 Urinary fistula

The evidence relating to diagnosis and treatment of urinary fistulae is generally low level and largely composed
of case series and other consensus statements. In particular, the epidemiology, aetiology, diagnosis, treatment
and prevention of obstetric and non-obstetric fistulae have been described in detail during the 2016 ICI
conference [702]. Most non-obstetric fistulae are iatrogenic in origin, with most caused by pelvic surgery
(e.g., hysterectomy for benign or malignant conditions, bowel resection, and urological surgery). The risks
during pelvic surgery increase relative to the complexity of the resection, the extent of primary disease, and
prior radiotherapy (especially for recurrent disease). When a fistula occurs following radiotherapy for primary
treatment, this may be an indication of tumour recurrence.

4.8.1 Epidemiology, aetiology and pathophysiology

4.8.1.1 Obstetric fistula
According to the WHO, fistulae affect > 2 million women, mostly from sub-Saharan African and Asian countries.
The pooled prevalence of fistula from population studies is 0.29/1000 pregnancies [703]. Poor quality obstetric
care, staff unaccountability, late referral, and poor nursing standards have been identified as health system
causes [703]. However, obstructed labour is poorly documented. The main individual risk factors include age at
first marriage, short stature, pregnancy with a male child, failure to attend antenatal care, low socio-economic
status, low social class, lack of employment, and illiteracy [704-706]. Obstetric fistulae have detrimental
consequences on global and individual health and are associated with malnutrition, sexual dysfunction, anxiety,
depression, insomnia, social isolation, worsening poverty, and suicide [707, 708].

4.8.1.2 Iatrogenic fistula

Poor obstetric care is usually responsible for VVF in the developing world. By contrast, in the developed world,
gynaecological or pelvic surgery is the main cause of VVF.

4.8.1.2.1 Post-gynaecological surgery

An injury to the urinary tract during hysterectomy for benign conditions (60–75%), hysterectomy for malignant
conditions (30%) and caesarean section (6%) are the main causes of postoperative VVF in the developed world
[709, 710]. The risk of pelvic organ fistula following hysterectomy ranges from 0.1 to 4% [711].

MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022 95
Fistulae may also occur as a result of primary or recurrent malignancy, or as a consequence of cancer
treatment by surgery, radiotherapy, and/or chemotherapy.

In a study including 536 women undergoing radical hysterectomy for invasive cervical cancer, bladder injury
occurred in 1.5% with VVFs forming in 2.6% and uretero-vaginal fistulae (UVFs) in 2.4% of cases [712]. Overall,
the rate of urogenital fistula appears to be ~9 times higher following radical hysterectomy for malignant disease
as compared to that following simple hysterectomy (abdominal or vaginal for benign conditions) [713]. Bladder-
sparing techniques during pelvic exenteration can increase the risk of fistula formation [714].

4.8.1.2.2 Radiation fistula

The risk of fistula seems to be higher for postoperative external radiation (1.9%) compared to intravaginal
brachytherapy (0.8%) [715], without any predictive factor being identified [716]. This is most likely due to
the heterogeneity of data regarding the tumour type and stage, the form of radiation, and the site and dose
delivered.

4.8.1.2.3 Rare causes of vesico-vaginal fistula

Foreign bodies such as pessaries, sex toys, cups etc. can be a cause of delayed presentation of VVF [717-
719]. Ketamine abuse has also been shown to be responsible for fistula formation [720].

4.8.1.3 Summary of evidence for epidemiology, aetiology and pathophysiology of urinary fistula

Summary of evidence LE
The risk of injury to the urinary tract and subsequent fistula formation is higher in women with 2
malignant disease undergoing radical surgery than in women with benign disease undergoing simple
surgical procedures.
The rate of fistula formation following radiotherapy for gynaecological cancer appears to be of the 4
same order as that following surgical treatment.

4.8.2 Classification
Due to the plethora of VVF classification systems, a consensual classification system needs to be adopted. The
Waaldijk and Goh classifications are widely used for diagnosis and follow-up [721-723]. They were originally
designed for obstetric fistulae and their use in iatrogenic fistulae is less relevant [724]. Waaldijk’s classification
is based on the size and site of the fistulae and divides them into three main categories: type 1 are VVFs
with no urethral involvement; type 2 are those that involve the urethra (and are sub-classified into those with
circumferential and non-circumferential urethral involvement); and type 3 are fistulae involving other parts of the
urinary tract. Goh’s classification also uses the presence or absence of urethral involvement to sub-categorise
VVFs and takes into account the degrees of fibrosis present. The WHO classification (Table 6) was originally
developed for obstetric fistulae and separates fistulae into simple and complex.

Table 6: Adapted WHO Classification of fistulae [703]*

Simple fistula with good prognosis Complex fistula with uncertain prognosis
• Single fistula < 4 cm • Fistula > 4 cm
• Vesico-vaginal fistula • Multiple fistula
• Closing mechanism not involved • Recto-vaginal mixed fistula, cervical fistula
• No circumferential defect • Closing mechanism involved
• Minimal tissue loss • Scarring
• Ureters not involved • Circumferential defect
• First attempt to repair • Extensive tissue loss
• Intravaginal ureters
• Failed previous repair
• Radiation fistula
*Although this classification was developed for obstetric fistula initially, it could be relevant for iatrogenic fistula
as well.

96 MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022
4.8.2.1 Recommendation for the classification of urinary fistula

Recommendation Strength rating

Use a classification system for urinary tract fistulae to try to standardise terminology in this Strong
subject area.

4.8.3 Diagnostic evaluation

Leakage of urine is the hallmark sign of a urogenital fistula. The leakage is usually painless, may be intermittent
if it is position dependent, but more usually is constant. Unfortunately, intraoperative diagnosis of a genito-
urinary or gastrointestinal injury is made in only about half of cases [725]. Diagnosis of VVF usually requires
clinical assessment often in combination with appropriate imaging or laboratory studies. Direct visual
inspection, cystoscopy, retrograde bladder filling with a coloured fluid, or placement of a tampon into the
vagina to identify staining may facilitate diagnosis of VVF. A double-dye test to differentiate between UVF and
VVF may be useful in some cases [710]. Testing the creatinine level in either the extravasated or collected fluid
will confirm fluid leakage as urine. Contrast-enhanced CT with late excretory phase reliably diagnoses urinary
fistulae and provides information about ureteric integrity and the possible presence of associated urinoma.
Magnetic resonance imaging, in particular with T2 weighting, also provides diagnostic information regarding
fistulae [726].

4.8.4 Disease management of fistula

4.8.4.1 Management of vesico-vaginal fistula
4.8.4.1.1 Conservative management
4.8.4.1.1.1 Spontaneous closure
The reported spontaneous closure rate is 13 ± 23% [727], although this applies largely to small fistulae (< 1 cm)
[702, 728]. Hence, immediate management is usually by urinary catheterisation or diversion; however, within
the first two weeks following fistula occurrence, surgical exploration and repair can be considered.

4.8.4.1.1.2 Pharmacotherapy
Several case reports describe a successful fistula closure rate following the induction of amenorrhoea
by oestrogen, oestrogen/progesterone combinations or luteinising hormone releasing hormone analogues
specifically for small (< 7 mm), uretero- or vesico-uterine fistulae following caesarean section [729-735]. One
RCT comparing the efficacy of using fibrin glue compared to Martius flap inter-positioning (n = 14; < 4 cm and
n = 5; > 5 cm) did not report significantly different outcomes between the two types of treatment [736].

4.8.4.1.1.3 Palliation and skin care

During the intervening period between diagnosis and repair, UI pads with the aim of prevention of skin
complications related to chronic urinary leakage can be provided and the use of a barrier cream or local
oestrogen can also be considered [737, 738].

4.8.4.1.1.4 Nutrition
Nutritional support is essential in patients with fistulae induced by malignant disease or radiotherapy [739], or
following diversion surgery [739-741].

4.8.4.1.1.5 Physiotherapy
Early involvement of the physiotherapist in preoperative management and rehabilitation of fistula patients
suffering from limb weakness, foot drop and limb contracture is essential [742, 743].

4.8.4.1.1.6 Antimicrobial therapy

Active infection in the genital or urinary tracts should be treated prior to surgical repair [744].

4.8.4.1.1.7 Counselling
Confident and realistic counselling by the surgeon is essential and the involvement of nursing staff or
counsellors with experience of fistula patients is also desirable.

4.8.4.1.2 Surgical management

4.8.4.1.2.1 Timing of surgery
Findings from uncontrolled case series suggest no difference in success rates for early (within three weeks) or
delayed (after three months) closure of VVF.

MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022 97
4.8.4.1.2.2 Surgical approaches
Vaginal procedures
There are two main types of closure techniques applied to the repair of urinary fistulae, the classical
saucerisation/partial colpocleisis [727] and the more commonly used dissection and repair in layers or flap-
splitting technique [745]. There are no data comparing their outcomes.

Abdominal procedures
Repair by the abdominal route is indicated when high fistulae are fixed at the vaginal vault and are inaccessible
via a vaginal approach. A transvesical repair has the advantage of being entirely extraperitoneal. A simple
transperitoneal repair is used less often although it is favoured by some using the laparoscopic approach. A
combined transperitoneal and transvesical procedure may be utilised for fistula repair following caesarean
section. There are no RCTs comparing abdominal and vaginal approaches. Results of secondary and
subsequent repairs are not as successful as the initial repair [746].

A single RCT compared trimming of the fistula edge with no trimming. There was no difference in success rates
but failed repairs in trimmed cases had larger recurrences than untrimmed cases, which were smaller [747].

Laparoscopic and robotic procedures

Small series (single figures) have reported using these techniques, but while laparoscopic repair is feasible with
and without robotic assistance, it is not possible to compare outcomes with alternative surgical approaches.

Tissue interposition
Tissue flaps are often added as an additional layer of repair during VVF surgery. Most commonly, such flaps
are utilised in the setting of recurrence after a prior attempt at repair, for VVF related to previous radiotherapy
(described later), ischaemic or obstetric fistulae, large fistulae, and finally those associated with a difficult or
tenuous closure due to poor tissue quality. However, there is no high-level evidence that the use of such flaps
improves outcomes for either complicated or uncomplicated VVF.

Postoperative management
There is no high-level evidence to support any particular practice in postoperative management but most
reported series used catheter drainage for > 10 days and longer periods in complex or radiation-associated
fistulae (up to three weeks). The performance of postoperative cystography prior to catheter removal can miss
a persistent fistula if not done with a micturition phase or if the fistula is located at the bladder neck.

4.8.4.1.3 Management of complications of vesico-vaginal fistulae

The complications of VVF repair are varied and can include:
• Persistence or recurrence of fistula;
• Persistence or recurrence of UI;
• Persistence of LUTS or occurrence of new LUTS, including de novo overactive bladder symptoms and/or
SUI;
• Infections: wound and UTIs/urosepsis;
• Ureteric obstruction (ligation, fibrosis or injury);
• Bladder outlet obstruction (meatal stenosis, urethral stricture or bladder neck obstruction);
• Bladder contracture;
• Vaginal stenosis;
• Sexual dysfunction (vaginismus/dyspareunia);
• Rare complications (granulomas/diverticulum formation);
• Neurological complications (foot drop/neurogenic bladder);
• Psychological trauma (social isolation/divorce/mental illness);
• Infertility.

The literature on the treatment and management of complications of fistula repairs is scarce and is mostly
experience-based. It is impossible to provide any specific evidence-based guidance.

4.8.4.2 Management of radiation fistulae

Modified surgical techniques are often required, and indeed, where the same techniques have been applied to
both surgical and post-radiation fistulae, the results from the latter have been consistently poorer [748]. Due to
the wide field abnormality surrounding many radiotherapy-associated fistulae, approaches include, permanent
urinary and/or faecal diversion [748, 749] or preliminary urinary and faecal diversion, with later undiversion
in selected cases following reconstruction. In cases where life expectancy is deemed to be short, ureteric
occlusion might be more appropriate.

98 MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022
4.8.4.3 Management of ureteric fistulae
4.8.4.3.1 General principles
Patients at higher risk of ureteric injury require experienced surgeons who can identify and protect the ureter
and its blood supply to prevent injury and recognise injury promptly. Immediate repair of any intraoperative
injury should be performed by observing the principles of debridement, adequate blood supply and tension-
free anastomosis with internal drainage using stents [750]. Delayed presentation of UUT injury should be
suspected in patients whose recovery after relevant abdominal or pelvic surgery is slower than expected, if
there is any fluid leak, and if there is any unexpected dilatation of the pelvicalyceal system.

While there is no evidence to support the use of one surgical approach over another, there is consensus that
repair should adhere to the standard principles of tissue repair and safe anastomosis, and be undertaken by
an experienced team. Conservative management is possible with internal or external drainage, endoluminal
management using nephrostomy and stenting where available, and early (< 2 weeks) or delayed (> 3 months)
surgical repair when required [751]. Functional and anatomical imaging should be used to follow-up patients
after repair to guard against development of ureteric stricture and deterioration in renal function.

4.8.4.3.2 Uretero-vaginal fistulae

Uretero-vaginal fistula occurring in the early postoperative phase predominantly after hysterectomy is the
most frequent presentation of UUT fistulae in urological practice. An RCT in 3141 women undergoing open or
laparoscopic gynaecological surgery found that prophylactic insertion of ureteric stents made no difference to
the low risk (1%) of ureteric injury [752].

Endoscopic management is sometimes possible by retrograde stenting, percutaneous nephrostomy and

antegrade stenting if there is pelvicalyceal dilatation, or ureteroscopic realignment [753]. However, the long-
term success rate is unknown. If endoluminal techniques fail or result in secondary stricture, the abdominal
approach to repair is standard and may require end-to-end anastomosis, reimplantation into the bladder using
psoas hitch or Boari flap, or replacement with bowel segments with or without reconfiguration. As a last resort,
nephrectomy may be considered, particularly in the context of a poorly functioning kidney and an otherwise
normal contralateral kidney [754-758].

4.8.4.3.3 Management of urethro-vaginal fistulae

4.8.4.3.3.1 Aetiology
Although urethro-vaginal fistulae are rare, most of them in adults have an iatrogenic aetiology. Causes include
surgical treatment of SUI with bulking agents or synthetic slings, surgery for urethral diverticulum and genital
reconstruction. Irradiation and even conservative treatment of prolapse with pessaries can lead to formation of
fistulae.

4.8.4.3.3.2 Diagnostic evaluation

Clinical vaginal examination, including the three-swab test, is often sufficient to diagnose UVF. Urethroscopy
and cystoscopy can be performed to assess the extent and location of the fistulae. In cases of difficult
diagnosis, VCUG or US can be useful. An 3D-MRI or CT scan is becoming utilised more widely to clarify
anatomy [759, 760].

4.8.4.3.3.3 Surgical management

Choice of surgery will depend on the size, localisation, and aetiology of the fistula and the amount of tissue
loss. Principles of reconstruction include identifying the fistula, creation of a plane between the vaginal wall and
urethra, watertight closure of the urethral wall, eventual interposition of tissue, and closure of the vaginal wall.

One case series reported that a vaginal approach yielded a success rate of 70% at first attempt and 92%
at second attempt, and that an abdominal approach only led to successful closure in 58% of cases [761]. A
vaginal approach required less operating time, had less blood loss and shorter hospitalisation.

Most authors have described surgical principles that are identical to those of VVF repair, and primary closure
rates of 53–95.4% have been described. A series of 71 women, treated for UVF reported that 90.1% of
fistulae were closed at the first vaginal intervention. Additionally, 7.4% were closed during a second vaginal
intervention. Despite successful closure, SUI developed in 52% of cases. Stress urinary incontinence patients
were treated with synthetic or autologous slings and nearly 60% became dry and an additional 32% improved.
Urethral obstruction occurred in 5.6% and was managed by urethral dilation or urethrotomy [762].

MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022 99
4.8.4.3.3.4 Flaps and neo-urethra
The simplest flap is a vaginal advancement flap to cover the urethral suture line. Labial tissue can be harvested
as a pedicled skin flap. This labial skin can be used as a patch to cover the urethral defect, but can also be
used to create a tubular neo-urethra [763, 764]. The construction of a neo-urethra has mostly been described
in traumatic aetiologies. In some cases, a transpubic approach has been used [765]. The numbers of patients
reported are small and there are no data on the long-term outcome of fistula closure and continence rates.
The underlying bulbo-cavernosus tissue can be incorporated in the pedicled flap and probably offers better
vascularisation and more bulking to the repair. This could allow a safer placement of a sling afterwards, in those
cases where bothersome SUI would occur postoperatively [766, 767].

4.8.4.3.3.5 Martius flap

In obstetrical fistula repair, the Martius labial bulbocavernosus muscle/fat flap was not found to have any
benefit. However, the Martius flap is still considered by some to be an important adjunctive measure in the
treatment of genitourinary fistulae for which additional bulking with well-vascularised tissue is needed [768].
The series of non-obstetrical aetiology are small and all of them are retrospective. There are no prospective
data, nor randomised studies [769]. The indications for Martius flap in the repair of UVF remain unclear.

4.8.4.3.3.6 Rectus muscle flap

Rectus abdominis muscle flaps have been described by some authors [770, 771].

4.8.4.3.3.7 Alternative approaches

An alternative retropubic retro-urethral technique has been described by Koriatim [772]. This approach allows a
urethro-vesical flap tube to be fashioned to form a continent neo-urethra.

4.8.4.4 Summary of evidence and recommendations for the management of urinary fistula

Summary of evidence LE
Spontaneous closure of surgical fistulae does occur, and appears more likely for small fistulae 3
although it is not possible to establish the rate with any certainty.
There is no evidence that the timing of repair makes a difference to the chances of successful closure 3
of a fistula.
There is no high-quality evidence of differing success rates for repair of VVFs by vaginal, abdominal, 3
transvesical, and transperitoneal approaches.
A period of continuous bladder drainage may be crucial to successful fistula repair but there is no 3
high-level evidence to support one regimen over another.
A variety of interpositional grafts can be used in either abdominal or vaginal procedures, although 3
there is little evidence to support their use in any specific setting.
Post-radiation fistula
Successful repair of irradiated fistulae may require prior urinary diversion and the use of non-irradiated 3
tissues to effect repair.
Ureteric fistula
Prophylactic ureteric stent insertion does not reduce risk of ureteric injury during gynaecological 2
surgery.
Antegrade endoluminal distal ureteric occlusion combined with nephrostomy tube diversion often 4
palliates urinary leakage due to malignant fistula in the terminal phase.
Urethro-vaginal fistula
Urethro-vaginal fistula repair may be complicated by SUI, urethral stricture and urethral shortening, 3
which may necessitate long-term follow-up.

Recommendations Strength rating

General
When reporting on outcomes after fistula repair, authors should make a clear distinction Strong
between fistula closure rates and postoperative urinary incontinence rates and the time at
which the follow-up was organised.
Do not routinely use ureteric stents as prophylaxis against injury during routine Strong
gynaecological surgery.

100 MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022
Suspect ureteric injury or fistula in patients following pelvic surgery if a fluid leak or Strong
pelvicalyceal dilatation occurs postoperatively, or if drainage fluid contains high levels of
creatinine.
Use three-dimensional imaging techniques to diagnose and localise urinary fistulae, Weak
particularly in cases with negative direct visual inspection or cystoscopy.
Manage upper urinary tract fistulae initially by conservative or endoluminal techniques Weak
where such expertise and facilities exists.
Surgical principles
Surgeons involved in fistula surgery should have appropriate training, skills and experience Weak
to select an appropriate procedure for each patient.
Attention should be given as appropriate to skin care, nutrition, rehabilitation, counselling Weak
and support prior to, and following, fistula repair.
Tailor the timing of fistula repair to the individual patient and surgeon requirements once any Weak
oedema, inflammation, tissue necrosis, or infection, are resolved.
Ensure that the bladder is continuously drained following fistula repair until healing is Weak
confirmed (expert opinion suggests: ten to fourteen days for simple and/or post-surgical
fistulae; fourteen to 21 days for complex and/or post-radiation fistulae).
Where urinary and/or faecal diversions are required, avoid using irradiated tissue for repair. Weak
Use interposition graft when repair of radiation-associated fistulae is undertaken. Weak
Repair persistent uretero-vaginal fistulae by an abdominal approach using open, Weak
laparoscopic or robotic techniques according to availability and competence.
Urethro-vaginal fistulae should preferably be repaired by a vaginal approach. Weak

4.9 Urethral diverticulum

A female urethral diverticulum is a sac-like protrusion composed of the entire urethral wall or only the urethral
mucosa, situated between the periurethral tissues and the anterior vaginal wall.

4.9.1 Epidemiology, aetiology, pathophysiology

Urethral diverticulum is an uncommon condition with an estimated prevalence of 1–6%. A prevalence of up
to 10% was reported among women with LUTS attending a tertiary referral centre [773]. However, as many
patients are asymptomatic or misdiagnosed, the true incidence is unknown [774-776]. Given the rarity of the
condition, most published series are small and single institutional. Urethral diverticulum is thought to arise from
repeated obstruction, infection, and subsequent rupture of periurethral glands into the urethral lumen, resulting
in an epithelialised cavity that communicates with the urethra [774].

Iatrogenic damage to the urethra may also play a role, as up to 20% of women with urethral diverticula
are noted to have a history of urethral surgery, dilation, or traumatic delivery [774, 777]. Iatrogenic urethral
diverticula formation associated with synthetic suburethral sling has also been reported [778-780].

4.9.2 Classification

Table 7: Classification system for female urethral diverticula based on characteristics*

Localisation Mid-urethral
Distal
Proximal
Full length
Configuration Single
Multi-loculated
Saddle shaped
Communication Mid-urethral
No communication visualised
Distal
Proximal
Continence Stress urinary incontinence
Continent
Post-void dribble
Mixed incontinence
*Limited LNS C3 classification of urethral diverticula [777, 781, 782].

MANAGEMENT OF NON-NEUROGENIC FEMALE LOWER URINARY TRACT SYMPTOMS (LUTS) - limited update March 2022 101