Evaluation of the Short- and Long-Term Outcome Predictors in Patients Undergoing Posterior Pelvic Exenteration: A Single-Center Experience

Abstract

Background: Pelvic exenteration (PE) is a radical surgery and is actually used to manage advanced cancers of differing origins. Risk factors for poor prognosis, correlated to this intervention, have been partially evaluated; thus it is useful to make an accurate assessment of the predictors correlated to short- and long-term outcomes with use of PE. Materials and Methods: A retrospective review was conducted involving 28 patients who underwent posterior PE, with curative intent; the procedure was performed for gynecologic and nongynecologic neoplasms at the University of Messina, Messina, Sicily, Italy. Results: The overall morbidity rate was 46.4%, the incidence of severe complications (grade 3/4) was 14%, and the perioperative mortality was 7.1%. A median survival of 29 months was recorded for the entire sample, with 3- and 5-year overall survivals of 37.5% and 16.7%, respectively. Conclusions: The only factor—when using a multivariate model—that was significantly associated with morbidity was Eastern Cooperative Oncology Group (ECOG) performance status. This study demonstrated that the origins of primary cancers did not affect patients' long-term survival significantly, while higher numbers of anastomoses, major complications, and, above all, higher ECOG scores were associated with shorter long-term survival. Based on the experience illustrated in this study, it is fundamentally important to select patients for treatment, paying special attention to their ECOG scores. (J GYNECOL SURG 32:84)

Introduction

Pelvic exenteration (PE) is a radical surgery that was first pioneered in gynecologic oncology in 1948 by Brunschwig,¹ and has been actually used in the management of advanced cancers of different origins. This surgery removes all the pelvic organs, including the uterus, ovaries, and vagina as well as the adjacent organs. Initially, the procedure was performed for purely palliative purposes and resulted in long-term survival of only a few patients.¹ Candidates for this procedure are those in whom curative alternatives are inferior or have been exhausted for treating advanced primary or locally recurrent cancers confined to the pelvis.² Improvements in surgical techniques, perioperative planning, and selection criteria have led to significant changes in perioperative morbidity and mortality after PE.^3–11 The most threatening complications involve the gastrointestinal (GI) and urinary systems, especially acute enteric complications, which can exceed 20%; fistulization; pelvic infection; sepsis; wound infection; and pyelonephritis that increase mortality and morbidity rates especially after PE for gynecologic cancers.^12,13

Today, PE is therefore considered to be a feasible procedure that offers certain patients a chance of life by shifting from a palliative procedure to a potentially curative one. In addition, given that risk factors for poor prognosis after PE have been partially evaluated,^5,14–16 it is worthwhile to assess predictors correlated to short-term outcomes accurately and study which predictors influence long-term survival.

In the study described in this article, a retrospective review was conducted involving patients who had undergone posterior pelvic exenteration (PPE), with curative intent, for treating gynecologic and nongynecologic neoplasms. These procedures were performed at the University of Messina, in Messina, Sicily, Italy. The aim of the study was to identify predictors influencing short- and long-term outcomes of the patients who underwent PPE.

Materials and Methods

Inclusion of patients and presurgical staging

Twenty-eight patients who underwent PPE, with curative intent, from 2000 to 2014, were included in the study. Fifteen patients had primary ovarian cancer, 2 patients had cervical cancer, 7 patients had colorectal cancer, and 4 had cancer of different origins (2 patients had uterine sarcoma, 1 patient had pseudomyxoma peritonei, and 1 had peritoneal mesothelioma).

All procedures followed for this study were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008.¹⁷ Informed consent was obtained from each patient for inclusion in the study.

Before surgery all the patients' cancers were staged radiologically via abdominal–pelvic ultrasound testing, computed tomography, and/or magnetic resonance imaging, to evaluate disease extent and to estimate the likely extent of necessary surgery. If the presence of metastatic disease was doubtful, positron emission tomography was performed. All of the cases were discussed at a multidisciplinary team meeting.

Hyperthermic intraperitoneal intraoperative chemotherapy (HIPEC) was performed in 15 patients because of peritoneal and pelvic carcinomatosis, using cisplatinum plus mytomicin C in 4 cases and cisplatinum plus doxorubicin in the other 11 cases.

Statistics

All participating physicians received extensive training in data collection techniques. Information about the study patients was collected via dedicated case-report forms and recorded in a central database in a fully anonymous fashion. Data collected included patient demographics, anthropometric measures, medical histories, clinical information, drug prescriptions, surgical procedures, and prognoses. Baseline characteristics of patients were expressed as medians together with range intervals for continuous variables and as absolute and percentage frequencies for categorical variables. Demographic and clinical characteristics of each patient were analyzed with a specific focus on postoperative morbidity or exitus. Postoperative morbidity was evaluated in accordance with the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE) v4.0.¹⁸

Cumulative incidence of postoperative morbidity was calculated as the ratio between the number of patients in which complications occurred and the number of individuals who underwent surgical procedures. Both total (grade 1/4) and severe (grade 3/4) postoperative morbidity incidence were expressed as rate per 100 patients, together with 95% confidence intervals (CIs). Postoperative mortality was evaluated as patients' death occurring within 60 days postsurgically. Overall survival time was calculated as the number of months from the date of surgery to death or to the end of follow-up. To identify predictors of postoperative morbidity, a univariate logistic regression model, using absence of complications as a comparator, was applied to assess the possible influence of each covariate of interest.

Patients' characteristics—such as age, body mass index (BMI), American Society of Anesthesiology score (ASA), Eastern Cooperative Oncology Group (ECOG) performance status, ascites, neoadjuvant chemotherapy (NACT), HIPEC technique, Peritoneal Cancer Index (PCI), number of anastomosis, duration of surgery, number of blood transfusions, albumin infusion, and Completeness of Cytoreduction score (CC score)—were evaluated to clarify the possible influence of these factors on the occurrence of complications. Moreover, predictors that emerged as significant when using the univariate model were included in a multivariate logistic regression model (adjusted odds ratio; OR).

A nonparametric two-tailed Rho Spearman test was used to assess possible associations among all the predictors that emerged as significant using the univariate model. This latter test was performed to avoid multicollinearity and to identify key predictors better so that these predictors could be included in the multivariate model. A receiver operating characteristic (ROC) curve by complications was used to assess the optimal duration of surgery to avoid postsurgical morbidity. To describe long-term survival time, a Kaplan–Meier time-to-event curve was used to estimate the survival time to event until the end of follow-up.

To assess differences in long-term survival among patients with different primary cancers, the lengths of time to death were compared using a log-rank test. Moreover, univariate Cox regression analysis was performed to assess the possible dependence of the event (death) by some potential predictors as age, BMI, ECOG score, NACT procedure, HIPEC procedure, PCI score, number of anastomosis, CC score, ascites, number of intraoperative transfusions, albumin infusions, complications (grade 1/4), or major complications (grade 3/4).

For real predictors, a multivariate Cox regression model was used, including all significant predictors found through the univariate approach. To show differences among patients in which complications occurred or did not occur, incidence ratio, hazard ratio, and mean survival time, each with a 95% CI, were evaluated. The lengths of time to death were compared with a log-rank test, and the relationships between death and complications, were calculated with the Cox proportional-hazards model. A two-tailed p-value was set at 0.05. ORs with 95% CIs were calculated for each covariate of interest. Statistical analyses were performed with SPSS.20.0 (SPSS Statistics, IBM).

Results

The median BMI of the patients was 24 (range: 17–38). The ECOG scores were 0 for 14 patients (50%), 1 for 12 patients (42.8%), and 2 for 2 patients (7.2%). ASA score was 1 for 6 patients (21.4%), 2 for 13 patients (46.4%), and 3 for 9 patients (32.2%). Ten patients (35.7%) were given neoadjuvant chemotherapy before PPE. Fifteen patients (53.5%) underwent HIPEC with a closed technique, because of peritoneal neoplastic diffusion. Eleven patients (39.3%) had preoperative ascites. The average PCI in patients with peritoneal involvement was 13 (range 2–24; Table 1).

Table 1.

Patients' Characteristics

Characteristics	Median (range or %)
Age (years)	59 (34–74)
BMI (kg/m²)	24 (17–38)
ECOG, # (%)
0	14 (50%)
1	12 (42.8%)
2	2 (7.2%)
ASA score, # (%)
1	6 (21.4%)
2	13 (46.4%)
3	9 (32.2%)
Treatment before PE, # (%)
NACT	10 (35.7%)
HIPEC, # (%)	15 (53.5%)
Histology, # (%)
Serous-cell carcinoma	15 (53.6%)
Adenocarcinoma	9 (32.1%)
Others	4 (14.3%)
FIGO stage at diagnosis, # (%)
II	2 (7.1%)
II C	6 (21.4%)
III B	9 (32.2%)
TNM^a stage
T4b N2a M1	2 (7.1%)
T4a N0 Mo	5 (17.9%)
Others	4 (14.3%)
Preoperative ascites, # (%)	11 (39.3%)
PCI	13 (2–24)

The TNM Staging of Malignant Tumours uses the following codes: T, for size of primary tumor and if it has invaded nearby tissues; N, for regional lymph nodes that are involved; M, for distant metastasis.

BMI, body mass index; ECOG, Eastern Cooperative Oncology Group; ASA American Society of Anesthesiology; PE, pelvic exenteration; NACT, neoadjuvant chemotherapy; HIPEC, hyperthermic intraperitoneal intraoperative chemotherapy; FIGO, International Federation of Gynecology and Obstetrics; PCI, Peritoneal Cancer Index.

The HIPEC technique was used as previously described in the current authors' reports.^19,20 The median number of intraoperatively given red blood cell units was 2 (range: 1–9). The median operation time was 446 minutes (range: 240–720). CC score was 0 in 24 patients (85.7%) and 1 in 4 patients (14.3%). The median hospital stay was 13 days (range: 7–63). All patients, except for the 1 patient who had pseudomyxoma, were given adjuvant chemotherapy.

The median survival of whole specimens was 29 months (95% CI: 19.5–38.5), with a 3-year and 5-year overall survival of 37.5 and 18.2 respectively, while the median survival analyzed for primitiveness was 28 months and 26 months for patients with ovarian and colorectal cancers, respectively, with 3-year and 5-year overall survivals of 50% and 16.7% for ovarian cancer, respectively, and 16.7% and 16.7% for colorectal cancer, respectively. Perioperative mortality was observed in 2 patients (7.1%).

The overall morbidity rate was 46.4% (95% CI: 28.0–64.9; 13 patients), while severe complications (grade 3/4) were observed in 4 patients (14%; 95% CI: 1.3–27.2). Wound infection, occurring in 9 patients, was the most common complication. Other common complications included 6 patients with fever, 5 patients with postoperative ileus, 4 patients with abdominal abscess, 3 patients with gastric stasis, 3 patients intestinal perforation, 3 patients with sepsis, 2 patients with pleural effusion, and 2 patients with arrhythmia (Table 2). Compared to patients without postsurgical morbidity, patients who had complications were older (OR: 1.21; 95% CI: 1.03–1.42; p = 0.018), with higher ASA scores (OR: 7.61; 95% CI: 2.58–36.6; p = 0.011), and higher ECOG performance status (OR: 18.92; 95% CI: 2.73–130.98; p < 0.01). Moreover, PCI (OR: 1.13; 95% CI: 1.00–1.27; p = 0.040), number of anastomosis (OR: 4.47; 95% CI: 1.09–18.30; p = 0.038), and duration of surgery (OR: 1.01; 95% CI: 1.0–1.01; p = 0.022), were directly associated with the occurrence of complications. Patients with complications needed, obviously, greater hospitalization times consistently (OR: 1.43; 95% CI: 1.07–1.91 p = 0.017).

Table 2.

Postoperative Complications

Complications	#
Patients with complications	13 (46.4%)
Anemia	1
Leukopenia	1
Arrhythmia	2
Wound infection	9
Intestinal perforation	3
Hemoperitoneum	1
Anastomotic leakage	1
Abdominal abscess	4
Urinary infection	1
Urinary leakage	1
Postoperative ileus	5
Gastric stasis	3
Pleural effusion	2
Respiratory distress	1
Fever	6
Sepsis	3
MOF	1
DIC	1

MOF, multi organ failure; DIC, disseminated intravascular coagulation; OR, odds ratio; CI, confidence interval; BMI, body mass index; ASA American Society of Anesthesiology; ECOG, Eastern Cooperative Oncology Group; NACT, neoadjuvant chemotherapy; HIPEC, hyperthermic intraperitoneal intraoperative chemotherapy; PCI, Peritoneal Cancer Index; CC Completeness of Cytoreduction.

The Spearman correlation test highlighted a significant correlation between ASA score and ECOG performance status (Spearman correlation coefficient [rs]: 0.759; p < 0.001) and between PCI and duration of surgery (rs: 0.744; p < 0.001). Thus, only the most significant variable between the two related factors (ECOG and PCI) was chosen, while ASA score and duration of surgery were excluded from the multivariate model to avoid a multicollinearity condition. The significant associations with the predictors evaluated in the univariate approach persisted only for ECOG performance status (OR: 18.89; 95% CI: 2.71–130.95; p < 0.01) when estimated as adjusted ORs, using the multivariate logistic regression model (Table 3). Furthermore, duration of surgery lasting up 320 or 390 minutes was associated with an absence of complications with a 92.3% or 69.2% of sensitivity and 53.3% or 73.3% of specificity, respectively, as shown by the ROC curve.

Table 3.

Predictors of Postsurgical Complications

	Univariate model			Multivariate model
Patients' characteristics	OR_crude	95% CI	p	OR_adj	95% CI	p
Age	1.21	(1.03–1.42)	0.018	1.12	0.99–1.33	0.209
BMI	0.98	(0.82–1.16)	0.797	—	—	—
ASA	7.61	(1.58–36.66)	0.011	—	—	—
ECOG	18.9	(2.73–130.98)	0.003	18.8	2.71–130.95	0.003
Ascites	1.71	(0.37–7.92)	0.490	—	—	—
NACT	1.72	(0.35–8.51)	0.507	—	—	—
HIPEC	1.83	(0.40–8.27)	0.433	—	—	—
PCI	1.13	(1.01–1.26)	0.040	0.92	0.77–1.14	0.468
Number of anastomoses	4.46	(1.09–18.3)	0.038	2.85	0.49–16.47	0.241
Duration of surgery	1.01	(1.00–1.01)	0.022	—	—	—
Number of transfusions	1.55	(0.93–2.57)	0.092	—	—	—
Albumin	8.00	(0.81–78.7)	0.075	—	—	—
CC score	2.33	(0.18–30.4)	0.518	—	—	—

OR, odds ratio; CI, confidence interval; BMI, body mass index; ASA American Society of Anesthesiology; ECOG, Eastern Cooperative Oncology Group; NACT, neoadjuvant chemotherapy; HIPEC, hyperthermic intraperitoneal intraoperative chemotherapy; PCI, Peritoneal Cancer Index; CC Completeness of Cytoreduction.

Overall perioperative mortality was observed in 2 patients and, thus, overall survival time was described in 26 patients using the Kaplan–Meier time-to-event curve (Fig. 1A). Primary cancer (ovarian, colon, other) did not influence patients' long-term survival time significantly (log rank: 0.076; p = 0.963; Fig. 1B), while median survival time of patients with postsurgical morbidity (20 months; 95% CI: 8.1–31.8) was significantly lower (log rank: 4.70; p = 0.03) respect to patients who did not have complications (46 months; 95% CI: 18.0–74.0; Fig. 1C).

FIG. 1.

Overall survival (A) or stratified by: primary cancer (B); postsurgical complications (C); or major postsurgical complications (D).

Major complications (grade 3/4) were inversely associated with long-term survival consistently, with a median survival time of 6 months and 30 months (95% CI: 23.9–36.1), respectively, in patients with or without major complications (log rank: 12.83; p < 0.001; Fig. 1D). Moreover, using the univariate Cox regression analysis, to assess the possible dependence of long-term survival time on potential predictors, ECOG performance status (hazard ratio [HR]: 3.64; 95% CI: 1.47–8.98), number of anastomosis (HR: 2.57; 95% CI: 1.18–5.59), and postoperative complications (HR: 2.68; 95% CI: 1.05–6.87) were related significantly to overall survival.

Nevertheless, a fully adjusted model, including all significant predictors found via the univariate approach, confirmed the results from the main analysis only for ECOG performance status (HR: 3.36; 95% CI: 1.36–8.31; Table 4). However, in patients free from postsurgical complications, ECOG performance status was not associated with overall survival (HR 1.25; 95% CI: 0.25–6.22; p = 0.79) while the risk of mortality increased significantly in patients who had complications (HR 4.66; 95% CI: 1.04–20.79; p = 0.04).

Table 4.

Predictors of Overall Survival Time

	Cox regression univariate model			Cox regression multivariate model ^a
Patients' characteristics	HR_crude	95% CI	p	HR_adj	95% CI	p
Age	1.07	(0.99–1.15)	0.08	—	—	—
BMI	0.95	(0.85–1.07)	0.42	—	—	—
ASA	1.76	(0.77–4.04)	0.18	—	—	—
ECOG	3.64	(1.47–8.98)	0.005	3.36	(1.36–8.31)	0.009
NACT	1.69	(0.63–4.57)	0.29	—	—	—
HIPEC	0.99	(0.38–2.56)	0.98	—	—	—
PCI	1.03	(0.98–1.09)	0.21	—	—	—
Number of anastomosis	2.57	(1.18–5.59)	0.017	1.77	(0.77–4.05)	0.18
CC score	3.61	(0.89–14.63)	0.07	—	—	—
Ascites	1.68	(0.66–4.25)	0.27	—	—	—
Number of transfusion	0.95	(0.77–1.18)	0.67	—	—	—
Albumin	2.54	(0.73–8.82)	0.14	—	—	—
Total complications (grades 1–4)	2.68	(1.05–6.87)	0.040	1.73	(0.60–5.00)	0.31

COX regression multivariate fully adjusted model, including all significant predictors at univariate approach.

HR, hazard ratio; CI, confidence interval; BMI, body mass index; ASA American Society of Anesthesiology; ECOG, Eastern Cooperative Oncology Group; NACT, neoadjuvant chemotherapy; HIPEC, hyperthermic intraperitoneal intraoperative chemotherapy; PCI, Peritoneal Cancer Index; CC Completeness of Cytoreduction.

Discussion

Brunschwig initially described the technique of PE as a palliative option for addressing advanced cervical cancer.¹ The procedure has also evolved to become a treatment option for locally advanced and recurrent pelvic malignancies, such as those of rectal, gynecologic, urologic, and sarcomatous origins. However, the procedure was purely palliative and the operation resulted in the long-term survival of only a few patients.¹ PE is classified into five groups: (1) anterior PE, which, in addition to resection of central pelvic organs, includes removal of the bladder and distal ureters bilaterally; (2) PPE, which involves removal of the central organs with the rectosigmoid colon (with or without the anal canal); (3) total PE, a combination of both anterior and PPEs; (4) extended exenteration, which includes abdominosacral resection, and (5) individualized approaches, such as rectal excision with concomitant radical prostatectomy with preservation of the bladder.²¹

After the initial description,¹ current improvements in surgical techniques, perioperative planning, selection criteria, anesthesia care, and perioperative care led to significant reductions in perioperative mortality, from >30% to more-acceptable rates of 0%–10%, with 5-year overall survival ranging from 30% to 60% in patients with gynecologic tumors and 25%–40% in patients with colorectal cancer.

However, reported morbidity rates were still high, ranging from 13% to 83% in various series.^3–11 The most threatening complications were those involving the GI and urinary systems, especially acute enteric complications, fistulization, pelvic infection, sepsis, wound infection, and pyelonephritis, which increased mortality and morbidity rates especially after PE for gynecologic cancers.^12,13 However, after this demolition-type operation, even if patients reported lingering GI symptoms and some persistent declines in physical function, most patients adjusted well, returning to almost baseline functioning within a year.²²

A number of clinical factors have been reported that affect poor prognosis in patients who underwent PE,^23–26 but the most important of these are represented by resection margin status, pelvic wall involvement and rectal involvement.^9,27–31 In addition as showed by Park et al. in their report,³² tumor size exceeding 4 cm was a predictive factor for recurrence, above all, after gynecologic surgery. In the current authors' study, the overall morbidity rate was 46.4%, the incidence of severe complications (grade 3/4) was 14%, and the perioperative mortality was 7.1%. The median survival recorded for the entire sample was 29 months, with 3-year and 5-year overall survival of 37.5% and 16.7%, respectively. In addition, analyzing the same parameters for neoplastic primitiveness showed that there was a median survival of 28 months for ovarian cancer with 3-year and 5-year overall survival of 50% and 16.7%, respectively, and a median survival of 26 months with 3-year and 5-year overall survival of 16.7% and 16.7%, respectively for colorectal cancer.

Among the predictors evaluated, it was demonstrated that patients with complications were more likely to be older, with higher ASA scores and ECOG performance status. Moreover, PCI, number of anastomoses were directly associated with the occurrence of complications; duration of surgery was less—but significantly—associated with the occurrence of complications. However the only factor in the multivariate model that was significantly associated with morbidity was ECOG performance status. Duration of surgery between 320 and 390 minutes was associated with absence of complications in 92.3% or 69.2% of cases, respectively. However, in this restricted sample, it was not possible to calculate an accurate cutoff.

Moreover the current authors noted that the origin of the primary cancer did not affect patients' long-term survival time significantly; higher ECOG performance status, higher numbers of anastomoses, and major complications were associated with shorter long-term survival. Nevertheless, the fully adjusted model confirmed the results from the main analysis only for ECOG performance status.

Conclusions

Although significant advances have been made in radiotherapy and chemotherapy, PE still remains the only chance for the control of locally advanced primary and recurrent pelvic malignancies that remain unresponsive to nonoperative therapies. Today, PE can play a curative role instead of only a palliative one. The current authors maintain, on the basis of the experience shown in this current study, that it is essential to make accurate selections of patients as PE candidates, especially with particular regard to ECOG performance status.

Footnotes

Disclosure Statement

The authors declare that they have no commercial associations that might create conflicts of interest in connection with this article.

References

Brunschwig

. Complete excision of pelvic viscera for advanced carcinoma: A one-stage abdominoperineal operation with end colostomy and bilateral ureteral implantation into the colon above the colostomy. Cancer, 1948; 1:177.

Kaur

, Joniau

, D'Hoore

, Vergote

. Indications, techniques and outcomes for pelvic exenteration in gynecological malignancy. Curr Opin Oncol, 2014; 26:514.

Boey

, Wong

, Ong

. Pelvic exenteration for locally advanced colorectal carcinoma. Ann Surg, 1982; 195:513.

Morley

, Hopkins

, Lindenauer

, Roberts

. Pelvic exenteration, University of Michigan: 100 patients at 5 years. Obstet Gynecol, 1989; 74:934.

Stanhope

, Webb

, Podratz

. Pelvic exenteration for recurrent cervical cancer. Clin Obstet Gynecol, 1990; 33:897.

Berek

, Howe

, Lagasse

, Hacker

. Pelvic exenteration for recurrent gynecologic malignancy: Survival and morbidity analysis of the 45-year experience at UCLA. Gynecol Oncol, 2005; 99:153.

Maggioni

, Roviglione

, Landoni

, et al. Pelvic exenteration: Ten-year experience at the European Institute of Oncology in Milan. Gynecol Oncol, 2009; 114:64.

Fleisch

, Pantke

, Beckmann

, et al. Predictors for long-term survival after interdisciplinary salvage surgery for advanced or recurrent gynecologic cancers. J Surg Oncol, 2007; 95:476.

Yoo

, Lim

, Seo

, Kang

, Yoo

, Kim

, Park

. Pelvic exenteration for recurrent cervical cancer: Ten-year experience at National Cancer Center in Korea. J Gynecol Oncol, 2012; 23:242.

10.

Kuhrt

, Chokshi

, Arrese

, Martin

Jr . Retrospective review of pelvic malignancies undergoing total pelvic exenteration. World J Surg Oncol, 2012; 10:110.

11.

Chiantera

, Rossi

, De Iaco

, et al. Morbidity after pelvic exenteration for gynecological malignancies: A retrospective multicentric study of 230 patients. Int J Gynecol Cancer, 2014; 24:156.

12.

Bladou

, Houvenaeghel

, Delpéro

, Guérinel

. Incidence and management of major urinary complications after pelvic exenteration for gynecological malignancies. J Surg Oncol, 1995; 58:91.

13.

Wydra

, Emerich

, Sawicki

, Ciach

, Marciniak

. Major complications following exenteration in cases of pelvic malignancy: A 10-year experience. World J Gastroenterol, 2006; 12:1115.

14.

Roos

, Van Eijkeren

, Boon

, Heintz

. Pelvic exenteration as treatment of recurrent or advanced gynecologic and urologic cancer. Int J Gynecol Cancer, 2005; 15:624.

15.

Sharma

, Odunsi

, Driscoll

, Lele

. Pelvic exenterations for gynecological malignancies: Twenty-year experience at Roswell Park Cancer Institute. Int J Gynecol Cancer, 2005; 15:475.

16.

Goldberg

, Sukumvanich

, Einstein

, Smith

, Anderson

, Fields

. Total pelvic exenteration: The Albert Einstein College of Medicine/Montefiore Medical Center experience (1987 to 2003). Gynecol Oncol, 2006; 101:261.

17.

Helsinki Declaration of 1975, revised in 2008.

18.

Cancer Therapy Evaluation Program, National Cancer Institute. Protocol Development. https://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm Accessed October 25, 2014.

19.

Saladino

, Fleres

, Mazzeo

, et al. The role of prophylactic hyperthermic intraperitoneal chemotherapy in the management of serosal involved gastric cancer. Anticancer Res, 2014; 34:2019.

20.

Macrì

, Saladino

, Trimarchi

, Bartolo

, Rossitto

, Cannaò

, et al. Cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy in elderly patients. In Vivo, 2011; 25:687.

21.

Nguyen

, Mc Gregor

, Freites

, et al. Exenterative pelvic surgery: Eleven year experience of the Swansea Pelvic Oncology Group. Eur J Surg Oncol, 2005; 31:1180.

22.

Rezk

, Hurley

, Carter

, et al. A prospective study of quality of life in patients undergoing pelvic exenteration: Interim results. Gynecol Oncol, 2013; 128:191.

23.

Einenkel

, Holler

, Hoffmeister

. Sonographic diagnosis and Endo-SPONGE assisted vacuum therapy of anastomotic leakage following posterior pelvic exenteration for ovarian cancer without using a protective stoma. J Gynecol Oncol, 2011; 22:131.

24.

Anthopoulos

, Manetta

, Larson

, Podczaski

, Bartholomew

, Mortel

. Pelvic exenteration: A morbidity and mortality analysis of a seven-year experience. Gynecol Oncol, 1989; 35:219.

25.

Rutledge

, Smith

, Wharton

, O'Quinn

. Pelvic exenteration: Analysis of 296 patients. Am J Obstet Gynecol, 1977; 129:881.

26.

Shingleton

, Soong

, Gelder

, Hatch

, Baker

, Austin

Jr . Clinical and histopathologic factors predicting recurrence and survival after pelvic exenteration for cancer of the cervix. Obstet Gynecol, 1989; 73:1027.

27.

Roberts

, Cavanagh

, Bryson

, Lyman

, Hewitt

. Major morbidity after pelvic exenteration: A seven-year experience. Obstet Gynecol, 1987; 69:617.

28.

Mitulescu

, Ungureanu

, Gluck

, et al. Total pelvic exenteration in the treatment of advanced pelvic cancer [in Romanian]. Chirurgia (Bucur), 2007; 102:143.

29.

Moriya

, Akasu

, Fujita

, Yamamoto

. Aggressive surgical treatment for patients with T4 rectal cancer. Colorectal Dis, 2003; 5:427.

30.

Yamada

, Ishizawa

, Niwa

, Chuman

, Akiba

, Aikou

. Patterns of pelvic invasion are prognostic in the treatment of locally recurrent rectal cancer. Br J Surg, 2001; 88:988.

31.

Mourton

, Chi

, Sonoda

, et al. Mesorectal lymph node involvement and prognostic implications at total pelvic exenteration for gynecologic malignancies. Gynecol Oncol, 2006; 100:533.

32.

Park

, Choi

, Jeong

, Chung

, Park

. The role of pelvic exenteration and reconstruction for treatment of advanced or recurrent gynecological malignancies: Analysis of risk factors predicting recurrence and survival. J Surg Oncol, 2007; 96:560.