Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis

YGYNO-975943; No. of pages: 15; 4C: Gynecologic Oncology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno

Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis Sherif A.M. Shazly a,d, Mohammad H. Murad b, Sean C. Dowdy c, Bobbie S. Gostout c, Abimbola O. Famuyide a,⁎ a

Minimally Invasive Gynecologic Surgery, Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, MN, USA Division of Preventive Medicine, Mayo Clinic, Rochester, MN, USA Division of Gynecologic Surgery, Department of Obstetrics and Gynecology, Mayo Clinic, Rochester, MN, USA d Department of Obstetrics and Gynecology, Women Health Hospital, Assiut University, Egypt b c

H I G H L I G H T S • Meta-analysis of 26 non-randomized studies comparing robotic, laparoscopic, and abdominal radical hysterectomy for women with stage IA1–IIA cervical cancer. • Robotic radical hysterectomy was associated with less estimated blood loss, febrile morbidity, and shorter hospital stay compared to abdominal approach. • Robotic radical hysterectomy and laparoscopic radical hysterectomy appear equivalent in intraoperative and short-term postoperative outcomes.

a r t i c l e

i n f o

Article history: Received 26 March 2015 Received in revised form 1 June 2015 Accepted 5 June 2015 Available online xxxx Keywords: Robotic Radical hysterectomy Laparoscopic hysterectomy Early stage cervical cancer Metaanalysis

a b s t r a c t Objective. To compare intraoperative and short-term postoperative outcomes of robotic radical hysterectomy (RRH) to laparoscopic and open approaches in the treatment of early stage cervical cancer. Methods. A search of MEDLINE, EMBASE (using Ovid interface) and SCOPUS databases was conducted from database inception through February 15, 2014. We included studies comparing surgical approaches to radical hysterectomy (robotic vs. laparoscopic or abdominal, or both) in women with stages IA1–IIA cervical cancer. Intraoperative outcomes included estimated blood loss (EBL), operative time, number of pelvic lymph nodes harvested and intraoperative complications. Postoperative outcomes were hospital stay and surgical morbidity. The random effects model was used to pool weighted mean differences (WMDs) and odds ratios (OR). Results. Twenty six nonrandomized studies were included (10 RRH vs abdominal radical hysterectomy [ARH], 9 RRH vs laparoscopic radical hysterectomy [LRH] and 7 compared all 3 approaches) enrolling 4013 women (1013 RRH, 710 LRH and 2290 ARH). RRH was associated with less EBL (WMD = 384.3, 95% CI = 233.7, 534.8) and shorter hospital stay (WMD = 3.55, 95% CI = 2.10, 5.00) than ARH. RRH was also associated with lower odds of febrile morbidity (OR = 0.43, 95% CI = 0.20–0.89), blood transfusion (OR = 0.12, 95% CI 0.06, 0.25) and wound-related complications (OR = 0.31, 95% CI = 0.13, 0.73) vs. ARH. RRH was comparable to LRH in all intra- and postoperative outcomes. Conclusion. Current evidence suggests that RRH may be superior to ARH with lower EBL, shorter hospital stay, less febrile morbidity and wound-related complications. RRH and LRH appear equivalent in intraoperative and short-term postoperative outcomes and thus the choice of approach can be tailored to the choice of patient and surgeon. © 2015 Elsevier Inc. All rights reserved.

1. Introduction In spite of universally adopted screening programs, cervical cancer remains the third most common malignancy among women worldwide [1]. Early stage cervical cancer is defined as International Federation of

⁎ Corresponding author at: Department of Obstetrics and Gynecology, Mayo Clinic Rochester, MN 55905, USA. E-mail address: [email protected] (A.O. Famuyide).

Gynaecology and Obstetrics (FIGO) stages IA1–IIA, and represents the majority of patients at the time of presentation. Fortunately, the probability of survival at these stages is generally high provided the disease is appropriately evaluated and managed [2]. Radical hysterectomy is the standard surgical procedure for the treatment of early stage cervical cancer. Although effective, radical hysterectomy is associated with serious complications, of which urinary tract-related morbidities predominate [3]. Traditionally, laparotomy had been the only available approach for radical hysterectomy. The promising outcomes and evolving

http://dx.doi.org/10.1016/j.ygyno.2015.06.009 0090-8258/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

2

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Studies abstracted from EMBASE, MEDLINE and SCOPUS databases (n = 447)

Articles excluded (irrelevant or non-comparative studies) (n = 415)

Relevant articles (n = 32)

Articles excluded (not meeting eligibility criteria) (n = 6)

Eligible studies (n = 26)

RRH vs. ARH (n = 10)

RRH vs. LRH (n = 9)

RRH vs. ARH/LRH (n = 7)

Fig. 1. Flow chart of study selection.

experience in minimally invasive surgery in the late 80's allowed surgeons to evaluate its validity in complex oncologic surgeries [3]. Initially, the laparoscopic approach was restricted to pelvic lymph node dissection as an adjuvant step to vaginal radical hysterectomy [4]. Subsequently, the effectiveness, safety, and benefits of short hospital stay associated with laparoscopic oncologic procedures were reported [5]. With further development of instruments and skill, laparoscopic surgery has proved to be at least as effective as traditional hysterectomy in managing benign gynecologic conditions [6]. For endometrial cancer treatment, the laparoscopic approach yielded comparable outcomes to laparotomy without adversely impacting survival [3,7]. In spite of these benefits, the adoption of a laparoscopic approach for gynecological malignancies was tempered by the longer operative time compared to laparotomy [7]. In 2008, the first case series of robotic radical hysterectomy was published, demonstrating potential advantages of the new technique over traditional laparoscopy, including improved visualization, enhanced articulation and ease of dissection, and lower rates of complications [8]. Since then, robotic surgery has been widely disseminated in gynecologic oncology and many studies were conducted to test and validate this approach for indications including cervical cancer [9–37]. Currently, four available approaches of radical hysterectomy (open or abdominal, vaginal, laparoscopic, robotic) are variably implemented with conflicting evidence. Evidence to support the use of robotic surgery over other approaches in early cervical cancer is still lacking. The aim of this systematic review and meta-analysis is to compare intraoperative and postoperative complications between robotic radical hysterectomy and other surgical methods in the treatment of early stage cervical cancer. Radical vaginal hysterectomy has been adopted only in certain specialized centers, and this was not evaluated in this meta-analysis.

2. Methods 2.1. Literature search A search was conducted for studies that compared robotic radical hysterectomy with open, laparoscopic approaches or both in treating early cervical cancer using three online databases: MEDLINE, EMBASE (with online Ovid interface) and SCOPUS. This was done in collaboration with an expert reference librarian and included studies from database inception to February 15, 2014. We used the terms “robotic OR robot OR da Vinci” AND “radical hysterectomy OR early cervical cancer OR stage IB, stage IB1, stage IB2, stage IIA, stage IIA1, stage IIA2, stage IIA cervical cancer.” Search was limited to “comparative research” using online filtering options. Additional search on references from review articles and studies was conducted and results were added to the initial search results. The detailed search strategy is appended (Appendix A). 2.2. Eligibility criteria and study selection The primary population of interest was women diagnosed with cervical cancer (IA1 to IIA) that were treated with radical hysterectomy with or without lymphadenectomy. Studies were selected by two independent reviewers who performed an initial screening of abstracts and a subsequent detailed review of the full texts of included articles. Prospective and retrospective studies comparing robotic surgery to laparoscopic, open, or both approaches were eligible for inclusion. Studies evaluating laparoscopically assisted vaginal hysterectomy, radical vaginal hysterectomy, radical trachelectomy for fertility preservation, recurrent cervical cancer cases, or single-armed studies were excluded. No studies were excluded based on sample size. Outcomes of interest were intraoperative and postoperative outcomes measures. (I) Intraoperative outcomes included estimated

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Author and year

Country

Type of the study

Data collection time frame

Sample size

Eligibility criteria

Histological types

Intervention

Corrado et al. (2015)

Rome, Italy

Retrospective

August 2010–December2012 for both arms (both approaches were available)

RRH (n = 30), mLRH (n = 30)

SCC, AC, ASC and clear cell carcinoma

A class B2 or C1 radical hysterectomy (according to the Querleu and Morrow classification)

Chen et al. (2014)

Taipei, Taiwan

Retrospective

2005–2013 (ARH, LRH), 2011–2013 (RRH)

RRH (n = 24), LRH (n = 32), ARH (n = 44)

SCC, AC. Other non-specified types represented 8.3% (RRH), 3.1% (LRH) and 2.3% (ARH) of cases

Radical hysterectomy,bilateral salpingo-oophorectomy, and bilateral pelvic lymphnode dissection

Kim et al. (2014)

Seoul, Korea

Retrospective

RRH(n = 23), LRH (n = 69)

Yim et al. (2014)

Seoul, Korea

Retrospective

January 2008–May 2013 (RRH). RRH cohort was matched to LRH cases conducted during the same period December 2009–May 2013 (both approaches were available during the same period)

Stage IA1 to IIA1 cervical cancer, no evidence of LN or distant metastasis (clinical or radiological)Women with a uterusN12 weeks, with pregnancy and who were unfit or had contraindications to pneumoperitoneum were excluded. Women with previous abdominal surgery were not contraindicated. Women with cervical cancer stage not greater than IIB and who did not receive preoperative brachytherapy or chemotherapy were included. Women chose their approach except women with previous pelvic surgeries (≥2 times) were offered RRH or ARH. Stage IA1 cervical cancer with LVSI and stage IIA

RRH(n = 60), LRH (n = 42)

New cases of cervical cancer with stage IIA2 or less disease, Gynecologic Oncologic Group performance score ≤ 1 and adequate financial capacity to afford the expenses of robotic surgery were included. Women with uterine size N 16 weeks or with previous 3 or more open abdominal surgeries were excluded.

SCC, AC, ASC. Other non-specified cases represented 4.4% and 5.8% of RRH and LRH groups respectively. SCC, AC and small cell type

Chong et al. (2013)

Daegu, Korea

Retrospective

January 2008–March 2012, November 2003–December 2006 for historical cohort

RRH (n = 50) and LRH (n = 50)

Stage IA2 to IIA2 cervical cancer

SCC, AC, ASC

Desille-Gbaguidi Tours, France et al. (2013)

Retrospective

2008–December 2011 for both arms (the study began from the purchase year of Da Vinci® system) October 2004 to December 2011 (RRH started in October 2009)

LRH (n = 27), RRH (n = 30)

Endometrial cancer treated with hysterectomy or early cervical cancer treated with radical hysterectomya

Not specified

Type II or III RH with pelvic LA (Piver classification). Para-aortic LNs were removed selectively according to pelvic LN results. Radical or modified hysterectomy with pelvic lymph node dissection ± paraaortic lymph node sampling (Type B and C Querleu and Morrow classification). Only 2 cases underwent nerve sparing. Type III nerve-sparing RH with pelvic LA (preservation of hypogastric nerve and inferior hypogastric plexus) Hysterectomy and RH with or without LA

ARH (n = 20), LAVHb (n = 20), RRH (n = 20) ARH (n = 51), RRH (n = 180) RRH (n = 25) and LRH (n = 25)

Early cervical cancer treated with radical hysterectomy

SCC and AC

RH and pelvic LA

Early stage cervical cancer or stage II endometrial cancer. Additional procedures including aortic LN or omentectomy and LN metastasis were excluded. Stage IB2–IIB cervicalcarcinoma with clinical responseafter 3 courses of chemotherapy

Not specified

SCC, AC, clear cell carcinoma

RH with pelvic LA (women requiring aortic LA were excluded) Class C1 (Querleu and Morrow) RH with LA, para-aortic LNs were removed if pelvic LNs were found positive.

ARH (n =

Early cervical cancer treated with radical hysterectomy

Not specified

RH with or without LA

Pilka et al. (2013) (in Czech)

Olomouc, Czech

Retrospective

Reynisson and Persson (2013) Vissa et al. (2013)

Lund, Sweden

Retrospective

Rome, Italy

Retrospective

Wright et al. (2012)

New York, USA

Retrospective

January 2001–December 2005 (ARH), December 2005–February 2012 (RRH) August 2010–December 2012 (RRH), data time frame for LRH was not specified (historic cohort prior to implementation of RRH) January 2006–March 2010 for the 3 arms. Data was extracted from a commercial data source “the Perspective database”

1610), LRH (n = 217), RRH (n = 67)

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Table 1 Characteristics, description of patients and interventions of included studies.

(continued on next page) 3

4

Author and year

Country

Type of the study

Gortchev et al. (2012)

Pleven, Bulgaria

Retrospective

Sert and albert (2011)

Oslo, Norway

Retrospective

Data collection time frame

Sample size

Eligibility criteria

Histological types

Intervention

(Premier, Charlotte, North Carolina). January 2006–December 2007 (LRH), May 2008–May 2010 (RRH), January 2006–May 2010 (ARH) November 2005–November 2009 (RRH), April 2004–August 2005 (LRH), March 2000–March 2004 (ARH)

RRH (n = 73), LAVHb (n = 46), ARH (n = 175) RRH (n = 35), LRH (n = 7), ARH (n

Stage 1B1 cervical cancer

SCC, AC

RH and pelvic LA (Type III hysterectomy)

SCC, AC, ASC

RH and pelvic LA—stage IA1 with LVSI and stage IA2 had type II RH. Stage IB1 had type III RH.

January 2007–November 2010 for the 3 arms (the 3 approaches were available within this time frame) January 2003–May 2010 for both arms (time frame for each arm was not specified)

RRH (n = 34), LRH (n = 31), ARH (n = 30) LRH (n = 76), RRH (n = 23)

Stage IA1 to IB1 cervical carcinoma. Tumor N 4 cm, uterine size N 12 cm, history of radiotherapy, pregnancy, histology other than AC or SCC or ASC, metastatic disease, pelvic or aortic LNs N 2 cm, histologically positiveLNs were excluded. Stage IA1 to IB2/IIA cervical cancer

SCC, AC, ASC

Stage IA1 with LVSI to IIA cervical cancer with no evidence of nodal involvement. Exclusion criteria were: pregnancy, metastasis, ovarian lesions, contraindications to anesthesia.

SCC, AC, ASC

June 2006–February 2009 (RRH), January 2002–February 2009 (ARH)

RRH (n = 32), ARH (n = 32)

Stage IA2 to IIB cervical cancer, no imaging evidence of nodal involvement, capacity to pay for surgery

SCC, AC

January 2007–June 2008 (RRH), the preceding 30 cases represented the ARH cohort (time frame was not specified) June 2005–July 2008 (RRH), 1995 and 2007 (ARH)

RRH (n = 15), ARH (n = 30)

Stages IA2–IB2 cervical cancer

SCC, AC

RH and pelvic LA in all patients (except one case in the robotic group who had sentinel LN biopsy “refused LA”) Stage IA2 and IB1 with primary lesion ≤ 2 cm were treated with Type II RH. Stage IB1 or IIA with tumors N 2 cm were treated with type III RH. LA was done including para-aortic LA if indicated. Modified RH with pelvic LA (type II hysterectomy) for IA2 and RH with pelvic LA (type III) for the remaining (most cases) Type III RH with pelvic LA

RRH (n = 63), ARH (n = 64)

Stage IA1–IIB cervical cancer. Endometrial carcinoma and metastatic diseases were excluded.

Type III RH with pelvic LA

Stage IA1 with LVSI on cone biopsy to IIA cervical cancer. Candidates for radical trachelectomy and women with positive sentinel LNs were excluded. Stages IA2 to IIA cervical cancer, newly diagnosed, no contraindication to surgery, no clinical or imaging evidence of nodal/parametrial involvement (stage IB2 received 3 courses of

SCC, AC. Others (6% and 16% in RRH and ARH respectively) were ASC or clear cell carcinoma. SCC and non-SCC (not specified)

SCC, AC, ASC. Others represented 10% of each armc

Modified RH (type B1 by Querleu and Morrow) for cervical cancer b 2 cm and type C1 (type III) RH for cervical

= 26)

Soliman et al. (2011)

Texas, USA

Retrospective

Tinelli et al. (2011)

Avellino, Italy and New York, USA

Retrospective

Nam et al. (2010)

Seoul, Korea

Geisler et al. (2010)

Toledo, OH, USA

Retrospective (matched case–control study) Retrospective

Cantrell et al. (2010)

North Carolina, USA

Retrospective

Halliday et al. (2010)

Montreal, Canada

Retrospective

January 2008–December 2009 (RRH), March 2003–December 2007 (ARH)

RRH (n = 16), ARH (n = 24)

Maggioni et al. (2009)

Milan, Italy

Retrospective

December 2007–March 2009 (RRH). Time frame for ARH was not specified (described as cases that were consecutive

RRH (n = 40), ARH (n = 40)

Type II and III RH with pelvic LA

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Table 1 (continued)

Miami, USA

Prospective compared to historical cohort

Lowe et al. (2009)

Chicago, USA

Retrospective

Boggess et al. (2008)

Chapel Hill, NC, USA

Retrospective

Magrina et al. (2008)

Arizona, USA

Retrospective

Nezhat et al. (2008)

New York, USA

Ko et al. (2008)

Boston, USA

Sert and Abeler (2007)

Oslo, Norway Pilot case control study

RRH (n = 51), ARH (n = 49)

neoadjuvant chemotherapy) Stages IA2 to IB2 cervical cancer

SCC, AC, ASC

Stage IB1 cervical cancer. BMI N 35, multiple prior surgeries, Extensive pelviabdominal adhesions or uterine size N 12 weeks were excluded from robotic surgery. Stage IA1 to IIA cervical cancerd

SCC, AC

cancer N 2 cm or unfavorable histologic type RH with pelvic and para-aortic LA

Type III RH

June 2005–November 2007 (RRH). Time frame for ARH was not specified (described as cases that were consecutive before implementation of robotic surgery). April 2003–September 2006 (RRH), November 1993–August 2006 (LRH and ARH)

RRH (n = 27), LRH (n = 31), ARH (n = 35)

Stage IA2–IB2 cervical cancer

SCC, AC, ASC

Retrospective

August 2000–June 2006 (LRH), April 2006–January 2008 (RRH)

RRH (n = 13), LRH (n = 30)

SCC, AC, ASC (1 case of glassy cell carcinoma in the LRH group)

Retrospective

August 2006–August 2007 (RRH), August 2004–August 2006 (ARH) November 2005–March 2006 (RRH), April 2004–August 2005 (LRH)

RRH (n = 16), ARH (n = 32) RRH (n = 7), LRH (n = 7)

Stages IA1 to IIA cervical cancer. Exclusion criteria were cervical lesion N 4 cm, uterine size N 12 cm, pregnancy, renal/hepatic dysfunctions or deficient bone marrow (no exclusion based on BMI or previous surgeries). Stages IA1 to IIA cervical cancer

SCC, AC

Type III RH

Stages IA1 to IB1 cervical cancer. Cases converted to laparotomy were excluded

SCC, AC

Stages IA1 and IA2 were treated with type II RH and IB1 was treated with type III RH with bilateral pelvic LA.

Type III RH with pelvic LA

Cervical cancer ≤ 2 cm with no LVSI was treated with modified RH, others were treated with RH, endometrial cancer with suspected or known cervical invasion was treated with RH, grade 3 tumor and/or outer third invasion only were treated with modified RH. RH with bilateral pelvic LA with or without para-aortic LA

ARH = Abdominal radical hysterectomy, LRH = Laparoscopic radical hysterectomy, mLRH = Mini-laparoscopic radical hysterectomy, RRH = Robotic radical hysterectomy, LAVH = Laparoscopic assisted vaginal hysterectomy, AC =

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Estape et al. (2009)

before implementation of robotic surgery). RRH (n = 32), August 2006–April 2008 ARH (RRH), July 2004–July 2006 (LRH), May 2002–July 2006 (ARH) (n = 14), LRH (n = 17) RRH (n = 7), July 2007–July 2008 (for both ARH arms). Both approaches were (n = 7) available during this period.

Adenocarcinoma, SCC = Squamous cell carcinoma, ASC = Adenosquamous carcinoma, LN = Lymph nodes, LVSI = Lymphovascular space involvement, BMI = Body mass index, RH = Radical hysterectomy, LA = Lymphadenectomy. a Only women with cervical cancer were included. b LAVH arm was excluded from analysis. c Other histologic types in RRH group: embryonal rhabdomyosarcoma, small cell carcinoma, villoglandular papillary adenocarcinoma. Other histologic types in ARH group: embryonal rhabdomyosarcoma, clear cell adenocarcinoma, neuroendocrine tumor, small cell carcinoma. d Four cases in the robotic group were beyond these stages, but not specified.

5

6

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Table 2 Methods of assessment of EBL and operative time and duration of follow-up after surgery. Author and year

Assessment of EBL

Assessment of operative time

Follow-up duration

Corrado et al. (2015)

Difference in the total amounts of suctioned and irrigation fluids The total volume of suctioned fluids

The beginning of skin incision to completing skin closure From skin incision and until all surgical staging procedures were completed From skin incision to skin closure (skin to skin) Time from the first skin incision to closure of the last port site Skin to skin

21.9 months(mLRH) and 29.5 months (RRH)a 13.9 months (RRH), 34.6 months (LRH), 37.1 months (ARH)b

Chen et al. (2014)

Kim et al. (2014) Yim et al. (2014)

Total amount of fluid in suction apparatus minus the volume of irrigation fluid N/A

Suctioned fluids plus weighed gauze with irrigation fluid subtracted Desille-Gbaguidi et al. (2013) N/A Pilka et al. (2013) (in Czech) N/A Chong et al. (2013)

47.1 months (RRH) and 58.2 months (LRH)a 25 months (RRH) and 19.5 months (LRH)a N/A

Reynisson and Persson (2013) Vissa et al. (2013) Wright et al. (2012) Gortchev et al. (2012) Sert and albert (2011)

N/A

N/A N/A The time between the onset of incision and the N/A last stitch on skin closure Skin to skin N/A

Suctioned fluid subtracted by irrigation fluids N/A N/A N/A

Skin to skin N/A Skin to skin Skin to skin

Soliman et al. (2011) Tinelli et al. (2011)

N/A The total amount of fluid in suction apparatus subtracted by the volume of irrigation fluid N/A N/A N/A N/A Difference between suctioned and irrigated fluid N/A

Skin to skin N/A

N/A N/A N/A 36 months (RRH), 56.4 months (LRH), 70 months (ARH)b N/A 24.5 months (RRH), 46.5 months (LRH)a

Skin to skin N/A Skin to skin N/A Skin to skin

15.3 months (RRH), 40.6 months (ARH)b N/A 12.2 months (RRH), 28 months (ARH)a N/A N/A

Insertion of Foley's catheter to closure of last trocar site Skin to skin Skin to skin Skin to skin

9.46 months (RRH), 31.39 months (LRH), 46.08 months (ARH) N/A N/A N/A

Nam et al. (2010) Geisler et al. (2010) Cantrell et al. (2010) Halliday et al. (2010) Maggioni et al. (2009) Estape et al. (2009) Lowe et al. (2009) Boggess et al. (2008) Magrina et al. (2008) Nezhat et al. (2008) Ko et al. (2008) Sert and Abeler (2007) a b

Determined by anesthetists (not detailed) N/A Difference between suctioned and irrigated fluid Sum of suctioned fluid and weighed sponges Skin to skin Determined by primary surgeons (not detailed) N/A N/A N/A

12 months (RRH), 29 months (LRH)b N/A N/A

Values in median. Values in mean.

blood loss (EBL) as approximately measured by irrigation-suction fluid difference, blood transfusion, operative time (skin-to-skin time), number of pelvic lymph nodes harvested and intraoperative complications. (II) Postoperative outcomes included length of hospital stay, individual and aggregate complications. The term ‘aggregate’ was defined to include both intra- and postoperative complications. Because there were many reported individual complications, they were categorized to facilitate analysis into wound related complications (infection, dehiscence and hernia), lymphovascular complications (lymphocyst and infected lymphocele, lymphedema, leg edema or vulvar edema), cuffrelated complications (vaginal evisceration, dehiscence and separation), febrile morbidity, thrombotic events (deep venous thrombosis or pulmonary emboli), abscess and infection (pelvic abscess or pelvic cellulitis), intestinal obstruction (including functional obstruction “ileus”), uretero-renal complications (ureteral stricture, ureteral fistula, hydroureter and hydronephrosis), urinary retention, and medical complications including cardiopulmonary events related to surgery and anesthesia. Recurrence and overall survival data were not evaluable because of missing data in the included studies. 2.3. Data abstraction Data abstraction was performed using a standardized form; these included study characteristics such as year, type of study and number of recruited patients, patient characteristics and selection frame including eligibility criteria, clinical staging and histopathology, surgical

intervention, comparison arms, intraoperative events, and postoperative complications. The methods of estimation of EBL and operative time were identified in each study. In addition, data was extracted from text, tables or figures. Minor discrepancies in data were adjudicated by consensus among reviewers. The risk of bias was assessed using the Newcastle-Ottawa Scale (NOS) [38].

2.4. Data analysis The differences in EBL, operative time, number of lymph nodes harvested and length of hospital stay (in days) were analyzed using mean, standard deviation (SD) and sample size for each group. The effect size was expressed as a weighted mean difference (WMD) with 95% confidence interval (CI) and two sided P values. Hozo's formula was used to convert data expressed as a median and range [39]. Binary outcomes (including individual and aggregate complications) were expressed as odds ratios (OR) and 95% CI. A random-effect model was used to pool outcomes across studies due to anticipated heterogeneity [40]. Heterogeneity was evaluated using I squared statistic and Cochrane Q test. An I squared value over 50% or Q test p value b 0.10 was considered to be consistent with substantial heterogeneity [41]. Statistical analysis was performed using Review Manager (RevMan) Version 5.3 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) [42].

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

7

I. EBL (in weighted mean difference) I.A. RRH versus ARH

I.B RRH versus LRH

II. Operative time (in weighted mean difference) II.A. RRH versus ARH

Fig. 2. Forest plots of effect estimates of intraoperative outcomes (estimated blood loss “EBL”, operative time, pelvic lymph nodes count, and intraoperative complications) between robotic (RRH), laparoscopic (LRH) and abdominal radical hysterectomy (ARH).

3. Results Four hundred and forty seven articles were initially retrieved from database search for comparative studies. We excluded 3 studies that evaluated different procedures or disease stages [43–45], 1 that included some cases of radical trachelectomy in the treatment cohort [9], 1 which was a protocol for a randomized controlled trial (RCT) [27], and 1 commentary [26]. Overall, 26 comparative studies that met the inclusion criteria were available for pooled analysis [10–25,28,29,31–34,36, 37,46,47]. Only 1 study included a mixed cohort of cervical and endometrial cancer cases. The study selection flowchart is depicted in Fig. 1.

The characteristics of these studies are shown in Table 1. All included studies were retrospective cohort design. Ten studies were conducted in the United States of America (USA) [10,11,16,17,21,22,24,33,36,47], 4 studies in Korea [13,20,25,37], 3 studies in Italy [14,23,46], 2 studies in Norway [31,32], 1 study in Sweden [29], 1 study in France [15], 1 study in Bulgaria [18], 1 study in Canada [19], 1 study in Czech [28], 1 study in Taiwan [12] and 1 study in both USA and Italy [34]. Overall, the studies had adequate methodological quality and low risk of bias, particularly in terms of cohort selection and outcome ascertainment. However, the comparability of cases and controls was less consistent (Appendix B). In all, 4013 women were included in the pooled analysis;

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

8

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

II.B. RRH versus LRH

III. Lymph nodes count (in weighted mean difference) III.A. RRH versus ARH

III.B RRH versus LRH

Fig. 2 (continued).

1013 women were treated by RRH, 710 by LRH and 2290 by ARH. The studies were conducted and published between 2007 and 2015. As shown in Table 1, RRH was compared to ARH in 10 of these studies [10,11,17,19,21–23,25,28,29], 9 studies compared RRH against LRH [13–15,20,31,34,37,46,47] and 7 studies compared all 3 approaches [12,16,18,24,32,33,36]. All women were treated with RH with or without LA; RH was either type II or III according to Piver classification [48] or type B1 and C1 using Querleu and Morrow classification [49]. Nerve-sparing technique was reported as the primary technique in one study [13] and as an occasional technique in another study [37].

Four studies included women with stage IIB [11,12,25,35]. However, they did not address a different management plan for these women and radical hysterectomy was conducted in all cases. Methods of assessment of outcomes and follow-up durations are summarized in Table 2. 3.1. RRH versus ARH Fig. 2 shows the forest plots of intraoperative events including EBL, operative time, lymph node harvested and intraoperative complications

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

9

IV. Intraoperative complications IV.A RRH versus ARH

IV.B RRH versus LRH

Fig. 2 (continued).

between RRH and ARH. RRH was associated with significantly lower EBL (ml) in comparison to ARH (WMD = 384.3, 95% CI = 233.7, 534.8). RRH had longer operative time (in minutes) than ARH although this difference was not statistically significant (WMD = 28.8, 95% CI = −2.15, 59.74). There was no significant difference in either the average number of pelvic lymph nodes harvested (WMD = −0.69, 95% CI = –3.37, 1.99) or in the

incidence of overall intraoperative complications (OR = 1.07, 95% CI = 0.55, 2.06) between RRH and ARH. Individually, both RRH and ARH were also comparable in the pooled odds of bladder injury (8/174 vs. 40/1690, OR = 1.03, 95% CI = 0.39, 2.77), bowel injury (2/139 vs. 8/ 1682, OR = 1.17, 95% CI = 0.23, 5.88) and ureter injury (4/141 vs. 33/ 1716, OR = 1.36, 95% CI = 0.44, 4.21).

Table 3 Meta-analysis estimates of individual complications. RRH versus ARH

Pelvic infection and abscess formation Cardiopulmonary complications Cuff related complications Febrile morbidity Intestinal obstruction Lymphovascular complications Thrombotic events Blood transfusion Readmission Reoperation Uretero-renal complications Urinary retention Wound related complications

RRH versus LRH

RRH

ARH

OR (95% CI)

P value

RRH

LRH

OR (95% CI)

P value

12/369 13/330 9/227 12/221 5/432 11/238 4/207 9/223 7/139 4/177 3/227 14/122 3/314

30/2026 129/1851 3/204 40/1750 14/1977 10/329 16/1867 310/1796 6/145 15/1721 29/1870 10/135 30/1903

1.58 [0.58, 4.35] 0.84 [0.34, 2.07] 1.79 [0.54, 5.93] 0.43 [0.20, 0.89] 0.71 [0.29, 1.71] 1.00 [0.41, 2.46] 2.27 [0.76, 6.82] 0.12 [0.06,0.25] 1.11 [0.36, 3.36] 0.62 [0.21, 1.82] 0.80 [0.24, 2.68] 1.35 [0.20, 8.85] 0.31 [0.13, 0.73]

0.37 0.71 0.34 0.02 0.44 1.00 0.14 b0.0001 0.86 0.38 0.72 0.76 0.008

3/123 5/160 6/224 6/323 4/119 13/293 3/149 12/349 3/59 – 13/287 6/112 2203

5/259 4/289 9/234 19/487 4/129 20/367 5/278 31/468 5/100 – 13/450 8/98 /8347

1.09 [0.23, 5.03] 1.83 [0.45, 7.47] 0.79 [0.19, 3.21] 0.47 [0.20, 1.09] 1.14 [0.28, 4.61] 0.80 [0.36, 1.74] 1.10 [0.28, 4.25] 0.61 [0.30, 1.26] 1.26 [0.31, 5.16] – 0.94 [0.42, 2.12] 0.62 [0.21, 1.86] 0.49 [0.13, 1.80]

0.92 0.40 0.74 0.08 0.86 0.57 0.89 0.18 0.74 – 0.89 0.39 0.28

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

10

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

I. Hospital stay I.A. RRH versus ARH

I.B. RRH versus LRH

Fig. 3. Forest plots of effect estimates of postoperative outcomes (hospital stay and complications) between robotic (RRH), laparoscopic (LRH) and abdominal radical hysterectomy (ARH).

Short-term postoperative outcomes are illustrated in Fig. 2. Hospital stay was significantly shorter in women treated with RRH as compared to ARH (WMD = − 3.55, 95% CI = − 5.00, − 2.10). Although it was not statistically significant, the pooled prevalence of aggregate complications was lower among women treated with RRH and the effect estimate was close to significance (OR = 0.56, 95% CI = 0.30, 1.03). In an attempt to determine the source of difference, analysis of individual postoperative complications was conducted (Table 3); both RRH and ARH were comparable in the pooled odds of pelvic infection and abscess formation, cardiopulmonary complications, cuff-related complications, intestinal obstruction/ileus, lymphovascular complications, thrombotic events, uretero-renal complications, urinary retention, readmission and reoperation. However, RRH was associated with significantly lower odds of febrile morbidity (OR = 0.43, 95% CI = 0.20–0.89), blood transfusion (OR = 0.12, 95% CI 0.06, 0.25) and wound-related complications (OR = 0.31, 95% CI = 0.13, 0.73) compared to ARH. 3.2. RRH versus LRH Although RRH was associated with less EBL than LRH, the mean difference was of borderline significance (WMD = − 44.31, 95% CI = − 89.19, 0.56). There was no significant difference in operative time or intraoperative complications. There was no significant difference in the odds of individual intraoperative complications including bladder injury (11/251 vs. 18/429, OR = 0.94, 95% CI = 0.42, 2.12), bowel injury (1/154 vs. 5/283, OR = 0.62, 95% CI = 0.10, 3.77) or

ureteral injury (5/124 vs. 3/310, OR = 2.59, 95% CI = 0.51, 13.10). The average number of lymph nodes harvested with RRH was greater than LRH although it was not statistically significant (Fig. 2). Postoperative parameters were also comparable between RRH and LRH. There were no significant differences in hospital stay (days) or aggregate complications (Fig. 3). Individual postoperative complications did not differ between both pooled cohorts. Although febrile morbidity was more recognized in the LRH cohort, the difference was not statistically significant (Table 3). In both comparisons (RRH vs ARH and RRH vs LRH), heterogeneity was substantial in the analyses of EBL, operative time, lymph nodes harvested, hospital length of stay and aggregate complication, making these pooled estimates less reliable. In contrast, comparative estimates of intraoperative complications were more reliable because of lack of statistically significant heterogeneity. 4. Discussion To our knowledge, this is the first systematic review and metaanalysis comparing RRH to alternative approaches in women with early stage cervical cancer. The results demonstrate that RRH was associated with lower EBL, shorter hospital stay, but longer operative time than ARH. In comparing RRH with ARH, the trend in these three measures shows a level of consistency across international centers that is uncommon in meta-analysis of surgical techniques. The difference in operative time was close to significance and may be explained in part by the learning curve that is associated with new technologies or approaches. Reynisson and Persson analyzed

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

11

II. Aggregate complications II.A. RRH versus ARH

II.B. RRH versus LRH

Fig. 3 (continued).

RRH in 6 sequential groups; surgical time was initially greater in RRH but was comparable or shorter after 60 cases of RRH were performed [29]. RRH was also associated with fewer postoperative complications than ARH, but this observation was of borderline statistical significance. Existing data are not adequate to assess disease survival or recurrence. Previous studies have evaluated the utility of robotic laparoscopic approach in gynecologic surgery. A meta-analysis of 22 studies published in 2010 analyzed data from a variety of gynecologic robotic surgeries including hysterectomy for benign and malignant conditions, myomectomy, sacrocolpopexy, fallopian tube reanastomosis, and adnexectomy. Robotic surgery was superior to laparotomy in terms of hospital stay and EBL and lower EBL when compared to laparoscopic surgery [50]. More specifically for gynecological malignancy, a systematic review was conducted to analyze comparative studies between robotic, laparoscopic and open approaches in women with endometrial cancer. Women who were eligible for the 8 studies included in the pooled

analysis (robotic = 589, laparoscopic = 396 and laparotomy = 606) underwent total hysterectomy, bilateral salpingo-oophorectomy and LA and were evaluated for perioperative outcomes. For women undergoing robotic surgery, estimated blood loss was lower and operative time longer, in comparison to women who had laparoscopic or abdominal approaches [51]. These findings were validated in a more recent meta-analysis published in 2013 with shorter length of stay and less frequent need for blood transfusion in the robotic cohort compared to laparoscopic and open approaches. Furthermore, the robotic approach resulted in fewer complications for women who underwent simple total hysterectomy with nodal staging [52]. Another meta-analysis of 22 studies that evaluated surgical treatment in women with endometrial cancer concluded that robotic surgery, although more efficient, had higher surgical complication rates than laparoscopic surgery [51]. Although sufficient evidence about the pros and cons of robotic simple hysterectomy is available in the literature, we considered radical surgeries to address whether the trend

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

12

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

of these differences might change as the procedure gets more complicated. Our meta-analysis suggests an advantage for robotic radical hysterectomy over the abdominal approach, but comparisons to laparoscopic radical hysterectomy showed equivalence. Thus the decision to choose one option over the other may be guided by availability of tools, surgical expertise and, in particular, by total costs of care. However, the literature on cost data is conflicting. Halliday reported significantly higher costs for ARH compared to RRH, not including amortization costs for the robot [19]. In contrast, Reynisson and Persson reported higher costs of RRH in comparison to ARH for the first 90 cases of RRH, alluding to the incremental costs of robotic surgery early in the learning curve. Thereafter, there was no significant cost difference between the 2 approaches [29]. Wright et al. compared the total costs of the 3 approaches concurrently; ARH had the lowest costs followed by RRH and LRH [36]. Finally, Desille-Gbaguidi showed that RRH was more expensive than LRH [15]. These conflicting results may be explained in part by differences in the methodology of calculating costs of care and by the different cost infrastructures in the countries of origin of these studies. There are limitations to this meta-analysis. Included studies are primarily nonrandomized and retrospective. Furthermore, pooled analysis was associated with significant heterogeneity for some measures which lowers confidence in the estimate, and publication bias could not be reliably evaluated because of the wide study heterogeneity [52]. Nevertheless, this is the first robust systematic

review and meta-analysis that addresses the utility of robotic approach in treating early stage cervical cancer with estimates that will be helpful in patient counseling and practice decision-making for equipment procurement.

5. Conclusion RRH is superior to ARH with lower EBL, shorter hospital stay, and less frequent wound-related and febrile morbidities. RRH is comparable to LRH in regard to short-term intraoperative and postoperative outcomes. Additional studies are necessary to evaluate longer term outcomes including survival and total costs of care.

Conflict of interest The authors state no conflict of interest.

Transparency document The Transparency document associated with this article can be found, in the online version.

Appendix A. Search strategy

Ovid MEDLINE(R) in-process & other non-indexed citations and Ovid MEDLINE(R) 1946 to present #

Searches

1 2 3 4 5

exp Hysterectomy/ exp Uterine Neoplasms/ exp Uterine Neoplasms/su or 1 Robotics/ 3 or hysterectom*.mp. [mp = title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 6 (4 or robot*.mp.) and 5 [mp = title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 7 (cervical adj3 (cancer* or malign* or carcinoma*)).mp. [mp = title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 8 7 and hysterectom*.mp. and (robot* or “da vinci”).mp. [mp = title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier] 9 6 or 8 10 9 and radical*.mp. [mp = title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

Results

Search type

24,741 105,472 35,528 14,549 47,265

Advanced Advanced Advanced Advanced Advanced

656 Advanced 40,626 Advanced 114 Advanced 656 Advanced 154

Embase 1988 to 2015 week 07 #

Searches

1 2 3 4 5

exp Hysterectomy/ exp Uterine Neoplasms/ exp Uterine Neoplasms/su or 1 Robotics/ 3 or hysterectom*.mp. [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword] 6 (4 or robot*.mp.) and 5 [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword] 7 (cervical adj3 (cancer* or malign* or carcinoma*)).mp. [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword] 8 7 and hysterectom*.mp. and (robot* or “da vinci”).mp. [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword] 9 6 or 8 10 9 and radical*.mp. [mp = title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]

Results 42,563 91,791 49,575 24,785 54,875

Search type Advanced Advanced Advanced Advanced Advanced

1776 Advanced 47,121 Advanced 255 Advanced 1776 Advanced 380

Web of science. (Radical SAME hysterectom*) AND (“da vinci” or robot*) and (study or studies or compar* or arm* or versus or vs or outcome*) 178.

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Checklist Selection Case definition Cases representativeness Selection of controls Definition of controls Comparability Case–control comparability Exposure Ascertainment of exposure Same ascertainment for cases and controls Non-response rate

Corrado et al. Chen et al. (2015) (2014)

Kim et al. (2014)

Yim et al. (2014)

Reynisson and Persson (2013)

Desille-Gbaguidi et.al. (2013)

Pilka et al. (2013)

Chong et al. (2013)

Vissa et al. (2013)

Wright et al. (2012)

Gortchev et al. (2012)

Sert and albert (2011)

Soliman et al. (2011)

✸ – ✸ ✸

✸ – ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

–

–

✸✸

–

–

–

–

–

✸

–

–

–

–

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

(continued on next page)

13

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Appendix B. Risk of bias assessment

14

Checklist Selection Case definition Cases representativeness Selection of controls Definition of controls Comparability Case–control comparability Exposure Ascertainment of exposure Same ascertainment for cases and controls Non-response rate

Tinelli et al. (2011)

Nam et al. (2010)

Geisler et al. (2010)

Cantrell et al. (2010)

Halliday et al. (2010)

Maggioni et al. (2009)

Estape et al. (2009)

Lowe et al. (2009)

Boggess et al. (2008)

Magrina et al. (2008)

Nezhat et al. (2008)

Ko et al. (2008)

Sert and Abeler (2007)

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

✸ ✸ ✸ ✸

–

✸✸

–

–

–

–

✸

–

✸

✸✸

–

–

–

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

✸ ✸

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

NA

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009

Appendix B . Risk of bias assessment (continued)

S.A.M. Shazly et al. / Gynecologic Oncology xxx (2015) xxx–xxx

References [1] S. Vaccarella, J. Lortet-Tieulent, M. Plummer, S. Franceschi, F. Bray, Worldwide trends in cervical cancer incidence: impact of screening against changes in disease risk factors, Eur. J. Cancer 49 (15) (2013) 3262–3273 (10//). [2] D. Saslow, D. Solomon, H.W. Lawson, M. Killackey, S.L. Kulasingam, J. Cain, et al., American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer, CA Cancer J. Clin. 62 (3) (2012) 147–172. [3] C. Pomel, D. Atallah, G.L. Bouedec, R. Rouzier, P. Morice, D. Castaigne, et al., Laparoscopic radical hysterectomy for invasive cervical cancer: 8-year experience of a pilot study, Gynecol. Oncol. 91 (3) (2003) 534–539 (12//). [4] D.F. Dargent, Laparoscopic surgery in gynecologic oncology, Surg. Clin. N. Am. 81 (4) (2001) 949–964. [5] P. Härkki-Sirén, J. Sjöberg, J. Mäkinen, P.K. Heinonen, M. Kauko, E. Tomás, et al., Finnish national register of laparoscopic hysterectomies: a review and complications of 1165 operations, Am. J. Obstet. Gynecol. 176 (1, Part 1) (1997) 118–122 (1//). [6] N. Johnson, D. Barlow, A. Lethaby, E. Tavender, L. Curr, R. Garry, Methods of hysterectomy: systematic review and meta-analysis of randomised controlled trials, BMJ 330 (7506) (2005) 1478. [7] S. Palomba, A. Falbo, R. Mocciaro, T. Russo, F. Zullo, Laparoscopic treatment for endometrial cancer: a meta-analysis of randomized controlled trials (RCTs), Gynecol. Oncol. 112 (2) (2009) 415–421 (2//). [8] J. Fanning, B. Fenton, M. Purohit, Robotic radical hysterectomy, Am. J. Obstet. Gynecol. 198 (6) (2008) 649.e1–649.e4 (6//). [9] K.C. Asciutto, G. Kalapotharakos, M. Löfgren, T. Högberg, C. Borgfeldt, Robot‐assisted surgery in cervical cancer patients reduces the time to normal activities of daily living, Acta Obstet. Gynecol. Scand. 94 (3) (2015) 260–265. [10] J.F. Boggess, P.A. Gehrig, L. Cantrell, A. Shafer, M. Ridgway, E.N. Skinner, et al., A case– control study of robot-assisted type III radical hysterectomy with pelvic lymph node dissection compared with open radical hysterectomy, Am. J. Obstet. Gynecol. 199 (4) (2008) 357.e1–357.e7. [11] L.A. Cantrell, A. Mendivil, P.A. Gehrig, J.F. Boggess, Survival outcomes for women undergoing type III robotic radical hysterectomy for cervical cancer: A 3-year experience, Gynecol. Oncol. 117 (2) (2010) 260–265 (5//). [12] C.-H. Chen, L.-H. Chiu, C.-W. Chang, Y.-K. Yen, Y.-H. Huang, W.-M. Liu, Comparing robotic surgery with conventional laparoscopy and laparotomy for cervical cancer management, Int. J. Gynecol. Cancer 24 (6) (2014) 1105–1111. [13] G.O. Chong, Y.H. Lee, D.G. Hong, Y.L. Cho, I.S. Park, Y.S. Lee, Robot versus laparoscopic nerve-sparing radical hysterectomy for cervical cancer: a comparison of the intraoperative and perioperative results of a single surgeon's initial experience, Int. J. Gynecol. Cancer 23 (6) (2013) 1145–1149. [14] G. Corrado, F. Fanfani, F. Ghezzi, A. Fagotti, S. Uccella, E. Mancini, et al., Minilaparoscopic versus robotic radical hysterectomy plus systematic pelvic lymphadenectomy in early cervical cancer patients. A multi-institutional study, Eur. J. Surg. Oncol. 41 (1) (2015) 136–141. [15] H. Desille-Gbaguidi, T. Hebert, J. Paternotte-Villemagne, C. Gaborit, E. Rush, G. Body, Overall care cost comparison between robotic and laparoscopic surgery for endometrial and cervical cancer, Eur. J. Obstet. Gynecol. Reprod. Biol. 171 (2) (2013) 348–352. [16] R. Estape, N. Lambrou, R. Diaz, E. Estape, N. Dunkin, A. Rivera, A case matched analysis of robotic radical hysterectomy with lymphadenectomy compared with laparoscopy and laparotomy, Gynecol. Oncol. 113 (3) (2009) 357–361. [17] J.P. Geisler, C.J. Orr, N. Khurshid, G. Phibbs, K.J. Manahan, Robotically assisted laparoscopic radical hysterectomy compared with open radical hysterectomy, Int. J. Gynecol. Cancer 20 (3) (2010) 438–442. [18] G. Gortchev, S. Tomov, L. Tantchev, A. Velkova, Z. Radionova, Robot-assisted radical hysterectomy—perioperative and survival outcomes in patients with cervical cancer compared to laparoscopic and open radical surgery, Gynecol. Surg. 9 (1) (2012) 81–88. [19] D. Halliday, S. Lau, Z. Vaknin, C. Deland, M. Levental, E. McNamara, et al., Robotic radical hysterectomy: comparison of outcomes and cost, J. Robot. Surg. 4 (4) (2010) 211–216. [20] T.-H. Kim, C.H. Choi, J.-K. Choi, A. Yoon, Y.-Y. Lee, T.-J. Kim, et al., Robotic versus laparoscopic radical hysterectomy in cervical cancer patients: a matched-case comparative study, Int. J. Gynecol. Cancer 24 (8) (2014) 1466–1473. [21] E.M. Ko, M.G. Muto, R.S. Berkowitz, C.M. Feltmate, Robotic versus open radical hysterectomy: a comparative study at a single institution, Gynecol. Oncol. 111 (3) (2008) 425–430. [22] M.P. Lowe, A.V. Hoekstra, A. Jairam-Thodla, D.K. Singh, B.M. Buttin, J.R. Lurain, et al., A comparison of robot-assisted and traditional radical hysterectomy for early-stage cervical cancer, J. Robot. Surg. 3 (1) (2009) 19–23. [23] A. Maggioni, L. Minig, V. Zanagnolo, M. Peiretti, F. Sanguineti, L. Bocciolone, et al., Robotic approach for cervical cancer: comparison with laparotomy: a case control study, Gynecol. Oncol. 115 (1) (2009) 60–64. [24] J.F. Magrina, R.M. Kho, A.L. Weaver, R.P. Montero, P.M. Magtibay, Robotic radical hysterectomy: comparison with laparoscopy and laparotomy, Gynecol. Oncol. 109 (1) (2008) 86–91. [25] E.J. Nam, S.W. Kim, S. Kim, J.H. Kim, Y.W. Jung, J.H. Paek, et al., A case–control study of robotic radical hysterectomy and pelvic lymphadenectomy using 3 robotic arms

[26] [27]

[28]

[29] [30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39] [40] [41] [42] [43]

[44]

[45]

[46]

[47]

[48] [49] [50]

[51]

[52]

15

compared with abdominal radical hysterectomy in cervical cancer, Int. J. Gynecol. Cancer 20 (7) (2010) 1284–1289. F. Nezhat, Minimally invasive surgery in gynecologic oncology: laparoscopy versus robotics, Gynecol. Oncol. 111 (2, Suppl.) (2008) S29–S32 (11//). A. Obermair, V. Gebski, M. Frumovitz, P.T. Soliman, K.M. Schmeler, C. Levenback, et al., A phase III randomized clinical trial comparing laparoscopic or robotic radical hysterectomy with abdominal radical hysterectomy in patients with early stage cervical cancer, J. Minim. Invasive Gynecol. 15 (5) (2008) 584–588. R. Pilka, R. Marek, P. Dzvinčuk, M. Kudela, D. Neubert, “ Learning curve” robotic radical hysterectomy compared to standardized laparoscopy assisted radical vaginal and open radical hysterectomy, Ceska Gynekol. 78 (1) (2013) 20–27. P. Reynisson, J. Persson, Hospital costs for robot-assisted laparoscopic radical hysterectomy and pelvic lymphadenectomy, Gynecol. Oncol. 130 (1) (2013) 95–99. B. Sert, V. Abeler, Robotic-assisted laparoscopic radical hysterectomy (Piver type III) with pelvic node dissection—case report, Eur. J. Gynaecol. Oncol. 27 (5) (2005) 531–533. B. Sert, V. Abeler, Robotic radical hysterectomy in early‐stage cervical carcinoma patients, comparing results with total laparoscopic radical hysterectomy cases. The future is now? Int. J. Med. Rob. Comput. Assist. Surg. 3 (3) (2007) 224–228. M.B. Sert, V. Abeler, Robot-assisted laparoscopic radical hysterectomy: comparison with total laparoscopic hysterectomy and abdominal radical hysterectomy; one surgeon's experience at the Norwegian Radium Hospital, Gynecol. Oncol. 121 (3) (2011) 600–604. P.T. Soliman, M. Frumovitz, C.C. Sun, R. dos Reis, K.M. Schmeler, A.M. Nick, et al., Radical hysterectomy: a comparison of surgical approaches after adoption of robotic surgery in gynecologic oncology, Gynecol. Oncol. 123 (2) (2011) 333–336. R. Tinelli, M. Malzoni, F. Cosentino, C. Perone, A. Fusco, E. Cicinelli, et al., Robotics versus laparoscopic radical hysterectomy with lymphadenectomy in patients with early cervical cancer: a multicenter study, Ann. Surg. Oncol. 18 (9) (2011) 2622–2628. E. Vizza, G. Corrado, E. Mancini, P. Vici, D. Sergi, E. Baiocco, et al., Laparoscopic versus robotic radical hysterectomy after neoadjuvant chemotherapy in locally advanced cervical cancer: a case control study, Eur. J. Surg. Oncol. 41 (2015) 142–147. J.D. Wright, T.J. Herzog, A.I. Neugut, W.M. Burke, Y.-S. Lu, S.N. Lewin, et al., Comparative effectiveness of minimally invasive and abdominal radical hysterectomy for cervical cancer, Gynecol. Oncol. 127 (1) (2012) 11–17. G.W. Yim, S.W. Kim, E.J. Nam, S. Kim, H.J. Kim, Y.T. Kim, Surgical outcomes of robotic radical hysterectomy using three robotic arms versus conventional multiport laparoscopy in patients with cervical cancer, Yonsei Med. J. 55 (5) (2014) 1222–1230. G. Wells, B. Shea, D. O'connell, J. Peterson, V. Welch, M. Losos, et al., The NewcastleOttawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Metaanalyses, 2000. S.P. Hozo, B. Djulbegovic, I. Hozo, Estimating the mean and variance from the median, range, and the size of a sample, BMC Med. Res. Methodol. 5 (1) (2005) 13. R. DerSimonian, N. Laird, Meta-analysis in clinical trials, Control. Clin. Trials 7 (3) (1986) 177–188. J. Higgins, S.G. Thompson, Quantifying heterogeneity in a meta‐analysis, Stat. Med. 21 (11) (2002) 1539–1558. Review Manager (RevMan) [Computer program], Version 53 Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration2014. A.M. Nick, M.M. Frumovitz, P.T. Soliman, K.M. Schmeler, P.T. Ramirez, Fertility sparing surgery for treatment of early-stage cervical cancer: open vs. robotic radical trachelectomy, Gynecol. Oncol. 124 (2) (2012) 276–280. J.C. Barnett, J.P. Judd, J.M. Wu, C.D. Scales Jr., E.R. Myers, L.J. Havrilesky, Cost comparison among robotic, laparoscopic, and open hysterectomy for endometrial cancer, Obstet. Gynecol. 116 (3) (2010) 685–693. J.F. Boggess, P.A. Gehrig, L. Cantrell, A. Shafer, M. Ridgway, E.N. Skinner, et al., A comparative study of 3 surgical methods for hysterectomy with staging for endometrial cancer: robotic assistance, laparoscopy, laparotomy, Am. J. Obstet. Gynecol. 199 (4) (2008) 360.e1–360.e9. E. Vizza, G. Corrado, E. Mancini, P. Vici, D. Sergi, E. Baiocco, et al., Laparoscopic versus robotic radical hysterectomy after neoadjuvant chemotherapy in locally advanced cervical cancer: a case control study, Eur. J. Surg. Oncol. 41 (1) (2013) 142–147. F.R. Nezhat, M.S. Datta, C. Liu, L. Chuang, K. Zakashansky, Robotic radical hysterectomy versus total laparoscopic radical hysterectomy with pelvic lymphadenectomy for treatment of early cervical cancer, JSLS 12 (3) (2008) 227–237. M.S. Piver, F. Rutledge, J.P. Smith, Five classes of extended hysterectomy for women with cervical cancer, Obstet. Gynecol. 44 (2) (1974) 265–272. D. Querleu, C.P. Morrow, Classification of radical hysterectomy, Lancet Oncol. 9 (3) (2008) 297–303. M. Reza, S. Maeso, J. Blasco, E. Andradas, Meta‐analysis of observational studies on the safety and effectiveness of robotic gynaecological surgery, Br. J. Surg. 97 (12) (2010) 1772–1783. G. Gaia, R.W. Holloway, L. Santoro, S. Ahmad, E. Di Silverio, A. Spinillo, Roboticassisted hysterectomy for endometrial cancer compared with traditional laparoscopic and laparotomy approaches: a systematic review, Obstet. Gynecol. 116 (6) (2010) 1422–1431. M. O'Neill, P.S. Moran, C. Teljeur, O.E. O'Sullivan, B.A. O'Reilly, M. Hewitt, et al., Robot-assisted hysterectomy compared to open and laparoscopic approaches: systematic review and meta-analysis, Arch. Gynecol. Obstet. 287 (5) (2013) 907–918.

Please cite this article as: S.A.M. Shazly, et al., Robotic radical hysterectomy in early stage cervical cancer: A systematic review and meta-analysis, Gynecol Oncol (2015), http://dx.doi.org/10.1016/j.ygyno.2015.06.009