Abstract
Objectives:
Chronic obstructive pulmonary disease (COPD) is a known risk factor for abdominal aortic aneurysm (AAA) growth and rupture. The impact of COPD on AAA sac behavior following endovascular AAA repair (EVAR) is poorly understood. This study aimed to determine the association between COPD and sac remodeling after EVAR.
Methods:
We identified all EVAR patients (2010–2021) in the Society for Vascular Surgery-Vascular Quality Initiative database. COPD severity (none, medication-treated, oxygen (O2)-dependent) was the primary exposure variable. The primary end point was 1-year sac growth post-EVAR. Mixed effects linear and logistic regression were used to assess the effects of COPD severity on sac remodeling, controlling for covariates. Cox proportional hazards regression and competing risks regression identified predictors of mortality and reintervention.
Results:
30,686 patients met inclusion criteria. COPD was present in 24% of patients (no COPD, 75.8% [n = 23,260], medication-treated COPD, 19.7% [n = 6,057], O2-dependent COPD, 4.5% [n = 1,369]). O2-dependent COPD was associated with persistent sac growth (mean difference: +0.55 mm, 95% confidence interval [CI] [0.05–1.05], P = 0.03) following EVAR implantation; however, medication-treated COPD was not (mean difference: −0.06 mm, 95% CI [−0.32 to 0.19]; P = 0.64). O2-dependent COPD patients did not experience increased reintervention rates (hazard ratio [HR] 0.82, 95% CI [0.65–1.02], P = 0.07) but were identified to have significantly increased mortality (HR 1.8, 95% CI [1.52–2.22], P < 0.0001).
Conclusion:
O2-dependent COPD was significantly associated with increased sac growth after EVAR. O2-dependent COPD did not correlate with increased reintervention, likely due to the disproportionately elevated mortality rates in this group. These findings highlight that patients with O2-dependent COPD have limited life expectancy and potentially less favorable sac remodeling, and elective EVAR should thus be reserved for only those at greatest risk for AAA-related mortality.
INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is a prevalent condition affecting nearly 400 million people worldwide.1 It is highly associated with long-standing smoking and varies in severity from mild and moderate cases, usually treated with inhaled medications (i.e., ‘‘medication-treated’’ COPD), to severe forms requiring oxygen therapy (i.e., ‘‘O2-dependent’’ COPD). Notably, nearly one-third of patients with abdominal aortic aneurysm (AAA) have concurrent COPD, which can increase the complexity of elective surgical decision-making.2
This relationship is underscored by the fact that COPD has been implicated to play an important role in the accelerated development of AAA due to its association with chronic inflammation, elastin degradation, and vascular smooth muscle cell dysfunction, among others.3,4 Furthermore, these biological underpinnings of AAA pathophysiology appear to correlate with COPD severity.5,6 Moreover, and perhaps most crucial, COPD has been linked to increased rates of AAA sac growth and rupture.7–10 Thus, COPD may increase the associated risks of untreated AAA, while also raising the perioperative risk of repair. Accordingly, there is an obvious appeal for applying less invasive endovascular aneurysm repair (EVAR) to minimize perioperative morbidity in this at-risk group, as reported in the Society for Vascular Surgery’s Clinical Practice Guidelines.11–13 However, COPD patients are still at increased risk of major adverse events during the perioperative period.14 While COPD is a known risk factor for accelerated preoperative AAA sac expansion, there are limited data surrounding its impact on sac behavior following repair.15,16 Consequently, these patients may remain at risk for continued sac expansion and the need for increased reintervention rates following EVAR.
Therefore, the purpose of this study was to characterize the impact of COPD on post-EVAR sac growth, reintervention, and mortality in contemporary practice.
METHODS
Cohort
The Vascular Quality Initiative (VQI) is a prospectively collected national set of 14 registries covering a range of vascular procedures, with more than 1,000 participating centers. This study retrospectively reviewed the national Endovascular AAA Repair registry database. The VQI’s Research Advisory Committee assented to the use of these data, and the University of Florida at Gainesville’s Institutional Review Board waived the need for informed consent, as all VQI records are deidentified of patient health information.
All patients who underwent either elective or nonelective EVAR between January 2010 and June 2021 were initially queried (N = 63,072). Patients who were missing critical data, including COPD status (N = 60) and postoperative endoleak (N = 735) were excluded. Records for subsequent (nonindex) procedures on an individual patient were also omitted (N = 79). Records were excluded if COPD was present but not treated (N = 5,819), and if 1-year postoperative sac growth data were not recorded (N = 25,693). After these exclusions, the remaining cohort of 30,686 patients were available for analysis.
The primary exposure variable was the presence of treated COPD. Patients were categorized into ‘‘no COPD,’’ ‘‘medication-treated COPD,’’ and ‘‘O2-dependent COPD.’’
End Points and Statistics
The primary end point of the study was aneurysm sac expansion after EVAR. Secondary outcomes were reintervention and long-term mortality. Aneurysm sac expansion was analyzed as a continuous variable, as well as a binary variable of sac expansion ≤5 mm vs. >5 mm, and as a binary variable of sac expansion ≤10 mm vs. >10 mm. For index hospitalization, ‘‘reoperation’’ is strictly a yes/no variable. Indication is not recorded. Reintervention for any reason during follow-up was counted as a reintervention which included index hospital reoperations (which are not subspecified in the registry), as well as reoperation events after the initial admission (note, in the VQI EVAR long-term follow-up file, the indication for reoperation includes sac growth, endoleak, device migration, limb occlusion, limb stenosis, rupture, infection, or ‘‘other’’). Mortality data are found in the VQI long-term follow-up data set.
Univariate comparison of the 3 exposure groups was carried out, including a comparison of demographics and baseline characteristics, operative characteristics, and outcomes. The included demographic characteristics were age, sex, cardiac and renal comorbidities, anemia, relevant preoperative medications, history of prior surgery, family history of AAA, and others. Body mass index (BMI) was calculated using the formula BMI = weight (kg)/height(m).2 Operative characteristics include maximum AAA diameter preoperatively, procedure time, type of anesthesia, and type of main body device. Univariate outcomes evaluated included type II endoleak, maximum amount of sac growth at last known follow-up, reintervention before discharge, any reintervention, 30-day and longterm mortality. Univariate comparisons of categorical variables were performed with chi-squared tests, and univariate comparisons of continuous variables were performed with Kruskal-Wallis tests.
Multivariable analysis was performed separately for primary and 2 secondary outcomes: sac expansion, reintervention, and long-term mortality. For all multivariable analysis, missing data points were imputed with the median or the mode, and overall rates of missingness were no greater than 1.5% for any included variable in any analysis (Supplemental Table I).
Mixed effects regression was used to identify the impact of COPD on sac growth as a continuous variable, and as a binary variable of sac growth >5 mm and sac growth >10 mm. Linear regression modeling was applied to the former, and logistic regression modeling to the latter. Due to the observation that many patients experienced mortality within 1–2 years of EVAR, a competing-risk model was used to assess the effect of COPD on long-term freedom from reintervention in the presence of a competing risk of death. Mixed effects Cox proportional hazards regression was used to assess the effects of COPD on mortality.
For all multivariable comparisons, model covariates included in-hospital type 2 endoleak, smoking, nonelective admission, age, gender, non-white race, ethnicity, BMI class, transfer status, hypertension, diabetes, coronary artery disease, congestive heart failure, prior lower extremity bypass, prior coronary artery bypass surgery, prior aneurysm repair, prior aortic surgery, prior lower extremity angioplasty or stent placement, prior major amputation, chronic kidney disease (creatinine >1.5 mg/dL or on dialysis), abnormal stress test, hemoglobin <12 mg/dL, preoperative aspirin, statin, beta-blocker, P2Y12 inhibitor use, family history of AAA, ejection fraction <50%, AAA diameter (mm), iliac aneurysm, surgeon assessment of unfitness for open aneurysm repair, use of regional versus general anesthesia, main aortic device configuration, and documentation of in-hospital nontype 2 endoleak (i.e., presence of type 1, 3, or 4 on completion imaging).
All statistical analyses were conducted using the R statistical software package (V.4.4.1, The R Foundation for Statistical Computing).
RESULTS
Demographics, Comorbidities, and Operative Characteristics
Many key demographic differences were present within the 3 groups (Table I). Patients with O2-dependent COPD were older (non-COPD, 73.3 ± 8.7 vs. medication-treated COPD, 72.9 ± 7.7 vs. O2-dependent COPD, 74.0 ± 7.7, P < 0.0001), more likely to be female (15.9% vs. 23.1% vs. 30.5%, P < 0.0001), and had increased prevalence of diabetes (19.4% vs. 21.0% vs. 24.3%, P < 0.0001). Smoking status also varied significantly between groups, with the O2-dependent COPD group most likely to be former smokers (56.2% vs. 55.2% vs. 70.3%, P < 0.0001) while the medication-treated COPD group were more likely to be current smokers (25.9% vs. 39.9% vs. 26.2%, P < 0.0001). COPD patients in both treatment groups were less likely to have a family history of AAA (9.4% vs. 7.9% vs. 7.2%, P < 0.0001). Patients with O2-dependent COPD were also more likely to have preoperative anemia (hemoglobin <12gm/dL) (17.0% vs. 18.2% vs. 25.1%, P < 0.0001) and be on preoperative anticoagulation (11.9% vs. 14.1% vs. 16.0%, P < 0.0001). Rates of preexisting chronic kidney disease and prior aortic repair were comparable between all groups. Nonelective index admission, a proxy for rupture or an acute aortic syndrome, was significantly higher in the O2-dependent COPD group (11.0% vs. 9.9% vs. 11.7%, P = 0.02).
Table I.
Demographic and baseline characteristics of patients with or without COPD who have undergone EVAR
| No COPD |
Medication-treated COPD |
O2-dependent COPD |
||
|---|---|---|---|---|
| Covariate | (N = 23,260, 75.8%) | (N = 6,057, 19.7%) | (N = 1,369, 4.5%) | P value |
| Age, years | 73.3 ± 8.7 | 72.9 ± 7.7 | 74.0 ± 7.7 | <0.001 |
| Female sex | 3,702 (15.9) | 1,399 (23.1) | 418 (30.5) | <0.001 |
| Non-white race | 2434 (10.5) | 531 (8.8) | 87 (6.4) | <0.001 |
| Hispanic | 696 (3.0) | 91 (1.5) | 22 (1.6) | <0.001 |
| BMI | 28.4 ± 5.4 | 28.2 ± 5.9 | 29.0 ± 7.2 | <0.001 |
| Smoking status | ||||
| Never | 4,164 (17.9) | 293 (4.8) | 48 (3.5) | <0.001 |
| Former | 13,071 (56.2) | 3,345 (55.2) | 962 (70.3) | <0.001 |
| Current | 6,025 (25.9) | 2,419 (39.9) | 359 (26.2) | <0.001 |
| Nonelective admission | 2,558 (11.0) | 598 (9.9) | 160 (11.7) | 0.024 |
| Hypertension | 19,078 (82.0) | 5,122 (84.6) | 1,168 (85.3) | <0.001 |
| Diabetes mellitus | 4,523 (19.4) | 1,272 (21.0) | 333 (24.3) | <0.001 |
| Coronary artery disease | 6,108 (26.3) | 1,948 (32.2) | 463 (33.8) | <0.001 |
| Congestive heart failure | 2,014 (8.7) | 951 (15.7) | 337 (24.6) | <0.001 |
| Creatinine >1.5 or on dialysis | 2,392 (10.3) | 638 (10.5) | 162 (11.8) | 0.176 |
| Hemoglobin <12 mg/dL | 3,963 (17.0) | 1,103 (18.2) | 343 (25.1) | <0.001 |
| Preoperative aspirin | 15,192 (65.3) | 3,981 (65.7) | 899 (65.7) | 0.818 |
| Preoperative statin | 16,580 (71.3) | 4,409 (72.8) | 1,002 (73.2) | 0.030 |
| Preoperative anticoagulation | 2,573 (11.9) | 797 (14.1) | 202 (16.0) | <0.001 |
| Prior aneurysm repair | 740 (3.2) | 195 (3.2) | 47 (3.4) | 0.872 |
| Family history of AAA | 2176 (9.4) | 479 (7.9) | 99 (7.2) | <0.001 |
Maximum mean aneurysm diameter at the time of repair varied significantly between groups, with the medication-treated COPD having the smallest mean aneurysm diameter and O2-dependent COPD patients having the largest (56.5 mm vs. 56.1 mm vs. 57.0 mm, P = 0.0001) (Table II). The incidence of being deemed unfit for open repair was 3-fold higher among O2-dependent COPD patients (11.9% vs. 18.3% vs. 34.2%, P < 0.0001). Furthermore, patients with O2-dependent COPD were more likely to undergo regional or local anesthesia instead of general anesthesia (8.4% vs. 10.2% vs. 17.6%, P < 0.0001). Procedure times were also significantly different, with O2-dependent COPD patients undergoing longer operations (127 min vs. 130 min vs. 134 min, P = 0.0001).
Table II.
Anatomic and operative characteristics of patients with and without COPD who have undergone EVAR
| No COPD |
Medication-treated COPD |
O2-dependent COPD |
||
|---|---|---|---|---|
| Covariate | (N = 23,260, 75.8%) | (N = 6,057, 19.7%) | (N = 1,369, 4.5%) | P value |
| Maximum AAA diameter, mm | 56.5 ± 12.1 | 56.1 ± 10.6 | 57.0 ± 10.5 | <0.001 |
| Iliac aneurysm | 5,973 (25.7) | 1,402 (23.1) | 312 (22.8) | <0.001 |
| Unfit for open repair | 2,776 (11.9) | 1,106 (18.3) | 468 (34.2) | <0.001 |
| Local/regional anesthesia | 1,960 (8.4) | 620 (10.2) | 241 (17.6) | <0.001 |
| Aortic main device | ||||
| Bifurcated infrarenal | 12,256 (52.7) | 3,209 (53.0) | 739 (54.0) | 0.620 |
| Bifurcated suprarenal | 3,062 (13.2) | 888 (14.7) | 166 (12.1) | 0.003 |
| Aortoiliac | 7,592 (32.6) | 1,877 (31.0) | 446 (32.6) | 0.049 |
| Aortoaortic | 350 (1.5) | 83 (1.4) | 18 (1.3) | 0.658 |
| Procedure time, min | 127 ± 64.2 | 130 ± 67.3 | 134 ± 70.4 | <0.001 |
| Conversion to open repair | 16 (0.1) | 1 (0.0) | 1 (0.0) | 0.316 |
Univariable Associations with Sac Growth, Reintervention, and Mortality
In unadjusted analysis, any diagnosis of treated COPD was statistically associated with postoperative sac diameter changes; however, the absolute differences were small since all patient groups had comparable mean sac regression (mean sac change −5.0 mm non-COPD vs. −5.5 mm medication-treated COPD vs. −5.0 mm O2-dependent COPD, P < 0.001) (Table III). In addition, subgroup analyses were performed among patients experiencing sac growth >5 mm and >10 mm during follow-up. Patients with sac growth >5 mm were less likely to have medication-treated COPD, although rates were similar in the non-COPD and O2-dependent COPD groups (6.5% vs. 5.4% vs. 6.4%, P = 0.006). However, there was no significant difference in the proportion of patients experiencing sac growth >10 mm (2.7% vs. 2.2% vs. 2.9%, P = 0.08).
Table III.
Univariate outcomes of patients with or without COPD who have undergone EVAR
| No COPD |
Medication-treated COPD |
O2-dependent COPD |
||
|---|---|---|---|---|
| Covariate | (N = 23,260, 75.8%) | (N = 6,057, 19.7%) | (N = 1,369, 4.5%) | P value |
| Sac growth (mm) | −5.0 ± 9.6 | −5.5 ± 9.1 | −5.0 ± 9.1 | <0.0001 |
| Sac growth >5 mm | 1,506 (6.5) | 325 (5.4) | 87 (6.4) | 0.006 |
| Sac growth >10 mm | 629 (2.7) | 134 (2.2) | 40 (2.9) | 0.079 |
| Reinterventiona | 1,310 (5.7) | 278 (4.6) | 51 (3.7) | <0.0001 |
| 30-day mortality | 2 (0.0) | 0 (0.0) | 0 (0.0) | <0.0001 |
| 1-year mortality | 144 (0.7) | 70 (1.3) | 31 (2.5) | <0.0001 |
| Any complication | 1,114 (4.8) | 361 (6.0) | 110 (8.1) | <0.0001 |
| Myocardial infarction | 193 (0.8) | 45 (0.7) | 7 (0.5) | 0.379 |
| Dysrhythmia | 401 (1.7) | 124 (2.0) | 33 (2.4) | 0.059 |
| Congestive heart failure | 104 (0.4) | 38 (0.6) | 15 (1.1) | 0.002 |
| Lung complication | 206 (0.9) | 92 (1.5) | 29 (2.1) | <0.0001 |
| Leg ischemia | 133 (0.6) | 61 (1.0) | 10 (0.7) | 0.001 |
| Intestinal ischemia | 76 (0.3) | 19 (0.3) | 7 (0.5) | 0.493 |
| Stroke | 37 (0.2) | 11 (0.2) | 2 (0.1) | 0.917 |
| Nontype II endoleak | 1,624 (7.0) | 413 (6.8) | 106 (7.7) | 0.479 |
| Type II endoleaka | 2,654 (11.4) | 582 (9.6) | 132 (9.6) | <0.0001 |
| Postoperative length of stay, days | 2.2 (4.5) | 2.3 (4.5) | 2.6 (3.5) | <0.0001 |
| Any intensive care unit stay | 7,136 (30.8) | 1,944 (32.3) | 552 (40.4) | <0.0001 |
Represents event occurrence by last known follow-up.
Since there were significant differences between the O2-dependent and medication-treated COPD cohorts in the incidence of sac growth >5 mm, further univariate analysis was performed comparing other characteristics of patients who experienced <5 mm sac growth and patients who experienced >5 mm sac growth. Mean AAA diameter at the time of repair differed (56.7 mm in <5 mm sac growth group vs. 52.3 mm in >5 mm sac growth group, P < 0.001); indicating that smaller aneurysms grew more after repair in this subset. The >5 mm sac growth group was also more likely to have a type II endoleak at last known follow-up (10.0% vs. 26.3%, P < 0.001), be older (73.1 vs. 74.7 years, P < 0.001), and were more likely to be on preoperative anticoagulation (12.2% vs. 18.3%, P < 0.001).
Neither type of treated COPD was associated with increased likelihood of long-term reintervention in univariate analysis (medication-treated COPD, hazard ratio [HR] 0.89, 95% confidence interval [CI] [0.80–1.00], P = 0.052 and O2-dependent COPD, HR 0.86, 95% CI [0.692–1.08], P = 0.189). Both types of COPD were associated with increased mortality (medication-treated COPD, HR 1.5, 95% CI [1.37–1.65], P < 0.001 and O2-dependent COPD, HR 2.4, 95% CI [2.05–2.72], P < 0.001).
Multivariable Associations with Sac Growth
Multivariable analysis of sac growth as a continuous variable was performed, with covariates as described in the methods section (Table IV). Non-COPD patients were used as the reference group. Medication-dependent COPD was not associated with sac growth compared with non-COPD (mean difference: −0.06 mm, 95% CI [−0.32 to 0.199], P = 0.64); however, O2-dependent COPD was associated with postoperative AAA sac growth after EVAR (mean difference: +0.55 mm, 95% CI [0.046–1.05], P = 0.03).
Table IV.
Multivariable analysis of COPD association with sac growth, reintervention, and all-cause mortality, with covariate data provided in supplemental tables
| Sac growtha |
Reinterventionb |
Mortalityc |
||||
|---|---|---|---|---|---|---|
| Covariate | HR [95% CI] | P-value | HR [95% CI] | P-value | HR [95% CI] | P-value |
| Non-COPD | REF | REF | REF | |||
| Medication-treated COPD | −0.06 [−0.322,0.199] | 064 | 0.92 [0.816,1.03] | 0.13 | 1.4[1.25, 1.52] | <0.001 |
| O2-dependent COPD | 0.55[0.046,1.05] | 0.03 | 0.82[0.652, 1.02] | 0.07 | 1.8[1.52.2.22] | <0.001 |
Results of linear regression model adjusted for patient and procedure factors shown in Supplemental Table I.
Results of a competing risks regression model adjusted for patient and procedure factors shown in Supplemental Table I and the competing risk of death.
Results of a Cox proportional hazards model adjusted for patient and procedure factors shown in Supplemental Table I.
Additional multivariable analysis was performed with the dependent variable of sac growth stratified into discrete binary categories of >5 mm versus <5 mm and >10 mm versus <10 mm (Supplemental Table II). Neither type of COPD was significantly associated with sac growth >5 mm or >10 mm. Factors associated sac growth >10 mm included preoperative history of prior AAA repair (odds ratio (OR) 1.6, 95% CI [1.09–2.30], P = 0.016), anemia (OR 1.2, 95% CI [1.03–1.48], P = 0.025), chronic kidney disease (OR 1.3, 95% CI [1.06–1.63], P = 0.012), and presence of concomitant iliac aneurysm (OR 1.4, 95% CI [1.23–1.69], P < 0.001).
Multivariable Associations with Reintervention
The effects of COPD on the likelihood of long-term reintervention were evaluated with a competing risks model, adjusting for patient and procedure factors as well as the competing risk of mortality. Cumulative incidence of reintervention and mortality by COPD group is shown in Figure 1. After adjustment for patient risk factors and the competing risk of mortality, O2-dependent COPD did not impact freedom from reintervention (HR 0.82, 95% CI [0.652–1.02], P = 0.073), nor did medication-treated COPD (HR 0.92, 95% CI [0.816–1.03], P = 0.130) (Table IV). Select covariates that were significantly associated with increased reintervention risk included nonelective admission (HR 1.7, 95% CI [1.44–1.90], P ≤ 0.001), female sex (HR 1.27, 95% CI [1.8–1.35], P = 0.001), transfer prior to admission (HR 1.3, 95% CI [1.11–1.56], P = 0.002), prior aortic surgery (HR 1.4, 95% CI [1.04–1.83], P = 0.025), and preoperative hemoglobin <12 mg/dL (HR 1.3, 95% CI [1.20–1.50], P ≤ 0.001). (Supplemental Table III).
Fig. 1.
Competing risks analysis curve of reintervention and survival in patients who have undergone EVAR.
Multivariable Associations with AllCause Mortality
After adjustment for patient risk factors, both medication-treated and O2-dependent COPD were associated with a 1.4x and 1.8x increase in mortality, respectively (HR 1.4, 95% CI [1.25–1.52], P < 0.001 and HR 1.8, 95% CI [1.52–2.22], P < 0.001) (Table IV). Select covariates associated with increased mortality included underweight status (BMI < 18.5: HR 1.4, 95% CI [1.12–1.84], P = 0.004), hemoglobin<12 mg/dL (HR 1.5, 95% CI [1.35–1.70], P ≤ 0.001) and being deemed unfit for open repair (HR 1.5, 95% CI [1.23–1.86], P ≤ 0.001). Notably, obesity was protective at almost all levels (Supplemental Table IV).
DISCUSSION
This study is among the first to examine COPD’s association with post-EVAR AAA sac growth and reintervention. Moreover, it elucidates some distinct differences in the postoperative natural history between O2-dependent and medication-treated COPD. This work offers 2 novel findings that may help anticipate EVAR outcomes in COPD patients. First, O2-dependent COPD does not appear to be associated with reintervention because reintervention among severe COPD patients is likely precluded by the associated longitudinal risk of increased allcause mortality. Second, O2-dependent COPD is significantly associated with an increased incidence of post-EVAR AAA sac diameter growth over time.
Previous research by our group explored the association between major adverse events and COPD, in patients stratified by severity; however, it did not specifically address reintervention or post-EVAR AAA sac remodeling.14 These findings also confirmed that EVAR is more commonly offered to COPD patients and that those with O2-dependent COPD have greater perioperative mortality, more major adverse events, and increased postoperative pulmonary complications. Our study builds on those prior findings by focusing on changes in AAA sac diameters, reintervention, and all-cause longer-term mortality.
Similarly, Qureshi et al. documented increased early mortality rates among COPD patients undergoing EVAR,15 which was largely attributed to pulmonary-related causes, while nonpulmonary mortality rates were comparable to subjects without COPD. Their study, which included 905 patients with a mean follow-up of 39.5 months, demonstrated similar reintervention rates between COPD and non-COPD patients. Our study’s results again align with these historical observations, confirming no significant differences in reintervention rates, but adding new insights about the impact of treated COPD on post-EVAR sac growth.
Notably, in Figure 1, our Kaplan-Meier analysis highlights that reintervention is more common in the 2-year period following EVAR, while mortality rates among COPD patients increase sharply between 1 and 2 years postoperatively. This supports our hypothesis that increased risk of mortality may lead to clinical decisions to delay or withhold reintervention surgery, even in the setting of documented sac growth. It is plausible that severe COPD, which is often progressive, might render patients less physiologically amenable to subsequent reintervention, prompting surgeons to defer additional procedures despite radiographic evidence of postoperative sac expansion. Although catheterbased interventions may be performed under moderate sedation, which poses less risk than general anesthesia and open surgical procedures, there is still a two-fold increase in procedural sedation adverse events in COPD patients, which may drive hesitancy around even minimally invasive procedures (Berzin, 2011 #31).
Importantly, Quereshi et al. also investigated sac growth following EVAR but did not stratify their analysis by COPD severity. They interestingly reported greater overall sac regression in the COPD group in distinction to our study’s findings. In contrast, our analysis revealed that persistent sac growth was associated only with the O2-dependent COPD group, underscoring the importance of COPD severity. These investigators attributed sac regression to smoking-related athero-occlusive lumbar artery disease, which may limit the development of type II endoleak. However, even when controlling for smoking history, our analysis demonstrated that patients with O2-dependent COPD still experienced an increased likelihood of sac expansion.17
Additional findings from this study suggest that surgeons may be more selective in considering EVAR among COPD patients. For example, COPD patients undergoing EVAR tend to be older, more likely to present with ruptured aneurysms, and have larger AAA diameters at the time of the procedure. This suggests that surgeons may be appropriately cautious in recommending elective procedures for COPD patients due to the higher perioperative morbidity and its impact on overall survival. The Society for Vascular Surgery guidelines do not address the prudence of offering EVAR in limited life expectancy, while the European Society for Vascular Surgery recommendation 67 explicitly does not recommend offering EVAR to patients with live expectancy less than 2–3 years.18 O2-dependent COPD patients have a range of reported life-expectancies from onset of oxygen therapy, with literature suggesting 2–5-year mean survival. Thus, some patients with O2-dependent COPD may be borderline candidates for elective EVAR based on life expectancy, as reported in prior VQI literature.19
Furthermore, because of the high physiological risk in some subgroups of COPD patients, general anesthesia is less frequently used. Instead, regional anesthetic strategies were more commonly employed when compared to subjects without COPD, a trend that has been facilitated in the era of percutaneous EVAR. Moreover, regional or local anesthesia is known to reduce pulmonary complications after EVAR, making it a potentially appealing option for O2-dependent COPD patients who face high aneurysm-related mortality risks.20
A source of ongoing clinical controversy following EVAR is the high prevalence of type II endoleaks and their long-term clinical significance. In the VQI, patients with type II endoleak were found to have no difference in 5-year mortality, and patients that underwent reintervention for endoleak did not have improved survival; however, other studies have demonstrated increased mortality over time.21,22 Our work also examined the incidence of type 2 endoleak and found that it is independently associated with sac growth, irrespective of COPD status. Additionally, we examined mortality among patients designated to be unfit for open repair by their operative surgeon. Specifically in this cohort, we found that these patients had a 50% increased risk of long-term post-EVAR mortality. This finding aligns with previous literature and highlights the need for caution when considering EVAR for patients who are not candidates for open repair.23
This study has several limitations that should be considered when interpreting the findings. First, 45.5% (N = 25,693 of 56,379) of patients who met initial inclusion criteria were ultimately excluded from the analysis because postsurgery sac growth data were unavailable. This may represent a source of bias. Second, because the analysis was conducted on a cohort of patients who had already undergone EVAR, we were unable to include a control group of patients with severe COPD managed without surgery. Consequently, the study may not fully capture the impact of the most moribund COPD cases on AAA outcomes, which could influence our observations. Moreover, COPD patients might be overrepresented among the population of patients undergoing EVAR due to the perceived increased associated perioperative risks of open surgery, such as prolonged intubation and pneumonia. In addition, registry data often provide reasonable quantitative insights on secular trends in care, but lacks more granular details, such as unique anatomical considerations or subtle clinical features, that are critical in the decision-making process between EVAR and other treatment options. Thus, the limitations of the data may impact the study’s ability to fully define the nuances of COPD severity and post-EVAR observations.
CONCLUSION
O2-dependent COPD was significantly associated with increased sac growth after EVAR. However, O2-dependent COPD did not correlate with an increased risk of reintervention, likely due to the disproportionately elevated mortality rates in this high-risk patient group. Additionally, our study suggests that surgeons may be appropriately deferring EVAR in patients with severe COPD. These findings justify a cautious clinical approach to potentially defer EVAR among the most moribund COPD patients and reserve intervention for those who are at the greatest risk of AAA-related mortality.
Supplementary Material
Funding:
KS Bellamkonda was supported by the Veterans Affairs Chief Resident for Quality and Safety program. Confiicts of interest: CI Ochoa Chaar is a consultant for EnVVEno medical and has IP USSN 10,524,891 B1. Meeting presentation: This work was presented as a poster at the Vascular Annual Meeting, 2023.
Footnotes
CREDIT AUTHORSHIP CONTRIBUTION STATEMENT
Ochoa Chaar: Writing e review & editing, Writing e original draft, Conceptualization. Kirthi S. Bellamkonda: Writing e review & editing, Writing e original draft, Conceptualization. Bjoern D. Suckow: Writing e review & editing. Jesse A. Columbo: Writing e review & editing. Gilbert R. Upchurch: Writing e review & editing. Benjamin Jacobs: Writing e review & editing. Cassius I. Ochoa Chaar: Writing e review & editing. Rebecca E. Scully: Writing e review & editing. Philip P. Goodney: Writing e review & editing. Salvatore T. Scali: Writing e review & editing, Supervision, Conceptualization. David H. Stone: Writing e review & editing, Supervision, Conceptualization.
SUPPLEMENTARY DATA
Supplementary data related to this article can be found at https://doi.org/10.1016/j.avsg.2025.02.019.
REFERENCES
- 1.Adeloye D, Song P, Zhu Y, et al. Global, regional, and national prevalence of, and risk factors for, chronic obstructive pulmonary disease (COPD) in 2019: a systematic review and modelling analysis. Lancet Respir Med 2022;10:447–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Müller V, Miszczuk M, Althoff CE, et al. Comorbidities associated with large abdominal aortic aneurysms. Aorta 2019;7: 108–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Takagi H, Umemoto T, Group A. A meta-analysis of the association of chronic obstructive pulmonary disease with abdominal aortic aneurysm presence. Ann Vasc Surg 2016;34:84–94. [DOI] [PubMed] [Google Scholar]
- 4.Bengtsson H, Bergqvist D, Ekberg O, et al. A population based screening of abdominal aortic aneurysms (AAA). Eur J Vasc Surg 1991;5:53–7. [DOI] [PubMed] [Google Scholar]
- 5.Meijer CA, Kokje VB, van Tongeren RB, et al. An association between chronic obstructive pulmonary disease and abdominal aortic aneurysm beyond smoking: results from a casee control study. Eur J Vasc Endovasc Surg 2012;44:153–7. [DOI] [PubMed] [Google Scholar]
- 6.Van Laarhoven C, Borstlap A, van Berge Henegouwen D, et al. Chronic obstructive pulmonary disease and abdominal aortic aneurysms. Eur J Vasc Surg 1993;7:386–90. [DOI] [PubMed] [Google Scholar]
- 7.Lindholt JS, Heickendorff L, Antonsen S, et al. Natural history of abdominal aortic aneurysm with and without coexisting chronic obstructive pulmonary disease. J Vasc Surg 1998;28:226–33. [DOI] [PubMed] [Google Scholar]
- 8.Choke E, Cockerill G, Wilson W, et al. A review of biological factors implicated in abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg 2005;30:227–44. [DOI] [PubMed] [Google Scholar]
- 9.Lederle FA, Wilson SE, Johnson GR, et al. Immediate repair compared with surveillance of small abdominal aortic aneurysms. New Engl J Med 2002;346:1437–44. [DOI] [PubMed] [Google Scholar]
- 10.Cronenwett J, Murphy T, Zelenock G, et al. Actuarial analysis of variables associated with rupture of small abdominal aortic aneurysms. Surgery 1985;98:472–83. [PubMed] [Google Scholar]
- 11.Chaikof EL, Dalman RL, Eskandari MK, et al. The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vasc Surg 2018;67: 2–77.e2. [DOI] [PubMed] [Google Scholar]
- 12.Compton CN, Dillavou ED, Sheehan MK, et al. Is abdominal aortic aneurysm repair appropriate in oxygen-dependent chronic obstructive pulmonary disease patients? J Vasc Surg 2005;42:650–3. [DOI] [PubMed] [Google Scholar]
- 13.Jonker FH, Schlösser FJ, Dewan M, et al. Patients with abdominal aortic aneurysm and chronic obstructive pulmonary disease have improved outcomes with endovascular aneurysm repair compared with open repair. Vascular 2009;17:316–24. [DOI] [PubMed] [Google Scholar]
- 14.Stone DH, Goodney PP, Kalish J, et al. Severity of chronic obstructive pulmonary disease is associated with adverse outcomes in patients undergoing elective abdominal aortic aneurysm repair. J Vasc Surg 2013;57:1531–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Qureshi MA, Greenberg RK, Mastracci TM, et al. Patients with chronic obstructive pulmonary disease have shorter survival but superior endovascular outcomes after endovascular aneurysm repair. J Vasc Surg 2012;56: 911–919.e2. [DOI] [PubMed] [Google Scholar]
- 16.Azofra EA, Marcos FÁ, Prendes CF, et al. Predictive factors of aneurysm sac growth in patients with a type II endoleak in the first post-EVAR control. Ann Vasc Surg 2020;68: 245–51. [DOI] [PubMed] [Google Scholar]
- 17.Higashiura W, Greenberg RK, Katz E, et al. Predictive factors, morphologic effects, and proposed treatment paradigm for type II endoleaks after repair of infrarenal abdominal aortic aneurysms. J Vasc Interv Radiol 2007;18:975–81. [DOI] [PubMed] [Google Scholar]
- 18.Wanhainen A, Van Herzeele I, Goncalves FB, 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;67:192–331. [DOI] [PubMed] [Google Scholar]
- 19.De Martino RR, Goodney PP, Nolan BW, et al. Optimal selection of patients for elective abdominal aortic aneurysm repair based on life expectancy. J Vasc Surg 2013;58: 589–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Van Orden K, Farber A, Schermerhorn ML, et al. Local anesthesia for percutaneous endovascular abdominal aortic aneurysm repair is associated with fewer pulmonary complications. J Vasc Surg 2018;68: 1023–1029.e2. [DOI] [PubMed] [Google Scholar]
- 21.Seike Y, Matsuda H, Shimizu H, et al. Nationwide analysis of persistent type II endoleak and late outcomes of endovascular abdominal aortic aneurysm repair in Japan: a propensity-matched analysis. Circulation 2022;145: 1056–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.DeMartino RR, Breite MD, Neal D, et al. Incidence, reintervention, and survival associated with type II endoleak at hospital discharge after elective endovascular aneurysm repair in the Vascular Quality Initiative. J Vasc Surg 2023;78:679–686.e1. [DOI] [PubMed] [Google Scholar]
- 23.De Martino RR, Brooke BS, Robinson W, et al. Designation as ‘‘unfit for open repair’’ is associated with poor outcomes after endovascular aortic aneurysm repair. Circ Cardiovasc Qual Outcomes 2013;6:575–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.