Laparoscopy versus laparotomy for FIGO stage I ovarian cancer

  • Comment
  • Review
  • Intervention

Authors


Abstract

Background

This is an updated version of the original review that was first published in the Cochrane Database of Systematic Reviews 2008, Issue 4. Laparoscopy has become an increasingly common approach to surgical staging of apparent early-stage ovarian tumours. This review was undertaken to assess the available evidence on the benefits and risks of laparoscopy compared with laparotomy for the management of International Federation of Gynaecology and Obstetrics (FIGO) stage I ovarian cancer.

Objectives

To evaluate the benefits and risks of laparoscopy compared with laparotomy for the surgical treatment of FIGO stage I ovarian cancer (stages Ia, Ib and Ic).

Search methods

For the original review, we searched the Cochrane Gynaecological Cancer Group Trials (CGCRG) Register, Cochrane Central Register of Controlled Trials (CENTRAL 2007, Issue 2), MEDLINE, EMBASE, LILACS, Biological Abstracts and CancerLit from 1 January 1990 to 30 November 2007. We also handsearched relevant journals, reference lists of identified studies and conference abstracts. For this updated review, we extended the CGCRG Specialised Register, CENTRAL, MEDLINE, EMBASE and LILACS searches to 6 December 2011.

Selection criteria

Randomised controlled trials (RCTs), quasi-RCTs and prospective cohort studies comparing laparoscopic staging with open surgery (laparotomy) in women with stage I ovarian cancer according to FIGO.

Data collection and analysis

There were no studies to include, therefore we tabulated data from non-randomised studies (NRS) for discussion.

Main results

We performed no meta-analyses.

Authors' conclusions

This review has found no good-quality evidence to help quantify the risks and benefits of laparoscopy for the management of early-stage ovarian cancer as routine clinical practice.

Plain language summary

Laparoscopy versus laparotomy (open surgery) for early-stage ovarian cancer

Stage I ovarian cancer is diagnosed when the tumour is confined to one or both ovaries, without spread to lymph nodes or other parts of the body. Approximately 25% of women with ovarian cancer will be diagnosed at an early stage, thus the diagnosis often occurs due to an accidental finding. The intention of surgical staging is to establish a diagnosis, to assess the extent of the cancer and to remove as much tumour as possible. The latter is particularly important as women with ovarian cancer survive for longer when all visible tumour has been removed.

We conducted this review in an attempt to clarify whether laparoscopy (keyhole surgery) is as safe and effective as laparotomy (open surgery) for early-stage ovarian cancer. We intended to include only high-quality studies that compared the two types of surgery.

We wanted to know whether women having laparoscopy survived as long as those having open surgery and whether there were differences in the time it took for the cancer to get worse. We were also interested to see how these different surgeries compared with regard to blood loss and other complications.

Unfortunately, we were unable to find any high-quality randomised trials comparing these approaches. Further research is needed.

Background

Description of the condition

This is an updated version of the original review that was first published in the Cochrane Database of Systematic Reviews 2008, Issue 4.

Ovarian cancer is the eighth most common cancer in women worldwide (Jemel 2011). A woman's risk of developing ovarian cancer before the age of 75 ranges from 0.5% in developing countries to 1% in developed countries (GLOBOCAN 2008; Jemel 2011). Just over a third of women with ovarian cancer are alive five years after diagnosis (EUROCARE 2003), largely because most women with ovarian cancer are diagnosed when the cancer is already at an advanced stage (Jemal 2008). International Federation of Gynaecology and Obstetrics (FIGO) stage I ovarian cancer (limited to the ovaries) is diagnosed in approximately 20% to 33% of women with ovarian cancer in developed countries (Maringe 2012) and diagnosis is usually made by accidental discovery at sonography, computerised tomography (CT scanning) or during laparoscopy. The incidence of accidental discovery of ovarian cancer at laparoscopy has been estimated to range from 0.65% (Wenzl 1996) to 0.9% (Muzii 2005) of premenopausal women and 3% of postmenopausal women who undergo the procedure for an adnexal mass (Muzii 2005), but may be higher depending on the selection criteria applied.

Most cancers of the ovary are epithelial (90%) with histological subtypes including serous (35%), endometrioid (10%), borderline (16%), mucinous (8%), clear cell (4%), undifferentiated and mixed epithelial (Kosary 2007). In general, the prognosis of ovarian tumours depends on the FIGO stage, tumour grade, histological subtype, age and the volume of residual disease after surgery (Benedet 2000), however for stage I tumours the most important prognostic indicators are considered to be the degree of differentiation (grade) and the occurrence of tumour rupture (Vergote 2001).

The standard management of women with ovarian cancer is comprehensive surgical staging by laparotomy, a midline abdominal incision that allows exposure of the entire abdomen. Comprehensive surgical staging includes a total hysterectomy, bilateral salpingo-oophorectomy, removal of all obvious sites of tumour, aspiration of cytological washings or ascites, omentectomy, retroperitoneal (pelvic and para-aortic) lymph node dissection or sampling and biopsy of all suspicious-looking areas including mesentery, liver and diaphragm (Benedet 2000; Schorge 2012). Systematic retroperitoneal lymph node dissection (RLND) may improve survival in stage I ovarian cancer by detecting microscopic disease (Chan 2007) and is considered a standard procedure in some centres (Schorge 2012), however the UK National Institute for Clinical Excellence (NICE) guidelines currently do not recommend RLND in stage I disease (NICE 2011).

A meta-analysis of four randomised controlled trials (RCTs) of adjuvant platinum-based chemotherapy, which included data from the International Collaborative Ovarian Neoplasm 1 (ICON1) trial (Trimbos 2003) and the Adjuvant Chemotherapy in Ovarian Neoplasm (ACTION) trial (Trimbos 2004), found that adjuvant chemotherapy significantly improved overall survival (OS) and progression-free survival (PFS) in women with early ovarian cancer (Winter-Roach 2012). However, it was considered not to be necessary in women with comprehensively staged, stage Ia or 1b grade 1 to 2 tumours, as subgroup analyses suggested that women who were optimally staged were unlikely to benefit from adjuvant chemotherapy. Hence, comprehensive surgical staging has an important impact on the subsequent management of women with early ovarian cancer, with adjuvant chemotherapy indicated when staging is considered to be inadequate (Elit 2004; Winter-Roach 2012).

Description of the intervention

The intention of surgical staging is to establish a diagnosis, to assess the extent of the disease and to remove as much gross tumour as possible (Schorge 2012). Surgical staging of ovarian cancer by laparoscopy is the same intra-abdominal procedure as that performed by laparotomy except that it involves two or more, much smaller, abdominal incisions, through which laparoscopic instruments are then inserted. Specimen retrieval bags are used to prevent spillage and possible seeding of cyst contents and to avoid contact with incision (port) sites. Cysts may be aspirated within the retrieval bag, or morcellated if solid, to facilitate extraction through the port sites (Ghezzi 2007). Larger specimens, like omentum, may be extracted through the vagina with the uterus after hysterectomy (Lee 2011; Park 2008a).

How the intervention might work

Several recent non-randomised studies (NRSs) in early ovarian cancer have reported that laparoscopic surgical staging is a safe and technically feasible procedure (Colomer 2008; Ghezzi 2009; Nezhat 2009; Park 2008b; Park 2010). The possible advantages of laparoscopy include smaller incisions, less blood loss, faster recovery, shorter hospital stay, fewer complications, less postoperative infection and a better visualisation of the tumour inside the abdomen as the laparoscopy image can be magnified (Gad 2011; Ghezzi 2007; Lee 2011). In addition, the shorter recovery period following laparoscopy means that chemotherapy can be commenced sooner compared with laparotomy (Ghezzi 2007; Nezhat 2009), potentially resulting in a favourable effect on survival.

However, laparoscopy has been associated with a higher rate of intraoperative cyst rupture for apparently benign (Muzii 2005) and borderline tumours (Fauvet 2005), which may result in upstaging of the unexpected ovarian cancer from stage Ia or 1b to Ic (Muzii 2005). It has been argued that some aspects of comprehensive surgical staging, particularly RLND, may be technically difficult to achieve via laparoscopy and, therefore, that laparoscopy should be restricted to women with pre-operative evidence of benign conditions only (Vergote 2004). Other disadvantages of laparoscopy may include longer operating times and the possibility of port-site metastases, although the risk of the latter in early disease is considered to be low (Schorge 2012). Furthermore, to facilitate laparoscopy, CO₂ is commonly used for pneumoperitoneum and has been shown to lower the peritoneal pH (Bergstrom 2008; Kuntz 2000) which may activate enzymes that increase tumour cell mitosis and growth factor production. In addition, mechanical damage to the mesothelium may occur with prolonged laparoscopic surgery, thereby increasing the risk of metastases in the abdominal cavity (Greene 1995; Volz 1999).

Why it is important to do this review

Laparoscopic surgical staging of stage I ovarian cancer remains controversial as it is unclear how the risks and benefits of this procedure compare with the conventional open approach by laparotomy. An earlier version of this systematic review, published in 2008, found insufficient evidence to evaluate laparoscopy for the management of early ovarian cancer as routine clinical practice. We continue to update this review with the aim of clarifying and consolidating the available evidence regarding this alternative surgical approach.

Objectives

To evaluate the benefits and harms of laparoscopy in the surgical treatment of FIGO stage I ovarian cancer (stages Ia, Ib and Ic) when compared with laparotomy.

Methods

Criteria for considering studies for this review

Types of studies

RCTs and quasi-RCTs. We also considered prospective cohort studies where the results had been adjusted for the baseline case mix using multivariate analyses, and excluded those with historical (non-concurrent) controls.

Types of participants

Women with stage I ovarian cancer defined by FIGO as follows.

  • Stage Ia: unilateral tumours

  • Stage Ib: bilateral tumours

  • Stage Ic: identified tumour spillage, tumour capsular penetration, positive peritoneal cytology

Types of interventions

Surgical staging via laparoscopy (experimental group) versus laparotomy (control group) for stage I ovarian cancer.

Types of outcome measures

Primary outcomes
  1. Overall survival (OS)

  2. Progression free survival (PFS)

Secondary outcomes
  1. Operating time

  2. Intraoperative tumour rupture

  3. Pelvic and para-aortic lymph node yield

  4. Size of omental specimen

  5. Estimated blood loss and the need for blood transfusion

  6. Comprehensive staging achieved by the allocated procedure (conversion to laparotomy)

  7. Surgical complications (immediate and delayed) including: injuries to the bladder, ureter, blood vessels, nerves, small bowel and colon; febrile morbidity; intestinal obstruction; haematomas and infections

  8. Length of hospital stay

  9. Time to adjuvant chemotherapy

  10. Systemic complications

  11. Abdominal wall recurrence: laparoscopy (port sites) and laparotomy (midline incision).

  12. Quality of life

Search methods for identification of studies

Electronic searches

We conducted searches to identify all published and unpublished RCTs and NRSs that compared laparoscopy and laparotomy for stage I ovarian cancer. The search strategies identified studies in all languages and, when necessary, we translated non-English language papers so that they could be fully assessed for potential inclusion in the review.

We searched the Cochrane Gynaecological Cancer Review Group (CGCRG) Trials Register, Cochrane Central Register of Controlled Trials (CENTRAL 2007, Issue 2), MEDLINE (January 1990 to November 2007), EMBASE (1990 to November 2007), LILACS (1990 to November 2007), Biological Abstracts (1990 to November 2007) and CancerLit (1990 to November 2007). For this updated version of the review, we extended these searches to 6 December 2011. (See Appendix 1, Appendix 2, Appendix 3 and Appendix 4 for the search strategies).

Searching other resources

We handsearched the citation lists of relevant publications and included studies, and contacted experts in the field to identify further trials. For the original review we also handsearched the following conferences and publications: Gynecologic Oncology, International Journal of Gynaecological Cancer, British Journal of Cancer, British Cancer Research Meeting, Annual Meetings of the International Gynaecologic Cancer Society, Annual Meetings of the American Society of Gynecologic Oncologists, Annual Meetings of the European Society of Medical Oncology (ESMO), and Annual Meetings of the American Society of Clinical Oncology (ASCO).

Data collection and analysis

Selection of studies

Two review authors sifted the searches and identified potentially eligible studies. All authors assessed the methodology of these potentially eligible studies according to the specific inclusion criteria. Review authors were not blind to the authors, institutions or journals of potentially relevant studies.

Data extraction and management

No studies fulfilled the inclusion criteria for this review. For future versions of this review, two authors will independently extract data from included trials to a pre-designed data collection sheet that includes the following information:

  • Study methodology: description of randomisation, blinding, number of study centres, study duration, length of follow-up and number of study withdrawals.

  • Participants: number, mean age, mean risk score.

  • Intervention: type of intervention, dose and schedule.

  • Outcomes:

    • We will extract data to allow for intention-to-treat (ITT) analysis where possible.

    • For dichotomous outcomes (e.g. number of lymph nodes, complications or deaths), we will extract outcome rates to estimate a risk ratio (RR).

    • For continuous outcomes (e.g. quality of life (QoL) measures and duration of treatment) we will extract means and standard deviations (SD) to estimate a mean difference (MD).

    • For time-to-event outcomes (e.g. overall survival) we will extract the log of the hazard ratio (log(HR)) and its standard error from trial reports. If these are not reported, we will attempt to estimate the log (HR) and its standard error using Parmar's methods (Parmar 1998).

Assessment of risk of bias in included studies

For future versions of this review we will assess the risk of bias in included studies using The Cochrane Collaboration's tool (Higgins 2011) and the following criteria:

  1. selection bias: random sequence generation and allocation concealment;

  2. performance bias: blinding of participants and personnel (patients and treatment providers);

  3. detection bias: blinding of outcome assessment;

  4. attrition bias: incomplete outcome data;

  5. reporting bias: selective reporting of outcomes;

  6. other possible sources of bias.

Results

Description of studies

Results of the search

The original search identified 706 citations, of which we retrieved 43 for detailed examination. We subsequently excluded 40 of these records and three NRSs (two case-control studies and one case series) were included in the original review (Ghezzi 2007; Hua 2005; Tozzi 2004; Figure 1). For this updated review, we excluded these NRSs but tabled their findings with other similar studies that were identified by the updated search (see Differences between protocol and review).

Figure 1.

Study flow diagram of original search 17 May 2007

From the updated search we identified 1395 records (1140 after de-duplication), 28 of which we screened for possible relevance. Of these, 11 new studies were identified for classification (Chen 2010; Chi 2005; Colomer 2008; Ghezzi 2009; Lee 2011; Nezhat 2009; Park 2008a; Park 2008b; Park 2010; Park 2011; Wu 2010; Figure 2). Park 2008b, Park 2010 and Park 2011 are extensions of the same series.

Figure 2.

Study flow diagram of updated search 30 November 2011

Included studies

There were no studies that met the inclusion criteria.

Excluded studies

Altogether we excluded 54 studies. None of these studies met the inclusion criteria in Types of studies. We have summarised the results of the relevant case series, case-control studies and retrospective cohort studies in three tables: Table 1, Table 2 and Table 3, respectively. None of the comparative NRSs reported adjusting results for baseline characteristics and we considered all of them to be at a high risk of selection bias and other bias (e.g. outcome assessment bias).

Table 1. Case series of comprehensive laparoscopic staging of early ovarian cancer (including fallopian tube cancer)
  1. Abbreviations: PFS = progression-free survival; OS = overall survival; n = number

    ¹Leblanc 2004 reported recurrences in confirmed stage I women only, i.e. excluding 8 women who were upstaged.

    ²Mean follow-up.

 No. of womenMean pelvic nodes (n)Mean para-aortic nodes (n)Median follow-up (months)Recurrences (n)PFS (%)OS (%)
Querleu 19948-8.6----
Pomel 199587.58.5----
Childers 199514------
Tozzi 20042419.419.646292100
Leblanc 2004421420543/34 ¹9198
Colomer 2008201811.324.7195100
Ghezzi 20092624.59.826.719696
Nezhat 20093614.812.255.9 ²392100
Chen 20104316.66.524.7393-
Table 2. Case-control studies of laparoscopic surgical staging versus open surgical staging of early ovarian cancer
  1. Abbreviations: LPS = laparoscopy; LPT, laparotomy; BMI = body mass index; PFS = progression-free survival; OS = overall survival; n = number; NA = not applicable; SD = standard deviation; NS = not specified.

    ¹Lymph nodes not specified as pelvic or para-aortic in origin.

    ²Right obturator nerve damaged and repaired in one patient.

Study name Hua 2005 Chi 2005 Ghezzi 2007
Period2002-20042000-20032003-2006
DesignProspective cohortRetrospective case-controlCase-control (with historical controls 1997-2003)
Intervention LPS LPTP value LPS LPTP value LPS LPTP value
Number of women1011-2030-1519-
Mean age4042-47.350.60.3155610.05
BMI (kg/m²)---24.625.40.6423.825.80.19
Mean operating time (min)298182 < 0.05321276 0.04377272 0.002
Mean blood loss (ml) (SD)280346 < 0.05235367 0.0032504000.28
Blood transfusion n (%)------1 (6.7)2 (10.5)1.0
Pelvic lymph nodes, n25 ¹27 ¹NS12.314.7NS25.225.10.96
Para-aortic nodes, n---6.79.2NS6.570.78
Omental specimen (cm³)---1863470.09---
Intra-operative tumour spillage, n (%)00----3 (20)2 (10.5)0.63
Postoperative complications n (%)2 (20%) ²7 (72.7%) < 0.010 (0)3 (10)-2 (13.3%)8 (42.1%)0.13
Hospital stay (days)---3.15.8 < 0.00137 0.001
Final diagnosis = stage I, n (%)------11 (73.3)13 (68.4)-
Tumour upstaged, n (%)00----4 (26.7)6 (31.6)1.0
Conversion to LPT, n---0NA-0NA-
Adjuvant chemotherapy, n (%)------11 (73.3)13 (68.4)-
Time to adjuvant chemotherapy (days)---------
Median follow-up in months (range)------16 (4-34)60 (32-108)-
Port-site/abdominal wound metastasis, n---0-----
Recurrences, n (%)------04 (21)-
OS, n (%)------15 (100)19 (100)-
Table 3. Retrospective cohort studies of laparoscopic surgical staging versus open surgery for early ovarian cancer
  1. Abbreviations: LPS = laparoscopy; LPT, laparotomy; BMI = body mass index; PFS = progression-free survival; OS = overall survival; n = number; NA = not applicable; NS = not specified.

    *These studies are expansions of the original data set (Park 2008b).

    ¹Wound infections, fever, ileus.

    ²Including one intra-operative great vessel injury repaired via a small abdominal incision.

    ³All these women were staged Ia/b compared with only five Ia/b and eight Ic cases in the LPT group.

    ⁴Including 10 lymphoceles in the LPT group. Two umbilical hernias occurred in the LPS group.

    ⁵One woman diagnosed as FIGO stage Ia grade 1 had severely disseminated disease 7 months postoperatively and died of the disease 15 months later. The other woman was stage Ia grade 2 at LPS and developed recurrence at the vaginal stump.

Study name Park 2008a Park 2008b Park 2010* Park 2011* Lee 2011
Study period2001-20062004-20072004-20082004-20102005-2010
Intervention LPS LPTP value LPS LPTP value LPS LPTP value LPS LPTP value LPS LPTP value
Number of women1719-1933-4076-84128-2687-
Mean age (years)43.248.90.15543.945.40.614--NS--NS4244.40.437
BMI (kg/m²)22.824.20.24723.222.70.578--NS--NS21.923.30.185
Mean operating time (min)303.8290.40.706221275 0.012230278 0.001207262 < 0.001228184 0.016
Mean blood loss (ml)231505 0.001240569 0.005301494 0.004252454 < 0.001230475 < 0.001
Blood transfusion, n (%)0 (0)2 (11)-1 (5)10 (30) 0.046 (15)23 (30)0.0716 (13)36 (28) 0.0120 (0)20 (23) 0.006
Median pelvic lymph nodes, n13.719.30.05227.233.90.079--NS--NS23.522.80.867
Median para-aortic nodes, n8.96.40.1876.68.80.324--NS--NS9.94.8 0.003
Omental specimen (cm³)---1602740.113--NS--NS---
Tumour size, mean (cm)4.04.50.6188.911.00.254------ 9.1 14.0 0.010
Complications n (%)0 (0)4 (21)¹-0.290---6 (7.1)25 (19.5) 0.013220⁴-
Return to bowel movement (days)3.82.0 < 0.0011.33.6 < 0.0011.73.6 < 0.0011.83.1 < 0.001---
Hospital stay (days)9.414.1 0.0028.914.5 0.0027.914.5 0.0026.313.5 < 0.0016.412.4 < 0.001
Final diagnosis = stage I, n16 ³13-15260.936------25820.212
Intra-operative tumour spillage, n (%)00-241.000--NS--NS013 (14.9) 0.037
Adjuvant chemotherapy, n (%)10 (59)17 (89)0.19615 (79)26 (79)-------17 (65.4)65 (74.7)0.453
Time (days) to adjuvant chemotherapy11.114.30.14012.817.6 0.04912.813.9 < 0.00115.820.7 < 0.0018.510.3 0.007
Upstaged, n (%)160.0924 (21)-----------
Conversion to LPT, n0NA-1NA----------
Port-site/abdominal wound metastases0--0--------00-
Median follow-up in months (range)19 (5 to 56)14 (5 to 61)-17 (2 to 40)23 (1 to 44)-------12 (1 to 42)25 (1 to 74)-
PFS, n (%)15 ⁵(88)19 (100)-19 (100)33 (100)-37 (92)71 (93)0.87666 (78)100 (78)0.87326 (100)79 (91)0.195
OS, n (%)16 (94)19 (100)-19 (100)33 (100)-38 (96)71 (94)0.84175 (89)110 (86)0.731---

Risk of bias in included studies

Not applicable.

Effects of interventions

There were no studies to include therefore no meta-analyses could be performed. Table 2 and Table 3 present the available data from the case-control and retrospective studies to date.

Discussion

Stage I ovarian cancer is a rare disease and the use of laparoscopy for surgical staging thereof is a relatively new field of clinical study, therefore data are scarce. Recent UK guidelines on the management of ovarian cancer do not consider laparoscopy as an approach to the surgical staging of early ovarian cancer (NICE 2011); however, the German Gynaecological Oncology Group (AGO) have cautiously included the option of this procedure in their recent guidelines, for selected patients and only when performed by expert laparoscopic oncology surgeons, pending further evidence (Mettler 2009).

Summary of main results

We found no randomised controlled trials (RCTs) to include in this review and from which to compare the risks and benefits of laparoscopy with the conventional open approach. Existing non-randomised evidence comparing these interventions is extremely limited and is particularly at risk of selection bias. We considered including case-control non-randomised studies (NRSs) in meta-analyses, however sample sizes were small, none of these studies reported performing statistical adjustments for baseline case mix using multivariate analyses (e.g. age, final FIGO stage, grade, tumour size, co-morbidity and adjuvant chemotherapy), the duration of follow-up varied widely, and the primary outcomes of this review (OS and PFS) were not consistently reported.

Overall completeness and applicability of evidence

Survival

According to Surveillance Epidemiology and End Results (SEER; Kosary 2007), five-year OS rates for stage Ia, Ib and Ic ovarian adenocarcinoma (excluding borderline tumours) are about 94%, 91% and 80% respectively. However, survival data relating to the surgical approach (laparoscopy versus laparotomy) in the existing literature are extremely limited: comparative studies of laparoscopy versus laparotomy for early ovarian cancer to date include three case-control studies (Table 2; Chi 2005; Ghezzi 2007; Hua 2005) and five retrospective cohort studies (Table 3; Lee 2011; Park 2008a; Park 2008b; Park 2010; Park 2011), three of which are expansions of the same case series (Park 2008b; Park 2010; Park 2011). Of these eight studies, six reported survival data but the length of follow-up varied widely or was not reported. Lee 2011 reported higher PFS rates in the laparoscopy group (100% for laparoscopy versus 91% for laparotomy) and did not report OS, however median follow-up was much shorter in the laparoscopy group compared with the laparotomy group (12 versus 25 months) and mean tumour size was significantly larger in the laparotomy group (P = 0.01). Park 2011 reported OS of 89% and 86% for laparoscopy and laparotomy respectively and PFS rates of 78% in each group, but did not report the duration of follow-up in each group. Ghezzi 2007 reported 100% OS in both groups, however the median duration of follow-up differed substantially between the groups (16 months in the laparoscopy group compared with 60 months in the laparotomy group).

Two studies conducted in women with early ovarian cancer (Park 2008a; Wu 2010) have reported unfavourable survival outcomes with laparoscopy. In Park 2008a (OS = 88% in the laparoscopy group versus 100% in the laparotomy group; Table 3), one woman who was diagnosed with FIGO stage Ia grade 1 ovarian cancer was shown to have severely disseminated disease at seven months and died of the disease 15 months later; the other woman had stage Ia grade 2 ovarian cancer at laparoscopy and developed recurrence at the vaginal stump. The extent of laparoscopy surgical staging was considered sufficient in the latter case, the tumour was not ruptured and retrieval bags were used. Wu 2010 reported data from a cohort of women with stage 1 ovarian cancer treated between 1984 and 2006 and found that those in whom the initial surgical approach was laparoscopic had significantly worse PFS and OS than those who underwent laparotomy (OS hazard ratio (HR) = 3.52), however, comprehensive staging was not the purpose of the laparoscopy in most of these women, therefore these results are difficult to interpret. In general, we consider the available survival data to be of a very low quality, hence it is not possible draw any conclusions regarding the relative effect of laparoscopic surgical staging compared with laparotomy on ovarian cancer survival from the existing literature.

Feasibility

Measures of the technical feasibility of laparoscopy have included pelvic and para-aortic lymph node yields, the size of the omental specimen, operating times and intra-operative tumour spillage. All case-control and cohort studies identified have reported statistically similar yields of retroperitoneal lymph nodes between their laparoscopy and laparotomy groups. Only two comparative studies (Chi 2005; Park 2008b) have reported mean omental specimen volumes, which were not statistically significantly different. With regard to operating times, some studies report significantly longer times with laparoscopy (Chi 2005; Ghezzi 2007; Hua 2005; Lee 2011) whilst others have reported significantly shorter times with laparoscopy (Park 2008b; Park 2010; Park 2011). These differences probably reflect differences in surgeons' skills and laparoscopic techniques between investigator teams.

Rupture or spillage of ovarian tumours during surgery has been reported to occur more frequently with laparoscopy than laparotomy (Romagnolo 2006) and has been identified as a prognostic indicator of disease-free survival (Vergote 2001). To date, six out of eight NRSs have compared rates of tumour spillage between laparoscopy and laparotomy groups (Hua 2005;Lee 2011; Park 2008a; Park 2008b; Park 2010; Park 2011). Five of these studies reported no significant difference between the two groups, and one study (Lee 2011) reported a statistically significantly higher rate of spillage in the laparotomy group (0% versus 14.9%; P = 0.037) which also had a significantly larger mean tumour size compared with the laparoscopy group. The definitions of spillage vary and the distinction between tumour rupture and puncture is not detailed in most studies (Ghezzi 2009). To properly assess these outcomes, technique and definitions need to be clearly defined in future studies. However, these limited data suggest that laparoscopy staging of early ovarian cancer is technically feasible when performed by experienced laparoscopic gynaecology oncology surgeons.

An inherent shortcoming of laparoscopy for surgical staging is the inability to palpate lymph nodes and other peritoneal surfaces (Colomer 2008; Park 2008a), however Chi 2010 argues that intra-operative direct visualisation and evaluation of nodes by palpation is inherently subjective. In a recent prospective study, of 111 women with apparent early ovarian cancer who underwent comprehensive staging by laparotomy that included retroperitoneal lymph node dissection (RLND), retroperitoneal nodal metastases were present in 13.5% of the women (Ditto 2012), which suggests that without RLND many women would be under-staged. However, systematic RLND may be associated with significant morbidity and is not a routine part of staging for early ovarian cancer in the UK, where clinical guidelines currently recommend retroperitoneal lymph node assessment with sampling of suspicious nodes (NICE 2011). Therefore, where RLND is not routine, lymph node palpation may play a crucial role in the decision-making process with regard to sampling. Another technically difficult part of the surgical staging procedure is the examination of the diaphragmatic peritoneum behind the liver and spleen and the dome of the liver (Park 2008a); this may be more difficult with laparoscopy, although it has been argued that isolated metastases to these areas are rare (Ghezzi 2009).

Safety

Surgical staging for ovarian cancer is a radical procedure that may be associated with severe intra-operative vascular, nerve, lymphatic, bowel and urinary tract complications. Common postoperative complications include wound infection, ileus, febrile morbidity and lymphoceles (Ghezzi 2007; Lee 2011; Park 2008a; Park 2008b). Three comparative studies in early ovarian cancer have reported significantly fewer postoperative complications with laparoscopy compared with laparotomy (Hua 2005; Lee 2011; Park 2011). The following complications have been reported in the laparoscopy participants of studies in early ovarian cancer: umbilical hernias (Lee 2011), retroperitoneal haematoma (Ghezzi 2007), vascular injury (Colomer 2008; Ghezzi 2007; Park 2008b), lymphoceles (Lee 2011; Nezhat 2009), obturator nerve damage (Hua 2005), bowel injury or obstruction (Nezhat 2009; Park 2008a) and ureter injury (Park 2008b). Estimated blood loss (EBL) in all case-control (Table 2) and comparative cohort studies (Table 3) has been statistically significantly less than in the laparoscopy groups compared with laparotomy groups, with the exception of one study (Ghezzi 2007). In these studies, rates of blood transfusion in laparoscopy groups ranged from 0% to 15%, whereas transfusions were necessary in up to 30% (Park 2010) of women who underwent laparotomy.

There have been several reports of the occurrence of abdominal wall metastases following laparoscopy for ovarian cancer (Childers 1994; Gleeson 1993; Leminen 1999). However, in the studies of laparoscopy in stage I ovarian cancer that have reported this outcome, no port-site metastases had occurred by the time of reporting in Chi 2005 (20 women), Park 2008a (17 women), Park 2008b (19 women), Nezhat 2009 (36 women) and Lee 2011 (26 women). Port-site metastases may be technique-related and limited mostly to patients with advanced disease (Chi 2005; Nezhat 2009). In a study of laparoscopic cytoreductive surgery for advanced ovarian cancer and in which no port-site metastases occurred, the authors attributed their results to a surgical technique that employed endoscopic bags to retrieve intact specimens and a layered closure of the trocar site (Nezhat 2010). Lee 2011 and Chi 2005 have also reported employing this technique to prevent port-site metastases.

Other outcomes

Lee 2011 evaluated the relative cost of laparoscopy compared with open surgery in women with early ovarian cancer and found that laparoscopy resulted in higher costs due to the cost of disposable instrumentation and direct material/operating room costs, but the cost of hospital stay was higher in the laparotomy group because the stay was longer. Where bed costs are higher, this difference in cost might be eliminated, however the median lengths of hospital stay in the laparotomy groups in most of the studies reporting this outcome seem excessive with a range of up to 14.5 days (Table 2; Table 3). Literature on the quality of life for women undergoing laparoscopy compared with laparotomy is scant, however Lee 2011 reported significantly lower postoperative pain scores in the laparoscopy group.

Agreements and disagreements with other studies or reviews

A meta-analysis of eight RCTs comparing laparoscopic surgical staging with laparotomy for endometrial cancer has shown the laparoscopic approach to be safe, with statistically significantly fewer postoperative complications than laparotomy, and similar rates of intra-operative complications (Zullo 2012). It is possible that similar conclusions may, in time, be drawn about laparoscopy and laparotomy for stage I ovarian cancer, however the evidence for this is not currently in the literature.

Authors' conclusions

Implications for practice

Due to technological advancements in instrumentation and an increase in laparoscopic surgical expertise, the role of laparoscopy in gynaecological cancers is expanding, however there is still wide regional variation in the laparoscopic skills and competence of gynaecological-oncology surgeons. We did not find any good evidence to recommend laparoscopy for the routine management of women with stage I ovarian cancer.

Implications for research

Survival data for patients with gynaecological malignancies managed by laparoscopy are still lacking. A major barrier to conducting randomised controlled trials (RCTs) in early ovarian cancer is the anticipated difficulty in recruiting sufficient numbers of participants (Ghezzi 2009). Other difficulties include standardising the quality of the surgery and the skill of the surgeons. Subsequent results from such trials may only be applicable to expert laparoscopic oncology surgeons. However, we understand, from a personal communication, that the Korean Gynecologic Oncology Group (KGOG) is currently developing a protocol for a RCT comparing laparoscopy with laparotomy for early ovarian cancer. Two recently reported Korean cohort studies (Lee 2011; Park 2011) recruited 325 women between them within the same six-year period (2004 to 2010), suggesting that a multicentre RCT is feasible. Participating institutions should be subgrouped according to whether retroperitoneal lymph node dissection or lymph node assessment with sampling is performed routinely. Outcomes of RCTs should include overall and progression-free survival, complications (intra-operative and postoperative), the use of adjuvant chemotherapy, patient satisfaction, quality of life and costs. It would be helpful if costs are reported separately for the preoperative, intraoperative and postoperative periods.

Acknowledgements

We would like to thank Jo Morrison, Gail Quinn, Clare Jess and Tracey Bishop of the Cochrane Gynaecological Cancer Review Group team that is based at the Royal United Hospital, Bath, UK for their help, advice and support throughout the review process; Jane Hayes for performing the updated search; and the library staff at the Royal United Hospital, Bath, UK, who obtained many articles for us. We would also like to thank Lidia Medieros for acting as the Contact Author on the original review.

Data and analyses

Download statistical data

This review has no analyses.

Appendices

Appendix 1. CENTRAL search strategy

#1  MeSH descriptor Ovarian Neoplasms explode all trees
#2  ovar* near/5 (cancer* or tumor* or tumour* or neoplas* or carcinoma* or malignan* or adenocarcinoma*)
#3  (#1 OR #2)
#4  MeSH descriptor Laparoscopy explode all trees
#5  laparoscop* or celioscop* or peritoneoscop* or (endoscop* near/5 abdom*)
#6  MeSH descriptor Laparotomy, this term only
#7  laparotom* or (abdom* near/5 (surg* or incision))
#8  (#4 OR #5 OR #6 OR #7)
#9  (#3 AND #8)

Appendix 2. MEDLINE search strategy

1   exp Ovarian Neoplasms/
2   (ovar* adj5 (cancer* or tumor* or tumour* or neoplas* or carcinoma* or malignan* or adenocarcinoma*)).mp.
3   1 or 2
4   exp laparoscopy/
5   (laparoscop* or celioscop* or peritoneoscop* or (endoscop* adj5 abdom*)).mp.
6   Laparotomy/
7   (laparotom* or (abdom* adj5 (surg* or incision))).mp.
8   4 or 5 or 6 or 7
9   3 and 8
10 randomized controlled trial.pt.
11 controlled clinical trial.pt.
12 randomized.ab.
13 placebo.ab.
14 clinical trials as topic.sh.
15 randomly.ab.
16 trial.ti.
17 exp cohort studies/
18 exp case-control studies/
19 comparative study/
20 (cohort* or prospective* or retrospective* or control* or longitudinal or follow-up).mp.
21 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20
22 9 and 21

key: [mp = protocol supplementary concept, rare disease supplementary concept, title, original title, abstract, name of substance word, subject heading word, unique identifier] 

Appendix 3. EMBASE search strategy

1   exp ovary tumor/
2   (ovar* adj3 (cancer* or tumor* or tumour* or neoplas* or carcinoma* or malignan* or adenocarcinoma*)).ti,ab.
3   1 or 2
4   laparoscopy/
5   (laparoscop* or celioscop* or peritoneoscop* or (endoscop* adj3 abdom*)).ti,ab.
6   laparotomy/
7   (laparotom* or (abdom* adj3 (surg* or incision))).ti,ab.
8   4 or 5 or 6 or 7
9   exp controlled clinical trial/
10 cohort analysis/
11 exp case control study/
12 exp comparative study/
13 (randomized or randomly or trial* or cohort* or prospective* or retrospective* or control* or longitudinal or follow-up).ti,ab.
14 9 or 10 or 11 or 12 or 13
15 3 and 8 and 14

key: [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]

Appendix 4. LILACS search strategy

(MH:"ovarian neoplasms" or (ovar$ and (cancer$ or tumor$ or tumour$ or tumour$ or neoplas$ or carcinoma$ or malignan$ or adenocarcinom$))) and (MH:"laparoscopy" or laparoscop$ or MH:"laparotomy" or laparotom$)

Feedback

Definition of studies, 7 June 2015

Summary

Name: Alexander Melamed

Comment: The term "prospective case-control" is used in this study is confusing. Case-control studies by definition are retrospective. A case-control study is one where patients with and without an outcome of interest are identified, and the presence or absence of a putative causal exposure is investigated among these "cases" and "controls." Perhaps what the authors of this study are referring to is what would conventionally be called a cohort study. For a useful discussion of the confusion in regarding the terminology please see Marshall T. What is a case-control study? International Journal of Epidemiology 2004;33:612–617.

I agree with the conflict of interest statement below:

I certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter of my feedback.

Reply

The authors would like to thank Dr Melamed for taking the time to write, and we accept his suggestion. We have amended the term prospective case-control to prospective cohort studies throughout the review.

Contributors

Theresa A Lawrie on behalf of the author team.

What's new

Last assessed as up-to-date: 23 November 2012.

DateEventDescription
16 June 2015Feedback has been incorporatedFeedback received and responded too.
1 April 2015AmendedContact details updated.

History

Protocol first published: Issue 3, 2005
Review first published: Issue 4, 2008

DateEventDescription
11 February 2015AmendedContact details updated.
27 March 2014AmendedContact details updated.
20 December 2012AmendedAuthorship order amended
12 September 2012New citation required but conclusions have not changedEleven newly identified studies excluded, no studies included. Conclusions unchanged.
30 November 2011New search has been performedSearch updated.
9 November 2010AmendedAuthor contact details amended.
14 May 2008AmendedConverted to new review format.

Contributions of authors

Lidia Medieros (LM), Daniela Rosa (DR), Mary Bozetti (MB), Maria Ines Rosa (MR), Alice Zelmanowicz (AZ) and Airton Stein (AS) contributed to the writing of the protocol and the original review. LM, DR, MR, MB and Maria Edelweiss sifted the searches, selected studies and extracted data for the original review (which included NRSs). Anaelena Ethur (AE) and Roselaine Zanini (RZ) contributed to the protocol and methods section. Tess Lawrie (TL) sifted the updated search and wrote the first draft of the updated review. All authors approved the final version.

Declarations of interest

None.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • National Institute for Health Research (NIHR), UK.

    This review received methodological, statistical and editorial support as part of the 10/4001/12 NIHR Cochrane Programme Grant Scheme: Optimising care, diagnosis and treatment pathways to ensure cost effectiveness and best practice in gynaecological cancer: improving evidence for the NHS.

Differences between protocol and review

For the original review (Medeiros 2008), we included NRSs and evaluated the quality of three studies (Ghezzi 2007; Hua 2005; Tozzi 2004) according to the STROBE and NOS tools. For the updated review, these studies were excluded, but we tabulated and discussed the data with other NRSs.

Characteristics of studies

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    EOC = epithelial ovarian cancer

Amara 1996A case series of 11 women with stage Ia to IIIc ovarian cancer who underwent laparoscopy staging
Berman 2003Narrative review
Bristow 2000Narrative review
Canis 1994Narrative review
Canis 1997A case series of 10 cases of laparoscopy for low malignant potential tumour and 15 cases of cancer, but stage not described
Canis 2000A series of laparoscopy for 28 cases of cancer and borderline tumour; stage not described
Chapron 1998Narrative review
Chen 2010A case series of 43 women who underwent laparoscopic surgical staging for apparent early ovarian cancer (see Table 1)
Chi 2005A retrospective case-control study of 50 women who underwent laparoscopic or open surgical staging for apparent early ovarian cancer (see Table 2)
Childers 1995A prospective case series of second-look laparoscopy to evaluate both intraperitoneal cavity and retroperitoneal lymph nodes and laparoscopic surgical staging. 14 women underwent laparoscopic staging of apparent early ovarian cancer; metastatic disease was discovered in 8 of these women (see Table 1).
Childers 1996A series of 138 cases of laparoscopy for suspicious ovarian masses (not surgical staging). Malignancies were discovered in 19 women.
Colomer 2008A prospective case series of laparoscopic surgical staging of 20 women with apparent early ovarian cancer (18 EOCs and 2 dysgerminomas) (see Table 1)
Darai1998A retrospective case series of 25 women with borderline ovarian tumours
de Poncheville 2001Retrospective study of surgery without para-aortic lymphadenectomy in stage I ovarian cancer. Investigators advocate surgery without lymphadenectomy for women with early ovarian cancer.
Dottino 1999Retrospective case series of 94 laparoscopic lymphadenectomies for gynaecological malignancies, including 14 women with ovarian cancer
Ghezzi 2007A case-control study of 15 women who underwent laparoscopic staging for early ovarian cancer versus 19 historical controls who had open surgery (see Table 2)
Ghezzi 2009A case series of 26 women who underwent laparoscopic staging for early ovarian cancer (see Table 1)
Goff 2006Narrative review
Hua 2005A small, prospective, case-control study of 10 women who underwent laparoscopic staging for early ovarian cancer versus 11 women who underwent laparotomy (see Table 2). No adjustments were made for baseline case mix.
Kadar 1995Included other cancers (endometrial, cervical, ovarian)
Klindermann 1995A survey conducted on laparoscopy in Germany
Leblanc 2004Cohort with other types of cancer (fallopian tube carcinoma) and in patients that were inadequately staged at the time of initial surgery for invasive ovarian carcinoma
Leblanc 2006Narrative review
Lee 2011A retrospective cohort study of laparoscopy in 26 women with early ovarian cancer versus 87 women who underwent open surgery (see Table 3). No adjustments were made for baseline case mix. Women in the laparotomy group had significantly larger tumours and longer follow-up.
Lécuru 2004Retrospective study of 105 women who underwent surgery for stage I ovarian cancer. 14 underwent laparoscopy only, 13 had a laparoscopy converted to laparotomy and 78 had a laparotomy. Comprehensive staging was less frequent in the laparoscopy group.
Maiman 1991Members and candidate members of the Society of Gynecologic Oncologists responded to a survey concerning the "laparoscopy management of ovarian neoplasm subsequently found be malignant"
Malik 1998Unexpected ovarian tumours were discovered in 11/292 women who underwent laparoscopy to evaluate adnexal masses
Maneo 2004Criteria for exclusion: 62 patients had fertility-sparing after surgery
Manolitsas 2001Narrative review
Mehra 2004A prospective case series of 32 women with ovarian and other gynaecological cancers who underwent laparoscopic retroperitoneal para-aortic lymphadenectomy
Nezhat 1992Series of cases
Nezhat 2009A case series of 36 women with early ovarian cancer who underwent complete laparoscopic staging (see Table 1)
Nezhat 2010A case series of women with advanced ovarian cancer who underwent complete laparoscopic staging
Park 2008aA retrospective cohort study of 17 versus 19 women with early ovarian cancer who underwent laparoscopic staging and laparotomy respectively (see Table 3). No adjustments were made for baseline case mix.
Park 2008bA retrospective cohort study of 19 versus 33 women with early ovarian cancer who underwent laparoscopic staging and laparotomy respectively (see Table 3)
Park 2010An expansion of the earlier study (Park 2008b; see Table 3)
Park 2011An expansion of Park 2008b and Park 2010 retrospective studies (see Table 3)
Parker 1990This study only included women with benign ovarian cysts
Pomel 1995A series of women with stage I ovarian carcinoma who underwent a laparoscopic procedure to complete their staging
Querleu 2003A retrospective study of laparoscopic restaging of 30 women with borderline ovarian tumours
Querleu 2006aNarrative review
Querleu 2006bMany types of tumours (cervical, vaginal, endometrial and ovarian carcinoma)
Reich 1990A case report
Romagnolo 2006A series of cases where laparoscopy was performed for suspected borderline ovarian tumours, and included fertility-sparing procedures
Rouzier 2005Narrative review
Spirtos 2005A case series of laparoscopy for all stages of ovarian cancer and other types of cancers
Tozzi 2004A case series of laparoscopic staging for early ovarian cancer (see Table 1)
Tozzi 2005Narrative review
Tropé 2006Narrative review
Vaisbuch 2005Narrative review
Vergote 2003Narrative review
Vinatier 1996Narrative review
Volz 1997Narrative review
Wenzl 1996A questionnaire was sent to all 97 Departments of Gynaecology in Austria to determine the frequency of discovering a malignant ovarian mass when laparoscopy is used to manage an adnexal mass
Wu 2010A retrospective cohort study of laparoscopy versus laparotomy, however surgical staging was not always the aim of the initial laparoscopy

Ancillary