Acute aortic dissection with entry tear at the aortic arch: long-term outcome

Abstract

OBJECTIVES

The goal was to evaluate outcomes after conservative or surgical treatment of acute aortic arch dissections.

METHODS

Between January 2009 and December 2018, patients with a diagnosis of acute aortic dissection were analysed. Aortic arch aortic dissection was defined as a dissection with an isolated entry tear at the aortic arch with no involvement of the ascending aorta.

RESULTS

Aortic arch dissection was diagnosed in 31 patients (age 59 ± 11 years). Surgical intervention was performed in 13 (41.9%) cases. Overall in-hospital mortality was 3% (n = 1), and all deaths occurred in the conservative group (n = 1; 6%), whereas the overall stroke rate was 3% (n = 1), and all strokes occurred in the group treated surgically (n = 1; 8%). Surgical repair was necessary for the following conditions: end-organ malperfusion (n = 9; 69%), impending rupture (n = 3; 23%) and dilatation of the aorta with ongoing pain refractory to medical treatment (n = 1; 8%). Overall survival at the end of the follow-up period was 71%, with 77% in the surgical group and 63% in the conservative group (P = 0.91). Freedom from surgical intervention was 71%, with 82% in the surgical and 63% in the conservative group (P = 0.21), and freedom from a neurological event was 88%, with 89% versus 89% (P = 0.68) in the surgical and conservative groups, respectively.

CONCLUSIONS

Aortic arch dissection is a rare pathological condition that is one of the most challenging decision-making entities. Patients manifesting an uneventful course not requiring a surgical intervention during a hospital stay were at a higher risk for aorta-related intervention during the follow-up period. The treatment modality had no impact on survival or on the incidence of a neurological event.

INTRODUCTION

Acute aortic dissection occurs when blood flow disrupts intimal integrity and enters the media layer. This event separates the aortic wall into 2 layers, which results in a relatively unstable aortic wall architecture that may result in wall rupture or rapid expansion of the diameter of the aorta [1, 2].

Depending on the extent of the dissection, 2 clinical classifications are used [3, 4], both of which reflect therapeutic decision-making based on anatomical-pathological presentations of the acute event. In Stanford type A aortic dissection (DeBakey types I and II), a surgical approach is generally indicated, whereas in an uncomplicated Stanford type B (DeBakey type III) dissection, medical therapy is recommended [5, 6]. Because none of the classifications consider the site of an intimal tear, in 1994 von Segesser et al. [7] proposed the term type non-A non-B. In these forms, the dissection is limited to the aortic arch or to a retrograde dissection arising from the descending aorta, which extends into the arch and stops before the ascending aorta. Forms with retrograde extension of dissection to the ascending aorta or proximal aortic arch require surgical treatment, even if the target of the intimal tear-oriented approach is not the ascending aorta or aortic arch [7–9].

Because a true non-A non-B aortic dissection with an intimal tear localized at the aortic arch, which is distal to the innominate artery (InA) and proximal to the left subclavian artery (LSA), is a rare pathological entity, there is limited information on the nature of this particular aortic pathological condition in the recent literature. The goal of this retrospective study was to determine, on the basis of a prospective evaluation of all acute aortic dissections referred to our emergency department, the incidence of acute aortic arch dissections with isolated entry at the aortic arch and to evaluate early and midterm outcomes after conservative compared with surgical treatment.

METHODS

Study population

The university hospital radiology database was searched for all patients with clinical suspicion of an acute aortic dissection diagnosed between January 2009 and December 2018. The study protocol was approved by the local ethics committee at the University of Basel, Basel, Switzerland. Written informed consent was waived due to the retrospective design of the study (EKNZ 2019-00077).

The cumulative caseload of patients with clinical suspicion of acute aortic dissection who had electrocardiography gated computed tomography angiography scans was 2277 cases; 194 cases had a diagnosis of type A aortic dissection. Type B aortic dissection was diagnosed in 312 patients, and 31 (9.9%) had aortic dissection with an entry tear at the aortic arch. Cases with the entry tear distal to zone 3 and a retrograde expansion of the dissection into the aortic arch were not included (Supplementary Material, Fig. S1).

To provide a clear definition of non-A non-B dissection with the entry tear at the aortic arch, the landmarks of the aortic arch were defined as follows: (i) the ascending aorta was defined as the segment between the sinotubular junction and the pericardial fold; (ii) the aortic arch was defined as the aortic segment extending from the pericardial fold to the aortic isthmus at the attachment of the ligamentum arteriosum, just distal to the LSA; and (iii) the descending aorta was considered to be the segment of the aorta distal to the attachment of the ligamentum arteriosum at the aortic isthmus [10, 11] (Fig. 1).

Figure 1:

Aortic arch morphopathological division. The proximal segment of the aortic arch lies between the transition of the ascending aorta and the space between the InA and the LCCA (segment 2A). The middle segment (2B) is the part of the aortic arch with the origin of the LCCA as a central element. The distal segment of the aortic arch is defined as the area at the origin of the LSA, with its distal border at the level of the ligamentum arteriosum (segment 2C). InA: innominate artery; LCCA: left common carotid artery; LSA: left subclavian artery.

Depending on where the entry tear was located, the aortic arch was divided into 3 subsegments. The most proximal segment (segment 2A, Figs 1 and 2A) of the aortic arch was between the transition of the ascending aorta (at the superior pericardial fold) and the area between the InA and left common carotid artery. The middle segment (segment 2B) was the part of the aortic arch that included the origin of the left common carotid artery, and the most distal segment (segment 3C) of the aortic arch was the area at the origin of the LSA, with its distal border at the level of the ligamentum arteriosum (Fig. 1). The classification, with strict consideration of anatomical landmarks, similar to the definition in the STORAGE guidelines [12].

Figure 2:

Entry tear location in aortic arch dissection. According to the aortic arch segments (A), the entry tear may be (a) in the area between the transition of the ascending aorta and aortic arch and the space between the innominate artery and left common carotid artery (segment 2A, B); (b) at the middle segment (segment 2B, C) of the aortic arch; and at (c) the distal segment of the aortic arch proximal to the level of the ligamentum arteriosum (segment 2C, D).

The data were obtained from a prospectively maintained institutional database. Follow-up was based on information obtained from our outpatient clinic register. From missing patients, with no register or report from the outpatient clinic, information was obtained first from telephone calls to patients and secondly from the referring physician’s office.

Clinical parameters

In-hospital mortality was defined as death before discharge. A neurological event was defined according to the Valve Academic Research Consortium [13], and a perioperative stroke was defined as any neurological deficit with or without evidence of cerebral injury in a computed tomography scan and/or from magnetic resonance imaging.

The dissection’s anatomy and organ perfusion were assessed; patients presenting initially with hypertension or pain only were treated conservatively. Patients with signs of end-organ malperfusion (defined by clinical, laboratory and imaging evidence) or aortic rupture were treated on an emergency basis. Endovascular treatment involved thoracic endovascular aortic repair (TEVAR) with (TEVAR zone 2) or without (TEVAR zone 3) carotid-subclavian bypass as well as isolated stenting of dissected visceral vessels. The hybrid approach included sternotomy for supra-aortic vessels debranching with Dacron bypasses anastomosed on the ascending aorta and TEVAR with a stent graft landing in the distal ascending aorta (TEVAR zone 0). Open surgery included aortic arch replacement usually with a frozen elephant trunk technique. The decision on an endovascular, hybrid or open approach was made by the thoracic aortic team, which included a cardiac surgeon, a vascular surgeon and a radiologist. Patients with no complications were admitted to the intensive care unit for blood pressure monitoring, hypertension and pain treatment according to recent guidelines [1].

Image analysis

Aortic diameters and the entry location were analysed from the computed tomography scans. The diameter of the aorta was analysed at the ascending aorta, aortic arch, thoracic aorta at the level of the tracheal bifurcation and abdominal aorta at the level of the coeliac trunk. Total and true and false lumen diameters were measured at each level at a point perpendicular to the corresponding aortic segment centre line and were determined using multiplanar reconstruction software.

The total and the true lumen diameters were measured directly, and the false lumen was calculated as the difference between the total and the true lumens.

Statistical methods

Continuous data were presented as mean ± standard deviation if the distribution was normal and as median with interquartile ranges if the distribution was skewed. Categorical variables were presented as a number with percentage. Comparisons were made using the t-test or the Kruskal–Wallis rank test in continuous variables, as appropriate, and the Fisher’s exact test was used in binary variables. Long-term outcomes such as a neurological event, need for reoperation and death were analysed as time-to-event using the Kaplan–Meier method, using log rank tests for comparisons. All confidence intervals (CIs) and P-values were two-sided; P-values below 0.05 were considered statistically significant. We did not correct for multiple testing. All analyses were performed using Stata 15 (Stata Corp., College Station, TX, USA).

RESULTS

The analysis included a total of 31 patients, with a mean age 59.0 ± 11 years; 8 (26%) were women. Patient demographics are presented in Table 1. In 19 (61%) patients the dissection entry tear was identified at the origin of the LSA (segment 2C, Figs 1 and 2D); in 11 (35%) the entry tear was between the area of the LSA and the InA (segment 2B, Figs 1 and 2C) and in 1 (3%) patient the entry tear was vis-à-vis the origin of the InA (segment 2A, Figs 1 and 2B) at the smaller curvature of the aortic arch.

Table 1:

Patient demographics at diagnosis of aortic arch dissection

	All patients (n = 31)	Conservative treatment (n = 18)	Surgical treatment (n = 13)	P-value
Age	59.0 ± 11.0	61.7 ± 11.1	55.8 ± 10.0	0.11
Gender (female)	8 (26)	7 (39)	1 (8)	0.095
History of myocardial infarction	2 (6)	2 (11)	0 (0)	0.49
Diabetes mellitus	0 (0)	0 (0)	0 (0)	1.00
Hypertension	27 (87)	16 (89)	11 (85)	1.00
Hypercholesterolaemia	11 (35)	10 (56)	1 (8)	0.008
COPD	3 (10)	3 (17)	0 (0)	0.24
Renal failure	3 (10)	1 (6)	2 (15)	1.00
History of stroke	1 (3)	0 (0)	1 (8)	1.00
PAD	2 (6)	1 (6)	0 (0)	1.00
Previous cardiac surgery	1 (3)	1 (6)	0 (0)	1.00
Marfan syndrome	3 (6)	1 (6)	2 (15)	0.55

Data are presented as categorical variables, given as counts and percentages or means with standard deviations.

COPD: chronic obstructive pulmonary disease; PAD: peripheral artery disease.

In a majority of patients, the dissection extended to the iliac arteries in 68% and to the abdominal aorta in 37% of cases. In 96% of cases the area of the aortic arch between the left common carotid artery and LSA (segments 2B and 2C) was involved in the dissection, whereas in 1 case the diagnosis was an entry tear at the segment corresponding to the InA origin (segment 2A). The diameters of the ascending aorta, the aortic arch and the thoracic aorta were under 45 mm at the acute event with the exception of 1 patient, in whom the diameter of the aortic arch was 46 mm and the upper interquartile range was below 40 mm (Table 2).

Table 2:

Aortic anatomy at diagnosis of aortic arch dissection

	All patients (n = 31)	Conservative treatment (n = 18)	Surgical treatment (n = 13)	P-value
Entry tear location
At InA (segment 2A)	1 (3)	1 (6)	0 (0)	1.00
At LCCA (segment 2B)	11 (35)	8 (44)	3 (23)	0.28
At LSA (segment 2C)	19 (61)	9 (50)	10 (77)	0.16
Bovine aortic arch	2 (6)	1 (6)	1 (8)	1.00
Diameter of corresponding aortic segment (mm)
AscAo	36.0 (34.8–38.9)	35.8 (34.1–39.5)	36.4 (35.4–37.9)	0.91
AoArch	31.3 (29.8–33.5)	30.6 (29.6–34.9)	32.0 (30.4–33.5)	0.68
True lumen	30.7 (29.6–32.4)	30.4 (29.6–32.5)	30.9 (29.5–32.2)	0.85
False lumen	0.0 (0.0–11.4)	0.0 (0.0–11.4)	0.0 (0.0–13.4)	0.95
ThAo	29.5 (28.2–32.1)	30.6 (27.9–33.8)	28.9 (28.3–32.0)	0.68
True lumen	19.9 (17.8–21.4)	19.9 (17.6–22.1)	19.8 (18.3–20.9)	1.00
False lumen	29.6 (26.5–36.2)	30.7 (27.0–38.6)	27.8 (25.8–31.3)	0.074
AbdAo	25.8 (23.7–29.6)	27.3 (24.7–30.6)	24.4 (23.3–28.1)	0.082
True lumen	20.3 (17.1–21.9)	21.1 (18.6–22.1)	17.6 (16.0–21.5)	0.057
False lumen	24.4 (22.3–27.5)	25.5 (22.8–31.1)	22.8 (18.6–26.6)	0.10

Categorical variables are presented as counts and percentages; continuous variables, as median values (first quartile–third quartile). Diameters are given in mm. ‘Bovine arch’ is a common arch origin of the innominate and the left common carotid arteries.

AbdAo: abdominal aorta; AoArch: aortic arch; AscAo: ascending aorta; InA: innominate artery; LCCA: left common carotid artery; LSA: left subclavian artery; ThAo: thoracic aorta.

Aortic repair during the initial hospital stay was performed in 13 (41.9%) patients. The reasons for urgent repair were as follows: end-organ malperfusion in 9 (69%), impending rupture in 3 (23%) and dilatation of the aorta with ongoing pain with hypertension refractory to medical therapy in 1 (8%) case. Ongoing pain, as a concomitant symptom in the surgical group, was present in 11 (25%) cases.

In the surgical group, TEVAR was conducted in 7 (26%) patients; 2 cases were conducted in zone 3; 4 cases, in zone 2 with carotid-subclavian bypass and 1 case was conducted in zone 1 with transposition of the left carotid artery and the LSA. In 1 case, a conventional elephant trunk procedure was done with replacement of the aortic arch. In 5 (38%) cases, dissection flap fenestration with angioplasty of the renal and the iliac arteries was conducted as an isolated intervention.

Overall, in-hospital mortality was 3% (n = 1); 1 patient in the conservative treatment group died (P = 1.0). The overall incidence of a neurological event during the hospital stay was 3% (n = 1), and it occurred in the surgical group (P = 1.0). New renal impairment was diagnosed in 3 (10%) cases: 2 (15%) were in the surgical and 1 (6%) was in the conservative group (P = 0.56). The incidence of myocardial infarction was 3% (n = 1), and this 1 case (6%) occurred in the conservative group. The mean hospital stay was 9.5 ± 7.1 days: 14.1 ± 8.7 in the surgical group and 6.3 ± 3.1 in the conservative group (P = 0.001) (Table 3).

Table 3:

Early outcome for conservative and surgical treatment strategies in aortic arch dissection

	All patients (n = 31)	Conservative treatment (n = 18)	Surgical treatment (n = 13)	P-value
Renal failure	3 (10)	1 (6)	2 (15)	0.56
Stroke	1 (3)	0 (0)	1 (8)	0.42
In-hospital deaths	1 (3)	1 (6)	0 (0)	1.00
Atrial fibrillation	4 (13)	1 (6)	3 (23)	0.28
Severe cardiac failure	1 (3)	0 (0)	1 (8)	0.42
Respiratory failure	4 (13)	2 (11)	2 (15)	1.00
Sepsis	1 (3)	0 (0)	1 (8)	0.42
Myocardial infarction	1 (3)	1 (6)	0 (0)	1.00
Hospital stay (days)	9.5 ± 7.1	6.3 ± 3.1	14.1 ± 8.7	0.001
ICU stay (days)	1.0 (1.0–3.0)	1.0 (1.0–1.0)	1.5 (1.0–3.8)	0.028

Categorical variables are presented as counts and percentages, continuous variables as mean ± standard deviation if normally distributed and median values (first quartile–third quartile) if skewed.

ICU: intensive care unit.

The follow-up period was 5.4 ± 3.9 years. Overall survival at 7 years was 71% (CI 38–89%), 77% (CI 31%–94%) in the surgical and 63% (CI 14%–89%) in the conservative group (Fig. 3). Overall freedom from aorta-related surgical intervention was 71% (CI 51%–85%), 82% (CI 46%–95%) in the surgical and 63% (CI 34%–82%) in the conservative group (Fig. 4). Overall freedom from neurological events at 7 years was 88% (CI 67%–96%), 89% (CI 43%–98%) in the surgical and 98% (CI 62%–97%) in the conservative group (Fig. 5).

Figure 3:

Kaplan–Meier survival curve. The 1-, 5- and 7-year survivals were 100%, 100% and 63% [95% confidence interval (CI) 14–89%] in the conservative group and 92% (95% CI 57–99%), 92% (95% CI 57–99%) and 77% (95% CI 31–94%); P = 0.91, respectively, in the surgical group.

Figure 4:

Kaplan–Meier curve for freedom from aortic-related reinterventions. The 1-, 5- and 7-year freedom from interventions was 83% [95% confidence interval (CI) 57–94%], 63% (95% CI 34–82%) and 63% (95% CI 34–82%) in the conservative group and 92% (95% CI 54–99%), 82% (95% CI 46–95%) and 82% (95% CI 46–95%), P = 0.21, respectively, in the surgical group.

Figure 5:

Kaplan–Meier curve for freedom from neurological events. The 1-, 5- and 7-year freedom from neurological events was 89% [95% confidence interval (CI) 62–94%], 89% (95% CI 62–97%) and 89% (95% CI 62–97%) in the conservative group and 100%, 89% (95% CI 43–98%) and 89% (95% CI 43–98%), P = 0.68, respectively, in the surgical group.

During the follow-up period, a total of 11 (35%) patients needed aortic reintervention, 2 (15%) from the surgical and 9 (50%) from the conservative group (P = 0.07). In 1 surgical patient, in whom TEVAR was performed in zone 2 during the initial hospital stay, a retrograde type A aortic dissection occurred during the follow-up period. In a second case, at the initial hospital stay, a dissection flap was fenestrated, and an open thoraco-abdominal aorta replacement was conducted during the follow-up period. Aortic remodelling is presented in Supplementary Material, Table S1.

In the conservative treatment group, in 5 cases, a TEVAR in zone 2 was performed with a carotid-subclavian artery bypass. In addition, 1 patient had a dissection membrane fenestration. The frozen elephant trunk procedure was performed in 2 cases, and a conventional aortic arch replacement was done in 1 case. In 1 patient, fenestration of the dissection membrane was performed as an isolated intervention due to visceral organ malperfusion.

DISCUSSION

This study evaluated the outcome of surgical and conservative treatment in isolated aortic arch dissection. At our institution, all patients without complications related to the aortic dissection, in accordance with the recommendations for type B aortic dissection, underwent conservative treatment [1]. Patients with signs of end-organ malperfusion or aortic rupture were treated on an emergency basis. Delayed intervention during the index hospital stay was performed with evidence of aortic dilatation and for ongoing pain with hypertension refractory to medical therapy. This decision-making strategy resulted in similar outcomes in regard to in-hospital major adverse events among the 2 treatment groups. At midterm there was no significant difference in survival compared with the incidence of neurological events between the 2 groups. Indeed, during the follow-up period, there was a clear tendency for an elevated intervention rate in the conservative group, where 50% of cases from the initial cohort needed a reintervention related to a late complication of the untreated aortic dissection. This result is not surprising. It is clear that, from the perspective of the pathology of the disease, aortic dilatation, with the potential to rupture, is a part of the natural history of untreated aortic dissection. The term acute aortic arch dissection as an independent entity was initially considered by von Segesser et al. [7] as non-A non-B aortic dissection and was defined either as an acute event located at the aortic arch or as a retrograde dissection of a type B dissection not involving the ascending aorta. The definition of non-A non-B aortic dissection is still a subject of discussion, and it seems that there is no clear consensus regarding the nomenclature or the treatment strategy. This situation is due mostly to limited reports focusing on this pathological condition and to the lack of clear agreement regarding their interpretation [8, 9, 14, 15]. Controversy is well reflected in recently published recommendations. On the one hand, aortic arch involvement with the exception of zone 0 is considered a type B pathological condition [16], whereas, on the other hand, non-A non-B aortic dissection is recognized as an independent entity, even in cases with evidence of retrograde involvement of the arch with entry at the distal thoracic aorta [17].

The presence of aortic arch involvement in association with a type B dissection with retrograde expansion is estimated to be >25%, and the presence of aortic arch dissection in this constellation is considered to be a complicated evolution of type B aortic dissection [8, 9, 14, 15, 18, 19]. Meanwhile, the clinical picture of the retrograde propagation of a type B dissection into the aortic arch may be considered in the literature as non-A non-B aortic dissection [7, 16, 17, 19, 20]. The propagation of retrograde dissection into the LSA in type B dissection occurs much more frequently than its extension beyond it. Consequently, counting LSA involvement in cases with evidence of distal thoracic entry, which is in reality a part of the type B pathology, as non-A non-B dissection, may lead to misinterpretation in treatment strategy [9, 14]. From our point of view, the isolated involvement of the LSA in retrograde dissection expansion is debatable as a criterion for the diagnosis of non-A non-B dissection. Instead the definition should include cases with a clear tear site at the aortic arch.

The extension of the dissection as such is not the most important pathological process. Rather, we suggest considering the location of the entry tear as a primary landmark for classification. The hypothesis that was offered in the past pointed out the entry tear-oriented surgical approach. In this setting, for example, the strategy for a retrograde type A dissection in cases with a distal tear in the thoracic aorta is not the same as that for an entry at the ascending aorta [7, 21–23].

To address this hypothesis, in our analysis, only cases with entry at the aortic arch were included. To create more clarity concerning borderline cases with an entry tear distal to the LSA at the descending aorta, the aortic arch as an anatomical-pathological unit was subdivided into the 3 segments (Fig. 1). The borders of the aortic arch were defined by natural landmarks: the proximal border at the transition between the ascending aorta and the aortic arch at the pericardial fold, and distal just at the origin of the LSA at the level of the ligamentum arteriosum [10–12]. In a normal case, this is placed vis-à-vis the origin of the LSA or distal (<1 cm) to it at the small curvature of the aortic arch. It is important to consider that, between defined landmarks, the aortic arch is adherent to the surrounding osseous and mediastinal tissue (Fig. 1). This situation may have an impact on dissection propagation as well on the locations of the entry tears [21]. The hypothesis is in part supported by past haemodynamic investigations. There is a net presence of pathological haemodynamic conditions, prone to intimal tear at the origin of the great vessel at the aortic arch. In contrast, shear stress, flow and pressure patterns present physiological patterns at the aortic isthmus and transition between the ascending aorta and the aortic arch [23–25]. These are regions where the aortic arch is transient to ‘mobile’ zones of the aorta.

A comparison of our results with those in the literature is limited by 2 important facts. One, the prevalence of non-A non-B aortic dissection as such is extremely rare: it is described only in a handful of reports with a limited case load. Additionally, the published reports do not distinguish between cases with the intimal tear distal to the LSA and those with the initial intimal tear at the aortic arch. Urbanski and Wagner [21] proposed an aggressive surgical approach in non-A non-B aortic dissection with an intimal tear in the aortic arch in order to avoid aorta-related adverse events during the follow-up period. However, the study was conducted on a small case load with only 8 patients, and an intimal tear in the aortic arch was diagnosed in just 1 patient [21]. In the largest series, which analysed 43 patients with non-A non-B dissection, in almost 50% of cases the entry tear was at the descending aorta [19]. In only a few cases was there evidence of aortic arch involvement via the retrograde expansion of the dissection membrane, which also included the retrograde involvement of the origin of the LSA. This fact suggests that in the descending entry subgroup of the non-A non-B group, defined by Rylski et al. [20] in a majority of cases, dissection was limited to the descending thoracic aorta, a pattern that is typical for type B aortic dissection. In a true arch entry subgroup, in almost 20% of cases a conservative approach was chosen. In the surgical approach within this subgroup, in ∼60% of cases, the entry tear was addressed. There was no difference in the outcomes of the 2 groups with aorta-related reinterventions in terms of survival or of the location of the entry tear. In addition, no subanalysis of conservative versus surgical therapy conducted.

Limitations

This was a single-centre study with all the limitations associated with a retrospective data collection process. Selection bias as such cannot be excluded. The results presented should be considered in light of the limited case load. The location of the re-entry tear was not assessed but may have an impact on outcome. This is especially true in the development of re-dissection and/or endoleak. Aortic arch dissection with the entry tear localized in the area between the InA and the LSA is a rare entity.

In acute aortic arch dissection, surgical intervention is required in complicated cases with end-organ malperfusion, aortic rupture or uncontrolled pain. Cases with an uneventful course may be treated conservatively, with excellent long-term results.

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

Conflict of interest: none declared.

Author contributions

Luca Koechlin: Conceptualization; Data curation; Investigation; Methodology; Validation; Visualization. Julia Schuerpf: Conceptualization; Data curation; Investigation; Methodology; Software. Jens Bremerich: Data curation; Supervision; Validation; Writing—review & editing. Gregor Sommer: Conceptualization; Data curation; Formal analysis; Software. Brigitta Gahl: Conceptualization; Data curation; Formal analysis; Resources; Writing—review & editing. Oliver Reuthebuch: Project administration; Resources; Software; Validation. Lorenz Gurke: Investigation; Methodology; Resources; Supervision. Edin Mujagic: Conceptualization; Data curation; Investigation; Methodology. Friedrich Eckstein: Conceptualization; Formal analysis; Supervision; Writing—review & editing. Denis A. Berdajs: Conceptualization; Data curation; Formal analysis; Methodology; Project administration; Supervision; Validation; Writing—original draft.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Anthony L. Estrera, Roman Gottardi and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

ABBREVIATIONS

CI

Confidence interval

InA

Innominate artery

LSA

Left subclavian artery

TEVAR

Thoracic endovascular aortic repair