The Clinical Impact of Thoracic Endovascular Aortic Repair in the Management of Thoracic Aortic Diseases

Abstract

Purpose:

Thoracic endovascular aortic repair (TEVAR) has been described to be superior to an open surgical approach, and previous studies have found superiority in TEVAR by reducing overall morbidity and mortality rates. This study aimed to describe the outcomes of TEVAR for patients with thoracic aortic disease at a high complexity.

Materials and Methods:

Descriptive study, developed by a retrospective review of a prospectively collected database. Patients aged above 18 years who underwent TEVAR between 2012 and 2022 were included. Patient demographics, perioperative data, surgical outcomes, morbidity, and mortality were described. Statistical and multivariate analyses were made. Statistical significance was reached when p values were <0.05.

Results:

A total of 66 patients were included. Male patients were 60.61% and the mean age was 69.24 years. Associated aortic diseases were aneurysms (68.18%), ulcer-related (4.55%), intramural-related hematoma (7.58%), trauma-related pathology (1.52%), and aortic dissection (30.30%). The mean hospital stay was 18.10 days, and intensive care unit was required for 98.48%. At 30 days, the mortality rate was 10.61% and the reintervention rate was 21.21%. Increased intraoperative blood loss (p=0.001) and male sex (p=0.04) showed statistical relationship with mortality. Underweight patients have 6.7 and 11.4 times more risk of complications and endoleak compared with higher body mass index values (p=0.04, 95% confidence interval [CI]=0.82–7.21) and (p=0.02, 95% CI=1.31–12.57), respectively.

Conclusion:

Thoracic endovascular aortic repair seems to be a feasible option for patients with thoracic aortic pathologies, with adequate rates of mortality and morbidity. Underweight patients seem to have an increased risk of overall morbidity and increased risk for endoleak. Further prospective studies are needed to prove our results.

Clinical Impact

Obesity and BMI are widely studied in the surgical literature. According to our study, there is a paradox regarding the outcomes of patients treated with TEVAR in terms of postoperative complications and mortality related to the body mass index. And shouldn’t be considered as a high-risk feature in terms of postoperative morbidity and mortality in this procedure.

Introduction

Thoracic aortic disease (TAD) is a heterogeneous disorder characterized by significant clinical differences between age groups, sex, and imaging findings.^1,2 Thoracic aortic disease has been commonly classified as either aneurysm or aortic dissection; given its heterogeneity, other subtypes of TAD have emerged with time.^1–5 The incidence of TAD also varies greatly between regions and per sex, with reports of 3 to 16.3 per 100 000 people/year.^1–5

This pathology is not usually treated in an emergency setting unless there is an acute complication. Whether thoracic endovascular aortic (TEVAR) repair is superior to an open surgical approach in an emergency and nonemergency setting has been debated; early observational studies have found superiority in TEVAR by reducing overall morbidity and mortality rates.^6–8 Specifically, TEVAR has demonstrated lower rates of spinal cord ischemia, respiratory failure, renal insufficiency, cardiac complications, transfusions, and paraplegia and shorter intensive care unit (ICU) stay.^9,10 Moreover, several studies report favorable long-term outcomes for TEVAR; Cheng et al ¹¹ reported an overall survival and aorta-specific survival rate at 11 years of 45.7% and 96.2%, respectively.

Initially implemented for the treatment of thoracic aortic aneurysms, TEVAR has evolved to treat a variety of aortic pathologies, including degenerative and traumatic thoracoabdominal aneurysms, penetrating atherosclerotic ulcers and intramural hematomas, aortic pseudoaneurysm, aortic aneurysm, type B aortic dissection, traumatic aortic transection, and aortoesophageal and aortobronchial fistula.^12,13 The use of TEVAR has increased exponentially since its first description in the early 1990s ⁷ ; thanks to innovations in endograft engineering design, it has become the predominant treatment for thoracic aortic repair. ¹⁴

Indications for TEVAR have changed over the years. In 2010, the American guidelines recommended endovascular stent grafting for patients with degenerative or traumatic aneurysms of the descending thoracic aorta >5.5 cm, saccular aneurysms, or postoperative pseudoaneurysms. ³ In 2012, the European guidelines recommended TEVAR for those with an aneurysm diameter exceeding 5.5 cm and rapid expansion of more than 5 mm in 6 months in patients with symptomatic thoracic aortic aneurysm. ¹⁵ The same year, the Food and Drug Administration (FDA) expanded its use to traumatic aortic transection, followed by European Association for Cardio-Thoracic Surgery (EACTS)/European Society of Cardiology (ESC/European Association of Percutaneous Cardiovascular Interventions (EAPCI) recommendation for complicated type B dissections.^16–18 Late open surgical conversion in TAD after a failed TEVAR occurs in 2% to 8% of cases.^19–22

Moreover, factors for increased risk of secondary aortic interventions after TEVAR in aortic dissections have been described, such as younger age, Marfan syndrome, acute dissection with an increased maximal aortic diameter on initial imaging, repair within 1 to 14 days, and arch procedures.^23,24 In addition, clinical risk factors predicting overall survival after TEVAR in TAD have also been reported, including arterial hypertension, renal failure, chronic obstructive pulmonary disease, atrial flutter, history of coronary bypass surgery, and associated abdominal aortic aneurysm. ²⁵

Despite progress in TEVAR, risks with endovascular manipulation persist, including stroke, spinal cord ischemia, device failure, unintentional great vessel coverage, access site complications, renal injury, endoleaks, guidewire injuries, retrograde dissections, and aortoesophageal and aortobronchial fistulas.^15,18,26 Nevertheless, TEVAR has transformed TAD prognosis. This study aimed to characterize patients with TAD managed with TEVAR at a high-complexity hospital and determine TEVAR’s postoperative outcomes from February 2012 to March 2022.

Materials and Methods

Study Population

With the institutional review board’s (IRB) approval and following Health Insurance Portability and Accountability Act (HIPAA) guidelines, a retrospective review of a prospectively collected database was conducted. All patients aged above 18 years who underwent endovascular repair of TAD between 2012 and 2022 were included. Patients with no surgical description, missing data, and follow-up <1 month were excluded. Ethical compliance with the Helsinki Declaration, current legislation on research Res. 008430-1993 and Res. 2378-2008 (Colombia), and the International Committee of Medical Journal Editors (ICMJE) were ensured under our Ethics and Research Institutional Committee (IRB) approval.

Statistical Analysis

Preoperative data included patients’ demographics and associated comorbidities. Clinical variables include preoperative renal function test, type, and size of aortic disease. Context of the procedure is either emergency or program. Operative variables include vascular approach, type of anesthesia used, intraoperative blood loss, and operative time. Outcomes analyzed were the requirement of reintervention, ICU requirement, hospital length of stay mortality, and morbidity rate at 30 days of follow-up.

Descriptive statistics of all study parameters were provided according to the nature of each variable. Data were analyzed using STATA 17 licensed version. The nature of the variables was assessed according to the Kurtosis/Skewness test. Continuous variables were summarized by means or medians and standard deviation or interquartile ranges (IQRs) according to their distribution. Categorical data were summarized by their frequency and proportion. The initial analysis includes independent associations between preoperative variables and surgical outcomes (mortality, in-hospital stay, morbidity, and presence of endoleak); for categorical variables, χ² and Fisher exact tests were performed; in cases of continuous data, 2-tailed t test, analysis of variance (ANOVA), or Mann-Whitney Wilcoxon test was performed when appropriate. For the association between continuous data, Pearson or Spearman tests were used according to variable distribution. A final multivariate analysis was performed including the independent variables that showed a relationship with outcomes in the initial analysis. Statistical significance was reached when p values were <0.05.

Body mass index (BMI) was classified as underweight if BMI <18.5 kg/m², normal weight if 18.5–24.9 kg/m², overweight if 25–29.9 kg/m², and obese if >30 kg/m²; a logistic regression was performed to evaluate the relationship with postoperative outcomes. All statistical analyses were performed with STATA 17 version.

Results

Demographic and Preoperative Characteristics

A total of 66 patients with TAD who underwent TEVAR were included. There were no patients with prior abdominal aortic aneurysm repair or prior ascending or arch repair. Male patients comprised the majority of the population (60.61%, n=40). The mean age was 69.24±10.90 years. The mean BMI was 25.49±4.03 kg/m²; underweight patients (>18.5 kg/m²) correspond to 6.06% (n=4), normal weight (18.5–24.9 kg/m²) to 34.85% (n=23), overweight (25–29.9 kg/m²) to 42.2% (n=28), and obese (>30 kg/m²) to 16.67% (n=11). Comorbidities were analyzed: type 2 diabetes mellitus was observed in 16.67% (n=11) of the patients, history of arterial hypertension was present in 84.85% (n=56) of the population, and coronary arterial disease was evidenced in 66.67% (n=44) of the patients. Smoking habit was reported in 39.39% (n=26). Associated aortic disease was classified into aortic aneurysms (68.18%, n=45), ulcer-related (4.55%, n=3), intramural-related hematoma (7.58%, n=5), trauma-related pathology (1.52%, n=1), and aortic dissection (30.30%, n=20).

The mean size of the aneurysms was 55.72±19.98 mm, and for dissections, the mean size was 51.42±18.62 mm. For the last one, the timing was evaluated; acute dissection was found in most cases (90%). Preoperative renal function was assessed with creatinine values and glomerular filtration rate (GFR) according to the Cockcroft-Gault formula. The median creatinine value was 1.02 mg/dL (IQR=0.8–1.8), and the median GFR was 67.5 mL/min/m² (IQR=47–80) (Table 1).

Table 1.

Preoperative Characteristics.

Variable	Result
Sex, % (n)
Male	60.61 (40)
Female	39.39 (26)
Age mean (SD)	69.24 (10.90)
Body mass index (SD)	25.49 (4.03)
Comorbidities, % (n)
Type 2 diabetes mellitus	16.67(11)
Arterial hypertension	84.85 (56)
Chronic obstructive pulmonary disease	24.24 (16)
Smoking history	39.39 (26)
Coronary arterial disease	66.67 (44)
Renal function tests, median (IQR)
Creatinine	1.02 (0.8–1.8)
Glomerular filtration rate	67.5 (47–80)
Type of aortic disease, % (n)
Aneurysm	68.18 (45)
Ulcer-related	4.55 (3)
Intramural hematoma-related	7.58 (5)
Trauma-related	1.52 (1)
Dissection	30.30 (20)
Acute dissection	90 (18)
Chronic dissection	10 (2)
Size of the defect, mean (SD)
Dissection	51.42 (18.62)
Aneurysm	55.79 (19.98)

Abbreviation: IQR, interquartile range.

Operative Characteristics

In most cases, patients were treated as an emergency procedure (51.52%, n=34) as these patients were all symptomatic. The vascular approach was analyzed; in the majority of the population, femoral access was preferred (96.97%, n= 64); in 2 patients, the vascular approach was infraclavicular due to a carotid-subclavian bypass. In those patients with a high risk of spinal cord ischemia, a peridural catheter was placed for drainage. In our series, this was necessary for 1 patient in whom perioperative paraplegia was evidenced. General anesthesia was preferred in most cases (93.94%, n=62). The median intraoperative blood loss was 200 mL (IQR=100–290). Intraoperative bleeding is considered multifactorial, especially considering the patients’ comorbidities (diabetes mellitus, arterial hypertension, etc). The surgical technique varied in only 2 cases, which does not allow to statistically evaluate these changes’ relationship with bleeding. In 100% of the cases, vascular bypass was not required; management was percutaneous with percutaneous closure devices, and vascular site complication rate was 0%. The median operative time was 146.02 minutes (IQR=125–146.02) (Table 2).

Table 2.

Operative Characteristics.

Variable	Result
Context of the procedure, % (n)
Emergency	51.52 (34)
Program	48.48 (32)
Vascular approach %(n)
Femoral	96.97 (64)
Infraclavicular	3.03 (2)
Type of anesthesia
Local	4.5 (3)
Spinal	0 (0)
Epidural	1.52 (1)
General	93.94 (64)
Intraoperative blood loss median (IQR)	200 (100– 290)
Operative time median (IQR)	146.02 (125–146.02)

Abbreviation: IQR, interquartile range.

Outcomes

The mean hospital length of stay was 18.10 days (IQR=10–23). Intensive care unit was required for 98.48% (n=65) of the patients with a median of 3 days (IQR=2–8). At 30 days of follow-up, the mortality rate was 10.61% (n=7) secondary to acute coronary syndrome (n=4), pneumonia with ventilatory failure and sepsis (n=2), and 1 patient due to endoleak. The reintervention rate was 21.21% (n=14), and the readmission rate was 0%. Regarding mortality, it was due to acute coronary syndrome (n=4), pneumonia with ventilatory failure and sepsis (n=2), and 1 patient secondary due to endoleak. Morbidity was observed in 15.15% (n=10) of the patients. One-year survival rate after the TEVAR procedure was 77.7% (n=51).

The endoleak rate was 10.61% (n=7), and in most cases it was type II (4 patients); aortoesophageal fístula was observed in 1 case. Vascular access thrombosis was observed in 3.03% (n=2) of the patients. One patient has perioperative paraplegia and was treated with cerebrospinal fluid drainage successfully (Table 3). Patients were evaluated with a median follow-up of 40 days (IQR=30–170).

Table 3.

Postoperative Outcomes.

Variable	Result
Outcomes
In-hospital stay, median (IQR)	15 (10–23)
Intensive care unit stay, median (IQR)	3(2–8)
Intensive care unit requirement, % (n)	98.48 (65)
Reintervention rate, % (n)	21.21 (14)
Mortality rate, % (n)	10.61 (7)
Morbidity rate, % (n)	15.15 (10)
Type of morbidity, % (n)
Aortoesophageal Fistula, % (n)	1.52 (1)
Parestesia, %(n)	0 (0)
Vascular access thrombosis, % (n)	3.03 (2)
Paraplegia	1.52 (1)
Endoleak, % (n)	10.61 (7)
Type IA	1
Type IB	1
Type II	4
Type III	1

Abbreviation: IQR, interquartile range.

Statistical Analysis

Associations between preoperative and surgical variables with postoperative outcomes were evaluated.

Mortality

T-test mean comparison was performed. Patients in the mortality group showed an increased age compared with the non-mortality group (mean 75.71 vs 68.47 years, respectively), with statistical significance (p=0.04). Types of aortic thoracic disease were assessed; however, there is no statistical relationship with mortality. Aneurysm size has a slight difference between groups, however with no statistical significance (mortality group 50.25 vs non-mortality group 56.39, p=0.56). Renal function was evaluated as well; using the Mann-Whitney Wilcoxon test, there is no statistical relationship between creatinine value or rate of glomerular filtration and mortality (p=0.6 and p=0.8, respectively). Intraoperative blood loss and operative time were evaluated; however, there is no relationship between these variables and mortality (p=0.3 and p=0.4, respectively). The presence of complications does not show a statistical relationship with mortality (p=0.9) (Table 4).

Table 4.

Independent Statistical Analysis.

Variable/outcome	Mortality	Hospital length of stay	Morbidity	Endoleak
Gender	0.06 (3.36)	0.7(0.4)	0.03 (4.61)	0.06 (3.36)
Age	0.04	0.1	0.50	0.69
BMI	0.63	0.77	0.00	0.00
Type 2 diabetes mellitus	0.21 (1.56)	0.09 (1.59)	0.75 (0.09)	0.85 (0.03)
Arterial hypertension	0.23 (1.39)	0.07 (1.66)	0.64 (0.21)	0.94 (0.00)
COPD	0.77 (0.07)	0.45 (1.02)	0.73 (0.11)	0.51 (0.42)
Smoking history	0.1 (2.06)	0.96 (0.52)	0.96 (0.00)	0.84 (0.03)
Coronary arterial disease	0.15 (1.55)	0.16 (1.41)	0.80 (0.05)	0.77 (0.07)
Creatinine	0.60	0.75	0.32	0.54
GFR	0.80	0.72	0.25	0.11
Aneurysm	0.29 (1.10)	0.16 (1.42)	0.38 (0.75)	0.84 (0.03)
Ulcer	0.54 (0.37)	0.01 (2.17)	0.45 (0.56)	0.54 (0.37)
Intramural hematoma	0.42 (0.64)	0.00 (2.33)	0.1 (2.59)	0.47 (0.50)
Trauma	0.72 (0.12)	0.99 (0.37)	0.67 (0.18)	0.72 (0.12)
Dissection	0.32 (0.95)	0.33 (1.15)	0.98 (0.00)	0.44 (0.58)
Size of aneurysm	0.56	0.13	0.81	0.81
Type of dissection	0.55 (1.16)	0.47 (1.02)	0.61 (0.98)	0.63 (0.91)
Context of the procedure	0.26 (1.24)	0.31 (1.18)	0.03 (4.68)	0.1 (1.65)
Vascular approach	0.62 (0.24)	0.96 (0.53)	0.54 (0.36)	0.62 (0.24)
Intraoperative blood loss	0.30	0.02	0.19	0.58
Operative time	0.42	0.64	0.23	0.62

Categorical data are expressed as p value (χ²); continuous data as p value (F value in the case of ANOVA test).

Abbreviations: ANOVA, analysis of variance; BMI, body mass index; COPD, chronic obstructive pulmonary disease; GFR, glomerular filtration rate.

Hospital length of stay

Analysis of variance or Spearman and Pearson correlations were performed according to the distribution of the data. Sex was evaluated; however, there was a statistical relationship with in-hospital stay (p=0.7). Comorbidities were assessed as well; nevertheless, there was no relationship between these variables and in-hospital stay (Table 4). The presence of aortic ulcer and intramural aortic hematoma showed a relationship with increase in hospital stay with p=0.01 and p=0.00, respectively. Operative variables were evaluated. There was a negative correlation between intraoperative blood loss and in-hospital stay which was statistically significant (rs=−0.27, p=0.02) (Table 4).

Morbidity

Sex showed a statistical relationship with the presence of any complication; female patients tended to have more complications compared with male patients (χ²=4.62, p=0.03). Context of the procedure was also evaluated: TEVAR under emergency conditions tended to have any complications compared with the programmed one (χ²=4.68, p=0.03). Patients with decreased BMI had significant differences in morbidity groups, with statistical significance (non-morbidity group mean 26.08 kg/m² vs morbidity group mean 22.19 kg/m², p=0.00). In terms of aneurysm size, there was a slight difference between morbidity groups; nevertheless, there was no statistical relationship (non-morbidity group mean 56.11 mm vs morbidity group mean 54.07, p=0.8). Renal function tests were evaluated as well. Creatinine value and GFR were not related to the presence of morbidities (p=0.32 and p=0.25, respectively). Operative characteristics were evaluated. Intraoperative blood loss and operative time did not show a statistically significant relationship with the presence of any complication (p=0.2 and p=0.1) (Table 4).

Endoleak

Mean comparisons were performed between patients with or without endoleak. Body mass index showed a statistical difference, with lower BMI in patients who present endoleak (endoleak group mean 21.68 kg/m² vs non-endoleak group mean 25.94, p=0.00). Aneurysm sizes varied between endoleak/non-endoleak groups; however, they failed to reach statistical significance (56 vs 53.5 mm, respectively; p=0.8). Renal function tests were also assessed; there was no relationship between creatinine levels of GFR with the presence of endoleak (p=0.51 and p=0.12, respectively). Operative variables such as intraoperative blood loss and operative time did not show any statistical relationship with the presence of endoleak (p=0.51 and p=0.62, respectively) (Table 4).

Multivariate analysis

All of the independent associations made in the initial analysis were included in a multivariate analysis to prove the related factors for each outcome. Increased intraoperative blood loss showed a statistical relationship with mortality (p=0.001, 95% CI=0.0–0.07); also, male sex was related to mortality with a p value of 0.04 (95% CI=0.38–0.7).

In terms of hospital stay, just the presence of intramural hematoma showed a statistical relationship with an increase in hospital stay, with a p value of 0.03 (95% CI=0.26–0.95). For morbidity, low BMI was related not only to the presence of any complication (p=0.05, 95% CI=0.04–0.9) but also to the presence of endoleak with a statistically significant value (p=0.01, 95% CI=0.04–0.6) (Table 4).

BMI

Outcomes such as mortality, reintervention, hospital length of stay, ICU stay, presence of any complication, and presence of endoleak were evaluated according to BMI. Patients with underweight have a 6.7 times more risk of the presence of any complications compared with higher BMI values with statistical significance (p=0.04, 95% CI=0.82–7.21). Patients with a BMI <18.5 kg/m² have 11.4 times more risk to present endoleak as postoperative complications compared with higher BMI values (p=0.02, 95% CI=1.31–12.57). Reintervention was not related to BMI values. Differences between in-hospital stay and ICU stay were slight between underweight, normal weight, overweight, or obese patients, but no statistical differences were found. Mortality did not show any relationship with nutritional status according to BMI value (Table 5).

Table 5.

BMI and Postoperative Outcomes.

Variable/outcome	Mortality	Complications	Endoleak	Reintervention	Hospital length of stay	ICU stay
BMI < 18.6 kg/m2	0.35	0.04*	0.02*	0.17	0.47	0.45
BMI 18.5–24.9 kg/m2	0.25	0.28	0.20	0.24	0.78	0.86
BMI 25–29.9 kg/m2	0.41	0.13	0.1	0.56	0.31	0.72
BMI >30 kg/m2	0.85	0.54	0.85	0.59	0.60	0.46

Abbreviations: BMI, body mass index; ICU, intensive care unit.

^*Statistically significant value.

Discussion

Minimally invasive procedures have appeared as a feasible option with lower morbidity and mortality rates. The continuous technological advances in grafts and equipment promoted an increase in its implementation. Due to lower rates of morbidity, mortality, length of stay, and blood loss compared with the open approach, TEVAR is the gold standard approach to treating thoracic aneurysms.^3,12,16–18 Some studies have shown TEVAR’s safety and feasibility, with acceptable complication rates between 10% and 22% in Fossaceca et al and Hellgren et al^27,28 akin to our results with a general morbidity rate of 15.15%. Fossaceca et al ²⁷ reported a mortality rate of around 7.5% in 53 patients who underwent TEVAR procedure, which is similar to our results of our population’s 30-day mortality rate of 10.61%.

In our study, endoleak was present in 7 patients with a total rate of 10.61% (1 type IA, 1 type IB, 4 type II, and 1 type III), and it was found that patients who presented endoleak had lower BMI with a significant difference (p=0.00), where the mean BMI of the endoleak group was 21.68 kg/m² and that of the non-endoleak group was 25.94 kg/m². This is similar to what has been reported in the literature, where it is evident that endoleaks after TEVAR have been reported in 15% to 30% of the previous series. ²⁹ However, with regard to BMI, there is scarce literature evaluating the association between BMI and the presence of endoleaks, which makes it difficult to compare this statement.

On the other hand, in our study, aneurysm sizes differed between patients with and without endoleak without reaching a statistical difference (56 vs 53.5 mm, respectively; p=0.8). This is similar to that found in the previous series, ³⁰ where the diameter (p=0.39), the length (p=0.22), and the diameter of the proximal neck (p=0.35) of the descending thoracic aortic aneurysm did not show a significant difference between the groups with or without endoleak, unlike the length of the proximal neck that was smaller in the group with endoleaks (p=0.04). ³⁰

Similarly, no significant relationship was found between renal function tests and creatinine levels of GFR with the presence of endoleaks (p=0.51 and p=0.12, respectively). In this regard, Alsac et al ³⁰ evaluated the incidence of endoleaks after TEVAR and found that there is no significant difference between the development of endoleak and the presence of comorbidities such as renal failure (p=0.82), where the non-endoleak group presented a rate of 13% and the endoleak group a rate of 15.4%, akin to elevated serum creatinine, which also did not reach a significant difference (p=0.2). ³⁰ In addition, some operative variables such as intraoperative blood loss (p=0.51) and operative time (p=0.62) did not reach a significant value to show any association with the presence of endoleaks in our study; nevertheless, evidence regarding this topic is scarce, thus supporting the idea that there is no significant relationship between these variables.

Regarding the impact of sex, 39.1% (n=26) were female patients, similar to Dias et al. ³¹ This is evidence that women are underrepresented in the vast majority of the studies. On the other hand, no significant association was found with respect to sex and length of hospital stay (p=0.7), which is different from previous studies such as that published by Jackson et al, ³² which showed that although the length of hospital stay in patients requiring ICU was not significantly different for women compared with men (p=0.515), there was a significant difference in the overall length of hospital stay for women (9.0±16.2 days) compared with men (4.7±5.8 days) (p<0.001). We found that this variable was related to extensive procedures. Also, it was related to more intraoperative bleeding and was also related to a longer recovery time.

According to what was published in the Global Registry for Endovascular Aortic Treatment (GREAT) study, ³³ a 2.8% rate of access-related complications was found, where women represented a 4.7% rate of complications, this being lower in men (1.8%). In our study, we found that sex showed a significant relationship with the presence of any complication, thus evidencing that women had more complications than men (χ²=4.62, p=0.03). Finally, in the multivariate analysis, male sex showed a higher relationship with respect to mortality (p=0.04, 95% CI=0.38–0.7), this being different from what was found in the literature where according to Lareyre et al ³⁴ no significant difference was found in overall mortality at a 2.2-year follow-up in women and men (26.9 vs 27.6%, p=0.58). Along the same lines, Deery et al ³⁵ found a higher 30-day and 1-year mortality for women.

Regarding mortality, it was found that there was no difference in associating intramural hematoma with increased mortality (p=0.42). Data regarding mortality in patients who underwent endovascular therapy for the management of intramural hematoma are scarce. According to what was found, the results of our study are similar to the literature, where it is possible to demonstrate that the mortality rate in these patients is 4.6% in the acute phase and 7.1% in a 3-year follow-up. ³⁶ In addition, the main limitation reported of the acute phase of intramural hematoma treatment is the secondary development of endoleaks and intimal ruptures/pseudoaneurysm formation, different from what was found in our research, where no association was found between the use of TEVAR for the treatment of intramural hematoma and consequent endoleaks (p=0.47). ³⁶

Previous researchers evaluated the postoperative outcomes of mortality and morbidity and their relationship with renal function. ³⁷ Okada et al ³⁸ reported that decreased level of GFR is an independent risk factor for 30-day mortality and also state that length of stay, reinterventions, and overall adverse events were worse in decreased renal function. ³⁸ Nevertheless, some authors describe that renal function is not associated with worse outcomes.^39,40 Our study shows no relationship between preoperative creatinine levels or GFR with any postoperative complication (morbidity. mortality, endoleak, length of stay).

The impact of BMI and its relationship with postoperative outcomes have been widely described in the worldwide surgical literature.^41–44 In most cases, for all surgical procedures, obesity is related to increased rates of postoperative morbidity and mortality according to some studies^41–43; nevertheless, Lavie et al ⁴⁵ in 2009, described the “Obesity paradox” in which patients with increased BMI show better prognosis for vascular pathologies such as arterial hypertension, among others. Authors such as Galal et al ⁴⁶ showed increased long-term mortality in underweight patients compared with overweight and obese patients who underwent major vascular surgery. ⁴⁶ In addition, Naazie et al ⁴⁴ identified that obese patients have the same 30-day and 1-year risk of mortality compared with normal-weight population. Also, it is evidenced that underweight patients were associated with higher postoperative complication (odds ratio=1.61, 95% CI=1.05–2.45, p=0.002), which is comparable with our results in which underweight population have a 6.7 times more risk of the presence of any complications compared with higher BMI patients with statistical significance (p=0.04, 95% CI=0.82–7.21). ⁴⁴

Limitations of this study include the retrospective nature and the absence of a comparative open surgery group and the relatively small sample size. Strengths include the long-term follow-up and that our study increases the evidence of TEVAR procedure in the Latin-American population and in contrast to some studies supports the feasibility and safety of this endovascular procedure in obese patients. Also, it suggests that in emergency patients, creatinine value is not related to increased morbidity, in-hospital length of stay, or mortality and could be performed independently of the renal function at the moment of the emergency.

Conclusion

The TEVAR procedure is a safe and feasible approach for patients with aortic pathologies, with acceptable rates of morbidity and mortality. According to our data, preoperative renal function is not related to postoperative outcomes. However, underweight patients have a 6.7-fold increased risk of overall morbidity and an 11-fold increased risk to present endoleak; obesity is not associated with poor postoperative outcomes. Further prospective studies are needed to prove our results.