Keywords
Pelvic exenteration - Rectus abdominis - Musculocutaneous flap
INTRODUCTION
Although perineal reconstruction following pelvic exenteration is a challenge for
plastic surgeons, it has not been performed as often as head and neck reconstruction
following malignancy due to the high rate of donor morbidity and the fact that reconstruction
is not indicated in all cases. Nevertheless, perineal reconstruction has its advantages
in preventing complications of pelvic exenteration by obliterating the pelvic dead
space; it also minimizes the donor scar by using the previous abdominal incision and
a vertical rectus abdominis musculocutaneous (VRAM) flap. However, few studies have
compared the complications and the outcomes following pelvic exenteration between
cases with and without a VRAM flap. This study aimed to compare the complications
and the outcomes following pelvic exenteration with or without VRAM flap coverage.
METHODS
All patients who underwent pelvic exenteration and immediate reconstruction with a
pedicled VRAM flap (n=9) between 2011 and 2013 at our center were retrospectively
reviewed. We reviewed the patient, tumor, and treatment characteristics; surgical
outcomes; and postoperative donor- and recipient-site complications. We also collected
information on patients who underwent pelvic exenteration without any reconstruction
surgery (n=9) between 2011 and 2013 to serve as a control group for comparison with
the VRAM group. Patients with incomplete data on postoperative complications were
excluded from this study.
The surgery was performed under general anesthesia in all cases, and the patient was
placed in the lithotomy position. Pelvic exenteration was first done according to
the range of the tumor. Using a hand-held Doppler ultrasound device, one can easily
identify the perforators. Skin paddle dimensions ranged from 5 cm×12 cm to 8 cm×13
cm, with the longest axis oriented vertically. The dimensions of the flap were tailored
to the size of the perineal defect. The flap was elevated vertically from the costal
margin to the level of the inguinal fold, and the longitudinal incision line was made
to include the pre-existing laparotomy incision ([Fig. 1]). Subcutaneous dissection was carried out to the level of the rectus sheath where
the lateral border was incised. The rectus muscle was then elevated off the posterior
sheath. Thereafter, the deep inferior epigastric vessels were identified. At this
point, the pedicle was mobilized to its origin and separated from all surrounding
tissue attachments to prevent kinking or compression. A tunnel was then made beneath
the pubic ramus, and the flap was delivered into the perineum. The muscle was divided
from the skin island superiorly by using electrocautery, and ligation of the superior
epigastric vessels was performed with vessel clips. The flap was rotated medially
and transposed onto the perineal defect to obliterate the maximum amount of dead space
possible and to prevent tension. A suction drainage system was placed in the pelvic
space. The skin paddle of the flap was then inset into the defect margin.
Fig. 1 A design of the incision line
A longitudinal incision line should include the pre-existing laparotomy incision.
Postoperatively, the flap was monitored by conventional techniques, which included
Doppler ultrasonography. The patients were placed on strict bed rest on an air mattress
for at least 96 hours. Postoperative positioning was in thigh abduction with a pillow
placed under the sacrum to limit pressure on the perineal wound during hospitalization.
Adequate perioperative prophylactic antibiotic therapy was administered for 7 days,
and intravenous prostaglandin was given to all patients until 5 days postoperatively.
For the comparison of complications between the two groups, we applied the classification
of surgical complications first proposed by Clavien et al. [1] According to the relevant criteria, each complication was classified into one or
more of six categories. The point value for each complication was compared between
the groups as described by the comprehensive complication index (CCI) [2].
RESULTS
Among the 9 patients in the VRAM flap group, one showed mild dehiscence (11%) and
another showed wound infection (11%). The overall characteristics of the patients
are listed in [Table 1]. In the patient with the wound infection, the flap was originally repositioned at
its primary site (abdomen), since there was no need for reconstruction after flap
elevation. After repositioning, signs of infection followed, and for the prevention
of peritonitis, flap excision was performed. Peritoneal complications that frequently
occur when direct closure is performed, including bowel obstruction, peritonitis,
perineal oozing, or seroma formation, did not occur in the patients who underwent
reconstruction. Other significant complications such as total necrosis or hematoma
also did not occur, and the patients reported being generally satisfied with their
appearance.
Table 1.
Demographic characteristics of the patients
|
No.
|
Age (yr)
|
Location
|
Histology
|
Resection
|
Complications
|
|
AC, adenocarcinoma; SCC, squamous cell carcinoma.
|
|
1
|
59
|
Rectum
|
AC
|
Posterior
|
Mild dehiscence
|
|
2
|
65
|
Cervix
|
SCC
|
Anterior
|
None
|
|
3
|
55
|
Cervix
|
SCC
|
Anterior
|
Partial necrosis
|
|
4
|
71
|
Cervix
|
SCC
|
Anterior
|
None
|
|
5
|
60
|
Cervix
|
SCC
|
Anterior
|
None
|
|
6
|
70
|
Cervix
|
SCC
|
Anterior
|
None
|
|
7
|
62
|
Cervix
|
SCC
|
Anterior
|
None
|
|
8
|
61
|
Cervix
|
SCC
|
Anterior
|
None
|
|
9
|
65
|
Rectum
|
AC
|
Posterior
|
None
|
In contrast, in the control group, peritonitis and bowel obstruction occurred in 1
case (11%) ([Table 2]).
Table 2.
Outline of postoperative complications
|
Characteristic
|
Pelvic exenteration with VRAM flap coverage
|
Pelvic exenteration without reconstruction
|
|
Values are presented as number (%).
VRAM, vertical rectus abdominis musculocutaneous.
|
|
No. of patients (n)
|
9
|
9
|
|
Dehiscence
|
1 (11)
|
0 (0)
|
|
Wound infection
|
1 (11)
|
0 (0)
|
|
Partial flap necrosis
|
1 (11)
|
0 (0)
|
|
Total flap necrosis
|
0 (0)
|
0 (0)
|
|
Seroma
|
0 (0)
|
0 (0)
|
|
Peritonitis
|
0 (0)
|
1 (11)
|
|
Bowel obstruction
|
0 (0)
|
1 (11)
|
Each complication was classified according to the Clavien-Dindo classification, and
the sum of the CCI values was compared between the groups. The average CCI values
were 12 and 16.2 in the reconstructed group and the control group, respectively.
Case 1
A 59-year-old woman with adenocarcinoma of the rectum underwent immediate perineum
and pelvic cavity reconstruction after posterior pelvic exenteration. A 7 cm×12 cm
right-pedicled VRAM flap was elevated and inset in the pelvic floor and the perineal
wound. Minor wound dehiscence in the gluteal fold developed on day 7 and was treated
with debridement and delayed repair. There were no significant donor-site complications.
At the 2-month follow-up examination, the integrity of the perineum was maintained,
and the patient was satisfied with the result ([Fig. 2]).
Fig. 2 A case of posterior exenteration
(A) Intraoperative view of flap harvest. (B) Immediate postoperative photograph of
the perineum. (C) Abdominal donor site after healing.
Case 2
A 65-year-old woman with squamous cell carcinoma of the cervix with bladder, vulva,
and rectum wall invasion underwent immediate perineal reconstruction after anterior
pelvic exenteration. An 8 cm×10 cm left-pedicled VRAM flap was used for pelvic wound
closure and obliteration of the pelvic dead space. The postoperative course was stable
without any significant complications. At the 1-month follow-up examination, the status
of the flap was acceptable and the patient was satisfied with the result ([Fig. 3]).
Fig. 3 A case of anterior exenteration
(A) The perineal defect before flap coverage with the flap inside the abdominal cavity.
(B) Immediately after insetting of the flap. (C) Healing at 6 months postoperatively.
Case 3
A 55-year-old woman with recurrent squamous cell carcinoma of the cervix had planned
to undergo pelvic exenteration and immediate reconstruction. At the beginning of the
surgery, VRAM flap elevation was done prior to performing pelvic exenteration. However,
after the procedure, there appeared to be sufficient perineal skin and therefore,
no need for flap reconstruction. The flap was therefore relocated to the original
site. Excessive tension was noted during the closure of the fascia performed by gynecologists,
and partial necrosis of the flap was observed on postoperative day 2. Pus and other
infection signs occurred the following day. For the prevention of pan-peritonitis
resulting from an infection of the surgical site, excision of the VRAM flap and anterior
rectus fascia reconstruction using an artificial mesh was performed on postoperative
day 4. Thereafter, wound dehiscence and sloughing of the lower abdomen were observed,
but healing was good after the debridement ([Fig. 4]).
Fig. 4 A case with flap removal and dehiscence
(A) The preoperative diagram on the left hemiabdomen. (B) The intraoperative view
of the deep inferior epigastric pedicle. The flap was relocated to the abdomen donor
site. (C) After flap removal due to excessive tension, mild wound dehiscence occurred
on the lower abdomen.
DISCUSSION
Pelvic exenteration is an operation performed for the treatment of pelvic organ malignancy.
Despite the destructive properties of the operation, there have been only limited
attempts to perform immediate postoperative reconstruction. This is primarily due
to the prolonged operating time required and the fact that primary closure is possible
in most cases. After pelvic exenteration, 38% of the patients had more than one life-threatening
surgical complication such as sepsis, peritonitis, bowel obstruction, and incarceration,
and 29% of the patients needed additional surgical treatment such as exploratory laparotomy,
bowel resection, and delayed repair [3]. In other published studies, mortality rates as high as 40% to 60% have been reported
for reoperation to correct intestinal and genitourinary complications following exenteration
[4]. This is related to the large, non-collapsible dead space and the irradiated, poorly
vascularized field ([Fig. 5]). Therefore, considering the frequency of complications after pelvic exenteration,
immediate steps to structurally prevent these complications can be a cost-effective
solution.
Fig. 5 A case of posterior pelvic exenteration without any reconstruction
Mild increased tension of the perineal skin flap was seen at 3 days postoperatively.
Pelvic prolapse of the small bowel can cause bowel adhesion, obstruction, and fistula
formation. Moreover, it can influence the recurrence of the cancer, as it limits the
range and the dose of the radiation treatment, thereby reducing the effectiveness
of the treatment. Dead space in the pelvis is the greatest source of postoperative
morbidity in those who have undergone exenteration, particularly following treatment
with radiation. An approximation of local tissues and primary wound healing is often
not possible because of the lack of mobility of the pelvis and soft tissues. Even
if closure is achieved without tension, the irradiated tissue may not heal. Delayed
perineal wound healing adds to patient discomfort, increased hospital stay, and the
possibility of additional operative procedures [5]. To prevent these complications, a tension-free coverage of dead space with a well-vascularized
tissue is required. The well-vascularized tissue is metabolically active and therefore
facilitates the delivery of oxygen, products of the immune system, and systemically
administered antibiotics and chemotherapeutics to the cells.
Furthermore, the creation of an abdominopelvic partition prevents small bowel prolapse,
which can lead to radiation enteritis, bowel obstruction, or fistula formation [5]. The use of silicone implants and absorbable meshes as abdominopelvic partitions
has been reported. However, procedures using artificial materials have several morbidities
such as inflammation and capsule formation [6]. Alternatively, an autologous tissue can be used. There are several options, including
omentum and myocutaneous flaps. The omentum can reach into the pelvis, bringing vascularized
tissue supplied by the gastroepiploic artery. However, it lacks tensile strength and
can result in herniation of the bowel through the perineal defect. Although it can
cover the raw surface of the pelvis, it does not effectively exclude the bowel; therefore,
morbidity remains high [3]. A myocutaneous flap is an ideal option, as it contains a mass of vascular tissue
and facilitates wound healing by providing a greater blood supply [7]. While skin necrosis can occur, the muscle is often viable and granulates like normal
tissue with wound closure in weeks compared with the months needed in situations where
no myocutaneous flap was brought into the wound.
Since McCraw et al. [8] introduced the use of the gracilis musculocutaneous pedicled flap for the reconstruction
of the pelvis and the perineum, it has been the workhorse for pelvic reconstructive
surgeons. However, experience has shown that the viability of the flap is unreliable,
particularly in the distal third; in large pelvic defects, the muscle often lacks
sufficient length and bulk and does not carry a skin paddle well. Although the morbidity
of exenterations is lessened with the use of a gracilis flap, it remains high. Partial
or total flap loss may occur in 10% to 24% of the patients. As many as 37% of the
patients require a subsequent procedure or experience delayed perineal wound healing,
particularly if the patient has undergone irradiation [9].
The rectus abdominis myocutaneous flap was first established by Mathes and Bostwick
[10]. Its use has since been extended to a wide variety of indications related to the
anterior trunk and groin region. The vascular supply of the rectus muscle is from
both the inferior and the superior epigastric vessels. A transpelvic modification
of this flap was originally described in 1984 by Shukla and Hughes [11]. It provides more rapid and suitable healing of the perineal wound postexenteration,
and prevents several postoperative complications.
Although there are other several perforator flap options that can cover a pelvic defect,
a myocutaneous flap should be the first preference because of the need for filling
the potential dead space in the pelvic cavity. Perforator flaps based on arteries
such as the pudendal artery, superior and inferior gluteal artery, and perineal artery
can be applied after pelvic exenteration; they are often limited to small-to-moderate
skin defects or vaginal reconstruction [12].
A limitation of this study was its small sample size, and for this reason, we could
not compare the data by using a conventional method of statistical analysis. Instead,
we used the method of classifying surgical complications and the CCI to compare the
severity of complications between the two study groups objectively.
As our case reports show, perineal reconstruction using a VRAM flap is an effective
surgical procedure for both anterior and posterior pelvic exenteration. Further, it
maintains the physical integrity of the patient and prevents major postoperative complications
such as bowel obstruction and infection by obliterating the dead space of the pelvic
cavity. Perineal reconstruction after pelvic exenteration using a VRAM flap can result
in a clinical outcome superior to that of the group without reconstruction in addition
to the prevention of major complications. However, as shown in one case, partial necrosis
can occur due to excessive tension at closure. To obtain stronger evidence to support
the use of this procedure, there should be additional research examining a larger
number of cases.
