Gastric cancer is one of the most common cancers and a main economic burden worldwide (1). According to the GLOBOCAN 2020 data, gastric cancer is the fifth most common malignancy and the fourth leading cause of cancer death, causing an estimated 768,793 deaths in 2020 globally (2). Surgical resection with lymphadenectomy is the cornerstone of multimodality curative treatment, and open gastrectomy (OG) has long been the gold standard worldwide (3). However, since Kitano et al. (4) reported the first laparoscopic gastrectomy (LG) for the treatment of early-stage distal gastric cancer in 1994, this minimally invasive technique has been rapidly developed and in the field of gastric cancer, especially for treatment of early gastric cancer (EGC) (5, 6).

Nowadays, the LG has gained growing popularity in the treatment of EGC since this minimally invasive technique has some definite benefits including lower postoperative complications, faster recovery, shortened postoperative length of stay, and better quality of life. Previously, several well-designed randomized controlled trials (RCTs) from China, Korea, and Japan demonstrated the beneficial short-term outcomes of laparoscopic distal gastrectomy (LDG) including less blood loss and postoperative pain, faster recovery, and shorter hospital stay, and similar oncologic safety to the open approach (7–10). However, despite the extensive use of laparoscopic surgery, whether this minimally invasive approach is beneficial for patients with advanced gastric cancer (AGC) remains controversial.

Recently, the CLASS-01 (11) and KLASS-02 trials (12) updated their results of long-term outcomes, indicating that locally AGC patients with LDG had similar long-term survival rates compared to open distal gastrectomy. Moreover, the LOGICA trial (13), which compared the LG with OG for treatment of AGC in the Western population, reported comparable outcomes including postoperative complications, length of hospital stay, R0 resection rate, lymph node yield, and 1-year overall survival (OS) rate. Therefore, in order to summarize the current high-quality evidence, we performed this meta-analysis of RCTs to compared the short- and long-term outcomes of LG versus OG for patients with AGC.

We conducted this meta-analysis according to the updated PRISRMA statement (14) ( Supplementary Material 1 ) and registered the protocol on PROSPERO (CRD 42021297141). A literature search was performed in PubMed, Embase, Scopus, and Cochrane Library for eligible RCTs in English from inception through December 2021. The search used broad search terms containing “gastric cancer,” “gastric carcinoma,” “stomach cancer,” “laparoscopic,” “laparoscopy,” “open gastrectomy,” and “RCT” (complete search strategies in Supplementary Material 2 ).

There is growing high-quality RCTs supporting the feasibility of LG for AGC, and its safety is confirmed in our study. In this up-to-date meta-analysis, we reviewed 12 RCTs with 4,101 patients to compare the short- and long-term outcomes of LG versus OG. The result shows that the LG significantly reduces the length of hospital stay and blood loss, whereas the number of retrieved lymph nodes was lower and surgical time was longer in the LG group. Furthermore, there were no differences between LG and OG groups in terms of postoperative complications, short-term mortality, and survival rate at 1, 3, and 5 years. The finding provides further evidence for the safety and efficacy of LG for the treatment of ACG.

To the best of our knowledge, this study is not the first meta-analysis of RCTs to compare the LG with OG for the treatment of gastric cancer. Recently, Vasavada and Patel (35) performed an updated meta-analysis of 11 RCTs (6 RCTs for EGC and 5 RCTs for AGC); the results demonstrated that the LG was associated with lesser wound-related complications without decreasing the length of hospital stay. However, for the patients with AGC, most trials analyzed in previous meta-analyses were non-randomized, which may increase the risk of potential selection and publication bias. Therefore, the current meta-analysis provides the most comprehensive and accurate analysis, since it sums up the up-to-date and high-quality RCTs in terms of comparing LG to OG for patients with AGC. Also, compared with previous meta-analyses, more RCTs with a long-term follow-up were included in it. In general, our results are in compliance with previous meta-analyses (36–40), showing that when compared to the open approach, the LG provides improved short-term outcomes and similar long-term outcomes in patients with AGC.

When choosing the clinical outcomes of our study, we compared the LG with OG on different levels in terms of safety (postoperative morbidity), difficulty (operative time, blood loss, number of retrieved lymph nodes), efficiency (length of hospital stay), and its long-term oncologic results (OS and DFS rates). The results of our meta-analysis indicate that the short-term outcomes consisting of blood loss and length of hospital stay are in favor of a laparoscopic approach, especially for total gastrectomy. Since the advanced laparoscopic approach provides a magnified surgical view while minimizing the length of the incision, a more delicate surgical manipulation of the organs, vessels, and nerves could be achieved during operation (23). In addition, the reduction of hospital stay may be a combined result of fewer blood loss during operation, faster postsurgical recovery of bowel function, and lighter postoperative pain (36).

The overall postoperative complications, including minor and major complications, were similar between the two surgical procedures in our study. However, a recent meta-analysis of data from 6 RCTs and 18 non-randomized trials found that LG was associated with a lower postoperative complication rate, with a significantly lower incidence of medical and minor surgical complications (36). The difference may result from several newly published RCTs, especially the CLASS-01 (16) and LOGICA trials (13). In the sensitivity analysis, we found a significant difference for the postoperative complications in the LG group after omitting the CLASS-01 trial (41). Considering the high risk of imprecision bias, more evidence about the effect of LG on postoperative complications is compellingly needed in the future. Furthermore, anastomotic leakage, the major postoperative complication of gastric surgery, was comparable between two groups. Notably, although the observed difference was meaningless from the statistical perspective, anastomotic leakage seemed to have a higher possibility to occur after laparoscopic surgery. It highlights that this potential risk should be put more emphasis on. There are studies that propose that the application of mini-laparotomy for extracorporeal anastomosis in laparoscopic surgery for AGC may result in increased surgical difficulty, which may increase the likelihood of anastomotic drawbacks on the other hand (41).

Based on the updated studies, LG requires a longer surgical time, which is in line with the results of all included trials. Compared with open approaches, laparoscopic techniques are more complex and less flexible. Frequently cleaning cameras and changing instruments during operation can also extend the surgical time (42, 43). In addition, it is a challenge to perform the dissection of enlarged or suprapancreatic lymph nodes through the laparoscopic approach, as it is difficult to follow the no-touch principle for laparoscopic lymphadenectomy at a deep lymph node station. Moreover, due to the restriction of the visibility and the narrowness of the peritoneal cavity, total omentectomy in LG is also hard to achieve compared with that in OG.

Although the result of our meta-analysis indicates that the number of retrieved lymph nodes was lower in the LG group, the mean number of retrieved lymph nodes in the LG group was 32.45 (95% CI 29.01 to 35.89). Based on the American Joint Committee on Cancer, an adequate dissection should include at least 15 lymph nodes for patients with gastric cancer to ensure accurate and robust N staging (44). A recent study demonstrated that an examined lymph node threshold of more than 30 was shown to be beneficial of survival for patients with gastric cancer and should be considered a clinical benchmark for practice (45). It is in accordance with the most crucial results of our study in terms of the long-term outcomes. The 1-, 3, and 5-year survival rates were similar among the two groups. When the surgical margins fulfilled the R0 resection criteria and adequate lymph node dissection could be achieved, long-term survival largely depended on the intrinsic biological characteristic of the cancer rather than the surgical approach (46, 47).

Furthermore, the learning curve was proved to have significant effects on most of the important surgical and short-term recovery outcome parameters (48). Yoo et al. performed a prospective study to estimate the learning curve of LG for EGC, indicating that surgeons who performed at least 50 cases of LG could achieve lower postoperative complications, more resected lymph nodes, shorter surgical time, and postoperative hospital stay (49). Thus, the LG should be restricted to specialist centers where adequate training and supervision could be provided during the learning curve.

However, some limitations of our meta-analysis must be acknowledged. First of all, the sample size of some included trials was relatively small, which may decrease the credibility of the results in our study or lead to small study effect bias (50). Secondly, our study did not consider the differences and potential impact of surgeons’ experience, perioperative care protocols, and surgical technique between studies, despite their application having been shown to be beneficial in many studies (51–53). Thirdly, most of all included trials were conducted in East Asia (eight in China, two in Korea), except one in Italy and one in Netherlands. Therefore, the generalizability of the findings to Western countries may be limited. Moreover, since the Western population has a comparatively low incidence of gastric cancer, higher body mass index, and more comorbidities (54), the results may not necessarily apply in the Western population.

Last but not least, there was significant heterogeneity in some pooled estimates, which might be explained by differences in sample sizes, surgeons’ experience, perioperative care protocols, surgical technique, pre- and postoperative chemotherapy, and other factors. Variations in sample size among studies were large, and some studies enrolled patients during a wide study interval, which may have introduced biases due to a progression in mastering the surgical skills and improvements in surgical instruments.

Our findings, which are contingent on rigorous meta-analyses of high-quality RCTs, suggest that LG offers improved short-term outcomes including shorter hospital stays and fewer blood loss, with comparable postoperative complications, short-term mortality, and long-term survival rates when compared to the open approach. However, considering the significant heterogeneity, more RCTs are needed to further evaluate the clinical outcomes of LG versus OG for patients with AGC. Furthermore, this updated meta-analysis could be the basis of future meta-analyses, as the inclusion criteria, statistical analysis, and short- and long-term outcomes were clearly defined and meticulously analyzed.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

JJ and SW conceived the idea, performed the analysis, and drafted the initial draft writing of this paper. GY, JW, and XX contributed to the collection and interpretation of data. KZ helped to frame the idea of the study and provided technical support. SW contributed to the revision of this paper and the final approval of the version to be published. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.