ORIGINAL ARTICLE
Long-term Oncologic Outcomes of Robotic Low Anterior Resection for Rectal Cancer A Comparative Study With Laparoscopic Surgery Eun Jung Park, MD, Min Soo Cho, MD, Se Jin Baek, MD, PhD, Hyuk Hur, MD, Byung Soh Min, MD, PhD, Seung Hyuk Baik, MD, PhD, Kang Young Lee, MD, PhD, and Nam Kyu Kim, MD, PhD
Objective: The aim of this study is to evaluate long-term oncologic outcomes of robotic surgery for rectal cancer compared with laparoscopic surgery at a single institution. Background: Robotic surgery is regarded as a new modality to surpass the technical limitations of conventional surgery. Short-term outcomes of robotic surgery for rectal cancer were acceptable in previous reports. However, evidence of long-term feasibility and oncologic safety is required. Methods: Between April 2006 and August 2011, 217 patients who underwent minimally invasive surgery for rectal cancer with stage I–III disease were enrolled prospectively (robot, n = 133; laparoscopy, n = 84). Median follow-up period was 58 months (range, 4–80 months). Perioperative clinicopathologic outcomes, morbidities, 5-year survival rates, prognostic factors, and cost were evaluated. Results: Perioperative clinicopathologic outcomes demonstrated no significant differences except for the conversion rate and length of hospital stay. The 5-year overall survival rate was 92.8% in robotic, and 93.5% in laparoscopic surgical procedures (P = 0.829). The 5-year disease-free survival rate was 81.9% and 78.7%, respectively (P = 0.547). Local recurrence was similar: 2.3% and 1.2% (P = 0.649). According to the univariate analysis, this type of surgical approach was not a prognostic factor for long-term survival. The patient’s mean payment for robotic surgery was approximately 2.34 times higher than laparoscopic surgery. Conclusions: No significant differences were found in the 5-year overall, disease-free survival and local recurrence rates between robotic and laparoscopic surgical procedures. We concluded that robotic surgery for rectal cancer failed to offer any oncologic or clinical benefits as compared with laparoscopy despite an increased cost. Keywords: laparoscopic surgery, oncologic outcome, rectal cancer, robotic surgery, total mesorectal excision (Ann Surg 2015;261:129–137)
M
inimally invasive surgery for colorectal cancer is a relatively new surgical modality, able to substitute for conventional open surgery on the basis of similar or better perioperative and oncologic outcomes.1–4 According to the long-term follow-up of the United Kingdom Medical Research Council Conventional versus Laparoscopic-Assisted Surgery in Colorectal Cancer (UK MRC CLASICC) trial, laparoscopic surgery was regarded as an alternative to conventional surgery.5,6
From the Section of Colon and Rectal Surgery, Department of Surgery, Yonsei University College of Medicine, Seoul, South Korea. Disclosure: The authors have no conflicts of interest or financial ties of disclosure. Reprints: Seung Hyuk Baik, MD, PhD, Section of Colon and Rectal Surgery, Department of Surgery, Yonsei University College of Medicine, Seoul, 120752, South Korea. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26101-0129 DOI: 10.1097/SLA.0000000000000613
Annals of Surgery r Volume 261, Number 1, January 2015
However, in rectal cancer, surgeons are faced with challenging conditions such as a narrow pelvic cavity, anatomical complexity, and restricted surgical view during laparoscopic surgery, although previous studies reported that laparoscopic rectal cancer surgery was feasible.2,5–7 Thus, the robotic system seems potentially suited for the surgical treatment of rectal cancer due to several theoretical advantages and has been used since it was adopted in 2001.8 However, there are not sufficient data to justify the adoption of the robotic system instead of laparoscopic surgery for rectal cancer. Several previous studies documented that robotic surgery is equivalent to laparoscopic surgery with respect to the short-term perioperative and oncologic outcomes.9–15 In rectal cancer surgery, the most important issue is definitely the long-term oncologic outcomes.9–16 To date, there are no reports to evaluate the long-term oncologic outcomes of robotic surgery compared with laparoscopic surgery for rectal cancer. We hypothesized that the long-term oncologic outcomes would be better in robotic rectal cancer surgery than in laparoscopic rectal cancer surgery. Therefore, this study aimed to evaluate the long-term oncologic outcomes of robotic surgery for rectal cancer compared with conventional laparoscopic surgery. Moreover, the difference of the cost between both procedures was evaluated.
METHODS Study Population and Patient Selection Between April 2006 and August 2011, 288 consecutive patients, who were diagnosed with rectal cancer (adenocarcinoma), underwent a low anterior resection (LAR) by a single surgeon (Baik SH) at Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. The first laparoscopic-assisted LAR and the first robotic-assisted LAR were performed on April 2006 and June 2006, respectively. Thus, the learning curve period of both procedures was equivalent. Before surgery, all patients were informed of the detailed characteristics of both robotic and laparoscopic surgical procedures, as explained by the nurse coordinator. It was explained that there was no evidence that one procedure is better than the other (although theoretical advantages of the robot in terms of facilitating pelvic dissection were explained) but that both have shown better short-term benefits over open approaches for the treatment of rectal cancer and that due to the health system in Korea, they would have to pay an additional US $6000 for choosing the robotic approach over the laparoscopic approach. After informed consent, therefore, the patients decided their preferred approach. Open surgery was performed when primary tumor invasion to other organs and structures was observed in preoperative imaging studies. Among these 288 patients, 50 patients who underwent open surgery, 20 patients with stage IV, and 1 patient lost to follow-up were excluded from the study population. Finally, 217 patients were enrolled in this study, which included robotic-assisted LAR (n = 133) www.annalsofsurgery.com | 129
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
using the da Vinci surgical system (Intuitive Surgical, Sunnyvale, CA) and laparoscopic-assisted LAR (n = 84) (Fig. 1). This study was approved by the Institutional Review Board of Severance Hospital.
Data Collection and Evaluation Parameters Data were collected from the Yonsei Colorectal Cancer Database for this study. Clinicopathologic results were evaluated retrospectively using this database, and oncologic outcomes of the
FIGURE 1. Flowchart of patient selection. TABLE 1. Patient Characteristics Robot (n = 133) Age∗ ,
yr Sex‡ Male Female Weight∗ , kg Height∗ , m BMI∗ , kg/m2 ASA score‡ 1 2 3 Tumor location from anal verge‡ Low (0–5 cm) Mid (5.1–10 cm) Upper (10.1–15 cm) History of abdominal surgery‡ Preoperative CRT‡ Yes No
Laparoscopy (n = 84)
59.2 ± 11.4 (32–86)
63.5 ± 11.2 (29–83)
86 (64.7) 47 (35.3) 62.5 ± 9.9 (37–86) 1.64 ± 0.07 (1.44–1.78) 23.1 ± 2.9 (14.6–32.8)
60 (71.4) 24 (28.6) 61.0 ± 9.4 (32–85) 1.63 ± 0.08 (1.44–1.80) 22.9 ± 2.8 (14.6–29.4)
94 (70.7) 31 (23.3) 8 (6.0)
51 (60.7) 28 (33.3) 5 (6.0)
33 (24.8) 60 (45.1) 40 (30.1) 10 (7.5)
16 (19.1) 37 (44.0) 31 (36.9) 11 (13.1)
15 (11.3) 118 (88.7)
10 (11.9) 74 (88.1)
P 0.007† 0.301§ 0.272† 0.256† 0.631† 0.263§
0.471§
0.176§ 0.888§
∗
Mean ± standard deviation (range). †Independent 2-sample t test. ‡n (%). §χ 2 test. ASA indicates American Society of Anesthesiologists; BMI, body mass index; CRT, chemoradiotherapy.
130 | www.annalsofsurgery.com
C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
TABLE 2. Operation Details and Perioperative Clinical Outcomes Operation time∗ , min Diverting ileostomy‡ Conversion‡ Intraoperative bleeding∗ , mL Intraoperative complications or events‡ Bladder injury Perforation of the rectum Disruption of colorectal anastomosis Days to 1st gas passing∗ , d Days to 1st bowel movement∗ , d Days to 1st soft diet∗ , d Length of hospital stay∗ , day
Robot (n = 133)
Laparoscopy (n = 84)
P
205.7 ± 67.3 (109–505) 29 (21.8) 0 (0.0) 77.6 ± 153.2 (0–700)
208.8 ± 81.2 (94–407) 20 (23.8) 6 (7.1) 82.3 ± 185.8 (0–1100)
0.766† 0.731§ 0.003¶ 0.841† 1.000¶
1 (0.8) 0 (0.0) 1 (0.8) 2.42 ± 0.93 (1–6) 3.61 ± 1.04 (1–7) 4.92 ± 1.35 (3–11) 5.86 ± 1.43 (4–14)
1 (1.2) 0 (0.0) 0 (0.0) 2.47 ± 1.10 (1–6) 4.07 ± 1.96 (1–8) 5.19 ± 1.62 (2–11) 6.54 ± 2.65 (3–25)
0.727† 0.143† 0.197† 0.035†
∗
Mean ± standard deviation(range). †Independent 2-sample t test. ‡n (%). §χ 2 test. ¶Fisher’s exact test.
TABLE 3. Perioperative and Long-term Postoperative Complications Robot (n = 133) Perioperative complications∗ Grade I‡ Ileus Voiding difficulty Wound infection Pleural effusion Retrograde ejaculation Grade II‡ Ascites Chyloperitoneum Blood transfusion due to bleeding Grade III‡ Anastomotic leakage Anastomotic site bleeding Ischemic colitis Intra-abdominal abscess Grade IV‡ Asthma attack Pneumonia Grade V‡ Long-term postoperative complications§ Grade I‡ Ileus Voiding difficulty Grade II‡ Grade III‡ Adhesion Incisional hernia Anastomotic stricture Rectovaginal/rectovesical fistula Grade IV‡ Grade V‡ Major perioperative complications‡ Major long-term complications‡
Laparoscopy (n = 84)
P 0.897†
11 (8.3) 3 5 2 0 1 5 (3.8) 2 2 1 9 (6.8) 6 2 1 0 1 (0.8) 0 1 0 (0.0)
7 (8.3) 3 3 0 1 0 4 (4.8) 0 1 3 6 (7.1) 3 3 0 0 2 (2.4) 1 1 0 (0.0)
3 (2.3) 1 2 0 (0.0) 2 (1.5) 1 0 1 0 0 (0.0) 0 (0.0) 10 (7.5) 2 (1.5)
2 (2.4) 1 1 0 (0.0) 3 (3.6) 0 2 1 0 0 (0.0) 0 (0.0) 8 (9.5) 3 (3.6)
0.687†
0.602¶ 0.378†
Major complications were defined as complications with a grade III and higher of the Clavien–Dindo classification. ∗ Complications within 30 days from operation date. †Fisher’s exact test. ‡n (%). §New complications 30 days after operation date ¶χ 2 test.
C 2014 Lippincott Williams & Wilkins
www.annalsofsurgery.com | 131
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
TABLE 4. Postoperative Pathologic Outcomes Robot (n = 133)
Laparoscopy (n = 84)
stage∗
TNM I II III Grade of differentiation∗ Well Moderate Poor Mucinous No. nodes analyzed§ PRM§, cm DRM§, cm Lymphovascular invasion∗ CRM∗ Noninvolved (>1 mm) Involved (≤1 mm)
P 0.255†
49 (36.8) 36 (27.1) 48 (36.1)
22 (26.2) 28 (33.3) 34 (40.5)
34 (25.5) 93 (69.9) 5 (3.8) 1 (0.8) 16.34 ± 8.79 (2–43) 11.17 ± 4.64 (4–35) 2.75 ± 2.14 (1–14) 43 (32.3)
19 (22.6) 60 (71.4) 2 (2.4) 3 (3.6) 16.63 ± 10.24 (2–49) 10.51 ± 5.65 (4–34) 2.87 ± 1.63 (1–9) 24 (28.6)
124 (93.2) 9 (6.8)
78 (92.9) 6 (7.1)
0.467‡
0.823¶ 0.352¶ 0.652¶ 0.559† 0.915†
∗
n (%). †χ 2 test. §Mean ± standard deviation (range). ‡Fisher’s exact test. ¶Independent 2-sample t test. DRM indicates distal resection margin; PRM, proximal resection margin.
entire enrolled patients were updated until May 2013. Insufficient data were reevaluated by electronic medical charts and telephone interviews were conducted, if needed. The preoperative status of patients was categorized by the American Society of Anesthesiologists Classification. The location of the tumor was divided as: low (below 5 cm), mid (5.1–10 cm) and upper rectum (10.1–15 cm). Conversion was defined as the unintended extension of laparotomy beyond the routine length of incision (4 cm) to complete the surgical procedures. Postoperative complications were stratified by the Clavien– Dindo classification of surgical complications.17 The total numbers of postoperative complications were counted for all cases related to morbidity. Major complications were defined as complications with a grade III and higher of the Clavien–Dindo classification.17 Longterm complications were defined as postoperative complications that occurred after the 30th postoperative day. For analysis of the pathologic outcomes, tumor node metastasis (TNM) stage (American Joint Committee on Cancer, seventh edition), grade of histologic differentiation, resection margins, numbers of harvested lymph nodes, lymphovascular invasion, and circumferential resection margin (CRM) were evaluated. Involved CRM was defined when the tumor was located 1 mm or less from the CRM.18 Examination of the resected specimens was performed using the protocol of the College of American Pathologists.19 Visual enhancement techniques were used when fewer than 12 nodes were evaluated. The preoperative clinical diagnosis was determined by abdominopelvic computed tomography (CT), chest CT, rectal magnetic resonance imaging, and transrectal ultrasonography. Tissue diagnosis was obtained by a colonoscopic biopsy. Local recurrence was defined as the relapse of the tumor at the primary site confirmed by radiological or histological evidence. Simultaneous local and systemic recurrences were counted as a local recurrence. Distant metastasis was considered as metastatic lesions that were diagnosed in other organs beyond the primary site. Total cost was evaluated and defined as total expenses that occurred from admission to discharge and also included all charges for the operation, perioperative management, and treatment. 132 | www.annalsofsurgery.com
Surgical Technique All cases of robotic-assisted LAR in this study were performed by the hybrid technique, which is composed of 2 surgical steps. The first step is performed by conventional laparoscopic instruments for the inferior mesenteric vessel ligation and left colon and splenic flexure mobilization. The second step is pelvic dissection using the robotic system on the basis of total mesorectal excision or tumor-specific mesorectal excision principles according to the tumor location from the anal verge.20,21 Tumor-specific mesorectal excision is a surgical method to dissect the mesorectum encircled by the proper fascia of the rectum below 4 cm from the lower margin of the tumor.22 Laparoscopic-assisted LAR was performed in the same manner as robotic surgery using laparoscopic instruments. Further detailed procedures of both robotic and laparoscopic surgical methods are described in our previous reports.21–23
Chemoradiotherapy and Follow-up Patients received chemoradiotherapy for stage T3–4N0 or N + M0 low- and mid-rectal cancers, with the majority of the patients undergoing postoperative treatment because of the standard protocols in place at our institution early in the study period. Thus, mainly, postoperative chemoradiotherapy was performed. The regimen was based on 5-fluorouracil and leucovorin. The discharged patients visited the outpatient clinic after 1 month, 3 months, and every 3 months for the first 3 years and then 6 months until 5 years after surgery. In the outpatient clinic, regular laboratory with carcinoembryonic antigen and a physical examination were performed. The patient was examined by colonoscopy 1 year after surgery and then 5 years. Chest and abdominopelvic computed tomographic scans were obtained every 6 months to detect local recurrence or systemic metastasis during the follow-up period.
Statistical Analysis Statistical analysis was performed using the SPSS program (Statistical Product and Service Solution 18 for Windows; SPSS Inc., Chicago, IL). The Student t test was used for continuous variables. C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
FIGURE 2. The 5-year overall survival rate between robotic and laparoscopic surgical procedures. The categorical variables were analyzed by the χ 2 test or Fisher’s exact test. The 5-year overall, disease-free survival rates and cumulative incidence of local recurrence were calculated by using the Kaplan– Meier method. Comparison of the survival and local recurrence rates between the 2 surgical methods was analyzed by the log-rank test. Univariate analysis of clinicopathological factors upon overall and disease-free survival was performed using the log-rank test to determine the prognostic value of the surgical methods (robot vs laparoscopy). All statistically significant factors determined by univariate analysis were conducted for multivariate analysis by the Cox proportional hazards regression model with a forward selection of variables. A P value less than 0.05 was considered statistically significant.
RESULTS Patient Characteristics
The mean age of robotic surgery was 59.2 ± 11.4 (mean ± standard deviation) years, and 63.5 ± 11.2 years for the laparoscopic group (P = 0.007). There were no significant differences between the groups in sex, weight, height, and body mass index. An American C 2014 Lippincott Williams & Wilkins
Society of Anesthesiologists score of 1 showed the highest distribution in both groups (70.7% vs 60.7%, robot vs laparoscopy) and was not significantly different between the groups. Tumor location and history of abdominal surgery were not different significantly between the groups. The rate of low– or mid–rectal cancer was 69.9% in the robotic group and 63.1% in the laparoscopic group. The rate of patients who underwent preoperative chemoradiotherapy was similar between the groups (Table 1).
Perioperative Clinical Outcomes, Complications
The mean operation time was 205.7 ± 67.3 minutes in robotic surgery and 208.8 ± 81.2 minutes in laparoscopic surgery (P = 0.766). There were no significant differences between the groups for intraoperative complications or bleeding or undergoing a diverting ileostomy. However, conversion occurred more frequently in the laparoscopic group (P = 0.003). There was no conversion in the robotic group compared with 6 cases in the laparoscopic group. The reasons for conversion of laparoscopic surgery were intraoperative pelvic sidewall bleeding (1 case), rectal perforation (1 case), tumor adhesion to the bladder (1 case), and severe narrow pelvic cavity (3 cases) (P = 0.003). Days to first gas passing, first bowel movement, and soft diet were not significantly different between the groups. However, patients who underwent laparoscopic surgery stayed in the hospital www.annalsofsurgery.com | 133
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
FIGURE 3. The 5-year disease-free survival rate between robotic and laparoscopic surgical procedures. longer than the robotic group, although the difference was only 0.68 days (P = 0.035) (Table 2). The occurrence rate of both perioperative and long-term complications was similar between robotic and laparoscopic surgical procedures (perioperative: P = 0.897, long term: P = 0.687). Anastomotic leakage rate was 4.5% in the robotic group and 3.6% in the laparoscopic group. Major complications were not significantly different. There were no 30-day mortalities in both groups (Table 3).
Postoperative Pathologic Assessment There was no significant difference between the robotic group and the laparoscopic group in the pathologic outcomes. TNM stage, histologic grade of differentiation, resection margins, and lymphovascular invasion showed similar results in both groups. The factors that reflected the quality of resected specimen such as the number of harvested lymph nodes, distal resection margin, and CRM were not significantly different. Number of nodes analyzed was 16.34 ± 8.79 (range: 2–43) in the robotic group and 16.63 ± 10.24 (range: 2–49) in the laparoscopic group (P = 0.823). Rate of positive CRM was 6.8% in the robotic group and 7.1% in the laparoscopic group and was not significantly different between the groups (P = 0.915) (Table 4). 134 | www.annalsofsurgery.com
Oncologic Outcomes
The mean follow-up period was 54.4 ± 17.3 (range: 4–80 months) (interquartile range: 49–66) months and the median followup period was 58 (range: 4–80 months) months. The 5-year overall survival rate was 92.8% in the robotic group and 93.5% in the laparoscopic group (P = 0.829). The 5-year overall survival according to stages of the robotic group was 97.8% in stage I, 94.2% in stage II, and 86.8% in stage III. In the laparoscopic group, the 5-year overall survival according to stages of the laparoscopic group was 95.0% in stage I, 100% in stage II, and 87.8% in stage III. The 5-year overall survival rate according to each stage was similar between the robotic and the laparoscopic groups (Fig. 2). The 5-year disease-free survival rate was 81.9% in the robotic group and 78.7% in the laparoscopic group (P = 0.547). The 5-year disease-free survival rates of the robotic group according to stages were 93.7% in stage I, 82.1% in stage II, and 70.3% in stage III. In the laparoscopic group, the 5-year disease-free survival rates according to stages were 83.1% in stage I, 92.9% in stage II, and 65.0% in stage III. There was no statistical difference of the 5-year disease-free survival according to each stage between the groups (Fig. 3). Cumulative incidence of local recurrence was 2.3% in the robotic group and 1.2% in the laparoscopic group (P = 0.649) C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
TABLE 5. Prognostic Factors of 5-Year Survival by Univariate Analysis n = 217 DFS (%)
FIGURE 4. Cumulative incidence of local recurrence rate of all patients with robotic or laparoscopic surgical procedures for rectal cancer: robotic 2.3% and laparoscopic 1.2% (P = 0.649).
(Fig. 4). Distant metastasis developed in 16 cases of robotic surgery and 14 cases of laparoscopic surgery, respectively. Port site or wound recurrences did not occur in any patients.
Prognostic Factors In univariate analysis, common factors that influence both the 5-year overall and disease-free survival are sex, TNM stage, and histologic grade. However, surgical methods between robotic and laparoscopic surgical procedures were not prognostic factors for the 5-year overall and disease-free survival rates in this study (Table 5). In multivariate analysis, the only prognostic factor impacting the 5-year disease-free survival was sex (P = 0.009, hazard ratio = 0.31), and prognostic factors impacting the 5-year overall survival were age (P = 0.003, hazard ratio = 7.65) and grade of differentiation [P = 0.002, hazard ratio = 16.29 (poor vs well), 15.03 (mucinous vs well)] (Table 6).
Cost
Total mean cost was US $10101.3 ± US $2804.8 for laparoscopic surgery, and US $12742.5 ± US $3509.9 for robotic surgery (P < 0.001). Under the mandatory Korean government medical insurance system (everyone has to enroll in the insurance system by law), the patient’s mean payment was US $4285.2 ± US $1255.1 for laparoscopic surgery and US $10029.4 ± US $2581.4 for robotic surgery (P < 0.001). The cost of robotic surgery was approximately 2.34 times higher than that of laparoscopic surgery. Robotic surgery is not covered by the health insurance system in Korea.
DISCUSSION In this study, the perioperative clinical outcomes were similar between robotic and laparoscopic surgical procedures. However, there were no patients who experienced conversion in robotic surgery. The reasons of conversion in laparoscopic surgery were severe narrow pelvic sidewall bleeding that might have been overcome if a more effective surgery was performed. Under this assumption, robotic surgery may have allowed for better visualization and a finer pelvic dissection within a narrow pelvic cavity and might have avoided conversion. The length of hospital stay was longer in the laparoscopic group for unclear reasons, although the significantly younger age of the patients in the robotic group (4.3 years younger, P = 0.007) may have influenced this outcome. C 2014 Lippincott Williams & Wilkins
Age, yr ≤65 >65 Sex Male Female BMI ≤25 >25 Tumor location Low (0–5 cm) Mid (5.1–10 cm) Upper (10.1–15 cm) ASA score 1 2 3 Surgical approach Robot Laparoscopy TNM stage I II III Histologic grade Well Moderate Poor Mucinous CRM Noninvolved Involved Conversion Yes No Lymphovascular invasion Yes No Neoadjuvant chemotherapy Yes No Perioperative morbidity Yes No
P
OS (%)
0.391 139 78
82.4 77.8
146 71
75.0 91.4
169 48
81.6 78.0
49 97 71
73.6 80.1 86.5
145 59 13
82.6 78.9 68.4
133 84
81.9 78.7
71 64 82
90.8 86.3 68.3
53 153 7 4
91.5 79.2 42.9 66.7
202 15
81.8 66.7
6 211
66.7 81.1
67 150
72.0 84.7
25 192
74.5 81.5
44 173
79.9 81.0
P 0.015
96.1 86.3 0.008
0.006 89.2 100
0.707
0.858 92.9 93.2
0.290
0.401 89.0 92.1 97.1
0.365
0.202 95.6 88.3 83.9
0.547
0.829 92.8 93.5
0.001
0.033 96.9 96.6 86.8
0.004
0.002 97.9 93.0 66.7 66.7
0.089
0.999 92.9 93.3
0.296
0.465 100 92.7
0.019
0.133 90.2 94.1
0.332
0.660 90.9 94.3
0.790
0.261 89.7 93.8
ASA indicates American Society of Anesthesiologists; BMI, body mass index; DFS, disease free survival; OS, overall survival.
Pathologic parameters that can access the quality of rectal surgery are CRM positivity and number of harvested lymph nodes of the resected specimen. The pathologic outcomes in this study showed no significant difference between robotic and laparoscopic surgical procedures. CRM involvement rate in this study was 6.8% in the robotic group and was comparable with other reports (0%– 16%).2,6,23–28 CRM positivity was not related to local recurrence. In the laparoscopic group, local recurrence occurred in 1 case with positive CRM. However, in the robotic group, local recurrence occurred in 3 cases with negative CRM. A positive CRM did not seem to be translated to local recurrence. These results were also observed in the previous CLASICC trial.6 In terms of the oncologic aspects, the 5-year disease-free survival rate of this study showed no significant difference between robot and laparoscopic surgical procedures (81.9% vs 78.7%, robot vs www.annalsofsurgery.com | 135
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
TABLE 6. Prognostic Factors of 5-Year Survival by Multivariate Analysis
Age, yr >65 vs ≤65 Sex Female vs male TNM stage II vs I III vs I Grade of differentiation Moderate vs well Poor vs well Mucinous vs well Lymphovascular invasion
Disease-Free Survival HR (95% CI)
Overall Survival P∗
—
— —
0.31 (0.13–0.74) 1.44 (0.47–4.44) 2.67 (0.94–7.54) 2.48 (0.83–7.40) 7.93 (1.73–36.28) 3.62 (0.39–33.28) 1.55 (0.76–3.16)
0.009 0.122 0.526 0.065 0.061 0.105 0.008 0.255 0.228
HR (95% CI) 7.65 (1.98–29.58)
Long-term Oncologic Outcomes of Robotic Low Anterior Resection for Rectal Cancer A Comparative Study With Laparoscopic Surgery Eun Jung Park, MD, Min Soo Cho, MD, Se Jin Baek, MD, PhD, Hyuk Hur, MD, Byung Soh Min, MD, PhD, Seung Hyuk Baik, MD, PhD, Kang Young Lee, MD, PhD, and Nam Kyu Kim, MD, PhD
Objective: The aim of this study is to evaluate long-term oncologic outcomes of robotic surgery for rectal cancer compared with laparoscopic surgery at a single institution. Background: Robotic surgery is regarded as a new modality to surpass the technical limitations of conventional surgery. Short-term outcomes of robotic surgery for rectal cancer were acceptable in previous reports. However, evidence of long-term feasibility and oncologic safety is required. Methods: Between April 2006 and August 2011, 217 patients who underwent minimally invasive surgery for rectal cancer with stage I–III disease were enrolled prospectively (robot, n = 133; laparoscopy, n = 84). Median follow-up period was 58 months (range, 4–80 months). Perioperative clinicopathologic outcomes, morbidities, 5-year survival rates, prognostic factors, and cost were evaluated. Results: Perioperative clinicopathologic outcomes demonstrated no significant differences except for the conversion rate and length of hospital stay. The 5-year overall survival rate was 92.8% in robotic, and 93.5% in laparoscopic surgical procedures (P = 0.829). The 5-year disease-free survival rate was 81.9% and 78.7%, respectively (P = 0.547). Local recurrence was similar: 2.3% and 1.2% (P = 0.649). According to the univariate analysis, this type of surgical approach was not a prognostic factor for long-term survival. The patient’s mean payment for robotic surgery was approximately 2.34 times higher than laparoscopic surgery. Conclusions: No significant differences were found in the 5-year overall, disease-free survival and local recurrence rates between robotic and laparoscopic surgical procedures. We concluded that robotic surgery for rectal cancer failed to offer any oncologic or clinical benefits as compared with laparoscopy despite an increased cost. Keywords: laparoscopic surgery, oncologic outcome, rectal cancer, robotic surgery, total mesorectal excision (Ann Surg 2015;261:129–137)
M
inimally invasive surgery for colorectal cancer is a relatively new surgical modality, able to substitute for conventional open surgery on the basis of similar or better perioperative and oncologic outcomes.1–4 According to the long-term follow-up of the United Kingdom Medical Research Council Conventional versus Laparoscopic-Assisted Surgery in Colorectal Cancer (UK MRC CLASICC) trial, laparoscopic surgery was regarded as an alternative to conventional surgery.5,6
From the Section of Colon and Rectal Surgery, Department of Surgery, Yonsei University College of Medicine, Seoul, South Korea. Disclosure: The authors have no conflicts of interest or financial ties of disclosure. Reprints: Seung Hyuk Baik, MD, PhD, Section of Colon and Rectal Surgery, Department of Surgery, Yonsei University College of Medicine, Seoul, 120752, South Korea. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26101-0129 DOI: 10.1097/SLA.0000000000000613
Annals of Surgery r Volume 261, Number 1, January 2015
However, in rectal cancer, surgeons are faced with challenging conditions such as a narrow pelvic cavity, anatomical complexity, and restricted surgical view during laparoscopic surgery, although previous studies reported that laparoscopic rectal cancer surgery was feasible.2,5–7 Thus, the robotic system seems potentially suited for the surgical treatment of rectal cancer due to several theoretical advantages and has been used since it was adopted in 2001.8 However, there are not sufficient data to justify the adoption of the robotic system instead of laparoscopic surgery for rectal cancer. Several previous studies documented that robotic surgery is equivalent to laparoscopic surgery with respect to the short-term perioperative and oncologic outcomes.9–15 In rectal cancer surgery, the most important issue is definitely the long-term oncologic outcomes.9–16 To date, there are no reports to evaluate the long-term oncologic outcomes of robotic surgery compared with laparoscopic surgery for rectal cancer. We hypothesized that the long-term oncologic outcomes would be better in robotic rectal cancer surgery than in laparoscopic rectal cancer surgery. Therefore, this study aimed to evaluate the long-term oncologic outcomes of robotic surgery for rectal cancer compared with conventional laparoscopic surgery. Moreover, the difference of the cost between both procedures was evaluated.
METHODS Study Population and Patient Selection Between April 2006 and August 2011, 288 consecutive patients, who were diagnosed with rectal cancer (adenocarcinoma), underwent a low anterior resection (LAR) by a single surgeon (Baik SH) at Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea. The first laparoscopic-assisted LAR and the first robotic-assisted LAR were performed on April 2006 and June 2006, respectively. Thus, the learning curve period of both procedures was equivalent. Before surgery, all patients were informed of the detailed characteristics of both robotic and laparoscopic surgical procedures, as explained by the nurse coordinator. It was explained that there was no evidence that one procedure is better than the other (although theoretical advantages of the robot in terms of facilitating pelvic dissection were explained) but that both have shown better short-term benefits over open approaches for the treatment of rectal cancer and that due to the health system in Korea, they would have to pay an additional US $6000 for choosing the robotic approach over the laparoscopic approach. After informed consent, therefore, the patients decided their preferred approach. Open surgery was performed when primary tumor invasion to other organs and structures was observed in preoperative imaging studies. Among these 288 patients, 50 patients who underwent open surgery, 20 patients with stage IV, and 1 patient lost to follow-up were excluded from the study population. Finally, 217 patients were enrolled in this study, which included robotic-assisted LAR (n = 133) www.annalsofsurgery.com | 129
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
using the da Vinci surgical system (Intuitive Surgical, Sunnyvale, CA) and laparoscopic-assisted LAR (n = 84) (Fig. 1). This study was approved by the Institutional Review Board of Severance Hospital.
Data Collection and Evaluation Parameters Data were collected from the Yonsei Colorectal Cancer Database for this study. Clinicopathologic results were evaluated retrospectively using this database, and oncologic outcomes of the
FIGURE 1. Flowchart of patient selection. TABLE 1. Patient Characteristics Robot (n = 133) Age∗ ,
yr Sex‡ Male Female Weight∗ , kg Height∗ , m BMI∗ , kg/m2 ASA score‡ 1 2 3 Tumor location from anal verge‡ Low (0–5 cm) Mid (5.1–10 cm) Upper (10.1–15 cm) History of abdominal surgery‡ Preoperative CRT‡ Yes No
Laparoscopy (n = 84)
59.2 ± 11.4 (32–86)
63.5 ± 11.2 (29–83)
86 (64.7) 47 (35.3) 62.5 ± 9.9 (37–86) 1.64 ± 0.07 (1.44–1.78) 23.1 ± 2.9 (14.6–32.8)
60 (71.4) 24 (28.6) 61.0 ± 9.4 (32–85) 1.63 ± 0.08 (1.44–1.80) 22.9 ± 2.8 (14.6–29.4)
94 (70.7) 31 (23.3) 8 (6.0)
51 (60.7) 28 (33.3) 5 (6.0)
33 (24.8) 60 (45.1) 40 (30.1) 10 (7.5)
16 (19.1) 37 (44.0) 31 (36.9) 11 (13.1)
15 (11.3) 118 (88.7)
10 (11.9) 74 (88.1)
P 0.007† 0.301§ 0.272† 0.256† 0.631† 0.263§
0.471§
0.176§ 0.888§
∗
Mean ± standard deviation (range). †Independent 2-sample t test. ‡n (%). §χ 2 test. ASA indicates American Society of Anesthesiologists; BMI, body mass index; CRT, chemoradiotherapy.
130 | www.annalsofsurgery.com
C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
TABLE 2. Operation Details and Perioperative Clinical Outcomes Operation time∗ , min Diverting ileostomy‡ Conversion‡ Intraoperative bleeding∗ , mL Intraoperative complications or events‡ Bladder injury Perforation of the rectum Disruption of colorectal anastomosis Days to 1st gas passing∗ , d Days to 1st bowel movement∗ , d Days to 1st soft diet∗ , d Length of hospital stay∗ , day
Robot (n = 133)
Laparoscopy (n = 84)
P
205.7 ± 67.3 (109–505) 29 (21.8) 0 (0.0) 77.6 ± 153.2 (0–700)
208.8 ± 81.2 (94–407) 20 (23.8) 6 (7.1) 82.3 ± 185.8 (0–1100)
0.766† 0.731§ 0.003¶ 0.841† 1.000¶
1 (0.8) 0 (0.0) 1 (0.8) 2.42 ± 0.93 (1–6) 3.61 ± 1.04 (1–7) 4.92 ± 1.35 (3–11) 5.86 ± 1.43 (4–14)
1 (1.2) 0 (0.0) 0 (0.0) 2.47 ± 1.10 (1–6) 4.07 ± 1.96 (1–8) 5.19 ± 1.62 (2–11) 6.54 ± 2.65 (3–25)
0.727† 0.143† 0.197† 0.035†
∗
Mean ± standard deviation(range). †Independent 2-sample t test. ‡n (%). §χ 2 test. ¶Fisher’s exact test.
TABLE 3. Perioperative and Long-term Postoperative Complications Robot (n = 133) Perioperative complications∗ Grade I‡ Ileus Voiding difficulty Wound infection Pleural effusion Retrograde ejaculation Grade II‡ Ascites Chyloperitoneum Blood transfusion due to bleeding Grade III‡ Anastomotic leakage Anastomotic site bleeding Ischemic colitis Intra-abdominal abscess Grade IV‡ Asthma attack Pneumonia Grade V‡ Long-term postoperative complications§ Grade I‡ Ileus Voiding difficulty Grade II‡ Grade III‡ Adhesion Incisional hernia Anastomotic stricture Rectovaginal/rectovesical fistula Grade IV‡ Grade V‡ Major perioperative complications‡ Major long-term complications‡
Laparoscopy (n = 84)
P 0.897†
11 (8.3) 3 5 2 0 1 5 (3.8) 2 2 1 9 (6.8) 6 2 1 0 1 (0.8) 0 1 0 (0.0)
7 (8.3) 3 3 0 1 0 4 (4.8) 0 1 3 6 (7.1) 3 3 0 0 2 (2.4) 1 1 0 (0.0)
3 (2.3) 1 2 0 (0.0) 2 (1.5) 1 0 1 0 0 (0.0) 0 (0.0) 10 (7.5) 2 (1.5)
2 (2.4) 1 1 0 (0.0) 3 (3.6) 0 2 1 0 0 (0.0) 0 (0.0) 8 (9.5) 3 (3.6)
0.687†
0.602¶ 0.378†
Major complications were defined as complications with a grade III and higher of the Clavien–Dindo classification. ∗ Complications within 30 days from operation date. †Fisher’s exact test. ‡n (%). §New complications 30 days after operation date ¶χ 2 test.
C 2014 Lippincott Williams & Wilkins
www.annalsofsurgery.com | 131
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
TABLE 4. Postoperative Pathologic Outcomes Robot (n = 133)
Laparoscopy (n = 84)
stage∗
TNM I II III Grade of differentiation∗ Well Moderate Poor Mucinous No. nodes analyzed§ PRM§, cm DRM§, cm Lymphovascular invasion∗ CRM∗ Noninvolved (>1 mm) Involved (≤1 mm)
P 0.255†
49 (36.8) 36 (27.1) 48 (36.1)
22 (26.2) 28 (33.3) 34 (40.5)
34 (25.5) 93 (69.9) 5 (3.8) 1 (0.8) 16.34 ± 8.79 (2–43) 11.17 ± 4.64 (4–35) 2.75 ± 2.14 (1–14) 43 (32.3)
19 (22.6) 60 (71.4) 2 (2.4) 3 (3.6) 16.63 ± 10.24 (2–49) 10.51 ± 5.65 (4–34) 2.87 ± 1.63 (1–9) 24 (28.6)
124 (93.2) 9 (6.8)
78 (92.9) 6 (7.1)
0.467‡
0.823¶ 0.352¶ 0.652¶ 0.559† 0.915†
∗
n (%). †χ 2 test. §Mean ± standard deviation (range). ‡Fisher’s exact test. ¶Independent 2-sample t test. DRM indicates distal resection margin; PRM, proximal resection margin.
entire enrolled patients were updated until May 2013. Insufficient data were reevaluated by electronic medical charts and telephone interviews were conducted, if needed. The preoperative status of patients was categorized by the American Society of Anesthesiologists Classification. The location of the tumor was divided as: low (below 5 cm), mid (5.1–10 cm) and upper rectum (10.1–15 cm). Conversion was defined as the unintended extension of laparotomy beyond the routine length of incision (4 cm) to complete the surgical procedures. Postoperative complications were stratified by the Clavien– Dindo classification of surgical complications.17 The total numbers of postoperative complications were counted for all cases related to morbidity. Major complications were defined as complications with a grade III and higher of the Clavien–Dindo classification.17 Longterm complications were defined as postoperative complications that occurred after the 30th postoperative day. For analysis of the pathologic outcomes, tumor node metastasis (TNM) stage (American Joint Committee on Cancer, seventh edition), grade of histologic differentiation, resection margins, numbers of harvested lymph nodes, lymphovascular invasion, and circumferential resection margin (CRM) were evaluated. Involved CRM was defined when the tumor was located 1 mm or less from the CRM.18 Examination of the resected specimens was performed using the protocol of the College of American Pathologists.19 Visual enhancement techniques were used when fewer than 12 nodes were evaluated. The preoperative clinical diagnosis was determined by abdominopelvic computed tomography (CT), chest CT, rectal magnetic resonance imaging, and transrectal ultrasonography. Tissue diagnosis was obtained by a colonoscopic biopsy. Local recurrence was defined as the relapse of the tumor at the primary site confirmed by radiological or histological evidence. Simultaneous local and systemic recurrences were counted as a local recurrence. Distant metastasis was considered as metastatic lesions that were diagnosed in other organs beyond the primary site. Total cost was evaluated and defined as total expenses that occurred from admission to discharge and also included all charges for the operation, perioperative management, and treatment. 132 | www.annalsofsurgery.com
Surgical Technique All cases of robotic-assisted LAR in this study were performed by the hybrid technique, which is composed of 2 surgical steps. The first step is performed by conventional laparoscopic instruments for the inferior mesenteric vessel ligation and left colon and splenic flexure mobilization. The second step is pelvic dissection using the robotic system on the basis of total mesorectal excision or tumor-specific mesorectal excision principles according to the tumor location from the anal verge.20,21 Tumor-specific mesorectal excision is a surgical method to dissect the mesorectum encircled by the proper fascia of the rectum below 4 cm from the lower margin of the tumor.22 Laparoscopic-assisted LAR was performed in the same manner as robotic surgery using laparoscopic instruments. Further detailed procedures of both robotic and laparoscopic surgical methods are described in our previous reports.21–23
Chemoradiotherapy and Follow-up Patients received chemoradiotherapy for stage T3–4N0 or N + M0 low- and mid-rectal cancers, with the majority of the patients undergoing postoperative treatment because of the standard protocols in place at our institution early in the study period. Thus, mainly, postoperative chemoradiotherapy was performed. The regimen was based on 5-fluorouracil and leucovorin. The discharged patients visited the outpatient clinic after 1 month, 3 months, and every 3 months for the first 3 years and then 6 months until 5 years after surgery. In the outpatient clinic, regular laboratory with carcinoembryonic antigen and a physical examination were performed. The patient was examined by colonoscopy 1 year after surgery and then 5 years. Chest and abdominopelvic computed tomographic scans were obtained every 6 months to detect local recurrence or systemic metastasis during the follow-up period.
Statistical Analysis Statistical analysis was performed using the SPSS program (Statistical Product and Service Solution 18 for Windows; SPSS Inc., Chicago, IL). The Student t test was used for continuous variables. C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
FIGURE 2. The 5-year overall survival rate between robotic and laparoscopic surgical procedures. The categorical variables were analyzed by the χ 2 test or Fisher’s exact test. The 5-year overall, disease-free survival rates and cumulative incidence of local recurrence were calculated by using the Kaplan– Meier method. Comparison of the survival and local recurrence rates between the 2 surgical methods was analyzed by the log-rank test. Univariate analysis of clinicopathological factors upon overall and disease-free survival was performed using the log-rank test to determine the prognostic value of the surgical methods (robot vs laparoscopy). All statistically significant factors determined by univariate analysis were conducted for multivariate analysis by the Cox proportional hazards regression model with a forward selection of variables. A P value less than 0.05 was considered statistically significant.
RESULTS Patient Characteristics
The mean age of robotic surgery was 59.2 ± 11.4 (mean ± standard deviation) years, and 63.5 ± 11.2 years for the laparoscopic group (P = 0.007). There were no significant differences between the groups in sex, weight, height, and body mass index. An American C 2014 Lippincott Williams & Wilkins
Society of Anesthesiologists score of 1 showed the highest distribution in both groups (70.7% vs 60.7%, robot vs laparoscopy) and was not significantly different between the groups. Tumor location and history of abdominal surgery were not different significantly between the groups. The rate of low– or mid–rectal cancer was 69.9% in the robotic group and 63.1% in the laparoscopic group. The rate of patients who underwent preoperative chemoradiotherapy was similar between the groups (Table 1).
Perioperative Clinical Outcomes, Complications
The mean operation time was 205.7 ± 67.3 minutes in robotic surgery and 208.8 ± 81.2 minutes in laparoscopic surgery (P = 0.766). There were no significant differences between the groups for intraoperative complications or bleeding or undergoing a diverting ileostomy. However, conversion occurred more frequently in the laparoscopic group (P = 0.003). There was no conversion in the robotic group compared with 6 cases in the laparoscopic group. The reasons for conversion of laparoscopic surgery were intraoperative pelvic sidewall bleeding (1 case), rectal perforation (1 case), tumor adhesion to the bladder (1 case), and severe narrow pelvic cavity (3 cases) (P = 0.003). Days to first gas passing, first bowel movement, and soft diet were not significantly different between the groups. However, patients who underwent laparoscopic surgery stayed in the hospital www.annalsofsurgery.com | 133
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
FIGURE 3. The 5-year disease-free survival rate between robotic and laparoscopic surgical procedures. longer than the robotic group, although the difference was only 0.68 days (P = 0.035) (Table 2). The occurrence rate of both perioperative and long-term complications was similar between robotic and laparoscopic surgical procedures (perioperative: P = 0.897, long term: P = 0.687). Anastomotic leakage rate was 4.5% in the robotic group and 3.6% in the laparoscopic group. Major complications were not significantly different. There were no 30-day mortalities in both groups (Table 3).
Postoperative Pathologic Assessment There was no significant difference between the robotic group and the laparoscopic group in the pathologic outcomes. TNM stage, histologic grade of differentiation, resection margins, and lymphovascular invasion showed similar results in both groups. The factors that reflected the quality of resected specimen such as the number of harvested lymph nodes, distal resection margin, and CRM were not significantly different. Number of nodes analyzed was 16.34 ± 8.79 (range: 2–43) in the robotic group and 16.63 ± 10.24 (range: 2–49) in the laparoscopic group (P = 0.823). Rate of positive CRM was 6.8% in the robotic group and 7.1% in the laparoscopic group and was not significantly different between the groups (P = 0.915) (Table 4). 134 | www.annalsofsurgery.com
Oncologic Outcomes
The mean follow-up period was 54.4 ± 17.3 (range: 4–80 months) (interquartile range: 49–66) months and the median followup period was 58 (range: 4–80 months) months. The 5-year overall survival rate was 92.8% in the robotic group and 93.5% in the laparoscopic group (P = 0.829). The 5-year overall survival according to stages of the robotic group was 97.8% in stage I, 94.2% in stage II, and 86.8% in stage III. In the laparoscopic group, the 5-year overall survival according to stages of the laparoscopic group was 95.0% in stage I, 100% in stage II, and 87.8% in stage III. The 5-year overall survival rate according to each stage was similar between the robotic and the laparoscopic groups (Fig. 2). The 5-year disease-free survival rate was 81.9% in the robotic group and 78.7% in the laparoscopic group (P = 0.547). The 5-year disease-free survival rates of the robotic group according to stages were 93.7% in stage I, 82.1% in stage II, and 70.3% in stage III. In the laparoscopic group, the 5-year disease-free survival rates according to stages were 83.1% in stage I, 92.9% in stage II, and 65.0% in stage III. There was no statistical difference of the 5-year disease-free survival according to each stage between the groups (Fig. 3). Cumulative incidence of local recurrence was 2.3% in the robotic group and 1.2% in the laparoscopic group (P = 0.649) C 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Outcomes of Robotic Rectal Cancer Surgery
TABLE 5. Prognostic Factors of 5-Year Survival by Univariate Analysis n = 217 DFS (%)
FIGURE 4. Cumulative incidence of local recurrence rate of all patients with robotic or laparoscopic surgical procedures for rectal cancer: robotic 2.3% and laparoscopic 1.2% (P = 0.649).
(Fig. 4). Distant metastasis developed in 16 cases of robotic surgery and 14 cases of laparoscopic surgery, respectively. Port site or wound recurrences did not occur in any patients.
Prognostic Factors In univariate analysis, common factors that influence both the 5-year overall and disease-free survival are sex, TNM stage, and histologic grade. However, surgical methods between robotic and laparoscopic surgical procedures were not prognostic factors for the 5-year overall and disease-free survival rates in this study (Table 5). In multivariate analysis, the only prognostic factor impacting the 5-year disease-free survival was sex (P = 0.009, hazard ratio = 0.31), and prognostic factors impacting the 5-year overall survival were age (P = 0.003, hazard ratio = 7.65) and grade of differentiation [P = 0.002, hazard ratio = 16.29 (poor vs well), 15.03 (mucinous vs well)] (Table 6).
Cost
Total mean cost was US $10101.3 ± US $2804.8 for laparoscopic surgery, and US $12742.5 ± US $3509.9 for robotic surgery (P < 0.001). Under the mandatory Korean government medical insurance system (everyone has to enroll in the insurance system by law), the patient’s mean payment was US $4285.2 ± US $1255.1 for laparoscopic surgery and US $10029.4 ± US $2581.4 for robotic surgery (P < 0.001). The cost of robotic surgery was approximately 2.34 times higher than that of laparoscopic surgery. Robotic surgery is not covered by the health insurance system in Korea.
DISCUSSION In this study, the perioperative clinical outcomes were similar between robotic and laparoscopic surgical procedures. However, there were no patients who experienced conversion in robotic surgery. The reasons of conversion in laparoscopic surgery were severe narrow pelvic sidewall bleeding that might have been overcome if a more effective surgery was performed. Under this assumption, robotic surgery may have allowed for better visualization and a finer pelvic dissection within a narrow pelvic cavity and might have avoided conversion. The length of hospital stay was longer in the laparoscopic group for unclear reasons, although the significantly younger age of the patients in the robotic group (4.3 years younger, P = 0.007) may have influenced this outcome. C 2014 Lippincott Williams & Wilkins
Age, yr ≤65 >65 Sex Male Female BMI ≤25 >25 Tumor location Low (0–5 cm) Mid (5.1–10 cm) Upper (10.1–15 cm) ASA score 1 2 3 Surgical approach Robot Laparoscopy TNM stage I II III Histologic grade Well Moderate Poor Mucinous CRM Noninvolved Involved Conversion Yes No Lymphovascular invasion Yes No Neoadjuvant chemotherapy Yes No Perioperative morbidity Yes No
P
OS (%)
0.391 139 78
82.4 77.8
146 71
75.0 91.4
169 48
81.6 78.0
49 97 71
73.6 80.1 86.5
145 59 13
82.6 78.9 68.4
133 84
81.9 78.7
71 64 82
90.8 86.3 68.3
53 153 7 4
91.5 79.2 42.9 66.7
202 15
81.8 66.7
6 211
66.7 81.1
67 150
72.0 84.7
25 192
74.5 81.5
44 173
79.9 81.0
P 0.015
96.1 86.3 0.008
0.006 89.2 100
0.707
0.858 92.9 93.2
0.290
0.401 89.0 92.1 97.1
0.365
0.202 95.6 88.3 83.9
0.547
0.829 92.8 93.5
0.001
0.033 96.9 96.6 86.8
0.004
0.002 97.9 93.0 66.7 66.7
0.089
0.999 92.9 93.3
0.296
0.465 100 92.7
0.019
0.133 90.2 94.1
0.332
0.660 90.9 94.3
0.790
0.261 89.7 93.8
ASA indicates American Society of Anesthesiologists; BMI, body mass index; DFS, disease free survival; OS, overall survival.
Pathologic parameters that can access the quality of rectal surgery are CRM positivity and number of harvested lymph nodes of the resected specimen. The pathologic outcomes in this study showed no significant difference between robotic and laparoscopic surgical procedures. CRM involvement rate in this study was 6.8% in the robotic group and was comparable with other reports (0%– 16%).2,6,23–28 CRM positivity was not related to local recurrence. In the laparoscopic group, local recurrence occurred in 1 case with positive CRM. However, in the robotic group, local recurrence occurred in 3 cases with negative CRM. A positive CRM did not seem to be translated to local recurrence. These results were also observed in the previous CLASICC trial.6 In terms of the oncologic aspects, the 5-year disease-free survival rate of this study showed no significant difference between robot and laparoscopic surgical procedures (81.9% vs 78.7%, robot vs www.annalsofsurgery.com | 135
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 1, January 2015
Park et al
TABLE 6. Prognostic Factors of 5-Year Survival by Multivariate Analysis
Age, yr >65 vs ≤65 Sex Female vs male TNM stage II vs I III vs I Grade of differentiation Moderate vs well Poor vs well Mucinous vs well Lymphovascular invasion
Disease-Free Survival HR (95% CI)
Overall Survival P∗
—
— —
0.31 (0.13–0.74) 1.44 (0.47–4.44) 2.67 (0.94–7.54) 2.48 (0.83–7.40) 7.93 (1.73–36.28) 3.62 (0.39–33.28) 1.55 (0.76–3.16)
0.009 0.122 0.526 0.065 0.061 0.105 0.008 0.255 0.228
HR (95% CI) 7.65 (1.98–29.58)
