Abstract
Background/Aim: Visceral adipose tissue index (VAI) and visceral-to-subcutaneous adipose tissue area ratio (VSR) are indices related to visceral fat mass, and intramuscular adipose tissue content (IMAC) is an index related to intramuscular fat. We investigated the impact of fat mass-related indices on the incidence of postoperative complications. Patients and Methods: Patients who underwent radical laparoscopic gastrectomy for c-stage I primary gastric cancer between April 2014 and December 2020 were included. We investigated the relationship between both VAI, VSR, IMAC and postoperative complications. For each value, the median was used as a cut-off point, the patients were divided into high- and low-fat groups, and background adjustments using propensity score matching analysis were used to compare these two groups. Results: Of the 490 patients, 245 (50.0%) were in the high-VAI group, 243 (49.6%) in the high-VSR group, and 239 (48.8%) in the high-IMAC group. After matching, intra-abdominal abscess and pancreatic fistula were greater in the high-VAI group than those in the low-VAI group (p=0.081 and p=0.081, respectively), and were also significantly greater in the high-VSR group (p=0.012 and p=0.025, respectively). However, anastomotic leakage was greater in the high-IMAC group than in the low-IMAC group (p=0.050). Conclusion: VAI and VSR, which reflect visceral fat mass, were useful in predicting intra-abdominal infections, whereas IMAC, which reflects intramuscular fat, was useful in predicting anastomotic leakage. Preoperative prediction using a combination of these factors may aid clinical assessment of complication risk.
Increased visceral fat mass, as a change associated with obesity, increases the postoperative complications of gastric cancer. High visceral fat mass prolongs the operation time and increases intraoperative blood loss (1-3) and postoperative infectious complications such as pancreatic fistula, intra-abdominal infection, and wound infection (4-10). Increasing obesity rates are an issue worldwide, and preoperative fat mass assessment is essential.
The relationship between indices reflecting fat mass and the chance of postoperative complications after laparoscopic gastrectomy has not been fully investigated. In addition to visceral fat mass measured by computed tomography (CT) cross-sectional area and intramuscular adipose tissue content (IMAC) (11) visceral-to-subcutaneous fat ratio (VSR) (12) has been shown to correlate with postoperative complications. However, the relationship between each index and specific complications after laparoscopic gastrectomy, such as surgical site infection (SSI), pancreatic fistula, anastomotic leakage, and intra-abdominal abscess, has not been thoroughly described. Because the clinical importance of increased complications varies, it is necessary to distinguish between them when performing preoperative interventions for increased fat mass. It is also important to consider, which of these indices is most predictive of which specific complications.
The purpose of this study was to clarify the relationship between previously reported indices related to fat mass and postoperative complications. Since multiple factors are involved in postoperative complications, we investigated each factor with background adjustment. If the complications predicted by these indices were different, it would be possible to take countermeasures for each.
Patients and Methods
Study participants. This was a retrospective study conducted at Ishikawa Prefectural Central Hospital (Kanazawa, Japan) and Juntendo University Urayasu Hospital (Chiba, Japan), which included patients who underwent laparoscopic gastrectomy for primary c-stage I gastric cancer, diagnosed according to the 15th edition of the Japanese Classification of Gastric Carcinoma, between April 2014 and December 2020. All experimental protocols described in this study were approved by the Institutional Ethical Review Committee (authorization number: 1761) and Juntendo University Urayasu Hospital (authorization number: E21-0126). We retrospectively collected clinical and laboratory data, including medical records and images, using the electronic patient record system of the hospital. The inclusion criteria were as follows: 1) primary early gastric cancer, 2) laparoscopic gastrectomy, and 3) preoperative body composition analysis data on CT images. We excluded patients with residual gastric cancer, cancers of other organs, open surgery, non-gastrectomy surgical procedures, and unresectable distant metastases. Patients who met the aforementioned criteria were divided into a low-fat group and a high-fat group. Postoperative outcomes were compared between the two groups after adjusting for patient background using propensity score matching (PSM).
Body composition. Muscle adiposity, visceral fat mass, subcutaneous fat mass, and skeletal muscle mass were measured on preoperative CT images using the graphic analysis software Ziostation (ZIOSOFT, Tokyo, Japan). For muscle adiposity, CT values [Hounsfield units (HUs)] of regions of interest were measured at the umbilical level and IMAC was calculated by dividing the CT value of the multifidus muscles with that of subcutaneous fat, as was done in previous studies (11, 13, 14). Visceral fat mass and subcutaneous fat mass, defined as Hus of −150 to −50, were measured at the umbilical level and skeletal muscle mass, defined as Hus of −29 to 150, at the level of the third lumbar vertebrae. The visceral fat and skeletal muscle mass measured on a single CT image slice were divided by the patient’s height in meters squared to obtain the visceral adipose tissue index (VAI) and skeletal muscle mass index (SMI) (15). Visceral adiposity was evaluated by the VSR, which was calculated as follows: VSR=visceral adipose tissue area (cm2)/subcutaneous adipose tissue area (cm2) (12).
Cut-off values for VAI, SMI, IMAC, and VSR were separately estimated for males and females based on the median for each group. The cut-off values for VAI were 33.59 cm2/m2 for males and 25.00 cm2/m2 for women. Patients with a VAI below the cut-off value were categorized as low-VAI, whereas those with a VAI above the cut-off value were classified as high-VAI. The cut-off values for SMI were 43.80 cm2/m2 for males and 33.44 cm2/m2 for women. Likewise, patients with an SMI below and above the cut-off value were categorized as having low or high SMI, respectively. The cut-off value for IMAC was −0.440 for men, −0.320 for women, and patients below the cut-off were categorized as low-IMAC and those above as high-IMAC. A higher IMAC indicated higher muscle adiposity. The cut-off value for VSR was 0.970 for men, 0.400 for women, and patients below the cut-off were categorized as low-VSR and those above as high-VSR.
Outcomes and analyses. The primary outcome was the Clavien–Dindo classification (CD) for grade 3 or higher severe postoperative complications, and the secondary outcomes were operating time, intraoperative blood loss, length of postoperative hospital stay, total postoperative complications, and infectious complications. Postoperative complications were defined as the occurrence of complication of CD grade 2 or higher within 30 postoperative days.
We performed PSM on low-fat and high-fat groups to adjust for differences in patient background and to reduce selection bias in a non-randomized study. The propensity score was estimated using a logistic regression model with the following covariates: sex, age, surgical procedure, lymph node dissection, comorbidities, serum albumin, C-reactive protein, anti-thrombotic drug use, and SMI; body mass index (BMI) was excluded. The nearest-neighbor matching method was applied, and one-to-one matching between the two groups was achieved. The caliper size was 0.20. After matching, postoperative outcomes were compared between the two groups. Patient characteristics and postoperative outcomes were compared using the Mann–Whitney U-test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables. Pearson’s correlation coefficient was used to analyze the correlation between the indices related to fat mass. We performed all statistical analyses using EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is based on R software (The R Foundation for Statistical Computing, Vienna, Austria) and R Commander (16). Statistical significance was set at p<0.05.
Results
Correlation between indices related to fat mass. There was no relationship between IMAC and VAI (r=0.029, p=0.516) or between VAI and VSR (r=−0.020, p=0.664). There was a weak negative relationship between IMAC and VSR (r=−0.105, p=0.020).
Patient characteristics according to VAI before and after PSM. A flowchart of the study is shown in Figure 1. A total of 490 patients who met the eligibility criteria were selected. After that, 245 (50.0%) and 245 (50.0%) patients, respectively, were categorized into the low- and high-VAI groups. Patient characteristics are shown in Table I. There were 97 patients in both groups after PSM. Before matching, the high-VAI group had a higher age (p<0.001), a higher BMI (p<0.001), a greater number of patients with D2 lymph node dissection (p=0.001), a greater number of patients with CKD (p=0.002) and diabetes (p<0.001), a lower low-SMI (p=0.002), a higher VAI (p<0.001), a higher VSR (p<0.001), and a higher IMAC (p<0.001) than the low-VAI group. After matching, there was no significant difference in factors that were not adjusted, other than BMI and VAI.
Study design.
Patient characteristics according to visceral adipose tissue index (VAI) before and after propensity score matching.
Comparison of postoperative outcomes according to VAI after matching. Comparisons of postoperative outcomes according to the VAI after matching are shown in Table II. The high-VAI group had a significantly longer operating time (p=0.001), more intraoperative blood loss (p<0.001), and longer postoperative hospital stay (p=0.011). No significant difference was observed in any of the postoperative complications, but intra-abdominal abscess and pancreatic fistula were greater in the high-VAI group (p=0.081 and p=0.081, respectively).
Comparison of postoperative outcomes according to visceral adipose tissue index (VAI) after matching.
Patient characteristics according to VSR before and after PSM. Of the 490 patients selected, 247 (50.4%) and 243 (49.6%) were categorized into the low- and high-VSR groups, respectively. Patient characteristics are shown in Table III. There were 112 patients in both groups after PSM. Before matching, the high-VSR group had a higher age (p<0.001), a greater number of patients with D2 lymph node dissection (p=0.033), a greater number of patients with CKD (p=0.002) and diabetes (p<0.001), a higher VAI (p<0.001), a higher VSR (p<0.001), and a higher IMAC (p=0.019) than the low-VSR group. After matching, there was no significant difference in factors that were not adjusted, other than BMI, VSR, and high-IMAC.
Patient characteristics according to visceral-to-subcutaneous adipose tissue area ratio (VSR) before and after propensity score matching.
Comparison of postoperative outcomes according to VSR after matching. Comparisons of postoperative outcomes according to the VSR after matching are shown in Table IV. The high-VSR group had a significantly longer operating time (p<0.001), longer postoperative hospital stay (p<0.001), and greater infectious complications (p=0.015). As for the specific infectious complications, intra-abdominal abscess and pancreatic fistula were significantly greater in the high-VSR group (p=0.012 and p=0.025, respectively).
Comparison of postoperative outcomes according to visceral-to-subcutaneous adipose tissue area ratio (VSR) after matching.
Patient characteristics according to IMAC before and after PSM. Of the 490 patients, 251 (51.2%) and 239 (48.8%) patients were categorized into the low- and high-IMAC groups, respectively. Patient characteristics are shown in Table V. There were 123 patients in both groups after PSM. Before matching, the high-IMAC group had a higher age (p<0.001), a higher BMI (p<0.001), a greater number of patients with CKD (p=0.023), diabetes (p<0.001), anti-thrombotic drug use (p=0.039), a higher VAI (p<0.001), a higher VSR (p=0.019), and a higher IMAC (p<0.001) than the low-IMAC group. After matching, there was no significant difference in factors that were not adjusted, other than BMI and IMAC.
Patient characteristics according to intramuscular adipose tissue content (IMAC) before and after propensity score matching.
Comparison of postoperative outcomes according to IMAC after matching. Comparisons of postoperative outcomes according to the IMAC after matching are shown in Table VI. The high-IMAC group had a significantly longer operating time (p=0.026), more intraoperative blood loss (p=0.001), and longer postoperative hospital stay (p=0.034). In the high-IMAC group, the total number of complications (p=0.046), severe complications (p=0.034), and infectious complications (p=0.047) were significantly greater. As for specific examples of complications, anastomotic leakage was greater in the high-IMAC group (p=0.050).
Comparison of postoperative outcomes according to intramuscular adipose tissue content (IMAC) after matching.
Discussion
In this study, we described the correlations between three indices of fat mass and postoperative complications after laparoscopic gastrectomy in patients with gastric cancer using PSM. High VAI and high VSR, indicating excess visceral fat, were superior in predicting intra-abdominal abscess and pancreatic fistula, while high-IMAC, indicating intramuscular fat, was associated with more anastomotic leakage and severe complications. These results indicate that visceral and intramuscular fat are each associated with different postoperative complications.
IMAC was not correlated with VAI and had a weak negative correlation with VSR. There was no correlation between VAI and VSR. Therefore, these indices are considered to be almost independent of each other.
It has been previously reported that pancreatic fistula and intra-abdominal abscess increase with increasing visceral fat content (4-10). In the present study, both VAI, which indicates the absolute amount of visceral fat, and VSR, which indicates the ratio of visceral fat to subcutaneous fat, were useful indices for predicting intra-abdominal infections. In contrast, IMAC was superior in detecting anastomotic leakage, which was reflected in the significantly higher incidence of severe complications. In addition to IMAC, there are other methods to measure intramuscular fat, such as muscle density, which have been reported to correlate with anastomotic leakage not only in patients after gastric cancer resection operations but also in patients after colorectal cancer resection operations (11, 17). The results were similar to those of a study that focused on laparoscopic gastrectomy. Both intra-abdominal infection and anastomotic leakage are problematic complications in the clinical course, and the results of this study suggest that a combined preoperative evaluation of these complications may be useful.
All of the indices used in this study can be measured with CT, which is commonly used in preoperative staging of cancer, and can be employed immediately in daily clinical practice without the need for more elaborate tests. Although the measurement of visceral fat is complicated by the need for image analysis software, it has the advantage of allowing the clinician to determine whether to place an intra-abdominal drain in anticipation of intra-abdominal infection. In the present study, we used the median cut-off values, which have the advantage of allowing us to judge the general trend by dividing the patients into two groups; however, whether it is appropriate as a general cut-off value requires a future multicenter validation study.
The limitations of this study include the following: 1) it was a two-center retrospective study, and 2) the validity of the cut-off values is not clear. It has been reported that a 4-week exercise intervention for obesity without caloric restriction significantly reduces body weight and visceral fat mass and prevents postoperative complications (18). It is necessary to establish which elements of the preoperative intervention, such as the program and its duration, helped prevent postoperative complications. In addition, muscle quality, which reflects intramuscular fat, has recently been emphasized in the diagnosis of sarcopenia, (19) reaffirming it as a potential target for intervention to be investigated. It is important to elucidate whether interventions such as nutrition, exercise, and pharmacotherapy are effective in treating the fat-related indices and reducing the associated postoperative complications described in this study.
Conclusion
VAI and VSR, reflecting visceral fat mass, were useful in predicting intra-abdominal infections, while IMAC, reflecting intramuscular fat, was useful in predicting anastomotic leakage and severe complications. Preoperative risk assessment using a combination of these factors may aid clinical decisions.
Footnotes
Authors’ Contributions
R. Matsui and N. Inaki equally contributed to the conception and design of the research; R. Matsui contributed to the 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 in relation to this study.
- Received July 12, 2022.
- Revision received August 7, 2022.
- Accepted August 8, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
