The outcomes of laparoscopic omentum-preserving gastrectomy compared to open surgery with omentectomy in gastric cancer patients: a propensity score matched study of 249 UICC stage 0–IV gastric cancer patients

All patients selected for this study had a complete preoperative work-up, including preoperative upper-gastrointestinal endoscopy. All patients had a verified stomach adenocarcinoma, and the exact tumor location was reported. In patients with early gastric cancer (EGC), tumor spotting was performed before the operation. A complete preoperative laboratory testing and chest X-ray were performed. In all patients, a chest and abdomen CT (computer tomography) scan was performed. A tumor board reviewed patients with complete preoperative staging before the operation. The eligibility criteria for perioperative treatment were as follows: UICC stage Ib gastric cancer or higher, age 18 years or older, ECOG 0 to 1, and adequate hepatic, renal, marrow, and cardiac function. Patients with a history of recent myocardial ischemia, uncontrolled angina, hypertension, cardiac arrhythmias, congestive heart failure, or other severe medical illnesses were taken under review by the tumor board. The regimen used for perioperative treatment was 5-FU–leucovorin–oxaliplatin–docetaxel (FLOT) four cycles pre- and four cycles post-operative. The precise doses for the regiment are provided in the as described elsewhere [9, 10]. The inclusion criteria for this study were adenocarcinoma diagnosed between 2015 and 2022, UICC tumor stages 0-IV, and complete follow-up. Patients with upfront unresectable gastric cancer, cancer in the stomach remnant, synchronous tumors, and patients lost during follow-up were excluded from the analysis. The demographic data, data regarding surgery, postoperative course, and pathological data were all stored in the hospital’s database. The follow-up data was collected from the national database. The follow-up and survival of the oncological patients in Slovenia is managed centrally in the National Oncological Database. This data is available for all clinicians treating cancer patients. The study was conducted according to the ethical directives of the Helsinki Declaration. Because of the retrospective study design, informed consent from the participants was waived by the Ethics committee.

Laparoscopic surgery was performed as follows: Briefly, upon trocar insertion, a thorough exploration of the abdominal cavity was performed. After the dissection of the right diaphragmatic pillar, the transaction of the phrenic-esophageal ligament, complete mobilization of the esophagus, fixation of the falciform ligament and the left liver lobe, we proceeded to the incision of the gastro-colic ligament starting 3 cm from the epiploic arcade. The mobilization of the esophagus was applied only in patients who received total gastrectomy, proximal subtotal gastrectomy, and the Ivor-Lewis procedure. The left gastroepiploic vessels were dissected at their take-off, taking special care not to injure the omental arteries. This was the point of gastric transection in subtotal gastrectomy. In total gastrectomy, we proceeded to dissection of the lymph node station No. 10. The dissection was continued up until the left diaphragmatic pillar was completely exposed. After completion of the lymphadenectomy of the splenic artery, the dissection was continued on the right side along the epiploic arcade. The gastroduodenal artery was exposed, and the distal part of the common hepatic artery, the right gastric artery, and the right gastroepiploic artery were identified. The anterior superior pancreaticoduodenal vein was the lowermost border of the No. 6 lymph node station dissection. The duodenum was transected with an EndoGIA 60 mm linear stapler with a purple cartridge (Medtronic, Minnesota, ZDA). The final step of the lymphadenectomy was the dissection of the lymph node stations 8, 11, 9 and 1. A 5–7 cm mini-laparotomy was made for specimen extraction. The reconstruction was mainly performed with a Roux-en-Y loop. An EndoGIA 60 mm linear stapled overlapping anastomosis was fashioned, and an air-leak test was performed [11, 12].

After surgery the patients were admitted for surveillance on the intensive care unit for two to three days. We discontinued the nasogastric tube and started clear fluids by mouth. The patients were allowed 100 ml clear fluids on day two, they were mobilized. The volume of fluids was slowly increased to day four, when they started with soft diet. The drains were removed on day 5 or when the secretion was below 200 ml per day. The patients were discharged on day 7.

Open surgery was performed similarly to laparoscopic surgery. The same lymph node stations were included in the lymphadenectomy as in laparoscopic surgery. In patients in the open group omentectomy was performed. This was done at the beginning of the dissection with the reflection of the omentum cranially and the incision of the gastrocolic ligament. The omentum was removed between the takeoff of the right and left gastroepiploic artery and resected en-block with the specimen.

The primary endpoint of the study was the 5-year overall survival. Overall survival was defined as the time from the operation until death from any cause. The secondary endpoints of the study were the morbidity according to Dindo-Claviene classification, the site of recurrence, the incidence of intestinal obstruction, and the length of hospital stay. Perioperative mortality was defined as death within 30 days of the procedure. Patients were regularly followed up after surgery. In the first two years after surgery, patients were followed every 4 months, then every 6 months until 5 years after surgery, then after once yearly. A complete physical exam was made, the tumor markers (CEA, Ca 19-9, Ca 72-4) were checked, and an abdominal ultrasound was performed at each control. In case of suspicion of tumor progression, a CT scan was performed. In patients with confirmed recurrence, the recurrence site was diagnosed with a CT scan, PET CT, operative exploration, or an autopsy.

We defined six parameters associated with significant prognostic value after gastrectomy (tumor location, type of operation, UICC stage, ASA score, perioperative chemotherapy, and distant metastases). The PSM was performed with these variables to adjust for their confounding effects. The propensity matching was performed with the SPSS program, which uses a logistic model to estimate the propensity scores with the six variables as co-factors and automatically performs the nearest matching with a caliper of 0.1 in a ratio of 1:1.

Descriptive statistics were used to summarize patients’ demographic and clinical data. Continuous variables were expressed as mean ± SD or median with interquartile range (IQR) where appropriate. Categorical variables were expressed as absolute values with percentages. Continuous variables were compared with Student’s t-test and Mann–Whitney’s U-test where applicable. Variables above the threshold p-value of 0.1 were included for multivariate analysis. The binary logistic regression model was used to determine the significant predictor for peritoneal recurrence. The Cox regression model was used for primary analysis and included covariates with a p-value of more than 0.1 in univariate analysis. Treatment effect estimates were expressed as hazard ratios with a 95% confidence interval. Both multivariate logistic models used the backward stepwise conditional regression. Kaplan–Meier curves were constructed to determine time-to-event end-points. Differences in survivals between groups were determined with the Log-rank and Breslow tests. A p-value of > 0.05 was selected as the level of significance. All statistical analyses were performed on SPSS for Windows 10 v. 25 (IBM, USA).

In total, 311 patients have been operated on for gastric cancer at our institution between 2015 and 2022. From these, 249 were eligible for our study (OPN; n = 143; LAP; n = 106). The included patients were divided into two groups: patients with open surgery and complete omentum resection and patients with laparoscopic gastrectomy and omentum preservation. After PSM, 202 patients were selected for the analysis (OPN; n = 101; LAP; n = 101). Of these patients, four patients were excluded because they were lost during follow-up, leaving 198 patients for the final analysis (LAP: n = 100; OPN: n = 98)(Patients’ demographics and tumor characteristics before and after PSM are presented in Table 1). Both groups were well balanced in the demographic and tumor characteristics after PSM.

In both groups, the total gastrectomy was the most prevalent procedure (OPN: 67.4% vs LAP: 54%; p = 0.266). Open surgery with omentectomy was on average 30 min faster compared to laparoscopic omentum-preserving gastrectomy (247.6 ± 65 in OPN vs 276 ± 59 in LAP; p = 0.002). D2 lymphadenectomy was performed in the same proportion of patients in both groups (LAP: 78% vs. OPN: 75.8%; p = 0.547). Meanwhile, the lymph node yield was significantly higher in the LAP group (31 ± 11 LNs vs. 25 ± 12 LNs; p = 0.002). The conversion rate in the Lap group was 16%. The reasons for conversion are listed in Table 2. In both groups, R0 resection was achieved in the same proportion (LAP: 96% vs. OPN: 98.9%; p = 0.503). Most of the patients had advanced gastric cancer (pT2 or higher) (LAP: 64% vs. OPN: 62.1%; p = 0.341). Patients with pathological stage T0 were included in both groups. These patients were diagnosed with gastric adenocarcinoma on preoperative biopsy and were found to have a complete tumor regression after chemotherapy. All preoperative biopsies were revised to confirm the diagnosis.

Patients in the LAP group had significantly less morbidity compared to OPN group (LAP: 17% vs. OPN: 23.4%; p = 0.009). There was no significant difference in the 30-day morbidity between groups (LAP: 1.1% vs. OPN: 0%; p = 0.317). Patients in the LAP group had a significantly shorter hospital stay compared to OPN group (9(6) days vs. 10(5) days; p = 0.009). Perioperative chemotherapy was administered in a similar proportion of patients in both groups (LAP: 49% vs. OPN 58.9%; p = 0.106).

In both groups, the most common site of recurrence was the peritoneum (LAP: 16% vs. OPN: 23.2%; p = 0.594). Patients in the LAP group had less frequent peritoneal recurrence compared to OPN group and more frequent recurrence in the lungs (LAP: 3% vs. OPN: 1.1%), while liver metastases were more common in the OPN group (LAP: 0% vs. OPN: 4.2%). All former differences did not reach the level of significance. In both groups, recurrence in local or distal LNs was exceedingly rare (LAP: 1% vs. OPN: 1.1%; p = 0.594). Also, there was no significant difference in the occurrence of intestinal obstruction. Intestinal obstruction was observed in 3.2% in OPN and in 3% in LAP group (p = 0.949).

The median follow-up period was 3.6 (5.8) years. There was no significant difference in the long-term survival between LAP and OPN groups (p = 0.950). The 5-year overall survival was 62.2% in LAP and 54.4% in OPN group (Fig. 1). From the included variables, the Cox proportional hazard model determined UICC (HR: 2.230; 95%CI 1.734–2.867; p < 0.0001) and age (HR: 1.057; 95%CI 1.030–1.086; p < 0.0001) to be significantly associated with long-term survival (Fig. 2).