Pelvic Exenteration: Oncological and Surgical Outcomes From a Single Tertiary Statewide Service

ABSTRACT

Background

Advancements in the multidisciplinary management of locally advanced and recurrent rectal cancer, as well as advanced pelvic malignancies, have significantly improved oncological outcomes. Pelvic exenteration is a complex procedure involving en bloc resection of the involved organs with the aim of achieving a microscopic negative margin (R0)¹. Since 2012, the Royal Brisbane and Women's Hospital (RBWH) has been performing pelvic exenteration surgery and is now the statewide referral centre in Queensland. This study aims to present our oncological and surgical outcomes since establishment.

Methods

This is a retrospective cohort study utilising data collected from medical records of patients who underwent pelvic exenteration at the RBWH between 1 January 2012 and 31 August 2023. Ethics approval was obtained from the Metro North Hospital and Health Service Human Research Ethics Committee (HREC 102257).

Results

A total of 139 patients underwent pelvic exenteration surgery over the 11‐year period. 117 of those were total exenteration, 22 were partial. The indications were locally advanced rectal cancer (56.8%), locally recurrent rectal cancer (22.3%), other primary or recurrent malignancy or benign, (10.8%, 7.9%, 2.2% respectively). The proportion of patients with an R0 resection was 90.4%. The overall survival (using Kaplan–Meier analysis) at 1, 2 and 5 years was 89.9%, 82.0%, and 62.9%, respectively. Disease‐free survival at 1, 2 and 5 years was 76.1%, 67.0% and 59.2%. Disease‐specific survival at 1, 2 and 5 years was 92.7%, 83.8% and 66.8%.

Conclusion

Our results were comparable to national data and PelvEx Collaborative group data and demonstrate that pelvic exenteration surgery can be safely performed in a lower volume centre with equivalent outcomes to higher volume centres.

1. Introduction

Over the last 30 years, the multidisciplinary management of locally advanced and recurrent rectal cancer, as well as other advanced pelvic malignancies has continued to evolve, leading to significant improvements in oncological outcomes [1]. Pelvic exenteration is a recognised treatment and remains the only potential curative option in select cases [2]. This radical surgery involves en bloc resection of the involved pelvic organs, often incorporating adjacent organs, regional lymph nodes, neurovascular structures, soft tissue and bone to achieve a microscopically negative resection margin (R0) [1, 2]. Given the complexity and extent of resection, post‐operative morbidity remains high, with 63% of patients experiencing a post‐operative complication and 90‐day mortality ranging from 0% to 8.6% [3, 4]. The available literature on surgical and survival outcomes predominantly concentrates on locally advanced and recurrent rectal cancer cohorts. However, long term data in this population show considerable heterogeneity with 5‐year disease‐free survival and overall survival ranging from 13%–89% and 12%–67.2% respectively [4].

The PelvEx Collaborative recently published surgical outcomes from 2017 to 2021, demonstrating improved R0 resection to 84.5% and 71.7% for locally advanced rectal cancer (LARC) and locally recurrent rectal cancer (LRRC), respectively [3]. These results may be attributed to the higher rate of bone resection, increasing from 8.2% to 19.6% in LARC and 20% to 41.9% in LRRC with greater need for pelvic reconstruction (32% for both cohorts) [3].

The Royal Brisbane and Women's Hospital has performed pelvic exenteration surgery since 2012 and is currently the state‐wide referral centre for pelvic exenteration in Queensland, Australia. In 2018, we published our series of the first 61 patients during the development of the pelvic exenteration service [5]. Despite reporting medium‐term data (median follow up of 24 months), results were comparable to the data from a recent publication by Solomon et al. in 2023 which evaluated over 1000 patients treated at the Royal Prince Alfred Hospital in Sydney [2]. Our series demonstrated an R0 resection of 91.5%. Whether this translated to improved long‐term survival was unknown [5].

Since inception, we have captured all patients that underwent pelvic exenteration from January 2012 to August 2023 allowing us to report oncological and survival outcomes up to 5 years post‐operatively. The aim of this study is to evaluate surgical, overall, disease‐free, and disease‐specific survival outcomes to contribute and compare to national and PelvEx Collaborative Data. We hope to demonstrate that pelvic exenterations can safely be performed in a lower volume statewide tertiary centre with acceptable morbidity and oncological results.

2. Methods

2.1. Study Design

This was a retrospective cohort study of patients who underwent total or partial pelvic exenteration at the Royal Brisbane and Women's Hospital, Queensland, Australia, between 1st January 2012 and 31st August 2023. Patient characteristics, oncological and surgical outcomes were collected from electronic and written medical records. Survival data was confirmed using linked data from the Australian Registry of Births, Deaths and Marriages. Ethics approval was obtained from the Metro North Hospital and Health Service Human Research Ethics Committee (HREC 102257).

2.2. Selection Criteria

The participants in this study included the medical information of all patients who underwent pelvic exenteration surgery due to advanced primary rectal cancer, other advanced primary malignancy, locally recurrent rectal cancer, other recurrent malignancy, or non‐malignant disease. Pelvic exenteration is defined as resection of the primary or recurrent tumour (e.g., rectum) with more than 50% of two or more organs from two (of five) different compartments and/or the resection of bone and major neurovascular pelvic sidewall structures [2, 6]. Pelvic exenteration was classified as complete (removal of tumour, with or without attached bone, and all remaining pelvic viscera within all anatomical compartments of the pelvis) or partial (removal of tumour, with or without attached bone, and en bloc resection of up to three (of five) anatomical compartments of the pelvis) [2, 6]. Diagnosis of malignancy and pre‐operative staging was based on histopathology and radiological evaluation. Each patient is discussed at the specialist multidisciplinary team meeting to determine patient suitability and peri‐operative optimisation.

2.3. Outcomes

Relevant clinical information collected included classification (e.g., primary rectal, recurrent rectal, other primary or recurrent malignancy, non‐malignant), tumour histology (e.g., adenocarcinoma, squamous cell carcinoma), tumour primary location, and neo‐adjuvant treatment (chemotherapy, radiation therapy, chemoradiotherapy, biologic therapy).

The operative data included organs resected, involvement of the pelvic side wall with resection of vascular structures, bone, nerve, as well as type of urinary conduit and pelvic reconstruction, all documented by the senior operating surgeon on the operation record. Post‐operative pathological assessment was based on the histopathology report completed by a Consultant Pathologist. The surgical margin—R0 was defined as a microscopically negative margin (> 1 mm).

Overall survival was calculated in months from the date of surgery to death or last clinical encounter. Disease recurrence was defined as local or distant disease recurrence confirmed on radiological imaging during follow up. Patients were censored from 30th March 2025.

2.4. Statistical Analysis

Statistical analyses were performed using IBM SPSS version 29. Descriptive data sets were reported as frequency (percentage) for categorical variables and median with interquartile range for continuous variables. Two sample t‐test was used for continuous variables and chi‐square test was used for categorical variables. Overall survival, disease‐free survival, and disease‐specific survival were estimated using the Kaplan–Meier survival method. Statistical analysis was completed in consultation with a qualified Statistician.

3. Results

Over the 11‐year study period (January 2012 to August 2023), in total 139 pelvic exenterations were performed. The indications were advanced primary rectal cancer (n = 79; 56.8%), locally recurrent rectal cancer (n = 31, 22.3%), other primary malignancy (n = 15; 10.8%), other recurrent malignancy (n = 11; 7.9%), and non‐malignant (n = 3; 2.2%). Patient characteristics and neoadjuvant treatment are summarised in Tables 1, 2, 3.

TABLE 1.

Patient demographics and clinical characteristics.

Pelvic exenterations (N = 139)
Variable, n (%)
Age, years a	57 (IQR 46–64.5)
Age, years
< 50	53 (38.1)
> 50	86 (61.9)
Sex
Male	85 (61.2)
Female	54 (38.8)
BMI, kg/m2 a	26.9 (IQR 22.8–31.6)
BMI ≥ 30	37 (26.6)
BMI < 30	102 (73.4)
ASA‐PS Classification
1	2 (1.4)
2	30 (21.6)
3	104 (74.8)
4	3 (2.2)

Abbreviations: ASA‐PS Classification, American Society of Anesthesiologists‐Physical Status Classification; BMI, Body Mass Index.

^a Median (IQR, interquartile range).

TABLE 2.

Patient preoperative pathological characteristics.

Pelvic exenteration (N = 139)
Variable, n (%)
Classification
Primary rectal cancer	79 (56.8)
Recurrent rectal cancer	31 (22.3)
Other primary malignancy	15 (10.8)
Other recurrent malignancy	11 (7.9)
Benign	3 (2.2)
Tumour histology
Adenocarcinoma	122 (89.1)
Squamous cell carcinoma	9 (6.5)
Other
Transitional cell	1 (0.7)
Leiomyosarcoma	2 (1.5)
Angioleiomyoma	1 (0.7)
Myxoid spindle cell	2 (1.5)
Tumour primary location (N = 137)
Rectum	110 (80.3)
Anus	7 (5.1)
Colon	12 (8.8)
Gynaecological	6 (4.4)
Prostate	1 (0.7)
Seminal vesicles	1 (0.7)

TABLE 3.

Preoperative treatment.

Pelvic exenteration (N = 137)
Variable, n (%)
Total neoadjuvant therapy	116 (84.7)
LCCRT	94 (68.6)
Definitive chemoradiotherapy	7 (5.1)
Radiotherapy alone	13 (9.5)
Biologic therapy	4 (2.9)

Abbreviation: LCCRT = long course chemoradiotherapy.

3.1. Surgical Outcomes

Surgical outcomes are presented in Table 4. In this study, complete pelvic exenteration was performed in 117 patients (84.2%), 22 were partial (15.8%). R0 resection was achieved in 90.4% (n = 123) of patients. In locally advanced rectal cancer, the R0 resection was 94.9%, in locally recurrent rectal cancer the R0 resection was 83.9%. The median length of admission was 36.5 (IQR 23–77) days in hospital.

TABLE 4.

Operative details.

Pelvic exenteration (N = 139)
Variable, n (%)
Pelvic exenteration
Total			117 (84.2)
Partial			22 (15.8)
Organs resected
Rectum			125 (89.9)
Anus			108 (77.7)
Colon			29 (20.1)
Small bowel			17 (12.2)
Vagina (partial, total)			24 (17.3),12 (8.6)
Uterus			34 (24.4)
Ovaries, Fallopian tubes (unilateral, bilateral)			18 (12.9), 21 (15.1)
Bladder Complete (partial)			120 (86.3)
Ureters			12 (8.6)
Prostate			76 (54.7)
Structures resected
Medial vascular surface of internal iliac artery and vein (left, right)			35 (25.2), 43 (30.9)
Internal iliac artery and vein (left, right)			40 (28.8), 35 (25.2)
Lateral pelvic sidewall (left, right)			32 (23), 24 (17.3)
Sacrum			68 (48.9)
Major Nerve (e.g., sciatic or sacral nerve root)			33 (23.7)
Urinary conduit (N = 120)
Ileal			41 (34.2)
Colonic			79 (65.8)
Pelvic reconstruction (N = 45)
Vertical rectus abdominis myocutaneous flap			29 (64.4)
Inferior gluteal artery perforator fasciocutaneous flap			5 (11.1)
Inferior gluteal artery myocutaneous flap			6 (13.3)
Gluteal rotation flap			2 (4.4)
Graft prosthesis			3 (6.6)
Operative time (minutes) a			717.5 (IQR 607.5–829)
Estimated blood loss (millilitres) a			3000 (IQR 1837.5–4500)

Resection	LARC (N = 79)	LRRC (N = 31)	Overall (N = 136)
R0	75 (94.9)	26 (83.9)	123 (90.4)
R1	2 (2.5)	5 (16.1)	11 (8.0)
R2	2 (2.5)	0	2 (1.5)
Pathological tumour response
Complete			12 (8.8)

^a Median (IQR, interquartile range).

Post‐operative post‐treatment pathological characteristics are summarised in Table 4. Complete pathological tumour response was seen in 12 patients (8.8%).

3.2. Postoperative Morbidity

Postoperative complications within 90 days of surgery are detailed in Table 5. The most frequent complications were prolonged ileus (30.2%, n = 42), venous thromboembolism (20.8%, n = 29), and intra‐abdominal or pelvic collection (25.2%, n = 36). Complications classified using Clavien‐Dindo grades III and above occurred in 43.2% of patients (n = 60); return to theatre was required in 33.1% of the cohort (n = 46); and the mortality rate was 2.2% (n = 3). Among patients who required urinary diversion (n = 120), 36.7% developed urosepsis (n = 44), 10 patients had a urine leak (8.3%), and 9 patients required revision due to conduit stoma complications (7.5%).

TABLE 5.

Postoperative complications within 90 days of surgery.

Complication (Total N = 139) Variable, n (%)
Clavien‐Dindo classification
I	5 (3.6)
II	51 (36.7)
IIIa	11 (7.9)
IIIb	36 (25.6)
IV	10 (7.2)
V	3 (2.2)
Specific complications
Midline wound infection	6 (4.3)
Perineal wound infection	23 (16.5)
Venous thromboembolism	29 (20.8)
Haemorrhage/blood transfusion	9 (6.5)
Intra‐abdominal collection	36 (25.2)
Managed with percutaneous drain (N = 35)	16 (45)
Sepsis	15 (10.8)
Anastomotic leak	1 (0.7)
Ileus	42 (30.2)
Small bowel obstruction	2 (1.4)
Small bowel perforation	1 (0.7)
Enterocutaneous fistula	3 (2.2)
Return to theatre	46 (33.1)
Stoma‐related complication (N = 131)
Colostomy revision	1 (0.8)
Urinary conduit complications (N = 120)
Urosepsis	44 (36.7)
Urine leak	10 (8.3)
Ureteric obstruction	16 (13.3)
Urolithiasis	6 (5)
Conduit revision	9 (7.5)
Myocutaneous flap complications (N = 42)
Total flap failure	1 (2.4)
Partial flap failure	4 (9.5)
Wound infection	6 (14.3)

3.3. Survival Outcomes

The survival outcomes are presented in Table 6. The overall survival (using Kaplan–Meier analysis) at 1, 2 and 5 years was 89.9%, 82.0%, and 62.9%, respectively (Figure 1). The overall 5‐year survival for advanced primary rectal cancer was 68.9%, compared to 52.6% for recurrent rectal cancer (Figure 2). Disease‐free survival at 1, 2 and 5 years was 76.1%, 67.0%, and 59.2%, respectively (Figure 3). Disease‐specific survival at 1, 2 and 5 years was 92.7%, 83.8% and 66.8%, respectively. The median follow‐up was 39 months (IQR 14–60).

TABLE 6.

Survival outcomes.

Time (months)	Overall survival	Survival LARC	Survival LRRC	Disease‐free survival	Disease‐specific survival
12	89.9	93.7	87.5	76.1	92.7
24	82.0	88.6	75.0	67.0	83.8
60	62.9	68.9	52.6	59.2	66.8
Median follow‐up (months) a			39 (IQR 14–60)

Note: Overall survival includes all classifications listed in Table 2.

Abbreviations: LARC, locally advanced rectal cancer; LRRC, locally recurrent rectal cancer.

^a Median (IQR, interquartile range).

FIGURE 1.

Overall survival.

FIGURE 2.

Survival by classification.

FIGURE 3.

Disease free survival.

4. Discussion

The role of pelvic exenteration has progressed over the last three decades with advances in perioperative staging through improved medical imaging technology, neoadjuvant treatment and surgical techniques, as well as the introduction of the multidisciplinary team to guide patient selection and management. A more radical surgical approach has been adopted to achieve a microscopically negative margin, intended to translate to improved survival outcomes [7].

This study evaluated the oncological and surgical outcomes of patients who underwent pelvic exenteration at a lower volume, single statewide tertiary centre. The results demonstrated excellent R0 resection rates with survival and post‐operative morbidity and mortality outcomes comparable to published reports and larger volume centres [2, 3, 4].

Locally advanced primary and recurrent rectal cancer comprise 82% (N = 114) of our cohort. Overall, R0 resection was achieved in 90.5% (n = 124) of patients. In locally advanced rectal cancer, R0 resection was 93.8%; in locally recurrent, the R0 resection was 82.4%. In comparison to national data, the Royal Prince Alfred (RPA) Hospital in Sydney recently published surgical, oncological, and quality of life data in over 1000 pelvic exenteration patients [2]. In locally advanced primary and recurrent rectal cancer, this group achieved an R0 resection of 89.2% and 74.1%, respectively [2].

In 2024, PelvEx Collaborative published updated results on surgical outcomes from 45 units between 2017 and 2021 [3]. The R0 resection in locally advanced rectal cancer and recurrent rectal cancer was 84.8% and 71.7% respectively [3].

We have consistently maintained our R0 resection rate since establishment of the service, from the first 60 patients to our current cohort. This has been aided by the establishment of our pelvic exenteration multi‐disciplinary team (MDT) meeting which involves both clinical and pathological review, with careful radiological and surgical planning. From inception, we have supported a radical approach as demonstrated in our high sacral and major nerve resection rates of 48.9% and 23.7%, respectively, as well as pelvic reconstruction (32.4%). The more recent PelvEx data attributes the improvement in R0 resection to increased number of bony and nerve resection, and consequent need for pelvic reconstruction [3]. Over the study period the preference for urinary diversion transitioned towards colonic conduits, in contrast to most Australian centres. This technique avoids a small bowel anastomosis and associated complications. In addition, in patients with central adiposity, left colonic conduits are less likely to be under tension when compared to the small bowel mesentery of an ileal conduit.

The overall survival of our cohort at 1, 2 and 5 years was 89.9%, 82.0%, 62.9%, respectively. Between locally advanced and recurrent rectal cancer, the outcomes were more favourable for primary disease with a 5‐year survival of 68.9%, in comparison to 52.6%. Disease‐free survival at 1, 2 and 5 years was 76.1%, 67.0% and 59.2%, respectively. Disease‐specific survival at 1, 2 and 5 years was 92.7%, 83.8% and 66.8%, respectively. At RPA, the 5‐year overall survival rates were 66.3% in advanced primary rectal cancer and 44.6% in locally recurrent rectal cancer [2].

There are several pathological factors that are thought to influence survival and recurrence, with most studies supporting an association between incomplete resection margin and reduced overall survival [7]. This is demonstrated in our higher R0 resection rate and corresponding 5‐year survival outcomes for locally advanced and recurrent rectal cancer, compared to national data.

The median length of stay (LOS) for our patients was 36.5 days (IQR 23–77), in comparison to the RPA Hospital; median length of hospital stay was 19 (IQR 14–27.8) days for the primary rectal group and 22 days (IQR 15–34) for recurrent [2]. PelvEx data reported a median LOS of 17 days (IQR 11–26) [3]. Our institution services a vast geographical area with a limited number of specialised Colorectal units across the state. The majority (56.7%) of patients are from a major city, 26.0% are from inner regional areas, with the remaining (17.3%) from outer regional areas [8]. These considerations influence our post‐operative management and length of hospital stay, as interhospital transfer back to the referring hospital is often not appropriate. This is reflected in the discharge destination of our patients; in total, 64.7% (N = 90) returned home, 30.9% (N = 43) were transferred back to their local hospital, and of those, 60.5% were transferring for rehabilitation. We are reviewing the current service delivery model to identify potential factors contributing to discharge delays. This has included improving communication and education to regional and rehabilitation centres.

Despite the survival outcomes, early post‐operative morbidity remains high, impacting recovery and length of stay for these patients. Our 30‐day and 90‐day mortality was 2.2% (n = 3) which was marginally higher than RPA at 1.4% [2]. PelvEx data reported a 90‐day mortality rate of 2.6% [3].

The most frequent complications observed were prolonged ileus (30.2%), intra‐abdominal infection (25.2%) and venous thromboembolism (20.8%). At our institution, an ongoing effort has been made to implement post‐operative screening and prevention strategies directed towards the most frequent complications seen. For example, to reduce the risk of venous thromboembolism, full length sequential compression devices are used intra‐operatively and in the early post‐operative period. A screening doppler ultrasound scan of bilateral lower limbs is performed at day 5 to assist early detection, in addition to standard chemical prophylaxis and early mobilisation. In addition, intravenous broad‐spectrum antibiotics are continued for 72 h post‐operatively given the higher risk of wound and intraabdominal infections.

The strengths of this study include the long‐term follow up and inclusion of disease‐free and disease‐specific survival outcomes providing a more accurate reflection of treatment effectiveness.

A limitation of this study is the long‐term retrospective data collection and risk of incomplete or missing data particularly prior to electronic storage of medical records in 2015. There is also a risk of selection bias as complete data is not available for the number of patients that did not proceed to exenteration surgery relative to the number referred to MDT. Sub‐group comparison was performed for resection rate and survival outcomes; this was not extended to other data points given the lower number of patients in our cohort. A relatively low proportion of cases were of gynaecological origin (4.4%); this is likely due to partial exenterations being performed independently by gynae oncologists. Our cohort represents patients referred to the statewide service and unfortunately would not capture these patients.

The Royal Brisbane and Women's Hospital has been the referral centre for pelvic exenterations since its establishment in 2012. Collaboration and support from our colleagues at the Royal Prince Alfred Hospital have been crucial for the development and growth of our service. The results presented will contribute to national and international data to further assist patient selection, treatment guidelines, and facilitate benchmarking across centres nationally and internationally.

The authors plan to submit further publications detailing the urological outcomes, long‐term surgical complications, and quality of life measures from this cohort's data set.

The review of eligible retrospective data for this study identified the value of prospective data collection; consequently, we have implemented prospective data collection of patients while maintaining the breadth of the dataset. This will facilitate ongoing review of clinical data for the purpose of quality improvement and safety, as well as reflect longitudinal changes in management and outcomes over a longer period of time.

5. Conclusion

The RBWH pelvic exenteration oncological and surgical outcomes are comparable to national data and PelvEx Collaborative international group data demonstrating that pelvic exenteration surgery can safely be performed in a lower volume statewide referral centre. The findings from this study support the expansion of multi‐disciplinary services available for patients in Queensland to improve our model of care and patient quality of life.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.