Predictive value of prognostic nutritional index on infection after radical gastrectomy: a retrospective study

Introduction

Gastric cancer (GC) is one of the most common and deadly malignancies worldwide (1,2). In China, the high incidence of GC has become a serious public health concern, with the country accounting for about half of all new cases in the world (3,4). Currently, radical surgery remains the gold standard treatment for GC, and relatively complete resection of the tumor can result in good outcomes. However, most patients experience postoperative complications due to their old age, advanced disease stage, declining physical function, and poor immunological status and surgical tolerance. Infection, the most common complication after radical gastrectomy, can lead to prolonged hospital stays, increased risk of intensive care unit (ICU) admission/rehospitalization, and a higher mortality rate (5-7).

Previous studies have shown that the incidence of infection after radical gastrectomy is 11.3–15.5%, with higher rates in the elderly and in immunocompromised patients (8,9). In a retrospective analysis, Xiao et al. (10) found that abdominal infection was the main postoperative complication (with an incidence rate of 10.2%) of patients undergoing radical surgery for GC, followed by pulmonary infection. Postoperative infection remains a challenge despite proactive preventive measures. Therefore, it is important to identify the risk factors and implement early interventions accordingly to minimize the occurrence of postoperative infection.

In recent years, nutrition- and inflammation-related indicators have been used to predict surgical risk and postoperative complications. One of these is the prognostic nutritional index (PNI), a simple tool calculated by combining the total peripheral blood lymphocyte count with the serum albumin (ALB) concentration. PNI can be used to assess the nutritional level and immune status of a patient and predict postoperative complications and survival time (11,12). Several clinical studies have found a close relationship between PNI and the prognosis of cancers, including gastric, colorectal, and pancreatic cancers (13,14). Shi et al. (15) further revealed that PNI score is an important factor in the occurrence and development of post-surgical infections. Another recent study showed that lower PNI is associated with increased susceptibility to postoperative infection of prostate cancer. Patients with severe malnutrition had more postoperative complications compared to well-nourished patients (16). However, few studies have investigated the relationship between PNI and infection after radical surgery for GC, and most of the available studies have small sample sizes and notable biases in study design, making it difficult to provide high-quality evidence for clinical practice.

In our current study, we explored the role of preoperative PNI in predicting the occurrence of infection after radical surgery for GC to inform the development of targeted measures for the further prevention and control of postoperative infections. We present the following article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-192/rc).

Methods

Study design

This is a single-center, retrospective, observational, and cohort study based on the hospital information system. It was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the ethics committee of Minhang Hospital, Fudan University (2021-012-01K). Informed consent was taken from all the patients.

Subjects

The demographic and clinicopathological data of patients who underwent radical surgery for GC in our center from December 2010 to December 2020 were retrieved from the hospital information system. The inclusion criteria were as follows: (I) all patients were diagnosed with primary GC by gastric mucosal histopathology; (II) patients were diagnosed with clinical TNM stage I–III, which met the indications for radical surgery; (III) patients were aged >18 years; and (IV) patients were assigned an American Society of Anesthesiologists (ASA) score of ≤III. The exclusion criteria were as follows: (I) remnant GC; (II) cardiac, cerebral, pulmonary, and/or other vital organ dysfunction; (III) other malignant tumors, hematological diseases, and/or autoimmune diseases; (IV) preoperatively confirmed acute and chronic infectious disease(s) that were still under treatment; (V) use of immunosuppressive drugs or hormones within 1 month before enrollment; (VI) a history of laparotomy or other major surgery within 1 month before enrollment; and (VII) loss to follow-up, perioperative deaths, or incomplete clinical data. A total of 1,111 postoperative inpatients were finally included and grouped according to their PNI scores (Figure 1). Pathological diagnosis and staging of the tumors were performed for each patient.

Figure 1 Flow chart of subject enrollment. PNI, prognostic nutritional index.

Radical surgery for GC

All subjects underwent laparoscopic or open surgery according to their condition after admission, and all operations were performed by the same surgical team. In the laparoscopic surgery group, patients were placed in a supine position after induction of general anesthesia and endotracheal intubation. A 1 cm incision was made at the lower edge of the umbilicus. After CO₂ pneumoperitoneum was established, the corresponding trocar was placed, and the abdominal cavity was explored to locate the tumor. In the open surgery group, a 5–8 cm incision was made in the middle of the abdomen for routine exploration, mobilization, and lymph node dissection. Depending on the location of the tumors, the surgical approaches included total gastrectomy, distal gastrectomy, and proximal gastrectomy, along with D2/D2 plus superior mesenteric lymph node dissection. Combined multi-organ resection was performed for patients with tumors invading adjacent organs. For total gastrectomy, the lymph nodes stations dissected were 1, 2, 3, 4d, 4sa, 4sb, 5–10, 11p, 12a, and 14v. For proximal gastrectomy, the lymph nodes stations dissected were 1–4, 7, 8a, 9, 10, and 11, with the preservation of vascular arcades on the lesser and greater curvatures. For distal gastrectomy, the lymph node stations dissected were 1, 3, 4d, 4sb, 5–9, 11p, 12a, and 14v. For proximal gastrectomy, end-to-side anastomosis between the anterior wall of the remnant stomach and the esophageal stump was performed. For distal gastrectomy, Billroth I or II GI reconstruction was performed. For total gastrectomy, Roux-en-Y GI reconstruction was performed. The abdominal cavity was washed using distilled water, and the tumor condition was examined. Antibiotics were applied perioperatively to prevent infection. Postoperatively, a small number of patients with advanced GC were given adjuvant chemotherapy with capecitabine plus oxaliplatin or with S-1 plus oxaliplatin according to the National Comprehensive Cancer Network (NCCN) guidelines.

Diagnostic criteria

Postoperative infections at various sites in patients undergoing radical surgery for GC were assessed according to the Diagnostic Criteria for Hospital-Acquired Infections (Trial) formulated by China’s Ministry of Health in 2001. The responsible physician monitored and recorded the patient’s body temperature, peripheral blood leukocyte count, and clinical signs. Intra-abdominal infection was defined as follows: (I) the presence of fever, abdominal pain, abdominal distension, or obvious signs of peritonitis, with an elevated neutrophil to lymphocyte ratio (NLR); (II) culture positive abdominal fluid (drainage fluid or puncture fluid); or (III) radiologically confirmed intra-abdominal infectious lesions (e.g., abscesses and purulent exudates). A diagnosis of pulmonary infection was made if the patient experienced persistent elevated body temperature (>38.0 ℃) for more than 24 hours after surgery, which might be accompanied by cough and sputum production, with or without positive sputum bacterial culture, and/or the presence of new infiltrates, solid changes, and other signs on chest images. Incisional infection was defined as postoperative infection involving the skin and subcutaneous tissue at the surgical site and featuring redness, swelling, heat, pain, and/or purulent discharge. Lower urinary tract infection was defined as cystitis and urethritis in postoperative patients, with pus and/or hematuria on routine urinalysis and pathogenic bacteria on urine culture. Catheter-associated infections were identified based on the patient’s clinical presentation (bacteremia, with hyperthermia and/or chills) and a positive tip culture. In all patients with infections, specimens of pathogenic bacteria were collected for isolation and culture of the pathogen, and antibiotics were administered.

Information collection

The data of patients who met the inclusion criteria were retrieved through the hospital information system. Baseline demographic information, pathologic features of tumors, preoperative laboratory measurements, and surgery-related indicators were collected by reviewing the patients’ medical records. The baseline demographic data included age, gender, body mass index (BMI), ASA score, history of abdominal surgery, chronic diseases, and smoking status. The pathologic features of tumors included histologic type, stage, location, and size. Surgery-related indicators consisted of surgical modality, operation time, intraoperative blood loss, blood transfusion, and neoadjuvant chemotherapy. The tumor stage and disease grade were classified according to the 7th edition of the TNM classification of the International Union Against Cancer (UICC), while gross staging and surgical approach and scope were based on the 3rd edition of the Japanese Gastric Cancer Treatment Guidelines.

The PNI was calculated as 10 × serum ALB value (g/dL) + 0.005 × peripheral lymphocyte count (per mm³). Based on a previous study, a cutoff value of 45 was used to divide patients into low and high PNI groups (17).

PSM

PSM was performed to balance the distribution of baseline characteristics. A logistic regression model was established with high/low PNI score as the dependent variables and with age, gender, BMI, ASA score, smoking status, neoadjuvant chemotherapy, tumor stage, tumor differentiation, tumor location, laboratory measurements, tumor size, surgical modality, operation time, and blood loss as independent variables. A 1:1 matching was performed according to the principle of proximity matching (caliper value: 0.05), without alternatives. The differences in each indicator between the high PNI group and low PNI group before and after matching were compared.

Statistical analysis

Statistical analysis was performed using the SPSS 22.0 software package (IBM Corp., Armonk, NY, USA). The normally distributed measurement data [presented as mean ± standard deviation (x¯±SD)], and the non-normally distributed measurement data [presented as median (interquartile spacing)] were compared using independent sample t-test or Mann-Whitney U test. The categorical variables were compared using chi-square test or Fisher’s exact probability test and are expressed as number of cases (%). Propensity score matching was performed to adjust for differences in baseline characteristics between high PNI and low PNI groups. The resulting score-matched pairs were analyzed subsequently. Univariate analysis was performed to evaluate potential factors for postoperative infection; subsequently, multivariate Logistic regression analysis was performed to obtain the odds ratio (OR) and 95% confidence interval (95% CI) for each factor. All P values reported are two-tailed. A P value of <0.05 was set as the threshold for statistical significance.

Results

Clinicopathological features before matching

A total of 1,111 patients [693 males (62.4%) and 418 females (37.6%), with an average age of 63.8 years] were included in this study. Most patients had ASA scores of 2 (74.3%) and stage I–II disease (72.9%). Laparoscopic surgery was performed in 59.6% of patients, while open surgery was performed in 40.4% of patients. Total gastrectomy was performed in 257 patients, partial gastrectomy in 775 patients, and combined multi-organ resection in 79 patients (including splenectomy in 33 cases, transverse colectomy in 18 cases, partial hepatectomy in 12 cases, and partial pancreatectomy in 16 cases). The mean operation time was 248.4±50.2 minutes, and the average intra-operative blood loss was 210.4±132.1 mL.

The patients were divided into a low PNI group (n=451) and a high PNI group (n=660), with the mean PNI value of patients in the low PNI group significantly lower than that of patients in the high PNI group (39.4±4.5 vs. 49.8±4.6). The clinicopathological features of these 2 groups before matching are summarized in Table 1. Differences in gender, number of smokers, previous surgical history, number of patients receiving neoadjuvant chemotherapy, and tumor location were not statistically significant between the 2 groups (all P>0.05), whereas differences in mean age, BMI, ASA score, tumor grade, tumor differentiation, preoperative laboratory measurements, tumor size, surgical modality, type of resection, mean operation time, and mean intraoperative blood loss were statistically significant (all P<0.05).

Clinicopathological features after matching

After PSM, the mean age, BMI, ASA score, tumor grade, tumor differentiation, tumor size, surgical modality, type of resection, mean operation time, and mean intraoperative blood loss were not significantly different between the low PNI group and the high PNI group (all P>0.05), although the ALB level, hemoglobin level, and lymphocyte count were still significantly lower in the low PNI group than in the high PNI group (all P<0.05). The mean PNI value was 50.4±3.7 in the high PNI group after matching, which was higher than that in the low PNI group. The clinicopathological features after matching are listed in Table 2.

Distribution of infections and detection of pathogenic bacteria after matching

Among the 812 patients, 101 experienced infections after surgery (Table 3), including intra-abdominal infections (n=46), pulmonary infections (n=27), and surgical site infections (n=25). In addition, 1 patient had a urinary tract infection, and 2 had catheter-related infections. The incidence of infection was 15.0% in the low PNI group, which was significantly higher than that in the high PNI group (P=0.026). The number of patients with noninfectious complications was also significantly higher in the low PNI group (P=0.029), although there was no statistical difference in the composition ratio of infected cases (P>0.05).

Microbiological specimens were collected for culture in 92 cases (91.1%), among which 77 (83.7%) were culture positive. A total of 82 pathogenic strains were identified, including 56 strains of gram-negative bacteria (68.3%), most of which were Escherichia coli; 23 strains of gram-positive bacteria (28.0%), most of which were Enterococcus faecalis; and 3 strains of fungi (3.7%), all of which were Candida albicans (Table 4).

Risk factors for infection in patients after radical surgery for GC

In the univariate analysis (Table 5), age ≥65 years (P=0.037), BMI ≥25 kg/m² (P=0.031), smoking (P=0.048), diabetes mellitus (P=0.023), low PNI value (P=0.026), combined multi-organ resection (P=0.017), operation time ≥240 minutes (P=0.009), blood loss ≥200 mL (P=0.011), and perioperative blood transfusion (P=0.021) were associated with the occurrence of infection after radical surgery for GC. Further multivariate logistic regression analysis (Table 6) showed that BMI ≥25 kg/m² (P=0.004), diabetes mellitus (P=0.042), PNI value <45 (P=0.037), combined multi-organ resection (P<0.001), operation time ≥240 minutes (P=0.023), and perioperative blood transfusion (P=0.025) were risk factors for infection after radical surgery for GC.

Discussion

Radical gastrectomy is the mainstay of curative treatment for GC. Most patients with early-stage GC and 40–50% of patients with advanced GC can be treated with radical resection. However, radical resection for GC is a complicated and time-consuming procedure with a high rate of postoperative infection. Postoperative infection poses a significant financial burden on patients and affects postoperative recovery; therefore it is clinically important to identify its risk factors (18). Many recent studies have demonstrated that the nutritional and immunological status of the human body is closely related to postoperative infection in a variety of diseases, and the value of nutritional assessment in oncology patients has increasingly been recognized (19,20). Many nutritional screening/assessment tools are available, including the Nutritional Risk Screen 2002, the Mini-Nutritional Assessment, and the Subjective Global Assessment. However, these tools are often cumbersome and susceptible to subjective factors, which may lead to less accurate results (21).

PNI is an objective nutritional assessment tool that is easily calculated based on the serum ALB concentration and the peripheral blood lymphocyte count. Serum ALB, which is synthesized by the liver, maintains colloid osmotic pressure and nutritional metabolism, and its level reflects the nutritional status of the body (22). Peripheral blood lymphocytes are involved in the destruction and apoptosis of tumor cells and thus constitute a critical component of anti-tumor immunity. A lower lymphocyte count suggests a decrease in the body’s anti-tumor immune function (23). Early studies have found that PNI value has potential as a prognostic marker of colorectal cancer, as patients with low PNI values tend to have poor long-term prognosis (14-16) Sakurai et al. (24) showed that PNI was useful in predicting prognosis and immunological and nutritional status in patients with GC, and that a PNI of <45 was predictive of anastomotic edema and peritoneal effusion. Although many studies have evaluated the role of PNI in patients with gastrointestinal (GI) cancer, most have focused on the prognostic value of PNI, with few studies analyzing the impact of PNI value on postoperative infection. In addition, baseline clinical features often differ among patients with different PNI scores. Thus, PSM analysis was adopted in this study to avoid potential confounding variables. Based on dimensionality reduction algorithms, PSM incorporates the information provided by multiple confounding factors into 1 propensity score and then stratifies, matches, or weights individuals from different comparison groups based on the propensity score to improve the balance of variables among nonrandomized groups. This method has been widely used in observational and nonrandomized studies. In our study, there were significant differences in the mean age, BMI, ASA score, tumor grade, tumor differentiation, surgical modality, and type of resection between the high and low PNI groups before matching, making a comparison of infection rates unfeasible. After matching, there were no significant differences in preoperative baseline data between these 2 groups (except for ALB level, hemoglobin level, and lymphocyte count), thus improving comparability.

Malnutrition leads to prolonged wound healing and altered immune function, both of which are associated with the occurrence of postoperative infections. In our study, postoperative infections occurred in 101 of 812 patients with GC. The most common infections were abdominal infection, pulmonary infection, surgical site infection, and urinary tract infection, which was consistent with the distribution of postoperative infection sites in patients with GC reported by Cai et al. (25). After one-to-one matching, we found that the incidence of infection was significantly higher in the low PNI group than in the high PNI group, suggesting that PNI can identify patients at high risk for postoperative infection. Our study also showed that the pathogenic bacteria were predominantly gram-negative bacteria, among which Escherichia coli was the most common pathogen, which may be related to the opening of the GI tract during the surgery. In a study of a large cohort of patients undergoing radical treatment for GC in China, the most common pathogens of abdominal infection after radical surgery were Escherichia coli, Klebsiella pneumoniae, Enterococci, and Candida albicans (26). This was similar to our findings. Many studies suggest that infections after radical surgery for GC are often caused by multidrug-resistant strains (2). Therefore, antimicrobial drugs must be applied rationally to reduce the occurrence of infection in clinical settings.

Infections after radical surgery for GC can be caused by multiple risk factors, which vary somewhat among different studies depending on the disease characteristics, surgical modality, and underlying conditions. The reported risk factors include diabetes mellitus, long operation time, perioperative blood transfusion, intraoperative blood loss, combined multi-organ resection, and high BMI (27-30). The univariate analysis in our current study showed that age ≥65 years, BMI ≥25 kg/m², diabetes mellitus, low PNI value, combined multi-organ resection, operation time ≥240 minutes, blood loss ≥200 mL, and perioperative blood transfusion were risk factors. Further multivariate analyses identified BMI ≥25 kg/m², diabetes mellitus, PNI score <45, combined multi-organ resection, operation time ≥240 minutes, and intra-operative blood loss as risk factors. The risk of postoperative infection was significantly higher in GC patients with a BMI of ≥25 kg/m² compared to that in patients with a normal BMI, probably due to the poor surgical exposure and long operation time in patients with obesity, as well as the high incidence of chronic diseases like diabetes and hypertension in such patients (31). According to Wang et al. (32), diabetes mellitus is an important risk factor for abdominal infection after radical surgery for GC because diabetic patients have more hyperactive catabolism and impaired protein synthesis, which results in poor healing of anastomosis and incision and thus high incidence rates of complications such as incisional infection and pulmonary infection. The development of surgical techniques has lowered the proportion of patients requiring blood transfusion during surgery, although some patients still require a blood transfusion to improve their anemic condition. However, as shown by Xu et al. (33), allogeneic blood transfusion mediates immunosuppression and impairs immune function, leading to the development of postoperative inflammatory reactions. Low PNI is a risk factor for infection after radical surgery for GC, which can be attributed to the following two reasons. First, reduced serum ALB level and decreased or insufficient activities of antibody synthetases contribute to decreased immunity and a high rate of infections (34,35). Second, lymphocytes are the prime movers of the immune system, and a low lymphocyte count indicates an inadequate immune response to a tumor. Therefore, based on the PNI score, clinicians may assess the risk of infection after radical surgery for GC, develop an intervention plan, and monitor the implementation of the plan. Patients with a PNI score of <45 can be actively treated with nutritional and immune interventions to reduce the incidence of postoperative infection.

Our current study has some limitations, including a retrospective, single-center design and possible selection bias. The negative impact of low preoperative PNI score on infection needs to be further validated in prospective studies.

In conclusion, infection is the most common complication after radical surgery for GC, with gram-negative bacteria being the most common pathogens. The rate of postoperative infection is higher in patients with low PNI values, and PNI <45 is a risk factor for postoperative infection. Therefore, appropriate nutritional interventions can be given for patients with low-PNI, thus reducing the incidence of postoperative infection.

Acknowledgments

Funding: This study is supported by Shanghai Minhang District Natural Science Research Project (2020MHZ040).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-192/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-192/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-192/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the ethics committee of Minhang Hospital, Fudan University (2021-012-01K). Informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Guggenheim DE, Shah MA. Gastric cancer epidemiology and risk factors. J Surg Oncol 2013;107:230-6. [Crossref] [PubMed]
2. Karimi P, Islami F, Anandasabapathy S, et al. Gastric cancer: descriptive epidemiology, risk factors, screening, and prevention. Cancer Epidemiol Biomarkers Prev 2014;23:700-13. [Crossref] [PubMed]
3. Jin G, Lv J, Yang M, et al. Genetic risk, incident gastric cancer, and healthy lifestyle: a meta-analysis of genome-wide association studies and prospective cohort study. Lancet Oncol 2020;21:1378-86. [Crossref] [PubMed]
4. Strong VE, Wu AW, Selby LV, et al. Differences in gastric cancer survival between the U.S. and China. J Surg Oncol 2015;112:31-7. [Crossref] [PubMed]
5. Ang TL, Fock KM. Clinical epidemiology of gastric cancer. Singapore Med J 2014;55:621-8. [Crossref] [PubMed]
6. Cheng Y, Zhang J, Zhang L, et al. Enteral immunonutrition versus enteral nutrition for gastric cancer patients undergoing a total gastrectomy: a systematic review and meta-analysis. BMC Gastroenterol 2018;18:11. [Crossref] [PubMed]
7. Xiao H, Zhou H, Liu K, et al. Development and validation of a prognostic nomogram for predicting post-operative pulmonary infection in gastric cancer patients following radical gastrectomy. Sci Rep 2019;9:14587. [Crossref] [PubMed]
8. Meta-analysis A. J Surg Res 2019;243:180-8. Dan Z, YiNan D, ZengXi Y, et al Thirty-Day Readmission After Radical Gastrectomy for Gastric Cancer:. [Crossref] [PubMed]
9. Xiao H, Wang Y, Quan H, et al. Incidence, Causes and Risk Factors for 30-Day Unplanned Reoperation After Gastrectomy for Gastric Cancer: Experience of a High-Volume Center. Gastroenterology Res 2018;11:213-20. [Crossref] [PubMed]
10. Xiao H, Quan H, Pan S, et al. Impact of peri-operative blood transfusion on post-operative infections after radical gastrectomy for gastric cancer: a propensity score matching analysis focusing on the timing, amount of transfusion and role of leukocyte depletion. J Cancer Res Clin Oncol 2018;144:1143-54. [Crossref] [PubMed]
11. Okadome K, Baba Y, Yagi T, et al. Prognostic Nutritional Index, Tumor-infiltrating Lymphocytes, and Prognosis in Patients with Esophageal Cancer. Ann Surg 2020;271:693-700. [Crossref] [PubMed]
12. Wang Z, Wang Y, Zhang X, et al. Pretreatment prognostic nutritional index as a prognostic factor in lung cancer: Review and meta-analysis. Clin Chim Acta 2018;486:303-10. [Crossref] [PubMed]
13. Yasar HA, Bir Yucel K, Arslan C, et al. The relationship between prognostic nutritional index and treatment response in patients with metastatic renal cell cancer. J Oncol Pharm Pract 2020;26:1110-6. [Crossref] [PubMed]
14. Tominaga T, Nagasaki T, Akiyoshi T, et al. Prognostic nutritional index and postoperative outcomes in patients with colon cancer after laparoscopic surgery. Surg Today 2020;50:1633-43. [Crossref] [PubMed]
15. Shi WK, Zhang XH, Zhang J, et al. Predictive ability of prognostic nutritional index in surgically resected gastrointestinal stromal tumors: a propensity score matching analysis. Jpn J Clin Oncol 2019;49:823-31. [Crossref] [PubMed]
16. Xiao H, Xiao Y, Chen P, et al. Association Among Blood Transfusion, Postoperative Infectious Complications, and Cancer-Specific Survival in Patients with Stage II/III Gastric Cancer After Radical Gastrectomy: Emphasizing Benefit from Adjuvant Chemotherapy. Ann Surg Oncol 2021;28:2394-404. [Crossref] [PubMed]
17. Sakurai K, Ohira M, Tamura T, et al. Predictive Potential of Preoperative Nutritional Status in Long-Term Outcome Projections for Patients with Gastric Cancer. Ann Surg Oncol 2016;23:525-33. [Crossref] [PubMed]
18. Shibasaki S, Suda K, Nakauchi M, et al. Non-robotic minimally invasive gastrectomy as an independent risk factor for postoperative intra-abdominal infectious complications: A single-center, retrospective and propensity score-matched analysis. World J Gastroenterol 2020;26:1172-84. [Crossref] [PubMed]
19. Arends J, Bachmann P, Baracos V, et al. ESPEN guidelines on nutrition in cancer patients. Clin Nutr 2017;36:11-48. [Crossref] [PubMed]
20. Sato R, Oikawa M, Kakita T, et al. The prognostic value of the prognostic nutritional index and inflammation-based markers in obstructive colorectal cancer. Surg Today 2020;50:1272-81. [Crossref] [PubMed]
21. Mao YS, Hao SJ, Zou CF, et al. Controlling Nutritional Status score is superior to Prognostic Nutritional Index score in predicting survival and complications in pancreatic ductal adenocarcinoma: a Chinese propensity score matching study. Br J Nutr 2020;124:1190-7. [Crossref] [PubMed]
22. Cheng YL, Sung SH, Cheng HM, et al. Prognostic Nutritional Index and the Risk of Mortality in Patients With Acute Heart Failure. J Am Heart Assoc 2017;6:004876. [Crossref] [PubMed]
23. Zhang H, Tao Y, Wang Z, et al. Evaluation of nutritional status and prognostic impact assessed by the prognostic nutritional index in children with chronic kidney disease. Medicine (Baltimore) 2019;98:e16713. [Crossref] [PubMed]
24. Sakurai K, Tamura T, Toyokawa T, et al. Low Preoperative Prognostic Nutritional Index Predicts Poor Survival Post-gastrectomy in Elderly Patients with Gastric Cancer. Ann Surg Oncol 2016;23:3669-76. [Crossref] [PubMed]
25. Cai J, Wei D, Gao CF, et al. A prospective randomized study comparing open versus laparoscopy-assisted D2 radical gastrectomy in advanced gastric cancer. Dig Surg 2011;28:331-7. [Crossref] [PubMed]
26. Xiao H, Xiao Y, Quan H, et al. Intra-abdominal infection after radical gastrectomy for gastric cancer: Incidence, pathogens, risk factors and outcomes. Int J Surg 2017;48:195-200. [Crossref] [PubMed]
27. Lian B, Chen J, Li Z, et al. Risk Factors and Clavien-Dindo Classification of Postoperative Complications After Laparoscopic and Open Gastrectomies for Gastric Cancer: A Single-Center, Large Sample, Retrospective Cohort Study. Cancer Manag Res 2020;12:12029-39. [Crossref] [PubMed]
28. Li L, Liu P, Wang J, et al. Clinicopathologic Characteristics and Risk Factors of Lymph Node Metastasis in Patients with Early Gastric Cancer in the Wannan Region. Med Sci Monit 2020;26:e923525. [Crossref] [PubMed]
29. Chen L, Yue C, Li G, et al. Clinicopathological features and risk factors analysis of lymph node metastasis and long-term prognosis in patients with synchronous multiple gastric cancer. World J Surg Oncol 2021;19:20. [Crossref] [PubMed]
30. Filip B, Toma C, Buna-Arvinte M, et al. Impact of Surgery and Early Postoperative Outcomes After Radical Gastrectomy for Cancer. Chirurgia (Bucur) 2018;113:478-85. [Crossref] [PubMed]
31. Liu M, Xing J, Arslan A, et al. Safety and efficacy of laparoscopic gastrectomy in obese patients with gastric cancer. Medicine (Baltimore) 2019;98:e17991. [Crossref] [PubMed]
32. Wang J, Zhao J, Zhang Y, et al. Early enteral nutrition and total parenteral nutrition on the nutritional status and blood glucose in patients with gastric cancer complicated with diabetes mellitus after radical gastrectomy. Exp Ther Med 2018;16:321-7. [Crossref] [PubMed]
33. Xu D, Fang X, Li Y, et al. Perioperative blood transfusion is one of the factors that affect the prognosis of gastric cancer. J BUON 2018;23:672-7. [PubMed]
34. Kubota T, Shoda K, Konishi H, et al. Nutrition update in gastric cancer surgery. Ann Gastroenterol Surg 2020;4:360-8. [Crossref] [PubMed]
35. Zhang Y, Zhu JY, Zhou LN, et al. Predicting the Prognosis of Gastric Cancer by Albumin/Globulin Ratio and the Prognostic Nutritional Index. Nutr Cancer 2020;72:635-44. [Crossref] [PubMed]

(English Language Editor: C. Gourlay)