Abstract
Background/Aim: Although the global leadership initiative on malnutrition (GLIM) criteria for malnutrition were published in 2018, few studies have correlated them with postoperative outcomes of gastric cancer. This study aimed to investigate the validity of predicting prognosis after gastrectomy in patients with advanced gastric cancer by dividing the patients according to the severity of malnutrition using the body mass index (BMI) and body weight loss (BWL), as defined by the GLIM criteria. Patients and Methods: This study included patients with c-stage II-III primary gastric cancer who underwent gastrectomy between April 2008 and June 2018. Patients were divided into normal, moderate malnutrition, and severe malnutrition groups according to the GLIM criteria of BMI and BWL. The primary endpoint was overall survival (OS). Results: Of the 512 patients, 340 (66.4%) were included in the normal group, 84 (16.4%) in the moderate group, and 88 (17.2%) in the severe group. Comparing the long-term prognosis in the three groups, the prognosis was worse in the moderate and severe groups than in the normal group for OS (p<0.001). Multivariate analysis for OS showed that moderate malnutrition [hazard ratio (HR)=1.689, 95% confidence interval (CI)=1.017-2.576, p=0.015] and severe malnutrition (HR=1.918, 95% CI=1.275-2.884, p=0.002) were independent poor prognostic factors. Conclusion: Using two simple indices from the GLIM criteria, preoperative BMI and BWL, this study found that moderate and severe malnutrition were independent poor prognostic factors for OS.
The global leadership initiative on malnutrition (GLIM) criteria were published in 2018 as a global consensus on malnutrition standards (1). It has been reported that the more severe the undernutrition, the more postoperative complications increase and the poorer the long-term prognosis (2, 3). However, the methods of screening for undernutrition and the targets of preoperative nutritional interventions may differ among countries and Institutions, and it has been challenging to standardize and generalize the screening for undernutrition. The establishment of diagnostic criteria with global consensus has the advantage of providing universal evaluation, but the validity of the diagnostic criteria for GLIM needs to be clarified.
There are few reports on whether the GLIM criteria are predictive of long-term prognosis after gastrectomy in patients with gastric cancer. Since the postoperative course of patients with gastric cancer differs from that of other cancers in that postoperative body weight loss (BWL) occurs (4, 5), preoperative assessment of nutritional status is particularly important. In the GLIM criteria, the severity of malnutrition was determined after the diagnosis of malnutrition was confirmed. Currently, the cut-off values are specified for body mass index (BMI) and BWL rate, but there is no consensus on muscle mass. In addition, the validity of the former two indices for malnutrition severity in predicting postoperative outcomes has not yet been clarified. Although systematic reviews have shown that reduced skeletal muscle mass is associated with poor long-term prognosis in patients with gastric cancer (6-8), we investigated whether a simple assessment of malnutrition severity using the GLIM definition other than muscle mass is a valid predictor of long-term prognosis.
This study aimed to clarify the impact of the severity of malnutrition as defined by the GLIM criteria on long-term prognosis after gastrectomy in patients with advanced gastric cancer. We hypothesized that the higher the severity of malnutrition defined by BMI and BWL, the poorer the long-term prognosis.
Patients and Methods
Patients. This was a single-institution retrospective cohort study; consecutive patients who underwent gastrectomy for primary stage II-III gastric cancer, diagnosed according to the 15th edition of the Japanese Classification of Gastric Carcinoma, between April 2008 and June 2018 were included. The inclusion criteria were as follows patients with 1) primary advanced gastric cancer and 2) gastrectomy. We excluded patients: 1) with residual gastric cancer; 2) cancers of other organs; 3) who underwent non-gastrectomy surgical procedures; 4) unresectable distant metastases; and 5) who received preoperative treatment. We included patients with CY1, which indicated positive ascites cytology in the absence of distant metastasis. The study protocol was approved by the Institutional Ethical Review Committee of Ishikawa Prefectural Central Hospital (authorization number: 1812).
Definition of malnutrition severity defined by the GLIM criteria. The parameters used to diagnose malnutrition severity as defined by the GLIM criteria in this study are shown in Table I. After diagnosing malnutrition using the GLIM criteria, the patients were classified as having moderate or severe malnutrition based on their BMI and BWL rate. Patients without malnutrition were defined as having normal nutrition. Patients who met the inclusion criteria were allocated to three groups: the normal nutrition group (normal group), moderate malnutrition group (moderate group), and severe malnutrition group (severe group).
Severity grade as defined by the GLIM criteria in the present study.
Postoperative chemotherapy. In p-stage II-III, S-1 was initiated as postoperative chemotherapy at 80-120 mg/m2/day, and the dose was reduced according to the guidelines if side effects were observed. It was continued for a maximum of one year, and no other treatment was administered until recurrence. Patients with p-stage IV and recurrence received treatment based on the Japanese Gastric Cancer Treatment Guidelines.
Body composition. Before surgery, we measured visceral fat area and skeletal muscle mass on plain CT images using the graphical analysis software Ziostation (ZIOSOFT, Tokyo, Japan). Visceral fat mass was measured at the umbilical level. Skeletal muscle mass was measured at the third lumbar vertebra level. Skeletal muscle mass measured on a single CT image slice was divided by the patient’s height in meters squared to obtain the skeletal muscle mass index (SMI) (9).
Cut-off values for SMI were separately estimated for men and women based on the median and 25th percentiles for each group. The cut-off values defined by the median for SMI were 41.87 cm2/m2 for males and 34.04 cm2/m2 for females. The cut-off values as defined by the 25th percentile for SMI were 37.33 cm2/m2 for males and 29.79 cm2/m2 for females. Patients with an SMI below or above the cut-off value were categorized as having a low or high SMI, respectively.
Outcomes. The primary outcome was overall survival (OS). The secondary outcomes were cancer-specific survival (CSS), other-cause survival (OCS), disease-free survival (DFS), the total number of postoperative complications, severe postoperative complications, infectious complications, continuation rate of postoperative chemotherapy for one year, and postoperative BWL. OS was defined as the period between surgery and death. DFS was defined as the period between surgery and recurrence or death. Postoperative complications that occurred within 30 days after surgery were defined as Clavien-Dindo classification (CD) grade ≥2. Severe complications were defined as a CD grade ≥3. Treatment failure was defined as discontinuation of postoperative chemotherapy within one year of treatment initiation.
Analyses. We used the Mann-Whitney U-test for continuous variables, the Chi-square test or Fisher’s exact test for categorical variables and the log-rank test for Kaplan-Meier survival analysis for OS, CSS, OCS, and DFS. We used Cox proportional hazards regression for univariate analysis to identify prognostic factors for OS with p-values <0.05, of which the multivariate analysis was performed to calculate hazard ratios (HRs). We performed all statistical analyses using EZR software (Saitama Medical Centre, Jichi Medical University, Saitama, Japan) (10). We set the statistical significance at p<0.05.
Results
Patient background. A total of 512 patients who met the eligibility criteria were selected: 340 (66.4%) were included in the normal group, 84 (16.4%) in the moderate group, and 88 (17.2%) in the severe group. Patient characteristics are shown in Table II. The moderate and severe groups were older (p=0.002), had a lower BMI (p<0.001), more advanced c-stage (p=0.022), lower preoperative albumin (p<0.001), lower SMI (p<0.001), and lower visceral fat area (p<0.001). Preoperative CRP was higher in the moderate group than in the normal and severe groups (p=0.002).
Patient characteristics.
Comparison of the postoperative outcomes. A comparison of postoperative outcomes according to the severity of malnutrition is shown in Table III. The moderate and severe groups had a higher percentage of serosal invasion than the normal group (p=0.005). There was no significant difference between the operating time and intra-operative blood loss. Infectious complications were significantly greater in the normal and moderate groups than in the severe group (p=0.035), including intra-abdominal abscess (p=0.045) and pancreatic fistula (p=0.009). The continuation rates of postoperative chemotherapy for one year were greater in the normal group than in the moderate and severe groups (p=0.001). The rates of postoperative BWL for one month, six months, and one year were significantly higher in the normal group than in the moderate and severe groups (p<0.001, p<0.001, and p<0.001, respectively).
Comparison of postoperative outcomes.
Long-term prognosis according to the severity of malnutrition. The median follow-up period was 41 months (interquartile range=17-60 months). The long-term survival curves divided by the severity of malnutrition are shown in Figure 1. Comparing the long-term prognosis in the three groups, the prognosis was worse in the moderate and severe groups than in the normal group for OS [HR=1.621, 95% confidence interval (CI)=1.359-1.934, p<0.001], CSS (HR=1.450, 95% CI=1.172-1.793, p<0.001), OCS (HR=2.210; 95% CI=1.598-3.057, p<0.001), and DFS (HR=1.399, 95% CI=1.182-1.654, p<0.001).
Survival curves for overall survival (OS), cancer-specific survival (CSS), other-cause survival (OCS), and disease-free survival (DFS) according to the severity of malnutrition, defined using the GLIM criteria. (a) OS (p<0.001), (b) CSS (p<0.001), (c) OCS (p<0.001), (d) DFS (p<0.001).
Prognostic factors for OS. The results of the analysis of prognostic factors for OS are shown in Table IV. In the univariate analysis, age >70 years (p<0.001), total gastrectomy (p<0.001), open surgery (p<0.001), D2 lymph node dissection (p=0.010), serosal invasion (p<0.001), lymph node metastasis (p<0.001), moderate and severe malnutrition (p<0.001 and p<0.020, respectively), low-SMI (p<0.001), and severe complications (p=0.028) were statistically significant. Multivariate analysis showed that age >70 years (HR=1.720, 95% CI=1.225-2.416, p=0.002), open surgery (HR=1.934, 95% CI=1.350-2.772, p<0.001), D2 lymph node dissection (HR=0.533, 95% CI=0.378-0.750, p<0.001), serosal invasion (HR=1.749, 95% CI=1.244-2.460, p=0.001), N3 lymph node metastasis (HR=2.583, 95% CI=1.829-3.648, p<0.001), moderate malnutrition (HR=1.689, 95% CI=1.107-2.576, p=0.015), and severe malnutrition (HR=1.918, 95% CI=1.275-2.884, p=0.002) were significant independent prognostic factors for OS.
Results of analyses of prognostic factors for overall survival.
Discussion
This study clarified that the severity of GLIM malnutrition, defined using two consensus measures of malnutrition severity, BMI, and weight loss, was associated with poor long-term prognosis after gastrectomy in patients with advanced gastric cancer. Comparison of the survival curves showed that OS, CSS, OCS, and DFS were poor in the moderate malnutrition and severe malnutrition groups. In addition, multivariate analysis of the poor prognostic factors showed that moderate and severe malnutrition were independent poor prognostic factors for OS. This is the first validation study to demonstrate the prognostic value of a simple assessment of malnutrition severity in patients with gastric cancer and postoperative weight loss, as defined by the GLIM criteria.
Moderate and severe malnutrition were independent poor prognostic factors for OS after gastrectomy in patients with advanced gastric cancer. Comparison of the Kaplan-Meier curves showed that moderate and severe malnutrition, as defined by the GLIM criteria, were equally associated with poor OS and DFS. This is because many patients in the moderate group had high preoperative CRP levels, which may have cachexia before surgery, and the continuation rate of chemotherapy was poor. Because the GLIM criteria define malnutrition by combining phenotypic criteria such as BMI and BWL and etiologic criteria consisting of reduced food intake and the presence of inflammation (1), it may be easier to select patients with cachexia. In addition, the severe group had the lowest SMI, which may have contributed to the poor OCS due to the larger number of patients with sarcopenia. The results of this study suggest that nutritional interventions are necessary during moderate malnutrition.
Frequency of postoperative complications did not correlate with severity of malnutrition. Infectious complications, such as intra-abdominal abscesses and pancreatic fistula, were more frequent in the normal group with a higher BMI. It has been reported that intra-abdominal infections and pancreatic fistula increase with high visceral fat content (11-14), and this study may have been influenced by the bias of cases with high visceral fat content in the normal group. The moderate group had the highest number of patients with high preoperative CRP levels and low albumin levels, and the number of patients with low-SMI was also high, which may explain the increase in the total number of postoperative complications. In a report that defined malnutrition severity using BMI and BWL as defined by the GLIM criteria, moderate or higher malnutrition was found to be a risk factor for severe complications (15), which is consistent with the results of this study. Although the severe group had the highest number of patients with low-SMI, a systematic review has shown that sarcopenia does not increase intra-abdominal infections (7), which is consistent with the results of this study. However, the percentage of severe complications among all complications was higher in patients with moderate and severe malnutrition (54.2% and 58.3%, respectively), and the impact of malnutrition diagnosis in the perioperative period was significant.
The GLIM criteria is a worldwide consensus standard for malnutrition, and the severity of malnutrition diagnosed in this study was defined using BMI and BWL. These were found to be a valid predictor of long-term prognosis in patients with gastric cancer. Although it has been reported that the assessment of malnutrition severity, including SMI, is useful for predicting prognosis (16, 17), there is still no definitive cut-off value for SMI, and it is too complicated for surgeons who have never measured the SMI to use in preoperative decision making. Although it has been stated that BMI should be reconsidered in Asia (1), the present study shows that the current BMI values are useful in predicting outcomes in Asians. This study is of great value in demonstrating the usefulness of only two simple indices for predicting complications and prognosis, which can be assessed during daily clinical practice.
There are several limitations to this study. First, it was a single-center retrospective cohort study. Second, the cut-off values of SMI are unclear and require validation in additional multicenter cohort studies. However, this is the first report to show the crucial impact of malnutrition severity as defined by the simple indices, BMI and BWL, on long-term postoperative prognosis in patients with advanced gastric cancer who exhibit postoperative body weight loss. Our findings imply that there is a high need for postoperative nutritional support in those with moderate and severe malnutrition. In the future, we would like to investigate whether a support system, including nutritional therapy, would lead to longer OS in such patients.
In conclusion, using two simple indices from the GLIM criteria, preoperative BMI and BWL, this study found that moderate and severe malnutrition were independent poor prognostic factors for OS.
Footnotes
Authors’ Contributions
R. Matsui and N. Inaki equally contributed to the conception and design of the research; R. Matsui and T. Tsuji contributed to the acquisition and analysis of the data; R. Matsui and N. Inaki contributed to the interpretation of the data; and R. Matsui and N. Inaki drafted the manuscript. All Authors critically revised the manuscript, agree to be fully accountable for ensuring the integrity and accuracy of the work, and read and approved the final manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest.
- Received June 6, 2022.
- Revision received July 2, 2022.
- Accepted July 7, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
