Advertisement

Editor's Choice – Variation in Intact Abdominal Aortic Aneurysm Repair Outcomes by Country: Analysis of International Consortium of Vascular Registries 2010 – 2016

Open ArchivePublished:June 15, 2021DOI:https://doi.org/10.1016/j.ejvs.2021.03.034

      Objective

      Outcomes for intact abdominal aortic aneurysm (AAA) repair vary over time and by healthcare system, country, and surgeon. The aim of this study was to analyse peri-operative mortality for intact AAA repair in 11 countries over time and compare outcomes by gender, age, and geographical location.

      Methods

      Prospective data on primary repair of intact AAA were collected from 11 countries through the International Consortium of Vascular Registries (ICVR) and analysed for two time periods, 2010 – 2013 and 2014 – 2016. The primary outcome was peri-operative mortality after endovascular aneurysm repair (EVAR) and open surgical repair (OSR). Multivariable logistic regression models were used to adjust for differences in patient characteristics.

      Results

      A total of 103 715 patients were included. The percentage of patients undergoing EVAR increased from 63.6% to 71.2% (p < .001) over the study period. This proportion varied by country from 35% in Hungary to 81% in the United States. Overall peri-operative mortality decreased from 2.1% to 1.6 % (p < .001). Mortality also declined significantly over time for both OSR 4.2% to 3.6 % (p = .002) and EVAR 1.0% to 0.7% (p = .002). Mortality was significantly higher for female than male patients (3.0% vs. 1.6% p < .001). The percentage of patients > 80 years old undergoing AAA repair remained constant at 23.6% (p = .91). Peri-operative mortality was higher for patients > 80 years than for those < 80 years old (2.7% vs. 1.6% p < .001). Forty-six per cent (n = 275) of all EVAR deaths occurred in the over 80s.

      Conclusion

      The proportion of AAA repairs performed using EVAR has increased over time. Peri-operative mortality continues to decline for both OSR and EVAR. Outcomes however were significantly worse for both women and those aged over 80, so efforts should be focused on these patient groups to further reduce elective AAA mortality rates.

      Keywords

      In this study of over 100 000 intact abdominal aortic aneurysm repairs, an increase in the proportion of patients undergoing endovascular aneurysm repair (EVAR) over time has been demonstrated. Peri-operative mortality rates are falling for both EVAR and open surgical repair. Outcomes however are poorer in women and octogenarians, and future efforts should be focused on improving outcomes for these patients.

      Introduction

      Outcomes after surgery vary between different healthcare systems and by country. These variations are well established in vascular surgery. Outcomes after elective abdominal aortic aneurysm (AAA) repair have been demonstrated previously to differ significantly in Europe, North America, and Australasia.
      • Beck A.W.
      • Sedrakyan A.
      • Mao J.
      • Venermo M.
      • Faizer R.
      • Debus S.
      • et al.
      Variations in abdominal aortic aneurysm care: a report from the international consortium of vascular registries.
      ,
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice – Assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      The importance of monitoring outcomes following AAA repair has been underlined by the recommendation in the European Society for Vascular Surgery (ESVS) AAA guidelines that vascular units performing aortic surgery should enter cases into a validated prospective registry to allow for monitoring of changes in practice and outcomes.
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor's choice – European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      The reporting of variation in outcomes after AAA repair is important for both patients and surgeons and is vital to drive quality improvement.
      • Behrendt C.A.
      • Venermo M.
      • Cronenwett J.L.
      • Sedrakyan A.
      • Beck A.W.
      • Eldrup-Jorgensen J.
      • Mani K.
      VASCUNET, the Vascular Quality Initiative and the International Consortium of Vascular Registries. VASCUNET, VQI and the International Consortium of Vascular Registries – Unique Collaborations for Quality Improvement in Vascular Surgery.
      The identification of high performing of units, regions, or countries enables them to share best practice principles and provides benchmark data for lower performing institutions to strive towards. This was clearly demonstrated after the publication of the 2008 VASCUNET report
      • Gibbons C.
      • Bjorck M.
      • Jensen L.P.
      • Laustsen J.
      • Lees T.
      • Moreno-Carriles R.
      • et al.
      The Second Vascular Surgery Database Report.
      that demonstrated the UK had the highest elective AAA mortality rate (7.5%) compared with countries in Europe and Australasia. This publication was the catalyst for the Vascular Society of Great Britain and Ireland to implement a Quality Improvement Programme in 2009, with the aim of halving the elective AAA mortality rate to 3.5% by 2014. A subsequent report in 2012 of over 8000 elective AAA repairs collected in the National Vascular Database between September 2008 and October 2010 demonstrated a mortality of 2.4%.
      • Mitchell D.
      • Hindley H.
      • Naylor A.R.
      • Wyatt M.
      • Loftus I.M.
      Outcomes after elective repair of infra-renal abdominal aortic aneurysm. A report from the Vascular Society.
      Data on the outcomes of intact AAA repairs from 11 counties collected between 2005 and 2013 also demonstrated improved outcomes and an overall falling peri-operative mortality. However, this also showed an increase in the mortality after open surgical repair (OSR) to 4.4%.
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice – Assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      This may in part be explained by the increased complexity of open AAA surgery, with anatomically less complex cases undergoing EVAR.
      The aim of this study was to analyse peri-operative mortality for intact AAA repair in contemporary data collected by the International Consortium of Vascular Registries
      • Scali S.T.
      • Beck A.W.
      • Sedrakyan A.
      • Mao J.
      • Venermo M.
      • Faizer R.
      • et al.
      Hospital volume association with abdominal aortic aneurysm repair mortality: Analysis of the international consortium of vascular registries.
      in 11 countries and compare outcomes by gender, age, and geographical location during two different time periods.

      Material and Methods

      Registry information

      Data from prospectively maintained vascular surgery registries on primary elective intact AAA repair by either OSR or endovascular aneurysm repair (EVAR) from 11 countries performed between 2010 and 2016 were combined and submitted to the Medical Device Epidemiology Network (MDEpiNet; www.mdepinet.org) Analytic Centre at Cornell University for analysis. De-identified patient level data were provided from eight registries (Australia, Denmark, Hungary, Finland, Malta, New Zealand, Sweden, and United States). Aggregate level data were submitted by three additional countries (Germany, Norway, UK), as differences in European and member state regulations did not allow for individual, patient level analyses of data from these nations. Case ascertainment rates have been reported previously at > 90% of aortic procedures performed in Denmark, Hungary, Malta, New Zealand, Sweden, and the UK, 80% in Norway and 63% in Australia.
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice – Assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      ,
      • Sawang M.
      • Paravastu S.C.V.
      • Liu Z.
      • Thomas S.D.
      • Beiles C.B.
      • Mwipatayi B.P.
      • et al.
      The relationship between aortic aneurysm surgery volume and peri-operative mortality in Australia.
      The Finnish registry collects complete data from the Helsinki region, but not the whole of Finland. The Vascular Quality Initiative (VQI) collects data from over 700 units in North America capturing about 23% of AAA repairs in the United States and the German Registry has over 130 participating centres. Further details on the vascular registries that contributed to this study are provided as supplementary information in the appendix. The mission of the International Consortium of Vascular Registries (ICVR) is to provide a collaborative platform through which registries and other stakeholders around the world can share data to improve vascular healthcare.

      Cohort

      Adult patients were eligible for the study if they had undergone elective OSR or EVAR for an intact AAA between January 2010 and December 2016. Patients undergoing complex open or endovascular AAA surgery, elective revision AAA surgery, and surgery for mycotic or infected AAAs were excluded. A complete dataset was created after combining data from the eight participating national vascular registries that provided patient level data and the three registries reporting aggregated results. Centres that participated in only one of the registries (EVAR or OSR) in the VQI were not included due to the need to calculate the proportion of EVAR procedures. Cases were excluded if the procedure type was not specified, or if patient age, sex, and post-operative mortality were not recorded: 155 cases were excluded (0.23%). Data on age, sex, comorbidities (history of medically treated diabetes, cardiac disease [ischaemic or congestive heart disease], renal impairment [creatinine ≥ 50 μmol/L]), as well as year of repair, operative approach (EVAR or open repair) and aneurysm diameter at time of repair were collected. Small aneurysm was defined as < 6 cm. Chronic obstructive pulmonary disease (COPD) data were not collected in two registries and therefore not available for risk adjustment. It was also not possible to investigate the influence of clamp site on OSR outcomes, as these data were not collected in all registries.

      Outcomes and statistical analysis

      The study was planned and performed in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational studies.
      The primary outcome measure was post-operative mortality. All registries except the Swedish (Swedvasc) and Danish Vascular registries collected in hospital mortality rates after AAA repair. In the Swedish registry, where a cross link to the population registry results in exact survival data, 30 day mortality was used as a proxy.
      Patient characteristics were examined by registry. Patient characteristics were examined by country and procedure approach (EVAR or OSR) and stratified by procedure era (2010 – 2013 and 2014 – 2016). These two time periods were determined before the data were analysed in accordance with the methodology of previous Vascunet studies.
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice – Assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      The proportion of EVAR procedures of all intact AAA repair repairs was also calculated. Mortality rates for each procedure were derived for intact EVAR and OSR across registries for the entire study period. Confidence intervals for mortality were derived using Wilson’s method. Differences in mortality and the proportion of EVAR procedures between time periods were compared using chi squared tests. EVAR and OSR cases were combined and the overall mortality was calculated following intact AAA repair procedures for all registries. The proportion of EVAR procedures was calculated out of all intact AAA repair procedures by registry. Aggregating data to registry level, the correlation between the proportion of EVAR reported by each registry and the overall mortality following intact AAA repair procedures was evaluated using a Pearson correlation coefficient. Subgroup analysis was calculated among patients with aneurysm diameters ≤ 6 cm, female patients, and octogenarians.
      Multivariable logistic regression models were used to obtain predicted adjusted mortality after EVAR and open repair for each registry that contributed individual level data. Multiple imputation was used to account for missing data. Data were complete for post-operative mortality, gender, age, and type of repair. A fully conditional specification method was used for arbitrary pattern to impute missing comorbidities and aneurysm diameter based on patient age, sex, procedure type, procedure year, and registry. Twenty imputations were performed and the mode of the predicted number of deaths was chosen for each registry. The parameter estimates were obtained by combining estimates from the twenty imputations using Rubin’s rule.
      • Rubin D.B.
      Multiple Imputation for Nonresponse in Surveys.
      Registries that contributed aggregated level data and had the statistical capacity (UK and Germany) applied the parameter estimates to their data and obtained predicted adjusted mortality. Missing covariates were imputed as average for this analysis. The assumption to interpret the predicted adjusted mortality for these two registries was that the effects of covariates on mortality would be similar in their registries to those in other registries based on which model was fitted. Comparisons between groups were performed using the chi squared test with statistical significance assumed at the p < .050 level. All analyses were performed using SAS 9.3 (Cary, NC).
      Ethics approval for the collection and analysis of vascular registry data was obtained on the basis of national regulations for each registry for this international collaborative project.

      Results

      The study population included a total of 103 715 intact AAA procedures with the majority being male (82.3%), and the prevailing technique for repair EVAR (67.3%). Patient characteristics by country are presented in Table 1 and by time period in Table 2. EVAR became more dominant over the study period, accounting for 63.6% of repairs in 2010 – 2013 and 71.2% in 2014 – 2016 (p < .001). Practice varied considerably between countries with the proportion of patients undergoing EVAR ranging from 35% in Hungary to 81% in the United States.
      Table 1Characteristics of patients with primary repair of intact abdominal aortic aneurysm (AAA) collected from 11 countries through the International Consortium of Vascular Registries (ICVR) for two time periods, 2010–2013 and 2014–2016
      AUSDKFINGERHUNMALNZNORSWEUKUSA
      Patients11 9333 9916227 1141 7221222 0654 0716 73434 31031 031
      Age
       <60416

      (3.5)
      211

      (5.3)
      47

      (7.6)
      695

      (9.8)
      208

      (12.1)
      3

      (2.5)
      65

      (3.1)
      233

      (5.7)
      219

      (3.3)
      915

      (2.7)
      2 211

      (7.1)
       60–692 733 (22.9)1 226

      (30.7)
      166

      (26.7)
      1 817

      (25.5)
      658

      (38.2)
      37

      (30.3)
      435

      (21.1)
      1 252

      (30.8)
      2 206

      (32.8)
      8 331

      (24.3)
      8 973

      (28.9)
       70–795 313

      (44.5)
      2 030

      (50.9)
      245

      (39.4)
      3 437

      (48.3)
      681

      (39.5)
      59

      (48.4)
      1 004

      (48.6)
      1 801

      (44.2)
      3 076

      (45.7)
      15 899

      (46.3)
      12 649

      (40.8)
       ≥803 471

      (29.1)
      524

      (13.1)
      164

      (26.4)
      1 165

      (16.4)
      175

      (10.2)
      23

      (18.9)
      561

      (27.2)
      785

      (19.3)
      1 233

      (18.3)
      9 165

      (26.7)
      7 198

      (23.2)
      Gender
       Male10 101

      (84.6)
      3 315

      (83.1)
      546

      (87.8)
      5 305

      (74.6)
      1 467

      (85.2)
      114

      (93.4)
      1 657

      (80.2)
      3 386

      (83.2)
      5 718

      (84.9)
      30 029

      (87.5)
      24 711

      (79.6)
       Female1 832

      (15.4)
      676

      (16.9)
      17

      (10.2)
      1 809

      (25.4)
      255

      (14.8)
      8

      (6.6)
      408

      (19.8)
      685

      (16.8)
      1 016

      (15.1)
      4 281

      (12.5)
      6 320

      (20.4)
      Diabetes
       No10 172

      (85.2)
      3 499

      (87.7)
      483

      (77.7)
      5 896

      (82.9)
      1 409

      (81.8)
      100

      (82.0)
      1 801

      (87.2)
      3 609

      (88.7)
      5 622

      (83.5)
      25 872

      (75.4)
      24 919

      (80.3)
       Yes1 760

      (14.7)
      451

      (11.3)
      114

      (18.3)
      1 218

      (17.1)
      313

      (18.2)
      22

      (18.0)
      264

      (12.8)
      462

      (11.3)
      876

      (13.0)
      4 610

      (13.4)
      6 069

      (19.6)
       Missing1

      (0.0)
      41

      (1.0)
      25

      (4.0)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      236

      (3.5)
      3 828

      (11.2)
      43

      (0.1)
      Cardiac
       No5 502

      (46.1)
      2 490

      (62.4)
      287

      (46.1)
      4 177

      (58.7)
      870

      (50.5)
      67

      (54.9)
      1 052

      (50.9)
      2 245

      (55.1)
      3 911

      (58.1)
      17 461

      (50.9)
      17 014

      (54.8)
       Yes6 430

      (53.9)
      1 421

      (35.6)
      306

      (49.2)
      2 937

      (41.3)
      852

      (49.5)
      55

      (45.1)
      1 013

      (49.1)
      1 826

      (44.9)
      2 535

      (37.6)
      13 538

      (39.5)
      13 981

      (45.1)
       Missing1

      (0.0)
      80

      (2.0)
      29

      (4.7)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      0

      (0.0)
      288

      (4.3)
      3 311

      (9.7)
      36

      (0.1)
      Renal impairment
       No10 939

      (91.7)
      3 254

      (81.5)
      478

      (76.8)
      6 888

      (96.8)
      0

      (0.0)
      110

      (90.2)
      1 886

      (91.3)
      3 820

      (93.8)
      4 707

      (69.9)
      29 033

      (84.6)
      28 373

      (91.4)
       Yes993

      (8.3)
      146

      (3.7)
      26

      (4.2)
      226

      (3.2)
      0

      (0.0)
      12

      (9.8)
      179

      (8.7)
      234

      (5.7)
      234

      (3.5)
      2 071

      (6.0)
      2 136

      (6.9)
       Missing1

      (0.0)
      591

      (14.8)
      118

      (19.0)
      0

      (0.0)
      1 722

      (100.0)
      0

      (0.0)
      0

      (0.0)
      17

      (0.4)
      1 793

      (26.6)
      3 206

      (9.3)
      522

      (1.7)
      Approach
       EVAR8 789

      (73.7)
      1 475

      (37.0)
      422

      (67.8)
      3 887

      (54.6)
      609

      (35.4)
      84

      (68.9)
      1 173

      (56.8)
      1 675

      (41.1)
      4 057

      (60.2)
      22 510

      (65.6)
      25 090

      (80.9)
       Open3 144

      (26.3)
      2 516

      (63.0)
      200

      (32.2)
      3 227

      (45.4)
      1 113

      (64.6)
      38

      (31.1)
      892

      (43.2)
      2 396

      (58.9)
      2 677

      (39.8)
      11 800

      (34.4)
      5 941

      (19.1)
      Small aneurysm
       No8 391

      (70.3)
      0

      (0.0)
      342

      (55.0)
      0

      (0.0)
      1 244

      (72.2)
      112

      (91.8)
      1 733

      (83.9)
      3 392

      (83.3)
      5 527

      (82.1)
      28 273

      (82.4)
      18 812

      (60.6)
       Yes3 324

      (27.9)
      0

      (0.0)
      52

      (8.4)
      0

      (0.0)
      462

      (26.8)
      10

      (8.2)
      331

      (16.0)
      678

      (16.7)
      1 205

      (17.9)
      2 479

      (7.2)
      11 781

      (38.0)
       Missing218

      (1.8)
      3 991

      (100.0)
      228

      (36.7)
      7 114

      (100.0)
      16

      (0.9)
      0

      (0.0)
      1

      (0.0)
      1

      (0.0)
      2

      (0.0)
      3 558

      (10.4)
      438

      (1.4)
      Data are presented as n (%). EVAR = endovascular aneurysm repair; AUS = Australia; DK = Denmark; FIN = Finland; GER = Germany; HUN = Hungary; MAL = Malta; NZ = New Zealand; NOR = Norway; SWE = Sweden; UK = United Kingdom; USA = United States of America.
      Table 2Characteristics of patients treated by endovascular aneurysm repair (EVAR) or open surgery for intact abdominal aortic aneurysm by time period
      2010–20132014–2016
      EVAR (n = 34 214)Open (n = 19 548)EVAR (n = 35 557)Open (n = 14 396)
      Age
       <601 119 (3.3)1 534 (7.8)1 294 (3.6)1 276 (8.9)
       60–697 441 (21.7)6 700 (34.3)8 325 (23.4)5 368 (37.3)
       70–7915 510 (45.3)8 769 (44.9)15 680 (44.1)6 235 (43.3)
       ≥8010 144 (29.6)2 545 (13.0)10 258 (28.8)1 517 (10.5)
      Gender
       Male29 143 (85.2)15 717 (80.4)29 896 (84.1)11 593 (80.5)
       Female5 071 (14.8)3 831 (19.6)5 661 (15.9)2 803 (19.5)
      Diabetes
       No25 921 (75.8)15 975 (81.7)29 040 (81.7)12 445 (86.4)
       Yes5 444 (15.9)2 302 (11.8)6 483 (18.2)1 931 (13.4)
       Missing2 849 (8.3)1 271 (6.5)34 (0.1)20 (0.1)
      Cardiac
       No16 159 (47.2)10 679 (54.6)19 411 (54.6)8 827 (61.3)
       Yes15 482 (45.3)7 792 (39.9)16 092 (45.3)5 528 (38.4)
       Missing2 573 (7.5)1 077 (5.5)54 (0.2)41 (0.3)
      Renal impairment
       No29 297 (85.6)16 475 (84.3)31 339 (88.1)12 377 (86.0)
       Yes2 276 (6.7)1 085 (5.6)2 245 (6.3)651 (4.5)
       Missing2 641 (7.7)1 988 (10.2)1 973 (5.5)1 368 (9.5)
      Small aneurysm
       No21 586 (63.1)12 518 (64.0)23 622 (66.4)10 100 (70.2)
       Yes6 925 (20.2)2 295 (11.7)9 181 (25.8)1 921 (13.3)
       Missing5 703 (16.7)4 735 (24.2)2 754 (7.7)2 375 (16.5)
      Data are presented as n (%).

      Peri-operative mortality

      Mortality decreased overall from 2.1% in 2010 – 2013 to 1.6% for 2014 – 2016 (p < .001) (Fig. 1). Mortality also declined significantly over time for OSR, 4.2% to 3.6 % (p = .002) and EVAR 1.0% to 0.7% (p = .002) (Fig. 2).
      Figure 1
      Figure 1Variation in intact abdominal aortic aneurysm (AAA) procedure combined mortality across registries by time period.
      Figure 2
      Figure 2Variation in mortality after (A) endovascular aneurysm repair (EVAR) and (B) open surgical repair (OSR) for intact abdominal aortic aneurysm across registries by time period.
      Low mortality rates in countries during the earlier time period were generally maintained, with marginal improvement when compared with the more contemporaneous period. For example, the United States improved from 1.3% to 1.2%, Sweden remained stable at 1.6%, and in Norway there was a slight increase in mortality from 1.4% to 1.5%. The greatest improvements were seen in the UK with mortality decreasing from 2.7% to 1.7%, Denmark 3.1% to 2.2%, and Australia 1.7% to 1.2 % (Fig. 1).
      Higher overall peri-operative mortality rates were observed in countries where a higher proportion of patients underwent OSR. Specifically, mortality in Hungary 3.4% (65% OSR), Denmark 2.7% (63% OSR), and Germany 2.7% (45% OSR) were higher relative than other nations. A weak correlation between the proportion of patients undergoing OSR and mortality was observed for outcomes based on the entire study period (correlation coefficient = –0.37, p = .26; correlation coefficient excluding Malta = –0.54, p = .11; Fig. 3).
      Figure 3
      Figure 3Correlation between proportion of endovascular aneurysm repair (EVAR) procedures and mortality after reported intact abdominal aortic aneurysm (AAA) repair procedures on the aggregated registry level. The size of the circles represents the number of total intact AAA repair procedures.

      Subgroup analysis

      Mortality was significantly higher for women than men (3.0% vs. 1.6% p < .001) and for both OSR (5.2% vs. 3.6% p < .001) and EVAR (1.5% vs. 0.7% p < .001), respectively (Fig. 4). In both time periods female patients were more likely than males to undergo OSR (19.6% vs. 14.8% 2010 – 2013 and 19.5% vs. 15.9% 2014 – 2016) and accounted for 19.5 % of all OSR cases. Conversely a higher proportion of men underwent endovascular repair 84.6%.
      Figure 4
      Figure 4Mortality for female patients, octogenarians, patients with aneurysm diameter < 6 cm after endovascular aneurysm repair (EVAR) and open surgical repair (OSR) for intact abdominal aortic aneurysm across registries.
      The percentage of patients over eighty undergoing AAA repair remained constant over time at 23.6% (p = .91). However, within this subgroup, the number of OSRs decreased from 25.1% (2010 – 2013) to 14.8% (2014 – 2016; p < .001). Peri-operative mortality was greater for patients over 80 years of age (2.7% vs. 1.6% p < .001) and for both OSR (9.3% vs. 3.3%) and EVAR (1.4% vs. 0.37%) (Fig. 4). Notably, 46% (n = 275) of all EVAR deaths occurred in the over 80 year old patient group.
      Peri-operative mortality for patients with pre-operative AAA diameters less than 6 cm was better than the entire cohort, 3% for OSR and 0.5% for EVAR (Fig. 4).

      Risk adjustment

      The risk adjusted analysis was consistent with the unadjusted analysis (Fig. 5). Variation in adjusted predicted mortality across registries was smaller, particularly for OSR (range of unadjusted mortality 2.5% – 7.9% vs. adjusted mortality: 3.2% – 5.3%), indicating that part of the variation in unadjusted mortality was due to differences in the characteristics of treated patients.
      Figure 5
      Figure 5Observed and predicted adjusted mortality after endovascular aneurysm repair (EVAR) and open surgical repair (OSR) for intact abdominal aortic aneurysm across registries. Logistic regression models adjusted for patient age, sex, procedure year, and comorbidities. Parameter estimates were applied to UK and German data with the assumption that the effects of covariates on mortality were similar in these registries to those in registries based on which the logistic regression models were developed.

      Discussion

      In this large study of more than 100 000 intact AAA repairs, a reduction in peri-operative mortality was observed for both OSR and EVAR over time from 2.1% between 2010 and 2013 to 1.6 % between 2014 and 2016. Although some of this reduction may reflect an increase in the proportion of patients undergoing endovascular repair, it is clear that mortality has decreased for both OSR (4.2% – 3.6%) and EVAR (1% – 0.7%). Within this large cohort, there was significant variation in both practice and outcomes by country. Not surprisingly, those countries that performed a higher proportion of OSR had greater overall mortality rates. It is interesting that despite eight of the countries in this study being European, practice is clearly not standardised and there remain significant variations in the interpretation of the ESVS AAA guidelines in clinical practice.
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor's choice – European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      Outcomes for the best performing countries, have demonstrated only marginal improvement over time, and it may be that the low mortality rates seen in the United States, Sweden, and Norway have little room for further improvement. This also points towards the need for other relevant quality improvement parameters for evaluation of AAA repair practice, including long term survival and need for re-intervention after EVAR.
      The most significant improvement in outcomes was observed in the UK with overall mortality falling from 2.7% to 1.7%. These improved outcomes probably reflect the residual impact of the AAA Quality Improvement Programme (QIP),
      • Mitchell D.
      • Hindley H.
      • Naylor A.R.
      • Wyatt M.
      • Loftus I.M.
      Outcomes after elective repair of infra-renal abdominal aortic aneurysm. A report from the Vascular Society.
      the regular feedback both at unit and surgeon level provided by the National Vascular Registry (NVR)
      • Waton S.
      • Johal A.
      • Heikkila K.
      • Cromwell D.
      • Boyle J.
      • Miller F.
      National Vascular Registry: 2019 annual report.
      and the reconfiguration of vascular services, which reduced the number of centres performing aneurysm surgery. The overall decline in OSR mortality is likely to have been influenced by the UK, which contributed more than a third of the patient cohort. Both the QIP and centralisation of vascular services in the UK were in part driven by a previous report from VASCUNET
      • Gibbons C.
      • Bjorck M.
      • Jensen L.P.
      • Laustsen J.
      • Lees T.
      • Moreno-Carriles R.
      • et al.
      The Second Vascular Surgery Database Report.
      that demonstrated high mortality after AAA repair in the UK, highlighting the importance of this type of comparative registry study in driving quality improvement. Outcomes for intact AAA repair have continued to improve in the UK, with a mortality rate of just 1.4% reported for 2016 – 2018.
      • Waton S.
      • Johal A.
      • Heikkila K.
      • Cromwell D.
      • Boyle J.
      • Miller F.
      National Vascular Registry: 2019 annual report.
      One of the most concerning findings of this study is the high mortality rate for both OSR and EVAR in female patients. This has previously been reported from a large series in the UK for both OSR and EVAR and consistent with the findings of this study identifying that women are more likely to undergo OSR than men.
      • Sidloff D.A.
      • Saratzis A.
      • Sweeting M.J.
      • Michaels J.
      • Powell J.T.
      • Thompson S.G.
      • et al.
      Sex differences in mortality after abdominal aortic aneurysm repair in the UK.
      A contemporary study from The Netherlands found that female sex was independently associated with peri-operative mortality, although only for OSR.
      • Indrakusuma R.
      • Jalalzadeh H.
      • Vahl A.C.
      • Koelemay M.J.W.
      • Balm R.
      Editor's Choice – Sex related differences in peri-operative mortality after elective repair of an asymptomatic abdominal aortic aneurysm in The Netherlands: a retrospective analysis of 2013 to 2018.
      Analogous to the findings, the authors also reported a higher proportion of female patients undergoing OSR, 30% compared with 21% of men. A recent meta-analysis has confirmed significantly higher 30 day mortality in women following EVAR.
      • Liu Y.
      • Yang Y.
      • Zhao J.
      • Chen X.
      • Wang J.
      • Ma Y.
      • et al.
      Systematic review and meta-analysis of sex differences in outcomes after endovascular aneurysm repair for infrarenal abdominal aortic aneurysm.
      The outcomes for intact AAA repair in patients over eighty reported in this study were significantly worse overall and particularly for mortality after OSR at 9.3%. Although peri-operative mortality for EVAR was lower at 1.4%, nearly half of all EVAR deaths in this study occurred in this group. Previous studies have demonstrated that long term survival following EVAR in this patient group is equivalent to an age matched population, being 50% at five years and less than 30% at 10 years.
      • Rueda-Ochoa O.L.
      • van Bakel P.
      • Hoeks S.E.
      • Verhagen H.
      • Deckers J.
      • Rizopoulos D.
      • et al.
      Survival after uncomplicated EVAR in octogenarians is similar to the general population of octogenarians without an abdominal aortic aneurysm.
      ,
      • Johal A.
      • Loftus I.M.
      • Boyle J.R.
      • Heikkila K.
      • Waton S.
      • Cromwell D.A.
      Long-term survival after endovascular and open repair of unruptured abdominal aortic aneurysm.
      Importantly, octogenarians were not recruited in randomised trials that identified the threshold for AAA repair,
      Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants.
      ,
      • Lederle F.A.
      • Johnson G.R.
      • Wilson S.E.
      • et al.
      Prevalence and associations of abdominal aortic aneurysm detected through screening. Aneurysm Detection and Management (ADAM) Veterans Affairs Cooperative Study Group.
      and were a small minority of those randomised in the trials comparing EVAR with OSR.
      • Patel R.
      • Sweeting M.J.
      • Powell J.T.
      • Greenhalgh R.M.
      EVAR Trial Investigators
      Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial.
      ,
      • Lederle F.A.
      • Stroupe K.T.
      • Kyriakides T.C.
      • Ge L.
      • Freischlag J.A.
      Open vs Endovascular Repair (OVER) Veterans Affairs Cooperative Study Group. Long-term cost-effectiveness in the veterans affairs open vs endovascular repair study of aortic abdominal aneurysm: a randomized clinical trial.
      Simultaneously, the octogenarian population has increased significantly in many countries over the past decades, and an increasing number of patients undergoing intact AAA repair are in this age group.
      • Lilja F.
      • Mani K.
      • Wanhainen A.
      Editor's Choice – Trend-break in abdominal aortic aneurysm repair with decreasing surgical workload.
      Treating aneurysms in the frail and elderly may not prolong life and may be unjustifiable in some cases.
      ,
      • Loftus I.M.
      • Haulon S.
      • Boyle J.R.
      NICE abdominal aortic aneurysm guidelines finally published: how will they influence aortic practice in the UK and beyond?.
      Certainly, these findings underscore the need for further study aimed at optimising treatments and outcomes in older patients.
      Although the observed decline in EVAR mortality to only 0.7% that is reported in this study should be applauded, it also raises the question as to whether death remains the best outcome measure following EVAR.
      • Boyle J.R.
      Is in hospital mortality following EVAR still a valid outcome measure?.
      The durability of EVAR has come under increasing scrutiny and longterm rupture rates following EVAR are more than three times greater than following OSR.
      • Johal A.
      • Loftus I.M.
      • Boyle J.R.
      • Heikkila K.
      • Waton S.
      • Cromwell D.A.
      Long-term survival after endovascular and open repair of unruptured abdominal aortic aneurysm.
      The quality of endovascular repair is perhaps better assessed by longer term outcome measures, such as re-intervention and AAA rupture. Although the society guidelines from North America and Europe both recommend EVAR as the primary repair modality for patients with anatomically suitable AAA,
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor's choice – European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      ,
      • Chaikof E.L.
      • Dalman R.L.
      • Eskandari M.K.
      • et al.
      The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm.
      the variation in use of EVAR between countries as demonstrated in this analysis highlights that some uncertainty remains in this field. Further, this is emphasised by the recently published NICE Guidelines in the UK.
      Contrary to society guidelines, NICE has recommended offering OSR for intact aneurysms, unless contraindicated by co-morbidity, anaesthetic risks or hostile abdomen. Whilst not questioning the early survival benefit of EVAR, NICE highlighted concerns related to the longterm outcomes and cost effectiveness of endovascular repair, whilst accepting the importance of patient choice.
      The durability of EVAR and associated high re-intervention rates remain a concern that have yet to be overcome by new technology.
      • Harrison S.C.
      • Winterbottom A.J.
      • Coughlin P.A.
      • Hayes P.D.
      • Boyle J.R.
      Editor's choice – Mid-term migration and device failure following endovascular aneurysm sealing with the Nellix stent graft system – a single centre experience.
      Capturing longer term outcomes in registries is clearly important and the recent addition of re-intervention and revision fields the NVR in the UK along with the capture of device specific data will deliver this in the future.
      • Mahase E.
      National registry will record patient outcomes from devices implanted during vascular surgery.
      While evaluation of longterm survival is possible in some vascular registries through linkage with population registries,
      • Sonesson B.
      • Björses K.
      • Dias N.
      • Rylance R.
      • Mani K.
      • Wanhainen A.
      • Resch T.
      Outcome after ruptured AAA repair in octo- and nonagenarians in Sweden 1994–2014.
      these data were unfortunately not available in this international registry collaboration.
      There are some limitations within this study. Although a number of countries with mature vascular registries have very high case ascertainment rates for AAA repair and some have undergone external validations by international validators and been demonstrated to be both accurate and reliable,
      • Venermo M.
      • Lees T.
      International Vascunet validation of the Swedvasc registry.
      • Altreuther M.
      • Menyhei G.
      International validation of the Danish Vascular Registry Karbase: a Vascunet report.
      • Bergqvist D.
      • Bjorck M.
      • Lees T.
      • Menyhei G.
      Validation of the VASCUNET registry – pilot study.
      • Eldrup N.
      • Debus E.S.
      International validation of the population based Malta Vascular Registry: A Vascunet report.
      this is not the case for all countries in this study. The VQI program is a collaboration of Society of Vascular Surgery Patient Safety Organisations, it is voluntary and does not have complete coverage of the United States and Canada, most recent data suggest it captures 23% of US AAA repairs.
      • Dansey K.D.
      • de Guerre L.E.V.M.
      • Swerdlow N.J.
      • Li C.
      • Lu J.
      • Patel P.B.
      • et al.
      Not all databases are created equal. A comparison of administrative data and quality improvement registries for abdominal aortic aneurysm repair.
      Given the focus on quality improvement by this group, the results from participating centres may be better than for the entirety of patients treated in these countries. Contemporary evidence from the United States confirms that this is indeed the case for both open intact AAA and ruptured AAA repairs, with significantly higher mortalities recorded in the National Inpatient Sample (NIS) than in the VQI (Intact NIS OSR mortality 5.4% vs. 3.5% for VQI).
      Validation of the Australian data in the Australasian vascular audit. Beiles CB, Bourke BM.
      Similarly, the datasets from Germany, Finland and Australia included in this study did not include all AAA repairs performed in these countries. Reported outcomes therefore may be prone to selection bias. Despite the Australian Vascular Audit being compulsory, evidence suggests that it only contains 63% of procedures, and therefore outcomes may be prone to selection bias.
      • Sawang M.
      • Paravastu S.C.V.
      • Liu Z.
      • Thomas S.D.
      • Beiles C.B.
      • Mwipatayi B.P.
      • et al.
      The relationship between aortic aneurysm surgery volume and peri-operative mortality in Australia.
      ,
      Validation of the Australian data in the Australasian vascular audit. Beiles CB, Bourke BM.
      Furthermore, the retrospective nature of the analysis performed in this study cannot account for all the potential differences in patient selection between countries and this makes comparisons difficult. Some confounders within the analysis have been adjusted for, but this adjustment was bound by the available data and adjustment for co-morbidity was limited to diabetes, cardiac disease, and renal impairment. There was a higher percentage of patients with small AAAs undergoing EVAR in the later period, and this may have contributed to improved outcomes, although this must be viewed in the context of a higher proportion of missing data on AAA diameter in the 2010 – 2013 period.
      In conclusion, the proportion of AAA repairs performed by EVAR has increased over time in this large international study. Peri-operative mortality continues to fall for both open AAA repair and EVAR. Importantly, outcomes are significantly worse for both women and those aged over 80 years. Individualised patient selection focused on aneurysm diameter, rupture risk, age, expected patient longevity and gender might lead to further improvement in outcomes.

      Conflicts of interest

      None.

      Funding

      The project and efforts were funded in part by the US Food and Drug Administration through grant 1U01FD005478 (PI Sedrakyan). Views expressed in the publication do not necessarily reflect the official policies of the Department of Health and Human Services; nor does any mention of trade names, commercial practices, or organisation imply endorsement by the US government.

      Acknowledgements

      Quiju Li for comments on the draft manuscript. Danica Marinac-Dabic, FDA Representative on the ICVR.

      References

        • Beck A.W.
        • Sedrakyan A.
        • Mao J.
        • Venermo M.
        • Faizer R.
        • Debus S.
        • et al.
        Variations in abdominal aortic aneurysm care: a report from the international consortium of vascular registries.
        Circulation. 2016; 134: 1948-1958
        • Budtz-Lilly J.
        • Venermo M.
        • Debus S.
        • Behrendt C.A.
        • Altreuther M.
        • Beiles B.
        • et al.
        Editor's choice – Assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
        Eur J Vasc Endovasc Surg. 2017; 54: 13-20
        • Wanhainen A.
        • Verzini F.
        • Van Herzeele I.
        • Allaire E.
        • Bown M.
        • Cohnert T.
        • et al.
        Editor's choice – European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
        Eur J Vasc Endovasc Surg. 2019; 57: 8-93
        • Behrendt C.A.
        • Venermo M.
        • Cronenwett J.L.
        • Sedrakyan A.
        • Beck A.W.
        • Eldrup-Jorgensen J.
        • Mani K.
        VASCUNET, the Vascular Quality Initiative and the International Consortium of Vascular Registries. VASCUNET, VQI and the International Consortium of Vascular Registries – Unique Collaborations for Quality Improvement in Vascular Surgery.
        Eur J Vasc Endovasc Surg. 2019; 58: 792-793
        • Gibbons C.
        • Bjorck M.
        • Jensen L.P.
        • Laustsen J.
        • Lees T.
        • Moreno-Carriles R.
        • et al.
        The Second Vascular Surgery Database Report.
        European Society for Vascular Surgery, 2008 (Available at:) ([Accessed April 2020])
        • Mitchell D.
        • Hindley H.
        • Naylor A.R.
        • Wyatt M.
        • Loftus I.M.
        Outcomes after elective repair of infra-renal abdominal aortic aneurysm. A report from the Vascular Society.
        The Abdominal Aortic Aneurysm Quality Improvement Programme, March 2012 (Available at:) ([Accessed April 2020])
        • Scali S.T.
        • Beck A.W.
        • Sedrakyan A.
        • Mao J.
        • Venermo M.
        • Faizer R.
        • et al.
        Hospital volume association with abdominal aortic aneurysm repair mortality: Analysis of the international consortium of vascular registries.
        Circulation. 2019; 140: 1285-1287
        • Sawang M.
        • Paravastu S.C.V.
        • Liu Z.
        • Thomas S.D.
        • Beiles C.B.
        • Mwipatayi B.P.
        • et al.
        The relationship between aortic aneurysm surgery volume and peri-operative mortality in Australia.
        Eur J Vasc Endovasc Surg. 2019; 57: 510-519
        • Rubin D.B.
        Multiple Imputation for Nonresponse in Surveys.
        John Wiley & Sons, New York2004
        • Waton S.
        • Johal A.
        • Heikkila K.
        • Cromwell D.
        • Boyle J.
        • Miller F.
        National Vascular Registry: 2019 annual report.
        The Royal College of Surgeons of England, London2019 (Available at:) ([Accessed April 2020])
        • Sidloff D.A.
        • Saratzis A.
        • Sweeting M.J.
        • Michaels J.
        • Powell J.T.
        • Thompson S.G.
        • et al.
        Sex differences in mortality after abdominal aortic aneurysm repair in the UK.
        Br J Surg. 2017; 104: 1656-1664
        • Indrakusuma R.
        • Jalalzadeh H.
        • Vahl A.C.
        • Koelemay M.J.W.
        • Balm R.
        Editor's Choice – Sex related differences in peri-operative mortality after elective repair of an asymptomatic abdominal aortic aneurysm in The Netherlands: a retrospective analysis of 2013 to 2018.
        Eur J Vasc Endovasc Surg. 2019; 58: 813-820
        • Liu Y.
        • Yang Y.
        • Zhao J.
        • Chen X.
        • Wang J.
        • Ma Y.
        • et al.
        Systematic review and meta-analysis of sex differences in outcomes after endovascular aneurysm repair for infrarenal abdominal aortic aneurysm.
        J Vasc Surg. 2020; 71: 283-296
        • Rueda-Ochoa O.L.
        • van Bakel P.
        • Hoeks S.E.
        • Verhagen H.
        • Deckers J.
        • Rizopoulos D.
        • et al.
        Survival after uncomplicated EVAR in octogenarians is similar to the general population of octogenarians without an abdominal aortic aneurysm.
        Eur J Vasc Endovasc Surg. 2020; 59: 740-747
        • Johal A.
        • Loftus I.M.
        • Boyle J.R.
        • Heikkila K.
        • Waton S.
        • Cromwell D.A.
        Long-term survival after endovascular and open repair of unruptured abdominal aortic aneurysm.
        Br J Surg. 2019; 106: 1784-1793
      1. Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants.
        Lancet. 1998; 352: 1649-1655
        • Lederle F.A.
        • Johnson G.R.
        • Wilson S.E.
        • et al.
        Prevalence and associations of abdominal aortic aneurysm detected through screening. Aneurysm Detection and Management (ADAM) Veterans Affairs Cooperative Study Group.
        Ann Intern Med. 1997; 126: 441-449
        • Patel R.
        • Sweeting M.J.
        • Powell J.T.
        • Greenhalgh R.M.
        • EVAR Trial Investigators
        Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial.
        Lancet. 2016; 388: 2366-2374
        • Lederle F.A.
        • Stroupe K.T.
        • Kyriakides T.C.
        • Ge L.
        • Freischlag J.A.
        Open vs Endovascular Repair (OVER) Veterans Affairs Cooperative Study Group. Long-term cost-effectiveness in the veterans affairs open vs endovascular repair study of aortic abdominal aneurysm: a randomized clinical trial.
        JAMA Surg. 2016; 151: 1139-1144
        • Lilja F.
        • Mani K.
        • Wanhainen A.
        Editor's Choice – Trend-break in abdominal aortic aneurysm repair with decreasing surgical workload.
        Eur J Vasc Endovasc Surg. 2017; 53: 811-819
      2. NICE guideline [NG156]. March 2020 (Available at:) ([Accessed April 2020])
        • Loftus I.M.
        • Haulon S.
        • Boyle J.R.
        NICE abdominal aortic aneurysm guidelines finally published: how will they influence aortic practice in the UK and beyond?.
        Eur J Vasc Endovasc Surg. 2020; 59: 697-698
        • Boyle J.R.
        Is in hospital mortality following EVAR still a valid outcome measure?.
        Eur J Vasc Endovasc Surg. 2019; 57: 520
        • Chaikof E.L.
        • Dalman R.L.
        • Eskandari M.K.
        • et al.
        The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm.
        J Vasc Surg. 2018; 67: 2-77
        • Harrison S.C.
        • Winterbottom A.J.
        • Coughlin P.A.
        • Hayes P.D.
        • Boyle J.R.
        Editor's choice – Mid-term migration and device failure following endovascular aneurysm sealing with the Nellix stent graft system – a single centre experience.
        Eur J Vasc Endovasc Surg. 2018; 56: 342-348
        • Mahase E.
        National registry will record patient outcomes from devices implanted during vascular surgery.
        BMJ. 2020; 370: m2981
        • Sonesson B.
        • Björses K.
        • Dias N.
        • Rylance R.
        • Mani K.
        • Wanhainen A.
        • Resch T.
        Outcome after ruptured AAA repair in octo- and nonagenarians in Sweden 1994–2014.
        Eur J Vasc Endovasc Surg. 2017; 53: 656-662
        • Venermo M.
        • Lees T.
        International Vascunet validation of the Swedvasc registry.
        Eur J Vasc Endovasc Surg. 2015; 50: 802-808
        • Altreuther M.
        • Menyhei G.
        International validation of the Danish Vascular Registry Karbase: a Vascunet report.
        Eur J Vasc Endovasc. 2019; 58: 609-613
        • Bergqvist D.
        • Bjorck M.
        • Lees T.
        • Menyhei G.
        Validation of the VASCUNET registry – pilot study.
        VASA. 2014; 43: 141-144
        • Eldrup N.
        • Debus E.S.
        International validation of the population based Malta Vascular Registry: A Vascunet report.
        Eur J Vasc Endovasc Surg Forum. 2020; 48: 5-7
        • Dansey K.D.
        • de Guerre L.E.V.M.
        • Swerdlow N.J.
        • Li C.
        • Lu J.
        • Patel P.B.
        • et al.
        Not all databases are created equal. A comparison of administrative data and quality improvement registries for abdominal aortic aneurysm repair.
        J Vasc Surg. 2021; 73: 874-888
      3. Validation of the Australian data in the Australasian vascular audit. Beiles CB, Bourke BM.
        ANZ J Surg. 2014; 84: 624-627

      Linked Article

      • Outcomes for Intact Abdominal Aortic Aneurysm Repair: What to do With Frailty and Quality of Life?
        European Journal of Vascular and Endovascular SurgeryVol. 62Issue 1
        • Preview
          In their paper, Boyle et al. combined the data from more than 100 000 patients from 11 registries in Europe, Australia, and the USA, and reported the outcomes of elective abdominal aortic aneurysm repair.1 Results were compared for age, sex, and country over two time periods (2010 – 2013 and 2014 – 2016). A small but reassuring reduction in peri-operative mortality for open (from 4.2% to 3.6%) and endovascular repair (EVAR; from 1.0% to 0.7%) was demonstrated over time.1
        • Full-Text
        • PDF
        Open Archive

      Comments

      Commenting Guidelines

      To submit a comment for a journal article, please use the space above and note the following:

      • We will review submitted comments as soon as possible, striving for within two business days.
      • This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
      • We require that commenters identify themselves with names and affiliations.
      • Comments must be in compliance with our Terms & Conditions.
      • Comments are not peer-reviewed.