Comparison of efficacy and safety between endoscopic and laparoscopic resections in the treatment of gastric stromal tumors: a systematic review and meta-analysis

Background: There are still clinical controversy on the efficacy and safety of endoscopic resection (ER) and laparoscopic resection (LR) in the treatment of gastrointestinal stromal tumors (GISTs). The present study aimed to evaluate the safety and efficacy of ER in the treatment of GISTs by comparing the relative outcomes of ER to LR. Methods: PubMed, Web of Science, Cochrane Library, and Embase were searched. Data were retrieved from January 2010 to January 2020 and subjected to a meta-analysis based on the intraoperative and postoperative outcomes of ER and LR. The intervention arm was treated by LR while the comparator arm was treated by ER. Relevant literature was selected based on the inclusion criteria, data was extracted, and quality evaluation of the included literature was carried out. The Newcastle-Ottawa Scale (NOS) was applied for assessing the quality of included studies. Heterogeneity between studies was assessed using the Cochrane.2 test and I2 statistic, and Funnel plots and Egger's test were used to detect publication bias. Results: The present analysis included 13 studies, comprising a total of 1,261 patients, (ER vs. LR: 543 vs. 718). The incidence rate of postoperative complications [odds ratio (OR), 0.400; P=0.001] was significantly lower in the ER group [3.3%; 95% confidence interval (CI), 0.015 to 0.055] than the LR group (8.9%; 95% CI, 0.03 to 0.17). The meta-analysis revealed that the recurrence rate following ER (1.7%; 95% CI, 0.005 to 0.033) was lower than that following LR (2.5%; 95% CI, 0.012 to 0.041). The R0 resection rate of ER (99%; 95% CI, 0.975 to 0.999) was similar to that of LR (100%; 95% CI, 0.995 to 1.000). No publication bias in this study (P>0.10), and the sensitivity analysis showed that the study was robust. Conclusions: ER was safer and more efficient than LR in terms of all the outcomes, except the R0 resection rate. Thus, ER should be considered the treatment of choice. However, attention should be paid to the surgical margin status following ER.