Early and Midterm Outcome of TEVAR for Thoracic Aortic Pseudoaneurysm

Backgrounds: The prognosis of thoracic aortic pseudoaneurysm (TAP) after thoracic endovascular aortic repair (TEVAR) remains unclear. This study investigates the early and midterm clinical outcome as well as relevant risk factors of TAP patients following TEVAR therapy. Methods: From July 2010 to July 2020, 37 eligible TAP patients who underwent TEVAR were selected into our research. We retrospectively explored their baseline, perioperative and follow-up data. Fisher exact test and Kaplan-Meier method were applied for comparing difference between groups. Risk factors of late survival were discerned using Cox regression analysis. Results: There were 29 men and 12 women, with the mean age as 59.5±13.0 years (range, 30-82). The mean follow-up time was 30.7±28.3 months (range, 1-89). For early result, early mortality ( ≦ 30days) happened in 3(8.1%) zone 3 TAP patients versus 0 in zone 4 (p= 0.028); acute arterial embolism of lower extremity and type II endoleak respectively occurred in 1(2.7%) case. For midterm result, survival at 3 months, 1 year and 5 years was 88.8±5.3%, 75.9±7.5% and 68.3±9.9%, which showed signicant difference between zone 2/3 versus zone 4 group (56.3±14.8% versus 72.9±13.2%, p= 0.013) and emergent versus elective TEVAR groups (0.0±0.0% versus 80.1±8.0%, p= 0.049). On multivariate Cox regression, lesions at zone 2/3 (HR 4.605, 95%CI 1.095-19.359), concomitant cardiac disease (HR 4.932, 95%CI 1.086-22.403) and emergent TEVAR (HR 4.196, 95%CI 1.042-16.891) were signicant independent risk factors for worse late clinical outcome. Conclusions:

retrospectively presents 37 patients diagnosed with TAP and treated by TEVAR therapy. We analyzed their clinical features, early and midterm results as well as prognostic factors.

Materials And Methods
This retrospective study is approved by the institutional research ethics committee of our hospital and waiver of informed consent form of individuals are granted.

Thoracic Endovascular Aortic Repair
Treatment with antihypertensive drugs and beta blockers to strictly maintain the systolic pressure at about 100 mm Hg and the heart rate at 60 to 70 beats/min was initiated on admission. Procedures were performed in the hybrid unit under general or local anesthesia. The right common femoral artery (CFA) was exposed through incision, approximately 5 cm, and then guidewire and catheters were inserted into CFA using puncture technique until reached appropriate sites. The left CFA was available when necessitating bilateral intervention. Digital subtraction angiography (DSA) of the thoracic aorta was performed to compare the lesion's dynamic characteristics with those seen on CTA. After intravenous administration of heparin sodium (0.5ml/kg), the stent graft with the diameter 10-20% oversized was advanced and prepositioned into the proper location. DSA was performed again to evaluate con guration of stent, endoleak and patency of branch vessels.
Physician-modi ed fenestration (PMF) technique was selected to preserve LSA when the distance from TAP to LSA was < 15 mm, the neck length of the TAP was < 15 mm and the diameter was > 5.5 cm or rapid aortic growth had occurred (> 1 cm/year). The detailed process of PMF technique was same as before (15). Generally, the outer sheath of the stent was moved 3-4cm backward to expose the stent, a scalpel and scissor were used to remove the member of pre-fenestration area, and then using a belt to constrain the stent to return the outer sheath to the original position. Adjust the stent-graft to t LSA ori ce under radioscopy. Balloon expandable stent for LSA was deployed when necessary. On the other hand, LSA was directly covered when lack of enough landing zone if the patient had dominant right vertebral artery and no history of left internal mammary artery graft or the need of left arm access dialysis.
For patients meeting the diagnostic criteria for infection, adjunctive procedure also includes the inserting of a 5F sheath into the contralateral groin to enable the catheter to be preplaced in the aneurysmal sac for perfusion of anti-TB drugs or sensitive antibiotics after deploying the stent graft(16). TEVAR was performed under local anesthesia when including perfusing drugs to the sac, for early recognition of related complications. Additionally, in our experience, for traumatic patients with pleural effusion, perioperative antibiotic was necessary for preventing postoperative infection.

Follow-up
Patients were scheduled for CTA and physical examination 1 months, 3 months, 6 months and 12 months after discharge, and annually afterwards to evaluate general condition, patency and position of vascular graft as well as early and late complications. Telephone interview was conducted semiannually.

Statistics
Statistics analysis was performed with SPSS 18.0 (SPSS Inc, Chicago, IL) and R software (R Core Team, Vienna, Austria). Fisher exact test and Kaplan-Meier method were applied for comparing difference between groups. Potential risk factors were incorporated into multivariate Cox regression analysis based on univariate Cox regression results (P < 0.1). Difference between each group would be considered signi cant if P value < 0.05.
Early mortality happened in three zone 3 TAP patients versus 0 in zone 4 (p = .028), and 2 of them received emergent TEVAR (Table 2). One case died of acute arterial embolism of the lower extremity 7 days postoperatively, whose preoperative CTA showed extensive aortic calci cation; one case with preoperative hydropericardium, hydrothorax and compressed main pulmonary artery died of pericardial tamponaed and compressed main pulmonary artery the day after TEVAR, whose etiology was atherosclerosis. The death of third case happened after discharge who had stable physical condition when discharged. Type II endoleak was observed in one zone 4 TEVAR case, which was left untreated and expected to thrombosed spontaneously. For other patients having received TEVAR within 30 days, there was no occurrence of ischemia of left upper extremity, stroke, paralysis, acute renal insu ciency, bowel ischemia or puncture complications, and the 30-days survival of them was 88.8 ± 5.3% (Table 2).

Midterm outcomes
The mean follow up was 30.7 ± 28.3 months (range, 1-89), and the overall survival at 3 months, 6 months, 1 year, 3 years and 5 years was 88.8 ± 5.3%, 85.8 ± 5.9%, 75.9 ± 7.5%, 75.9 ± 7.5% and 68.3 ± 9.9%, and no death occurred in zone 4 group within 6 months, which showed signi cant difference between zone 2/3 and zone 4 groups (p = .013) ( Table 2, Fig. 2A). Six new adverse event happened during midterm follow up, 3 of which had known cause. One case developed aortoesophageal stula (AEF) due to esophageal foreign body, manifested as "hematochezia for 2 days, 79g/l hemoglobin". He received emergent TEVAR and strict perioperative anti-infection treatment, but subsequently, he suffered from recurrent fever, was treated by exploratory thoracotomy and aortoesophageal repair 5 months later, but nally died of sepsis because of stent infection after 57 months. The other case was also diagnosed with AEF caused by previous stent grafting due to aortic dissection three years ago. His massive hematemesis needed emergent TEVAR, but he died of bacteremia 11 months after TEVAR. The last case had satisfactory follow-up results but died of other disease 9 months later. There was no complication or reintervention in other patients. For special case, the patient following PMF remains alive after 45-months follow up. Representative midterm postoperative (5 years) CTA results of a trauma patient was showed in Fig. 1 (E-H).
The trauma group had the most satisfactory result: no death happened and no complication was observed during 79-months follow up (Fig. 2C)

Discussion
In this paper, we described early and midterm results of TAP patients following TEVAR therapy. As a procedure with less invasiveness, short procedure duration and less bleeding, TEVAR is an ideal alternative of open surgery for aortic pathologies at distal aortic arch and descending aorta, especially for emergent situation with combined traumatic injury, severe infection and aortic rupture. TEVAR should be considered for TAP considering its emergency and complexity.
Atherosclerotic thoracic TAP accounts for the largest proportion (13,35.1%) in our series. Actually, atherosclerotic aortic pseudoaneurysm generally originates from penetrating aortic ulcer (PAU) (2). We usually make the judgment based on patients' advanced age, the gravity of aortic calci cation by CTA and medical history of aortic ulcer or intramural haematoma by CTA results. A study showed that the survival of atherosclerotic TAP after TEVAR has a gradual yearly decreasing trend with 5-years survival lower than 40% due to extensive atherosclerosis of the aortic wall (17). The trend is unobvious in our results with 68-months survival as 67.7 ± 13.4%. In addition, one atherosclerotic TAP case in our series with extensive aortic calci cation died of acute embolism of lower extremity. Therefore, like well-studied periprocedural stroke due to the displacement and embolization of aortic debris after TEVAR, perioperative embolism of lower extremity also deserves more investigation.
Blunt trauma is a mostly reported pathogenesis, and it has been a trend to use TEVAR to treat trauma related aortic pathologies. 7 (18.9%) trauma TAP cases in our series all received TEVAR therapy, and no complication and death was observed during the follow up (Fig. 2C). The result is understandable due to healthier and younger vessel nature based on trauma patients' relative younger ages (52.9 ± 9.6 years, 44 to 68) than atherosclerotic patients (66.9 ± 8.1 years, 56 to 82). According to the literatures, the line between nonoperative management (NOM) and operative treatment for grade III (pseudoaneurysm) blunt traumatic aortic injury (BTAI) remains blurred. Some studies suggested that BTAI with pseudoaneurysm could be managed by NOM instead of instant operative treatment (9, 18). However, there was 10% risk of TAP requiring surgical therapy after NOM (18), and postdischarge follow-up was restricted to < 1 year in most studies (19). Based on our series, we suggest that instant TEVAR for traumatic TAP is reasonable.
Infected TAP is a tough condition with high mortality (20)(21)(22)(23). This microorganisms reach the vessel wall through various ways including direct invasion and spread, feeding vessels, lymphatic vessels and iatrogenic factors. Despite the controversy of treatment with arti cial graft exposed to infected tissue, many studies have shown that endovascular repair is a palliative treatment in acute phase, a temporary bridge to secondary open surgery (20,(24)(25)(26). For patients only undergoing endovascular treatment, lifelong appropriate antibiotic therapy, or regular and full-course anti-tuberculous chemotherapy is recommended (27). 4 (10.8%) infectious cases originated from AEF, which was caused by esophageal foreign bodies in 2 and previous stent grafting in 2. Blood culture was positive for staphylococcus aureus in one case. Moreover, there were 7 (18.9%) tuberculosis TAP patients having received TEVAR in our series, and satisfactory early and midterm outcome had been obtained: patients gained weight during follow up, with 1 month and 51 months survival as 85.7 ± 13.2% and 71.4 ± 17.1%.
For patients with infected TAP, innovative adjunctive procedure in addition to TEVAR was adopted in our center, which might contribute to the satisfactory outcome and has been published recently for the rst time (16). After the deployment of aortic stent sealing the aneurysm sac, directly delivering the antibiotic agent to the aneurysmal sac via pre-placed catheter would ensure that the drugs would permeate into nearby tissues continued and slowly instead of being swept away by blood stream. Moreover, in situ administration might improve drug bioavailability compared to oral or intravenous administration.
Systemic vasculitis is another intractable pathogenesis of TAP, and the outcome of them was unsatisfactory albeit following immunotherapy (Fig. 2C). Some experience on TEVAR for systemic vasculitis has been reported, which is superior to open surgery on preventing immediate complications and aortic anatomical pseudoaneurysm. However, recurrent TAP remains to be a continuing problem because of aortic wall injury or mechanical force at the edge of the stent triggering vascular in ammation (28). It is suggested that adjunctive perioperative and postoperative immunosuppressive treatment is essential (29,30).
Two cases having received TEVAR due to previous aortic dissection developed distal TAP with secondary AEF, and the time to the development of AEF after TEVAR respectively was 39 months and 8 years. A multicenter study showed that the occurrence of AEF after TEVAR was 1.5% and the median time was three months (31). The incidence of distal TAP is a joint result of the unsolved underlying aortic in ammation at the time of TEVAR or mismatched oversize proportion of the stent, plus the mechanical force at the edge of the stent. The mechanical compression and secondary invasion from TAP might be the cause of AEF considering 2-years dysphagia history of our patient.
Symptomatic cases accounts for 83.8% (31/37) percent of overall TAP patients. The most common symptom was chest/back pain or tightness (56.8%, 21/37). It was back pain and distress in another study (17). Actually, hemoptysis, hematemesis, hematochezia and melena are strikingly noticeable as quite common symptoms of respiratory and digestive diseases. They occurred due to aortopulmonary stula (APF) or AEF even with a visible intrabronchial or intraesophageal mass (24,(32)(33)(34)(35)(36)(37). In such situations, endoscopic clamping biopsy poses a risk of massive bleeding. There was no detectable APF in our patients manifested as hemoptysis, albeit in one patient who complained of persistent hemoptysis for even seven years after TEVAR. Under this circumstances, respiratory disorder is more likely to be considered.
Our results suggest that patients with zone 4 TAP have a signi cantly more favorable midterm outcome than zone 3 TAP. A similar study also indicated the existence of better death/survival ratio among patients with middle thoracic aneurysms than among patients with aneurysms situated at proximal aorta (38). On one hand, when the ori ce of LSA that was only partially covered by the stent, local uid dynamics showed remarkable disturbance (39). On the other hand, zone 4 aneurysms, with su cient proximal landing zone, could receive a straightforward endovascular surgery without secondary endovascular intervention and thus avoid relevant complications, having greater chance of satisfactory survival (38). However, a single-center research on TEVAR for thoracic TAP, with 24.0 ± 18.5months (range, 1.5 to 67.3) follow-up, showed no evidence about this (17). It might because that they only focused on univariate effect and ignored the interaction between covariates. In univariate analysis, they found that surgery opportunity, emergent or elective, had signi cantly distinctive effect on the survival, which is consistent with our nding.
Clinical indications for emergent/urgent TEVAR includes ruptured descending thoracic aortic aneurysms, acute BTAI and complicated acute type B aortic dissections (40). In our series, TAP patient with AEF manifested as massive haematemesis or hematochezia was also indications. Although TEVAR therapy could prevent death by exsanguination, the survival outcome is not desirable due to high risk of infectious complications and failure to solve fundamental cause. AEF patients following open management also had poor outcome including implantation of vascular prosthesis and patch repair (41). The latest research indicated that in situ aortic replacement by cryopreserved aortic homograft and concomitant primary closure of the esophagus is a feasible and promising therapy for primary and secondary AEF (42).
TAP patients with cardiac disease (16.2% of the series) as an independent risk factor were proved to pose a 4.9-fold increased risk to worse late survival outcome. Patients with coronary disease also developed aortic calci cation, particularly the occurrence of atherosclerotic aortic aneurysm reached 46.3%. A research suggested that cardiac catheterization before TEVAR is necessary for evaluating coronary arteries' condition and predicting the degree of aortic calci cion (43).
In this series of TAP, the mortality at 1 and 5 years was 75.9 ± 7.5 and 68.3 ± 9.9, which was a satisfactory result comparing to open surgery (44). The mortality of over 10% at 3 months might links to patients' complex condition or early complication of TEVAR. As a research showed that once emergent cases survived TEVAR procedure, the survival of them become stable after 6 months (17).
Although there are several results in this report, there are also limitations. Actually, the effectiveness of endovascular therapy is underestimated because we only calculated all-cause mortality and concomitant cardiac disease was proved signi cantly associated with TAP patients' survival. It is a retrospective observational study based on single center experience with a long time span and limited cases. Cohort study and randomized controlled trial are needed to further demonstrate our results.

Conclusion
In conclusion, TEVAR therapy is effective and safe with satisfactory early and midterm clinical outcome for TAP patients. Lesions at zone 2/3, concomitant cardiac disease and emergent TEVAR might be independent risk factors for their midterm survival outcome.