Acute type B aortic dissection, characterized by an in-hospital mortality rate of 13% and a significant 40% mortality rate in cases of acute complications, poses a significant threat to patient well-being [1,2]. While conventional management typically involves medical therapy, surgical intervention becomes imperative in complex cases, such as malperfusion, aortic rupture, refractory hypertension, and persistent pain. In recent years, endovascular treatment has emerged as a preferred approach to open aortic repair in the early stages, due to its association with reduced morbidity and mortality rates [3-5].

However, despite these advances, there remains a dearth of comprehensive reporting on the long-term outcomes of endovascular treatment, particularly concerning morphological analysis and the characteristics of aortic remodeling—a crucial factor influencing the risk of late aortic events. Existing studies on aortic remodeling after endovascular treatment for acute complicated type B aortic dissection encounter limitations stemming from the heterogeneity of remodeling measurements, diverse patient conditions (acute, subacute, or chronic), extensiveness of treatment, and restricted sample size [6-11].

This study reports the clinical outcomes of endovascular treatment in patients with acute complicated type B aortic dissections, with a follow-up period extending beyond 6 years. The primary focus is the complex process of aortic remodeling post-treatment. The insights gained from this research are intended to inform and enhance future treatment protocols for patients with this challenging vascular pathology.

A retrospective investigation utilizing data from the Songklanagarind Hospital database was conducted to scrutinize patients with acute complicated type B aortic dissections treated with thoracic endovascular aortic repair (TEVAR). The study encompassed the period from January 2010 to January 2022, and a comprehensive review of electronic medical records was conducted. Initially, 38 patients were considered for inclusion in this study. However, computed tomography (CT) images for six patients were inaccessible via the hospital’s Picture Archiving and Communication System. As a result, the final study cohort comprised of 32 patients who met the inclusion criteria. Data compilation included variables such as age, sex, body mass index (BMI), comorbidities, tobacco use, history of previous aortic interventions, American Society of Anesthesiologists (ASA) class, aortic details, complications, in-hospital metrics, and follow-up outcomes.

1) Procedure

General anesthesia was administered, and a totally percutaneous approach to the common femoral arteries was performed, along with systemic heparinization in accordance with routine protocol. Typically, the thoracic stent-graft oversized by 0% to 10%. The primary entry tear was covered with a proximal sealing zone of 1.5-2.0 cm. In situations where achieving proximal sealing posed challenges, additional procedures, such as bypass, chimney, or fenestration of the arch branch, were performed. Coverage of the left subclavian artery without bypass was considered permissible if the patient exhibited hemodynamic instability and had a left vertebral artery diameter of less than 5 mm. For patients with malperfusion, a balloon-expandable stent graft was utilized. Surgical success was defined as the successful deployment of the stent graft at an appropriate location, effectively covering the primary tear without any proximal endoleak.

2) Follow-up

Follow-up computed tomography angiography (CTA) was conducted as part of the routine protocol at intervals of 1, 6, and 12 months, and annually thereafter. During each follow-up visit, comprehensive clinical data were meticulously documented, including the resolution of pathological conditions and symptoms, occurrence of complications, instances of re-intervention, and mortality rates.

3) Term, definition, and imaging measurement protocol

All preoperative and follow-up whole aorta CT angiograms underwent meticulous evaluation, involving measurements of the maximum true and false lumen (TL and FL) diameters, proximal and distal extension of dissection, vessel malperfusion, and endoleak during follow-up CT assessments. These measurements, which are crucial for discerning aortic remodeling, were performed by two proficient vascular surgeons (SW and PJ) [12].

The aorta is generally divided into three distinct segments for classification purposes: Section 1 encompasses the thoracic-abdominal aorta located proximally above the celiac trunk, corresponding to zones 3-5 according to established reporting standards; Section 2 comprises the abdominal aorta between the celiac trunk and renal artery, corresponding to zones 6-8; and Section 3 includes the abdominal aorta positioned distally below the renal artery, corresponding to zones 9-11. In this study, the focus was on Section 1, which encompasses the primary entry tear and stent graft placement [8,13] (Fig. 1).

Figure 1. Imaging measurement protocol was demonstrated. CA, celiac artery; SMA, superior mesenteric artery; RRA, right renal artery; LRA, left renal artery.

① Pretreatment TL and FL diameter

Assessed by measuring the maximum diameter of the TL and FL in pretreatment axial CTA images within the aortic Section 1.

② Post-treatment TL and FL diameters

Evaluated by measuring TL and FL diameters using post-treatment axial CTA images, following a routine interval protocol. These measurements were taken at the same locations as the pretreatment assessments.

③ Favorable lumen remodeling

Defined as an increase of more than 20% in the posttreatment TL diameter compared to the pretreatment TL diameter, coupled with a decrease of more than 20% in the post-treatment FL diameter compared to the pretreatment FL diameter.

④ Unfavorable lumen remodeling

Defined as a post-treatment aortic diameter measurement that did not meet the criteria for favorable aortic remodeling, as previously outlined.

⑤ Total FL thrombosis

Defined as the absence of any contrast fill-in within the FL observed on post-treatment CTA images of the aortic Section 1.

⑥ Partial FL thrombosis

Defined as the presence of any contrast fill-in within the FL observed on post-treatment CTA images of the aortic Section 1.

⑦ Aortic remodeling

Aortic remodeling was categorized into four groups based on a combination of lumen remodeling and FL thrombosis type; (1) Ideal aortic remodeling: This category comprises patients exhibiting favorable lumen remodeling and total FL thrombosis. (2). Average aortic remodeling: Patients in this category demonstrated favorable lumen remodeling along with partial FL thrombosis. (3) Undesirable aortic remodeling: Patients in this group displayed unfavorable lumen remodeling in conjunction with partial FL thrombosis. (4) Fair aortic remodeling: This category encompasses patients exhibiting unfavorable lumen remodeling and total FL thrombosis.

4) Statistical analysis

Continuous variables with a normal distribution were expressed as mean±standard deviation, while non-normally distributed variables were presented as median and interquartile range. Categorical variables were expressed as percentages. Survival analyses, including overall survival, disease-specific survival, and reintervention-free survival were performed using Kaplan-Meier curves. Additionally, a Cox regression model was used to perform multivariate survival analyses considering pertinent patient factors.

The study protocol was approved by the Human Research Ethics Committee of the Prince of Songkla University (No. REC.65-352-10-3). All statistical analyses were performed using RStudio version 1.1.456 (RStudio, Inc.), ensuring a rigorous and meticulous analytical approach in line with scholarly standards.

Over the past decade, 32 patients with acute complicated type B aortic dissections who underwent TEVAR at our institution were included in this study. The majority of the patients were male, with an average age of 57 years and an average BMI of 25 kg/m². Hypertension was the predominant comorbidity. The ASA classification was predominantly Class 3, and three patients presented with aortic rupture. TEVAR was most commonly indicated due to malperfusion, particularly affecting the left renal and left lower limb vessels. The pretreatment TL and FL diameters averaged 12.9 and 27.9 mm, respectively, with more than half of the patients exhibiting distal extension of the tear to aortic zone 10 (Table 1).

Table 1 . Patients’ demographic features (n=32).

Patient’s characteristic	Number
Age (y)	57.6±11.9
Male	27 (84.4)
BMI (kg/m2)	25 (23.4, 26.9)
Smoking	26 (81.3)
Active	12 (37.5)
Non-active	14 (43.8)
Hypertension	29 (90.6)
Diabetes	4 (12.5)
COPD	1 (3.1)
CAD	4 (12.5)
CKD	3 (9.3)
Stage 2	1 (3.1)
Stage 3	1 (3.1)
Stage 4	1 (3.1)
Previous aortic intervention	3 (9.4)
ASA class
2	1 (3.1)
3	25 (78.1)
4	6 (18.8)
Indication for TEVAR
Rupture	3 (9.3)
Malperfusion	16 (50.0)
Right renal artery	3 (18.8)
Left renal artery	7 (43.8)
Celiac artery	1 (6.3)
Superior mesenteric artery	1 (6.3)
Right limb	4 (25.0)
Left limb	7 (43.8)
Spinal cord	1 (6.3)
Persistent pain	1 (3.1)
Uncontrolled hypertension	2 (6.3)
Progressive disease	13 (40.6)
Pretreatment TL diameter (mm)	12.9
Pretreatment FL diameter (mm)	27.9
Distal extension tear zone
5	3 (9.4)
7	2 (6.3)
9	5 (15.6)
10	18 (56.2)
11	4 (12.5)

Values are presented as number (%) or mean±standard deviation..

COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; CKD, chronic kidney disease; ASA, American Society of Anesthesiologists; TEVAR, thoracic endovascular aortic repair; TL, true lumen; FL, false lumen..

The Valiant thoracic stent (Medtronic Vascular) was the primary choice for almost all patients, with proximal stent deployment most frequently in zones 2 and 3, and distal stent landing predominantly in zone 5. The mean stent length was 185 cm. The average operative time was 235.0 minutes, with a median radiation time of 12 minutes. The median contrast and radiation doses were 72.5 mL and 200 mGy, respectively. A total of 68.8% of the patients underwent concomitant procedures, mainly related to achieving proximal sealing, with carotid-subclavian bypass and left subclavian coverage being the most common (Table 2).

Table 2 . Operative information.

Operative information	Number
Valiant thoracic stent	29 (90.6)
Proximal stent zone
1	2 (6.3)
2	17 (53.1)
3	12 (37.5)
4	1 (3.1)
Distal stent zone
5	31 (96.9)
6	1 (3.1)
Stent length (mm)	185 (150, 200)
Operative time (min)	235.0±88.5
Contrast doses (mL)	72.5 (51.3, 117.0)
Radiation time (min)	12.0 (9.0, 19.0)
Radiation doses (mGy)	200±114
Concomitant procedure	22 (68.8)
Proximal sealing related
Carotid-subclavian bypass	10 (45.5)
Left subclavian artery coverage	4 (18.2)
Carotid-carotid with carotid-subclavian bypass	2 (9.1)
Fenestrated left subclavian artery	2 (9.1)
Chimney left subclavian artery	1 (4.5)
Chimney left common carotid with carotid-subclavian bypass	1 (4.5)
Visceral vessel related
PTA common iliac artery	2 (9.1)
Celiac stenting	1 (4.5)
SMA stenting	1 (4.5)
Iliac stenting	1 (4.5)
Fem-fem bypass	1 (4.5)

Values are presented as number (%) or mean±standard deviation..

PTA, percutaneous transluminal angioplasty; SMA, superior mesenteric artery..

The median hospitalization duration was 14.0 days, including 3.0 days in the intensive care unit and 1.0 day on mechanical ventilation. The in-hospital mortality rate was 9.4%. Among the deceased, one patient suffered hypoxic arrest due to upper airway obstruction resulting from a neck hematoma, another succumbed to septic shock caused by severe ventilator-associated pneumonia, and the final case involved multiorgan failure induced by severe reperfusion syndrome following prolonged limb ischemia. Notably, all deceased patients exhibited signs of malperfusion, with none presenting with ruptures. Complications occurred in 21.9% of the cases, including temporary acute kidney injury (AKI), wound bleeding, pneumonia, and permanent spinal cord ischemia. Reoperation was needed in 6.3% of the cases, with two patients requiring re-exploration of the neck to control bleeding. Immediate endoleaks were observed in 12.5% of cases, predominantly type 2 (Table 3).

Table 3 . In-hospital outcomes.

In-hospital outcome	Number (%)
ICU stay (d)	3.0 (1.0, 7.5)
Ventilator day (d)	1.0 (0, 7.7)
Hospital stay (d)	14.0 (7.0, 21.5)
Mortality	3 (9.4)
Complication	7 (21.9)
AKI (temporary)	3 (9.4)*
Wound bleeding	2 (6.3)
Pneumonia	2 (6.3)
Spinal cord ischemia (permanent)	1 (3.1)*
Reoperation	2 (6.3)
Neck exploration with stop bleeding	2 (6.3)
Immediate endoleak	4 (12.5)
Type 2	3 (75.0)
Type 1A	1 (25.0)

ICU, intensive care unit; AKI, acute kidney injury..

*One patient had AKI and spinal cord ischemia..

Table 4 shows the average changes in the aortic TL and FL diameters over a 5-year follow-up period. Of the 32 patients, 75% (24/32) demonstrated favorable lumen remodeling, while 50% (16/32) achieved total FL thrombosis. Half of the total patients exhibited favorable lumen remodeling coupled with total FL thrombosis, categorized as “ideal aortic remodeling.” Meanwhile, one-fourth of patients who displayed favorable lumen remodeling but partial FL thrombosis were classified as having “average aortic remodeling.” Conversely, 25% of patients with unfavorable lumen remodeling coupled with partial FL thrombosis were categorized as having “undesirable aortic remodeling.” Notably, no instances were recorded in which unfavorable aortic remodeling coincided with total FL thrombosis (Table 5).

Table 4 . Change in the aorta during the follow-up period.

Table 5 . Pattern of aortic change and grouping of pattern remodeling of the aorta.

Survival outcomes over the spans of 6 months, 1 year, 3 years, and 6 years are depicted in the Kaplan-Meier curve, with rates of 78% (95% confidence interval [CI], 65-93), 75% (95% CI, 61-91), 65% (95% CI, 50-84), and 44% (95% CI, 27-70), respectively (Fig. 2A). Conversely, the aorta-related survival rates at the same intervals were 87% (95% CI, 76-99), 87% (95% CI, 76-99), 83% (95% CI, 71-97), and 75% (95% CI, 58-97) (Fig. 2B). Reintervention-free survival exhibited rates of 96% (95% CI, 90-100), 96% (95% CI, 90-100), 71% (95% CI, 56-92), and 71% (95% CI, 56-92) over the specified time points (Fig. 2C).

Figure 2. Kaplan-Meier curves estimated (A) overall survival, (B) aorta-related survival, and (C) reintervention-free survival.

Most reinterventions were associated with the progression of the dissecting aneurysm, necessitating proximal and distal stent extension. There was only one case of retrograde type A dissection requiring open surgical repair. Notably, patients with unfavorable lumen remodeling demonstrated lower survival rates compared to those with favorable lumen remodeling. This trend persisted in patients with partial FL thrombosis compared to those with total FL thrombosis (Fig. 3). Within the distinct aortic remodeling groups, the Kaplan-Meier curve revealed a statistically significant survival difference. Specifically, the ideal remodeling group exhibited the highest survival rates, while the undesirable remodeling group had the lowest (P-value=0.0075). Similarly, poor aorta-related survival was observed in the cohort with undesirable aortic remodeling (P-value=0.0039; Fig. 4).

Figure 3. Kaplan-Meier curves estimated overall survival by (A) aortic remodeling, and (B) FL thrombosis. TL, true lumen; FL, false lumen.

Figure 4. Kaplan-Meier curves estimated (A) overall survival and (B) aorta-related survival by the pattern of aortic remodeling.

In Cox regression analysis, after adjusting for sex, ASA class, concomitant procedure, and rupture status factors, the undesirable aortic remodeling was identified as the sole factor with a statistically significant impact on predicting survival (adjusted hazard ratio 4.37, P-value=0.021; Table 6).

Table 6 . Cox regression analysis for predicting survival by patient factors.

	Crude HR (95% CI)	adjusted HR (95% CI)	P (Wald’s test)
Pattern: ref.=ideal
Average	1.58 (0.44-5.70)	1.34 (0.30-5.93)	0.697
Undesirable	5.13 (1.66-15.85)	4.37 (1.25-15.25)	0.021
Female	0.29 (0.04-2.19)	0.13 (0.01-1.39)	0.092
ASA class 4	1.02 (0.29-3.60)	1.25 (0.20-7.66)	0.812
Concomitant procedure	1.99 (0.64-6.16)	1.90 (0.47-7.73)	0.368
Rupture	1.64 (0.37-7.36)	4.01 (0.54-29.67)	0.174

HR, hazard ratio; CI, confidence interval; ASA, American Society of Anesthesiologists..

This study provides a comprehensive analysis of the TEVAR used for acute complicated type B aortic dissection at our institution. The primary indication was malperfusion, with nearly 70% of the cases requiring concomitant procedures. In-hospital complications occurred in approximately 21.9%, including AKI, bleeding, and pneumonia, while the in-hospital mortality rate was 9.4%. The overall survival rates at 6 months, 1 year, 3 years, and 6 years were 78%, 75%, 65%, and 44%, respectively. Conversely, aorta-related survival rates mirrored values of 87%, 87%, 83%, and 75%, and reintervention-free survival rates were 96%, 96%, 71%, and 71% across the corresponding time intervals. Significantly, patients who exhibited favorable lumen remodeling and total FL thrombosis demonstrated markedly better survival outcomes.

Complicated type B aortic dissection, albeit uncommon, is associated with a high in-hospital mortality rate. Despite the prominence of endovascular stent grafts as a pivotal treatment option, studies consistently report in-hospital mortality rates ranging from 7% to 11%, aligning with our center’s rate of 9.4% [13,14]. Comparisons with the International Registry of Acute Aortic Dissection (IRAD) reveal similar 5-year survival rates, emphasizing the comparable nature of our 5-year aorta-related survival rate of 83%. However, the overall survival rate of 56% is relatively low, primarily attributable to non-aortic-related causes, which account for over half of the recorded mortalities [13,15].

In-hospital complications, as reported by the IRAD, align closely with our findings, with complication rates reaching as high as 22.7%, and acute renal failure accounting for one-fourth of these complications. The similarity in outcomes may be attributed to shared indications for surgery, particularly the focus on renal artery malperfusion [13,15]. Notably, the low incidence of spinal cord ischemia, a significant concern in endovascular graft placement for complicated type B dissections, is reported at 1%-3% in recent studies [15,16]. Despite its rarity, spinal cord ischemia remains a critical and devastating condition due to the multiple risks associated with endovascular graft placement in complicated type B dissections. These risks include left subclavian artery coverage, extensive aortic involvement, hypotension, and spinal artery perfusion from the FL. Consequently, shorter stent coverage or concomitant procedures to preserve the LSCA were judiciously considered and selected to minimize the risk of spinal cord ischemia.

While much emphasis has been placed on the short- and mid-term outcomes of TEVAR for complicated type B aortic dissection, the long-term durability of stent grafts and reintervention-free survival remains a subject of debate. The IRAD reports a reintervention-free survival rate of 69.4% at 5 years, reflecting a notable decline in reintervention after 3 years, consistent with our study’s plateau rate of 71% from 3 to 6 years. Our study highlights that aneurysmal degeneration, rather than stent graft-induced distal re-dissections, is the primary cause of reintervention [13,16,17].

Aortic remodeling assessments following TEVAR have employed various methodologies, including measurement of diameter, area, and volume, as well as the evaluation of FL thrombosis. Consistent with the existing literature, our study reaffirms the association between favorable lumen remodeling and superior survival rates. Notably, we introduced a refined remodeling metric that combines aortic diameter change with FL thrombosis, which has proven to be a precise predictor of survival after TEVAR. An undesirable aortic remodeling pattern emerged as the strongest predictor of poor outcomes, as demonstrated by the Cox regression analysis, which indicated a hazard ratio of 4.37. In our study, most deaths within the undesirable remodeling group were aorta-related, with the majority occurring as in-hospital deaths. These deaths primarily resulted from postoperative complications that arose as sequelae, leading to mortality shortly after the procedure. Additionally, one patient experienced progression of the FL, resulting in mesenteric ischemia. Consequently, the Kaplan-Meier survival curve for aorta-related survival closely the overall survival curve, indicating a similar pattern in their trajectories. In our study, patients with partial FL thrombosis showed significant lower survival compared to those with total FL thrombosis. This trend may be explained by the persistent presence of dynamic blood flow within the FL, which promotes the progression of dissecting aneurysms and often necessitates subsequent reintervention. These factors likely have a detrimental effect on patient survival outcomes in the later stages of the disease. Future research should delve into the organization and management of factors influencing aortic remodeling to achieve favorable patterns and consequently enhance survival outcomes [7-11,16,18-20].

However, this study has several limitations. Its retrospective and descriptive nature poses challenges in controlling the treatment settings and patient parameters. A limited number of patients, inherent to the rarity of the disease, affected the statistical power of the analysis. Moreover, heterogeneity in data recording and unclear information in medical records presented challenges in the analysis phase. Future research should advocate systematic and intensive follow-up protocols to ensure comprehensive data management.

This study provides supporting evidence that, despite the escalating prominence of endovascular treatment as a pivotal alternative to open surgery in managing acute complicated type B aortic dissection, the associated mortality rate persists at a notable level. Nevertheless, extended examination reveals that disease-specific survival rates over an extended period are deemed acceptable. Importantly, this study identified aortic remodeling patterns, a composite metric combining changes in aortic lumen diameter with the presence of FL thrombosis, as the primary factor in predicting survival outcomes after TEVAR.

The authors thank the International Affairs Office of the Faculty of Medicine, Prince of Songkla University, for editing the English language of the manuscript.

None.

The authors have nothing to disclose.

Concept and design: all authors. Analysis and interpretation: SW, PJ. Data collection: SW. Writing the article: SW, PJ. Critical revision of the article: SW, PJ. Final approval of the article: all authors. Statistical analysis: none. Obtained funding: none. Overall responsibility: SW.