Sie sind auf Seite 1von 8

CE: T.M.

; ANNSURG-D-17-02059; Total nos of Pages: 8;


ANNSURG-D-17-02059

ORIGINAL ARTICLE

Efficacy and Safety of Complete Mesocolic Excision in Patients


With Colon Cancer
Three-year Results From a Prospective, Nonrandomized, Double-blind,
Controlled Trial
Zhidong Gao, MD,  y Chao Wang, MD,  z Yancheng Cui, MD,  Zhanlong Shen, MD,yz Kewei Jiang, MD,  z
Danhua Shen, MD,§ Yi Wang, MD,ô Siyan Zhan, PhD,jj Peng Guo, MD,  Xiaodong Yang, MD, 
Fangfang Liu, PhD,§ Kai Shen, MD,  Bin Liang, MD,  Mujun Yin, MD,  Qiwei Xie, MD,  Youli Wang, MD, 
Shan Wang, MD, PhD,  y and Yingjiang Ye, MD, PhD  z

deposits (97.2% vs 91.6%, log-rank P < 0.022) were also associated with
Objective: The aim of the study was to evaluate the oncological outcomes of
improved LRFS.
complete mesocolic excision (CME) in colon cancer patients.
Conclusions: Our findings demonstrate that, compared with NCME, CME
Summary Background Data: CME is considered a standard procedure for
improves 3-year LRFS without increasing surgical risks.
colon cancer patients. However, previous evidence regarding the effect of
CME on prognosis has fundamental limitations that prevent it from being Keywords: colon cancer, complete mesocolic excision, outcome, safety,
fully accepted. survival
Methods: Patients who underwent radical resection for colon cancer were
(Ann Surg 2018;xx:xxx–xxx)
enrolled between November 2012 and March 2016. According to the prin-
ciples of CME, patients were stratified into 2 groups based on intraoperative
surgical fields and specimen photographs. The primary outcome was local
recurrence-free survival (LRFS). The clinicopathological data and follow-up T here has been a gradual shift from incidents of rectal to colon
tumors among colorectal cancer cases.1 Consequently, surgeons
have focused on the global standardization and quality control of
information were collected and recorded. The final follow-up date was
April 2016. The trial was registered in ClinicalTrials.gov (identifier: colon cancer surgery. Owing to simple anatomical structure and
NCT01724775). concise surgical technique, colon cancer prognosis 30 years ago was
Results: There were 220 patients in the CME group and 110 patients in the significantly better than that of rectal cancer.2 However, a recent
noncomplete mesocolic excision (NCME) group. Baseline characteristics study showed that rectal cancer patients had a significantly reduced
were well balanced. Compared with NCME, CME was associated with a local recurrence rate and an improved 5-year relative survival
greater number of total lymph nodes (24 vs 20, P ¼ 0.002). Postoperative compared with those with colon cancer.3 One reason was the
complications did not differ between the 2 groups. CME had a positive effect introduction of total mesorectal excision, which led to revolutionary
on LRFS compared with NCME (100.0% vs 90.2%, log-rank P < 0.001). improvements in rectal cancer surgery, thereby improving progno-
Mesocolic dissection (100.0% vs 87.9%, log-rank P < 0.001) and nontumor sis.4 In contrast, colon cancer surgery remained stagnant in Western
countries until Hohenberger et al5 proposed complete mesocolic
excision (CME) as a standardized surgical technique in 2009. CME is
based on the principles of total mesorectal excision, where the entire
From the Department of Gastrointestinal Surgery, Peking University People’s embryological mesocolon, from the viscera to the parietal planes, is
Hospital, Beijing, China; yKey Laboratory of Colorectal Cancer Diagnosis and
Treatment Research, Beijing, China; zLaboratory of Surgical Oncology, removed in the horizontal direction and the main artery pedicle is
Peking University People’s Hospital, Beijing, China; §Department of Pathol- ligated at the root in the vertical direction. CME advocated elimi-
ogy, Peking University People’s Hospital, Beijing, China; ôDepartment of nating tumor tissue by removing the entire mesentery, which is
Radiology, Peking University People’s Hospital, Beijing, China; and associated with reduced local recurrence rates and better survival.6
jjDepartment of Epidemiology, Peking University Health Science Center,
Beijing, China. A recent study showed that CME is more effective for improving
This work was supported by the Specialized Research Fund for the Science and specimen quality and survival, but has a higher complication rate.7
Technology Project of Beijing (Z161100000516104), the Doctoral Program of However, evidence regarding CME is limited and remains contro-
Higher Education (20130001120064), and the Capital Health Development versial due to a lack of high-quality prospective studies. In addition,
Research Project (TG-2015-002) of China.
The authors report no conflicts of interest. previous reports of substantial intraoperative injury have compli-
Supplemental digital content is available for this article. Direct URL citations cated the acceptance of CME.8
appear in the printed text and are provided in the HTML and PDF versions of To investigate whether CME is safe and effective for colon
this article on the journal’s Web site (www.annalsofsurgery.com). cancer treatment, we performed a prospective, nonrandomized,
Reprints: Yingjiang Ye, MD, PhD, Laboratory of Surgical Oncology, Department
of Gastrointestinal Surgery, Peking University People’s Hospital, 11 Xizhimen double-blind, controlled trial.
Nan Street, Xicheng District, Beijing 100044, P. R. China.
E-mail: yeyingjiang@pkuph.edu.cn; Shan Wang, MD, PhD, Key Laboratory of
Colorectal Cancer Diagnosis and Treatment Research, Department of Gastro- METHODS
intestinal Surgery, Peking University People’s Hospital, 11 Xizhimen Nan
Street, Xicheng District, Beijing 100044, P. R. China. Ethical Considerations
E-mail: shanwang@pkuph.edu.cn. The study was approved by the Ethics Committee of Peking
Copyright ß 2018 Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 0003-4932/16/XXXX-0001 University People’s Hospital (Beijing, China) and complied with the
DOI: 10.1097/SLA.0000000000003012 guidelines of the responsible government agency. The trial was

Annals of Surgery  Volume XX, Number XX, Month 2018 www.annalsofsurgery.com | 1

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Gao et al Annals of Surgery  Volume XX, Number XX, Month 2018

registered with ClinicalTrial.gov (NCT01724775) in 2012. All par- Statistics


ticipants provided written informed consent.
Sample Size Calculation
Trial Design and Study Participants The primary outcome of the ESCME trial was 5-year LRFS.
The calculation assumed that 9.0% of patients in the NCME group
Study Design would develop local recurrence compared with 1.0% in the CME
The Efficacy Study of Complete Mesocolic Excision group (absolute CME upstaging rate: 8.0%). The sampling rate was
(ESCME) is an ongoing prospective, nonrandomized, double-blind, 1:2. To control the probability of a type I error at a ¼ 0.05, a sample
controlled trial to determine the efficacy and safety of CME for a size of 107 patients was required for the NCME group, with a
duration of up to 5 years from enrollment in colon cancer patients. statistical power of 80.0% and a 5.0% significance level for a total
Colon cancer patients at Peking University People’s Hospital (Bei- of 321 subjects. Enrollment of up to 350 subjects was anticipated
jing, China) were recruited between November 2012 and November over a 5-year period.
2017. The primary objective of the ESCME trial was to assess the 5-
year local recurrence-free survival (LRFS) of CME. The safety of Statistical Analyses
CME was also assessed as a secondary objective. Continuous variables are presented as the mean (standard
This medium-term (3 yr) outcome report includes patients deviation) or median (first to third quartile) and compared using
enrolled between November 2012 and March 2016. All patients were Pearson’s or Spearman’s rank analysis. Categorical variables are
followed up until April 2016. presented as the frequencies with percentages. Baseline character-
istics were compared using the Student t test, Mann–Whitney U test,
Inclusion and Exclusion Criteria x2 test, or Fisher exact test, as appropriate. The correlations between
The inclusion criteria were as follows: age >18 years, endo- specimen morphometry [large bowel length (LBL), distance from
scopic biopsy proven primary colon cancer (including mucinous the tumor (DVT) or large bowel (DVB) to the central vascular
adenocarcinoma and signet-ring cell carcinoma), and acceptance of ligation, and area of mesentery (AM)] and body indications [height,
open surgery with written informed consent. Exclusion criteria weight, body mass index (BMI), and body surface area (BSA)] were
included emergency surgery, synchronous or metachronous multiple analyzed using Spearman’s rank analysis (rs  0.20 was considered
primary colorectal cancer, preoperative neoadjuvant chemoradio- acceptable).
therapy, distant metastasis, and other malignant diseases. Univariate and multivariate analyses were conducted for the
assessment of risk factors. In the univariate analysis, variables were
Surgery and Specimen Retrieval compared using the Student t test, Mann–Whitney U test, x2 test,
High-resolution digital color photographs of the surgical field Fisher exact test, or univariate logistic regression, as appropriate.
were immediately taken after removing specimens from the abdom- Variables that were significant in the univariate analysis were
inal cavity. Specimens were stretched without tension on a table and included in the multivariate analysis. Linear or logistic regression
images in 3 planes were obtained (Supplemental Figure 1A–D, models were used based on the distribution of the dependent var-
http://links.lww.com/SLA/B501). Two third-party experts who had iables. The determination of multicollinearity was also performed as
not participated in patient management evaluated the images from 2 required using the independent variables. Regression coefficients [b
perspectives: (1) the surgical field to assess whether the main values/odds ratios (OR)] and 95.0% confidence intervals (CIs)
vascular trunk was exposed and to determine whether the surgery were calculated.
was a central ligation (Supplemental Figure 1A, http://links. Overall survival (OS), disease-free survival (DFS), LRFS, and
lww.com/SLA/B501), and (2) the anterior, posterior, and cross- metastasis-free survival (MFS) curves were created using the
sectional aspects of the specimen to assess whether the visceral Kaplan–Meier method and compared by the log-rank test. OS
and parietal fascia were intact and to determine whether the surgery was calculated from the date of surgery to the date of death from
was a mesocolic dissection (Supplemental Figure 1B–D, http:// any cause. DFS was calculated from the date of surgery to the date of
links.lww.com/SLA/B501). After the evaluation, the third-party the first recurrence (local, distant, or mixed) or death from any cause.
experts divided the patients into 2 groups: the CME group (mesocolic LRFS was calculated from the date of surgery to the date of the first
dissection and central ligation satisfied simultaneously) and the local recurrence. MFS was calculated from the date of surgery to the
noncomplete mesocolic excision (NCME) group (mesocolic dissec- date of the first metastasis.
tion or central ligation satisfied individually, or neither). Specimen All statistical analyses were conducted using SPSS for Win-
morphometry was performed using ImageScope version 10 (Aperio dows (version 20.0; IBM Corp., Armonk, NY) and R (version 3.3.3;
Technologies Inc., Vista, CA) according to the protocol devised by The R Foundation for Statistical Computing, Vienna, Austria). A 2-
West et al.9 sided P < 0.05 was considered statistically significant.
Both third-party experts and the surgeons were blinded. The
third-party experts did not have access to the patients’ information or RESULTS
know who had performed the operation and the surgeons were
blinded to the results of the evaluation. Patient Characteristics
In total, 346 patients were enrolled between November 2012
Specimen Processing and March 2016. Sixteen patients were excluded due to insufficient
After the photographs were taken, a colorectal surgeon clas- third-party evaluation. Three-hundred thirty patients were included
sified lymph nodes as pericolic, intermediate, or main lymph nodes in the final analysis (CME group, N ¼ 220; NCME group, N ¼ 110)
according to the Japanese Society for Cancer of the Colon and (Supplemental Figure 3, http://links.lww.com/SLA/B501). All
Rectum classification of lymph node stations (Supplemental patients in the CME group satisfied both the mesocolic dissection
Table 1 and Supplemental Figure 2, http://links.lww.com/SLA/ and central ligation criteria. In the NCME group, 19 patients satisfied
B501).10 Pericolic lymph nodes were divided into 3 subgroups based the mesocolic dissection criteria only, 63 patients satisfied the central
on distance from the tumor: <5.0 cm, 5.0 to 10.0 cm, and >10.0 cm. ligation criteria only, and 28 patients satisfied neither the mesocolic

2 | www.annalsofsurgery.com ß 2018 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Annals of Surgery  Volume XX, Number XX, Month 2018 Outcomes of Complete Mesocolic Excision

TABLE 1. Pathological Characteristics and Morphological Measurements


Characteristic or Measurement CME (N ¼ 220) NCME (N ¼ 110) P
Tumor diameter, cm, median (range) 5 (4–6) 5 (4–7) 0.135
Number of total LNs, median (range) 24 (17–35) 20 (16–27) 0.002
Number of positive LNs, median (range) 0 (0–2) 0 (0–1) 0.545
Histological type, N (%)
Adenocarcinoma 204 (92.7) 100 (90.9)
Mucinous adenocarcinoma 16 (7.3) 10 (9.1) 0.563
Tumor differentiation, N (%)
Low 57 (25.9) 28 (25.5)
Moderate 153 (69.6) 76 (69.1)
High 10 (4.5) 6 (5.4) 0.936
Tumor deposits, N (%) 19 (8.6) 7 (6.4) 0.470
Lymphovascular invasion, N (%) 69 (31.4) 24 (21.8) 0.069
Cecum and ascending colon cancer
Area of mesentery, mm2, median (range) 13,052.4 (10,894.1–15,571.5) 9,093.6 (7,206.9–10,549.1) <0.001
Distance from the VT to the tumor, mm, mean (SD) 129.2 (21.6) 113.2 (24.6) <0.001
Distance from the VT to the large bowel, mm, median (range) 87.7 (80.5–100.9) 70.6 (65.0–86.4) <0.001
Large bowel length, mm, median (range) 238.4 (208.1–269.2) 229.9 (199.0–266.8) 0.203
Sigmoid colon cancer
Area of mesentery, mm2, median (range) 10,317.4 (8,299.9–13,160.3) 7,475.5 (5,996.7–8,980.4) <0.001
Distance from the VT to the tumor, mm, mean (SD) 143.5 (25.5) 121.3 (28.1) <0.001
Distance from the VT to the large bowel, mm, median (range) 101.5 (90.6–121.7) 81.9 (69.8–101.2) <0.001
Large bowel length, mm, median (range) 207.7 (182.1–249.5) 211.8 (185.3–263.6) 0.560

P < 0.05.
CME indicates complete mesocolic excision; LN, lymph node; NCME, noncomplete mesocolic excision; SD, standard deviation; VT, vascular tie.

dissection nor central ligation criteria. Pathological staging was Pattern and Risk Factors of Lymph Node Metastasis
reported as stages I, IIA, IIB, IIC, IIIA, IIIB, and IIIC in 28 The incidence of metastatic lymph nodes in the pericolic,
(8.5%), 45 (13.6%), 89 (27.0%), 32 (9.7%), 4 (1.2%), 85 (25.8%), intermediate, and main groups was 36.8% (N ¼ 81), 11.4% (N ¼ 25),
and 47 (14.2%) patients, respectively. In addition, 157 patients and 5.9% (N ¼ 13), respectively. Within the pericolic lymph nodes,
received adjuvant chemotherapy. The baseline characteristics were the incidence of metastasis in the <5.0, 5.0 to 10.0, and >10.0 cm
well balanced between the groups (Supplemental Table 2, http:// subgroups was 36.8% (N ¼ 81), 9.1% (N ¼ 20), and 3.6% (N ¼ 8),
links. lww.com/SLA/B501). respectively. All patients who had pericolic lymph nodes metastases
of 5.0 to 10.0 or >10.0 cm also had synchronous pericolic lymph
Pathological Characteristics and Specimen nodes metastases <5.0. For stage III disease, the incidence of
Morphometry metastatic lymph nodes in the pericolic, intermediate, and main
Patients’ pathological characteristics are summarized in groups was 86.2% (N ¼ 81), 26.6% (N ¼ 25), and 13.8% (N ¼ 13),
Table 1. There were a significantly greater number of total lymph respectively. Within the pericolic lymph nodes, the incidence of
nodes in the CME group than in the NCME group (24 vs 20, P ¼ metastasis in the <5.0, 5.0 to 10.0, and >10.0 cm subgroups was
0.002). However, there was no significant difference in the number of 86.2% (N ¼ 81), 21.3% (N ¼ 20), and 8.5% (N ¼ 8), respectively.
positive lymph nodes between the groups. Due to the small sample The overall incidence of lymph node skip metastasis was 2.7% (N ¼
size and heterogeneity in specimen morphometry, only the cecum 6). Patients with right-sided colon cancer (N ¼ 4, 66.7%) were most
and ascending colon and sigmoid colon were analyzed between the 2 likely to develop skip metastases. In the multivariate logistic regres-
groups in the specimen morphological assessment. Specimens in the sion analysis, only data from the CME group were used. Main lymph
CME group had greater DVT, DVB, and AM in the cecum and node metastases were associated with lymphovascular invasion (LVI)
ascending colon and sigmoid colon, compared with those in the (OR: 5.109, 95.0% CI, 1.092–23.910; P ¼ 0.038) and positive lymph
NCME group. However, there was no significant difference in the nodes (OR: 1.476, 95.0% CI, 1.257–1.733; P < 0.001) (Table 3).
LBL (cecum and ascending colon and sigmoid colon) between the
2 groups. Short-term Outcomes and Risk Factors of
In the correlation analysis, height, weight, BMI, and BSA were Postoperative Surgical Complications
independently associated with AM in both the cecum and ascending Patients in the CME group had less intraoperative blood loss
colon and sigmoid colon. LBL, however, did not correlate with these and a shorter time to first flatus, drainage removal time, time to soft
body measurements (Fig. 1A, B). diet, and length of postoperative hospital stay compared with patients
in the NCME group. By contrast, no significant differences in
Predictors of the Number of Lymph Nodes operative time, time to first bowel motion, or postoperative compli-
Harvested cations were observed between the groups (Table 4). In the multi-
In the multivariate linear regression analysis, CME, right- variate logistic regression analysis, intraoperative blood loss (OR:
sided colon cancer, pathological (p) T3 to 4 stage disease, and tumor 1.004, 95.0% CI, 1.001–1.006; P ¼ 0.002) was the only factor
diameter were associated with a greater number of lymph nodes associated with postoperative surgical complications (Only grade 2
harvested. By contrast, advanced age and a higher BMI were and higher postoperative surgical complications, according to the
associated with fewer lymph nodes harvested (Table 2). Clavien–Dindo classification,11 were calculated.) (Table 5).

ß 2018 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 3

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Gao et al Annals of Surgery  Volume XX, Number XX, Month 2018

FIGURE 1. Correlation analysis between specimen morphometry and body measurements for (A) the cecum and ascending colon
and (B) the sigmoid colon. The Spearman’s rank correlation coefficient is indicated in the boxes and by the size of the circle. The
color of the circle indicates the direction of the correlation (blue: positive correlation; red: negative correlation). AM indicates area of
mesentery; BMI, body mass index; BSA, body surface area; DVB, distance from the large bowel to the central vascular ligation; DVT,
distance from the tumor to the central vascular ligation; LBL, large bowel length.

Survival Analyses during follow-up, including one from the CME group due to acute
Five patients (1.5%) who were lost to follow-up were heart failure and another from the NCME group due to recurrence.
excluded. In total, 325 patients (CME group, 218; NCME group, No significant differences in 3-year OS (97.2% vs 98.3%, log-rank P
107) were included in the survival analysis. The median follow-up ¼ 0.645), DFS (92.2% vs 90.0%, log-rank P ¼ 0.335), or MFS
time was 20.5 (range, 1.0–41.0) months. Two patients (0.6%) died (92.2% vs 95.8%, log-rank P ¼ 0.486) rates were observed between

TABLE 2. Predictors of the Number of Total Lymph Nodes (LNs)


Multivariate Analysis
Factor No. of LNs, Median (Range) P ß (95.0% CI) SE Standardized Coefficient ß P
Group
NCME 20 (16–27) Ref.
CME 24 (17–35) 0.002 3.616 (0.886–6.346) 1.388 0.143 0.010
Sex
M 24 (18–33) —
F 20 (16–29) 0.003 — — — —
Tumor side
Right 25 (19–35) <0.001 4.379 (1.917–6.840) 1.251 0.180 0.001
Left 18 (15–24) Ref.
pT stage
1–2 17 (12–24) Ref.
3–4 22 (17–33) <0.001 3.975 (0.065–7.886) 1.988 0.100 0.046
Age, yrs — <0.001 0.148 (0.234 to 0.062) 0.044 0.165 0.001
Tumor diameter, cm — <0.001 1.207 (0.611–1.803) 0.303 0.208 <0.001
BMI, kg/m2 — 0.001 0.444 (0.794 to 0.093) 0.178 0.126 0.013
Area of mesentery, mm2 — <0.001 — — — —
Large bowel length, mm — <0.001 — — — —

P < 0.05.
BMI indicates body mass index; CI, confidence interval; CME, complete mesocolic excision; F, female; M, male; NCME, noncomplete mesocolic excision; p, pathological; Ref.,
reference; SE, standard error.

4 | www.annalsofsurgery.com ß 2018 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Annals of Surgery  Volume XX, Number XX, Month 2018 Outcomes of Complete Mesocolic Excision

TABLE 3. Logistic Regression Analysis of Risk Factors of Main Lymph Node (LN) Metastasis
Main LNs Univariate Analysis Multivariate Analysis
Factor Negative Positive OR (95.0% CI) P OR (95.0% CI)
Lymphovascular invasion, N (%)
Y 60 (29.0) 9 (69.2) 5.512 (1.635–18.587) 0.006 5.109 (1.092–23.910)
N 147 (71.0) 4 (30.8) Ref. Ref.
Tumor deposits, N (%)
Y 15 (7.2) 4 (30.8) 5.689 (1.566–20.660) 0.008 —
N 192 (92.8) 9 (69.2) Ref. —

Number of positive LNs, median (range) 0 (0–1) 7 (5–10) 1.500 (1.275–1.765) <0.001 1.476 (1.257–1.733)

P < 0.05
Data were only from CME group.
CI indicates confidence interval; N, no; OR, odds ratio; Ref., reference; Y, yes.

the groups. However, CME had a positive effect on LRFS as et al14,15 reported an association between the dissection plane and
compared with NCME (100.0% vs 90.2%, log-rank P < 0.001) survival in 2003 and 2008, respectively. Hohenberger et al5 finally
(Fig. 2A–D). united these 2 key points, which included the horizontal (mesocolic
In a subgroup analysis of Union for International Cancer dissection) and vertical (central ligation) directions, and proposed
Control stage, no significant differences in OS, DFS, or MFS rates CME as a standard surgery for colon cancer. However, previous
were observed between the treatment groups. With respect to LRFS, studies6,16,17 comparing the effect of CME to NCME lacked quality
there was no recurrence in stage I disease in either of the 2 groups. control. Even in the CME group, a mesocolic plane was not achieved
However, compared with NCME, CME had a positive effect on stage in nearly a fifth of all surgeries,16,18 and central ligation was
II (100.0% vs 90.0%, log-rank P ¼ 0.031) and stage III (100.0% vs accomplished less successfully than surgeons had anticipated,19
90.0%, log-rank P ¼ 0.004) disease. In the univariate analysis, CME which may bias the actual effect of CME. The ESCME trial was
(100.0% vs 90.2%, log-rank P < 0.001), mesocolic dissection the first registered study with strict quality control measures to
(100.0% vs 87.9%, log-rank P < 0.001), and nontumor deposits investigate the effect of CME on prognosis. Our findings suggest
(97.2% vs 91.6%, log-rank P < 0.022) had a positive effect on LRFS. that CME is associated with better short- and long-term outcomes
than NCME, especially in stage III disease, pT3 to 4 tumors, and
cases with LVI.
DISCUSSION Previous studies9,20 have shown that CME is associated with
In 1909, Jamieson and Dobson12 first introduced lymph node better surgical specimen quality, including greater LBL, greater
dissection in colon cancer surgery. Since then, 2 main methods, the DVT, and larger AM. Our findings demonstrate that, expect for
‘‘no-touch’’ isolation technique13 and D3 surgery,10 have extended LBL, DVT, DVB, and AM were positively associated with BMI and
the region of lymphadenectomy. Meanwhile, the mesocolic plane BSA. Therefore, it was considered inappropriate to compare speci-
was not considered important in surgery until Bokey et al and West mens from different countries. After adjusting for the influence of

TABLE 4. Short-term Outcomes


Outcome CME (N ¼ 220) NCME (N ¼ 110) P
Operative time, min, median (range) 175 (150–210) 180 (150–215) 0.135
Intraoperative blood loss, mL, median (range) 80 (50–100) 110 (50–200) <0.001
Abdominal drainage volume during postoperative days 1–3, mL, median (range) 243 (131–489) 190 (114–360) 0.099
Time to first flatus, d, median (range) 4 (3–4) 4 (3–5) 0.048
Time to first bowel motion, d, median (range) 5 (4–6) 5 (4–7) 0.647
Drainage removal time, d, median (range) 7 (6–9) 8 (7–10) 0.016
Time to soft diet, d, median (range) 6 (5–7) 7 (5–8) 0.015
Length of postoperative hospital stay, d, median (range) 11 (9–14) 12 (10–14) 0.014
Postoperative surgical complications, N (%)
Wound complications 8 (3.6) 4 (3.6) 1.000
Intestinal obstruction 21 (9.5) 8 (7.3) 0.492
Bleeding 1 (0.5) 0 (0.0) 1.000
Anastomotic leakage 2 (0.9) 1 (0.9) 1.000
Chyle leakage 11 (5.0) 1 (0.9) 0.068
Postoperative nonsurgical complications, N (%)
Respiratory complications 7 (3.2) 2 (1.8) 0.723
Cardiac complications 4 (1.8) 2 (1.8) 1.000
Renal failure 0 (0.0) 1 (0.9) 0.333
Cerebrovascular accident 1 (0.5) 0 (0.0) 1.000

P < 0.05.
Only grade 2 and higher postoperative surgical complications according to Clavien–Dindo classifications were calculated.
CME indicates complete mesocolic excision; NCME, noncomplete mesocolic excision.

ß 2018 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 5

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Gao et al Annals of Surgery  Volume XX, Number XX, Month 2018

TABLE 5. Logistic Regression Analysis of Risk Factors of Postoperative Surgical Complications


Surgical Complications Univariate Analysis Multivariate Analysis
Yes No
Factor (N ¼ 50) (N ¼ 280) OR (95.0% CI) P OR (95.0% CI) P
Sex, N (%)
M 34 (19.4) 141 (80.6) 2.125 (1.122–4.025) 0.021 — —
F 16 (10.3) 140 (89.7) Ref. —
Intraoperative blood loss, mL, median (range) 100 (50–200) 100 (50–150) 1.003 (1.001–1.005) 0.008 1.004 (1.001–1.006) 0.002

P < 0.05.
Only grade 2 and higher postoperative surgical complications according to Clavien–Dindo classifications were calculated.
CI indicates confidence interval; F, female; M, male; OR, odds ratio; Ref., reference.

race, CME was found to be associated with greater DVT, DVB, and number of total lymph nodes compared with NCME (24 vs 20, P ¼
AM, but not LBL. 0.002). A greater number of harvested lymph nodes are an important
Lymphadenectomy is one of the core features of colon cancer factor in support of CME according to 2 main concepts. The first is
surgery. In the present study, CME was associated with a greater that many previous studies21,22 have shown that extended lymph

FIGURE 2. Kaplan–Meier plots of 3-year (A) overall survival, (B) disease-free survival, (C) local recurrence-free survival, and (D)
metastasis-free survival. CME indicates complete mesocolic excision.

6 | www.annalsofsurgery.com ß 2018 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Annals of Surgery  Volume XX, Number XX, Month 2018 Outcomes of Complete Mesocolic Excision

node resection improves long-term survival. The second is stage DFS rate in stage II and stage III disease. However, Merkel et al6
migration. CME may remove residual tumors in NCME that make demonstrated that CME had significantly improved the 5-year
patients receive chemotherapy through upstaging from stage II to cancer-related survival in stage III, but not stage I and stage II
stage III disease, which would reduce the time to recurrence and disease. Our study showed that there were no significant differences
improve prognosis.23 However, both in the present study and in the in the 3-year OS, DFS, and MFS rates between the 2 groups
population-based study with the largest sample size,16 the propor- according to the different stages of disease. The only positive effect
tions of patients with different stages of disease did not differ of CME, as compared with NCME, was on the 3-year LRFS rate in
between the treatment groups. In addition, our study revealed that stage II and stage III, but not stage I disease. Hence, CME may be
CME did not improve the number of positive lymph nodes compared more suitable for stage III disease and pT3 to 4 tumors.
with NCME. Our study has several limitations. First and foremost, unlike
Many surgeons, especially from the United States, value D2 CME, which has detailed surgical requirements, including preserva-
surgery for colon cancer treatment.24 One reason is that apical lymph tion of the mesocolic plane and the central ligation, NCME has no
node metastases occur in 7.0% to 8.0% of patients with stage III uniform definition.7 In China, many colorectal surgeons have utilized
disease.25,26 However, some surgeons regard apical lymph node Japanese D3 surgery over the last 2 decades. Therefore, the types of
metastasis as being the same as distant metastasis and suggest that NCME used on patients in our study may be more radical than
extended lymphadenectomy may not influence survival because conventional surgery in some Western countries. Meanwhile, the 3
apical lymph node metastasis is associated with poor progno- NCME subgroups were confounding factors that may lead to less
sis.27–29 Another reason is that lymph node skip metastases, espe- meaningful comparisons between CME and NCME. Due to limited
cially central lymph node involvement in the absence of pericolic or sample size, direct comparisons between CME and the different
intermediate lymph node involvement, occur in 1.0% to 5.0% of NCME subgroups could not be made. However, through univariate
cases.28,30,31 In the present study, the incidence of lymph node analysis, this study found mesocolic dissection to have a positive
metastasis in the main group was 13.8% (N ¼ 13) for stage III effect on LRFS and central ligation did not influence prognosis.
disease, which was marginally higher than previous studies. This Second, this study is not a randomized controlled trial, which may
may be due to the high proportion of patients with stage III disease in lead to selection bias. However, it is almost impossible to perform a
our study (N ¼ 94, 42.7%), compared with Japanese (N ¼ 4587, randomized controlled trial of CME. One important reason is that
36.7%)32 and Danish (N ¼ 140, 38.5%) studies.16 In addition, in our patients at a high risk of apical lymph node metastasis and more
study, main lymph node metastasis was associated with LVI (OR: advanced-stage disease cannot be made to accept D2 surgery in
5.109, P ¼ 0.038) and the number of positive lymph nodes (OR: China due to ethical considerations. Another is the technical difficult
1.476, P < 0.001). This suggests that CME is particularly necessary of randomization because surgeons with excellent technique and
for patients with stage III disease and cases with LVI. skills tend to perform CME. Lastly, the primary outcome of our study
Surgical safety is an important factor contributing to the lack was LRFS. Thus, the actual effect of CME on OS and DFS may need
of a comprehensive acceptance of CME. CME must be performed to be observed for longer or a greater sample size may be required.
close to organs and vasculature that are not commonly fully exposed In conclusion, this study provides evidence in support of the 3-
in NCME. Bertelsen et al8 reported that there was a significantly year LRFS benefits of CME, as compared with NCME, in patients
higher rate of injury to other organs during CME compared with with colon cancer. Furthermore, CME was associated with a shorter
conventional surgery (9.1% vs 3.6%, P < 0.001). The latest data from hospital stay without increasing surgical risk. This study also sug-
Merkel et al6 also showed a higher incidence of postoperative gests that CME may be more suitable for stage III disease, pT3 to 4
morbidity in the period after CME. However, there were no signifi- tumors, and patients with LVI.
cant differences in postoperative complications between the CME
and NCME groups in our study. This is mainly due to the compre- ACKNOWLEDGMENTS
hensive implementation of Japanese D3 surgery in China over the last The authors thank Professor Tao Wu from the Department of
2 decades. Chinese colorectal surgeons were familiar with the Epidemiology at Peking University Health Science Center (Beijing,
technique of central ligation before the concept of CME was intro- China) for performing the sample size calculation; Dr. Zhuang Tao
duced. A non-negligible phenomenon in our study was the high from the Beijing Medical Development Clinical Research Institute
incidence of chyle leakage in the CME group compared with the (Beijing, China) for the careful statistical review; and Mr. Shuqiang
NCME group (5.0% vs 0.9%, P ¼ 0.068), which is due to the Mao, Dr. Fan Liu, Dr. Yuanpei Lin, Mr. Xuesong Zhao, Dr. Chang
extended region of the lymphadenectomy. Chyle leakage delayed not Wang, Dr. Yongbo An, Mr. Yushi Zhou, Mr. Haoran Zhang, Mr.
only the time to diet, but also the time to chemotherapy, which could Dafang Liu, and Mr. Yongbai Li from the Surgical Oncology Labo-
potentially influence prognosis. ratory at Peking University People’s Hospital (Beijing, China) for
The effect of CME on long-term survival remains controver- collecting the clinical data. The authors also thank Editage (www.e-
sial. Advocators generalized CME based on several reasons, one ditage.cn) for English language editing.
being the inadequate pathological examinations that could not detect
all the lymph node metastases. Rosenberg et al33 detected occult REFERENCES
tumor cells in the regional lymph nodes of 51.8% of patients (N ¼ 44)
1. Cucino C, Buchner AM, Sonnenberg A. Continued rightward shift of colo-
with node-negative colorectal cancer on routine histopathological rectal cancer. Dis Colon Rectum. 2002;45:1035–1040.
analysis using reverse transcriptase polymerase chain reaction. A 2. McDermott FT, Hughes ES, Pihl E, et al. Comparative results of
meta-analysis34 revealed that the molecular detection of tumor cells surgical management of single carcinomas of the colon and rectum:
in the regional lymph nodes was associated with disease recurrence a series of 1939 patients managed by one surgeon. Br J Surg. 1981;68:
and a poor survival in node-negative colorectal cancer. Other reasons 850 –855.
included the phenomenon of skip metastasis and stage migration as 3. Birgisson H, Talback M, Gunnarsson U, et al. Improved survival in cancer of
the colon and rectum in Sweden. Eur J Surg Oncol. 2005;31:845–853.
mentioned above. Undetected lymph node and skip metastases may
4. Quirke P, Steele R, Monson J, et al. Effect of the plane of surgery achieved on
result in stage migration in the CME group, but early recurrence in local recurrence in patients with operable rectal cancer: a prospective study
the NCME group. Results from the population-based study with the using data from the MRC CR07 and NCIC-CTG CO16 randomised clinical
largest sample size16 indicated that CME had improved the 4-year trial. Lancet. 2009;373:821–828.

ß 2018 Wolters Kluwer Health, Inc. All rights reserved. www.annalsofsurgery.com | 7

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
CE: T.M.; ANNSURG-D-17-02059; Total nos of Pages: 8;
ANNSURG-D-17-02059

Gao et al Annals of Surgery  Volume XX, Number XX, Month 2018

5. Hohenberger W, Weber K, Matzel K, et al. Standardized surgery for colonic 20. Kobayashi H, West NP, Takahashi K, et al. Quality of surgery for stage III
cancer: complete mesocolic excision and central ligation—technical notes and colon cancer: comparison between England, Germany, and Japan. Ann Surg
outcome. Colorectal Dis. 2009;11:354–364. Oncol. 2014;21(Suppl. 3):S398–S404.
6. Merkel S, Weber K, Matzel KE, et al. Prognosis of patients with colonic 21. Kotake K, Honjo S, Sugihara K, et al. Number of lymph nodes retrieved is an
carcinoma before, during and after implementation of complete mesocolic important determinant of survival of patients with stage II and stage III
excision. Br J Surg. 2016;103:1220–1229. colorectal cancer. Jpn J Clin Oncol. 2012;42:29–35.
7. Wang C, Gao Z, Shen K, et al. Safety, quality and effect of complete 22. Ovrebo K, Rokke O. Extended lymph node dissection in colorectal cancer
mesocolic excision vs non-complete mesocolic excision in patients with surgery. Reliability and reproducibility in assessments of operative reports. Int
colon cancer: a systemic review and meta-analysis. Colorectal Dis. J Colorectal Dis. 2010;25:213–222.
2017;19:962–972. 23. Sjo OH, Merok MA, Svindland A, et al. Prognostic impact of lymph node
8. Bertelsen CA, Neuenschwander AU, Jansen JE, et al. Short-term outcomes harvest and lymph node ratio in patients with colon cancer. Dis Colon Rectum.
after complete mesocolic excision compared with ’conventional’ colonic 2012;55:307–315.
cancer surgery. Br J Surg. 2016;103:581–589. 24. Vogel JD, Eskicioglu C, Weiser MR, et al. The American Society of Colon and
9. West NP, Sutton KM, Ingeholm P, et al. Improving the quality of colon cancer Rectal Surgeons Clinical Practice Guidelines for the treatment of colon cancer.
surgery through a surgical education program. Dis Colon Rectum. Dis Colon Rectum. 2017;60:999–1017.
2010;53:1594–1603. 25. Kanemitsu Y, Komori K, Kimura K, et al. D3 lymph node dissection in right
10. General rules for clinical and pathological studies on cancer of the colon, hemicolectomy with a no-touch isolation technique in patients with colon
rectum and anus. Part I. Clinical classification. Japanese Research Society for cancer. Dis Colon Rectum. 2013;56:815–824.
Cancer of the Colon and Rectum. Jpn J Surg. 1983;13:557–573. 26. Kawada H, Kurita N, Nakamura F, et al. Incorporation of apical lymph node
11. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification status into the seventh edition of the TNM classification improves prediction
of surgical complications: five-year experience. Ann Surg. 2009;250: of prognosis in stage III colonic cancer. Br J Surg. 2014;101:1143–1152.
187–196. 27. Ang CW, Tweedle EM, Campbell F, et al. Apical node metastasis indepen-
12. Jamieson JK, Dobson JF VII. Lymphatics of the colon: with special reference dently predicts poor survival in Dukes C colorectal cancer. Colorectal Dis.
to the operative treatment of cancer of the colon. Ann Surg. 1909;50: 2011;13:526–531.
1077–1090. 28. Lan YT, Lin JK, Jiang JK, et al. Significance of lymph node retrieval from the
13. Turnbull RB Jr, Kyle K, Watson FR, et al. Cancer of the colon: the influence of terminal ileum for patients with cecal and ascending colonic cancers. Ann Surg
the no-touch isolation technic on survival rates. Ann Surg. 1967;166:420–427. Oncol. 2011;18:146–152.
14. West NP, Morris EJ, Rotimi O, et al. Pathology grading of colon cancer 29. Emmanuel A, Haji A. Complete mesocolic excision and extended (D3)
surgical resection and its association with survival: a retrospective observa- lymphadenectomy for colonic cancer: is it worth that extra effort? A review
tional study. Lancet Oncol. 2008;9:857–865. of the literature. Int J Colorectal Dis. 2016;31:797–804.
15. Bokey EL, Chapuis PH, Dent OF, et al. Surgical technique and survival in 30. Liang JT, Lai HS, Huang J, et al. Long-term oncologic results of laparoscopic D3
patients having a curative resection for colon cancer. Dis Colon Rectum. lymphadenectomy with complete mesocolic excision for right-sided colon
2003;46:860–866. cancer with clinically positive lymph nodes. Surg Endosc. 2015;29:2394–2401.
16. Bertelsen CA, Neuenschwander AU, Jansen JE, et al. Disease-free survival after 31. Bertelsen CA, Kirkegaard-Klitbo A, Nielsen M, et al. Pattern of colon cancer
complete mesocolic excision compared with conventional colon cancer surgery: lymph node metastases in patients undergoing central mesocolic lymph node
a retrospective, population-based study. Lancet Oncol. 2015;16:161–168. excision: a systematic review. Dis Colon Rectum. 2016;59:1209–1221.
17. Galizia G, Lieto E, De Vita F, et al. Is complete mesocolic excision with 32. Watanabe T, Muro K, Ajioka Y, et al. Japanese Society for Cancer of the Colon
central vascular ligation safe and effective in the surgical treatment of right- and Rectum (JSCCR) guidelines 2016 for the treatment of colorectal cancer.
sided colon cancers? A prospective study. Int J Colorectal Dis. 2014;29: Int J Clin Oncol. 2018;23:1–34.
89–97. 33. Rosenberg R, Hoos A, Mueller J, et al. Prognostic significance of cytokeratin-
18. Bertelsen CA, Bols B, Ingeholm P, et al. Can the quality of colonic surgery be 20 reverse transcriptase polymerase chain reaction in lymph nodes of node-
improved by standardization of surgical technique with complete mesocolic negative colorectal cancer patients. J Clin Oncol. 2002;20:1049–1055.
excision? Colorectal Dis. 2011;13:1123–1129. 34. Rahbari NN, Bork U, Motschall E, et al. Molecular detection of tumor cells in
19. Munkedal DLE, Rosenkilde M, Nielsen DT, et al. Radiological and pathologi- regional lymph nodes is associated with disease recurrence and poor survival
cal evaluation of the level of arterial division after colon cancer surgery. in node-negative colorectal cancer: a systematic review and meta-analysis.
Colorectal Dis. 2017;19:O238–O245. J Clin Oncol. 2012;30:60–70.

8 | www.annalsofsurgery.com ß 2018 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

Das könnte Ihnen auch gefallen