University of Groningen Advances in complex endovascular aortic surgery Dijkstra, Martijn Leander

Advances in complex endovascular aortic surgery

Dijkstra, Martijn Leander

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from

it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date:

2018

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Dijkstra, M. L. (2018). Advances in complex endovascular aortic surgery. University of Groningen.

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

chapter 7

One-Year Outcomes of Endovascular Aneurysm Repair

in High-Risk Patients Using the Endurant Stent-Graft :

Comparison of the ASA Classifi cation and SVS/AAVS

Medical Comorbidity Grading System for the Prediction of

Mortality and Adverse Events

M.L. DIJKSTRA,

_{, S.M. VAN STERKENBURG,}

_{J.W. LARDENOYE,}

_C.J.

ZEEBREGTS,

_{AND M.M. REIJNEN,}

_{FOR THE ENGAGE INVESTIGATORS}

1 _{Department of Surgery, Rijnstate Hospital, Arnhem, the Netherlands} 2 _{Department of Surgery, Division of Vascular Surgery, University Medical Center Groningen, University of}

Groningen, the Netherlands

absTracT

Purpose: To evaluate the outcome and survival of patients with extensive

co-morbid conditions after endovascular aneurysm repair (EVAR) and objectify which of 2 medical co-morbidity classifications is more accurate in predicting adverse outcomes.

Methods: All 1263 patients (mean age 73.1 years; 1129 men) treated using

the Endurant Stent Graft system and entered in the prospective global post-marketing ENGAGE registry (ClinicalTrials.gov identifier NCT00870051) were grouped using the American Society of Anesthesiologists (ASA) classification and the Society for Vascular Surgery/ American Association for Vascular Surgery (SVS/ AAVS) medical co-morbidity grading system. Patients assigned to ASA III and IV and SVS/AAVS 2 and 3 categories were considered high risk. Primary outcome was 1-year all-cause mortality. Secondary outcomes included technical and clinical success, major adverse events (MAE), aneurysm rupture, endoleaks, and secondary endovascular procedures. One-year follow-up of the entire ENGAGE cohort was the endpoint of the study.

Results: A total of 1263 patients were included. The overall technical success rate

was high, the lowest being 97.4 % in the ASA I group. The overall 30-day and 1-year Kaplan-Meier survival estimates were 98.7 % and 92.5 %, respectively. All cause 1-year mortality was higher in the ASA III and IV groups, but this did not reach statistical significance (5.2 % and 5.7 % for ASA I and II vs. 9.0 % and 9.9 % for ASA III and IV, p = 0.12). In the SVS/AAVS groups, 1-year all-cause mortality significantly increased with the SVS/AAVS score to 11.3 % in the SVS/ AAVS 3 group (p = 0.002). There were significantly more MAEs in the SVS/AAVS 3 group at 1 year (p < 0.001 for group 1 vs 3 and group 2 vs 3).

Conclusion: Endovascular aneurysm repair has evolved, and high technical

suc-cess and low mortality and morbidity can be achieved in high-risk patients. When treating high-risk patients, the peri-operative risks should always be weighed against the expected gains. In contrast to the ASA classification, the SVS/AAVS medical co-morbidity grading system is a useful tool to predict occurrence of MAEs and 1-year survival in patients undergoing EVAR.

7

inTroducTion

Endovascular aneurysm repair (EVAR) has become the predominant treatment

for abdominal aortic aneurysms (AAA).1 _{Compared with open surgery, the}

technique is associated with lower 30-day mortality and morbidity.2–4

However, since patients presenting with an AAA frequently have extensive co-morbid conditions, long-term survival is largely dependent on non-aneurysm-related

events.5–7 _{This was also seen in the hallmark EVAR-2 trial,}8 _{which showed that}

aneurysm-related mortality was significantly lower after EVAR, though there was no benefit in overall survival. Treatment was, however, associated with a

consid-erable 30-day mortality of 9.0 %.8 _{This has led to the belief that a conservative}

strategy might be warranted in patients with extensive comorbid disease. Several grading systems are used to objectify patient co-morbidity and predict outcome. The American Society of Anesthesiologists (ASA) physical status clas-sification is most commonly employed due to its relative ease of use and broad

applicability.9

Another frequently applied classification system in vascular surgery is the Society for Vascular Surgery/American Association for Vascular Surgery

(SVS/AAVS) medical comorbidity grading system.10 _{Both have been evaluated}

after EVAR, but it is unclear which of the two is more accurate in predicting 1-year overall mortality. Other scoring systems, such as the Acute Physiology and Chronic Health Evaluation (APACHE) score and Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity (POSSUM) system, have either been proven to be less accurate (APACHE) or not very practical for

use in daily practice (POSSUM).11,12

The vascular POSSUM score (V-POSSUM) has been evaluated in both EVAR and open aneurysm repair but does not accurately predict mortality on an individual level. The V-POSSUM may, however, be of use

for interhospital comparisons and surgical audits.13

The high 30-day mortality observed in the EVAR-2 trial might not be in accor-dance with current practice. EVAR has evolved since its introduction in numerous ways, including patient selection, improved devices, increased operator experi-ence, improved perioperative cardiopulmonary care, and aggressive treatment of perioperative adverse events. Recent studies reporting the outcome of EVAR in a subgroup of high-risk patients show acceptable morbidity and mortality; it

should be noted, however, that these were not randomized controlled trials.14,15

Given this reduction in perioperative mortality, the aim of this study was to evaluate the outcome of EVAR in a large cohort of patients with extensive comorbid conditions (ASA ≥III and/or SVS/AAVS grade ≥ 2), specifically in terms of technical success, morbidity, and 1-year mortality. It was hypothesized these

patients can be safely treated using EVAR with an acceptable morbidity and 1-year survival. Second, an attempt was made to objectify which classification system is more accurate in predicting adverse outcomes.

MeThods

study design

A total of 1263 patients (mean age 73.1 years; 1129 men) treated with the Endurant stent-graft (Medtronic Vascular, Santa Rosa, CA, USA) were enrolled in the prospective, global, postmarketing ENGAGE registry between March 2009 and April 2011. The majority of the 79 participating centers (> 75 %) obtained ethics committee (EC) approval; for the remainder of the sites, the EC was either notified or EC approval was waived as the device was commercially available and standard of care treatment was followed for the patients. Full study design, selection criteria, data collection, and a detailed description of the Endurant

sys-tem have been previously published.16,17 _{The trial was registered on the National}

Institutes of Health website (ClinicalTrials.gov; identifier NCT00870051).

classification systems

The patients were categorized using both the ASA and SVS/ AAVS systems by the treating or implanting physician; the ENGAGE registry did not supply any guidance on how to determine either classification. The ASA physical status

clas-sification9 _{categorizes patients as healthy (level I) or having mild systemic disease}

(level II), severe systemic disease (level III), life-threatening severe systemic disease (level IV), or life expectancy <  24 hours without operation (level V). The SVS/

AAVS medical comorbidity grading system10

scores major (cardiac, pulmonary, and renal) and minor (age and hypertension) risk factors individually from 0 to 3 with specified weights applied to each factor’s score. The sum of the scores is divided by 10 to restore the 0 to 3 point scale. Patients assigned to ASA III and IV and SVS/AAVS 2 and 3 categories were considered high risk.

Baseline patient characteristics and aneurysm anatomy variables are depicted in Table 1 for both classifications. In 1 patient, the ASA classification was missing, while the SVS/AAVS classifications were missing in 46 cases. The SVS/AAVS 0 group consisted of only 1 patient and was therefore not included in the statisti-cal analysis.

7

Table 1.

Baseline Characteristics for the ASA and SVS/AA

VS Classifications. a V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) p Age, y 70.3 ± 7.4 72.1 ± 8.1 74.1 ± 8.1 74.5 ± 7.3 < 0.001: I vs III, I vs IV , II vs III, II vs IV 63.2 ± 4.8 72.1 ± 5.4 78.5 ± 8.0 < 0.001 all Men 94.8 (73/77) 89.0 (469/527) 90.6 (475/524) 83.6 (112/134) 0.004: I vs III, I vs IV , II vs III, II vs IV 93.4 (155/166) 90.2 (559/620) 86.3 (371/430) 0.02: 1 vs 3 Primary tr eatment indication Aneurysm diameter < 0.001: I vs III, I vs IV , II vs III, II vs IV 0.11 ≥ 1.5x normal infrar enal aorta 9.1 (7/77) 3.4 (18/527) 2.3 (12/524) 3.0 (4/134) 6.6 (11/166) 2.7 (17/620) 2.8 (12/430) 4–5 cm (≥ 5-mm ↑ in 6 mo) 9.1 (7/77) 8.2 (43/527) 4.6 (24/524) 3.7 (5/134) 7.8 (13/166) 6.9 (43/620) 4.7 (20/430) > 5 cm 79.2 (61/77) 84.6(446/527) 91.4 (479/524) 93.3 (125/134) 82.5 (137/166) 87.7 (544/640) 90.2 (388/430) Other 2.6 (2/77) 3.8 (20/527) 1.7 (9/524) 0.0 (0/134) 3.0 (5/166) 2.6 (16/620) 2.3 (10/430)

Baseline symptoms None

72.7 (56/77) 80.1 (422/527) 87.2 (457/524) 91.8 (123/134) < 0.001: I vs III, I vs IV , II vs III, II vs IV 80.7 (134/166) 84.5 (524/620) 84.4 (363/430) 0.47 Abdominal pain 24.7 (19/77) 14.0 (74/527) 6.5 (34/524) 6.0 (8/134) 14.5 (24/166) 10.3 (64/620) 9.8 (42/430) 0.23 Back pain 3.9 (3/77) 4.9 (26/527) 6.3 (33/524) 3.0 (4/134) 3.6 (6/166) 5.0 (31/620) 5.6 (24/430) 0.62

Table 1.

Baseline Characteristics for the ASA and SVS/AA

VS Classifications. a (continued) V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) p Other 1.3 (1/77) 3.4 (18/527) 2.1 (11/524) 0.7 (1/134) 3.0 (5/166) 1.9 (12/620) 3.0 (13/430) 0.48

Risk factors Tobacco use

36.5 (27/74) 52.6 (270/513) 47.2 (242/513) 51.1 (68/133) 0.004: I vs II, I vs IV 62.2 (102/164) 54.8 (332/606) 34.9 (147/421) < 0.001: 1 vs 3, 2 vs 3 Hypertension 47.4 (36/76) 71.8 (371/517) 82.2 (429/522) 79.4 (104/131) < 0.001: I vs II, I vs III, I vs IV , II vs III 60.8 (101/166) 75.5 (468/620) 80.0 (344/430) < 0.001: 1 vs 2, 1 vs 3 Hyperlipidemia 27.0 (20/74) 57.9 (285/492) 65.6 (322/491) 70.5 (93/132) < 0.001: I vs II, I vs III, I vs IV , II vs III, II vs IV 56.7 (89/157) 59.7 (355/595) 62.0 (251/405) 0.50 Diabetes mellitus 7.9 (6/76) 13.4 (70/521) 24.3 (126/518) 26.2 (34/130) < 0.001: I vs III, I vs IV , II vs III, II vs IV 13.4 (22/164) 19.6 (121/617) 20.0 (85/424) 0.15 Cancer 14.3 (11/77) 14.5 (76/523) 26.5 (135/510) 24.2 (32/132) < 0.001: I vs III, II vs III, II vs IV 18.2 (30/165) 21.3 (131/614) 20.2 (85/420) 0.66 Alcoholism 5.3 (4/76) 2.7 (14/514) 4.1 (21/511) 0.8 (1/128) 0.016 3.6 (6/166) 4.1 (25/605) 1.4 (6/417) 0.05: 2 vs 3 Car diac disease 20.8 (16/77) 44.6 (235/527) 63.0 (330/524) 70.9 (95/134) < 0.001: I vs II, I vs III, I vs IV , II vs III, II vs IV 34.3 (57/166) 53.9 (334/620) 60.9 (262/430) < 0.001 all Pulmonary disease 10.8 (8/74) 15.1 (79/522) 32.4 (168/518) 46.1 (59/128) < 0.001: I vs III, I vs IV , II vs III, II vs IV , III vs IV 12.7 (21/166) 26.0 (161/620) 28.8 (124/430) < 0.001: 1 vs 2, 1 vs 3

7

Table 1.

Baseline Characteristics for the ASA and SVS/AA

VS Classifications. a (continued) V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) p Renal insuf ficiency 3.9 (3/77) 8.8 (46/522) 21.0 (109/520) 27.1 (36/133) < 0.001: I vs III, I vs IV , II vs III, II vs IV 1.8 (3/166) 14.8 (92/620) 22.1 (95/430) < 0.001 all Car otid artery disease 6.0 (4/67) 8.2 (37/450) 13.9 (61/438) 13.0 (13/100) 0.002: II vs III 5.7 (8/141) 10.6 (55/518) 13.1 (48/366) 0.053 CV/neur ologic disease 2.6 (2/77) 9.7 (51/527) 14.9 (78/524) 21.6 (29/134) < 0.001: I vs II, I vs III, I vs IV , II vs III, II vs IV 9.6 (16/166) 11.5 (71/620) 15.6 (67/430) 0.06 Vascular disease 14.3 (11/77) 24.3 (128/527) 32.3 (169/524) 61.9 (83/134) < 0.001 all 22.3 (37/166) 32.3 (200/620) 31.4 (135/430) 0.04: 1 vs 2, 1 vs 3

Aneurysm morphology Maximum diameter

, mm 59.9 ± 15.2 59.4 ± 11.5 61.0 ± 11.5 61.3 ± 10.1 0.002: II vs III, II vs IV 58.5 ± 11.4 59.9 ± 11.4 61.7 ± 12.1 0.004: 1 vs 3, 2 vs 3 Neck diameter , mm 22.8 ± 3.4 23.7 ± 3.5 23.9 ± 3.5 23.6 ± 3.6 0.008 23.1 ± 3.2 23.7 ± 3.4 23.9 ± 12.2 0.052 Neck length, mm 28.1 ± 11.0 28.5 ± 13.1 25.4 ± 12.1 26.8 ± 10.8 0.002: I vs III, II vs III 27.6 ± 13.0 27.0 ± 12.5 26.9 ± 12.2 0.89

Neck angle, deg

30.1 ± 28.1 28.7 ± 23.8 32.9 ± 23.9 27.4 ± 19.3 0.001: II vs III, III vs IV 25.4 ± 21.1 30.1 ± 24.8 32.7 ± 23.3 0.002: 1 vs 3, 2 vs 3

Abbreviations: ASA, American Society of Anesthesiologists; CV

, cerebrovascular; ICU, intensive care unit; SVS/AA

VS, Society for V ascular Surger y/American Associa-tion for V ascular Surger y.

aContinuous data are presented as the means ± standard deviation; categorical

data are given as the percentage (count/sample). Classification

data were missing

in

1 patient for the ASA and in 46 patients for the SVS/AA

VS. SVS/

AA

VS group 0 had 1 patient and was not included in the analysis.

definitions and outcomes

Technical success was defined as successful delivery and deployment of the

Endurant stent-graft in the planned location, with no unintentional coverage of the internal iliac arteries or any visceral aortic branches and with removal of the delivery system.

Major adverse events were defined as death from any cause and/or the

occur-rence of bowel ischemia, myocardial infarction, paraplegia, procedural blood loss ≥ 1000 mL, renal failure, respiratory failure, or stroke.

Primary outcome of the present sub-analysis based on the ASA and SVS/AAVS classifications was all-cause mortality at 30 days and 1 year in the entire cohort. Secondary outcomes included technical and clinical success rates, major adverse events (MAEs), aneurysm rupture, endoleaks, and secondary endovascular pro-cedures related to the AAA or Endurant device.

statistical analysis

Nonparametric Kruskal-Wallis tests were used for continuous outcomes to compare all categories, while chi-square tests were used to compare categorical variables. The logrank test was used for time-to-event analyses. If a significant result was found, paired comparisons of 2 categories at a time were completed. Mortality was analyzed at the 1-year time point and as a time-to-event endpoint using Kaplan-Meier estimates and log-rank testing. A 2-sided p  <  0.05 was considered statistically significant. No adjustments for multiple comparisons were made due to the exploratory nature of the analyses. If the categories were mutually exclusive, only 1 p-value was given; if the categories could overlap, a separate p-value for each category was given. All statistical analyses were performed with SAS software (version 9.13; SAS Institute, Cary, NC, USA).

resulTs

The mean age was significantly higher in the high-risk groups for both clas-sifications (ASA I/II vs III/IV, p < 0.001; SVS/AAVS 1 vs 2/3, p < 0.001). Patients were predominantly male in all groups. Indications for implant were similar in the SVS groups. In ASA I and II groups, there were significantly more implants due to rapid growth or an aortic diameter 1.5 times the normal infrarenal aorta (p < 0.001 for I vs III, I vs IV, II vs III, and II vs IV; Table 1).

7

greater frequency in the ASA IV (78.9 %) group compared to ASA I (59.7 %). Consequently, spinal anesthesia was used significantly less in the ASA IV group (5.3 %) compared with the other groups. In 13.2 % of the cases, local anesthesia was used. In the SVS/AAVS groups, there was no significant difference between groups in terms of anesthesia type. In ASA III and IV groups, the procedure time was significantly longer compared to ASA I and II (p < 0.001). Procedure details are shown in Table 2. Procedure time was also significantly longer in the high risk SVS/AAVS groups.

The overall technical success rate was high, the lowest being 97.4 % in the ASA I group. There was no difference in technical success among the groups. In the ASA IV group, there were significantly more type II endoleaks at completion angiography (25.6 %, p < 0.001; Table 2), but this difference disappeared at 1 year (p = 0.46). In the SVS/AAVS group, there were more endoleaks (p > 0.05), especially type II, in the high-risk groups, but this did not result in an increase in secondary endovascular procedures or late aneurysm rupture. The duration of hospital stay significantly increased with the SVS/AAVS score, from 5.3 ± 4.0 days in the lowest to 7.6 ± 6.9 in the highest group (p < 0.001). Surprisingly, the hospital stay was significantly shorter in the ASA IV group.

All-cause 30-day mortality was very low in all groups, with the ASA III group having the highest percentage (1.7 %). In the SVS/AAVS groups, SVS/AAVS 3 had the highest 30-day mortality (1.9 %). The 30-day and 1-year overall Kaplan-Meier survival estimates were 98.7 % and 92.5 %, respectively (Figure  1A). Differences in 30-day MAEs were significantly more conversions to open surgery in the ASA I group (2.6 %, p < 0.001 for groups I vs II and I vs III; Table 3) and more periprocedural blood loss (> 1000 mL) in the SVS/AAVS 3 group (2.6 %;

p < 0.001 for groups 2 vs 3; Table 3). There was no difference in the incidence

of secondary endovascular procedures (Table 3).

All-cause 1-year mortality (Figure 1B) was higher in ASA III and IV groups, but this did not reach statistical significance (5.2 % and 5.7 % for ASA I and II vs 9.0 % and 9.9 % for ASA III and IV, respectively; p = 0.12). In the SVS/AAVS 3 group, there was a significant increase in the occurrence of ≥ 1 MAEs (p < 0.001 for groups 1 vs 3 and 2 vs 3). In the SVS/AAVS groups (Figure 1C), the 1-year all-cause mortality significantly increased with the SVS/AAVS score, to 11.3 % in the SVS/AAVS 3 group (p = 0.002).

Analysis of all-cause mortality after 1 year (n = 94) showed various causes of death, mostly not aneurysm related. Of the 94 deaths after 1 year, 25 (26.5 %)

Table 2.

Pr

ocedur

e Details for the ASA and the SVS/AA

VS Classifications. a V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) P

Type of anesthesia General

59.7 (46/77) 58.8 (310/527) 61.6 (323/524) 78.9 (105/133) < 0.001: I vs IV , III vs IV 56.6 (94/166) 60.6 (376/620) 65.5 (281/429) 0.10 Spinal 18.2 (14/77) 26.0 (137/527) 20.8 (109/524) 5.3 (7/133) < 0.001: I vs IV ,

II vs III, III, II vs IV, III vs IV

23.5 (39/166) 22.1 (137/620) 19.1 (82/429) 0.37 Epidural 3.9 (3/77) 6.1 (32/527) 11.1 (58/524) 6.8 (9/133) 0.01: II vs III 7.8 (13/166) 8.2 (51/620) 8.6 (37/429) 0.95 Local 19.5 (15/77) 13.1 (69/527) 12.6 (66/524) 12.8 (17/133) 0.042 15.7 (26/166) 13.4 (83/620) 12.6 (54/429) 0.61 Pr ocedur e time, min 81.9 ± 48.5 91.3 ± 39.8 107.8 ± 47.8 108.9 ± 42.3 < 0.001: I vs II, I vs III, I vs IV , II vs III, II vs IV 86.7 ± 42.7 99.1 ± 46.4 105.8 ± 43.7 < 0.001 all

Estimated blood loss, mL

215.6 ± 352.4 192.1 ± 207.2 220.8 ± 212.1 219 ± 197.8 < 0.01: I vs III, I vs IV , II vs III, II vs IV 210.6 ± 251.2 198.7 ± 198.3 227.0 ± 244.9 0.55 Volume of contrast, mL 113.4 ± 62.1 133.4 ± 70.2 137.9 ± 70.7 111.7 ± 64.7 < 0.001: I vs II, I vs III, II vs IV , III vs IV 121.3 ± 69.0 134.2 ± 69.0 132.1 ± 70.5 0.06 Hospital stay , d 7.7 ± 7.1 6.2 ± 6.9 7.0 ± 6.4 5.4 ± 5.3 < 0.001: I vs II, I vs IV , II vs III, II vs IV , III vs IV 5.3 ± 4.0 6.2 ± 6.9 7.6 ± 6.9 < 0.001: 1 vs 3 2 vs 3 ICU stay , h 6.8 ± 12.6 10.5 ± 59.3 10.6 ± 25.6 8.7 ± 24.4 0.003: II vs III 5.6 ± 11.6 10.3 ± 45.3 12.2 ± 48.1 0.24 Technical success 97.4 (75/77) 99.2 (523/527) 99.4 (512/524) 97.8 (131/134) 0.013 99.4 (165/166) 98.7 (612/620) 99.3 (427/430) 0.55

7

Table 2.

Pr

ocedur

e Details for the ASA and the SVS/AA

VS Classifications. a (continued) V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) P Clinical success 94.8 (73/77) 98.1 (517/527) 97.9 (513/524) 96.3 (129/134) 0.022 98.2 (163/166) 99.8 (603/620) 97.7 (420/430) 0.77

Endoleak, all treated

c 13.3 (10/75) 8.2 (43/524) 10.3 (54/524) 6.0 (8/133) 0.021 6.7 (11/165) 8.4 (52/616) 11.0 (47/429) 0.19 I 6.7 (5/75) 4.4 (23/524) 7.8 (41/524) 6.0 (8/133) 3.6 (6/165) 4.9 (30/616) 8.4 (36/429) II 1.3 (1/75) 2.3 (12/524) 2.1 (11/524) 0.0 (0/133) 2.4 (4/165) 1.9 (12/616) 1.9 (8/429) III 5.3 (4/75) 1.1 (6/524) 0.6 (3/524) 0.0 (0/133) 0.0 (0/165) 1.6 (10/616) 0.7 (3/429) IV 0.0 (0/75) 0.2 (1/524) 0.0 (0/524) 0.0 (0/133) 0.0 (0/165) 0.2 (1/616) 0.0 (0/429)

Endoleak, all untr

eated 9.3 (7/75) 12.2 (64/524) 17.2 (90/524) 30.8 (41/133) < 0.001: I vs IV , II vs III, II vs IV , III vs IV 12.1 (20/165) 15.6 (96/616) 17.5 (75/429) 0.27 I 2.7 (2/75) 1.0 (5/524) 1.1 (6/524) 0.8 (1/133) 0.6 (1/165) 1.1 (7/616) 1.4 (6/429) II 6.7 (5/75) 9.0 (47/524) 13.5 (71/524) 25.6 (34/133) 10.3 (17/165) 11.7 (72/616) 13.3 (57/429) III 0 (0/75) 0.2 (1/524) 0.4 (2/524) 0.8 (1/133) 0.6 (1/165) 0.3 (2/616) 0.2 (1/429) IV 0 (0/75) 1.5 (8/524) 2.1 (11/524) 2.3 (3/133) 0.6 (1/165) 1.9 (12/616) 2.1 (9/429)

Abbreviations: ASA, American Society of Anesthesiologists; ICU, intensive care unit; SVS/AA

VS, Society for V

ascular Surger

y/American Association for V

ascular Sur

-ger

y.

aContinuous data are presented as the means ± standard deviation; categorical

data are given as the percentage (count/sample). Classification

data were missing

in

1 patient for the ASA and in 46 patients for the SVS/AA

VS. SVS/AA

VS group 0 had 1 patient and was not included in the analysis.

Table 3.

Adverse Events and Survival for the ASA and the SVS/AA

VS Classifications. a V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) p

30-Day MAE All cause mortality

1.3 (1/77) 0.9 (5/527) 1.7 (9/524) 0.7 (1/134) 0.67 0.6 (1/166) 1.1 (7/620) 1.9 (8/430) 0.41 Multiple events 5.2 (4/77) 3.0 (16/527) 4.4 (23/524) 4.5 (6/134) 0.60 4.2 (7/166) 2.4 (15/620) 6.0 (26/430) 0.01: 2 vs 3 Bowel ischemia 0 (0/77) 0.2 (1/527) 0.4 (2/524) 0 (0/134) 0.80 0 (0/166) 0.2 (1/620) 0.5 (2/430) 0.49 Myocar dial infar ction 0 (0/77) 1.1 (6/527) 1.3 (7/524) 1.5 (2/134) 0.77 0.6 (1/166) 0.8 (5/620) 1.9 (8/430) 0.22 Paraplegia 0 (0/77) 0 (0/527) 0 (0/524) 0 (0/134) — 0 (0/166) 0 (0/620) 0 (0/430) — Pr

ocedural blood loss

≥ 1000 mL 3.9 (3/77) 1.1 (6/527) 1.1 (6/524) 2.2 (3/134) 0.21 2.4 (4/166) 0.5 (3/620) 2.6 (11/430) 0.01: 1 vs 2, 2 vs 3 Renal failur e 0 (0/77) 0 (0/527) 0.8 (4/524) 0 (0/134) 0.13 0 (0/166) 0.5 (3/620) 0.2 (1/430) 0.57 Respiratory failur e 0 (0/77) 0 (0/527) 0 (0/524) 0 (0/134) — 0 (0/166) 0 (0/620) 0 (0/430) — Str oke 0 (0/77) 0.2 (1/527) 0.2 (1/524) 0 (0/134) 0.94 0.6 (1/166) 0 (0/620) 0.2 (1/430) 0.21

Conversion to open surgery

2.6 (2/77) 0 (0/527) 0 (0/524) 0 (0/134) < 0.001, I vs II, I vs III 0.6 (1/166) 0.0 (0/620) 0.2 (1/430) 0.27 Aneurysm ruptur e 0 (0/77) 0 (0/527) 0 (0/524) 0 (0/134) — 0 (0/166) 0 (0/620) 0 (0/430) —

Secondary endovascular procedur

e 2.6 (2/77) 2.1 (11/527) 2.1 (11/524) 0 (0/134) 0.39 1.8 (3/166) 2.3 (14/620) 1.6 (7/430) 0.76 to corr

ect type I/III

endoleak 0 (0/77) 0.2 (1/527) 0.8 (4/524) 0 (0/134) 0.36 0 (0/166) 0.5 (3/620) 0.5 (2/430) 0.67 365-Day MAE

7

Table 3.

Adverse Events and Survival for the ASA and the SVS/AA

VS Classifications. a (continued) V ariable ASA I (n = 77) ASA II (n = 527) ASA III (n = 524) ASA IV (n = 134) p b SVS/AA VS 1 (n = 166) SVS/AA VS 2 (n = 620) SVS/AA VS 3 (n = 430) p All-cause mortality 5.2 (4/77) 5.7 (30/523) 9.0 (47/520) 9.9 (13/131) 0.12 4.8 (8/166) 5.9 (36/615) 11.3 (48/426) 0.002: 1 vs 3, 2 vs 3 Multiple events 9.1 (7/77) 9.2 (48/523) 13.1 (68/520) 15.3 (20/131) 0.10 7.8 (13/166) 8.6 (53/615) 17.1 (73/426) < 0.001: 1 vs 3, 2 vs 3 Bowel ischemia 0 (0/77) 0.2 (1/523) 0.4 (2/520) 0 (0/131) 0.80 0 (0/166) 0.2 (1/615) 0.5 (2/426) 0.49 Myocar dial infar ction 1.3 (1/77) 1.9 (10/523) 2.1 (11/520) 3.1 (4/131) 0.82 1.2 (2/166) 1.6 (10/615) 3.1 (13/426) 0.20 Paraplegia 0 (0/77) 0 (0/523) 0.0 (0/520) 0 (0/131) — 0 (0/166) 0 (0/615) 0 (0/426) — Renal failur e 0 (0/77) 0.8 (4/523) 1.7 (9/520) 0.8 (1/131) 0.34 0 (0/166) 1.0 (6/615) 1.9 (8/426) 0.13 Respiratory failur e 0 (0/77) 0 (0/523) 0 (0/520) 0.8 (1/131) 0.04: II vs IV , III vs IV 0 (0/166) 0.2 (1/615) 0 (0/426) 0.62 Str oke 0 (0/77) 0.4 (2/523) 0.6 (3/520) 1.5 (2/131) 0.40 0.6 (1/166) 0.5 (3/615) 0.5 (2/426) 0.98

Conversion to open surgery

2.6 (2/77) 0.4 (2/527) 0.4 (2/524) 0.7 (1/134) 0.09 0.6 (1/166 0.5 (2/430) 0.5 (3/620) 0.98 Aneurysm ruptur e 0 (0/77) 0 (0/523) 0.4 (2/524) 0 (0/134) 0.28 0 (0/166) 0 (0/620) 0.5 (2/430) 0.16

Secondary endovascular procedur

e 3.9 (3/77) 7.0 (37/527) 6.1 (32/524) 3.0 (4/134) 0.29 5.4 (9/166) 6.5 (40/620) 5.3 (23/430) 0.73 to corr

ect type I/III

endoleak 0 (0/77) 1.5 (8/527) 1.9 (10/524) 0 (0/134) 0.27 1.2 (2/166) 1.3 (8/620) 1.9 (8/430) 0.72

Abbreviations: ASA, American Society of Anesthesiologists; MAE, major adverse event; SVS/AA

VS, Society for V

ascular Surger

y/American Association for V

ascular

Surger

y.

aContinuous data are presented as the means ± standard deviation; categorical

data are given as the percentage (count/sample). Classification

data were missing

in

1 patient for the ASA and in 46 patients for the SVS/AA

VS. SVS/AA

VS group 0 had 1 patient and was not included in the analysis.

were caused by cardiac pathology, 23 (24.5 %) by cancer, and 15 (16.0 %) by pulmonary pathology. The remaining patients died from various causes, includ-ing sepsis, gastrointestinal bleedinclud-ing (not aneurysm-related), neurological events, and renal failure.

discussion

This cohort study shows that high-risk patients can be treated by EVAR using the Endurant stent-graft system with a high technical success and low 30-day and 1-year morbidity and mortality. The ASA score was not useful in predicting

Figure 1. Kaplan-Meier analyses for all-cause mortality to 1 year in (A) the entire cohort

(n = 1263), (B) the ASA (American Society of Anesthesiologists Society) groups, and (C) the SVS/ AAVS (Society for Vascular Surgery/American Association for Vascular Surgery) groups.

7

correlation with not only overall survival but also the overall incidence of MAEs. Advances in patient selection, operative technique, and perioperative care may all have contributed to significant improvements in patients with extensive comorbid conditions. The overall <  2 % perioperative mortality in the present study was considerably lower compared to the reported 9 % 30-day mortality

in the EVAR-2 trial.8 _{At 1 year, the all-cause mortality was 9.9 % in the ASA IV}

group and 11.3 % in the SVS/AAVS 3 group, which are in contrast to the EVAR-2

trial,8 _{where the 1-year all-cause mortality in the EVAR group was as high as}

22.3 % (19.2 % for patients who did not undergo intervention). Our findings are

in accord with a recent study by Tadros et al,18 _{which showed improved patient}

outcomes due to advances in EVAR, such as improved patient selection and preoperative planning, better devices, growing operator experience and familiar-ity with EVAR, and advances in perioperative cardiopulmonary care.

A high technical success rate was achieved in all groups without significant differences between them. In the ASA groups, there was a difference in the occurrence of (uncorrected) endoleaks (Table  2), but there was no difference in the need for secondary endovascular procedures or aneurysm rupture. There were more MAEs in the SVS/AAVS 3 group compared with SVS/AAVS 1 or 2 (p < 0.001) and a greater incidence of endoleaks, but the latter was not statisti-cally significant. Importantly, this trend did not result in a corresponding differ-ence in secondary endovascular procedures or ruptures.

The overall hospital stay was higher than expected, with the lowest (5.3 days) in the SVS/AAVS 1 group. As anticipated the SVS/AAVS score showed a significant correlation with the length of hospital stay due to a slower recovery in this vulnerable patient group. To our surprise, the opposite was true for the ASA score, where patients in the highest category had the shortest hospital stay. This indicates that the ASA score does not seem to be a valid tool for the prediction of outcome after EVAR.

The ASA classification is the most widely used medical comorbidity scoring system in surgical patients. However, in this study, there was no statistical difference in ei-ther 30-day (p = 0.67) or 1-year mortality (p = 0.12) between the different groups. In general, it seems AAA patients who meet the criteria for surgical repair are by definition not ASA I (“a normal healthy patient”). Yet, in this series, 77 patients were classified as being ASA I. This can be partly explained by the known large

interobserver variability of the ASA classification.9 _{Also, the ENGAGE registry does}

were assigned by the treating physicians, and these were the data analyzed. An analysis excluding ASA I patients was conducted, but this did not alter the results (data not shown). Therefore, for the purpose of predicting outcome in patients

undergoing EVAR, the ASA classification seems to fall short.19 _{The SVS/AAVS}

clas-sification, on the other hand, although less frequently used and reported, seems to predict 1-year survival quite accurately, with 1-year mortality rising from 4.8 % in the SVS/AAVS 1 group to 11.3 % in the SVS/AAVS 3 group (p = 0.002).

Perioperative morbidity seems hard to predict based on our data. There were, however, significantly more MAEs in the SVS/AAVS 3 group at 1 year, and there seems to be a nonsignificantly higher incidence of myocardial infarction in the high SVS/AAVS groups. A significant difference was found in the 1-year respiratory failure rate, which was significantly higher in the ASA IV group. The inability of either classification to predict morbidity is in part due to the very low incidence of all MAEs.

General anesthesia was the preferred modality; only a small proportion (13 %) of the cohort had local anesthesia. Earlier studies have shown a benefit in periop-erative hemodynamics when using local anesthesia, which may be the preferred

option in high-risk patients.20 _{In our study, the highest incidence of local}

anes-thesia was in the ASA I group (19.5 %). The choice of anesanes-thesia was based on the treating physician and anesthesiologist and probably also on local experience and preference. More liberal use of local anesthesia may further improve patient outcomes after EVAR, especially when treating patients with extensive comorbid conditions. Interestingly, ASA III and IV patients also tended to have longer operative times. One could hypothesize, for instance, in high-risk patients that a surgeon’s preference may be to perform femoral cutdown for more controlled access, leading to slightly longer procedures. However, the ENGAGE registry does not include specific data besides the duration of the procedure. This again is a limitation of the registry design.

The ENGAGE registry represents the “real-world” outcome after EVAR with the Endurant device and could serve as a benchmark for future studies. The large sample size and the cutoff point at 1-year follow-up as the study endpoint strengthens the robustness of the estimates for this study. Patient selection, along with many of the earlier mentioned factors, may have improved the outcome of EVAR in any patient category, as well as the high-risk patients. The ENGAGE registry does employ a high degree of clinical rigor for a postmarket study. For example, all data undergo independent review and verification by investigators, site monitoring is routine, protocol endpoints are 100 % monitored, and 30-day MAEs and all deaths are adjudicated by an independent clinical events committee.

7

limitations

First, the ENGAGE registry failed to provide guidance on the classification systems. These systems are well known, but nevertheless, a significant portion of patients was classified as ASA 1, which cannot be true in case of an AAA. This, however, clearly reflects the current clinical use of the system by treating physicians. This misjudgment could have affected the results, but an analysis excluding these patients did not change the outcome.

Second, the registry focuses on 1 endograft only. It seems reasonable to as-sume, however, the choice of device does not affect the relation between risk assessment and outcome. Also, the present study was a subgroup analysis and should therefore be interpreted with caution. Furthermore, as it is not a randomized controlled trial, bias may have occurred. Data collection was done prospectively, but retrospectively analyzed. Duration of follow-up (limited to 1 year) and patients lost to follow-up may have influenced the results, although the Kaplan-Meier method was used for evaluation of survival to account for this. With regard to collection of mortality data, there is indeed a possible bias. Deaths, as all other data, were reported by the treating physicians. An overview of the deaths shows a large number of causes; as reported, most were unrelated to the aneurysm (eg, neurologic, cardiac, and pulmonary). A number of patients were reported to be dead when contacted for routine follow-up visits, or hos-pital records showed the patient had died. In these cases, either the general physician or the nursing home was contacted to supply additional information on the cause of death.

Ultimately, the treating physician determines who to treat, when to treat, and whether a patient is “high risk,” and interobserver variability may be high. The exploratory nature of this study lends itself to facilitate and guide the thought processes for more specific and focused future studies. With that in mind, it should be noted that statistically significant results from an observational cohort are less robust evidence compared to results from randomized controlled trials.

conclusion

The presented results in patients with extensive medical comorbidities show EVAR using the Endurant stent-graft can be performed with high technical suc-cess rates and low perioperative morbidity and mortality. This may have

Of course, when treating high-risk patients, the perioperative risks should always be weighed against the expected gains. In contrast to the ASA classification, the SVS/AAVS medical comorbidity grading system is a useful tool to predict occurrence of MAEs and 1-year survival in patients undergoing EVAR. Further prospective randomized controlled trials are needed to confirm these findings.

7

references

1. Dua A, Kuy S, Lee CJ, et al. Epidemiology of aortic aneurysm repair in the United States from 2000 to 2010. J Vasc Surg. 2014; 59: 1512–1517.

2. Zeebregts CJ, Geelkerken RH, van der Palen J, et al. Outcome of abdominal aortic aneurysm repair in the era of endovascular treatment. Br J Surg. 2004; 91: 563–568. 3. Prinssen M, Verhoeven EL, Buth J, et al. A randomized trial comparing conventional

and endovascular repair of abdominal aortic aneurysms. N Engl J Med. 2004; 351: 1607–1618.

4. EVAR trial participants. Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet. 2005; 365: 2179–2186.

5. United Kingdom EVAR Trial Investigators, Greenhalgh RM, Brown LC, et al. Endo-vascular versus open repair of abdominal aortic aneurysm. N Engl J Med. 2010; 362: 1863–1871.

6. De Bruin JL, Baas AF, Buth J, et al.; DREAM Study Group. Long-term outcome of open or endovascular repair of abdominal aortic aneurysm. N Engl J Med. 2010; 362: 1881–1889.

7. Brown LC, Powell JT, Thompson SG, et al. The UK EndoVascular Aneurysm Repair (EVAR) trials: randomized trials of EVAR versus standard therapy. Health Technol Assess. 2012; 16: 1–218.

8. EVAR trial participants. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aorticaneurysm (EVAR trial 2): randomised controlled trial. Lancet. 2005; 365: 2187–2192.

9. Mak PH, Campbell RC, Irwin MG. The ASA Physical Status Classification: interob-server consistency. American Society of Anesthesiologists. Anaesth Intensive Care. 2002; 30: 633–640.

10. Chaikof EL, Fillinger MF, Matsumura JS, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg. 2002; 35: 1061–1066.

11. Lazarides MK, Arvanitis DP, Drista H, et al. POSSUM and APACHE II scores do not predict the outcome of ruptured infrarenal aortic aneurysms. Ann Vasc Surg. 1997; 11: 155–158.

12. Tambyraja AL, Murie JA, Chalmers RT. Prediction of outcome after abdominal aortic aneurysm rupture. J Vasc Surg. 2008; 47: 222–230.

13. Kapma M, Kahmann O, Van Stijn I, et al. Evaluation of risk prediction models, V-POSSUM and GAS, in patients with acute abdominal aortic rupture treated with EVAR or an open procedure [published online January 16, 2014]. J Cardiovasc Surg (Torino).

14. Sobocinski J, Maurel B, Delsart P, et al. Should we modify our indications after the EVAR-2 trial conclusions? Ann Vasc Surg. 2011; 25: 590–597.

15. Pol RA, Zeebregts CJ, van Sterkenburg SM, et al. Outcome and quality of life after endovascular abdominal aortic aneurysm repair in octogenarians. J Vasc Surg. 2014; 60: 308–317.

16. Böckler D, Fitridge R, Wolf Y, et al. Rationale and design of the Endurant Stent Graft Natural Selection Global Postmarket Registry (ENGAGE): interim analysis at 30 days of the fi rst 180 patients enrolled. J Cardiovasc Surg (Torino). 2010; 51: 481–491.

17. Stokmans RA, Teijink JA, Forbes TL, et al. Early results from the ENGAGE registry: real-world performance of the Endurant Stent Graft for endovascular AAA repair in 1262 patients. Eur J Vasc Endovasc Surg. 2012; 44: 369–375.

18. Tadros RO, Faries PL, Ellozy SH, et al. The impact of stent graft evolution on the results of endovascular abdominal aortic aneurysm repair. J Vasc Surg. 2014; 59: 1518–1527.

19. Conners MS, Tonnessen BH, Sternbergh WC, et al. Does ASA classifi cation impact success rates of endovascular aneurysm repairs? Ann Vasc Surg. 2002; 16: 550–555. 20. Bettex DA, Lachat M, Pfammatter T, et al. To compare general, epidural and local

anesthesia for endovascular aneurysm repair (EVAR). Eur J Vasc Endovasc Surg. 2001; 21: 179–184.

21. Brown PM, Zelt DT, Sobolev B. The risk of rupture in untreated aneurysms: the impact of size, gender, and expansion rate. J Vasc Surg. 2003; 37: 280–284.