Nationwide analysis of EVAR explantation outcomes in Korea: A comprehensive dataset study

Abstract

Objective

Endovascular aortic aneurysm repair (EVAR) is the primary treatment for abdominal aortic aneurysms, constituting 70%–80% of interventions. Despite initial benefits, long-term studies show increased mortality. Using nationwide data, this study assesses outcomes of EVAR, open aortic repair (OAR), and EVAR explantation (EE) in Korea, while exploring characteristics of late open conversion, including the rising EE incidence.

Methods

Employing the National Health Insurance Service database, covering health-related data for nearly 50 million Koreans, the study spanned from 2002 to 2020. Patients with AAA diagnoses (I71.3 or I71.4) were categorized into OAR, EVAR, and EE groups based on procedural codes. Statistical analyses, including t-tests, Fisher’s exact tests, Cox proportional hazard models, and multivariate Cox regression, assessed baseline characteristics, mortality risks, and factors within the EE group.

Results

The analysis encompassed 26,195 patients, with 66.19% in the EVAR group, 31.87% in the OAR group, and 1.94% in the EE group. EVAR cases steadily increased from 2002 to 2018. Survival rates favored EVAR, followed by OAR and EE. 30-day survival was lower in EE than EVAR. Multivariate analysis for EE revealed no risk factors for 30-days survival but identified age, chronic kidney disease, high Charlson Comorbidity Index scores, and less than 6 months since EVAR as risk factors for overall mortality.

Conclusion

Rising EE trends with increased EVAR adoption, particularly evident in the Korean dataset, underscore inferior outcomes. This highlights the critical need for strategic initial treatment decisions and timely interventions to enhance overall results and mitigate the unfavorable EE incidence.

Keywords

Abdominal aortic aneurysm endovascular aneurysm repair complication

Introduction

In recent decades, endovascular aortic aneurysm repair (EVAR) has emerged as the standard procedure for abdominal aortic aneurysms (AAA), constituting 70% to 80% of aneurysm repairs.¹ Numerous prospective randomized clinical trials consistently demonstrated EVAR’s initial superiority over conventional open aortic repair (OAR) in elective AAA treatment, resulting in reduced mortality and morbidity during perioperative and short-term follow-up periods.² However, despite these early benefits, long-term follow-up studies have uncovered a loss of this advantage, revealing increased overall mortality and, notably, higher aneurysm-related morbidity and mortality four to 8 years post-intervention.³

Recent literature reports the incidence of endoleaks, particularly type I (5%) and type II (20%–40%).⁴ Endoleaks after EVAR significantly increase the risk of potential aneurysmal rupture.⁵ Additionally, complications such as aortic graft infections add complexity to EVAR outcomes.⁶ While short-term benefits over OAR are apparent, long-term concerns arise due to reduced durability, increased reinterventions, and the potential for rupture.³ Current recommendations emphasize considering endovascular or laparoscopic approaches for addressing endoleak, particularly when the sac increases by more than 10 mm.⁵ If these interventions prove unsuccessful, pursuing late open conversion (LOC) is recommended, in line with the reported overall LOC incidence of 5.3% in recent meta-analyses.⁷ The purpose of this study is to utilize data from the National Health Insurance Service (NHIS) to compare the outcomes of EVAR, OAR, and EVAR explantation (EE) in Korea. Additionally, the study aims to investigate the characteristics of LOC, including EE, which has been on the rise in Korea.

Methods

The study was approved by the IRB (Institutional Review Board) of Korea University Guro hospital (No. K2022-0122-001). This research, based on anonymized national data, was exempted by the IRB from obtaining consent forms from individual patients.

Study population

We utilized the database provided by the NHIS which includes health care utilization, health screening, socio-demographic variables, and mortality of approximately 50 million Koreans. The NHIS is a government insurance system covering nearly 97% of the entire Korean population. Approximately 3% of individuals fall under the classification of Medical Aid (MA) and were not included in the NHIS and our study.⁸ Once the research receives final approval from the National Health Insurance Service Research Review Committee, access to the data is granted through the National Health Insurance Service’s Big Data Sharing Service (https://nhiss.nhis.or.kr).

Study design and diagnostic code

Our study focused on patients registered in Korea’s NHIS from 2002 to 2020. We finalized our cohort by selecting patients with the AAA diagnosis code from this database. The diagnostic codes were based on the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10 code). The diagnostic codes for AAA were included with I71.3 (AAA, ruptured) or I71.4 (AAA, without mention of rupture). Then, we divided this cohort into OAR group, EVAR group, and EE group. The cohort was categorized into three groups based on procedural codes: the OAR group comprised patients with OAR codes (O0223, O0224, or O2034) without prior EVAR codes, the EVAR group included patients with EVAR codes (M6611 or M6612) without subsequent OAR codes, and the EE group was defined that OAR codes occurred following the initial EVAR code. The comorbidities and Charlson Comorbidity Index (CCI) were identified based on ICD-10 codes in these patients; hypertension, dyslipidemia, diabetes mellitus, ischemic heart disease, cerebrovascular disease, chronic kidney disease, congestive heart failure, and chronic obstructive pulmonary disease. The calculation method for CCI was adopted as described in a previously referenced study.⁹ The procedure codes and diagnostic codes for comorbidities mentioned above are detailed in Supplementary Table 1.

Statistics

In baseline characteristics (Table 1), continuous values for each group were represented by their median, and comparisons between groups were conducted using the ANOVA. Categorical variables were assessed using Fisher’s exact test or the chi-square test. The Cox proportional hazard model was utilized to explore the relationship between each group and mortality. Through this method, we evaluated the risk of mortality with hazard ratios adjusted by age, sex, and CCI (Table 2). Multivariate analysis using Cox regression was conducted within the EE group to identify risk factors associating both overall and 30-day mortality. Variables with a univariable p-value of 0.1 or less were included in the multivariate analysis (Table 3). The Kaplan-Meier method was employed to generate survival curves for each group and facilitate their comparative analysis (Figure 2).

Table 1.

Baseline Characteristics of AAA Patients in EVAR, OAR, and EE groups.

	EVAR (n = 17338)	OAR (n = 8349)	EE (n = 508)	p Value
Male	13935 (80.37)	6773 (81.12)	423 (83.27)	0.1175
Age, y	71.64±10.56	68.81±10.98	71.80±10.99	<.001
Age ≥75^a	7640 (44.07)	2708 (32.44)	241 (47.44)	<.001
Hypertension	14772 (85.2)	7015 (84.02)	458 (90.16)	<.001
Dyslipidemia	12015 (69.3)	5101 (61.1)	382 (75.2)	<.001
DM	6550 (67.78)	3090 (37.01)	191 (37.6)	0.4921
IHD	6675 (38.5)	3121 (37.38)	206 (40.55)	0.1217
CVD	4594 (26.5)	1883 (22.55)	117 (23.03)	<.001
CKD	421 (2.43)	110 (1.32)	32 (6.3)	<.001
CHF	2809 (16.2)	1049 (12.56)	118 (23.23)	<.001
COPD	2392 (13.8)	1174 (14.06)	62 (12.2)	0.4706
CCI	5.06±0.54	4.67±0.91	5.48±0.79	<.001
CCI ≥3^b	14295 (82.45)	6695 (80.19)	436 (85.83)	<.001

EVAR, endovascular aortic aneurysm repair; OAR, open aortic repair; EE, EVAR explantation; DM, Diabetes mellitus; IHD, Ischemic heart disease; CVD, Cerebrovascular disease; CKD, Chronic kidney disease; CHF, Congestive Heart failure; COPD, Chronic obstructive pulmonary disease; CCI, Charlson Comorbidity Index. Data are expressed as numbers (%) and means ± SD.

^aAge ≥75 means the number of patients who are 75 years of older.

^bCCI ≥3 means the number of patients with a CCI score of 3 or higher.

Table 2.

Comparisons of overall and 30-day mortality among AAA patients in EVAR, OAR, and EE groups.

	Unadjusted		Adjusted by Age, Sex, CCI
	Hazard Ratio	p value	Hazard Ratio	p-value
Overall mortality
EE vs OAR (Ref. OAR)	1.682 (1.476–1.917)	<.0001	1.325 (1.161–1.511)	<.0001
EE vs EVAR (Ref. EVAR)	1.47 (1.294–1.67)	<.0001	1.397 (1.229–1.587)	<.0001
OAR vs EVAR (Ref. EVAR)	0.792 (0.761–0.823)	<.0001	0.936 (0.900–0.974)	<.0001
30-days mortality
EE vs OAR (Ref. OAR)	1.115 (0.848–1.465)	0.4366	0.967 (0.735–1.273)	0.8128
EE vs EVAR (Ref. EVAR)	2.106 (1.605–2.765)	<.0001	2.099 (1.599–2.756)	<.0001
OAR vs EVAR (Ref. EVAR)	1.885 (0.483–0.583)	<.0001	2.114 (0.430–0.521)	<.0001

AAA, abdominal aortic aneurysm; EVAR, endovascular aortic aneurysm repair; OAR, open aortic repair; EE, EVAR explantation; CCI, Charlson Comorbidity Index. Data are expressed as hazard ratio (95% confidence interval).

Table 3.

Analysis of uni- and multi-variable risk factors for overall mortality and 30-days mortality in the EVAR explantation group.

	Overall mortality		30-days mortality
	Crude HR	Adjusted HR	Crude HR	Adjusted HR
Male	1.12 (0.80–1.57)		0.87 (0.44–1.72)
Age	1.04 (1.02–1.05)^d	1.04 (1.02–1.05)^d	1.01 (0.98–1.03)
Age ≥75^a	1.77 (1.37–2.29)^d		1.15 (0.68–1.95)
Hypertension	1.56 (0.95–2.55)		1.93 (0.60–6.18)
Dyslipidemia	1.06 (0.79–1.41)		1.09 (0.59–2.03)
DM	1.25 (0.97–1.61)		1.31 (0.77–2.23)
IHD	1.21 (0.93–1.55)		0.83 (0.48–1.44)
CVD	1.41 (1.06–1.88)^d	1.06 (0.78–1.44)	1.38 (0.77–2.46)
CKD	2.26 (1.48–3.45)^d	1.83 (1.18–2.85)^d	1.52 (0.61–3.82)
CHF	1.37 (1.03–1.83)^d	1.07 (0.78–0.15)	0.72 (0.36–1.42)
COPD	1.44 (1.02–2.04)^d	1.22 (0.86–1.73)	1.24 (0.59–2.63)
CCI	1.12 (1.08–1.16)^d	1.08 (1.03–1.13)^d	1.07 (0.99–1.17)
CCI ≥3^b	4.57 (1.70–12.30)^d		2.71 (0.38–19.62)
Duration ≥ 6M^c	1.36 (1.04–1.78)^d	1.38 (1.05–1.82)^d	1.09 (0.60–1.98)

EVAR, endovascular aortic aneurysm repair; HR, hazard ratio; DM, Diabetes mellitus; IHD, Ischemic heart disease; CVD, Cerebrovascular disease; CKD, Chronic kidney disease; CHF, Congestive Heart failure; COPD, Chronic obstructive pulmonary disease; CCI, Charlson Comorbidity Index. The crude HR reflects the univariable analysis of risk factors, while the adjusted HR takes into account the multivariable analysis. Data are expressed as hazard ratio (95% confidence interval).

^aAge ≥75 means the number of patients who are 75 years of older.

^bCCI ≥3 means the number of patients with a CCI score of 3 or higher.

^cDuration ≥ 6M means patients with a duration of 6 months or more from EVAR to EVAR explantation.

^dmeans p < 0.05.

Results

A total of 26,195 patients were included in the analysis, with 17,338 (66.19%) in the EVAR group, 8349 (31.87%) in the OAR group, and 508 (1.94%) in the EE group. Among the three groups, EVAR exhibited the largest distribution, steadily increasing from 2002 to 2018, with a slight decrease from 2018 to 2020. OAR demonstrated a consistent annual increase after 2022, comprising only 725 cases in 2020, accounting for approximately 40% of the 1850 EVAR cases during the same period. EE has shown a steady rise every year since 2005, starting with one case in 2005 and reaching 95 cases in 2020. As of 2020, among patients undergoing treatment for AAA, EVAR represented 69.2% (1850 patients), OAR 27.2% (725), and EE 3.6% (95) (Figure 1).

Figure 1.

Annual Trends in Endovascular Aortic Aneurysm Repair (EVAR), Open Aortic Repair (OAR), and EVAR Explantation (EE) in Korea.

The mean age was 70.7 ± 10.8 years, with the OAR group showing a slightly younger average age of 68.8 ± 11.0 years. Comorbidities such as diabetes mellitus, ischemic heart disease, and COPD did not demonstrate significant differences among the three groups. Cerebrovascular disease showed distributions of 26.5% in the EVAR group, 23.0% in the EE group, and 22.6% in the OAR group, with the highest prevalence observed within the EVAR group. Additionally, hypertension, hyperlipidemia, chronic kidney disease, and congestive heart failure were more prevalent in the EE group compared to the other groups. Assessing the Charlson Comorbidity Index (CCI) scores, the proportion of scores equal to or greater than 3 was highest in the EE group at 85.8%, followed by 82.6% in the EVAR group and 80.2% in the OAR group. The median time from EVAR to EE was 1246.75 days (Table 1).

Overall survival rates showed EVAR as the highest, followed by the OAR group and then the EE group. However, around 18 months into observation, the OAR group demonstrated the highest survival rates (Figure 2). 30-day survival was 94.7% in the EVAR group, 90.3% in the OAR group, and 89.1% in the EE group. One-year survival was 83.5% in the EVAR group, 82.0% in the OAR group, and 73.2% in the EE group. Three-year survival was 69.8% for EVAR group, 73.5% for OAR group, and 57.2% for EE group. The EE group had the lowest survival across all intervals.

Figure 2.

Survival probabilities and survival numbers over time according to Endovascular Aortic Aneurysm Repair (EVAR), Open Aortic Repair (OAR), and EVAR Explantation (EE) in Korea.

We calculated the risk of mortality for each group after adjusting for age, sex, and CCI (Table 2). While 30-days survival did not differ between the EE and OAR groups, the EE group was 2.10 times higher than the EVAR group, with statistical significance. The OAR group was 1.89 times higher than the EVAR group. For overall mortality, the EE group was 1.33 times higher than OAR and 1.40 times higher than EVAR (Table 2). We performed a multivariate analysis with cox regression to determine the impact on survival within the EE group. There were no risk factors in multivariate analysis for 30-days survival. However, age, chronic kidney disease, CCI score, and less than 6 months since EVAR were risk factors for overall mortality (Table 3).

Discussion

With over 97% of Koreans in Korean territories covered by NHIS for medical services, the NHIS-based analysis is representative of Korea.⁸ This study provides comprehensive initial data on EE in Korea, with over 1000 annual EVAR procedures (Figure 1), representing early findings in Asia. Among 26,195 individuals treated for AAA in Korea, the 30-days mortality rates for EVAR, OAR, and EE groups were 5.3%, 9.7%, and 10.9%, respectively, with the difference becoming more pronounced over 1-year and 3-year survival periods. Overall mortality, not procedure-related mortality, was compared in this study. Due to a lack of detailed EE cause analysis, direct comparisons with other studies may not be appropriate. Nonetheless, the study indicated unfavorable outcomes for EE compared to other treatments, with 30-days mortality similar to reported results in the US.¹⁰

Outcomes of EVAR explantation

The main cause of aneurysmal rupture after EVAR is often attributed to untreated or undetected type 1A endoleak.¹¹ Some studies have reported outcomes for managing endoleak after EVAR, utilizing approaches such as Fenestrated EVAR or surgical EE.¹² In cases of EVAR failure, which include EE, LOC may become the only practical option, indicating that LOC is necessary when other interventions are not feasible or effective for addressing a failing EVAR.¹³ The morbidity and mortality rates for LOC are reported to be higher than standard procedures, primarily due to additional technical challenges encountered during surgery, particularly in terms of access, aortic cross-clamping, and stent graft removal.¹⁴ In a Canadian study, elective EE was associated with lower mortality rates than treatment for infection or rupture, emphasizing the recommendation to consider surgery for individuals with an indication for EE before the onset of symptoms or rupture.¹⁵

Studies on late open conversion

Most studies on LOC after EVAR have primarily consisted of single-center case series. However, recently, two meta-analyses have reported, consolidating findings from these individual studies. Both groups highlighted higher perioperative mortality in cases of infection or rupture. The Italian group reported a 30-days mortality of 19.4% for infection and 32.6% for rupture, while the Canadian group reported 15% for infection and 33% for rupture.^16,17 Based on data from the US National Surgical Quality Improvement Program, the 30-day mortality rates for 32,164 individuals treated for AAA were reported as 1.7% for EVAR, 4.2% for OAR, and 10.0% for LOC.¹⁰ In a recent multicenter study in Europe, the 30-day mortality for LOC in noninfected EVAR failure was reported as 6.1% for elective cases and 28.3% for rupture.¹⁸

In this study, EVAR was performed, and, on average, EE was undertaken 3.41 years later. The EE group exhibited a higher incidence of various morbidities. The authors considered that comorbidities might potentially serve as risk factors for both EE and mortality, and they entertained the idea that the condition of AAA patients could have changed over the 3-year period. Due to elevated comorbidity and inherent surgical risks, outcomes for EE are less favorable, with several studies indicating higher mortality rates, particularly in cases of rupture.¹⁸ Efforts should be made to minimize the occurrence of EE, and if necessary, its performance at an appropriate time becomes crucial. Recent guidelines have favored OAR if the patient’s condition allows.¹⁹ Selecting the appropriate modality for the treatment of AAA patients and determining the optimal timing for intervention are crucial.

This study has its limitations. It focused on patients who underwent EE, whether partial or complete, after EVAR. This study, as a nationwide big data analysis, did not sufficiently capture clinical information such as whether the surgeries or procedures were elective or emergency, the presence of an aneurysmal rupture, the patient’s status in terms of shock, or whether there was graft infection following EVAR. The presence of rupture is a critical factor in AAA research. Therefore, although we conducted a sample survey on the presence of rupture among our AAA patients, we did not cite the rupture status as a risk factor in this study due to discrepancies between the diagnostic codes in Korea’s NHIS and the actual medical records regarding rupture presence at our institution. In particular, Jo et al. recently reported a multicenter study on how the I71.3 code reflects ruptured AAA in Korea and reported that the rate of true ruptured AAA was only 43%.²⁰ Many studies, however, target LOC involving a return to open repair after EVAR. Although these concepts are closely related, LOC predominantly entails EE, but may include a small subset of cases involving open proximal fixation, open endoleak repair, and similar procedures. Furthermore, this study utilized big data from Korea but did not obtain detailed information on specific aspects such as rupture or infection.

Conclusion

The Korean dataset analysis indicates a rising trend in EE aligned with the increased adoption of EVAR for AAA. Comparative evaluations with EVAR and OAR demonstrate less favorable outcomes for EE, especially in terms of 30-day and overall mortality rates. To mitigate EE incidence, the strategic choice of the initial treatment modality is essential. Moreover, precise timing for treatment in high-morbidity EE cases is crucial. The study underscores the need for well-designed research to address the evolving challenges in EE management, offering valuable insights for clinicians navigating the complexities of EVAR-related complications.

Supplemental Material

Supplemental Material - Nationwide analysis of EVAR explantation outcomes in Korea: A comprehensive dataset study

Supplemental Material for Nationwide analysis of EVAR explantation outcomes in Korea: A comprehensive dataset study by Hyo Kee Kim, Pyoung Jae Park, Jee Hyun Park, Young Ju Oh, Cheol Woong Jung, and Heungman Jun in Vascular

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a Korea University grant in 2022.

ORCID iD

Heungman Jun

Supplemental Material

Supplemental material for this article is available online.

References

Schermerhorn

Buck

O'Malley

, et al. Long-term outcomes of abdominal aortic aneurysm in the medicare population. N Engl J Med 2015; 373(4): 328–338.

United Kingdom

ETI

Greenhalgh

Brown

, et al. Endovascular versus open repair of abdominal aortic aneurysm. N Engl J Med 2010; 362(20): 1863–1871.

Patel

Sweeting

Powell

, et al. Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial. Lancet 2016; 388(10058): 2366–2374.

Wanhainen

Van Herzeele

Bastos Goncalves

, et al. Editor’s choice -- European society for vascular surgery (ESVS) 2024 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms. Eur J Vasc Endovasc Surg 2024; 67(2): 192–331.

Moll

Powell

Fraedrich

, et al. Management of abdominal aortic aneurysms clinical practice guidelines of the European society for vascular surgery. Eur J Vasc Endovasc Surg 2011; 41(Suppl 1): S1–S58.

Argyriou

Georgiadis

Lazarides

, et al. Endograft infection after endovascular abdominal aortic aneurysm repair: a systematic review and meta-analysis. J Endovasc Ther 2017; 24(5): 688–697.

Goudeketting

Fung Kon Jin

PHP

Unlu

, et al. Systematic review and meta-analysis of elective and urgent late open conversion after failed endovascular aneurysm repair. J Vasc Surg 2019; 70(2): 615–628 e7.

Song

Jung

Song

, et al. Background and data configuration process of a nationwide population-based study using the Korean national health insurance system. Diabetes Metab J 2014; 38(5): 395–403.

Quan

Sundararajan

Halfon

, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005; 43(11): 1130–1139.

10.

Ultee

Soden

Zettervall

, et al. Conversion from endovascular to open abdominal aortic aneurysm repair. J Vasc Surg 2016; 64(1): 76–82.

11.

Chaikof

Dalman

Eskandari

, et al. The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vasc Surg 2018; 67(1): 2–77 e2.

12.

Dias

Farivar

Steenberge

, et al. Management of failed endovascular aortic aneurysm repair with explantation or fenestrated-branched endovascular aortic aneurysm repair. J Vasc Surg 2018; 68(6): 1676–1687 e3.

13.

Wanhainen

Verzini

Van Herzeele

, et al. Editor's choice - European society for vascular surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms. Eur J Vasc Endovasc Surg 2019; 57(1): 8–93.

14.

Perini

de Troia

Tecchio

, et al. Infrarenal endograft clamping in late open conversions after endovascular abdominal aneurysm repair. J Vasc Surg 2017; 66(4): 1048–1055.

15.

Dubois

Harlock

Gill

, et al. A Canadian multicenter experience describing outcomes after endovascular abdominal aortic aneurysm repair stent graft explantation. J Vasc Surg 2021; 74(3): 720–728 e1.

16.

Perini

Gargiulo

Silingardi

, et al. Twenty-two year multicentre experience of late open conversions after endovascular abdominal aneurysm repair. Eur J Vasc Endovasc Surg 2020; 59(5): 757–765.

17.

Mazzaccaro

Righini

Giannetta

, et al. Factors associated with perioperative mortality after late open conversion for failed endovascular aortic repair. J Cardiovasc Surg 2023; 64(3): 297–303.

18.

Lopez Espada

Behrendt

Mani

, et al. Editor’s choice - the VASCUNExplanT project: an international study assessing open surgical conversion of failed non-infected endovascular aortic aneurysm repair. Eur J Vasc Endovasc Surg 2023; 66(5): 653–660.

19.

Powell

Wanhainen

. Analysis of the differences between the ESVS 2019 and NICE 2020 guidelines for abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2020; 60(1): 7–15.

20.

Seong

Ahn

, et al. Validation of I71.3 code for ruptured abdominal aortic aneurysm in Korea: misplaced diagnosis in claims data. Ann Surg Treat Res 2023; 104(3): 170–175.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB