Robotic or laparoscopic pelvic exenteration for gynecological malignancies: feasible options to open surgery

Pelvic exenteration, either curative or palliative, means not only the partial or complete evisceration of the pelvic cavity but also, most often, the creation of stomas [1, 2]. Considered a mutilating procedure, it is strongly associated with high morbidity and psychosocial impacts on patients [3, 4]. However, in selected cases, it has emerged as the only possibility to increase life expectancy, improve the chances of survival, and relieve symptoms [5, 6]. In the field of gynecology, pelvic exenteration has evolved from a salvage procedure to radical resection [1, 7, 8], mainly for the treatment of cervical cancer (CC) [7, 8], either locally advanced stage IVa or centrally recurrent. However, patients with other relapsed gynecological cancers may benefit from this extensive procedure [9].

In an effort to lower the associated surgical risk and the psychoemotional impact, minimally invasive pelvic exenteration—either laparoscopic or robotic—has been established as a feasible option [1, 10, 11], possibly with a higher rate of acceptance. Patients benefit from the same survival chances, lower morbidity and mortality rates, shorter hospital stays, and faster recovery [12, 13, 14]. Similar oncological results are expected [15].

In centers where robotic surgery is an alternative and the learning curve has been overcome, robot-assisted procedures are preferred over laparoscopies, which are more challenging from a technical standpoint [10, 15, 16]. However, laparoscopy has a higher implementation rate owing to its lower cost and wider availability.

The main focus of this paper was to acknowledge the feasibility of a minimally invasive surgery (MIS) approach for pelvic exenterations [17] and prove that it is a useful tool for treating patients with primary locally advanced or recurrent gynecologic malignancies [18, 19].

Despite the lack of publications and data for interpretation, this study also aimed to analyze overall survival (OS) and possibly identify parameters that could negatively impact patient prognosis.

This study was designed as a retrospective observational analysis of a single team’s experience from Jan 2008 to Jan 2022. The study group included 12 consecutive patients who presented with locally advanced CC stage IVa or pelvic recurrences of both cervical and other gynecologic malignancies, such as uterine or vulvar cancer. There were no selection criteria for the cases; all patients were informed of either option (open or minimally invasive) and accepted the latter unconstrained.

All patients provided written informed consent prior to surgical intervention, and the study was approved by the Local Review Board for retrospective use. Data were collected from the department’s electronic database, available at the Fundeni Clinical Institute.

Age, associated pathologies, diagnosis (including stage on admission), and treatment plan were recorded. Imaging tools, such as computed tomography (CT) and magnetic resonance imaging (MRI), as well as histopathological results following biopsy, were used to diagnose patients and establish surgical indications.

The performed exenterations were classified as anterior (AP) or total (TP) and were either robotic or laparoscopic. In contrast to laparoscopy [15], the robotic approach offers better visualization and easier manipulation of instruments and involves placing the cart between the patient’s legs, thereby improving access to the pelvic structures [20]. Surgery began with a sharp dissection lymphadenectomy on the right side [21]. The right ureter was then divided towards the bladder. The same steps were performed on the left side after rectosigmoid mobilization [15]. In cases where TP pelvic exenteration was performed, an Endo GIA stapler was used to transect the colon. Otherwise, surgery was continued with a type V radical hysterectomy (according to the Piver-Rutledge-Smith classification, 1974) [22], including bilateral salpingo-oophorectomy with concomitant resection of the bladder, which was later removed through the vaginal opening [23].

If TP exenteration was performed, dissection was resumed downwards to the level of the endopelvic fascia with en bloc removal of the supralevator tissues [24]. Specimens were retrieved through a perineal incision. An omental patch was placed to ensure that the patient did not develop empty pelvic syndrome. Perineal reconstruction can be performed using either myocutaneous flaps or mesh techniques [25, 26]; our team prefers alloplastic grafts.

The urinary diversions were all incontinent, either reconstructive (Bricker ileal conduit) or cutaneous ostomies [27]. The ureters were placed underneath the mesentery to prevent adhesions.

Operating time was defined as the time from skin incision to the end of the operating procedure, and hospital stay was calculated from the day of surgery until discharge. Blood loss was estimated using a suction bottle. Postoperative complications were classified according to their time of occurrence and scaled using the Clavien-Dindo (CD) score [28]. Early complications were defined as having developed during the same admission as surgery and were subdivided into minor-CD 1 and 2 and major-CD scores of 3 and 4. Late complications were classified.

Complications can be psychological (e.g., depressive disorder syndrome) or ostomy-related (colostomy or ureterostomy). Furthermore, a major problem is the perineal incision, which can become either infected or dehiscent. In addition, chemotherapy (CHT) that induces electrolyte imbalances and vascular injuries can cause postoperative issues.

Histological findings, resection margins (R0-for microscopically negative), and the presence of residual tumor tissue were noted in the Pathology Department. Follow-ups were possible in compliant patients. A cut-off of 24 months was set to analyze the data. OS was defined as the interval from the day of surgery to the date of death or to the end of the set cutoff date if the patient was still alive. Disease-free survival (DFS) was defined as the length of time after pelvic exenteration until relapse or the development of metastases.

Statistical analyses were performed using IBM SPSS Statistics for Windows (version 23.0; IBM Corp., Armonk, NY, USA). Categorical variables were noted as percentages, and continuous variables as mean ± standard deviation or median and range. Survival rates were represented with the help of Kaplan-Meier curves using the log-rank test to compare the groups. Statistical significance was set at p-value <0.05.

Between Jan 2008 and Jan 2019, 1,273 pelvic exenterations were performed at the Fundeni Clinical Institute, of which 983 were for gynecological malignancies. Nevertheless, only 12 patients with locally advanced stage IVa CC or recurrent cancers underwent minimally invasive pelvic exenterations because only a specialized team that has already overcome the learning curve can perform such pluri-visceral resection (Fig. 1).

Fig. 1
Patients undergoing pelvic exenteration (Jan 2008–Jan 2022).
MIS, minimally invasive surgery.

• Click for larger image
• Download as PowerPoint slide

Their average age was 58.18 years (range: 40–73). Four patients presented with associated cardiovascular disease, three with anemia, one with type II diabetes mellitus, and one with Basedow-Graves disease.

Five patients had locally advanced CC-stage IVa-all with bladder invasion. They were included in a treatment scheme consisting of concurrent chemoradiotherapy (CRT) or radiotherapy (RT) alone (according to the oncological plan available at the time) and were referred for surgery due to cancer persistence, according to imaging or biopsy results, or both.

Seven patients experienced recurrence after vulvar cancer (one), endometrial cancer (one), and CC (five). They all received prior treatment, including surgery and RT ± CHT, and were classified as centro-pelvic recurrences involving the bladder with/without the rectum. According to the histopathological report from the initial surgery, only two specimens were free of residual lesions. Recurrences occurred after a median of 15 months (6–32), and were included in the treatment scheme established by a multidisciplinary team.

Six anterior and total pelvic exenterations each each were performed. Nine of these procedures were performed robotically (6 AP and 3 TP), and 3 patients were treated laparoscopically. The urological reconstructions consisted of 8 double-barrel ureterostomies (6 robotic) and 4 Bricker urinary diversions (3 robotic).

All TP exenterations involved terminal colostomies. No intraoperative complications occurred during the study period. After a brief laparoscopic exploration, one patient who was not included in the series underwent open surgery because of severe tissue adhesions.

The mean operative time was 360±30.7 minutes for AP exenterations and 440±40.7 minutes for TP exenterations. The mean blood loss was 350±35 mL, and no intraoperative transfusions were needed.

The mean hospital stay was 18 days (6–38).

Nine patients (75%) underwent R0 resections with clear margins, 10 specimens showed residual tumor tissue, and 16.7% had no evidence of disease.

Early complications included one major event: a patient with a urinoma requiring percutaneous drainage (CD 3a) and one minor event: a patient with a perineal wound infection (CD 2). The perioperative morbidity was calculated as 16.6%. No deaths related to the surgical procedures were recorded.

Minor late complications included urinary diversion-related pathology in five patients with acute pyelonephritis, three patients with hydronephrosis (two patients presented with both complications), and thrombophlebitis in two patients. Three patients developed major late complications comprising of the following conditions: consequent CHT administration inducing acute renal failure (CD4a), development of carcinosis with consecutive bowel obstruction that required surgical intervention (CD 3b), and an iliac artery fistula (CD 3a) resolved by stent placement.

The DFS was, on average, 12 months, ranging from 10 to 14 months, and the OS was 20 months (1–127).

1. Follow-up and prediction factors

Regarding OS, 50% of the patients were still alive at the cutoff point. A statistically significant p-value was not obtained because of the small number of cases. However, some trends were easily observed; therefore, some prognostic factors could be identified.

Age was not a significant parameter that influenced patient survival. However, females within the age interval of 50–69 years were at a higher risk of developing adverse events, whereas only 16.6% of females under 50 or above 70 years of age did not reach the cut-off (p=0.634).

Patients with recurrent malignancies had worse outcomes, as 57.1% failed to reach the 24-month cut-off. In contrast, patients with stage IVa CC had higher OSs (p=0.537);—3 out of 6 patients who lived past the cut-off were part of this group (Fig. 2).

Fig. 2
Survival in months for patients undergoing minimally invasive pelvic exenteration: 0 is the stage IVa CC category and 1 is the relapsed malignancies category (p=0.537).

• Click for larger image
• Download as PowerPoint slide

Regarding the type of surgery performed, patients who underwent TP had a worse prognosis than those who underwent AP. TP exenteration cases had lower OSs, with 66.7% of patients not surviving beyond the cutoff (p=0.111).

Minor (early and late) complications not only impact survival rates but also affect the quality of life, involving redundant hospital admissions. Four of six patients in this subgroup had shorter survival times than the 24-month cut-off (p=0.878).

Major complications were also associated with negative outcomes (p=0.759). Patients who required ICU admission or surgical re-intervention had the shortest survival times, notably lower than the median OSs (1 and 4 months, respectively, versus 20 months).

Considering the 24-month follow-up, 4 out of 6 patients who failed to reach this cut-off were diagnosed with distant metastases or local recurrence (p=0.169). In contrast, patients with R0 disease had improved outcomes (p=0.735), while patients with R0 disease who did not develop relapse had the longest survival of 127 months and 108 months, respectively. In fact, the first patient in the series, having been carefully selected, had the longest survival rate.

Laparoscopic pelvic exenteration was first described in 2003 by Pomel et al. [29], followed by robot-assisted procedures and full robotic interventions, which were systematically performed to prove the benefits of the technique [14]. In light of the 2018 LACC study, MIS was associated with worse oncological outcomes and faced an important backlash; however, according to the latest publications, things appear to have more nuances. In contrast, robotic surgery is suggested to have the same oncological benefits, whereas the 90-day surgery-related morbidity is not influenced [30, 31]. As indications for pelvic exenteration have expanded, surgery has been established as safe and effective (with a 40% 5-year survival rate) [1, 16, 32]. Conventionally, surgery plays an important role [32], but according to the latest treatment guidelines (NCCN, version 1. 2021), patients with locally advanced stage IVa disease should be treated with CRT and brachytherapy [33]. However, high relapse rates after curative-intent CRT have been reported in several studies. Residual tumor tissue found in histopathological specimens after definitive CRT could be associated with surgical intervention efficacy in achieving better local control [34, 35, 36]. Furthermore, other relapsed gynecological malignancies have been treated in this manner, making MIS pelvic exenteration a useful tool [7, 21].

In terms of difficulties encountered, urinary reconstruction is one explanation for why laparoscopy is more challenging [37], making robotic surgery the preferred approach. 3D stereoscopic vision offers greater visualization, and articulated wristed robotic instruments, along with the harmonic scalpel, reduce tremors and can achieve better control of the pelvic sidewall vessels [1, 10, 15, 38]. However, robotic interventions are still expensive procedures [1]. The operating time is influenced by the setup and docking steps, favoring laparoscopy, but Bizzarri et al. [12] reported quite the opposite, describing a significantly lower operating time for the robotic approach. The success of this surgical intervention is highly dependent on the operating team, the number of patients treated, and the learning curve [10]. Several studies have shown that minimally invasive pelvic exenteration is a feasible procedure [1, 12, 14].

By the end of 2018, 163 minimally invasive pelvic exenterations were reported, 155 of which were for gynecologic malignancies. Bizzarri et al. [12] presented a 23 case series along with a literature review proposing that minimally invasive techniques give pelvic exenteration an upgrade as long as a prior patient selection is performed. In 2019, an editorial published by a Chinese team described their experience of 23 robot-assisted laparoscopic cases, compared the results with the literature, and reinforced the idea that such an alternative is possible [10]. Under these circumstances, our experience with 12 cases cannot be undermined compared to the number of TP surgical interventions performed worldwide.

In 2020, an American team examined short-term outcomes in a minimally invasive group and analyzed the cost-effectiveness of the technique, length of hospital stay, and complication rates. They concluded that patients in this category had lower discharge costs, shorter hospital stays, and developed major postoperative issues less frequently than those who underwent open surgery approach [14].

Later, an article by Cianci et al. [1] focusing on robotic procedures only described 53 such cases in a systematic review and stated that the operative time is reduced owing to the technical enhancements that robotic surgery offers. A decrease in blood loss and hospital stay was also noted in comparison to the open approach, but further investigation would have to be conducted in order to assess patient late outcomes.

The late postoperative complications were mainly associated with urinary diversion [10]. Continent reconstruction is reserved only for young patients and is associated with the development of a fistula and consequent septic shock. Largely used techniques, such as the Bricker and Wallace incontinent diversion, are preferred to reduce the frequency of postoperative issues.

Regarding oncological outcomes, the disease-free survival (DFS) is expected to be approximately 11 months. Several studies have reported OS rates at 2 and 5 years of 40.7% and 27.0%, respectively [12, 14].

According to the literature, the cervix is the most common primary site (85.8%) [12] of gynecological malignancies addressed by MIS pelvic exenteration [1, 7]. Our experience with 12 consecutive cases revealed that 83.3% of patients were treated for CC, half of whom had stage Iva disease. These patients were all treated with neoadjuvant therapy; however, several studies have shown that concurrent CRT or RT alone is not always sufficient to achieve a disease-free status. A closer look at the statistics noted in other publications shows that more than 30% [12, 39] of patients present with recurrence after curative-intent CRT or RT. In contrast, our classification describes relapses in patients who have already undergone surgery.

Nevertheless, the choice of MIS resides in the fact that patients are given a better chance to cope with such a diagnosis and the mutilating procedure [40, 41].

Although the share of laparoscopy in the literature is higher (82.8%) [12] than that in our series (25%), the robotic approach was preferred for multiple reasons - easier pelvic dissection and swift overcoming the learning curve.

As reports show, almost 80% [12] of the reviewed minimally invasive exenterations were AP exenterations. In our patients, the extension of the disease to the posterior structures was generally associated with tumor relapse rather than primary cancer; therefore, only 50% of the resections were AP.

Regarding urinary diversions, the ileal conduit (Bricker/Wallace technique) and ureterosigmoidostomies are the main reconstructive approaches (81.6%) [12], as described in the literature, whereas only 33.3% of our cases benefitted from such a conduit. This is one explanation for why we reported many minor late complications (54.5% of our patients developed hydronephrosis and acute pyelonephritis) [41]. However, such techniques require a longer operative time and are considered additional risk factors for patients with associated cardiovascular diseases.

No significant difference compared with other publications was noted concerning operating time and estimated blood loss [12, 42]. Hospital stays varied greatly among the cited publications. A closer look at the time range revealed that all patients largely fall within an interval of 5 to 37 days, with a median of 10–11 days [12, 39]. Our patients registered a slightly higher median of 18 days, and the first two patients were discharged later for close monitoring or assistance in caring for the perineal wound.

Early morbidity, including both minor and major complications, was not associated with pelvic exenteration. However, major events also have an impact on OS. Major early complications accounted for 8.7% of postoperative issues in previous studies [12], whereas our patients scored 8.3%. The overall post-operative mortality rate was 0.6%, and no death [12].

An R0 resection has been reported achievable in 92.9% of patients [12], while in our study, it was found to be 75%. Considering the heterogeneity of cases, it is not surprising that patients with primary advanced or relapsed CC, but with no residual tissue on histological specimens, have the best survival rates.

Additionally, although different histological types of cancers potentially have different survival rates, “advanced gynecologic pelvic disease” has very similar outcomes, and a standardized surgical procedure was performed in all cases. Therefore, the oncological results may be interpreted together. DFS (11.8 months) and 2-year OS (50%) in our patients were consistent with literature citations [11, 12, 33]; however, data are generally scarce, and the number of patients enrolled in the studies is small. Extensive follow-up is necessary to further assess relapse, metastasis development, and long-term patient outcomes in the long run [1, 12, 14].

Minimally invasive pelvic exenteration is safe and feasible, providing patient selection is appropriate. Primary advanced-stage IVa CC or relapsed gynecological malignancies benefit from the same oncological results as open surgery but with lower morbidity. The choice between laparoscopy and the robotic approach depends on the hospital setting and the surgeons’ learning curve and habits; however, regarding urinary reconstruction, an ileal conduit is preferable. R0 resection is mandatory, and careful follow-up is required, especially for patients who fall into the risk categories of relapsed cancer, TP exenteration, and major complications.