Comparison of laparoscopic versus open gastrectomy with D2 lymph node dissection for advanced gastric cancer: A meta-analysis

For many decades, D2 procedure has been accepted in the far-east as the standard treatment for both early (EGC) and advanced gastric cancer (AGC). In case of AGC, the debate on the extent of nodal dissection has been open for many years in order to highlight the safety and efficacy of treatment, hence this study. A comprehensive literature research was performed in PubMed to identify studies that compared laparoscopicassisted gastrectomy (LAG) and open gastrectomy (OG) with D2 lymph node dissection (D2-LND) for treatment of AGC for the last five years. Data of interest were checked and subjected to meta-analysis with RevMan 5.3 software. The pooled risk ratios (RR) and weighted mean difference (WMD) with 95% confidence intervals (CI) were calculated. Overall, 19 studies were included in this meta-analysis. LG had some advantages over OG, including shorter hospitalization (WMD -2.31; 95% CI -4.09 to -0.53; P = 0.01), less blood loss (WMD -120.49; 95% CI -174.27 to -66.71; P < 0.01), faster bowel recovery (WMD -0.55; 95% CI -0.86 to -0.24; P ˂ 0.01) and earlier ambulation (WMD -0.75; 95% CI -1.38 to -0.11; P = 0.02). In terms of surgical and oncological safety, LG could achieve similar lymph nodes (WMD, -0.94, 95% CI, -2.95 to 1.06; P=0.36), a lower complication rate [odds ratio (OR)=0.80; 95%CI, 0.68-0.97; P=0.02], and overall survival (OS) and disease-free survival (DFS) comparable to OG. In conclusion, for AGCs both techniques (LAG and OG) appeared comparable in shortand longterm results. More time was needed to perform LAG; nonetheless, it had some advantages in achieving faster postoperative recovery over OG. In order to clarify this controversial issue ongoing trials and future studies are needed.


Introduction
With the third highest rate of mortality, cancer of the stomach is one of the most malignant tumors in the world 1 . In fact, around three-quarters of all new cases and deaths occur in Eastern Asia, Eastern Europe and South America, and almost close to half (42 %) in China 2 . Despite considerable progress in the diagnosis and treatment of gastric cancer, it remains a major health problem and requires a multidisciplinary approach in which surgery is the cornerstone in the treatment of gastric cancer 3 , which includes conventional open gastrectomy (OG) and laparoscopy-assisted gastrectomy (LAG).
For a long time in history, conventional open gastrectomy (OG) surgery remains the treatment of choice for gastric cancer. A complete resection along with lymph node dissection was recognized as broad as the only one cure for gastric cancer 4 .
But since its first description in 1991 by Kitano 5 , laparoscopic gastrectomy (LG) has experienced rapid development and has been widely accepted and widely used to treat EGC nowadays, especially in East Asia( China, Japan and Korea) 6,7 .
With curative intent, the D2 lymph node dissection has been widely applied in traditional open surgery for locally advanced gastric cancer (AGC) 8 , however, the use of this procedure in LG for AGC is discussed. The oncological safety of LG in AGC treatment has not been well evaluated, and it remains to be confirmed whether laparoscopic surgery can still guarantee the advantage of minimal invasion, whether it increases perioperative complications and mortality, and whether it can achieve the same degree of radicalism as open surgery. To this end, we conducted a meta-analysis by comparing LAG with OG with D2 lymph node dissection for AGC with regard to their short-and long-term outcomes in order to determine the current status of LAG.

Search strategy:
A comprehensive PubMed search from January 2011 to February 2016 was performed using the following Mesh search headings and text words: "laparoscopy-assisted gastrectomy", "laparoscopicassisted gastrectomy", "open gastrectomy", "conventional gastrectomy", "gastric cancer", "gastric carcinoma", "advanced gastric cancer", "D2 dissection". Logical combinations of these and related terms (stomach, neoplasm) were used to maximize sensitivity and only studies published in English were included. Title and abstracts of each identified publication were screened, and only publications that reported the full texts for the clinical outcomes of this analysis were further retrieved. Reference lists of systematic reviews or meta-analysis were additionally checked to identify potential eligible studies.

Study eligibility:
Studies following below criteria were selected: (1) RCTs and nonrandomized comparative studies (2) Patients with advanced gastric carcinoma and/or with locally advanced were acceptable, without limitations for race, age or gender and where the patients in the LG and OG groups were compared; and the staging system was based on the individual reports (3) Studies comparing the short and long-term outcomes of LADG and OG with D2 LND.

Quality assessment:
For all the included studies 9 , to assess their quality we used the modified Newcastle-Ottawa Scale. The total score was 9 stars, and the quality of each study was graded as level 1 (0-5 stars) or level 2 (6-9 stars). Studies ≥ 5 stars were considered as highquality studies.

Data synthesis and statistical analysis:
The Proximal, distal, total and subtotal gastrectomies were used as surgical options. Gastrointestinal tract reconstruction modalities included Billroth-I/ II, Roux-en-Y anastomosis, eosophago-gastrostomy, and eosophago-jejunostomy. None of the studies had reported the use of neo-adjuvant chemoradiotherapy, but one study 28 reported the use of chemo-radiotherapy in postoperative period. The follow-up ranged from 30 days to 5 years. Detailed information on study characteristics is shown in Table 1 and quality assessment in Table 2, all study had a score of ≥ 6 stars.
The quality of each study was graded as low level (total score 0-5) or high level (total score 6-9).

a) Selection: (1) Assignment for treatment:
One star was assigned if details of criteria for assignment of patients to treatments provided.
(2) One star was assigned if the laparoscopic-assisted distal gastrectomy group was representative of patients for gastric cancer; no star was assigned if groups of patients were selected or selection of the group was not described. (3) One star was assigned if the open distal gastrectomy group was representative of patients for gastric cancer; no star was assigned if groups of patients were selected or selection of the group was not described.

b)
Comparability: Comparability variables were as follows: 1, age; 2, sex; 3, depth of tumor invasion on preoperative diagnosis; 4, extent of lymphadenectomy; 5, median or mean followup; 6, American Society of Anesthesiologists status; 7, tumor size; 8, postoperative pathologic stage; and 9, histological type. (4) Two stars were assigned if the groups were all comparable for the variables 1-5; 1 star was assigned if one of these five characteristics was not reported, even if there were no other differences between the groups, and other characteristics had been controlled for; and no star was assigned if the two groups differed. (5) Two stars were assigned if the groups were all comparable for the variables 6-9; 1 star was assigned if one of these four characteristics was not reported, even if there were no other differences between the groups, and other characteristics had been controlled for; and no star was assigned if the two groups differed.

c)
Outcome assessment: (6) one star was assigned if primary outcome parameters were clearly reported. (7) One star was assigned if more than 90% of patients were followed up.

1) Comparison of intraoperative effects
The duration of operation time in LAG group was 59 Results of analyses of intraoperative effects are shown in Table 3.
LG laparoscopic-assisted gastrectomy, OG open gastrectomy, LN lymph node, WMD weighted mean difference    The comparison outcomes of short-term results between LG and OG is summarized in Table 4.

2) Analyses of short-term results
LG laparoscopic-assisted gastrectomy, OG open gastrectomy, WMD weighted mean difference, OR odds ratio  Results of meta-analysis of long-term effects are presented in Table 5. In our study, we can observe that LAG demonstrated several advantages for AGC treatment. LAG group patients showed significantly less blood loss, faster recovery, earlier ambulation and shorter hospital stay compared to those treated by OG.

3) Analyses of long-term effects
These advantages are consistent with the findings of fast track surgery and benefits to AGC patient's recovery 35, 36 . Less blood loss found in LAG group is same as for scholars who found that it could reduce the risk of adverse effects such as acute lung injury, hypothermia 37 . Faster bowel function recovery and shorter postoperative hospital findings are identical with previous meta-analysis done by Ding, Wang, Qiu and Huang 38-41 .
We also found in this study that the operation time for LAG is longer, explainable by the skill and the familiarity of surgeons with the laparoscopic system influencing the length of operation time. Some studies reporting the learning curve of LAG In this study, we found that both, overall and specific complication rates in LAG groups were much lower. As we already know, LG has the inherent benefit of minimal invasiveness, reducing the risk of causing massive tissue and organ damage during an operation, therefore, would lead to fewer complications (15% in LAG group versus 17.6% in OG group). Similarly, compared to OG group, we found fewer medical complications in LAG group -medical complications such as respiratory and cardiovascular events, pulmonary embolism, deep venous thrombosis and non-surgical infections are potential life threatening events. Here, we found a rate of 4.67 % medical complication for LG patients, which is significantly lower compared to 8.9 % in OG group. This difference could also be attributed to the benefits of laparoscopic surgery as, not only the LG patients had lesser hospital stay and quicker recovery, but also lesser probability to acquire a nosocomial infection 47 ; also these sunsets of patients were capable of early mobilization/ambulation than OG patients, thereby reducing the risk of developing hypostasis and deep venous thrombosis 48 .
From the oncology security standpoint, LG is comparable to OG. In fact, at specific stages of the disease, an adequate LN dissection is very important for prognosis in the treatment of gastric cancers not only to reduce the possibility of recurrence and metastasis 49 , but also renders survival benefits 50 . As D2 Lymphadenectomy is already considered as a method of choice for treating AGC in East Asia 51 , it leads us to believe that the success of this practice is an essential part of the radical resection for the AGC treatment 52 . And as we found in our metaanalysis, the number of LNs retrieved in LG with D2 lymphadenectomy did not differ much from that observed in the OG suggesting that LG had lymphadenectomy efficiency comparable to that of OG, which matched with Ding's 38 Wang's 39 and Quan's 53 meta-analysis.
Another variable that also affects the oncological results is resection margin distance. It is well established that the main objective of radical resection is complete removal of tumor mass and that, in many cancers, a close correlation exists between positive resection margin and increased risk of local recurrence and consequently, decreased OS and DFS 54 . As suggested by recent studies, the status of surgical margin could be considered as an independent prognostic factor for GC patients 55,56 thereby ensuring complete resection of the tumor tissues by a sufficient distance between the edge of the tumor and the resection margin, and also reduced the risk of a positive resection margin 57 .
Thus it may in part reflect the possibility of healing of surgical procedures in the evaluation of the distance of the resection margin. Here, we found that there is no difference in the distance between the proximal and distal resection margin for the two techniques, suggesting that both, LAG and OG, possessed comparable curability and oncological safety. Deng et al. 58 suggested in their study that the size of the tumor should also be considered an important clinic pathology factor in order to improve the accuracy of the prediction of the prognosis of the GC patients. When compared to Quan's meta-analysis 53 , we observed that the tumor size in OG was much larger than in LG. There is significant heterogeneity among the articles in terms of tumor sizes.
For long-term results, in order to evaluate the direct effects of surgical interventions, OS, DFS and recurrence were used. Our study showed that different types of survival seemed almost the same for LG as for OG. The results for 1-, 3-, and 5-year OS and 5-year RFS and 5-year DFS were separately compared, showing that the effects of treatment with LAG were no comparable to OG. Unlike Chen et al. analysis 59 , recurrences observed in LG Group was less, which apparently corresponded to the studies performed by Huang 41 and Quan 53 and we theorize that this variation in finding is relative to the nature of the studies included for the analysis. In order to broaden the case pool and eliminate confounding bias at the same time, only recently published articles were included and rejected those where EGC and AGC were treated together.
And as for the interpretation of the conclusions of this study, we do not overlook certain limitations which are as follow. Firstly, only one RCT was included in our meta-analysis and most of the studies were carried out retrospectively. Secondly, ranging from a total of 8 to 1078 patients, the case volumes of the selected studies varied greatly. In such cases, the experience of surgeons would greatly influence the comparisons of surgical results, leading to a misinterpretation of the results of surgical procedures, and explains the high heterogeneity among the studies. This scenario necessitates future trials with prospective design and multi-center participation.

Conclusions
This study has allowed us to demonstrate that in both short-and long-term, LG with D2 LND could be as effective as OG to treat AGC patients. It has also been shown that LG had several advantages such as minimal invasion, faster recovery and shorter hospitalization. Currently, there are ongoing RCTs studying the value of LG over OG, but in the waiting of the publication of their findings, and based on our analysis, as well as the previous meta-analyses, we can conclude that LG may very well be applied for the treatment of AGC, especially if experienced surgeons perform it.