Introduction

Vesicovaginal fistula (VVF) is an abnormal connection between the urinary bladder and the vagina, which causes leakage of urine in the vagina. Although a rare entity, urogenital fistula is caused mainly by surgery, radiation therapy, or malignancy in the Western world [1]. It occurs due to obstetric complications such as prolonged and obstructed labor in developing parts of the world. It remains an important but neglected topic that the World Health Organization has referred to as a forgotten disease [2,3,4]. The incidence of VVF ranges from 0.3 to 2% [5]. At least 3 million women worldwide are believed to have an untreated vesicovaginal fistula, with the majority of them from Africa and Southern Asia. In Africa, 30,000 to 130,000 women develop vesicovaginal fistula annually [1]. Women having VVF are continuously damp from urine leakage and sometimes suffer genital ulceration, infections, and an unpleasant smell. Approximately 20% of women with fistula often develop unilateral or bilateral foot drop that restricts their daily activities [6].

In women with this disorder, it causes physical, social, and psychological effects. VVF prevention and management can be supported by knowledge of the disease, professional birth attendance, surgical care, along with therapeutic support. Addressing the rising public health concerns of VVF, various charitable and non-governmental organizations are developing management programs and establishing particular centers for the care of patients with VVF [3, 7].

The majority of reports for VVF consisted of case series and experiences of health professionals. Whereas the existing studies were not specific, with studies mostly focused on obstetric fistulas as mainstream. In this study, we systematically reviewed the existing literature of the last decade, discussing the occurrence of vesicovaginal fistula, its etiology, surgical approach, and outcomes after developing VVF.

Material and Methods

Protocol

Our systematic review and meta-analysis were conducted according to the MOOSE guidelines after registration in PROSPERO (CRD42020215772) [8].

Eligibility Criteria

We included cross-sectional studies, case-control studies, cohort studies, and case series (more than 20 patients) with women diagnosed with vesicovaginal fistula during 2010–2020 and excluded studies with women diagnosed as other causes of urinary incontinence and pregnant women. We also excluded the study with inadequate data and results. In addition, letters to the editor, viewpoints, and experiences were also excluded in the study.

Search Strategy

We used electronic databases like PubMed, PubMed Central, Scopus, and Embase to search relevant articles from 2010 to 2020 using terms like “vesicovaginal fistula”, “VVF” and “gynecological fistula” with appropriate Boolean operators. The detailed search strategy is included in the supplementary file.

Study Selection

Two reviewers (PJ and PK) independently screened the title and abstract of imported studies, and any arising conflict was solved by the third reviewer (GM). A full-text review was done independently by GM and PK. Data were extracted for both quantitative and qualitative synthesis. The conflicts were resolved by taking the opinion of the third reviewer (PJ). The screening was performed with the help of Covidence [9].

Data Extraction

Relevant data, including study characteristics, quality, and endpoints, were extracted onto a standardized form designed in Excel. Our outcomes were the prevalence of overall genitourinary fistulas, vesicovaginal fistulas among different genitourinary fistula, anatomical types of vesicovaginal fistula, and gynecological etiology of vesicovaginal fistula, the surgical approach for closure, and success of closure of the vesicovaginal fistula. We extracted the data from included studies based on our outcomes of interest.

Methodologic Quality

The quality of individual articles was evaluated using the Joanna Briggs Institute (JBI) critical appraisal. In addition, the risk of bias was assessed. Two of the authors had independently assessed the design of each study, the number of patient included outcomes of VVF, included risk factors, and if the outcome as mentioned earlier were measured. Disagreements were resolved by discussion with a third person.

Data Analysis

Data were analyzed using CMA-3 [41]. The proportion was used as a measure of effects, and the I2 test measured heterogeneity. The random/fixed-effect model was used based on heterogeneity.

Sensitivity Analysis

Sensitivity analysis was done by excluding individual studies to observe the impact of individual studies.

Subgroup Analysis

Subgroup analysis was performed while evaluating the outcome of interest as appropriate. In addition, less commonly reported results were tabulated in supplementary files.

Publication Bias

Publication bias across the study was assessed using Egger’s funnel plot using the MD and 1/SE values for appropriate outcomes.

Results

We identified a total of 8288 studies after thorough database searching and a total of 1875 duplicates were removed. We screened 6413 studies and excluded 6014 studies. After assessing 399 studies for full-text eligibility, 368 were excluded for definite reasons (Fig. 1). The remaining 31 studies were included in the qualitative summary and quantitative analysis (Table 2 and Supplementary file 2).

Fig. 1
figure 1

PRISMA flow diagram

Full size image

Quantitative Analysis

Total of 31 studies were included in the analysis. There was no study from an apparently normal population investigating genitourinary fistula, but two studies evaluated the prevalence of genitourinary fistula (GUF) among risk groups and showed 12.3% (CI: 1.5–56%) (Supplement file 3, Fig. 1).

Rate of VVF Among GUF

Fifteen studies reported the VVF among the GUF series they have studied. Pooling the data using the random effect model showed 76.57% of cases were VVF among GUF (proportion, 0.7657; CI, 0.6542–0.8496) (Fig. 2). Sensitivity analysis to gauge the impact of individual studies in the overall result was conducted by excluding individual studies and showed no significant change after excluding particular studies (Supplement file 3, Fig. 2).

Fig. 2
figure 2

Rate of vesico-vaginal fistula (VVF) among genitourinary fistula (GUF) studied in different studies

Full size image

Common Anatomical Types of Fistula Reported

In most studies, there were no clear specifications of different anatomical types of VVF rather classified overall GUF, so while pooling anatomical types of all GUF pooled. Pooling of data from six studies reporting a common anatomical type of fistula using a random-effect model showed supra-trigonal in 55.70% (Proportion, 0.5570; CI, 0.3439–0.7510; I2, 93.87), trigonal in 27.54% (Proportion, 0.2754; CI, 0.1811–0.3952; I2, 83.86) (Fig. 3). Rest, less commonly reported fistula were circumferential, juxta-cervical, juxta-urethral, etc. (Supplement file 3, Table 1).

Fig. 3
figure 3

Commonly reported anatomical types of fistula

Full size image
Table 1 JBI assessment of included studies
Full size table

Obstetric Fistula

Obstetric etiology was commonly reported etiology in most of the studied fistula population.

Cesarean Section

Pooling of data from 19 studies reporting a cesarean section using a random-effect model showed 19.29% (proportion, 0.1929; CI, 0.1326–0.2721; I2, 97.78) (Fig. 4). Sensitivity analysis to gauge the impact of the individual study on the cesarean section as etiology was carried out by excluding individual studies and showed no significant change after excluding particular studies (Supplement file 3, Fig. 3).

Fig. 4
figure 4

Cesarean section as culprit etiology for fistula among GUF cases reported in various studies

Full size image

Obstructed Labor

Pooling of data from 13 studies reporting an obstructed labor using a random-effect model showed 31.14% (proportion, 0.3114; CI, 0.1823–0.4786; I2, 96.80) (Fig. 5). Sensitivity analysis to gauge the impact of the individual study on obstructed labor as etiology was carried out by excluding individual studies and showed no significant change after excluding a particular study (Supplement file 3, Fig. 4). Other less commonly reported obstetric etiology of fistula were vaginal delivery, cesarean hysterectomy, instrumental delivery, etc. (Supplement file 3, Table 2). Most obstetric fistulae were iatrogenic in origin, and the commonly reported were cesarean section, cesarean hysterectomy, instrumental deliveries, etc. (Supplement file 3, Table 3).

Fig. 5
figure 5

Obstructed labor as culprit etiology for fistula among GUF cases reported in various studies

Full size image
Table 2 Qualitative summary
Full size table

Gynecological Etiology of Fistula

Among gynecological etiology, hysterectomy (vaginal, abdominal) was the commonly reported etiology. Less widely reported gynecological etiologies include radiation therapy for cancer, different gynecological procedures, and cancer (Supplement file 3, Table 4).

Among 16 studies reporting hysterectomy, pooling of data using a random-effect model showed 46.52% of fistula associated with hysterectomy (proportion, 0.4652; CI, 0.3617–0.5719, I2, 95.72) (Fig. 6). Sensitivity analysis to gauge the impact of the individual study on hysterectomy as etiology was carried out by excluding individual studies and showed no significant change after excluding a particular study (Supplement file 3, Fig. 5).

Fig. 6
figure 6

Hysterectomy as culprit etiology for fistula among GUF cases reported in various studies

Full size image

Surgery for Fistula Closure

Different types of surgical treatment were employed as a definitive treatment of fistula closure. Due to the unavailability of data on surgical treatment of VVF, the management of GUF was only reported in most studies, so pooling was done for the management of GUF. Surgical approach for closure includes the vaginal approach, abdominal approach, combined abdominal and vaginal, laparoscopic approach, and less commonly employed procedures were diversion techniques, etc. (Supplement file 3, Table 5).

The abdominal approach was reported in 17 studies. Pooling of data showed that 49.50% of the surgical closure was done by the abdominal approach (proportion, 0.4950; CI, 0.3723–0.6182; I2, 93.55) (Fig. 7). Sensitivity analysis to gauge the impact of the individual study on the abdominal approach for fistula closure was carried out by excluding individual studies and showed no significant change after excluding particular studies (Supplement file 3, Fig. 6).

Fig. 7
figure 7

Abdominal approach for surgery among GUF

Full size image

A vaginal approach for fistula closure was reported in 14 studies. Pooling of data showed 42.31% of procedures carried out by a vaginal approach (proportion, 0.4231; CI, 0.3182–0.5354) (Fig. 8). Sensitivity analysis to gauge the impact of the individual study on the vaginal approach for fistula closure was carried out by excluding individual studies (Supplement file 3, Fig. 7).

Fig. 8
figure 8

Vaginal approach for surgery among GUF

Full size image

Successful Closure of the Fistula

Twenty-three studies reported successful closure of fistula in their outcome. In 87.09% of the surgeries (proportion, 0.8709; CI, 0.8439–0.8938), a successful closure of fistula was reported (Fig. 9). Sensitivity analysis on successful fistula closure by excluding individual studies showed no differences (Supplement file 3, Fig. 8). Among operated cases, 82.69% were successful and continent surgeries (Proportion, 0.8269; CI, 0.7393–0.8895; I2, 83.39) (Supplement file 2, Fig. 9.). Sensitivity analysis on successful and continent surgeries by excluding individual studies showed no significant differences (Supplement file 3, Fig. 10).

Fig. 9
figure 9

Successful surgery among GUF

Full size image

Publication Bias

Included studies showed some publication bias for the respective outcome. Supplementary file 3, Fig. 11 showed publication bias of reporting VVF among fistula using Egger’s funnel plot.

Discussion

Vesicovaginal fistulas have a significant impact on the patient’s physical, social, and mental well-being. They have remained a concealed condition as it affects most of the overlooked population of women in the rural parts of the world. It can stigmatize a woman in society and lower her self-confidence and outlook towards life. A paper labels obstetric fistula to be the neglected condition of poverty [42]. There is a need for effective measures to prevent this condition by properly identifying the etiology, its occurrence, and risk factors in the community. Furthermore, there is a need for proper universal education, empowerment of women with accessible and improved medical services.

We found that the vesicovaginal fistula is the most common type of genitourinary fistula, and it accounted for 76.57% of various types of genitourinary fistula. This is concordant with Hillary’s systematic review, which mentions vesicovaginal fistula as the most common type of fistula [3]. We found that the prevalence of genitourinary fistula (GUF) among the risk group is 12.3% (CI: 1.5–56%). However, this estimate was based on just two studies, and the lack of inclusion of normal women of reproductive age group makes our finding hard to generalize. Among the different types of vesicovaginal fistula, the common types were supra-trigonal in 55.70%, followed by trigonal in 27.54%, and other types including circumferential, juxta-cervical, and juxtaurethral. VVF can be classified on various bases like the fistula site, etiology, involvement of continent mechanism, size of fistula, and clinical examination. Classification of fistula into types aids in the decision-making about the management of the patients, adjunct treatments, and follow-up guidance.

The pooling of data from our study showed that the primary etiology of the fistula was obstructed labor and C-section among obstetric etiology, and history of gynecological surgery among gynecological etiology. This aligns with a review that points out the common cause of VVF in developed countries to be pelvic surgery [3]. In cases of underdeveloped countries, prolonged obstructive labor is noted to be the most common etiology (95.2%), followed by cesarean section (9%) and instrumental delivery (2%) [3]. There is a significant discrepancy in VVF’s reported incidence and causes between the developed (0.3%) and developing nations (2%) [43]. These figures suggest the need for more intensive studies in this area, especially in developing countries, due to its relatively high incidence and preventable etiology. There is a lack of adequate studies done in these nations reporting on vesicovaginal fistula.

The timing of repair of the vesicovaginal fistula is widely debated, dependent on the status of surrounding tissues. Early repair is preferred in the case of instrumental delivery or cesarean section when the tissue is healthy. However, in cases of gynecological surgery, a 6–12-week delay allows dissipation of most granulation tissue, increasing the possibility of a successful repair. This review shows that most of the research displayed that the surgery successfully treated the fistula, with 87.09% having urinary continence post-surgery. Rajamaheswari et al. [44] demonstrated the successful vaginal and abdominal repair outcome as 86.7% and 100%, respectively. The study also concluded that most supratrigonal VVF showed comparable results when approached vaginally or abdominally [44]. Another study by El-Azab [45] noted that the success rate for a vaginal approach was 91%, whereas an abdominal repair was 84%. The preferred approach for surgical repair relies on the surgeon’s familiarity, location of the fistula, space in the vaginal cavity, need for procedures like ureteric reimplantation, and feasibility of getting necessary interposition flaps. Both routes have their advantages and drawbacks. Our study found a higher rate of abdominal approach for the correction of the fistula than the vaginal approach. Usually, the abdominal route is chosen when the vaginal repair is contraindicated. The vaginal approach was used in 42.31% of patients with vesicovaginal fistula based on our study, which is far lower than Hillary’s review in which 71% and 81% of repair of lower urinary tract fistula were done transvaginally [3]. There are multiple advantages with a vaginal repair, such as shorter operative time, decreased hospital stay, reduced blood loss, and avoidance of abdominal and bladder incisions. However, both studies pointed out a lack of randomized trials to effectively compare the benefits of transabdominal and transvaginal approaches, which could provide an important area of study for future research [44, 45].

It is important to implement guidelines on safe obstetric practice and good surgical practice in gynecological surgeries that would help reduce the genitourinary fistula. However, one of the limitations of our review could be the inability to correctly portray the incidence and prevalence rates because many cases occur in developing nations where there is a lack of proper diagnosis, documentation, and treatment modalities available. Additionally, most studies did not clearly report the outcome of VVF separately, instead, they reported the outcome of overall GUF so we could not fully dissect the details of VVF alone. Also, our review was limited to English-language articles alone. Thus, we recommend formulating national policies that disseminate the information about the condition among middle-aged women, proper identification and documentation of the cases seen, proper maternal prenatal, natal, and postnatal care, and the provision of proper technologies and resources for its treatment.

Selecting the abdominal or vaginal approach of vesicovaginal fistula repair may be biased by the surgeon’s basic specialization, whether gynecologist or urologist. Thus, another variable of study would be a basic specialization or specialty unit carrying out the repair.

Conclusion

Vesicovaginal fistula is the most common type of genitourinary fistula. Still, there is a significant discrepancy in the incidence and causes of VVF between developed and developing nations, and obstructed labor leads to the most common cause in developing countries. Though we have noticed that both vaginal and abdominal approaches are almost equally used to repair a fistula, both show favorable outcomes. This could be the result of bias of operating surgeons’ preference based on their initial training. More robust studies and improved reporting of cases should be encouraged to improve the data in the future.