Clinical Study On Post-endovascular Repair Syndrome Of Aortic Aneurysm

ObjectiveAfter endovascular aortic repair(EVAR), the occurrence of post-endovascular repair syndrome of aortic aneurysm disease(PERS) manifests as fever, leukocytosis, decrease in hemoglobin(HGB) and platelet(PLT), or coagulation disorders. And even occur some adverse events such as heart or cerebrovascular events, acute kidney injury, multiple organ failure, or death. It is not clear what the pathogenesis of PERS, the relationship between PERS and the postoperative adverse events, or the risk factors for PERS is. This study was designed retrospectively to describe the clinical feature of PERS, establish PERS risk score and analyze the risk factors for PERS to set up the early warning mechanism or do some early intervention for prevention or mitigation of PERS leading to the postoperative adverse events.MethodsThe clinical data of aortic dilatation from January 1, 1997 to December 31, 2014 were analyzed retrospectively in our center, including type B aortic dissection(TBAD), thoracic aortic aneurysm(TAA), thoracoabdominal aortic aneurysm(TAAA), and abdominal aortic aneurysm(AAA).Part I: The database of aortic dilatation disease was established. By comparing preoperative variables with the extreme value of corresponding variables in postoperative 7 days, Analysis of variation in the variables and investigation on clinical feature of PERS were performed.Part II: This is a case-control study design. Through binary logistic stepwise regression, the relationship between the postoperative adverse events and PERS was analyzed to identify the predictive factors for PERS risk score.Part III: Using the predictive factors for PERS risk score, the risk scoring model for prediction of PERS was established by ROC curve, to analyze its correlation with the postoperative adverse events and verify the rationality of this scoring model.Part IV: This is a case-control study design. The risk factors of PERS were explored by logistic regression analysis.Part I: Inflammatory reaction occurred after EVAR, characterized by the significant increasing of the body temperature(36.49±0.44 VS 37.63±0.57, P<0.001) and the WBC(6.62±1.65 VS 12.54±3.92, P<0.001). Postoperative HGB(129.10±17.66 VS108.10±18.76P<0.001) was significantly reduced. Coagulation disorder appeared, which was manifestedas the significant decreasing of PLT(189.26±69.88 VS 142.18±60.13, P<0.001), the significant increasing of fibrinogen(4.22±1.94 VS 3.75±1.54, P<0.001) and international normalized ratio(1.18±0.47 VS 1.04±0.16, P<0.001), and significant prolonging of the activated partial thromboplastin time(46.96±23.77 VS 37.80±8.69, P<0.001), thrombin time(21.99±26.87 VS 16.78±1.99, P<0.002) and prothrombin time(15.06±10.16 VS 13.28±1.61, P=0.003). And blood glucose(Glu) and protein metabolism disorders arisen(Glu =8.03±2.62 VS 5.55±1.59,P<0.001;albumin =34.00±4.00 VS 38.93±4.37, P<0.001;prealbumin =175.98±54.41 VS 227.45±64.47, P<0.001).In addition, the function of the vital organs such as liver and kidney was also significantly affected(total bilirubin=12.39±12.15 VS 14.81±8.16, P<0.001; lactate dehydrogenase=192.74±67.07 VS 235.39±174.35, P<0.001; alanine aminotransferase= 31.26±77.88 VS 23.01±31.11, P=0.005; aspartate aminotransferase=37.23±147.92 VS 24.76±51.84, P=0.029; serum creatinine=95.89±80.62 VS 108.64±91.51, P<0.001).Part II: Postoperative temperature, WBC, HGB, PLT, Glu, albumin and prealbumin were chosen as alternative predictive factors of PERS risk scoring model. By stepwise regression analysis, and controlling the different pathological diagnosis, age, sex and graft material properties, the predictive factors of PERS risk scoring model were WBC(OR=1.10, CI: 1.04-1.17, P<0.001), Glu(OR=1.09, CI: 1.01-1. 88, P=0.0498), HGB(OR=0.95, CI: 0.93-0.96, P<0.001) and PLT(OR=0.983, CI: 0.978-0.988, P<0.001). According to ROC curve analysis, the cutoff values of WBC, HGB, PLT and Glu were 15.39, 94, 98 and 8.0 respectively.By ROC curve analysis, the combination HGB and PLT could effectively predict the postoperative adverse events, and additional WBC and(or) Glu did not significantly increase the predictability of the postoperative adverse events.Part III: By ROC curve analysis, th first model for PERS was score =(-0.0545) × HGBx-0.0171 × PLTx, which was divided into normal, mild, moderate and severe according to the score value.The second model was established according to the cutoff value of HGB and PLT, which was divided into normal, mild and severe.The third model was established according to the decreasing extent of HGB and PLT.The ROC curve showed the same prediction effectiveness of the 3 models(AUC =0.81, P>0.05). By ordered logistic regression analysis, all the 3 models were closely related to the postoperative adverse events(model 1: OR=3.6, CI: 2.79-4.63, P<0.001; model 1: OR=7.17, CI: 5.07-10.15, P<0.001; model 1: OR= 6.4, CI: 4.52-9.08, P<0.001).Because the value of HGB and PLT were taken according to the different clinical decreasing extent, and did not been affected by specific study population, the model 3 coud be used more widely. So it was reasonable for the prediction effectiveness of PERS.According to the model 3, the incidence of PERS was 73%, including mild(50.8%), moderate(17.8%) and severe(4.4%).Part IV: By the model 3, in the preoperative basic condition, cerebrovascular disease(OR=2.13, CI: 1.23-3.70, P=0.007), chronic renal insufficiency(OR=3.26, CI:1.33-7.98, P=0.010) and ASA classification(OR=1.44, CI:1.11-1.85, P=0.005) were independent risk factors, and preoperative PLT(OR = 0.61, CI: 0.52-0.70, P<0.001) and HGB(OR=0.44, CI:0.37-0.51, P<0.001) were independent protective factors of PERS. Duing the period of EVAR, the blood loss(OR=1.57, CI: 1.34-1.84, P<0.001) and duration of EVAR(OR=1.41,CI: 1.18-1.68, P<0.001) were independent risk factors for PERS. In endovascular graft-related information, the number of aortic graft(OR=1.68, CI: 1.32-2.14, P<0.001) and the number of endovascular graft for branch artery(OR=1.31, CI: 1.09-1.58, P=0.004) were independent risk factors for PERS. In the degree of change of the risk factors for the postoperative adverse events, the decreasing degree of PLT(OR=1.16, CI:1.01-1.34, P=0.037) and HGB(OR=2.12, CI: 1.82-2.48, P<0.001), and the increasing degree of FBG(OR=1.37, CI: 1.19-1.59, P<0.001) were an independent risk factor, and increasing the degree of WBC(OR=0.85, CI: 0.74-0.98, P=0.024) was an independent protective factor of PERS.In the pathological morphology of aortic dilatation disease, the maximum diameter of aortic aneurysm(OR=1.27, CI: 1.04-1.55, P=0.019), the difference of aortic aneurysm sac between preoperative and postoperative period, and the length of aortic graft were risk factors for PERS.The length of postoperative hospital stay and hospitalization expenses had a very significant statistical difference(P<0.001) in the difference division of PERS. And with the aggravation of PERS, all the two variables increased accordingly.ConclusionsPERS can be characterized by inflammatory response, decreased hemoglobin, coagulation disorder, and metabolic disorder of glucose or protein, which may be divided into different levels including mild, moderate and severe. The occurrence of PERS is closely related to the postoperative adverse events. According to the PERS controllable risk factors, early warning mechanism may be established to do early intervention such as follows: in the preoperative period of EVAR, to correct anemia and low platelets, and improve the physical condition of patients and renal insufficiency; in the intraoperative period, as far as possible to shorten the operation time and reduce blood loss; in the postoperative period, to control effectively inflammatory response and appropriate blood glucose level, and to correct the decreasing amplitude of hemoglobin or platelet; to reduce the number of and the actual length of graft-stents in aorta; to treat multisite aortic aneurysm in different periods. These preventive measures may be effective to prevent or reduce the postoperative adverse events caused by PERS.