Total parietal peritonectomy for 61 patients: a retrospective study

Objective: To evaluate the morbidity of total parietal peritonectomy (TPP) during cytoreduction surgery, and its impact on the site of recurrence of different peritoneal surface malignancies (PSM). Methods: We led a retrospective study in a French tertiary cancer institution (Centre Oscar Lambret Lille) experienced in treating PSM over a 6-year period from 2012 to 2018. All patients underwent a total parietal peritonectomy during a debulking surgery for PSM including ovarian cancer, appendiceal pseudomyxoma peritonei or peritoneal mesothelioma. Results: Among the 61 patients included in this study, 49 patients(80.3%) had an ovarian cancer. The rate of complete tumor resection reached 86.9% with almost 69% of surgeries being highly complex. 73.8% were transfused during the surgical procedure. The median length of hospital stay was 10 days including 7 days in Intensive Care Unit. Overall, 19 patients (31.1%) had an early postoperative complication, including 3 with a grade IIIB complication of Clavien Dindo classification. With a median follow-up of 30 months, the estimated disease-free survival in the ovarian cancer subgroup who had an initial peritonectomy (n = 42) was 84.7% at 1 year and 12.0% at 3 year. The main site of first and second recurrence was peritoneal (42% and 14%). Conclusion: TPP is a safe surgical procedure to treat peritoneal surface malignancies and their recurrences with a low rate of grade IIIB morbidity and no treatment-related death and allow optimal surgery. In this study there is no atypical recurrence site, such as abdominal muscle involvement.


Introduction
Peritonectomy has been used for the treatment of peritoneal surface malignancies (PSM) including colorectal cancer [1,2], appendiceal pseudomyxoma peritonei [3], and peritoneal mesothelioma [4]. Baratti et al. [5] was the first to evaluate the concept of total parietal peritonectomy (TPP) in peritoneal mesothelioma. Pelvic peritonectomy has been applied in advanced ovarian cancer with peritoneal carcinomatosis and has increased the rate of complete cytoreduction up to 60% [6].
As a technique, peritonectomy has been described and standardized first by Sugarbaker in 1995 [7] for colon cancer then modified by other groups depending on histopathology [8]. According to Sugarbaker [9], only abnormal peritoneum should be excised. But the extensive resection of tumors showed contradictory outcomes. Kyriazanos et al. [10], described the "cocoon technique" a standardized approach to achieve total extraperitoneal peritonectomy. Usually coupled with Hyperthermic Intraperitoneal Chemoperfusion (HIPEC), extensive cytoreduction surgery (CRS) has been criticized for its morbidity and mortality rates along with its benefits for all patients [11]. On the other hand, other studies have shown significant improvements in rates of survival and acceptable rates of morbidity [12,13].
In order to obtain an optimal cytoreduction, surgeons perform a TTP to resect all visible disease, since any residual disease represents a potential risk for recurrence, especially in chemo-resistant tumors.
However, removing only visible residual disease following neoadjuvant chemotherapy (NACT) and not "normally appearing peritoneum" that was previously affected before chemotherapy, showed a high rate of recurrence of 70% with the commonest site being the peritoneum, at previous disease sites [14]. Therefore, it may be hypothesized that complete removal of the peritoneum affected prior to NACT may prevent or delay the recurrence in these patients.
Another hypothesis of the TPP benefits is the removal of a potential site of disease recurrence.
We performed this observational study to evaluate primarily the morbidity and mortality related to TPP performed during CRS. We also evaluated the impact of this surgery on the site of recurrence of different PSM, essentially the ovarian cancer.

Study design
We conducted a retrospective and descriptive cohort study in a French tertiary cancer institution (Centre Oscar Lambret -Lille) over a 6-year period from 2012 to 2018.
All patients included in the present study underwent a TTP during a debulking surgery for PSM and treatments were validated by the tumor-board committee. Data were collected from hospital patient files without direct interaction with the patients for the research purpose. The study protocol was approved by the institutional review board and conformed to the French ethical standards and 2008 Helsinki declaration.

Surgical procedures
The goal of CRS was to obtain a complete cytoreduction (no macroscopic residual disease). The disease extent was quantified using Sugarbaker's Peritoneal Cancer Index (PCI) [7]. Completeness of cytoreduction score (CC-score) was used to describe the residual disease status.
TPP was defined as removal of the complete peritoneum (pelvic, bilateral anteroparietal peritoneum, bilateral upper quadrant peritoneum) using electric cautery or blunt dissection, as described in details by Yokosu et al. [16]. A TPP was performed in three different cases. First, to resect all sites of peritoneal disease in primary debulking surgery (PDS) prior to administration of a chemother-apy. Second, to resect all sites of residual disease in interval debulking surgery (IDS). And finally, to remove potential sites of disease recurrence.
Small and scattered localizations on the visceral surface were resected by local excision and/or electrocoagulation. In case of massive and/or deeply infiltrating disease, visceral resections were performed, including cholecystectomy, splenectomy, sigmoid, right or total colectomy. Clinically suspicious lymph nodes (para-aortic and pelvic) were sampled and submitted to pathologic examination.

Systemic chemotherapy
All patients with ovarian cancer received at least 6 cycles of a platinum compound (Carboplatin or Cisplatin) and Taxane (Paclitaxel or Docetaxel) before and/or after surgery depending on the resectability of the tumor.

Follow up and definition of recurrence
The occurrence of a recurrence was assessed by a clinical examination completed by biological (increase in the tumor marker CA-125), radiological examinations (thoraco-abdominopelvic CT scan, PET-scan) and/or histological evidence if possible.
The date of diagnosis of the first recurrence was collected, as well as the site of recurrence (Liver, Pleural, peritoneal carcinomatosis…) and the date of the latest news or death if any.

Data collection
Consecutive patients with PSM including carcinomatosis of gastrointestinal (colon cancer) or gynecologic origin, mostly advanced ovarian cancer, fallopian tube and primary peritoneal tumors such as mesothelioma (DMPM) or pseudomyxoma peritonei, undergoing PDS or IDS with or without intraperitoneal chemotherapy (IPC) were included. In the ovarian subgroup, we included patients with FIGO stage I-IV-B. Patients with stage IV-A (pleural effusion and/or lung metastasis) were operated only if they showed complete response to systemic chemotherapy and the cytology became negative. Patients with stage IV-B who had resectable liver metastasis, inguinal lymph node metastasis and spleen metastasis lesions were also included. We only collected data for patients whose records included complete information.
The following clinical and paraclinical items were collected: age at diagnosis, body mass index (BMI), WHO performance status [17], preoperative CA 125 level. The tumor characteristics were also detailed: International Federation of Gynaecology and Obstetrics (FIGO) [18] stage for ovarian cancer, histological type [15] (serous, mucinous, endometrioid, clear-cell, papillary, mixed, non-epithelial adenocarcinoma, pseudomyxoma, mesothelioma and serous borderline tumors), peritoneal disease spread according to the PCI [7] at time of diagnosis, CRS results according to the CCR score [7], presence or absence of micro-or macro-scopic peritoneal lesions as long as the presence or absence of identified genetic mutations.
Surgical characteristics at time of peritonectomy included the Aletti Score [19] to classify the difficulty of the procedure in High, Moderate and Low risk. The severity of postoperative complications were described using the Clavien-Dindo classification [20] based on the therapy used to correct them.

Statistical analysis
Analysis of all the included patients was performed. Results were presented using descriptive statistics: numbers and proportions for categorical variables, and median and ranges, for continuous variables.
Time-to-event analyses were calculated from the date of diagnostic in patients with peritonectomy performed as part of initial surgery and not for a recurrence. Overall survival (OS) was defined as the time interval from the date of initial diagnosis to the date of death from any case. Patient alive were censored at the date of last news. Diseasefree survival (DFS) was defined as the time interval from the date of initial diagnosis to the date of the first tumor recurrence or death due to any cause. Patient alive without recurrence were censored at the date of the last news. Data for survival curves were calculated using the Kaplan Meier method. OS and DFS were assessed according to the histopathology but also to the context of the TPP (when undergoing a debulking surgery at initial diagnosis or to treat a recurrence). Data were analyzed using the STATA v15.0 software (Timberlake Consultants Limited, Richmond upon Thames, UK).

Results
From 2004 to 2018, 61 patients underwent a TPP at our center. Patient characteristics are summarized in Table 1. The median age was 62 years (29-80) and 82% had a good performance status. Eighty percent of the patients had ovarian cancer, mainly High Grade Serous Ovarian Cancer (HGSOC), 13% pseudomyxoma, and 2% mesothelioma. Fifty three patients (87%) had the peritonectomy at initial diagnosis and 8 patients (13%) had a recurrent disease with CCR-0 resection obtained in 87% of all patients.
Forty eight patients (84.2%) had a peritonectomy for macroscopic lesions and histology confirmed the disease. Only two patients had a preventive peritonectomy, where no macroscopic lesions were visualized and pathology report did not show any lesions. On the contrary, 8 patients had a complete peritonectomy for "supposedly" macroscopic lesions and histology report came back negative.
The characteristics of the surgery are available in Table 2. The complexity of the surgical procedures was assessed using the Aletti score: 42 patients (68.9%) were classified as "high" in terms of surgical difficulties, often associated with a more efficient debulking. The median operating time was 310 min with 82% percent having an ex-traperitoneal peritonectomy and 66.7% lymphadenectomy. Eighteen patients (32.7%) had at least one complication during their surgery with the most common being bleeding (55.6%). Other complications were pleural (22.2%), ureter or bladder (16.7%) and digestive injuries (5.6%). The median blood loss was around 1200 mL with forty-five patients (73.8%) being transfused during the operating time. The median hospitalization stay after the surgery was 10 days (6-50), including 7 days (2-17) in Intensive Care Unit (ICU) with a median parenteral nutrition of 6 days (2-21). The early post-operative complications affected 19 patients (31.1%) and are detailed in Table 3. The main complications were medical complications like infections, renal failure, etc (11.5%), wound dehiscence and evisceration (6.6%) and lymphoceles (6.6%). Three patients were reoperated: one for a leakage from a bowel anastomosis complicated by peritonitis, one for a subphrenic abscess and one for internal bleeding. Other complications were in terms of frequency: parietal wall infection (3.3%), thromboem-bolism (1.6%) and bleeding (1.6%).
In the ovarian cancer subgroup who had an initial peritonectomy (n = 42), Table 4, almost 73.8% of patients had a FIGO stage IIIC, 7.1% stage IVA and 16.7% stage IVB. At time of diagnosis, the median of CA125 was 1236 UI/L (8-30780) and 69% had ascites whereas only 16.7% had it at time of peritonectomy. With a median PCI of 20  at the time of diagnosis, 54.8% of patients received a NACT before undergoing an IDS in order to optimize the cytoreduction of the cancer.
Among all patients, 32 (54.1%) had a recurrence (Table 5) of their cancer, 29 in the ovarian subgroup and 3 in the Pseudomyxoma Peritoneii subgroup. Out of 49 patients in the ovarian subgroup, 21 had a peritoneal recurrence (42%), 10 had distant metastasis (20%) and 3 lymph nodes recurrence (6%). Of them, 9 patients had a second recurrence after treating the first one. Again, the predilection of recurrence was peritoneal (14%), distant metastasis (6%) and lymph node (4%). Only 2 patients (4%) had a third recurrence with peritoneal lesions in all of them. The only patient who had a preventive peritonectomy during IDS did not have any per-or post-operative complications but had a liver recurrence without peritoneal lesions after almost 1 year.
In Table 6, HIPEC regimens and site of recurrences were detailed. Different HIPEC protocols were used depending on the histology studies: irinotecan and oxaliplatin heated to 42-43 • C for 30-40 min for pseudomyxoma, endometrioid carcinoma and mucinous carcinoma, cisplatin and doxorubicin heated to 41-43 • C for 60-90 min for mesothelioma and cisplatin heated to 40-42 • C for 90 min for high-grade serous adenocarcinoma. Two patients with pseudomyxoma were given mitomycin C because of a higher risk of bleeding and allergy. Patients who received the irinotecan and oxaliplatin protocol were also given an infusion of 5-fluouracil (5-FU) along with folinic acid 1 hour prior to HIPEC. Patients who received the Cisplatin protocol were given an IV perfusion of sodium thiosulfate for renal protection.
In the ovarian subgroup who had an initial peritonectomy, 8 patients had an IPC during surgery: 5 HGSOC, 2 mucinous and 1 endometrioid ovarian cancer. All patients had a FIGO stage IIIC. Only 2 patients had a Grade 3 complication in the postoperative period: one patient with an infected lymphocele drained under CT-Scan guidance and one patient re-operated for post-operative bleeding. Five patients had a recurrence out of 8:1 peritoneal and 3 at the level of lymph nodes and 1 patient died 2 months after surgery from general status alteration with liver and lung metastasis.
At a median follow-up of 30 months (95% Confidential Interval (CI): 23-38 months), 12 deaths occurred in the whole cohort (n = 61). In the ovarian subgroup, 23 recurrences occurred leading to death in 6 patients. In this group, the median DFS was 19.1 months (15.5-26.6 months) with

Discussion
Over the last decades, concepts about the treatment of PSM and specifically advanced-stage ovarian cancer have evolved. The optimal surgical treatment has changed over time from residual tumour of less than 1 cm [21] to the complete removal of all macroscopic disease with no residual tumour [22,23], even in selected recurrence cases [24,25]. This change was independently associated with better overall survival and disease-free survival. Even though systematic complete peritonectomy is not usually recommended unless there is a widespread peritoneal involvement, it can be a necessary step for optimal surgery except in colorectal and gastric cancer carcinomatosis, where high-volume disease is a common contraindication for treatment.
Since a visual examination concurred in only 8% with the final pathologic findings, a macroscopically normal peritoneum (no obvious residual tumor) may hide a tumor beneath it with up to 50% in the region surrounding the tumor nodule and 35% in uninvolved regions [26]. According to Baratti et al. [5], selective peritonectomy of macroscopically involved peritoneum resulted in 50% of patients with residual tumour remaining on parietal surfaces. Sinukumar et al. [27] in the INDESPO study found residual disease in 23.3% of the patients with normal-looking areas of the peritoneum in the IDS. In our study, we could not study this no-tion, since we did a complete peritonectomy regardless of visual assessment. But on the contrary, 12% of supposedly affected lesions, were histologically normal.
Complete peritonectomy was associated with high morbidity (about 40%) and significant mortality (around 4%), especially when combined with visceral resection, and even in experienced hands. In a German study, the rate of postoperative complications for the same surgical technique was 54.5%, higher than the rate in our study [28]. In our study, around 28% of patients had per-operative complications like haemorrhage, bladder injury, diaphragmatic resection and bowel injury and 74% of patients being blood transfused. After 30 days, 30% had postoperative complications with grade 3-4 morbidity observed in 11%. Our rate of complication is slightly higher than other studies like the INDEPSO study [27]. This can be explained partly by the complexity of the surgery represented by the high Aletti Score in 69% of patients. The same reason can also explain the long median ICU stay (7 days) and median parenteral feeding duration (8 days).
In the ovarian cancer group, the overall survival was 77% at 3 years and the median disease-free survival was 19.1 months (15.5-26.6). Compared to other studies, like Vergote et al. [29] and Bakrin et al. [30], our patients did not have a worse overall survival. Bakrin   the survival rates [36,37]. In our study, HIPEC was used in nineteen patients, with eight of them having an advanced ovarian cancer and eight having pseudomyxoma peritonei. As expected, the rate of grade 3-4 postoperative morbidity in the HIPEC group (16.6%) was slightly higher than the rest of the patients (11%). Another significant thing is the high rate of recurrence in the ovarian subgroup which reached 50% with a preferential site of recurrence being the lymph nodes.
The limitations of our study are firstly linked to its retrospective character. There is a potential selection-bias of diseases associated with confounding risk factors. Another limitation is the heterogeneity of the studied group: it included patients with different peritoneal histology and different timing of surgery (initial or recurrence).
However, the strength of this study consists in the limited number of studies discussing this topic. Another strong point is the detailed description of the morbidity and sites of recurrence after TTP.

Conclusions
Complete peritonectomy is a feasible surgical procedure to treat PSM and their recurrences with a low rate of grade 3-4 morbidity and an acceptable operating time. Even though it does not prevent peritoneal recurrence, this technique did not show to be a risk factor for distant recurrences, especially parietal wall metastasis. It should be known and mastered by surgeons to increase the rate of optimal surgery which can affect the survival rates.

Author contributions
RS , AB, SM , FN, EL, DH designed the research study. RS, AB, FK and DH performed the Data Collection. SM and OA analyzed the data. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate
All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Centre Oscar Lambret (approval code: CEC-2022-001).