Corresponding author: Koichiro Murakami, nagaokakyohp-mura@athena.ocn.ne.jp
DOI: 10.31662/jmaj.2025-0474
Received: October 7, 2025
Accepted: May 3, 2026
Advance Publication: July 17, 2026
Background: Post-obstetric rectovaginal fistula significantly impairs women’s quality of life, and the indication for diverting stoma creation remains controversial. Despite numerous international reports indicating that diverting the stoma is unnecessary for post-obstetric rectovaginal fistula, many institutions continue to perform stoma creation in clinical practice.
Methods: We retrospectively analyzed 46 cases of post-obstetric rectovaginal fistula treated with perineal body reconstruction over 3 years. Patient demographics, fistula characteristics, treatment history at other institutions, surgical outcomes, and postoperative course were evaluated. The primary outcome was the absence of fistula recurrence during a 12-month follow-up period.
Results: The median age was 34 years, the median body mass index was 21, the median fistula size was 2 mm, and the median duration from onset to surgery was 268 days. The majority of cases involved severe obstetric trauma with grade 4 perineal lacerations. Seven patients (15.2%) were recommended stoma creation at other institutions, and a total of 15 surgeries were performed at other institutions. Stable surgical outcomes were achieved with a median operative time of 130 minutes and a median blood loss of 250 mL. During the 12-month follow-up period, favorable results were obtained in 44 cases (95.7%), with one missed concomitant high fistula subsequently treated successfully via a transvaginal approach, and one suspected recurrence reported by telephone at 24 months (after completion of follow-up) without clinical confirmation.
Conclusions: Our institution’s original perineal body reconstruction technique achieved a high cure rate without requiring a diverting stoma creation. Typical post-obstetric rectovaginal fistulas develop at the lowest point where rectal pressure is highest, making perineal body reconstruction effective for the majority of cases. However, in patients with extensive perineal laceration scarring on digital rectal examination, preoperative fistulography should be considered to rule out concomitant high fistulas. Diverting stoma creation for post-obstetric rectovaginal fistula is generally unnecessary.
Key words: obstetric trauma, rectovaginal fistula, perineal body reconstruction, diverting stoma, fistula closure
Rectovaginal fistula significantly impairs women’s quality of life. Obstetric trauma, particularly sphincter injury following grade 3-4 perineal lacerations, is a major etiology because the close proximity of the vagina and rectum can lead to fistula formation. The indication for diverting stoma creation in treating this condition has been debated for many years.
International literature predominantly reports that diverting stoma creation is unnecessary for post-obstetric rectovaginal fistula. Pinto et al. reported achieving a 95% cure rate without requiring stoma creation in 113 cases. Similarly, Tsang et al. questioned the necessity of stoma creation and concluded that treatment is possible with appropriate surgical techniques without requiring a stoma.
However, in clinical practice in Japan, diverting stoma creation is proposed and performed for post-obstetric rectovaginal fistula at university hospitals and many high-volume centers. This indicates a significant discrepancy between literature-based evidence and actual clinical practice.
Our institution has established an original treatment strategy using perineal body reconstruction without requiring a diverting stoma for post-obstetric rectovaginal fistula, achieving favorable treatment outcomes. This study validates the efficacy of our treatment strategy through retrospective analysis of 46 cases and reconsiders the necessity of stoma creation.
We included 46 cases of post-obstetric rectovaginal fistula that underwent surgery at our institution between July 2022 and June 2025. All cases involved rectovaginal fistula resulting from perineal lacerations during delivery. Fistulas due to malignancy, inflammatory bowel disease, or radiation were excluded.
We conducted a retrospective analysis using medical records. The evaluation items included patient demographics (age, delivery mode, perineal laceration grade, and duration from onset to surgery), fistula characteristics (fistula size), treatment history at other institutions (recommendation for stoma creation, actual stoma creation, and number of surgeries for rectovaginal fistula), and intraoperative variables (operative time, blood loss, and number of perineal body sutures).
The primary outcome was the absence of fistula recurrence during the 12-month follow-up period. The 12-month follow-up was set as the standard observation period for all patients who completed follow-up. Secondary outcomes included postoperative complications (including wound infection, fecal incontinence, and chronic pain), changes in fecal incontinence severity (assessed by Wexner score), and improvement in sexual function.
All procedures were performed with patients in the jackknife (prone) position under saddle block anesthesia. Our institution’s original perineal body reconstruction was performed in all cases following these steps:
1. Transverse perineal incision and subcutaneous dissection (Figure 1)
2. Adequate dissection of rectal and vaginal walls, confirmation of fistula presence (Figure 2)
3. Excision of fistulous tract, including vaginal and rectal walls, when a fistula is confirmed (Figure 3)
4. Reliable closure of the rectal wall with vertical mattress sutures and repair of the vaginal wall with continuous sutures
5. Confirmation that bilateral perineal bodies are adequately dissected with good mobility, allowing midline crossing (Figure 4)
6. Three-dimensional repair of the perineal body using 3-0 absorbable sutures (Figure 5)
7. Vertical closure of the perineal area (Figure 6)
Primary fistula closure was performed in all cases without requiring stoma creation. No drains were placed.
Continuous variables are expressed as median (range), and categorical variables as number of cases (%). All statistical analyses were performed using R version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria).
This study was approved by the institutional review board of Tesseikai Neurosurgical Hospital (approval number: 2025-88) and was conducted in accordance with the ethical standards of the Declaration of Helsinki. Informed consent was obtained from all participants.
Patient demographics for the 46 cases are shown in Table 1. The median age was 34 years (26-62 years). Grade 4 perineal laceration was most common with 18 cases (39.1%), followed by no recorded grade in 17 cases (37.0%), grade 2 in 7 cases (15.2%), and grade 3 in 3 cases (6.5%). The median duration from onset to surgery was 268 days.
Table 1. Patient Demographics and Clinical Characteristics (n = 46).
| Parameter | Median (range) or number (%) |
|---|---|
| Age (years) | 34 (26-62) |
| Fistula size (mm) | 2 (0-20) |
| Duration from onset to surgery (days) | 269 (45-6,271) |
| Grade 4 perineal laceration | 18 (39.1) |
| Grade 3 perineal laceration | 3 (6.5) |
| Grade 2 perineal laceration | 7 (15.2) |
| Perineal laceration grade unknown | 17 (37.0) |
| Previous surgery at other institutions | 15 surgeries in 11 patients (23.9%) |
| Stoma recommendation from other institutions | 7 patients (15.2%) |
The median fistula size was 2 mm (0-20 mm). While relatively small fistulas were common, cases with large fistulas were also included.
The median operative time was 130 minutes (84-198 minutes), the median blood loss was 250 mL (55-991 mL), and the median number of sutures was 14 (8-18) (Table 2). No serious intraoperative complications occurred.
Table 2. Surgical Outcomes and Postoperative Results (n = 46).
| Parameter | Median (range) or number (%) |
|---|---|
| Operative time (minutes) | 130 (84-198) |
| Blood loss (mL) | 250 (55-991) |
| Number of sutures | 14 (8-18) |
| Postoperative complications | 10 (21.7%) |
| Recurrence | 2 (4.3%) |
| Follow-up period (months) | 12 (6-12) |
Postoperative complications occurred in 10 cases (21.7%). Complications included fecal incontinence in five cases, missed concomitant high fistula in one case, chronic pain in one case, and subcutaneous infection in two cases. Subcutaneous infections were treated with Seton drainage in one case and conservative antibiotic therapy in another. Regarding fecal incontinence, all affected patients had severe preoperative incontinence (Wexner score ≥12) that improved postoperatively.
During the 12-month follow-up period, favorable results were obtained in 44 cases (95.7%) with perineal body reconstruction alone. Among the two cases that did not achieve the primary outcome, one had a missed concomitant high fistula approximately 10 cm proximal to the anus (identified on postoperative fistulography), which was subsequently treated successfully via a transvaginal approach at 18 months. The other reported suspected recurrence by telephone at 24 months (after completion of the standard follow-up period) without clinical confirmation, as the patient was unable to visit due to residing in a distant area.
Sexual function improved in all cases where assessment was possible, with two patients achieving postoperative pregnancy and vaginal delivery, and three patients delivering via cesarean section.
This study demonstrates that our institution’s original perineal body reconstruction technique achieved a 95.7% success rate (44/46 cases) in treating post-obstetric rectovaginal fistula without requiring a diverting stoma during the 12-month follow-up period. The technique proved effective even in challenging cases, including those in which stoma creation was recommended at other institutions (7 cases) and those with multiple prior surgical failures (11 patients with 15 prior surgeries).
International literature is predominantly negative regarding the necessity of stoma creation for post-obstetric rectovaginal fistula. Pinto et al. reported a 95% cure rate without a stoma in 113 cases. Tsang and Rothenberger concluded that stoma creation is not routinely necessary. Hull et al. reported an 85% success rate in the non-stoma group, with no significant difference compared to the stoma group. Our results (95.7% success rate) are consistent with these reports.
Several surgical techniques have been described for rectovaginal fistula repair. The endorectal advancement flap (ERAF), first described by Rothenberger in 1982, has reported success rates of 65%-82%. While ERAF provides mucosal coverage, it does not address the underlying perineal body defect. Muscle flap repairs, such as the gracilis muscle flap, have reported success rates of approximately 72% but carry significant donor site morbidity. The transvaginal approach, as demonstrated in one of our cases, can be effective for high fistulas that are not accessible via the perineal route.
Our perineal body reconstruction technique offers several advantages: (1) it addresses both the fistula and the underlying anatomical defect; (2) it simultaneously improves sphincter function; (3) it restores normal perineal anatomy; and (4) it avoids the morbidity associated with stoma creation or muscle flap harvest.
Based on our experience, we propose the following rationale for patient selection and surgical technique. Typical post-obstetric rectovaginal fistulas develop at the lowest point of the rectovaginal septum, where rectal intraluminal pressure is highest during defecation and straining. At this location, the fistula is in close proximity to the disrupted perineal body, making perineal body reconstruction with robust tissue interposition an essential approach for the majority of post-obstetric fistulas. Conversely, for fistulas extending higher than the upper margin of the perineal body (approximately 3 cm from the anus), the rectal intraluminal pressure is significantly lower, and our experience suggests that fistula tract excision with careful suturing of the rectal and vaginal walls alone—without tissue interposition—is sufficient to achieve healing.
However, in rare cases with extensive perineal lacerations, concomitant high fistulas may develop in addition to the typical low fistula. Because most post-obstetric rectovaginal fistulas are located just above the dentate line and are typically not visualized on contrast studies, we had not routinely performed preoperative fistulography. The missed high fistula case taught us that this approach requires modification in selected patients. In that case, fistula tract excision and careful suturing of the rectal and vaginal walls without tissue interposition achieved complete healing at 18 months.
We now recommend that when digital rectal examination reveals extensive perineal laceration scarring, preoperative fistulography or other imaging studies (such as endoanal ultrasound or magnetic resonance imaging) should be actively performed to rule out concomitant high fistulas before proceeding with perineal body reconstruction alone. Patients with identified high fistulas should be counseled that alternative or additional surgical approaches may be necessary.
This study has several limitations. First, this is a single-center retrospective study without a control group. Second, one suspected recurrence was reported after completion of the 12-month follow-up period and could not be clinically confirmed due to the patient residing in a distant area. Although we could not confirm the actual cause, we hypothesize that a delayed local infection, such as a suture abscess, may have occurred beyond the follow-up period; if undrained, this could have led to perineal body dehiscence. Third, one case of missed high fistula highlights the importance of preoperative imaging in selected patients. Fourth, selection bias may exist. Fifth, the possibility of delayed local infections, such as suture abscess occurring beyond the standard 12-month follow-up period, suggests that extended follow-up of 18 or 24 months should be considered for patients who are able to attend regular follow-up visits. Future prospective comparative trials through multicenter collaboration are needed.
Our institution’s original perineal body reconstruction technique achieved a high cure rate of 95.7% in 46 cases of post-obstetric rectovaginal fistula without requiring a diverting stoma during the 12-month follow-up period. Typical post-obstetric rectovaginal fistulas develop at the lowest point where rectal pressure is highest, and perineal body reconstruction is effective for the majority of these low fistulas. However, in rare cases with extensive perineal lacerations, concomitant high fistulas may coexist. Therefore, when digital rectal examination reveals extensive perineal laceration scarring, preoperative fistulography or other imaging studies should be actively performed to rule out concomitant high fistulas. For high fistulas not accessible via the perineal approach, alternative techniques such as the transvaginal approach can achieve successful outcomes. Additionally, considering the possibility of delayed local infections beyond the standard follow-up period, extended follow-up of 18 or 24 months should be considered for patients who are able to attend regular visits.
Our study results, consistent with international literature, demonstrate that stoma creation for post-obstetric rectovaginal fistula is generally unnecessary. Radical treatment is possible with appropriate surgical techniques and careful patient selection, significantly contributing to improved patient quality of life.
We thank all patients who participated in this study and the medical staff who contributed to their care.
Conceptualization, Methodology, Surgical procedures, Data curation, Writing - original draft, Writing - review & editing, Project administration: Koichiro Murakami.
Methodology, Data analysis, Writing - review & editing, Validation: Yumi Nishimura.
Methodology, Investigation, Data collection, Writing - review & editing, Supervision: Tomoyuki Mizukuro.
None
This study was approved by Tesseikai Neurosurgical Hospital Institutional Review Board (Approval No. 2025-88).
All procedures were performed under saddle block anesthesia in the jackknife position. Written informed consent for publication of surgical images was obtained from all patients. Patient privacy was maintained through appropriate anonymization of identifying features.
Pinto RA, Peterson TV, Shawki S, et al. Are there predictors of outcome following rectovaginal fistula repair? Dis Colon Rectum. 2010;53(9):1240-7.
Tsang CBS, Rothenberger DA. Rectovaginal fistulas. Surg Clin North Am. 1997;77(1):95-114.
Hull TL, El-Gazzaz G, Gurland B, et al. Surgeons should not hesitate to perform episioproctotomy for rectovaginal fistula secondary to cryptoglandular or obstetrical origin. Dis Colon Rectum. 2011;54(1):54-9.
Maeda K, Wada N, Shida A. Treatment of rectovaginal fistula. J Anus Rectum Colon. 2023;7(2):52-62.
Dawes AJ, Jensen CC. Rectovaginal fistulas secondary to obstetrical injury. Clin Colon Rect Surg. 2021;34(1):28-39.
Keighley MRB. Enterovaginal fistula. Surgery of the anus, rectum and colon. London: W B Saunders; 1993. p. 2076-93.
Hashimoto K, Mizukuro T, Fujino M. Surgical management of rectovaginal fistula: current strategies and outcomes. J Jpn Soc Gynecol Surg. 2020;31:31-3. Japanese.
John OL, Norman FM, Mitchel BB. Surgical approaches to postobstetrical perineal body defects (rectovaginal fistula and chronic third- and fourth-degree). Clin Obstet Gynaecol. 2020;53(1):134-44.
Rothenberger DA, Christenson CE, Balcos EG, et al. Endorectal advancement flap for treatment of simple rectovaginal fistula. Dis Colon Rectum. 1982;25(4):297-300.
de Parades V, Dahmani Z, Blanchard P, et al. Endorectal advancement flap with muscular plication: a modified technique for rectovaginal fistula repair. Colorectal Dis. 2011;13(8):921-5.
Kobayashi H, Sugihara K. Successful management of rectovaginal fistula treated by endorectal advancement flap: report of two cases and literature review. Springerplus. 2015;4(1):21.
Wexner SD, Ruiz DE, Genua J, et al. Gracilis muscle interposition for the treatment of rectourethral, rectovaginal, and pouch-vaginal fistulas: results in 53 patients. Ann Surg. 2008;248(1):39-43.
Park SO, Hong KY, Park KJ, et al. Treatment of rectovaginal fistula with gracilis muscle flap transposition: long-term follow-up. Int J Colorectal Dis. 2017;32(7):1029-32.
Download PDF