Why Open Surgical Repair Remains Essential: Insights into the Suitability of Endovascular Aneurysm Repair in Ruptured Abdominal Aortic Aneurysms from a Tertiary Referral Center in South Korea

Abstract

Background

Several studies have demonstrated that emergency endovascular aneurysm repair (eEVAR) has become the standard treatment for ruptured abdominal aortic aneurysm (rAAA) compared to open surgical repair (OSR). This study investigated the feasibility of eEVAR in rAAA patients and analyzed the outcomes of OSR.

Methods

At our institution, all patients (n=58) presenting with rAAA underwent OSR. We conducted a retrospective review of surgically treated rAAA patients between February 2006 and March 2021. Patients with impending rupture or isolated iliac aneurysm rupture were excluded.

Results

Anatomic measurements based on preoperative computed tomography indicated that 28 patients (48.3%) could have been candidates for eEVAR. Reasons for unsuitability included (1) proximal neck length <10 mm (n=21, 51.7%); (2) proximal neck angulation >60° (n=11, 19.0%); (3) iliac artery diameter <5 mm (n=6, 10.3%); and (4) proximal neck diameter >32 mm (n=3, 5.2%). The 30-day mortality rate for OSR was 17.2% (n=10). The 1-year and 5-year survival rates were 53.4%±6.5% and 33.4%±6.3%, respectively. Multivariable logistic regression analysis revealed that a high preoperative serum lactate level (>5 mmol/L) and the presence of bowel ischemia were significant risk factors for 30-day mortality (odds ratio [OR], 11.95; 95% confidence interval [CI], 1.53–93.08; p<0.018; and OR, 15.28; 95% CI, 1.60–146.18; p<0.018, respectively).

Conclusion

More than half of rAAA patients were not candidates for eEVAR due to various anatomical reasons. OSR demonstrated favorable short- and long-term outcomes and remains a viable standard treatment for rAAA.

Introduction

Ruptured abdominal aortic aneurysm (rAAA) has an incidence ranging from 2.86 to 14.3 per 100,000 individuals per year and represents a surgical emergency with an overall mortality rate—including prehospital and pretreatment deaths—of approximately 80% [1]. More than half of patients do not survive to reach the operating room even in the 2000s, and the postoperative mortality of open surgical repair (OSR) for rAAA remains as high as 36.6% in recent meta-analyses of prospective trials [2-4].

Meanwhile, minimally invasive endovascular approaches have been widely adopted for the treatment of abdominal aortic aneurysms (AAAs) in patients with suitable anatomy because of their shorter hospitalization durations and lower morbidity. Consequently, the hypothesis that emergency endovascular aneurysm repair (eEVAR) might improve outcomes for rAAA was raised, and numerous retrospective studies along with a few randomized controlled trials have demonstrated comparable early postoperative mortality rates between eEVAR and OSR [5-9]. However, because eEVAR is technically limited in anatomically unsuitable patients, these studies may not represent the entire rAAA patient population.

As a dedicated aortic center, our policy has been to perform OSR on all rAAA cases regardless of anatomical conditions. In this study, we investigated the feasibility of eEVAR in rAAA patients and analyzed the outcomes of OSR.

Methods

Patients

A total of 58 patients who underwent emergency surgery for rAAA at Seoul National University Bundang Hospital from 2006, when computed tomography (CT) scans became available, to 2021 were enrolled. The mean patient age was 73.17±11.18 years, and 50 patients (86%) were men. Patients without an overt rupture of the abdominal aorta or those with isolated common iliac artery aneurysm rupture were excluded. Medical records and CT images were retrospectively reviewed.

Baseline patient characteristics are presented in Table 1. The mean follow-up period was 2.6±3.6 years.

Table 1.

Patient characteristics and preoperative conditions (N=58)

Characteristic	Value
Age (yr)	73.17±11.18
Male sex	50 (86.2)
Body mass index (kg/m2)	23.47±3.53
Smoking (current)	20 (34.5)
Comorbidities
Hypertension	41 (70.7)
Diabetes mellitus	12 (20.7)
Current smoker	20 (34.5)
Dyslipidemia	7 (12.1)
Stroke	9 (15.5)
Chronic kidney disease	7 (12.1)
Chronic obstructive pulmonary disease	4 (6.9)
Coronary artery disease	11 (19.0)
Preoperative conditions
pH	7.33 (7.20–7.38)
pH <7.2	11 (19.0)
Serum lactate (mmol/L)	3 (1.6–7.4)
Lactate >5 mmol/L	18 (31.0)
Hematocrit (%)	28.95 (24.6–35.1)
Hematocrit <25%	16 (27.6)
Serum creatinine (mg/dL)	1.50 (1.11–1.74)
Serum creatinine >1.5 mg/dL	27 (46.6)
PT INR	1.25 (1.14–1.43)
PT INR >1.5	12 (20.7)
SBP at ER admission	113 (94–140)
Lowest SBP during ER stay	82 (65.5–102.5)
Shock status (lowest SBP <70 mm Hg)	17 (29.3)
Altered consciousness	20 (34.5)
Syncope	9 (15.5)
Response to voice	6 (10.3)
Response to pain	3 (5.2)
Unresponsive	2 (3.4)
Bowel ischemia	6 (10.3)
ER stay duration (min)a)	69 (45–105)
Vasoconstrictor use in ER	8 (13.8)
Beta-blocker use in ER	5 (8.6)
Calcium-channel blocker use in ER	16 (27.6)
Time from OR admission to surgical incision (min)	35 (25–45)
Lowest SBP at OR before surgical incision (min)	66.5 (55–80)
Time from surgical incision to proximal clamping (min)	29 (25–48)
Lowest SBP before proximal clamping during operation (min)	62 (54–68)

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

PT, prothrombin time; INR, international normalized ratio; SBP, systolic blood pressure; ER, emergency room; OR, operating room.

^a)Limited to patients (n=41) who were diagnosed using computed tomography at outside hospitals.

Approval was obtained from the Institutional Review Board (IRB) of Seoul National University Bundang Hospital (IRB no., B-2202-741-105), and individual patient consent was waived due to the retrospective study design.

Surgical procedure

Management of rAAA patients began upon their arrival in the emergency room. Blood pressure was carefully maintained within the lower normal range to minimize bleeding from the ruptured aneurysm. Intravenous calcium-channel blockers and beta-blockers were administered to control systolic blood pressure below 100 mm Hg [10-12]. To prevent deterioration in vital signs following anesthetic induction, surgery was initiated immediately. After entering the peritoneal cavity, the proximal neck of the rAAA was rapidly cross-clamped using a combination of blunt and sharp dissection. Changes in blood pressure, the use of vasoconstrictors or antihypertensive agents, and the interval from emergency room arrival to surgical incision and proximal clamping are summarized in Table 1. If the distal anastomosis had to be performed at the level of the internal iliac artery (IIA) and access to the IIA was technically challenging, the IIA was sacrificed. In cases where rAAA was accompanied by bowel ischemia, massive retroperitoneal hematoma, or severe mesenteric edema, surgeons from the Department of General Surgery assisted with evaluation and management. When necessary, a Hartmann procedure was performed.

EVAR feasibility evaluation and clinical outcomes

We adhered to the general EVAR feasibility criteria provided by the manufacturers of Food and Drug Administration-approved commercial devices and the recent Society for Vascular Surgery (SVS) guidelines [13]. The following conditions were considered unsuitable for EVAR: (1) proximal neck length <10 mm, (2) proximal neck angulation >60°, (3) iliac artery diameter <5 mm, and (4) proximal neck diameter >32 mm (Fig. 1). When a patient met 1 or more of these conditions, duplicate counts were recorded. The primary endpoint of this study was 30-day mortality. Given that previous randomized controlled trials failed to demonstrate the superiority of EVAR—despite large retrospective studies showing better survival with eEVAR compared to OSR—we also analyzed whether EVAR feasibility was associated with clinical outcomes of OSR in rAAA using logistic regression analysis. Long-term survival was evaluated based on outpatient clinic records and mortality statistics from the Ministry of the Interior and Safety of Korea.

Fig. 1.

Anatomical conditions regarded as unsuitable for endovascular aneurysm repair and representative computed tomographic images.

Statistical analysis

Statistical analysis was performed using IBM SPSS ver. 20.0 (IBM Corp.). Categorical variables are presented as numbers (%) and continuous variables as means±standard deviations or medians (interquartile ranges), as appropriate. Logistic regression analysis was used to identify risk factors for 30-day mortality. Variables with p-values less than 0.2 in the univariable analysis were entered into the multivariable analysis. A p-value of less than 0.05 was considered statistically significant.

Results

CT assessment of AAA and EVAR feasibility

The maximal AAA diameter was 76.81±19.27 mm. Among all 58 patients, 28 (48.3%) were categorized as the “EVAR feasible” group, while the remaining 30 (51.7%) had at least 1 reason rendering them unsuitable for EVAR (Table 2). The most common reason for EVAR unsuitability was a short proximal neck (n=21). Additionally, 11 patients had multiple factors precluding EVAR.

Table 2.

Preoperative computed tomography assessment of abdominal aortic aneurysms for endovascular aneurysm repair suitability

Variable	Value
Maximal aortic diameter (mm)	76.81±19.27
50–59	9 (15.5)
60–69	9 (15.5)
70–80	15 (25.9)
80–90	8 (13.8)
90–99	12 (20.7)
≥100	5 (8.6)
Any unsuitable anatomical factors	30 (51.7)
Short proximal neck	21 (36.2)
Angulated neck	11 (19.0)
Unsuitable ilio-femoral access	6 (10.3)
Large neck diameter	3 (5.2)
Combined	11 (19.0)

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

Operative results and early clinical outcomes

The median operation time was 190 minutes (interquartile range, 165–235 minutes), and the median intraoperative estimated blood loss was 2,300 mL (interquartile range, 1,300–3,500 mL) (Table 3). The overall early (≤30 days) mortality rate among all patients was 17.2%. In the anatomically unsuitable group, 6 out of 30 patients died. Bowel ischemia was the most frequent cause of death (n=6), while the remaining 4 deaths were due to intractable bleeding. To address concerns regarding abdominal compartment syndrome, 5 patients underwent delayed laparotomy closure with an initially open abdomen. Two patients required delayed bowel resection due to postoperative abdominal compartment syndrome, and 1 patient—who did not exhibit abdominal compartment syndrome—developed superior mesenteric artery thrombotic occlusion combined with massive mesenteric ischemia, necessitating bowel resection. When comparing early mortality rates before and after 2017—after a dedicated aortic surgeon assumed responsibility for rAAA management—the early mortality rate decreased to 7.7% (2 out of 26) after 2017, compared to 25.0% (8 out of 32) before 2017.

Table 3.

Operative results and early postoperative clinical outcomes

Variable	Value
Operative results
Operation time (min)	190 (165–235)
Estimated blood loss (mL)	2,300 (1,300–3,500)
Transfused RBCs in OR (packs)	6 (4–10)
Transfused FFP in OR (packs)	3 (3–6)
Transfused PC in OR (packs)	10 (0–18)
Postoperative clinical outcomes
30-day mortality	10 (17.2)
Major complications	17 (29.3)
Bowel infarction requiring bowel resection	9 (15.5)
Reoperation for bleeding control	6 (10.3)
Delayed laparotomy closure	5 (8.6)
Abdominal compartment syndrome	2 (3.4)
Myocardial infarction	1 (1.7)
Reintubation	5 (8.6)
Tracheostomy	3 (5.2)
Mechanical ventilation (hr)	12.92 (8–28)
ICU stay (hr)	44.98 (28.88–101.17)
Total hospital stay (day)	18 (11–29)

Values are presented as median (interquartile) or number (%).

RBCs, red blood cells; OR, operating room; FFP, fresh frozen plasma; PC, platelet concentrates; ICU, intensive care unit.

Univariable and multivariable logistic regression analyses demonstrated that a preoperative serum lactate level exceeding 5 mmol/L (odds ratio [OR], 11.95; 95% confidence interval [CI], 1.53–93.08; p=0.018) and the presence of preoperative bowel ischemia (OR, 15.28; 95% CI, 1.60–146.18; p=0.018) were independent risk factors for early mortality (Table 4). EVAR feasibility was not associated with early mortality.

Table 4.

Logistic regression analysis for 30-day mortality after open surgical repair for ruptured abdominal aortic aneurysms

Variable	Univariable			Multivariable
	OR (95% CI	p-value		OR (95% CI)	p-value
Age (yr)	1.05 (0.97–1.14)	0.212		1.01 (0.90–1.15)	0.833
Hypertension	1.82 (0.34–9.61)	0.482
Diabetes	0.95 (0.17–5.19)	0.953
Cerebrovascular accident	0.56 (0.06–5.02)	0.601
End-stage renal disease	0.78 (0.08–7.28)	0.826
Chronic obstructive pulmonary disease	1.67 (0.16–17.90)	0.673
Coronary artery disease	2.14 (0.45–10.11)	0.335
EVAR suitability	0.67 (0.176–2.67)	0.566
Serum creatinine >1.5 mg/dL	6.11 (1.17–31.91)	0.032		7.89 (0.88–70.85)	0.065
Hematocrit <25%	1.15 (0.259–5.15)	0.851
Serum lactate >5 mmol/L	12.00 (2.23–64.49)	0.004		11.95 (1.53–93.08)	0.018
pH <7.2	2.14 (0.454–10.11)	0.335
PT INR >1.5	1.86 (0.40–8.62)	0.429
Shock status	1.80 (0.44–7.40)	0.418
Bowel ischemia	23.00 (3.50–151.00)	0.001		15.28 (1.60–146.18)	0.018

OR, odds ratio; CI, confidence interval; EVAR, endovascular aneurysm repair; PT, prothrombin time; INR, international normalized ratio.

Long-term survival

During follow-up (beyond 30 days post-surgery), 16 patients died, of whom 9 belonged to the EVAR non-feasible group. Two patients discharged to a nursing facility died 1 month after discharge from unknown causes. Aside from 1 case of delayed ischemic colitis and 1 case of graft infection, the remaining 14 deaths were unrelated to surgery; among these, 1 was due to pneumonia, 1 occurred due to malignancy, and 12 were classified as deaths from unknown causes, as confirmed by mortality statistics collected after 1 year postoperatively. The overall 1-year and 5-year survival rates were 53.4%±6.5% and 33.4%±6.3%, respectively (Fig. 2).

Fig. 2.

Overall survival after open surgical repair for ruptured abdominal aortic aneurysms.

Discussion

In this study, we demonstrated that more than half of rAAA patients might not be suitable for eEVAR due to unfavorable anatomy, while the outcomes of primary OSR were satisfactory, with a 30-day mortality rate of 17.2%. Although the hypothesis that EVAR suitability might impact survival after OSR was not confirmed, other preoperative clinical factors—including age, serum creatinine and lactate levels, and the presence of bowel ischemia—were found to be associated with 30-day mortality.

EVAR has become widely adopted as the primary treatment option for elective AAA repair worldwide. Yei et al. [14] reported that EVAR, which was performed approximately 1.1 times as often as OSR from 2004 to 2008, was conducted at a rate 8.5 times higher than OSR from 2014 to 2018 in their study of the Medicare-matched SVS Vascular Quality Initiative and the Vascular Implant Surveillance and Interventional Outcomes Network.

In addition to its improved early periprocedural outcomes in non-ruptured AAA, EVAR began to be applied in the setting of rAAA. Large international multicenter retrospective studies have even suggested that endovascular treatment is superior to open repair [5,15]. However, all randomized controlled trials comparing eEVAR and OSR for rAAA have shown no significant difference between the 2 treatments [7-9,16]. Furthermore, meta-analyses including these trials have failed to demonstrate the superiority of eEVAR [4,17]. These clinical trials have inherent limitations, as patients are randomized after CT evaluation for EVAR suitability, which may introduce selection bias. Although the IMPROVE Trial randomized patients before assessing EVAR suitability, over 30% of those allocated to eEVAR ultimately underwent OSR due to unsuitable anatomy.

In the AJAX trial enrollment process, 240 patients were excluded before randomization due to unfavorable anatomy, and only 155 patients were included after EVAR suitability assessment, indicating that rEVAR is anatomically feasible in less than 40% of cases [8]. Other single-center retrospective studies have reported EVAR feasibility in rAAA to be 23.0% in a Korean group and 50.5% in a Swedish group [18,19]. Consistent with these findings, 51.3% of our rAAA patients were deemed not feasible for EVAR.

Violations of the instructions for use (IFU) during EVAR have been associated with subsequent sac enlargement, reintervention, and compromised long-term survival in elective non-ruptured AAA [20-22]. In the context of eEVAR for rAAA, IFU violation is a significant risk factor for in-hospital mortality, a higher incidence of Type 1a endoleak, and an increased likelihood of intraoperative conversion to open repair [23]. Given these clinical implications and the low incidence of EVAR feasibility, our center has adopted a policy of treating all rAAA cases with OSR, despite performing EVAR in elective non-ruptured AAA.

In fact, eEVAR often requires more time to control extravasation, especially in centers that do not stock a complete range of prostheses in all sizes. In contrast, OSR can achieve hemostasis immediately after cross-clamping, and the graft can be trimmed and tailored to a wide range of configurations [24]. Even after stent graft deployment, eEVAR carries a risk of intra-abdominal bleeding due to residual endoleak, which occurs in approximately 20% of cases; these patients are frequently converted to OSR due to shock or the risk of abdominal compartment syndrome.

Population-based studies have reported static OSR results in rAAA from 53%, potentially justifying the adoption of eEVAR for rAAA [1]. However, our early mortality rate was 17.2%, with a further reduction to 7.7% after 2017 compared to 25.0% before 2017. Since bowel ischemia was the most common cause of death in our study, rapid isolation and cross-clamping of the proximal neck are critical in OSR for rAAA to minimize blood loss and prevent abdominal compartment syndrome. This procedure is best performed by specialized and experienced surgeons, which likely explains the improved outcomes observed after dedicated aortic surgeons assumed responsibility for rAAA management. Karthikesalingam et al. [15] also reported that teaching hospitals with high bed capacity had lower postoperative mortality regardless of treatment modality.

The symptoms of rAAA—sudden, severe abdominal pain or syncope due to hypotension—often lead to diagnosis in the emergency room of the nearest primary or secondary healthcare facility, where definitive treatment may not be available. This necessitates emergency transfer, requiring a decision between a specialized referral center capable of both OSR and EVAR and a nearby vascular center that offers only EVAR. A population-based study in the United Kingdom found that a 1-stop emergency vascular service yields better survival outcomes despite longer travel distances [3]. In light of the critical role of experienced surgeons in OSR and the low feasibility of EVAR in real-world settings, rAAA patients should be referred to centers where dedicated aortic surgeons can provide emergency OSR.

Limitations

This study has several limitations. As a retrospective observational study from a single center, our results may not be readily generalizable. Furthermore, this single-arm study does not allow for a direct comparison between OSR and eEVAR outcomes. Most of our patients were referred from distant primary hospitals, and the death of more critically ill patients during transfer may have introduced selection bias. Additionally, patients who could have been treated with eEVAR at a primary or closer hospital may have been excluded, further contributing to selection bias. However, because our surgical approach for rAAA is applied regardless of anatomical constraints, selection bias based on anatomical conditions is likely less significant compared to biases related to hemodynamic stability permitting transfer.

Conclusion

More than half of rAAA patients were not candidates for eEVAR due to unsuitable anatomy. OSR for rAAA performed at a specialized referral center by dedicated aortic surgeons yielded favorable clinical outcomes which were superior to those reported for eEVAR in the current literature. Given the importance of OSR performed by experienced surgeons and the low feasibility of EVAR, centers equipped with dedicated aortic surgeons to provide emergency OSR should be prioritized in the triage of rAAA patients.

Funding Statement

Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.