Optimizing compliance, efficiency, and safety during surveillance of small abdominal aortic aneurysms
Article Outline
Background
Outcome data documenting safety for observation of small abdominal aortic aneurysms (AAA 4.0 to 5.4 cm) are lacking outside of large clinical trials but requires near perfect patient compliance. This study describes a clinical pathway for AAA surveillance using a prospective database utilizing a nurse practitioner oversight to provide efficient use of clinic visits while maintaining a high level of patent participation.
Methods
Over a 7-year period (June 1999 through June 2006), 334 patients were enrolled in an AAA surveillance pathway at our academic veterans hospital. To minimize patient travel, clinic visitation was reserved for an initial examination with patient education and for discussion of intervention options in patients demonstrating AAA growth (>5.4 cm or expansion >1 cm/yr) during follow-up. Biannual ultrasound or CT imaging was scheduled and results discussed (after physician review) via telephone or “same day” direct patient contact. An electronic database was used to update patient information and plan follow-up.
Results
Compliance with the AAA surveillance pathway was achieved in 98.5% of patients, with only three patients (0.9%) lost to follow-up and two others (0.6%) choosing early repair at civilian institutions. At a mean interval of 29 months (±20 mo), surgical repair was performed in 225 (67%) patients by open (n = 143) or endovascular (n = 82) techniques for AAA growth to >5.4 cm (n = 219) or expansion by >1cm/yr (n = 6). One hundred six patients currently remain in surveillance. A single AAA rupture resulting in death occurred during surveillance (0.3%) and perioperative mortality (<60 days) was 0.9% in patients needing intervention for AAA growth. Cumulative aneurysm-related mortality was 0.9% for patients compliant with the AAA surveillance pathway.
Conclusions
Use of a prospectively-maintained surveillance database managed by a non-physician provider with a reliance on telephone contact resulted in a high degree of patient compliance, reduced unnecessary patient travel, and provided practical clinic use. Limited additional resources were needed to implement our pathway and a similar approach may prove useful for large volume hospital, clinic, or practice systems.
The United Kingdom (UK) Small Aneurysm Trial and the Aneurysm Detection and Management Study (ADAM) provided an initial valuable foundation for recommended management of small abdominal aortic aneurysms (AAA).1, 2, 3, 4, 5 Both trials demonstrated that observation could be safely employed in compliant patients for 4.0 to 5.4 cm diameter AAAs with a low associated annual risk of rupture (0.6%3 to 1.0%2) compared with immediate elective open repair. Outside of prospective trials requiring patient consent at entry to comply with an assigned treatment regime, there are few reports detailing the safety of small AAA surveillance in general medical practice.6, 7 Valentine et al6 showed that failure of patient compliance with surveillance occurred in one-third of a regional veterans population. “Lost to follow-up” was associated with a greater than 10% risk of AAA rupture and a lower overall survival rate compared to compliant patients. Failure of follow-up was due to numerous patient-related factors and physician/medical center deficiencies including transportation/travel distance difficulties, poor patient motivation, inadequate patient counseling, and reliance on primary care for patient education and nonoperative AAA management. Based on similar concerns at our academic Veterans Administration (VA) medical center, in 1999 the vascular surgery division devised and directed a clinical pathway for small AAA surveillance utilizing vascular practitioners to assume management of a prospectively-maintained patient AAA database, provide patient education, and coordinate surveillance imaging with a focus on minimizing clinic follow-up visits by “one-stop” imaging and evaluation, and reliance on telephone communication. This report describes the implementation of an AAA surveillance clinical pathway aimed at improving patient participation while providing satisfactory clinical outcomes for aneurysm related morbidity and mortality.
Methods
Population
The James A. Haley Veterans Hospital (JAHVH) is a 327-bed tertiary care, academic medical center in Tampa, Florida with a “catchment” range of 250 miles spanning eight counties over central Florida and serving a veteran population of approximately 436,000. Prior to 1998, there was no organized method to track small AAAs (<5.5 cm) for patients other than those enrolled in the observation arm of the ADAMS trial (n = 32), and most of ongoing surveillance was relegated to the primary practitioner physician assigned to each veteran’s care.
AAA imaging and measurement
In mid 1999, we implemented a clinical pathway to manage all patients found to have aortic aneurysms within our hospital system. Our vascular surgery service receives between four to eight new consultations for aortoiliac ectasia/aneurysm per month. The initial imaging exam as ordered by the primary VA practitioner was variable and included patients who were being screened for AAA, patients with known AAA but were new to the Veterans Health System, or patients being imaged for other abdominal pathology.
All initial imaging studies (abdominal ultrasound [US], computed tomography [CT], or magnetic resonance imaging [MRI]) were physician reviewed (M.R.B., P.A.A.) in order to arrive at an accurate description of the aneurysm location and size prior to any clinic visitation. Aneurysm size was determined by recording the maximal cross-sectional diameter as measured perpendicular to the direction or tortuosity of the aorta. In general, 3 to 4 cm diameter AAA were followed by yearly US, and 4 to 5.4 cm diameter AAA were followed biannually with US or CT scans. Due to some disparity and variability (up to 10 mm vs CT measures) in US studies interpreted in our hospital’s radiology department,8, 9 at least a single CT scan was performed for all AAA >4.5 cm to confirm accurate aortic sizing and included chest, abdominal and pelvic imaging of the entire aorto-iliac system. Fusiform AAA >5.4 cm, saccular AAA >5.0 cm, iliac aneurysms >4.0 cm, and any AAA with >1.0 cm annual growth documented by CT scan and occurring in adequate medical risk patients were considered for either endovascular (if anatomically feasible) or open repair. Occasional patients with extremely high medical risk and not felt to have anatomy amenable to endovascular repair were imaged at 6-month intervals until AAA growth to 6.0 cm. Almost no patient was denied intervention for a large AAA based solely on severity of comorbidities. Although we performed surveillance for many aortic pathologies including chronic type B dissections, para-anastomotic/recurrent aneurysms, postendovascular repairs, thoracic and thoracoabdominal aneurysm, pararenal aortic and isolated iliac (de novo) aneurysms, and 3 to 4 cm AAA, this report focused on management of 473 veterans (99% male) entering the pathway with infrarenal AAA >4.0 cm. This included 85 (18%) known AAA experiencing growth from a smaller size, 94 (20%) patients with AAA suggested by physical exam, 23 (5%) AAA found as a result of familial screening, and 271 (57%) AAA found incidentally by various imaging studies done for nonvascular symptomatology. Patients presenting on their initial consultation with symptomatic aneurysms or size >5.4 cm who were good operative risk were excluded from final outcome analysis.
Clinical pathway
As part of their clinical duties on our vascular service, an advanced registered nurse practitioner (ARNP) receives, evaluates and responds to all new AAA consults, schedules all imaging studies, and reviews all completed studies with attending vascular surgeons (M.R.B., P.A.A.) to confirm AAA measurements. After imaging review, the surveillance or treatment plan is updated and the ARNP arranges needed clinic visitations, maintains telephone contact with patients during surveillance or intervention planning, and enters and updates patient and AAA information into a database. A comprehensive medical record is available for all veterans through a national computerized patient record system (CPRS) that also permits tracking of AAA consultations and ongoing communication with referring practitioners and primary care physicians. In general, after surgeon imaging review of >4.0 cm AAA imaging studies, initial telephone contact was made with the patient by the ARNP, and brief (<5 min) education of AAA disease and recommended treatment was provided. All patients with AAA >5.4 cm were seen within 2 weeks in clinics by attending surgeons and the contacting ARNP for operative consideration. While not mandatory, patients with 4.0 to 5.4 cm AAA were offered an initial clinic appointment for further education, physical examination, and discussion of planned surveillance. Patient’s family members or close friends were encouraged to attend clinic visits to assure more complete questioning and potential retention of discussed information. Printed AAA educational materials were provided and the patients were given contact phone numbers to the vascular office and ARNP for addressing further issues.
For AAAs continued under surveillance, serial US or CT studies were scheduled by the ARNP either at our hospital or other VA outpatient clinics with imaging capabilities to accommodate potential patient transportation limitations and minimize travel distances. Appointment scheduling was arranged by telephone contact and confirmations mailed to patients. Most of the completed studies could be viewed at a hospital computer workstation of the electronic storage system for imaging performed at most VA-related facilities within the VISN 8 (central Florida) network. After physician review of imaging, study results, further management plan, and surveillance scheduling were conveyed to patients via phone by the ARNP within 1to 2 weeks. For some studies performed at our hospital and immediately available for review, patients were seen in our clinic/office the same day of imaging and results discussed. If a patient failed to return telephone messages left for them or missed an imaging study, a letter was mailed with detailed instructions for further planned follow-up. Further attempts at direct telephone contact were continued at monthly intervals with “lost to follow-up” designated after more than 3 months of failed telephone and mailed contact. Clinic visitations were avoided during surveillance of small AAA unless specifically requested by patients. The AAA surveillance clinical pathway is detailed diagrammatically in Fig 1.
Fig 1.
Abdominal aortic aneurysm surveillance algorithm.
Surveillance database
Using a standard Microsoft Access program (Redmond, Wash), a database template was created that prospectively records patient demographics (name, date of birth, social security number), date of initial study and imaging modality, current AAA diameter, date of next imaging and status of study ordering/scheduling, and a comments box where we have detailed additional imaging findings. Through menu drop-down options specific database queries can be performed including listing of AAA studies due for review and listing of all AAA >5.0 cm (nearing consideration for intervention). The database was queried every 1 to 2 weeks to generate a list of patients due for AAA imaging review and missed appointments could be tracked and rescheduled by the ARNP. The AAA surveillance database was maintained on a shared drive system and could be accessed through any hospital-based computer by log-in of vascular service personnel.
Study design
Prospective information including patient demographics and AAA surveillance history was used along with retrospective medical record review including perioperative reports and clinic visits to analyze late outcomes and causes of death determined. Differences in the presence of medical comorbidities between AAA patients remaining in surveillance, and those who required operative repair according to AAA growth or the development of symptoms were analyzed using χ2 calculations. Noncompliance with the clinical pathway was defined as missing no more than two consecutive imaging appointments as recommended by the vascular surgeon’s surveillance plan. Patients were considered lost to follow-up if they declined AAA surveillance, left the VA Health System, or died. Life table estimates of completeness of follow-up, patient survival, and aneurysm-related mortality were performed by standard methods.
Results
Of 473 patients initially referred with AAA >4.0 cm, 139 had AAA >5.4 cm and were considered for immediate operative repair. Five of these patients had symptomatic intact or contained ruptures and underwent urgent open (n = 4) or endovascular (n = 1) repair. Sixty-four open and 62 endovascular repairs were electively performed for asymptomatic, intact AAAs >5.4 cm in 126 adequate risk patients. Eight extremely high risk patients with asymptomatic AAAs >6.0 cm required open repair for inadequate endovascular anatomy. Perioperative (<60 days) mortality for those requiring immediate repair was 5.8% (n = 8) with two of these deaths involving ruptured AAAs. Overall mortality for immediate repair AAA at our facility during the study period has been 2.5% with the majority of deaths resulting from cardiopulmonary conditions.
The remaining 334 patients presenting with AAA 4.0 to 5.4 cm were entered into the surveillance pathway during the study interval. Mean age (+/− SD) of the overall patient cohort with AAA >4.0 cm was 74 + 8 years (range 50 to 91 years) with no difference in the average age of patients presenting with AAA >5.4 cm and those entering surveillance for AAA 4.0 to 5.4 cm. Table I describes the patient comorbidities for the cohort. Conditions traditionally associated with cardiovascular disease (ie, myocardial infarction, hyperlipidemia, and hypertension) reached statistical significance comparing immediate repairs to patients enrolled in surveillance suggesting that immediate repair patients had poorer control of cardiovascular risk factors as compared to surveillance patients. Duration of surveillance for enrollees ranged from 6 to 120 months (mean 34 months + 18). The mean size of AAA at the time of enrollment into surveillance was 4.6 cm + 5 with 27% of AAA being >5.0 cm upon their initial vascular consultation.
Table I. Patient comorbidities for AAA consultations
| Characteristic | Immediate repair (CT > 5.4 cm) | Surveillance group (CT/US 4.0 – 5.4 cm) | ||
|---|---|---|---|---|
| (n = 473) | (n = 139) | (n = 334) | P value | |
| Male | 471 (99%) | |||
| Coronary artery disease | 273(58%) | 111(80%) | 162(49%) | <.001 |
| Myocardial infarction⁎ | 222(47%) | 93(67%) | 129(39%) | <.001 |
| Congestive heart failure† | 9(2%) | 4(3%) | 5(2%) | NS |
| Hypertension | 397(83%) | 136(98%) | 261(78%) | <.001 |
| Hyperlipidemia | 341(72%) | 98(71%) | 243(71%) | <.001 |
| Diabetes mellitus | 80(17%) | 30(22%) | 50(15%) | NS |
| Severe COPD | 33(7%) | 14(10%) | 19(6%) | NS |
| Active smoking | 170(36%) | 92(66%) | 78(23%) | <.001 |
| Any smoking history | 378(80%) | 138(99%) | 240(70%) | <.001 |
| Renal insufficiency‡ | 16 (3%) | 4(3%) | 12(4%) | NS |
⁎Remote history >6 months. |
†American Heart Association class III/IV. |
‡Creatinine clearance <50 mg dL. |
During biannual surveillance imaging, 225 (68%) patients experienced AAA growth to > 5.4 cm (n = 218) or more rapid expansion (>1 cm/yr) (n = 6) necessitating surgical intervention at a mean interval of 29 + 20 months after entry into surveillance. Two of the six patients experiencing expansion of intact AAAs presented with symptoms and required urgent repairs. Aneurysm repair was accomplished by open surgical graft replacement in 143 patients and endovascular techniques in 82 cases. Two AAA ruptures occurred in the surveillance cohort (0.6%). One high-risk patient with a 5.3 cm AAA presented with a contained rupture and survived emergent open repair. The other rupture occurred in a patient with a 5.1 cm AAA who died at home within 24 hours after evaluation at a civilian hospital for complaint of back pain that had improved before discharge. No late AAA ruptures occurred after endovascular or open repairs. As anticipated, AAA growth necessitating operative management for patients remaining in the AAA clinical pathway increased with time. Operative conversion rates at 1, 3, 5, and 7 years were 38%, 65%, 70%, and 80%, respectively. Over 7 years of AAA surveillance, the cumulative rate of operative conversion in the surveillance population was 68%. Fig 1 illustrates operative intervention for the surveillance group annually.
Aneurysm-related mortality was low for the surveillance cohort. Two patients undergoing elective open AAA repair died from perioperative cardiac events resulting in an overall operative (<60 days) mortality of 0.9% in patients requiring intervention (n = 225) during follow-up. Including the single fatal AAA rupture without intervention occurring during surveillance (0.3%), cumulative aneurysm-related mortality was 0.9% (3/329) for patients compliant with the AAA surveillance pathway.
By comparison, non-aneurysm related death occurred in 22 patients in the surveillance cohort (Table II). Cardiac (2.4%) and cancer-related (2.1%) causes were the most frequent contributors to late mortality. All cancer-related deaths occurred after at least 1 year of surveillance with three of the seven cancer deaths attributed to metastatic disease. Patient survival during surveillance was 92.4%/year. Fig 2 summarizes both mortality and compliance failure.
Table II. Aneurysm-related and non-aneurysm related mortality for patients not lost to follow-up
| Group | All cause mortality | |||||
|---|---|---|---|---|---|---|
| Aneurysm-related† | Cardiac | Pulmonary | Cancer | Other | Total | |
| (n = 329)⁎ | 3 (0.9%) | 8 (2.4%) | 3 (0.9%) | 7 (2.1%) | 4 (1%) | 25 (7.6%) |
| Surveillance w/repair (n = 223) | 2 | 4 | 3 | 5 | 4‡ | 18(8.1%) |
| Surveillance (n = 106) | 1 | 4 | 0 | 2 | 0 | 7(6.6%) |
⁎334 patients: five patients who left surveillance program. |
†Deaths include two postoperative cardiac deaths and one surveillance ruptured AAA. |
‡One death each: age >85, sepsis, bowel perforation, and renal failure. |
Fig 2.
Operative conversions of surveillance enrollees per annum with all-cause lost to follow-up.
Patient compliance with the AAA surveillance pathway was 98.5% during follow-up. Three patients (0.9%) were lost to follow-up because of patient relocation out of the region. Two (0.6%) chose elective repair for their asymptomatic <5.5 cm AAA at civilian hospitals. The remaining 79 patients with small AAAs have continued in surveillance. Surveillance enrollment with annual surveillance mortality and compliance is shown in Fig 3.
Fig 3.
Surveillance enrollment with surveillance mortality and compliance failure per annum for the study period.
Discussion
Evidence from prospective clinical trials indicate that AAA surveillance is safe and cost effective.1, 2, 3, 4, 5, 10, 11, 12 With the recent recommendations and Medicare approved reimbursement for AAA screening, vascular centers can expect an increase in practitioner and patient awareness of the condition.13 The threshold size associated with a safe recommendation for continued AAA observation is based on numerous prospective clinical and population-based natural history studies demonstrating a significant incremental increase in the risk of rupture for AAA between 5.0 and 6.0 cm.1, 2, 3, 4, 5, 10, 14, 15, 16, 17 We adopted surveillance guidelines validated in the ADAM trial with continued observation used for 4.0 to 5.4 cm AAA and repair recommended for AAA >5.4 cm.3, 4 Debate is now focused on demonstrating the optimal method for durable AAA repair (endovascular or open). Our clinical pathway does not stratify patients for a particular type of operative repair but relies strictly on operative planning based on AAA size or expansion criteria. In this series, virtually no patient was denied an operative repair based on risk profile alone. Whether or not patterns AAA expansion can be predicted by clinical pathways such as this remains unanswered but ongoing small aneurysm trials such as the Positive Impact of Endovascular Options for Treating Aneurysms Early (PIVOTAL) trial should provide important insight as to the benefit early endovascular repair.
In both the UK Small Aneurysm and ADAM trials,1, 2, 3, 4 at least two-thirds of patients entering surveillance eventually required open repair for AAA growth or development of aneurysm-related symptoms. Our experience was similar with 68% of patients receiving endovascular or open repair at a mean interval of 29 months that also mirrored the rate of disease progression seen in those landmark randomized trials. In addition, the AAA rupture rate (0.6%) seen during surveillance of small AAAs using our clinical pathway was even less than the reported <1% annual rupture risk found in the UK and ADAM trials. When combined with only two operative fatalities after interventions required during AAA follow-up, the overall aneurysm-related mortality in our series was extremely low (0.9%). During an equivalent length of study follow-up as in UK and ADAM trials, overall patient survival was higher in our experience with non-aneurysm related deaths far outnumbering those attributed to AAA.
The most important finding of our study is the high retention/compliance rate (98.5%) of patients achieved during surveillance of small AAA (4.0 to 5.4 cm). This was facilitated through an efficient clinical pathway incorporating continuity of AAA care assured from a single (vascular) provider, daily non-physician provider oversight, prospective database tracking of patients, “patient friendly” scheduling of surveillance imaging to reduce repetitive or unnecessary travel and clinic visitation, and telephone contact. Such high patient compliance rates have been achieved in prospective trials requiring patient consent (and obligation) to participate and study coordinators to initiate and arrange follow-up evaluations. Use of a non-physician “facilitator” (our ARNP) in our pathway was critical to creating and maintaining patient rapport and motivation during follow-up by playing a role similar to that of research study coordinators in formal prospective trials. A 30-day audit of our clinical pathway revealed that on average our ARNP staff dedicated 20% of a 40-hour work week towards the completion of aneurysm consultations and review of current AAA in surveillance including updating enrollment information for the surveillance database which currently holds 240 small AAA(s). Equally important is a centralized records system which allows for system-wide, shared information within VA health care facilities and reduces the chance of duplicated or unnecessary studies. By developing a familiarity with the patient and his medical history, we can promote a higher degree of customer satisfaction and motivation which we believe contributed to enhanced retention in our surveillance program.
We believe that these methods would be applicable in other high-volume hospital, clinic, or practice systems to optimize patient safety by “centralizing” small AAA surveillance. Although a potential limitation of the study lies in how to apply our AAA or other similarly-designed clinical pathways outside of the controlled veterans health care environment, several options for offset of added resource expenditure (eg, salary support for ARNP, database management) exist. Regional and possibly statewide registries could be developed in a similar manner to those established for disease (ie, cancer) or outcome conditions (ie, mortality). The cost of such programs could be underwritten by large third party healthcare providers or regional hospital systems with a focus on reduction of resources for those institutions. An overall reduction of reimbursements related to clinic visits would follow because fewer visits are necessary within the surveillance pathway. Larger hospital or medical systems could see more global cost advantages in reducing numbers of patients experiencing AAA rupture, emergent/urgent interventions, and extended intensive care and hospital stays.12 With increasing public/physician awareness and expansion of AAA screening with now available reimbursement for directed ultrasound studies, additional efforts aimed at implementing surveillance pathways would seem a logical extension to improve AAA care. Even if not directly performed by vascular surgeons, high-volume hospital systems, multi-specialty practice groups, or health maintenance organizations could “centralize” AAA surveillance for multiple physician providers, eliminate individual provider responsibility for arranging and assuring completeness of follow-up, and thereby allow the potential benefits of the pathway to extend to a larger “pool” of patients. Some revenue generation could also com from non-physician provider telephone contact with patients as these services remain billable (CPT codes 99371-3) and for patients requiring repair for AAA growth during surveillance, documentation of intervention planning from CT images by the surgeon/operator is also reimbursable (CPT G0288). Patients experiencing AAA growth during surveillance managed “centrally” could then be sent back to the referring/assigned providers and their preferred vascular surgeon/“interventionalist” for appropriate AAA treatment. Other common pathological entities that require serial imaging studies (eg, small pulmonary nodules or low probability mammographic breast lesions) could be “centrally” managed by similar surveillance methods with the resource “tools” described in this report. Reduction in the variability of care quality, efficiency of disease management, and improved patient compliance may all result in safer outcomes.
Conclusion
Implementation of a clinical pathway for nonoperative management of small AAA (4.0 to 5.4 cm) was associated with high patient compliance, low rupture risk, and extremely low cumulative aneurysm-related mortality. Aneurysm surveillance can be efficiently performed to achieve optimal patient safety and could be associated with long-term resource utilization and cost benefits.
Author contributions
References
- . Mortality results for randomized controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. Lancet. 1998;352:1649–1655
- . Long-term outcomes of immediate repair compared with surveillance of small abdominal aortic aneurysms. N Engl J Med. 2002;346:1437–1444
- Immediate repair compared with surveillance of small abdominal aortic aneurysm. N Engl J Med. 2002;346:1437–1444
- The aneurysm detection and management study screening program validation cohort and final results. Arch Intern Med. 2000;160:1425–1430
- . Guidelines for the treatment of abdominal aortic aneurysm: report of subcommittee of The Joint Council of the American Association for Vascular Surgery and Society for Vascular Surgery. J Vasc Surg. 2003;37:1106–1117
- . Watchful waiting in cases of small abdominal aortic aneurysms–appropriate for all patients?. J Vasc Surg. 2000;32:441–450
- Late results of small untreated abdominal aortic aneurysms. Ann Vasc Surg. 1991;5:359–362
- The limitations of ultrasound in surveillance of small abdominal aortic aneurysms. Br J Surg. 1991;78:614–616
- Abdominal aortic aneurysm evaluation: comparison of US, CT, MRI and angiography. Magn Reson Imaging. 1990;8:199–204
- . The Multicenter Aneurysm Screening Study (MASS) into the effect of abdominal aortic aneurysm screening on mortality in men: a randomized controlled trial. Lancet. 2002;360:1531–1539
- . Health service costs and quality of life for early elective surgery or ultrasonographic surveillance for abdominal aortic aneurysm. Lancet. 1998;352:1656–1660
- . Hospital costs and benefits of screening for abdominal aortic aneurysms (Results from a randomized population screening trial). Eur J Vasc Endovasc Surg. 2002;23:55–60
- . Screening for abdominal aortic aneurysm: a best-evidence systematic review for the US Preventive Services Task Force. Ann Intern Med. 2005;142:203–211
- . Risk factors for aneurysm rupture in patients kept under ultrasound surveillance. Ann Surg. 1999;230:289–297
- Growth rates and risk of rupture of abdominal aortic aneurysms. Br J Surg. 1998;85:1674–1680
- Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. JAMA. 2002;287:2968–2972
- . Learning from the last ultrasound: a population-based study of patients with abdominal aortic aneurysm. Arch Intern Med. 1997;157:2064–2068
Competition of interest: none.
PII: S0741-5214(07)00514-9
doi:10.1016/j.jvs.2007.03.038
© 2007 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
