Meta-analysis and meta-regression of the total endovascular aortic repair in aortic arch
Abstract
Summary:Background: The total endovascular repair of the aortic arch is becoming more and more an acceptable alternative to the open repair in selected patients. The aim of the present study is to perform a meta-analysis of the available data on the outcomes of the various endovascular techniques used to treat the pathologies in this challenging anatomical area. Patients and methods: An extensive electronic search in PubMed/MEDLINE, Science Direct Databases, and the Cochrane Library was conducted. All papers published up to January 2022 on the endovascular techniques in the aortic arch (chimney-thoracic endovascular aortic repair (ChTEVAR), fenestrated/branched grafts as custom made devices (CMD) and surgeon modified TEVAR (SM TEVAR) providing information about at least one of the essential outcomes defined in the inclusion criteria. Results: Out of the 5078 studies found through the search in the databases and registers, 26 studies with a total number of patients of 2327 with 3497 target vessels were included in the analysis. The studies reported a high technical success rate with an estimated proportion of 95.8% (95% CI, 93–97.6%). Moreover, the pooled estimation of the early type Ia/III endoleak was 8.1% (95% CI, 5.4–12.1%). The pooled mortality was 4.6% (95 CI, 3.2–6.6%) with a significant heterogeneity and the stroke had an estimated proportion (major and minor combined) of 4.8% (95% CI, 3.5–6.6%). A meta-regression analysis showed no significant variation between the groups in mortality (P=.324) showed however a significant difference between the therapeutic methods regarding stroke P<.001 (lower rate in ChTEVAR and SM vs. CMD). Conclusions: The present meta-analysis could demonstrate good short- and long-term outcomes of the multiple total endovascular repair methods used in the aortic arch.
Introduction
As an alternative to the open repair, especially in patients who are assessed to be unfit for surgery the endovascular repair opened the door for treating aortic arch pathologies with encouraging results [1, 2].
The implementation of the endovascular procedures in this anatomical challenging zone paved the road for the development of several endovascular techniques like chimney (or parallel stents, snorkel, periscopes…) thoracic endovascular aortic repair (ChTEVAR), fenestrated/branched stent-grafts as custom made devices (CMD), surgeon modified TEVAR (SM TEVAR or back-table physician-modified grafts) and in situ fenestration.
However, the relatively small sample size of the published studies on the beforementioned procedures and the limited available outcome analysis of these procedures done in landing zone 0 and 1 beyond the left subclavia artery drove us to perform a dedicated review and meta-analysis of the available data. Moreover, we performed a regression analysis (meta-regression) to comparing the outcomes of these different endovascular methods.
Since the in situ-fenestration procedure has been extensively discussed in two recent reviews [3, 4], we decided not to include the studies describing this method in our review. However, we will discuss the in-situ fenestration in the light of the results of these review articles.
Methods
Search strategy (review protocol)
This review of the current data on the endovascular repair procedures of aortic arch (including chimney/snorkel technique, fenestrated and branched custom-made devices and surgeon modified devices) diseases was performed according to “The PRISMA 2020 statement: an updated guideline for reporting systematic reviews” [5].
We conducted an extensive search in PubMed/MEDLINE, Science Direct Databases, and the Cochrane Library. The search included all papers published up to January 2022 in any language. Medical Subject Headings terms used included “aortic arch,” “chimney stent grafts”, “chimney graft”, “chimney”, “periscope”, “snorkel”, “fenestrated”, branched”, “custom made”, “surgeon modified”, “physician modified”, “thoracic”, “arch”, “thoracic aorta”, “endovascular repair”, “TEVAR.” We looked also in the reference list of the retrieved studies for relevant articles. The review is registered in the International prospective register of systematic reviews (PROSPERO) under the ID: CRD42022367109.
Eligibility and inclusion criteria
A study was considered eligible according to the following inclusion criteria:
- (1)Describing an endovascular repair of the arch using chimney, fenestrated/branched (custom-made devices) or surgeon modified stent graft with or without adjunct extra-anatomic debranching.
- (2)Reporting of a case series of ≥10 patients in whom the brachiocephalic artery and/or the left common carotid artery was/were target vessels for the endovascular repair.
- (3)Providing information about at least one of the essential outcomes: early and late type I/III endoleak, 30-day mortality rate, early and late patency, incidence of perioperative stroke, re-intervention and retrograde aortic dissection.
Exclusion criteria
- (1)Case reports or case series with <10 patients.
- (2)Technical notes or case series without describing relevant outcomes.
- (3)Experimental works (in-vitro or ex-vivo).
Study selection
Articles were selected at the title and abstract level, and the selected papers were critically evaluated for eligibility before inclusion. Articles were excluded if they did not describe the type of the used technique and at least one of the basic outcomes or if the data were not original. Only the most recent report from each center was analyzed in case of duplicate publications on the same population of patients (Figure 1).
Data extraction
Two reviewers have independently extracted the following data from each study: Number of patients, number of target vessels, landing zones urgency of treatment, type of aortic lesion treated, type of stent grafts or stents used, technical success (defined as successful deployment of the main stent graft as well as the bridging stents/chimney grafts with complete exclusion of the aortic arch disease and without any type I or III endoleak on completion angiography), median length of follow-up, early and late patency, early and late type I /III endoleak, mortality rates, complications (retrograde A dissection), development of perioperative stroke and spinal cord ischemia.
Early patency and early endoleaks were defined as those identified during the first 30 postoperative days.
Outcomes
Primary outcomes were technical success, 30-day mortality, type I/III endoleak, stroke and spinal cord ischemia.
The secondary outcomes were the mortality, reintervention and patency during the follow-up.
Statistical analysis
Continuous variables were reported as median with quartiles (25% and 75%) or mean±standard deviation. The pooled proportions of event rates of the significant outcomes were calculated using the Freeman-Tukey transformation (arcsine square root transformation) to calculate the weighted summary proportion under the random-effects model (DerSimonian and Laird) [6]. Forest plot graphs were used to illustrate the weighted outcomes as well as the pooled estimation with the 95% confidence interval (CI).
We used the Cochran Q and I2 tests to assess the heterogeneity. Cochrane Q was calculated, and P<.05 was used to indicate the presence of heterogeneity. An I2 value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity (I2 between 30% and 60% indicates a moderate heterogeneity, >60% indicates substantial heterogeneity, and >75% indicates considerable heterogeneity).
Because of the design of this meta-analysis including single arm studies only, we assessed the bias of the eligible studies using the Egger regression analysis [7] (we considered P value <.05 significant with one-sided Egger test).
Additionally, we performed a meta-regression analysis (random-effect meta-regression) to compare the outcomes of the different studies while adjusting for the effect of the used therapy method. We considered a P<.05 as significant for the meta-regression.
The statistical analysis was performed using the Comprehensive Meta-Analysis Package V3 (Biostat, Englewood, NJ) statistical software.
Results
“Studies” characteristics
We have identified 5078 studies through the search in the databases and registers. After screening and assessment of the records, only 45 reports were identified as “eligible” for the analysis. Eventually, 26 studies [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34] with a total number of patients of 2327 with 3497 target vessels were included in the analysis (Figure 1). The number of patients varied between the studies regarding the operative method: 916 (39%) had ChTEVAR, 559 (24%) had CMD (fenestrated or branched) and 852 (37%) had SM-TEVAR.
The patients included in the analyzed studies were treated due to aortic dissection in 50% of cases, 37% aneurysms, 7% penetrating aortic ulcers, 2% intra mural hematoma, 2% pseudo-aneurysms and 2% due to endoleak Ia or traumatic aortic injury. Significantly more patients with aneurysm were treated with CMD and more dissections were managed with ChTEVAR or SM TEVAR (both P<.001).
The details of the included studies and the outcomes are summarized in Tables I and II.
The proximal landing zone was 0 in 12%, 84%, 21% and I in 23%, 12%, 15% and II in 65%, 4% and 64% in the ChTEVAR, CMD and SM TEVAR groups respectively.
The results of the meta-analysis demonstrated as an estimated pooled proportion stratified according to the used procedure as well as the heterogeneity and bias are shown in Table III.
Operative outcomes with meta-analysis
The studies reported a relatively high technical success rate between 84% and 100%. An exception was this reported by Dueppers et al. [13] with 37%. The overall estimated proportion of technical success for all procedures together reached 95.8% (95% CI: 93–97.6%) (Figure 2). Both the bias (P>.001) and heterogeneity between studies (Q. P<.001; I2=80.8%) were significant.
Moreover, the pooled estimation of the early type Ia/III endoleak was 8.1% (95% CI: 5.4–12.1%) but with a significant publication bias and heterogeneity (Q. P<.0001; I2=82% and P=.032, respectively) (Figure 3).
Regarding 30-day mortality, the overall estimated proportion was 4.6% (95% CI: 3.2–6.6%) with a significant heterogeneity (Q. P<.001; I2=58%) but not the bias (P=.084). However, the mortality as reported in the analyzed studies varied between 0% and 18.5% (Electronic supplementary material [ESM] 1).
Another important outcome here is the periprocedural stroke. We found an estimated proportion of the stroke (major and minor combined) of 4.8% (95% CI: 3.5–6.6%). Similar to “mortality” the bias was not significant (P=.358) and the heterogeneity was (Q. P .001; I2=50%) (ESM 2). The spinal cord ischemia (SCI) has an estimated rate of 2% (95% CI: 1.4–2.8%) without significant heterogeneity (Q. P=.67; I2=0%) or bias (P=.126).
Furthermore, the in the meta-analysis included studies reported a 0.9% event rate of retrograde type A aortic dissection (rAAD) as a complication resulting in a pooled estimation of 1.9% (95% CI: 1.3–2.7%) without a significant heterogeneity (Q. P=.831; I2=0%) or publication bias (P=.279) (ESM 3).
The re-interventions were required in 237 patients (10.2%) yielding a pooled estimated rate of 11.4% (95% CI: 7.8–16.4%) with a significant heterogeneity (Q. P<.001; I2=85%) and bias (P=.031) (ESM 4).
Interestingly a high rate of patency was reported during follow-up (all the studies but one [26] provided information) with an estimated event rate of 97.6% (95% CI: 95.8–98.7%). However, publication bias (P<.001) and heterogeneity (Q. P<.001; I2=67%) were both significant.
Meta-regression
A meta-regression analysis using the endovascular methods as moderators and the CMD as a reference group showed no significant variation between the groups in mortality (P=.324) or the rAAD (P=.219), showed however a significant difference between the therapeutic methods regarding stroke P<.001 (lower rate in ChTEVAR and SM vs. CMD). Similarly, the SCI was significantly more frequent in the CMD group compared to SM TEVAR group (P=.007) and showed a tendency to have a higher rate compared to ChTEVAR (P=.06).
Regarding endoleak, the SM and CMD groups have significantly lower endoleak rate than ChTEVAR group (P<.001 and P=.04, respectively). Moreover, the SM group had less endoleak than the CMD (P=.007). Furthermore, all three methods showed no significant difference in re-intervention rate during follow-up (P=.621).
Discussion
The present meta-analysis offers a summary with a detailed analysis of the up to date published data regarding the endovascular procedures the most used in the aortic arch focusing on the feasibility, safety and the operative outcomes.
In this analysis, we included only studies with at least 10 patients in whom two or more supra aortic vessels were endovascularly vascularized. The reason for this decision was the complexity and the frequent periprocedural complications of endovascular repair in zones 1 and 0 compared to more distal, proximal landing zones.
We found a high technical success estimated rate ≥94% throughout all the three procedural types (Table III). However, a significant heterogeneity in ChTEVAR group and publication bias in the CMD group were detected.
Despite the wide variation of the reported 30-day mortality in the analyzed studies between 0% and 18.5%, the estimated proportion was low with 4.6% (95% CI: 3.2–6.6%) particularly when compared to this reported by Liakopoulos et al. in their experience with frozen elephant truck (FET) in treating aortic arch diseases (11.1% for non-acute A dissections) [35] and to the mortality rate (8.8%) described in a large review of FET with over 3000 patients [36].
Other important outcomes, the periprocedural stroke and the SCI were estimated to be 4.8% (95% CI: 3.5–6.6%) and 2% (95% CI: 1.4–2.8%) respectively, in all procedures combined. These rates are clearly lower than the prevalence of stroke and SCI varying between 8.8–15.2% and 5.6–8.8% reported in most FET series [35, 37, 38]. However, a meta-analysis of arch hybrid repair and open repair from 2013 showed a lower prevalence of stroke with 7.6% and 6.2%, respectively [39].
Although the SM TEVAR and CHTEVAR showed in the meta-regression better stroke rate than the CMD, we should mention that the proximal landing zones 0 and I were clearly more frequent in the CMD compared to the ChTEVAR (35%) and SM TEVAR (36%) (ESM 5) groups, and this in turn can explain the higher calculated stroke rate in the CMD group [40].
Although the stroke rates in all groups were lower than that reported in the standard TEVAR [41], the cerebrovascular accidents still a major challenge of the aortic arch endovascular repair due to the risk of atherosclerotic and gaseous embolization. Thus, some authors recommend the temporary clamping of the carotid arteries during manipulation/deployment of the stent graft and to flush the devices with CO2 before saline to reduce the air trapped in the graft [29, 42].
Interestingly, the SCI was also higher in the CMD group with an incidence of >5% reported in three series; Tsilimparis et al. [29] reported a 5.6% rate with one case of permanent and two of temporary paraplegia without providing more information about the length of aortic covering, the use of spinal drainage or subclavian reconstruction. Kurimoto et al. [20] reported SCI in 3 patients (5.4%), in whom the distal end of the stent graft was at the level of Th 10, 8 and 6, one with a covered, not reconstructed left subclavian artery and two with occluded iliac arteries. An even higher rate (6.7% – only one patient) was mentioned by O’Callaghan et al. [25]. This patient developed lower limb weakness after removal of the spinal drainage and recovered after reinserting it. In the same study, the authors reported a 5.6% SCI between the 18 patients who received ChTEVAR. The covered subclavia artery in the patient who developed the ischemia was not reconstructed due to the urgency of case and without improvement after the delayed revascularization.
From the systematic analysis of Prendes et al. [4] only one study [43] met the inclusion criteria we used for this meta-analysis. In this study, 148 patients with 183 arch vessels were treated with in-situ laser fenestration of a thoracic stent-graft (using 980-nm laser system calibrated to deliver pulses at 18 W of energy in 3 seconds) due to aortic arch pathologies with 97.3% technical success.
The reported 30-day mortality was 2.9% with a stroke rate of 5% similar to the estimated pool proportion of our study groups.
There were seven endoleaks (4.7%: one type Ib, three type IIIC, and three type II), three retrograde dissections (2.0%), and five strokes (3.4%). No branch stent graft occlusion or spinal ischemia was reported.
To provide a comparative overview of the 3 methods we analyzed we performed a meta-regression analysis. This analysis showed a significant higher rate of endoleak Type I/III than SM TEVAR and CMD despite the significant heterogeneity of the ChTEVAR studies (Q. P<.001; I2=90%). The reported endoleak in the ChTEVAR series ranged between 5.6% [25] and 16.4% [16] except for the very high 78.8% endoleak rate reported by Dueppers et al. [13]. However, in this study (included 33 patients) no standardized method was used; chimneys, periscopes, sandwich and debranching with different target vessels, proximal neck lengths and diverse lengths of bridging stents. To reduce the incidence of gutter endoleaks in ChTEVAR a sealing zone distal to the last chimney graft of more than 10 mm was suggested [44].
The re-intervention was frequently needed in 11.4% (pooled) of patients but not significantly varied between the different methods used: ChTEVAR 11.5%, CMD 13.4% and SM TEVAR 8.4%.
Reasons for re-intervention were new endoleaks, stent graft migration, stent induced new entry (SINE), bridging stents occlusion, aneurysmal degeneration and new false lumen formation.
Although the re-intervention seemed to be relatively frequent, the reported patency during follow up was remarkably high; ≈98% as a pool estimation in ChTEVAR and CMD groups (with a significant publication bias) and 97% in the SM TEVAR group.
The most feared complication in the aortic arch is the rAAD. Fortunately, the incidence of this catastrophic complication still low with a pooled estimation of 2%. However, higher incidences (5.6–6.7%) were reported in several works [8, 17, 22, 25]. To reduce the incidence of the rAAD in TEVAR procedures; avoidance of post-ballooning, abstain from the use of stent grafts with proximal bare springs and from the excessive oversizing of the aortic stent grafts especially in patients with type B aortic dissection were suggested [45, 46].
Finally, although the CMDs (as fenestrated or branched) are going to be more and more implemented in the treatment of arch pathologies, the long time required for their production limits their use in the urgent situation. Therefore, the SM TEVAR and ChTEVAR still find their place in the endovascular armamentarium with acceptable outcomes, may be until an off-the-shelf fenestrated/branched device is available.
Limitations
An important limitation is the absence of data directly comparing the different endovascular methods in the aortic arch in term of a case-control study due to the nature of the patients and the pathologies treated using the endovascular method as an alternative to open repair.
Similarly, direct comparison between endovascular and open repair does not exist. Being applied mostly in older patients suffering from severe comorbidities and frailty, long-term results of endovascular repair of the aortic arch are lacking and against this background, favorable outcomes of open repair seem somehow unsurprising.
Moreover, the significant heterogeneity and bias seen in many outcomes of the reviewed study reduces the robustness of the reported outcomes.
Conclusions
The present meta-analysis could demonstrate acceptable short- and long-term outcomes of the multiple total endovascular repair methods (ChTEVAR, SM TEVAR and CMD) used in the aortic arch. Despite the higher stroke rate seen in the CMD groups, this method was used mainly in zone 0 and 1 compared to the other methods. The ChTEVAR had significantly more perioperative endoleaks.
However, the study is limited by the heterogeneity and publication bias and the absence of studies directly comparing the different methods with each other.
Electronic supplementary material
The electronic supplementary material (ESM) is available with the online version of the article at https://doi.org/10.1024/0301-1526/a001061
References
1 Current options and recommendations for the treatment of thoracic aortic pathologies involving the aortic arch: an expert consensus document of the European Association for Cardio-Thoracic surgery (EACTS) and the European Society for Vascular Surgery (ESVS). Eur J Cardiothorac Surg. 2019;55(1):133–62.
2 . A current systematic evaluation and meta-analysis of chimney graft technology in aortic arch diseases. J Vasc Surg. 2017;66(5):1602–10.e2.
3 . Endovascular in situ fenestration technique of aortic arch pathology: a systematic review and meta-analysis. Ann Vasc Surg. 2021;76:472–80.
4 . A systematic review of experimental and clinical studies reporting on in situ laser fenestration of aortic endografts. J Vasc Surg. 2022;75(2):740–52.e1.
5 The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Rev Esp Cardiol. 2021;74(9):790–9.
6 , Meta-analysis in the design and monitoring of clinical trials. Stat Med. 1996;15(12):1237–48. discussion 49–52.
7 . Systematic reviews in health care: investigating and dealing with publication and other biases in meta-analysis. BMJ. 2001;323(7304):101–5.
8 . Long-term outcomes after thoracic endovascular aortic repair using chimney grafts for aortic arch pathologies: 10 years of single-center experience. Ann Vasc Surg. 2021;72:400–8.
9 European Multicenter Registry for the performance of the chimney/snorkel technique in the treatment of aortic arch pathologic conditions. Ann Thorac Surg. 2016;101(6):2224–30.
10 . Double homemade fenestrated stent graft for total endovascular aortic arch repair. J Vasc Surg. 2019;70(4):1031–8.
11 . Double fenestrated physician-modified stent-grafts for total aortic arch repair in 50 patients. J Vasc Surg. 2021;73(6):1898–905.e1.
12 Case series of aortic arch disease treated with branched stent-grafts. Br J Surg. 2018;105(4):358–65.
13 Long-term results of total endovascular repair of arch-involving aortic pathologies using parallel grafts for supra-aortic debranching. J Vasc Surg. 2022;75(3):813–23.e1.
14 Fenestrated and scalloped endovascular grafts in zone 0 and zone 1 for aortic arch disease. Ann Vasc Surg. 2020;69:360–5.
15 Global experience with an inner branched arch endograft. J Thorac Cardiovasc Surg. 2014;148(4):1709–16.
16 Outcomes of chimney technique for aortic arch diseases: a single-center experience with 226 cases. Clin Interv Aging. 2019;14:1829–40.
17 Clinical outcomes of thoracic endovascular aneurysm repair using commercially available fenestrated stent graft (Najuta endograft). J Vasc Surg. 2015;62(6):1473–8.
18 . Early and midterm results of thoracic endovascular aortic repair using a branched endograft for aortic arch pathologies: a retrospective single-center study. JTCVS Tech. 2020;4:17–25.
19 . Handmade fenestrated stent grafts to preserve all supra-aortic branches in thoracic endovascular aortic repair. J Thorac Cardiovasc Surg. 2020;160(3):629–39.e1.
20 Thoracic endovascular aortic repair for challenging aortic arch diseases using fenestrated stent grafts from zone 0. Ann Thorac Surg. 2015;100(1):24–32. discussion 3.
21 Thoracic endovascular repair for aortic arch pathologies with surgeon modified fenestrated stent grafts: a multicentre retrospective study. Eur J Vasc Endovasc Surg. 2021;62(5):758–66.
22 Long-term outcomes of thoracic endovascular repair for aortic arch dissection using customized single-branched fenestrated stent-graft. Vasc Endovasc Surg. 2021;55(6):577–85.
23 Outcomes of total endovascular aortic arch repair with surgeon-modified fenestrated stent-grafts on zone 0 landing for aortic arch pathologies. J Endovasc Ther. 2022;29(1):109–16.
24 Endovascular repair by customized branched stent-graft: a promising treatment for chronic aortic dissection involving the arch branches. J Thorac Cardiovasc Surg. 2015;150(6):1631–8.e5.
25 . Outcomes for supra-aortic branch vessel stenting in the treatment of thoracic aortic disease. J Vasc Surg. 2014;60(4):914–20.
26 Long-term clinical outcomes of thoracic endovascular aortic repair for arch aneurysms with the Najuta thoracic stent-graft system. Ann Vasc Dis. 2020;13(4):384–9.
27 Endovascular treatment for aortic arch pathologies: chimney, on-the-table fenestration, and in-situ fenestration techniques. J Thorac Dis. 2020;12(4):1437–48.
28 Multicenter global early feasibility study to evaluate total endovascular arch repair using three-vessel inner branch stent-grafts for aneurysms and dissections. J Vasc Surg. 2021;74(4):1055–65.e4.
29 Single-center experience with an inner branched arch endograft. J Vasc Surg. 2019;69(4):977–85.e1.
30 . Fenestrated endovascular repair for diseases involving the aortic arch. J Vasc Surg. 2020;71(5):1464–71.
31 Endovascular treatment of post type A chronic aortic arch dissection with a branched endograft: early results from a retrospective international multicenter study. Ann Surg. 2021;273(5):997–1003.
32 . A challenging treatment for aortic arch aneurysm with fenestrated stent graft. Ann Thorac Surg. 2017;104(6):1915–22.
33 Off-the-shelf devices for treatment of thoracic aortic diseases: midterm follow-up of TEVAR with chimneys or physician-made fenestrations. J Endovasc Ther. 2020;27(1):132–42.
34 Fenestrated thoracic endovascular aortic repair using physician-modified stent grafts (PMSGs) in zone 0 and zone 1 for aortic arch diseases. Cardiovasc Intervent Radiol. 2019;42(1):19–27.
35 Single-center experience with the frozen elephant trunk procedure in 111 patients with complex aortic disease. J Thorac Dis. 2020;12(10):5387–97.
36 Frozen elephant trunk does not increase incidence of paraplegia in patients with acute type A aortic dissection. J Thorac Cardiovasc Surg. 2020;159(4):1189–96.e1.
37 Long-term experience with the E-vita Open hybrid graft in complex thoracic aortic diseasedagger. Eur J Cardiothorac Surg. 2017;51(2):329–38.
38 Single-centre experience with the frozen elephant trunk technique in 251 patients over 15 years. Eur J Cardiothorac Surg. 2017;52(5):858–66.
39 . A systematic review and meta-analysis of hybrid aortic arch replacement. Ann Cardiothorac Surg. 2013;2(3):247–60.
40 . Analysis of stroke after TEVAR involving the aortic arch. Eur J Vasc Endovasc Surg. 2012;43(3):269–75.
41 Thoracic Endovascular Aortic Repair of Aortic Arch Pathologies with the Conformable Thoracic Aortic Graft: Early and 2 year Results from a European Multicentre Registry. Eur J Vasc Endovasc Surg. 2016;51(6):791–800.
42 . Air embolism during TEVAR: carbon dioxide flushing decreases the amount of gas released from thoracic stent-grafts during deployment. J Endovasc Ther. 2017;24(1):84–8.
43 Early and midterm outcomes of in situ laser fenestration during thoracic endovascular aortic repair for acute and subacute aortic arch diseases and analysis of its complications. J Vasc Surg. 2020;72(5):1524–33.
44 , A shorter aortic arch sealing ring (<10 mm) distal to the most distal chimney seems to be an additional important factor contributing to a gutter endoleak in aortic arch TEVAR. Ann Vasc Surg. 2019;61:116–23.
45 Retrograde ascending aortic dissection during or after thoracic aortic stent graft placement: insight from the European registry on endovascular aortic repair complications. Circulation. 2009;120(Suppl 11):S276–81.
46 Retrograde type A aortic dissection after endovascular stent graft placement for treatment of type B dissection. Circulation. 2009;119(5):735–41.
