Application Of Low-dose Retrospective ECG-gated CTA In Diagnosis And Prognostic Evaluation Of Aortic Dissection

Part Ⅰ Association of Imaging Features and Delayed Adverse Events:Evaluated by Retrospective ECG-gated CTA in Uncomplicated Aortic DissectionObjectiveLow-dose retrospective ECG-gated CT angiography(CTA)enables multiphase images reconstruction within the whole cardiac cycle with slightly higher radiation dose than traditional static imaging and might realize the dynamic assessment of the aortic wall and intimal oscillations.However,few studies have concentrated on the evaluation of aortic dissection by retrospective ECG-gated CTA and whether imaging features are associated with delayed adverse event(DAE).The current study aimed to investigate the feasibility of low-dose retrospective ECG-gated CTA for quantitative evaluation of acute uncomplicated type B aortic dissection(uTBAD),and analyze the predictive value of CTA morphological parameters(dynamic and static parameters)in the occurrence of DAE.Materials and Methods1.Study PopulationFrom January 2010 to June 2018,patients with type B aortic dissection who performed low-dose retrospective ECG-gated CTA were retrospectively enrolled in our department.Exclusion criteria were as follows:(1)subacute or chronic aortic dissection(≥14 days);(2)complicated aortic dissection within 48 h after symptom onset(≥1 indications:aortic rupture,malperfusion,aneurysmal degeneration,refractory hypertension,recurrent pain);(3)complete data were not available to retrieve from the Hospital Medical Records System.Finally,the remaining 87 uTB-AAD patients were included for analysis.2.Image AnalysisAll eligible patients were performed low-dose retrospective ECG-gated CTA with a dual-source CT scanner.The ECG-gated tube current modulation technique was used:full tube current was centered at one phase(70%R-R interval),and 20%of the peak tube current was at the remaining R-R intervals.Multiphase images were reconstructed through the entire aorta from 0 to 95%R-R intervals with an increment of 5%.Areas of the true lumen and aorta at origin level of five main branch arteries(upper thoracic descending aorta(UTDA),celiac trunk artery,superior mesenteric artery,renal artery,inferior mesenteric artery)were measured,as well as short-axis diameters of the descending aorta and false lumen at the UTDA origin level.CTA parameters were as follows:(1)Dynamic parameters:obtained from multiphase images(0-95%R-R intervals),including relative true lumen area(RTLA),ratio of true lumen relative area(r-RTLA),the maximum diameter of the descending aorta,false lumen,and primary entry tear.The equations are as follows:RTLAn(%)=(TLAn/AAn)x 100%r-RTLA(%)=(RTLAmax-RTLAmin/RTLAmax)× 100%where TLA represents ture lumen area,AA represents aortic area,and n represents a specific phase of the R-R interval.(2)Static parameters:acquired from single-phase images(70%R-R interval),including RTLA in 70%R-R phase(RTLA70%)at the five branch artery levels,descending aortic diameter,false lumen diameter,entry tear size,number of intimal tears,circumferential extent of the false lumen at the five levels,location and configuration of the false lumen,false lumen branches and patency,and length of dissection.3.Study EndpointsOutcome analysis comprised incidences of DAE and early mortality within 3 to 14 days since symptom occurring.DAE encompassed malperfusion,aortic rupture,aneurysmal degeneration,refractory hypertension and recurrent pain.Results1.Patient CharacteristicsAmong the 87 acute uTBAD patients(71.3%male;age,52.2 ± 10.6 years),26 patients(29.9%)were developed DAE.In terms of the 26 patients who developed DAE,two patients(7.7%)died before any interventions.There existed significant difference between DAE and non-DAE groups in terms of treatment(P<0.001)2.r-RTLA at different regions of the aortaThe mean estimated effective radiation dose was 12.1 ± 5.2(4.5-25.1)mSv.There existed significant differences in r-RTLA among the five aortic branch origin levels(P=0.002).The r-RTLA at inferior mesenteric artery level was significantly larger than that of UTDA and superior mesenteric artery levels(all P<0.05).3.CTA Morphological Features and DAEDynamic parameters and DAE:smaller values of RTLAmin(P=0.010),larger values of r-RTLA at the UTDA level(P<0.001)and at the renal artery level(P=0.016)demonstrated higher incidences of DAE;maximum diameter of the descending aorta(P<0.001),the false lumen(P=0.008),and entry tear size(P=0.007)were positively associated with the occurrence of DAE.Static parameters and DAE:diameter of the descending aorta(P=0.002),false lumen diameter(P=0.042),and entry size(P=0.018)measured in 70%R-R interval were significantly larger in the DAE group than in the non-DAE group.No significant differences were discovered in terms of RTLA70%between two groups(all P>0.05)4.Diagnostic Performance of CTA Parameters for Detecting DAEr-RTLA at the UTDA level was discovered to have the largest area under the curve(area under curve(AUC)=0.839,P<0.001),and this dynamic parameter yielded the highest diagnostic accuracy(81.6%)in detecting DAE at an optimal cutoff value of 38.3%.Performance of dynamic characteristics was superior to static features in the detection of DAE(AUC:0.887 vs.0.723,P<0.001).ConclusionLow-dose retrospective ECG-gated CTA is an effective method for quantitative evaluation of aortic dissection.Dynamic morphological features obtained from multiple images could better evaluate risk of developing DAE in patients with acute uTBAD patients than static characteristics obtained from single-phase images.These dynamic features might aid in risk stratification and early targeted interventions of high-risk patients.Part Ⅱ Value of Retrospective ECG-gated CT Angiography in Predicting Renal Injury and TEVAR Outcomes in Complicated Aortic DissectionObjectiveThe formation and development of aortic dissection is strongly associated with hemodynamics.Movement of the internal flap could result in partial or complete collapse of the true lumen,resulting in dynamic ischemia of the viscera.Additionally,radiologic signs might allow for early detection of visceral malperfusion prior to clinical indicators.However,relationship between intimal motion and outcomes of thoracic endovascular aortic repair(TEVAR)in patients with complicated type B aortic dissection(cTBAD)is poorly understood.We sought to investigate the abilities of dynamic variables obtained from retrospective CT angiography(CTA)for predicting preoperative renal injury and TEVAR outcomes,and estimate TEVAR outcomes of cTBAD patients in different phases.Materials and Methods1.Assessment of Acute Renal Injury before TEVAR by Retrospective CTA1.1 Patient DemographicsWe retrospectively enrolled patients who underwent TEVAR for TBAD from January2002 to December 2017 in our department.Exclusion criteria included the following:(1)no CTA examination within 48 hours from admission;(2)patients who received TEVAR within 48 hours of admission;(3)the intimal flap involved the kidney vessels or did not reach the plane under the renal artery origin;(4)patients with a history of renal injury;(5)unavailability of complete data for retrieval from the hospital medical records system.1.2 Image AnalysisRetrospective ECG-gated CTA was performed by dual-source CT.Images were generally reconstructed at every 5%R-R interval from 0%to 95%.Area of the true lumen and total abdominal aorta were computed automatically at the superior aspect of the renal artery ostia,as well as short-axis diameters of the aorta and false lumen at the level of the upper thoracic descending aorta.Ratios of true lumen area to abdominal aorta area were calculated to evaluate the relative area of true lumen(RTLA)in each phase.CTA morphological features included minimum RTLA(RTLAmin),maximum RTLA(RTLAmax),relative change of RTLA(r-RTLA),maximum diameter of the descending aorta,maximum diameter of false lumen,maximum size of the primary entry tear,reentry tear at the renal artery origin level,accessary renal artery,diameters of renal arteries,blood supply of kidneys.1.3 Study EndpointsThe primary endpoint was the occurrence of AKI preoperatively.AKI was defined according to Kidney Disease Improving Global Outcomes Guidelines:a maximum serum creatinine(SCr)level increase of more than 1.5 times the baseline level during the first week preoperatively or SCr>0.3 mg/dl(≥26.5 mmol/L)within 48 hours of admission.2.Role of Retrospective CTA in Predicting TEVAR Outcomes2.1 Patient DemographicsBetween March 2009 and June 2018,all patients who underwent TEVAR for cTBAD were retrospectively enrolled.Exclusion criteria included the following:(1)chronic dissection(>90 days);(2)aortic dissection secondary to trauma,iatrogenic injury,or intramural hematoma;(3)unavailability of low-dose retrospective CTA images before TEVAR;(4)unavailability of complete data for retrieval from the hospital medical records system;(5)incomplete follow-up.Finally,the analysis included 79 patients who received TEVAR for complicated hyperacute(<2 days),acute(2-14 days),or subacute(15-90 days)dissection.2.2 Image AnalysisAreas of the true lumen and aorta were acquired at the tracheal bifurcation at each R-R interval over the whole heart cycle.Relative true lumen area(RTLA)was calculated every 5%R-R interval.Parameters that reflect the state of intimal motion were evaluated,including the true lumen collapse(TLC)and difference between the maximum and minimum RTLA(D-TLA).The TLC was assessed by calculating the times at which the RTLA was<25%in a cardiac cycle(TLC25%)and<50%in a cardiac cycle at the tracheal bifurcation throughout the cardiac cycle.Furthermore,changes of false lumen area and status between the latest follow-up and initial CT angiography images after TEVAR were documented.2.3 Study EndpointsThe primary endpoints were early mortality and early adverse events(early-AEs)within 30 days after TEVAR,and the secondary endpoints were late mortality and late adverse events associated with dissection from>30 days postoperatively to the study end date.Results:1.Assessment of Acute Renal Injury before TEVAR by Retrospective CTA1.1 Demographics and Clinical FeaturesA total of 108 consecutive participants were involved in this study.The mean age was 50.3 ± 9.4 years and 79(73.1%)were male.Forty-three(39.8%)patients developed AKI before TEVAR.Systolic blood pressure(SBP)and diastolic blood pressure on admission were considerably higher in patients with AKI than those in the non-AKI group(SBP:P<0.001;diastolic blood pressure:P=0.015).Patients with AKI had a significantly higher value of maximum SCr compared to the non-AKI group(P<0.001).1.2 Morphometric Characteristics in CTAThe mean estimated effective radiation dose was 14.5± 6.2(5.1-22.3)mSv.Group-averaged TLA in the AKI group was smaller than that of the non-AKI group in each R-R interval(all P ≤ 0.001).RTLAmin(0.2 ± 0.1 vs.0.4 ± 0.1,P<0.001)and RTLA,ax(0.4±0.2 vs.0.6 ± 0.1,P<0.001)in patients with AKI were significantly smaller than non-AKI patients.Subjects with AKI exhibited larger values in r-RTLA(%)(P<0.001),maximum diameters of the descending aorta(P=0.023)and primary entry tear(P=0.012).No significant difference was discovered between two groups in terms of accessory renal artery of left kidney and right kidney,diameter of left renal artery and right renal artery between AKI and non-AKI groups(all P>0.05).1.3 Risk Predictors of AKISBP(Odds ratio(OR)=1.037;P=0.001)and r-RTLA(OR=1.050;P=0.001)were considered independent factors for AKI preoperatively.The value of the area under the curve for SBP and r-RTLA were 0.761 and 0.788,respectively.1.4 Outcomes after TEVARPatients with AKI before TEVAR had a significant increase in adverse events compared to the non-AKI group(P=0.001).The AKI group was more likely to suffer from acute renal failure than the non-AKI group(P=0.013).Patients with larger r-RTLA had a higher incidence of acute renal failure(P=0.002).2.Role of Retrospective CTA in Predicting TEVAR Outcomes2.1 Patient CharacteristicsThe mean age of 79 enrolled patients was 49.9±1 1.9 years,77.2%patients were male.There were 31 patients in hyperacute phase,32 patients in acute phase and 16 patients in subacute phase.2.2 TEVAR OutcomesThe median follow-up time was 339 days(interquartile range:41-566 days).The overall early mortality and early-AEs rate was 13.9%and 24.1%,respectively.Patients received TEVAR in hyperacute setting had markedly higher mortality and adverse event rates in early and late terms compared with acute and subacute groups(all P<0.05).No significant difference was found between acute and subacute groups in the incidence of mortality or adverse event(all P>0.05)2.3 CTA Morphologic CharacteristicsSpearman’s correlation coefficients between D-TLA and TEVAR indications were r=0.420(P<0.001)for aortic rupture and r=0.229(P=0.042)for malperfusion.TLC25%showed moderate correlation to malperfusion(r=0.406,P<0.001).A longer time of TLC25%(P=0.049)and a larger D-TLA(P<0.001)were correlated to an increased early death.Similarly,a larger D-TLA(P=0.001)and a maximum false lumen diameter>22 mm(P=0.030)occurred more frequently in patients with early-AEs than those without early-AEs.2.4 COX proportional hazard modelsD-TLA was considered as an independent predictor of early mortality(hazard ratio[HR]=1.153,P<0.001).Both larger D-TLA(HR=1.128,P<0.001)and maximum diameter of false lumen>22 mm(HR=3.226,P=0.023)were independently correlated with increased incidences of early-AEs.2.5 Predict TEVAR OutcomesArea under curve of D-TLA was 0.849 for predicting early mortality and 0.742 for early-AEs with a best cutoff value of 21.5%.The D-TLA>21.5%group exhibited worse survival and event-free survivals in follow-up than the D-TLA≤21.5%group(all P<0.001).2.6 Aortic Remodeling PostoperativelyA decrease in false lumen area was observed in 51 patients(81.0%).No significant differences between the D-TLA>21.5%and D-TLA≤21.5%groups were observed in terms of aortic remodeling after TEVAR(all P>0.05).ConclusionLow-dose retrospective ECG-gated CTA could realize quantitative assessment of dynamic motion of the intimal flap.CTA dynamic morphological features obtained from multiphase images are correlated with acute kidney injury pre-operative and worse post-operative outcomes,which might be helpful for future risk stratification before TEVAR and optimal management.Additionally,patients receiving TEVAR in the hyperacute phase have significantly worse outcomes than those in the acute and subacute phases.