- Email: [email protected]
Water dynamics during solid-state fermentation by Aspergillus oryzae YH6
Received: 18 May 2018
Revised: 13 January 2019
Accepted: 28 January 2019
DOI: 10.1002/rcs.1988
REVIEW ARTICLE
Robotic‐assisted minimally invasive esophagectomy versus the conventional minimally invasive one: A meta‐analysis and systematic review | Liang Yao5,6† Dacheng Jin1,2,3,4† Kehu Yang3,4 | Yunjiu Gou2
|
Jun Yu2
|
Rong Liu5
|
Tiankang Guo3
|
1
Department of Clinical Medicine, Gansu University of Traditional Chinese Medicine, Lanzhou, China
Abstract Background:
Conventional video‐assisted minimally invasive esophagectomy (MIE)
2
Department of Thoracic Surgery, Gansu Province People's Hospital, Lanzhou, China
is safe and associated with low rates of morbidity and mortality, but the two‐
3
dimensional monitor reduces eye‐hand harmony and viewing yield. Robotic‐assisted
Institution of Clinical Research and Evidence Based Medicine, Gansu Province People's Hospital, Lanzhou, China
4
Evidence‐Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou, China
minimally invasive esophagectomy (RAMIE) with its virtual reality simulators offers a realistic three‐dimensional environment that facilitates dissection in the narrow working space, but it is expensive and requires longer operative time. Therefore, the aim of this meta‐analysis was to assess the safety and feasibility of RAMIE versus MIE in
5
The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing, China
patients with esophageal cancer. Material and Methods:
PubMed, EMBASE, Cochrane library, and Chinese Biomed-
6
Clinical Division, Hong Kong Baptist University, Hong Kong, China
ical Literature databases were systematically searched up to 21 September 2018 for case‐controlled studies that compared RAMIE with MIE.
Correspondence Yunjiu Gou, Gansu Province People's Hospital, Lanzhou, China. Email: [email protected] Kehu Yang, Lanzhou University, Lanzhou, China. Email: [email protected]
Result:
Eight case‐controlled studies involving 1862 patients (931 under RAMIE
and 931 under MIE) were considered. No statistically significant difference between the two techniques was observed regarding R0 resection rate (OR = 1.1174, P = 0.8647), conversion to open (OR = 0.7095, P = 0.7519), 30‐day mortality rate (OR = 0.8341, P = 0.7696), 90‐day mortality rate (OR = 0.3224, P = 0.3329), in‐ hospital mortality rate (OR = 0.3733, P = 0.3895), postoperative complications, number of harvested lymph nodes (mean difference [MD] = 0.8216, P = 0.2039), operation time (MD = 24.3655 min, P = 0.2402), and length of stay in hospitals (LOS) (MD = −5.0228 day, P = 0.1342). The meta‐analysis showed that RAMIE was associated with a significantly fewer estimated blood loss (EBL) (MD = −33.2268 mL, P = 0.0075). And the vocal cord palsy rate was higher in the MIE group compared with RAMIE, and the difference was significant (OR = 0.5696, P = 0.0447). Conclusion:
This meta‐analysis indicated that RAMIE and MIE display similar feasi-
bility and safety when used in esophagectomy. However, randomized controlled studies with larger sample sizes are needed to evaluate the benefit and harm in patients with esophageal cancer undergoing RAMIE.
Abbreviations: CI, confidence interval; EBL, estimated blood loss; LOS, length of stay; MD, mean difference; MIE, minimally invasive esophagectomy; NOS, Newcastle‐ Ottawa scale; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy; SD, standard deviation † First two authors D.C.J. and L.Y. contributed equally to this work and were considered as co‐first authors.
Int J Med Robotics Comput Assist Surg. 2019;15:e1988. https://doi.org/10.1002/rcs.1988
wileyonlinelibrary.com/journal/rcs
© 2019 John Wiley & Sons, Ltd.
1 of 11
2 of 11
1
JIN
I N T RO D U CT I O N
|
2.2
Esophageal cancer is one of the most common neoplasms of the 1
|
ET AL.
Study selection
Studies selected for this meta‐analysis were included according to the
digestive system. At present, esophagectomy is an efficient way to
following criteria: Case‐control studies that compared RAMIE and MIE
treat esophageal cancer. With the improvements in video‐assisted sur-
in the treatment of esophageal neoplasm, original studies, esophagec-
gery, conventional minimally invasive esophagectomy (MIE) has gained
tomy, and reporting at least one of the outcome measures mentioned
2-5
Multiple single hospitals have
below. After screening of the titles, abstracts and full text were
reported that this procedure is safe, resulting in lower rates of morbid-
screened by one researcher, and studies were excluded if the
more attention in recent years.
ity and mortality and transthoracic esophagectomy performed through
abovementioned standards were not confirmed or had a low‐quality
a reduced thoracic wound.6-11 Several articles demonstrated that rad-
score (<7) according to the Newcastle‐Ottawa scale (NOS).
ical surgery using MIE to treat esophageal carcinoma can narrow the rates of pulmonary complications and lower the mortality rate.4,5,9 However, conventional MIE has also some limitations. The two‐
2.3
|
Outcomes of interest
dimensional monitor reduces eye‐hand harmony and viewing yield. Difficulties increase when dealing with complex surgical procedures.12
The outcome of interest consisted of operative time (min), estimated
Furthermore, the high technical complexity of a minimally invasive
blood loss (EBL) (mL), length of stay (LOS) (days), number of harvested
procedure may lead to a complanate learning curve.13 Thus, these fac-
lymph nodes, mortality after operation, R0 resection rate, and compli-
tors may limit the development of this technique.
cations. Operative time included the total and thoracic operative time.
Robotic‐assisted minimally invasive esophagectomy (RAMIE) is a relatively new minimally invasive technique. The first robotic‐assisted
Postoperative mortality was defined as death within 30 days, 90 days, and in‐hospital after the esophagectomy.
two‐stage, three‐field esophagectomy was reported by Kernstine et al in 2004.14 RAMIE is offering more advantages than conventional MIE. RAMIE offers realistic three‐dimensional immersion environments that should theoretically facilitate dissection in the narrow working space up to the mediastinum, and it is technically feasible and safe in terms of oncological outcomes.15-20 Nevertheless, the high costs, longer operation time, and the lack of haptic feedback are the most common disadvantages revealing the need of more progresses in this technology.2123
Up to now, no meta‐analysis is available reporting the comparison
between RAMIE and MIE, and no randomized controlled trials were published. However, several case‐control studies show comparable
2.4
|
Data extraction and quality assessment
Two investigators independently extracted the data from the included studies according to the parameters mentioned above and summarized them. Selected studies were all case‐control studies; thus, NOS was used for grading31 (Table 2). All included studies had a high‐ quality score according to NOS. Discrepancies were resolved by discussion.
outcomes and survival after esophagectomy.24-30 Therefore, the aim of our meta‐analysis was to evaluate the safety and effectiveness of RAMIE versus MIE in patients who underwent esophagectomy.
2.5
|
Publication bias
If the number of the included studies was more than 10, the publication bias was evaluated. However, less than 10 studies were included
2 2.1
MATERIAL AND METHODS
|
|
in our studies.32
Literature search 2.6
|
Statistical analysis
A comprehensive literature search was conducted to find all relevant trials using the following databases: PubMed, EMBASE, Cochrane Library,
The meta‐analysis was performed using R software (R x64 3.4.2, The
and the Chinese Biomedical Literature Database. The following search
Cochrane Collaboration, Oxford, UK). Dichotomous variables were
terms were used: “Esophageal Neoplasm,” “Neoplasm, Esophageal,”
pooled using the odds ratio (OR) with a 95% confidence interval (CI).
“Esophagus Neoplasm,” “Esophagus Neoplasms,” “Cancer of Esopha-
Because the events were rare, the random effect was used in this
gus,” “Esophagus Cancer,” “esophagectomy,” “oesophagectomy,” and
meta‐analysis. Continuous variables were pooled using the mean dif-
“robot*.” We used medical subject heading terms (MeSH) and free
ference (MD) with a 95% CI. P values <0.05 were considered statisti-
words as keywords. To identify if our search was comprehensive, the
cally significant. In studies that only reported medians and ranges, the
characteristics of both the control group and study language were not
mean and standard deviation (SD) were assessed using the means of
considered. Two authors independently evaluated the included studies,
the method provided by Hozo et al.33 The studies' statistical heteroge-
and disagreements were resolved through discussion. The latest search
neity was evaluated through the I2 statistic.34 I2 value <25% indicated
was made on 21 September 2018.
low heterogeneity, and a value >50% indicated high heterogeneity.
JIN
3 of 11
ET AL.
2.7
|
Ethical approval
3.1
|
LOS
The article did not involve ethical approval because our study is a sec-
The total LOS after operation was indicated in six studies. Four of
ondary research, not involved with patients recruited.
them showed a shorter LOS in patients who underwent RAMIE than MIE. He et al29 and Deng et al28 had a shorter LOS in the MIE group. Pooled data analysis demonstrated no statistically significant
3
|
RESULTS
difference between these two techniques (MD = −0.5381 day, 95% CI [−2.4750; 1.3987]; P = 0.5861), with a high heterogeneity
A total number of 953 articles were found by the initial literature
(I2 = 77.9%; P = 0.0004) (Figure 2A).
search (Figure 1). One article was added after finding it through other sources. After duplicate removal, 906 results remained. After a careful check of titles and abstracts, 869 articles were excluded, because the excluded studies were case reports or did not compare the two types
3.2
|
Operative time
of surgery at the same time. The 37 remaining articles were further assessed for eligibility by a full examination of the text. Finally, eight
Operative time was reported in five studies. In particular, three studies
relevant studies were included in this meta‐analysis. This meta‐
indicated a longer operative time with RAMIE, and two studies
analysis included 1862 patients, in which, 931 were subjected to
showed a shorter operative time with RAMIE compared with MIE.
RAMIE and 931 were subjected to MIE. All studies were of high‐
Pooled data analysis demonstrated no statistically significant differ-
quality scores based on NOS. The characteristics and methodological
ence
quality assessment scores of the included studies are summarized in
(MD = 24.3655 min, 95% CI [−16.2929; 65.0238]; P = 0.2402), with
Tables 1 and 2.
a heterogeneity (I2 = 48.7%; P = 0.05) (Figure 2B).
FIGURE 1
Flow chart of selection for included studies
in
operative
time
between
these
two
techniques
4 of 11
JIN
TABLE 1
ET AL.
Characteristics of the selected studies included in the meta‐analysis
Study
Year
Study Type
Data Source
Country
Design
Group
Park et al
2016
RO
OD
South Korea
CCS
RAMIE
62
57 5
64.3 ± 8.0
MIE
43
40 3
64.3 ± 8.0
23.3 ± 3.1
16
15 1
65(53‐86)
21.3(17.5‐26.3)
Number
Sex (M/F)
Age y (Mean ± SD)
BMI (kg/m2) (Mean ± SD) 23.5 ± 2.8
Suda et al
2012
RO
OD
Japan
CCS
RAMIE MIE
20
15 5
64.5(50‐79)
20.4 (14.9‐24.8)
Weksler et al
2012
RO
OD
American
CCS
RAMIE
11
83
58.7 ± 8.5
27.1
MIE
26
20 6
64.3 ± 11.3
27.9
34
32 2
56.76 ± 8.39
NA
Chao et al
2018
RO
OD
TaiWan
CCS
RAMIE MIE
34
33 1
53.47 ± 8.69
NA
Deng et al
2018
RO
OD
China
CCS
RAMIE
42
35 7
60.7 ± 6.9
NA
MIE
42
33 9
61.8 ± 9.5
NA
He et al
2018
RO
OD
China
CCS
RAMIE
27
20 7
61 ± 8
21.5 ± 2.7
27
20 7
61.6 ± 9.8
21.9 ± 2.8
Babatunde et al
2016
RO
OD
Canada
CCS
RAMIE
170
142 28
56/64/70
NA
MIE
170
143 27
56/63/69
NA
RAMIE
569
471 98
62.9 ± 9.6
NA
MIE
569
489 80
62.8 ± 9.3
NA
MIE
Weksler et al
2017
RO
NCDB
American
CCS
Abbreviations: BMI, body mass index; CCS, case‐control study; IL, Ivor Lewis; MIE, minimally invasive esophagectomy; MK, McKeown; NA, not available; NCDB, The National Cancer Data Base; NOS, Newcastle‐Ottawa Scale; OD, original data; RAMIE, robotic‐assisted minimally invasive esophagectomy; RO, retrospective observational.
3.3
|
Harvested number of lymph node
3.6
|
Vocal cord palsy
In all studies, the number of harvested lymph nodes was reported. In
Six studies showed the vocal cord palsy rate. The meta‐analysis indi-
five studies, it was shown that RAMIE patients were subjected to
cated that MIE has a higher rate of vocal cord palsy incidence than
the removal of a higher number of lymph nodes compared with MIE
RAMIE (OR = 0.5696, 95% CI [0.3288; 0.9868]; P = 0.0447) with a
patients. Moreover, in two studies, the same number of harvested
low heterogeneity (I2 = 0%; P = 0.4932) (Figure 5).
lymph nodes was reported, while in one study, it was reported that RAMIE patients were subjected to the removal of a smaller number of lymph nodes compared with MIE patients. Pooled data analysis
3.7
|
Conversion rate
showed that the differences were not statistically significant (MD = 0.8216, 95% CI [−0.4458; 2.0890]; P = 0.2039), with a high
Two studies reported the conversion rate. The meta‐analysis indicated no significant differences between these two techniques (OR = 0.7095,
2
heterogeneity (I = 50%; P = 0.0501) (Figure 3A).
95% CI [0.0845; 5.9588]; P = 0.7519) with a low heterogeneity (I2 = 0%; P = 0.9742) (Figure 6).
3.4
|
EBL 3.8
|
Mortality after operation
EBL was described in six studies. The meta‐analysis indicated that RAMIE had a lower EBL than MIE. (MD = −33.2268 mL, 95% CI
Postoperative mortality was defined as death within 30 days, 90 days,
[−57.5870; −8.8666]; P = 0.0075) with a low heterogeneity (I = 0;
and in‐hospital. In six studies, the 30‐day mortality rate was reported,
P = 0.7076) (Figure 3B).
in two studies the 90‐day mortality rate was reported, and the in‐
2
hospital mortality rate was reported in two studies. Pooled data analysis showed that differences were not statistically significant between these mortality rates. The 30‐day mortality rate (OR = 0.8341, 95% CI
3.5
|
R0 resection rate
[0.2478; 2.8079]; P = 0.7696) had a low heterogeneity (I2 = 0%; P = 0.8310), the 90‐day mortality rate (OR = 0.3224, 95% CI
Two studies reported R0 resection rate. The meta‐analysis indicated
[0.0326; 3.1879]; P = 0.3329) had a heterogeneity of I2 = 0%
no statistically significant differences between these two techniques
(P = 0.9974), and the in‐hospital mortality rate (OR = 0.3733, 95% CI
(OR = 1.1174, 95% CI [0.6760; 1.8472]; P = 0.8647) with a low het-
[0.0396; 3.5224]; P = 0.3895) had a heterogeneity of I2 = 0%
2
erogeneity (I = 0%; P = 0.7802) (Figure 4).
(P = 0.9045) (Table 3).
JIN
5 of 11
ET AL.
TABLE 1 Upper Thoracic
Study Park et al
Suda et al
Mid Thoracic
Neoadjuvant Chemoradiation
Tumor Size (cm)
Surgical Procedures
NOS (Score) 8
15 (24.2%)
39(62.9%)
101.6 ± 17.1
8(12.9%)
NA
MK
7(16.3%)
9 (20.9%)
27(62.8%)
106.7 ± 13.8
4(9.3%)
NA
MK
7
9
NA
6
3.75(0.9‐8.0) 3.65(0.0‐7.0)
NA
2
12
6
NA
17
NA
NA
NA
NA
4
NA
IL
NA
NA
NA
NA
10
NA
IL
10(29.4%)
15(44.1%)
9(26.5%)
NA
17(50%)
NA
MK
10(29.4%)
19(55.9%)
9(26.5%)
NA
17(50%)
NA
MK
NA
NA
NA
NA
NA
NA
MK
NA
NA
NA
NA
NA
NA
MK
1(3.7)
18(66.6)
8(29.6)
94.6 ± 13.8
NA
NA
MK
3(11.1)
15(55.6)
9(33.3)
92.9 ± 23
NA
NA
MK
Chao et al
Deng et al
FEV1 (pred%, Mean ± SD)
8(12.9%)
2
Weksler et al
Lower Thoracic
He et al
Babatunde et al
NA
156
14
NA
160
Weksler et al
NA
513
NA
527
NA
7
NA
NA
70.6% (120)
20.0/35.0/50.0
NA
10
NA
70.6% (120)
22.5/35.0/50.0
NA
NA
NA
405 (71.2%)
33.7(17‐57)
NA
NA
401 (70.5%)
35.0 (15‐55)
NA
7
7
8
7
7
8
Abbreviations: BMI, body mass index; CCS, case‐control study; IL, Ivor Lewis; MIE, minimally invasive esophagectomy; MK, McKeown; NA, not available; NCDB, The National Cancer Data Base; NOS, Newcastle‐Ottawa Scale; OD, original data; RAMIE, robotic‐assisted minimally invasive esophagectomy; RO, retrospective observational.
TABLE 2
The Newcastle‐Ottawa scale Selection 1
Study
2
Comparability 3
4
1
Exposure 1
2
3
Quality Score
Park et al 2016
*
*
0
*
**
*
*
*
8
Suda et al 2012
*
*
0
*
*
*
*
*
7
Babatunde et al 2016
*
*
0
*
*
*
*
*
7
Weksler et al 2012
*
*
0
*
*
*
*
*
7
Chao et al 2018
*
*
0
*
*
*
*
*
7
Deng et al 2018
*
*
0
*
**
*
*
*
8
He et al 2018
*
*
0
*
*
*
*
*
7
Weksler et al 2017
*
*
0
*
**
*
*
*
8
*One point.
3.9
|
Postoperative complications
perform esophagectomy. This meta‐analysis did not show statistically significant differences between MIE and RAMIE regarding operative
Postoperative complications included the anastomotic leak, empyema,
time, mortality rate, hospital stay, harvested lymph nodes, R0 resec-
pneumonia, arrhythmia, and chylothorax. Table 4 presents the summa-
tion rate, and some complications. However, we demonstrated that
rized results. No statistically significant differences were observed in
the intraoperative blood loss in RAMIE was less when compared with
postoperative complications.
MIE. Regarding the vocal cord palsy rate, the MIE group had a higher rate than the RAMIE group. Patients with esophageal cancer always have a high cancer met-
4
|
DISCUSSION
astatic rate on the recurrent laryngeal nerve (RLN) lymph node. Hence, having an accurate lymph node dissection during the esoph-
Over the past decade, an increasing number of medical institutions
agectomy is significantly important. Our meta‐analysis indicated no
have reported progress on using RAMIE. To our knowledge, this is
significant difference in the number of harvested lymph nodes. How-
the first meta‐analysis comparing these two surgical techniques to
ever, four studies24,25,28,29 demonstrated that RAMIE allows
6 of 11
FIGURE 2 or MIE
JIN
ET AL.
A, LOS meta‐analysis in patients who underwent RAMIE or MIE; B, operative time meta‐analysis in patients who underwent RAMIE
FIGURE 3 A, Number of harvested lymph nodes meta‐analysis in patients who underwent RAMIE or MIE; B, EBL meta‐analysis in patients who underwent RAMIE or MIE
FIGURE 4
R0 resection rate meta‐analysis in patients who underwent RAMIE or MIE
achieving a higher mean number of harvested lymph nodes along the 25
rate. Park et al24 concluded that RAMIE was superior to MIE in
stated that without increasing RLN palsy and pul-
lymph node dissection. Moreover, Deng et al28 and He et al29 con-
monary complications rates, RAMIE could still fulfill a high removal
cluded that RAMIE had more advantages than MIE's lymph node
RLN area. Chao
JIN
7 of 11
ET AL.
FIGURE 5
Vocal cord palsy rate meta‐analysis in patients who underwent RAMIE or MIE
FIGURE 6
Conversion rate meta‐analysis in patients who underwent RAMIE or MIE
TABLE 3
Mortality after operation for RAMIE versus MIE
Mortality
No. of Studies
RAMIE
MIE
I2
Effect Measure (OR)
Pooled Effect 95%CI
P Value 0.7696
30‐day mortality
6
335
335
0
0.8341
[0.2478; 2.8079]
90‐day mortality
2
61
61
0
0.3224
[0.0326; 3.1879]
0.3329
In‐hospital mortality
3
61
80
0
0.3733
[0.0396; 3.5224]
0.3895
Abbreviations: CI, confidence interval; MIE, minimally invasive esophagectomy; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy.
TABLE 4
Summary of complications for RAMIE versus MIE Effect Measure (OR)
Analysis Model
Pooled effect 95%CI
P Value
0
0.98
RE
[0.3434; 3.1936]
0.95
0
0.43
RE
[0.1091; 1.7063]
0.23
89
0
0.45
RE
[0.0774; 2.5751]
0.37
61
80
0
0.55
RE
[0.1712; 1.7710]
0.32
192
192
3.1%
2.06
RE
[0.8047; 5.2793]
0.13
Complications
No. of Studies
RAMIE
MIE
Pneumonia
5
176
172
Arrhythmia
3
54
73
Chylothorax
3
85
Empyema
3 6
Anastomotic leakage
I2
Abbreviations: CI, confidence interval; MIE, minimally invasive esophagectomy; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy; RE, random effect.
dissection; however, these results were not significantly different.
results when RAMIE was used according to the following evidences:
This may be limited by the relatively low sample size. Weksler
First, compared with the traditional two‐dimensional monitor, the
et al27 and Babatunde et al30 concluded that the number of lymph
three‐dimensional self‐controlled magnified view provided by the
nodes dissected by RAMIE and MIE was similar, which was consis-
robotic camera allows a better visualization of the upper mediasti-
tent with our findings. The dissection of nodes located ventrally to
num. Second, even in a limited anatomical space, RAMIE provides
the left RLN is the most challenging part of the surgical procedure.
a meticulous surgical assistance. Third, the third robot arm which is
Several novel approaches have been recently proposed during MIE
controlled by the operator with the application of a stable and
in order to improve the operative exposure and reduce RLN injury.
self‐controllable traction can ensure a safer and easier approach to
However, in spite of the improvement, the left RLN lymph node
nodal dissection.
remains difficult to reach using conventional MIE. In this case, the
During the lymph node dissection throughout the thoracic surgery,
guarantee of a high‐quality dissection is extremely needed, and this
the excessive exposure of the RLN may be the main cause of vocal cord
requires experienced assistants.
palsy. Bilateral RLN injury may cause dyspnea and endanger the
Despite our meta‐analysis, evidences are lacking demonstrating
patient's life. Therefore, it is extremely important to reduce the inci-
a better performance of RAMIE; we conferred good oncologic
dence of RLN injury. However, some present studies reported that the
8 of 11
JIN
ET AL.
incidence of vocal cord palsy was high in both RAMIE and MIE.21,24,26
It is difficult to precisely measure blood loss because its measure-
This result may be due to the use of a flexible laryngoscope by the
ments might be associated with surgeons' estimation. In addition,
doctors even in the absence of any clinical manifestation. As a result,
the possibility that the surgeons are reporting a reduced blood loss
the incidence of patients with vocal cord palsy was higher than normal.
cannot be ignored. Six studies have been analyzed in terms of blood
In the study we included, patients with RLN injury showed unilateral
loss, but the patient sample size was too small to have statistical
injury. Thus, patients' symptoms were in all cases vocal cord paralysis.
results. More high‐quality reports using a large database should be
Our meta‐analysis showed that MIE had higher RLN injury than
performed to identify patients who underwent esophagectomy
RAMIE. Among them,24-26,28 the rates of vocal cord palsy in the MIE
because of esophageal cancer, together with a documentation of
group were significantly higher when compared with that in the
the surgical approach.
RAMIE group, which was consistent with our findings. This result
The mortality rate is a safe outcome, and in our meta‐analysis it
might be due to the advantage in the lymph node dissection under
resulted in no significant differences between MIE and RAMIE. Post-
the RAMIE, which allows a better performance in a good field of vision.
surgical mortality was defined as death within 30 days, 90 days, and
However, due to the limited number of studies we included, this view
in‐hospital stays. Of the 30 mortality rates, we included six studies.
needs to be further proven in the future. Therefore, our conclusion
In two of the studies, deaths were not reported. In one study, three
was that, although this was a positive outcome according to our
deaths were reported in both the RAMIE and MIE groups. No intra-
results, it is not sufficient to confirm that MIE's lymph node dissection
operative deaths were reported in the studies that were included in
is not effective.
this study. The cause of death was mostly due to other diseases, such
In our meta‐analysis, no significant differences were found in
as chronic diseases, cardiovascular diseases etc. Similarly, 90 mortality
operative time between the two techniques. Weksler et al27
and in‐hospital mortality were associated for this reason. In general,
reported that the operative time did not increase significantly with
previous studies confirmed no differences in death rate between
the use of the robot because of the faster thoracic mobilization.
RAMIE and MIE, and the safety of RAMIE and MIE has been widely
Most of the articles have a comparison of this intraoperative out-
confirmed, which is consistent with our findings. In 2017, Weksler
come and always show a positive result.24,29,35,36 Suda et al26 and
et al38 published an article comparing open esophagectomy, MIE,
27
concluded that the surgical time of MIE was longer
and RAMIE. A total of 1707 patients with esophageal cancer were
when compared with that of RAMIE. We speculated that the setup
Weksler et al
included in Weksler's study after a propensity matching method.
time of the robot was excluded and that simply the effective surgical
Weksler concluded that the 30‐day mortality rate in RAMIE was
time was compared. The question that comes to mind is, whether
higher when compared with that in the MIE group (5.6 vs 2.8); how-
the beginning of anesthesia in robotic surgery should wait until the
ever, no significant differences were observed in 90‐day mortality.
robot is fully installed. The shorter length of the anesthetic may
Weksler believed that the mortality was greater than 20% when
accelerate a patient's recovery after surgery. If, in the future, the
low‐volume surgeons operated in low‐volume hospitals.39,40 Although
effective surgical time of the robot can be proven to be shorter than
MIE clearly demonstrated benefits in reducing postoperative morbid-
that of MIE, the advantages of the robot will be highlighted. The
ity, it is unclear whether MIE and RAMIE decrease short‐term mortal-
data extracted from the included studies show differences regarding
ity among patients undergoing esophagectomy. In other studies that
the operative time because of many different reasons. First of all,
were included, we confirmed that there was no difference in mortality
different understanding in the robotic‐assisted system by different
between RAMIE and MIE at 30 days, 90 days, and during hospital
surgeons always resulted in a discrepancy in the operative time.
stay. Similarly, both RAMIE and MIE have been widely confirmed to
Learning curve was connected with surgeons' experience and may
be safer than open. This is consistent with the conclusion that RAMIE
be a significant element contributing to the difference in operative
and MIE have the same postoperative mortality.
time among RAMIE users.37 Moreover, a robotic‐assisted procedure
Although few reports have been published on robotic esophagec-
implies a machine time setup that affects the entire operative time.
tomy, it has been widely used in reflux disease and achalasia. Mi
Blood loss is also a very important index in surgery. In this
et al41 demonstrated through meta‐analysis that gastric fundus fold-
meta‐analysis, RAMIE had a lower EBL than MIE. In six studies, clear
ing had a less complication occurrence in robotic surgery when com-
data were given on blood loss. In all studies, it was concluded that
pared with conventional endoscopic surgery. In addition, Maeso
the intraoperative blood loss in the RAMIE group was lower when
et al42 concluded that the probability of perforations of Heller
28
compared with that in the MIE group. Deng et al
showed that
myotomies was smaller, but that there were no significant differences
RAMIE had significantly less blood loss than MIE. It was believed
between Nissen fundoplications and gastric bypasses. Heller
that the meticulous dissection was performed under excellent visual-
myotomies appeared to significantly reduce intraoperative mucosal
ization of the robot, indicating less invasiveness of RAMIE. These
perforation. Thus, it seems that the robot can better handle esopha-
findings were consistent with our results. He et al29 believed that
geal pathology. Gastric bypass may be more likely conversion to open
RAMIE and MIE had the same safety and effectiveness in intraoper-
with a robot. The authors suggested that these results may be limited
ative and short‐term outcomes. Although in other studies, it has
by the size of the sample. In two studies, the conversion to thoracot-
been concluded that the blood loss of RAMIE was less when com-
omy was reported. Park et al24 had one case in each of the two
pared with that of MIE, differences were not statistically different.
groups; however, details were not given. Weksler described one case
JIN
9 of 11
ET AL.
in the MIE group, in which the patient suffered from chronic obstruc-
when clinical and functional outcomes were considered; thus,
tive pulmonary disease. He could not tolerate single lung ventilation
observer bias is an important limitation. Third, all the included studies
and required intermittent ventilation of the right lung. Bleeding during
were case‐control studies. Six of the other seven studies were retro-
the MIE and surgery was terminated. None of the other five studies
spective studies with small samples, which made the meta‐analysis
involved conversion to thoracotomy. Other important postoperative
subjected to the bias of the study design. More high‐quality studies
complications such as anastomotic leak, empyema, arrhythmia, pneu-
comparing RAMIE with MIE are expected in the future. Therefore,
monia, and chylothorax showed no significant differences between
we are looking forward to more high‐quality meta‐analysis publication
the two techniques (Table 4). Among them, the results of the anasto-
in the near future.44
motic leak had a high heterogeneity. We analyzed that the cause of the high heterogeneity may be associated with the different anastomotic methods. Park et al19 and Suda et al21 did not give the details of anastomotic methods. Chao et al25 and Weksler et al27 mentioned that Stapler anastomosis was used in both RAMIE and MIE groups. Pneumonia is a major complication of esophagectomy. However, the samples we included in this meta‐analysis were extremely small, resulting in a not statistically significant difference. In other words, other possible sources of heterogeneity might be the generic definition of pneumonia, the unknown patients' smoking habit, and the chronic pulmonary disease. Therefore, we need a more precise and persuasive definition of pneumonia and more high‐quality articles to
5
|
CO NC LUSIO NS
This study indicated that RAMIE and MIE mainly display similar effects and safety in the treatment of esophageal cancer. In lymph node dissection, RAMIE could reduce the risk of nerve damage due to its good visual field and flexibility, and the intraoperative blood loss is less than MIE. With the development of robotic‐assisted techniques and thoracic surgeons gaining experience, RAMIE might be beneficial for an increasing number of patients. However, randomized controlled studies with larger sample sizes are needed to evaluate the benefit and harm of RAMIE used in patients with esophageal cancer.
perform a further research. Two articles had analyzed the long‐term survival. Park et al19
ACKNOWLEDGEMENTS
demonstrated that the 5‐year overall survivals were not different between the two groups (69% in the RAMIE group vs 59% in the MIE group; P = 0.737). The 5‐year freedom from locoregional recurrence was 88% in the RAMIE group and 74% in the MIE group. However, the difference was not statistically significant (P = 0.1). The 5‐ year freedom from distant recurrence was not different between the two groups (72% in the RAMIE group and 71% in the MIE group; 28
P = 0.594). Weksler et al
demonstrated that the median survival
The authors thank Liang Yao, Peijing Yan (Clinical Research and Evidence Based Medicine of Gansu Province People's Hospital) for their help and support in the methodology and meta process. This work was supported by Health Industry Plan of GanSu Province (grant number GSWSKY2017‐56) and the Hospital Internal Scientific Research Foundation of Gansu Province People's Hospital (grant number 16GSSY3‐1).
was not significantly different among the two groups, with a median survival of 48 months (95% CI: 34 to 55) after RAMIE, 49 months (95% CI: 32 to 55) after MIE. The latest international guidelines for patients with esophageal and esophagogastric junction cancer do not include recommendations 43
for RAMIE.
AUTHOR CONT R IBUT IONS Experimental ideation and design: D.C.J., L.Y., and Y.J.G. Experiment implementation: D.C.J., L.Y., R.L., and T.K.G. Data analysis: D.C.J., K. H.Y., L.Y., and J.Y. Manuscript writing: D.C.J. and L.Y.
National Comprehensive Cancer Network only recom-
mended MIE in patients with esophageal cancer because MIE is still
ORCID
a better known treatment option, while the use of a new technology
Dacheng Jin
https://orcid.org/0000-0002-4213-3544
such as RAMIE may take more time for surgeons. In some publications, initial effectiveness and safety of RAMIE have been demonstrated with short‐term outcomes. Nevertheless, long‐term overall survival data were not reported yet. No technical problems with RAMIE have been reported, and this result encourages a further exploration of RAMIE potentials. The costs are also an important issue regarding the use of RAMIE. High costs may result an obstacle in considering possible benefits of
RE FE RE NC ES 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet‐Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65(2):87‐108. 2. Lagergren J, Lagergren P. Recent developments in esophageal adenocarcinoma. CA Cancer J Clin. 2013;63(4):232‐248. 3. Paul S, Altorki N. Outcomes in the management of esophageal cancer. J Surg Oncol. 2014;110(5):599‐610.
robotic surgery. In order to accelerate developments and revise costs, we expect more competitors entering the market. In this way, researchers may evaluate the real cost effect of robotic assistance.
4. Mao T, Fang W, Gu Z, Guo X, Ji C, Chen W. Comparison of perioperative outcomes between open and minimally invasive esophagectomy for esophageal cancer. Thorac Cancer. 2015;6(3):303‐306.
Several limitations exist in this meta‐analysis. First, some outcomes between these two techniques had high heterogeneity and could not be significantly decreased even using a random‐effect model. Second, all the included studies did not use a blinding method
5. Takeuchi H, Miyata H, Ozawa S, et al. Comparison of short‐term outcomes between open and minimally invasive esophagectomy for esophageal cancer using a nationwide database in Japan. Ann Surg Oncol. 2017;24(7):1821‐1827.
10 of 11
JIN
ET AL.
6. Barbour AP, Cormack OMM, Baker PJ, et al. Long‐term health‐related quality of life following esophagectomy: a nonrandomized comparison of thoracoscopically assisted and open surgery. Ann Surg. 2017; 265(6):1158‐1165.
24. Park S, Hwang Y, Lee HJ, Park IK, Kim YT, Kang CH. Comparison of robot‐assisted esophagectomy and thoracoscopic esophagectomy in esophageal squamous cell carcinoma. J Thorac Dis. 2016;8(10): 2853‐2861.
7. Ahmadi N, Crnic A, Seely AJ, et al. Impact of surgical approach on perioperative and long‐term outcomes following esophagectomy for esophageal cancer. Surg Endosc. 2018;32(4):1892‐1900.
25. Chao YK, Hsieh MJ, Liu YH, Liu HP. Lymph node evaluation in robot‐ assisted versus video‐assisted thoracoscopic esophagectomy for esophageal squamous cell carcinoma: a propensity‐matched analysis. World J Surg. 2018;42(2):590‐598.
8. Briez N, Piessen G, Torres F, Lebuffe G, Triboulet JP, Mariette C. Effects of hybrid minimally invasive oesophagectomy on major postoperative pulmonary complications. Br J Surg. 2012;99(11):1547‐1553. 9. Luketich JD, Pennathur A, Awais O, et al. Outcomes after minimally invasive esophagectomy: review of over 1000 patients. Ann Surg. 2012;256(1):95‐103. 10. Fuchs HF, Harnsberger CR, Broderick RC, et al. Mortality after esophagectomy is heavily impacted by center volume: retrospective analysis of the nationwide inpatient sample. Surg Endosc. 2017;31(6): 2491‐2497. 11. Luketich JD, Alvelo‐Rivera M, Buenaventura PO, et al. Minimally invasive esophagectomy: outcomes in 222 patients. Ann Surg. 2003; 238(4):486‐494. 12. Tekkis PP, Senagore AJ, Delaney CP, Fazio VW. Evaluation of the learning curve in laparoscopic colorectal surgery: comparison of right‐ sided and left‐sided resections. Ann Surg. 2005;242(1):83‐91. 13. Tapias LF, Morse CR. Minimally invasive Ivor Lewis esophagectomy: description of a learning curve. J Am Coll Surg. 2014;218(6): 1130‐1140. 14. Kernstine KH, DeArmond DT, Karimi M, et al. The robotic, 2‐stage, 3‐ field esophagolymphadenectomy. J Thorac Cardiovasc Surg. 2004;127 (6):1847‐1849. 15. Diana M, Marescaux J. Robotic surgery. Br J Surg. 2015;102(2): e15‐e28. 16. Kim DJ, Park SY, Lee S, Kim HI, Hyung WJ. Feasibility of a robot‐ assisted thoracoscopic lymphadenectomy along the recurrent laryngeal nerves in radical esophagectomy for esophageal squamous carcinoma. Surg Endosc. 2014;28(6):1866‐1873. 17. Ishikawa N, Kawaguchi M, Inaki N, Moriyama H, Shimada M, Watanabe G. Robot‐assisted thoracoscopic hybrid esophagectomy in the semi‐prone position under pneumothorax. Artif Organs. 2013; 37(6):576‐580. 18. Wee JO, Bravo‐Iniguez CE, Jaklitsch MT. Early experience of robot‐ assisted esophagectomy with circular end‐to‐end stapled anastomosis. Ann Thorac Surg. 2016;102(1):253‐259. 19. Park SY, Kim DJ, Do YW, Suh J, Lee S. The oncologic outcome of esophageal squamous cell carcinoma patients after robot‐assisted thoracoscopic esophagectomy with total mediastinal lymphadenectomy. Ann Thorac Surg. 2017;103(4):1151‐1157.
26. Suda K, Ishida Y, Kawamura Y, et al. Robot‐assisted thoracoscopic lymphadenectomy along the left recurrent laryngeal nerve for esophageal squamous cell carcinoma in the prone position: technical report and short‐term outcomes. World J Surg. 2012;36(7):1608‐1616. 27. Weksler B, Sharma P, Moudgill N, Chojnacki KA, Rosato EL. Robot‐ assisted minimally invasive esophagectomy is equivalent to thoracoscopic minimally invasive esophagectomy. Dis Esophagus. 2012;25(5):403‐409. 28. Deng HY, Huang WX, Li G, et al. Comparison of short‐term outcomes between robot‐assisted minimally invasive esophagectomy and video‐ assisted minimally invasive esophagectomy in treating middle thoracic esophageal cancer. Dis Esophagus. 2018;31(8):1‐7. 29. He H, Wu Q, Wang Z, et al. Short‐term outcomes of robot‐assisted minimally invasive esophagectomy for esophageal cancer: a propensity score matched analysis. J Cardiothorac Surg. 2018;13(1):52‐56. 30. Yerokun BA, Sun Z, Yang CJ, et al. Minimally invasive versus open esophagectomy for esophageal cancer: a population‐based analysis. Ann Thorac Surg. 2016;102(2):416‐423. 31. Wells GA, Shea B, O'Connell D, Peterson J, Welch V, Losos M(2014) The Newcastle‐Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta‐analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm. Accessed 2009 Oct 19. 32. Egger M, Smith GD, Schneider M, Minder C. Bias in meta‐analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629‐634. 33. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5(13):1‐10. 34. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta‐analyses. BMJ. 2003;327(7414):557‐560. 35. Wei S, Chen M, Chen N, Liu L. Feasibility and safety of robot‐assisted thoracic surgery for lung lobectomy in patients with non‐small cell lung cancer: a systematic review and meta‐analysis. World J Surg Oncol. 2017;15(1):98‐106. 36. Luo RB, Montalvo D, Suzuki T, Sandler BJ, Jacobsen GR, Horgan S. Complications of robotic foregut surgery: risks and advantages. Minerva Chir. 2017;72(1):44‐60. 37. Chen PD, Wu CY, Hu RH, et al. Robotic major hepatectomy: is there a learning curve? Surgery. 2017;161(3):642‐649.
20. Boone J, Borel Rinkes IHM, van Hillegersberg R. Robot‐assisted thoracolaparoscopic esophagolymphadenectomy for esophageal cancer. Surg Endosc. 2007;21(12):2342‐2343.
38. Weksler B, Sullivan JL. Survival after esophagectomy: a propensity‐ matched study of different surgical approaches. Ann Thorac Surg. 2017;104(4):1138‐1146.
21. Suda K, Nakauchi M, Inaba K, Ishida Y, Uyama I. Robotic surgery for upper gastrointestinal cancer: current status and future perspectives. Dig Endosc. 2016;28(7):701‐713.
39. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349(22):2117‐2127.
22. Bansal SS, Dogra T, Smith PW, et al. Cost analysis of open radical cystectomy versus robot‐assisted radical cystectomy. BJU Int. 2018;121(3):437‐444.
40. Li X, Wei L, Shang W, et al. Trace and evaluation systems for health services quality in rural and remote areas: a systematic review. J Public Health. 2018;26(2):127‐135.
23. Almamar A, Alkhamesi NA, Davies WT, Schlachta CM. Cost analysis of robot‐assisted choledochotomy and common bile duct exploration as an option for complex choledocholithiasis. Surg Endosc. 2018;32 (3):1223‐1227.
41. Mi J, Kang Y, Chen X, Wang B, Wang Z. Whether robot‐assisted laparoscopic fundoplication is better for gastroesophageal reflux disease in adults: a systematic review and meta‐analysis. Surg Endosc. 2010; 24(8):1803‐1814.
JIN
11 of 11
ET AL.
42. Maeso S, Reza M, Mayol JA, et al. Efficacy of the Da Vinci surgical system in abdominal surgery compared with that of laparoscopy: a systematic review and meta‐analysis. Ann Surg. 2010;252(2):254‐262. 43. Ajani JA, D' Amico TA, Almhanna K, et al. Esophageal and esophagogastric junction cancers, version 1.2015. J Natl Compr Canc Netw. 2015;13(2):194‐227. 44. Yao L, Sun R, Chen YL, et al. The quality of evidence in Chinese meta‐ analyses needs to be improved. J Clin Epidemiol. 2016;74:73‐79.
How to cite this article: Jin D, Yao L, Yu J, et al. Robotic‐ assisted minimally invasive esophagectomy versus the conventional minimally invasive one: A meta‐analysis and systematic review. Int J Med Robotics Comput Assist Surg. 2019;15: e1988. https://doi.org/10.1002/rcs.1988
Revised: 13 January 2019
Accepted: 28 January 2019
DOI: 10.1002/rcs.1988
REVIEW ARTICLE
Robotic‐assisted minimally invasive esophagectomy versus the conventional minimally invasive one: A meta‐analysis and systematic review | Liang Yao5,6† Dacheng Jin1,2,3,4† Kehu Yang3,4 | Yunjiu Gou2
|
Jun Yu2
|
Rong Liu5
|
Tiankang Guo3
|
1
Department of Clinical Medicine, Gansu University of Traditional Chinese Medicine, Lanzhou, China
Abstract Background:
Conventional video‐assisted minimally invasive esophagectomy (MIE)
2
Department of Thoracic Surgery, Gansu Province People's Hospital, Lanzhou, China
is safe and associated with low rates of morbidity and mortality, but the two‐
3
dimensional monitor reduces eye‐hand harmony and viewing yield. Robotic‐assisted
Institution of Clinical Research and Evidence Based Medicine, Gansu Province People's Hospital, Lanzhou, China
4
Evidence‐Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou, China
minimally invasive esophagectomy (RAMIE) with its virtual reality simulators offers a realistic three‐dimensional environment that facilitates dissection in the narrow working space, but it is expensive and requires longer operative time. Therefore, the aim of this meta‐analysis was to assess the safety and feasibility of RAMIE versus MIE in
5
The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing, China
patients with esophageal cancer. Material and Methods:
PubMed, EMBASE, Cochrane library, and Chinese Biomed-
6
Clinical Division, Hong Kong Baptist University, Hong Kong, China
ical Literature databases were systematically searched up to 21 September 2018 for case‐controlled studies that compared RAMIE with MIE.
Correspondence Yunjiu Gou, Gansu Province People's Hospital, Lanzhou, China. Email: [email protected] Kehu Yang, Lanzhou University, Lanzhou, China. Email: [email protected]
Result:
Eight case‐controlled studies involving 1862 patients (931 under RAMIE
and 931 under MIE) were considered. No statistically significant difference between the two techniques was observed regarding R0 resection rate (OR = 1.1174, P = 0.8647), conversion to open (OR = 0.7095, P = 0.7519), 30‐day mortality rate (OR = 0.8341, P = 0.7696), 90‐day mortality rate (OR = 0.3224, P = 0.3329), in‐ hospital mortality rate (OR = 0.3733, P = 0.3895), postoperative complications, number of harvested lymph nodes (mean difference [MD] = 0.8216, P = 0.2039), operation time (MD = 24.3655 min, P = 0.2402), and length of stay in hospitals (LOS) (MD = −5.0228 day, P = 0.1342). The meta‐analysis showed that RAMIE was associated with a significantly fewer estimated blood loss (EBL) (MD = −33.2268 mL, P = 0.0075). And the vocal cord palsy rate was higher in the MIE group compared with RAMIE, and the difference was significant (OR = 0.5696, P = 0.0447). Conclusion:
This meta‐analysis indicated that RAMIE and MIE display similar feasi-
bility and safety when used in esophagectomy. However, randomized controlled studies with larger sample sizes are needed to evaluate the benefit and harm in patients with esophageal cancer undergoing RAMIE.
Abbreviations: CI, confidence interval; EBL, estimated blood loss; LOS, length of stay; MD, mean difference; MIE, minimally invasive esophagectomy; NOS, Newcastle‐ Ottawa scale; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy; SD, standard deviation † First two authors D.C.J. and L.Y. contributed equally to this work and were considered as co‐first authors.
Int J Med Robotics Comput Assist Surg. 2019;15:e1988. https://doi.org/10.1002/rcs.1988
wileyonlinelibrary.com/journal/rcs
© 2019 John Wiley & Sons, Ltd.
1 of 11
2 of 11
1
JIN
I N T RO D U CT I O N
|
2.2
Esophageal cancer is one of the most common neoplasms of the 1
|
ET AL.
Study selection
Studies selected for this meta‐analysis were included according to the
digestive system. At present, esophagectomy is an efficient way to
following criteria: Case‐control studies that compared RAMIE and MIE
treat esophageal cancer. With the improvements in video‐assisted sur-
in the treatment of esophageal neoplasm, original studies, esophagec-
gery, conventional minimally invasive esophagectomy (MIE) has gained
tomy, and reporting at least one of the outcome measures mentioned
2-5
Multiple single hospitals have
below. After screening of the titles, abstracts and full text were
reported that this procedure is safe, resulting in lower rates of morbid-
screened by one researcher, and studies were excluded if the
more attention in recent years.
ity and mortality and transthoracic esophagectomy performed through
abovementioned standards were not confirmed or had a low‐quality
a reduced thoracic wound.6-11 Several articles demonstrated that rad-
score (<7) according to the Newcastle‐Ottawa scale (NOS).
ical surgery using MIE to treat esophageal carcinoma can narrow the rates of pulmonary complications and lower the mortality rate.4,5,9 However, conventional MIE has also some limitations. The two‐
2.3
|
Outcomes of interest
dimensional monitor reduces eye‐hand harmony and viewing yield. Difficulties increase when dealing with complex surgical procedures.12
The outcome of interest consisted of operative time (min), estimated
Furthermore, the high technical complexity of a minimally invasive
blood loss (EBL) (mL), length of stay (LOS) (days), number of harvested
procedure may lead to a complanate learning curve.13 Thus, these fac-
lymph nodes, mortality after operation, R0 resection rate, and compli-
tors may limit the development of this technique.
cations. Operative time included the total and thoracic operative time.
Robotic‐assisted minimally invasive esophagectomy (RAMIE) is a relatively new minimally invasive technique. The first robotic‐assisted
Postoperative mortality was defined as death within 30 days, 90 days, and in‐hospital after the esophagectomy.
two‐stage, three‐field esophagectomy was reported by Kernstine et al in 2004.14 RAMIE is offering more advantages than conventional MIE. RAMIE offers realistic three‐dimensional immersion environments that should theoretically facilitate dissection in the narrow working space up to the mediastinum, and it is technically feasible and safe in terms of oncological outcomes.15-20 Nevertheless, the high costs, longer operation time, and the lack of haptic feedback are the most common disadvantages revealing the need of more progresses in this technology.2123
Up to now, no meta‐analysis is available reporting the comparison
between RAMIE and MIE, and no randomized controlled trials were published. However, several case‐control studies show comparable
2.4
|
Data extraction and quality assessment
Two investigators independently extracted the data from the included studies according to the parameters mentioned above and summarized them. Selected studies were all case‐control studies; thus, NOS was used for grading31 (Table 2). All included studies had a high‐ quality score according to NOS. Discrepancies were resolved by discussion.
outcomes and survival after esophagectomy.24-30 Therefore, the aim of our meta‐analysis was to evaluate the safety and effectiveness of RAMIE versus MIE in patients who underwent esophagectomy.
2.5
|
Publication bias
If the number of the included studies was more than 10, the publication bias was evaluated. However, less than 10 studies were included
2 2.1
MATERIAL AND METHODS
|
|
in our studies.32
Literature search 2.6
|
Statistical analysis
A comprehensive literature search was conducted to find all relevant trials using the following databases: PubMed, EMBASE, Cochrane Library,
The meta‐analysis was performed using R software (R x64 3.4.2, The
and the Chinese Biomedical Literature Database. The following search
Cochrane Collaboration, Oxford, UK). Dichotomous variables were
terms were used: “Esophageal Neoplasm,” “Neoplasm, Esophageal,”
pooled using the odds ratio (OR) with a 95% confidence interval (CI).
“Esophagus Neoplasm,” “Esophagus Neoplasms,” “Cancer of Esopha-
Because the events were rare, the random effect was used in this
gus,” “Esophagus Cancer,” “esophagectomy,” “oesophagectomy,” and
meta‐analysis. Continuous variables were pooled using the mean dif-
“robot*.” We used medical subject heading terms (MeSH) and free
ference (MD) with a 95% CI. P values <0.05 were considered statisti-
words as keywords. To identify if our search was comprehensive, the
cally significant. In studies that only reported medians and ranges, the
characteristics of both the control group and study language were not
mean and standard deviation (SD) were assessed using the means of
considered. Two authors independently evaluated the included studies,
the method provided by Hozo et al.33 The studies' statistical heteroge-
and disagreements were resolved through discussion. The latest search
neity was evaluated through the I2 statistic.34 I2 value <25% indicated
was made on 21 September 2018.
low heterogeneity, and a value >50% indicated high heterogeneity.
JIN
3 of 11
ET AL.
2.7
|
Ethical approval
3.1
|
LOS
The article did not involve ethical approval because our study is a sec-
The total LOS after operation was indicated in six studies. Four of
ondary research, not involved with patients recruited.
them showed a shorter LOS in patients who underwent RAMIE than MIE. He et al29 and Deng et al28 had a shorter LOS in the MIE group. Pooled data analysis demonstrated no statistically significant
3
|
RESULTS
difference between these two techniques (MD = −0.5381 day, 95% CI [−2.4750; 1.3987]; P = 0.5861), with a high heterogeneity
A total number of 953 articles were found by the initial literature
(I2 = 77.9%; P = 0.0004) (Figure 2A).
search (Figure 1). One article was added after finding it through other sources. After duplicate removal, 906 results remained. After a careful check of titles and abstracts, 869 articles were excluded, because the excluded studies were case reports or did not compare the two types
3.2
|
Operative time
of surgery at the same time. The 37 remaining articles were further assessed for eligibility by a full examination of the text. Finally, eight
Operative time was reported in five studies. In particular, three studies
relevant studies were included in this meta‐analysis. This meta‐
indicated a longer operative time with RAMIE, and two studies
analysis included 1862 patients, in which, 931 were subjected to
showed a shorter operative time with RAMIE compared with MIE.
RAMIE and 931 were subjected to MIE. All studies were of high‐
Pooled data analysis demonstrated no statistically significant differ-
quality scores based on NOS. The characteristics and methodological
ence
quality assessment scores of the included studies are summarized in
(MD = 24.3655 min, 95% CI [−16.2929; 65.0238]; P = 0.2402), with
Tables 1 and 2.
a heterogeneity (I2 = 48.7%; P = 0.05) (Figure 2B).
FIGURE 1
Flow chart of selection for included studies
in
operative
time
between
these
two
techniques
4 of 11
JIN
TABLE 1
ET AL.
Characteristics of the selected studies included in the meta‐analysis
Study
Year
Study Type
Data Source
Country
Design
Group
Park et al
2016
RO
OD
South Korea
CCS
RAMIE
62
57 5
64.3 ± 8.0
MIE
43
40 3
64.3 ± 8.0
23.3 ± 3.1
16
15 1
65(53‐86)
21.3(17.5‐26.3)
Number
Sex (M/F)
Age y (Mean ± SD)
BMI (kg/m2) (Mean ± SD) 23.5 ± 2.8
Suda et al
2012
RO
OD
Japan
CCS
RAMIE MIE
20
15 5
64.5(50‐79)
20.4 (14.9‐24.8)
Weksler et al
2012
RO
OD
American
CCS
RAMIE
11
83
58.7 ± 8.5
27.1
MIE
26
20 6
64.3 ± 11.3
27.9
34
32 2
56.76 ± 8.39
NA
Chao et al
2018
RO
OD
TaiWan
CCS
RAMIE MIE
34
33 1
53.47 ± 8.69
NA
Deng et al
2018
RO
OD
China
CCS
RAMIE
42
35 7
60.7 ± 6.9
NA
MIE
42
33 9
61.8 ± 9.5
NA
He et al
2018
RO
OD
China
CCS
RAMIE
27
20 7
61 ± 8
21.5 ± 2.7
27
20 7
61.6 ± 9.8
21.9 ± 2.8
Babatunde et al
2016
RO
OD
Canada
CCS
RAMIE
170
142 28
56/64/70
NA
MIE
170
143 27
56/63/69
NA
RAMIE
569
471 98
62.9 ± 9.6
NA
MIE
569
489 80
62.8 ± 9.3
NA
MIE
Weksler et al
2017
RO
NCDB
American
CCS
Abbreviations: BMI, body mass index; CCS, case‐control study; IL, Ivor Lewis; MIE, minimally invasive esophagectomy; MK, McKeown; NA, not available; NCDB, The National Cancer Data Base; NOS, Newcastle‐Ottawa Scale; OD, original data; RAMIE, robotic‐assisted minimally invasive esophagectomy; RO, retrospective observational.
3.3
|
Harvested number of lymph node
3.6
|
Vocal cord palsy
In all studies, the number of harvested lymph nodes was reported. In
Six studies showed the vocal cord palsy rate. The meta‐analysis indi-
five studies, it was shown that RAMIE patients were subjected to
cated that MIE has a higher rate of vocal cord palsy incidence than
the removal of a higher number of lymph nodes compared with MIE
RAMIE (OR = 0.5696, 95% CI [0.3288; 0.9868]; P = 0.0447) with a
patients. Moreover, in two studies, the same number of harvested
low heterogeneity (I2 = 0%; P = 0.4932) (Figure 5).
lymph nodes was reported, while in one study, it was reported that RAMIE patients were subjected to the removal of a smaller number of lymph nodes compared with MIE patients. Pooled data analysis
3.7
|
Conversion rate
showed that the differences were not statistically significant (MD = 0.8216, 95% CI [−0.4458; 2.0890]; P = 0.2039), with a high
Two studies reported the conversion rate. The meta‐analysis indicated no significant differences between these two techniques (OR = 0.7095,
2
heterogeneity (I = 50%; P = 0.0501) (Figure 3A).
95% CI [0.0845; 5.9588]; P = 0.7519) with a low heterogeneity (I2 = 0%; P = 0.9742) (Figure 6).
3.4
|
EBL 3.8
|
Mortality after operation
EBL was described in six studies. The meta‐analysis indicated that RAMIE had a lower EBL than MIE. (MD = −33.2268 mL, 95% CI
Postoperative mortality was defined as death within 30 days, 90 days,
[−57.5870; −8.8666]; P = 0.0075) with a low heterogeneity (I = 0;
and in‐hospital. In six studies, the 30‐day mortality rate was reported,
P = 0.7076) (Figure 3B).
in two studies the 90‐day mortality rate was reported, and the in‐
2
hospital mortality rate was reported in two studies. Pooled data analysis showed that differences were not statistically significant between these mortality rates. The 30‐day mortality rate (OR = 0.8341, 95% CI
3.5
|
R0 resection rate
[0.2478; 2.8079]; P = 0.7696) had a low heterogeneity (I2 = 0%; P = 0.8310), the 90‐day mortality rate (OR = 0.3224, 95% CI
Two studies reported R0 resection rate. The meta‐analysis indicated
[0.0326; 3.1879]; P = 0.3329) had a heterogeneity of I2 = 0%
no statistically significant differences between these two techniques
(P = 0.9974), and the in‐hospital mortality rate (OR = 0.3733, 95% CI
(OR = 1.1174, 95% CI [0.6760; 1.8472]; P = 0.8647) with a low het-
[0.0396; 3.5224]; P = 0.3895) had a heterogeneity of I2 = 0%
2
erogeneity (I = 0%; P = 0.7802) (Figure 4).
(P = 0.9045) (Table 3).
JIN
5 of 11
ET AL.
TABLE 1 Upper Thoracic
Study Park et al
Suda et al
Mid Thoracic
Neoadjuvant Chemoradiation
Tumor Size (cm)
Surgical Procedures
NOS (Score) 8
15 (24.2%)
39(62.9%)
101.6 ± 17.1
8(12.9%)
NA
MK
7(16.3%)
9 (20.9%)
27(62.8%)
106.7 ± 13.8
4(9.3%)
NA
MK
7
9
NA
6
3.75(0.9‐8.0) 3.65(0.0‐7.0)
NA
2
12
6
NA
17
NA
NA
NA
NA
4
NA
IL
NA
NA
NA
NA
10
NA
IL
10(29.4%)
15(44.1%)
9(26.5%)
NA
17(50%)
NA
MK
10(29.4%)
19(55.9%)
9(26.5%)
NA
17(50%)
NA
MK
NA
NA
NA
NA
NA
NA
MK
NA
NA
NA
NA
NA
NA
MK
1(3.7)
18(66.6)
8(29.6)
94.6 ± 13.8
NA
NA
MK
3(11.1)
15(55.6)
9(33.3)
92.9 ± 23
NA
NA
MK
Chao et al
Deng et al
FEV1 (pred%, Mean ± SD)
8(12.9%)
2
Weksler et al
Lower Thoracic
He et al
Babatunde et al
NA
156
14
NA
160
Weksler et al
NA
513
NA
527
NA
7
NA
NA
70.6% (120)
20.0/35.0/50.0
NA
10
NA
70.6% (120)
22.5/35.0/50.0
NA
NA
NA
405 (71.2%)
33.7(17‐57)
NA
NA
401 (70.5%)
35.0 (15‐55)
NA
7
7
8
7
7
8
Abbreviations: BMI, body mass index; CCS, case‐control study; IL, Ivor Lewis; MIE, minimally invasive esophagectomy; MK, McKeown; NA, not available; NCDB, The National Cancer Data Base; NOS, Newcastle‐Ottawa Scale; OD, original data; RAMIE, robotic‐assisted minimally invasive esophagectomy; RO, retrospective observational.
TABLE 2
The Newcastle‐Ottawa scale Selection 1
Study
2
Comparability 3
4
1
Exposure 1
2
3
Quality Score
Park et al 2016
*
*
0
*
**
*
*
*
8
Suda et al 2012
*
*
0
*
*
*
*
*
7
Babatunde et al 2016
*
*
0
*
*
*
*
*
7
Weksler et al 2012
*
*
0
*
*
*
*
*
7
Chao et al 2018
*
*
0
*
*
*
*
*
7
Deng et al 2018
*
*
0
*
**
*
*
*
8
He et al 2018
*
*
0
*
*
*
*
*
7
Weksler et al 2017
*
*
0
*
**
*
*
*
8
*One point.
3.9
|
Postoperative complications
perform esophagectomy. This meta‐analysis did not show statistically significant differences between MIE and RAMIE regarding operative
Postoperative complications included the anastomotic leak, empyema,
time, mortality rate, hospital stay, harvested lymph nodes, R0 resec-
pneumonia, arrhythmia, and chylothorax. Table 4 presents the summa-
tion rate, and some complications. However, we demonstrated that
rized results. No statistically significant differences were observed in
the intraoperative blood loss in RAMIE was less when compared with
postoperative complications.
MIE. Regarding the vocal cord palsy rate, the MIE group had a higher rate than the RAMIE group. Patients with esophageal cancer always have a high cancer met-
4
|
DISCUSSION
astatic rate on the recurrent laryngeal nerve (RLN) lymph node. Hence, having an accurate lymph node dissection during the esoph-
Over the past decade, an increasing number of medical institutions
agectomy is significantly important. Our meta‐analysis indicated no
have reported progress on using RAMIE. To our knowledge, this is
significant difference in the number of harvested lymph nodes. How-
the first meta‐analysis comparing these two surgical techniques to
ever, four studies24,25,28,29 demonstrated that RAMIE allows
6 of 11
FIGURE 2 or MIE
JIN
ET AL.
A, LOS meta‐analysis in patients who underwent RAMIE or MIE; B, operative time meta‐analysis in patients who underwent RAMIE
FIGURE 3 A, Number of harvested lymph nodes meta‐analysis in patients who underwent RAMIE or MIE; B, EBL meta‐analysis in patients who underwent RAMIE or MIE
FIGURE 4
R0 resection rate meta‐analysis in patients who underwent RAMIE or MIE
achieving a higher mean number of harvested lymph nodes along the 25
rate. Park et al24 concluded that RAMIE was superior to MIE in
stated that without increasing RLN palsy and pul-
lymph node dissection. Moreover, Deng et al28 and He et al29 con-
monary complications rates, RAMIE could still fulfill a high removal
cluded that RAMIE had more advantages than MIE's lymph node
RLN area. Chao
JIN
7 of 11
ET AL.
FIGURE 5
Vocal cord palsy rate meta‐analysis in patients who underwent RAMIE or MIE
FIGURE 6
Conversion rate meta‐analysis in patients who underwent RAMIE or MIE
TABLE 3
Mortality after operation for RAMIE versus MIE
Mortality
No. of Studies
RAMIE
MIE
I2
Effect Measure (OR)
Pooled Effect 95%CI
P Value 0.7696
30‐day mortality
6
335
335
0
0.8341
[0.2478; 2.8079]
90‐day mortality
2
61
61
0
0.3224
[0.0326; 3.1879]
0.3329
In‐hospital mortality
3
61
80
0
0.3733
[0.0396; 3.5224]
0.3895
Abbreviations: CI, confidence interval; MIE, minimally invasive esophagectomy; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy.
TABLE 4
Summary of complications for RAMIE versus MIE Effect Measure (OR)
Analysis Model
Pooled effect 95%CI
P Value
0
0.98
RE
[0.3434; 3.1936]
0.95
0
0.43
RE
[0.1091; 1.7063]
0.23
89
0
0.45
RE
[0.0774; 2.5751]
0.37
61
80
0
0.55
RE
[0.1712; 1.7710]
0.32
192
192
3.1%
2.06
RE
[0.8047; 5.2793]
0.13
Complications
No. of Studies
RAMIE
MIE
Pneumonia
5
176
172
Arrhythmia
3
54
73
Chylothorax
3
85
Empyema
3 6
Anastomotic leakage
I2
Abbreviations: CI, confidence interval; MIE, minimally invasive esophagectomy; OR, odds ratio; RAMIE, robotic‐assisted minimally invasive esophagectomy; RE, random effect.
dissection; however, these results were not significantly different.
results when RAMIE was used according to the following evidences:
This may be limited by the relatively low sample size. Weksler
First, compared with the traditional two‐dimensional monitor, the
et al27 and Babatunde et al30 concluded that the number of lymph
three‐dimensional self‐controlled magnified view provided by the
nodes dissected by RAMIE and MIE was similar, which was consis-
robotic camera allows a better visualization of the upper mediasti-
tent with our findings. The dissection of nodes located ventrally to
num. Second, even in a limited anatomical space, RAMIE provides
the left RLN is the most challenging part of the surgical procedure.
a meticulous surgical assistance. Third, the third robot arm which is
Several novel approaches have been recently proposed during MIE
controlled by the operator with the application of a stable and
in order to improve the operative exposure and reduce RLN injury.
self‐controllable traction can ensure a safer and easier approach to
However, in spite of the improvement, the left RLN lymph node
nodal dissection.
remains difficult to reach using conventional MIE. In this case, the
During the lymph node dissection throughout the thoracic surgery,
guarantee of a high‐quality dissection is extremely needed, and this
the excessive exposure of the RLN may be the main cause of vocal cord
requires experienced assistants.
palsy. Bilateral RLN injury may cause dyspnea and endanger the
Despite our meta‐analysis, evidences are lacking demonstrating
patient's life. Therefore, it is extremely important to reduce the inci-
a better performance of RAMIE; we conferred good oncologic
dence of RLN injury. However, some present studies reported that the
8 of 11
JIN
ET AL.
incidence of vocal cord palsy was high in both RAMIE and MIE.21,24,26
It is difficult to precisely measure blood loss because its measure-
This result may be due to the use of a flexible laryngoscope by the
ments might be associated with surgeons' estimation. In addition,
doctors even in the absence of any clinical manifestation. As a result,
the possibility that the surgeons are reporting a reduced blood loss
the incidence of patients with vocal cord palsy was higher than normal.
cannot be ignored. Six studies have been analyzed in terms of blood
In the study we included, patients with RLN injury showed unilateral
loss, but the patient sample size was too small to have statistical
injury. Thus, patients' symptoms were in all cases vocal cord paralysis.
results. More high‐quality reports using a large database should be
Our meta‐analysis showed that MIE had higher RLN injury than
performed to identify patients who underwent esophagectomy
RAMIE. Among them,24-26,28 the rates of vocal cord palsy in the MIE
because of esophageal cancer, together with a documentation of
group were significantly higher when compared with that in the
the surgical approach.
RAMIE group, which was consistent with our findings. This result
The mortality rate is a safe outcome, and in our meta‐analysis it
might be due to the advantage in the lymph node dissection under
resulted in no significant differences between MIE and RAMIE. Post-
the RAMIE, which allows a better performance in a good field of vision.
surgical mortality was defined as death within 30 days, 90 days, and
However, due to the limited number of studies we included, this view
in‐hospital stays. Of the 30 mortality rates, we included six studies.
needs to be further proven in the future. Therefore, our conclusion
In two of the studies, deaths were not reported. In one study, three
was that, although this was a positive outcome according to our
deaths were reported in both the RAMIE and MIE groups. No intra-
results, it is not sufficient to confirm that MIE's lymph node dissection
operative deaths were reported in the studies that were included in
is not effective.
this study. The cause of death was mostly due to other diseases, such
In our meta‐analysis, no significant differences were found in
as chronic diseases, cardiovascular diseases etc. Similarly, 90 mortality
operative time between the two techniques. Weksler et al27
and in‐hospital mortality were associated for this reason. In general,
reported that the operative time did not increase significantly with
previous studies confirmed no differences in death rate between
the use of the robot because of the faster thoracic mobilization.
RAMIE and MIE, and the safety of RAMIE and MIE has been widely
Most of the articles have a comparison of this intraoperative out-
confirmed, which is consistent with our findings. In 2017, Weksler
come and always show a positive result.24,29,35,36 Suda et al26 and
et al38 published an article comparing open esophagectomy, MIE,
27
concluded that the surgical time of MIE was longer
and RAMIE. A total of 1707 patients with esophageal cancer were
when compared with that of RAMIE. We speculated that the setup
Weksler et al
included in Weksler's study after a propensity matching method.
time of the robot was excluded and that simply the effective surgical
Weksler concluded that the 30‐day mortality rate in RAMIE was
time was compared. The question that comes to mind is, whether
higher when compared with that in the MIE group (5.6 vs 2.8); how-
the beginning of anesthesia in robotic surgery should wait until the
ever, no significant differences were observed in 90‐day mortality.
robot is fully installed. The shorter length of the anesthetic may
Weksler believed that the mortality was greater than 20% when
accelerate a patient's recovery after surgery. If, in the future, the
low‐volume surgeons operated in low‐volume hospitals.39,40 Although
effective surgical time of the robot can be proven to be shorter than
MIE clearly demonstrated benefits in reducing postoperative morbid-
that of MIE, the advantages of the robot will be highlighted. The
ity, it is unclear whether MIE and RAMIE decrease short‐term mortal-
data extracted from the included studies show differences regarding
ity among patients undergoing esophagectomy. In other studies that
the operative time because of many different reasons. First of all,
were included, we confirmed that there was no difference in mortality
different understanding in the robotic‐assisted system by different
between RAMIE and MIE at 30 days, 90 days, and during hospital
surgeons always resulted in a discrepancy in the operative time.
stay. Similarly, both RAMIE and MIE have been widely confirmed to
Learning curve was connected with surgeons' experience and may
be safer than open. This is consistent with the conclusion that RAMIE
be a significant element contributing to the difference in operative
and MIE have the same postoperative mortality.
time among RAMIE users.37 Moreover, a robotic‐assisted procedure
Although few reports have been published on robotic esophagec-
implies a machine time setup that affects the entire operative time.
tomy, it has been widely used in reflux disease and achalasia. Mi
Blood loss is also a very important index in surgery. In this
et al41 demonstrated through meta‐analysis that gastric fundus fold-
meta‐analysis, RAMIE had a lower EBL than MIE. In six studies, clear
ing had a less complication occurrence in robotic surgery when com-
data were given on blood loss. In all studies, it was concluded that
pared with conventional endoscopic surgery. In addition, Maeso
the intraoperative blood loss in the RAMIE group was lower when
et al42 concluded that the probability of perforations of Heller
28
compared with that in the MIE group. Deng et al
showed that
myotomies was smaller, but that there were no significant differences
RAMIE had significantly less blood loss than MIE. It was believed
between Nissen fundoplications and gastric bypasses. Heller
that the meticulous dissection was performed under excellent visual-
myotomies appeared to significantly reduce intraoperative mucosal
ization of the robot, indicating less invasiveness of RAMIE. These
perforation. Thus, it seems that the robot can better handle esopha-
findings were consistent with our results. He et al29 believed that
geal pathology. Gastric bypass may be more likely conversion to open
RAMIE and MIE had the same safety and effectiveness in intraoper-
with a robot. The authors suggested that these results may be limited
ative and short‐term outcomes. Although in other studies, it has
by the size of the sample. In two studies, the conversion to thoracot-
been concluded that the blood loss of RAMIE was less when com-
omy was reported. Park et al24 had one case in each of the two
pared with that of MIE, differences were not statistically different.
groups; however, details were not given. Weksler described one case
JIN
9 of 11
ET AL.
in the MIE group, in which the patient suffered from chronic obstruc-
when clinical and functional outcomes were considered; thus,
tive pulmonary disease. He could not tolerate single lung ventilation
observer bias is an important limitation. Third, all the included studies
and required intermittent ventilation of the right lung. Bleeding during
were case‐control studies. Six of the other seven studies were retro-
the MIE and surgery was terminated. None of the other five studies
spective studies with small samples, which made the meta‐analysis
involved conversion to thoracotomy. Other important postoperative
subjected to the bias of the study design. More high‐quality studies
complications such as anastomotic leak, empyema, arrhythmia, pneu-
comparing RAMIE with MIE are expected in the future. Therefore,
monia, and chylothorax showed no significant differences between
we are looking forward to more high‐quality meta‐analysis publication
the two techniques (Table 4). Among them, the results of the anasto-
in the near future.44
motic leak had a high heterogeneity. We analyzed that the cause of the high heterogeneity may be associated with the different anastomotic methods. Park et al19 and Suda et al21 did not give the details of anastomotic methods. Chao et al25 and Weksler et al27 mentioned that Stapler anastomosis was used in both RAMIE and MIE groups. Pneumonia is a major complication of esophagectomy. However, the samples we included in this meta‐analysis were extremely small, resulting in a not statistically significant difference. In other words, other possible sources of heterogeneity might be the generic definition of pneumonia, the unknown patients' smoking habit, and the chronic pulmonary disease. Therefore, we need a more precise and persuasive definition of pneumonia and more high‐quality articles to
5
|
CO NC LUSIO NS
This study indicated that RAMIE and MIE mainly display similar effects and safety in the treatment of esophageal cancer. In lymph node dissection, RAMIE could reduce the risk of nerve damage due to its good visual field and flexibility, and the intraoperative blood loss is less than MIE. With the development of robotic‐assisted techniques and thoracic surgeons gaining experience, RAMIE might be beneficial for an increasing number of patients. However, randomized controlled studies with larger sample sizes are needed to evaluate the benefit and harm of RAMIE used in patients with esophageal cancer.
perform a further research. Two articles had analyzed the long‐term survival. Park et al19
ACKNOWLEDGEMENTS
demonstrated that the 5‐year overall survivals were not different between the two groups (69% in the RAMIE group vs 59% in the MIE group; P = 0.737). The 5‐year freedom from locoregional recurrence was 88% in the RAMIE group and 74% in the MIE group. However, the difference was not statistically significant (P = 0.1). The 5‐ year freedom from distant recurrence was not different between the two groups (72% in the RAMIE group and 71% in the MIE group; 28
P = 0.594). Weksler et al
demonstrated that the median survival
The authors thank Liang Yao, Peijing Yan (Clinical Research and Evidence Based Medicine of Gansu Province People's Hospital) for their help and support in the methodology and meta process. This work was supported by Health Industry Plan of GanSu Province (grant number GSWSKY2017‐56) and the Hospital Internal Scientific Research Foundation of Gansu Province People's Hospital (grant number 16GSSY3‐1).
was not significantly different among the two groups, with a median survival of 48 months (95% CI: 34 to 55) after RAMIE, 49 months (95% CI: 32 to 55) after MIE. The latest international guidelines for patients with esophageal and esophagogastric junction cancer do not include recommendations 43
for RAMIE.
AUTHOR CONT R IBUT IONS Experimental ideation and design: D.C.J., L.Y., and Y.J.G. Experiment implementation: D.C.J., L.Y., R.L., and T.K.G. Data analysis: D.C.J., K. H.Y., L.Y., and J.Y. Manuscript writing: D.C.J. and L.Y.
National Comprehensive Cancer Network only recom-
mended MIE in patients with esophageal cancer because MIE is still
ORCID
a better known treatment option, while the use of a new technology
Dacheng Jin
https://orcid.org/0000-0002-4213-3544
such as RAMIE may take more time for surgeons. In some publications, initial effectiveness and safety of RAMIE have been demonstrated with short‐term outcomes. Nevertheless, long‐term overall survival data were not reported yet. No technical problems with RAMIE have been reported, and this result encourages a further exploration of RAMIE potentials. The costs are also an important issue regarding the use of RAMIE. High costs may result an obstacle in considering possible benefits of
RE FE RE NC ES 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet‐Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65(2):87‐108. 2. Lagergren J, Lagergren P. Recent developments in esophageal adenocarcinoma. CA Cancer J Clin. 2013;63(4):232‐248. 3. Paul S, Altorki N. Outcomes in the management of esophageal cancer. J Surg Oncol. 2014;110(5):599‐610.
robotic surgery. In order to accelerate developments and revise costs, we expect more competitors entering the market. In this way, researchers may evaluate the real cost effect of robotic assistance.
4. Mao T, Fang W, Gu Z, Guo X, Ji C, Chen W. Comparison of perioperative outcomes between open and minimally invasive esophagectomy for esophageal cancer. Thorac Cancer. 2015;6(3):303‐306.
Several limitations exist in this meta‐analysis. First, some outcomes between these two techniques had high heterogeneity and could not be significantly decreased even using a random‐effect model. Second, all the included studies did not use a blinding method
5. Takeuchi H, Miyata H, Ozawa S, et al. Comparison of short‐term outcomes between open and minimally invasive esophagectomy for esophageal cancer using a nationwide database in Japan. Ann Surg Oncol. 2017;24(7):1821‐1827.
10 of 11
JIN
ET AL.
6. Barbour AP, Cormack OMM, Baker PJ, et al. Long‐term health‐related quality of life following esophagectomy: a nonrandomized comparison of thoracoscopically assisted and open surgery. Ann Surg. 2017; 265(6):1158‐1165.
24. Park S, Hwang Y, Lee HJ, Park IK, Kim YT, Kang CH. Comparison of robot‐assisted esophagectomy and thoracoscopic esophagectomy in esophageal squamous cell carcinoma. J Thorac Dis. 2016;8(10): 2853‐2861.
7. Ahmadi N, Crnic A, Seely AJ, et al. Impact of surgical approach on perioperative and long‐term outcomes following esophagectomy for esophageal cancer. Surg Endosc. 2018;32(4):1892‐1900.
25. Chao YK, Hsieh MJ, Liu YH, Liu HP. Lymph node evaluation in robot‐ assisted versus video‐assisted thoracoscopic esophagectomy for esophageal squamous cell carcinoma: a propensity‐matched analysis. World J Surg. 2018;42(2):590‐598.
8. Briez N, Piessen G, Torres F, Lebuffe G, Triboulet JP, Mariette C. Effects of hybrid minimally invasive oesophagectomy on major postoperative pulmonary complications. Br J Surg. 2012;99(11):1547‐1553. 9. Luketich JD, Pennathur A, Awais O, et al. Outcomes after minimally invasive esophagectomy: review of over 1000 patients. Ann Surg. 2012;256(1):95‐103. 10. Fuchs HF, Harnsberger CR, Broderick RC, et al. Mortality after esophagectomy is heavily impacted by center volume: retrospective analysis of the nationwide inpatient sample. Surg Endosc. 2017;31(6): 2491‐2497. 11. Luketich JD, Alvelo‐Rivera M, Buenaventura PO, et al. Minimally invasive esophagectomy: outcomes in 222 patients. Ann Surg. 2003; 238(4):486‐494. 12. Tekkis PP, Senagore AJ, Delaney CP, Fazio VW. Evaluation of the learning curve in laparoscopic colorectal surgery: comparison of right‐ sided and left‐sided resections. Ann Surg. 2005;242(1):83‐91. 13. Tapias LF, Morse CR. Minimally invasive Ivor Lewis esophagectomy: description of a learning curve. J Am Coll Surg. 2014;218(6): 1130‐1140. 14. Kernstine KH, DeArmond DT, Karimi M, et al. The robotic, 2‐stage, 3‐ field esophagolymphadenectomy. J Thorac Cardiovasc Surg. 2004;127 (6):1847‐1849. 15. Diana M, Marescaux J. Robotic surgery. Br J Surg. 2015;102(2): e15‐e28. 16. Kim DJ, Park SY, Lee S, Kim HI, Hyung WJ. Feasibility of a robot‐ assisted thoracoscopic lymphadenectomy along the recurrent laryngeal nerves in radical esophagectomy for esophageal squamous carcinoma. Surg Endosc. 2014;28(6):1866‐1873. 17. Ishikawa N, Kawaguchi M, Inaki N, Moriyama H, Shimada M, Watanabe G. Robot‐assisted thoracoscopic hybrid esophagectomy in the semi‐prone position under pneumothorax. Artif Organs. 2013; 37(6):576‐580. 18. Wee JO, Bravo‐Iniguez CE, Jaklitsch MT. Early experience of robot‐ assisted esophagectomy with circular end‐to‐end stapled anastomosis. Ann Thorac Surg. 2016;102(1):253‐259. 19. Park SY, Kim DJ, Do YW, Suh J, Lee S. The oncologic outcome of esophageal squamous cell carcinoma patients after robot‐assisted thoracoscopic esophagectomy with total mediastinal lymphadenectomy. Ann Thorac Surg. 2017;103(4):1151‐1157.
26. Suda K, Ishida Y, Kawamura Y, et al. Robot‐assisted thoracoscopic lymphadenectomy along the left recurrent laryngeal nerve for esophageal squamous cell carcinoma in the prone position: technical report and short‐term outcomes. World J Surg. 2012;36(7):1608‐1616. 27. Weksler B, Sharma P, Moudgill N, Chojnacki KA, Rosato EL. Robot‐ assisted minimally invasive esophagectomy is equivalent to thoracoscopic minimally invasive esophagectomy. Dis Esophagus. 2012;25(5):403‐409. 28. Deng HY, Huang WX, Li G, et al. Comparison of short‐term outcomes between robot‐assisted minimally invasive esophagectomy and video‐ assisted minimally invasive esophagectomy in treating middle thoracic esophageal cancer. Dis Esophagus. 2018;31(8):1‐7. 29. He H, Wu Q, Wang Z, et al. Short‐term outcomes of robot‐assisted minimally invasive esophagectomy for esophageal cancer: a propensity score matched analysis. J Cardiothorac Surg. 2018;13(1):52‐56. 30. Yerokun BA, Sun Z, Yang CJ, et al. Minimally invasive versus open esophagectomy for esophageal cancer: a population‐based analysis. Ann Thorac Surg. 2016;102(2):416‐423. 31. Wells GA, Shea B, O'Connell D, Peterson J, Welch V, Losos M(2014) The Newcastle‐Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta‐analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm. Accessed 2009 Oct 19. 32. Egger M, Smith GD, Schneider M, Minder C. Bias in meta‐analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629‐634. 33. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5(13):1‐10. 34. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta‐analyses. BMJ. 2003;327(7414):557‐560. 35. Wei S, Chen M, Chen N, Liu L. Feasibility and safety of robot‐assisted thoracic surgery for lung lobectomy in patients with non‐small cell lung cancer: a systematic review and meta‐analysis. World J Surg Oncol. 2017;15(1):98‐106. 36. Luo RB, Montalvo D, Suzuki T, Sandler BJ, Jacobsen GR, Horgan S. Complications of robotic foregut surgery: risks and advantages. Minerva Chir. 2017;72(1):44‐60. 37. Chen PD, Wu CY, Hu RH, et al. Robotic major hepatectomy: is there a learning curve? Surgery. 2017;161(3):642‐649.
20. Boone J, Borel Rinkes IHM, van Hillegersberg R. Robot‐assisted thoracolaparoscopic esophagolymphadenectomy for esophageal cancer. Surg Endosc. 2007;21(12):2342‐2343.
38. Weksler B, Sullivan JL. Survival after esophagectomy: a propensity‐ matched study of different surgical approaches. Ann Thorac Surg. 2017;104(4):1138‐1146.
21. Suda K, Nakauchi M, Inaba K, Ishida Y, Uyama I. Robotic surgery for upper gastrointestinal cancer: current status and future perspectives. Dig Endosc. 2016;28(7):701‐713.
39. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349(22):2117‐2127.
22. Bansal SS, Dogra T, Smith PW, et al. Cost analysis of open radical cystectomy versus robot‐assisted radical cystectomy. BJU Int. 2018;121(3):437‐444.
40. Li X, Wei L, Shang W, et al. Trace and evaluation systems for health services quality in rural and remote areas: a systematic review. J Public Health. 2018;26(2):127‐135.
23. Almamar A, Alkhamesi NA, Davies WT, Schlachta CM. Cost analysis of robot‐assisted choledochotomy and common bile duct exploration as an option for complex choledocholithiasis. Surg Endosc. 2018;32 (3):1223‐1227.
41. Mi J, Kang Y, Chen X, Wang B, Wang Z. Whether robot‐assisted laparoscopic fundoplication is better for gastroesophageal reflux disease in adults: a systematic review and meta‐analysis. Surg Endosc. 2010; 24(8):1803‐1814.
JIN
11 of 11
ET AL.
42. Maeso S, Reza M, Mayol JA, et al. Efficacy of the Da Vinci surgical system in abdominal surgery compared with that of laparoscopy: a systematic review and meta‐analysis. Ann Surg. 2010;252(2):254‐262. 43. Ajani JA, D' Amico TA, Almhanna K, et al. Esophageal and esophagogastric junction cancers, version 1.2015. J Natl Compr Canc Netw. 2015;13(2):194‐227. 44. Yao L, Sun R, Chen YL, et al. The quality of evidence in Chinese meta‐ analyses needs to be improved. J Clin Epidemiol. 2016;74:73‐79.
How to cite this article: Jin D, Yao L, Yu J, et al. Robotic‐ assisted minimally invasive esophagectomy versus the conventional minimally invasive one: A meta‐analysis and systematic review. Int J Med Robotics Comput Assist Surg. 2019;15: e1988. https://doi.org/10.1002/rcs.1988