Abstract

Background: This study aims to examine factors influencing mortality and reintervention among endovascular aortic aneurysm repair patients treated over nearly two decades in the first hybrid operating room of Türkiye.

Methods: Between December 2006 and December 2020, a total of 372 patients (349 males, 23 females; mean age: 70.4±8.2 years; range, 45 to 89 years) who underwent endovascular aortic aneurysm repair were retrospectively analyzed. Independent predictors of 30-day, one-year, and five-year mortality rate were identified. The Kaplan-Meier analyses compared outcomes across surgical approaches and graft types.

Results: Mortality rates were 6.5% at 30 days, 14.2% at one year, and 21.8% at five years. Multivariate analysis identified postoperative complications (hazard ratio [HR]=3.68, p=0.034), elevated creatinine (HR=1.58, p=0.004), prolonged intubation (HR=1.01, p<0.001), and increased blood transfusion volume (HR=1.17, p=0.003) as independent predictors of 30-day mortality. Advanced age emerged as a significant predictor of five-year mortality (HR=1.09, p<0.001). Elective procedures and bi-iliac grafts demonstrated significantly improved survival across all timeframes (p<0.001).

Conclusion: This real-world analysis identifies several procedural and patient-related factors associated with mortality following endovascular aortic aneurysm repair. Postoperative care indicators consistently demonstrated stronger independent associations with mortality than anatomical or demographic factors, highlighting the importance of comprehensive perioperative care optimization in endovascular aortic aneurysm repair management.

Over the last two decades, endovascular aortic repair (EVAR) emerged as the first treatment option in patients with abdominal aortic aneurysms (AAAs) with suitable anatomy.[1,2] According to the EVAR 1 trial results, EVAR showed an early survival benefit compared to open repair. Though, EVAR procedure is burdened by a substantially higher rate of secondary procedures.[3] The EVAR 1 trial demonstrated a 5% reintervention rate in the first four years after EVAR.[3] The long-term outcomes of EVAR are still under-researched, particularly in real-world settings beyond randomized clinical trials.

First hybrid cardiovascular surgery operating room of Türkiye was established at our institution in 2007. Since then, all the endovascular procedures are carried out solely by cardiovascular surgeons without any regular collaboration with interventional radiologists or cardiologists. The EVAR procedures constitute a significant portion of the hybrid interventions, as we published our mid-term results previously.[4,5] In the present study, we aimed to provide a comprehensive analysis of nearly 20 years of real-world EVAR data, examining variables such as age, aortic diameter, surgical timing, graft type, and device brand to determine their impact on survival and complication rates.

Methods

This single-center, retrospective cohort study was conducted at İzmir Atatürk Training and Research Hospital, Department of Cardiovascular Surgery between December 2006 and December 2020. A total of 501 patients who underwent EVAR at our center were evaluated. After excluding 129 patients who were lost to follow-up during the postoperative period, a total of 372 patients (349 males, 23 females; mean age: 70.4±8.2 years; range, 45 to 89 years) were included in the analysis. The pre- and postoperative data, as well as outpatient clinic attendance information, were obtained from the hospital information system. Written informed consent was obtained from each patient. The study protocol was approved by the İzmir Katip Çelebi University Ethics Committee (Date: 19.11.2020, No: 1068). The study was conducted in accordance with the principles of the Declaration of Helsinki.

Statistical analysis
Study power analysis and sample size calculation were performed using the G*Power version 3.1.9.7 software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) to evaluate the statistical power of the relationship between surgical status (emergency/elective) and five-year mortality. With an effect size of w=0.189, ?=0.05, and a sample size of 372 patients, the analysis yielded a power of 0.954. This value confirms the adequacy of our sample size in evaluating the effect of surgical type on mortality.

Statistical analysis was performed using the Jamovi version 2.3.28 and JASP version 0.18.3 software (R Foundation for Statistical Computing, Vienna, Austria). Continuous data were expressed in mean ± standard deviation (SD) or median (min-max), while categorical data were expressed in number and frequency. Data normality was assessed using the Shapiro-Wilk test or Kolmogorov-Smirnov/ Anderson-Darling test. Categorical variables were analyzed using the Pearson chi-square or Fisher exact tests. For continuous variables, independent samples t-test or Mann-Whitney U test was applied based on normality assumptions. Survival analyses were conducted using the Kaplan-Meier method to evaluate the impact of different surgical approaches, graft t ypes, a nd d evice brands on survival outcomes. To identify independent predictors of mortality, Cox proportional hazards regression analyses were performed at 30-day, one-year, and five-year follow-up periods. First, univariate analyses were conducted and variables showing significant associations in the univariate analyses were subsequently included in the multivariate models. Due to insufficient mortality events in non-Medtronic™ categories (no deaths on Days 30, only three deaths at one year), the device brand variable could not be incorporated into 30-day and one-year mortality models. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for each variable. A p value of <0.05 was considered statistically significant.

Results

Hypertension was the most prevalent comorbidity, followed by hyperlipidemia, coronary artery disease, and chronic obstructive pulmonary disease (COPD). A ruptured AAA was present in 10.8% of cases, and 20.7% underwent emergency procedures. The median aortic diameter measured 63.0 mm. Medtronic™ devices (Medtronic Inc., MN, USA) were used most frequently, followed by Endologix™ (Endologix Inc., CA, USA) and Gore™ (Gore Inc., DE, USA). Bi-iliac grafts were applied in nearly four out of five procedures, while uni-iliac grafts were used less commonly. Endoleak was observed in 13.1% of patients, with Type 1A and Type 1B being the most frequent subtypes. Postoperative complications were noted in over a quarter of the cohort, and 16.7% required secondary interventions. The observed mortality rates at 30 days, one year, and five years were 6.5%, 14.2%, and 21.8%, respectively, with a median follow-up time of 17 (range, 1 to 149) months.

Considering the relationship between demographic and clinical variables with mortality, age was found to be significantly higher in patients who died across all mortality timeframes (p<0.001). Similarly, the presence of ruptured AAA and emergency surgical intervention were significantly associated with higher mortality rates (p<0.001). Other demographic variables and comorbidities did not demonstrate statistically significant associations with mortality (p>0.05) (Table 1).

Table 1. Comparison of demographic and clinical characteristics by 30-day, 1-year, and 5-year mortality

Analysis of surgical characteristics revealed that abdominal aortic diameter was significantly larger in patients with 30-day (p=0.002), one-year (p=0.001), and five-year (p=0.004) mortality. Device brand analysis demonstrated that Medtronic™ devices were significantly more prevalent in patients with 30-day (p=0.025) and one-year (p=0.004) mortality. Uni-iliac graft procedures were associated with significantly higher mortality rates across all timeframes (30-day: p=0.003; one-year: p<0.001; five-year: p=0.001). Additionally, postoperative complications, blood transfusion requirements, intubation duration, and length of stay in the intensive care unit (ICU) were all significantly elevated in non-survivors across all mortality timeframes (p<0.05) (Table 2).

Table 2. Comparison of surgical characteristics by 30-day, 1-year, and 5-year mortality in patients undergoing EVAR

Laboratory parameter evaluation demonstrated significantly elevated creatinine (p<0.001), international normalized ratio (INR) (p=0.018), and aspartate aminotransferase (AST) (p<0.001) levels, along with significantly decreased platelet counts (p=0.002) in patients with 30-day mortality. In one-year mortality analysis, INR (p=0.009) and AST (p=0.008) values remained significantly elevated, while five-year mortality analysis revealed only creatinine levels to be significantly higher (p=0.003) (Table 3).

Table 3. Comparison of laboratory parameters by 30-day, 1-year, and 5-year mortality in patients undergoing EVAR

In the analysis of endoleak development, no significant differences were observed between patients with and without endoleak in terms of abdominal aortic diameter, device brand, surgical timing, or graft configuration. However, the abdominal aortic diameter was significantly smaller in patients with Type 1 endoleak (p=0.042), while no statistically significant differences were found for Type 2 or Type 3 endoleaks with respect to these variables (p>0.05 for all comparisons).

Kaplan-Meier survival analyses clearly demonstrated the effect of surgical approaches on long-term outcomes. The mean survival times were significantly longer in patients undergoing elective surgery compared to emergency cases across all mortality timeframes (Table 4, Figure 1). Similarly, while examining the impact of graft type on survival, patients with bi-iliac grafts demonstrated significantly longer survival times compared to those with uni-iliac grafts across all mortality intervals (Table 4, Figure 2). While assessing the influence of device brand on survival, a statistically significant difference was observed in short and mid-term follow-up (30 days: p=0.034; 1 year: p=0.005), whereas this significance was lost in long-term follow-up (5 years: p=0.106). As demonstrated in Figure 3, Cook™ (Cook Medical LLC, IN, USA) and Anaconda™ (Anaconda Biomed S.L., Barcelona, Spain) brands exhibited better survival curves compared to Medtronic™, although this difference lost statistical significance at the five-year follow-up (Table 4).

Table 4. Kaplan-Meier survival analysis comparing 30-day, 1-year, and 5-year mortality by surgical timing, graft type, and device brand

Figure 1. Kaplan-Meier survival curves comparing 30-day mortality between elective and emergency EVAR procedures (p<0.001).
EVAR: Endovascular aortic repair.

Figure 2. Kaplan-Meier analysis comparing 1-year survival between bi-iliac and uni-iliac grafts in EVAR procedures (p<0.001).
EVAR: Endovascular aortic repair.

Figure 3. Comparison of 5-year survival rates among different device brands in EVAR procedures (p=0.106).
EVAR: Endovascular aortic repair

To elucidate the independent predictors of mortality at different time points, we performed Cox proportional hazards regression analysis (Table 5). The multivariate analysis revealed several consistent predictors across different timeframes. For 30-day mortality, postoperative complications, elevated creatinine levels, prolonged intubation duration, and increased blood transfusion volume emerged as independent predictors. In the univariate analysis, age, ruptured AAA, emergency surgical status, and abdominal aortic diameter all demonstrated significant associations with 30-day mortality; however, these relationships did not persist in the multivariate model.

Table 5. Cox proportional hazards regression analysis for 30-day, 1-year, and 5-year mortality in patients undergoing endovascular aortic repair

The analysis of one-year mortality identified advanced age, prolonged intubation duration, and increased blood transfusion requirements as independent predictors. While graft type showed a significant association in the univariate analysis, this association did not maintain statistical significance in the multivariate model.

The five-year mortality analysis demonstrated that advanced age, postoperative complications, prolonged intubation duration, and increased blood transfusion volume maintained their significant impact on long-term survival. Notably, device brand showed a significant association with five-year mortality in the univariate analysis, but this effect was attenuated in the multivariate model. Device brand could not be analyzed in the 30-day and one-year mortality models due to the absence or limited number of mortality events in non-Medtronic™ device categories during these timeframes.

Discussion

In our study, we examined the long-term mortality outcomes of patients undergoing EVAR and assessed the effects of various patient and procedure-related factors. Our findings support the safety of EVAR in older populations while highlighting significant differences between emergency and elective surgeries. Within this context, the alignment of our study with existing literature and its novel contributions are emphasized.

One of our primary findings is the increase in long-term mortality rates following EVAR, particularly in older patients and those with larger abdominal aortic diameters. This observation, consistent with prior studies, underscores the sensitivity of EVAR outcomes to age and aortic diameter.[6] Similarly, in the study by Bulder et al.,[7] higher long-term mortality risk was reported among women following AAA repair, with the impact of age, comorbidity, and complications being particularly pronounced.

Comparing emergency and elective surgeries, our study demonstrated a lower mortality rate for elective procedures, emphasizing the importance of timely and planned EVAR implementation. This finding is corroborated by Khan and Fisher's[8] study, which associated emergency EVAR with higher mortality, particularly within older age groups.

Regarding device type, our study identifies the Medtronic™ device as associated with higher mortality rates compared to other brands. Nevertheless, this situation has a simple explanation related to our daily practice since only Medtronic™ as a company holds a base located in our city. Thus, almost all the time the emergency cases were/are carried out using Medtronic™ devices, unless other brands are in our own armamentarium by chance on that particular day. While graft type showed a significant association in the univariate analysis, this association did not maintain statistical significance in the multivariate model at all timeframes. This finding might also be crucial in highlighting the impact of device selection on patient outcomes and may provide a basis for further research in this area.[9]

The influence of graft type in EVAR procedures was also analyzed. Our study shows that bi-iliac grafts contribute to higher survival rates, a result supported by similar findings in the literature, such as those by Li et al.,[10] where bi-iliac grafts were linked to lower complication rates. This finding also has a simple explanation since almost all the uni-iliac cases were emergency cases with hemodynamic instability and fast deployment of the device was crucial. In addition, the only available uni-iliac graft was that of Medtronic™ before 2018. Recently, we prefer to use bi-iliac devices whenever possible and available while attempting to deal with the hemodynamic instability using resuscitative balloon occlusion of the aorta (REBOA).

Patient-related factors, such as age and comorbidities, should be considered in relation to long-term mortality. The particularly low survival rates in elderly patients underscore the critical role of age in EVAR outcomes.[11] Comorbidities, particularly hypertension, COPD, and chronic kidney disease, further increase mortality risk and necessitate carefully planned follow-up protocols for these groups.

Among procedural factors, whether EVAR is performed under emergency or elective conditions directly affects the patient's postoperative care requirements. In our study, elective surgeries were associated with longer survival, reinforcing the importance of personalized surgical timing. This is also observed in the study by Enriquez de Salamanca Gambara et al.,[12] where higher mortality was reported in emergency EVAR cases.

Nonetheless, there are some limitations to this study. One notable limitation is its retrospective design, which may inherently limit the ability to establish causal relationships between variables. While data were rigorously collected from a well-documented cohort, the reliance on historical records can introduce biases, including inconsistencies in record-keeping and potential misclassification of clinical events. Additionally, although our sample is sizeable, it is confined to a single center, which may limit the generalizability of the findings to other populations and healthcare settings. The variations in device brands and types within the study group also present a challenge in controlling for manufacturer-specific differences that could impact outcomes. Another limitation involves the follow-up duration for some patients, which may not fully capture late-onset complications associated with EVAR, particularly in terms of secondary interventions or aneurysmrelated mortality beyond five years. Moreover, the study did not incorporate a randomized design, indicating that the choice of graft type or device may have been influenced by surgeon preference or patient-specific characteristics rather than random allocation. Finally, while laboratory and clinical markers were thoroughly analyzed, the absence of more recent imaging data on aneurysm morphology limits our ability to assess certain structural predictors of endoleak. Future studies with a prospective, multi-center approach and standardized device selection criteria could help validate these findings and offer a broader context for interpreting EVAR outcomes.

In conclusion, this study provides a detailed examination of long-term outcomes following endovascular aortic aneurysm repair in patients treated in a hybrid operating room setting. Our findings underscore the relationship between several preoperative and intraoperative factors, such as patient age, abdominal aortic diameter, and procedural urgency, and both early and extended postoperative mortality rates. Notably, patients undergoing elective endovascular aortic aneurysm repair demonstrated markedly better survival outcomes compared to those receiving emergency intervention. Device selection also emerged as a key factor, with specific brands linked to differing mortality rates and survival times, indicating a potential impact of device choice on long-term outcomes, although multivariate analysis did not confirm this impact.

The study further highlights the survival advantage associated with bi-iliac graft use over uni-iliac grafts, suggesting that bi-iliac grafting could serve as a preferable approach to minimize complications like endoleak. Additionally, laboratory markers such as elevated creatinine and aspartate aminotransferase levels were associated with increased mortality, providing potential biomarkers for identifying high-risk patients preoperatively. Our results also indicate that postoperative complications, intensive care unit duration, and blood transfusion requirements correlate strongly with poorer outcomes, emphasizing the importance of optimized perioperative management. By identifying these predictors, this study contributes valuable insights that may guide future clinical decisions and patient management strategies, ultimately enhancing survival rates and quality of life for patients undergoing endovascular aortic aneurysm repair.

Acknowledgments: The authors extend their gratitude to Ali Karakoç and Gökhan Karakoç from Model Statistics CRO and Clinical Trials Center (www.modelistatistik.com) for their invaluable contributions to the statistical analysis in this study.

Data Sharing Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author Contributions: Control/supervision: I.Y., H.I., L.Y., A.G.; Data collection and processing: I.Y., H.I., S.Y.; Design, analysis and interpretation, idea/concept, writing the article, critical review: I.Y., H.I., S.Y., O.G., H.Ç., L.Y., A.G.; Literature review, references and fundings, materials: I.Y., H.I.

Conflict of Interest: The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding: The authors received no financial support for the research and/or authorship of this article.

References

1) Cicala N, Perini P, Catasta A, Fornasari A, Ucci A, Freyrie A. Systematic review and meta-analysis of incidence, indications, and outcomes of early open conversions after EVAR for abdominal aortic aneurysms. Int Angiol 2024;43:271-9. doi:10.23736/S0392-9590.24.05153-8.

2) Pratesi C, Esposito D, Apostolou D, Attisani L, Bellosta R, Benedetto F, et al. Guidelines on the management of abdominal aortic aneurysms: Updates from the Italian Society of Vascular and Endovascular Surgery (SICVE). J Cardiovasc Surg (Torino) 2022;63:328-52. doi: 10.23736/ S0021-9509.22.12330-X.

3) Patel R, Sweeting MJ, Powell JT, Greenhalgh RM; EVAR trial investigators. 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;388:2366-74. doi:10.1016/S0140-6736(16)31135-7.

4) Gunes T, Yilik L, Yetkin U, Yurekli I, Ozcem B, Yazman S, et al. A comparison of open conventional and endovascular surgical therapies in abdominal aortic aneurysm repair. Turk Gogus Kalp Dama 2012;20:515-23

5) Yazman S, Yürekli İ, Yılık L, Yetkin U, İner H, Güneş T, et al. Our mid-term results of endovascular repair of abdominal aortic aneurysms. Turk Gogus Kalp Dama 2016;24:274-80 doi: 10.5606/tgkdc.dergisi.2016.12320.

6) Conway R, Byrne D, O"Riordan D, Silke B. Short- and longterm mortality following acute medical admission. QJM 2023;116:850-4. doi: 10.1093/qjmed/hcad181.

7) Bulder RMA, van der Vorst JR, van Schaik J, Bedene A, Lijfering WM, Bastiaannet E, et al. Persistent high longterm excess mortality after elective AAA repair especially in women: A large population-based study. Ann Surg 2023;278:815-22. doi: 10.1097/SLA.0000000000006044.

8) Khan K, Fisher R. Long term survival after standard EVAR. BJS 2021;108(Suppl 6):znab259.1128. doi: 10.1093/ bjs/znab259.1128

9) Cosentino N, Resta ML, Somaschini A, Campodonico J, D'Aleo G, Di Stefano G, et al. ST-segment elevation acute myocardial infarction complicated by cardiogenic shock: Early predictors of very long-term mortality. J Clin Med 2021;10:2237. doi: 10.3390/jcm10112237.

10) Li Y, He C, Zhang H, Zhang X, Zhang X, Zhang T. Endovascular repair for abdominal aortic aneurysm in mainland China: A systematic review and meta-analysis. Ann Vasc Surg 2023;89:338-52. doi: 10.1016/j.avsg.2022.09.058.

11) Wang G, Sun Y, Lin Z, Fei X. Elective endovascular vs open repair for elective abdominal aortic aneurysm in patients ?80 years of age: A systematic review and metaanalysis. Vasc Endovascular Surg 2023;57:386-401. doi:10.1177/15385744221149911.

12) Enriquez de Salamanca Gambara R, Sanz-García A, Martín-Conty JL, Polonio-López B, Del Pozo Vegas C, Martín-Rodríguez F, et al. Long-term mortality in patients transferred by emergency medical services: Prospective cohort study. Prehosp Disaster Med 2023;38:352-9. doi:10.1017/S1049023X23005800.