Pelvic exenteration has become an important procedure, offering a potential curative modality for patients with recurrent gynecologic malignancies.1 Although cure rates have been reported as 50% or greater, the associated morbidity is considerable.2 Most patients undergoing pelvic exenteration have traditionally had limited or no vaginal function if vaginal reconstruction has not been performed. Body image and sexual function are known to be significantly affected in such patients.3,4 Patients with vaginal capacity after pelvic exenteration tend to report fewer problems with many aspects related to quality of life, including physical health, psychosocial and sexual function, and medical and marital interactions.4
During the last several years, many approaches to vaginal reconstruction have been used, including gracilis myocutaneous flaps,5 rectus abdominis flaps,6–9 split-thickness skin grafts into the omental flap,10 and neovagina created from bowel.11 The rectus abdominis flaps have several advantages compared with the others, including versatility, reliability, and lack of need for dilatation.6 However, in some cases, the abdominal wall can be difficult to close secondary to the large defect created. Most surgeons recommend a pedicle size of 12–15 cm in width.6,9 To make closure of the abdominal wall defect easier and yet still obtain an adequate neovagina, we have modified the technique as reported herein.
MATERIALS AND METHODS
After exenteration is performed, the levator muscles are reapproximated in the midline with interrupted absorbable sutures. An abdominal incision centered over the left or right rectus abdominis muscle is created to outline a flap 7–8 cm in width and about 10 cm in length (Fig. 1A). The skin incision is carried down to isolate the flap consisting of the skin, subcutaneous fat, anterior rectus sheath, and it is anchored on the rectus muscle. The superior aspect of the rectus muscle is usually broader than the inferior portion, and the lateral border must be carefully identified. The superior aspect of the muscle is transected, and the muscle is sharply dissected off the posterior rectus sheath. Care should be taken not to disturb the inferior epigastric vessels during the course of this dissection. The rectus muscle and overlying subcutaneous tissues and skin are mobilized to achieve tension-free neovagina creation, and the flap is rotated about the inferior attachment and brought into the pelvis (Fig. 1B–E). The modified vertical rectus abdominis myocutaneous flap is made into the shape of a cone using 3–0 absorbable subcuticular sutures, which tapers going superficial to deep with a defect left posteriorly. The flap is then sewn into position using interrupted absorbable sutures, attaching it to the anterior and lateral introitus (Fig. 1E). A 6- × 1-cm full-thickness skin graft is then harvested from a strip of skin adjacent to the lower abdominal incision. The skin graft is meshed to a 3:1 ratio (Fig. 1F). The graft is then placed posteriorly and sutured to the myocutaneous flap with interrupted absorbable sutures. Next, an inflatable vaginal stent (Mentor Corporation, Goleta, CA) with central drainage tubing is inserted and inflated. The tubing can then be connected to a bulb suction device. The mold is secured and left in the neovagina for approximately 7 days. To reduce the risk of prolapse, the colostomy is placed on the side of the abdominal wall with the rectus muscle in situ, and the urostomy is placed on the contralateral side. The abdominal wall is closed with interrupted, permanent, monofilament sutures.
Between March 1998 and March 2004, 18 consecutive women underwent pelvic exenteration and neovagina creation with a modified vertical rectus abdominis myocutaneous flap. Institutional Review Board approval was obtained from both the University of Iowa and the University of Texas MD Anderson Cancer Center for this study. The mean patient age was 54 years (range 36–77 years). The body mass index ranged from 18 to 44 (mean 27). The cancer diagnoses included 10 patients with recurrent cervical cancer, 6 with recurrent vaginal or vulvar cancer, and 1 each with recurrent endometrial and recurrent rectal cancer. All patients had been treated previously with pelvic and/or para-aortic radiotherapy. Thirteen of 18 patients had an infralevator exenteration performed, whereas 5 had a supralevator exenteration. Twelve patients underwent a total pelvic exenteration, 5 had an anterior exenteration, and 1 had a posterior exenteration. Eight patients had continent urinary conduits (Miami pouch), whereas 9 had incontinent urinary diversions. Among patients who underwent a total pelvic exenteration, the colostomy was brought through the rectus muscle on the side that was not the donor to the vertical rectus abdominis myocutaneous flap, and the urinary conduit was placed on the donor side. For example, after a total pelvic exenteration, the vertical rectus abdominis myocutaneous flap was harvested from the right side and the colostomy was brought through the left side. Patients who underwent an anterior exenteration had the urinary conduit placed through the rectus muscle on the nondonor side.
Although formal time records for the vaginal reconstruction part of the surgery were not kept, based on our last 2 cases, approximately 15 minutes are required for the creation of the skin graft. The vaginal mold was left in place for 7 to 8 days. The length of stay in the hospital ranged from 8 to 42 days (mean 21 days; median 18 days). Twelve of 18 patients had some kind of a postoperative complication. These included 2 patients with sepsis, 2 with line sepsis, 1 with myocardial infarction, 1 with a deep venous thrombosis, 1 with a small bowel obstruction, 1 with an ileus, 2 with a pelvic abscess, 1 with aspiration pneumonia and a superficial wound separation, and 1 with leakage from the right ureter that was managed with a percutaneous nephrostomy. These complications were thought to be related to the overall exenterative surgery and not specifically related to the modified vertical rectus abdominis myocutaneous flap. There were no postoperative deaths. The mean follow-up time for all patients is 2.8 years (range 0.25–5.9 years). None of the 13 patients with colostomies developed hernias or prolapse of their colostomies at last follow-up. Two patients developed a partial stricture of the neovagina, but this was minimal and did not affect function. One patient had a partial distal flap loss that ultimately resulted in a well-epithelialized neovagina. Among all patients, the length of neovagina ranged from 6 to 9 cm.
All of the patients were interviewed by one of the members of the surgical team. Eight patients at last follow-up were sexually active (including the patient with partial flap loss). Of the sexually active patients, 1 had recurrent disease and was receiving chemotherapy. Two other patients have died from recurrent disease, and 2 were alive with recurrence. The other 6 patients had no evidence of disease, were not sexually active, but had a viable neovagina. Among the patients who were not sexually active, 5 did not have a sexual partner.
Despite advances in adjuvant radiation and chemotherapy, pelvic exenteration still remains an important salvage therapy for recurrent or advanced malignancy. Hawighorst–Knapstein and colleagues4 have reported that patients who undergo reconstructive surgery after pelvic exenteration have significantly improved sexual, marital, and psychological quality of life and improved body image. Thus, most surgeons recommend pelvic reconstruction when possible.
We describe here a modification to the rectus abdominis myocutaneous flap for neovagina creation initially detailed by Tobin and Day.6 Subsequently, others have reported using the rectus abdominis myocutaneous flap for vaginal reconstruction.7–9,12 The advantages to a rectus abdominis myocutaneous flap are a straightforward technique, reliable viability, a large arc of rotation, an epigastric donor site that does not interfere with colostomy or urinary conduit placement, and the lack of need for continuous dilatation as compared with split-thickness skin grafting.6 The advantage over a gracilis flap is its better reliability, avoidance of painful thigh scarring, and decreased risk of prolapse.5–6,12 Gracilis flaps are associated with a 11–37% failure rate because of vascular compromise and risk of perineal prolapse.13–16 The disadvantage of a traditional vertical rectus abdominis myocutaneous neovagina flap is the large abdominal wall defect required to create the neovagina.6,9,12 Smith and colleagues9 reported that as many as 23% of patients required Marlex mesh to accomplish fascial closure with the traditional rectus abdominis myocutaneous flap. Another problem with larger flaps is that the flap may not fit into the pelvis of obese patients or in patients with a thick abdominal wall. Our modification allows a much smaller donor site to be used, and the remainder of the neovagina is completed with a small full-thickness skin graft taken from the inferior abdominal incision to make up the tissue needed for adequate diameter of the neovagina. We are able to do this by creating a conical shape with the myocutaneous flap rather than creating a cylinder. We were able to close the abdominal wall primarily in all patients using the modified vertical rectus abdominis myocutaneous flap. Furthermore, the modified vertical rectus abdominis myocutaneous flap leaves only a single midline surgical scar at closure. Another method of avoiding a large anterior rectus fascia defect is to use a horizontal approach of the inferiorly based rectus abdominis myocutaneous flap or a transverse rectus abdominis myocutaneous flap. However, the incision with these approaches can be difficult to close and can be a cosmetic problem.
Pelvic and vaginal reconstruction is associated with many advantages, including reduction of postoperative complications such as intestinal fistulae.17–19 Pelvic reconstruction may allow improved healing in the pelvis as a result of potentially improved blood supply in the pelvis with the flap. Overall, the traditional rectus abdominis flaps have been reliable for pelvic reconstruction.7,9 However, Carlson and colleagues8 have reported 3 episodes of necrosis of the subcutaneous and cutaneous portions of the flap. All 3 of these cases were associated with either prior or intraoperative vascular injury/compromise in the region of the flap.8 Among our patients, only 1 patient had a partial distal flap loss, which eventually resulted in a well-epithelialized neovagina. Contraindications for a rectus abdominis myocutaneous flap should include prior compromise of inferior epigastric vessels (for example, prior a Maylard or Cherney incision). If there is concern about the vascular viability, then preoperative angiography or color flow duplex scanning can be performed to characterize the blood flow to the regions to be used for the myocutaneous flap.20
Data are limited regarding sexual function after exenterative surgery.7,9,21 Ratliff and colleagues21 reported that 47.5% of women were sexually active after pelvic exenteration and neovagina creation with gracilis myocutaneous graft. Similar rates of sexual activity have been reported in other studies with rectus abdominis myocutaneous flaps.7,9 However, most studies have not assessed preoperative sexual function as well as satisfaction with postoperative sexual function. Our results are similar with regard to sexual function but lack a formal and detailed analysis. Our future work will focus on preoperative and postoperative sexual function as well as differences between sexual activities in women with gracilis versus modified rectus myocutaneous neovagina creation.
In summary, we have described a modification of the vertical rectus abdominis myocutaneous flap that appears to be both flexible and reliable. This technique offers the gynecologic surgeons another option for pelvic reconstruction with all of the advantages of a myocutaneous flap without the difficulty of closing a large abdominal wall defect.
1. Brunschwig A. Complete excision of pelvic viscera for advanced carcinoma. Cancer 1948;1:177–83.
2. Morley GW, Hopkins MP, Lindenauer SM, Roberts JA. Pelvic exenteration, University of Michigan: 100 patients at 5 years. Obstet Gynecol 1989;74:934–42.
3. Brown RS, Haddox V, Posada A, Rubio A. Social and psychological adjustment following pelvic exenteration. Am J Obstet Gynecol 1972;114:162–71.
4. Hawighorst-Knapstein S, Schonefussrs G, Hoffmann SO, Knapstein PG. Pelvic exenteration: effects of surgery on quality of life and body image-a prospective longitudinal study. Gynecol Oncol 1997;66:495–500.
5. Morrow CP, Lacey CG, Lucas WE. Reconstructive surgery in gynecologic cancer employing the gracilis myocutaneous pedicle graft. Gynecol Oncol 1979;7:176–87.
6. Tobin GR, Day TG. Vaginal and pelvic reconstruction with distally based rectus abdominis myocutaneous flaps. Plast Reconstr Surg 1988;81:62–73.
7. Carlson JW, Soisson AP, Fowler JM, Carter JR, Twiggs LB, Carson LF. Rectus abdominis myocutaneous flap for primary vaginal reconstruction. Gynecol Oncol 1993;51:323–9.
8. Carlson JW, Carter JR, Saltzman AK, Carson LF, Fowler JM, Twiggs LB. Gynecologic reconstruction with a rectus abdominis myocutaneous flap: an update. Gynecol Oncol 1996;61:364–8.
9. Smith HO, Genesen MC, Runowicz CD, Goldberg GL. The rectus abdominis myocutaneous flap. Modifications, complications, and sexual function. Cancer 1998;83:510–20.
10. Morley GW, Lindenauer SM, Youngs D. Vaginal reconstruction following pelvic exenteration: surgical and psychological considerations. Am J Obstet Gynecol 1973;116:996–1002.
11. Wesley JR, Coran AG. Intestinal vaginoplasty for congenital absence of the vagina. J Pediatr Surg 1992;27:885–9.
12. Benson C, Soisson AP, Carlson J, Culbertson G, Hawley-Bowland C, Richards F. Neovaginal reconstruction with a rectus abdominis myocutaneous flap. Obstet Gynecol 1993;81:871–5.
13. Cain JM, Diamond A, Tamimi HK, Greer BE, Figge DC. The morbidity and benefits of concurrent gracilis myocutaneous graft with pelvic exenteration. Obstet Gynecol 1989;74:185–9.
14. Lacey CG, Stern JL, Feigenbaum S, Hill EC, Braga CA. Vaginal reconstruction after exenteration with use of gracilis myocutaneous flaps: The University of California, San Francisco experience. Am J Obstet Gynecol 1988;158:1278–82.
15. Soper JT, Larson D, Hunter VG, Berchuck A, Clarke-Pearson DL. Short gracilis myocutaneous flaps for vulvovaginal reconstruction after radical pelvic surgery. Obstet Gynecol 1989;74:823–7.
16. Copeland LJ, Hancock KC, Gershenson DM, Stringer CA, Atkinson EN, Edwards CL. Gracilis myocutaneous vaginal reconstruction concurrent with total pelvic exenteration. Am J Obstet Gynecol 1989;160:1095–101.
17. Soper JT, Berchuck A, Creasman WT, Clarke-Pearson DL. Pelvic exenteration: factors associated with major surgical morbidity. Gynecol Oncol 1989;35:93–8.
18. Rutledge FN, Smith JP, Wharton JT, O'Quinn AG. Pelvic exenteration: Analysis of 296 patients. Am J Obstet Gynecol 1977;129:881–92.
19. Miller B, Morris M, Gershenson DM, Levenback CL, Burke TW. Intestinal fistulae formation following pelvic exenteration: a review of the University of Texas M. D. Anderson Cancer Center experience. Gynecol Oncol 1995;56:207–10.
20. Pacifici A, Tinti A, Flamini FO, Cordellini M, Trapassi S, Dominici C. Colour flow duplex scanning: an accurate, non-invasive technique for preoperative evaluation of the vascular supply of the rectus abdominis myocutaneous flap. Scand J Plast Reconstr Surg Hand Surg 1995;29:319–24.
21. Ratliff CR, Gershenson DM, Morris M, Burke TW, Levenback C, Schover LR, et al. Sexual adjustment of patients undergoing gracilis myocutaneous flap vaginal reconstruction in conjunction with pelvic exenteration. Cancer 1996;78:2229–35.