Delayed presentation of postcaesarean section vesicouterine fistula repaired robotically
- Edward Ramez Latif ,
- Mohammed Kamil Quaraishi ,
- Davon Mitchell and
- Edward Streeter
- Department of Urology, East Kent Hospitals University NHS Foundation Trust, Canterbury, UK
- Correspondence to Dr Edward Ramez Latif; elatif@wollemia.com.au
Abstract
A premenopausal patient in her late forties presented with a 15-year history of urinary incontinence starting shortly after a caesarean section performed for her fourth delivery and more recently associated episodic light haematuria and passage of clots per vagina. The haematuria was intermittent over several months and associated with per-vaginal bleeding. She had symptomatic anaemia. Flexible cystoscopy and MRI showed a vesicouterine fistula. She underwent a robotic-assisted hysterectomy and multi-layered repair of the bladder with omental interposition. She remained in the hospital for 4 days, though recovered well and underwent catheter removal at 14 days following a normal cystogram. At 3 months, she was well with no incontinence or urinary symptoms. This case demonstrates the need for vigilance in assessing patients with persistent incontinence, particularly in the context of prior caesarean section. It highlights the feasibility of robotic surgery for correcting these defects in a society where minimally invasive surgery is becoming the norm, and cosmesis is a priority for patients.
Background
The global incidence of the urogenital fistula is hard to accurately calculate. The WHO estimates 50 000–1 00 000 women are diagnosed with urogenital fistulae annually.1 Fistulae can occur between any part of the urological and genital tracts and include vesicovaginal, urethrovaginal, ureterovaginal, vesicocervical and vesicouterine fistulae. A 2013 meta-analysis identified a pooled prevalence of 0.29 per 1000 women of reproductive age suffered a urogenital fistula.2 In well-resourced countries, 83.2% of fistulae result from surgery, whereas in low-resourced countries 95.2% occur following childbirth.3
While obstetric fistulae are relatively common, vesicouterine fistula is a rare disease affecting women postcaesarean section with the cases reported largely from developing countries.4 5 As a consequence it does not figure highly in the differential diagnosis for vaginal bleeding, urinary incontinence or haematuria. While it is a rare problem, it is a significant one affecting young women and requiring complex treatment. In 1957 Youssef first described his syndrome, which presents with the classic triad of cyclic haematuria, amenorrhoea and preserved urinary continence, and is characteristic of vesicouterine fistula,6 7 though some sufferers have urinary incontinence.8 9 Józwik and Józwik have developed a classification system of three types which accounts for variations in presentation.10
Traditionally such fistulas are managed with excision of the fistulous tract and closure of the defects with or without hysterectomy depending on the feasibility of uterine preservation and the patients’ desire for further pregnancy. In the UK 120 patients annually undergo surgery for lower tract fistulae of which vesicouterine fistulae represent a small fraction, the majority being vesicovaginal and urethrovaginal. Most are treated in units undertaking an average of 1 repair every 10 years.11
Case presentation
Patient demographics
The patient is a premenopausal woman in her late-40s who migrated from rural Pakistan to the UK in the past decade. She has had four children, all by caesarean section.
Medical history
Aside from her documented gynaecological history the patient has a longstanding history of anaemia.
Clinical presentation and assessment
She had suffered from persistent urinary incontinence for 15 years, starting shortly after her last caesarean section. This was previously diagnosed as stress urinary incontinence. This had a profound impact on her quality of life, resulting in a reluctance to engage with her community and embarrassment. In the past several months she was passing clots per vagina, and both the urinary incontinence and normally voided urine had been intermittently haematuric with the normal void very lightly bloodstained. The dominant complaint was vagina bleeding and incontinence.
After developing symptomatic anaemia with fatigue and dizziness, she sought medical attention from her general practitioner who found her haemoglobin was 84 g/dL and referred her to the hospital for further assessment. Pelvic examination, including speculum by gynaecology showed fluid passing through the cervical os consistent with urine. No cervical lesions were seen.
A flexible cystoscopy by urology showed a suspected fistulous tract of approximately 1 cm diameter with surrounding erythematous mucosa, and the differential diagnosis was a locally advanced malignancy. An MRI of the pelvis was performed clearly demonstrating a fistulous tract between the posterior bladder wall and the anterior uterus with no malignant features. In the context of the onset of incontinence shortly after caesarean section, and the absence of other causes of vesicouterine fistulae such as radiotherapy and malignancy, the diagnosis of delayed presentation of the iatrogenic vesicouterine fistula was formed.
After the consultation, a robotic hysterectomy and repair of the fistulous tract were planned given she had no desire to have further children, and the fistula was large and chronic in nature.
Investigations
Several investigations were performed to come to a definitive diagnosis.
Imaging
-
Pelvic ultrasound: Generally unreliable in the assessment for vesicouterine fistula, though modified techniques have been reported.12 This patients’ ultrasound was normal, with no fistula visualised. Pelvic ultrasound as a single modality of investigation is not recommended.
-
MRI: An MRI scan was performed with contrast. This gives a more distinct appreciation of the soft tissues when compared with CT. The MRI clearly demonstrated fluid within the uterine cavity with a fistulous tract between the posterosuperior bladder and the uterus. The endometrium was thickened, but no overt malignant change was seen. (figure 1) The bladder wall maintained a normal thickness.
Sagittal T2-weighted non-contrast MRI image. Fluid is seen in both bladder and uterus communicating via the fistulous tract (arrow).
Invasive
-
Flexible cystoscopy: An invasive investigation performed to assess the bladder for (1) source of haematuria, (2) cause of incontinence and (3) exclusion of malignancy. In this patients’ case, a clear fistulous tract was seen (figure 2). The prominent bullous erythema around the tract was thought to be the likely source of the light haematuria.
A cystoscopic image of the fistulous tract (arrow) with surrounding erythematous mucosa.
Investigations not used in this case, but that may also be applicable include:
Clinical assessment: The three pad test
- Intravesical instillation of methylene blue or oral phenazopyridine (pyridium) then packing the vagina and assessing for colouration on the pads can help to differentiate ureterovaginal fistulas and vesicouterine from vesicovaginal fistulas. The colour of fluid on the pad, and the order of the pads affected indicate the level of the fistula. In vesicouterine or ureterovaginal fistula the pad in the proximal third of the vagina should be discoloured as urine passes through the cervix/fistula, while in vesicovaginal fistula the middle and lower third vaginal pads should be discoloured. Care must be taken in patients with stress incontinence as leakage per urethra may also discolour the lower pad.
Imaging
-
Contrast enhanced CT: This investigation can be used for identification of fistulae, though in our patients case where malignancy was considered a possibility it was felt an MRI would provide more information. Where the working diagnosis is an iatrogenic fistula, CT is a reasonable diagnostic test.
-
Cystogram: Retrograde cystogram can also identify fistulae with opacification of the uterine cavity following administration of contrast through a catheter.
Per vaginal fluid
-
Measure Creatinine in the fluid passing through the vagina. As it is urine the level should exceed that of serum creatinine by a degree of 100’s–1000’s.13
Invasive
-
Rigid cystoscopy: In cases where concern over malignant fistula persists this facilitates biopsy if histology is required to guide treatment.
Differential diagnosis
In this patient, the working diagnosis was delayed presentation of iatrogenic vesicouterine fistula. The principal differential diagnosis needing exclusion was malignancy in the pelvic viscera, particularly bladder cancer, cervical cancer, uterine cancer and endometrial cancer.
Ultimately the diagnosis of the benign vesicouterine fistula was reached through consideration of the history which showed a temporal relationship with the original caesarean section, as well as the investigations which included an MRI showing no compelling features of malignancy and a cystoscopy which revealed some erythema through no typical features of bladder cancer, or extravesical malignancy invading the bladder.
Surgical excision of the uterus and fistulous tract was performed and the histology confirmed no malignancy.
Treatment
In discussion with the urology team, gynaecology team and the patient, a plan was made to perform a robotic-assisted hysterectomy and excision of the fistulous tract with bladder repair.
A laparoscopic approach has been described in the literature.8 14–20 However, reports of robotic procedures are exceedingly rare.21–23
There are several descriptions of non-excisional repair/closure of fistulae, however, given the time period over which this fistula had been present, the size of the fistula, the well-epithelialised tract, and the need for histology, it was felt the best option remained excision with hysterectomy en bloc.24–28
Based on local expertise and clinician experience, the procedure was performed robotically. A DaVinci Si Surgical System robotic platform (Intuitive Surgical, Sunnyvale, California, USA) was used with four ports and an additional Air Seal iFS port for assistance (ConMed, Utica, New York, USA). The ports were placed with a supra-umbilical camera, and a standard infraumbilical working port array.
At the commencement of the procedure, a rigid cystoscopy was performed. The ureteric orifices were confirmed to be several centimetres from the fistula. The fistula was then circumscribed with a Collings knife (Cook Medical Bloomington, Indiana, USA) endoscopically to mark the planned excision margins.
The robot was then docked, and the vesicouterine plane developed. The fibrotic fistulous tract was identified and the bladder opened with the tract excised en bloc with the uterus sparing the cervix (figure 3). Subtotal hysterectomy was performed as evidence shows it is faster and associated with less blood loss than total hysterectomy.29 In addition, it avoided having two long suture lines crossing. Hysterectomy was then completed and the specimen delivered through the camera port (figure 4). The bladder was closed with 2 layers of continuous 2–0 polyglactin suture (Vicryl) (Ethicon, Somerville, New Jersey, USA). The Os of the cervix was closed with 2-0 polyglactin suture. A pedicalised omental tongue was sutured into the plane between the two suture lines with further polyglactin suture. A leak test performed with 250 mL showed no evidence of leakage from the bladder.
Intraoperative image showing the bladder open and the endoscopically marked planned resection margin (arrow).
The uterus and fallopian tubes with arrow indicating the small section of bladder excised and fistulous tract.
The patient was transferred to the ward and recovered over 3 nights. A drain left overnight had drained only 15 mL of fluid in 24 hours, so was removed. She had minimal abdominal pain, though some bloating and constipation. She was discharged with a catheter in place and returned at 2 weeks for a Cystogram which showed no evidence of urine leak (figure 5); thus, the catheter was removed.
The cystogram showing resolution of the fistula with no urine leak.
Outcome and follow-up
The patient was followed up at 3 months. Her haemoglobin had recovered and was 9.8 g/dL, a substantial improvement in her initial presentation. She had no further haematuria, per vaginal bleeding or urinary incontinence. She reported a marked improvement in her quality of life.
Final histology demonstrated a fistulous tract with endometrial lining. There was no evidence of malignancy.
She subsequently moved home to another city several hours away and will be followed up locally.
Discussion
Given this is a rare pathology, there are no published guidelines specifically addressing vesicouterine fistula from any Urological or Gynaecological society or the WHO. While there are guides on the management of vesicovaginal fistula13 these fistulas by their nature require a different management strategy.
Consequently, surgeons treating vesicouterine fistulae must turn to the literature where case reports and small case series represent the documented knowledge on this condition. These case reports are for fistulae that have been present for a shorter time period. They are principally focused on laparoscopic and open repairs. In most cases, the recommended treatment is uterine sparing. In this context, our case is unique as uterine preservation was not desired by the patient, and the fistula was present for over 15 years.
Learning points
-
Early recognition of obstetric fistula offers the potential for a prompt repair.
-
In patients with cyclic haematuria, amenorrhoea and persistent urinary incontinence, a vesicouterine fistula must be considered and excluded.
-
MRI is an excellent diagnostic test, though cystoscopy is critical in establishing diagnosis and ruling out malignancy.
-
Minimally invasive repair with robotic techniques is feasible and appropriate and should be considered in all patients suitable for treatment, where the equipment is available.
Footnotes
-
Contributors ERL: wrote the case report and obtained patient consent. MKQ: edited and updated the section on investigations. DM: completed referencing and edited images. ES: performed the operation, wrote the section on treatment and edited the discussion.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Patient consent for publication Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2020. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer