- Email: [email protected]
Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations
YGYNO-976604; No. of pages: 6; 4C: 3 Gynecologic Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations D Cibula a,⁎, M Zikan a, D Fischerova a, R Kocian a, A Germanova a, A Burgetova b, L Dusek c, Z Fartáková a, M Schneiderová e, K Nemejcová d, J Slama a a
Gynecologic Oncology Center, Department of Obstetrics and Gynecology, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic Department of Radiodiagnostics, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic Institute for Biostatistics and Analyses; Faculty of Medicine, Masaryk University, Brno, Czech Republic d Department of Pathology, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic e Department of Surgery, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic b c
H I G H L I G H T S • • • •
Our approach is inventive in the reconstruction goal and the modification of the flap. Pelvic floor reconstruction was associated with a marked decrease of severe post-operative morbidity and improvement of quality of life. The MRAM group showed fewer complications, including reoperations, related to symptoms of the empty pelvis syndrome. There was no case of flap necrosis, no paraastomal hernia and only one case of the incisional hernia.
a r t i c l e
i n f o
Article history: Received 22 November 2016 Received in revised form 8 January 2017 Accepted 11 January 2017 Available online xxxx Keywords: Pelvic exenteration Empty pelvic syndrome Pelvic floor reconstruction Muscular flap
a b s t r a c t Objective. To describe the technique and report experiences with pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after extensive pelvic procedures. Methods. Surgical technique of MRAM harvest and transposition is carefully described. The patients in whom pelvic floor reconstruction with MRAM after either infralevator pelvic exenteration and/or extended lateral pelvic sidewall excision was carried out were enrolled into the study (MRAM group, n = 16). Surgical data, post-operative morbidity, and disease status were retrospectively assessed. The results were compared with a historical cohort of patients, in whom an exenterative procedure without pelvic floor reconstruction was performed at the same institution (control group, n = 24). Results. Both groups were balanced in age, BMI, tumor types, and previous treatment. Substantially less patients from the MRAM group required reoperation within 60 days of the surgery (25% vs. 50%) which was due to much lower rate of complications potentially related to empty pelvis syndrome (1 vs. 7 reoperations) (p = 0.114). Late post-operative complication rate was substantially lower in the MRAM group (any grade: 79% vs. 44%; grade ≥ 3: 37% vs. 6%) (p = 0.041). The performance status 6 months after the surgery was ≤1 in the majority of patients in MRAM (81%) while in only 38% of patients from the control group (p = 0.027). There was one incisional hernia in MRAM group while three cases were reported in the controls. Conclusions. Pelvic floor reconstruction by MRAM in patients after pelvic exenterative procedures is associated with a substantial decrease in postoperative complications that are potentially related to empty pelvis syndrome. © 2017 Elsevier Inc. All rights reserved.
Introduction
⁎ Corresponding author at: Gynecologic Oncology Center, Department of Obstetrics and Gynecology, Apolinarska 18, Prague 2, 120 00 Czech Republic. E-mail address: [email protected] (D. Cibula).
Extensive surgical procedures, including pelvic exenterations (PE) or extended lateral pelvic sidewall excisions (ELSE), belong to a standard spectrum of procedures in gynecologic oncology even in the 21st century. In spite of new developments in radiotherapy, it remains the only option for management with a curative potential for a selected group of
http://dx.doi.org/10.1016/j.ygyno.2017.01.014 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
2
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
patients. Long-term survival oscillates above 50% in the majority of groups after PE performed for mixed pelvic cancers, which is an outstanding perspective for those who would have otherwise lived less than a year with a quickly deteriorating quality of life in most cases [1–6]. The leading indication remains local pelvic tumor recurrence, especially after previous radiotherapy, while less frequently these procedures are performed as a primary treatment or for palliation. Improved perioperative care has minimized the mortality associated with these procedures; at the same time, in spite of new operative techniques and sophisticated intraoperative and postoperative care, morbidity remains the biggest challenge. The most challenging are serious postoperative complications requiring reoperations, which often have long-term consequences for patients' quality of life. Empty pelvis syndrome is one of the major causes of severe post-operative morbidity after pelvic exenterations [7]. The syndrome entails a variety of symptoms, including abscess formation, bowel obstruction caused by firm adhesions to the denuded structures of the pelvic side wall, permanent discharge, bowel perforation and fistulas, all with one common cause: the dead space in the pelvis [8]. While similar spaces after the removal of organs in other regions of the abdominal cavity, for instance after splenectomy, liver or bowel resection, are easily filled by the bowels and other organs, the space deep in the pelvis remains empty due to short bowel mesentery, especially if the empty space is extended laterally beyond the endopelvic fascia. A variety of reconstructive procedures using the omentum, absorbable mashes, or silicon expanders have been used to fill in the dead space and create a mechanical support for the small bowels with mostly unsatisfactory results. Lower morbidity was, however, universally reported after either perineal reconstruction or neovagina formation by myocutaneous flaps, most often using the rectus abdominis muscle [9–11]. Since the main purpose of such reconstructions is to cover a large perineal defect or to form a neovagina, the flap must be composed of the muscle, anterior fascial sheath and skin. MRAM flap procedure has been performed in our institution for a pelvic reconstruction in patients who have undergone PE and/or ELSE, and in whom no vaginal or perineal reconstruction is planned, since 2012. Since MRAM transposition is carried out primarily for pelvic floor reconstruction, but not for vaginal or perineal reconstruction, it does not require skin and anterior sheath resection. MRAM is composed of the rectus abdominis muscle, posterior fascial sheath and peritoneum, while the anterior fascial sheath and the skin are preserved intact. Unilateral or bilateral flaps are transposed into the horizontal position in the pelvis at the level of the former pelvic diaphragm, so a new and compact pelvic muscle floor is formed in order to fill in the dead space in the pelvis, create an artificial diaphragm for small bowel support and form a mechanical barrier preventing the small bowels to adhere deeply in the pelvis to the denuded obturator fossa, large vessels, and pelvic bones. The aim of this paper is to describe the technique and report experiences with pelvic floor reconstruction. Intra-operative and post-operative morbidity were compared to the historical group of patients who underwent exenterative procedures without pelvic floor reconstruction.
All intra-operative and early postoperative complications were classified using a Dindo-Clavien 5 grade system [12]. In short, grades 1 and 2 are characterized by a need for pharmacological treatment; grade 3 required surgical or radiological intervention; grade 4 stands for lifethreatening complication; and grade 5 for a death of a patient. Complications which occurred during the surgical procedure were reported as intra-operative while those presented within 30 days after the surgery as early post-operative. Performance status was evaluated according to the ECOG scale, before the surgery and at 6 ± 2 months after the surgery. Preoperative marking of the location of the ostomy/ies was routinely performed by a stomal therapist; usually, two potential sites were marked, one at the optimal (usual) site, and the other one shifted 3– 5 cm laterally for the case of harvesting the MRAM. Anatomy of the MRAM flap. The rectus abdominis muscle is a paired muscle running vertically on each side of the anterior wall of the abdomen. The muscle is contained in the rectus sheath consisting of the aponeurosis of the lateral abdominal muscles; above the arcuate line, the internal oblique aponeurosis splits to envelop the rectus abdominis muscle, while below the arcuate line the aponeurosis is localized only anterior to the muscle (Fig. 1). The rectus abdominis has two dominant vascular pedicles. The inferior epigastric vessels, branches of external iliac vessels, which run and branch on the posterior surface of the muscle, and the superior epigastric vessels. Both vascular systems are richly connected to each other so the entire muscle can be fed by only one preserved pedicle. Inferior epigastric vessels serve as the vascular pedicle for the MRAM flap. Technique of MRAM flap harvest and transposition. MRAM was harvested either on one side or bilaterally, in its short version (until the level of umbilicus) or long version (until its attachment to ribs). The type of MRAM and number of flaps (unilateral or bilateral) was determined by the size and the shape of defect after the exenterative part of the surgery. Usually, the total infralevatoric PE and PE combined with ELSE required a bilateral flap to cover the whole pelvis and to fill in the deep pelvic defect below the reconstructed muscle floor. First, the muscle was exposed medially by opening the linea alba and rectus sheath medially and then the muscle was carefully dissected from the anterior rectus sheath. Since exenterative procedure was usually performed from the lower midline laparotomy, it had to be
Methods Patient selection and clinical assessment. All patients in whom pelvic floor reconstruction with MRAM after extensive pelvic procedure, either infralevator PE and / or ELSE, was carried out were enrolled into the study. The patients in whom rectus muscle flap was used for vaginal or perineal reconstruction were not included in the study since in these cases the flap was composed of a complete fascial sheath and skin. Surgical data, post-operative complications, and disease status were retrospectively assessed. The results were compared with a historical cohort of patients, in whom an exenterative procedure without pelvic floor reconstruction was performed.
Fig. 1. Anatomy of vascular supply and fascia rectus sheath below and above the arcuate line.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
extended up to the xiphoideus if a long version was executed. Maximum care was taken not to open the anterior fascia, especially at the level of intersections, where the dissection is more demanding. Next, the muscle was divided cranially, either at the level of the umbilicus (short version) or at the attachment to the ribs (long version), using monopolar diathermy. Finally, the flap was freed laterally, at a lateral margin of the muscle, starting from the cranial edge and heading caudally until above the course of the previously detected epigastric vessels. While the anterior surface of the final flap is formed by the exposed muscle, dorsally the muscle is covered by the peritoneum and posterior rectus sheath above arcuate line. It is important to prevent a separation of the muscle and the fascia / peritoneum, especially below the arcuate line, since the feeding fragile vessels run on the dorsal aspect of the muscle. The attachment of the muscle to the pubic bone was preserved, and inferior epigastric vessels were not dissected down to the external iliac vessels (Fig. 2). After the bleeding was controlled, the flap was carefully rotated at 180o into the pelvis. In the case of bilateral flaps, their placement was adjusted to the shape of the empty space. In most cases, both flaps were positioned next to each other in a U-shape, forming one muscle floor. In a few cases, usually after simultaneous vulvectomy or extensive vaginectomy that caused a large space deep in the pelvis, one flap was placed straight towards the sutured perineal skin, filling the deep part of the pelvic space, while the U-shaped diaphragm was created by the other side flap, mimicking the pelvic floor. Flaps were fixed in their final positions with a few stitches, attaching the whole thickness of the flap (muscle, fascia, peritoneum) to the suitable structures in the pelvis (pre-sacral fascia, obturator fascia, iliac fascia) (Fig. 3). Finally, the integrity of the preserved anterior rectus sheath was carefully checked and any defect was sutured before the laparotomy was finally closed. The fascia was not augmented by any alloplastic mesh. Statistical analysis Standard summary statistics were used to describe primary data. Fisher exact test and Mann/Whitney U test were applied to test statistical significance of differences between groups of patients. Results MRAM was used in 11 consecutive women after PE (infralevatoric in all cases), in 3 cases after ELSE, and in 2 cases after a combined procedure (PE + ELSE). MRAM was harvested bilaterally in 7 cases (44%) as short (N = 2) or long (N = 5) version, and unilaterally in 9 cases
Fig. 2. Harvested bilateral MRAM.
3
Fig. 3. New pelvic floor reconstructed by bilateral MRAM.
(56%), as a short version (N = 4 on the left side; N = 1 on the right side) or long version (N = 2 on the left side; N = 2 on the right side), respectively, with the preservation of the anterior fascial sheath in all cases. The control group was composed by the historical cohort of patients after PE performed at the same institution but without pelvic floor reconstruction or neovagina formation. Characteristics of the groups after PE with (MRAM) and without (Control) MRAM flap transposition is shown in Table 1. Both groups were similar in median age, BMI, performance status before the procedure, distribution of tumor types, and histological types, and in the proportion of patients who previously received radiotherapy, chemotherapy, or previous surgery. The significantly longer follow-up in the control group reflects its historical nature. The majority of procedures were indicated as salvage treatment for tumor recurrence (Table 2). ELSE procedures have been performed in our center since 2011, so the majority appear in the MRAM group (5 out of 6). Operation time was not longer in the MRAM group, and blood loss was substantially lower in the MRAM group, which reflects the use of advanced bipolar technology in more recent cases. Both groups were balanced in the number of days spent in ICU, the length of hospital stay, and interval to the beginning of peristalsis (Table 3). Substantially more patients from the control group required reoperation during the 60 days after the surgery (50% vs. 25%) (p = 0.188). Lower reoperation rate was caused by less interventions due to complications potentially related to empty pelvis syndrome, such as the formation of pelvic abscess, GI fistula, or small bowel obstruction caused by the adhesions deeply in the pelvis, which appeared only in 1 case in the MRAM group while in 7 cases (29%) in the control group (p = 0.114). There were substantial differences in late post-operative morbidity between the groups (p = 0.041) (Table 3). More patients in the control group experienced late postoperative complications (79% vs. 44%), and even more importantly, a bigger proportion of complications were categorized as grade ≥ 3 (37% vs. 6%). Performance status, evaluated 6 months after the surgery according to the ECOG scale, was ≤1 in the majority of patients in MRAM (81%) while only in 38% in the control group (p = 0.027). There was only one incisional hernia in the MRAM group while 3 cases were reported in the control group. No case of parastomal hernia appeared in any of the groups. Neither partial nor total flap necrosis was found in any case in the MRAM group. The higher proportion of patients was without evidence of disease at the time of last follow-up. However, this reflects a shorter interval from the surgery and most likely it is not influenced by the reconstructive part of the surgery (Table 3).
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
4
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 1 Characteristics of the whole group after exenterative procedures and both subgroups with (MRAM) or without (Controls) pelvic floor reconstruction. Characteristics1 Age BMI Performance status before surgery 0 1 2 Tumor type Cervix Vagina Vulva Endometrium Urinary bladder Rectum Histology Squamous cancer Adenocancer Other Previous treatment RT in primary treatment RT for relapse CHT in primary treatment CHT for relapse Previous simple or rad hyst Other surgery in primary treatment* Follow-up (months)
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
p2
51 (26; 73) 26 (19; 49)
50 (26; 73) 27 (19; 49)
56 (27; 71) 25 (21; 29)
0.240 0.635
N = 20 (50%) N = 18 (45%) N = 2 (5%)
N = 13 (54%) N = 9 (38%) N = 2 (8%)
N = 7 (44%) N = 9 (56%) N=0
0.493
N N N N N N
N N N N N N
N N N N N N
0.971
= = = = = =
18 (45%) 10 (25%) 7 (18%) 3 (8%) 1 (2%) 1 (2%)
= = = = = =
11 (46%) 6 (25%) 4 (17%) 2 (8%) 0 1 (4%)
= = = = = =
7 (44%) 4 (25%) 3 (19%) 1 (6%) 1 (6%) 0
N = 30 (75%) N = 7 (18%) N = 3 (7%)
N = 18 (75%) N = 4 (17%) N = 2 (8%)
N = 12 (75%) N = 3 (19%) N = 1 (6%)
0.999
N = 19 (48%) N = 4 (10%) N = 7 (18%) N = 9 (23%) N = 12 (30%) N = 4 (10%) 17 (6; 99)
N = 12 (50%) N = 2 (8%) N = 3 (13%) N = 4 (17%) N = 8 (33%) N = 3 (13%) 35 (6; 99)
N = 7 (44%) N = 2 (12%) N = 4 (25%) N = 5 (31%) N = 4 (25%) N = 1 (6%) 9 (6; 36)
0.755 0.999 0.407 0.441 0.729 0.638 b 0.001
1 absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables; 2 p-value of Fisher's exact test or Mann-Whitney U test; RT = radiotherapy; CHT = chemotherapy; rad hyst = radical hysterectomy; * including transureteral tumor resection, exploratory laparotomy, proximal vaginectomy, paraaortic lymphadenectomy.
Discussion We have been performing MRAM in our institution since 2012. In the first few years, the flap was used selectively only in patients with large defects. Since 2015, based on a substantial improvement of postoperative complication rate, there is a tendency to create the artificial muscle floor in nearly all patients after either infralevatoric exenteration or ELSE procedures. In this article, we report a group of 16 cases with MRAM. To be able to document the importance of pelvic floor reconstruction, we compare the results with a control group after a similar procedure, without MRAM. Even though we only present a limited number of patients, there is a substantial decrease in the number of
Table 2 Description of surgical data. Characteristics1 Indication for surgery Palliation Primary treatment Recurrence treatment Type of procedure Anterior exenteration Posterior exenteration Total exenteration ELSE combined with PE ELSE Operation time (min) Blood loss (ml)
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
N = 2 (5%) N = 7 (18%) N = 31 (77%)
N = 1 (4%) N = 3 (13%) N = 20 (83%)
N = 1 (6%) N = 4 (25%) N = 11 (69%)
0.590
N = 8 (20.0%) N = 9 (23%)
N = 5 (21%) N = 6 (25%)
N = 3 (19%) N = 3 (19%)
0.232
N = 20 (50%) N=3
N = 13 (54%) N=1
N = 7 (44%) N=2
p2
N = 3 (8%) N=0 351 (116; 555) 372 (116; 480)
N = 3 (19%) 341 (233; 555) 0.978
750 (200; 5000)
550 (200; 5000)
1300 (250; 4500)
0.025
1 absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables; 2 p-value of Fisher's exact test or Mann-Whitney U test.
reoperations for reasons associated with empty pelvis syndrome, as well as a decrease of severe late post-operative complications. The postoperative complication rate is reported in a broad range between 30 and 90% [1–6,13] after PE, which reflects the lack of a classification system rather than any real differences between institutions. Although after these demanding operations there is an array of potential sources of morbidity, the complications can be clustered into two most frequent and most severe groups, those associated with urinary diversion and those caused by so-called empty pelvis syndrome [7,14]. The space remaining after tissue removal in the bone pelvis cannot be filled by any tissue or organs since the mesentery is too short to allow the bowels to fall. It often results in the formation of an abscess, hematoma or lymphocele, causing a permanent discharge, chronic infection, bowel irritation, bowel perforation, and fistulas. Moreover, the removal of pelvic floor muscles and pelvic organs causes an absence of mechanical support for the bowels, which, together with bowel adhesions deep in the pelvis, may lead to chronic ileus and bowel obstruction. Several techniques have been tested for pelvic floor reconstruction, including the omental flap [15–17], myocutaneous flaps based on the rectus abdominis muscle [9,18–20], the thigh flaps [21] or the transverse abdominis muscle [18], absorbable mesh [22] or silicon expanders [23]. The use of rectus abdominis flap after a pelvic exenteration was reported by many other authors as a myocutaneous flap used primarily for perineal reconstruction or for neovagina formation [9,10,11,19– 20]. The biggest cohorts were recruited from rectal cancer patients, in whom the greatest challenge after PE is usually to cover a large perineal defect, while a lateral pelvic side wall resection is less extensive than in gynecologic tumors. Flaps are therefore used in these cases primarily to cover the perineal defect. In gynecological malignancies, it was shown that the vaginal reconstruction after a pelvic exenteration could dramatically lower the postoperative morbidity. The morbidity was compared in a series of 64 patients with a pelvic exenteration; 29 of them had vaginal reconstruction with TRAM and 8 with Singapore fascio-cutanous flap [11]. In those with neovagina formation, there were fewer cases with pelvic abscess (20% vs. 7%) and small bowel fistula (20% vs. 3%).
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
5
Table 3 Post-operative morbidity and disease status. Characteristics1 ICU (days) Hospital stay (days) Bowel peristalsis (days) Reoperation in 60 days A) Related to urinary diversion Ureteral anastomotic leak Small bowel obstruction caused by adhesions to the pouch Ureteral stricture B) Related potentially to pelvis empty syndrome Pelvis abscess GI fistula Small bowel obstruction caused by deep pelvis adhesions C) Other causes Wound complications Early postop. Complications Maximum grade 1 2 3 4 Performance status at 6 mo. 0–1 2 3 4 Late postop. Complications Maximum grade 1–2 3–4 Incisional hernia Disease status AWD DOD NED 1
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
p2
7 (2; 35) 17 (8; 133) 5 (1; 15) N = 16 (40%) N = 7 (18%) N = 1 (3%) N = 5 (13%)
6 (2; 33) 18 (8; 133) 5 (2; 15) N = 12 (50%) N = 4 (17%) N = 1 (4%) N = 2 (8%)
7 (5; 35) 17 (14; 57) 5 (1; 11) N = 4 (25%) N = 3 (19%) N=0 N = 3 (19%)
0.824 0.455 0.131 0.188 0.999 0.999 0.373
N N N N N
N N N N N
N N N N N
0.999 0.114 0.508 0.999 0.631
= = = = =
1 (3%) 8 (20%) 2 (5%) 1 (3%) 5 (13%)
= = = = =
1 (4%) 7 (29%) 2 (8%) 1 (4%) 4 (17%)
= = = = =
0 1 (6%) 0 0 1 (6%)
N = 1 (3%) N = 32 (80%)
N = 1 (4%) N = 19 (80%)
N=0 N = 12 (75%)
0.999 0.999
N N N N
= = = =
8 (20%) 17 (43%) 12 (30%) 3 (7%)
N N N N
= = = =
4 (17%) 9 (38%) 8 (33%) 3 (12%)
N N N N
= = = =
4 (25%) 8 (50%) 4 (25%) 0
0.463
N N N N N
= = = = =
22 (55%) 10 (25%) 6 (15%) 2 (5%) 26 (65%)
N N N N N
= = = = =
9 (38%) 9 (38%) 4 (17%) 2 (8%) 19 (79%)
N N N N N
= = = = =
13 (81%) 1 (6%) 2 (12%) 0 7 (44%)
0.027
N = 16 (40%) N = 10 (25%) N = 4 (10%)
N = 10 (42%) N = 9 (37%) N = 3 (13%)
N = 6 (38%) N = 1 (6%) N = 1 (6%)
N = 5 (13%) N = 10 (25%) N = 25 (62%)
N = 4 (17%) N = 8 (33%) N = 12 (50%)
N = 1 (6%) N = 2 (12%) N = 13 (82%)
absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables;
Our approach was inventive in terms of the reconstruction goal, the modification of the flap, and the technique of the flap placement in the pelvis. The primary goal in our patients was pelvic floor reconstruction. Since the purpose was not to cover a perineal defect, both the skin and the anterior sheath were preserved, so even a bilateral flap would not compromise the anterior abdominal wall. It is the purpose of the flap that determines its placement in the pelvis. If the main aim is a perineal reconstruction, the flap must be placed vertically deeply to the pelvis. It is usually insufficient to simultaneously create a compact artificial pelvic floor and completely cover the denuded area of the obturator fossa, sacral bone, and other exposed structures in the pelvis. The main and only purpose of the flap transposition in our group was the pelvic floor reconstruction. One or two flaps were inserted horizontally into the pelvis at the level of the former pelvic floor, forming a U-shaped compact structure. If the pelvic defect was deep, the second flap from the other side was partially transposed below the new pelvic floor to fill in the deeper space. The newly formed pelvic floor was sufficient to create support for the small bowels, fill in the dead space and cover the denuded structures of the pelvic sidewall. Due to the complexity of each PE case, the number of factors influencing postoperative morbidity, and the limited number of procedures, a prospective design of the trial comparing patients with and without pelvic floor reconstruction is not an option. A control group in our study was composed by patients after PE without pelvic floor reconstruction performed in the same institution. It is predominantly a historical cohort, in which the majority of cases (15/24) were operated before 2012. The time periods overlap only partially, and the operation technique as well as post-operative care comparison might be biased. The improvement of the clinical practice is obvious when blood loss is
2
0.041 0.190 0.638 0.201
p-value of Fisher's exact test or Mann-Whitney U test.
compared. Blood loss is significantly lower in the more recent group of patients with MRAM. Operation time is identical in both groups, even though MRAM requires around 60 min of additional time. The decrease in blood loss and shrinking of operation time is most likely a result of the advanced bipolar technology we have been using since 2009. On the other hand, it is unlikely that these developments in surgical techniques or intensive care can prevent reoperations and other serious complications that are caused by a too-large unfilled defect in the pelvis, the absence of mechanical support of the bowels, and the formation of adhesions to the exposed structures of the pelvic sidewall. It should also be emphasized that the radicality of the procedures has risen simultaneously, as shown by the predominant occurrence of ELSE procedures in the MRAM group (5 of 6 cases). Rectus abdominis flaps are rarely associated with the flap loss [11, 18]. In a large series of 100 patients who received the rectus abdominis flap for vaginal or perineal reconstruction, there were only 2 cases of total flap necrosis, while in other smaller cohorts the rate was as high as 14%. There was no case of total or partial flap necrosis in our group. Another potential downside of the rectus abdominis flap is a weakening of the abdominal wall and a higher risk of donor-side complications. A risk of incisional hernias between 3 and 11% was reported after RAM flap transposition [9,19–20]. When using the myoperitoneal flap (MRAM), a complete anterior fascia sheath with rich vascular supply from subcutaneous tissue is preserved, which secures the integrity of the anterior abdominal wall. It is reassuring that we only had one patient with incisional hernia (not requiring any surgery) and no case of parastomal hernia in our cohort. In our series, MRAM was used in 16 consecutive women after anterior, total, or posterior PE, and / or ELSE. In all cases, a short (up to the
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
6
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
umbilicus level) or long (up to the ribs) MRAM was transposed to the pelvis unilaterally or bilaterally with the preservation of the anterior fascia sheath. Pelvic floor reconstruction by MRAM was associated with substantially fewer reoperations in 60 days due to complications potentially related to pelvis empty syndrome, as well as with significantly lower risk of serious late postoperative complications, and an improved performance status 6 months after the operation. Conclusion Pelvic floor reconstruction by MRAM after extensive pelvic exenterative procedures is associated with a decrease in the number of reoperations due to complications potentially related to empty pelvis syndrome, a lower occurrence of severe late postoperative complications, and a better performance status 6 months after the surgery. The experience from one institution did not show an increase in procedure related morbidity such as flap necrosis or incisional hernia. Conflict of interest There are no conflicts of interest.
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
Acknowledgements This work was supported by Charles University in Prague (UNCE 204024 and PRVOUK-P27/LF1/1) and by the project of Ministry of Health of the Czech Republic (MZ CR – RVO VFN64165). References [1] A. Maggioni, G. Roviglione, F. Landoni, V. Zanagnolo, M. Peiretti, N. Colombo, L. Bocciolone, R. Biffi, L. Minig, C.P. Morrow, Pelvic exenteration: ten-year experience at the European Institute of Oncology in Milan, Gynecol. Oncol. 114 (2009) 64–68. [2] V. Chiantera, M. Rossi, P. De Iaco, C. Koehler, S. Marnitz, G. Ferrandina, F. Legge, F. Parazzini, G. Scambia, A. Schneider, G.F. Vercellino, Survival after curative pelvic exenteration for primary or recurrent cervical cancer: a retrospective multicentric study of 167 patients, Int. J. Gynecol. Cancer 24 (2014) 916–922. [3] A.M. Schmidt, P. Imesch, D. Fink, H. Egger, Indications and long-term clinical outcomes in 282 patients with pelvic exenteration for advanced or recurrent cervical cancer, Gynecol. Oncol. 125 (2012) 604–609. [4] A. Brunschwig, H.R. Barber, Pelvic exenteration combined with resection of segments of bony pelvis, Surgery 65 (1969) 417–420. [5] D.Q. Nguyen, A.D. McGregor, O. Freites, N.D. Carr, J. Beynon, A.M. El-Sharkawi, M.G. Lucas, Exenterative pelvic surgery–eleven year experience of the Swansea pelvic oncology group, Eur. J. Surg. Oncol. 31 (2005) 1180–1184. [6] A. Urh, P.T. Soliman, K.M. Schmeler, S. Westin, M. Frumovitz, A.M. Nick, B. Fellman, D.L. Urbauer, P.T. Ramirez, Postoperative outcomes after continent versus
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
incontinent urinary diversion at the time of pelvic exenteration for gynecologic malignancies, Gynecol. Oncol. 129 (2013) 580–585. G.L. Goldberg, P. Sukumvanich, M.H. Einstein, H.O. Smith, P.S. Anderson, A.L. Fields, Total pelvic exenteration: the Albert Einstein College of Medicine/Montefiore medical center experience (1987 to 2003), Gynecol. Oncol. 101 (2006) 261–268. P.S. Ghosh, H. Fawzi, Empty pelvis syndrome, J. Obstet. Gynaecol. 24 (2004) 714–715. R.E. Horch, W. Hohenberger, A. Eweida, U. Kneser, K. Weber, A. Arkudas, S. Merkel, J. Göhl, J.P. Beier, A hundred patients with vertical rectus abdominis myocutaneous (VRAM) flap for pelvic reconstruction after total pelvic exenteration, Int. J. Color. Dis. 29 (2014) 813–823. D.B. Chessin, J. Hartley, A.M. Cohen, M. Mazumdar, P. Cordeiro, J. Disa, B. Mehrara, B.D. Minsky, P. Paty, M. Weiser, W.D. Wong, J.G. Guillem, Rectus flap reconstruction decreases perineal wound complications after pelvic chemoradiation and surgery: a cohort study, Ann. Surg. Oncol. 12 (2005) 104–110. M. Jurado, A. Bazán, J.L. Alcázar, E. Garcia-Tutor, Primary vaginal reconstruction at the time of pelvic exenteration for gynecologic cancer: morbidity revisited, Ann. Surg. Oncol. 16 (2009) 121–127. D. Dindo, N. Demartines, P.A. Clavien, Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey, Ann. Surg. 240 (2) (2004 Aug) 205–213. H.J. Yoo, M.C. Lim, S.S. Seo, S. Kang, C.W. Yoo, J.Y. Kim, S.Y. Park, Pelvic exenteration for recurrent cervical cancer: ten-year experience at National Cancer Center in Korea, J. Gynecol. Oncol. 23 (2012) 242–250. S. Marnitz, S. Dowdy, M. Lanowska, A. Schneider, K. Podratz, C. Köhler, Exenterations 60 years after first description: results of a survey among US and German gynecologic oncology centers, Int. J. Gynecol. Cancer 19 (2009) 974–977. C.S. Hultman, M.A. Sherrill, E.G. Halvorson, C.N. Lee, J.F. Boggess, M.O. Meyers, B.A. Calvo, H.J. Kim, Utility of the omentum in pelvic floor reconstruction following resection of anorectal malignancy: patient selection, technical caveats, and clinical outcomes, Ann. Plast. Surg. 64 (2010) 559–562. J.M. Hay, A. Fingerhut, J.C. Paquet, Y. Flamant, Management of the pelvic space with or without omentoplasty after abdominoperineal resection for carcinoma of the rectum: a prospective multicenter study, Eur J Surg 163 (1997) 199–206. G.J. Poston, S.R. Smith, W.N. Baker, Retrocolic pelvic omentoplasty in abdominoperineal excision of the rectum, Ann. R. Coll. Surg. Engl. 73 (1991) 229–232. J.T. Soper, L.J. Havrilesky, A.A. Secord, A. Berchuck, D.L. Clarke-Pearson, Rectus abdominis myocutaneous flaps for neovaginal reconstruction after radical pelvic surgery, Int. J. Gynecol. Cancer 15 (2005) 542–548. R.J. Chokshi, M.P. Kuhrt, D. Arrese, E.W. Martin Jr., Reconstruction of total pelvic exenteration defects with rectus abdominus myocutaneous flaps versus primary closure, Am. J. Surg. 205 (2013) 64–70. T.A. Creagh, L. Dixon, F.A. Frizelle, Reconstruction with vertical rectus abdominus myocutaneous flap in advanced pelvic malignancy, J. Plast. Reconstr. Aesthet. Surg. 65 (2012) 791–797. R.A. Nelson, C.E. Butler, Surgical outcomes of VRAM versus thigh flaps for immediate reconstruction of pelvic and perineal cancer resection defects, Plast. Reconstr. Surg. 123 (2009) 175–183. A.O. Momoh, A.M. Kamat, C.E. Butler, Reconstruction of the pelvic floor with human acellular dermal matrix and omental flap following anterior pelvic exenteration, J. Plast. Reconstr. Aesthet. Surg. 63 (2010) 2185–2187. G.C. Guimarães, G. Baiocchi, B.M. Rossi, F.O. Ferreira, S. Aguiar, W.T. Nakagawa, A. Lopes, The use of silicone expander and cecal transposition after pelvic exenteration, Eur. J. Surg. Oncol. 33 (2007) 586–589.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations D Cibula a,⁎, M Zikan a, D Fischerova a, R Kocian a, A Germanova a, A Burgetova b, L Dusek c, Z Fartáková a, M Schneiderová e, K Nemejcová d, J Slama a a
Gynecologic Oncology Center, Department of Obstetrics and Gynecology, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic Department of Radiodiagnostics, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic Institute for Biostatistics and Analyses; Faculty of Medicine, Masaryk University, Brno, Czech Republic d Department of Pathology, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic e Department of Surgery, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic b c
H I G H L I G H T S • • • •
Our approach is inventive in the reconstruction goal and the modification of the flap. Pelvic floor reconstruction was associated with a marked decrease of severe post-operative morbidity and improvement of quality of life. The MRAM group showed fewer complications, including reoperations, related to symptoms of the empty pelvis syndrome. There was no case of flap necrosis, no paraastomal hernia and only one case of the incisional hernia.
a r t i c l e
i n f o
Article history: Received 22 November 2016 Received in revised form 8 January 2017 Accepted 11 January 2017 Available online xxxx Keywords: Pelvic exenteration Empty pelvic syndrome Pelvic floor reconstruction Muscular flap
a b s t r a c t Objective. To describe the technique and report experiences with pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after extensive pelvic procedures. Methods. Surgical technique of MRAM harvest and transposition is carefully described. The patients in whom pelvic floor reconstruction with MRAM after either infralevator pelvic exenteration and/or extended lateral pelvic sidewall excision was carried out were enrolled into the study (MRAM group, n = 16). Surgical data, post-operative morbidity, and disease status were retrospectively assessed. The results were compared with a historical cohort of patients, in whom an exenterative procedure without pelvic floor reconstruction was performed at the same institution (control group, n = 24). Results. Both groups were balanced in age, BMI, tumor types, and previous treatment. Substantially less patients from the MRAM group required reoperation within 60 days of the surgery (25% vs. 50%) which was due to much lower rate of complications potentially related to empty pelvis syndrome (1 vs. 7 reoperations) (p = 0.114). Late post-operative complication rate was substantially lower in the MRAM group (any grade: 79% vs. 44%; grade ≥ 3: 37% vs. 6%) (p = 0.041). The performance status 6 months after the surgery was ≤1 in the majority of patients in MRAM (81%) while in only 38% of patients from the control group (p = 0.027). There was one incisional hernia in MRAM group while three cases were reported in the controls. Conclusions. Pelvic floor reconstruction by MRAM in patients after pelvic exenterative procedures is associated with a substantial decrease in postoperative complications that are potentially related to empty pelvis syndrome. © 2017 Elsevier Inc. All rights reserved.
Introduction
⁎ Corresponding author at: Gynecologic Oncology Center, Department of Obstetrics and Gynecology, Apolinarska 18, Prague 2, 120 00 Czech Republic. E-mail address: [email protected] (D. Cibula).
Extensive surgical procedures, including pelvic exenterations (PE) or extended lateral pelvic sidewall excisions (ELSE), belong to a standard spectrum of procedures in gynecologic oncology even in the 21st century. In spite of new developments in radiotherapy, it remains the only option for management with a curative potential for a selected group of
http://dx.doi.org/10.1016/j.ygyno.2017.01.014 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
2
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
patients. Long-term survival oscillates above 50% in the majority of groups after PE performed for mixed pelvic cancers, which is an outstanding perspective for those who would have otherwise lived less than a year with a quickly deteriorating quality of life in most cases [1–6]. The leading indication remains local pelvic tumor recurrence, especially after previous radiotherapy, while less frequently these procedures are performed as a primary treatment or for palliation. Improved perioperative care has minimized the mortality associated with these procedures; at the same time, in spite of new operative techniques and sophisticated intraoperative and postoperative care, morbidity remains the biggest challenge. The most challenging are serious postoperative complications requiring reoperations, which often have long-term consequences for patients' quality of life. Empty pelvis syndrome is one of the major causes of severe post-operative morbidity after pelvic exenterations [7]. The syndrome entails a variety of symptoms, including abscess formation, bowel obstruction caused by firm adhesions to the denuded structures of the pelvic side wall, permanent discharge, bowel perforation and fistulas, all with one common cause: the dead space in the pelvis [8]. While similar spaces after the removal of organs in other regions of the abdominal cavity, for instance after splenectomy, liver or bowel resection, are easily filled by the bowels and other organs, the space deep in the pelvis remains empty due to short bowel mesentery, especially if the empty space is extended laterally beyond the endopelvic fascia. A variety of reconstructive procedures using the omentum, absorbable mashes, or silicon expanders have been used to fill in the dead space and create a mechanical support for the small bowels with mostly unsatisfactory results. Lower morbidity was, however, universally reported after either perineal reconstruction or neovagina formation by myocutaneous flaps, most often using the rectus abdominis muscle [9–11]. Since the main purpose of such reconstructions is to cover a large perineal defect or to form a neovagina, the flap must be composed of the muscle, anterior fascial sheath and skin. MRAM flap procedure has been performed in our institution for a pelvic reconstruction in patients who have undergone PE and/or ELSE, and in whom no vaginal or perineal reconstruction is planned, since 2012. Since MRAM transposition is carried out primarily for pelvic floor reconstruction, but not for vaginal or perineal reconstruction, it does not require skin and anterior sheath resection. MRAM is composed of the rectus abdominis muscle, posterior fascial sheath and peritoneum, while the anterior fascial sheath and the skin are preserved intact. Unilateral or bilateral flaps are transposed into the horizontal position in the pelvis at the level of the former pelvic diaphragm, so a new and compact pelvic muscle floor is formed in order to fill in the dead space in the pelvis, create an artificial diaphragm for small bowel support and form a mechanical barrier preventing the small bowels to adhere deeply in the pelvis to the denuded obturator fossa, large vessels, and pelvic bones. The aim of this paper is to describe the technique and report experiences with pelvic floor reconstruction. Intra-operative and post-operative morbidity were compared to the historical group of patients who underwent exenterative procedures without pelvic floor reconstruction.
All intra-operative and early postoperative complications were classified using a Dindo-Clavien 5 grade system [12]. In short, grades 1 and 2 are characterized by a need for pharmacological treatment; grade 3 required surgical or radiological intervention; grade 4 stands for lifethreatening complication; and grade 5 for a death of a patient. Complications which occurred during the surgical procedure were reported as intra-operative while those presented within 30 days after the surgery as early post-operative. Performance status was evaluated according to the ECOG scale, before the surgery and at 6 ± 2 months after the surgery. Preoperative marking of the location of the ostomy/ies was routinely performed by a stomal therapist; usually, two potential sites were marked, one at the optimal (usual) site, and the other one shifted 3– 5 cm laterally for the case of harvesting the MRAM. Anatomy of the MRAM flap. The rectus abdominis muscle is a paired muscle running vertically on each side of the anterior wall of the abdomen. The muscle is contained in the rectus sheath consisting of the aponeurosis of the lateral abdominal muscles; above the arcuate line, the internal oblique aponeurosis splits to envelop the rectus abdominis muscle, while below the arcuate line the aponeurosis is localized only anterior to the muscle (Fig. 1). The rectus abdominis has two dominant vascular pedicles. The inferior epigastric vessels, branches of external iliac vessels, which run and branch on the posterior surface of the muscle, and the superior epigastric vessels. Both vascular systems are richly connected to each other so the entire muscle can be fed by only one preserved pedicle. Inferior epigastric vessels serve as the vascular pedicle for the MRAM flap. Technique of MRAM flap harvest and transposition. MRAM was harvested either on one side or bilaterally, in its short version (until the level of umbilicus) or long version (until its attachment to ribs). The type of MRAM and number of flaps (unilateral or bilateral) was determined by the size and the shape of defect after the exenterative part of the surgery. Usually, the total infralevatoric PE and PE combined with ELSE required a bilateral flap to cover the whole pelvis and to fill in the deep pelvic defect below the reconstructed muscle floor. First, the muscle was exposed medially by opening the linea alba and rectus sheath medially and then the muscle was carefully dissected from the anterior rectus sheath. Since exenterative procedure was usually performed from the lower midline laparotomy, it had to be
Methods Patient selection and clinical assessment. All patients in whom pelvic floor reconstruction with MRAM after extensive pelvic procedure, either infralevator PE and / or ELSE, was carried out were enrolled into the study. The patients in whom rectus muscle flap was used for vaginal or perineal reconstruction were not included in the study since in these cases the flap was composed of a complete fascial sheath and skin. Surgical data, post-operative complications, and disease status were retrospectively assessed. The results were compared with a historical cohort of patients, in whom an exenterative procedure without pelvic floor reconstruction was performed.
Fig. 1. Anatomy of vascular supply and fascia rectus sheath below and above the arcuate line.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
extended up to the xiphoideus if a long version was executed. Maximum care was taken not to open the anterior fascia, especially at the level of intersections, where the dissection is more demanding. Next, the muscle was divided cranially, either at the level of the umbilicus (short version) or at the attachment to the ribs (long version), using monopolar diathermy. Finally, the flap was freed laterally, at a lateral margin of the muscle, starting from the cranial edge and heading caudally until above the course of the previously detected epigastric vessels. While the anterior surface of the final flap is formed by the exposed muscle, dorsally the muscle is covered by the peritoneum and posterior rectus sheath above arcuate line. It is important to prevent a separation of the muscle and the fascia / peritoneum, especially below the arcuate line, since the feeding fragile vessels run on the dorsal aspect of the muscle. The attachment of the muscle to the pubic bone was preserved, and inferior epigastric vessels were not dissected down to the external iliac vessels (Fig. 2). After the bleeding was controlled, the flap was carefully rotated at 180o into the pelvis. In the case of bilateral flaps, their placement was adjusted to the shape of the empty space. In most cases, both flaps were positioned next to each other in a U-shape, forming one muscle floor. In a few cases, usually after simultaneous vulvectomy or extensive vaginectomy that caused a large space deep in the pelvis, one flap was placed straight towards the sutured perineal skin, filling the deep part of the pelvic space, while the U-shaped diaphragm was created by the other side flap, mimicking the pelvic floor. Flaps were fixed in their final positions with a few stitches, attaching the whole thickness of the flap (muscle, fascia, peritoneum) to the suitable structures in the pelvis (pre-sacral fascia, obturator fascia, iliac fascia) (Fig. 3). Finally, the integrity of the preserved anterior rectus sheath was carefully checked and any defect was sutured before the laparotomy was finally closed. The fascia was not augmented by any alloplastic mesh. Statistical analysis Standard summary statistics were used to describe primary data. Fisher exact test and Mann/Whitney U test were applied to test statistical significance of differences between groups of patients. Results MRAM was used in 11 consecutive women after PE (infralevatoric in all cases), in 3 cases after ELSE, and in 2 cases after a combined procedure (PE + ELSE). MRAM was harvested bilaterally in 7 cases (44%) as short (N = 2) or long (N = 5) version, and unilaterally in 9 cases
Fig. 2. Harvested bilateral MRAM.
3
Fig. 3. New pelvic floor reconstructed by bilateral MRAM.
(56%), as a short version (N = 4 on the left side; N = 1 on the right side) or long version (N = 2 on the left side; N = 2 on the right side), respectively, with the preservation of the anterior fascial sheath in all cases. The control group was composed by the historical cohort of patients after PE performed at the same institution but without pelvic floor reconstruction or neovagina formation. Characteristics of the groups after PE with (MRAM) and without (Control) MRAM flap transposition is shown in Table 1. Both groups were similar in median age, BMI, performance status before the procedure, distribution of tumor types, and histological types, and in the proportion of patients who previously received radiotherapy, chemotherapy, or previous surgery. The significantly longer follow-up in the control group reflects its historical nature. The majority of procedures were indicated as salvage treatment for tumor recurrence (Table 2). ELSE procedures have been performed in our center since 2011, so the majority appear in the MRAM group (5 out of 6). Operation time was not longer in the MRAM group, and blood loss was substantially lower in the MRAM group, which reflects the use of advanced bipolar technology in more recent cases. Both groups were balanced in the number of days spent in ICU, the length of hospital stay, and interval to the beginning of peristalsis (Table 3). Substantially more patients from the control group required reoperation during the 60 days after the surgery (50% vs. 25%) (p = 0.188). Lower reoperation rate was caused by less interventions due to complications potentially related to empty pelvis syndrome, such as the formation of pelvic abscess, GI fistula, or small bowel obstruction caused by the adhesions deeply in the pelvis, which appeared only in 1 case in the MRAM group while in 7 cases (29%) in the control group (p = 0.114). There were substantial differences in late post-operative morbidity between the groups (p = 0.041) (Table 3). More patients in the control group experienced late postoperative complications (79% vs. 44%), and even more importantly, a bigger proportion of complications were categorized as grade ≥ 3 (37% vs. 6%). Performance status, evaluated 6 months after the surgery according to the ECOG scale, was ≤1 in the majority of patients in MRAM (81%) while only in 38% in the control group (p = 0.027). There was only one incisional hernia in the MRAM group while 3 cases were reported in the control group. No case of parastomal hernia appeared in any of the groups. Neither partial nor total flap necrosis was found in any case in the MRAM group. The higher proportion of patients was without evidence of disease at the time of last follow-up. However, this reflects a shorter interval from the surgery and most likely it is not influenced by the reconstructive part of the surgery (Table 3).
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
4
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 1 Characteristics of the whole group after exenterative procedures and both subgroups with (MRAM) or without (Controls) pelvic floor reconstruction. Characteristics1 Age BMI Performance status before surgery 0 1 2 Tumor type Cervix Vagina Vulva Endometrium Urinary bladder Rectum Histology Squamous cancer Adenocancer Other Previous treatment RT in primary treatment RT for relapse CHT in primary treatment CHT for relapse Previous simple or rad hyst Other surgery in primary treatment* Follow-up (months)
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
p2
51 (26; 73) 26 (19; 49)
50 (26; 73) 27 (19; 49)
56 (27; 71) 25 (21; 29)
0.240 0.635
N = 20 (50%) N = 18 (45%) N = 2 (5%)
N = 13 (54%) N = 9 (38%) N = 2 (8%)
N = 7 (44%) N = 9 (56%) N=0
0.493
N N N N N N
N N N N N N
N N N N N N
0.971
= = = = = =
18 (45%) 10 (25%) 7 (18%) 3 (8%) 1 (2%) 1 (2%)
= = = = = =
11 (46%) 6 (25%) 4 (17%) 2 (8%) 0 1 (4%)
= = = = = =
7 (44%) 4 (25%) 3 (19%) 1 (6%) 1 (6%) 0
N = 30 (75%) N = 7 (18%) N = 3 (7%)
N = 18 (75%) N = 4 (17%) N = 2 (8%)
N = 12 (75%) N = 3 (19%) N = 1 (6%)
0.999
N = 19 (48%) N = 4 (10%) N = 7 (18%) N = 9 (23%) N = 12 (30%) N = 4 (10%) 17 (6; 99)
N = 12 (50%) N = 2 (8%) N = 3 (13%) N = 4 (17%) N = 8 (33%) N = 3 (13%) 35 (6; 99)
N = 7 (44%) N = 2 (12%) N = 4 (25%) N = 5 (31%) N = 4 (25%) N = 1 (6%) 9 (6; 36)
0.755 0.999 0.407 0.441 0.729 0.638 b 0.001
1 absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables; 2 p-value of Fisher's exact test or Mann-Whitney U test; RT = radiotherapy; CHT = chemotherapy; rad hyst = radical hysterectomy; * including transureteral tumor resection, exploratory laparotomy, proximal vaginectomy, paraaortic lymphadenectomy.
Discussion We have been performing MRAM in our institution since 2012. In the first few years, the flap was used selectively only in patients with large defects. Since 2015, based on a substantial improvement of postoperative complication rate, there is a tendency to create the artificial muscle floor in nearly all patients after either infralevatoric exenteration or ELSE procedures. In this article, we report a group of 16 cases with MRAM. To be able to document the importance of pelvic floor reconstruction, we compare the results with a control group after a similar procedure, without MRAM. Even though we only present a limited number of patients, there is a substantial decrease in the number of
Table 2 Description of surgical data. Characteristics1 Indication for surgery Palliation Primary treatment Recurrence treatment Type of procedure Anterior exenteration Posterior exenteration Total exenteration ELSE combined with PE ELSE Operation time (min) Blood loss (ml)
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
N = 2 (5%) N = 7 (18%) N = 31 (77%)
N = 1 (4%) N = 3 (13%) N = 20 (83%)
N = 1 (6%) N = 4 (25%) N = 11 (69%)
0.590
N = 8 (20.0%) N = 9 (23%)
N = 5 (21%) N = 6 (25%)
N = 3 (19%) N = 3 (19%)
0.232
N = 20 (50%) N=3
N = 13 (54%) N=1
N = 7 (44%) N=2
p2
N = 3 (8%) N=0 351 (116; 555) 372 (116; 480)
N = 3 (19%) 341 (233; 555) 0.978
750 (200; 5000)
550 (200; 5000)
1300 (250; 4500)
0.025
1 absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables; 2 p-value of Fisher's exact test or Mann-Whitney U test.
reoperations for reasons associated with empty pelvis syndrome, as well as a decrease of severe late post-operative complications. The postoperative complication rate is reported in a broad range between 30 and 90% [1–6,13] after PE, which reflects the lack of a classification system rather than any real differences between institutions. Although after these demanding operations there is an array of potential sources of morbidity, the complications can be clustered into two most frequent and most severe groups, those associated with urinary diversion and those caused by so-called empty pelvis syndrome [7,14]. The space remaining after tissue removal in the bone pelvis cannot be filled by any tissue or organs since the mesentery is too short to allow the bowels to fall. It often results in the formation of an abscess, hematoma or lymphocele, causing a permanent discharge, chronic infection, bowel irritation, bowel perforation, and fistulas. Moreover, the removal of pelvic floor muscles and pelvic organs causes an absence of mechanical support for the bowels, which, together with bowel adhesions deep in the pelvis, may lead to chronic ileus and bowel obstruction. Several techniques have been tested for pelvic floor reconstruction, including the omental flap [15–17], myocutaneous flaps based on the rectus abdominis muscle [9,18–20], the thigh flaps [21] or the transverse abdominis muscle [18], absorbable mesh [22] or silicon expanders [23]. The use of rectus abdominis flap after a pelvic exenteration was reported by many other authors as a myocutaneous flap used primarily for perineal reconstruction or for neovagina formation [9,10,11,19– 20]. The biggest cohorts were recruited from rectal cancer patients, in whom the greatest challenge after PE is usually to cover a large perineal defect, while a lateral pelvic side wall resection is less extensive than in gynecologic tumors. Flaps are therefore used in these cases primarily to cover the perineal defect. In gynecological malignancies, it was shown that the vaginal reconstruction after a pelvic exenteration could dramatically lower the postoperative morbidity. The morbidity was compared in a series of 64 patients with a pelvic exenteration; 29 of them had vaginal reconstruction with TRAM and 8 with Singapore fascio-cutanous flap [11]. In those with neovagina formation, there were fewer cases with pelvic abscess (20% vs. 7%) and small bowel fistula (20% vs. 3%).
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
5
Table 3 Post-operative morbidity and disease status. Characteristics1 ICU (days) Hospital stay (days) Bowel peristalsis (days) Reoperation in 60 days A) Related to urinary diversion Ureteral anastomotic leak Small bowel obstruction caused by adhesions to the pouch Ureteral stricture B) Related potentially to pelvis empty syndrome Pelvis abscess GI fistula Small bowel obstruction caused by deep pelvis adhesions C) Other causes Wound complications Early postop. Complications Maximum grade 1 2 3 4 Performance status at 6 mo. 0–1 2 3 4 Late postop. Complications Maximum grade 1–2 3–4 Incisional hernia Disease status AWD DOD NED 1
Total (N = 40)
Control (N = 24)
MRAM (N = 16)
p2
7 (2; 35) 17 (8; 133) 5 (1; 15) N = 16 (40%) N = 7 (18%) N = 1 (3%) N = 5 (13%)
6 (2; 33) 18 (8; 133) 5 (2; 15) N = 12 (50%) N = 4 (17%) N = 1 (4%) N = 2 (8%)
7 (5; 35) 17 (14; 57) 5 (1; 11) N = 4 (25%) N = 3 (19%) N=0 N = 3 (19%)
0.824 0.455 0.131 0.188 0.999 0.999 0.373
N N N N N
N N N N N
N N N N N
0.999 0.114 0.508 0.999 0.631
= = = = =
1 (3%) 8 (20%) 2 (5%) 1 (3%) 5 (13%)
= = = = =
1 (4%) 7 (29%) 2 (8%) 1 (4%) 4 (17%)
= = = = =
0 1 (6%) 0 0 1 (6%)
N = 1 (3%) N = 32 (80%)
N = 1 (4%) N = 19 (80%)
N=0 N = 12 (75%)
0.999 0.999
N N N N
= = = =
8 (20%) 17 (43%) 12 (30%) 3 (7%)
N N N N
= = = =
4 (17%) 9 (38%) 8 (33%) 3 (12%)
N N N N
= = = =
4 (25%) 8 (50%) 4 (25%) 0
0.463
N N N N N
= = = = =
22 (55%) 10 (25%) 6 (15%) 2 (5%) 26 (65%)
N N N N N
= = = = =
9 (38%) 9 (38%) 4 (17%) 2 (8%) 19 (79%)
N N N N N
= = = = =
13 (81%) 1 (6%) 2 (12%) 0 7 (44%)
0.027
N = 16 (40%) N = 10 (25%) N = 4 (10%)
N = 10 (42%) N = 9 (37%) N = 3 (13%)
N = 6 (38%) N = 1 (6%) N = 1 (6%)
N = 5 (13%) N = 10 (25%) N = 25 (62%)
N = 4 (17%) N = 8 (33%) N = 12 (50%)
N = 1 (6%) N = 2 (12%) N = 13 (82%)
absolute and relative frequencies for categorical variables; median supplemented by min-max range for continuous variables;
Our approach was inventive in terms of the reconstruction goal, the modification of the flap, and the technique of the flap placement in the pelvis. The primary goal in our patients was pelvic floor reconstruction. Since the purpose was not to cover a perineal defect, both the skin and the anterior sheath were preserved, so even a bilateral flap would not compromise the anterior abdominal wall. It is the purpose of the flap that determines its placement in the pelvis. If the main aim is a perineal reconstruction, the flap must be placed vertically deeply to the pelvis. It is usually insufficient to simultaneously create a compact artificial pelvic floor and completely cover the denuded area of the obturator fossa, sacral bone, and other exposed structures in the pelvis. The main and only purpose of the flap transposition in our group was the pelvic floor reconstruction. One or two flaps were inserted horizontally into the pelvis at the level of the former pelvic floor, forming a U-shaped compact structure. If the pelvic defect was deep, the second flap from the other side was partially transposed below the new pelvic floor to fill in the deeper space. The newly formed pelvic floor was sufficient to create support for the small bowels, fill in the dead space and cover the denuded structures of the pelvic sidewall. Due to the complexity of each PE case, the number of factors influencing postoperative morbidity, and the limited number of procedures, a prospective design of the trial comparing patients with and without pelvic floor reconstruction is not an option. A control group in our study was composed by patients after PE without pelvic floor reconstruction performed in the same institution. It is predominantly a historical cohort, in which the majority of cases (15/24) were operated before 2012. The time periods overlap only partially, and the operation technique as well as post-operative care comparison might be biased. The improvement of the clinical practice is obvious when blood loss is
2
0.041 0.190 0.638 0.201
p-value of Fisher's exact test or Mann-Whitney U test.
compared. Blood loss is significantly lower in the more recent group of patients with MRAM. Operation time is identical in both groups, even though MRAM requires around 60 min of additional time. The decrease in blood loss and shrinking of operation time is most likely a result of the advanced bipolar technology we have been using since 2009. On the other hand, it is unlikely that these developments in surgical techniques or intensive care can prevent reoperations and other serious complications that are caused by a too-large unfilled defect in the pelvis, the absence of mechanical support of the bowels, and the formation of adhesions to the exposed structures of the pelvic sidewall. It should also be emphasized that the radicality of the procedures has risen simultaneously, as shown by the predominant occurrence of ELSE procedures in the MRAM group (5 of 6 cases). Rectus abdominis flaps are rarely associated with the flap loss [11, 18]. In a large series of 100 patients who received the rectus abdominis flap for vaginal or perineal reconstruction, there were only 2 cases of total flap necrosis, while in other smaller cohorts the rate was as high as 14%. There was no case of total or partial flap necrosis in our group. Another potential downside of the rectus abdominis flap is a weakening of the abdominal wall and a higher risk of donor-side complications. A risk of incisional hernias between 3 and 11% was reported after RAM flap transposition [9,19–20]. When using the myoperitoneal flap (MRAM), a complete anterior fascia sheath with rich vascular supply from subcutaneous tissue is preserved, which secures the integrity of the anterior abdominal wall. It is reassuring that we only had one patient with incisional hernia (not requiring any surgery) and no case of parastomal hernia in our cohort. In our series, MRAM was used in 16 consecutive women after anterior, total, or posterior PE, and / or ELSE. In all cases, a short (up to the
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014
6
D. Cibula et al. / Gynecologic Oncology xxx (2017) xxx–xxx
umbilicus level) or long (up to the ribs) MRAM was transposed to the pelvis unilaterally or bilaterally with the preservation of the anterior fascia sheath. Pelvic floor reconstruction by MRAM was associated with substantially fewer reoperations in 60 days due to complications potentially related to pelvis empty syndrome, as well as with significantly lower risk of serious late postoperative complications, and an improved performance status 6 months after the operation. Conclusion Pelvic floor reconstruction by MRAM after extensive pelvic exenterative procedures is associated with a decrease in the number of reoperations due to complications potentially related to empty pelvis syndrome, a lower occurrence of severe late postoperative complications, and a better performance status 6 months after the surgery. The experience from one institution did not show an increase in procedure related morbidity such as flap necrosis or incisional hernia. Conflict of interest There are no conflicts of interest.
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
Acknowledgements This work was supported by Charles University in Prague (UNCE 204024 and PRVOUK-P27/LF1/1) and by the project of Ministry of Health of the Czech Republic (MZ CR – RVO VFN64165). References [1] A. Maggioni, G. Roviglione, F. Landoni, V. Zanagnolo, M. Peiretti, N. Colombo, L. Bocciolone, R. Biffi, L. Minig, C.P. Morrow, Pelvic exenteration: ten-year experience at the European Institute of Oncology in Milan, Gynecol. Oncol. 114 (2009) 64–68. [2] V. Chiantera, M. Rossi, P. De Iaco, C. Koehler, S. Marnitz, G. Ferrandina, F. Legge, F. Parazzini, G. Scambia, A. Schneider, G.F. Vercellino, Survival after curative pelvic exenteration for primary or recurrent cervical cancer: a retrospective multicentric study of 167 patients, Int. J. Gynecol. Cancer 24 (2014) 916–922. [3] A.M. Schmidt, P. Imesch, D. Fink, H. Egger, Indications and long-term clinical outcomes in 282 patients with pelvic exenteration for advanced or recurrent cervical cancer, Gynecol. Oncol. 125 (2012) 604–609. [4] A. Brunschwig, H.R. Barber, Pelvic exenteration combined with resection of segments of bony pelvis, Surgery 65 (1969) 417–420. [5] D.Q. Nguyen, A.D. McGregor, O. Freites, N.D. Carr, J. Beynon, A.M. El-Sharkawi, M.G. Lucas, Exenterative pelvic surgery–eleven year experience of the Swansea pelvic oncology group, Eur. J. Surg. Oncol. 31 (2005) 1180–1184. [6] A. Urh, P.T. Soliman, K.M. Schmeler, S. Westin, M. Frumovitz, A.M. Nick, B. Fellman, D.L. Urbauer, P.T. Ramirez, Postoperative outcomes after continent versus
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
incontinent urinary diversion at the time of pelvic exenteration for gynecologic malignancies, Gynecol. Oncol. 129 (2013) 580–585. G.L. Goldberg, P. Sukumvanich, M.H. Einstein, H.O. Smith, P.S. Anderson, A.L. Fields, Total pelvic exenteration: the Albert Einstein College of Medicine/Montefiore medical center experience (1987 to 2003), Gynecol. Oncol. 101 (2006) 261–268. P.S. Ghosh, H. Fawzi, Empty pelvis syndrome, J. Obstet. Gynaecol. 24 (2004) 714–715. R.E. Horch, W. Hohenberger, A. Eweida, U. Kneser, K. Weber, A. Arkudas, S. Merkel, J. Göhl, J.P. Beier, A hundred patients with vertical rectus abdominis myocutaneous (VRAM) flap for pelvic reconstruction after total pelvic exenteration, Int. J. Color. Dis. 29 (2014) 813–823. D.B. Chessin, J. Hartley, A.M. Cohen, M. Mazumdar, P. Cordeiro, J. Disa, B. Mehrara, B.D. Minsky, P. Paty, M. Weiser, W.D. Wong, J.G. Guillem, Rectus flap reconstruction decreases perineal wound complications after pelvic chemoradiation and surgery: a cohort study, Ann. Surg. Oncol. 12 (2005) 104–110. M. Jurado, A. Bazán, J.L. Alcázar, E. Garcia-Tutor, Primary vaginal reconstruction at the time of pelvic exenteration for gynecologic cancer: morbidity revisited, Ann. Surg. Oncol. 16 (2009) 121–127. D. Dindo, N. Demartines, P.A. Clavien, Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey, Ann. Surg. 240 (2) (2004 Aug) 205–213. H.J. Yoo, M.C. Lim, S.S. Seo, S. Kang, C.W. Yoo, J.Y. Kim, S.Y. Park, Pelvic exenteration for recurrent cervical cancer: ten-year experience at National Cancer Center in Korea, J. Gynecol. Oncol. 23 (2012) 242–250. S. Marnitz, S. Dowdy, M. Lanowska, A. Schneider, K. Podratz, C. Köhler, Exenterations 60 years after first description: results of a survey among US and German gynecologic oncology centers, Int. J. Gynecol. Cancer 19 (2009) 974–977. C.S. Hultman, M.A. Sherrill, E.G. Halvorson, C.N. Lee, J.F. Boggess, M.O. Meyers, B.A. Calvo, H.J. Kim, Utility of the omentum in pelvic floor reconstruction following resection of anorectal malignancy: patient selection, technical caveats, and clinical outcomes, Ann. Plast. Surg. 64 (2010) 559–562. J.M. Hay, A. Fingerhut, J.C. Paquet, Y. Flamant, Management of the pelvic space with or without omentoplasty after abdominoperineal resection for carcinoma of the rectum: a prospective multicenter study, Eur J Surg 163 (1997) 199–206. G.J. Poston, S.R. Smith, W.N. Baker, Retrocolic pelvic omentoplasty in abdominoperineal excision of the rectum, Ann. R. Coll. Surg. Engl. 73 (1991) 229–232. J.T. Soper, L.J. Havrilesky, A.A. Secord, A. Berchuck, D.L. Clarke-Pearson, Rectus abdominis myocutaneous flaps for neovaginal reconstruction after radical pelvic surgery, Int. J. Gynecol. Cancer 15 (2005) 542–548. R.J. Chokshi, M.P. Kuhrt, D. Arrese, E.W. Martin Jr., Reconstruction of total pelvic exenteration defects with rectus abdominus myocutaneous flaps versus primary closure, Am. J. Surg. 205 (2013) 64–70. T.A. Creagh, L. Dixon, F.A. Frizelle, Reconstruction with vertical rectus abdominus myocutaneous flap in advanced pelvic malignancy, J. Plast. Reconstr. Aesthet. Surg. 65 (2012) 791–797. R.A. Nelson, C.E. Butler, Surgical outcomes of VRAM versus thigh flaps for immediate reconstruction of pelvic and perineal cancer resection defects, Plast. Reconstr. Surg. 123 (2009) 175–183. A.O. Momoh, A.M. Kamat, C.E. Butler, Reconstruction of the pelvic floor with human acellular dermal matrix and omental flap following anterior pelvic exenteration, J. Plast. Reconstr. Aesthet. Surg. 63 (2010) 2185–2187. G.C. Guimarães, G. Baiocchi, B.M. Rossi, F.O. Ferreira, S. Aguiar, W.T. Nakagawa, A. Lopes, The use of silicone expander and cecal transposition after pelvic exenteration, Eur. J. Surg. Oncol. 33 (2007) 586–589.
Please cite this article as: D. Cibula, et al., Pelvic floor reconstruction by modified rectus abdominis myoperitoneal (MRAM) flap after pelvic exenterations, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.01.014