Abstract
Background/Aim: Laparoscopic gastrectomy (LG) may have greater clinical benefits as a less invasive surgery for elderly patients with gastric cancer (GC). Therefore, we aimed to evaluate the survival benefit of LG in elderly patients with GC, especially focusing on preoperative comorbidities, and nutritional and inflammatory status. Patients and Methods: Data collected from 115 patients aged ≥75 years with primary GC who underwent curative gastrectomy, comprising 58 patients who underwent open gastrectomy (OG) and 57 patients who underwent LG, were retrospectively reviewed (total cohort), and 72 propensity-matched patients (matched cohort) were selected for survival analysis. The aim of the study was to determine short- and long-term outcomes, and the clinical markers to identify a population who may benefit from LG in elderly patients. Results: The complication and mortality rates as a short-term outcome in the total cohort and overall survival (OS) as a long-term outcome in the matched cohort did not differ significantly between the groups. In the total cohort, advanced tumor stage and ≥3 comorbidities were independent factors for poor prognosis in terms of OS [hazard ratio (HR)=3.73, 95% confidence interval (CI)=1.78-7.78, p<0.001 and HR=2.50, 95% CI=1.35-4.61, p<0.01, respectively]. The surgical approach was not an independent risk factor for postoperative complications (grade ≥III) and OS. In subgroup analysis of the total cohort, patients with a neutrophil-lymphocyte ratio (NLR) ≥3 in the LG group demonstrated a trend toward greater OS (HR=0.26, 95% CI=0.10-0.64, interaction p<0.05). Conclusion: LG might offer greater survival benefits than OG in frail patients such as those with high NLR.
- Gastric cancer
- laparoscopic gastrectomy
- open gastrectomy
- elderly patients
- neutrophil–lymphocyte ratio
- overall survival
Gastric cancer (GC) is the fifth most frequently diagnosed cancer and the third leading cause of cancer-related deaths worldwide (1). It is particularly prevalent in East Asian countries, such as Japan, Korea, and China. To treat GC, surgery is the most effective treatment for resectable GC. Due to its minimally invasive nature, laparoscopic gastrectomy (LG) is now performed broadly for early and advanced GC (2-5), even in elderly patients (6-8).
With the rapid aging of the global population, including the Japanese population, the number of patients with frailty is increasing. The frailty is a common geriatric syndrome causing vulnerability to stressors that leads to disorders and is a strong predictor of adverse outcomes in surgical patients (9). Patients with frailty are at an increased risk of postoperative complications, chemotherapy intolerance, disease progression, and death. Although the gold standard definition of frailty is lacking due to its complex and multidimensional status (10, 11), comorbidities, and the nutritional and inflammatory status are associated with frailty (12-14) and strongly correlate with postoperative morbidities and mortality in elderly patients with cancer, including GC (15-17).
Since elderly patients are at a higher risk of postoperative complications associated with gastrectomy (18) due to frailty caused by reduced functional reserve, increased comorbidities and poor nutritional status, a minimally invasive approach is preferred. Hence, less invasive LG may be suitable for elderly patients, which is associated with a lower inflammatory response and lesser impact on the immune function (19, 20). Randomized controlled trials to evaluate the safety and efficacy of LG compared to those of OG in elderly patients with locally advanced GC are currently in progress (21, 22). Since elderly patients need an individualized optimization of treatment due to frailty, upfront identification of patients who are more likely to benefit from LG is needed. Some previous studies have proposed factors, such as the modified frailty index which use laboratory data (i.e., albumin, hematocrit, creatinine) (23), prognostic nutritional index (PNI) (24), anemia, Charlson comorbidity index, and combined resection (25) that predict the outcome of LG in elderly patients with GC. However, only one study has investigated the populations that most benefited from LG compared to OG (26). Thus, it remains inconclusive whether LG is an appropriate surgical approach for elderly patients with GC, and the prognostic markers of this population have yet to be elucidated.
In the present study, to address the knowledge gap in the literature, our primary aim of the study was short- and long-term outcomes in elderly patients with GC who underwent OG and LG; the secondary aim was determining the independent risk factors and clinical markers to identify a population that are best suited for LG, with a special focus on preoperative comorbidities and hematologic parameters that reflect the nutritional and inflammatory status.
Patients and Methods
Patient characteristics in the total cohort. The flowchart of the study selection process is shown in Figure 1. The medical records of 115 patients aged ≥75 years with primary GC who underwent curative gastrectomy (pR0) at Kyushu Central Hospital between January 2012 and February 2020 were retrospectively reviewed. The participants comprised 58 patients who underwent open gastrectomy (OG group) and 57 patients who underwent laparoscopic gastrectomy (LG group) (total cohort). Patients who underwent non-curative gastrectomy (R1/2), gastrectomy for remnant GC, and gastrectomy after endoscopic submucosal dissection were excluded. Short- and long-term outcomes were compared between the two groups. The follow-up duration was calculated from the date of surgery to the date of death or last follow-up. Patients lost during follow-up were censored at the time of the last follow-up. Follow-up data were obtained from the electronic medical records of Kyushu Central Hospital. Tumor stage classification and assessment were performed in accordance with the Japanese Classification of Gastric Carcinoma (15th Edition), reflecting the 8th American Joint Committee on Cancer/Union for International Cancer Control tumor–node–metastasis classification (27). Activities of daily living and postoperative complications were assessed according to the American Society of Anesthesiologists-Physical Status Scale (28) and Clavien–Dindo classification (29), respectively. The proportions of the missing data were carcinoembryonic antigen (CEA) 3.5%, carbohydrate antigen 19-9 (CA19-9) 5.2%, C-reactive protein (CRP), Glasgow prognostic score (GPS) and CRP–albumin ratio (CAR) 0.9%, and total cholesterol, controlling nutritional status (CONUT) score 6.1%. There were no missing data for comorbidities, other clinicopathological variables, or nutritional and inflammatory status. Patients with missing data were included in this analysis. The research reported in this paper was in compliance with the Declaration of Helsinki. This study was approved by the Ethics Committee of Kyushu Central Hospital of the Mutual Aid Association of Public School Teachers (approval number: 218). The Ethics Committee of Kyushu Central Hospital of the Mutual Aid Association of Public School Teachers waived the need for written informed consent due to the retrospective nature of this study. The participants had the option to opt out through the hospital website of Kyushu Central Hospital.
Flowchart of the selection process for the study. OG: Open gastrectomy; LG: laparoscopic gastrectomy.
Preoperative comorbidities and nutritional and inflammatory statuses. Comorbidities in each patient were classified based on the following 10 categories: cardiac, pulmonary, liver, renal, cerebral, hypertensive, diabetes/endocrine, vascular, and immune diseases along with other types of cancer, as described in our previous study (30) (Table I). Patients with renal disease were divided into two groups based on the chronic kidney disease classification as reported previously (31): renal disease present [glomerular filtration rate (GFR) <45 ml/min/1.73 m2, stages G3b-5] or renal disease absent (GFR ≥45 ml/min/1.73 m2, stages G1-3a). The severity of comorbidities was not assessed. The hematologic scoring system reflecting the nutritional and inflammatory statuses, including PNI, GPS, CONUT score, neutrophil–lymphocyte ratio (NLR), platelet–lymphocyte ratio (PLR), lymphocyte–monocyte ratio (LMR), and CAR were observed.
Categories of preoperative comorbidities.
Propensity score matching. We performed propensity score matching (PSM) to control the selection bias and balance some covariates that might be associated with the prognosis between the OG and LG groups, and 36 pairs of patients were analyzed for survival (matched cohort) (Figure 1). The patients’ propensity scores were estimated using a logistic regression model based on the following variables: depth of invasion, lymph node metastasis, serum albumin level, and CRP level. Patients in the OG and LG groups were matched in a 1:1 ratio using the nearest propensity score on the logit scale with a caliper of 0.20. Subsequently, the patients’ clinicopathological characteristics, comorbidities, and systemic nutritional and inflammatory statuses were re-evaluated.
Statistical analysis. The associations between the variables were tested using the chi-squared test, Student’s t-test, or the Mann–Whitney U-test. The prognostic value of each variable was analyzed using the univariate log-rank test. Multivariate analysis was performed using the Cox’s proportional hazards model to identify the independent factors that were predictive of patient survival. Each predictive variable with a p-Value of <0.05 in univariate analysis was assessed for multicollinearity, and variables for which the correlation coefficients (Spearman) with the other factors were <0.4 were included in multivariate analysis. Subgroup analyses were conducted according to the surgical approach. Statistical significance was set at p<0.05. All statistical analyses were performed using the JMP14.2.0 software (SAS Institute, Cary, NC, USA).
Results
Patient characteristics in the total and matched cohorts. The clinicopathological characteristics of the total and matched cohorts are summarized in Table II. In the total cohort, 58 (50.4%) and 57 (49.6%) patients were in the OG and LG groups, respectively. The median age was 82 years (range=75-94 years) in the OG group and 81 years (range=75-94 years) in the LG group (p=0.64). Less advanced depth of invasion, a negative lymph node status, less advanced stage, and lower CEA levels were more frequently observed in the LG group than in the OG group (p<0.05, p<0.01, p<0.001 and p<0.01, respectively). In the matched cohort, no significant differences were observed between the two groups.
Clinicopathological characteristics in Japanese patients with GC aged ≥75 years who underwent gastrectomy in the total and matched cohorts.
Preoperative comorbidities and systemic nutritional and inflammatory status in the total and matched cohorts. The comparisons of preoperative comorbidities and systemic nutritional and inflammatory statuses of the total and matched cohorts are summarized in Table III. In the total cohort, the serum albumin level was significantly higher, and the CRP level was lower in the LG than the OG group (p<0.05 and p<0.01, respectively), resulting in significantly lower GPS and CAR in the LG group (p<0.01 and p<0.05, respectively). Regarding preoperative comorbidities, the incidence of each disease and number of comorbidities were not significantly different. In the matched cohort, no significant differences were observed in preoperative comorbidities or systemic nutritional and inflammatory statuses between the groups.
Preoperative comorbidities, and nutritional and inflammatory status in patients with GC aged ≥75 years who underwent gastrectomy in the total and matched cohorts.
Operative outcomes in the total cohort. The operative outcomes of the total cohort are presented in Table IV. The dominant surgical procedure in both groups was distal gastrectomy (69.0% in the OG group and 82.5% in the LG group), and the distribution of surgical procedures was not statistically different. The LG group had a significantly longer operative time, less blood loss, and a shorter postoperative hospital stay (p<0.0001 for all). The complication rate (all grades and grade ≥III) and the 30- and 90-day mortality rates did not differ significantly between the OG and LG groups. The follow-up periods after surgery were 3.3±2.6 years in the OG group and 3.0±1.9 years in the LG group.
Operative outcome in patients with GC aged ≥75 years who underwent gastrectomy in the total cohort.
Survival in the total and matched cohorts. In the total cohort, the 5-year overall survival (OS) and disease-specific survival (DSS) rates in the LG group were significantly higher than those in the OG group (OS: 69.3% vs. 43.1%, p<0.05; DSS: 90.0% vs. 71.9%, p<0.05) (Figure 2A, B). In the matched cohort, there were no statistically significant differences in the 5-year OS and DSS rates between the OG and LG groups (OS: 49.9% vs. 63.5%, p=0.37; DSS: 82.1% vs. 86.9%, p=0.52) (Figure 2C, D).
Overall survival (A) and disease-specific survival (B) curves in patients with gastric cancer aged ≥75 years who underwent gastrectomy in the total cohort. Overall survival (C) and disease-specific survival (D) curves in patients with gastric cancer aged ≥75 years who underwent gastrectomy in the matched cohort. Survival rates were calculated by the Kaplan-Meier method and compared between the open gastrectomy (OG) and laparoscopic gastrectomy (LG) groups with the log-rank test.
Cause of death in the total cohort. In the total cohort, the mortality rate in the follow-up period was 58.6% in the OG group and 28.1% in the LG group. The proportion of GC deaths in the OG group was higher than that in the LG group (24.1% vs. 7.0%, p<0.05) (Table V). The proportion of deaths from other cancers and non-cancerous diseases was not significantly different between the two groups (8.6% vs. 1.8%, p=0.21 and 20.7% vs. 19.3%, p=0.85, respectively).
Cause of death in patients with GC aged ≥75 years who underwent gastrectomy in the total cohort.
Independent risk factors for postoperative complications and overall survival in the total cohort. In the total cohort, multivariate analysis showed that preoperative pulmonary disease and diabetes/endocrine disease were independent risk factors for postoperative complications (grade ≥III) [hazard ratio (HR)=5.78, 95% confidence interval (CI)=1.54-21.6, p<0.01 and HR=4.07, 95% CI=1.11-14.9, p<0.05, respectively) (Table VI). Advanced tumor stage (stages II, III) and ≥3 comorbidities were independent factors for poor prognosis in terms of OS (HR=3.73, 95% CI=1.78-7.78, p<0.001 and HR=2.50, 95% CI=1.35-4.61, p<0.01, respectively) (Table VII). The surgical procedure and surgical approach were not an independent risk factor for postoperative complications (grade ≥III) and OS (Table VI and Table VII).
Univariate and multivariate analyses of clinicopathological factors, and postoperative complication (≥Grade III) in patients with GC aged ≥75 years who underwent gastrectomy in the total cohort.
Univariate and multivariate analyses of clinicopathological factors and OS in patients with GC aged ≥75 years who underwent gastrectomy in the total cohort.
Subgroup analysis for postoperative complications and overall survival according to the surgical approach in the total cohort. Subgroup analysis for postoperative complications (grade ≥III) and OS according to the surgical approach in the total cohort is shown in Figure 3 and Figure 4. In the overall patients, the HR of LG compared with OG was 0.70 (95% CI=0.21-2.35) in postoperative complications (grade ≥III) and 0.51 (95% CI=0.28-0.92) in OS. There were no significant interactions for postoperative complications (grade ≥III) in almost all subgroups; however, patients with positive lymph node metastasis in the LG group showed a favorable trend compared to those in the OG group (HR=0.13, 95% CI=0.01-2.41, interaction p<0.05) (Figure 3). There were no significant interactions for OS in almost all subgroups; however, patients with NLR ≥3 in the LG group demonstrated a trend toward longer OS compared to those in the OG group (HR=0.26, 95% CI=0.10-0.64, interaction p<0.05) (Figure 4).
Subgroup analysis of postoperative complications (grade ≥III) in patients with gastric cancer aged ≥75 years who underwent gastrectomy in the total cohort. HR: Hazard ratio.
Subgroup analysis of overall survival in patients with gastric cancer aged ≥75 years who underwent gastrectomy in the total cohort. HR: Hazard ratio.
Discussion
In this study, we demonstrated that LG for elderly patients with GC was equivalent to OG in terms of both short- and long-term outcomes and that the surgical approach (OG or LG) was not an independent factor for postoperative complications (grade ≥III) and OS. However, NLR was found to be a suitable candidate biomarker to identify the subset of patients who would most benefit from LG.
In our study, the LG group had a longer operative time, less blood loss, and shorter postoperative hospital stay, similar to previous studies (32, 33). The outcomes of the LG group were not inferior to those of OG group in terms of both short-term (complication rate and mortality) and long-term outcomes, as previously reported (32-34). The 30- and 90-day mortality rates of 0% and 3.4% in the OG group, respectively, and of 3.5% and 3.5% in the LG group, respectively, were acceptable compared to the mortality rates of previous studies, which reported short-term mortality rates of 0.0%-19.7% in elderly patients with GC who had undergone gastrectomy (32, 35-39). Regarding the cause of death, a previous study showed that death from respiratory disease occurred more frequently in the OG group than in the LG group, which might have caused unfavorable OS in the OG group (32). Although the proportion of death from pulmonary disease was more frequently observed in the OG group than in the LG group, the difference was not significant in our study.
In our study, preoperative ≥3 comorbidities were independent risk factors for OS. A previous study showed that multiple comorbidities were significant risk factors for postoperative death from other disease in elderly patients with GC who underwent gastrectomy (17). Therefore, surgeons should pay attention to patients’ preoperative comorbidities and take early preventive measures to reduce the occurrences of adverse outcomes. Surgical approach (OG or LG) was not an independent risk factor for postoperative complications (grade ≥III) and OS. Hence, LG might be a suitable alternative treatment option for elderly patients with GC.
Chronic inflammation plays an important role in the occurrence of frailty (40). Previous studies have shown that high levels of inflammatory parameters, such as CRP and interleukin-6, are associated with aspects of frailty, such as poor physical performance and muscle strength in elderly individuals (41, 42). NLR has recently emerged as a biomarker of systematic inflammation and may represent the balance between pro-tumoral inflammatory status and antitumoral immune response (43-45). A high NLR is associated with poor prognosis in patients with various cancers, including GC (45-47). In addition, NLR is associated with frailty in community-dwelling older adults (13) and older adults with cancer (12).
In our subgroup analysis, patients with NLR ≥3 in the LG group demonstrated a trend toward longer OS, probably because elderly patients with high NLR might be frail due to high inflammation and low immune status and might benefit from less invasive surgery. Our findings point toward the possible clinical application of NLR as it is simple, low cost, and conveniently feasible through a routine blood test. Thus, NLR may be a candidate biomarker to determine frailty and identify patients who would benefit from LG. A previous study showed that LG improved the survival of elderly patients with stage II GC with a good immune prognostic index score, which is a parameter based on the derived NLR and serum lactate dehydrogenase (26). The association between preoperative inflammatory and immune status and survival after LG requires further investigation.
This study has several limitations. Firstly, this was a retrospective observational study conducted at a single institution with a small sample size and a short follow-up period. Therefore, these results may not be definitive. In addition, although missing data might cause bias, they were relatively small in our study. Secondly, to reflect frailty, even though we focused on preoperative comorbidities and systemic nutritional and inflammatory statuses, other factors, such as the Eastern Cooperative Oncology Group Performance Status Scale, may be more suitable. Thirdly, information on the severity of each comorbidity was lacking. Although the severity of comorbidities substantially affects frailty, we believe that the number of comorbidities comprehensively reflects the systemic physical status and frailty of each patient. Lastly, our finding regarding the improved prognosis of elderly patients with a high NLR in the LG group needs to be interpreted with caution because of the results of the subgroup analysis.
Despite these limitations, we believe that our study using real-world data from a regional central hospital in Japan, which has one of the highest incidences of GC (48), has important clinical implications. For elderly patients with frailty, such as those with high NLR, LG may be suitable for GC as well as perioperative rehabilitation, nutritional management, and more minimally invasive surgery such as sentinel node navigation surgery.
Conclusion
The surgical approach was not an independent factor for both short- and long-term outcomes in elderly patients with GC. However, elderly patients with high NLR in the LG group demonstrated a trend toward longer survival. Less invasive surgery such as LG might offer greater survival benefits than OG in frail patients such as those with high NLR which indicates the high inflammation and low immune status. Our study findings will help in the improved screening of elderly patients with GC for the better management of adverse outcomes associated with gastrectomy.
Acknowledgements
The Authors thank Hiroki Yoshida (Data Seed Inc.) for the statistical support and Editage (www.editage.com) for English language editing. This work was supported in part by the Shinnihon Foundation of Advanced Medical Treatment Research (to S.I.).
Footnotes
↵# Current address: Department of Surgery, Fukuoka Tokushukai Hospital, 4-5, Sugukita, Kasuga-shi, Fukuoka, 816-0864, Japan.
Authors’ Contributions
Conceptualization, Shuhei Ito; Data curation, Shuhei Ito, Kippei Ohgaki; Data analysis, Shuhei Ito; Supervision, Eisuke Adachi, Yoichi Ikeda, Yoshihiko Maehara; Writing – original draft, Shuhei Ito; Writing – review & editing, Kippei Ohgaki, Tetsuro Kawazoe, Shota Sato, Shunji Ikeda, Keisei Kakizoe, Huanlin Wang, Toshihiko Nakamura, Shinichiro Maehara, Eisuke Adachi, Yoichi Ikeda, Yoshihiko Maehara. All Authors have read and agreed to the final version of this manuscript.
Conflicts of Interest
The Authors declare no competing interests in this work.
- Received February 19, 2023.
- Revision received March 5, 2023.
- Accepted March 10, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.










