Laparoscopy-Assisted Versus Open Gastrectomy for Gastric Cancer: A Comprehensive Systematic Review and Meta-Analysis Based on Randomized Controlled Trials

Gastric cancer is a major health problem worldwide [1,2]. It is now the fifth common cancer and the second leading cause of cancer-related mortality in the world, accounting for more the 10% of the annual cancer deaths globally [3-6]. There is global variation in the incidence of gastric cancers with higher incidence in Southeast Asia specially Korea and Japan and much lower incidence in Africa, Australia and USA. The treatment of gastric cancer depends upon the stage of the disease. One of the commonest staging systems is the American Joint Committee on Cancer (AJCC) gastric cancer staging system AJCC 7th edition, known as the TNM classification [7,8]. In stage I to III, radical surgery in the form of gastrectomy and lymph nodes dissection is the mainstay of the treatment and the aim is cure of the disease. In stage IV, palliative treatment is recommended.


Introduction
Gastric cancer is a major health problem worldwide [1,2]. It is now the fifth common cancer and the second leading cause of cancer-related mortality in the world, accounting for more the 10% of the annual cancer deaths globally [3][4][5][6]. There is global variation in the incidence of gastric cancers with higher incidence in Southeast Asia specially Korea and Japan and much lower incidence in Africa, Australia and USA. The treatment of gastric cancer depends upon the stage of the disease. One of the commonest staging systems is the American Joint Committee on Cancer (AJCC) gastric cancer staging system -AJCC 7th edition, known as the TNM classification [7,8]. In stage I to III, radical surgery in the form of gastrectomy and lymph nodes dissection is the mainstay of the treatment and the aim is cure of the disease. In stage IV, palliative treatment is recommended.
Laparoscopy-assisted gastrectomy for gastric cancer was first introduced in 1991 and reported by Kitano et al in 1994 [9]. Since that time, it gains acceptance due to its advantages over open gastrectomy as a result of minimal invasion. In 1997, a study published by Goh et al showed that laparoscopy-assisted gastrectomy was superior to conventional open surgery because of less pain, faster recovery and better cosmetic outcomes [10]. Laparoscopy-assisted gastrectomy is frequently used in Asian countries mainly Korea and Japan, where screening programs resulting in early diagnosis and better prognosis than other countries [11][12][13][14][15]. In spite of its advantages, it remains controversial and still not become an alternative to conventional open gastrectomy because of the complexity of the procedure especially in establishing the continuity of the digestive tract (reconstruction) and extended lymph nodes dissection. Furthermore, the oncological safety (cancer clearance) of laparoscopic-assisted gastrectomy is not yet well-established [16]. Another concern is the post-site recurrence, associated with pneumoperitoneum and visceral manipulation, which have been report by Azagra et al. [17].
The continuing controversy led researchers focus on this area by conducting RCTs and non-RCTs comparing laparoscopyassisted gastrectomy with open gastrectomy [18][19][20][21][22][23][24][25][26][27][28][29][30][31]. But the sample size of these studies is not adequate to define the superiority of LAG over the open technique. Several metaanalyses were conducted in the last decay, some of them include few RCTs, they include data from 734, 390 and 732 patients respectively [32][33][34]; they may have included a false positive errors and they lacked statistical significance to reach a final conclusions because of the relatively small sample size of these meta-analyses [16]; Other meta-analyses combined both RCTs and non-RCTs, but, Publication bias is likely to be greater for non-RCTs because the quality of evidence is much lower than that of RCTs. Consequently, the results of meta-analyses including non-RCTs should be interpreted with caution [35]. Furthermore, most of the previous meta-analyses did not reported on both early and advanced gastric cancer. Additional RCTs have been published and will strengthen the current evidence. Therefore, it is necessary to carry out a meta-analysis by pooling of the results of all available published RCTs up to date to evaluate the safety and effectiveness of LAG over open gastrectomy in gastric cancer, this will help surgeons in clinical decision making.

Materials and Methods
The meta-analysis was performed in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-analysis statement (PRISMA) statement [36].

Literature Search Strategy
A comprehensive systematic literature search using Mendeley, PubMed, EMBASE, the Cochrane Library and Google Scholar databases for studies published until May 2019. Only articles published in English language were included in this meta-analysis. No language restriction was provided. The Following Medical Subject Headings (MeSH) AND free-text terms were used: gastric cancer; gastric carcinoma; stomach cancer; gastric neoplasm's; stomach neoplasm's; laparoscopic-assisted; laparoscopy-assisted; minimally invasive; open gastrectomy; conventional gastrectomy; laprarotomy; Randomized controlled trial. Logistic combinations between these terms were used to maximize sensitivity. The PubMed database was used to search for additional studies using authors' names. Trials were identified also by using related-articles function in the PubMed. The search was extended further by searching the reference lists of all retrieved articles and previous meta-analyses.

Study Selection
The author screens the primary data from the studies collected in the electronic search. The following eligibility criteria were applied for inclusion of the studies in the metaanalysis: i.
Be a published randomized controlled trial.
ii. Studies comparing laparoscopy-assisted with open gastrectomy for gastric cancer, regardless of the type of gastrectomy performed (total gastrectomy, distal gastrectomy and proximal gastrectomy) and regardless of the tumor stage ( early or advanced gastric cancer).
iii. Histological proven of adenocarcinoma of the stomach through endoscopic biopsy iv. Trials reporting at least one of the following outcome measures: operative outcomes (operative time, intraoperative blood loss, and transfused patients), postoperative outcomes( post-operative analgesic consumption, time to first ambulation, time to first flatus, time to first oral intake, length of post-operative hospital stay, overall post-operative morbidity, post-operative surgical and medical complications { abdominal abscess, anastomotic leakage, anastomotic stenosis, bleeding, gastroparesis, ileus, intestinal obstruction, pancreatic fistula, pancreatitis, wound infection and pneumonia}, incidence of reoperation, and mortality), and Oncological outcomes( number of harvested lymph nodes, tumor recurrence and metastasis).
Studies was excluded from this meta-analysis if a.
They were non-RCTs. h. The outcomes of interests were not reported for the two techniques. If two or more articles were published by the same team in the same institution, then the better quality or the most recent publication was included in the meta-analysis to avoid including the same patients.

Data Extraction
Data were extracted using a data extraction forms designed for this study. The following data were extracted from each included study: the first authors' name, year of publication, country, study center, number of participants, number of subject operated on each technique, type of gastrectomy, extent of lymphadenctomy, type of reconstruction, follow-up (months), and patients characteristics including age, sex, BMI, ASA score, co-morbidities, pathological tumor/metastasis stage, tumor size, tumor location, histological type, adjuvant treatment, and the outcomes of interest.

Methodological Quality Assessment
The quality and risk of bias for each included RCT was evaluated using the Cochrane Risk of Bias tool which is recommended in the Cochrane Collaboration Handbook 5.1.0 [37]. The following domains were assessed: Random sequence generation; allocation concealment; blinding of participants and personnel (performance bias); blinding of the outcome assessment (detection bias); incomplete outcome data (attrition bias); selective reporting; and other bias. The risk of bias in each domain was examined and classified as low, high or unclear.

Quality of Evidence
We assessed the quality of evidence of the study outcomes using the GRADE assessment tool as provided by the Cochrane Collaboration. The overall quality of evidence was scored as very low, low, moderate and high according to the evaluation of the risk of bias, inconsistency, indirectness, imprecision and other considerations.

Statistical Analysis
Statistical analysis was performed using Review Manager 5.3 (Nordic Cochrane center, Copenhagen, Denmark). The odd ratios (ORs) and 95% confidence intervals (CIs) were calculated for dichotomous variables including transfused patients, overall post-operative morbidity, short-term mortality, post-operative surgical and medical complications, incidence of reoperation and tumor recurrence and metastasis. Weighted mean differences (WMDs) with 95% confidence intervals (CIs) were calculated for continuous variables including operative time, intra-operative blood loss, post-operative analgesic consumption, time to first ambulation, time to first flatus, time to first oral intake, length of post-operative hospital stays, number of harvested lymph nodes.

Time to first Ambulation
Data from six included papers included 2197 patients was available to calculate the weighted mean difference for the time to ambulation [20,23,70,71,76,77]. The time was shorter in the LAG group than in the OG group (MD -0.49, 95% CI -0.89 to -0.09, P < 0.02) with significant heterogeneity among trials (I² = 95%, P < 0.00001). ( Figure 4B).
Number of harvested lymph nodes.

Discussion
The number of gastric cancer patients undergoing laparoscopy-assisted gastrectomy has been increasing worldwide. However, it cannot be recommended for routine management of gastric cancer before its superiority over OG is not guaranteed. The current meta-analysis examined whether LAG is an acceptable and safe alternative to OG for gastric cancer patients from a clinical perspective. The strengths of this metaanalysis including adequate power with 5198 participants, in addition, we followed the PRISMA and GRADE evidence profile, all included studies were RCTs of good quality, operations were done by both eastern and western surgeons. the trials included patients of all ages and both genders furthermore, we included patients of gastric cancer from all pathological stages (early and advanced disease). So, the results are likely to be representative of the population of gastric cancer patients presented to the reporting centers.
The present study revealed that LAG offers less intraoperative blood loss, faster bowel recovery, shorter hospital stay, less frequency of analgesic consumption and less post-operative overall morbidity. Those benefits are attributed to the minimal invasiveness, which contributed to enhanced recovery after surgery. The less pain, better cosmetic effect and faster recovery could produce not only organic, but also psychological benefits to the patients. The less blood loss in LAG may be justified by the magnified view through the monitor, which allows careful dissection to prevent bleeding, thus preventing interference with surgical vision by blood accumulation [11].
No procedure associated differences were found in the incidence of reoperation and post-operative mortality, which supports the safety of LAG. However, western patients have been found to have higher post-operative mortality and morbidity rates compared with eastern patients [81]. This could be due to the fact that western patients have average higher BMI than eastern patients, which increase technical complexity of LAG with prolonged operative time, increased blood loss and risk of inadequate lymph node dissection. The second reason is that surgeons in eastern countries specially in Korea and Japan might be more experienced in the surgical management of gastric cancer than surgeons in the west. The third reason is that eastern patients presented with early stage tumors with a better prognosis, thanks to the screening programs in those countries. Moreover, eastern patients are younger, healthier with lower rates of cardiovascular disease and lower risks of post-operative thromboembolic events than western patients.
This meta-analysis suggests that LAG has longer operative time as compared to OG. The finding was in line with many previous meta-analyses [11,48,54]. LAG with lymphadenectomy is a new, technically challenging and time-consuming procedure even for experienced surgeons who reached a learning curve plateau. A multivariate analysis revealed that operative time was affected by patients and tumor characteristics regardless of the surgeon's experience initially, followed by gradual decrease in the operative time once proficiency in the laparoscopic surgery has been achieved [82]. Therefore, in LAG, consideration to patients and tumor characteristics is much more important to solve this problem.
Oncological safety is very important in the surgical treatment of gastric cancer. Studies indicated that lymph node status is one of the most critical independent predictors of patient's gastrectomy for gastric malignancy [83,84]. Therefore, lymphadenectomy is important in the treatment of gastric cancer. The insufficient lymph node dissection for node-positive cases increases the potential risk of tumor recurrence [85]. In contrast to previous meta-analyses that shows reduction in the number of harvested lymph nodes in the laparoscopic group [34,54], this meta-analysis clearly shows that there is no statistically significant difference in the number of lymph node harvested and the tumor recurrence/metastasis between the two groups, indicating oncological equivalence of both LAG and OG . Thus, LAG can be used as an alternative to OG for gastric cancer.
The current meta-analysis has few limitations. Firstly; sixteen out of the seventeenth included trials were conducted in east Asia [18][19][20][21][22][23][70][71][72][73][74][75][76][77][78][79], and only one trial conducted in Italy [80]. Thus , the included participants might not reflect the general patient's population ,furthermore the application of the results to western patients should be performed with caution. Secondly, none of the included trials reported quality of life scores or cost effectiveness, which are areas of concern. Finally, the long-term outcomes could not be assessed (Summary of findings). ⨁⨁⨁◯ MODERATE a *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

Laparoscopy-assisted gastrectomy compared to open gastrectomy for gastric cancer
CI: Confidence interval; MD: Mean difference; OR: Odds ratio GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect Explanations a. Because of the nature of the procedure, its very difficult to perform blinging.
b. Unexplained statistically significant heterogeneity among included trials.

Conclusion
Despite the fact that LAG is a technically demanding and timeconsuming procedure, the results of our meta-analysis suggests that LAG can be used as a safe and acceptable alternative to OG for gastric cancer with similar mortality rates, comparable oncological results, less intra-operative blood loss, decreased post-operative morbidity, faster recovery and earlier hospital discharge. Further large well-designed multi-center RCTs are required to evaluate its long-term oncological outcomes.