Impact of postoperative complications on long-term survival of gastric cancer patients: results from a high-volume institution in China

Background This study was aimed to evaluate the impact of postoperative complications (POCs) on long-term survival for gastric cancer (GC) patients with curative resection. Methods From January 2009 to December 2014, a total of 1667 GC patients with curative gastrectomy were analyzed. Patients with any complications Clavien–Dindo (CD) grade II or higher were divided into complication group. Independent risk factors for the development of POCs and the relationship between POCs and long-term survival (excluding death within 90 days after surgery) were analyzed. Results Overall POCs CD ≥ 2 were diagnosed in 285 (17.10%) patients including infectious complications (ICs) in 231 (13.9%) and noninfectious complications (NICs) in 78 (4.68%) patients. Age ≥ 65 (P = 0.003), presence of comorbidity (P = 0.019), extensive lymphadenectomy (P = 0.027) and perioperative blood transfusion (P = 0.040) were independent risk factors of POCs. Multivariate analysis identified that presence of POCs (P < 0.001) was an independent prognostic factor and further analysis by complication type demonstrated that the deteriorated overall survival was mainly caused by ICs (P = 0.007) rather than NICs (P = 0.075), moreover, among all complications, pulmonary infection (P < 0.001) was the only significant prognostic factor.


Introduction
Gastric cancer (GC) is the fifth most frequently diagnosed cancer and the third leading cause of cancer death worldwide [1]. Gastrectomy with curative resection is the most effective treatment strategy which provides the chance to dramatically extend the long-term survival of gastric cancer patients. However, surgery for gastric cancer still remains technically demanding and the following postoperative complications (POCs) have been reported to be with a wide range of incidence of 7-46% [2][3][4][5].
Recent studies have shown that POCs increase the length of hospital stay and early mortality [6,7].
Moreover, POCs decrease the overall survival (OS) and disease free survival (DFS) in several types of cancer like lung, breast and colon [8][9][10], and this effect might is mainly strengthened by infectious complications (ICs) not noninfectious complications (NICs) [11][12][13]. Therefore, the prevention and treatment of POCs may play a crucial role to improve the surgical outcomes of cancer patients.
Currently, in gastric cancer, although increasing numbers of studies have suggested that presence of POCs is adversely correlated with long-term survival [14][15][16][17][18], several studies reported from both eastern and western countries have negative findings [19][20][21][22]. What's more, which kind of specific complication in gastric cancer would have a crucial effect on the long-term survival has only been discussed in few studies [21,23] and remains far from resolved.
The aim of this study was to explore the relationship between POCs and the long-term survival,and to identify factors associated with the development of POCs. Hospital. The inclusion criteria included: 1) histologically proven gastric adenocarcinomas; 2) with radical surgical resection (R0); 3) without preoperative therapy; 4) no distant metastasis. The exclusion criteria of our study included: 1) with other synchronous or metachronous (within 5 years) cancers; 2) remnant gastric cancer; 3) with harvested number of lymph node less than 15; 4) with emergency treatment. Additionally, patients died within 90 days (all the death was directly associated with serious intra-or post-operative complications) after surgery were also excluded to avoid exaggerating the effect of POC on long-term survival. Finally, 1667 patients who underwent gastrectomy with potentially curative resection were enrolled into this study, as shown in Fig. 1.

Patients and ethical issues
Potentially curative resection is regarded as gastrectomy with R0 resection combined with adequate lymphadenectomy. The surgery was performed by experienced surgeons according to the Japanese gastric cancer treatment guidelines [24]. The resected specimens were pathologically classified according to the JGCA classification [25], and staged with the updated AJCC 8th TNM system [26].

Assessment of postoperative complications
In the present study, the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 [27], which is enough exhaustive in terms of postoperative morbidities, was used to define complications, and any morbidity classified as Clavien-Dindo (CD) grade II or higher that developed during hospitalization or within 30 days after surgery was identified as a POC using Clavien-Dindo grade system [28,29]. However, grade I complications were not evaluated considering the potential description bias in the patient records. If a patient suffered more than one POC, the highest-ranked complication was used for grade analysis.

Follow-up
The follow-up was mainly performed by out-patient visit, meanwhile, routine telephone interviews were regarded as important supplementary methods. The last update of follow-up information was Jan

Statistical Analysis
The analyses were all performed with software IBM SPSS Statistics version 23.0 (International Business Machines Corporation, Armonk, NY, USA). Statistical analysis was performed using descriptive statistics for demographic data. For group-wise comparisons, Student t test (normal distribution) or Mann-Whitney U test (nonparametric distributions) were used. To compare nominal data, the Chi-square test or the Fisher exact test was used. The subsequent multivariate logistic regression analysis was performed to detect independent risk factors of POCs. The Kaplan-Meier method and log-rank test were used to calculate survival rates and compare the survival rates respectively. Univariate and multivariate Cox proportional hazards regression models were used to analyze the hazard ratios for overall survival. P value less than 0.05 was considered to be statistically significant.

Predictors related to occurrence of POCs
Relationships between the occurrence of POCs and clinicopathological parameters are shown in   (Table 3).

Discussion
Several previous studies have reported the relationship between POCs and long-term survival in gastric cancer, however, the results still remain controversial [11, 15-17, 20-23, 30-36]. Additionally, these studies failed to exclude patients who died in a short postoperative period. Of note, POCs increase the early mortality, which would overshadow the real influence of complications on long-term survival of cancer patients [37,38]. In the present study, 1667 GC patients with curative resection were analyzed and 17.10% of them occurred POCs C-D ≥ II (excluding deaths within 90 days after surgery). Older age, comorbidities, extensive lymph node dissection and perioperative blood transfusion were associated with a high incidence of POCs. With respect to the relationship of complication and survival, we revealed (1)  With respect to the mechanism of presence of complications on poor survival, Accumulated evidences [17,40,41] had indicated that the surgical stress, especially in major surgery, would induce body inflammatory response which could be enhanced and prolonged by POCs. It's also well established that postoperative inflammatory response has a contribution on host immunosuppression by mainly suppressing cell-mediated immunity [42,43], especially natural killer cells and cytotoxic T lymphocytes are compromised [43], which promotes the proliferation and metastasis of residual tumor cells. Furthermore, numerous researches have confirmed that ICs have a direct effect on cancer cell metastatic ability by activating bacterial antigen-mediated processes [44,45]. Indeed, in our study, a remarkable difference in overall survival between patients stratified by presence of complication in p-Stage III is likely to reflect the quantity of residual tumor cells which would cause early recurrence.

Consent for publication:
Not applicable.

Availability of data and materials:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests:
The authors declare that they have no competing interests.    Figure 1 The flow diagram of gastric cancer patients enrolled in this study

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. TableS1.docx