Surgical Treatment And Prognostic Analysis Of Aortic Dissection Involving Proximal Z1-3 Zone

Aortic dissection is a rare but life-threatening and critically ill condition.The early mortality of dissection only involving the descending aorta(Stanford type B aortic dissection,TBAD)is relatively lower than that of Stanford type A aortic dissection(TAAD),which involves the ascending aorta.However,TBAD still faces the risk of poor organic malperfusion or rupture in the acute phase,and aneurysmal expansion in the chronic phase.Up to now,international guidelines have consistently recommended that thoracic endovascular aortic repair(TEVAR)should be regarded as the preferred treatment for complicated TBAD with malperfusion or aortic rupture,and the first-line treatment for uncomplicated TBAD(uTBAD)remains the best medical treatment(BMT).However,an increasing number of researches suggested that uTB AD with high-risk features had a poor prognosis after receiving BMT alone,and TEVAR was more effective than BMT alone.Therefore,the indications for TEVAR treatment of TBAD are constantly expanding.There is no consensus on the optimal timing of TEVAR for uTBAD patients.Some scholars recommend that TEVAR be postponed to the subacute phase(15-90 days after symptom onset),which might reduce the risk of complications such as retrograde type A aortic dissection,cerebral infarction,and spinal cord ischemia after intervention in the acute phase,and also leverage the advantage of positive aortic remodeling induced by intervention in the subacute phase.However,the 2022 clinical practice guideline on the management of TBAD by the Society of Thoracic Surgeons/American Association for Thoracic Surgery indicates that TEVAR in 24 hours-90 days after symptom onset may be reasonable for acute uTBAD with high-risk features.The first part of this thesis was based on the experience of TBAD treatment in our center and aimed to explore the impact of intervention timing on early outcomes and mid-to long-term prognosis of acute,subacute uTBAD patients.With the continuous enrichment and improvement of epidemiological data and operative techniques,aortic dissection with proximal involvement of the aortic arch(zone 1-2)has gained more and more attention.In order to standardize academic communication and clinical practice,European and American scholars have proposed new aortic dissection classification schemes based on the classic DeBakey classification and Stanford classification.As of now,surgical treatment options for aortic dissection with proximal involvement of zone 1-2 include open surgery,hybrid arch repair,and endovascular repair.Most of the existing literature focuses on the mid-to long-term efficacy of a certain treatment regimen,with few clinical research reports on the strategy of HENDO(Hybrid-Endovascular-Open aortic arch repair).The second part of this thesis reported the mid-to long-term efficacy of HENDO in treating aortic dissection with proximal involvement of zone 1-2.Although the indications of TEVAR continue to expand and TEVAR has become the preferred surgical option for TBAD,it still faces the risk of certain complications such as retrograde type A aortic dissection,stent-graft migration,and aortic dilation of landing zone.The third part of this thesis explored the risk factors affecting proximal seal-related complications and consequently constructed a risk prediction model based on the data of patients with aortic dissection involving zone 1-3 and receiving TEVAR.The abstract of each part is as follows:Part One:Timing and clinical outcomes of thoracic endovascular aortic repair for acute and subacute uncomplicated Stanford type B aortic dissectionObjective:This study aimed to explore the impact of the intervention timing,from the onset of symptoms to the thoracic endovascular aortic repair(TEVAR),on the early and mid-to long-term outcomes of patients with acute or subacute uncomplicated Stanford type B aortic dissection(uTBAD).Methods:This retrospective cohort study included consecutive patients with acute or subacute uTBAD who received TEVAR at the vascular centre of our hospital from January 1st,2015 to December 31st,2019.Based on the timing from symptom onset to TEVAR,the cohort was divided into 3 groups(≤3 days,4-14 days,and 15-90 days).The early outcomes included early mortality,aortic-related early events,immediate type Ⅰa endoleak,early cardio/cerebrovascular events,renal function deterioration,pulmonary infection,and access complications,etc.The late outcome was all-cause mortality and aortic-related late events during the follow-up.We compared the effects of different intervention timing on early and late outcomes of uTBAD through univariate analyses and multivariate regression analyses and depicted the Kaplan-Meier curves.Results:This study included 304 patients(men,83.9%;men age,52.8±11.8 years).The median clinical follow-up was 49.0(37.0,63.0)months.Of the 304 TEVARs,74(24.3%)were performed within 3 days after dissection onset,179(58.9%)in 4-14 days,and 51(16.8%)in 15-90 days.No significant difference was noted in the early mortality(1.4%vs 1.7%vs 0.0%),aortic-related early events(0.0%vs 0.6%vs 0.0%),immediate type Ia endoleak(5.4%vs 6.7%vs 5.9%),early cardio/cerebrovascular events(2.7%vs 2.8%vs 3.9%),ventilation support time≥48 hours(1.4%vs 2.8%vs 2.0%),and access complications(0.0%vs 0.6%vs 2.0%)among the 3 groups(P>0.05).However,rates of early renal function deterioration(12.2%vs 4.5%vs 9.8%)and pulmonary infection(8.1%vs 6.7%vs 2.0%)were higher after TEVAR within the first 3 days than those during 4-14 days and 15-90 days.The survival probabilities of the 3 groups(≤3 days,4-14 days,and 15-90 days)at 5 years were 93.7%(95%CI:87.9%-99.9%),92.2%(95%CI:88.2%-96.4%),and 88.3%(95%CI:75.6%-100.0%),respectively;the freedom from aortic-related late events at 5 years were 85.2%(95%CI:76.0%-95.5%),91.7%(95%CI:87.6%-96.0%),77.9%(95%CI:66.4%-91.5%),respectively;the rates oflatetype Iaendoleak were2.7%,1.1%,and 0.0%,respectively(P>0.05).Multiple Cox regression revealed that the intervention timing of TEVAR was not associated with the mid-to long-term follow-up outcomes of the all-cause mortality and the freedom of aortic-related late events(adjusted P>0.05).Conclusion:Patients with acute or subacute uTBAD who undertook TEVAR had relatively low risks of postoperative mortality,good mid-to long-term survival probabilities and freedom from aortic-related late events.Compared to TEVAR in the subacute phase(15-90 days),TEVAR in the acute phase(4-14 days)would not statistically increase the risk of early mortality,early morbidities,mid-to long-term all-cause mortality and aorticrelated late events.However,intervention timing within the first 3 days was potentially associated with the tendency of increasing renal function deterioration and pulmonary infection.Therefore,it is more reasonable that TEVAR should be postponed after the first 3 days of symptom onset.In the future,it is necessary to design and conduct multi-centre,prospective clinical trials with large sample sizes or real-world registry studies to produce evidence of high quality.Part Two:Clinical prognosis of HENDO system in the treatment of aortic arch dissection with proximal extension involving the zone 1-2Objective:This study aimed to report the early and mid-to long-term outcomes of the Hybrid-Endovascular-Open aortic arch repair(HENDO)system for the treatment of aortic arch dissection(also named non-A non-B aortic dissection,NANB)with proximal extension involving the zone 1-2.Methods:Consecutive patients with NANB who underwent endovascular repair,hybrid arch repair,and open surgical repair in our hospital between January 2015 and December 2019 were retrospectively analyzed and compared.The early outcomes included early mortality,immediate type Ia endoleak,early cardio/cerebrovascular events,and renal function deterioration,etc.The late outcomes were all-cause mortality and aortic-related late events during the mid-to long-term follow-up.Kaplan-Meier curves were depicted to analyze survival from all-cause mortality and freedom from aortic-related late events in the endovascular,hybrid,and open groups.Propensity score matching and subgroup stratification(stratified by proximal dissection extension:zone 1 and zone 2)were performed as sensitivity analyses to compare the outcomes among the three treatment patterns after controlling for major confounders.Results:This study included 151 patients(men,79.5%;mean age,47.3±10.5 years).The median clinical follow-up was 40.0(25.0,56.0)months,all-cause mortality of the whole cohort was 5.3%,accumulated survival probability at 5 years was 91.4%(95%CI:83.4%100.0%);aortic-related events of the whole cohort was 15.2%,and accumulated survival free from aortic-related events at 5 years was 79.7%(95%CI:70.9%-89.6%).The cohort was divided into three groups with 72(47.7%)in the endovascular group,46(30.5%)in the hybrid group,and 33(21.8%)in the open group.No significant difference was noted in the early mortality(1.4%vs 2.2%vs 3.0%,P=0.791)between 3 groups;the incidence of immediate type Ⅰa endoleak was significantly greater(20.8%vs 6.5%,P=0.002)in the endovascular group than that of the hybrid group;however,the endovascular group had significantly less renal function deterioration(4.2%vs 26.1%vs 24.2%;P=0.001),shorter operation time(97.5 min vs 273.5 min vs 312.0 min,P<0.001),shorter hospitalization(6.0 days vs 8.0 days vs 11.0 days,P<0.001),shorter intensive care unit(ICU)stay(24.0 h vs 48.0 h vs 69.0 h,P<0.001),shorter ventilation support time(8.0 h vs 15.5 h vs 17.0h,P<0.001),a lower proportion of blood transfusion(1.4%vs 37.0%vs 63.6%,P<0.001)and fewer expenses(179.0 thousand yuan vs 245.0 thousand yuan vs 172.0 thousand yuan,P<0.001)compared with hybrid arch repair and open surgical repair.No significant difference was noted in the mid-to long-term all-cause mortality(5.6%vs 4.3%vs 3.0%,P=1.0)and aortic-related late events(16.7%vs 15.2%vs 12.1%,P=0.834)between the endovascular,hybrid,and open groups.The rates of late endoleak in total between the endovascular,hybrid,and open groups showed no significant difference(9.7%vs 4.3%vs 6.1%,P=0.630);the rates of late type Ⅰa endoleak in the endovascular and hybrid group were 6.9%and 2.2%respectively.The propensity score matching analyses and stratification analyses displayed similar results as the primary analyses.Conclusions:After surgical treatment based on the HENDO system,patients with proximal involvement of zone 1-2 aortic dissection(NANB)can achieve relatively low early mortality,good mid-to long-term survival probability and freedom from aortic-related events.For patients with NANB in our study,endovascular repair,hybrid arch repair,and open surgical repair have comparable mid-to long-term survival probabilities and freedom from aortic-related events.Compared with hybrid arch repair and open surgical repair,endovascular repair exhibits advantage of minimally invasion in terms of operation time,postoperative hospitalization,ICU stay,ventilation support time,blood transfusion,hospitalization expenses,and renal function deterioration.Therefore,it is necessary and beneficial to promote the HENDO system,integrate endovascular,hybrid and open repair techniques,and establish individualized therapy for aortic dissection involving the aortic arch.Part Three:Risk factors of proximal seal-related events after TEVAR for acute,subacute aortic dissection with proximal extension in zone 1-3 and nomogram model establishmentObjective:This study aimed to identify the risk factors of proximal seal-related events(PSE)after thoracic endovascular aortic repair(TEVAR)for acute,subacute aortic dissection with proximal extension in zone 1-3,and establish a nomogram predictive model.Methods:This retrospective study consecutively involved patients with acute or subacute aortic dissection with proximal extension in zone 1-3 treated by TEVAR from January 1st,2015 to December 31st,2019 at the vascular centre of our hospital.As a composite endpoint,proximal seal-related events(PSE)included type Ⅰa endoleak,retrograde type A aortic dissection(RTAD),30-day ascending aortic intramural hematoma thickness,proximal stent-induced new entry(SINE),stent-graft migration and transaortic dilation at the proximal landing zone.Patients were divided into the PSE and no-PSE groups,and univariate analyses and multivariable Logistic analyses were used to identify the independent risk factors related to PSE.The nomogram predictive model was constructed based on the risk factors identified by multivariable Logistic analyses.To assess its discriminative ability,the receiver operating characteristic(ROC)curve was depicted and the area under the curve(AUC)was calculated.Calibration ability was evaluated by drawing the calibration curve.Clinical utility was evaluated using decision curve analysis(DCA).The bootstrap resampling method was applied for the internal validation of the predictive model,and the ROC curve and the calibration curve after internal validation were plotted again.Results:This study included 413 patients(men,85.2%;mean age,51.2 ± 11.9 years).The median clinical follow-up was 49.0(36.5,62.0)months and the median radiological follow-up was 7.0(0.0,25.0)months.During follow-up,50(12.1%)patients occurred PSEs,including 42 cases of immediate type Ⅰa endoleak(including 9 persistent type Ⅰa endoleak),3 new-onset type Ⅰa endoleak,3 RTAD,1 ascending aortic intramural hematoma thickness,5 proximal SINE,3 stent-graft migration,and 2 transaortic dilation.The multivariable Logistic regression showed that bird beak configuration(odds ratio[OR]=4.246),unhealthy proximal landing(OR=3.937),chimney technique(OR=27.647),insufficient proximal landing length(10-20mm,OR=4.377;<10mm,OR=9.166),and tortuosity of zone 1-4(OR=1.037)were independent risk factors of PSE(P<0.05).Based on the above 5 independent risk factors of PSE,a nomogram predictive model was established.The AUC of the model was 0.846(95%confidence interval[CI]:0.788-0.904),the AUC was 0.842(95%CI:0.783-0.900)after internal validation by means of bootstrap resampling;the slope of the calibration curves before and after resampling was approximately 1.These proved the nomogram predictive model had good performance of discrimination and calibration.DCA curve revealed that the net benefit value of the model was good when the risk threshold was set between 5%and 80%.Conclusions:This study explored and identified 5 independent risk factors that affected the occurrence of PSE after TEVAR in patients with aortic dissection proximally extending to zone 1-3,including bird beak configuration,insufficient proximal landing length,unhealthy proximal landing,excessive tortuosity of zone 1-4,and the chimney technique.Based on these risk factors,a nomogram predictive model with good discrimination and calibration was constructed to identify patients at high risk of PSE and assist in the establishment of individualized strategies.