[Comparison between laparotomy and laparoscopic operation for patients with gastric gastrointestinal stromal tumors:...

Objective: To evaluate the safety, feasibility and efficacy of laparoscopic surgery for gastric gastrointestinal stromal tumors (G-GISTs). Methods: This study is a multicenter observational study. Between 1998 and 2017, 1072 patients with G-GISTs underwent laparoscopic and open resection were retrospectively analyzed clinicopathological data of G-GISTs patients from 12 Chinese hospitals (including Guangdong Provincial People's Hospital).

(1) Pathologically confirmed G-GIST by immunohistochemistry and/or mutation analysis; (2) Underwent laparoscopic or open resection; (3) Achieved R0 resection; (4) Age ≥18 years.

(1) Metastatic or recurrent G-GIST; (2) Concurrent other malignancies; (3) Received neoadjuvant imatinib therapy; (4) Intraoperative conversion to laparotomy. Propensity score matching (PSM) was performed at a 1:1 ratio (caliper width = 0.1×SD) with matched variables: age, sex, body mass index (BMI), admission time, surgical approach, combined organ resection, intraoperative tumor rupture, tumor location, tumor size, mitotic count, tumor necrosis, CD117, CD34 expression, and modified NIH risk stratification.

long-term oncological outcomes (overall survival [OS] and recurrence-free survival [RFS]).

Perioperative outcomes (tumor rupture, operation time, incision length, time to first flatus, time to semiliquid diet, postoperative complications, hospital stay). Results: Among 1 072 enrolled patients (laparoscopic group: 574; open group: 498), the mean age was (57.8±11.6) years (522 males, 550 females). Before PSM, significant differences existed in age, BMI, admission time, tumor location, surgical approach, tumor size, mitotic count, necrosis, CD34, and modified NIH risk stratification (all P<0.05). After PSM (279 patients per group), baseline characteristics were balanced (all P>0.05). After PSM, compared with the open group, the laparoscopic group had a shorter incisional length [(4.2±0.1) cm vs. (13.0±0.2) cm, t=43.063, P<0.001], less blood loss [(49.0±3.7) ml vs. (148.7±20.2) ml, t=4.851, P<0.001], shorter time to first flatus [(3.4±0.1) days vs. (4.7±0.2) days, t=6.151, P<0.001], shorter time to semiliquid diet [(5.7±0.2) days vs. (7.7±0.2) days, t=7.559, P<0.001], and a shorter postoperative hospital stay [(7.8±0.2) days vs. (10.7±0.3) days, t=7.642, P<0.001]. No differences were observed in operation time, tumor rupture, or complication rates (all P>0.05). Post-PSM 5-year OS rates were 96.8% (in the laparoscopic group) vs. 95.1% (in the open group) (χ²=0.060, P=0.807); 5-year RFS rates were 94.1% vs. 92.5% (χ²=1.908, P=0.168). Cox analysis showed that surgical approach was not a significant factor for OS (P>0.05). Mitotic count >10/50 HPF was an independent risk factor for OS (P<0.05). Tumor size >5 cm was an independent risk factor for RFS (P<0.05). Conclusions: Comparing to the laparotomy, laparoscopic operation for G-GISTs had superior perioperative outcomes without compromising oncologic outcomes.