There has been a recent increase in the use of totally laparoscopic total gastrectomy (TLTG) for gastric cancer. However, there is no scientific evidence to determine which esophagojejunostomy (EJS) technique is the best. In addition, both short- and long-term oncological results of TLTG are inconsistent. We reviewed 25 articles about TLTG for gastric cancer in which at least 10 cases were included. We analyzed the short-term results, relationships between EJS techniques and complications, long-term oncological results, and comparative study results of TLTG. TLTG was performed in a total of 1170 patients. The mortality rate was 0.7%, and the short-term results were satisfactory. Regarding EJS techniques and complications, circular staplers (CSs) methods were significantly associated with leakage (4.7% vs. 1.1%, p < 0.001) and stenosis (8.3% vs. 1.8%, p < 0.001) of the EJS as compared with the linear stapler method. The long-term oncological prognosis was acceptable in patients with early gastric cancers and without metastases to lymph nodes. Although TLTG tended to increase surgical time compared with open total gastrectomy and laparoscopy-assisted total gastrectomy, it reduced intraoperative blood loss and was expected to shorten postoperative hospital stay. TLTG is found to be safer and more feasible than open total gastrectomy and laparoscopy-assisted total gastrectomy. At present, there is no evidence to encourage performing TLTG for patients with advanced gastric cancer from the viewpoint of long-term oncological prognosis. Although the current major EJS techniques are CS and linear stapler methods, in this review, CS methods are significantly associated with leakage and stenosis of the EJS.
circular stapler;esophagojejunostomy;gastric cancer;linear stapler;totally laparoscopic total gastrectomy
Laparoscopy-assisted distal gastrectomy (LADG) for gastric cancer was first performed by Kitano et al1 in 1994 and showed satisfying short-term results.2 The long-term results of LADG for early gastric cancer were also favorable,3 and LADG was reported to be surgically less invasive and lead to an earlier recovery than open surgery.4 ; 5
The number of reports on laparoscopy-assisted total gastrectomy (LATG) for gastric cancer has increased with the advancement of techniques for lymphadenectomy and reconstructive procedures for the upper stomach.6; 7; 8; 9 ; 10 However, esophagojejunostomy (EJS) via minilaparotomy in LATG is relatively difficult because of the limited angle of the direct view, depending on the patients somatotype and obesity index, and totally laparoscopic total gastrectomy (TLTG) has become more commonly used for intracorporeal anastomosis performed under pneumoperitoneum. As surgical techniques progressed from LATG to TLTG, new EJS techniques have been devised for TLTG.11; 12; 13; 14 ; 15 TLTG has been widely performed in Japan and Korea and has shown favorable short-term performance results.16; 17; 18 ; 19
EJS in TLTG is a very important surgical procedure because it is associated with the risk of anastomotic leakage, bleeding, and stenosis.20 ; 21 No scientific conclusion can be drawn at present regarding which procedure reduces the postoperative complications of EJS in TLTG because no clear evidence is available based on well-designed randomized controlled trials (RCTs). We retrospectively reviewed reports on various techniques for EJS in TLTG, compared various EJS techniques and complications, and investigated short- and long-term oncological results and comparative study results of TLTG.
This review included TLTG or totally laparoscopic degastrectomy for gastric or remnant gastric cancer, using Roux-en Y reconstruction. As with surgical procedures via minilaparotomy, the removal of resected specimens and performing jejunojejunostomy were permitted. EJS had to be performed intracorporeally under pneumoperitoneum to satisfy the determination of TLTG. Regarding EJS techniques, this review targeted the so-called single stapling technique (SST), double stapling technique (DST), and hemidouble stapling technique (HDST) using circular staplers (CSs), as well as the functional end-to-end anastomosis (FETEA) and overlap methods using linear staplers (LSs), and the hand-sewn (HS) method, and classified procedures into these six types. Reports that did not meet the above criteria or those using several or unknown EJS techniques were excluded from the analyses of EJS techniques.
We analyzed short-term results, relationships between EJS techniques and complications, as well as long-term oncological results and comparative study results of TLTG.
This review included only English articles identified by the term “totally laparoscopic” or a combination of “laparoscopic” and “total gastrectomy” in the PubMed online database. On July 15, 2013, a final search of PubMed was performed, and we selected and reviewed original articles describing the clinical results of TLTG performed in at least 10 cases. Reports using animal data or those including robotic surgeries were excluded.
Data were collected and analyzed using StatMate IV for Windows (ATMS Co., Ltd., Tokyo, Japan). The Student t test was used to compare the continuous variable, and the Chi-square test was used to compare the categorical variable. A p value < 0.05 was considered significant.
In 2005, Dulucq et al22 reported the first case series of TLTG for gastric cancer (n = 8), which was a prospective single-center study of TLTG and other surgical procedures. Since then, TLTG has become a more commonly used surgical procedure for gastric cancer, and many reports focusing on reconstruction techniques in EJS and short-term results have been published in recent years. Our initial literature search found 629 articles; however, only 25 of them satisfied the conditions described above. 11; 12; 13; 14; 15; 16; 17; 18; 19; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37 ; 38
Table 111; 12; 13; 14; 15; 16; 17; 18; 19; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35 ; 36 summarizes the reports cited in this review. Of the 25 articles cited in this review, only two were prospective studies and the remaining 23 were retrospective studies. Three articles described comparative studies of TLTG and open total gastrectomy (OTG),16; 17 ; 19 including one comparing TLTG versus LATG.18 Sixteen articles focused on surgical and short-term results only, and nine articles described long-term oncological results.13; 17; 23; 25; 27; 29; 35; 36 ; 37
Author | Publication year | Study design | Comparative study | Total number | EJS method |
---|---|---|---|---|---|
Huscher et al23 | 2007 | Retrospective | No | 11 | FETEA |
Topal et al16 | 2008 | Prospective | TLTG vs. OTG | 38 | SST/DST |
Usui et al11 | 2008 | Retrospective | TLTG vs. TLTG with J pouch | 23 | SST |
Ziqiang et al24 | 2008 | Retrospective | No | 14 | FETEA |
Okabe et al12 | 2009 | Retrospective | No | 16 | FETEA |
Jeong and Park13 | 2009 | Retrospective | No | 16 | DST |
Shinohara et al25 | 2009 | Retrospective | No | 55 | FETEA |
Kinoshita et al14 | 2010 | Retrospective | No | 10 | SST |
Bracale et al26 | 2010 | Retrospective | No | 67 | FETEA |
Inaba et al15 | 2010 | Retrospective | No | 53 | Overlap |
Marangoni et al27 | 2012 | Retrospective | No | 53 | DST |
Moisan et al17 | 2012 | Prospective | TLTG vs. OTG | 31 | HS |
Nunobe et al28 | 2011 | Retrospective | No | 41 | DST |
Tsujimoto et al29 | 2012 | Retrospective | TLTG vs. LPG | 15 | Overlap |
Jeong et al30 | 2012 | Retrospective | TLTG vs. LDG | 118 | SST/DST |
Lee et al31 | 2012 | Retrospective | No | 27 | FETEA |
Shim et al32 | 2013 | Retrospective | SST vs. DST vs. HDST vs. Overlap | 48 | SST/DST/HDST/Overlap |
Kim et al33 | 2012 | Retrospective | No | 124 | FETEA |
Yoshikawa et al34 | 2013 | Retrospective | No | 20 | SST |
Kim et al18 | 2013 | Retrospective | TLTG vs. LATG | 90 | FETEA |
Kim et al19 | 2013 | Retrospective | TLTG vs. OTG | 139 | FETEA |
Ebihara et al35 | 2013 | Retrospective | No | 65 | FETEA |
Lafemina et al36 | 2013 | Retrospective | No | 17 | DST |
Kim et al37 | 2013 | Retrospective | No | 36 | SST |
Nagai et al38 | 2013 | Retrospective | Early group vs. recent group | 94 | Overlap |
DST = double stapling technique; EJS = esophagojejunostomy; FETEA = functional end-to-end anastomosis; HDST = hemidouble stapling technique; HS = hand-sewn; LATG = laparoscopy assisted total gastrectomy; LDG = laparoscopic distal gastrectomy; LPG = laparoscopic proximal gastrectomy; OTG = open total gastrectomy; SST = single stapling technique; TLTG = totally laparoscopic total gastrectomy.
As for EJS techniques, SST, DST, HDST, FETEA, overlap, and HS methods were used in seven, seven, one, 10, four, and one articles, respectively.
Table 21; 2; 3; 4; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 17; 18; 19; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37 ; 38 summarizes the backgrounds of patients reported in the articles cited in this review. TLTG was performed in a total of 1170 patients, and of the 1094 with data on sex ratios, TLTG was more often performed on men (760 men vs. 334 women). Some studies reported mean values, and the others used median values for age and body mass index; when mean values were used, the mean age and mean body mass index were calculated to be 62.2 years and 23.2 kg/m2, respectively. A history of open abdominal surgery was described in seven articles, and 92 patients (92/559, 16.5%) had a history of open abdominal surgery. In 462 patients with data on clinical stage, the numbers of patients in each of the TNM classification (7th edition) by the Union for International Cancer Control were 1/317/65/54/25 in the order of clinical stages 0/I/II/III/IV, respectively.
Author | Publication year | Total number | Sex M/F | Age (y) Mean ± SD or median (range) | BMI (kg/m2) Mean ± SD or median (range) | Previous abdominal surgery | TNM classification UICC, 7th edition (0/I/II/III/IV) | Surgical indication |
---|---|---|---|---|---|---|---|---|
Huscher et al23 | 2007 | 11 | ND | ND | ND | ND | 0/4/1/2/4 | Advanced included |
Topal et al16 | 2008 | 38 | 23/15 | 68.0 (37–85) | 24.0 (17–30) | ND | 0/17/7/10/4 | Advanced included |
Usui et al11 | 2008 | 23 | 18/5 | 67.7 ± 11.5 | 23.7 ± 11.5 | ND | ND | Under T2 and N1 |
Ziqiang et al24 | 2008 | 14 | 9/5 | 57.4 ± ND | ND | ND | ND | Advanced included |
Okabe et al12 | 2009 | 16 | 11/5 | 70.0 (39–81) | 20.8 (16.7–27.6) | ND | ND | Any T and N0 |
Jeong and Park13 | 2009 | 16 | 10/6 | 59.0 ± ND | 23.0 ± ND | 0 | ND | Under T2 and N0 |
Shinohara et al25 | 2009 | 55 | 41/14 | 59.0 (29–80) | ND | ND | 0/17/12/16/10 | Advanced included |
Kinoshita et al14 | 2010 | 10 | 9/1 | 63.7 (45–80) | 22.4 (18.0–26.0) | ND | ND | Any T and N0 |
Bracale et al26 | 2010 | 67 | 45/22 | 66.9 ± ND | ND | ND | 0/35/15/12/5 | Advanced included |
Inaba et al15 | 2010 | 53 | 40/13 | 59.4 (30–82) | 22.0 (15.0–32.4) | ND | ND | Advanced included |
Marangoni et al27 | 2012 | 13 | ND | ND | ND | ND | 1/3/4/5/0 | Advanced included |
Moisan et al17 | 2012 | 22 | ND | 67.0 (29–83)a | 26.0 (19–30)a | 10a | ND | Advanced included |
Nunobe et al28 | 2011 | 41 | 31/10 | 65.8 ± 1.6 | 23.8 ± 0.6 | 9 | 0/41/0/0/0 | Under T1 and N0 |
Tsujimoto et al29 | 2012 | 15 | 10/5 | 65.8 ± 14.3 | 20.8 ± 3.8 | ND | 0/10/2/1/2 | Under T3 and N1 |
Jeong et al30 | 2012 | 118 | 77/41 | 63.7 ± 11.0 | 23.2 ± 3.6 | 2 | ND | Under T2 and N0 |
Lee et al31 | 2012 | 27 | 16/11 | 59.1 ± ND | 24.6 ± ND | ND | 0/23/4/0/0 | ND |
Shim et al32 | 2013 | 48 | 33/15 | 56.7 ± ND | 24.3 ± ND | ND | ND | ND |
Kim et al33 | 2012 | 124 | 77/47 | 57.4 ± ND | 23.6 ± ND | 27 | ND | Advanced included |
Yoshikawa et al34 | 2013 | 20 | ND | ND | ND | ND | ND | ND |
Kim et al18 | 2013 | 90 | 61/29 | 58.0 ± 10.8 | 23.2 ± 2.9 | 21 | ND | Advanced included |
Kim et al19 | 2013 | 139 | 86/53 | 58.0 (30–84) | 23.6 (13.6–32.4) | 23 | ND | Advanced included |
Ebihara et al35 | 2013 | 65 | 45/20 | 65.9 ± 10.2 | 23.5 ± 4.0 | ND | 0/65/0/0/0 | Stage I |
Lafemina et al36 | 2013 | 17 | 40/8 | 64.0 (55–70)b | 27.1 (24.0–30.5)b | ND | ND | Advanced included |
Nagai et al38 | 2013 | 94 | 64/30 | 66.0 ± ND | 21.6 ± ND | ND | 0/71/19/4/0 | Advanced included |
Kim et al37 | 2013 | 36 | 24/12 | 60.9 ± 11.4 | 23.4 ± 3.4 | ND | 0/31/1/4/0 | Under T1 and N0 |
Total or mean | 1170 | 760/334 | 62.2 c | 23.2 d | 92 | 1/317/65/54/25 |
BMI = body mass index; ND = not described in the article; OTG = open total gastrectomy; SD = standard deviation; TLTG = totally laparoscopic total gastrectomy; UICC = Union for International Cancer Control.
a. Nine laparoscopic subtotal gastrectomy cases were included in these data.17
b. Thirty-one OTG cases were included in these data.36
c. Mean age of the TLTG cases was calculated using every data of the articles, except for median value.
d. Mean BMI of the TLTG cases was also calculated using every data of the articles, except for median value.
Table 2 also presents data containing clear surgical indication of TLTG. Thirteen reports had no limitations on the indication of TLTG for advanced gastric cancer, and the others limited indication to cases of T3 or less and N1 or less gastric cancer.
Table 311; 12; 13; 14; 15; 16; 17; 18; 19; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37 ; 38 provides the surgical results of TLTG. Some studies used median values, and others used mean values to report the surgical results. The mean surgical time and mean blood loss calculated from reported mean values were 254.2 minutes and 114.0 mL, respectively. For lymphadenectomy, the numbers of patients classified as D0/D1/D1+/D2 in the 14th edition of the Japanese Classification of Gastric Carcinoma by the Japan Gastric Cancer Association were 0/13/238/270, respectively, and some of the patients with advanced gastric cancer concomitantly underwent pancreatosplenectomy or splenectomy. Twenty-three patients were converted to OTG, mostly owing to intraoperative accidental symptoms as well as uncontrollable bleeding and difficulties in EJS techniques. The mean number of dissected lymph nodes calculated from the reported mean values was 39.5.
Author | Publication year | Surgical time (min) Mean ± SD or median (range) | Blood loss (mL) Mean ± SD or median (range) | Lymphadenectomy (D0/D1/D1+/D2) | Conversion to OTG | Harvested lymph nodes Mean ± SD or median (range) |
---|---|---|---|---|---|---|
Huscher et al23 | 2007 | 304.0 ± 83.0 | ND | 0/2/0/9 | 3 a | 35.0 ± 18.0 a |
Topal et al16 | 2008 | 187.0 (120–360) | 10.0 (5–400) | 0/0/0/38 | 0 | 17.0 (0–90) |
Usui et al11 | 2008 | 305.9 ± 57.6 | 77.5 ± 71.7 | 0/0/22/1 | 0 | ND |
Ziqiang et al24 | 2008 | 255.1 ± ND | 107.5 ± ND | 0/0/0/14 | 0 | ND |
Okabe et al12 | 2009 | 325.0 ± 68.0 | 195.0 ± 197.0 | ND | 0 | 47.0 ± 13.0 |
Jeong and Park13 | 2009 | 194.0 ± ND | 170.0 ± ND | 0/0/16/0 | 0 | 33.0 ± ND |
Shinohara et al25 | 2009 | 406.0 (200–865) | 102.0 (20–694) | 0/0/0/55 | 0 | 46 (17–106) |
Kinoshita et al14 | 2010 | 257.0 ± ND | 69.0 ± ND | ND | 0 | 43.3 ± ND |
Bracale et al26 | 2010 | 249.0 ± ND | ND | 0/5/0/62 | 7 | ND |
Inaba et al15 | 2010 | 373.4 ± 105.0 | 146.5 ± 325.3 | ND | 0 | ND |
Marangoni et al27 | 2012 | 260.0 ± ND | 125.0 ± ND | ND | 1 | 26.0 ± ND |
Moisan et al17 | 2012 | 250.0 (160–240) b | 100.0 (50–500) b | 0/3/5/23 b | 1 b | 35 (9–68) b |
Nunobe et al28 | 2011 | 298.6 ± 10.1 | 85.9 ± 15.2 | 0/0/0/41 | 0 | 42.8 ± 2.3 |
Tsujimoto et al29 | 2012 | 236.4 ± 43.4 | 51.2 ± 58.0 | 0/3/7/5 | 0 | 38.6 ± 15.4 |
Jeong et al30 | 2012 | 292.0 ± 88.0 | 256.0 ± 207.0 | 0/0/99/19 | 1 | 41.0 ± 16.0 |
Lee et al31 | 2012 | 126.2 ± 21.3 | ND | 0/0/27/0 | 0 | 33.7 ± 16.2 |
Shim et al32 | 2013 | A:229.1 ± 45.7 c | ND | ND | 0 | ND |
B:226.5 ± 51.4 c | ||||||
C:209.0 ± 39.4 c | ||||||
D:205.5 ± 33.1 c | ||||||
Kim et al33 | 2012 | F:189.0 ± 46.3 d | ND | ND | 0 | 42.6 ± 15.5 d |
S:148.3 ± 51.9 d | 37.4 ± 15.7 d | |||||
Yoshikawa et al34 | 2013 | 297.1 ± ND | ND | ND | 0 | ND |
Kim et al18 | 2013 | 166.4 ± 47.5 | ND | ND | 0 | 43.1 ± 17.2 |
Kim et al19 | 2013 | 144.0 (72–345) | ND | ND | 0 | 37 (2–94) |
Ebihara et al35 | 2013 | 271.5 ± 64.7 | 85.2 ± 143.2 | 0/0/62/3 | 1 | 30.2 ± 12.4 |
Lafemina et al36 | 2013 | 230 (190–277) | 250 (150–450) | ND | 9 | 22 (17–28) |
Nagai et al38 | 2013 | E:341.4 ± 75.7e | 70.2 ± 77.3e | ND | 0 | 53.4 ± 21.0e |
R:368.0 ± 94.6e | 80.4 ± 115.0e | 47.0 ± 18.8e | ||||
Kim et al37 | 2013 | 227.1 ± 57.6 | 77.1 ± 71.7 | ND | 0 | 37.9 ± 10.9 |
Total or mean | 254.2 f | 114.0 g | 0/13/238/270 | 23 | 39.5 h |
DST = double stapling technique; HDST = hemidouble stapling technique; ND = not described in the article; OTG = open total gastrectomy; SD = standard deviation; SST = single stapling technique; TLTG = totally laparoscopic gastrectomy.
a. Eighty-nine laparoscopic subtotal gastrectomy cases were included in these data.23
b. Nine laparoscopic subtotal gastrectomy cases were included in these data.17
c. Shim et al32 reported four types of EJS techniques: types A, B, C, and D denote SST, DST, HDST, and overlap methods, respectively.
d. Kim et al33 reported the comparison of TLTG cases between first 70 cases (F) and subsequent 54 cases (S).
e. Nagai et al38 reported the comparison of TLTG cases between early period (E) and recent period (R).
f. Mean surgical time was calculated using every data of the articles, except for median value.
g. Mean blood loss was calculated using every data of the articles, except for median value.
h. Average of harvested lymph nodes was also calculated using every data of the articles, except for median value.
Table 411; 12; 13; 14; 15; 16; 17; 18; 19; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37 ; 38 provides the postoperative results of TLTG. Postoperative complications included leakage of the EJS (n = 33), leakage of the duodenal stump (n = 19), anastomotic bleeding (n = 33), postoperative pancreatic fistula (n = 16), and stenosis of the EJS (n = 38). The mortality rate was only 0.7% (8/1170). Based on articles with mean values, the time to the first flatus and time to restart oral intake were 3.3 days and 5.0 days, respectively, and the mean postoperative hospital stay was 12.0 days.
Author | Publication year | EJS leakage | Stump leakage | Bleeding | Pancreatic fistula | EJS stenosis | Mortality | Time to flatus (d) Mean ± SD or median (range) | Time to intake (d) Mean ± SD or median (range) | Hospital stay (d) Mean ± SD or median (range) |
---|---|---|---|---|---|---|---|---|---|---|
Huscher et al23 | 2007 | 0 | 2 | 2 | 0 | 0 | 2 | 3.4 ± 1.0 a | 5.9 ± 4.5a | 11.4 ± 4.5a |
Topal et al16 | 2008 | 2 | 0 | 0 | 0 | 0 | 1 | ND | ND | 11.0 (6–73) |
Usui et al11 | 2008 | 0 | 0 | 0 | 0 | 0 | 0 | ND | ND | 11.2 ± 5.3 |
Ziqiang et al24 | 2008 | 0 | 0 | 0 | 1 | 0 | 0 | 3.9 ± ND | 4.9 ± 1.0 | ND |
Okabe et al12 | 2009 | 0 | 0 | 0 | 0 | 1 | 0 | ND | 3.0 (ND) | 11.0 (ND) |
Jeong and Park13 | 2009 | 0 | 0 | 0 | 0 | 0 | 0 | ND | ND (3–5) | 11.0 (8–14) |
Shinohara et al25 | 2009 | 2 | 0 | 0 | 7 | 0 | 0 | ND | 3.0 (3–6) | 14.0 (9–25) |
Kinoshita et al14 | 2010 | 0 | 0 | 0 | 0 | 0 | 0 | ND | 4.0 (2–10) | 13.0 (8–24) |
Bracale et al26 | 2010 | 4 | 3 | 5 | 0 | 2 | 1 | 4.7 ± ND | ND | 12.4 (8–45) |
Inaba et al15 | 2010 | 2 | 1 | 0 | 3 | 1 | 0 | ND | ND | 14.4 ± ND |
Marangoni et al27 | 2012 | 0 | 0 | 0 | 0 | 0 | 1 | ND | ND | 11.0 (ND) |
Moisan et al17 | 2012 | 2 | 2 | 1 | 0 | 0 | 0 | ND | 4.0 (2–13) | 7.0 (4–59) |
Nunobe et al28 | 2011 | 2 | 0 | 0 | 3 | 3 | 0 | ND | 2.7 ± 0.5 | 16.9 ± 1.5 |
Tsujimoto et al29 | 2012 | 0 | 0 | 0 | 0 | 0 | 0 | 3.0 ± 1.3 | 4.1 ± 2.2 | 13.5 ± 9.1 |
Jeong et al30 | 2012 | 9 | 1 | 11 | 0 | 4 | 2 | 2.9 ± 0.8 | 3.5 ± 4.5 | 11.9 ± 11.9 |
Lee et al31 | 2012 | 0 | 0 | 3 | 0 | 0 | 0 | ND | ND | 8.1 ± ND |
Shim et al32 | 2013 | A:2 b | 1 b | 0 b | 0 b | 5 b | 0 b | ND | ND | 10.3 ± ND b |
B:2 b | 0 b | 0 b | 0 b | 4 b | 0 b | 8.4 ± ND b | ||||
C:1 b | 1 b | 1 b | 0 b | 1 b | 0 b | 9.3 ± ND b | ||||
D:0 b | 0 b | 0 b | 0 b | 0 b | 0 b | 8.8 ± ND b | ||||
Kim et al33 | 2012 | F:2 c | 1 c | 1 c | 0 c | 5 c | 0 c | 3.2 ± 0.9 c | 8.6 ± 9.6 c | 12.7 ± 11.5 c |
S:0 c | 0 c | 3 c | 0 c | 1 c | 0 c | 3.3 ± 0.9 c | 5.6 ± 4.6 c | 8.8 ± 5.8 c | ||
Yoshikawa et al34 | 2013 | 0 | 0 | 0 | 0 | 0 | 0 | ND | ND | ND |
Kim et al18 | 2013 | 0 | 1 | 2 | 0 | 4 | 0 | 3.4 ± 1.0 | 4.5 ± 1.8 | 7.9 ± 4.3 |
Kim et al19 | 2013 | 0 | 1 | 3 | 0 | 2 | 0 | 3.0 (2–6) | 3.0 (3–46) | 7.0 (5–72) |
Ebihara et al35 | 2013 | 0 | 1 | 0 | 1 | 3 | 1 | 1.9 ± 0.7 | 4.6 ± 1.8 | 21.4 ± 13.5 |
Lafemina et al36 | 2013 | 1 | 2 | 1 | 0 | 1 | 0 | ND | ND | 8.0 (6–9) |
Nagai et al38 | 2013 | E:2 d | 0 d | 0 d | 0 d | 0 d | 0 d | ND | 5.5 ± 2.5 d | 16.7 ± 9.5 d |
R:0 d | 1 d | 0 d | 1 d | 0 d | 0 d | 4.9 ± 4.7 d | 14.2 ± 12.1 d | |||
Kim et al37 | 2013 | 0 | 1 | 0 | 0 | 0 | 0 | 3.1 ± 0.7 | ND | 9.2 ± 8.7 |
Total or Mean | 33 | 19 | 33 | 16 | 38 | 8 | 3.3 e | 5.0 f | 12.0 g |
DST = double stapling technique; EJS = esophagojejunostomy; HDST = hemidouble stapling technique; ND = not described in the article; SD = standard deviation; SST = single stapling technique; TLTG = totally laparoscopic gastrectomy.
a. Eighty-nine laparoscopic subtotal gastrectomy cases were included in these data.23
b. Shim et al32 reported four types of EJS techniques: types A, B, C, and D denote SST, DST, HDST, and overlap methods, respectively.
c. Kim et al33 reported the comparison of TLTG cases between first 70 cases (F) and subsequent 54 cases (S).
d. Nagai et al38 reported the comparison of TLTG cases between early period (E) and recent period (R).
e. Mean time to first flatus was calculated using every data of the articles, except for median value.
f. Mean time to oral intake was calculated using every data of the articles, except for median value.
g. Average of postoperative hospital stay was also calculated using every data of the articles, except for median value.
Table 5 summarizes the EJS techniques used in the cited articles. For methods of anvil insertion and purse-string suture placement, only representative methods are listed in Table 5. With EJS techniques using LSs, disruption of the esophageal hiatus and slippage of the EJS site into the lower mediastinum are possible. The HS method is highly feasible to use for intracorporeal anastomosis, providing a sufficient view under pneumoperitoneum, and has the advantage of not requiring an entry hole closure. EJS techniques used in 101 patients, 139 patients, 14 patients, 553 patients, 176 patients, 31 patients, and 156 patients were SST, DST, HDST, FETEA, overlap, HS, and unknown methods, respectively.
SST | DST | HDST | FETEA | Overlap | HS | p | |
---|---|---|---|---|---|---|---|
Anvil insertion | Endo PSI EndoStitch Hand-sewn, etc. | Orvil EST, etc. | Orvil, etc. | – | – | – | |
Double stapling | – | + | + | – | – | – | |
Hiatus destroy | – | – | – | + | + | – | |
Anastomosis in mediastinum | – | – | – | + | + | – | |
Necessity of entry hole closure | + | + | + | + | + | – | |
Diameter of anastomosis (mm) | < 30 | < 30 mm | < 30 | > 30 | > 30 | 20–30 | |
Number (n) a | 101 | 139 | 14 | 553 | 176 | 31 | |
Duration of anastomosis Mean ± SD or median (range) | 6.0 ± ND d 43.2 ± 11.5 f P: 8.9 ± 5.1 g A: 6.4 ± 3.6 g | 54.0 ± ND c 42.8 ± 11.3 f | 37.0 ± 7.1 f | 42.5 ± ND b 44.0 ± ND e | 34.3 ± 6.4 f | 0.041f | |
Complications (n, %) | |||||||
EJS leakage | 6, 5.9 | 5, 3.6 | 1, 7.1 | 6, 1.1 | 2, 2.3 | 2, 6.5 | 0.009 |
Bleeding | 4, 4.0 | 1, 0.7 | 1, 7.1 | 15, 2.7 | 0, 0 | 1, 3.2 | 0.110 |
EJS stenosis | 12, 11.9 | 8, 5.8 | 1, 7.1 | 12, 2.2 | 1, 0.6 | 0, 0 | < 0.001 |
Stump leakage | 3, 3.0 | 2, 1.4 | 1, 7.1 | 8, 1.4 | 2, 1.1 | 2, 6.5 | 0.198 |
Pancreatic fistula | 0, 0 | 3, 2.2 | 0, 0 | 9, 1.6 | 4, 2.3 | 0, 0 | 0.680 |
Mortality (n, %) | 0, 0 | 1, 0.7 | 0, 0 | 4, 0.7 | 0, 0 | 0, 0 | 0.803 |
DST = double stapling technique; EJS = esophagojejunostomy; EST = efficient purse-string stapling technique; FETEA = functional end-to-end anastomosis; HDST = hemidouble stapling technique; HS = hand-sewn; ND = not described in the article; SD = standard deviation; SST = single stapling technique.
a. A total of 156 cases were excluded from these series, because descriptions of EJS techniques and breakdown of methods were unclear.
b. Ziqiang et al24 reported these data; the range was 32–66 minutes.
c. Jeong and Park13 reported these data; the range was 38–75 minutes.
d. Kinoshita et al14 reported these data, but this was the mean time for purse-string by hand-sewn suturing. Its range was 5–7 minutes.
e. Bracale et al26 reported these data; the range was 38–54 minutes.
f. Shim et al32 reported four types of EJS techniques: types A, B, C, and D denote SST, DST, HDST, and overlap methods, respectively.
g. Kim et al37 reported time for purse-string suture (P) and time for anvil placement (A).
For the time required to perform EJS, Kinoshita et al14 reported 6 minutes on average from purse-string suture placement using the HS method to the insertion of the anvil using the SST method. Kim et al37 also reported that the mean times for purse-string suture and for anvil placement in the SST method were 8.9 minutes and 6.4 minutes, respectively. Jeong and Park13 reported that EJS using the DST method took 54 minutes on average. Ziqiang et al24 reported that EJS using the FETEA method took 42.5 minutes on average, and Bracale et al26 also reported that it took 44.0 minutes on average, accounting for 17.7% of the entire surgical time. Shim et al32 compared four EJS techniques (SST, DST, HDST, and overlap methods) and showed that EJS took 43.2 ± 11.5 minutes, 42.8 ± 11.3 minutes, 37.0 ± 7.1 minutes, and 34.3 ± 6.4 minutes, respectively, with a significantly shorter time required for the overlap method (p = 0.041).
Regarding the major postoperative complications associated with each method, the incidence rates of leakage of the EJS were 5.9% (6/101), 3.6% (5/139), 7.1% (1/14), 1.1% (6/553), 2.3% (2/176), and 6.5% (2/31) using the SST, DST, HDST, FETEA, overlap, and HS methods, respectively, showing high rates with HDST, HS, and SST methods (p = 0.009). The incidence rates of anastomotic bleeding were also 4.0% (4/101), 0.7% (1/139), 7.1% (1/14), 2.7% (15/553), 0%, and 3.2% (1/31) using the SST, DST, HDST, FETEA, overlap, and HS methods, respectively, with no significant difference between the six methods (p = 0.110). The incidence rates of stenosis of the EJS were 11.9% (12/101), 5.8% (8/139), 7.1% (1/14), 2.2% (12/553), 0.6% (1/176), and 0%, using SST, DST, HDST, FETEA, overlap, and HS methods, respectively, showing the highest rate with the SST method (p < 0.001). No significant difference was detected for the incidences of leakage of the duodenal stump, postoperative pancreatic fistula, and mortality.
In the analysis using the stapling device (Table 6), the incidence of leakage of the EJS was significantly higher in the CS methods than in the LS methods (4.7% vs. 1.1%, p < 0.001). The incidence of stenosis of the EJS was also significantly higher in the CS methods than in the LS methods (8.3% vs. 1.8%, p < 0.001). No significant difference was detected for the incidence of anastomotic bleeding.
CS methods SST DST HDST | LS methods FETEA Overlap | p | |
---|---|---|---|
Number (n) | 254 | 729 | |
Complication (n, %) | |||
EJS leakage | 12, 4.7 | 8, 1.1 | < 0.001 |
Bleeding | 6, 2.4 | 15, 2.1 | 0.777 |
Stenosis of EJS | 21, 8.3 | 13, 1.8 | < 0.001 |
CS = circular stapler; DST = double stapling technique; EJS = esophagojejunostomy; FETEA = functional end-to-end anastomosis; HDST = hemidouble stapling technique; LS = linear stapler; SST = single stapling technique.
Table 713; 17; 23; 25; 27; 29; 35; 36 ; 37 summarizes nine articles that followed the long-term oncological prognosis of patients who underwent TLTG. The use of various prognostic indexes in these articles made it difficult for us to appropriately evaluate the results. The observation periods varied widely (1–160 months) among reports, and several reports included many cases of advanced gastric cancer. For case series including advanced gastric cancer patients, the incidences of recurrence were 11 patients (20.0%) in the report by Shinohara et al,25 11 patients (22.0%) in the report by Marangoni et al,27 five patients (16.1%) in the report by Moisan et al,17 and 14 patients (29.2%) in the report by Lafemina et al.36 In the reports of Tsujimoto et al29 and Ebihara et al,35 in which strict conditions were established for the indication of TLTG, no recurrence was reported during the observation period.
Author | Publication year | Surgical indication | Follow-up period Mean ± SD or median (range) | Recurrence, n (%) | Survival rates (%) |
---|---|---|---|---|---|
Huscher et al23 | 2007 | Advanced included | 57.6 ± 44.5a | 31,a 31.0a | OS = 59%,b DFS = 57%b |
Jeong and Park13 | 2009 | Under T2 and N0 | 6.4 ± ND | ND | ND |
Shinohara et al25 | 2009 | Advanced included | 16.0 (7–130) | 11, 20.0 | ND |
Marangoni et al27 | 2012 | Advanced included | 10.0 (3–26) | 11, 22.0 | ND |
Moisan et al17 | 2012 | Advanced included | 28.0 (ND) | 5, 16.1 | OS = 82.3,c DFS = 79.4c |
Tsujimoto et al29 | 2012 | Under T3 and N1 | 18.9 ± ND | 0, 0 | ND |
Ebihara et al35 | 2013 | Stage I | 37.0 (11–68) | 0, 0 | ND |
Lafemina et al36 | 2013 | Advanced included | 18.0 (11–34) | 14, 29.2 | ND |
Kim et al37 | 2013 | Under T1 and N1 | 13.2 (ND) | ND | ND |
DFS = disease-free survival; ND = not described in the article; OS = overall survival; SD = standard deviation.
a. Eighty-nine laparoscopic subtotal gastrectomy cases were included in these data.23
b. These were the 5-year OS and DFS rates.23
c. These were the 3-year OS and DFS rates.17
Regarding survival time, Moisan et al17 reported 3-year disease-free survival (DFS) and 3-year overall survival (OS) rates of 79.4% and 82.3%, respectively. When the analysis was limited to patients with early gastric cancer, the 3-year DFS and 3-year OS rates were pegged at 81.7% and 93.3%, respectively. Huscher et al23 reported 5-year DFS and 5-year OS rates of 57.0% and 59.0%, respectively.
Table 816; 17; 18 ; 19 provides the overview and conclusions of three TLTG versus OTG comparative studies and one TLTG versus LATG comparative study. The surgical time of TLTG was significantly longer than that of OTG in two articles (Topal et al16: 187.0 minutes vs. 150.0 minutes, p = 0.0003; Moisan et al 17: 250.0 minutes vs. 210.0 minutes, p = 0.007, mean). However, TLTG was associated with significantly less blood loss in two articles (Topal et al 16: 10.0 mL vs. 175.0 mL, p = 0.0001; Moisan et al 17: 100.0 mL vs. 300.0 mL, p = 0.001, median) and a significantly shorter postoperative hospital stay in two articles (Moisan et al 17: 7.0 days vs. 10.5 days, p = 0.001; Kim et al 19: 7.0 days vs. 8.0 days, p < 0.001, median), with no difference in postoperative complications compared to OTG. All articles concluded that TLTG was safe and feasible compared with OTG and LATG. 16; 17; 18 ; 19
Author | Publication year | Comparison | Surgical time (min) Mean ± SD or median (range) | Blood loss (ml) Mean ± SD or median (range) | Postoperative complicationsn (%) | Hospital stay Mean ± SD or median (range) | Conclusion for TLTG |
---|---|---|---|---|---|---|---|
Topal et al16 | 2008 | Prospective | 187.0 (120–360) | 10.0 (5–400) | 15, 39.4 | 11.0 (6–73) | Safe and feasible |
TLTG vs. OTG | 150.0 (120–360) | 175.0 (50–1400) | 9, 40.9 | ND | |||
(38 vs. 22) | p = 0.0003 | p < 0.0001 | p = 0.913 | p = 0.847 | |||
Moisan et al17 | 2012 | Prospective | 250.0 (160–420) | 100.0 (50–500) | 7, 22.5 | 7.0 (4–59) | Safe and feasible |
TLTG vs. OTG | 210.0 (135–390) | 300.0 (200–1400) | 4, 12.9 | 10.5 (6–37) | |||
(31 vs. 31) | p = 0.007 | p < 0.001 | p = 0.506 | p = 0.001 | |||
Kim et al19 | 2009 | Retrospective | 144.0 (72–345) | ND | 14, 10.0 | 7.0 (5–72) | Safe and feasible |
TLTG vs. OTG | 137.0 (65–355) | ND | 45, 21.7 | 8.0 (2–34) | |||
(139 vs. 207) | p = 0.381 | ND | p = 0.005 | p < 0.001 | |||
Kim et al18 | 2011 | Retrospective | 166.4 ± 47.5 | ND | 10, 11.1 | 7.9 ± 4.3 | Safe and feasible |
TLTG vs. LATG | 158.5 ± 45.5 | ND | 4, 16.0 | 9.5 ± 7.5 | |||
(90 vs. 23) | p = 0.461 | ND | p = 0.500 | p = 0.198 |
LATG = laparoscopy-assisted total gastrectomy; ND = not described in the article; OTG = open total gastrectomy; SD = standard deviation; TLTG = totally laparoscopic total gastrectomy.
At present, low invasive treatments for gastric cancer include endoscopic procedures (such as endoscopic mucosal resection and endoscopic submucosal dissection) and laparoscopic gastrectomy (LG).39 As is the trend for laparoscopic surgery for various organs,40 ; 41 laparoscopic-assisted surgery has been switched to totally laparoscopic surgery in LG to allow greater magnification, a wider view, and practical use of more refined surgical techniques. As the noninferiority of LATG to OTG has been demonstrated,42; 43 ; 44 many reports about TLTG, as cited in this review, have been published. The critical point in TLTG is intracorporeal EJS techniques, if special techniques such as combined resection of adjacent organs45 and extended lymphadenectomy25 ; 46 are not taken into consideration.47 In this review, EJS techniques using CSs were significantly more associated with leakage and stenosis of the EJS, as compared with those using LSs. When SST and DST methods were used for esophageal cancer and colorectal cancer surgery, anastomotic stenosis was considered a relatively frequent complication and occurred in 2.4–10.0% of patients.48; 49 ; 50 As EJS is performed using longitudinally long devices in LS methods, a wider diameter of anastomosis can be secured, although there are concerns about the possible onset of gastroesophageal reflux disease owing to the disruption of the esophageal hiatus, slippage of the EJS site into the mediastinum, and severe consequences of leakage of the EJS.14; 34 ; 51 By contrast, the HS method is a simple and low-cost surgical technique, although currently not mainstream.17 ; 52 According to the time for anastomosis, it may not be a good comparative item in this study, as the devices used and the surgeons' experience and surgical technique cannot be at the same level. The current major EJS techniques of TLTG are obviously CS and LS methods, and which one is superior to the other remains to be determined. To correctly answer the question, analysis of clinical results from well-planned RCTs of EJS techniques in TLTG between CS and LS methods is warranted.
Based on the short-term results of comparative studies of TLTG versus OTG or LATG, it has been concluded that TLTG is, at present, safe and feasible.16; 17; 18 ; 19 The short-term results of TLTG from the 25 articles cited in the present review are substantially favorable and almost equivalent to the short-term results of OTG.30; 31; 32; 33; 34; 35; 36; 37 ; 38 Topal et al16 have suggested that the concomitance of splenectomy (p = 0.006) and the number of dissected lymph nodes (p = 0.042) are surgical factors related to the onset of postoperative complications of TLTG. Surgeons with only recent experience in performing TLTG more often encounter complications (p = 0.032), 16 and so even surgeons who are well experienced in LG should not readily switch to TLTG for advanced gastric cancer.
Regarding the long-term oncological prognosis investigated in this review, TLTG produced extremely satisfying results in patients with early gastric cancer and without metastases to lymph nodes.29 ; 35 However, the global effectiveness of LADG has been demonstrated only for early gastric cancer,3 ; 5 and therefore close attention should be paid to the long-term oncological results of LATG and TLTG in Japan. Most of the articles on TLTG were relatively new, and only a few articles included long-term results.17 ; 23 In addition, the disease stages of cases included in the survey varied widely, and variable prognostic factors specific to advanced gastric cancer, such as peritoneal dissemination, were not taken into consideration at all, and thus the data used in this review were insufficient for accurate analysis of the long-term oncological prognosis.36 In the future, long-term oncological effectiveness or noninferiority of TLTG for early gastric cancer should be demonstrated based on RCTs using the same method as that used for LADG.
We investigated various EJS techniques in TLTG, as well as short-term results, long-term oncological results, and comparative study results of TLTG. At present, TLTG is believed to be a safe and feasible surgical technique for surgeons with a steep learning curve, comparable to OTG and LATG.52; 53 ; 54 However, because of the difficulty of TLTG when applied to advanced gastric cancer requiring combined resection of adjacent organs and extended lymphadenectomy and its association with an increased risk of serious perioperative complications,25 the risks and benefits should be weighed prior to performing TLTG. The potential effects of TLTG on the long-term oncological prognosis have not been determined and warrant further investigation.55 For EJS techniques, leakage and stenosis of the EJS may occur more frequently in CS methods than in LS methods. Although CS and LS methods for EJS are the major techniques used in TLTG at present, further research is necessary to establish which reconstruction techniques are suitable for EJS.
Published on 26/05/17
Submitted on 26/05/17
Licence: Other
Are you one of the authors of this document?