A, Cumulative survival after endovascular aneurysm repair in patients with or without systemic inflammatory disease (log rank, P = .99; Gehan-Breslow-Wilcoxon; P = .86). B, Increased type II endoleaks and late sac expansion in patients with systemic inflammatory disease. Endoleaks: P = .03 (log-rank test) and hazard ratio of 3.179 (95% CI, 1.203-6.210). Late sac expansion: P = .02 (log-rank test) and hazard ratio of 6.087 (95% CI, 1.037-9.380). The slightly reduced numbers of patients at year 0 reflect patients with an endoleak at the completion of the operative procedure who were not included. The solid line indicates sac expansion; dashed line, endoleak. C, Decreased sac shrinkage in patients with systemic inflammatory disease. P < .001 (log-rank test) and hazard ratio of 0.106 (95% CI, 0.032-0.350). D, Sac size shrinkage percentage in patients with and without systemic inflammatory disease at 1 and 5 years after endovascular aneurysm repair. Patients without autoinflammatory disease had significant sac size shrinkage after 5-year follow-up compared with first-year follow-up computed tomographic scan. The error bars indicate SEM.
aP < .05 (Bonferroni post hoc analysis; P < .001).
eTable 1. Operative Characteristics
eTable 2. Data Supporting Figure, A
eTable 3. Medications Associated With Significant Endoleak After EVAR
eTable 4. Data Supporting Figure, B
eTable 5. Univariable Analysis of Systemic Factors Associated With Significant Endoleak After EVAR
eTable 6. Data Supporting Figure, C
Sherif Y. Shalaby, Trenton R. Foster, Michael R. Hall, Kirstyn E. Brownson, Penny Vasilas, Daniel G. Federman, Hamid R. Mojibian, Alan Dardik. Systemic Inflammatory Disease and Its Association With Type II Endoleak and Late Interventions After Endovascular Aneurysm Repair. JAMA Surg. 2016;151(2):147–153. doi:10.1001/jamasurg.2015.3219
Abdominal aortic aneurysms are associated with chronic inflammation within the aortic wall, and previous studies have suggested that chronic inflammation may be a consequence of a dysregulated and persistent autoimmune response. Persistent aortic remodeling after aneurysm repair could place the patient at risk for endoleak or sac rupture.
To determine whether patients with systemic inflammatory disease and large aneurysms have persistent aortic remodeling after endovascular aneurysm repair (EVAR).
Design, Setting, and Participants
The records of all patients who underwent EVAR between July 2002 and June 2011 at the Veterans Affairs Connecticut Healthcare System were included in this retrospective review. Patients were considered to have a systemic inflammatory disease when confirmed by a referring specialist. Post-EVAR surveillance was performed by yearly imaging.
Endovascular aneurysm repair.
Main Outcomes and Measures
Significant endoleak, defined as endoleak and sac diameter increase of 0.5 cm or greater.
A total of 51 of 79 patients (65%) had a systemic inflammatory disease. These patients had similar comorbid conditions compared with patients without inflammation but significantly greater numbers of major postoperative complications after EVAR (23.5% vs 3.6%; P = .02) and overall postoperative complications after EVAR (27.5% vs 7.1%; P = .03). Patients with a history of systemic inflammatory disease developed more endoleaks (45.1% vs 17.9%; P = .02) and late sac expansion (51.0% vs 21.4%; P = .01) and required more interventions (21.6% vs 3.6%; P = .03) during long-term follow-up. Systemic inflammatory disease was significantly associated with significant endoleak (odds ratio, 5.18; 95% CI, 1.56-17.16; P = .007).
Conclusions and Relevance
Patients with systemic inflammatory disease are at high risk for postoperative complications, type II endoleak, sac expansion, and additional interventions after EVAR. Additional strategies for improving the efficacy of EVAR in these patients may be warranted.
Abdominal aortic aneurysm (AAA) is common, with a prevalence of 3% in people older than 60 years.1 The pathogenesis of AAA is associated with extracellular matrix remodeling, vascular smooth muscle cell apoptosis and necrosis, oxidative stress, and infiltration of inflammatory cells, often associated with connective tissue disease, resulting in aortic wall degeneration.2- 6 Several studies have suggested that chronic inflammation in the aortic wall may be a consequence of a dysregulated and persistent autoimmune inflammatory response against aortic wall antigens.7,8
Systemic inflammatory diseases have been characterized as forming a spectrum of inflammatory and autoimmune components, with some patients having largely autoinflammatory subtypes, such as inflammatory bowel disease, osteoarthritis, or gout; some patients having largely autoimmune subtypes, such as lupus, type 1 diabetes mellitus, or Goodpasture syndrome; and some patients having mixed subtypes, such as ankylosing spondylitis or psoriasis.9- 12 A common mechanism of these diseases includes overproduction of various inflammatory cytokines, such as interleukin 1, interleukin 18, interleukin 6, tumor necrosis factor–α, and type I interferon.13 These cytokines are produced by activated immune cells, including autoreactive T lymphocytes and plasma cells, that result in the formation of autoantibodies and inflammation.14 Interleukin-17–producing Th(17) T cells may promote inflammation and tissue destruction in autoimmune diseases and may also confer susceptibility to allergy.15,16 Because autoinflammatory, autoimmune, and allergic diseases have common pathophysiology and are considered a disease spectrum, they are now being treated with similar agents.17 In addition, localized autoimmune diseases, such as Sjögren syndrome, are frequently associated with systemic inflammatory diseases, highlighting the systemic nature of seemingly localized inflammation.18
It is possible that systemic inflammatory disease produces an environment of persistent aortic inflammation and remodeling, resulting in aneurysmal degeneration; it is also possible that aortic aneurysms continue to remodel after endovascular aneurysm repair (EVAR) in the systemic inflammatory environment, as EVAR does not remove the aortic wall. We hypothesized that patients with systemic inflammatory disease and large aneurysms may have persistent aortic inflammation and remodeling after EVAR, potentially decreasing the efficacy of this minimally invasive procedure. Accordingly, we examined the outcome after EVAR in veterans, a group of patients with more comorbid diseases compared with nonveterans with similar diagnoses, forming a high-risk population.19,20 We previously reported that veterans with metabolic syndrome (MetS) have more type II endoleaks after EVAR,21 and we reanalyzed these data to determine whether these patients had systemic inflammatory diseases and how these systemic inflammatory diseases affect remodeling after EVAR.
The medical records of all patients who underwent EVAR between July 2002 and June 2011 at the Veterans Affairs Connecticut Healthcare System were reviewed with approval of the system’s institutional review board, waiving informed consent for individual patients because patient-identifiable information was not recorded in this retrospective review. The presence of a systemic inflammatory disease was defined by diagnosis and/or treatment of an inflammatory disease. The presence of an autoinflammatory disease was defined by diagnosis of osteoarthritis, gout, pseudogout, or inflammatory bowel disease.14,22,23 The presence of allergic disease was defined by allergic rhinitis, eczema, or seasonal rhinitis. Medical management prescribed by a rheumatologist or other disease-related specialist included any active or prior corticosteroid use, via oral or injection route of administration, as well as medications used to specifically treat autoimmune disease.
Patient demographics, comorbid conditions, and laboratory data were determined at the time just prior to EVAR. Heart disease was defined as the presence of coronary artery disease, congestive heart failure, or atrial fibrillation; previous vascular intervention was defined as a history of vascular open or endovascular procedure. Metabolic syndrome was defined using the modified National Cholesterol Education Program and World Health Organization criteria (eg, having ≥3 of the following criteria: diagnosis of hypertension [systolic blood pressure >130 mm Hg or diastolic blood pressure >90 mm Hg on 3 measurements during at least a 6-month period] or significant pharmacologic antihypertensive medication use; at least 1 measured serum triglyceride level >150 mg/dL [to convert to millimoles per liter, multiply by 0.0113] and pharmacologic treatment; a low serum level of high-density lipoprotein level measuring <50 mg/dL for women or <40 mg/dL for men [to convert to millimoles per liter, multiply by 0.0259]; a fasting blood glucose level of >110 mg/dL [to convert to millimoles per liter, multiply by 0.0555] or a diagnosis of diabetes mellitus requiring pharmacologic treatment; and a body mass index [BMI, calculated as weight in kilograms divided by height in meters squared] of >30 at the time of the intervention).19 The MetS score was defined as the number of MetS parameters present in an individual patient and ranged from 0 to 5, and a MetS score of 3 or greater was considered to indicate the presence of MetS.
Operative measures included aneurysm size, defined as the largest aneurysm diameter on computed tomographic (CT) scan prior to intervention; planned perioperative interventions, defined as planned adjuncts to the EVAR procedure that would be performed before, during, or after the procedure, including coil embolization, bypass, or iliac limb extension; and urgent presentation, defined as a symptomatic or ruptured aneurysm. Endoleaks were defined using the Society for Vascular Surgery criteria.24 Excluder (W. L. Gore and Associates), AneuRx (Medtronic), Zenith (Cook), Talent (Medtronic), Endologix, Endurant (Medtronic), and Ancure (Guidant Endovascular Solutions) grafts were used at the discretion of the surgeon during the time of the study.
Perioperative mortality and morbidity rates were determined within 30 days postoperatively. Major complications included myocardial infarction, ischemic colitis, atrial fibrillation, acute kidney injury, temporary or permanent dialysis, acute respiratory distress syndrome, blood stream infection, or spinal cord ischemia. Total complications also included wound infection, buttock claudication, deep vein thrombosis, or pulmonary embolism. Perioperative myocardial ischemia was determined by a serum troponin rise greater than the normal range or electrocardiogram changes consistent with ischemia. Acute kidney injury was defined as a creatinine rise of more than 0.3 mg/dL (to convert to micromoles per liter, multiply by 88.4) or 1.5 times the preoperative baseline.
Surveillance after EVAR was performed by annual radiologic imaging by CT, magnetic resonance imaging, or duplex unless the sac diameter diminished to less than 3 cm, in which case the frequency of surveillance was lengthened at the discretion of the surgeon, typically to every 5 years. Complications related to EVAR included sac rupture, stent fracture, migration, and endoleak. Significant endoleak was defined as an endoleak associated with increased sac diameter. Sac shrinkage or expansion was reported with a change of 0.5 cm or greater compared with previous CT imaging.
Data were recorded as mean (SEM) and compared using the unpaired t test; categorical data were compared using the χ2 test. Multivariable analysis was performed using logistic regression including variables that were selected a priori for clinical significance. Survival data were analyzed with Kaplan-Meier statistics and compared using the log-rank and Gehan-Breslow-Wilcoxon tests. P values less than .05 were considered signiﬁcant. Statistical analysis was conducted using GraphPad Prism and Minitab.
To determine whether patients with advanced aneurysmal disease also had a history of systemic inflammatory disease, the records of patients treated with EVAR at the VA Connecticut between 2002 and 2011 were reviewed. A total of 51 of 79 patients (65%) had 1 or more diagnoses of systemic inflammatory disease, including autoinflammatory, autoimmune, and allergic disease subtypes (Table 1). Patients frequently had multiple different systemic inflammatory diseases, with 27 patients (53%) having only 1 diagnosis, 18 (35%) having 2 diagnoses, and 6 (12%) having 3 diagnoses; all patients received treatment for their diagnoses. Patients with systemic inflammatory disease had similar demographics and comorbid conditions to patients undergoing EVAR but without a history of systemic inflammatory disease, other than increased BMI and increased need for more than 1 antihypertensive medication (Table 2).
Patients with systemic inflammatory disease had a similar mean aneurysm diameter at the time of EVAR compared with patients without a history of systemic inflammatory disease; in patients with systemic inflammatory disease, there were increased numbers of planned perioperative interventions associated with the EVAR procedure (eg, coil embolization) but this was not statistically significant (eTable 1 in the Supplement). Only 1 patient presented urgently with symptoms consistent with a contained rupture; all other patients were asymptomatic and treated electively. Endoleaks (type II) at the completion of the operative procedure were noted at similar frequency between patients with and without inflammatory disease (eTable 1 in the Supplement).
Perioperative (<30 days) mortality was similar between patients with or without a history of systemic inflammatory disease (Table 3). There was an increased number of early major (P = .02) and overall (P = .03) complications in patients with a history of systemic inflammatory disease (Table 3). No patient had spinal cord ischemia, deep venous thrombosis, or pulmonary embolism postoperatively.
The mean follow-up time after the EVAR was 5.2 years; patients were followed up yearly with imaging, generally CT (97%) or magnetic resonance imaging or duplex (3%). Overall survival was not significantly different between patients with or without systemic inflammatory disease (Figure, A; eTable 2 in the Supplement). Patients with systemic inflammatory disease had a median (SD) survival of 6.9 (0.4) years, which was not different than patients without systemic inflammatory disease (7.5 [0.6] years; P = .99). No patient had a ruptured aneurysm after EVAR.
A total of 43 patients (54.4%) developed endoleaks at any time after EVAR, and patients with systemic inflammatory disease developed more endoleaks (68.7%) compared with patients without systemic inflammatory disease (28.6%; P < .001; Table 3). Endoleaks were classified at 1 year after EVAR; 28 patients (35.4%) had endoleaks, most of which were type II endoleaks. Patients with systemic inflammatory disease had more type II endoleaks after EVAR compared with patients without systemic inflammatory disease (33.3% vs 7.1%; P = .01; Table 3). Medication use was not related to development of an endoleak, although active treatment of gout was less likely to be associated with significant endoleak (eTable 3 in the Supplement). Endoleak with late sac expansion (eg, a significant endoleak) occurred in 32 patients (40.5%) after EVAR and was more frequent in patients with systemic inflammatory disease (51.0% vs 21.4%; P = .01; Table 3). Patients with systemic inflammatory disease more frequently required an intervention for significant endoleak compared with patients without systemic inflammatory disease (21.6% vs 3.6%; P = .03); 11 of 12 type II endoleaks (92%) in all patients that required intervention had a systemic inflammatory disease (Table 3).
Kaplan-Meier analysis showed that endoleak was more likely to occur over time in patients with systemic inflammatory disease (hazard ratio, 3.18; P = .03; log-rank test; Figure, B; eTable 4 in the Supplement); similarly, late sac expansion was also more likely to occur in patients with systemic inflammatory disease (hazard ratio, 6.09; P = .02). Systemic factors that were associated with a significant type II endoleak during long-term follow-up after EVAR included systemic inflammatory disease, several components of MetS, and planned perioperative intervention (eTable 5 in the Supplement). Multivariable logistic regression confirmed that a history of systemic inflammatory disease was associated with the development of a significant endoleak after EVAR (odds ratio, 5.18; 95% CI, 1.56-17.16; P = .007; Table 4). Similarly, sac shrinkage was uniformly present after EVAR in patients without a history of systemic inflammatory disease, but sac shrinkage was much less common in patients with systemic inflammatory disease (Figure, C; eTable 6 in the Supplement). At 1 and 5 years after EVAR, patients without systemic inflammatory disease had significant sac shrinkage compared with patients with systemic inflammatory disease (odds ratio −2.71; P = .002; Figure, D).
We report that in veterans with large AAAs treated with EVAR, 65% of patients had a systemic inflammatory disease. Patients with a systemic inflammatory disease have increased perioperative complications and develop more endoleaks and late sac expansion and require additional interventions during follow-up compared with patients without a systemic inflammatory disease. However, patients with systemic inflammatory disease have similar long-term survival compared with patients without a systemic inflammatory disease. These results suggest that EVAR requires more intense surveillance, more interventions, and more costs to be efficacious in patients with a history of systemic inflammatory disease.
We observed that patients with a history of systemic inflammatory disease developed more endoleaks and late sac expansion after EVAR compared with patients without such a history (Table 3; Figure, B). We have previously reported that patients with MetS develop more type II endoleaks after EVAR.21 However, a history of systemic inflammatory disease may be a more significant predictor of significant endoleak (eg, endoleak and late sac expansion) than MetS (Table 4). Although there may be subtle differences in inflammatory processes between patients with MetS and patients with systemic inflammatory diseases, it is not currently clear whether these differences might be clinically significant. Similarly, a history of smoking was not associated with difference in rates of endoleak (eTable 3 in the Supplement); as such, the importance of particular subtypes of inflammation on the development of endoleak is not clear. The similar mean AAA diameters in patients without and with systemic inflammatory diseases (eTable 1 in the Supplement) suggest that the inflammatory processes associated with systemic inflammatory diseases may be more significant for post-EVAR remodeling compared with the inflammatory processes associated with early AAA development and expansion. However, all these data suggest the importance of inflammation in aortic wall remodeling even after technically successful EVAR.
We also report that patients with systemic inflammatory disease had increased major and total perioperative complications after EVAR (Table 3) despite similar baseline characteristics (Table 2) and operative course (eTable 1 in the Supplement) as patients without systemic inflammatory disease. Patients with systemic inflammatory disease are at increased risk for cardiovascular events compared with the general population; for example, a study identified autoimmune diseases, such as psoriatic arthritis, psoriasis, and rheumatoid arthritis, as increasing the risk for cardiovascular events.25 Patients with autoinflammatory disease, such as gout, also are at increased risk for all-cause and cardiovascular mortality.26 As such, patients with systemic inflammatory disease are at increased baseline risk for cardiovascular events and therefore may be at increased risk for postoperative complications. Patients with systemic inflammatory disease had a higher mean BMI compared with patients without systemic inflammatory disease (Table 2); it is possible that higher BMI may contribute to inflammation.
There were several limitations to this study. The retrospective nature of this study did not allow confirmation of the systemic inflammatory state or its severity of activity with testing such as the erythrocyte sedimentation rate or high-sensitivity C-reactive protein levels; however, there was no difference in the development of endoleak between patients actively treated or not (eg, patients with or without active disease) (eTable 2 in the Supplement). We were also unable to detect some systemic inflammatory diseases, such as type 1 diabetes mellitus, that present early in life, as these conditions would prevent enlistment in the armed forces and thus prevent the patient from seeking care in the VA health care system. In addition, as expected with older veterans, our population consisted exclusively of men; women may have different prevalence of inflammatory and autoimmune diseases and different prevalence of AAA, with different outcomes after EVAR. Lastly, the limitations of this retrospective report need confirmation in prospective studies; serial examination of post-EVAR proximal neck diameter changes would likely be more meaningful in a group of patients treated with uniform types of proximal fixation and sealing.
Patients with systemic inflammatory diseases have increased type II endoleak and late sac expansion, as well as the need for interventions after EVAR, consistent with increased aortic wall remodeling despite technically successful aneurysm exclusion and prevention of rupture. These data suggest that systemic inflammatory disease is a risk factor for failure of EVAR and that patients with systemic inflammatory disease require increased surveillance and maintenance of their endografts, or perhaps adjunctive pharmacological therapy, to maintain procedural success. It is possible that these patients might benefit from alternative strategies to prevent aneurysm rupture.
Corresponding Author: Alan Dardik, MD, PhD, Department of Surgery, Yale University School of Medicine, 10 Amistad St, Room 437, PO Box 208089, New Haven, CT 06520-8089 (email@example.com).
Accepted for Publication: June 18, 2015.
Published Online: October 21, 2015. doi:10.1001/jamasurg.2015.3219.
Author Contributions: Dr Shalaby had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Shalaby, Foster, Hall, Dardik.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Shalaby, Vasilas, Federman, Dardik.
Critical revision of the manuscript for important intellectual content: Shalaby, Foster, Hall, Brownson, Federman, Mojibian, Dardik.
Statistical analysis: Shalaby, Foster.
Obtained funding: Dardik.
Administrative, technical, or material support: Foster, Hall, Brownson, Vasilas, Federman, Dardik.
Study supervision: Federman, Dardik.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported with the resources and the use of facilities at the VA Connecticut Healthcare System, West Haven.
Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Previous Presentation: This study was presented at the 2015 Association of VA Surgeons Annual Meeting; May 5, 2015; Miami Beach, Florida.