| Part I Comparison of color duplex ultrasound and multi-slice computed tomography angiography in the detection of endoleak following endovascular graftexclusion of aortic dissectionObjective: Comparison of color duplex ultrasound and multi-slice computed tomography angiography in the detection of endoleak following endovascular graftexclusion of aortic dissection after 1 month of postoperative.Materials and methods: Collected in Sichuan Provincial People’s hospital on January 1,2014 to October 31,2016,34 patients with endovascular graft exclusion of aortic dissection after 1 month of postoperative were retrospectively analyzed.Among them(mean age:55.91±11.80;age range :38-81 years old),including 25 cases(73.5%)of men,9 cases(26.5%)of women.All of the patients did multi-slice CT angiography and color doppler ultrasound examination at the same time.Besides,intervals of the two kinds of examination was not more than 1 week.CT images and ultrasound images were evaluated by the radiograph reading group which was consisted of two experienced high qualification doctors.Before evaluation,any member of the radiograph reading group didn’t know the results of each other.Regarded multi-slice CT angiography imaging results as the gold standard of postoperative endoleak detection,we can calculate color doppler ultrasonic detection of endoleak of the true positive rate and false positive rate and false negative rate,sensitivity,specificity,accuracy,negative predictive value,positive predictive value.Data analysis used matching chi-square test,p < 0.05 for the difference had statistical significance.Kappa index further evaluated that multi-slice CT angiographyimaging and color doppler ultrasonic detected the consistency of endoleak.Results: Multi-slice CT angiography detected in 11 cases of endoleak,endoleak rate was 32.4%,which had 10 cases(29.4%)type I endoleak,1 case(2.9%)type III endoleak,without type II endoleak.In the 10 cases endoleak of type I,5 cases appeared in the proximal stent(type Ia),3 cases appeared in the distal stent(type Ib),2 cases appeared both in the proximal and distal stent(type Ia and Ib).Color doppler ultrasound detected 8 cases of endoleak were all type I,5 cases of endoleak were located in the proximal stent(type Ia),2 cases were located in the distal stent(type Ib),1 case was located both in the proximal and distal stent(type Ia and Ib).Among 8 cases of endoleak,6 cases were consistency of multi-slice computed tomography angiography findings,2 cases of color doppler ultrasound showed positive while multi-slice computed tomography were negative.CTA detected 1 case of type III endoleak while color doppler ultrasound showed negative.Compared with multi-slice CT angiography results,color doppler ultrasound endoleak detection of true positive rate was 54.5%(6/11),false positive rate was 8.7%(2/23),the false negative rate was 45.4%(5/11),the sensitivity was 54.5%(6/11),specificity was 91.3%(21/23),the accuracy was 79.4%(27/34),positive predictive value and negative predictive value were 75%(6/8)and 80.8%(21/26)respectively.The consistency of two kinds of examination method was medium(k=0.494).None of the 34 cases patients after endovascular graft exclusion of aortic dissection with stents were implanted in failure,disintegration,displacement and deformation.To some extent,all false lumen of patients didn’t have thrombosis and support coverage area of true blood flow were unobstructed.Stent intimal mural thrombus formation of all patients was not found.Due to the lack of 6 exception hospital treatment of patients with preoperative image data while 28 patients had complete preoperative image data,postoperative MS-CTA images showed that the size of dissecting aneurysm didn’t expand,true cavity pressure reduced and recovered with different degree.Conclusion:Both multi-slice computed tomography and color dopplerultrasonography can detect endoleak after endovascular graft exclusion of aortic dissection.Considering the cumulative dose of radiation and cost,color doppler ultrasonography can be a preliminary screening,but the sensitivity and positive predictive value in detecting endoleak were low.Multi-slice CT angiography is still as the main imaging examination method of endoleak follow-up.Part II Type I endoleak after endovascular graft exclusion of Standford B type of aortic dissection: the predictive factors of preoperative multi-slice computed tomography imagingObjective: This study aimed to analyze preoperative multi-slice CT angiography image feature parameters and discussed whether different characteristic parameters can predict type I endoleak of Standford type B aortic dissection after 1 month of endovascular graft exclusion.Materials and methods: Collected in Sichuan Provincial People’s hospital on January 1,2014 to October 31,2016,26 patiens of Standford type B aortic dissection after 1 month of endovascular graft exclusion did multi-slice CT follow-up examination.The patients preoperative imaging parameters of multi-slice CT angiography image were analyzed retrospectively.Among them(mean age :58.88±11.28;age range:40-81 years old),including 18 cases(69.23%)of men and 8 cases(30.77%)of women.The internal time from the first time of preoperative CTA examination to operation was between 1-22 days.All patients did multi-slice CT angiography after 1 month of postoperative,according to the results of endoleak detection,we can divide the patients into two groups: group of endoleak(LG:n=10)and without endoleak group(NLG:n= 16).Preoperative CTA images characteristic parameters included: the width of the initial entry;the true lumen diameter,falselumen diameter,aortic diameter and the true lumen ratio,false lumen ratio,true and false lumen ratio at the different levels such as:the proximal entry level,the maximum diameter of aortic aneurysm level,left subclavian artery opening level,trachea juga bifurcate level,diaphragm level,the maximum diameter from diaphragm to celiac axis area level,celiac axis to renal artery area level,renal artery to the bifurcation of iliac blood vessels area level;aortic calcification,false lumen thrombosis,the proximal and distal stent of aortic calcification and false lumen thrombosis,from celiac axis to renal artery level of aortic calcification and false lumen thrombosis.The image characteristics and the relevant measurement data of all patients with preoperative CTA images and postoperative CTA images after 1 month of EVGE were analyzed by two experienced high qualification doctor radiograph reading group.When analysing the preoperative CTA images,any member of read group did not know the postoperative CTA images results.Data analysis used independent sample t-test,Fisher’s exact probability method,the area under ROC curve was used to determine the optimal threshold of preoperative CTA characteristic parameters and calculated the degree of sensitivity,specificity,with p<0.05 for the difference had statistical significance.Results: 26 cases of Standford type B aortic dissection patien after1 month of EVGE did CTA examination.16 patients without endoleak,10 patients with endoleak,all endoleaks were type I,5 cases of endoleak were located in the proximal stent(type Ia),3 cases were located in the distal stent(type Ib),2 cases were located both in the proximal and distal stent(type Ia and Ib),without type II endoleak.Among preoperative CTA characteristic parameter evaluation: the true and false lumen ratio(T/F)between the two groups at the level of the maximum diameter from celiac axis to renal artery area had statistical difference(endoleak group: 1.11±0.50cm;noendoleak group: 0.67± 0.41;p=0.04;the area under the ROC curve=0.77,the critical point cutoff value of 0.68,sensitivity 75%,specificity 83.3%).The true lumen diameter,false lumen diameter,aortic diameter and the true lumen ratio,false lumen ratio,true and false lumen ratio between the two groups at the following didn’t havestatistical difference:the proximal entry level,the maximum diameter of aortic aneurysm level trachea juga bifurcate level,diaphragm level,the maximum diameter from diaphragm to celiac axis area level,renal artery to the bifurcation of iliac blood vessels area level.The aortic diameter between the two groups at the left subclavian artery opening level didn’t have statistical difference.The true lumen diameter,false lumen diameter,aortic diameter and the true lumen ratio,false lumen ratio between the two groups at the celiac axis to renal artery area level didn’t have statistical difference.The aortic calcification,false lumen thrombosis,the proximal and distal stent of aortic calcification and false lumen thrombosis,from celiac axis to renal artery level of aortic calcification and false lumen thrombosis between the two groups didn’t have statistical difference.Conclusion: The characteristic parameter of preoperative multi-slice computed tomography angiography image:the true and false lumen ratio(T/F)at the level of the maximum diameter from celiac axis to renal artery area can predict type I endoleak of Standford type B aortic dissection after 1 month of endovascular graft exclusion. |
PDF Full Text Request