The association between body mass index (BMI) and clinical outcomes remains unclear among patients with resectable gastric cancer.

To investigate the relationship between BMI and long-term survival of gastric cancer patients.

This retrospective study included 2526 patients who underwent radical gastrectomy for gastric cancer between September 2013 and June 2018. The patients were divided into four groups: Group A (low BMI, < 18.5 kg/m²), group B (normal BMI, 18.5-24.9 kg/m²), group C (overweight, 25-29.9 kg/m²), and group D (obese, ≥ 30 kg/m²). Clinicopathological findings and survival outcomes were recorded and analyzed.

Preoperative weight loss was more common in the low-BMI group, while diabetes was more common in the obese group. Upper-third gastric cancer accounted for a large proportion of cases in the higher BMI groups. Major perioperative complications tended to increase with BMI. The 5-year overall survival rates were 66.4% for group A, 75.0% for group B, 77.1% for group C, and 78.6% for group D. The 5-year overall survival rate was significantly lower in group A than in group C (P = 0.008) or group D (P = 0.031). Relative to a normal BMI value, a BMI of < 18.5 kg/m² was associated with poor survival (hazard ratio: 1.558, 95% confidence interval: 1.125-2.158, P = 0.008).

Low BMI, but not high BMI, independently predicted poor survival in patients with resectable gastric cancer.

Gastric cancer is the fifth most common cancer and the third leading cause of cancer-related death[1]. Gastric cancer patients often experience malnutrition, which may lead to obvious weight loss before surgery[2], particularly at advanced stages of the disease[3]. However, obesity is becoming increasingly common in both Western and Eastern countries[4,5]. Some studies have examined the association between body mass index (BMI) and the prognosis of gastric cancer[6-15], although the long-term outcomes remain unclear for patients with different BMIs.

Low preoperative BMI is associated with poor long-term outcomes among patients with gastric cancer[6,7], which may highlight the clinical importance of preoperative weight loss. However, some studies have claimed that BMI is not a risk factor for poor survival[8,9], while others have suggested that overweight/obese gastric cancer patients have a higher risk of postoperative complications and experience poorer outcom-es[10,11]. Moreover, different studies have indicated that a high BMI is not associated with an increased risk of perioperative complications[12,13], and that patients with gastric cancer and a high BMI have comparable or better long-term outcomes, relative to individuals with a normal BMI[14,15]. Considering the discrepancies in these findings, this retrospective study aimed to clarify the relationship between preoperative BMI and long-term prognosis among patients with resectable gastric cancer.

Obesity is associated with cancer mortality, and the malnutrition status of cancer patients may also be related to their long-term prognosis. However, the relationships between specific BMI groupings and cancer survival are less clear. The results of the present study suggest that low BMI independently predicted poor long-term survival, while high BMI was associated with major perioperative complications but not long-term survival.

Our group with high BMI had an increased proportion of diabetes, upper-third gastric cancer, and perioperative major complications. Similarly, a meta-analysis suggested that high BMI was associated with an increased incidence of cardiac carcinoma. In this context, obese patients are more likely to have esophagogastric junction disruption or an augmented gastroesophageal pressure gradient, which could induce reflux[19]. In addition, gastroesophageal reflux disease is strongly associated with cardiac or esophageal adenocarcinoma[20]. However, this distribution pattern was not observed in other studies[6,12].

Several studies have indicated that high BMI is associated with increased risks of major complications and perioperative mortality[9,20,21], which is consistent with our findings. Obese patients typically have poor surgical field visibility, as well as an increased possibility of oozing, which can complicate the dissection of lymph nodes and formation of an anastomosis. However, high BMI was not associated with an increased risk of perioperative complications in some medical centers[8,22-24], which could be related to high volumes and experienced surgeons reducing the risks associated with radical gastrectomy in obese patients[12]. In our study, fewer eLNs were retrieved in the higher BMI groups (groups C and D), although the only significant difference was observed between groups B and C, and these two groups had similar 5-year OS outcomes. Dhar et al[10] reported that a higher BMI was associated with a higher risk of local recurrence and shorter recurrence-free survival, which they attributed to the difficulty in achieving adequate lymphadenectomy in obese patients with gastric cancer. In addition, it can be challenging to determine the actual number of lymph node dissections in obese patients, considering the difficulty involved in isolating lymph nodes within the abundant intra-abdominal fat. Nevertheless, as indicated above, it is possible that experienced surgeons in high-volume centers might be able to perform appropriate radical lymphadenectomy and achieve favorable oncological outcomes even for obese patients. In our study, the median number of eLNs was similar for groups B and C, and the difference might be attributable to the difference in the range, rather than a difference in the median value. Nevertheless, we observed that an inadequate number of eLNs (< 30) was an independent predictor of poor long-term outcomes, which is consistent with the increasing number of studies[25-27] that suggest that a higher number of retrieved LNs is associated with improved long-term outcomes. The 8^th edition of the AJCC TNM cancer staging guidelines[17] recommend that a minimum of 16 LNs, but preferably ≥ 30 LNs, be assessed during gastric cancer surgery. In addition, Deng et al[28,29] demonstrated that an insufficient number of eLNs may be a risk factor for postoperative recurrence in patients with LN-negative gastric cancer. We emphasize the importance of standard radical gastrectomy procedures and a sufficient number of eLNs for obese patients to achieve favorable short-term and oncological outcomes. Less experienced surgeons should perform D2 lymphadenectomy for obese patients with gastric cancer until they have sufficient experience.

In our study, low BMI (< 18.5 kg/m²) was an independent predictor of poor 5-year OS, although comparable 5-year OS rates were observed between the low and normal BMI groups. Patients with advanced gastric cancer are more likely to experience preoperative weight loss and malnutrition, which is associated with decreased survival. Moreover, gastrectomy leads to postoperative weight loss[15], and it is possible that overweight/obese patients would reach a more appropriate body weight after surgery, which could improve their long-term prognosis. Our multivariate analysis revealed that high BMI was not associated with poor survival, and a similar result was observed in a retrospective study of 427 Japanese patients with gastric cancer by Wada et al[7]. In the present study, stratified analyses revealed a difference in the 5-year OS rates when we compared the low-BMI and overweight groups of patients with pTNM stage II–III disease. Nevertheless, the 5-year OS rates were comparable between the four BMI-based groups of patients with pTNM stage I gastric cancer. The discrepancies between the results of the overall and stratified analyses may be related to the limited numbers of patients in groups A and D, which might have biased the stratified analysis.

Previous studies have supported different conclusions regarding the association between BMI and long-term prognosis[6-15]. Our results suggest that patients with obese and normal BMI have comparable 5-year OS rates, while conflicting results have been reported[11,13]. We speculate that these differences might be related to the specific group divisions, as only two BMI-based groups (cut-off: 25 kg/m²) were used in the studies by Lianos et al[11] and Shimada et al[13], who concluded that high BMI independently predicted a poor prognosis. Three BMI-based groups (< 18.5 kg/m², 18.5-24.9 kg/m², and ≥ 25 kg/m²) were used in a Chinese and a Japanese study[7,9], which revealed that low BMI, rather than high BMI, was associated with poor survival, similar to our findings. The main difference between these studies was that a BMI of < 25 kg/m² was analyzed either as a single group or by separating into two groups (< 18.5 kg/m² and 18.5-24.9 kg/m²). Furthermore, Shimada et al[13] reported average BMI values of 26.4 ± 2 kg/m² in the obese group and 22 ± 2.2 kg/m² in the normal-BMI group, which are clearly different from the values in our study. The differences in average BMI values might explain the conflicting conclusions. Nevertheless, there has been a recent shift toward the general opinion that high BMI does not affect long-term outcomes among gastric cancer patients, particularly those who are treated in high-volume medical centers[12].

The present study has several limitations. First, we did not analyze data regarding adjuvant chemotherapy use, which might have influenced the patients’ outcomes. Second, the surgical approach was not considered in this study, although surgical procedures can influence postoperative nutritional status and long-term outcomes, especially after total or near-total gastrectomy[30-32]. Third, the stratified analysis might have been biased based on the small numbers of patients in groups A and D.

A low BMI (< 18.5 kg/m²) was an independent risk factor for poor long-term survival after radical gastrectomy for gastric cancer. However, a high BMI was not a risk factor for poor survival in this setting.