CASE REPORT
Vesicosacrofistulization After Robotically Assisted Laparoscopic Sacrocolpopexy Mallika Anand, MD, Staci L. Tanouye, MD, and John B. Gebhart, MD, MS
Abstract: Diskitis after sacrocolpopexy for pelvic organ prolapse has been increasingly reported in the literature. We present a case of vesicosacrofistulization resulting in diskitis and osteomyelitis after robotically assisted laparoscopic sacrocolpopexy performed at an outside institution. A 70-year-old woman with uterovaginal prolapse and stress urinary incontinence underwent robotic supracervical hysterectomy with sacrocolpopexy and transobturator sling placement at an outside hospital. Postoperatively, she had recurrent urinary tract infections; by 3 months postoperatively, fevers and leg and back pain had developed. She was given a diagnosis of L5-S1 spondylodiskitis. After 3.5 weeks of intravenous antibiotic therapy failed, further evaluation revealed a fistulous tract to the sacrum. She was transferred to our institution and underwent sacrocolpopexy mesh removal, L5-S1 debridement, antibiotic treatment, and physical therapy. One year after this repair surgery, she has returned to her usual activities with no current symptoms of infection, prolapse, urinary incontinence, or back pain. Vesicosacrofistulization is a serious complication of sacrocolpopexy that can result in diskitis and osteomyelitis. Prevention involves avoiding placing mesh on the bladder and at the L5-S1 disk space during open or minimally invasive sacrocolpopexy. A high index of suspicion for diskitis, even several months after surgery, should be maintained to expedite evaluation. If fistulization of pelvic structures to the sacrum is suspected, a multidisciplinary evaluation and treatment approach should be considered to optimize patient care. Key Words: diskitis, mesh, osteomyelitis, sacral fistula, sacrocolpopexy (Female Pelvic Med Reconstr Surg 2014;20: 180Y183)
T
he recently published ‘‘Guidelines for privileging and credentialing physicians for sacrocolpopexy for pelvic organ prolapse’’ by the American Urogynecologic Society’s Guidelines Development Committee1 emphasized a requisite understanding of potential complications that could occur from sacrocolpopexy. Sacral diskitis, osteomyelitis, erosion into pelvic organs, fistula formation, and sinus tract formation were among the possible complications highlighted. We present the case of a patient with vesicosacrofistulization resulting in diskitis and osteomyelitis after robotically assisted laparoscopic sacrocolpopexy performed at an outside institution.
REPORT OF A CASE A 70-year-old woman with symptomatic uterovaginal prolapse and stress urinary incontinence had undergone robotically assisted laparoscopic supracervical hysterectomy, sacrocolpopexy, From the Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, MN. Reprints: John B. Gebhart, MD, MS, Department of Obstetrics and Gynecology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. E-mail: [email protected]. The authors have declared they have no conflicts of interest. Copyright * 2014 by Lippincott Williams & Wilkins DOI: 10.1097/SPV.0000000000000033
180
www.fpmrs.net
and placement of a transobturator sling with postoperative cystoscopy at an outside facility. Her medical history was significant for chronic back pain and polymyalgia rheumatica, and she was taking prednisone. Preoperatively, she had received 2 g of cefazolin for antibiotic prophylaxis. Her postoperative course was complicated by recurrent urinary tract infections (Escherichia coli and Enterococcus faecalis) that were unresponsive to outpatient therapy. Three months postoperatively, she became febrile with progressively worsening back and leg pain precluding ambulation and had experienced a 5.9-kg weight loss. She was admitted to her local hospital for further evaluation. Physical examination revealed a soft, nontender abdomen, left costovertebral angle tenderness, motor weakness and hyperreflexia of the left lower extremity, and a nontender vertebral spine. Laboratory analysis showed an increased erythrocyte sedimentation rate of 115 mm/h (reference range, 0Y29 mm/h), a white blood cell count of 5.7 109/L (reference range, 3.5Y10.5 109/L), a decreased hemoglobin value of 9.4 g/dL (reference range, 12.0-15.5 g/dL), and a decreased albumin value of 2.6 g/dL (reference range, 3.5Y5.0 g/dL). Urine culture revealed persistent E. faecalis infection. Computed tomography (CT) demonstrated L5-S1 spondylodiskitis and vertebral osteomyelitis. Suspicion for communication from the hysterectomy site to the prevertebral region was noted at that time. Magnetic resonance imaging (MRI) confirmed the spinal changes but did not demonstrate a fistulous tract (Fig. 1). Over several weeks, empiric broad-spectrum intravenous (IV) antibiotics were administered, initially with ampicillin and sulbactam, followed by vancomycin and piperacillin and tazobactam, then vancomycin and ertapenem for suspected mesh infection, without improvement. Blood cultures were negative. A second CT 3 1/2 weeks after starting the antibiotics demonstrated little improvement in the spondylodiskitis and vertebral osteomyelitis. She was referred to the Division of Gynecologic Surgery at our institution for further evaluation. Review of the outside CT images confirmed L5-S1 spondylodiskitis with a fistulous tract to the vagina (Fig. 2) with possible bladder involvement (Fig. 3). Vaginal speculum examination did not demonstrate erosion of mesh into the vagina or vesicovaginal fistula. Cystourethroscopy revealed mesh and suture erosion along the posterior bladder wall superior to the interureteric ridge. A urine culture was obtained. Given the findings of intravesical mesh and failure of conservative management, the plan was made to proceed with surgical mesh excision and vertebral debridement jointly with the Division of Orthopedic Surgery. Consultation with the Division of Infectious Diseases was obtained, and perioperative antibiotic management with continued broad-spectrum IV antibiotics, including antifungal coverage, was recommended. Bilateral external ureteral stents were placed cystoscopically to enable palpation of the ureters throughout the dissection. A midline laparotomy incision was made, the bowel packed, and colon mobilized. The retropubic space was entered, and an intentional cystotomy was made in the bladder dome and carried
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
Postsacrocolpopexy Vesicosacrofistulization
FIGURE 3. Computed tomographic image demonstrating inflammatory changes (arrow) possibly involving the posterior bladder wall.
FIGURE 1. Magnetic resonance imaging signal consistent with inflammation of the L5-S1 vertebral bodies, prevertebral soft tissue, and intervertebral disk space.
inferiorly to the level of the intravesical mesh, which was elliptically excised (Fig. 4). The bladder was mobilized off the vagina, and ULTRAPRO mesh (Ethicon, Somerville, NJ) and 2-0 GORE-TEX sutures (Gore Medical, Newark, DE) were freed from the anterior and posterior vagina and cervical stump. The pelvic peritoneum overlying the mesh was opened, revealing densely adherent mesh along the right pelvic sidewall. A right ureterolysis was performed and the mesh freed to the lumbosacral junction. A small portion of the mesh with 2-0 GORE-TEX sutures at the sacral attachment point was temporarily left in place for ease of identification by the orthopedic surgery team.
The orthopedic surgery team then performed anterior L5-S1 diskectomy and sharp debridement of the L5-S1 disk space, obtained cultures, and obtained biopsy samples of the L5 and S1 vertebrae. The bladder was closed in 2 layers and reperitonealized. A deep drain was placed in the right pelvis. The abdomen was copiously irrigated and closed, the stents removed, and a transurethral Foley catheter placed. Urinalysis and culture revealed bacterial (Leuconostoc lactis and Lactobacillus) and fungal (Candida glabrata) infections. Vertebral spine cultures revealed C. glabrata infection. Infectious disease specialists recommended continuation of antibacterial IV antibiotics for 4 weeks in addition to a 2-week oral course; IV antifungal antibiotics were continued for a total of 12 weeks postoperatively. She was ambulating postoperatively and discharged to an inpatient facility close to her home, where she continued to receive IV antibiotics. She was discharged home 1 month later and worked with a physical therapist for 2 weeks. Follow-up plain radiography at the time and 1 year after the repair surgery revealed reassuring postoperative changes of her spine. As of 1 year after the mesh excision, she is without vertebral pain and is ambulating reasonably well with the exception of a chronic hip condition. She had 3 urinary tract infections that were treated over the past year, after which she was started on suppressive therapy with oral nitrofurantoin by her local physician, without further symptoms. At present, she is free of symptoms of infection, urinary incontinence, bladder spasms, or pelvic organ prolapse.
COMMENT
FIGURE 2. Computed tomographic image demonstrating a thick-walled, air-filled sinus tract (arrows) extending from the vaginal cuff to the lumbosacral junction. * 2014 Lippincott Williams & Wilkins
Fistulization to the sacrum after sacrocolpopexy for pelvic organ prolapse is an infrequent complication but is serious because it can result in diskitis and osteomyelitis. Fistulization between the bladder and sacrum has been observed after pelvic irradiation for gynecologic2 and colorectal3 cancers; to our knowledge, no other cases of vesicosacrofistulization have been reported after sacrocolpopexy for pelvic organ prolapse. Previously reported cases of fistulization or sinus tract formation to the sacrum resulting in diskitis and/or osteomyelitis after surgery for pelvic organ prolapse have consisted of vaginosacral fistulae occurring after open abdominal sacrocolpopexy,4,5 laparoscopic sacrocervicopexy,6 or robotically assisted laparoscopic sacrocolpopexy.7 Presenting symptoms have included fever,5 back pain,4Y7 pelvic pain,4,7 leg pain6,7 or weakness,5,7 falling,7 fatigue,5 and low appetite.5 Although www.fpmrs.net
181
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Anand et al
Female Pelvic Medicine & Reconstructive Surgery
FIGURE 4. Excised sacrocolpopexy mesh. A, Anterior attachment with excised portion of bladder; P, posterior vaginal attachment; S, sacral attachment.
some patients with vaginosacral fistulae also experienced vaginal discharge5,7 or spotting,4 our patient had refractory urinary tract infections, consistent with her vesicosacrofistulization. In one case of sacrocolpopexy-associated bladder mesh erosion without fistulization, the presenting symptoms consisted of hematuria and urinary stone formation.8 The time course of presentation in all cases was highly variable, ranging from several weeks to 14 months.4Y6,8 Speculum examination may reveal mesh eroding into the vagina.4,5,7 For our patient, the result of intraoperative cystoscopy was negative at the time of the initial surgery, but diagnostic cystoscopy revealed mesh eroding into the bladder 4 months postoperatively. There should be a low threshold to pursue diagnostic cystoscopy in patients who have undergone surgery for prolapse or urinary incontinence.9 Vaginal cultures may reveal candidiasis,6 which is nonspecific; blood cultures should be obtained.10 There was no leukocytosis in any of the previously reported cases, but a mild leukocytosis may be present in one third of patients.11 As in our case, the erythrocyte sedimentation rate can be increased6; both erythrocyte sedimentation rate and C-reactive protein serve as markers to aid in diagnosis and treatment response.10 Computed tomography or MRI may demonstrate L5-S1 diskitis,6,7 osteomyelitis,5,7 epidural abscess,5Y7 and the fistulous tract.5Y7 In our case, CT performed at an outside institution revealed the fistulous tract extending to the pelvic viscera, whereas MRI revealed only diskitis and osteomyelitis. In general, MRI is the most sensitive diagnostic imaging modality for diskitis/osteomyelitis.10,11 whereas CT may be useful for preoperative planning and biopsy.11 Biopsy of the L5-S1 disk was performed by an interventional radiologist in one case6 and is recommended to tailor antibiotic therapy when blood cultures are negative.10,11
182
www.fpmrs.net
&
Volume 20, Number 3, May/June 2014
Once a fistula to the sacrum has been diagnosed, an aggressive, multidisciplinary treatment plan involving specialists in gynecologic surgery, infectious disease, orthopedic surgery, and possibly neurosurgery should be undertaken. In 4 of 5 cases, initial conservative management with broad-spectrum IV antibiotics failed,5Y7 and complete resection of the sacrocolpopexy mesh was ultimately necessary in all of the cases.4Y7 In general, surgical treatment for diskitis is warranted in the setting of failed conservative (ie, antibiotic) therapy, worsening symptoms or deformity, an epidural abscess with mass effect or septic embolization, or significant bony involvement.11 Sacral debridement by an orthopedic surgeon may be necessary depending on the degree of vertebral involvement; if the infection extends beyond the posterior annulus into the spinal canal, neurosurgery may be necessary. Even in cases of isolated intravesical mesh erosion after sacrocolpopexy in which fistulization has not occurred, open, complete excision of transvesical mesh is generally recommended,8,9 although cystoscopic mesh excision may be attempted.8 Postoperatively, appropriate antibiotic therapy should be continued for several weeks.10,11 Long-term follow-up is essential to ensure complete recovery. Fistulization to the sacrum is hypothesized to occur from mesh erosion resulting in ascending infection with epidural abscess formation.6 In the case of our patient, intravesical mesh erosion was present, but no epidural abscess was observed at the time of imaging. Minimizing risk factors for mesh and suture erosion distally and diskitis proximally might reduce the risk of vesicosacral fistulization. Smoking is the greatest demographic risk factor for mesh erosion after sacrocolpopexy.12 The use of expanded polytetrafluoroethylene mesh also increases the risk of mesh erosion.12 Data have been controversial or incomplete in determining whether concurrent total or supracervical hysterectomy is a risk factor when performed with open or minimally invasive sacrocolpopexy.13 General risk factors for postoperative diskitis include direct entry into the disk during the index surgery, prior radiotherapy, diabetes mellitus, immunosuppression, corticosteroid use, malnutrition, renal or liver disease, increased operative time, and lack of adherence to sterile technique.10,11 Long-term prednisone use in our patient may have put her at risk for diskitis and contributed to a delay in presentation. The single greatest risk factor for sacrocolpopexyassociated diskitis is thought to be direct seeding of the L5-S1 disk space by placement of the sacrocolpopexy stitch into the disk.14,15 A recent study by Abernethy et al14 suggests that placement of the stitch 5 mm inferior to the bony prominence of S1, or true sacral promontory, should avoid the disk space. Good et al15 demonstrated that the visual sacral promontory, or most acute angle at the lumbosacral junction, represents the angle of descent between the L5 and S1 vertebrae, and, therefore, the intervertebral disk. This suggests that avoiding suture placement into the L5-S1 intervertebral disk may be challenging without haptic feedback to ensure placement over a bony structure and to control the depth of suture placement.15 These considerations must be kept in mind as robotically assisted laparoscopic sacrocolpopexy becomes increasingly performed for pelvic organ prolapse.
ACKNOWLEDGMENT The authors thank Mark A. Nathan, MD, Department of Radiology, Mayo Clinic, Rochester, MN, for his assistance in selecting the radiographic images. * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
REFERENCES 1. American Urogynecologic Society’s Guidelines Development Committee. Guidelines for privileging and credentialing physicians for sacrocolpopexy for pelvic organ prolapse. Female Pelvic Med Reconstr Surg 2013;19(2):62Y65. 2. Lentz SS, Homesley HD. Radiation-induced vesicosacral fistula: treatment with continent urinary diversion. Gynecol Oncol 1995;58(2):278Y280. 3. Tonus C, Kolotas C, Appel P, et al. Colorectales CarcinomVBehandlungsalternative durch intraoperative Radiotherapie mittels Afterloading-flab-Technik? [in German]. Chirurg 1999;70(1):43Y47. 4. Unger JB. A persistent sinus tract from the vagina to the sacrum after treatment of mesh erosion by partial removal of a GORE-TEX soft tissue patch. Am J Obstet Gynecol 1999;181(3):762Y763.
Postsacrocolpopexy Vesicosacrofistulization
8. Patsner B. Mesh erosion into the bladder after abdominal sacral colpopexy. Obstet Gynecol 2000;95(6 Pt 2):1029. 9. Frenkl TL, Rackley RR, Vasavada SP, et al. Management of iatrogenic foreign bodies of the bladder and urethra following pelvic floor surgery. Neurourol Urodyn 2008;27(6):491Y495. 10. Cheung WY, Luk KD. Pyogenic spondylitis. Int Orthop 2012;36(2):397Y404. Epub 2011 Oct 28. 11. Skaf GS, Domloj NT, Fehlings MG, et al. Pyogenic spondylodiscitis: an overview. J Infect Public Health 2010;3(1):5Y16. Epub 2010 Feb 19. 12. Cundiff GW, Varner E, Visco AG, et al.; Pelvic Floor Disorders Network. Risk factors for mesh/suture erosion following sacral colpopexy. Am J Obstet Gynecol 2008;199(6):688.e1Y688.e5. Epub 2008 Oct 31.
5. Hart SR, Weiser EB. Abdominal sacral colpopexy mesh erosion resulting in a sinus tract formation and sacral abscess. Obstet Gynecol 2004;103(5 Pt 2):1037Y1040.
13. Matthews CA, Carroll A, Hill A, et al. Prospective evaluation of surgical outcomes of robot-assisted sacrocolpopexy and sacrocervicopexy for the management of apical pelvic support defects. South Med J 2012;105:274Y278.
6. Downing KT. Vertebral osteomyelitis and epidural abscess after laparoscopic uterus-preserving cervicosacropexy. J Minim Invasive Gynecol 2008;15(3):370Y372.
14. Abernethy M, Vasquez E, Kenton K, et al. Where do we place the sacrocolpopexy stitch? A magnetic resonance imaging investigation. Female Pelvic Med Reconstr Surg 2013;19(1):31Y33.
7. Muffly TM, Diwadkar GB, Paraiso MF. Lumbosacral osteomyelitis after robot-assisted total laparoscopic hysterectomy and sacral colpopexy. Int Urogynecol J 2010;21(12):1569Y1571. Epub 2010 Jun 8.
15. Good MM, Abele TA, Balgobin S, et al. Preventing L5-S1 discitis associated with sacrocolpopexy. Obstet Gynecol 2013;121(2 Pt 1):285Y290.
* 2014 Lippincott Williams & Wilkins
www.fpmrs.net
183
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Vesicosacrofistulization After Robotically Assisted Laparoscopic Sacrocolpopexy Mallika Anand, MD, Staci L. Tanouye, MD, and John B. Gebhart, MD, MS
Abstract: Diskitis after sacrocolpopexy for pelvic organ prolapse has been increasingly reported in the literature. We present a case of vesicosacrofistulization resulting in diskitis and osteomyelitis after robotically assisted laparoscopic sacrocolpopexy performed at an outside institution. A 70-year-old woman with uterovaginal prolapse and stress urinary incontinence underwent robotic supracervical hysterectomy with sacrocolpopexy and transobturator sling placement at an outside hospital. Postoperatively, she had recurrent urinary tract infections; by 3 months postoperatively, fevers and leg and back pain had developed. She was given a diagnosis of L5-S1 spondylodiskitis. After 3.5 weeks of intravenous antibiotic therapy failed, further evaluation revealed a fistulous tract to the sacrum. She was transferred to our institution and underwent sacrocolpopexy mesh removal, L5-S1 debridement, antibiotic treatment, and physical therapy. One year after this repair surgery, she has returned to her usual activities with no current symptoms of infection, prolapse, urinary incontinence, or back pain. Vesicosacrofistulization is a serious complication of sacrocolpopexy that can result in diskitis and osteomyelitis. Prevention involves avoiding placing mesh on the bladder and at the L5-S1 disk space during open or minimally invasive sacrocolpopexy. A high index of suspicion for diskitis, even several months after surgery, should be maintained to expedite evaluation. If fistulization of pelvic structures to the sacrum is suspected, a multidisciplinary evaluation and treatment approach should be considered to optimize patient care. Key Words: diskitis, mesh, osteomyelitis, sacral fistula, sacrocolpopexy (Female Pelvic Med Reconstr Surg 2014;20: 180Y183)
T
he recently published ‘‘Guidelines for privileging and credentialing physicians for sacrocolpopexy for pelvic organ prolapse’’ by the American Urogynecologic Society’s Guidelines Development Committee1 emphasized a requisite understanding of potential complications that could occur from sacrocolpopexy. Sacral diskitis, osteomyelitis, erosion into pelvic organs, fistula formation, and sinus tract formation were among the possible complications highlighted. We present the case of a patient with vesicosacrofistulization resulting in diskitis and osteomyelitis after robotically assisted laparoscopic sacrocolpopexy performed at an outside institution.
REPORT OF A CASE A 70-year-old woman with symptomatic uterovaginal prolapse and stress urinary incontinence had undergone robotically assisted laparoscopic supracervical hysterectomy, sacrocolpopexy, From the Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, MN. Reprints: John B. Gebhart, MD, MS, Department of Obstetrics and Gynecology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. E-mail: [email protected]. The authors have declared they have no conflicts of interest. Copyright * 2014 by Lippincott Williams & Wilkins DOI: 10.1097/SPV.0000000000000033
180
www.fpmrs.net
and placement of a transobturator sling with postoperative cystoscopy at an outside facility. Her medical history was significant for chronic back pain and polymyalgia rheumatica, and she was taking prednisone. Preoperatively, she had received 2 g of cefazolin for antibiotic prophylaxis. Her postoperative course was complicated by recurrent urinary tract infections (Escherichia coli and Enterococcus faecalis) that were unresponsive to outpatient therapy. Three months postoperatively, she became febrile with progressively worsening back and leg pain precluding ambulation and had experienced a 5.9-kg weight loss. She was admitted to her local hospital for further evaluation. Physical examination revealed a soft, nontender abdomen, left costovertebral angle tenderness, motor weakness and hyperreflexia of the left lower extremity, and a nontender vertebral spine. Laboratory analysis showed an increased erythrocyte sedimentation rate of 115 mm/h (reference range, 0Y29 mm/h), a white blood cell count of 5.7 109/L (reference range, 3.5Y10.5 109/L), a decreased hemoglobin value of 9.4 g/dL (reference range, 12.0-15.5 g/dL), and a decreased albumin value of 2.6 g/dL (reference range, 3.5Y5.0 g/dL). Urine culture revealed persistent E. faecalis infection. Computed tomography (CT) demonstrated L5-S1 spondylodiskitis and vertebral osteomyelitis. Suspicion for communication from the hysterectomy site to the prevertebral region was noted at that time. Magnetic resonance imaging (MRI) confirmed the spinal changes but did not demonstrate a fistulous tract (Fig. 1). Over several weeks, empiric broad-spectrum intravenous (IV) antibiotics were administered, initially with ampicillin and sulbactam, followed by vancomycin and piperacillin and tazobactam, then vancomycin and ertapenem for suspected mesh infection, without improvement. Blood cultures were negative. A second CT 3 1/2 weeks after starting the antibiotics demonstrated little improvement in the spondylodiskitis and vertebral osteomyelitis. She was referred to the Division of Gynecologic Surgery at our institution for further evaluation. Review of the outside CT images confirmed L5-S1 spondylodiskitis with a fistulous tract to the vagina (Fig. 2) with possible bladder involvement (Fig. 3). Vaginal speculum examination did not demonstrate erosion of mesh into the vagina or vesicovaginal fistula. Cystourethroscopy revealed mesh and suture erosion along the posterior bladder wall superior to the interureteric ridge. A urine culture was obtained. Given the findings of intravesical mesh and failure of conservative management, the plan was made to proceed with surgical mesh excision and vertebral debridement jointly with the Division of Orthopedic Surgery. Consultation with the Division of Infectious Diseases was obtained, and perioperative antibiotic management with continued broad-spectrum IV antibiotics, including antifungal coverage, was recommended. Bilateral external ureteral stents were placed cystoscopically to enable palpation of the ureters throughout the dissection. A midline laparotomy incision was made, the bowel packed, and colon mobilized. The retropubic space was entered, and an intentional cystotomy was made in the bladder dome and carried
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
Postsacrocolpopexy Vesicosacrofistulization
FIGURE 3. Computed tomographic image demonstrating inflammatory changes (arrow) possibly involving the posterior bladder wall.
FIGURE 1. Magnetic resonance imaging signal consistent with inflammation of the L5-S1 vertebral bodies, prevertebral soft tissue, and intervertebral disk space.
inferiorly to the level of the intravesical mesh, which was elliptically excised (Fig. 4). The bladder was mobilized off the vagina, and ULTRAPRO mesh (Ethicon, Somerville, NJ) and 2-0 GORE-TEX sutures (Gore Medical, Newark, DE) were freed from the anterior and posterior vagina and cervical stump. The pelvic peritoneum overlying the mesh was opened, revealing densely adherent mesh along the right pelvic sidewall. A right ureterolysis was performed and the mesh freed to the lumbosacral junction. A small portion of the mesh with 2-0 GORE-TEX sutures at the sacral attachment point was temporarily left in place for ease of identification by the orthopedic surgery team.
The orthopedic surgery team then performed anterior L5-S1 diskectomy and sharp debridement of the L5-S1 disk space, obtained cultures, and obtained biopsy samples of the L5 and S1 vertebrae. The bladder was closed in 2 layers and reperitonealized. A deep drain was placed in the right pelvis. The abdomen was copiously irrigated and closed, the stents removed, and a transurethral Foley catheter placed. Urinalysis and culture revealed bacterial (Leuconostoc lactis and Lactobacillus) and fungal (Candida glabrata) infections. Vertebral spine cultures revealed C. glabrata infection. Infectious disease specialists recommended continuation of antibacterial IV antibiotics for 4 weeks in addition to a 2-week oral course; IV antifungal antibiotics were continued for a total of 12 weeks postoperatively. She was ambulating postoperatively and discharged to an inpatient facility close to her home, where she continued to receive IV antibiotics. She was discharged home 1 month later and worked with a physical therapist for 2 weeks. Follow-up plain radiography at the time and 1 year after the repair surgery revealed reassuring postoperative changes of her spine. As of 1 year after the mesh excision, she is without vertebral pain and is ambulating reasonably well with the exception of a chronic hip condition. She had 3 urinary tract infections that were treated over the past year, after which she was started on suppressive therapy with oral nitrofurantoin by her local physician, without further symptoms. At present, she is free of symptoms of infection, urinary incontinence, bladder spasms, or pelvic organ prolapse.
COMMENT
FIGURE 2. Computed tomographic image demonstrating a thick-walled, air-filled sinus tract (arrows) extending from the vaginal cuff to the lumbosacral junction. * 2014 Lippincott Williams & Wilkins
Fistulization to the sacrum after sacrocolpopexy for pelvic organ prolapse is an infrequent complication but is serious because it can result in diskitis and osteomyelitis. Fistulization between the bladder and sacrum has been observed after pelvic irradiation for gynecologic2 and colorectal3 cancers; to our knowledge, no other cases of vesicosacrofistulization have been reported after sacrocolpopexy for pelvic organ prolapse. Previously reported cases of fistulization or sinus tract formation to the sacrum resulting in diskitis and/or osteomyelitis after surgery for pelvic organ prolapse have consisted of vaginosacral fistulae occurring after open abdominal sacrocolpopexy,4,5 laparoscopic sacrocervicopexy,6 or robotically assisted laparoscopic sacrocolpopexy.7 Presenting symptoms have included fever,5 back pain,4Y7 pelvic pain,4,7 leg pain6,7 or weakness,5,7 falling,7 fatigue,5 and low appetite.5 Although www.fpmrs.net
181
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Anand et al
Female Pelvic Medicine & Reconstructive Surgery
FIGURE 4. Excised sacrocolpopexy mesh. A, Anterior attachment with excised portion of bladder; P, posterior vaginal attachment; S, sacral attachment.
some patients with vaginosacral fistulae also experienced vaginal discharge5,7 or spotting,4 our patient had refractory urinary tract infections, consistent with her vesicosacrofistulization. In one case of sacrocolpopexy-associated bladder mesh erosion without fistulization, the presenting symptoms consisted of hematuria and urinary stone formation.8 The time course of presentation in all cases was highly variable, ranging from several weeks to 14 months.4Y6,8 Speculum examination may reveal mesh eroding into the vagina.4,5,7 For our patient, the result of intraoperative cystoscopy was negative at the time of the initial surgery, but diagnostic cystoscopy revealed mesh eroding into the bladder 4 months postoperatively. There should be a low threshold to pursue diagnostic cystoscopy in patients who have undergone surgery for prolapse or urinary incontinence.9 Vaginal cultures may reveal candidiasis,6 which is nonspecific; blood cultures should be obtained.10 There was no leukocytosis in any of the previously reported cases, but a mild leukocytosis may be present in one third of patients.11 As in our case, the erythrocyte sedimentation rate can be increased6; both erythrocyte sedimentation rate and C-reactive protein serve as markers to aid in diagnosis and treatment response.10 Computed tomography or MRI may demonstrate L5-S1 diskitis,6,7 osteomyelitis,5,7 epidural abscess,5Y7 and the fistulous tract.5Y7 In our case, CT performed at an outside institution revealed the fistulous tract extending to the pelvic viscera, whereas MRI revealed only diskitis and osteomyelitis. In general, MRI is the most sensitive diagnostic imaging modality for diskitis/osteomyelitis.10,11 whereas CT may be useful for preoperative planning and biopsy.11 Biopsy of the L5-S1 disk was performed by an interventional radiologist in one case6 and is recommended to tailor antibiotic therapy when blood cultures are negative.10,11
182
www.fpmrs.net
&
Volume 20, Number 3, May/June 2014
Once a fistula to the sacrum has been diagnosed, an aggressive, multidisciplinary treatment plan involving specialists in gynecologic surgery, infectious disease, orthopedic surgery, and possibly neurosurgery should be undertaken. In 4 of 5 cases, initial conservative management with broad-spectrum IV antibiotics failed,5Y7 and complete resection of the sacrocolpopexy mesh was ultimately necessary in all of the cases.4Y7 In general, surgical treatment for diskitis is warranted in the setting of failed conservative (ie, antibiotic) therapy, worsening symptoms or deformity, an epidural abscess with mass effect or septic embolization, or significant bony involvement.11 Sacral debridement by an orthopedic surgeon may be necessary depending on the degree of vertebral involvement; if the infection extends beyond the posterior annulus into the spinal canal, neurosurgery may be necessary. Even in cases of isolated intravesical mesh erosion after sacrocolpopexy in which fistulization has not occurred, open, complete excision of transvesical mesh is generally recommended,8,9 although cystoscopic mesh excision may be attempted.8 Postoperatively, appropriate antibiotic therapy should be continued for several weeks.10,11 Long-term follow-up is essential to ensure complete recovery. Fistulization to the sacrum is hypothesized to occur from mesh erosion resulting in ascending infection with epidural abscess formation.6 In the case of our patient, intravesical mesh erosion was present, but no epidural abscess was observed at the time of imaging. Minimizing risk factors for mesh and suture erosion distally and diskitis proximally might reduce the risk of vesicosacral fistulization. Smoking is the greatest demographic risk factor for mesh erosion after sacrocolpopexy.12 The use of expanded polytetrafluoroethylene mesh also increases the risk of mesh erosion.12 Data have been controversial or incomplete in determining whether concurrent total or supracervical hysterectomy is a risk factor when performed with open or minimally invasive sacrocolpopexy.13 General risk factors for postoperative diskitis include direct entry into the disk during the index surgery, prior radiotherapy, diabetes mellitus, immunosuppression, corticosteroid use, malnutrition, renal or liver disease, increased operative time, and lack of adherence to sterile technique.10,11 Long-term prednisone use in our patient may have put her at risk for diskitis and contributed to a delay in presentation. The single greatest risk factor for sacrocolpopexyassociated diskitis is thought to be direct seeding of the L5-S1 disk space by placement of the sacrocolpopexy stitch into the disk.14,15 A recent study by Abernethy et al14 suggests that placement of the stitch 5 mm inferior to the bony prominence of S1, or true sacral promontory, should avoid the disk space. Good et al15 demonstrated that the visual sacral promontory, or most acute angle at the lumbosacral junction, represents the angle of descent between the L5 and S1 vertebrae, and, therefore, the intervertebral disk. This suggests that avoiding suture placement into the L5-S1 intervertebral disk may be challenging without haptic feedback to ensure placement over a bony structure and to control the depth of suture placement.15 These considerations must be kept in mind as robotically assisted laparoscopic sacrocolpopexy becomes increasingly performed for pelvic organ prolapse.
ACKNOWLEDGMENT The authors thank Mark A. Nathan, MD, Department of Radiology, Mayo Clinic, Rochester, MN, for his assistance in selecting the radiographic images. * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Female Pelvic Medicine & Reconstructive Surgery
&
Volume 20, Number 3, May/June 2014
REFERENCES 1. American Urogynecologic Society’s Guidelines Development Committee. Guidelines for privileging and credentialing physicians for sacrocolpopexy for pelvic organ prolapse. Female Pelvic Med Reconstr Surg 2013;19(2):62Y65. 2. Lentz SS, Homesley HD. Radiation-induced vesicosacral fistula: treatment with continent urinary diversion. Gynecol Oncol 1995;58(2):278Y280. 3. Tonus C, Kolotas C, Appel P, et al. Colorectales CarcinomVBehandlungsalternative durch intraoperative Radiotherapie mittels Afterloading-flab-Technik? [in German]. Chirurg 1999;70(1):43Y47. 4. Unger JB. A persistent sinus tract from the vagina to the sacrum after treatment of mesh erosion by partial removal of a GORE-TEX soft tissue patch. Am J Obstet Gynecol 1999;181(3):762Y763.
Postsacrocolpopexy Vesicosacrofistulization
8. Patsner B. Mesh erosion into the bladder after abdominal sacral colpopexy. Obstet Gynecol 2000;95(6 Pt 2):1029. 9. Frenkl TL, Rackley RR, Vasavada SP, et al. Management of iatrogenic foreign bodies of the bladder and urethra following pelvic floor surgery. Neurourol Urodyn 2008;27(6):491Y495. 10. Cheung WY, Luk KD. Pyogenic spondylitis. Int Orthop 2012;36(2):397Y404. Epub 2011 Oct 28. 11. Skaf GS, Domloj NT, Fehlings MG, et al. Pyogenic spondylodiscitis: an overview. J Infect Public Health 2010;3(1):5Y16. Epub 2010 Feb 19. 12. Cundiff GW, Varner E, Visco AG, et al.; Pelvic Floor Disorders Network. Risk factors for mesh/suture erosion following sacral colpopexy. Am J Obstet Gynecol 2008;199(6):688.e1Y688.e5. Epub 2008 Oct 31.
5. Hart SR, Weiser EB. Abdominal sacral colpopexy mesh erosion resulting in a sinus tract formation and sacral abscess. Obstet Gynecol 2004;103(5 Pt 2):1037Y1040.
13. Matthews CA, Carroll A, Hill A, et al. Prospective evaluation of surgical outcomes of robot-assisted sacrocolpopexy and sacrocervicopexy for the management of apical pelvic support defects. South Med J 2012;105:274Y278.
6. Downing KT. Vertebral osteomyelitis and epidural abscess after laparoscopic uterus-preserving cervicosacropexy. J Minim Invasive Gynecol 2008;15(3):370Y372.
14. Abernethy M, Vasquez E, Kenton K, et al. Where do we place the sacrocolpopexy stitch? A magnetic resonance imaging investigation. Female Pelvic Med Reconstr Surg 2013;19(1):31Y33.
7. Muffly TM, Diwadkar GB, Paraiso MF. Lumbosacral osteomyelitis after robot-assisted total laparoscopic hysterectomy and sacral colpopexy. Int Urogynecol J 2010;21(12):1569Y1571. Epub 2010 Jun 8.
15. Good MM, Abele TA, Balgobin S, et al. Preventing L5-S1 discitis associated with sacrocolpopexy. Obstet Gynecol 2013;121(2 Pt 1):285Y290.
* 2014 Lippincott Williams & Wilkins
www.fpmrs.net
183
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
