The effects of epidural anesthesia and analgesia on cluster of differentiation and cytokines in patients after gastric cancer resection.

Background: Epidural use can prevent peri-operative neuro-endocrine stress responses, improve pain after surgery , and reduce opioid volume which lead to immunosuppression. Methods: Forty cases of patients with gastric cancer were nally enrolled into the study. Patients who received general anesthesia (GA group, n=20), a combination of general anesthesia and peri-operatve epidural use (EGA group, n=20), were given intravenous analgesia or epidural analgesia, respectively. We collected visual analogue scale ( VAS ) scores, length of hospital stay, the rst passage of atus time and incidence of postoperative nausea and vomiting (PONV). We also collected the Cluster of Differentiation (CD)3+, CD4+, CD8+, CD4+/CD8+, Interleukin (IL)-4, IL-6, Interferon (IFN)-γ before surgery, as well as the postoperative days 1, 3 ,7. Results: VAS scores and PONV in GA group were higher than EGA group in the postoperative 3 days. CD3+, CD4+, CD4+/CD8+ were declined in the postoperative 3 days, and nearly recovered to the baseline in the seventh days after surgery in both of the two groups. CD3+ in the GA group decreased more than EGA group. IL-4, IL-6, IFN-γ were increased in the postoperative 3 days, and nearly recovered to the baseline in the seventh days after surgery in both of the two groups, IL-4, IL-6 in the GA group increased more than EGA group. IFN-γ in the EGA group increased more than GA group. Conclusions: A combination of general anesthesia and peri-operatave epidural use can relieve postoperative pain and PONV. A combination of general anesthesia and peri-operatave epidural use decreases immunosuppression in gastric cancer resection.


Background
The stress response induced by surgery can activate the immune regulation mechanism during a systemic in ammatory reaction. [1][2][3][4] There are many important contributors to the tumor microenvironment, such as cytokines, chemokines, in ammatory mediators, which exists in many stages of progression to metastasis. [5][6][7] Some clinical factors, such as general anaesthetics, postoperative pain, and opioid analgesia, have been recognized as immunosuppressive and have in uenced the development of tumours. [8][9][10][11][12][13][14] In some studies, epidural anaesthesia (EA) was associated with improved overall survival in patients with gastric cancer; in other studies, EA did not improve overall survival. [15][16][17] EA may reduce cytokines and neuro-endocrine stress immune stimulation, prevent nerve impulses, decrease excitability of the sympathetic adrenal medulla axis, reduce cortisol production, and improve the function of T lymphocytes. [18,19] Among cytokines, interleukin (IL)-4, IL-6, and interferon (IFN)-γ play prominent roles in chronic in ammation, autoimmunity, infectious diseases and cancer, where they often act as diagnostic or prognostic indicators of disease activity and response to therapy. [20,21] Therefore, we designed this study to explore the interaction between epidural use and the aforementioned cytokines in patients with gastric cancer.

Methods
Patient identi cation and exclusion: This study was a single-centre, randomized, observer-blinded study. Inclusion criteria were: (1) American Society of Anesthesiologists (ASA) I-, (2) gastric cancer was diagnosed bygastroscopy and biopsy in our hospital, (3) patients underwent radical resection of gastric cancer in our hospital. Exclusion criteria were: (1) emergency operations, (2) laparoscopic procedures, (3) neoadjuvant treatment,(4) abnormally white blood cell count, (5) peri-operative transfusion, (6) severe heart, lung, liver, kidney, or endocrine diseases, After con rming understanding of the recruiter's description of the trial, all the patients signed informed consent forms. Anaesthesia and analgesia protocols were standardized as follows. Patients were randomly assigned to the general anaesthesia (GA) group or epidural anaesthesia (EGA) group by computer-generated codes at the beginning of the study.
The clear primary outcome of this study was in ammatory response, which was evaluated via the concentration of IL-6. Previously published studies on IL-6 suggested that its standard deviation (SD) in vivo is in the order of 7.6 pg/ml. Twenty patients were calculated based on detecting a reduction of 6.8pg/ml deviation on a-value 20% and a power of 95%. To compensate for potential dropouts, we enrolled Twenty-ve patients.
Anaesthesia technique and grouping method: On arrival to the operating room, patients were monitored via electrocardiogram and blood pressure as well as pulse oximetry. Patients randomly assigned to the Oesophageal temperature was monitored and maintained above 36℃. Patient-controlled intravenous analgesia (PCIA) with sufentanil (0.5 μg/ml) was available for 72-120 hours. The PCIA protocol of the sufentanil group (S group) consisted of 0.5 μg/ml sufentanil (total 300 ml) with a background infusion rate of 4 ml h-1, a bolus dose of 3 ml, and a lockout time of 15 min. Patients in the EGA group were given T8-10 epidural anaesthesia before general anaesthesia. An infusion of 0.5% ropivacaine (AstraZeneca Pharmaceutical Co., Shanghai, China) was administered during surgery, and the loading dose (0.5% ropivacaine, 5-7ml) depended on the height and weight of the patient. GA was induced with 0.05 mg/kg midazolam, 0.2 μg/kg sufentanil, 1-1.5 mg/kg propofol, and 0.15 mg/kg cisatracurium, and anaesthesia was maintained with propofol when the bispectral index (BIS) was 40-60. Patient-controlled epidural analgesia (PCEA) with a combination of 0.2% ropivacaine and 0.5 μg/ml sufentanil was available for 72-120 hours. Acute rescue analgesic medications could be used when the VAS score was more than 4 and a bolus of 100 mg tramadol, administered via intravenous injection, was required more than 3 times.
Indicator and data: A physician who was blinded to the group assignment assessed postoperative pain intensity using the visual analogue scale (VAS) on postoperative days 1, 2, and 3. The demographic data, cancer stage, degree of differentiation, duration of the operation, length of hospital stay, time of the rst passage of atus and incidence of postoperative nausea and vomiting (PONV) were recorded. We also collected data on cluster of differentiation markers (CD)3+, CD4+, CD8+, and CD4+/CD8+, Interleukin (IL)-4, IL-6, and Interferon (IFN)-γ on the day before surgery, day (d) 0, and on postoperative days 1, 3, and 7.
Statistical approach: Statistical analysis was performed using SPSS version 22.0 for Windows (IBM Corp., USA). Normally distributed data were expressed as the means ± SDs. Categorical variables were described using frequencies and were analysed using the 2 test. Fisher's exact test was used for small sample sizes (expected frequencies <5). We checked for normality of the data with the Shapiro Wilk test and used a one-way ANOVA between the two groups. The results with P<0.05 were considered statistically signi cant.

Patient characteristics
Between November 12, 2017, and December 15, 2017, fty patients were screened for this study, and forty patients were ultimately included in this study. Five patients with metastases or peri-operative transfusions were excluded. Five patients had their analgesic regimen modi ed because of hypotension and severe nausea ( Figure 1). The two groups were similar with respect to age, height, weight, gender, ASA grade, cancer stage and degree of differentiation, duration of surgery, surgical procedure and surgical manner between the two groups (Table 1).

Association between epidural use and postoperative variables
The VAS scores in the GA group were higher than in the EGA group on postoperative day 3, and the incidence of PONV in the EGA group was lower (P<0.05, Table 2). There were no differences in days of analgesia, the time of the rst passage of atus, and length of hospital stay. CD3+, CD4+, and CD4+/CD8+ T cells declined on postoperative day 3 (P<0.05) and nearly recovered to baseline seven days after surgery in both groups. CD3+ T cells decreased more in the GA group than in the EGA group (P<0.05, Table 3). IL-4, IL-6, and IFN-γ increased on postoperative day 3 (P<0.05; Figures 2, 3, and 4) and nearly recovered to baseline seven days after surgery in both groups. IL-4 and IL-6 increased more in the GA group than in the EGA group (P<0.05, Figures 2 and 3). IFN-γ increased more in the EGA group than in the GA group (P<0.05, Figure 4).

Discussion
In our study, we found a bene cial effect of epidural use. The VAS scores in the GA group were higher than those in the EGA group on postoperative day three. The incidence of PONV was higher in the GA group. It has been proven that epidural anaesthesia can provide better analgesia after gastric cancer surgery, which was consistent with our study. [22] Epidural administration of anaesthesia and analgesia is considered a technique with risk of complications, such as neuraxial haematoma, hypotension, pruritus; the subjective experience of anaesthetists often leads to the failure of epidural anaesthesia and analgesia. [23,24] In our study, there were two patients in the EGA group and one patient in the GA group who could not use EA because of hypotension.
Many studies have reported that opioids impair the peri-operative immune system and increase vascular permeability, and the use of epidural anaesthesia may prevent these peri-operative immunosuppressive changes during major surgery because it can decrease neuro-endocrine stress responses. [18,[25][26][27][28] Immune surveillance is the primary indicator for stopping the metastasis of tumours, and immunosuppression may destroy the defensive barrier. [29] Clinical events that may lead to immunosuppressive changes facilitated by surgery include injury, pain, and use of anaesthetic medications. [30][31][32] CD3+ T cells, CD4+ T cells, and CD8+ T cells are the main cells which take part in antitumour immunity. [33] The ratio of CD4+/CD8+ cells decreases as the serum cortisol levels increase. [34,35] In our study, CD3+, CD4+, and CD4+/CD8+ T cells were inhibited and had a negative trend. All patients' CD3+ and CD4+ T cells decreased to different degrees on the rst postoperative day, but in the GA group, patients' CD3+ T cells signi cantly decreased and recovered to preoperative levels until the seventh postoperative day. The change in CD8+ T cells was not statistically signi cant. These results are consistent with some prior studies, [36,37]they described a lymphocyte depressing factor that was present in the serum of patients in the GA group but was absent in the serum of patients in the EGA group.
Many cytokines can modulate the immune system. Pro-in ammatory cytokines may favour tumour progression. Activated biological cascades lead to immunosuppression, which affects the immune response and decreases IL-4 and IFN-γ. [38,39] As a pro-in ammatory factor, IL-4 stimulates the proliferation of B cells and participates in the differentiation of Th2 cells. The combination of the cytokine IL-6 with prostaglandin 2 can reduce the production of the immune factor IL-2 by Th1 cells and affect the activation of NK cells. [40] IL-6 has a major in uence on the proliferation, survival and metastatic properties of cancerous cells, and studies using preclinical mouse models indicate that treatments targeting IL-6 or its receptor display therapeutic e cacy as anticancer agents. [20]In our study, the pro-in ammatory cytokines IL-4 and IL-6 increased more in the GA group than in the EGA group, which may indicate that immune function was less suppressed in the EGA group. IFN-γ can kill tumour cells directly and transform Th0 cells into Th1 cells, which play a large role in the activation of NK cells and T cells. The level of IFN-γ expression embodies the memory ability of this antitumour cytokine. [41,42] In our study, IFN-γ increased more in the EGA group than in the GA group, which may indicate that immune function was less suppressed in the EGA group. Therefore, in our study, EA was able to relieve postoperative pain and PONV, was able to decrease immunosuppression, and may have reduced in ammatory-associated metastasis in gastric cancer.
There were a few limitations in our study. First, we only measured some T cell subsets and some important immune factors. The immune system is a complex system. There are many immune cells, such as NK cells, and other immune factors, such as IL-2 and IL-12, which may be investigated in the future because these antitumourigenic cytokines were increased by EA. Second, these patients' long-term outcomes, such as long-term survival, recurrence rates and metastasis rates, could not be obtained at present. We will evaluate these results in 3 to 5 years. Hospital. This study was registered prospectively at http://www.chictr.org.cn/index.aspx on October 10, 2017 (Registered ChiCTR-INR-17012939). The purpose of the study was explained to the family of patients under the study. The written consent was obtained from study participants. The patients were informed that the care to be given was not be compromised in any way and con dentiality was assured. Name and other identifying information were not used in the study.
Not applicable.

Availability of data and material
All data generated or analysed during this study are included in this published article, and can be freely available to any scientists wishing to use. The data supporting our ndings can be found in http://www.chictr.org.cn/index.aspx (Registered ChiCTR-INR-17012939), but identifying patient data would not be shared.

Competing Interests
The authors declare that they have no con icts of interest.  EGA=epidural anesthesia combined with general anesthesia group.
ASA=American Society of Anesthesiologists.

Figure 3
Comparison of IL-6 between the general anaesthesia group and epidural combined with general anaesthesia group on day (d) 0, the day before surgery, and on postoperative days 1, 3, and 7.

Figure 4
Comparison of IFN-γ between the general anaesthesia group and epidural combined with general anaesthesia group on day (d) 0, the day before surgery, and postoperative days 1, 3, and 7.

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