The role of omentectomy in the surgical management of mucinous and serous borderline ovarian tumors

2.1 Information sources and search strategy

A systematic literature search was conducted in September 2024 using two electronic databases, PubMed and Scopus, to retrieve relevant studies published from their inception to September 2024. The search strategy was standardized across both databases, focusing on studies related to BOTs and the omentum. The search terms included “borderline ovarian tumors” and “omentum”, along with additional terms of neoplasms (e.g., neoplasm, tumor, tumour) and surgical management (e.g., omentectomy, removal, resection). Boolean operators “AND” and “OR” were used to refine the search and ensure comprehensive retrieval of relevant studies. The search strategy was designed to capture a broad range of studies addressing the clinical significance of omentectomy and its role in BOT management. The database search strategies are presented in Table 1.

2.2 Data selection, extraction, and management

The search results were imported into Covidence software (https://www.covidence.org/) for systematic screening. Covidence is an online tool that automatically identifies and excludes duplicates from both databases, with additional duplicates manually excluded during the screening process. The screening was conducted independently by two reviewers (Samuel Haataja & Elina Urpilainen) in two stages: (1) title and abstract screening and (2) full-text review. Any disagreements were resolved through discussion until a consensus was reached.

Studies were included if they provided information on the omentum and omentectomy in the BOT context. Inclusion criteria included publications in English, appropriate study designs, and relevance to the research question. Exclusion criteria included non-English publications, an inappropriate study design (e.g., case reports, meta-analyses), and a lack of information on the omentum or omentectomy. Additionally, studies that did not explicitly address the omentum’s role in BOT management were excluded from the final analysis (Fig. 1). The following information was collected from each study: (1) number of patients, (2) BOT subtype, (3) completeness of staging, (4) definition of omentectomy, (5) omentectomy, (6) peritoneal biopsies, (7) recurrence and survival outcomes, and (8) additional relevant results.

Fig. 1.Flow chart. Identification of studies and the screening process.

2.3 Assessment of methodological quality and risk of bias

The database search identified 25 cohort studies and 15 case series. The number of cases in some case series was small, but cohort studies tended to be considerably larger. The inclusion and exclusion criteria for BOT cases and the definition of omentectomy varied between studies. Only three studies discussed possible bias extensively, and publication bias was difficult to estimate. However, no studies were excluded based on these reasons.

Methodological quality was assessed using the relevant Joanna Briggs Institute critical appraisal checklists for cohort [13] and case series [14] study designs. This included an examination of the study population selection method, measures of outcome and relevant exposures, identification and management of potential confounding variables, validity and reliability of the outcome measurement, and follow-up strategies [13, 14].

Methodological quality varied between studies. Inclusion and exclusion criteria were specified; however, confounding factors were not defined or addressed in most studies. There were considerable variations in how to handle follow-up. Some studies experienced a significant loss in follow-up, and the reasons for this were not adequately described or explored. An assessment of individual studies can be found in Supplementary Tables 1 and 2.

3.1 Study identification and selection

A total of 548 studies were identified from PubMed and Scopus. After removal of duplicates (both automatically and manually), 412 studies were included in title and abstract screening. Following this, 76 studies were selected for full-text review, resulting in the inclusion of 40 studies in the final analysis. The most common reasons for exclusion during the screening process were a wrong study design (n = 12), the full text being unavailable (n = 13), and a lack of relevant data regarding the omentum (n = 8).

3.2 Characteristics of included studies

A total of 40 studies (Supplementary Table 3 (Ref. [1, 2, 4, 5, 6, 8, 9, 10, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46])) were included in this systematic literature review, published between 1993 and 2024, with three publications from the 20th century. All studies were published in English. Most were conducted in Europe (n = 20), with some in North America (n = 7) and Asian countries (n = 6). The study periods in the included studies ranged from 1965 to 2022.

3.3 Omentectomy and omental disease in sBOTs and mBOTs

International guidelines recommend performing omentectomy as a part of surgical staging for BOTs [11]. Among the 40 studies included in this review, 39 (97.5%) reported performing omentectomy or other surgical procedures involving the omentum. Eleven reported performing infracolic omentectomy, while 23 reported performing omentectomy without further specification. Omental biopsies were performed in nine studies. Other surgical procedures, such as greater omentum resection [26, 31, 39], omental excision [34], and partial omentectomy [30], were less common. Omentectomy data were collected regardless of whether it was performed during initial or restaging surgery (Supplementary Table 3). There were no frozen sections done from the omentum in any of the included studies. In one study concerning FSS reported less pregnancies after omentectomy [1]. The other studies did not report complications in terms of omentectomy.

In studies reporting omentectomy separately for sBOTs and mBOTs, it was performed more frequently for sBOT patients. In Bendifallah et al.’s [2] study, omentectomy was performed in 42.9% of sBOT patients compared to 12.9% of mBOT patients. Guo et al. [5] and Şahin et al. [39] reported similar findings, with omentectomy rates of 74.6% and 52.4%, and 71.1% and 41.2%, for sBOT and mBOT patients, respectively. In Zilliox et al.’s [46] study, omentectomy was performed for all seven sBOT and six mBOT patients included. Kavak Cömert et al. [9], De Decker et al. [10], and Gungorduk et al. [23] only included mBOT patients, with 53.3%, 62.2%, and 40.1% omentectomy rates, respectively. In 17 studies, the number of performed omentectomies was not reported separately by subtype (Supplementary Table 3).

Among the 37 studies that reported performing omentectomy, 19 (51.4%) reported the presence of omental disease: with sBOT in 15 studies (78.9%) and with mBOT in 3 studies (15.8%). In three studies, omental disease was observed without further specifying the subtype [22, 24, 39]. In seven studies including both sBOT and mBOT patients, omental disease was exclusively or more frequently observed in patients with sBOT [2, 4, 5, 6, 16, 19, 29]. In four of these studies, omental disease was observed solely in sBOT patients [4, 16, 19, 29]. Examining both sBOT and mBOT patients, Bendifallah et al. [2] and Guo et al. [5] observed higher omental disease rates in sBOT patients. Bendifallah et al. [2] reported omental involvement in 45.1% of sBOT compared to 3.5% of mBOT patients who had undergone omentectomy. Similarly, Guo et al. [5] found omental involvement after performing omentectomy in 18.3% of sBOT and 5.4% of mBOT cases (Supplementary Table 3).

Four studies exclusively included mBOT patients [9, 10, 23, 27]. Kavak Cömert et al. [9] and De Decker et al. [10] found no omental disease among the 40 and 44 patients who had undergone omentectomy, respectively. Gungorduk et al. [23] reported omental disease in 3 of 163 (1.8%) patients who had undergone omentectomy. Jung et al. [27] did not report the results of the omentectomy procedure performed. The omental disease rate among mBOT patients was similarly low in other studies, with Guo et al. [5] reporting the highest rate at 5.4% (Supplementary Table 3).

Guo et al. [5] found that patients undergoing omentectomy were more frequently diagnosed at higher FIGO stages, and omental and peritoneal implants were significantly more common in sBOT patients compared to other subtypes. Similarly, Trillsch et al. [41] reported patients undergoing omentectomy diagnosed at higher FIGO stages and presenting with more peritoneal implants. In contrast, Lecointre et al. [31] found no association between infragastric omentectomy and the FIGO stage of the tumor. Kristensen et al. [4] observed advanced-stage tumors only in sBOTs. Regarding the location of extra-ovarian disease, Şahin et al. [39] found the omentum to be the most frequent site of BOT implants. In contrast, Gungor et al. [24] found the omentum and appendix to be the least involved organs.

3.4 Omental disease and upstaging

Seven studies reported upstaging due to omental disease [4, 5, 6, 16, 28, 29, 38], all exclusively in sBOT patients. Among the four studies that included only mBOT patients, Jung et al. [27] reported upstaging in 16 (4.4%) patients in the staging group but did not provide data regarding the extra-ovarian disease location.

Guo et al. [5] found that upstaging was more frequent in sBOTs compared to non-sBOTs, highlighting a significant difference. Similarly, Kristensen et al. [4] and Daraï et al. [6] found a higher risk of upstaging in sBOT patients. Daraï et al. [6] specifically reported a higher risk of upstaging in sBOT patients with initial FIGO stage I or II disease. Kristensen et al. [4] and Camatte et al. [16] observed omental upstaging only in sBOT patients. Kristensen et al. [4] found noninvasive implants in the omentum, leading to the upstaging of three sBOT patients with presumed stage I disease and one sBOT patient with presumed stage II disease to stage IIIA. Kane et al. [28] and Nezhat et al. [38] exclusively studied sBOT patients and found upstaging due to omental disease. Kane et al. [28] reported that 4 of 18 sBOT patients with initial stage I disease were upstaged to stage IIIA due to microscopic implants found on the omentum or in the peritoneal biopsy.

3.5 Peritoneal biopsies and peritoneal implants

The ESMO-ESGO consensus guidelines recommend using the term metastasis for extraovarian disease in mBOT, while the term implant should be used for sBOT [11]. Three studies exclusively included sBOT patients with peritoneal implants [20, 21, 39]. Of the 37 other studies, peritoneal biopsies were performed in 28 (75.7%). Peritoneal biopsies were more frequently performed in sBOT compared to mBOT patients. For example, Bendifallah et al. [2] reported peritoneal biopsies performed for 76.4% of sBOT patients, compared to 34.7% of mBOT patients. Similarly, Şahin et al. [39] reported biopsies performed for 71.1% and 41.2% of sBOT and mBOT patients in the staging group, respectively. Data on peritoneal biopsy rate by subtype were not available in most studies (Table 2, Ref. [1, 2, 4, 5, 6, 8, 9, 10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46]).

In the four studies exclusively including mBOT patients [9, 10, 23, 27], peritoneal metastases were observed only in Gungorduk et al.’s [23] study, with noninvasive implants found in 12 of 364 patients (3.3%). Jung et al. [27] reported performing peritoneal biopsies but reported no results [27] (Table 2).

In seven studies including both sBOT and mBOT patients, the prevalence of peritoneal implants was higher, or peritoneal implants were only observed in sBOT [2, 4, 16, 24, 29, 39, 45]. Bendifallah et al. [2] observed peritoneal implants in 30.2% of sBOT and 6.7% of mBOT patients with peritoneal biopsies performed. Şahin et al. [39] reported peritoneal implants in 13 of 76 (17.1%) sBOT and 2 of 51 (3.9%) mBOT patients, with invasive implants only seen in sBOTs. However, data on peritoneal biopsy rate were not available in this study. Similarly, Westermann et al. [45] reported peritoneal implants in 101 of 312 (32.4%) sBOT and 2 of 129 (1.6%) mBOT patients, with invasive implants only seen in sBOTs and peritoneal biopsies performed for 93.9% of the whole study population (Table 2).

Kristensen et al. [4] found noninvasive implants in sBOT patients only. In contrast, Fotopoulou et al. [22] reported higher rates of invasive peritoneal implants in mBOT compared to sBOT patients (33.3% vs. 12.8%), while noninvasive implants were more frequently observed in sBOT patients (25.6% vs. 8.3%). Peritoneal biopsies were performed for 46 of the 51 patients [22]. Regarding both peritoneal and omental findings, Guo et al. [5] found that both omental and peritoneal implants were more common in women with sBOT, with a statistically significant difference. Similarly, Kumpulainen et al. [29] observed peritoneal and omental implants in sBOT patients only.

3.6 Recurrence and survival outcomes

Recurrences were observed in 35 studies. In eight studies, recurrences were observed only in sBOT patients or were more frequent in sBOT compared to mBOT patients, with recurrence rates ranging from 3.9% to 29.2% in sBOT and 2.0% to 14.8% in mBOT [1, 2, 5, 16, 24, 31, 39, 45]. In contrast, two studies reported a higher recurrence rate in mBOT patients: Damak et al. [17] (38.1% vs. 27.8%) [17] and Kærn et al. [15] (10.1% vs. 4.0% [15]; Supplementary Table 3).

Among studies exclusively including mBOT patients, recurrences were observed in three out of four studies. Kavak Cömert et al. [9] observed one recurrence among 75 patients (1.3%) with no deaths during a 51-month follow-up. Gungorduk et al. [23] reported recurrence in 24 out of 364 mBOT patients (6.6%), with no significant difference between recurrence and completeness of staging. Jung et al. [27] reported recurrence rate of 3.5% in 432 patients, while De Decker et al. [10] observed no recurrences in mBOT patients (Supplementary Table 3).

In advanced-stage BOTs, recurrence rates were higher. Westermann et al. [45] reported recurrence rate of 5.9% for FIGO stage I patients compared to 40.0% for FIGO stages II–IV patients with a significant difference. Bendifallah et al. [2] reported recurrence rates of 19.1%, 33.3% and 29.7% for patients with FIGO stages I, II, and III BOT, respectively. In a study by Shih et al. [40], 21% of patients with advanced-stage sBOT experienced recurrence, while Falcone et al. [20, 21] reported recurrence rates of 54.0% and 85.0%, respectively, in patients with advanced stage (FIGO stages II and III) sBOTs. Shih et al. [40] reported that only patients with omental disease or metastases to multiple sites developed recurrent disease, with a recurrence rate of 28.0% (Supplementary Table 3).

Results regarding the impact of omentectomy on recurrence and survival varied between studies. Qi et al. [1] found omentectomy to have no effect on recurrence. Gungorduk et al. [23] reported no differences in progression-free survival or overall survival (OS) in patients undergoing omentectomy. Nasioudis et al. [38] reported no advantage in OS for patients with stage I BOT undergoing omentectomy. In this study, five-year OS rates for sBOT and mBOT patients were 66.4% and 84.5%, respectively [37]. Menczer et al. [35] also found no difference between omentectomy and OS. In contrast, Trillsch et al. [41] found omentectomy to be associated with improved progression-free survival, even when adjusting for FIGO stage. Similarly, Bendifallah et al. [2] observed improved five-year OS with complete staging.

Regarding surgical staging in general, Guvenal et al. [25] found no difference in survival rates based on the extensiveness of surgical staging. Similarly, Guo et al. [5] found no significant difference in recurrence rates between complete and incomplete surgical staging (14.0% vs. 13.7%). However, complete surgical staging was associated with significantly lower recurrence rate in FIGO stage > I patients (26.0 vs. 44.8%), while no significant difference was observed in FIGO stage I patients [5]. Şahin et al. [39] found that surgical staging or presence of implants were not associated with recurrence. Regarding peritoneal biopsies, Camatte et al. [16] found that lack of complete peritoneal staging in presumed stage I BOTs increased recurrence rates. Falcone et al. [21] observed that invasive peritoneal implants had impact on disease-free survival (DFS) but had no effect on OS.

3.7 Surgical approach: staging, conservative treatment, and surgical route

Of the 40 included studies, 16 (40%) reported performing complete surgical staging, including omentectomy and peritoneal biopsies (Supplementary Table 3 and Table 2). Both laparoscopic and laparotomic approaches were used in 18 studies [1, 2, 5, 6, 9, 16, 20, 21, 22, 23, 26, 30, 31, 33, 34, 41, 43, 46]. Several studies reported performing conservative surgery, also referred to as FSS, generally defined as preserving the uterus and at least one ovary [1, 2, 5, 6, 9, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26, 28, 29, 34, 38, 41, 42, 43, 46].

FSS on recurrence and survival was evaluated. Kristensen et al. [4] and Gungorduk et al. [23] found no significant association between FSS and recurrence, while six studies reported higher recurrence rates in FSS patients, compared to non-FSS patients [1, 17, 31, 33, 39, 42]. For example, Qi et al.[1] reported higher recurrence rates in patients undergoing FSS compared to non-FSS treatment (10.2% vs. 2.5%), while Lecointre et al. [31] reported 17.3% in FSS compared to 3.1% in non-FSS patients. Despite higher recurrence rates, Loizzi et al. [33] found no significant difference in OS between the FSS and non-FSS groups.

Regarding surgical staging and survival outcomes, Bendifallah et al. [2] reported higher five-year OS in the complete staging group compared to incomplete staging (98.4% vs. 93.8%). In contrast, Guo et al. [5] reported higher five-year OS in the incomplete staging group, despite higher five-year DFS in the complete staging group. Westermann et al. [45] reported significantly shorter DFS after FSS compared to patients undergoing radical surgery. Kristensen et al. [4] recommended staging surgery including omentectomy for sBOT patients and radical treatment for sBOT patients without fertility-preserving desire. Daraï et al. [6] recommended performing restaging operations for sBOT patients.