If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Editor's Choice – Systematic Review and Meta-Analysis of Sex Specific Differences in Adverse Events After Open and Endovascular Intact Abdominal Aortic Aneurysm Repair: Consistently Worse Outcomes for Women
Previously, reports have shown that women experience a higher mortality rate than men after elective open (OAR) and endovascular (EVAR) repair of abdominal aortic aneurysm (AAA). With recent improvements in overall AAA repair outcomes, this study aimed to identify whether sex specific disparity has been ameliorated by modern practice, and to define sex specific differences in peri- and post-operative complications and pre-operative status; factors which may contribute to poor outcome.
Methods
This was a systematic review, meta-analysis, and meta-regression of sex specific differences in 30 day mortality and complications conducted according to PRISMA guidance (Prospero registration CRD42020176398). Papers with ≥ 50 women, reporting sex specific outcomes, following intact primary AAA repair, from 2000 to 2020 worldwide were included; with separate analyses for EVAR and OAR. Data sources were Medline, Embase, and CENTRAL databases 2005 – 2020 searched using ProQuest Dialog.
Results
Twenty-six studies (371 215 men, 65 465 women) were included. Meta-analysis and meta-regression indicated that sex specific odds ratios (ORs) for 30 day mortality were unchanged from 2000 to 2020. Mortality risk was higher in women for OAR and more so for EVAR (OR [95% CI] 1.49 [1.37 – 1.61]; 1.86 [1.59 – 2.17], respectively) and this remained following multivariable risk adjustment. Transfusion, pulmonary complications, and bowel ischaemia were more common in women after OAR and EVAR (OAR: ORs 1.81 [1.60 – 2.04], 1.40 [1.28 – 1.53], 1.54 [1.36 – 1.75]; EVAR: ORs 2.18 [2.08 – 2.29] 1.44 [1.17 – 1.77], 1.99 [1.51 – 2.62], respectively). Arterial injury, limb ischaemia, renal and cardiac complications were more common in women after EVAR (ORs 3.02 [1.62 – 5.65], 2.13 [1.48 – 3.06], 1.46 [1.22 – 1.72] and 1.19 [1.03 – 1.37], respectively); the latter was associated with greater mortality risk on meta-regression.
Conclusion
Increased mortality risk for women following AAA repair remains. Women had a higher incidence of transfusion, pulmonary and bowel complications after EVAR and OAR. Higher mortality risk ratios for EVAR may result from cardiac complications, additional arterial injury, and embolisation, leading to renal and limb ischaemia. These findings indicate possible causes for observed outcome disparities and targets for quality improvement.
This study demonstrates that the risk of worse outcomes in women (vs. men) after elective abdominal aortic aneurysm repair has not ameliorated with time and the risk ratio for mortality is higher for EVAR than open repair. Sex specific differences varied with type of repair, allowing insight into the increased mortality risk differential for EVAR. Better pre-operative optimisation might reduce the higher risk of systemic complications in women. Arterial injury, renal, and other ischaemic complications (limb and bowel ischaemia), which probably contribute to mortality, are higher in women, especially after EVAR, and may be avoidable with strategic planning.
Introduction
Higher operative mortality for women (vs. men) undergoing open (OAR) or endovascular (EVAR) abdominal aortic aneurysm (AAA) repair has been reported worldwide.
Effect modification of sex and age for the outcome relationship in abdominal aortic aneurysm treatment: German Diagnosis Related Groups Statistics from 2005 to 2014.
This discrepancy remains following multivariable risk factor adjustment, indicating that differences in age or pre-operatively identified comorbid status do not adequately explain the increase in mortality risk for women.
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Salient analysis of key sex specific differences is hindered by substantial under representation of women in major trials, disparate reporting of comorbidities and outcomes, and frequent pooled reporting of intact and ruptured AAA repair.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.
Further investigation, and a greater understanding of key contributory factors, such as sex specific differences in peri- and post-operative complications, is needed.
Additionally, over the last 20 years, improvement in pre-operative optimisation, patient selection, and multidisciplinary supportive care have improved outcomes dramatically for AAA repair.
Advances in available devices and techniques for minimally invasive endovascular repair have also considerably reduced peri-operative mortality, can be performed under local anaesthesia, and reduce recovery time.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
However, while these advances are advantageous to both sexes, it is unclear whether improvements have been conferred equally, with some reports suggesting a widening of the mortality risk differential.
A recent meta-analysis, conducted by Liu et al., explored sex specific differences after EVAR for infrarenal AAA, demonstrating an increased risk of mortality, limb ischaemia, cardiac, and renal complications for women.
However, because of the inclusion of patient cases from an outdated era, prior to key advances in endovascular techniques, low profile devices, and anaesthesia and intensive care strategies, these results may not be reflective of modern practice.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Midterm pivotal trial results of the Talent Low Profile System for repair of abdominal aortic aneurysm: analysis of complicated versus uncomplicated aortic necks.
Indeed, more recent studies have reported that despite increased peri-procedural complications for women, there is no difference in survival following EVAR.
Additionally, without comparison with OAR, it was not possible to assess whether differences in the complication risks reported by Liu et al. were a result of disadvantages specifically associated with EVAR affecting women disproportionately.
This study aimed to explore the hypothesis that, in modern practice, women continue to be at greater risk of death and post-operative complications from elective AAA repair (with an increased risk differential for EVAR over OAR), as well as to examine potential pre-, peri-, and post-operative causes for this disparity. This was performed using a systematic review, meta-analysis, and meta-regression, examining sex specific differences in mortality, post-operative complications, and pre-operative status, for both OAR and EVAR for intact AAA.
Methods
The protocol for this systematic review and meta-analysis was registered with the International Prospective Register of Systematic Reviews (PROSPERO, CRD42020176398). The study has been conducted and reported in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines (Supplementary Table S1).
Chapter 10: Analysing data and undertaking meta-analyses.
in: Higgins J.P.T. Thomas J. Chandler J. Cumpston M. Li T. Page M.J. Welch V.A. Cochrane Handbook for Systematic Reviews of Interventions version 6.1. 2020
An update of a comprehensive search strategy ProQuest Dialog was created for MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials (CENTRAL) using a combination of controlled vocabulary (MeSH or EMTREE) terms and free text terms, with a final search data of 31 January 2021.
Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.
Papers with ≥ 50 women, reporting sex specific outcomes, following intact, primary AAA repair, with separate reporting for EVAR and OAR, from 2000 to 2020 worldwide were included. The detailed search strategy, inclusion and exclusion criteria are given in Table 1. The criterion of ≥ 50 women was established by a senior statistician, for an earlier set of reviews, to reduce bias and the confidence intervals on the odds ratios (ORs) (which relate to both the number of events and the size of the study).
Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.
Manual searches of further information sources, such as the references from relevant published studies or reviews, conference abstracts, and supplementary notes, were also conducted. If necessary, authors were contacted to provide clarification or identify additional data.
Potentially eligible reports were assessed by two independent reviewers, with discrepancies resolved by a third author and pre-specified data extracted (Supplementary Text S2). Publication quality and risk of bias was assessed using the Newcastle Ottawa Scale.
Wells G, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for Assessing The Quality Of Nonrandomised Studies In Meta-Analyses. Ottawa Hospital Research Institute. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp [Accessed 3 Feb 2021].
The methodology for statistical analysis was conducted in accordance with the Cochrane Handbook for Systematic Reviews, using RevMan 5.3 software and RStudio.
In the absence of significant heterogeneity of study design and methodology (e.g. heterogeneity in definition of outcome variable), quantitative pooled meta-analyses of these data were conducted using the Mantel-Haenszel (dichotomous data), inverse variance (continuous data), or generic inverse variance (risk adjusted mortality ORs) method with a random effects model.
Where sufficient data were identified (at least eight studies), meta-regression analyses were conducted to examine aggregate (study level) effects of significant covariates (e.g. midpoint of study, difference in age, and log odds of significant complications), on between study heterogeneity for log odds of operative mortality. A mixed effects model was performed, with studies weighted by a combination of the inverse of within study variance and the residual heterogeneity.
For each meta-regression, the change in outcome per unit increase in covariable (βi, slope co-efficient (se, standard error)), the strength of association (p value), and the residual heterogeneity (tau2) are reported. A p value of < .050 was considered to be statistically significant. Heterogeneity was assessed using a chi square test and the I2 statistic, and publication bias was assessed using funnel plots, Egger’s test and Duval and Tweedie’s trim and fill procedure.
Sensitivity analyses were undertaken to evaluate the stability of results (Supplementary Text S2).
Results
Study selection
A total of 26 studies were eligible for inclusion. All studies were retrospective (in one case a retrospective analysis of data from prospective randomised controlled trials
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
), 18 reported OAR and 25 reported EVAR, with a total of 371 215 men and 65 465 women (Fig. 1, Table 2). All studies had a low to moderate risk of bias on the Newcastle Ottawa scale (Table 1, Supplementary Fig. S3). For each reported variable, studies and supplementary data were examined to establish sufficient consistency in definitions to enable quantitative synthesis of data (Supplementary Tables S4 and S5).
Figure 1Preferred reporting items for Systematic Reviews and Meta-Analysis (PRISMA) flow diagram for search for studies reporting sex specific outcomes after open or endovascular (EVAR) repair of abdominal aortic aneurysm.
Table 2Summary of included studies reporting sex specific outcomes after open aortic repair (OAR) or endovascular aneurysm repair (EVAR) of abdominal aortic aneurysm (AAA)
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Effect modification of sex and age for the outcome relationship in abdominal aortic aneurysm treatment: German Diagnosis Related Groups Statistics from 2005 to 2014.
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.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Similar 5-year outcomes between female and male patients undergoing elective endovascular abdominal aortic aneurysm repair with the Ovation stent graft.
Data are presented as median or median (interquartile range), or mean ± standard deviation, unless stated otherwise. NO = Newcastle Ottawa score; IRAAA = infrarenal AAA; cAAA = complex (juxta, para, or supra renal) AAA; ACE = Anevrysme de l’aorte abdominale, Chirurgie vs. Endoprothese trial; DAK = Gesundheit insurance database; DREAM = Dutch Randomised Endovascular Aneurysm Management trial; DSAA = Dutch Surgical Aneurysm Audit; ENGAGE = Endurant Stent Graft Natural Selection Global Postmarket registry; EVAR-1= Endovascular vs. open repair of abdominal aortic aneurysm trial; GFSO = German Federal Statistical Office; HES = Hospital Episode Statistics; JACSM= JApan Committee for Stentgraft Management registry; NVR = National Vascular Registry; NSQIP = American College of Surgeons’ National Surgical Quality Improvement Program; ONS = Office of National Statistics, UK; OVER = Open vs. Endovascular Repair trial; Swedvasc = Swedish vascular registry; SVS-VQI = Society for Vascular Surgery, Vascular Quality Initiative; VISCOAP = Washington State’s Vascular Interventional Surgical Care and Outcomes Assessment Program; VSGNE = Vascular Study Group of New England database.
∗ Reported outcomes: Mortality: 1a, 30 d; 1b, in hospital; 1c long term. Peri/post-operative complications: 2a, blood transfusion, 2b, arterial injury, 2c, limb ischaemia, 2d, bowel ischaemia, 2e, cardiac, 2f, respiratory, 2g, neurological and 2h, renal complications, 2i, re-intervention and 2j, endoleaks. Duration of treatment: 3a operative time, 3b length of intensive care and 3c hospital stay. Pre-operative comorbidities and risk factors: 4a cardiovascular, 4b, respiratory, 4c, renal, 4d, diabetes, 4e, peripheral vascular disease, 4f, smoker, 4g, comorbidity score.
Overall, 13 studies (12 374 women, 50 387 men) reported 30 day mortality for OAR, ranging from 3.64% – 9.53% for women and 1.35% – 6.81% for men. Cumulative analysis demonstrated statistically significantly higher odds of death for women, without evidence of heterogeneity (OR [95% CI] 1.49 [1.37 – 1.61], p < .001, I2 = 0) (Fig. 2). Reported 30 day mortality outcomes for EVAR, from 15 non-conflicting studies (18 780 women, 93 160 men), ranged from 0 – 6.29% for women and 0.57% – 2.67% for men. Cumulative analysis demonstrated an even greater differential mortality risk for EVAR (OR 1.86 [1.59 – 2.17], p < .001, I2 = 23) (Fig. 2). Pooled analysis of 30 day and/or in hospital mortality reporting demonstrated similar results (OAR OR 1.47 [1.38 – 1.57], p < .001, I2 = 0; EVAR OR 1.77 [1.51 – 2.08], p < .001, I2 = 48) (Supplementary Fig. S6).
Effect modification of sex and age for the outcome relationship in abdominal aortic aneurysm treatment: German Diagnosis Related Groups Statistics from 2005 to 2014.
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Figure 2Comparison of 30 day mortality for men and women following (A) open (OAR) and (B) endovascular aneurysm repair (EVAR). M-H = Mantel-Haenszel; OR = odds ratio; CI = confidence interval.
Few papers reported risk adjusted ORs for 30 day/in hospital mortality, but cumulative analysis revealed consistent outcomes for OAR (OR 1.45 [1.21 – 1.72], p < .001, I2 = 19) and EVAR (OR 1.86 [1.43 – 2.40], p < .001, I2 = 16), with only minor heterogeneity (Fig. 3).
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
While women were slightly older at the time of operation (OAR mean difference (MD) 2.31 years [1.80 – 2.83], p < .001, I2 = 88; EVAR MD 2.57 years [2.09 – 3.05], p < .001, I2 = 91), sufficient data reporting enabled meta-regression for EVAR, which showed no statistically significant association between increased age difference and log mortality odds for women (n = 11, βi = 0.086 (se = 0.14), p = .53, tau2 = 0.02) (Table 2).
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Similar 5-year outcomes between female and male patients undergoing elective endovascular abdominal aortic aneurysm repair with the Ovation stent graft.
Women were also likely to be operated at a smaller aortic diameter than men; the difference was attenuated for EVAR and statistically significant heterogeneity was observed (OAR MD −4.93 mm [95% CI −6.71 – −3.14], p < .001, I2 = 95; EVAR MD −2.10 mm [95% CI −2.66 – −1.55], p < .001, I2 = 84)
Similar 5-year outcomes between female and male patients undergoing elective endovascular abdominal aortic aneurysm repair with the Ovation stent graft.
(Table 2). Assessment of the influence of other comorbidities on mortality was limited by heterogeneity, secondary to variation in definitions and reporting of comorbidities and patient selection, although cardiac disease appeared more common in men (OAR OR 0.72 [0.59 – 0.88]; EVAR OR 0.65 [0.48 – 0.87]) and respiratory disease more common in women (OAR OR 1.51 [1.23 – 1.86]; EVAR OR 1.26 [1.08 – 1.46]). Full data for these and other comorbidities are given in Supplementary Table S5.
Figure 3Comparison of risk adjusted odds ratios (OR) of 30 day mortality for men and women following (A) open (OAR) and (B) endovascular aneurysm repair (EVAR). CI = confidence interval; IV = inverse variance.
To investigate whether sex specific differences in mortality had changed over time, meta-regression, using study midpoint as a covariate, was conducted. While the mortality OR of women for both OAR and EVAR have increased slightly over the last 20 years, this finding was statistically non-significant (OAR: n = 15, βi = 0.087 (se = 0.019), p = .65, tau2 = 0; EVAR: n = 18, βi = 0.014 (se = 0.034), p = .69, tau2 = 0.05) (Fig. 4).
Effect modification of sex and age for the outcome relationship in abdominal aortic aneurysm treatment: German Diagnosis Related Groups Statistics from 2005 to 2014.
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Figure 4Meta-regression examining association between midpoint of study and log odds of 30 day/in hospital mortality for women vs. men following (A) open (OAR) and (B) endovascular aneurysm repair (EVAR). No statistically significant trend in log odds of female mortality over time was observed. Studies are labelled by first author and year of publication.
Operative time (minutes) was statistically significantly lower for women compared with men for OAR (MD −14.98 [−18.45 – −11.51], p < .001, I2 = 0), and not statistically significantly different for EVAR (MD 4.33 [−0.99 – 9.65], p = .11, I2 = 79) (Supplementary Table S4).
Statistically significant sex specific differences in post-operative outcomes are illustrated in Fig. 5 (and Supplement S4). No complications were more common in men than women. The difference for cardiac complications, defined as myocardial ischaemia (with inclusion of congestive cardiac failure in three studies), was not statistically significant for women undergoing OAR, but was statistically significantly higher for women after EVAR, (OR 1.19 [1.03 – 1.37], p < .001, I2 = 0).
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Respiratory complications (pneumonia or acute respiratory distress), were observed to be statistically significantly more common for women after both forms of repair (OAR OR 1.40 [1.28 – 1.53], p < .001, I2 = 0; EVAR OR 1.44 [1.17 – 1.77], p < .001, I2 = 14).
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Renal complications, defined as acute renal failure requiring intense medical treatment or dialysis (or for Trenner et al. a coded diagnosis of renal infarction), were not more common for women following OAR, but were statistically significantly more likely following EVAR (OR 1.46 [1.25 – 1.72], p < .001, I2 = 0).
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
An increase in the risk of neurological complication (acute cerebrovascular accident) was not observed for women following OAR, whereas an increase for EVAR was observed (OR 2.08 [1.01 – 4.29], p < .050, I2 = 54), but was not robust to sensitivity analysis.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Risk of bowel ischaemia (mesenteric thrombus or embolism or colonic ischaemia) was statistically significantly greater for women for both OAR and EVAR (OAR OR 1.52 [1.34 – 1.74] p < .001, I2 = 0; EVAR OR 1.99 [1.51 – 2.62], p < .001, I2 = 2).
Lower limb ischaemia (arterial occlusion and/or major amputation), was not more likely for women for OAR, but was statistically significantly more likely for women following EVAR (OR 2.13 [1.48 – 3.06], p < .001, I2 = 75), although substantial heterogeneity was observed.
An increased length of hospital stay was reported for women for both OAR and EVAR, although substantial heterogeneity was observed (OAR MD 1.19 [0.65 – 1.73], p < .001, I2 = 94; EVAR MD 1.14 [0.85 – 1.44], p < .001, I2 = 95, days).
Figure 5Comparison of 30 day complications for men and women following (A) open (OAR) and (B) endovascular aneurysm repair (EVAR). M-H = Mantel-Haenszel; CI = confidence interval; OR = odds ratio.
Meta-regression to assess the influence of sex specific differences in post-operative complications on between study heterogeneity for mortality outcomes was limited by paucity of data. For OAR, insufficient data reporting prevented any evaluations. For EVAR, eight studies reported cardiac outcomes, for which a strong association between increased cardiac complications and increased female mortality was observed (βi = 2.96 (se = 1.27), p = .020, tau2 = 0.00), eight studies also reported renal complications, for which the positive association with mortality was almost statistically significant (βi = 2.50 (se = 1.31), p = .056, tau2 = 0.01) (Supplementary Fig. S7).
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Insufficient data were available for multivariable regression or assessment of other covariables.
Endograft complications
Although type 1 endoleak (OR 1.75 [1.56 – 1.96], p < .001, I2 = 0) and any endoleak (OR 1.11 [1.02 – 1.22], p < .020, I2 = 14) were more likely to occur following EVAR for women, no statistically significant difference in the likelihood of re-intervention was observed (Supplementary Table S4).
Similar 5-year outcomes between female and male patients undergoing elective endovascular abdominal aortic aneurysm repair with the Ovation stent graft.
Egger’s test suggested the presence of funnel plot asymmetry for publication bias for OAR, but cumulative findings were robust to trim and fill analysis (adjusted OR 1.45 [95% CI 1.36 – 1.55], p < .001) (Supplementary Figs and Text S8). No significant suggestion of publication bias was observed for EVAR (p = .899). Sensitivity analyses for all analyses of mortality and post-operative outcomes were robust, with the exception of the following: (1) incidence of cardiac complications following OAR were heavily weighted by data from Germany and on removal of these data a small but statistically significant increase in cardiac complications was observed for women (OR 1.27 (95% CI 1.04 – 1.55), p = .020, I2 = 32); (2) the increase in neurological complications following EVAR for women was not robust to sensitivity analysis; (3) incidence of any endoleak was heavily weighted by data from the USA and not robust on sensitivity analysis. The sex specific differences in 30 day mortality persisted in both studies reporting infrarenal AAAs alone and studies including juxta-, para-, or suprarenal AAA, although in the latter group of studies ORs for the sex specific disparity may have increased slightly for EVAR (Supplementary Table S4). Despite significant heterogeneity, sensitivity analysis was robust for the pre-operative variables of sex specific differences in age and AAA diameter.
Discussion
This systematic review demonstrated a significantly higher 30 day mortality risk for women following OAR, with an increased risk differential for EVAR (OR 1.49 vs. OR 1.86, respectively), which meta-regression showed has not been ameliorated by modern practice. The study also demonstrated that sex specific differences varied with type of repair, allowing insight into the increased mortality risk differential for EVAR. For both OAR and EVAR, women were at increased risk of respiratory complications, bowel ischaemia, and need for blood transfusion: no post-operative complication was more common in men and procedure times were similar for both sexes. Following EVAR, women were also at increased risk of cardiac and renal complications as well as limb ischaemia and arterial injury. These findings provide a shocking presentation of gender disparity in AAA repair outcomes and raise questions regarding procedure specific causes for the inequalities observed.
It is tempting to attribute the increased risks for women undergoing EVAR to technical difficulties secondary to challenging anatomy (e.g., more tortuous access vessels and shorter angulated aneurysmal necks). However, as the outcomes in this paper are for the subset of women deemed anatomically suitable for EVAR, and operative times were comparable, it is unlikely that anatomical differences alone can sufficiently explain the disparities in outcomes observed.
Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.
Moreover, the development of low profile delivery devices to facilitate challenging access, and innovative solutions for hostile neck anatomy have greatly reduced many of the traditional obstacles to endovascular repair.
Results of the ANCHOR prospective, multicenter registry of EndoAnchors for type Ia endoleaks and endograft migration in patients with challenging anatomy.
Abdominal aortic aneurysm treatment in women. Excellent long-term results for endovascular aneurysm repair morphologic changes in renal and superior mesenteric arteries after fenestrated endovascular aortic repair.
This may in part be the result of differences in systemic perfusion or blood loss (noted by Ultee et al. as a risk factor for bowel ischaemia [OR 2.0]), in operative strategies (e.g., proximal extension or clamp location, hypogastric artery exclusion), or caused by increased embolisation.
Emboli are an important factor in the pathogenesis of post-operative bowel (and renal) ischaemia, and as EVAR is associated with twice the number of emboli as OAR, increased embolisation in women could contribute to the greater risk differential observed for EVAR.
Acute kidney injury is also associated with a two fold increase in mortality and cardiovascular morbidity, and may contribute to the increased mortality risk for women following EVAR.
Sex specific differences in kidney injury may also result from embolism, anatomical differences (e.g., closer proximity of the renal arteries, orificial disease, angulation), selection of suboptimal endovascular strategies (e.g., renal artery coverage), or iatrogenic injury (e.g., contrast nephropathy or intimal damage).
Renal protective strategies such as pre-hydration, avoidance of renotoxic medication, judicious use of contrast, and a careful operative approach with renal preservation, can greatly reduce the risk of renal injury and should be strictly implemented for women.
EVAR access complications are associated with four times the odds of peri-operative death, while conversion to open, rather than planned cutdown, is associated with a higher risk of major complication.
In this study double the risk of limb ischaemia (and transfusion) for women undergoing EVAR, combined with a three fold increased risk of arterial injury and without evidence of increased operation time (indicative of increased complexity) was observed. Therefore it is certainly possible that limb access complications, with an iatrogenic component, are likely to have contributed to the increased mortality risk for women, and present a target for quality improvement through emphasis on strategic planning.
Differential sex specific mortality risk remained following adjustment for confounding risk factors, and analysis of pre-operative comorbid status, although very limited by heterogeneity, did not demonstrate clear evidence of increased vulnerability among women undergoing AAA repair overall. However, the increased risk of respiratory complications and longer length of stay may suggest a role for pre-operative respiratory physiotherapy and exercise.
In addition, while men were more often reported to have pre-operative cardiac disease, following EVAR, women were more likely to suffer cardiac complications. This may reflect failure to recognise and investigate comorbidity before EVAR, secondary to reduced index of suspicion and unconscious or systemic bias, which limits the ability to optimise the patient’s health status prior to repair and may increase the risk of adverse events.
Sex differences in treatments, relative survival, and excess mortality following acute myocardial infarction: national cohort study using the SWEDEHEART registry.
Better standardisation and reporting of comorbidity assessment is needed to allow future studies to robustly analyse how this affects the differential in sex specific mortality risk.
This meta-analysis was limited because of heterogeneity in patient selection, threshold for treatment, publication bias, and a lack of systematic reporting of comorbidity and complications, which prevented further meta-regression. Although cumulative analysis of retrospective data cannot offer the same level of detail or control for confounding factors as a well designed prospective analysis, the present findings of increased mortality and complications for women were robust to sensitivity analysis and warrant further detailed investigation. While women were more likely to have complications and more likely to die in hospital, it was not possible to directly assess “failure to rescue” (FTR, death directly attributed to a significant post-surgical complication) on cumulative analysis, as this was not reported as a specific outcome within included studies. Female sex has been reported previously as an independent risk factor for FTR among patients aged ≥ 80 years undergoing AAA repair (OR 1.95).
Therefore, FTR could be considered a useful outcome for future studies evaluating sex specific disparity. A further limitation of this study was that to reduce heterogeneity, analysis was limited to intact AAA repair, and therefore does not include analysis of sex specific differences in patients with an index presentation of AAA rupture.
Women have been overlooked in the field of vascular surgery, and under represented in key randomised trials, which form the foundation of evidence based care.
Therefore, it remains possible that advances in modern practice, with best practice guidance and endovascular therapy primarily based on evidence for AAA repair in men, are not suitable for women.
For example, the introduction of screening in men only, leading to timely identification and treatment, may have contributed to widening of the risk differential.
Greater understanding of the differences in the pathophysiology of AAAs in women, pertinent risk factors, and differences in provision of care could enable a tailored pathway and narrowing of this outcome disparity.
This study demonstrates that the risk of mortality following AAA repair is consistently higher for women. This may be secondary to sex specific differences in post-operative complications, which differ for OAR and EVAR, or unequal pre-operative recognition and optimisation of risk factors. Such glaring disparity cannot and should not remain unchallenged. This study indicates key areas to target including reduction in arterial injury and emboli and improved renal protection. Just as changes to pre-operative optimisation and dedicated multidisciplinary care have radically improved overall outcomes for all patients with AAA repair; an intelligent approach to quality improvement combined with proportionate inclusion of women in prospective research, has the potential to dramatically reduce the gender gap following AAA repair.
Conflict of interest
None.
Funding
None.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Effect modification of sex and age for the outcome relationship in abdominal aortic aneurysm treatment: German Diagnosis Related Groups Statistics from 2005 to 2014.
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.
Sex-related trends in mortality after elective abdominal aortic aneurysm surgery between 2002 and 2013 at National Health Service hospitals in England: less benefit for women compared with men.
Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years.
Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.
Midterm pivotal trial results of the Talent Low Profile System for repair of abdominal aortic aneurysm: analysis of complicated versus uncomplicated aortic necks.
Chapter 10: Analysing data and undertaking meta-analyses.
in: Higgins J.P.T. Thomas J. Chandler J. Cumpston M. Li T. Page M.J. Welch V.A. Cochrane Handbook for Systematic Reviews of Interventions version 6.1. 2020(updated September 2020). Cochrane. Available at:
Wells G, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for Assessing The Quality Of Nonrandomised Studies In Meta-Analyses. Ottawa Hospital Research Institute. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp [Accessed 3 Feb 2021].
Similar 5-year outcomes between female and male patients undergoing elective endovascular abdominal aortic aneurysm repair with the Ovation stent graft.
Results of the ANCHOR prospective, multicenter registry of EndoAnchors for type Ia endoleaks and endograft migration in patients with challenging anatomy.
Abdominal aortic aneurysm treatment in women. Excellent long-term results for endovascular aneurysm repair morphologic changes in renal and superior mesenteric arteries after fenestrated endovascular aortic repair.
Sex differences in treatments, relative survival, and excess mortality following acute myocardial infarction: national cohort study using the SWEDEHEART registry.
00445-7&id=gr1.jpg){kind=link}
00445-7&id=gr2.jpg){kind=link}
00445-7&id=gr3.jpg){kind=link}
00445-7&id=gr4.jpg){kind=link}
00445-7&id=gr5.jpg){kind=link}
