Aortic arch replacement with descending thoracic aorta stenting for chronic dissection: early outcomes

Abstract

Background

The frozen elephant trunk (FET) is used to treat patients with arch and descending aortic diseases. This study evaluates the early clinical outcomes and postoperative aortic remodelling in patients with chronically dissected aorta who were treated with the FET procedure.

Methods

A retrospective analysis of prospectively collected data from 23 patients who underwent aortic arch replacement with descending thoracic aorta stenting using a branched hybrid FET prosthesis at Southampton General Hospital between 2017 and 2022. All patients received Thoraflex™ prostheses. Patient characteristics, surgical outcomes, follow-up data, and postoperative aortic remodelling, including false lumen thrombosis and changes in aortic dimensions, were assessed using CT imaging.

Results

The mean age was 63.8 ± 12.1 years, 17(74%) were men and two patients (9%) had a connective tissue disorder. Ten (43%) had re-do surgery and one patient had concomitant thoracic endovascular aneurysm repair. There was no in-hospital or 30-day mortality. One patient (4%) suffered a stroke and one (4%) developed paraplegia. False lumen thrombosis was achieved in all patients at the level of the stent. The mean follow-up was 3.5 ± 1.8 years. Aortic size stabilized in 91% of patients, with a reduction in false lumen diameter and an increase in true lumen diameter in all patients. Four patients (17%) required further aortic interventions.

Conclusion

The FET technique demonstrates favourable early clinical outcomes and satisfactory aortic remodelling in patients with chronic aortic dissection. The procedure effectively promotes false lumen thrombosis, stabilizes the aortic size and facilitates downstream procedures, if required, therefore making it a viable treatment option.

Trial registration

The study was conducted as a service evaluation, for which local institutional approval was obtained (Service Evaluation Number SEV/0649 obtained on 06 February 2024).

Background

The frozen elephant trunk (FET) technique was introduced over 20 years ago as a treatment for aortic arch and proximal descending thoracic aortic aneurysms [1]. Over the last decade, the prosthesis has evolved, with the development and use of the branched FET [2, 3]. Initially employed for aneurysm repair, its indications have expanded to include both acute and chronic aortic dissection pathology [4]. However, data on aortic remodelling following the implantation of branched FET in chronic dissection remain limited.

Chronic aortic dissection is defined by the International Registry of Acute Aortic Dissection as a dissection persisting beyond one month from the acute presentation [5]. In patients treated medically, or those with residual disease following surgical intervention, the weakened and dissected aorta may degenerate, leading to aneurysm formation. If left untreated, this can result in aortic rupture [6, 7].

Southampton General Hospital, a high-volume tertiary cardiac surgery centre in the United Kingdom, began implanting branched FET prostheses in 2015. By 2017, we extended the use of this prosthesis to treat patients with chronic aortic dissection.

This study aims to elucidate the clinical outcomes of patients with chronic aortic dissection treated with branched FET and to evaluate short term outcomes and the extent of distal aortic remodelling following implantation.

Methods

We conducted a retrospective analysis of prospectively collected data from all patients who underwent aortic arch replacement with stenting of the descending aorta using a hybrid branched FET prosthesis for chronic aortic dissection at our institution between June 1, 2017, and June 1, 2022. All patients received Thoraflex™ (Terumo Corporation, Tokyo, Japan) prostheses, a multi-branch hybrid device composed of two parts: a vascular graft for the aortic arch, which includes three side branches for the head and neck vessels, and a stented segment positioned in the descending aorta.

Indication for surgery

The most common indication for surgery in these patients with chronic aortic dissection was the development of an aortic arch aneurysm involving the distal arch in 17 patients (74%). In 12 patients (52%), the aneurysm extended into the descending thoracic aorta. Proximal descending aortic aneurysm formation following chronic dissection was the indication in 6 patients (26%). This was planned as a first-stage procedure to trigger proximal aneurysmal regression, with or without subsequent downstream TEVAR or open surgery.

Surgical technique

The operations were performed on cardiopulmonary bypass, with arterial cannulation via the right subclavian artery in all but one case, where the brachiocephalic trunk was used. The operations were conducted under hypothermic circulatory arrest with anterior cerebral perfusion.

The prostheses were anastomosed to either zone 2 or zone 1 of the aortic arch [8]. The stent was deployed antegradely into the true lumen of the descending thoracic aorta (DTA), and successful deployment was assessed using intraoperative angioscopy with an Ambu^® aScope™ (Ambu A/S, Ballerup, Denmark).

After completing the distal cuff anastomosis of the prosthesis and the left subclavian branch, distal body perfusion on bypass was initiated through the prosthesis’ perfusion limb.

Prosthesis sizing

The prosthesis size was selected by adding 10–15% to the average length of the true lumen’s long and short crescent in the initial cases. In later cases, prosthesis size was estimated by adding 15–20% to the true lumen diameter, calculated from its perimeter measurement.

The prosthesis length was determined by measuring the distance along the centreline of the aortic curvature, from the intended implantation zone to ensure complete coverage of any primary tear or aneurysm, and the shortest stent satisfying these criteria was selected.

Patient follow-up

All patients underwent a postoperative CT aortogram before discharge. Follow-up visits occurred 6–8 weeks post-discharge, with additional surveillance at 6 months and annually thereafter. At each surveillance visit, a CT aortogram was performed. For patients with progressive distal aortic dilatation, follow-up intervals were shortened, depending on the aorta’s diameter and the rate of enlargement.

Data collection

Patient characteristics were obtained from the “Hospital Integrated Clinical Support System” database, which monitors risk-adjusted outcomes for our unit. Data is entered prospectively and audited monthly. Mortality data and information on any subsequent interventions were extracted from this system, along with data from the Aortic Surveillance Clinic.

Aortic imaging analysis

All patients underwent CT aortogram preoperatively, postoperatively, and during follow-up visits. Thrombosis of the false lumen, along with the diameters of the aorta, true lumen, and false lumen, was assessed at three levels: the proximal stented descending aorta (segment A), the distal descending thoracic aorta to the celiac axis (segment B), and the abdominal aorta from the celiac axis to the iliac bifurcation (segment C), as previously described [9].

Image analysis was performed using Sectra Workstation IDS7 Version 24.2 (Sectra AB, Linköping, Sweden). The maximal diameter of each aortic segment was measured in the axial (orthogonal) plane after aligning the long axis of the aorta in the coronal and sagittal planes [10].

All image analysis was performed by one person (HK) to reduce inter-observer bias. Segmental thrombosis was defined as thrombosis involving 80% or more of the segment. Any descending aortic segment reaching or exceeding 45 mm was considered aneurysmal. Changes in lumen diameter were expressed as both absolute values (in millimetres) and percentages relative to preoperative measurements.

Positive aortic remodelling was defined as a statistically significant (p value < 0.05) increase in true lumen diameter accompanied by a decrease in the false lumen and/or overall aortic diameter for each of the segments A, B and C. Aortic stabilization was defined as the absence of significant change in true lumen, false lumen, and overall aortic (< 3 mm) diameters following surgery.

Statistical methods

The descriptive results for continuous variables are expressed as Mean ± one Standard Deviation. The categorical variables are expressed as the number of cases and percentages. The continuous variables with normal distribution were compared with student’s t-test. Survival and freedom from death and re-intervention curves were generated constructing Kaplan-Meier and actuarial survival curves. Cumulative incidence curve of the re-intervention was constructed with death and re-intervention as competing risks. The statistical analysis was performed with SPSS version 26.

Results

Patient characteristics

Between June 1, 2017, and June 1, 2022, a total of 23 patients underwent aortic arch replacement with descending thoracic aorta stenting using a branched hybrid stent-graft as a frozen elephant trunk for chronic aortic dissection at Southampton General Hospital. The patient characteristics are summarized in Table 1.

Table 1.

Patients’ characteristics

Patients’ Number			23
Mean Age, [range] (Years)			63.8 ± 12.1 [33 to 79]
Female Gender			6 (26%)
Hypertension			19 (83%)
Diabetes Mellitus			2 (9%)
Hypercholesterolemia			4 (17%)
COPD / Asthma			2 (9%)
Previous Cardiac Surgery			10 (43%)
Pathology	Previous type A (DeBakey I) dissection Previous type B dissection Previous type A and B dissection		13 (56%) 8 (35%) 2 (9%)
NYHA class III or IV			2 (9%)
Mean Logistic Euroscore			16.1 ± 9.2
Median time to surgery (months)			48
Location of primary tear	Ascending aorta Aortic arch Segment A of the descending aorta Segment B of the descending aorta No tear identified preoperatively		7 (30%) 6 (26%) 4 (17%) 3 (13%) 3 (13%)
Distal re-entry tear	Present distal re-entry tear No distal re-entry tear identified		12 (52%) 11 (48%)

The characteristics of the study patients with aortic arch replacement and descending thoracic aorta stenting (frozen elephant trunk technique) for chronic dissection

COPD: chronic obstructive pulmonary disease, NYHA: New York Heart Association

Values: n (%) or mean ± standard deviation

Two patients (9%) were diagnosed with connective tissue disorders, specifically Loeys-Dietz syndrome. An additional four patients (17%) had a positive family history of aortic disease, but current genetic screening did not reveal any known mutations.

Operative and postoperative course

Operative and postoperative results are summarized in Table 2. There were no in-hospital deaths, and the rates of neurological and spinal complications were low. No cases of visceral ischemia occurred during the perioperative period.

Table 2.

The operative and the postoperative results

Median prosthetic stent size (mm)			30
Stent length (mm)		100 mm 150 mm	4 (17%) 19 (83%)
Cardiopulmonary bypass time (min)			268 ± 68
Cardiac ischemic time (min)			123 ± 47
Hypothermic circulatory arrest time (min)			66 ± 18
Hypothermic circulatory arrest temperature (°C)			20.1 ± 1.4
Concomitant aortic valve replacement			2 (9%)
Concomitant aortic root replacement			3 (13%)
Concomitant coronary artery bypass grafting			2 (9%)
Concomitant TEVAR			1 (4%)
30-Day / in-hospital mortality			0
Re-exploration for bleeding			1 (4%)
Re-intubation			2 (9%)
Intensive Therapy Unit stay (days)			10.1 ± 11.7
Permanent neurological deficit			1 (4%)
Paraplegia			1 (4%)
Renal failure requiring temporary hemofiltration			3 (13%)
New atrial fibrillation			8 (35%)
Postoperative in hospital stay (days)			27.3 ± 22.1

The operative and the postoperative data of the patients with aortic arch replacement and descending thoracic aorta stenting (frozen elephant trunk technique) for chronic dissection

TEVAR: thoracic endovascular aortic repair

Values: n (%) or mean ± standard deviation

Follow-up

The mean follow-up period was 3.5 ± 1.8 years (range: 5 months to 6.6 years). During follow-up, three patients (13%) died at 5 months, 4.5 years, and 5 years postoperatively. The causes of death were aspiration pneumonia in a patient with a prior stroke, a subdural hemorrhage following a fall, and an infection in an endovascular aortic repair (EVAR) prosthesis in the abdominal aorta (the patient, aged 83, received palliative care).

Kaplan-Meier survival curves, cumulative incidence of distal aortic intervention, and freedom from death and intervention curves are shown in Fig. 1. The estimated 6-year survival rate was 69%, and the cumulative risk of aortic intervention was 31%. Freedom from death or re-intervention at 6 years was estimated at 60%. None of the patients experienced stent-related complications like thrombosis, stent-induced new entry tears, or stent migration.

Fig. 1.

(A) Kaplan-Meier survival curve and cumulative incidence of distal aortic intervention with death as a competing factor. (B) Freedom from death and distal intervention curve

Postoperative interventions

Four patients (17%) required further aortic interventions distal to the stent. Three patients underwent TEVAR procedures at 1, 1.5, and 2.5 years postoperatively. None of these patients developed type 1a or type 3 endoleaks. One patient had a thoracoabdominal aortic replacement using left heart bypass, 2 years after the initial operation.

The indications for further intervention were the failure of the descending thoracic aneurysm to regress in two patients, who subsequently received the preplanned TEVAR as a second-stage procedure. Two additional patients exhibited progressive dilatation of the distal descending thoracic and abdominal aortas, necessitating further intervention (one TEVAR and one thoracoabdominal aortic replacement). There was no mortality associated with the distal re-interventions.

False lumen thrombosis

The rates of false lumen thrombosis before and after surgery, as well as during follow-up, are shown in Fig. 2. Preoperatively, only 9 patients (39%) had thrombosis of the false lumen in segment A (the proximal descending thoracic aorta above the eighth thoracic vertebra). After surgery, complete false lumen thrombosis was achieved in all patients in this segment (around the stent), and this was maintained throughout the follow-up period.

Fig. 2.

The rate of false lumen thrombosis on the preoperative and the surveillance CT scans. The assessment was performed at three levels: DTA around the stent (segment A), the DTA distal to the stent (segment B), and the abdominal aorta beyond the coeliac axis (segment C)

In the descending thoracic aorta distal to the stent (segment B), the false lumen was thrombosed in 8 patients (35%) preoperatively, increasing to 11 patients (48%) immediately after surgery. During follow-up, an additional 5 patients experienced false lumen thrombosis, resulting in 16 patients (70%) with thrombosis in this segment by the last follow-up, without requiring further intervention.

In the abdominal aorta (segment C), the false lumen was thrombosed in 7 patients (30%) preoperatively, increasing to 10 patients (43%) immediately after surgery. During follow-up, two additional patients experienced false lumen thrombosis, resulting in 12 patients (52%) with thrombosis in this segment by the last follow-up.

Aortic diameter and remodelling

Changes in aortic diameter, as well as the diameters of the true and false lumens, are depicted in Fig. 3. These data demonstrate a stable total aortic diameter in the majority of patients throughout follow-up. There was no statistically significant change in the mean diameter measurements during the follow-up in all three segments. The false lumen diameter decreased over time, with an average reduction of 25% in segment A and 17% in segment B, two years post-surgery compared to preoperative measurements (from 43 ± 19 mm to 33 ± 15 mm for segment A and from 26 ± 19 mm to 22 ± 14 mm for segment B, p < 0.05 for both). Additionally, the true lumen diameter increased over time, with an 80% increase in segment A and a 53% increase in segment B, two years after surgery, compared to preoperative measurements (from 13 ± 5 mm to 24 ± 5 mm for segment A and from 15 ± 5 mm to 21 ± 5 mm for segment B, p < 0.05 for both). No patient suffered distal aortic rupture in the follow-up period. Changes in the abdominal aorta (segment C) were less pronounced. The aortic diameter remained stable, and although there was a trend toward reduction of the false lumen and expansion of the true lumen, these changes did not reach statistical significance.

Fig. 3.

Changes in the true lumen diameter (I), the false lumen diameter (II), and the aortic diameter (III). The changes were expressed in mm (left) and as a percentage (right) when compared with the preoperative diameter. The assessment was performed at three levels: DTA around the stent (segment A), the DTA distal to the stent (segment B), and the abdominal aorta beyond the coeliac axis (segment C)

Discussion

The main finding of this study is that the early clinical outcomes of aortic arch replacement with stenting of the descending aorta (using the FET technique with a hybrid prosthesis) are favourable. Complete false lumen thrombosis occurred around the stent in all patients. Additionally, 70% of patients achieved full thrombosis of the false lumen in the descending thoracic aorta, distal to the stent, without any further intervention. Aortic size stabilized in 91% of patients, with a decrease in false lumen diameter and an increase in true lumen size. This demonstrates satisfactory postoperative aortic remodelling at the level of the stent and in the descending thoracic aorta distal to the stent (segments A and B). Thus, it successfully achieved the treatment goal of decompressing the false lumen, halting its expansion, and inducing regression of the false lumen aneurysm around the stent, and in many cases distal to, the stent.

The treatment of chronic aortic dissection with FET using various prostheses has been evaluated in previous studies, although outcomes and postoperative aortic remodelling data remain limited [9, 11–14]. In our cohort, there was no in-hospital or 30-day mortality, and the rate of neurological and spinal cord complications was on the lower end of the spectrum compared to other published reports on FET [9, 11, 15, 16].

High percentage of false lumen thrombosis was reported in other studies [9, 11], though it occurred several months after surgery. In our cohort, all patients had false lumen thrombosis in the pre-discharge scans, an encouraging finding given that false lumen thrombosis is associated with improved outcomes, even if it is not curative [14, 17–20].

In the thoracic aorta (segments A and B), false lumen thrombosis was associated with stabilization or reduction of both the aortic and false lumen diameters in most patients (70%). However, a subgroup of patients (two patients) continued to experience progressive dilatation of the descending thoracic aorta distal to the stent. One of these patients had full thrombosis of the false lumen, nevertheless his distal thoracic and upper abdominal aorta continued to dilate and required a second stage intervention with a TEVAR on his DTA, then a third stage intervention with an EVAR on his upper abdominal aorta. The second patient who was known to have connective tissue disorder (Loeys-Dietz syndrome) and had several residual distal aortic fenestrations (about 3–4 cm distal to the stent and several smaller fenestrations around the renal arteries. The patient underwent a successful thoraco-abdominal aneurysm surgical repair. For these patients, the FET provided an optimal landing zone for subsequent TEVAR or anastomotic sites for thoracoabdominal surgery, which could be performed using left heart bypass rather than full cardiopulmonary bypass. Furthermore, the FET facilitated enlargement of the true lumen, simplifying any necessary follow-up interventions such as surgery or TEVAR.

Although no distal stent graft-induced new entry was observed in our cohort, these events have been documented in the literature [21] and some patients required subsequent interventions in the downstream aorta. Thus, all patients undergoing such procedures must receive diligent follow-up care in an aortic surveillance clinic, underscoring the importance of ongoing monitoring. We propose that a multidisciplinary approach, involving aortic MDTs, specialized aortovascular surgeons performing operations, and dedicated surveillance clinics, plays a pivotal role in achieving reduced long-term aortic mortality and morbidity.

Conclusion

In conclusion, our study demonstrates favourable early clinical outcomes for patients with chronic aortic dissection, both postoperatively and in the short-term follow-up. Aortic remodelling, particularly in the descending thoracic aorta, was positive in the majority of patients.

Limitations

This study is limited by small number of patients, its retrospective nature and single-centre design. Multicentre studies and registry analyses are needed to determine if these findings are consistent across other institutions.

Abbreviations

COPD

Chronic obstructive pulmonary disease

CT scan

Computerized Tomography scan

DTA

Descending thoracic aorta

EVAR

Endovascular aortic repair

FET

Frozen elephant trunk

ITU

Intensive therapy unit

NYHA

New York Heart Association

TEVAR

Thoracic endovascular aortic repair

Author contributions

Hassan Kattach (Data curation; Formal analysis; Methodology; Project administration; Software; Visualization; Writing of the original draft), Amit Modi (Conceptualization; Investigation; Methodology; Supervision; Review and editing), Mohammed Hadi (Review and editing), Iain Wilson (Review and editing), Robert Allison (Review and editing), Benjamin Patterson (Review and editing), Ian Nordon (Review and editing), Geoff M. Tsang (Conceptualization; Methodology; Project administration; Supervision; Review and editing).

Funding

The study was funded by the Department of Cardiothoracic Surgery, Southampton General Hospital. No commercial or external funding supported this study.

Data availability

The data that support the findings of this study are available from the University Hospital Southampton NHS foundation Trust, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of University Hospital Southampton NHS foundation Trust.

Declarations

Consent for publication

The manuscript does not contain any individual person’s data.

Competing interests

The authors declare no competing interests.

Ethical approval

In accordance with the National Research Ethics Service, this retrospective study—using data collected during routine patient care—did not require research ethics committee approval. The study was conducted as a service evaluation, for which the University Hospital Southampton NHS Foundation Trust’s approval was obtained (Service Evaluation Number SEV/0649, granted on 06 February 2024).