reoperative pelvic surgery · several controversies exist regarding surgery for locally recurrent...

Scientifi c American Surgery DOI 10.2310/7800.2047

11/14

© 2014 Decker Intellectual Properties Inc

gastrointestinal tract and abdomen

R E O P E R A T I V E P E L V I C S U R G E R Y

Eric J. Dozois, MD, FACS, FASCRS, and Daniel I. Chu, MD

Reoperative pelvic surgery is technically challenging and carries with it signifi cant potential risk. The inherent risks of any pelvic operation, such as bleeding and injury to critical structures such as the ureters, are magnifi ed by the oblitera-tion of previous embryonic fusion planes and anatomic relationships within the confi nes of a narrow, deep opera-tive fi eld. This topic review addresses the surgical indica-tions, techniques, and pitfalls when dealing with recurrent pelvic pathology and provides an overview of safe and effective approaches to reoperative pelvic surgery.

Pelvic pathology that requires reoperative surgery in the gastrointestinal tract usually involves four surgical themes: (1) recurrent malignancies, for which we use recurrent rectal cancer as an example; (2) complications from ileal pouch-anal anastomoses (IPAAs) for infl ammatory bowel disease; (3) complications from low pelvic anastomoses; and (4) pal-liative situations. For any of these indications, the goals of reoperative pelvic surgery are twofold: (1) resection/repair of the primary indication, which could include pelvic exen-teration for recurrent rectal cancers to resection/repair of a problematic coloanal anastomosis, and (2) reconstruction when possible, which could include complex soft tissue and genitourinary constructs to restoration of gastrointestinal continuity.

Pelvic Anatomy

An advanced understanding of pelvic anatomy that includes soft tissue, gastrointestinal, genitourinary, neuro-vascular, and osseous structures is essential in the evaluatio n and management of complex pelvic pathology. Previous surgery, pelvic radiation, and large tumors can distort the well-defi ned embryologic fusion planes such as the presa-cral space, making pelvic dissections particularly diffi cult and dangerous. Several important neural and vascular struc-tures, such as the pelvic autonomic plexus and iliac vessels, respectively, are located in the pelvis, and injury or removal of them as part of en bloc resection may have important physiologic sequelae as well as long-term neurologic and musculoskeletal consequences.

Bleeding is one of the most signifi cant risks of reoperative pelvic surgery. The arterial and venous anatomy associated with the rectum, as well as the urologic and gynecologic organs, must be clearly delineated so that mobilization and/or ligation of vessels necessary to remove these organs can be safely performed. When dissection is carried out in the lateral pelvic sidewall, for example, multiple branches of the internal iliac arteries and veins will be encountered, result-ing in major blood loss when not approached carefully. In the previously irradiated pelvis, the vascular dissection necessary for en bloc tumor removal or sacral resection can be facilitated by an experienced vascular surgeon as vessels are often densely fi brosed and yet highly fragile.

If sacrectomy is to be performed, the surgeon must be familiar with the relationship between the thecal sac, sacral

nerve roots, sciatic nerve, piriformis, obturator internus muscles, and ligaments, including the sacrotuberous and sacrospinous. For extended sacropelvic resections, an expe-rienced orthopedic oncologic spine surgeon greatly enhance s the ability to preserve important nerve roots and ensure adequate oncologic musculoskeletal margins.

Operative Management

recurrent malignancy: rectal cancer

Despite modern management, locally recurrent rectal cancer remains a signifi cant problem, with the incidence of recurrence as high as 33% in some series.1–3 Only approxi-mately 20% of patients with recurrent disease, however, may be amenable to repeat curative resection.4 Although the incidence of metastatic disease in recurrent rectal cancer approaches 70%, up to 50% of patients die with local disease only.2,5–9 In addition, although most local recurrences occur within the fi rst 2 to 3 years following initial curative resection, a small number do recur after longer periods of time.10,11 The 5-year survival following surgery for recurrent rectal carcinoma is reported to be as high as 58%. Morbidity, however, is reported to be between 20 and 80%, with most postoperative complications consisting of wound complications.3,12,13

Several controversies exist regarding surgery for locally recurrent rectal cancer. These controversies include the effectiveness and response to adjuvant chemoradiotherapy, intraoperative electron beam radiotherapy (IOERT), and operative parameters such as the timing of resection. These factors affect not only the prognosis of the patients but also patient selection and fi nal outcomes. The selection of patients for recurrent pelvic operation includes a wide range of variables, including the patient’s fi tness for surgery, the extent of recurrent disease, the presence of metastatic dis-ease, and the intent of the resection (cure versus palliation). The extent and magnitude of the operation must be discussed with the patient. The possibility of a permanent colostomy and urostomy and the complications of reopera-tive pelvic surgery, such as hemorrhage, poor wound heal-ing, and urinary and sexual dysfunction, should be reviewed in detail.

Operative Planning

Patients being considered for resection should undergo evaluation and staging by a multidisciplinary team. Physical examination of the rectum, inguinal nodal basins, and vagina in women helps determine the extent of recurrence and resectability. Evaluation of the remaining colon via endoscopy is imperative to rule out synchronous tumors and adequacy of bowel for reconstruction. Staging of the extent of locoregional disease is best accomplished with computed tomography (CT) or magnetic resonance imaging (MRI). Positron emission tomography (PET) is useful to

Scientifi c American Surgery

11/14

gastro reoperative pelvic surgery — 2

assist in distinguishing tumor recurrence from postradiation scar and to rule out distant metastatic disease. In the setting of recurrent pelvic tumor and previous treatment, MRI may better delineate the tumor in relation to the musculoskeletal anatomy.

Evaluation of all outside records, including operative notes, chemotherapy regimens, and total radiation doses, provides invaluable information for both perioperative and intraoperative management. In reoperative pelvic surgery, it is imperative that the technical details of any previous operations be thoroughly reviewed as this information helps formulate a strategy for resection and reconstruction while decreasing ambiguity at the time of reoperation.

Radiation history is important in the evaluation of patients with recurrent rectal cancer because treatment with chemoradiation is a signifi cant part of the multidisciplinary management of recurrent disease. The type and extent of radiation are crucial in decision making as maximum allow-able dosages need to be calculated. In a multicenter study, Valentini and colleagues reported an 8.5% complete patho-logic response and a 29% downstaging following reirradia-tion.14 Das and colleagues similarly reported an improved 3-year survival of up to 66% in patients with recurrent rectal cancer treated with preoperative chemoradiation.15 If addi-tional external-beam radiation is administered, plans for surgical intervention are expedited, thus allowing for the synergistic benefi ts of both external and intraoperative radiation.

Without any treatment, mean survival is approximately 8 months in patients with locally recurrent rectal cancer. Chemoradiation may alleviate symptoms, but 5-year sur-vival remains poor at less than 5%.16–19 The prognostic factors associated with poor outcomes after resection for recurrent rectal cancer are listed here [see Table 1], but sur-gery remains the best chance for cure in patients with recur-rent rectal cancer. Although absolute and relative contrain-dications for exenterative surgery are reported [see Table 2], management algorithms continue to evolve, with a para-digm shift from palliation to cure in patients with recurrent, locally advanced, and even metastatic disease.

Resection with negative microscopic margins (R0 resec-tion) and appropriate regional lymphadenectomy is the goal of reoperative surgery for pelvic recurrence. R1 and R2 resection rates of up to 58%, however, have been reported

after reoperative surgery for recurrent rectal carcinomas. In a Mayo Clinic series, the reported curative negative resec-tion margin was 45% in 394 patients who underwent surgi-cal exploration for recurrent rectal carcinoma. These patients had a statistically signifi cant improved 5-year survival of 37% compared with 16% in patients in whom negative margins were not achieved.20 MD Anderson Cancer Center similarly achieved negative margins in 76% of 85 patients undergoing resection for recurrent rectal cancer.21 The 5-year disease-free survival was 46%, and multivariate analysis showed that an R1 resection was associated with a negative prediction of survival. Surgical margins therefore remain the most signifi cant factor for long-term survival when operating for recurrent rectal carcinoma.2,20,21

In the published literature, multiple classifi cation schemas exist that categorize the extent and boundaries of pelvic exenterations. For the purposes of this topic review, we classify exenteration into (1) anterior, (2) posterior, or (3) combined anteroposterior resections using the rectum or neorectum (if a previous low anterior resection or coloanal anastomosis was performed) as a focal point of the malig-nant process. An anterior exenteration includes resection of the rectum or neorectum and any involved viscera in front of the rectum, such as the reproductive organs, the bladder, or both [see Figure 1]. A posterior exenteration includes the rectum or neorectum and any tissue posterior and lateral, such as musculoskeletal tissue and neurovascular structures [see Figure 2]. Based on this defi nition, we therefore consider sacrectomy a potential component of a posterior exentera-tion. An anteroposterior exenteration involves a combina-tion of anterior and posterolateral structures, including the bony sacropelvic components [see Figure 3].

Operative Technique

Anterior exenteration We use ureteral stents for all reoperative cases and when extensive pelvic radiation has been given. The Lloyd-Davies position is used to allow access to the perineum. Care is taken to protect the extremi-ties from nerve injury with adequate padding of both the tucked arms and lateral lower extremities in the stirrups. A midline exploratory laparotomy is used to confi rm the absence of extrapelvic disease and determine local resect-ability. Any suspicious extrapelvic lesions are biopsied and sent for frozen-section analysis. Biopsy-proven extrapelvic disease may preclude pelvic resection for cure.

The left colon is mobilized and transected at the appropri-ate level for subsequent end colostomy. The entrance to the

Table 1 Prognostic Factors Negatively Impacting Outcomes Following Surgery for

Recurrent Rectal CancerInability to achieve R0 resection35

Type of primary surgery (worse for APR vs LAR)20,85

Rising CEA86

Advanced stage of primary tumor, i.e., T4, nodal disease42

Recurrence-free interval < 12 months42

High sacral involvement (above S2)48

Multiple points of tumor fixation within pelvis20

Major blood transfusion, neural invasion35

Cortical and marrow sacral involvement35

Presence of pain preoperatively—sciatica42

APR = abdominoperineal resection; CEA = carcinoembryonic antigen; LAR = low anterior resection.

Table 2 Absolute and Relative Contraindications for Exenterative Surgery

Absolute contraindicationsUnresectable distant metastasisBilateral pelvis sidewall diseasePoor performance status

Relative contraindicationsExtensive unilateral sidewall diseaseInability to achieve R0 resectionSacral involvement above S3Encasement of iliac vessels


11/14


llealconduit

Vag.cuff

Factor© MAYO

2011

S1S2

Factor© MAYO

2011

Figure 1 Anterior exenteration: en bloc resection of rectum or neorectum and genitourinary and/or gynecologic structures and postresection defect requiring reconstruction. Reproduced with permission from the Mayo Clinic.

Figure 2 Posterior exenteration: en bloc resection of rectum or neorectum and sacrum and a postresection defect requiring reconstruction. Reproduced with permission from the Mayo Clinic.

pelvis must be completely cleared of extraneous tissue for optimal pelvic exposure. Any adherent tissue (often small bowel) to the pelvic tumor should be resected en bloc. Deep pelvic dissection to the presacral space then proceeds poste-riorly along the distal aorta and continues caudally over the iliac vessels and ureters [see Figure 4]. Vesseloops are used to retract the ureters and vascular structures. In the reoperative pelvis, dense fi brosis and distorted anatomy require careful, meticulous dissection to avoid inadvertent injuries and major bleeding complications. The most at-risk region for signifi cant bleeding occurs during mobilization of the left common iliac vein, and this dissection should proceed with caution. Posterior lumbar branches, if not identifi ed, can also lead to signifi cant blood loss if avulsed.

The anterior and lateral lines of resection (decided on during preoperative review of imaging) are delineated, and the involved structures and organs are mobilized widely for subsequent en bloc resection. The presacral space is further developed, and the dissection is carried distally along the anterior sacrum while staying anterior to the Waldeyer fascia. The dissection is then carried laterally along the pelvic sidewall on each side with careful protection of the lumbosacral plexus and internal iliac vessels.

In women, anterior involvement may require resection of the posterior wall of vagina, which can be approached transvaginally and transabdominally. In contrast, anterior fi xation in men often requires cystoprostatectomy due to invasion of the trigone and prostate. Partial cystectomy may


11/14


be suffi cient in rare cases. When the bladder is resected, the ureters are mobilized as close to the bladder as possible to provide adequate length for urinary conduit reconstruction. To mobilize the bladder, the space of Retzius is entered to fully mobilize the anterior portion of the bladder. The blood supply to the bladder is taken by serially ligating the arteries and veins along the pelvic sidewall, coursing back and forth from the internal iliac vessels. The wings of peritoneum to the bladder are then taken down until the bilateral vasa are identifi ed and clipped. We then isolate and trace both ureters to the bladder before clipping and bringing them out

of the pelvis. The distal margins are sent and confi rmed to be negative by pathology. With the ureters both isolated and out of the pelvis, we then take down the pedicles to the bladder and prostate and continue the anterior dissection by opening the endopelvic fascia bilaterally. The dorsal venous complex is subsequently ligated. We then reach the urethra, and after delivering the catheter into the wound, we transect the urethra.

Construction of an end colostomy and urinary conduit completes the procedure. In women who require posterior vaginal wall removal, reconstruction may be necessary [see Reconstruction, below]. For anterior resection, soft tissue reconstruction to fi ll a void in the pelvis is used selectively.

Posterior exenteration (sacrectomy) First stage: anterior approach After placement of ureteral stents, a midline exploratory laparotomy is used to confi rm the absence of extrapelvic disease and determine local resectability. The left colon is mobilized and transected at the appropriate level to be used later for colostomy, exposing the presacral space. Deep pelvic dissection then proceeds posteriorly along the lower aorta and continues distally over the iliac vessels and ureters. The anterior and lateral lines of resection are delin-eated, and the involved structures and organs are mobilized widely for subsequent en bloc resection. Frozen-section biopsies are taken as needed to establish that margins are negative at the level of the proximal sacral transection.

Vascular exposure often requires mobilization of the lower aorta and vena cava, in addition to the iliac arteries and veins. Circumferential mobilization of the common and external iliac arteries may be needed to facilitate exposure of the veins. The internal iliac artery branches are ligated and divided distal to the takeoff of the posterior division of the superior gluteal artery branch to preserve blood fl ow to the gluteal fl aps [see Figure 5]. Multiple internal iliac vein branches are ligated after control of the main trunk(s) of the internal iliac vein. The branches are divided before ligation

S1S2

Factor© MAYO

2011 llealconduit

Vag.cuff

Superior glutealartery

D.F.© MAYO

2010

Internal iliacartery

Lateral sacralartery

Median sacralartery

Figure 3 Anteroposterior exenteration: en bloc resection of both anterior and posterior structures and postresection defect requiring reconstruc-tion. Reproduced with permission from the Mayo Clinic.

Figure 4 Anatomy of presacral space after rectal mobilization. The Waldeyer fascia is not shown to illustrate the neurovascular structures. Reproduced with permission from the Mayo Clinic.


11/14


a

Mediansacralartery

Superiorglutealartery

Internaliliacvein

Internaliliac

artery

Lateralsacralartery

b

D.F.© MAYO

2010

c

Figure 5 (a) Vascular exposure and dissection. Internal iliac artery branches are ligated, preserving gluteal blood fl ow. (b) Ligation of the lateral sacral branches of the internal iliac vein, all the way down to the distal sacrum. (c) High ligation of the internal iliac vein. Reproduced with permission from the Mayo Clinic.

of the main trunk to avoid venous distention of the branche s, which can lead to troublesome bleeding. Lateral and middle sacral vein branches, which drain into the posterior aspect of the left common iliac vein and caval confl uence, are ligated and divided. Suture ligature is preferable for short, broad-based internal iliac vein branches. The vascular dis-section is carried along both sides of the sacrum onto the pelvic fl oor.

During this anterior stage of the operation, a colostomy and ileal or colonic urinary conduit are constructed as neede d. A rectus abdominis myocutaneous fl ap is then prepared for subsequent perineal reconstruction. Using intraoperative fl uoroscopy, the sacral level of transection is determined and anterior osteotomies are performed. A screw is placed in the osteotomy site to facilitate alignment for fl uoroscopy [see Figure 6]. Prior to closing and turning the patient prone, a thick Silastic mesh is placed anterior to the sacrum and posterior to the vessels and soft tissue struc-tures, to protect against injury when blind osteotomies are performed during the second stage (prone) of the procedure [see Figure 7]. The abdomen is then closed, stomas are matured (fecal, urinary), and the patient is turned and placed in the prone position.

Second stage: posterior approach The second stage of the procedure is carried out with the patient in the prone position. A posterior midline incision is made along the midportion of the sacrum, and the gluteus maximus muscles

D.F.© MAYO

2010

Figure 6 Unicortical transverse osteotomy, marked with a “screw” (arrow) to facilitate alignment with fl uoroscopy during posterior dissection. Reproduced with permission from the Mayo Clinic.


11/14


are dissected away from the sacral attachments. The sacro-spinous and sacrotuberous ligaments are divided to access the pelvic cavity posteriorly [see Figure 8]. The piriformis muscle is divided while protecting the sciatic and pudendal nerves. Laminectomy, dural sac ligation, and sacral resec-tion are then carried out [see Figure 9]. Final osteotomies are performed based on imaging studies that ensure tumor-free margins. After resection, the specimen is examined by the pathologist and the surgical team to assess for gross mar-gins. The posterior wound is then reconstructed with a ped-icled myocutaneous rectus fl ap [see Figure 10]. Depending on the extent of resection and estimated operative time for each stage, the entire operation may be performed in 1 day or, increasingly in our experience, on separate operating days.

Combined anteroposterior exenteration When a combined anteroposterior component exenteration is performed, the

steps outlined above apply. The sequence of this extensive procedure depends on multiple factors, but the anterior exenteration is typically completed fi rst, followed by the posterior exenteration. When high sacral or extended sacro-pelvic resection is done, the operation is best executed in separate stages on distinct operating days. Our typical approach would be to start with the anterior dissection on day 1, complete the posterior approach with removal of tumor on day 3, and conclude with the spinopelvic reconstruction on day 5.

Reconstruction Reconstruction after exenteration typi-cally involves the perineum and the gastrointestinal and/or genitourinary tracts. In women, vaginal reconstruction may be necessary when extended resections include the posterior vagina. Perineal wound complications after extended pelvic resections represent a considerable source of morbidity for patients. Nonhealing perineal wounds after extended pelvic resections are reported to occur in 7 to 66% of patients,22 and the addition of radiotherapy has been associated with major wound complications in 41% of patients after perineal resection for rectal cancer.23 Exenterative surgery that includes sacropelvic resections represent some of the most challenging wounds to fi ll and cover.

Omental fl ap The main advantage of the omental pedi-cled fl ap after exenteration is its ability to fi ll the pelvic dead space.24 Filling this space with vascularized tissue may decrease the risk of pelvic sepsis and prevent small bowel from adhering to the raw surfaces, which may cause small bowel obstructions. The omentum is mobilized and a right or left omental pedicle is constructed based on the greatest length and bulk. Spreading and lengthening techniques can be performed by detaching the gastroepiploic arcade from the stomach on the origin side of the right or left gastroepi-ploic pedicle. The gastroepiploic artery and vein on the opposite side are divided and ligated. The right gastroepi-ploic vessels are preferred because they are larger and have more epiploic branches than the left. A window can be fashioned through the transverse mesocolon to allow a more direct route to the pelvis.

Vertical rectus abdominis myocutaneous (VRAM) fl ap The VRAM fl ap provides a well-vascularized, bulky tissue paddle that can be transposed to the pelvis to fi ll dead space and reconstruct the perineum. This fl ap has become our primary choice for soft tissue reconstruction following exenteration. Chessin and colleagues studied 59 patients who underwent a VRAM or primary closure following abdominoperineal resection for anorectal cancer treated with preoperative radiation.22 The patients who were recon-structed with a VRAM fl ap had signifi cantly lower rates of perineal wound complications (15.8% versus 44.1%). More-over, patients had no abdominal wall hernias or infections associated with the fl ap harvest site. Radice and colleagues compared primary closure with pedicled omentum and VRAM fl ap and found that the VRAM fl ap signifi cantly reduced wound infections and length of stay compared with the other types of closure.25

The VRAM fl ap is mobilized by raising a supraumbilical skin paddle with underlying fat and rectus muscle. The anterior rectus sheath is taken with the fl ap, and the poste-rior rectus sheath is left for abdominal wall closure. The

D.F.© MAYO

2010

Figure 7 Placement of Silastic mesh to protect anterior vasculature when posterior osteotomies are performed blindly. Reproduced with permission from the Mayo Clinic.

Internalpudendalnerve &vessels

Sciatic nerve

Piriformismuscle

Sacrotuberous ligament

Sacrospinousligament

D.F.© MAYO

2010

Figure 8 Division of the sacrospinous and sacrotuberous ligaments and the piriformis muscle. Reproduced with permission from the Mayo Clinic.


11/14


perforating vessels through the anterior rectus sheath are carefully dissected to minimize the fascial resection. The blood supply to the VRAM is the deep inferior epigastric artery and vein, which are mobilized as a pedicle. Transab-dominal passage of the VRAM fl ap, sized to reconstruct the perineal wound, is done by rotating the pedicle into the

pelvis with the skin of the pedicle facing the perineum. It is important to avoid signifi cant twisting and tension when securing the fl ap to the perineum [see Figure 11].

Gluteus maximus myocutaneous fl ap The gluteus maximus myocutaneous fl ap is based on the inferior gluteal artery pedicle. The gluteus maximus fl ap can be constructed as a

D.F.© MAYO

2010

a

b c

S1

D.F.© MAYO

2010

Figure 10 Transperineal delivery of pedicled myocutaneous rectus fl ap for perineal reconstruction. Inset shows fl ap reconstruction site after healing takes place. Reproduced with permission from the Mayo Clinic.

Figure 9 Posterior sacral osteotomy (a), laminectomy (b), and dural sac ligation (c). Dashed lines outline margins of bony resection. Reproduced with permission from the Mayo Clinic.

Figure 11 Harvest of a vertical rectus abdominis myocutaneous fl ap (VRAM) on a long pedicle to reach the pelvis for reconstruction. Reproduced with permission from the Mayo Clinic.


11/14


are elevated based on the vascular muscle distribution sup-plied by the superfi cial femoral artery. This axial-pattern myocutaneous fl ap is based on branches of the femoral ves-sels, including lateral and medial circumfl ex arteries, which arise from branches of the profunda femoris artery. The long myocutaneous fl ap includes the bulk of the quadriceps fem-oris and can provide signifi cant coverage of pelvic defects.

The total fi llet hemipelvectomy fl ap uses the majority of the thigh musculature. The blood supply to this large fl ap is based on preservation of the superfi cial and profunda femoris vessels, allowing for a substantial amount of muscle preservation. The circumfl ex vessels of the profunda femoris artery perforate the adductor magnus muscle to the poste-rior and lateral compartments of the thigh and play a role in supporting a total thigh fi llet fl ap. The total thigh fi llet fl ap provides the largest amount of vascularized tissue when massive defects require soft tissue reconstruction [see Figure 14].

Urinary reconstruction When bladder resection is required, the construction of a functional urinary outlet must be established. Options include continent procedures (Kock pouch, Indiana pouch) or incontinent urostomies. Because exenteration for malignancies often precludes continent procedures, options for urinary reconstruction include ureterocolostomy or, more commonly, an ileal or colon conduit. If the superior rectal artery is ligated but the inferior mesenteric artery and its sigmoidal branches are preserved to ensure blood supply, then the sigmoid colon becomes an excellent urinary conduit as it can be placed in either upper abdominal quadrant as necessary. This approach additionally avoids a small bowel anastomosis as required with construction of an ileal conduit. If the inferior mesenteric artery (IMA) has been taken, however, we do construct an ileal conduit.

Vaginal reconstruction In women who undergo extended anterior exenteration, a portion of the posterior wall of the vagina may need to be removed to achieve negative margins. A variety of techniques have been described for neovaginal reconstruction after anterior pelvic exenteration. We prefer the rectus abdominis myocutaneous (RAM) fl ap for neovaginal reconstruction because of its consistent blood

local advancement or a V-Y advancement myocutaneous fl ap. The technique involves elevation and advancement of the superior half of the muscle and overlying skin. Two myocutaneous gluteus maximus fl aps are prepared, leaving the lower 50% of the underlying muscle intact. Both fl aps are dissected free and raised. After deepithelialization of its media part, one of the two fl aps is then advanced into the pelvic defect. After partial sacrectomy, even large cavities can be fi lled completely. The second fl ap is advanced toward the midline and sutured in layers to the fi rst fl ap, resulting in a solid midline reconstruction. Both donor defects are closed using a V-Y technique. For very large perineal defects following sacrectomy, the combination of an inferiorly based VRAM and bilateral fasciocutaneous V-Y advancement fl aps may be necessary for adequate closure.

Gracilis fl ap When a VRAM fl ap is not available, the gracilis muscle may be used to partially fi ll a pelvic defect. The gracilis fl ap has an inconsistent blood supply and lim-ited arc of rotation; additionally, the lack of bulk may be insuffi cient for obliteration of a large pelvic dead space. The gracilis fl ap is harvested by making a longitudinal incision over the muscle of the medial thigh. Its blood supply is centered on the proximal pedicle of the profunda femoris artery, and the distal pedicle originating from the saphenous artery is ligated. The muscle and skin paddle, if constructed, are mobilized and rotated into the perineal wound [see Figure 12].

Total thigh fi llet fl ap Thigh fi llet fl aps can be constructed following hemipelvectomy using soft tissue from the ampu-tated limb [see Figure 13]. The posterior hemipelvectomy fl ap is designed as a myocutaneous fl ap based on the superior and inferior gluteal vessels. By preserving the gluteal ves-sels, large posterior fl aps incorporating the gluteal muscle can be rotated for coverage of the pelvic defect after the bony pelvis has been removed. If oncologically appropriate, preservation of the internal iliac vessels will increase poste-rior fl ap blood supply and decrease fl ap necrosis. Limited resection of the bony pelvis allows for preservation of the sacral perforators.

Long anterior hemipelvectomy fl aps are myocutaneous fl aps of the anterior compartment musculature. These fl aps

Figure 12 Gracilis fl ap rotating into the perineal wound. Repro-duced with permission from the Mayo Clinic.

Figure 13 Total thigh fi llet fl ap after mobilization off the femur. Reproduced with permission from the Mayo Clinic.


11/14


supply and its ability to obliterate pelvic dead space [see Figure 15]. Soper and colleagues reviewed 32 patients who had neovaginal reconstruction with a RAM fl ap after exen-terative surgery (85% received radiation) and found a fl ap complication rate of 32%, with only one patient developing vaginal stenosis.26 Neovaginal reconstruction with a RAM fl ap has decreased perineal complications and maintains preservation of the psychosexual self-image of the patient.

Sacropelvic reconstruction When tumors extend posteri-orly into the sacrum or deep into the lateral sidewall and through the sciatic notch, sacropelvic resection, including hemipelvectomy, will be required to achieve negative margins. The type of hemipelvectomy performed depends on the extent of the tumor. The four types of sacropelvic resections include total sacrectomy (type 1), hemisacrectomy (type 2), hemisacrectomy + external hemipelvectomy (type 3), and total sacrectomy + hemipelvectomy (type 4), with types 1 and 3 being most commonly performed [see Figure 16].

Troubleshooting

Pelvic exenterations are technically diffi cult, and surgeons must be equipped to deal with inevitable problems. In reoperative pelvic surgery, signifi cant danger exists with

massive hemorrhage from presacral veins and the major iliac vessels. In the event of massive presacral vein bleeding, strategies used to stop the hemorrhage include packing, muscle patch, argon beam coagulation, thumbtacks, and bone wax.27–29 We have observed that packing, occasionally with the use of hemostatic agents such as Surgicel, controls presacral bleeding in the majority of cases. If bleeding continues, our next technique is to harvest a piece of rectus abdominis muscle and then to cauterize the muscle over the bleeding site; the resultant coagulum is remarkably hemo-static. Thumbtacks are reserved for situations when the origin of hemorrhage can be clearly visualized and stopped with point pressure; if inaccurately placed, tacks can make the bleeding much worse.

Iliac artery and vein bleeding can be deadly and must be dealt with in an expeditious but controlled manner. Direct pressure can temporize bleeding while exposure is obtained. Proximal and distal control should be achieved. Vascular sutures and equipment must be readily available from the very beginning of the case to use for suture repair and/or ligation. For extended sacral resections, consideration should be given to controlling and ligating the internal iliac artery early in the procedure.30 In the previously irradiated pelvis, we think it is also important to collaborate with experienced vascular surgeons given the technical diffi culty working with fi brosed vessels.

During dissection in the reoperative pelvis, it is essential to immediately identify normal anatomic structures. The genitourinary, vascular, and osseous structures such as the ureters, distal aorta, and sacral promontory, respectively, can provide landmarks to help guide the descent into the narrow confi nes of the pelvis. The typical anterior and posterior planes are often obliterated due to scar or fi brosis, and orientation to critical structures is paramount as sharp dissection with a knife, heavy scissors, or osteal elevators may be the only means to make any forward progress. Guidance from the perineum, such as application of vaginal retractors, distention of the bladder, or a rigid proctoscope through the neorectum, can also assist with understanding the spatial relationships in the deep pelvis.

Complications

The mortality in selected series after exenterative pelvic surgery ranges from 0 to 9% overall.20,31–36 Causes of death include uncontrolled hemorrhage at the time of surgery, renal failure, pelvic sepsis, and cardiovascular compromise.37,38 Between 24 and 68% of patients developed complications, most of which were related to failure of perineal wound healing.12,31,34,37–39 The incidence of complications increases with the extent of resection.37,39 Patients undergoing sacrec-tomy, hemipelvectomy, or IOERT have even higher rates of complications, most of which are related to the posterior wound.20 The most signifi cant reported side effects of IOERT include dose-dependent peripheral neuropathy (16 to 34%) and ureteral stenosis (6%).40

Well-described postoperative complications include pelvic abscesses, small bowel obstructions, fi stula formation, perineal wound problems, and ureteral obstruction. In a series by Henry and colleagues, 53% developed complica-tions after resection for pelvic recurrences.41 Nearly a third of the complications in this series were related to infection.

Figure 14 Healed total thigh fi llet fl ap used to cover the defect. Reproduced with permission from the Mayo Clinic.


11/14


urinary complications. Double-J stent placement, if possible, is the best way to address hydronephrosis caused by ure-teral obstruction versus nephroureteral stenting if an endo-scopic stent is unable to be placed. Finally, percutaneous nephrostomy tubes can be used if a complete urinary diver-sion is needed to heal a leak.

Outcome Evaluation

Oncologic outcomes after exenteration are diffi cult to compare because of the lack of consistent classifi cation of these procedures. Moreover, published series often include a heterogeneous patient population. When recurrent disease is confi ned to the axial and anterior locations (male and fe-male genitourinary organs, including the bladder), an R0 resection is frequently possible and therefore leads to better survival. Moore and colleagues found that 72% of patients with a recurrent axial or anterior recurrence were able to achieve an R0 resection.42 This fi nding is in contrast to lat-eral and posterior recurrences, which have lower rates of R0 resection: 50 and 36%, respectively.42 Stocchi and colleagues noted a 3-year survival of 45% with en bloc resection involv-ing the urinary tract.43 Anterior recurrences are less likely to

Among these patients, there were 9.2% pelvic abscesses, 5.7% abdominal wound infections, and 3.4% perineal wound infections. The majority of superfi cial wound infections can be successfully managed with local drainage and wound care. Percutaneous drainage of pelvic abscesses is commonl y performed with catheters placed either anteriorly or posteri-orly. Abscesses low in the pelvis, which are not amenable to CT-guided percutaneous drainage, may be addressed through the perineum or transanally in the operating room. Chessin and colleagues noted a decrease in the perineal wound complication rate, from 44 to 16%, with the use of a RAM fl ap to close the perineum.22 The increased use of vascularized tissue fl aps will continue to decrease infection-related complications.

Urologic complications are also common with ureteral obstruction and leak occurring with construction of urinary conduits.38 Permanent bladder dysfunction requiring inter-mittent catheterization or cystectomy may also occur after posterior exenterations. Renal insuffi ciency resulting from ureteral stenosis secondary to ureterolysis and pelvic irra-diation has been described.3 A CT urogram with delayed contrast imaging of the urinary system can diagnose most

a b

Figure 15 (a) In-setting of a vertical rectus abdominis myocutaneous fl ap (VRAM) to close the large perineal defect after exenteration that includes resection of the posterior wall of the vagina. (b) Healed fl ap with closure of the posterior vagina. Reproduced with permission from the Mayo Clinic.


11/14


become fi xed to the bony pelvis and are more amenable to complete excision and, therefore, improved survival.

The ability to achieve an R0 resection with lateral and pos-terior recurrences is more diffi cult because of the potential for increased points of fi xation.20,42 Recent series have been able to obtain R0 resections in 51 to 100% of patients, which included lateral and posterior recurrences.35,37,39,44–46 The 5-year survival rate for this heterogeneous patient population has ranged from 20 to 39%. An R0 resection was the most important variable in predicting 5-year survival [see Table 3].20,37 Similar results have been reported by our group for recurrent rectal cancers involving the aortoiliac axis.47 Major vascular reconstructions involving the aorta and iliac arteries and veins were performed to achieve an R0 resection in over 50% of these patients, with reported 4-year survival and disease-free survival rates of 55% and 45%, respectively.

Oncologic outcomes for lower sacral resection have resulted in 5-year survival rates as high as 37% in some patients.20 These results from lower sacral involvement have prompted questions on whether similar outcomes could be expected in patients with high sacral involvement. Our initial experience with high sacrectomy, defi ned as above the third sacral body, showed promising results with an R0 resection in all nine patients: 30-day mortality of zero and median survival of 31 months.39 Wanebo and colleagues reported two series of patients with recurrent rectal cancer who underwent high sacral resection, achieving 5-year sur-vival rates of 31 and 33%, respectively.1,48 Our group recentl y reported survival outcomes in 30 patients with locally recur-rent rectal cancer undergoing extended sacral resection and demonstrated a 46% 5-year survival and 43% 5-year disease-free survival with R0 resection, achieved in 93% of patients.46 Moreover, there were no operative deaths and an acceptable morbidity. Surgery provided excellent local tumor control, with only one patient having re-recurrence in the pelvis.

Improved survival in patients with recurrent rectal cancer is possible when IOERT is added to resection compared with patients who do not receive IOERT.49 When IOERT is used as part of a multimodality approach, a 15% increase in survival has been demonstrated.3 Although there are no randomized trials comparing surgery alone and surgery with IOERT, there seems to be a benefi t not only for patients with negative margins but also in those with microscopic and grossly positive margins. Haddock and colleagues recently published the Mayo Clinic’s experience with multi-modal therapy for locally advanced recurrent colorectal cancer that included the use of IOERT.50 In our practice, the dose of IOERT given at the time of surgery is dependent on the resection margin [see Table 4]. Survival estimates at 5 years were 46%, 27%, and 16% for R0, R1, and R2 resection, respectively. For carefully selected patients, exenterative resection of pelvic recurrence can provide a 5-year survival of up to 39%, which is similar to that seen after resection of solitary metastasis to the lung, liver, and brain.37

For re-recurrent rectal cancer, our experience suggests that an R0 re-resection provides outcomes similar to those of surgery for fi rst-time recurrent rectal cancer.51 From 47 patients with re-recurrent rectal cancer, we achieved R0, R1, and R2 resection margins in 60%, 32%, and 8% of patients, respectively. Approximately 81% of patients required non-bowel organ resection, including seven sacral resections. Thirty-day mortality was nil, and 5-year overall survival and disease-free survival were 33% and 27%, respectively.

SacralResection

ExternalHemipelvectomy

© MAYO2011

1 2

3 4

Figure 16 Sacropelvic resection classifi cation (types 1 to 4). Repro-duced with permission from the Mayo Clinic.

Table 3 Survival Following Exenteration for Recurrent Rectal CancerStudy, Year Number of Patients Type of Exenteration R0 (%) Survival

Saito et al, 200344 85 Posterior, combined 51 5 yr (39%)

Moriyan et al, 200487 57 Posterior, combined 84 3 yr (54%)5 yr (36%)

Melton et al, 200688 29 Anterior, posterior, combined 62 2 yr (63%)5 yr (20%)

Schurr et al, 200889 72; 45 resected Anterior, posterior, combined 82 Overall median: 54.9 mo

Colibaseanu et al, 201446 30 High posterior 93 5 yr (46%)


11/14


and guides drainage if indicated. Pelvic MRI can be particu-larly useful to elucidate fi stula tracts and the extent of sepsis. A pouchogram can assess the architecture of the pouch and demonstrate an anastomotic leak. Pouchoscopy provides information about the health of the lining of the pouch and can determine whether Crohn disease is present. An exami-nation under anesthesia may be necessary to complete the assessment of the pouch, ensure control of sepsis, and deter-mine the strategy for pouch salvage. Confi rming a diagnosis of Crohn disease is important because biologic therapy, such as infl iximab, may convert a patient with Crohn disease to a surgical candidate or suppress the disease enough to allow a pouch revision to be more successful.55,56

Although some minor reconstructive options may be performed transanally, most major pouch reconstruction options require an abdominal approach with pelvic explora-tion. Reports of reoperative pouch reconstruction success vary from 48 to 93%, with success in large part depending on the patient’s underlying disease (Crohn disease versus UC) and whether the complication is a mechanical versus an infection-related cause.52 Patients with a history of pelvic sepsis or those with a delayed diagnosis of Crohn disease often experience minimal success from reoperative pouch surgery.

Although perineal approaches such as ileal pouch advancements are well described, an abdominal approach is typically required to safely free the pouch from the sacrum and pelvic fl oor, especially in the setting of previous pelvic sepsis.57,58 When reoperating in the setting of previous pelvic sepsis, excision of all infected tissue and control of all exist-ing sinuses and fi stulas is imperative to prevent recurrence.

Operative Technique

The patient is typically placed in the Lloyd-Davies position with all extremities safely padded to avoid nerve injuries. Bilateral ureteral stents may be placed to ease the identifi cation of the ureters during surgery. A lower midline incision is made, and the abdomen is explored. The incision should be limited if possible to the level of the umbilicus as it allows the small bowel to be packed cephalad under the intact upper abdominal wall. The small bowel is often found to be adhesed to pelvic structures and to the retroperitoneu m, which requires tedious and careful lysis. The ureters and gonadal vessels should be identifi ed and looped with Silas-tic vessel loops. The course of the ureters is typically normal until the level of the pelvic brim and then becomes more medial as it enters a previously operated pelvis; therefore, identifi cation of the ureters in the lower abdomen above the pelvic brim is a good strategy to begin identifying normal anatomy.

The ileal pouch is found deep within the pelvis, often densely adherent to the anterior, lateral, and posterior walls of the pelvis. Additional proximal small bowel often fi lls the pelvis, and a signifi cant amount of time is inevitably spent lysing adhesions and obtaining orientation to achieve proper exposure. Embryonic fusion planes are nonexistent; therefore, dissection of the pouch is bloody and fraught with risk to injuring nearby structures such as the iliac and gonadal vessels, ureters, pelvic nerves, and presacral veins. It is often helpful to identify anatomic landmarks, including

Reoperative pelvic surgery for patients with recurrent or even re-recurrent rectal cancers can therefore provide signifi cant benefi t to patients, and situations thought to be nonoperable should be carefully reviewed for the possibility of an R0 resection.

complications following ipaa

Reoperative pelvic surgery in patients with infl ammatory bowel disease typically occurs in patients who have compli-cations following IPAA for ulcerative colitis (UC). Indica-tions for surgery [see Table 5] can be divided into mechanical or infection-related problems.52 Mechanical complications include anastomotic strictures refractory to anal dilation, long efferent limbs, pouch septa, or a twisted pouch. Infection-related complications from anastomotic dehiscence and pelvic sepsis may occur in approximately 25% of patients following IPAA, leading to fi brosis, chronic fi stula, and abscesses.53,54 The goals of reoperative surgery in the setting of a failing pouch will vary depending on several factors including underlying pathology (i.e., Crohn disease or malignancy) and the presence of sepsis. The type of reop-erative pouch surgery therefore varies from partial recon-struction, defi ned as pouch repair without disconnecting the anastomosis, to complete reconstruction, defi ned as pouch disconnection with redo/repair of the pouch and reanasto-mosis, to permanent pouch excision with end ileostomy.

Operative Planning

When patient symptoms suggest complications in the pouch following surgery for UC, a thorough evaluation should be undertaken to clarify the source of the problem. Imaging such as CT and MRI evaluates for pelvic infection

Table 4 Intraoperative Radiotherapy Doses Related to Resection Margin

Margin IOERT Dose (Gy)

R0 (< 5 mm) 750–1,250

R1 1,000–1,500

Localized R2 1,500–2,000

IOERT = intraoperative electron beam radiotherapy.

Table 5 Indication for Reoperative Pouch Surgery

MechanicalStricture: neoterminal ileum, pouch, anastomoticLong efferent limb with S pouchPouch septumTwisted pouchRedundant blind limb

Infection relatedFistula/chronic sinus tract: pouch-vaginal, pouch-sacral,

pouch-perirectal, pouch–abdominal wall, pouch–small bowel

Pelvic abscessAnastomotic dehiscenceRetained rectal mucosa


11/14


the normal proximal ureters, distal aorta, and common iliac vessels, before proceeding deeper into the pelvis. Iliac veins such as the left common iliac are at particular risk for injury due to their fragile thin wall, relative immobility, and trajec-tory across the midline. The dissection plane should remain anterior to the Waldeyer fascia to avoid the presacral veins.

The pouch mesentery is tethered by its main vascular sup-ply, usually from the superior mesenteric artery (SMA), and follows the curvature of the sacrum from the promontory. The surgeon should be able to raise this mesentery off the posterior wall to completely encircle the pouch mesentery. A clear path to the pouch can then be appreciated. Lateral and anterior dissection of the pouch continues to the pelvic fl oor, with care to preserve the ureters as they approach the bladder neck and genitourinary structures such as the pros-tate or vagina. Once the pouch is suffi ciently exposed, either partial reconstruction is performed, or, in cases of complete reconstruction or pouch excision, the pouch is transected sharply with a knife or scissors at the pelvic fl oor and delivered out of the abdominal fi eld.

For complete pouch reconstruction, the old pouch is excised and a new J pouch, approximately 12 to 15 cm in length, is constructed using sequential fi rings of a stapling device. Presuming that tension-free reach is feasible, a hand-sewn IPAA is then constructed using Lone Star retractors for perineal exposure. Circumferential, interrupted, full-thickness bites of absorbable sutures secure the apex of the new ileal pouch to the distal anal canal with a posterior orientation of the pouch mesentery.

For pouch excisions, we perform a mucosectomy of the remaining anal canal from the dentate line to remove any residual mucosa and close the proximal end of the anal canal using several layers of interrupted absorbable sutures placed transanally. The pouch is transected at a healthy-appearing point on the prepouch ileum, and a tension-free end ileostomy is constructed at a premarked stoma site.

Troubleshooting

The loss of normal anatomic planes in cases of pelvic sepsis, with chronic fi stulas and abscesses, makes reopera-tive pouch surgery technically challenging, and dissection may appear to be blind at certain points. The best strategy that the surgeon can employ in these circumstances is to constantly assess where the critical structures such as the ureters, bladder, and gynecologic and neurovascular struc-tures are or may be. It can be helpful to perform an endos-copy through the pouch to better delineate the size and orientation of the pouch and its efferent/afferent limbs, which may aid with pelvic dissection.

When performing complete pouch reconstruction, the neoterminal ileum must have suffi cient reach to the pelvis for a tension-free anastomosis. The same technical principles used during index pouch construction to obtain reach should be employed: complete mobilization of the small bowel mesentery off the retroperitoneum to the root of the SMA at the duodenum, sequential scoring of the peritoneum along the axis of the new pouch mesentery, and ligation of mesenteric vessels only if collateral circulation is readily identifi able to the new pouch. The apex of the pouch should once again easily reach the inferior aspect of the pubis

symphysis to estimate adequate length. The options avail-able to provide suffi cient reach for a new pouch are some-what limited due to the restricted vascular supply, and if there is suffi cient doubt regarding a tension-free reach, then an end ileostomy should be constructed.

Complications

Postoperative complications after reoperative pouch sur-gery include loss of the ileal pouch if not initially intended, pouch dysfunction, infection, hemorrhage, and/or injury to surrounding structures, such as the ureters and pelvic nerves. The preponderance of complications tends to fall in the same category as the indication for the reoperation. In an early Mayo Clinic study, postoperative infectious complica-tions such as recurrent fi stulas or abscesses were observed in up to 59% of patients after reoperative surgery for infection-related pouch problems such as chronic fi stulas.59 For mechanical problems such as anastomotic strictures, steno-sis recurred in 52% of patients after the initial reoperation. The need for more than one operation was observed in up to 73% of patients. Approximately 69% of patients who underwent more than one reoperative procedure ultimately lost their pouch compared with a 14% chance of pouch loss in those patients who only underwent one reoperative sur-gery. Careful initial pouch construction and a single, defi ni-tive reoperative surgery for pouch-related complications if they occur are therefore critical to long-term pouch success.

Outcome Evaluation

Many technical approaches to pouch reoperation, with varying success, have been described.60 Shawki and col-leagues reported on 76 patients who underwent reoperation for pouch complications.61 The most common indication for reoperation was sepsis and sepsis-related complications. Of 23 patients who required reoperative pouch surgery, seven patients underwent complete pouch reconstruction and 16 patients underwent pouch revision. The pouch salvage rate in these patients was 69%. The most common reason for pouch failure was a diagnosis of Crohn disease and pelvic sepsis. MacLean and colleagues described 63 combined abdominoperineal pouch reconstructions in 57 patients, of whom 30% underwent pouch reconstruction and 70% underwent revision.62 The primary indication for reopera-tion was pouch-vaginal fi stula and pelvic sepsis. Pouch revision success was achieved in 67% of patients.

Mathis and colleagues reported 51 patients with an initial diagnosis of UC, of whom 22 patients were subsequently proved to have Crohn disease.52 Sixty-fi ve percent under-went pouch reoperation for infectious complications and 35% underwent reoperation for mechanical complications. Complete reconstruction was performed in 43% and partial reconstruction in 57%. Pouch survival following reconstruc-tion was 93% at 1 year and 89% at 5 years. Most failed pouches (75%) occurred in patients with a later diagnosis of Crohn disease. Pouch survival in patients with a mechanical indication was 91% compared with pouch survival of 79% in those with an infectious indication. Although this is often a technically demanding surgery, these results indicate that a signifi cant proportion of pouches may be salvaged with an experienced team.


11/14


low anastomotic complications

Similar to reoperative pouch surgery, the primary indica-tions for reoperative surgery for low anastomotic complica-tions are for mechanical or infection-related complications. Mechanical problems are usually related to anastomotic strictures secondary to ischemia or surgical technique. Infection-related complications are inevitably related to anastomotic dehiscence. The overall incidence of anasto-motic leaks following low anterior resection is reported from 1 to 19%, with signifi cant challenges in management when they occur.63–66 Apart from substantial patient morbidity and the risk of mortality, resultant function is signifi cantly worse and the cost of treatment of the anastomotic leak has been shown to be substantial.67–69 Of even more concern has been the fi nding that local recurrence is increased and survival is decreased following anastomotic leak in the setting of surgery for colorectal cancer.70–73 Anastomotic leaks may also lead to chronic fi brosis and scarring and thus ultimately lead to chronic complications, including decreased neorectal capacitance, incontinence, or stricture.64,74

Operative Planning

The location, size, and containment of an anastomotic leak typically dictate how a patient clinically presents. Patients with free anastomotic leaks within the peritoneal cavity present with classic signs of hemodynamic compromise and peritonitis; however, patients with very low, well-contained leaks within the pelvis may present subclinically and pose a diagnostic challenge. In general, patients with clinical dete-rioration, sepsis, or diffuse peritonitis require immediate surgical intervention. Patients who are not systemically ill with subclinical leaks may not require immediate reoperation and can often be managed with CT- or ultrasonography-guided percutaneous drain placement for cavities greater than 3 cm versus aspiration for small cavities less than 3 cm.

Management algorithms have been developed to address anastomotic leaks.75 For leaks with extensive peritoneal contamination, signifi cant breakdown of the anastomosis, or evidence of ischemia, the operative plan requires takedown of the anastomosis with complete diversion in the form of an ileostomy or colostomy. If the anastomotic ring is mostly intact and there are no signs of bowel ischemia, then the strategy includes preserving the anastomosis, draining the region, and placing a proximal diverting loop ileostomy, which may salvage the anastomosis. Local repairs are typi-cally not successful; however, an omental fl ap in addition to proximal diversion may assist in healing of the pelvic anastomosis. Other novel techniques with limited support-ing data include covered stents, vacuum-assisted closure devices, or over-the-scope clips.76,77

The critical goal in reoperative pelvic surgery for infection-related complications is to control sepsis. Salvage or recon-struction of the anastomosis will be most successful 6 to 12 months later. In very low anastomotic leaks with a con-necting abscess, transanal drainage may be performed in the operating room under direct vision. In this instance, the anastomosis is opened and the abscess is drained, followed by placement of a catheter in the defect for irrigation. In the management of pelvic abscesses using drains, serial sino-grams must reveal no fi stula and resolution of the abscess

cavity before confi dence in the integrity of the anastomosis. If a persistent fi stula to bowel remains after months, an eventual operation may be required for anastomotic revision or diversion. A diverting proximal stoma is often required in conjunction with local measures to control sepsis.

When reoperative surgery is performed for a failed anas-tomosis, it is important to consider the tissue quality of the pelvic fl oor, sphincter function, and remaining rectum or anus. In motivated patients with intact sphincter function, reoperation with attempted coloanal anastomoses may lead to a successful outcome. In the setting of chronic pelvic sep-sis, the remaining rectal stump is often fi brotic and a high risk for a new anastomosis, but the anal canal is typically soft and pliable, providing a point for a coloanal anastomo-sis. Preoperative manometry is useful to assess sphincter function. Patients must also be aware that bowel length may limit the surgeon’s ability to create a low anastomosis and that a permanent ostomy may be required. In addition, a diverting loop ileostomy is required for at least 3 months if a low pelvic anastomosis can be created.

Operative Technique

The technical strategy of reoperative surgery for low anastomotic complications is similar to the approach used for recurrent pelvic malignancies and reoperative pouch surgery. In brief, the patient is placed in the Lloyd-Davies position to provide access to the perineum, and all extremi-ties are safely padded. Bilateral ureteral stents may be placed to ease the identifi cation of the ureters during surgery. A lower midline incision is made, and the abdomen is explored. The small bowel is packed into the upper abdo-men, and the pelvis is exposed. Critical structures, including the ureters, gonadal vessels, and iliac vessels, should be immediately identifi ed and tagged if necessary. The distal bowel should then be mobilized to the level of the anasto-mosis. It is helpful to know what type of anastomosis was made (hand sewn, stapled, end to end, or end to side) to guide the dissection. The use of a fi nger or instruments such as a rigid proctoscope or vaginal retractor may also assist with identifying important pelvic structures.

Defi nitive surgical options depend on the indication for the reoperative surgery. For mechanical problems such as a benign stricture, resection with (1) no reanastomosis versus (2) reanastomosis and possible proximal diversion requires exposure beyond the original anastomosis to allow room for distal transection. This distal margin may be at the level of the pelvic fl oor. In cases of very low anastomoses, it may be necessary to perform the distal transection from a transanal approach with an intersphincteric dissection to meet the abdominal dissection plane.

For infection-related problems such as an anastomotic leak, the surgical option depends on the extent of contami-nation, time to reoperation, degree of dehiscence, patient stability, and overall clinical assessment. These options include (1) attempted primary repair/buttressing with wide drainage and possible proximal diversion, (2) resection with reanastomosis and possible proximal diversion, or (3) resec-tion with no reanastomosis. Although fecal diversion may not prevent another postoperative complication, it is impor-tant to consider as it may decrease the magnitude of morbid-ity should another anastomotic dehiscence or deep surgical space infection recur.


11/14


Troubleshooting

Reconstruction of gastrointestinal continuity can pose a major challenge after resection of a previous low anastomo-sis. When bowel length may not seem adequate for a tension-free anastomosis, several maneuvers can be employed before resorting to an end ileostomy or colostomy [see Table 6]. These include (1) primary maneuvers: mobilization of the embryonic fusion planes and detachment of periorgan ligaments such as the splenic and hepatic fl exures; (2) secondary maneuvers: directed ligation of vascular pedicles such as the inferior mesenteric artery (IMA) and vein (IMV) that restrict mobility [see Figure 17]; and (3) tertiary maneu-vers, such as retroileal colorectal anastomoses or bowel rotations from the proximal transverse or right colon [see

Figure 18]. These techniques may have already been per-formed during the index anastomotic construction, but if not, these principles should be applied to help reestablish gastrointestinal continuity.

Complications

The spectrum of postoperative complications after reoperative pelvic surgery for anastomotic complications is determined by the indication for reoperation and the reop-erative procedure. For infection-related indications such as anastomotic leaks, early studies observed high morbidity and mortality (72%) with diverting loop colostomy and drainage versus complete anastomotic resection and end colostomy, which had no reported morbidity and mortality

Table 6 Mobilization Techniques for Diffi cult ReconstructionsManeuver Goals Example

Primary Separation of embryonic fusion planes Cattell and Mattox maneuvers

Division of peri-organ “ligaments” Splenic flexure mobilization

Secondary Ligation of vascular pedicles Ligation of the inferior mesenteric vein and artery

Preservation of collateral blood supply Preserving the middle colic artery to supply ileal pouch

Tertiary Extended resection to mobile proximal bowel Deloyers procedure and retroileal reconstruction

Stoma construction End ileostomy or colostomy

a

IMA

Factor© MAYO

2014

b

IMA

LCV

c

Figure 17 Increasing colon length with primary and secondary maneuvers. (a) Primary maneuvers such as splenic fl exure mobilization provide additional reach for the proximal colon, but maximal reach is restricted by the inferior mesentery artery (IMA). (b) After high ligation of the IMA, the colon is further restricted by the inferior mesenteric vein (IMV). Transection of the IMV must be performed proximal to the confl uence of the left colic vein (LCV). (c) Ligation of the IMV proximal to the LCV provides maximum colon reach to the pelvis for tension-free colorectal or co-loanal anastomoses. Reproduced with permission from the Mayo Clinic.


11/14


(0%).78 These early fi ndings applied the management prin-ciples used in perforated diverticulitis, which is most often treated with an end colostomy (the Hartmann procedure).

More recent studies after reoperative surgery for anasto-motic complications have reported higher morbidity, from 26 to 55%.79–81 Genser and colleagues looked at 50 consecu-tive patients who underwent reoperative surgery for anasto-motic strictures (40%) and infection-related indications such as chronic pelvic sepsis (28%) and fi stula (6%).80 All patients underwent resection and anastomosis, with 54% and 46% of these patients receiving a new colorectal or coloanal anastomosis, respectively. Mortality was nil, but morbidity approached 26%. Morbidity was generally minor (Dingo grade < 3) and included intraoperative injury to the bladder, prolonged ileus, urinary tract infections, and one readmis-sion for dehydration from a diverting ileostomy. Only one case of presacral hemorrhage (Dingo grade 4) required reoperation.

Other groups have also reported no mortality after reop-erative surgery but have shown associated morbidity.79,81 Lefevre and colleagues studied 34 patients who underwent reoperative surgery for mechanical and infection-related indications.81 The mean operating time was 279 minutes and morbidity was 55%, with a 12% anastomotic leak rate after

reanastomosis. Pitel and colleagues reported on 66 patients comprising 44 colorectal and 22 coloanal anastomoses.79 Mortality was again nil, and morbidity was 32%, consisting of a 7.7% anastomotic leak rate, 9.3% readmission rate, and 15.4% reintervention rate. Although these studies demonstrate a signifi cant risk of morbidity after reoperative surgery for anastomotic complications, the morbidity appears to be relatively minor and manageable, with no reported mortality.

Outcome Evaluation

Reoperative surgery for anastomotic complications can be safe and effective. For mechanical complications such as benign strictures, functional outcomes are excellent. Schlegel and colleagues observed 27 patients who underwent resec-tion and reanastomosis for low anastomotic strictures and found no recurrent strictures and satisfactory functional results up to 28.7 months postoperatively.82 Other series have reported restenosis rates in up to 4% of patients but similarly acceptable, long-term functional results.80,81

In the series by Genser and colleagues that included both mechanical and infection-related anastomotic complications, all patients successfully underwent reoperation and up to 94% of patients ultimately had their intestinal continuity

a

MC

RC

IC

SMA

Factor© MAYO

2014

b c

Figure 18 Tertiary maneuvers. (a) Resection of colon to the proximal transverse colon and ligation of the middle colic artery (MC) prepares for two maneuvers that allow for a tension-free, low anastomosis. (b) Deloyers procedure. Counterclockwise rotation of the remaining right colon around the axis of the superior mesenteric artery (SMA) may allow for a tension-free low pelvic anastomosis. (c) Retroileal reconstruction. Tun-neling the remaining right colon through the ileal mesentery may allow for a tension-free low pelvic anastomosis. Reproduced with permission from the Mayo Clinic. IC = ileocolic artery; RC = right colic artery.


11/14


reestablished.80 The median number of bowel movements was two during the day and zero at night, and 70% of patients had less than three stools a day up to 21 months on follow-up. Pitel and colleagues reported a median follow-up of 35.7 months and long-term success rates in 78.8% of their 66 redo colorectal and coloanal patients with a median Wexner score of 8.79 These studies suggest that although technically demanding, reoperative pelvic surgery for anas-tomotic complications is associated with resolution of the original complication, acceptable morbidity, and satisfactory long-term functional outcomes.

palliative reoperative pelvic surgery

Palliative reoperative pelvic surgery may be necessary in select patients to relieve pain from recurrent cancers, malig-nant bowel obstructions, chronic abscesses, and/or fi stulas that negatively affect quality of life. Brophy and colleagues reported on 35 patients who underwent palliative exentera-tion procedures for recurrent pelvic carcinoma with refrac-tory symptoms including pain, bleeding, obstruction, and fi stulas.83 Operative morbidity and mortality were 47% and 3%, respectively. The median overall survival was 20 months, and improvement in quality of life was reported in 88% of patients. Esnaola and colleagues prospectively assessed pain in 45 patients with unresectable recurrent rectal cancer from which 30 patients (67%) underwent a palliative resection.84 Patients who did not undergo palliative resection reported signifi cantly more pain beyond the third month of treat-ment. These studies suggest that reoperative pelvic surgery has a benefi cial role for patients even in palliative situations.

Conclusions

Reoperative pelvic surgery is a challenging but important way to address debilitating and potentially life-threatening issues such as recurrent malignancies, ileal pouch complica-tions, low anastomotic problems, and palliative situations. The safety and effectiveness of reoperative surgery are dic-tated by key technical principles, including early identifi ca-tion of critical structures, ability to control bleeding with a variety of methods, and understanding reconstruction strat-egies from a gastrointestinal, genitourinary, and musculo-skeletal perspective. Multidisciplinary teams may need to be involved in complex cases of reoperative surgery, but when executed properly, a successful outcome with acceptable low morbidity and satisfactory long-term functional results can be produced.

Financial Disclosures: Eric J. Dozois, MD, and Daniel I. Chu, MD, have no relevant fi nancial relationships to disclose.

References

1. Wanebo HJ, Koness RJ, Vezeridis MP, et al. Pelvic resection of recurrent rectal cancer. Ann Surg 1994;220:586–95; discussion 595–7.

2. Pacelli F, Tortorelli AP, Rosa F, et al. Locally recurrent rectal cancer: prognostic factors and long-term outcomes of multimodal therapy. Ann Surg Oncol 2010;17:152–62.

3. Bouchard P, Efron J. Management of recurrent rectal cancer. Ann Surg Oncol 2010;17:1343–56.

4. Sagar PM, Pemberton JH. Surgical management of locally recurrent rectal cancer. Br J Surg 1996;83:293–304.

5. McDermott FT, Hughes ES, Pihl E, et al. Local recurrence after potentially curative resection for rectal cancer in a series of 1008 patients. Br J Surg 1985;72:34–7.

6. Pilipshen SJ, Heilweil M, Quan SH, et al. Patterns of pelvic recurrence following defi nitive resections of rectal cancer. Cancer 1984;53:1354–62.

7. Rich T, Gunderson LL, Lew R, et al. Patterns of recurrence of rectal cancer after potentially curative surgery. Cancer 1983;52:1317–29.

8. Gunderson LL, Sosin H. Areas of failure found at reopera-tion (second or symptomatic look) following “curative sur-gery” for adenocarcinoma of the rectum. Clinicopathologic correlation and implications for adjuvant therapy. Cancer 1974;34:1278–92.

9. Mannaerts GH, Rutten HJ, Martijn H, et al. Abdominosacral resection for primary irresectable and locally recurrent rectal cancer. Dis Colon Rectum 2001;44:806–14.

10. Palmer G, Martling A, Cedermark B, et al. A population-based study on the management and outcome in patients with locally recurrent rectal cancer. Ann Surg Oncol 2007;14:447–54.

11. Bakx R, Visser O, Josso J, et al. Management of recurrent rectal cancer: a population based study in greater Amster-dam. World J Gastroenterol 2008;14:6018–23.

12. Heriot AG, Byrne CM, Lee P, et al. Extended radical resec-tion: the choice for locally recurrent rectal cancer. Dis Colon Rectum 2008;51:284–91.

13. Jimenez RE, Shoup M, Cohen AM, et al. Contemporary out-comes of total pelvic exenteration in the treatment of colorectal cancer. Dis Colon Rectum 2003;46:1619–25.

14. Valentini V, Morganti AG, Gambacorta MA, et al. Preop-erative hyperfractionated chemoradiation for locally recur-rent rectal cancer in patients previously irradiated to the pelvis: a multicentric phase II study. Int J Radiat Oncol Biol Phys 2006;64:1129–39.

15. Das P, Delclos ME, Skibber JM, et al. Hyperfractionated accelerated radiotherapy for rectal cancer in patients with prior pelvic irradiation. Int J Radiat Oncol Biol Phys 2010;77:60–5.

16. Deziel DJ, Kiel KD, Kramer TS, et al. Intraoperative radia-tion therapy in biliary tract cancer. Am Surg 1988;54:402–7.

17. Cummings BJ, Rider WD, Harwood AR, et al. Radical exter-nal beam radiation therapy for adenocarcinoma of the rectum. Dis Colon Rectum 1983;26:30–6.

18. Danjoux CE, Gelber RD, Catton GE, et al. Combination chemo-radiotherapy for residual, recurrent or inoperable carcinoma of the rectum: E.C.O.G. study (EST 3276). Int J Radiat Oncol Biol Phys 1985;11:765–71.

19. Rhomberg W, Eiter H, Hergan K, et al. Inoperable recurrent rectal cancer: results of a prospective trial with radiation therapy and razoxane. Int J Radiat Oncol Biol Phys 1994;30:419–25.

20. Hahnloser D, Nelson H, Gunderson LL, et al. Curative potential of multimodality therapy for locally recurrent rectal cancer. Ann Surg 2003;237:502–8.

21. Bedrosian I, Giacco G, Pederson L, et al. Outcome after curative resection for locally recurrent rectal cancer. Dis Colon Rectum 2006;49:175–82.


11/14


40. Gunderson LL, Nelson H, Martenson JA, et al. Locally advanced primary colorectal cancer: intraoperative electron and external beam irradiation +/– 5-FU. Int J Radiat Oncol Biol Phys 1997;37:601–14.

41. Henry LR, Sigurdson E, Ross EA, et al. Resection of isolated pelvic recurrences after colorectal surgery: long-term results and predictors of improved clinical outcome. Ann Surg Oncol 2007;14:2000–9.

42. Moore HG, Shoup M, Riedel E, et al. Colorectal cancer pelvic recurrences: determinants of resectability. Dis Colon Rectum 2004;47:1599–606.

43. Stocchi L, Nelson H, Sargent DJ, et al. Is en-bloc resection of locally recurrent rectal carcinoma involving the urinary tract indicated? Ann Surg Oncol 2006;13:740–4.

44. Saito N, Koda K, Takiguchi N, et al. Curative surgery for local pelvic recurrence of rectal cancer. Dig Surg 2003;20:192–9; discussion 200.

45. Schurr P, Lentz E, Block S, et al. Radical redo surgery for local rectal cancer recurrence improves overall survival: a single center experience. J Gastrointest Surg 2008;12:1232–8.

46. Colibaseanu DT, Dozois EJ, Mathis KL, et al. Extended sacropelvic resection for locally recurrent rectal cancer: can it be done safely and with good oncologic outcomes? Dis Colon Rectum 2014;57:47–55.

47. Abdelsattar ZM, Mathis KL, Colibaseanu DT, et al. Surgery for locally advanced recurrent colorectal cancer involving the aortoiliac axis: can we achieve R0 resection and long-term survival? Dis Colon Rectum 2013;56:711–6.

48. Wanebo HJ, Antoniuk P, Koness RJ, et al. Pelvic resection of recurrent rectal cancer: technical considerations and out-comes. Dis Colon Rectum 1999;42:1438–48.

49. Dresen RC, Gosens MJ, Martijn H, et al. Radical resection after IORT-containing multimodality treatment is the most important determinant for outcome in patients treated for locally recurrent rectal cancer. Ann Surg Oncol 2008;15:1937–47.

50. Haddock MG, Miller RC, Nelson H, et al. Combined modal-ity therapy including intraoperative electron irradiation for locally recurrent colorectal cancer. Int J Radiat Oncol Biol Phys 2011;79:143–50.

51. Colibaseanu DT, Mathis KL, Abdelsattar ZM, et al. Is cura-tive resection and long-term survival possible for locally re-recurrent colorectal cancer in the pelvis? Dis Colon Rectum 2013;56:14–9.

52. Mathis KL, Dozois EJ, Larson DW, et al. Outcomes in patients with ulcerative colitis undergoing partial or complete reconstructive surgery for failing ileal pouch-anal anastomosis. Ann Surg 2009;249:409–13.

53. Scott NA, Dozois RR, Beart RW Jr, et al. Postoperative intra-abdominal and pelvic sepsis complicating ileal pouch-anal anastomosis. Int J Colorectal Dis 1988;3:149–52.

54. Williams NS, Johnston D. The current status of mucosal proctectomy and ileo-anal anastomosis in the surgical treat-ment of ulcerative colitis and adenomatous polyposis. Br J Surg 1985;72:159–68.

55. Matzke GM, Kang AS, Dozois EJ, et al. Mid pouch stricture-plasty for Crohn’s disease after ileal pouch-anal anastomo-sis: an alternative to pouch excision. Dis Colon Rectum 2004;47:782–6.

56. Colombel JF, Ricart E, Loftus EV Jr, et al. Management of Crohn’s disease of the ileoanal pouch with infl iximab. Am J Gastroenterol 2003;98:2239–44.

22. Chessin DB, Hartley J, Cohen AM, et al. Rectus fl ap recon-struction decreases perineal wound complications after pelvic chemoradiation and surgery: a cohort study. Ann Surg Oncol 2005;12:104–10.

23. Bullard KM, Trudel JL, Baxter NN, et al. Primary perineal wound closure after preoperative radiotherapy and abdom-inoperineal resection has a high incidence of wound failure. Dis Colon Rectum 2005;48:438–43.

24. Liebermann-Meffert D. The greater omentum. Anatomy, embryology, and surgical applications. Surg Clin North Am 2000;80:275–93, xii.

25. Radice E, Nelson H, Mercill S, et al. Primary myocutaneous fl ap closure following resection of locally advanced pelvic malignancies. Br J Surg 1999;86:349–54.

26. Soper JT, Secord AA, Havrilesky LJ, et al. Rectus abdominis myocutaneous and myoperitoneal fl aps for neovaginal reconstruction after radical pelvic surgery: comparison of fl ap-related morbidity. Gynecol Oncol 2005;97:596–601.

27. Kandeel A, Meguid A, Hawasli A. Controlling diffi cult pelvic bleeding with argon beam coagulator during laparo-scopic ultra low anterior resection. Surg Laparosc Endosc Percutan Tech 2011;21:e21–3.

28. Civelek A, Yegen C, Aktan AO. The use of bonewax to con-trol massive presacral bleeding. Surg Today 2002;32:944–5.

29. Harrison JL, Hooks VH, Pearl RK, et al. Muscle fragment welding for control of massive presacral bleeding during rectal mobilization: a review of eight cases. Dis Colon Rectum 2003;46:1115–7.

30. Yeo HL, Paty PB. Management of recurrent rectal cancer: practical insights in planning and surgical intervention. J Surg Oncol 2014;109:47–52.

31. Kakuda JT, Lamont JP, Chu DZ, et al. The role of pelvic exenteration in the management of recurrent rectal cancer. Am J Surg 2003;186:660–4.

32. Estes NC, Thomas JH, Jewell WR, et al. Pelvic exenteration: a treatment for failed rectal cancer surgery. Am Surg 1993;59:420–2.

33. Magrini S, Nelson H, Gunderson LL, et al. Sacropelvic resection and intraoperative electron irradiation in the man-agement of recurrent anorectal cancer. Dis Colon Rectum 1996;39:1–9.

34. Law WL, Chu KW, Choi HK. Total pelvic exenteration for locally advanced rectal cancer. J Am Coll Surg 2000;190:78–83.

35. Ike H, Shimada H, Ohki S, et al. Outcome of total pelvic exenteration for locally recurrent rectal cancer. Hepatogas-troenterology 2003;50:700–3.

36. Melton GB, Paty PB, Boland PJ, et al. Sacral resection for recurrent rectal cancer: analysis of morbidity and treatment results. Dis Colon Rectum 2006;49:1099–107.

37. Weber KL, Nelson H, Gunderson LL, et al. Sacropelvic resection for recurrent anorectal cancer. A multidisciplinary approach. Clin Orthop Relat Res 2000;(372):231–40.

38. Moriya Y, Akasu T, Fujita S, et al. Total pelvic exenteration with distal sacrectomy for fi xed recurrent rectal cancer in the pelvis. Dis Colon Rectum 2004;47:2047–53; discussion 2053–4.

39. Dozois EJ, Privitera A, Holubar SD, et al. High sacrectomy for locally recurrent rectal cancer: can long-term survival be achieved? J Surg Oncol 2011;103:105–9.


11/14


57. Fazio VW, Tjandra JJ. Pouch advancement and neoileoanal anastomosis for anastomotic stricture and anovaginal fi stula complicating restorative proctocolectomy. Br J Surg 1992;79:694–6.

58. Fazio VW, Tjandra JJ. Transanal mucosectomy. Ileal pouch advancement for anorectal dysplasia or infl ammation after restorative proctocolectomy. Dis Colon Rectum 1994;37:1008–11.

59. Galandiuk S, Scott NA, Dozois RR, et al. Ileal pouch-anal anastomosis. Reoperation for pouch-related complications. Ann Surg 1990;212:446–52; discussion 452–4.

60. Fazio VW, Wu JS, Lavery LC. Repeat ileal pouch-anal anas-tomosis to salvage septic complications of pelvic pouches: clinical outcome and quality of life assessment. Ann Surg 1998;228:588–97.

61. Shawki S, Belizon A, Person B, et al. What are the outcomes of reoperative restorative proctocolectomy and ileal pouch-anal anastomosis surgery? Dis Colon Rectum 2009;52:884–90.

62. MacLean AR, O’Connor B, Parkes R, et al. Reconstructive surgery for failed ileal pouch-anal anastomosis: a viable surgical option with acceptable results. Dis Colon Rectum 2002;45:880–6.

63. Golub R, Golub RW, Cantu R Jr, et al. A multivariate analysis of factors contributing to leakage of intestinal anastomoses. J Am Coll Surg 1997;184:364–72.

64. Heald RJ, Leicester RJ. The low stapled anastomosis. Br J Surg 1981;68:333–7.

65. Vignali A, Fazio VW, Lavery IC, et al. Factors associated with the occurrence of leaks in stapled rectal anastomoses: a review of 1,014 patients. J Am Coll Surg 1997;185:105–13.

66. Petersen S, Freitag M, Hellmich G, et al. Anastomotic leak-age: impact on local recurrence and survival in surgery of colorectal cancer. Int J Colorectal Dis 1998;13:160–3.

67. Fielding LP, Stewart-Brown S, Blesovsky L, et al. Anasto-motic integrity after operations for large-bowel cancer: a multicentre study. Br Med J 1980;281:411–4.

68. Ansari MZ, Collopy BT, Hart WG, et al. In-hospital mortal-ity and associated complications after bowel surgery in Victorian public hospitals. Aust N Z J Surg 2000;70:6–10.

69. Braga M, Vignali A, Zuliani W, et al. Laparoscopic versus open colorectal surgery: cost-benefi t analysis in a single-center randomized trial. Ann Surg 2005;242:890–5, discus-sion 895–6.

70. Branagan G, Finnis D. Prognosis after anastomotic leakage in colorectal surgery. Dis Colon Rectum 2005;48:1021–6.

71. Merkel S, Wang WY, Schmidt O, et al. Locoregional recur-rence in patients with anastomotic leakage after anterior resection for rectal carcinoma. Colorectal Dis 2001;3:154–60.

72. Chang SC, Lin JK, Yang SH, et al. Long-term outcome of anastomosis leakage after curative resection for mid and low rectal cancer. Hepatogastroenterology 2003;50:1898–902.

73. Walker KG, Bell SW, Rickard MJ, et al. Anastomotic leakage is predictive of diminished survival after potentially

curative resection for colorectal cancer. Ann Surg 2004;240:255–9.

74. Nesbakken A, Nygaard K, Lunde OC. Outcome and late functional results after anastomotic leakage following mesorectal excision for rectal cancer. Br J Surg 2001;88:400–4.

75. Phitayakorn R, Delaney CP, Reynolds HL, et al. Standard-ized algorithms for management of anastomotic leaks and related abdominal and pelvic abscesses after colorectal surgery. World J Surg 2008;32:1147–56.

76. Geiger TM, Miedema BW, Tsereteli Z, et al. Stent placement for benign colonic stenosis: case report, review of the litera-ture, and animal pilot data. Int J Colorectal Dis 2008;23:1007–12.

77. Glitsch A, von Bernstorff W, Seltrecht U, et al. Endoscopic transanal vacuum-assisted rectal drainage (ETVARD): an optimized therapy for major leaks from extraperitoneal rectal anastomoses. Endoscopy 2008;40:192–9.

78. Mileski WJ, Joehl RJ, Rege RV, et al. Treatment of anasto-motic leakage following low anterior colon resection. Arch Surg 1988;123:968–71.

79. Pitel S, Lefevre JH, Tiret E, et al. Redo coloanal anastomosis: a retrospective study of 66 patients. Ann Surg 2012;256:806–10; discussion 810–1.

80. Genser L, Manceau G, Karoui M, et al. Postoperative and long-term outcomes after redo surgery for failed colorectal or coloanal anastomosis: retrospective analysis of 50 patients and review of the literature. Dis Colon Rectum 2013;56:747–55.

81. Lefevre JH, Bretagnol F, Maggiori L, et al. Redo surgery for failed colorectal or coloanal anastomosis: a valuable surgical challenge. Surgery 2011;149:65–71.

82. Schlegel RD, Dehni N, Parc R, et al. Results of reoperations in colorectal anastomotic strictures. Dis Colon Rectum 2001;44:1464–8.

83. Brophy PF, Hoffman JP, Eisenberg BL. The role of palliative pelvic exenteration. Am J Surg 1994;167:386–90.

84. Esnaola NF, Cantor SB, Johnson ML, et al. Pain and quality of life after treatment in patients with locally recurrent rectal cancer. J Clin Oncol 2002;20:4361–7.

85. Tilney HS, Tekkis PP, Sains PS, et al. Factors affecting circumferential resection margin involvement after rectal cancer excision. Dis Colon Rectum 2007;50:29–36.

86. Schneebaum S, Arnold MW, Young D, et al. Role of carcino-embryonic antigen in predicting resectability of recurrent colorectal cancer. Dis Colon Rectum 1993;36:810–5.

87. Moriya Y, Akasu T, Fujita S, et al. Total pelvic exenteration with distal sacrectomy for fi xed recurrent rectal cancer in the pelvis. Dis Colon Rectum 2004;47:2047–53; discussion 2053–4.

88. Melton GB, Paty PB, Boland PJ, et al. Sacral resection for recurrent rectal cancer: analysis of morbidity and treatment results. Dis Colon Rectum 2006;49:1099–107.

89. Schurr P, Lentz E, Block S, et al. Radical redo surgery for local rectal cancer recurrence improves overall survival: a single center experience. J Gastrointest Surg 2008;12:1232–8.

reoperative pelvic surgery · several controversies exist regarding surgery for locally recurrent...

Documents