Comparison between total laparoscopy and laparoscopy-assisted distal gastrectomy for gastric cancer

Methods

This meta-analysis was conducted from April to July 2013 in Sichuan Cancer Hospital, Chengdu, China. We searched PubMed, EMBASE, and the China Knowledge Resource Integrated Database (CNKI), and no restrictions were applied. The following key words were used: ‘gastric’ or ‘stomach,’ ‘cancer’ or ‘tumor’ or ‘carcinoma,’ ‘laparoscopy,’ or ‘laparoscopic,’ ‘laparoscopy-assisted’ or ‘laparoscopic-assisted,’ and ‘gastrectomy.’ The last search was conducted in May 2013. A total of 1,169 articles were selected (Figure 1). The collected studies for this meta-analysis met all of the following inclusion criteria: (1) contained both TLDG and LADG for gastric cancer; (2) study designs were retrospective case-control and prospective, non-random studies; and (3) sufficient data were provided. The main exclusion criteria included: (1) contained either TLDG or LADG alone for gastric cancer; (2) no sufficient data; (3) contained cases of total and proximal gastrectomy. According to the pre-specified inclusion and exclusion criteria, all data were extracted independently. Crosschecking was used to discover differences, and disagreements were resolved through discussion. The following data from each study were extracted: the name of the first author, year of publication, nationality of the corresponding author, study period, numbers of TLDG and LADG procedures, characteristics of patients, operation time, blood loss, postoperative hospital stay, lymph nodes, time to first flatus, and postoperative complications. When duplicate articles and data were discovered, the article that was published most recently, or contained more study subjects was included.

Figure 1

Flow chart of the selecting process of articles. LADG - laparoscopy-assisted distal gastrectomy, TLDG - total laparoscopy distal gastrectomy. CNKI - China Knowledge Resource Integrated Database.

The RevMan 5.2 statistical package (The Cochrane Collaboration, www.cochrane.org) was used to perform statistical analysis of the data. Heterogeneity was assessed using the chi-squared test with significance set at p<0.10, and measured using the I² index. When p>0.10, the fixed effects model was used. When p<0.10, the random effect model was used. For qualitative variables (surgical complications), the odds ratio (OR) was used to assess the outcomes of these studies. For quantitative variables (operation time, blood loss, hospitalization days, lymph nodes, time to first flatus) the mean difference (MD) with 95% confidence interval (CI) was used to assess the outcomes of these studies. P<0.05 was considered as statistically significant. The study was approved by the Institution Review Board.

Results

In the process of searching the literature, only one prospective, multicenter study was found. We included both retrospective and prospective studies that were of high quality. A total of 9 studies were included in our meta-analysis, comprising 2,046 patients; 846 of these patients underwent TLDG and 1,200 underwent LADG.6,9-16 The main characteristics of the 9 studies were not statistically different, this information is listed in Table 1.

Table 1

Characteristics of studies of total laparoscopy distal gastrectomy (TLDG) and laparoscopy-assisted distal gastrectomy (LADG) for gastric cancer.

Operation time

Operation time was measured in all 9 studies. The meta-analysis results had shown that there was no significant difference for TLDG and LADG for gastric cancer in operation time. The heterogeneity test for operation time was: I²=95%, p<0.00001, and the test for overall effect was: Z=0.28, p=0.78 (Table 2).

Table 2

Meta-analysis results of operation time, time to first flatus, postoperative hospital stay. Mean difference are shown with 95% confidence intervals (CIs).

Blood loss

There were 6 studies that provided data on blood loss (784 patients). Our results revealed that blood loss of TLDG was less than that of LADG. Meanwhile, a significant difference was displayed. The heterogeneity test for blood loss was: I²=61%, p=0.02, and the test for overall effect was: Z=2.07, p=0.04 (Figure 2).

Figure 2

Forest plot corresponding to blood loss, standardized mean differences (SMDs) are shown with 95% confidence interval (CI). LADG - laparoscopy-assisted distal gastrectomy, TLDG - total laparoscopy distal gastrectomy, IV - inverse variance, df - degrees of freedom

Number of harvested lymph nodes

All studies (2,046 patients) provided data on the number of harvested lymph nodes. The MD of harvested lymph nodes came from TLDG minus LADG. Moreover, the result displayed the harvested lymph nodes was greater of TLDG than of LADG. Forest plot revealed that there was a significant difference between TLDG and LADG in the dissected lymph nodes. The heterogeneity test for the number of harvested lymph nodes was: I²=75%, p=0.0001, and the test for overall effect was: Z=2.26, p=0.02 (Figure 3).

Figure 3

Forest plot corresponding to number of harvested lymph nodes, mean differences with 95% confidence interval (CI). LADG - laparoscopy-assisted distal gastrectomy, TLDG - total laparoscopy distal gastrectomy, IV - inverse variance, df - degrees of freedom

Postoperative complications

Not all studies provided intact data regarding postoperative complications. One study only analyzed the anastomosis-related complications, and one study did not provide any data on postoperative complications. We collected data on the overall postoperative complications from 7 studies including a total of 1,607 patients. Forest plot results revealed there was no significant difference between TLDG and LADG. Moreover, we also analyzed the main postoperative complications, such as anastomotic leakage, wound problems, and postoperative bleeding (Table 3). Interestingly, these were no differences between the 2 groups.

Table 3

Meta-analysis results of postoperative complications, and odds ratio (OR) with 95% confidence interval (CI).

Discussion

Laparoscopy gastrectomy (LG) was widely accepted for treating early gastric cancer in Japan and Korea.17-19 Many studies demonstrated the safety and feasibility of LG, and most surgeons performed LADG as the first step of LG.20-22 During the process of LADG, when the surrounding omentum is divided and the blood vessels are clipped, surgeons must pull the stomach out of the abdomen through a small abdominal wall incision. Then, the surgeons cut off the stomach carcinoma, and perform the anastomosis extracorporeally. This process has technical difficulties, especially in cases of obese patients.6,15 However, compared with open surgery, LADG was less invasive, and was associated with shorter hospital stays and faster recovery time,1 except for several trocar puncture wounds, LADG only requires a small abdominal incision to complete the surgery. Along with the accumulated surgeon experience and improved anastomotic equipment came the development of TLDG. Compared with LADG with extra-corporeal stomach dissection and gastroenterostomy, the specimen was extracted via a smaller incision, and anastomosis was performed intracorporeally during TLDG. The mini laparotomy was considered to be less invasive and more favorable. Existing research compared the short-term clinical outcomes between TLDG and LADG for gastric cancer, but inconsistent conclusions were reported. For example, some studies reported that TLDG can shorten the postoperative hospital stay, while other researchers do not agree. Therefore, we performed this meta-analysis to compare the clinical outcomes between TLDG and LADG to confirm whether TLDG should be recommended for distal gastrectomy.

In our studies, compared with LADG, TLDG had lower blood loss and more harvested lymph nodes. The smaller abdominal incision of TLDG may be the reason for lower blood loss. We cannot confirm why TLDG resulted in a greater number of harvested lymph nodes, but most studies observed this, and a statistically significant difference was displayed. The D2 lymph node dissection is the standard radical surgery for gastric cancer. Based on this radical surgery, the number of lymph nodes harvested should be the same in both TLDG and LADG. More high quality clinical trials are needed to confirm this inference. Although the mean difference in the operation time between TLDG and LADG was 2.15 in the current study, -there was no significant difference.

Many studies revealed that TLDG can shorten the postoperative recovery time and reduce postoperative complications. Our meta-analysis results revealed that no significant differences existed in the postoperative recovery time and complications between TLDG and LADG. In the studies from our search, only one reported the comparative outcome of total cost.16 Due to the requirement for more staplers, the cost of TLDG was higher than that of LADG. More studies are needed to compare surgery costs between LADG and TLDG. Due to the heterogeneity of the included studies, the meta-analysis results should be treated with caution.

Several limitations were discussed in our study. First, the sample size was low in some of the studies, and most of the studies included in this meta-analysis were retrospective studies, only one prospective, non-random study was included. Compared with randomized controlled trials, factors in the retrospective studies were not controlled, thereby decreasing the reliability of the results. Second, LADG and TLDG were performed from 2005-2012. Due to the experience of these surgeons and the fact that the learning curve revealed large differences, the clinical outcomes varied greatly. Therefore, it may be better to combine these studies using a random-effect model.

In summary, this meta-analysis provides evidence that TLDG significantly reduces intraoperative bleeding and harvests more lymph nodes. However, long-term follow-up outcome is not clear, especially regarding the recurrence rate and overall survival rate. At the same time, more prospective, random trials are needed to confirm the clinical outcomes of TLDG.