Long-Term Complication: Colovaginal Fistula Nine Years after Laparoscopic Sacrocolpopexy

Abstract

Background:

Colovaginal fistula (CVF) is a rare, late complication of laparoscopic sacrocolpopexy that significantly impairs quality of life. We report a case occurring 9 years postoperatively to highlight its presentation and management.

Case:

A G3P3 woman in her seventies presented with yellow, foul-smelling vaginal discharge 9 years after laparoscopic total hysterectomy with sacrocolpopexy. Colonoscopy revealed a rectovaginal fistula; histology excluded diverticular disease. A multidisciplinary team performed laparoscopic removal of the eroded mesh and sigmoid resection with primary anastomosis. At 1-year follow-up the vagina was completely healed and the patient reported no further abnormal discharge.

Conclusion:

CVF may arise almost a decade after sacrocolpopexy. A high index of suspicion, prompt multidisciplinary surgery and long-term surveillance are essential for optimal outcomes.

INTRODUCTION

With the global population aging, the incidence of pelvic organ prolapse (POP) is steadily increasing. The estimated lifetime risk of symptomatic vaginal prolapse requiring surgical intervention is now up to 13%.¹ Sacrocolpopexy is one of the most commonly performed procedures for the management of apical POP, particularly following the U.S. Food and Drug Administration (FDA)’s discontinuation of mesh products for transvaginal repairs. This technique provides superior outcomes and is associated with relatively few complications.² Colovaginal fistula (CVF) is a rare but devastating complication of laparoscopic sacrocolpopexy, with incidence in the existing literature limited to isolated case reports. As laparoscopic and robotic-assisted laparoscopic approaches to sacrocolpopexy gain popularity, research into the prevention and management of CVF will become increasingly invaluable.³ In this report, we present a case of CVF diagnosed 9 years after laparoscopic sacrocolpopexy and describe the multidisciplinary approach used for its successful management.

CASE PRESENTATION

The patient is a G3P3 woman in her seventies who underwent laparoscopic total hysterectomy with sacrocolpopexy for POP 9 years prior at another institution. Her medical history was unremarkable. Notably, she had undergone transvaginal partial mesh removal approximately 1 year after the sacrocolpopexy due to vaginal mesh exposure, after which she experienced occasional mild vaginal discharge. She presented to our center with a chief concern of a 20-day history of yellow, foul-smelling vaginal discharge.

On physical examination, the patient was well-appearing, had a body mass index of 18.6 kg/m², and was normotensive. The external genitalia and the vaginal epithelium appeared atrophic and examination of the vaginal vault revealed purulent, bloody discharge without visible mesh exposure and fistula tract. On POP Quantification (POP-Q) examination, which assesses prolapse by measuring 6 defined vaginal points (anterior: Aa, Ba; posterior: Ap, Bp; apex: C, D) relative to the hymen during maximal straining, no evidence of anterior, apical, or posterior vaginal prolapse was identified (Table 1).

Table 1.

The Pelvic Organ Prolapse Quantification Staging System (cm)

Aa −3	Ba −3	C −7
Gh 5	Pb 2.5	TVL 7
Ap −3	Bp −3	D −

To establish a definitive diagnosis, contrast-enhanced pelvic magnetic resonance imaging (MRI) was performed (Figure 1). The study revealed a tubular tract arising from the superolateral aspect of the vaginal cuff and communicating with the upper rectum. The canal extended toward the right upper quadrant, where a fibrotic band was identified. Marked enhancement of the fistulous tract wall was observed after contrast administration, with mild mural thickening. These findings are consistent with a formed upper rectum-to-vaginal cuff fistula, accompanied by an adjacent adhesion superolateral to the fistulous track. The findings demonstrated the absence of a pelvic abscess and ruled out cervical or colorectal cancer as the underlying cause of the rectovaginal fistula in this patient.

Figure 1.

Contrast-enhanced pelvic MRI. Labels 1 to 3 (in red): 1—bladder, 2—vaginal stump, 3—upper rectum. The yellow arrow indicates a tubular communication between the upper right edge of the vaginal cuff and the upper rectum.

To distinguish between the presence of a vesicovaginal or rectovaginal fistula, a methylene blue dye test was performed. Instillation of methylene blue dye into the bladder showed no leakage from the vaginal vault. However, dye introduced into the rectum leaked into the vaginal vault, indicating a potential rectovaginal fistula.

To precisely determine the location of the fistula, enteroscopy was performed by the colorectal surgery team. During colonoscopy of the transverse colon, a 6°F catheter was inserted through the vaginal fistula, and methylene blue was injected. Exudation of the dye was observed at the distal sigmoid colon, identifying a fistula approximately 3 mm in diameter, surrounded by mucosal congestion and edema (Figure 2). These findings confirmed the diagnosis of a CVF.

Figure 2.

Fiber colonoscopy identifying the precise location of the colovaginal fistula. (A) The red arrow indicates the fistula in the distal sigmoid colon, surrounded by mucosal congestion and edema. (B) A 6F catheter is inserted through the vaginal fistula to delineate its tract. (C) Injection of methylene blue outlines the fistula, measuring approximately 3 mm in diameter.

SURGICAL PROCEDURE

After discussion and planning with the colorectal surgery team, laparoscopic exploration was performed. Upon introducing the laparoscope, we observed that the sigmoid colon was tightly adhered to the right vaginal stump. After lysis of adhesions and mobilization of the colon, a sigmoid colon fistula could be seen with a sinus tract connecting the sigmoid colon to the right vaginal apex. Mesh was identified at the area of the fistula between the sigmoid colon and vagina, and the surrounding tissues at the vaginal side were indurated and inflamed. The mesh appeared contracted into a cord-like structure under tension, extending from the vaginal apex along the right uterosacral ligament to the presacral area. Additionally, the right ureter was adherent to the right uterosacral ligament (Figure 3).

Figure 3.

Laparoscopic exploration findings. (A) Labels 1 to 5 (in yellow): 1—fistula, 2—bladder, 3—vaginal stump, 4—sigmoid colon, 5—right uterosacral ligament. (B) The yellow arrow indicates the fistula in the sigmoid colon. (C) The yellow arrow highlights the inflamed mesh located between the sigmoid colon and the vaginal stump. (D) Labels 1 and 2 (in yellow): 1—right ureter, 2—right uterosacral ligament.

To facilitate effective removal of the band of mesh, the right uterosacral ligament was carefully dissected away from the right ureter using sharp dissection. Using a cup-shaped uterine manipulator to lift the vaginal stump, because of the infection the mesh could be easily dissected from the vaginal wall, then the mesh was transected near the anterior aspect of the sacrum. Thorough irrigation of the pelvic and abdominal cavities using saline was performed, and the vaginal stump was closed with full-thickness suturing using 2/0 absorbable sutures, thereby sealing the vaginal fistula. Following closure of the vagina, the bladder was instilled with methylene blue and no staining of the tissues was seen. Cystoscopy confirmed the integrity of the bladder, with no evidence of injury.

Subsequently, a colorectal surgeon performed rectosigmoid resection with primary anastomosis near the site of the sigmoid fistula. The resected rectosigmoid segment measured approximately 11 cm in length, and histological analysis showed edema of the intestinal wall with fibrous tissue hyperplasia in the subserosal layer and minimal inflammatory exudate on the serosal surface. Chronic inflammatory cell infiltration was observed in the stroma. The fistula tissue was partially covered with squamous epithelium, displaying focal epithelial loss, granulation tissue proliferation, fibrinous exudate, proliferative fibrous tissue, and exposed muscularis propria of the intestinal wall, but pathological examination revealed no evidence of diverticulitis, diverticular disease, or Crohn’s disease (Figure 4). The patient experienced a smooth uneventful postoperative recovery; cefazolin was initiated 1 week postoperatively for therapeutic antimicrobial coverage, and she resumed a regular diet by postoperative day 7. Intravaginal estrogen ointment was applied for 3 months postoperatively. At a 1-year follow-up, on physical examination the vagina was fully healed and the patient was no longer complaining of any abnormal vaginal discharge. There was no recurrence of vaginal prolapse, and normal bowel movements, without any episodes of diarrhea or constipation.

Figure 4.

Pathological examination findings. (A) Gross appearance of the resected rectosigmoid segment. (B–D) Hematoxylin and eosin staining (H&E). Histological analysis showing edema of the intestinal wall with fibrous tissue hyperplasia in the subserosal layer and minimal inflammatory exudate on the serosal surface. Chronic inflammatory cell infiltration was observed in the stroma. The fistula tissue was partially covered with squamous epithelium, displaying focal epithelial loss, granulation tissue proliferation, fibrinous exudate, proliferative fibrous tissue, and exposed muscularis propria of the intestinal wall.

DISCUSSION

CVFs are relatively rare and even more rarely are associated with previous sacrocolpopexy. Acute andiverticulitis is the most common etiology and accounts for 79% of cases in the literature based on imaging and intraoperative findings.⁴ Interestingly, a history of hysterectomy is a significant risk factor for CVF formation in patients with diverticulitis. Studies report that 71%–95% of patients with CVF had undergone prior hysterectomy,^4–7 likely due to the absence of a protective uterus, which allows inflammation and infection to extend more directly to the vaginal wall. In this case, the patient who developed the complication had undergone a total hysterectomy with sacrocolpopexy at another center. Internationally, sacrocolpopexy is more commonly combined with supracervical hysterectomy or uterine suspension, which offers better long-term safety.⁸ One case report in the literature describes a spontaneous CVF, thought to be secondary to the gastrointestinal effects of systemic lupus erythematosus (SLE) in a patient with history of prior hysterectomy.⁹ Asymptomatic diverticular disease is another proposed patholophysiology for CVF formation.¹⁰

The incidence of CVF following sacrocolpopexy is unknown. However, there are only a handful of case reports published in the literature. A sigmoid-vaginal fistula occurring 4 years after laparoscopic sacrocolpopexy was reported in 2018 in a 66-year-old G3P3 woman with eosinophilic enteritis and colonic diverticula, suggesting that inflammatory bowel disease may contribute to mesh erosion.¹¹

Colovesical fistulas can be associated with various types of mesh augmented POP repairs. A comprehensive PubMed search using the terms “colovaginal fistula,” “rectal mesh erosion after sacrocolpopexy,” “colovaginal fistula sacrocolpopexy,” “rectovaginal fistulas sacrocolpopexy,” and “anovaginal fistulas sacrocolpopexy” revealed no additional cases since the 2018 report. However, searches from 2014 to 2024 using terms such as “pelvic mesh and fistula,” “ureterovaginal fistula and sacrocolpopexy,” and “vesicovaginal fistula and sacrocolpopexy” identified various genital tract fistulas associated with pelvic mesh. Among these, one case of CVF was attributed to anterior mesh-augmented pelvic floor repair,¹² while a ureterovaginal fistula¹³ and a vesicovaginal fistula¹⁴ were associated with sacrocolpopexy. Additionally, 2 cases of rectovaginal fistula^15,16 were linked to transvaginal mesh surgery, and 2 cases of urethrovaginal fistula^17,18 were associated with synthetic sling placement. Other reported fistulas, including vaginocutaneous,¹⁹ ischiorectal,²⁰ and cutaneous²¹ fistulas were all attributed to transvaginal mesh implantation. These findings emphasize the delayed presentation of genital tract fistulas after pelvic mesh surgery, underscoring the need for long-term follow-up in these patients. Therefore, standardized techniques for mesh placement should be employed intraoperatively, and long-term surveillance of these patients is mandatory to minimize late complications and prevent late-onset fistula formation.

Infection of the implanted mesh may result from direct contamination during surgery or ascending infection.²² In this case, the patient experienced vaginal mesh exposure, necessitating partial removal via the vaginal route. Postoperatively, the patient presented with abnormal vaginal discharge and severe mesh inflammation. The mesh may have then acted as a wick, providing a surface that facilitated bacterial migration from the vaginal vault into deeper planes, leading to sigmoid-vaginal apex adhesions and subsequent CVF. These findings suggest the mesh functioned as both an infectious nidus and a conduit for bacterial spread, with postoperative local inflammation contributing critically to fistula formation.

Complete excision of the infected mesh is challenging due to altered anatomy and scarring from prior sacrocolpopexy. However, complete mesh removal is desirable whenever feasible, as retention of the mesh may predispose to recurrence unless the infection is unequivocally localized to the distal segment.²³ In this case, neither preoperative imaging nor intraoperative endoscopic assessment identified a presacral abscess. The infection was localized to the vaginal cuff, without extension into the presacral space. Considering the rich vascularity of the sacral fixation site and the need to avoid major hemorrhagic complications, we opted to leave the mesh tail attached to the sacrum in situ.

CVF is the third most common type of surgical fistula requiring repair.²⁴ Its management necessitates a multidisciplinary approach and comprehensive diagnostic evaluation. Successful treatment significantly enhances the patient’s quality of life.²⁵ Rectosigmoid resection with primary anastomosis remains a safe and effective treatment option, with success rates of up to 95% in patients whom are healthy enough to undergo primary anastomosis.⁴ For patients unable to undergo surgical intervention, endoscopic management with over-the-scope (OTS) clips has shown promise, achieving long-term closure in 50% of cases.^26–28 In this patient, laparoscopic rectosigmoid resection with primary anastomosis was successfully performed, leading to complete resolution of symptoms.

The rising of mesh exposure following total hysterectomy and sacrocolpopexy increases over time, from 6.6% at 1 year to at least 10% at 5 years.²⁹ With the increasing adoption of total hysterectomy and sacrocolpopexy, CVF may become more common in the future. Therefore, long-term follow-up and careful evaluation of surgical materials and techniques are essential to mitigate risks and optimize patient outcomes.

CONCLUSIONS

CVF is a rare complication of sacrocolpopexy. It can be successfully managed through a multidisciplinary laparoscopic transabdomninal approach. Excision of the involved mesh with complete removal when feasible is necessary. Due to delayed presentation, long-term surveillance of patients should be considered to facilitate early detection and prevention of late-onset fistula formation.