Available online at www.sciencedirect.com
ScienceDirect EJSO xx (2015) 1e8
www.ejso.com
Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer G. Corrado a,*, G. Cutillo a, G. Pomati b, E. Mancini a, I. Sperduti c, L. Patrizi b, M. Saltari b, C. Vincenzoni a, E. Baiocco a, E. Vizza a a
b
Gynecologic Oncology Unit, “Regina Elena” National Cancer Institute, Rome, Italy Department of Surgery, Section of Gynaecology and Obstetrics, Tor Vergata University, Rome, Italy c Scientific Direction, “Regina Elena” National Cancer Institute, Rome, Italy Accepted 30 April 2015 Available online - - -
Abstract Objective: To compare different techniques of minimally invasive surgery (laparoscopy and robotics) to abdominal surgery in order to identify the optimal surgical technique in the treatment of endometrial cancer. Methods and materials: A single-institutional, matched, retrospective, cohort study was performed. All patients with clinical stage I or occult stage II endometrial cancer who underwent robotic hysterectomy, bilateral salpingo-oophorectomy lymphadenectomy from August 2010 and December 2013 were identified. Surgical and oncological outcomes were compared with patients matched by age, body mass index, tumor histology, and grade, who underwent abdominal or laparoscopic surgery between January 2001 and December 2013. Results: Three groups were identified: 177 laparotomies (group A), 277 laparoscopies (group B) and 72 robotics (group C). There were no statistically significant differences between the three groups in terms of age, BMI and FIGO stage. The operative time was shortest in group B (p ¼ 0.0001). Blood loss and transfusions were equivalent in group B and C, while they were greater in group A (p ¼ 0.0001). The intraoperative, early and late postoperative complications, rate of conversion, the re-intervention and median hospital stay were lower in group C. The rate of recurrence and death from disease was similar in all three groups. Conclusions: Minimally invasive surgery was superior to abdominal surgery in terms of surgical outcomes. Robotic surgery was superior to laparoscopy in terms of intra- and post-operative complications, conversion rates, length of hospital stay and re-interventions. In terms of oncological outcomes the three groups were equivalent. Ó 2015 Elsevier Ltd. All rights reserved.
Keywords: Robotic hysterectomy; Laparoscopic hysterectomy; Abdominal hysterectomy; Endometrial cancer; Minimally invasive surgery
Introduction Endometrial cancer (EC) is the most common cancer of the female genital system. In 2013, 49.560 new cases of endometrial cancer were diagnosed with a 3% death rate in the USA.1 Traditionally, the main treatment of endometrial cancer is surgery where it includes abdominal total * Corresponding author. Department of Oncological Surgery, Gynecologic Oncology Unit, “Regina Elena” National Cancer Institute e IFO, Via Elio Chianesi 53, 00144, Rome, Italy. Tel.: þ39 06 5266 5172; fax: þ39 06 5266 5173. E-mail address: [email protected] (G. Corrado).
hysterectomy, salpingo-oophorectomy and eventually pelvic and/or paraortic lymphadenectomy. In 1993, Childers2 was the first to propose the laparoscopic approach in performing complex surgical procedures such us total hysterectomy. The Gynecologic Oncologic group confirmed the superiority of laparoscopy compared to laparotomy in a randomized study (LAP-2),3 in terms of complications and hospital stay. Afterwards, laparoscopy became the favorite surgical technique to employ in the endometrial cancer staging. Moreover, in the last few years, minimally invasive surgery has acquired a leading role in gynecologic oncology.4
http://dx.doi.org/10.1016/j.ejso.2015.04.020 0748-7983/Ó 2015 Elsevier Ltd. All rights reserved. Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
2
G. Corrado et al. / EJSO xx (2015) 1e8
Recently, the robotic system Da Vinci (Intuitive Surgical IncÒ, 1266 Kifer Road, Building 101 Sunnyvale, CA) has been introduced. Its use in Gynecology was first approved in 2005 by the Food and Drug Administration. Robotic surgery allows surgeons to perform complex procedures, achieving three-dimensional vision, more accurate hand movements and better control of the instruments.5 However, the actual benefits received by robotic surgery compared to laparoscopy are not yet clear, especially considering the high cost of purchase and maintenance of the robot. The aim of this study is to compare three surgical techniques: laparotomy, laparoscopy and robotic surgery in order to identify the best method in terms of surgical outcome and survival in the staging of EC. Material and methods This is a retrospective study comparing surgical and oncological outcomes of three surgical techniques in the treatment of EC: laparotomy, laparoscopy and robotic. All the patients were treated by the same surgical team between January 2001 and December 2013 at the Gynaecologic Oncologic Unit, “Regina Elena” National Cancer Institute, Rome, Italy. All laparoscopic and robotic operations were performed by the same surgeon (E.V.) and his assistant (G.C.). Study design All patients with an histologically confirmed diagnosis of EC, both endometrioid and non-endometrioid, were included in the study. Before surgery, all patients were submitted to a clinical and instrumental evaluation, consisting of collecting medical history, and undergoing a physical examination, vaginal-pelvic examination, chest X-ray, an ultrasound scan and pelvic magnetic resonance imaging scan. Based on the clinical and instrumental evaluations, all the patients underwent either a type A or type B1 hysterectomy according to the QuerleueMorrow classification.6 Pelvic lymphadenectomy were performed only when risk factors (myometrial invasion more than 50%, high grading and non-endometrioid histotype) were detected at the intraoperative histological examination. Paraaortic lymphadenectomy is not routinely performed unless pelvic lymph nodes are confirmed to have metastatic disease on frozen section evaluations in order to determine the field of postoperative radiation. With positive pelvic nodes we performed aortic lymphadenectomy until inferior mesenteric artery, if positive lymph nodes at this level we extend until the left renal vein. Minimally invasive surgery (MIS) was introduced in our hospital in January 2004 and, thereafter, only patients with anesthesiological contraindications to minimally invasive surgery underwent abdominal hysterectomy. In our Institute robotic surgery began in August of 2010 and up to December 2011 all endometrial cancers that could be operated in
minimally invasive surgery, if the patient gave his consent, was operated by robotics. From January 2012 to date, to contain the high cost of robotic surgery, we decided internal guidelines to our Institute in which the patients with endometrial cancer at high risk of recurrence or obese BMI>30 kg/ m2 underwent robotic surgery. Informed consent for abdominal or MIS (laparoscopic or robotic) was obtained from all patients in accordance with the local and international legislation (declaration of Helsinki).7 All the data were collected independently from an internal review board. Patient characteristics were recorded, including: age, body mass index (BMI), histology, FIGO stage,8 grading, prior abdominal surgery and concomitant pathology such us other tumors, hypertension and diabetes. Then, the intra-operative parameters were recorded: operative time, blood loss, number of retrieved lymph nodes, transfusions, conversion rate and intra-operative complications. Operative time was calculated from the time of the first surgical incision to skin closure. Hematic blood loss was evaluated by the difference in the total amount of suction and irrigation fluids. Postoperative parameters included early postoperative complications (in the first 30 days after surgery) and later postoperative complications (more than 30 days after surgery), type of adjuvant therapy (radiotherapy and/or chemotherapy), median follow up in one month, recurrence, 3-years disease free survival (DFS) and 3-years overall survival (OS). Adjuvant therapy was tailored to the pathologic findings at primary operation after multidisciplinary tumor board (gynecologic oncology, pathology, radiation oncology, medical oncology) discussion. Treatment was based on the results of prospective, randomized clinical trials and National Comprehensive Cancer Network Guidelines.9 All the information concerning the follow up was collected over telephone calls to the patients. Statistical analysis Descriptive statistics was used to describe the patient characteristics. Continuous variables were compared using the ManneWhitney test and categorical variables were compared using chi-square test or Fisher exact test, as appropriate. All significance was defined at the p < 0.05 level. The SPSS (SPSS Inc., Chicago, IL, USA) statistical program was used for all analyses. Survival was calculated by the KaplaneMeier product-limit method from the date of surgery until the time of death for any causes (OS), relapse (DFS), or last visit (OS and DFS). Results Patient characteristics (Table 1) A total of 526 patients underwent endometrial cancer staging between January 2001 and December 2013 at our Institute: 177 abdominal hysterectomies (Group A), 277
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
G. Corrado et al. / EJSO xx (2015) 1e8 Table 1 Patients characteristics. Characteristics
Group A (range,%)
Group B (range,%)
Group C (range,%)
N patients Median age (years) Prior abdominal surgery Concomitant phatology Median BMI (kg/m2) Histology Adenosquamous Adenocarcinoma Serous Clear cell Type of surgery Class A Class B Pelvic lymphadenectomy FIGO stage IA IB II IIIA IIIB IIIC1 IIIC2 IVB Grading G1 G2 G3
177 64 (35e90) 90 (50.8%)
277 72 62 (28e86) 63 (38e88) 0.08 124 (44.8%) 42 (58.3%) 0.02
132 (74.5%) 181 (65.3%) 52 (72.2%) 28 (17e80)
P value
0.1
29 (17e59)
29 (20e42) 0.06 0.62 5 (2.8%) 4 (1.4%) 2 (2.7%) 158 (89.3%) 250 (90.2%) 60 (83.3%) 7 (3.9%) 11 (3.9%) 4 (5.5%) 7 (3.9%) 12 (4.3%) 6 (8.3%) 0.0001 143 (80.8%) 259 (93.5%) 69 (95.8%) 34 (19.2%) 18 (6.5%) 3 (4.1%) 64 (36.1%) 90 (32.5%) 27 (37.5%) 0.89 0.27 94 (53.1%) 45 (25.4%) 12 (6.8%) 7 (3.9%) 4 (2.2%) 4 (2.3%) 5 (2.8%) 6 (3.4%)
171 (61.7%) 66 (23.8%) 15 (5.4%) 13 (4.7%) 2 (0.7%) 6 (2.1%) 1 (0.3%) 3 (1.1%)
39 (54.1%) 18 (25%) 8 (11.1%) 2 (2.7%) 1 (1.4%) 1 (1.4%) 2 (2.7%) 1 (1.4%) 0.05
31 (17.5%) 93 (52.5%) 53 (29.9%)
59 (21.2%) 10 (13.8%) 150 (54.1%) 38 (52.7%) 68 (24.5%) 24 (33.3%)
BMI: body mass index. Bold indicates statistically significant values.
laparoscopic hysterectomies (Group B) and 72 robotic hysterectomies (Group C) were performed. No significant statistical differences were found between the three Groups in terms of age, BMI, histology, grading, FIGO stage and concomitant pathology. There was a higher prevalence of patients who underwent prior abdominal surgery in Group C (p ¼ 0.02) whereas type B1 radical hysterectomy was more frequent in patients in Group A (p ¼ 0.0001). Intra-operative characteristics (Tables 2 and 3) As it is shown in Table 2, median operative time was lower in group B rather than group A and C (p ¼ 0.0001). Blood loss were similar in group B and C, but lower than group A (p ¼ 0.0001). Pelvic lymphadenectomy was performed in 36.1%, 32.5% and 37.5% of patients in Group A, B and C, respectively. No significant differences were detected in the median of pelvic lymph nodes retrieved. Para-aortic lymphadenectomy was performed in a small number of patients in the three Groups. Intra-operative complications were higher in Group A (3.4%) than both Group B (2.6%) and C (1.3%). This difference was statistically significant (p ¼ 0.0001).
3
Transfusion rate was much higher in Group A compared to MIS (p ¼ 0.0001), while it was similar in Groups B and C. In Group B, 5.4% of conversions to abdominal surgery occurred compared to 1.4% and 2.7% of conversions to laparotomy and laparoscopy, respectively of the Group C. Moreover, one of the conversions to laparoscopy was due to a malfunctioning of a robotic arm. On the other hand, the other conversion to laparoscopy was due to the removal of a bulky aortic lymph node in one of our first robotic aortic lymphadenectomies. The conversion to abdominal surgery was needed to suture a bowel lesion. Postoperative characteristics (Tables 2 and 3) Early postoperative complications occurred in 7.9%, 3.9% and 1.4% in Groups A, B and C, respectively. Later postoperative complications were higher in Group A (4.5%), but they were similar in Groups B and C (3.6% vs 2.7%). No re-interventions were needed in Group C, while 4.5% and 5% of the patients underwent reintervention in Groups A and B, respectively. The median hospital stay was lower in Group C (3 days) than Groups A and B (p ¼ 0.0001). Oncological outcomes (Table 4) As it is shown in Table 4, no significant differences were found between the three Groups regarding the rate of patients who underwent adjuvant therapy (p ¼ 0.08). Similarly, no significant differences were detected in terms of modality of adjuvant therapy: radiotherapy (p ¼ 0.33), chemotherapy (p ¼ 0.18) or sequentially radiotherapy and chemotherapy (p ¼ 0.30). Median follow up was 78 months in Group A (range 1e159), 47 months in Group B (range 4e140) and 34 months in Group C (range 4e45). 11.9%, 7.2% and 6.9% of the patients had recurrence in Groups A, B and C, respectively with a median disease free survival of 29.5 months (range, 2e62 months) in Group A, 9 months (range, 2e53 months) in Group B and 12 months (range, 6e18 months) in Group C. The 3-year OS was 86.7%, 91.7% and 91.5% in Groups A, B and C, respectively (Fig. 1A) and the 3-year DFS was 92.1%, 88.4% and 91.5% in Groups A, B and C, respectively (Fig. 1B). The Kaplan Meyer curves show no significant differences in OS and DFS between the three Groups (p ¼ 0.95 and p ¼ 0.15, respectively), even if, some of the patients were lost to follow up above all in Group A unfortunately. Discussion Our study confirmed that MIS (laparoscopy or robotics) was more adequate/effective than abdominal surgery in terms of surgical outcome in the staging of endometrial cancer. Above all robotic surgery was safer than
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
4
G. Corrado et al. / EJSO xx (2015) 1e8
Table 2 Surgical outcomes. Characteristics
Group A (range,%)
Group B (range,%)
Group C (range,%)
P
Median operative time (min) Median blood loss (mL) Median pelvic lymph nodes Median aortic lymph nodes Major intraoperative complications Major early postoperative complications Major late postoperative complication Blood transfusion Conversion to laparotomy Conversion to laparoscopy Re-intervention Median hospital stay (days)
120 200 16 14 6 14 8 24 e e 8 7
100 100 12 9 7 11 10 3 15 e 14 4
115 100 16 6 1 1 2 1 1 2 0 3
0.0001 0.0001 0.89 0.0001 0.0001 0.0001 0.0001 0.0001 0.003
(45e420) (40e500) (5e58) (3e47) (3.4%) (7.9%) (4.5%) (13.5%)
(4.5%) (4e34)
(35e410) (5e600) (5e43) (4e34) (2.5%) (3.9%) (3.6%) (1.1%) (5.4%) (5%) (2e20)
(60e325) (5e250) (8e38) (3e9) (1.4%) (1.4%) (2.7%) (1.4%) (1.4%) (2.7%)
0.05 0.0001
(2e10)
Bold indicates statistically significant values.
laparoscopy in the treatment of endometrial cancer in terms of intraoperative and postoperative complications, hospital stay, conversions and re-intervention. On the other hand, robotic surgery was equivalent to laparoscopy in terms of blood loss and equivalent to abdominal surgery and laparoscopy regarding oncological outcomes. Surgical outcome The results of our study are not completely consistent with those of previous studies except for a few aspects. In Table 3 Intraoperative and postoperative complications. Type of complication
Group A (%)
Group B (%)
Group C (%)
Intraoperative Bowel injury Bladder injury Ureteral injury Nerves injury Vascular injury Postoperative short-term (£30 days) Haemoperitoneum Ureteral fistula Bladder fistula Bowel perforation Lymphocele Vaginal cuff dehiscence Laparotomic wound dehiscence Multiple Infections Pulmonary embolism Postoperative long-term (>30 days) Constipation Vaginal cuff dehiscence Laparocele Lymphocele Lymphedema Paresthesia and sciatic pain Multiple Insensitivity bladder
6 (3.4%) 2 e e 1 3 14 (7.9%)
7 (2.5%) 1 1 1 e 4 11 (3.9%)
1 (1.4%) 1 e e e e 1 (1.4%)
2 1 1 4 1 e 4 e 1 e 8 (4.5%)
4 2 e 2 e 1 e 1 e 1 10 (3.6%)
e e e e e e e e 1 e 2 (2.7%)
e e 2 1 2 2 e 1
e 4 1 e 1 3 1 e
1 e 1 e e e e e
contrast to our study, in a meta-analysis involving 3403 patients, Ran L et al.10 reported shorter operative time and lower blood loss, but higher complication rates in robotic surgery than laparoscopy. Moreover, in the meta-analysis of Ran L et al.10 the hospital stay and transfusion rates were equivalent in robotic surgery and laparoscopy. Instead, in our study, hospital stay was shorter in robotic group than in laparoscopy group (Table 2), but this result could be linked to the higher complication rates that we obtained in laparoscopy Group (Table 3). Surprisingly, in Group A we had more intestinal perforations than the other Groups, while in Group B the most common complication was hemoperitoneum. In Group C, the only complication was due to infection on 3rd postoperative day which required administration of antibiotic therapy. The use of minimally invasive surgery, however, has allowed us to completely eliminate abdominal wound dehiscence that in our cases represented 2.3% of the early complications of Group A. Table 5 shows the main studies11e18 comparing the surgical and oncological outcomes of the three approaches: laparotomy, laparoscopy and robotic in the staging of EC. Our results regarding the complication rates, median number of pelvic lymph nodes retrieved and conversion rates Table 4 Oncological outcomes. Adjuvant therapy
Group A (%)
Group B (%)
Group C (%)
P
None CT RT RT þ CT Median follow-up (months) Recurrence NED AWD DOD Other cause of dead Lost follow-up
101 (57.1%) 20 (11.3%) 57 (32.2%) 12 (6.8%) 78 (1e159)
169 (61%) 12 (4.3%) 75 (27%) 21 (7.6%) 47 (4e140)
36 (50%) 4 (5.5%) 25 (34.7%) 8 (11.1%) 34 (4e45)
0.08 0.05 0.18 0.3 0.001
15 (11.9%) 105 (83.3%) 0 5 (4%) 16 (12.7%) 51 (28.8%)
20 (7.2%) 237 (85.5%) 7 (2.5%) 8 (2.9%) 0 25 (9%)
5 (6.9%) 66 (91.6%) 1 (1.3%) 4 (5.5%) 1 (1.4%) 0
CT: chemotherapy; RT: radiotherapy; DOD: dead of disease; AWD: alive with disease; NED: no evidence of disease. Bold indicates statistically significant values.
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
G. Corrado et al. / EJSO xx (2015) 1e8
5
Figure 1. Survival analysis of robotic, laparoscopy and laparotomy for endometrial cancer. A) 3 year overall survival. B) 3 year disease free survival.
were similar in the review of literature; while Boggess JF et al.17 and Jung YW et al.16 found a shorter operative time in robotic surgery than laparoscopy surgery. Instead, our study did not show any significant differences in blood loss between robotic and laparoscopy surgery. Operative time was found to be significantly longer in laparotomy than robotic surgery, but shorter in laparoscopy than robotic surgery. Probably, the discrepant results regarding operative time and blood loss could be explained by the learning curve, as extensive experience in laparoscopic surgery helps surgeons to improve surgical outcomes in robotic surgery. Certainly, ample experience in laparoscopic surgery renders the transition from traditional laparoscopic surgery to robotic surgery easier, consequently improving the initial surgical outcomes.15 In our results, blood loss were similar in group B and C, but lower than group A (p ¼ 0.0001). This finding may have been influenced by the more radical surgery performed in group A compared to groups B and C. This more radical surgery in the group A is explained by the fact that we performed abdominal hysterectomy before the results of ILIADE study19 in 2009 that showed that Class II hysterectomy did not improve locoregional control and survival compared to class I hysterectomy in FIGO stage I of endometrial cancer. We have not had any vaginal cuff leakage or vaginal dehiscence that are an important complication described in robotic procedures.20 In the robotic group, the vaginal vault was closed always with single stitches using the vaginal way so as to reduce to a minimum the risk of vaginal dehiscence. In our study, the number of aortic lymph nodes removed was lower in groups of minimally invasive surgery compared to the laparotomy. This result is due to the fact that over the years the role of the aortic lymphadenectomy in endometrial cancer was gone resizing. In fact, even if in
1988 the International Federation of Obstetrics and Gynecology recommended surgical staging for endometrial cancer patients, 27 years later, the role of lymph node dissection remains controversial. Actually, data of literature on patterns of lymphatic dissemination in EC show that isolated para-aortic dissemination (with negative pelvic nodes) is rare (usually
ScienceDirect EJSO xx (2015) 1e8
www.ejso.com
Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer G. Corrado a,*, G. Cutillo a, G. Pomati b, E. Mancini a, I. Sperduti c, L. Patrizi b, M. Saltari b, C. Vincenzoni a, E. Baiocco a, E. Vizza a a
b
Gynecologic Oncology Unit, “Regina Elena” National Cancer Institute, Rome, Italy Department of Surgery, Section of Gynaecology and Obstetrics, Tor Vergata University, Rome, Italy c Scientific Direction, “Regina Elena” National Cancer Institute, Rome, Italy Accepted 30 April 2015 Available online - - -
Abstract Objective: To compare different techniques of minimally invasive surgery (laparoscopy and robotics) to abdominal surgery in order to identify the optimal surgical technique in the treatment of endometrial cancer. Methods and materials: A single-institutional, matched, retrospective, cohort study was performed. All patients with clinical stage I or occult stage II endometrial cancer who underwent robotic hysterectomy, bilateral salpingo-oophorectomy lymphadenectomy from August 2010 and December 2013 were identified. Surgical and oncological outcomes were compared with patients matched by age, body mass index, tumor histology, and grade, who underwent abdominal or laparoscopic surgery between January 2001 and December 2013. Results: Three groups were identified: 177 laparotomies (group A), 277 laparoscopies (group B) and 72 robotics (group C). There were no statistically significant differences between the three groups in terms of age, BMI and FIGO stage. The operative time was shortest in group B (p ¼ 0.0001). Blood loss and transfusions were equivalent in group B and C, while they were greater in group A (p ¼ 0.0001). The intraoperative, early and late postoperative complications, rate of conversion, the re-intervention and median hospital stay were lower in group C. The rate of recurrence and death from disease was similar in all three groups. Conclusions: Minimally invasive surgery was superior to abdominal surgery in terms of surgical outcomes. Robotic surgery was superior to laparoscopy in terms of intra- and post-operative complications, conversion rates, length of hospital stay and re-interventions. In terms of oncological outcomes the three groups were equivalent. Ó 2015 Elsevier Ltd. All rights reserved.
Keywords: Robotic hysterectomy; Laparoscopic hysterectomy; Abdominal hysterectomy; Endometrial cancer; Minimally invasive surgery
Introduction Endometrial cancer (EC) is the most common cancer of the female genital system. In 2013, 49.560 new cases of endometrial cancer were diagnosed with a 3% death rate in the USA.1 Traditionally, the main treatment of endometrial cancer is surgery where it includes abdominal total * Corresponding author. Department of Oncological Surgery, Gynecologic Oncology Unit, “Regina Elena” National Cancer Institute e IFO, Via Elio Chianesi 53, 00144, Rome, Italy. Tel.: þ39 06 5266 5172; fax: þ39 06 5266 5173. E-mail address: [email protected] (G. Corrado).
hysterectomy, salpingo-oophorectomy and eventually pelvic and/or paraortic lymphadenectomy. In 1993, Childers2 was the first to propose the laparoscopic approach in performing complex surgical procedures such us total hysterectomy. The Gynecologic Oncologic group confirmed the superiority of laparoscopy compared to laparotomy in a randomized study (LAP-2),3 in terms of complications and hospital stay. Afterwards, laparoscopy became the favorite surgical technique to employ in the endometrial cancer staging. Moreover, in the last few years, minimally invasive surgery has acquired a leading role in gynecologic oncology.4
http://dx.doi.org/10.1016/j.ejso.2015.04.020 0748-7983/Ó 2015 Elsevier Ltd. All rights reserved. Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
2
G. Corrado et al. / EJSO xx (2015) 1e8
Recently, the robotic system Da Vinci (Intuitive Surgical IncÒ, 1266 Kifer Road, Building 101 Sunnyvale, CA) has been introduced. Its use in Gynecology was first approved in 2005 by the Food and Drug Administration. Robotic surgery allows surgeons to perform complex procedures, achieving three-dimensional vision, more accurate hand movements and better control of the instruments.5 However, the actual benefits received by robotic surgery compared to laparoscopy are not yet clear, especially considering the high cost of purchase and maintenance of the robot. The aim of this study is to compare three surgical techniques: laparotomy, laparoscopy and robotic surgery in order to identify the best method in terms of surgical outcome and survival in the staging of EC. Material and methods This is a retrospective study comparing surgical and oncological outcomes of three surgical techniques in the treatment of EC: laparotomy, laparoscopy and robotic. All the patients were treated by the same surgical team between January 2001 and December 2013 at the Gynaecologic Oncologic Unit, “Regina Elena” National Cancer Institute, Rome, Italy. All laparoscopic and robotic operations were performed by the same surgeon (E.V.) and his assistant (G.C.). Study design All patients with an histologically confirmed diagnosis of EC, both endometrioid and non-endometrioid, were included in the study. Before surgery, all patients were submitted to a clinical and instrumental evaluation, consisting of collecting medical history, and undergoing a physical examination, vaginal-pelvic examination, chest X-ray, an ultrasound scan and pelvic magnetic resonance imaging scan. Based on the clinical and instrumental evaluations, all the patients underwent either a type A or type B1 hysterectomy according to the QuerleueMorrow classification.6 Pelvic lymphadenectomy were performed only when risk factors (myometrial invasion more than 50%, high grading and non-endometrioid histotype) were detected at the intraoperative histological examination. Paraaortic lymphadenectomy is not routinely performed unless pelvic lymph nodes are confirmed to have metastatic disease on frozen section evaluations in order to determine the field of postoperative radiation. With positive pelvic nodes we performed aortic lymphadenectomy until inferior mesenteric artery, if positive lymph nodes at this level we extend until the left renal vein. Minimally invasive surgery (MIS) was introduced in our hospital in January 2004 and, thereafter, only patients with anesthesiological contraindications to minimally invasive surgery underwent abdominal hysterectomy. In our Institute robotic surgery began in August of 2010 and up to December 2011 all endometrial cancers that could be operated in
minimally invasive surgery, if the patient gave his consent, was operated by robotics. From January 2012 to date, to contain the high cost of robotic surgery, we decided internal guidelines to our Institute in which the patients with endometrial cancer at high risk of recurrence or obese BMI>30 kg/ m2 underwent robotic surgery. Informed consent for abdominal or MIS (laparoscopic or robotic) was obtained from all patients in accordance with the local and international legislation (declaration of Helsinki).7 All the data were collected independently from an internal review board. Patient characteristics were recorded, including: age, body mass index (BMI), histology, FIGO stage,8 grading, prior abdominal surgery and concomitant pathology such us other tumors, hypertension and diabetes. Then, the intra-operative parameters were recorded: operative time, blood loss, number of retrieved lymph nodes, transfusions, conversion rate and intra-operative complications. Operative time was calculated from the time of the first surgical incision to skin closure. Hematic blood loss was evaluated by the difference in the total amount of suction and irrigation fluids. Postoperative parameters included early postoperative complications (in the first 30 days after surgery) and later postoperative complications (more than 30 days after surgery), type of adjuvant therapy (radiotherapy and/or chemotherapy), median follow up in one month, recurrence, 3-years disease free survival (DFS) and 3-years overall survival (OS). Adjuvant therapy was tailored to the pathologic findings at primary operation after multidisciplinary tumor board (gynecologic oncology, pathology, radiation oncology, medical oncology) discussion. Treatment was based on the results of prospective, randomized clinical trials and National Comprehensive Cancer Network Guidelines.9 All the information concerning the follow up was collected over telephone calls to the patients. Statistical analysis Descriptive statistics was used to describe the patient characteristics. Continuous variables were compared using the ManneWhitney test and categorical variables were compared using chi-square test or Fisher exact test, as appropriate. All significance was defined at the p < 0.05 level. The SPSS (SPSS Inc., Chicago, IL, USA) statistical program was used for all analyses. Survival was calculated by the KaplaneMeier product-limit method from the date of surgery until the time of death for any causes (OS), relapse (DFS), or last visit (OS and DFS). Results Patient characteristics (Table 1) A total of 526 patients underwent endometrial cancer staging between January 2001 and December 2013 at our Institute: 177 abdominal hysterectomies (Group A), 277
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
G. Corrado et al. / EJSO xx (2015) 1e8 Table 1 Patients characteristics. Characteristics
Group A (range,%)
Group B (range,%)
Group C (range,%)
N patients Median age (years) Prior abdominal surgery Concomitant phatology Median BMI (kg/m2) Histology Adenosquamous Adenocarcinoma Serous Clear cell Type of surgery Class A Class B Pelvic lymphadenectomy FIGO stage IA IB II IIIA IIIB IIIC1 IIIC2 IVB Grading G1 G2 G3
177 64 (35e90) 90 (50.8%)
277 72 62 (28e86) 63 (38e88) 0.08 124 (44.8%) 42 (58.3%) 0.02
132 (74.5%) 181 (65.3%) 52 (72.2%) 28 (17e80)
P value
0.1
29 (17e59)
29 (20e42) 0.06 0.62 5 (2.8%) 4 (1.4%) 2 (2.7%) 158 (89.3%) 250 (90.2%) 60 (83.3%) 7 (3.9%) 11 (3.9%) 4 (5.5%) 7 (3.9%) 12 (4.3%) 6 (8.3%) 0.0001 143 (80.8%) 259 (93.5%) 69 (95.8%) 34 (19.2%) 18 (6.5%) 3 (4.1%) 64 (36.1%) 90 (32.5%) 27 (37.5%) 0.89 0.27 94 (53.1%) 45 (25.4%) 12 (6.8%) 7 (3.9%) 4 (2.2%) 4 (2.3%) 5 (2.8%) 6 (3.4%)
171 (61.7%) 66 (23.8%) 15 (5.4%) 13 (4.7%) 2 (0.7%) 6 (2.1%) 1 (0.3%) 3 (1.1%)
39 (54.1%) 18 (25%) 8 (11.1%) 2 (2.7%) 1 (1.4%) 1 (1.4%) 2 (2.7%) 1 (1.4%) 0.05
31 (17.5%) 93 (52.5%) 53 (29.9%)
59 (21.2%) 10 (13.8%) 150 (54.1%) 38 (52.7%) 68 (24.5%) 24 (33.3%)
BMI: body mass index. Bold indicates statistically significant values.
laparoscopic hysterectomies (Group B) and 72 robotic hysterectomies (Group C) were performed. No significant statistical differences were found between the three Groups in terms of age, BMI, histology, grading, FIGO stage and concomitant pathology. There was a higher prevalence of patients who underwent prior abdominal surgery in Group C (p ¼ 0.02) whereas type B1 radical hysterectomy was more frequent in patients in Group A (p ¼ 0.0001). Intra-operative characteristics (Tables 2 and 3) As it is shown in Table 2, median operative time was lower in group B rather than group A and C (p ¼ 0.0001). Blood loss were similar in group B and C, but lower than group A (p ¼ 0.0001). Pelvic lymphadenectomy was performed in 36.1%, 32.5% and 37.5% of patients in Group A, B and C, respectively. No significant differences were detected in the median of pelvic lymph nodes retrieved. Para-aortic lymphadenectomy was performed in a small number of patients in the three Groups. Intra-operative complications were higher in Group A (3.4%) than both Group B (2.6%) and C (1.3%). This difference was statistically significant (p ¼ 0.0001).
3
Transfusion rate was much higher in Group A compared to MIS (p ¼ 0.0001), while it was similar in Groups B and C. In Group B, 5.4% of conversions to abdominal surgery occurred compared to 1.4% and 2.7% of conversions to laparotomy and laparoscopy, respectively of the Group C. Moreover, one of the conversions to laparoscopy was due to a malfunctioning of a robotic arm. On the other hand, the other conversion to laparoscopy was due to the removal of a bulky aortic lymph node in one of our first robotic aortic lymphadenectomies. The conversion to abdominal surgery was needed to suture a bowel lesion. Postoperative characteristics (Tables 2 and 3) Early postoperative complications occurred in 7.9%, 3.9% and 1.4% in Groups A, B and C, respectively. Later postoperative complications were higher in Group A (4.5%), but they were similar in Groups B and C (3.6% vs 2.7%). No re-interventions were needed in Group C, while 4.5% and 5% of the patients underwent reintervention in Groups A and B, respectively. The median hospital stay was lower in Group C (3 days) than Groups A and B (p ¼ 0.0001). Oncological outcomes (Table 4) As it is shown in Table 4, no significant differences were found between the three Groups regarding the rate of patients who underwent adjuvant therapy (p ¼ 0.08). Similarly, no significant differences were detected in terms of modality of adjuvant therapy: radiotherapy (p ¼ 0.33), chemotherapy (p ¼ 0.18) or sequentially radiotherapy and chemotherapy (p ¼ 0.30). Median follow up was 78 months in Group A (range 1e159), 47 months in Group B (range 4e140) and 34 months in Group C (range 4e45). 11.9%, 7.2% and 6.9% of the patients had recurrence in Groups A, B and C, respectively with a median disease free survival of 29.5 months (range, 2e62 months) in Group A, 9 months (range, 2e53 months) in Group B and 12 months (range, 6e18 months) in Group C. The 3-year OS was 86.7%, 91.7% and 91.5% in Groups A, B and C, respectively (Fig. 1A) and the 3-year DFS was 92.1%, 88.4% and 91.5% in Groups A, B and C, respectively (Fig. 1B). The Kaplan Meyer curves show no significant differences in OS and DFS between the three Groups (p ¼ 0.95 and p ¼ 0.15, respectively), even if, some of the patients were lost to follow up above all in Group A unfortunately. Discussion Our study confirmed that MIS (laparoscopy or robotics) was more adequate/effective than abdominal surgery in terms of surgical outcome in the staging of endometrial cancer. Above all robotic surgery was safer than
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
4
G. Corrado et al. / EJSO xx (2015) 1e8
Table 2 Surgical outcomes. Characteristics
Group A (range,%)
Group B (range,%)
Group C (range,%)
P
Median operative time (min) Median blood loss (mL) Median pelvic lymph nodes Median aortic lymph nodes Major intraoperative complications Major early postoperative complications Major late postoperative complication Blood transfusion Conversion to laparotomy Conversion to laparoscopy Re-intervention Median hospital stay (days)
120 200 16 14 6 14 8 24 e e 8 7
100 100 12 9 7 11 10 3 15 e 14 4
115 100 16 6 1 1 2 1 1 2 0 3
0.0001 0.0001 0.89 0.0001 0.0001 0.0001 0.0001 0.0001 0.003
(45e420) (40e500) (5e58) (3e47) (3.4%) (7.9%) (4.5%) (13.5%)
(4.5%) (4e34)
(35e410) (5e600) (5e43) (4e34) (2.5%) (3.9%) (3.6%) (1.1%) (5.4%) (5%) (2e20)
(60e325) (5e250) (8e38) (3e9) (1.4%) (1.4%) (2.7%) (1.4%) (1.4%) (2.7%)
0.05 0.0001
(2e10)
Bold indicates statistically significant values.
laparoscopy in the treatment of endometrial cancer in terms of intraoperative and postoperative complications, hospital stay, conversions and re-intervention. On the other hand, robotic surgery was equivalent to laparoscopy in terms of blood loss and equivalent to abdominal surgery and laparoscopy regarding oncological outcomes. Surgical outcome The results of our study are not completely consistent with those of previous studies except for a few aspects. In Table 3 Intraoperative and postoperative complications. Type of complication
Group A (%)
Group B (%)
Group C (%)
Intraoperative Bowel injury Bladder injury Ureteral injury Nerves injury Vascular injury Postoperative short-term (£30 days) Haemoperitoneum Ureteral fistula Bladder fistula Bowel perforation Lymphocele Vaginal cuff dehiscence Laparotomic wound dehiscence Multiple Infections Pulmonary embolism Postoperative long-term (>30 days) Constipation Vaginal cuff dehiscence Laparocele Lymphocele Lymphedema Paresthesia and sciatic pain Multiple Insensitivity bladder
6 (3.4%) 2 e e 1 3 14 (7.9%)
7 (2.5%) 1 1 1 e 4 11 (3.9%)
1 (1.4%) 1 e e e e 1 (1.4%)
2 1 1 4 1 e 4 e 1 e 8 (4.5%)
4 2 e 2 e 1 e 1 e 1 10 (3.6%)
e e e e e e e e 1 e 2 (2.7%)
e e 2 1 2 2 e 1
e 4 1 e 1 3 1 e
1 e 1 e e e e e
contrast to our study, in a meta-analysis involving 3403 patients, Ran L et al.10 reported shorter operative time and lower blood loss, but higher complication rates in robotic surgery than laparoscopy. Moreover, in the meta-analysis of Ran L et al.10 the hospital stay and transfusion rates were equivalent in robotic surgery and laparoscopy. Instead, in our study, hospital stay was shorter in robotic group than in laparoscopy group (Table 2), but this result could be linked to the higher complication rates that we obtained in laparoscopy Group (Table 3). Surprisingly, in Group A we had more intestinal perforations than the other Groups, while in Group B the most common complication was hemoperitoneum. In Group C, the only complication was due to infection on 3rd postoperative day which required administration of antibiotic therapy. The use of minimally invasive surgery, however, has allowed us to completely eliminate abdominal wound dehiscence that in our cases represented 2.3% of the early complications of Group A. Table 5 shows the main studies11e18 comparing the surgical and oncological outcomes of the three approaches: laparotomy, laparoscopy and robotic in the staging of EC. Our results regarding the complication rates, median number of pelvic lymph nodes retrieved and conversion rates Table 4 Oncological outcomes. Adjuvant therapy
Group A (%)
Group B (%)
Group C (%)
P
None CT RT RT þ CT Median follow-up (months) Recurrence NED AWD DOD Other cause of dead Lost follow-up
101 (57.1%) 20 (11.3%) 57 (32.2%) 12 (6.8%) 78 (1e159)
169 (61%) 12 (4.3%) 75 (27%) 21 (7.6%) 47 (4e140)
36 (50%) 4 (5.5%) 25 (34.7%) 8 (11.1%) 34 (4e45)
0.08 0.05 0.18 0.3 0.001
15 (11.9%) 105 (83.3%) 0 5 (4%) 16 (12.7%) 51 (28.8%)
20 (7.2%) 237 (85.5%) 7 (2.5%) 8 (2.9%) 0 25 (9%)
5 (6.9%) 66 (91.6%) 1 (1.3%) 4 (5.5%) 1 (1.4%) 0
CT: chemotherapy; RT: radiotherapy; DOD: dead of disease; AWD: alive with disease; NED: no evidence of disease. Bold indicates statistically significant values.
Please cite this article in press as: Corrado G, et al., Surgical and oncological outcome of robotic surgery compared to laparoscopic and abdominal surgery in the management of endometrial cancer, Eur J Surg Oncol (2015), http://dx.doi.org/10.1016/j.ejso.2015.04.020
G. Corrado et al. / EJSO xx (2015) 1e8
5
Figure 1. Survival analysis of robotic, laparoscopy and laparotomy for endometrial cancer. A) 3 year overall survival. B) 3 year disease free survival.
were similar in the review of literature; while Boggess JF et al.17 and Jung YW et al.16 found a shorter operative time in robotic surgery than laparoscopy surgery. Instead, our study did not show any significant differences in blood loss between robotic and laparoscopy surgery. Operative time was found to be significantly longer in laparotomy than robotic surgery, but shorter in laparoscopy than robotic surgery. Probably, the discrepant results regarding operative time and blood loss could be explained by the learning curve, as extensive experience in laparoscopic surgery helps surgeons to improve surgical outcomes in robotic surgery. Certainly, ample experience in laparoscopic surgery renders the transition from traditional laparoscopic surgery to robotic surgery easier, consequently improving the initial surgical outcomes.15 In our results, blood loss were similar in group B and C, but lower than group A (p ¼ 0.0001). This finding may have been influenced by the more radical surgery performed in group A compared to groups B and C. This more radical surgery in the group A is explained by the fact that we performed abdominal hysterectomy before the results of ILIADE study19 in 2009 that showed that Class II hysterectomy did not improve locoregional control and survival compared to class I hysterectomy in FIGO stage I of endometrial cancer. We have not had any vaginal cuff leakage or vaginal dehiscence that are an important complication described in robotic procedures.20 In the robotic group, the vaginal vault was closed always with single stitches using the vaginal way so as to reduce to a minimum the risk of vaginal dehiscence. In our study, the number of aortic lymph nodes removed was lower in groups of minimally invasive surgery compared to the laparotomy. This result is due to the fact that over the years the role of the aortic lymphadenectomy in endometrial cancer was gone resizing. In fact, even if in
1988 the International Federation of Obstetrics and Gynecology recommended surgical staging for endometrial cancer patients, 27 years later, the role of lymph node dissection remains controversial. Actually, data of literature on patterns of lymphatic dissemination in EC show that isolated para-aortic dissemination (with negative pelvic nodes) is rare (usually
