Abstract
Background: The aim of this study was to evaluate the impact of postoperative infectious complications on long-term outcomes after curative resection of gastric cancer. Patients and Methods: Patients who underwent curative gastrectomy with lymphadenectomy for gastric cancer at Yokohama City University and Kanagawa Cancer Center from January 2000 to August 2015 were retrospectively selected from medical records. Clinicopathological factors between patients with and without infectious complications were compared. Prognostic factors of long-term survival were analyzed by univariate and multivariate Cox proportional hazards analyses. Results: A total of 2,254 patients were eligible for inclusion in the present study. Fifty-eight patients had postoperative infectious complications (IC group); 2,196 had no postoperative infectious complications (NC group). In the IC group, the median age (p=0.031), body mass index (p=0.004), American Society of Anesthesiologists physical status (p=0.006) and percentage of male patients (p<0.001) were higher in comparison to the NC group. The operation time was longer (p<0.001) and the incidence of intestinal-type histology was higher (p=0.017) in the IC group. The 5-year overall survival rates of the IC and NC groups were 59.8% and 83.2%, respectively (p<0.001). Univariate and multivariate analyses demonstrated that postoperative infectious complications were a significant risk factor for poorer overall survival (hazard ratio=2.38; 95% confidence interval=1.47-3.85, p<0.001). Conclusion: Perioperative management is necessary to reduce the incidence of postoperative infectious complications and improve the survival of patients after curative resection of gastric cancer.
In 2018, gastric cancer was the fifth most common type of cancer, with 1,033,701 new cases, and, with 781,631 gastric-cancer related deaths, was the third-leading cause of cancer death, (1). Complete surgical resection is indispensable for curing localized gastric cancer. However, 20-60% patients develop recurrent disease, even if complete resection is achieved (2, 3). Although several postoperative adjuvant chemotherapies have been established, their effect on the long-term prognosis remains insufficient. The risk of recurrence and mortality after surgery may be influenced by the aggressiveness of the tumor and the extent of surgery (4-7). Several studies have reported that postoperative complications have a negative impact on the long-term prognosis of various types of malignancies, including gastrointestinal ones (8-11). The morbidity and mortality rates after radical gastric surgery are reported to be 20-30% and 1-5%, respectively (12, 13). As with other malignancies, several studies have reported that postoperative complications such as anastomotic leakage have a negative impact on the long-term prognosis of gastric cancer patients (14, 15). In addition, postoperative infectious complications are reported to be an independent prognostic factor in gastric cancer patients (16, 17).
Although there is room for discussion as to why postoperative complications affect long-term prognosis, several studies have reported that prolonged inflammation affects host immunosuppression, leading to the growth of micrometastases, and increased deaths due to other causes. Goldfarb et al. reported that treatment aimed at the perioperative enhancement of cell-mediated immunity with the simultaneous inhibition of excessive catecholamine and prostaglandin responses was successful in limiting postoperative immunosuppression and metastatic progression (18). In addition, Dunn et al. suggested that the adaptive immune system can function by identifying and eliminating nascent tumor cells in experimental models (19).
However, the extent to which postoperative infectious complications influence the long-term prognosis of patients with gastric cancer is still unknown. Moreover, there have been no multicenter large data studies in the past. The aim of this study was to determine how much postoperative infectious complications influence long-term prognosis in patients after curative resection of gastric cancer.
Patients and Methods
Patients. Data were retrospectively collected on patients with histologically proven gastric adenocarcinoma and who underwent curative gastrectomy with D2 or D1+ lymphadenectomy as the first treatment at Yokohama City University and Kanagawa Cancer Center between January 2000 and August 2015.
In principle, D1+ and D2 gastrectomy was performed for clinical T1 and T2–T4 disease, respectively, according to the Japanese Gastric Cancer Treatment Guidelines version 3 (20). Pathological stage was histopathologically diagnosed according to the third English edition of the Japanese Classification of Gastric Carcinoma (21).
The patients who underwent radical gastrectomy between January 2000 and August 2006 underwent surgery alone. After August 2006, patients with pathological stage II and III disease received S-1 adjuvant treatment based on the Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC), which became the standard treatment for advanced gastric cancer (20, 22).
Patients were followed up at outpatient clinics. The dates of recurrence and death were collected from the follow-up data. All patients underwent hematological tests including carcinoembryonic antigen and carbohydrate antigen 19-9 tumor markers, physical examinations and computed tomographic examination regularly until 5 years after surgery.
Definition of postoperative infectious complications. Grade 3-5 postoperative complications during hospitalization or within 30 days after surgery were retrospectively determined according to the Clavien–Dindo classification (23). In order to exclude the possibility of a description bias in the patient's records and the overestimation of infectious complications, grade 1 and 2 complications were not included. Anastomotic leakage was diagnosed clinically and radiologically based on a water-soluble swallow test. Pancreatic fistula was diagnosed when the amylase content of the drain output was three times higher than the serum level after postoperative day 3 or when fever/an increase in inflammatory markers such as white blood cells and C-reactive protein was observed with fluid collection without air density around the pancreas on computed tomography. Wound infection was diagnosed when signs of infection, such as redness and swelling or purulent discharge from the wound were observed. Other infectious complications were defined as being present in patients who received antibiotic therapy for an infection or suspected infection due to pneumonia, lymphatic fistula, abdominal abscess and other causes. Patients were classified into two groups according to the presence (IC group) and absence (NC group) of postoperative infectious complications.
Statistical analyses. Fisher's exact test or the chi-squared test and the Mann–Whitney U-test for continuous variables were used to compare the IC and NC groups. Recurrence-free survival (RFS) was defined as the period between surgery and recurrence or death, whichever came first. Overall survival (OS) was defined as the period between surgery and death. The Kaplan–Meier method was used to calculate the OS and RFS curves, and compared by the log-rank test. Cox's proportional hazard model was used to evaluate the OS by univariate and multivariate analyses. p-Values of less than 0.05 were considered statistically significant. All statistical analyses were performed using SPSS software (ver. 23.0; IBM Corp., Armonk, NY, USA). This retrospective study was approved by the Institutional Review Board of the Yokohama City University (IRB number: B160707003) and Kanagawa Cancer Center (2016.epidemiologic study-22).
Results
Patient characteristics. A total of 2,254 patients were eligible for inclusion in the present study. The median age was 65 years, 68% of the patients were male. Twenty-four percent of the tumors were located in the upper third of the stomach. Patients underwent open surgery and D2 lymphadenectomy 75% and 80%, respectively. Fifty-eight patients had infectious complications of grade 3 or more and were classified into the IC group; 2,079 had infectious complications grade 2 or less and were classified into the NC group. Table I summarizes the patients' clinicopathological characteristics. The median age (p=0.031), percentage of male patients (p<0.001) and body mass index (p=0.004) of the IC group were higher in comparison to the NC group. D2 lymphadenectomy was performed more frequently (p=0.005) in the NC group, while the operative time was longer (p<0.001) in the IC group. However, infectious complications were not associated with pathological T factor (p=0.314) or pathological N factor (p=0.916).
Postoperative infectious complications. Fifty-eight out of 2,254 patients had postoperative complications; the total number of events was 75 (3.3%). The overall surgical mortality rate was 0.2% (five patients). The details are shown in Table II. The most frequent postoperative infectious complication was anastomotic leakage (1.4%), followed by surgical site infection (0.8%) and pancreatic fistula (0.6%). Clavien–Dindo grade 3 and 4 complications were observed in 63 (2.7%) and seven (0.3%) patients, respectively.
Survival analyses. The median follow-up time was 60 (range=0-177) months considering the 2,254 patients. Recurrence was seen in 289 patients. Two hundred and thirty-eight out of the 406 patients who died in this period died of gastric cancer. The 5-year RFS rates of the IC and NC groups were 59.3% and 81.2%, respectively (p=0.002). The corresponding 5-year OS rates were 59.8% and 83.2%, respectively (p<0.001). The RFS and OS curves are shown in Figures 1 and 2. The RFS and OS were significantly worse in the IC group with an unadjusted hazard ratio (HR) for infectious complications of 2.06 [95% confidence interval (CI) 1.30-3.23; p=0.002) and 2.32 (95% confidence interval (CI)=1.45-3.73; p<0.001], respectively. The univariate and multivariate analyses demonstrated that infectious complications were a significant prognostic factor for both RFS (HR=1.41, 95% CI=1.02–1.96; p=0.038) and OS (HR=2.38, 95% CI=1.47–3.85, p<0.001), respectively (Tables III and IV).
Discussion
The present study aimed to evaluate the impact of postoperative infectious complications on long-term outcomes after curative resection of gastric cancer using a large multicentral dataset. The major finding of the study was that postoperative infectious complications, which induce a prolonged inflammatory response after surgery, had a negative impact on both RFS and OS after curative resection of gastric cancer. Thus, careful treatment management to reduce the incidence of postoperative complications might be necessary.
The present study showed that the adjusted HR of OS in patients with postoperative infectious complications was 2.38 (95% CI=1.47-3.85, p<0.001). Kubota et al. examined 1,395 patients and reported that postoperative complications, including not only infectious but also non-infectious complications, had an obvious impact on both OS and disease-specific mortality in patients with gastric cancer. They reported that in regard to OS, patients with postoperative complications had an adjusted HR of 1.88 (95% CI=1.26-2.80, p=0.0018) (17). In addition, Tokunaga et al. examined 765 patients and reported that intra-abdominal infectious complications adversely affected OS and RFS in patients with gastric cancer. They reported that the hazard ratio for those with intra-abdominal infectious complications was 2.44 (95% CI=1.475-4.060, p<0.001) (24). Thus, our study confirmed the results of previous studies.
When comparing the patient background factors between patients with and without IC, there were some differences. Firstly, delays in the initiation, or lack of compliance with adjuvant treatment was a possible reason for the increased rate of recurrence and worse survival of patients with postoperative infectious complications (17, 25). Kubota et al. reported that the planned number of treatment cycles was administered to 62.5% and 66.4% of patients with and without complications, respectively. In addition, adjuvant treatment was initiated at 7 and at 6 weeks after surgery in patients with and without complications, respectively. In the present study, patients with postoperative infectious complications tended to lack compliance with adjuvant treatment, as this was observed in 595 patients for whom adjuvant treatment was indicated (HR=3.357, 95% CI=0.890-12.659, p=0.074). Secondly, systemic immunosuppression induced by postoperative infectious complications might have worsened comorbidities such as cardiovascular diseases, pulmonary diseases, renal dysfunction and hepatic dysfunction, leading to death from other causes (17, 26-32). Kubota et al. reported that causes of non-cancer death in patients with pStage I/II disease included cardiovascular disease, cerebrovascular disorders, and pulmonary diseases. In the present study, among the 1,542 patients with comorbidities, patients observed postoperative infectious complications tended to die of other cause. Thirdly, mediators induced by inflammation might enhance the migration and invasion of malignant cells, leading to metastasis (18, 19, 33). However, the abovementioned mechanism was not evaluated in the present study, and further studies are needed to analyze this possibility.
On the basis that prolonged inflammation possibly affects immunosuppression, leading to the growth of micrometastases and increasing death due to other causes, we did not include non-infectious complications such as anastomotic stenosis, liver dysfunction and renal dysfunction, which seem to have no induction of an inflammatory response. In addition, patients who received antibiotic therapy not only for obvious infectious complications but also for suspected infection were considered to have grade 2 complications, which might have led to the overestimation of postoperative infectious complications. Thus, we set the criteria for inclusion in the postoperative infectious complications group as Clavien–Dindo grade 3 or more, although there were no specific criteria for this cut-off value. In present study, Clavien–Dindo grade 3 or more postoperative infectious complications were observed in 58 (2.6%) patients. The most frequent postoperative infectious complication was anastomotic leakage (1.4%), followed by surgical site infection (0.8%) and pancreatic fistula (0.6%). The morbidity rate in this study was 0.2%, which was similar to those in other large studies (34, 35).
In terms of the follow-up period and sample size, the present results are considered to be reliable. However, several limitations were associated with present study. Firstly, this was a retrospective cohort study that might have a selection bias. Secondly, since the study period was long (2000–2015), perioperative management such as surgical procedures, perioperative care, and adjuvant treatment might have changed over the years in which the data were collected (34, 36). Thirdly, as the extent of biological response to postoperative complications varies depending on individual characteristics, the data used to evaluate the degree of the inflammatory response (e.g. the white blood cell count and C-reactive protein value) were not collected. Thus, the effect of the extent of the inflammatory response on the prognosis is unknown.
In conclusion, the presence of postoperative infectious complications had an obvious impact and was an independent prognostic factor for both OS and RFS after curative resection of gastric cancer. It is necessary to manage perioperative care in order to reduce the incidence of postoperative infectious complications.
Acknowledgements
The Authors express their sincere gratitude to Ms. Minako Igarashi, Ms. Yuka Maruyama, Ms. Mariko Yamauchi, Ms. Akiko Yoshida, Ms. Natsumi Sato and Ms. Rika Takahashi for their excellent data management in this study.
Footnotes
↵* These Authors contributed equally to this study.
Authors' Contributions
Substantial contributions to conception and design were made by YM, TA and YR. YM, Substantial contributions to acquisition of data, or analysis and interpretation of data were made by TA, MJ, KK, KK, SS, MN, TH, TY, HT, TS, TO, HC, TO, NY, TY, and YR. Drafting the article or revising it critically for important intellectual content were made by YM, TA, TH, TY, NY, TY and YR. Final approval of the version to be published have given by TA, YR, and MM. Each Author participated sufficiently in the work to take public responsibility for appropriate portions of the content; and agreed to be 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. All Authors read and approved the final article.
Conflicts of Interest
The Authors declare no conflicts of interest in association with the present study.
- Received March 30, 2020.
- Revision received June 15, 2020.
- Accepted June 16, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved