Comparison of short-term outcomes and perioperative systemic immunity of laparoscopy-assisted and open radical gastrectomy for gastric cancer


  • Source of funding Internal support from the Scientific Research Foundation granted by West China Hospital, Sichuan University (No.141040132), China.

  • Contribution of authors X.Z. Chen designed and managed the study and assisted with the laparoscopic surgery, data collection, statistical analysis, and manuscript writing; J.K. Hu provided academic instruction and performed laparoscopy-assisted surgery; B. Zhang, Z.X. Chen, and J.P. Chen performed the open surgery; J. Liu and L.L. Wang performed the laboratory tests; K. Yang and C. Yang assisted with data collection and handling; and Z.G. Zhou performed academic instruction.

  • Conflict of interest None to declare.

Jian-Kun Hu, M.D., Ph.D., Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Guo Xue Xiang, No.37, Chengdu 610041, Sichuan Province, China Email:


Objective: To compare the perioperative systemic immunity of laparoscopy-assisted and open radical gastrectomy for gastric cancer.

Methods: Patients with gastric adenocarcinoma proven by endoscopy and biopsy were eligible, while patients with preoperative staging of T4, N2–3, or M1 were excluded. Eligible patients willing to undertake laparoscopic surgery in the consecutive cohort were assigned to the laparoscopy-assisted gastrectomy (LAG) group, while concurrent patients were assigned to the conventional open gastrectomy (OG) group. All operations were performed with the intention of radical resection. Various immunological parameters were tested in peripheral venous blood collected at preoperative 1st day and postoperative 2nd day (POD2) and 7th day (POD7). SPSS 13.0 software was used for statistical analysis.

Results: Thirty patients were included, 15 each in the LAG and OG groups. The general characteristics and short-term outcomes (harvested lymph nodes number, hospital stay, complications, and mortality rate) of the two groups were comparable, but the operation time was significantly longer in LAG (P = 0.001). Moreover, intergroup comparisons indicated no significant differences between the groups in levels of neutrophils, T-lymphocytes, natural killer cells, IgG, IgM, IgA, C3, C4, interleukin-6, or interleukin-10 at any time point (P > 0.05). However, there was a gradual decrease in natural killer cell count in the LAG group up to POD7 (P = 0.008).

Conclusion: The changes in systemic immunity markers were comparable between laparoscopy-assisted and open gastrectomy for gastric cancer. However, there was a trend of suppression of natural killer cells in the laparoscopy-assisted gastrectomy group.


Minimally invasive surgery has become commonplace in oncology, and laparoscopic resection for colorectal cancer is practiced globally (1, 2). Laparoscopy-assisted gastrectomy (LAG) for gastric cancer was proposed in the past decade, and several randomized controlled trials have indicated the feasibility and safety of this technique (3).

Disturbance of immunity is always a concern about laparoscopic techniques for oncologic surgery (4). The establishment of CO2 pneumoperitoneum can induce intraperitoneal hypertension and hypercarbonemia (5–8). Complex surgical procedures can prolong the operating time and anesthesia (9, 10). These are unfavorable factors for postoperative immunity and increase the underlying risk of metastasis, seeding, or recurrence of gastric cancer. However, lessening surgical trauma might lighten the disturbance of immunity caused by operative stress. Therefore, the LAG technique's effects on immunity are still controversial. Whether LAG affects the survival outcome of gastric cancer is also still unclear.

The present clinical controlled study aimed to compare and map the perioperative changes of various systemic immunological components of LAG and conventional radical open gastrectomy (OG) for gastric cancer, including non-specific immunity, specific immunity, the complement system, and cytokines.


The study protocol was approved by the Ethics Committee of Biomedical Research, West China Hospital, Sichuan University, China (SN. 2008–3). It was also registered in the Chinese Clinical Trial Register (ChiCTR), a primary register of the WHO International Clinical Trials Registry Platform (SN. ChiCTR-TRC-00000373) (11). All included patients provided informed consent.


Patients with gastric adenocarcinoma verified by endoscopy and biopsy were potentially eligible. Inclusion criteria were: 1) preoperative 64 multi-detector spiral computed tomography assessment indicating resectable disease of TNM stage within cT1N0M0-cT3N1M0, as determined using Japanese Gastric Cancer Association (JGCA) classifications (12); 2) WHO performance score <2; 3) age ranging from 18 to 75; and 4)  BMI ≤ 30 kg/m2. Exclusion criteria were: 1) other type of stomach malignancy, e.g., lymphoma or gastrointestinal stromal tumor; 2) any previous malignancy or synchronous malignancy in other organs; 3) severe co-morbidity leading to intolerance of major abdominal surgery; 4) previous operation on upper abdomen; 5) emergency operation due to major hemorrhage or perforation of gastric cancer; and 6) receipt of neo-adjuvant chemotherapy.


This was a prospective non-randomized clinical controlled trial, and blinding was not feasible. In the series, eligible patients who were willing to undergo minimally invasive surgery were assigned to the LAG group. Concurrent eligible patients who did not ask for laparoscopic surgery were assigned to the conventional open gastrectomy (OG) group. The allocation was balanced by surgeons without personal preference.

Surgical techniques

Both LAG and OG were performed by a single gastrointestinal surgical team (Chen JP, Chen ZX, Hu JK, Zhang B and Chen XZ) that had wide experience with open and laparoscopic procedures. All operations were performed with the intention of radical resection. The resection pattern of distal, proximal, or total gastrectomy was dependent on the site of stomach tumor. Intraoperative frozen section examination was carried out to ensure R0 resection in all the cases. D2 or D2+ lymphadenectomy according to the JGCA classification was adopted in both LAG and OG groups (12). Reconstruction patterns were Billroth-II anastomosis (anterior to transverse colon) for distal gastrectomy, esophago-gastric remnant anastomosis for proximal gastrectomy, and Roux-en-Y anastomosis (anterior to transverse colon) for total gastrectomy. All LAG were performed by a single surgeon, Hu JK, and assistant surgeon Chen XZ.

Immunological parameters and laboratory tests

Fasting peripheral venous blood samples were collected on the mornings of the preoperative 1st (PRE) and postoperative 2nd (POD2) and 7th days (POD7). We expected that the POD2 levels might represent the initial stress of operation and anesthesia, while the POD7 levels might represent the end of the early postoperative period. The specimens were immediately transferred to and tested in the Lab of Clinical Immunology and the Lab of Clinical Biochemistry at West China Hospital, Sichuan University, China.

Neutrophils and lymphocytes with EDTA were counted by XE-2100 Flow Cytometer (Sysmex America, Inc., USA). The proportions of the T-lymphocyte sub-population and of natural killer cells with heparin were analyzed with an Epics XL Flow Cytometer (Beckman Coulter, Inc., USA). The levels of immunoglobulins (IgG, IgM, IgA) and complements (C3, C4) without anticoagulant were analyzed by IMMAGE Immunochemistry Analyzer (Beckman Coulter, Inc., USA). The levels of interleukins (IL-6, IL-10) without anticoagulant were tested by ELISA assay kits (Immunotech, Inc., France) and detected by Benchmark Microplate Reader (Bio-Rad Laboratories, Inc., USA).


Statistical analysis was performed by SPSS 13.0 software (SPSS, Inc., USA). For continuous data with normal distribution, the one-way ANOVA test was used, while for data not conforming to normal distribution, the Mann-Whitney U test was used. For categorical data, the chi-square test or Fisher's exact test was used. A two-sided P value of less than 0.05 was considered to indicate statistical significance. Intragroup comparisons were performed to compare POD2 or POD7 levels to baseline in each group. Intergroup comparisons were to compare two groups based on baseline levels minus the POD2 or POD7 level.


Patient information

The study included a total of 30 patients enrolled between August 2008 and March 2009 (15 each in the LAG and OG groups). The general characteristics of the two groups were fairly comparable (P > 0.05) (Table 1).

Table 1.  General information on included patients
 LAG (n = 15)OG (n = 15)P value
  1. § In OG group, 3 patients were classified as stage IV due to No. 8p lymph node metastasis.

Gender (male/female)12 / 312 / 31.000
Age (years)52.3 ± 12.254.3 ± 12.30.669
Site of tumor  0.878
Maximum tumor diameter (cm)4.53 ± 3.004.63 ± 1.650.911
Pathological TNM stage (JGCA)  0.175
 Ia (T1N0)20 
 Ib (T1N1/T2N0)12 
 II (T1N2/T2N1/T3N0)33 
 IIIa (T2N2/T3N1)33 
 IIIb (T3N2)64 
 IV (TXN3M0) §03 
Resection pattern  0.670
 Distal gastrectomy98 
 Proximal gastrectomy21 
 Total gastrectomy46 

Short-term clinical outcomes

The operation time was significantly longer in the LAG group, at 54.2 min (P = 0.001) (Table 2). Though the average number of lymph nodes harvested in this group was lower by 11.2 (P = 0.028), for each case we achieved the aim of dissecting at least 15 lymph nodes, meeting the criteria for radical gastrectomy. The postoperative hospital stay was about 1.4 days shorter for the LAG group (P = 0.019).

Table 2.  Short-term clinical outcomes
 LAG (n = 15)OG (n = 15)P value
  1. In the LAG group, there was one case of cardiac sudden death at postoperative 15th day and one case of gastric paralysis. In the OG group, there was one case of intraoperatively supraventricular arrhythmia, one of postoperative pneumonia, and two of gastric paralysis.

Operation time (min)293.5 ± 31.6239.3 ± 23.70.001
Lymph nodes harvested 29.7 ± 11.0 40.9 ± 15.10.028
Postoperative hospital stay (day)  9.5 ± 1.4 10.9 ± 1.60.019
Postoperative complications2 (13.3%)4 (26.7%)0.651
Mortality1 (6.7%)0 (0%)1.000

The postoperative complication rates were 13.3% (2/15) and 26.7% (4/15) in LAG and OG, respectively, without a significant difference (P = 0.651) (Table 2). One patient died, of cardiac sudden death at postoperative 15th day, and there was no significant difference in mortality between the two groups (P = 1.000).

Systemic immunological changes


Neutrophils represent non-specific immunity and immune response to surgical stress. Neutrophil counts were significantly increased at POD2 in both groups (P < 0.001), and then regressed slightly, but were still significantly higher than baseline levels at POD7 in both groups (P < 0.001 for LAG, P = 0.001 for OG) (Figure 1a). However, intergroup comparisons found there were no statistical differences between two groups at either POD2 (P = 0.323) or POD7 (P = 0.489) time point, respectively.

Figure 1.

Error bars of perioperative changes in immune cell counts: (a) neutrophils (×109/L); (b) T-lymphocytes (×109/L); (c) natural killer cells (×109/L).

T-lymphocyte sub-populations

T-lymphocytes are the most important component of cellular immunity. Intragroup comparison found that T-lymphocyte counts were temporarily decreased at POD2 (P = 0.001 for LAG, P < 0.001 for OG compared to baselines) and then rebounded to near–baseline levels (P = 0.252 for LAG, P = 0.851 for OG) in both groups (Figure 1b). Intergroup comparisons found no significant difference between the two groups in the change from baseline at either POD2 (P = 0.379) or POD7 (P = 0.246), respectively. The CD4+ and CD8+ sub-populations of T-lymphocytes showed a similar pattern, and no significant difference was found between groups (data not shown).

Natural killer cells

Natural killer (NK) cells are a type of cytotoxic lymphocyte that constitutes a major component of the innate immune system. In the LAG group, the mean NK cell count was significantly lower at POD7 than at baseline (P = 0.008). In the OG group, the NK cell count was lowest at POD2, but was not significantly different from the baseline (P = 0.267). However, although there appeared to be a discrepancy between the two groups, intergroup comparison showed there was no significant difference between the changes from baseline of the two groups at either POD2 (P = 0.273) or POD7 (P = 0.133) time point, respectively (Figure 1c).


Immunoglobulins play an essential role in humoral immunity. IgG levels were significantly lower at POD2 (P = 0.030 for LAG, P = 0.001 for OG) and returned to baseline levels at POD7 (P = 0.258 for LAG, P = 0.433 for OG) (Figure 2a). In the LAG group, the IgM level was slightly decreased at POD2 without significant difference to baseline and had a potential trend of rebound at POD7 (P = 0.204). However, in the OG group, the IgM level was significantly higher than baseline at POD7 (P = 0.020) (Figure 2b). IgA levels in both groups showed a small, insignificant inhibition at POD2 (P = 0.469 for LAG, P = 0.109 for OG), and near-baseline levels at POD7 (P = 0.932 for LAG, P = 0.873 for OG) (Figure 2c). There were no significant differences between the two groups in the changes from baseline of IgG, IgM, or IgA (P > 0.05).

Figure 2.

Error bars of perioperative changes of immunoglobulins: (a) IgG (g/L); (b) IgM (g/L); (c) IgA (g/L).


The complement system is a biochemical cascade that is part of the innate immune system. C3 levels were decreased at POD2 in both groups, but the difference was only significant in the LAG group (P = 0.041). Subsequently, C3 levels returned to baseline levels in both groups (P = 0.647 for LAG, P = 0.720 for OG compared to baseline) (Figure 3a). C4 levels decreased slightly but not significantly at POD2 for both groups (P = 0.163 for LAG, P = 0.152 for OG), then returned to baseline levels at POD7 (P = 0.624 for LAG, P = 0.833 for OG) (Figure 3b). The changes in C3 and C4 were similar in both groups, without significant differences by intergroup comparison (P > 0.05).

Figure 3.

Error bars of perioperative changes of complements: (a) C3 (g/L); (b) C4 (g/L).


Neither IL-6 nor IL-10 levels accorded with a normal distribution. Hence, the Mann-Whitney U test was used to analyze the changes in IL-6 and IL-10. In the OG group, the IL-6 level was significantly increased at POD2 and POD7 (P < 0.001 and P = 0.024, respectively) relative to baseline (Figure 4a). In the LAG group, the IL-6 level increased at POD2 (P = 0.003), but returned to baseline level at POD7 (P = 0.129) (Figure 4a). IL-10 levels significantly increased in the LAG and OG groups at POD2 (P = 0.002 and P = 0.017, respectively), and subsequently regressed to baseline levels at POD7 (P = 0.039 for LAG, P = 0.028 for OG) (Figure 4b). The intergroup comparisons did not find any difference of the changes of IL-6 or IL-10 (P > 0.05) (Figure 4).

Figure 4.

Box plots of perioperative changes of interleukins: (a) IL-6 (pg/L); (b) IL-10 (pg/L).


Gastric cancer is a common malignancy, and is a major health burden globally (13, 14). Although the incidence of gastric cancer is decreasing in western countries, its incidence remains high in east Asian countries such as Japan, Korea, and China (15–17). Surgical resection is currently the accepted treatment for gastric cancer (18), and laparoscopic surgery has gradually spread and prevailed in the surgical management of gastric cancer, especially in east Asia (19–22).

However, there have been few investigations on perioperative immunological alterations induced by LAG in patients with gastric cancer. Fujii et al. demonstrated that changes in T-lymphocyte populations did not differ significantly between LAG and OG (23). Similarly, our clinical study indicated that LAG's influence on systemic immunity is equivalent to that of open surgery in various aspects. Non-specific immunity appeared to be normally and rapidly stimulated and to remain high for the entire early postoperative period. Cellular and humoral immunity trended to be temporarily inhibited, but mostly recovered with time. The LAG group's earlier recovery of IL-6 may mean that less inflammatory response occurred in that group.

NK cells are thought to play a major role in tumor rejection (24). Recent studies have shown that NK cells play a particularly important role in suppressing tumor growth and metastasis (25). Da Costa et al. showed in animals that stimulation of tumor growth after open surgery is due to suppression of NK cell activity (26). Tang et al. and Sandoval et al. found both laparoscopic and open surgery could similarly suppress the natural antitumoral cellular immunity (27, 28). Wichmann et al. found that although there was significant decrease of NK cell counts postoperatively in both laparoscopic and open groups, the count was significantly higher in the laparoscopic group than in the open group (29). It seems that laparoscopic surgery has an advantage in preserving natural antitumoral immunity. However, we found that LAG led to a decreased post-surgical NK cell count, but that there was no significant suppression in the OG group. Therefore, LAG's influence on NK cell counts is still unclear, and requires further investigation.

In short, the changes in systemic immunity were apparently similar and comparable between laparoscopy-assisted and open gastrectomy for gastric cancer. However, we recommend further investigation of the influence of LAG on natural antitumoral immunity.


We thank the Gastrointestinal Volunteer Team (GIVOLT) based on the Multi-disciplinary Treatment Project of Gastrointestinal Tumors, West China Hospital, Sichuan University, China.

Conflict of Interest

All the authors have no conflicts of interest or financial ties to disclose.