The Effect Of Perioperative Eras Pathway Management On Short- And Long-term Outcomes Of Gastric Cancer Patients (A Single-center Retrospective Study, Propensity Score Matching Method)

Objective:To compare postoperative short-term outcomes and long-term prognosis between perioperative ERAS and conventional pathways protocols in gastric cancer patients.Methods: This is a single institute retrospective cohort study,all patients were pathologically proved to be gastric adenocarcinoma,underwent standard radical gastrectomy with D2 lymphadenectomy during the period of 2007-2012.19 cases received preoperative chemotherapy,15 cases of neuroendocrine carcinoma,12 cases of squamous cell carcinoma,1 case of lymphoma,9 cases of stromal tumor,97 cases existing distant metastasis,154 cases of combined organ resection,298 cases of palliative surgery,58 cases of ascites cytology positive,55 cases of synchronous / metachronous multiple primary carcinoma,42 cases of gastric stump cancer,28 cases of ASA grade Ⅳwere excluded among all 2795 cases.Total 2124 cases were eligible to be analysed and divided into ERAS groups and Non-ERAS group(Traditional programme group)according to the different perioperative pathway protocol.Propensity score matching method(in SPSS,24.0 version,IBM Company)was used to balance the baseline characteristics including age,sex,BMI,tumor location,surgical approach,etc.Two groups were matched in a 1:1 ratio.There were 521 cases per group after matched.The short-term clinical outcomes(postoperative complications,length of hospital stay,blood loss,30-day re-admission rate,etc.)and overall 5-year survival rates were compared between the two groups.The follow-up management of patients after operation was mainly by telephone,out-patient follow-up,checking the records of patients on HIS(hospital information system),etc.The cut-off time was June 2017,and the median follow-up time was 1900 days(30 days 3797 days).For the continuous variables,such as age,length of hospitalization,number of lymph node dissection and tumor size,which accord with normal distribution,they are expressed in the form of mean addition and minus standard deviation,and analyzed by t-test of independent samples.Otherwise,it is expressed as median and quartile spacing and analyzed by rank sum test.The chi-square test(χ~2)or Fisher exact probability method was used to classified variables such as sex,tumor location,operation mode and so on.The survival analysis uses the Kaplan-Meier method to draw the survival curve,tested by the Log-Rank.All statistical tests were two-sided,and P values less than 0.05 were considered significant.Results: The incidence of overall postoperative complications was similar between the two groups(ERAS group = 18.4%,non-ERAS group = 19.4%,P = 0.69).Complications such as anastomotic fistula(Non-ERAS = 1.7%,ERAS = 2.5%,p = 0.39),abdominal bleeding(Non-ERAS = 0.8%,ERAS= 1.0%,p = 0.74),duodenal stump fistula(Non-ERAS = 0.8%,ERAS = 0.6%,p = 0.71),intestinal obstruction(Non-ERAS = 1.2%,ERAS = 1.5%,p = 0.69),gastroparesis(Non-ERAS = 2.1%,ERAS = 1.9%,p = 0.83)were also similar.However,the incidence of SSI(Non-ERAS = 6%,ERAS = 3.3%,p = 0.04),atelectasis(Non-ERAS = 7.3%,ERAS = 4.0%,p = 0.02),thromboembolic diseases(Non-ERAS = 2.5%,ERAS = 0.4%,p < 0.01)is lower in ERAS group.The number of lymph node harvested [Non-ERAS group = 29(11),ERAS group = 32(13),p < 0.01)],operation time [Non-ERAS group = 160(50),ERAS group = 145(13),p < 0.01)],intraoperative bleeding volume(ML)[Non-ERAS group = 300(200)],ERAS group = 200(200),p < 0.01),hospitalization days[Non-ERAS groups = 10(2),ERAS = 7(3),p <0.01] and expenses(RMB)[Non-ERAS = 42214(5139),ERAS = 39789(3969),p <0.01] were superior in ERAS group.There were no significant difference in unplanned reoperation in both groups(Non-ERAS = 3.1%,ERAS = 2.1%,p = 0.33),30-day readmission rate(Non-ERAS = 6.1%,ERAS = 5.6%,p = 0.69)and perioperative mortality(Non-ERAS = 0.4%,ERAS = 0.2%,p = 0.56,P =0.56).The 5-year overall survival rates of Non-ERAS group and ERAS group were 66.2% and 72.8% respectively(Log Rank=7.28,P=0.007).The subgroup analysis found that 5-year OS rates of stage I were 93.4% and 92.7%(Log Rank = 0.11,P=0.73),these of stage Ⅱ and Ⅲ were 82.2% vs 75.2%(Log Rank=7.08,P=0.007)and 47.6% vs 35.7%(Log Rank=5.40,P=0.02)in ERAS group and non-ERAS group respectively.Conclusions: Perioperative ERAS pathway management is safe and feasible for patients with gastric cancer,without increasing the incidence of complications and 30-day readmission rate.This protocol can improve the prognosis of patients with gastric cancer.