Is concomitant splenectomy beneficial for the long-term survival of patients with gastric cancer undergoing curative gastrectomy? A single-institution study

Background Curative resection is the treatment of choice for gastric cancer, but it is unclear whether gastrectomy should also include splenectomy. We retrospectively analyzed long-term survival in patients in our hospital who underwent gastrectomy plus splenectomy (G + S) or gastrectomy alone (G-A) for gastric cancer. Methods We identified 214 patients who underwent surgery with curative intent between 1980 and 2003. Of these, 100 underwent G + S, and 114 underwent G-A. The primary endpoint was 5-year overall survival (OS). Results Median follow-up was 18 months in patients who underwent G + S, and 26.5 months in patients who underwent G-A. The 5-year OS rate was significantly higher in patients who underwent G-A (33.8%; 95% CI 24.2 to 43.4%) than in those who underwent G + S (28.8%; 95% CI 19.6 to 38.0%) (log-rank test, P = 0.013). Conclusions Splenectomy does not benefit patients undergoing gastrectomy for gastric cancer. Routine splenectomy should be abandoned in patients undergoing radical resections for gastric cancer.


Background
Gastric cancer remains the most common cause of cancer-related deaths worldwide, with poor patient prognosis and a lack of adequate treatment methods. Although surgical resection is the primary treatment method for gastric cancer, more extensive surgery is accompanied by greater risks of surgery-related morbidity and mortality [1]. The mortality rate for patients undergoing gastrectomy surgery in western countries often exceeds 5%, and may be as high as 16% [2], although the rate in Japanese patients has been reported as being less than 2% [3]. Moreover, the efficacy of concomitant splenectomy remains unclear. As our understanding of the role of the spleen in the immunological defenses of the body and in the prevention of sepsis has increased, the value of routine removal of the spleen, en bloc with the stomach, during the course of radical, potentially curative resection for gastric cancer has been reassessed.
In this study, we retrospectively analyzed long-term survival data of patients who underwent gastrectomy, with or without splenectomy, for gastric cancer at a single institution. We found that operative complication rates were significantly higher in patients undergoing concomitant splenectomy, and it carried no benefit to patients already undergoing gastrectomy. Consequently, we believe, its routine use in radical resection for gastric cancer should be abandoned.

Patients
We identified 214 patients with histologically confirmed gastric cancer who had undergone radical surgery at the First Affiliated Hospital of China Medical University between 1980 and 2003, Patients were included in the study if they: 1) had histologically confirmed gastric cancer; 2) had undergone curative surgery; 3) had a complete medical record available; and 4) had never received neoadjuvant therapies. The number of patients at every period of diagnosis and the number treated by each surgeon were roughly equal.
All patients were followed up by mail or telephone interviews, and the final follow-up was in December 2008. Clinical, surgical, and pathological findings at the time of surgery and at each follow-up were collected and recorded in the database.

Ethics approval
The study protocol was approved by the Ethics Committee of China Medical University. [The patients signed the informed consent about operation and so on routinly, we don't need another consent about this study any more].

Surgical procedures and classifications of gastric cancer
All operations were performed at the First Affiliated Hospital of China Medical University. Surgical procedures and pathological assessments were standardized in accordance with the Japanese classification of gastric cancer [4]. All patients underwent standard total or distal sub-total gastrectomy, depending on the location and macroscopic appearance of the primary tumor.

Endpoints and follow-up
The primary endpoint was 5-year overall survival (OS) rate. OS was calculated from the date of surgery until the date of death or final follow-up contact. Patients remaining alive at the date of final follow-up were censored at that point. Patients were followed up every 6 months for the first 5 years after surgery, and every 12 months thereafter.

Statistical analyses
OS was analyzed in all eligible patients. Survival curves were determined by the Kaplan-Meier method, and compared by the log-rank test. Potential prognostic factors were entered into a Cox regression model. For univariate analyses, prognostic factors of interest and treatment group were deemed covariates in the Cox regression model. Subgroups were analyzed in this model to evaluate interactions between treatment and subgroup. All P-values were two-sided, with P < 0.05 considered significant. All statistical analyses were performed using SPSS software (version 16.0; SPSS Inc., Chicago, IL, USA).

Results
Of the 214 patients who underwent gastrectomy for gastric cancer from 1980 to 2003, 100 patients (median age 59 years) underwent gastrectomy plus splenectomy (G + S), while the remaining 114 (median age 55.5 years), underwent gastrectomy alone (G-A). All patients were followed up for at least 5 years, until December 19, 2008.
The baseline demographic and clinical characteristics of the two groups were similar (Table 1). Of the 214 patients, 13 (6.1%) had early gastric cancer (EGC) confined to the submucosa or mucosa. All patients underwent curative resection. Total gastrectomy was performed in 60 of the 100 (60%) patients who underwent G + S, and in 68 of the 114 (60%) who underwent G-A. Complications were more common (P = 0.005) in the G + S group than in the gastrectomy-alone group ( Table 1; Table 2). A total of 26 patients, 12% in each group, received adjuvant chemotherapy after surgery. The median follow-up durations were 18 months in the G + S group and 26.5 months in the G-A group. By the end of follow-up, 87 of the 100 patients (87%) who underwent G + S and 83 of the 114 (73%) who underwent G-A had died. Kaplan-Meier analysis of OS in these two groups showed a significant between-group difference ( Figure 1). The 5-year OS rates were 33.8% (95% CI 24.2 to 43.4%) in the G-A group and 28.8% (95% CI 19.6 to 38.0%) in the G + S group (p = 0.013 by the log-rank test).
The hazard ratio (HR) for death was 1.456 (95% CI 1.076 to 1.970; P = 0.015) in the G + S relative to the G-A group (Table 3). After adjustment for nine baseline variables (age, sex, tumor size, Borrmann type, T stage, lymph-node stage, TNM stage, complications and type of gastrectomy) using Cox regression analysis, the HR was 1.777 (95% CI 1.137 to 2.777; P = 0.012) ( Table 3). As expected, multivariate analysis showed that Borrmann type IV and advanced TNM stages were significantly associated with poor survival (Table 3).
G-A was significantly more beneficial than G + S in men, and in patients with tumor size >6 cm, Borrmann type III tumors, or those who underwent subt-otal gastrectomy (Table 4). We therefore would not recommend routine splenectomy in patients with gastric cancer who require sub-total gastrectomy, unless the tumor lies close to or directly invades the splenic hilum. Our results indicate that splenectomy should not be included in the treatment of patients with curable gastric cancer.

Discussion
The spleen is an organ that protects the host against infection and perhaps also against tumor micrometastases, and knowledge about its immunological functions has increased markedly in recent years. Although lymph nodes in the hiluma of the spleen may be affected by gastric tumors, the spleen itself is seldom affected. Our retrospective comparison showed that OS was significantly higher and complication rates significantly lower in patients undergoing G-A than in those undergoing G + S for curative resection of gastric cancer. More extended surgery is associated with increased risks of operative morbidity and mortality. We found that removal of the spleen during the course of resection for gastric carcinoma significantly increased operative morbidity rates compared with spleen preservation (24% versus 10%, P < 0.05). The increased morbidity after splenectomy may due to the higher incidence of infectious complications after gastrectomy. Mortality rates in western countries for patients undergoing gastrectomy often exceed 5% and may be as high as 16% [2]. Thus, G + S should not be considered the standard curative surgical procedure in patients with gastric cancer unless the addition of splenectomy is found to significantly benefit patients by reducing operative morbidity and mortality rates or improving long-term survival. In the current study, no evidence was found that removal of the spleen led to any increase in 5-year OS rate after potentially curative resection. We found that the 5-year OS rates were 28.8% in patients undergoing G + S and 33.8% in patients undergoing G-A, similar to previous findings [4][5][6][7]. Splenectomy was also shown to negatively influence survival, and in addition, increased length of hospital stay and probability of death [8][9][10].
In addition to the extent of the surgery, the operative skill and experience of the surgeon(s) and the workload of cases are also important factors for survival rates [11,12]. Many studies have reported a relationship between the number of cases treated in a hospital and the outcomes of cancer treatment [12][13][14][15][16][17]. Moreover, the uniformity of treatment is also important. Our study was carried out in a hospital that performs a high volume of dissections for gastric cancer, with low morbidity and mortality rates. All participating surgeons were from the same department, which minimizes the variation in individual operating skill and management, and balances the comparisons between the two groups without bias from the skill of individual surgeons.
Because G-A is associated with lower mortality and adequate PS when performed in selected institutions with sufficient surgical experience and good postoperative management, we recommend that most patients with curable gastric cancer should undergo total or subtotal gastrectomy combined with radical lymphadenectomy, a type of surgery compatible with the preservation of the spleen.

Conclusions
Splenectomy does not benefit patients undergoing radical resections for gastric cancer, and should not be performed.