Association Between Postoperative Atrial Fibrillation And Esophagectomy

Objective Postoperative atrial fibrillation(POAF)is the most common complication of esophageal cancer resection.In recent years,minimally invasive surgery has been widely used in esophageal cancer resection,of which robot assistant minimally invasive esophageal cancer resection(RAMIE)is a clinical hot spot.However,it is not clear whether there is a correlation between the surgical procedure and POAF.The purpose of this study is to evaluate the association between surgical procedure and POAF,in addition its impact on prognosis.Method This study is divided into two stages.The first stage: a single-center cohort study,which included patients diagnosed with esophageal cancer and underwent tumor resection in a medical center from January2005 to April 2012.Patients were divided into open esophagectomy(OE)group and minimally invasive esophagectomy(MIE)group,in which MIE included thoracoscopic and thoracoscopic combined laparoscopic.The primary outcome variable was POAF,and the secondary outcome variables were length of hospital stay(LOS),one-year mortality and long-term mortality(last follow-up: May 31,2014).Demographic,comorbidities,drugs,preoperative evaluation,tumor and operation-related information were collected as covariates.Multivariate logistic regression was used to model POAF,one-year mortality rate and long-term mortality rate,and negative binomial regression was used to model LOS.The correlation between surgical procedures and POAF after covariate correction and its impact on prognosis were observed.The second stage: a national multi-center cross-sectional study,using the national inpatient sample(NIS)to collect patients diagnosed with esophageal cancer and undergoing surgical resection from 2011 to 2014.First of all,they were divided into standard minimally invasive esophageal cancer resection group(SMIE),OE group and robot-assisted esophageal cancer resection group(RAMIE).SMIE includes thoracoscopy,laparoscopy and mediastinoscopy.RAMIE is defined as esophagetomy assisted by Da Vinci robot.The main outcome variable is POAF.Secondly,patients were divided into POAF group and control group.Secondary outcome variables were in-hospital mortality,LOS and medical expenses.The international classification of disease,9th revision code(ICD-9)was used to collect age,gender,race,complications,radiotherapy and chemotherapy,medical expenses payment,patient location,hospital volume and its geographical location,and hospital teaching as covariates.Multivariate logistic regression was used to model POAF and in-hospital mortality.Negative binomial regression and generalized linear model were used to model LOS and medical expenses,respectively.To observe the correlation between surgical procedures and POAF,and the influence of POAF on LOS,in-hospital mortality and medical expenses.Use ICD-10 code to collect the variables,using NIS2016 database.Analysis was conducted with merging data of 2011-2014 and 2016.Comparable results were reached,suggesting the robustness of this study.Result484 patients with esophageal cancer(421 in OE group vs.63 in MIE group)were included in the first stage.A total of 53 patients had POAF.The median LOS was 12 days.9.3%(45 patients)died within one year postoperation,and the long-term mortality rate was 22.1%(107 patients).There was no statistical difference between the outcome variables: POAF,OE(11.4%)vs.MIE(7.9%),P = 0.52;LOS,OE(median 12.0 days)vs.MIE(median 13.0 days),P = 0.30;One-year mortality rate,OE(9.0%)vs.MIE(11.1%),P = 0.60;Long-term mortality,OE(22.8%)vs.MIE(17.5%),P = 0.34.After covariate adjusted,MIE is associated with a reduced POAF risk(Adjusted odds ratio [a OR]: 0.04,95% CI: 0.01,0.24,P < 0.01),and negatively correlated with LOS(adjusted incidence rate ratio [airr]: 0.77,95% CI 0.68,0.87,P < 0.01).There was no correlation between surgical methods and mortality.In the second stage,3181 patients with esophageal cancer were enrolled and divided into SMIE group(174 cases),OE group(2728 cases)and RAMIE group(279 cases).POAF was found in 25.4%(808 cases).There were statistical differences in POAF among groups: SMIE group(8.0%)vs.OE group(26.6%)vs RAMIE(24.7%),X2 =29.71,P < 0.01.According to incidence of POAF,the patients were divided into groups(2373 cases in the control group vs 808 cases in the POAF group).The median LOS was 10.0 days,3.5% of patients(112 cases)died during hospitalization,and the median medical expense was 39,780.2 $.There were statistical differences in LOS,in-hospital mortality and medical expenses between groups: LOS,median 10.0 days in control group vs.13.0days in POAF group,P < 0.01;In-hospital mortality rate,control group(2.8%)vs.POAF group(5.7%),P < 0.01;The median medical expenses in the control group was37374.3 $ vs.48835.2$ in the poaf group,P < 0.01.After covariate adjusted,compared with SMIE,OE and RAMIE increased POAF risk(OE,a OR: 6.02,95% CI: 1.92,18.92,P < 0.01;RAMIE,a OR: 5.23,95% CI: 1.58,17.26,P = 0.01).After covariate adjusted,POAF increased the risk of in-hospital death(a OR: 2.01,95%CI: 1.06,3.81,P = 0.03),prolonged LOS(a IRR: 1.23,95%CI 1.10,1.37,p < 0.01),and increased medical expenses(control group,Margin = 47465.2$,95% CI : 44905.5$,50024.9$;POAF group,Margin = 61947.4$,95% CI:56198.9$,67695.8$).The 2016 data retrieved by ICD-10 was merged with the 2011-2014 data,the results were similar to those before.After covariate adjusted,compared with SMIE,OE and RAMIE increased POAF risk(OE,a OR: 4.93,95% CI: 1.80,13.54,P < 0.01;RAMIE,a OR: 4.13,95% CI: 1.46,11.69,P = 0.01).Compared with the control group,POAF increased the risk of in-hospital death(a OR: 1.80,95% CI: 1.05,3.08,P = 0.03),prolonged LOS(a IRR: 1.21,95%CI 1.10,1.34,P < 0.01),and increased medical expenses(control group,Margin = 48147.7$,95% CI : 45859.9$,50435.5$).POAF group,Margin = 60880.8$,95% CI:55704.3$,66057.3$).Conclusion Compared with OE,MIE can be used as an alternative procedure that does not increase the long-term mortality but can reduce the risk of POAF.Further research shows that RAMIE may increase the risk of POAF compared with SMIE,which in turn leads to a prolonged LOS,increased mortality and increased medical expenses.Sensitivity analysis indicates the robustness of the conclusion.