Blunt traumatic thoracic aortic injuries: a retrospective cohort analysis of 2 decades of experience

Abstract

OBJECTIVES

The aim of this study was to analyse and report the changes in the management of blunt traumatic aortic injuries (BTAIs) in a single centre during the last 2 decades.

METHODS

A retrospective analysis of all patients diagnosed with BTAI from January 1999 to January 2020 was performed. Data were collected from electronic/digitalized medical history records.

RESULTS

Forty-six patients were included [median age 42.4 years (16–84 years), 71.7% males]. The predominant cause of BTAI was car accidents (54.5%, n = 24) and all patients presented with concomitant injuries (93% bone fractures, 77.8% abdominal and 62.2% pelvic injuries). Over 70% presented grade III or IV BTAI. Urgent repair was performed in 73.8% of patients (n = 31), with a median of 2.75 h between admission and repair. Thoracic endovascular repair (TEVAR) was performed in 87% (n = 49), open surgery (OS) in 10.9% (n = 5) and conservative management in 2.1% (n = 1). Technical success was 82.6% (92.1% TEVAR, 79% OS). In-hospital mortality was 19.5% (17.5% TEVAR, 40% OS). Of these, 3 died from aortic-related causes. Seven (15.2%) required an early vascular reintervention. The median follow-up was 34 months (1–220 months), with 19% of early survivors having a follow-up of >10 years. Only 1 vascular reintervention was necessary during follow-up: secondary TEVAR due to acute graft thrombosis. Of the patients who survived the initial event, 6.7% died during follow-up, none from aortic-related causes.

CONCLUSIONS

Even with all the described shortcomings, in our experience TEVAR for BTAI proved to be feasible and effective, with few complications and stable aortic reconstruction at mid-term follow-up. With the current technical expertise and wide availability of a variety of devices, it should be pursued as a first-line therapy in these challenging scenarios.

INTRODUCTION

Blunt traumatic aortic injury (BTAI) is a catastrophic injury associated to high mortality rates, normally occurring after sudden deceleration, and usually seen in high speed motor vehicle accidents and falls from great heights [1]. In the Advanced Trauma Life Support, it is identified as one of the 8 ‘life-threatening chest injuries’ and has been implicated as the second most common cause of death in trauma patients [2, 3]. Estimates suggest that <25% of patients with such an injury reach the hospital, and of those who do, 50% die within 24 h [3, 4].

During the last decades, there has been a paradigm shift in the management of BTAI, with thoracic endovascular aortic repair (TEVAR) becoming the primary line of treatment. The 2011 Society for Vascular Surgery (SVS) clinical practice guidelines for the management of BTAI recommend TEVAR over open repair or non-operative management, given better survival rates and lower rates of paraplegia [3, 5]. An SVS endorsed severity classification system was proposed by Azizzadeh et al., including 4 grades of BTAI: type I (intimal tear), type II (intramural haematoma), type III (pseudoaneurysm) and type IV (rupture, complete transection) [3, 5, 6]. The SVS guidelines recommend expectant management with serial imaging for grade I injuries and urgent TEVAR for grades II–IV [3]. Similarly, the 2017 European Society for Vascular Surgery Guidelines recommend TEVAR as first-line treatment, with urgent repair of patients with free ruptures or large periaortic haematomas [1]. Although the superiority of TEVAR has been established, controversy still surrounds certain aspects, including the optimal timing of the intervention, graft selection, management of the subclavian artery and long-term durability of these repairs.

The objective of this study was to analyse and report this centre’s experience with BTAI, looking specifically at 30-day all-cause and aortic injury-related mortality. Furthermore, we wanted to determine potential differences in management strategies through the last 2 decades, subdividing patients treated between 1999 and 2006, 2007 and 2013 and 2014 and 2020.

MATERIALS AND METHODS

Ethical statement

Approval from the ethical board was not required, given the retrospective nature of the study and absence of any identifiable patient information reported.

Study design

A retrospective analysis of all consecutive patients coded with the diagnosis of traumatic aortic injury from January 1990 to January 2020 in a tertiary vascular centre was performed. Medical histories were examined, and patients without sufficient information in their medical history records (patients treated before the year 1999) were excluded due to the lack of data. The ‘Strengthening the Reporting of Observational Studies in Epidemiology’ statement guidelines were used for the preparation of this study and the assessment of data collection. Furthermore, the statistical and data reporting guidelines for the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery were followed accordingly [7].

The primary outcome of this study was 30-day all-cause and aortic injury-related mortality. Secondary outcomes were to evaluate differences in management between 1999 and 2009 and 2010 and 2020, looking specifically into trends regarding the type of preferred treatment, the timing of the aortic procedure and the degree of oversizing.

Study variables

Variables were divided into 3 categories: preoperative data, intraoperative and 30-day short-term results and mid- to long-term data. Preoperative variables included were comorbidities, psychiatric disorders, cause of trauma, concomitant injuries and preoperative computed tomography angiography (CTA) (grade of the BTAI according to Azizzadeh et al.’s [6] classification, aortic diameters and type of aortic arch). CTA evaluation was performed with the Aquarius iNtuition TeraRecon workstation and Syngo Imaging. Aortic diameters were measured at the proximal and distal landing zones as specified in the surgical report and postoperative CTA. Treatment strategy (open, endovascular or conservative management), time between admission and aortic surgical repair, timing of concomitant surgical treatment, management of the subclavian artery, stent graft diameters in patients undergoing TEVAR and degree of proximal and distal oversizing and postoperative complications were also collected. Long-term outcomes assessed were death, vascular reinterventions and CTA availability at 5 and 10 years of follow-up. Long-term data were collected via hospital records and telephone interviews. However, due to the antiquity of some of the patients, in some cases, demographic and contact details were unavailable. After the telephonic interview, a vascular consultation with aortic imagining was arranged. The study end date was defined to be the 31st of January 2020 and the follow-up index was calculated for each patient to increase the validity of the study [8].

Statistical analysis

Statistical analysis was performed with IBM^® SPSS^® Statistics Version 26. Variables with over 25% of missing values were not included in this study. Normality was evaluated using the Shapiro–Wilk W test. Normally distributed variables were expressed as mean ± standard deviation, while variables that did not follow a normal distribution were expressed with medians and ranges or interquartile ranges (IQR). Mid- to long-term outcomes were evaluated as time-to-event outcomes with survival analysis.

RESULTS

Preoperative data

A total of 69 patients were identified with the coded diagnosis of BTAI. Of these patients, 21 were excluded due to missing patient-data, all of whom treated prior to 1999, and 2 due to wrongly coded diagnoses (1 penetrating aortic injury and 1 ruptured penetrating aortic ulcer), making the total number of patients in this study 46. The median age was 42.4 years (16–84 years) and 71.7% (n = 33) were males. The most prevalent concomitant pathologies were hypertension (37%, n = 17) and dyslipidaemia (18.5%, n = 8) and 10.9% (n = 5) had prior coronary artery disease. An underlying psychiatric condition was observed in 10.9% (n = 5), mainly schizophrenia. The predominant cause of BTAI was car accidents (54.5%, n = 24) and all patients presented with concomitant injuries (Table 1). Diagnosis of the BTAI was made with CTA. Over 70% of patients presented grade III or IV BTAI lesions, with a similar distribution of the lesions in the aortic isthmus and in the descending aorta. Preoperative CTA lesion and aortic characteristics are represented in Table 2.

Table 1:

Characteristics of blunt traumatic aortic injury patients

	Global (n = 46)
Causes of injury , n (%)
− Car accidents	24 (54.5)
− Motorcycle accidents	10 (22.7)
− Accidental precipitation injuries	6 (13.6)
− Suicide attempt	3 (6.8)
Concomitant injuries , n (%)
− Thoracic injuries	43 (95.6)
− Bone fractures	42 (93.3)
− Abdominal injuries	35 (77.8)
− Pelvic injuries	28 (62.2)
− Brain injuries	19 (59.1)
− Carotid dissection	4 (9.1)
− Spinal cord injury	4 (8.9)
Arrived intubated	24 (52.9)
ISS, mean ± SD	43.8 ± 2.3

IQR: interquartile range; ISS: injury severity score; SD: standard deviation.

Table 2:

Preoperative computed tomography angiography aortic and lesion morphology

	Global (n = 46)
Grade of aortic injury , n (%)
− Type I	−3 (6.5)
− Type II	−9 (20)
− Type III	−16 (35.5)
− Type IV	−17 (37.7)
Location of aortic injury , n (%)
− Aortic arch
− Outer curvature	−4 (9.4)
− Inner curvature	−20 (43.8)
− Descending aorta	−21 (46.9)
Type of aortic arch , n (%)
− Type I	−24 (53.1)
− Type II	−14 (31.3)
− Type III	−7 (15.6)
Proximal aortic diameter (mm), mean ± SD	24 ± 0.7
Distal aortic diameter (mm), mean ± SD	21 ± 1.2

SD: standard deviation.

Procedural operative data and short-term outcomes

Patients underwent urgent repair in 73.8% of cases (n = 31), while 26.2% (n = 11) were performed in a delayed manner. The median time between admission and aortic treatment was 2.75 h (range 30 min to 432 h). Over 90% of patients required concomitant surgical procedures. In 64.3% of patients with additional procedures, aortic repair was the first intervention to be performed. Thoracic endovascular repair (TEVAR) was performed in 87% (n = 40), open surgery in 10.9% (n = 5) and conservative management in 1 patient. Technical success was 82.6% (92.1% for TEVAR, 79% for open surgery). Patients underwent general anaesthesia in 95% of cases, and TEVAR was performed under local anaesthesia in 5%. All-cause in-hospital mortality was 19.5% (n = 9) (17.5% TEVAR vs 40% open surgery). Aortic-related mortality was 6.5% (5% TEVAR vs 20% open repair). Ten procedural-related complications occurred [7 after TEVAR (17.5%) and 3 after open surgery (60%)].

In patients undergoing TEVAR, the proximal landing zone was Ishimaru zone 2 in 22 patients, zone 3 in 12 patients and zone 4 in 2 patients. Concomitant revascularization of the left subclavian artery was not performed in any of the 22 patients who received primary coverage. Of these, 2 patients (9.1%) required a carotid-subclavian bypass due to postoperative arm ischaemia. Endografts used for repair included Cook Zenith TX2 (26/40) (Cook Medical, Bjæverskov, Denmark), Cook Zenith Alpha (1/40) (Cook Medical), Valiant (4/40) (Medtronic PLC, Dublin, Ireland) and TAG^® (4/40) (W.L. Gore & Associates, Delaware, USA), depending on the physician preference and availability. All of the implanted endografts had a straight configuration, with diameters ranging between 22 and 36 mm. The most frequently implanted diameter was 22 mm (29.8%), with over 80% of devices under 30 mm. In 97.1% of patients, only 1 endograft was necessary. The mean proximal oversizing was 11.5 ± 2% and the median distal oversizing was 29.4% (IQR, 18.2–44.4%). Percutaneous femoral access was performed in 25.7% of patients and surgical cut-down in 74.3%. Seven patients developed procedural-related complications and required an early reintervention. Two patients required a carotid-subclavian bypass, 2 a femoral endarterectomy due to infrainguinal limb ischaemia and 1 patient developed colonic ischaemia. Two patients underwent immediate conversion to open surgery: 1 patient presented haemodynamic instability with a large intrathoracic haematoma, not resolved after endograft placement. Thoracotomy and drainage of the haematoma was performed after TEVAR, without resolution of the situation and intraoperative death. The second patient was treated in 1999, and one of the first TEVAR cases performed in this centre and had a stent fracture with secondary aortic occlusion. Urgent conversion with thoracotomy and prosthetic substitution of the thoracic aorta was performed; however, the patient died shortly after the intervention.

Five patients underwent open repair, with the last open surgery performed in 2003. Two patients died shortly after the operation: 1 due to massive haemodynamic shock and aortic bleeding and the other from multiorgan failure in the postoperative period. Of the remaining 3 patients, 1 developed spinal cord ischaemia after aortic replacement, as well as hoarseness and left vocal paralysis. No 30-day or in-hospital vascular reinterventions were recorded in these patients.

Mid- and long-term follow-up

Follow-up was quantified from the date of the procedure to their last available consultation, and the study end date was defined as the 31st of January 2020. Of the 37 patients who left the hospital alive, 16 were lost during the follow-up (43.2%), 13 during the first year, while 3 patients were lost after having been followed for over 30 months. Regarding the follow-up index, it was 1 in 43.5% (20 patients), between 0.90 and 0.99 in 6.5% and between 0.50 and 0.89 in 6.5%. Twenty patients (43.5%) had an FUI of <0.5, of whom half (10 patients or 21.7%) were 0.01, and lost shortly after discharge. This was mainly due to the relocation of patients from this hospital to other smaller clinics or trauma-specialized centres. The FUI is represented in Figs. 1 and 2.

Figure 1:

A graphic representation of the follow-up index.A distribution of the follow-up index of each patient according to the date of admission. Red dots represent patients who died during the follow-up, while blue dots represent patients who are still alive. Patients coded in blue with a 1 follow-up index completed the study (study end date: 31 January 2020).

Figure 2:

A histogram of the follow-up index distribution, with 10 patients presenting an index of less 0.01 and 20 patients an index of 1.

In total, long-term data were available for 24 patients. The median follow-up for these patients was 34 months (range, 1–220 months), 8 patients have over 5 years of follow-up and 7 patients had >10-year follow-up (Table 3). Two patients died during this period, 1 from a stroke 1 year after the procedure and another from cancer. No vascular deaths were recorded. The global survival of these patients is depicted with Kaplan–Meier curves in Fig. 3. Only 1 vascular-related reintervention was documented 12 months after the index procedure, a case of acute stent graft thrombosis, resulting in a subtotal lumen stenosis and requiring redo TEVAR (Fig. 4).

Table 3:

Patient follow-up

	Group 1 (1999–2006)	Group 2 (2007–2013)	Group 3 (2014–2020)
Total number of patients	23	7	16
Number of TEVAR, n (%)	17 (74)	7 (100)	16 (100)
Deaths, n (%)	7 (30)	2 (28)	2 (12.5)
Loss of FU before 1 year, n (%)	7 (30)	3 (43)	8 (50)
FU >1 year, n (%)	9 (39)	2 (29)	7 (44)
FU >3 years, n (%)	8 (35)	1 (14)	3 (19)
FU >5 years, n (%)	6 (26)	1 (14)	1 (6)
FU >10 years, n (%)	6 (26)	1 (14)	0 (0)

TEVAR: thoracic endovascular repair; FU: follow-up

Figure 3:

Kaplan–Meier curve, limited at 36 months of follow-up. Upper and lower confidence interval limits have been included, as has the total number of patients at risk.

Figure 4:

Computed tomography angiography of stent graft thrombosis. (A) Axial view and (B) Sagittal view.

Comparison between 1999 and 2006, 2007 and 2013 and 2014 and 2020 cohorts

A total of 23 patients were admitted with BTAI during 1999 and 2006, 7 patients between 2007 and 2013 and 16 patients between 2014 and 2020. The 5 patients treated with open surgery and the case treated conservatively were all performed before 2006, with the last open intervention for BTAI being performed in 2003. Notable differences between the tertiles included the number of interventions performed in an ‘urgent, non-deferrable setting’, rising from 43.5% before 2006 to 86% between 2007 and 2013 and to 100% from 2014 onwards; with a concomitant shortening of the time between patient-admission and the aortic intervention. There was also a tendency for increased coverage of the LSA and zone 2 sealing, despite the similar distribution of injury location and increased use of percutaneous femoral access. Finally, in-hospital mortality rates and all-cause complication rates appear to be lower, especially from 2014 onwards, with a lower aortic-related complication rate (Table 4).

Table 4:

Comparison between 1999 and 2006, 2007 and 2013 and 2014 and 2020

	1999–2006 (n = 23)	2007–2013 (n = 7)	2014–2020 (n = 16)	Global (n = 46)
Age, median (IQR)	40 (29 to 47)	56 (24 to 64)	57 (25 to 63)	42 (27 to 59)
Males (%), n (%)	21 (91.3)	5 (71.4)	7 (43.8)	33 (71.7)
Hypertension (%), n (%)	8 (34.7)	5 (71.4)	4 (25)	17 (37)
Dislipidaemia (%), n (%)	3 (13)	2 (28.6)	2 (12.5)	7 (15.2)
PAD (%), n (%)	2 (8.7)	1 (14.2)	1 (6.3)	4 (8.7)
Psychiatric disorders (%), n (%)	0 (0)	2 (28.6)	3 (18.8)	5 (10.9)
CAD (%), n (%)	2 (8.7)	2 (28.6)	1 (6.3)	5 (10.9)
Renal failure (%), n (%)	0 (0)	0 (0)	1 (6.3)	1 (3.6)
ISS, median (IQR)	43 (42 to 46)	50 (27 to 55)	43 (41 to 50)	43 (41 to 50)
Urgent procedures (%), n (%)	10 (43.5)	6 (85.7)	16 (100)	(69.6)
Time between admission and intervention (h), median (IQR), n (%)	24 (2 to 120)	1.8 (1.8 to 80)	2 (1 to 3.9)	2.7 (1.7 to 27.7)
TEVAR (%), n (%)	17 (73.9)	7 (100)	16 (100)	40 (87)
Coverage of the LSA (%)—only TEVAR, n (%)	6 (35.2)	4 (57.1)	12 (75)	22 (55)
Percutaneous femoral access (%) (only TEVAR), n (%)	1 (5.8)	2 (28.5)	7 (43.6)	10 (25)
Proximal oversizing, median (IQR)	5 (1 to 17)	5 (4 to 16)	10 (7 to 18)	8 (5 to 18)
Distal oversizing, median (IQR)	22 (−8 to 25)	36 (14 to 40)	32 (19 to 44)	29 (17 to 44)
In-hospital mortality (%), n (%)	6 (26.1)	2 (28.6)	1 (6.3)	9 (20)
All-cause in-hospital complications, n (%)	23 (100)	7 (100)	13 (81.3)	43 (93.5)
Aortic-related complications, n (%)	7 (30.5)	3 (42.9)	0 (0)	10 (21.7)

CAD: coronary artery disease; ISS: injury severity score; IQR: interquartile range; LSA: left subclavian artery; PAD: peripheral arterial disease; TEVAR: thoracic endovascular repair.

DISCUSSION

The 2011 SVS and 2017 European Society for Vascular Surgery guidelines recommend TEVAR over open repair for the treatment of patients with BTAI due to its lower 30-day mortality and spinal cord ischaemia rates, making it the first treatment option for these patients [1, 3]. The results of this study support this recommendation, with short-term all-cause mortality rates of 19.5% (17.5% TEVAR vs 40% open surgery) and aortic-related mortality rates of 6.5% (5% TEVAR vs 20% open repair). The 2 cases of technical failure and conversion to open repair after TEVAR date back to 1999, when TEVAR was still a relatively novel procedure. Although little doubt exists over the benefits of TEVAR, controversies regarding the optimal timing of the aortic repair, appropriate endograft selection and the long-term durability of these procedures remain unsolved.

The 2011 SVS guidelines suggest that delayed TEVAR (repair after the first 24 h of care) should be exclusively reserved for patients with prohibitive operative risk, in contrast to previously published studies that report up to 65% mortality reduction rates in patients undergoing delayed repair [1, 9, 10]. A retrospective study from the American National Trauma Data Bank including 507 patients undergoing TEVAR for BTAI (378 early TEVAR, 129 late TEVAR) found a mortality rate of 11.9% in the early group vs 5.4% in the delayed group (odds ratio 2.36; 95% confidence interval of 1.03–5.35), maintained after adjusting for age, ISS and admission pathology. However, the grade of the aortic injury was not available and could be an important confounder of these results [11]. In our centre, in accordance to the 2011 SVS guidelines, a trend for more prompt aortic intervention was observed during the last 15 years.

Traditional endograft designs were developed for the treatment of aneurysmal pathology, typically affecting larger sized aortas, although in recent years some thoracic grafts have received FDA approval for their specific use in BTAI (e.g. C-TAG, which received FDA approval for the treatment of BTAI in 2012). BTAIs are more frequent in younger patients, who have a lower prevalence of vascular pathology, and who are frequently in a state of haemodynamic shock. This makes selection of the optimal endograft size a difficult process, in which excessive oversizing is sometimes unavoidable. The literature suggests adequate oversizing for BTAI treatment to be <10%, with the deployment of suboptimal devices or incorrectly oversized devices causing distal migration, infolding and/or thrombosis [12]. However, adequate size selection is not always possible. A study evaluating oversizing in patients undergoing TEVAR for BTAI reported a mean proximal oversizing of 19% and a mean distal oversizing of 27%, while other larger multicentre studies have reported varying percentages between 0% and 60% [5, 13]. The smallest available thoracic device for BTAI is currently the Valiant Navion™ Thoracic Stent Graft System (Medtronic PLC, Dublin, Ireland), which has a proximal diameter of 20 mm [14]. However, this is a relatively new stent graft, with the publication of its 30-day pivotal results in 2019 and no available long-term data [15, 16]. A few years ago, the TRANSFIX study evaluated 50 patients undergoing TEVAR with the Zenith Alpha (Cook Medical, Bloomington, Inc.) for the treatment of BTAI. After a mean follow-up of 21 months, incidental finding of thrombus formation within the stent graft, most commonly at the distal part of the device, was observed in 15 patients (30%), leading to the removal of BTAI as an indication from its IFUs. This, alongside the recent urgent recall of all 18–20–22 mm Zenith Alpha devices from the market in 2017, sheds even further doubt of optimal device selection in these patients [17, 18].

In our study, the specific endograft device was available in 35 of the 40 patients who underwent TEVAR. In 5 patients the endoprosthesis used was not recorded. Of these, 2 patients were intervened in 1999 and died perioperatively (1 from stent fracture, with posterior aortic occlusion, requiring an urgent thoracotomy, who died 11 days postoperatively; and the other patient died from sigmoid perforation and colonic ischaemia on the 15th postoperative day). A third patient, treated in the year 2000, presented with the partial occlusion of the graft and required conversion to open surgery. In this case, the diameter and length of the stent graft were documented (22 × 22 × 150), but unfortunately, the actual type the stent graft was not recorded. A fourth patient, intervened in 2007, suffered from on-the-table intraoperative death from massive bleeding. Although the stent graft was implanted and the BTAI was excluded, the patient required subsequent urgent thoracotomy, drainage of pericardiac tamponade and CPR and died on the operative table. A last patient in whom the type of endografts used was not reported was treated in year 2003 and presented no early graft-related complications. Albeit these complications, and the consideration than not all endografts used has specific BTAI approval, it is possible that they were, nonetheless, the best available treatment strategy. Regarding oversizing, the median proximal oversizing was 5% (IQR 1–17) between 1999 and 2006, 5% (IQR, 4–16) between 2007 and 2013 and 10% (IQR 7–18) between 2014 and 2020; while the median distal oversizing was 22% (IQR −8 to 25) between 1999 and 2006, 36% (IQR, 14–40) between 2007 and 2013 and 29% (IQR 17–44) between 2014 and 2020. Only 1 thrombotic complication was recorded, occurring in a patient who underwent TEVAR with a 26 mm × 26 mm straight Cook Zenith TX2^® (Cook Medical, Bloomington, Inc.) device in zone 2 and primary coverage of the subclavian artery. His native aorta was 19 mm in diameter (37% proximal and 44% distal oversizing). After 1 year of follow-up, and following cessation of aspirin therapy without consultation, he was admitted with 99% stent graft thrombotic occlusion and he successfully underwent immediate redo TEVAR with a 24-mm Cook Zenith TX^® (Fig. 4). He has not presented any further complications.

Long-term follow-up in these patients is difficult and in our study, ∼50% of patients were lost during the first 12 months. They frequently require multiple surgical interventions, passing through the hands of multiple specialists. Many patients are also young and fit and are more likely to travel, with a subsequent decrease in probability of attending mandatory annual or bi-annual follow-up. Finally, a substantial percentage of patients have psychiatric disorders, in which BTAI is secondary to suicide attempts. Arranging serial CTA or magnetic resonance and follow-up consultations with multiples specialists can prove incredibly difficult in this subgroup of BTAI patients. In our study, the most significant loss of follow-up was observed within the first year. This could be secondary to the transfer of patients to trauma centres after management of the BTAI or discharge without specific vascular surgery recommendations. This loss of follow-up is one of the key points that we will try to improve in future patients, giving stricter recommendations and emphasizing the need for follow-up vascular surgery consultations. Fortunately, the rate of long-term procedural-related complications appears to be low. Piffaretti et al. [19] report on 35 patients with a mean follow-up of 66 months, 13 (30.2%) of whom had >5 years follow-up, and only 1 endograft-related complication. Similarly, in this study, 7 patients had over 10-year follow-up (15.2%), with only 1 endograft-related complication. Urgnani et al. [20] report on 20 patients with a mean follow-up of 58 months; with 2 cases of stent fracture. Ziza et al. report on 223 TEVARs (26% in BTAI patients) with a mean follow-up of 43.4 months, with an estimated complication free survival rate at 12, 34, 60 and 120 months of 73 ± 3%, 64 ± 4%, 62 ± 4% and 57 ± 5%, respectively; with the best survival rate in the BTAI group, in which no late death was observed and the long-term mortality was 1.7% [21].

Limitations

The two-decade long and retrospective nature of this study made the data collection process challenging. Furthermore, clinical practices changed during these 2 years, potentially affecting patient outcomes. Unfortunately, in this study, there were >30 variables with large numbers of missing values could not be properly assessed, including some baseline comorbidities, haemodynamic variables, pre-operative and operative data and long-term follow-up. In addition, it would have been of great interest to compare results by device generation and specific changes in management strategies. However, given the retrospective nature of this study, this was not possible. To mitigate the loss of follow-up, patients were contacted telephonically and a new vascular consultation with an imagining study was recommended. However, not all patients agreed to this. A proportion of patients had gone aortic imaging in other centres and although they shared the radiological report and findings, a few of the long-term imagining tests could not be independently assessed. Therefore, the loss of follow-up is a significant issue to be taken into account, particularly difficult in these patients.

CONCLUSION

Even with all the described shortcomings, in our experience TEVAR for BTAI proved to be feasible and effective, with few complications and stable aortic reconstruction at mid-term follow-up. With the current technical expertise and wide availability of a variety of devices, it should be pursued as a first-line therapy in these challenging scenarios.

Conflict of interest: none declared.

ABBREVIATIONS

BTAIs

Blunt traumatic aortic injuries

CTA

Computed tomography angiography

IQR

Interquartile ranges

OS

Open surgery

SVS

Society for Vascular Surgery

TEVAR

Thoracic endovascular repair

Author contributions

Carlota Fernandez Prendes: Conceptualization; Data curation; Formal analysis; Methodology; Resources; Writing—original draft. Jan Stana: Resources; Writing—review & editing. Karina Domingos Schneidwind: Conceptualization; Data curation; Methodology; Resources. Barbara Rantner: Resources; Writing—review & editing. Nikolaos Konstantinou: Data curation; Investigation; Resources. Jan Bruder: Data curation; Resources. Christian Kammerlander: Data curation; Resources. Ramin Banafsche: Supervision; Writing—review & editing. Nikolaos Tsilimparis: Conceptualization; Data curation; Resources; Supervision; Writing—review & editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Gabriele Piffaretti and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.