Abstract
Background
Enterovaginal fistulas represent a serious complication of various diseases and therapeutic procedures, often associated with complicated clinical courses and massive impairment of quality of life. As underlying conditions and procedures are multifarious, therapeutic approaches are challenging and have to be tailored individually. As the therapeutic management is complex and individualized, multiple surgical interventions might be necessary.
Methods
The aim of this study was to identify possible predictors for outcome in the treatment enterovaginal fistula patients. The study was realized as a retrospective analysis. Ninety-two patients treated with enterovaginal fistulas between 2004 and 2016 were analyzed. Patient characteristics, therapeutic data, and endoscopic findings were stratified according to etiology, closure rate and time, as well as recurrence of fistula. Main outcome measure was the overall rate of fistula closure.
Results
Overall therapeutic success rate was 67.4%. Postoperatively derived fistulas were most frequent (40.2%), mainly after rectal surgery (59.5%). Postoperative and non-IBD-inflammation associated fistulas had better outcome than IBD-, radiotherapy-, and tumor-related fistulas (p = 0.001). Successful fistula closure was observed more frequently after radical surgical interventions, best results observed after transabdominal surgery (p < 0.001). Fistula recurrence was also less frequently observed after radical surgical therapies (p = 0.029). A temporary stoma was associated with higher incidence of fistula closure (p = 0.013) and lower incidence of fistula recurrence (p = 0.042) in the postoperative subgroup, as well as shortened therapy period in all groups (p = 0.031).
Conclusion
Enterovaginal fistulas are a result of various etiologies, and treatment should be adjusted accordingly. A very sustainable, rapid, and persistent therapeutic success can be expected after radical surgical approaches with temporary diverting stoma. This is especially true for postoperatively derived fistulas.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Enterovaginal fistulas are rare afflictions possibly leading to severe clinical symptoms with critical impairment of quality of life for affected women. A variety of predisposing conditions and surgical procedures have been described to potentially cause enterovaginal (EV) fistulas [1,2,3,4, 9,10,11]. The postpartum development of entero- or rectovaginal fistulas is most common in up to 88% of all EV fistula events [1, 2]. Crohn’s disease with perineal disease manifestation causes only approximately 0.2–2.1% of all enterovaginal fistulas [2, 9]. Yet, the rate of Crohn’s disease-related EV fistulas might rise up to 10% in case of previous rectal resections. Likewise, entero- or rectovaginal fistulas are a relevant phenomenon after rectal surgery for malignant and benign diseases as well as after pouch procedures in patients with ulcerative colitis and familial adenomatous polyposis [5, 7, 8]. Concurrently with the introduction of stapler hemorrhoidectomy and transanal stapled rectal resection for obstructed defecation, rectovaginal fistulas have been associated with these procedures [2,3,4, 6]. Likewise, surgical interventions using stapler devices for functional pelvic floor disorders have been reported to result in rectovaginal fistulas [2,3,4, 6]. Other than obstetric fistulas, gynecologic malignancies or radiation therapy for gynecologic tumors may lead to fistula development [10, 11].
As etiologic conditions and procedures vary wildly, therapeutic approaches are challenging and have to be individually tailored [14]. As the therapeutic management is complex and individualized, multiple surgical interventions might be necessary. Besides well-described local surgical approaches including mucosa flap, surgical repair might require a diverting stoma, rectal resection, or even complex reconstruction with muscle transposition [2, 12,13,14,15,16,17,18]. Nowadays, endoscopic advances such as OTS-clips (over the scope clips) offer new therapeutic prospects; yet, their clinical success and role has to be further evaluated [19].
While several studies have already focused on clinical features of selected EV fistula patients, data regarding the genesis and treatment proportions in EV fistula patients in real life settings is still scarce. Moreover, general conditions of the patients affected (e.g., comorbidities) should be taken into account when choosing treatment algorithm.
Therefore, the aim of the present study is to analyze long-term outcomes after various treatment modalities of EV fistulas of diverse origin and thus to identify possible predictors for treatment outcome.
Methods
Study design
The study was designed as a monocentric, observational cohort-study at a single academic reference center for Surgical Endoscopy. The study was approved by the Institutional Review Board of the LMU University of Munich (protocol number EK-LMU 19–062).
Study population
One hundred eighteen patients who presented with entero-, colo-, recto-, or anovaginal fistula at the Department of General, Visceral and Transplantation Surgery of the Ludwig-Maximilians University Munich, between 01/2004 and 12/2016 were included into primary analysis (Fig. 1). These patients were stratified according to their respective fistula’s cause. Finally, patients were categorized into the successfully treated fistula cohort (SFT) if closure of the fistula was achieved during the observation period. If persistence of the fistula was detected during the observation period, respective patients were labeled as unsuccessful therapy cohort (UFT).
Data sources
Demographic data, information on clinical history, clinical examinations, and therapeutic information were extracted from the clinical documentation system, clinical charts, endoscopic reports, and anaesthesiology reports.
Outcomes
Overall rate of fistula closure was defined as primary endpoint. In addition, the following secondary endpoints are analyzed: rate of fistula closure depending on etiology, on therapeutic approach, and on existence of a diverting stoma; fistula recurrence after initial therapy success depending on etiology and treatment; and potential further influential factors on therapy success such as comorbidities.
Successful fistula closure was defined as missing clinical signs of enterovaginal fistula, meaning no secretion, no pus, and no newly occurred abscess. Unsuccessful treatment was defined as clinical evidence of fistula, e.g., persistent secretion, detectable fistula course, and newly occurred abscess.
Patient- and treatment-specific variables
The following parameters were assessed at time of first fistula treatment: age, Charlson Comorbidity Index (CCI), serum albumin levels, preexisting anorectal diseases and previous surgeries, cause of EV fistula, antibiotic treatment, treatment modalities, number of therapeutic approaches applied per patient (therapy load), number of operations performed in curative intent (operative load), fistula recurrence after initial therapy success, and creation of a diverting stoma.
The Charlson Comorbidity Index (CCI) is an index to predict a mortality risk over time for comorbid conditions [20].
Statistical analysis
This study was carried out as an explorative study. Descriptive statistical analysis was performed, and relations between fistula etiology, therapeutic success, and further underlying conditions were reviewed. In further steps inter-group differences were quantified by performing comparative analysis calculating Student’s t-test and Fisher’s exact test for single-group differences, as well as chi-square test for multiple group differences. Odds ratios have been calculated applying the Babtista-Pike method. For statistical analysis and graphical presentation, SPSS statistical software package (version 25, IBM, Chicago, Ill) and GraphPad Prism (version 8.4.2, GraphPad Software, San Diego, CA) were used. p value (two-tailed) of < 0.05 was regarded as statistically significant.
Results
Study population
Primary database query revealed a total of 118 patients who presented with enterovaginal fistula to our Surgical Endoscopy Unit between 2004 and 2016. 19 patients had to be excluded due to insufficient primary data, whereas 7 patients were excluded due to loss of follow-up. Hence, 92 patients (78%) were included into the final analysis. The algorithm of patient case selection for the analysis, as well as detailed number of patients stratified by fistula etiology is shown in Fig. 1.
Nineteen patients presented with an anovaginal fistula, 20 with a rectovaginal, and 12 with a sigmoidovaginal fistula. Two patients suffered from enterovaginal fistula, and for 39, a specific localization of the fistula origin within the colon and rectum was not documented. Median follow-up was 18.9 months.
The majority of patients was affected by postoperative fistulas (40.2%), and only 3.3% presented with postpartum fistulas. 59.5% of postoperative fistulas occurred after rectal resection, 27.0% after gynecological surgeries, and 13.5% after combined interventions. 44.6% of the patients did not have any history of pre-existing anorectal diseases. Of all postoperative fistulas, 12 (32.4%) occurred directly at the anastomotic site.
Overall closure rate was 67.4%, whereas fistula therapy was unsuccessful in 32.6%.
Details on primary therapeutic approaches as well as on fistula etiology and main demographic data are given in Tables 1 and 2 and Fig. 1.
Therapy success and risk for recurrence depending on fistula etiology
Concerning fistula etiology, postoperative and non-IBD inflammatory fistula (e.g., diverticulitis) had the best therapeutic outcome, which was significantly better as compared to fistula of other etiologies (Fig. 2A). Postpartum fistulas showed a 100% therapeutic success. Tumor- and radiotherapy-related fistulas demonstrated the least favorable outcome with a high proportion of unsuccessful therapies. Likewise, IBD-related fistulas were related to impaired therapeutic outcome.
No difference could be detected between etiology-specific subgroups regarding risk for fistula recurrence. Concerning postoperative fistulas, there was a tendency towards better results in case of primary gynecological surgeries compared to primary rectal resections leading to enterovaginal fistulas, though the groups were small, and the comparison did not reach statistical significance (p = 0.1893; Fig. 2C).
Details on therapy outcome with regard to fistula etiology are given in Table 2 and Fig. 2.
Therapy success depending on co-morbidities
A higher CCI-score was associated with a clear tendency towards poorer therapeutic outcome (p = 0.0752; Fig. 2D). However, no difference in serum albumin-levels was detected between the SFT and UFT patient cohorts. Pre-existing anorectal disorders did not show any influence on therapeutic outcomes (Fig. 2B). Details are shown in Table 2 and Fig. 2.
Therapy success depending on type of therapeutic approach
Regarding the first therapeutic approach, patients undergoing abdominal operation for fistula closure showed significantly better results with a closure rate of 69.2% as compared to local operative therapy, endoscopic therapy, conservative approach, or stoma alone (Table 3, Fig. 3A).
In case more than one treatment effort was needed due to fistula recurrence or persistence, a transabdominal approach as the most invasive treatment option still showed the best results with a closure rate of 62.9%. A local surgical approach also led to favorable closure rates (24.2%) (Fig. 3B). Moreover, a local surgical approach showed better closure rates than an endoscopic, conservative, or stoma alone treatment regimens.
Endoscopic interventions showed poor closure rates with only 5.1% success rate in case of primary approach and 4.8% in the overall analysis for all therapeutic options.
A diverting stoma alone was never sufficient for fistula closure. Yet, this approach was used for symptom control in case of tumor related fistula and a palliative setting.
There was no difference in fistula recurrence rates depending on the treatment chosen as the first curative treatment (Fig. 3C). Nevertheless, recurrence was monitored significantly less frequently in case of a more radical therapeutic approach at any time during the therapy sequence (Fig. 3D). Details on fistula closure rates and recurrence rates with regard to the therapeutic regimen are shown in Tables 3 and 4 and Fig. 3. Odds ratios for different therapeutic approaches are given in Table 5.
Stoma vs. no stoma
Overall analysis did not reveal an advantage for a temporary stoma with regard to fistula closure rates (Fig. 4A). However, there was a tendency towards lower recurrence rates in case of a temporary stoma. Besides, fistula closure was achieved significantly faster in case of a temporary enteral diversion (p = 0.0312; Fig. 4B). Especially the group of postoperative fistulas showed a clear and significant benefit for both, primary fistula closure and recurrence in case of an applied stoma. This was independent of the treatment regimen chosen otherwise (Fig. 4C, D). Details on the effect of a temporary enteral diversion for fistula closure and recurrence rates are shown in Fig. 4, as well as Tables 5 and 6.
Discussion
The presented study analyzes a large cohort of patients with enterovaginal fistulas. This condition, independent of its genesis, leads to crucial impairment of women’s quality of life and often is a challenge with frequent need for several therapeutic interventions. The aim of the presented study was to analyze therapeutic success independent of fistula genesis, to evaluate specific surgical techniques, and to identify potential predictors on surgical outcome.
The presented patient population is heterogenous as enterovaginal fistulas are caused by several underlying conditions. With an overall closure rate of 67.4%, our own results are comparable or better than previously reported in the literature [2, 21,22,23]. Differences might be due to differences in fistula origin between our cohort and that reported previously [23]. In our analysis, we demonstrate that in case of postoperative fistulas, therapy success is likely and even especially favorable after preceding gynecological surgeries. Postpartum fistulas, in our own patient population had a success closure-rate of 100%, the results being underlined by a previously published series with similar closure rates [24]. Favorable results in non-IBD associated inflammatory by radical treatment of the infectious focus alone, are in line with previous reports [18, 21]. Tumor- and radiation-associated fistulas were found to have the most unfavorable closure rates, just like IBD-associated fistulas with impaired therapeutic success.
Concerning different treatment regimens, more invasive treatment options like an abdominal surgical approach showed the best results even in cases of recurrent fistula. This might also include discontinuity resections with secondary reconstruction at a later point. These results are concordant with data reported by Corte et al. in 2015, who also found higher rates of fistula closure when major surgical procedures were preferred [21]. Similarly, a local surgical approach was likely to lead to therapy success as well. Even though our results suggest a rather radical treatment to show higher association with satisfying fistula closure and recurrence rates, the time point of using invasive treatments within the curative therapy sequence does not seem to be important. Therefore, the primary use of less invasive treatment options might still be applicable for selected patients, especially for those with higher comorbidities. Although an older study by Lowry and co-workers found two or more prior surgical approaches for fistula closure to be associated with a higher relative risk for recurrence, our own data still suggests that in case of initial treatment failure, the use of a more invasive therapy is still helpful and reasonable at a later point without significant impairment in outcome [22]. Thus, we advocate an individualized step-up approach, rather than a maximal invasive treatment in all patients for treatment of enterovaginal fistulas.
Whereas Corte and colleagues identified a diverting stoma as an independent factor for treatment success in case of rectovaginal fistula, we did not confirm this [21]. Nevertheless, our results suggest that a temporary stoma shortens time to fistula closure. However, this was barely significant with regard to the whole population and could not be concluded in a subgroup-analysis. The latter seems mainly caused by large differences in the time intervals until fistula closure. Patients with a postoperative fistula profited most from a stoma whereas patients with inflammatory genesis, such as diverticulitis, did not in the same manner.
Among patients with unsuccessful treatment attempts, the CCI score was higher. This finding is in line with previous results as a higher CCI is a risk factor for anastomotic leakage in colorectal surgery [25, 26]. Therefore, in case of a high CCI score in patients with enterovaginal fistula, the treatment regime has to be planned carefully, and emphasis has to be put on optimization of comorbidities prior to treatment.
There are some limitations of this study. The retrospective character does not allow for standardization of the documentation of fistula characteristics, such as height and length. However, a prospective study with regard to therapy strategies for enterovaginal fistulas might be impossible to realize since many individual aspects play a substantial role in the therapeutic decision-making. Especially heterogeneity of the disease and patient cohort leads to a variety of treatment options. The presented patient population differs to some extent from those previously published, e.g., with a much lower proportion of postpartum fistulas. However, the analysis of this large patient population—as compared to previous studies—allowed to perform subgroup analysis concerning etiology-specific aspects and success rate of different treatment approaches.
Our study substantiates the multifarious character of enterovaginal fistulas regarding their genesis, as well as possible therapeutic options, and strengthens the fact that the therapeutic approach needs to be tailored individually according to fistula etiology, localization, and individual patient characteristics [2, 14, 23, 27, 28].
The results suggest to favor invasive treatment options, as well as the use of a diverting stoma, in order to achieve beneficial therapy success and low recurrence rates. In detail, the results of our study propose that in case of inflammatory origin of the fistula, surgical treatment of the infectious focus alone will likely lead to success. In case of postoperative enterovaginal fistulas a radical abdominal re-operation including a temporary stoma is most favorable with high closure-rates and will shorten the time interval to fistula closure. If this cannot be performed, a local surgical intervention will also likely lead to therapeutic success and should be favored over the sole use of endoscopic or conservative treatment options. For those patients, an individual step-up approach (local treatment first, a more radical approach in case of fistula recurrence) is a valuable alternative treatment option (Fig. 5).
Data availability
Not applicable
References
Homsi R, Daikoku NH, Littlejohn J, Wheeless CR Jr (1994) Episiotomy: risk of dehiscence and rectovaginal fistula. Obstet Gynecol Surv 49(12):803–808
Ommer A, Herold A, Berg E, Fürst A, Schiedeck T, Sailer M (2012) Deutsche S3-Leitlinie: Rektovaginale Fistel. Ger Med Sci 10:DOC 15. https://doi.org/10.3205/000166
Naldini G (2011) Serious unconventional complications of surgery with stapler for haemorrhoidal prolapse and obstructed defaecation because fo rectocoele and rectal untussusception. Colorectal Dis 13:323–327
Martelucci J, Talento P, Carriero A (2011) Early complications after stapled transanal rectal resection performed using the Contour® Transtar™ device. Colorectal Dis 13:1428–1431
Matthiessen P, Hansson L, Sjödahl R, Rutegard J (2010) Anastomotic-vaginal fistula (AVF) after anterior resection of the rectum for cancer – occurrence and risk factors. Colorect Dis 12(4):351–357
Gagliardi G, Pesatori M, Altomare DF, Binda GA, Bottini C et al (2008) Results, outcome predictors and complications after stapled transanal rectal resection for obstructed defecation. Dis Colon Rectum 51:186–195
Gecim IE, Wolff BG, Pemberton JH, Devine RM, Dozios RR (2000) Does technique of anastomosis play any role in developing late perianal abscess or fistula? Dis Colon Rectum 43(9):1241–1245
Lolohea S, Lynch AC, Robertson GB, Frizelle FA (2005) Ileal pouch-anal anastomosis-vaginal fistula: a review. Dis Colon Rectum 48(9):1802–1810
Andreani SM, Dang HH, Grondona P, Khan AZ, Edwards DP (2007) Rectovaginal fistula in Crohn’s disease. Dis Colon Rectum 50:2215–2222
Narayanan P, Nobbenhuis M, Reynolds KM, Sahdev A, Reznek EH, Rockall AG (2009) Fistulas in malignant gynecologic disease: etiology, imaging, and management. Radiographics 29(4):1074–1084
Anderson JR, Spence AJ, Parks TG, Bond EB, Burrows BD (1984) Rectovaginal fistulae following radiation treatment for cervical carcinoma. Ulster Med J 53(1):84–87
Fürst A, Schmidbauer CH, Swol-Ben J, Ieselnieks I, Schwandner O, Agha A (2008) Gracilis transposition for repair of recurrent anovaginal and rectovaginal fistulas in Corhn’s disease. Int J Colorectal Dis 23:349–353
van der Hagen SJ, Soeters PB, Baeten CG, van Gemert WG (2001) Laparoscopic fistula excision and omentoplasty for high rectovaginal fistulas: a prospective study of 40 patients. Int J Colorectal Dis 26:1463–1467
Kröpil F, Raffel A, Renter MA, Schauer M, Rehders A, Eisenberger CF, Knoefel WT (2010) Individualisiserte und differenzierte Therapie von rektovaginalen Fisteln. Zentralbl Chir 135:307–311
Delancey JOL, Miller NF, Berger MB (2010) Surgical approach to postobstetrical perineal body defects (rectovaginal fistula and chronic third and fourth-degree lacerations). Clin Obstet Gynecol 53(1):134–144
Khanduja KS, Padmanabhan A, Kerner BA, Wise WE, Aguilar PS (1994) Reconstruction of rectovaginal fistula with sphincter disruption by combining rectal mucosal advancement flap and anal spincteroplasty. Dis Colon Rectum 42(11):1432–1437
Khanduja KS, Yamashita HJ, Wise WE, Aguilar PS, Hartmann EF (2007) Delayed repair of obstetric injuries of the anorectum and vagina. Dis Colon Rectum 37(4):344–349
Kraemer M, Kara D (2016) Laparoscopic surgery of benign entero-vesical or entero-vaginal fistulae. Int J Colorectal Dis 31:19–22
Prosst RL, Joos AK (2016) Short-term outcomes of a novel endoscopic clipping device for closure of the internal opening of 100 anorectal fistulas. Tech Coloproctol 20:753–758
Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40(5):373–383
Corte H, Maggiore L, Treton X, Lefevre JH, Ferron M, Panis Y (2015) Rectovaginal fistula: what is the optimal strategy? Annals Surg 5(262):855–861
Lowry AC, Thorson AG, Rothenberger DA, Goldberg SM (1988) Repair of simple rectovaginal fistulas. Dis Colon Rectum 31(9):676–678
Ryoo SB, Oh HK, Han EC, Kwon YH, Song I, Moon SH, Choe EK, Park KJ (2019) Outcomes of surgical treatments for rectovaginal fistula and prognostic factors for successful closure: a single-center tertiary hospital experience. Ann Surg Treat Res 97(3):149–156
Reisenauer Ch (2019) Operative Therapie bei postpartelen rektovaginalen Fisteln: die gynäkologische Sicht. Zntrealblatt Chir 144:380–386
Tian Y, Xu B, Yu G, Li Y, Liu H (2017) Comorbidity and the risk of anastomotic leak in Chinese patients with colorectal cancer undergoing colorectal surgery. Int J Colorectal Dis 32(7):947–953
Sparreboom CL, van Groningen JT, Lingsma HF, Wouters MWJM, Menon AG, Kleinrensink GJ, Jeekel J, Lange JF (2018) Dutch ColoRectal Audit group. Different risk factors for early and late colorectal anastomotic leakage in a nationwide audit. Dis Colon Rectum 61(11):1258–1266
Schlöricke W, Zimmermann M, Hoffmann M, Laubert T, Nolde J, Hildebrand P, Bruch HP, Bouchard R (2012) Chirurgische Versorgung und Prognose rektovaginaler Fisteln in Abhängigkeit ihrer Genese. Zentralblatt Chir 137:390–395
Hauch A, Ramamoorthy S, Zelhart M, Dobke M (2020) Refining approaches to surgical repair of rectovaginal fistulas. Annals Plastic Surg 84(4):250–256
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
Moritz Drefs: primary author of manuscript, acquisition of data, statistical analysis of data, and analysis and interpretation of results. Sebastian Schömer Cuenca: acquisition of data, analysis and interpretation of results, and critical review of manuscript. Ulrich Wirth: interpretation of data, interpretation of results, statistical counseling of study, and critical review of manuscript. Florian Kühn: conceptual review of study, interpretation of results, and critical review of manuscript. Maria Burian: conceptual review of study, interpretation of results, and critical review of manuscript. Jens Werner: conceptual review of study, critical review of statistical analysis, analysis and interpretation of results, and critical review of manuscript. Petra Zimmermann: conceptual development of study, review of statistical analysis, analysis and interpretation of results, critical review of manuscript, and possible impact on clinical management.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Preliminary results were presented at the annual meeting of the German Association for General and Visceral Surgery, 2020.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Drefs, M., Schömer Cuenca, S., Wirth, U. et al. Predictors of outcome for treatment of enterovaginal fistula. Int J Colorectal Dis 38, 187 (2023). https://doi.org/10.1007/s00384-023-04453-2
Accepted:
Published:
DOI: https://doi.org/10.1007/s00384-023-04453-2