Early and midterm outcomes of endovascular treatment for chronic aneurysmal aortic dissection: a retrospective study

Abstract

Background:

The aim of this study was to evaluate endovascular treatment for enlarged Stanford type B chronic aneurysmal aortic dissection (CAAD). The conventional treatment for CAAD is open repair; however, the operative mortality is high in extensive prosthetic graft replacements.

Methods:

A retrospective single-center study was conducted on 74 consecutive patients who underwent endovascular treatment for CAAD in the past 8.5 years. In the partial exclusion (PE) group, entry sites in close proximity to the maximum diameter of CAAD were closed using a stent graft and reentry sites were left without closure. In the complete exclusion (CE) group, we attempted to close all entry and reentry sites.

Results:

A total of 43 patients (PE group) and 31 patients (CE group) were included with mean ages of 59 and 63 years, respectively. Operative mortalities of 2.3% and 0% were observed in the PE and CE groups, respectively. Complete tear closure was successful in 17 of 31 patients (54.8%) in the CE group. In the PE group, complete thrombosis of the false lumen was achieved in only one case (2.3%). Freedom rates from reentry closure were 90.2%, 86.9%, and 78.2% at 1, 3, and 5 years, respectively. The diameter of the true lumen/aorta changed from 16.9/62.9 mm to 30.2/53.6 mm and from 13.7/55.1 mm to 25.8/51.0 mm in the aortic arch and descending thoracic aorta, respectively. The freedom rates from secondary intervention in successful and unsuccessful CE cases were 92.9% and 69.1%, respectively, at 1 year and 92.9% and 53.7%, respectively, at 3 years.

Conclusion:

Endovascular treatment for CAAD had favorable early and midterm outcomes.

Introduction

In Stanford type B acute aortic dissection, in particular complicated cases with malperfusion and rupture, entry closure using stent grafts (SGs) is useful. However, for uncomplicated Stanford type B acute aortic dissection, there is still some room for discussion regarding treatment using SGs.^1,2 At present, complicated cases, such as those mentioned above or those with an aortic diameter of ⩾40 mm at onset are managed by acute-phase SG surgery. However, endovascular treatment for uncomplicated cases remains controversial; therefore, conservative therapy is usually selected. Meanwhile, several postoperative cases of Stanford type A or type B aortic dissection that are not associated with organ ischemia or rupture and that are given conservative therapy transition to chronic aortic aneurysmal dissection (CAAD) eventually require treatment.^2,3

Generally, 20–50% of Stanford type B aortic dissections under conservative therapy transition to CAAD during the chronic phase.^4,5 The conventional treatment for CAAD is open repair; however, there have been many cases of extensive aortic aneurysms that were difficult to treat. Accordingly, surgery has been commonly performed in stages. In cases of extensive prosthetic graft replacement, the operative mortality has been reported to be approximately 8–11%.^6,7 In addition, the risk of paraplegia after extensive prosthetic replacement performed in one-stage surgery is relatively high, whereas surgery performed in several stages results in repeat surgery and increased surgical risks.^8,9 Under these circumstances, endovascular treatment by entry closure using SG for CAAD was introduced and has become widespread in recent years. We have been performing this procedure at our institution in Japan. Therefore, this study was aimed at evaluating the early and midterm outcomes of this procedure and to discuss the validity of our treatment plan for CAAD.

Methods

Treatment plan

First, the position of the aortic tear was confirmed preoperatively by a contrast-enhanced computed tomography (CT) scan. Usually, an entry site was observed in the aortic arch or the descending thoracic aorta, whereas the reentry site was seen at a location that was distal to the entry site. The entry site was closed using SG [Figure 1(a) and (b)]. Depending on the case, endovascular treatment was also performed to close other tears. Branch tears of the celiac artery were closed by coil embolization ± SGs [Figure 1(c)]. Small tears were closed using an Amplatzer vascular plug [AVP; St. Jude Medical, Inc., St. Paul, MN, USA; Figure 1(d)], whereas intimal tears caused by dissection of the main abdominal branches (e.g. renal arteries) were closed by bridging the intimal hole and the branch by a covered stent [Figure 1(e)] or by the combined use of a fenestrated SG and a covered stent [Figure 1(f)]. Branch tears of the superior mesenteric artery [Figure 1(g)] or carotid artery [Figure 1(h)] were closed by a self-expanding bare stent. Tears in the iliac artery or the infra-renal abdominal aorta were closed using SGs.

Figure 1.

Various methods of tear closure during endovascular treatment for CAAD.

(a) Angiography of entry closure using stent graft

(b) A CT before and after entry closure: expansion of the true lumen, thrombosis of the false lumen, and reduction of the false lumen can be seen

(c) When reentry is close to celiac artery, celiac trunk is embolized with coil to prevent back flow from celiac artery and stent graft is deployed to cover the reentry

(d) Reentry closure using AVP

(e) A covered stent (VIABAHN^®: WL Gore & Associates, Inc., Flagstaff AZ, USA) was deployed from intimal tear to the renal artery

(f) When the intimal tear is larger than 6–8 mm, fenestrated stent graft is deployed and covered stent (VIABAHN^®) is deployed from fenestration to the renal artery

(g) An SMA stent is deployed to cover the intimal tear close to SMA

(h) A carotid artery stent is deployed to cover the intimal tear in carotid artery

AVP, Amplatzer vascular plug; CAAD, chronic aneurysmal aortic dissection; CT, computed tomography; SG, stent graft, SMA, superior mesenteric artery.

Entry closure alone was performed if the region of maximum CAAD diameter was located in the entry area (partial exclusion; PE group), whereas reentry closure was performed in addition to entry closure if the region of maximum expansion was located distal to the entry site (complete exclusion; CE group).

Postoperative protocol

The patients underwent outpatient follow-up observation at 1 month, 6 months, and 1 year postoperatively and every 6 months thereafter. Patients were examined and underwent contrast-enhanced CT to determine the presence of thrombosis in the false lumen and to measure the diameter of the aortic arch, descending thoracic aorta, and abdominal aorta, including the diameters of the true and false lumens. The descending thoracic aorta was measured at the maximum aortic diameter at the level of the left ventricle, whereas the diameter of the abdominal aorta was measured at the level of the celiac artery.

Study design

A retrospective study was performed on 74 patients who had undergone endovascular intervention for Stanford type B CAAD at our hospital during a period of 8 years and 7 months from July 2006 to February 2015. Patients whose clinical course had elapsed at least 1 year after the onset of dissection, excluding those with acute dissection, were included. Patients with the acute thrombosed type and those with ulcer-like projections on the CAAD were also excluded. Among the eligible patients, 43 were classified as the PE group and 31 as the CE group. The CE group was further subdivided into CE success and CE attempt groups. Furthermore, the factors to gain favorable results such as complete thrombosis of false lumen after endovascular treatment were analyzed using univariate and multivariate analysis. The institutional review board at Jikei Univesity School of Medicine, Japan, approved this study [no. 29-332 (8948)], and written informed consent from each patient was waived because of the retrospective design.

No power calculation was conducted for the study sample size beforehand since all cases within the study period were included.

Diagnostic imaging

A preoperative contrast-enhanced CT was performed in all cases to identify in detail the position of the entry site and other tears (reentry). The maximum short-axis diameter of the CAAD and the access conditions were also evaluated.

Definitions

In both the PE and CE groups, cases in which the target entry sites were successfully closed were defined as technical successes. In the CE group, cases in which the absence of blood flow in the false lumen was observed on CT images were defined as CE successes, whereas those with residual blood flow in the false lumen were defined as CE attempts. Depending on the degree of thrombosis of the false lumen on postoperative CT images, cases were classified as either complete or partial thrombosis.

Chronic kidney disease was defined as an estimated glomerular filtration rate of <45 ml/min/1.73 m². Chronic obstructive pulmonary disease was defined as having a forced expiratory volume in one second of <700 ml or emphysematous changes on performing chest CT with the requirement of home oxygen therapy.

Statistical analysis

Continuous variables were presented as mean ± standard deviation. The two groups were compared using Student’s t test and the Chi-square test as appropriate. Operation-related variables, such as the duration of surgery, were presented as median values, and the Mann–Whitney U test was performed for statistical analysis. The overall survival (OS) rate of freedom from secondary intervention and the rate of freedom from aneurysm-related death after treatment were analyzed using the Kaplan–Meier method. The log-rank test was performed to compare the two groups using the Kaplan–Meier method. The factors of complete thrombosis of false lumen were examined using logistic regression models for each outcome. Related variables that were significant in the univariate analyses underwent multivariate analysis by the variable reduction method. A p-value <0.05 was considered to be statistically significant. All analyses were performed using SPSS version 20.0 (IBM SPSS, Armonk, NY, USA).

Results

Early outcomes

No significant differences were observed between the two groups in terms of age, sex, duration from dissection onset to treatment, the maximum short-axis diameter of the CAAD, the preoperative condition, and the follow-up observation period (Table 1) even though the duration from dissection onset to treatment tended to be longer in the CE group than it was in the PE group. Operative death occurred in 1 out of 74 patients (1.4%): 2.3% in the PE group; and 0% in the CE group. Technical success in the PE and CE groups were 97.6% and 96.8%, respectively (Table 2). The device that was most commonly used to perform entry closure was the TAG (WL Gore & Associates, Inc., Flagstaff AZ, USA), and no significant difference in the frequency of use was observed between the two groups.

Table 1.

Profiles of patients.

	PE group	CE group	p-value
Number of patients	43	31
Age	58.9 ± 11.4	62.5 ± 12.4	n.s.
Sex (male)	34 (79.12%)	25 (60.6%)	n.s.
Interval from onset to treatment (months)	80.2 (12–250)	90.7 (12–384)	n.s.
Emergent cases	2 (4.6%)	2 (6.5%)	n.s.
Aneurysm size (mm)	67.1 (55–75)	65.0 (57–87)	n.s.
hypertension	38 (88.4%)	28 (90.3%)	n.s.
diabetes	4 (9.3%)	3 (9.7%)	n.s.
CAD	8 (18.6%)	5 (16.1%)	n.s.
CKD	8 (18.6%)	8 (25.8%)	n.s.
COPD	5 (11.6%)	4 (12.9%)	n.s.
Follow-up period (months)	40.5 ± 23.5	40.4 ± 25.5	n.s.

CAD, coronary artery disease; CE, complete exclusion; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; n.s., not significant; PE, partial exclusion.

Table 2.

Operative data.

	PE group (n = 43)	CE group (n = 31)	p-value
Operation time (min)	200 (95–603)	221.5 (110–772)	p < 0.05
Blood loss (g)	200 (50–3300)	200 (50–2400)	n.s.
Contrast used (ml)	345 (150–800)	350 (75–665)	n.s.
Fluoro time (min)	37.5 (14–119)	39 (20–111)	n.s.
Technical success	97.6%	96.8%	n.s.
Mortality	1 (2.3%)	0	n.s.
Stroke	0	0	n.s.
Heart/ respiratory failure	0	0	n.s.
Renal insufficiency	0	0	n.s.
Spinal cord ischemia	0	1 (3.2%)	n.s.
Stent graft deployed			n.s.
TAG	39	26
TX2	3	4
others	1	1

CE, complete exclusion; n.s., not significant; PE, partial exclusion.

In the patient who died in the PE group, the tear in the arch could not be closed; therefore, the carotid artery was concomitantly reconstructed using the Chimney method. Furthermore, because of the fact that this patient previously underwent coronary artery bypass grafting using the left internal thoracic artery, the SG was extended proximally after creating a bypass from the left common carotid artery to the left subclavian artery. However, the entry site could not be completely closed, and the blood pressure dropped during surgery. Unfortunately, the patient died from cardiac failure on the same day due to prolonged hypotension. In one patient in the CE group, we attempted to close a renal artery tear; however, a new intimal tear formed. In this case, conversion to open surgical repair was performed the next day, and the patient was eventually discharged from the hospital without any major complications. Prosthetic graft replacement with reconstruction of abdominal branches was performed, and prosthetic graft could be anastomosed to prior SGs (Figure 2). No central nervous or spinal cord complications were observed except for one case of delayed paraplegia on postoperative day (POD) 2 in CE group.

Figure 2.

Prosthetic graft is anastomosed to prior stent graft (white arrows: anastomotic site).

The duration of surgery was significantly longer in the CE group; however, no significant differences were observed between the groups in terms of blood loss volume, volume of contrast agent used, and fluoroscopy duration (Table 2). There were 17 cases of CE success (54.8%) in the CE group. When the CE success and CE attempt groups were compared, no differences were observed with regard to age, sex, and number of tears closed; however, the duration from the onset of dissection to treatment tended to be longer in the CE attempt group. Moreover, the number of tears, particularly in the abdominal branches, was significantly higher in many cases in the CE attempt group. Ultimately, three patients (21.4%) in the CE attempt group required open conversion (Table 3). During subsequent CT imaging, we were able to confirm complete thrombosis of the false lumen in three more patients and finally in 20 patients in the CE group (64.5%). In the PE group, thrombosis of the false lumen was complete in one patient (2.3%) and partial in the remaining 42 patients (97.7%).

Table 3.

Comparison of the CE success group and CE attempt group.

	CE success group (n = 17)	CE incomplete group (n = 14)	p-value
Age	63.0 ± 15.0	61.2 ± 9.2	n.s.
Sex (male)	14 (82.3%)	11 (78.6%)	n.s.
Interval from dissection to treatment (months)	71.5 ± 59.3	114.1 ± 96.7	n.s.
Number of intimal tears	2.5 ± 0.7	4.0 ± 2.0	p < 0.05
Number of tears closed	2.5 ± 0.7	3.1 ± 2.0	n.s.
Number patients with tears involving abdominal branches	3 (17.6%)	12 (85.7%)	p < 0.05
Open conversion	0	3 (21.4%)	n.s.
Spinal cord ischemia	0	0	n.s.

CE, complete exclusion; n.s., not significant.

Aortic remodeling

On reviewing aortic remodeling after treatment, gradual shrinkage of entire aorta and expansion of true lumen were observed when complete thrombosis was achieved. On the other hand, in the PE group it was also observed that the closure of entry sites that were mainly located on the aortic arch caused rapid expansion of the true lumen of the aortic arch and the descending thoracic aorta. We also observed gradual shrinkage of the diameter of the thoracic aorta. However, even though we observed an expansion of the true lumen of the abdominal aorta, no changes or gradual expansion in the aortic diameter were observed. In addition, we discovered that the expansion of the true lumen was approximately 1.8-fold in the thoracic aorta when compared with the 1.3-fold expansion of the true lumen in the abdominal aorta. These results indicated that the degree of expansion was also limited (Figure 3) in the PE group.

Figure 3.

Changes in aortic diameter after entry closure.

n.s., not significant.

Midterm outcomes

Postoperative follow up was possible for all patients. The results were favorable, with OS rates of 97.2%, 90.1%, and 82.9% at 1, 3, and 5 years, respectively [Figure 4(a)]. In the PE group, the OS rates at 1, 3, and 5 years were 97.4%, 94.3%, and 94.3%, respectively. In the CE group, the OS rates at 1, 3, and 5 years were 93.3%, 86.0%, and 73.9%, respectively. The results tended to be higher in the PE group; however, no significant differences were observed between the two groups [p = 0.0622; Figure 4(b)]. In addition, the rates of freedom from aneurysm-related death at 1, 3, and 5 years were 100%, 96.9%, and 96.9%, respectively, in the PE group and 96.7%, 96.7%, and 96.7%, respectively, in the CE group. Results in both groups were favorable, and no significant differences were observed between the two groups (Figure 5). Furthermore, the rates of freedom from secondary intervention at 1, 3, and 5 years were found to be 73.4%, 61.5%, and 40.8%, respectively, in the PE group and 79.9%, 65.4%, and 53.3%, respectively, in the CE group. The rates in the CE group tended to be higher; however, no significant differences were observed between the two groups (p = 0.282; Figure 6).

Figure 4(a).

Overall survival of CAAD patients.

CAAD, chronic aneurysmal aortic dissection, CE, complete exclusion; n.s., not significant; PE, partial exclusion.

Figure 4(b).

Comparison of overall survival among CAAD patients.

CAAD, chronic aneurysmal aortic dissection, CE, complete exclusion; n.s., not significant; PE, partial exclusion.

Figure 5.

Freedom from aneurysm-related death.

CE, complete exclusion; n.s., not significant; PE, partial exclusion.

Figure 6.

Freedom from secondary intervention among CAAD patients.

CAAD, chronic aneurysmal aortic dissection; CE, complete exclusion; n.s., not significant; PE, partial exclusion.

The rates of freedom from secondary intervention at 1, 3, and 5 years were 93.3%, 84.0%, and 58.0%, respectively, in the CE success group and 63.4%, 42.3%, and 42.3%, respectively, in the CE attempt group (Figure 7). Upon comparing the CE subgroups, the results were more favorable in the CE success group than they were in the CE attempt group (p = 0.0185). A comparison of freedom from secondary intervention at 1, 3, and 5 years according to the degree of thrombosis of the false lumen showed rates of 92.9%, 92.9%, and 65.0%, respectively, in the complete thrombosis group and 71.2%, 53.7%, and 41.4%, respectively, in the partial thrombosis group (p = 0.022, Figure 8).

Figure 7.

Freedom from secondary intervention in the CE subgroups.

CE, complete exclusion; PE, partial exclusion.

Figure 8.

Freedom from secondary intervention according to degree of false lumen thrombosis.

CE, complete exclusion; PE, partial exclusion.

Secondary intervention in the PE group included proximal extension in 10 patients, distal reentry closure in 8 patients (Figures 9 and 10), repair of a new intimal tear at the distal edge of the SG in 2 patients, and additional treatment at the same entry site in 1 patient with insufficient entry closure. The rates of freedom from reentry closure at 1, 3, 5, and 7 years were favorable at 90.2%, 86.9%, 78.2%, and 67.0%, respectively (Figure 11). In the PE group, additional treatment was performed in six patients, and ultimately, complete thrombosis of the false lumen was achieved in seven patients. Complete thrombosis of the false lumen was achieved in 27 out of 74 patients (36.5%). The mean number of surgical procedures required to achieve complete thrombosis of the false lumen was 1.68 (range, 1–4).

Figure 9.

Typical example of CAAD treatment.

(a) Preoperative angiogram show large entry site in the aortic arch and narrowing of the true lumen.

(b) Angiogram 1 year after entry closure shows expansion of the true lumen of the aortic arch through the abdominal aorta. A tear can be also observed close to the abdominal branches and right external iliac artery.

(c) Angiogram 1 year after entry closure shows tears in the right renal and external iliac arteries.

(d) Angiogram after entry closure: a covered stent is placed in the right renal artery and the reentry site is closed by bridging the intimal tear and renal artery.

SG is placed from the common iliac artery to the external iliac artery to close the reentry site. Contrast-enhancement of the false lumen is no longer observed.

CAAD, chronic aneurysmal aortic dissection; CT: computed tomography; SG: stent graft.

Figure 10.

Three-dimensional contrast-enhanced CT images before and after CAAD treatment.

(a) Preoperative CT image reveals the entry site and dissection in the aortic arch through the right external iliac artery.

(b) After entry closure, contrast-enhancement of the false lumen can be no longer be observed in the aortic arch, but is evident from the tear close to the abdominal branches and the right external iliac artery.

(c) After complete closure of all tears in the abdominal branches and right external iliac artery, blood flow through the false lumen is completely stopped.

CAAD, chronic aneurysmal aortic dissection; CT: computed tomography.

Figure 11.

Freedom from reentry closure.

Factors to obtain complete thrombosis of false lumen

We performed a univariate analysis using factors to obtain a complete thrombosis of false lumen. In the univariate analysis, fewer intimal tears, fewer tears located on abdominal aorta, and absence of tears involving visceral branches were significant patient-related factors to obtain a complete thrombosis of false lumen. We performed a logistic regression analysis to identify independent factors to obtain a complete thrombosis of false lumen, absence of tear involving visceral branches was an independent factor to obtain a complete thrombosis of false lumen (Table 4).

Table 4.

Factors to obtain complete thrombosis of false lumen.

Variables	Univariate analysis			Multivariate analysis
	Odds ratio	95% CI	p-value	Odds ratio	95% CI	p-value
Patient-related factors
Age	1.304	0.9890–1.089	0.131
Male sex	0.856	0.2350–3.1143	0.813
Period after onset	0.996	0.9878–1.004	0.391
hypertension	0.600	0.1337–2.692	0.5,0
dyslipidemia	0.717	0.2352–2.188	0.559
diabetes	1.275	0.2252–7.216	0.784
CAD	1.492	0.3984–5.587	0.553
COPD	1.667	0.3714–7.479	0.505
CKD	2.300	0.6962–7.598	0.172
Number of intimal tears	0.236	0.0966–0.5784	0.002	0.742	0.1816–3.032	0.678
Number of tears located on abdominal aorta	0.133	0.0466–0.3791	0.000	0.363	0.0775–1.696	0.197
Tears involving visceral branches	0.024	0.0053–0.1077	0.000	0.053	0.0106–0.2621	0.000
Procedure-related factors
CE	173.4	18.907–1590.3	0.000

CAD, coronary artery disease; CE, complete exclusion; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease.

Discussion

When treating uncomplicated Stanford type B acute aortic dissections, the use of SG during the acute phase is still controversial.^1,2 Even though conservative therapy has been the routine management procedure for such cases, it has been made clear that many of these cases evolve to CAAD and end up requiring treatment.^2,3 Such CAAD lesions tend to be extensive, and outcomes of conventional prosthetic graft replacement are not satisfactory.^6–9 In this study, all CAAD entry sites were closed using SG^1,2 and the need for reentry closure was based on each individual case. In such CAAD cases, large tears that can become the entry site are often located in the aortic arch. Nevertheless, the rate of surgical death after entry closure by SG was reported to be approximately 3%;¹⁰ therefore, it is a relatively safe procedure when compared with open repair. In this study, the rate of surgical death in endovascular treatment was favorable at 1.4%. Therefore, we believe that entry closure using SG is an extremely well tolerated procedure.

Despite the fact that treating CAAD by entry closure using SG may be well tolerated, there is room for discussion regarding its efficacy. It is believed that prompt treatment during the early phase of dissection facilitates more remodeling by entry closure.¹¹ Accordingly, there have been cases that were successfully and completely treated by the endovascular closure of entry sites alone immediately after onset. However, the efficacy of such a procedure is unclear when CAAD has occurred. Therefore, in this study, we chose patients in whom at least 1 year had elapsed since the onset of dissection and who had developed CAAD.

The final objective of CAAD treatment is complete thrombosis of the false lumen and the eventual disappearance of the false lumen. Actually, gradual shrinkage of entire aorta and expansion of true lumen were observed when complete thrombosis was achieved. This positive remodeling of entire aorta ensures low rate of secondary intervention and aneurysm-related death. In this study, despite rarely achieved complete thrombosis of the false lumen by closing the entry site alone to treat CAAD, entry closure prevented CAAD from persistent enlargement, especially in the aortic arch close to entry site. If the entry sites were located on the maximum expansion area, only entry closure was performed. Even though there were few cases that required reentry closure even if the entry sites were successfully closed initially, the rate of freedom from reentry closure during follow up was generally favorable. Therefore, it appears that firmly closing the entry sites was just as effective in preventing CAAD expansion as entry closure in both the acute and chronic phases of aortic dissection. For cases in which the aortic arch entry sites were present in an area of maximum expansion and without much expansion in the distal abdominal aorta or below, it appeared that performing entry closure alone was sufficient. In contrast, if the descending thoracic aorta and its distal aorta are expanded, additional reentry closure may be needed after entry closure for the same time or the second stage. If contrast-enhanced imaging is performed after closing the entry site, a large volume of blood flow from the entry site through the false lumen disappears; this would make the flow from the reentry site into the false lumen obvious. We presumed that additional reentry closure may be necessary if the false lumen blood flow from this reentry site toward the descending thoracic aorta is marked. Endovascular reentry closure may be ideal, however it in not a firm treatment for reentry closure. In fact, in this study, the success rate of endovascular reentry site closure was 54.8%. Open repair is feasible after endovascular entry closure, moreover TEVAR for the purpose of entry closure will make the invasiveness of second operation lower.

To achieve complete thrombosis, it is important to minimize blood flow through the false lumen. In this univariate analysis, fewer intimal tears, fewer tears located on abdominal aorta, and absence of tears involving visceral branches were significant factors to obtain a complete thrombosis of false lumen. In the multivariate analysis, absence of tears involving visceral branches was an independent factor to obtain a complete thrombosis of false lumen. The infallible method of closing tears, including those in the abdominal branches, needs to be further considered. However, significantly fewer secondary interventions were performed, and fewer complications occurred when additional closure of the reentry site successfully achieved complete thrombosis of the false lumen when compared with cases wherein closure was performed on the entry site alone. Therefore, closing both entry and reentry sites should be attempted when possible. Because CAAD needs more extensive treatment, if treatment is performed as a single-stage surgery, it is essential to exercise caution, such as maintaining a relatively high blood pressure and securing cerebrospinal fluid drainage, to minimize the risk of paraplegia.^12,13

It is important to note that the closure of the entry site alone can expand the true lumen of the aortic arch through the thoracic descending aorta and reduce the aortic diameter; these findings suggest that preventing the subsequent expansion of a CAAD is an effective management practice. In fact, we were able to obtain satisfactory rates of freedom from reentry closure after entry closure. In cases where complete thrombosis of the false lumen was not achieved, the need for secondary intervention was highly likely when compared with cases where complete thrombosis was achieved. Because endovascular treatment has the advantage of repeated performance without significantly increasing surgical risk, entry closure may be a safe and useful method of managing CAAD. Endovascular entry closure should be a first line treatment for CAAD patients. After the endovascular entry closure is performed, endovascular reentry closure should be performed if necessary. When endovascular reentry closure does not work or it seems to be difficult, such as CAAD with tears involving abdominal branches, open repair (prosthetic graft replacement with reconstruction of abdominal branches) should be performed.

Endovascular entry closure for uncomplicated Stanford type B acute aortic dissection will be accepted for positive remodeling of the aorta in the near future.

Conclusion

Even though endovascular intervention to treat CAAD had a high rate of secondary intervention, its midterm outcomes were generally favorable. Therefore, this could be considered to be one of the effective treatment options for CAAD. Closing the entry site alone was effective in preventing subsequent expansion of CAAD. However, an attempt to perform additional closure of the reentry site to achieve complete thrombosis of the false lumen led to fewer secondary interventions and decreased complications.

Limitations

This retrospective study has several limitations, including the small number of patients, potential selection bias and restricted external validity because of the single-center analysis. For instance, the devices and operative techniques used were based on the preferences of the surgeon. Therefore, the outcomes may be biased according to the learning curve.