Does the diameter of abdominal aortic aneurysm influence late survival following abdominal aortic aneurysm repair? A systematic review and meta-analysis

Abstract

Background

Studies reporting the influence of preoperative abdominal aortic aneurysm diameter on late survival following abdominal aortic aneurysm repair have not been consistent.

Aim: To report the influence of abdominal aortic aneurysm diameter on overall long-term survival following abdominal aortic aneurysm repair.

Methods

Embase, Medline and the Cochrane electronic databases were searched to identify articles reporting the influence of abdominal aortic aneurysm diameter on late survival following open aneurysm repair and endovascular aneurysm repair published up to April 2015. Data were extracted from multivariate analysis; estimated risks were expressed as hazard ratio.

Results

A total of 2167 titles/abstracts were retrieved, of which 76 studies were fully assessed; 19 studies reporting on 22,104 patients were included. Preoperative larger abdominal aortic aneurysm size was associated with a worse survival compared to smaller aneurysms with a pooled hazard ratio of 1.14 (95% CI: 1.09–1.18), per 1 cm increase in abdominal aortic aneurysm diameter. Subgroup analysis of the different types of repair was performed and the hazard ratio (95% CI), for open aneurysm repair and endovascular aneurysm repair were 1.08 (1.03–1.12) and 1.20 (1.15–1.25), respectively, per 1 cm increase. There was a significant difference between the groups p < 0.02.

Conclusions

This meta-analysis suggests that preoperative large abdominal aortic aneurysm independently influences overall late survival following abdominal aortic aneurysm repair, and this association was greater in abdominal aortic aneurysm repaired with endovascular aneurysm repair.

Keywords

Abdominal aortic aneurysm endovascular aneurysm repair systematic review survival factors hazard rates size

Introduction

Abdominal aortic aneurysm (AAA) rupture is associated with high surgical mortality worldwide,¹ yet AAA can be effectively managed by either open aneurysm repair (OAR) or endovascular aneurysm repair (EVAR) in an elective setting with a 10-fold decrease in perioperative mortality. Regardless of the repair method chosen, results from four randomized trials and two large cohort studies indicate that survival is similar between EVAR and OAR.^2,3

Previous research has identified several independent prognostic variables associated with lower survival following AAA repair with varying size effects. These include demographic such as age, gender and clinical comorbidities including cardiac, renal or pulmonary impairment.⁴ Furthermore, population studies have also shown that an increase in infra-renal abdominal aortic size above the normal aortic diameter or the presence of an AAA (defined as greater than or equal to 3 cm) are independent predictors of increased cardiovascular and overall late mortality.^5,6

Koskas and Kfeifer⁷ were the first to report that the diameter of the AAA negatively predicted late survival following OAR in a multicenter prospective trial consisting of 794 patients. The hazard ratio (HR), 95% confidence interval (CI) was 1.1 (95% CI: 1.04–1.08) for each 1 cm increase in AAA diameter.⁷ Following the emerging use of EVAR in the early 2000s, endovascular registries from Europe (EUROSTAR) and the United States (Lifeline registry) both reported that larger AAA were independently associated with worse late survival.^8,9 However, other studies have observed no difference in late survival between small and large AAA diameter.^10,11

Since late survival following AAA repair is determined by preoperative factors such as AAA diameter, quantifying the individual risk is necessary in clinical decision-making. This systematic review and meta-analysis aimed to report the influence of preoperative AAA diameter on late survival of patients undergoing elective AAA repair.

Methods

The systematic review was conducted according to the preferred reporting items for systematic reviews and meta-analysis (PRISMA).¹² Individual patient consent was not possible, as this review will analyze published literature only. The Health and Disability Ethics committee has approved this project. This topic was defined in the PICOT¹³ format as: the population are patients undergoing elective AAA repair (via either OAR or EVAR); intervention and comparison: preoperative AAA diameter, Outcome: all cause mortality and time: greater than or equal to one year.

Search strategy

EMBASE, Medline and the Cochrane Library Database were searched via the electronic OVID SP database. With the assistance of a clinical librarian, “exploded” medical subject headings (MeSH) terms for MEDLINE and Cochrane, and EMTREE terms for EMBASE were used to broaden the key word search: “abdominal aortic aneurysm,” “size,” “blood vessel diameter,” “long term survival” and “survival rate” along with their synonyms. Two independent researchers (MK and JR) conducted the search and when disagreement arose the reviewers met to resolve any issues.

There was no date restriction and no limitations on publication language or study type applied to the search. The last search was completed in April/May 2015. An additional manual search of articles was conducted using references from relevant articles and review papers. The journals of Annals of Vascular Surgery, European Journal of Endovascular and Vascular Surgery, Journal of Endovascular Therapy, Journal of Vascular Surgery and Vascular were searched for any relevant articles published “online first.” Abstracts of conference proceedings were searched for full text publication. Eligible titles or abstracts where imported into Endnote X7 (Thomson Reuters) library and full text articles were obtained.

Study selection

Inclusion and exclusion selection criteria

Two independent reviewers adhered to the following inclusion criteria: any studies with greater than a 100 patients, reporting survival data and information about preoperative AAA diameter following elective AAA repair (OAR or EVAR) with at least one year follow-up with the primary endpoint of outcome being all-cause mortality. There were no restrictions to operative repair methods and complex repairs (suprarenal clamping/visceral debranching or fenestrations). The exclusion criteria were studies that were solely limited to small AAA (<5 cm), rupture AAA, first generation/early custom-made stent grafts, high-risk patients and octogenarians.

Study selection and data extraction

Two researchers independently performed the data extraction. Studies that met the inclusion criteria were entered into a Microsoft Excel spreadsheet. When studies from large registries, or known databases were included, only the most recent study that contained the largest number of patients or the study that kept AAA diameter as a continuous variable was used in the analysis. Study authors were contacted when clarification was required.

The quality of the observational studies (cohort) was assessed using the Newcastle-Ottawa Scale (NOS).¹⁴ This tool employs a 9-point system that assesses three domains: patient selection, comparability of the study groups and the ascertainment of study outcome. Studies with a score of 9 indicate a lower risk of bias whereas scores of 7–8 indicate medium risk of bias and a score of ≤6 indicates a higher chance of bias.

Statistical analysis

Reported baseline risk factors were extracted from baseline tables, group means and standard deviations (SD) were weighted and combined for descriptive purposes. A meta-analysis of time-to-event data was undertaken. Reported HR (statistically significant and non-significant) from multivariate Cox proportional models were extracted from individual studies. Pooled estimates together with 95% CI were calculated using a random effects model, chosen due to expected heterogeneity among the studies. Heterogeneity was expressed with the I² statistic with greater than 25%, 50% and 75% defined as low, moderate and high degrees of heterogeneity. Statistical significance was set at a p-value 0.05. Sub group analyses were performed according to a priori groupings related to type of repair (EVAR vs OAR). The meta-analysis was performed using Review Manager (RevMan) [Computer program] Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012. Meta-regression was performed in R using the package metafor, with heterogeneity estimated using the DerSimonian–Laird method with inverse variance weights.^15,16

Results

The literature search flow chart results are summarized in Figure 1. Of 2167 titles, 72 were thought to be relevant after screening titles and abstracts. Following duplication removal, 66 articles were assessed of which 10 met the inclusion criteria.^17–26 The hand searching identified seven additional studies^7,11,27–30 and searching vascular journals for ahead of print/“Online first” articles identified two additional studies.^31,32 Nineteen articles were included in the systematic review of which three did not include sufficient data to allow inclusion into the meta-analysis. One study provided descriptive results only²⁷ and two studies included univariate (unadjusted) risk ratios.^25,26

Figure 1.

PRISMA Diagram showing flowchart of study selection.

Nineteen studies published between 1989 and 2015 were included in this review. Ten studies were from the United States, eight were from Europe and one was from Australia. Repairs of AAA were done with EVAR in 10 studies and OAR in 5 studies. Three studies included both OAR and EVAR and one study did not state the method of repair. “Raw” data from one study was obtained and analyzed according to type of repair.³¹ The average (SD) NOS was 7.7 (1.4) indicating a high overall quality of observational studies (Table 1).

Table 1.

Study characteristics of included studies.

First author	Year	Country	Area	NOS	N at start	Repair type	Duration of study	Average follow-up (months)	Adjustors
Beck et al.³³	2009	USA	VSGNNE	7	639	EVAR	2003–2007	12	Age, COPD, supra-renal clamp and renal impairment
Brady et al.²⁰	2001	UK	Multicenter	9	1139	NR	1991–1995	43.2	Age, sex, PVD, angina, IHD smoking status, cholesterol, diabetes, hypertension, BMI, WBC, creatinine and COPD
Brewster et al.²⁷	2006	USA	Massachusetts	6	873	EVAR	1995–2005	27	NR
Carlisle and Swart²⁵	2007	UK	Torquay	5	130	OAR	1999–2006	35^a	NR
De Bruin et al.¹⁷	2014	Netherlands/ Belgium	DREAM (RCT post hoc)	9	351	EVAR and OAR	Seven years	76.8	Age, sex, reintervention, tobacco, statin, antiplatelet, hypertension, cardiac, pulmonary, renal and carotid disease
Diehm et al.²²	2007	USA/Switzerland	NR	7	711	EVAR	1994–2006	48.3	Age, hemoglobin, diabetes, statin use, renal and pumonary scores,
Grootenboer et al.¹⁸	2013	Europe	Eurostar Registry Multicentre	8	9227	EVAR	1996–2006	12.6^a	Age, sex, ASA scale, cardiovascular comorbidities, AAA and treatment characteristics
Hertzer and Mascha¹⁹	2005	USA	Cleveland Clinic	9	855	OAR	1976–2003	NR	Age, preliminary coronary revascularization, COPD, renal, impairment and graft configuration
Huang et al.³²	2015	USA	Mayo Clinic	9	1116	EVAR and OAR	2000–2011	91.2^a	Age, sex, type of repair, history of cancer and surgical risk score
Kabbani et al.²¹	2014	USA	Detroit	9	245	OAR	1986–2013	54^a	Age, smoking, IHD, COPD, renal, hyperlipidemia, renal impairment, diabetes, history of smoking and stroke
Keith et al.²⁶	2013	USA	Alabama	8	740	EVAR	2000–2011	40	NR
Khashram et al.³¹	2015	Australia	Brisbane	9	1340	EVAR and OAR	1990–2013	78 OAR and 48 EVAR	Age, sex, ASA scale, type of repair. COPD, PVD, renal impairment and HF
Koskas and Kieffer⁷	1997	France	Multicentre	9	794	OAR	1989	60	Age, thrombosis of IVC, iliac involvement, carotid artery occlusion, IHD, renal impairment, neck aneurysm and incision method
Lomazzi et al.³⁰	2011	Italy	Varese	7	235	EVAR	2000–2008	26.3	Age, ASA scale and sex
Mastracci et al.²⁴	2010	USA	Cleveland Clinic	8	412	EVAR	1998–2005	48	Age, HF, COPD, statin and aspirin use,
Roger et al.²⁹	1989	USA	Olmstead, Rochester	9	131	OAR	1971–1987	NR	Age, sex, diabetes, uncorrected IHD, hypertension, creatinine, history of stroke, cancer and smoking
Saratzis et al.¹¹	2013	Greece	Thessaloniki	7	383	EVAR	2008–2011	34	Age, WBC, hemoglobin, creatinine, cancer history, C reactive protein and endoleak type 1
Tsilimparis et al.²³	2012	Germany	Berlin	6	119	EVAR	Four years	34	Age, current smoking, IHD and renal function
Zairns²⁸	2005	USA	IDE trial (LIFELINE Registry)	7	2664	EVAR	Five year	34	Age, sex, IHD, HF, hypertension, COPD, diabetes and renal failure

RCT: randomized controlled trial; VSGNNE: Vascular Study Group of Northern New England; IDE: investigative device exemption; NOS: Newcastle-Ottawa Scale; IHD: ischemic heart disease; COPD: chronic obstructive pulmonary disease; PVD: peripheral vascular disease; ASA: American society of anesthesiology; BMI: body mass index; WBC: white blood count; HF: heart failure.

Median follow-up.

A total of 22,104 patients were included in the systematic review. Baseline demographics, clinical risk factors and AAA diameter descriptions are presented in Table 2; 13 studies reported AAA diameter as a continuous variable (per cm, mm or 5 mm increments), 1 study transformed aortic sizes into a logarithmic scale²⁰ and 4 studies categorized AAA diameter;^17,23,26,33 15 studies that comprise 20,205 patients, provided baseline mean/median aneurysm diameter measurements (range 5.1–6.4 cm).

Table 2.

Patient and AAA baseline characteristics.

Author	Age (SD)	Males (%)	No of groups	AAA diameter in mm (SD)	Method of measurement	Units/ presentation	IHD^a	Smoking history^b	Renal impairment	Diabetes	COPD
Beck et al.³³	74 (NR)	82	1	57.0 (10)	NR	≥65 mm	31.0	90.1	7.0	18.0	39.0
Brady et al.²⁰	NR	NR	1	51.0 (32–92)	US	per logarithmic unit (0.8 cm)	NR	NR	NR	NR	NR
Brewster et al.²⁷	75.7 (7.6)	81.4	1	56.8 (10.6)	CT	NR	57.2	65.1	7.0	12.4	24.1
Carlisle and Swart²⁵	NR	NR	1	NR	NR	per 10 mm	NR	NR	NR	NR	NR
De Bruin et al.¹⁷	70.2 (6.7)	91.7	1	60.4 (8.7)	CT	≥70 mm	44.2	59.0	8.3	10.0	23.1
Diehm et al.²²	75.8 (7.5)	90.9	1	58.2 (10.6)	NR	per 1 mm	45.0	77.6	21.8	15.0	32.1
Grootenboer et al.¹⁸	72.2 (7.8)	93.2	1	58.1 (12.2)	CT	per 1 mm	60.4	48.7	19.1	13.0	41.9
Hertzer and Mascha¹⁹	69 (47–94)	86	1	59.0	CT and US	per 10 mm	NR	NR	NR	NR	6.2
Huang et al.³²	73 (7.6)	86	1	58.0 (14.9)	CT, MRI or US	per 1 cm	NR	NR	NR	NR	NR
Kabbani et al.²¹	71 (7.9)	68.8	1	64 (19)	NR	per 1 cm	42.0	85.0	N/A	14.0	36.0
Keith et al.²⁶	71.4	84.6	3	NR	CT	<5, 5–5.9 and >6 cm	52.5	73.5	10.8	16.2	27.4
Khashram et al.³¹	72.4 (7.4)	81.7	1	59.4 (11.9)	CT	per 1 cm	47.3	90.4	8.1	13.0	27.5
Koskas and Kieffer⁷	67.8 (8.4)	92.9	1	NR	NR	per 1 cm	42.8	NR	10.8	9.2	21.2
Lomazzi et al.³⁰	71.9 (8)	92.7	1	NR	CT	per 1 mm	48.0	NR	11.0	NR	53.2
Mastracci et al.²⁴	75.4 (7.5)	88	1	64.0 (11.6)	CT	per 5 mm	58.0	19^b	28.0	16.0	38.0
Roger et al.²⁹	NR	79	1	60.0 (NR)	NR	per 1 cm	42.7	44.0^b	12.0	8.0	NR
Saratzis et al.¹¹	68.9 (8.2)	91.7	1	61.8 (14)	CT	per 1 cm	9.7	72.6	N/A	18.8	NR
Tsilimparis et al.²³	71.3 (8.2)	91.6	2	58 (34–93)	CT	≥60 mm	36.1	47.0	14.3	14.3	20.0
Zairns²⁸	73.1 (7.8)	88.6	1	55.8 (10.2)	CT	per 1 mm	82.7	NR	3.4	12.4	29.2
Range	68.9–75.8	68.8–93.2		51–64 mm			9.7–82.7	19.0–90.4	7.0–28.0	8.0–18.8	6.2–53.2

SD: standard deviation; NR: not reported; IHD: ischemic heart disease; COPD: chronic obstructive pulmonary disease.

Various definitions used.

Current smokers.

The full meta-analysis included 16 studies and 19,722 patients. All of the studies adjusted for age while several also adjusted for comorbid conditions (Table 1). Larger AAA diameter measured prior to AAA repair was associated with lower reported survival compared with smaller aneurysms. A 1-cm increase in AAA diameter was associated with a pooled HR of 1.14 (95% CI: 1.10–1.18), I² = 48% (Figure 2). Excluding four studies with either categorical^17,23,33 or logarithmic²⁰ AAA diameter conversions did not influence the overall risk – HR 1.13 (95% CI: 1.09–1.18), I² = 50% for each increase in 1 cm of AAA diameter.

Figure 2.

Meta-analysis of the effect of AAA size of late mortality following elective AAA repair (per 1 cm increase).

Thirteen studies were included in a subgroup analysis according to AAA repair type that included an equal weight (50%) for each category (Figure 3). EVAR was associated with a significantly higher mortality risk compared with OAR for each 1 cm increase in AAA diameter, pooled HR was EVAR 1.20 (95% CI: 1.15–1.25), I² = 0% and OAR 1.08 (95% CI: 1.03–1.12), I² = 12%, respectively (Figure 3). This subgroup analysis excluded two studies that included both EVAR and OAR in the same analysis,^17,32 one study that categorized AAA diameter²³ and one that did not report how the AAA was repaired.²⁰

Figure 3.

Subgroup analysis of AAA size according to type of repair (per 1 cm increase, EVAR: endovascular aneurysm repair, OAR: open aneurysm repair).

Meta-regression was undertaken to determine if the between-study heterogeneity could be accounted for by the mid-year of study or duration of follow-up; 16 studies contributed to mid-year of the study and 15 studies contributed to the duration of follow-up. There was an association for a decrease in log(HR) by duration of follow up (β = 0.998, 95% CI: 0.996–1.000, I² = 24%, p < 0.013; Figure 4). There was no evidence of a change in log(HR) by mid-year study (β = 1.004, 95% CI: 0.999–1.010, I² = 47%, p = 0.13).

Figure 4.

Log(HR) of AAA size vs study follow-up. The solid line represents the regression line and the shaded gray area corresponds to the 95% CI. Each circle represents a different study and the size of the circle is proportional to the weight of study (inverse variance = 1/standard error² of the HR).

Discussion

Individualizing patient treatment to their specific demographics and comorbidities is warranted to improve long-term outcomes. This systematic review found that larger AAAs were independently associated with a lower survival following elective AAA repair and this observation was greater for EVAR than OAR.

AAA size has also been shown to be an important independent predictor of perioperative morbidity and mortality. Schouten et al.³⁴ reported on 500 patients undergoing elective open AAA repair and larger AAA diameter had a significantly higher risk of cardiovascular complications and mortality. A recent report from the Vascunet database included 5895 patients from six countries and concluded that the size of AAA was associated with higher perioperative mortality with an adjusted odds ratio of perioperative mortality for a 1-cm increase in AAA size in patients undergoing either OAR or EVAR of 1.14 (95% CI: 1.03–1.27) and 1.28 (95% CI: 1.06–1.55), respectively.³⁵ The diameter of AAA at the time of repair has also been associated with a decreased five-year survival as demonstrated by a recent meta-regression analysis including 13,281 patients treated between 1978 and 2011.³⁶

Why are larger AAAs associated with worse survival?

Based on the results from this meta-analysis, there appears to be two factors that could explain why larger AAAs may have a worse survival. First, this association was found in both types of repairs; therefore, a biological cause seems plausible. Larger AAA might exhibit more inflammatory mediators or larger size AAA might be associated with more advanced cardiovascular disease.^8,23,34 Five studies provided group comparison between small and large AAA. The results from three large studies^8,9,26 suggest that patients with larger AAA were older and had a greater burden of cardiovascular disease than patients with small AAA. In the two other smaller studies,^10,23 there was no difference in morbidities between the groups. However, patient co-morbidities were adjusted for in the survival models and the influence of AAA size was an independent predictor of late survival.

Second, the effect estimate of AAA size of EVAR treatment was significantly higher in this analysis compared to the OAR group.

Nevertheless, this does not adequately explain why the association was greater in EVAR than OAR. Results from the Lifeline and EUROSTAR registries have also shown that an increase in AAA diameter was independently associated with a higher AAA related mortality, higher rupture post repair, re-intervention and surgical conversion to OAR.^8,9,28 One might speculate that each re-intervention might have an additive mortality risk.

Roger et al.²⁹ were the first to include AAA size in a multivariate model, but AAA diameter was not a significant mortality predictor in their study. Almost a decade later, Koskas and Kfieifer⁷ were the first to show that preoperative AAA size was an independent predictor of poor late survival. Interestingly, this finding appeared to generate little discussion, including within the reporting paper. It was not until subsequent EVAR data began to emerge, highlighting morphological aortic neck and iliac artery differences between small and large AAA that interest in this area has increased.^8,9

Limitations

This systematic review has a number of potential limitations related to study selection and potential biases in reporting and publication. Specifically, as AAA diameter (>5.5 cm) has not been subjected to any randomized trials, it is possible that higher risk patients were selected to undergo EVAR and hence were more likely to experience higher late mortality. However, it was notable that three studies^7,19,29 in the OAR group recruited patients prior to the EVAR era, and AAA diameter remained a negative predictor of late survival.

Furthermore, potential confounders that were not adjusted for in the analysis might have contributed to the findings of this review. For example, other anatomical features of AAAs such as short proximal necks or juxta/supra-renal aneurysms have been associated with worse outcomes and lower survival following AAA repair. Some of the earlier studies might have included first generation endovascular stent grafts that were associated with higher mortality and endoleak rates and might have influenced the results.³⁷ We excluded studies that only included such grafts to minimize the selection bias.

The outcome of interest in this analysis was overall patient survival during the reported follow-up period. This period ranged from 12 to 91.2 months and the metaregression at univariable level suggests an association of a decreased risk with larger AAA during follow-up. In addition, the long-term success of the index EVAR procedure and variations in endoleak rates and reporting is not accounted for. Future studies comparing endoleak rates, AAA diameter and the effect on aneurysm related mortality is warranted to better understand this relationship.

In addition, this review contains a mixture of study designs: single surgeon series, tertiary single center studies, national registries, treatment specific registries and post hoc randomized trials. Therefore, heterogeneity is expected in such an analysis. We were able to reduce heterogeneity by grouping studies into type of repair (OAR and EVAR).

As more AAA are repaired with endovascular therapy and open repair is reserved for a selected population,³⁸ estimating the influence of AAA diameter in relation to repair method will be susceptible to selection biases. We recommend that future studies reporting on late predictors of survival following AAA repair should include AAA diameter within the predictive modeling. In addition, AAA measurements should be kept as continuous data.³⁹ When applicable, separating the analysis according to type of repair might shed further light on the influence of repair type in relation to late survival.

Conclusion

In conclusion, best estimate from the published literature appears to indicate that preoperative AAA diameter influences late survival following elective AAA repair. Larger AAA is associated with poorer survival and this association is greater for EVAR than OAR.

The inclusion of AAA diameter in the clinical decision-making process, therefore, seems warranted when considering the most appropriate surgical management option for individualizing patient care.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Stewart

Khanduri

McCord

. Global disease burden of conditions requiring emergency surgery. Brit J Surg 2014; 101: 9–22.

Paravastu

Jayarajasingam

Cottam

. et al. Endovascular repair of abdominal aortic aneurysm. Cochrane Database Sys Rev 2014; 1: Cd004178.

Stather

Sidloff

Dattani

. Systematic review and meta-analysis of the early and late outcomes of open and endovascular repair of abdominal aortic aneurysm. Brit J Surg 2013; 100: 863–872.

Khashram

Williman

Hider

. Systematic review and meta-analysis of factors influencing survival following abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2016; 51: 203–215.

Newman

Arnold

Burke

. Cardiovascular disease and mortality in older adults with small abdominal aortic aneurysms detected by ultrasonography: the cardiovascular health study. Ann Intern Med 2001; 134: 182–190.

Norman

Pearce

. Infrarenal aortic diameter predicts all-cause mortality. Arterioscler Thromb Vasc Biol 2004; 24: 1278–1282.

Koskas

Kieffer

. Long-term survival after elective repair of infrarenal abdominal aortic aneurysm: results of a prospective multicentric study. Ann Vasc Surg 1997; 11: 473–481.

Peppelenbosch

Buth

Harris

. Diameter of abdominal aortic aneurysm and outcome of endovascular aneurysm repair: does size matter? A report from EUROSTAR. J Vasc Surg 2004; 39: 288–297.

Zarins

Crabtree

Bloch

. Endovascular aneurysm repair at 5 years: does aneurysm diameter predict outcome? J Vasc Surg 2006; 44: 920–930.

10.

Wang

Carpenter

. EVAR in small versus large aneurysms: does size influence outcome? Vasc Endovasc Surg 2009; 43: 244–251.

11.

Saratzis

Sarafidis

Melas

. Impaired renal function is associated with mortality and morbidity after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2013; 58: 879–885.

12.

Moher

Liberati

Tetzlaff

. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009; 339: 2535.

13.

Rios

Thabane

. Association between framing of the research question using the PICOT format and reporting quality of randomized controlled trials. BMC Med Res Methodol 2010; 10: 11.

14.

Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp (2008, accessed 1 February 2015).

15.

R Development Core Team. R: a language and environment for statistical computing, Veinna: R Foundation for Statistical Computing, 2015.

16.

Viechtbauer

. Conducting meta-analyses in R with the metafor package. J Stat Soft 2010; 36: 1–48.

17.

De Bruin

Baas

Heymans

. Statin therapy is associated with improved survival after endovascular and open aneurysm repair. J Vasc Surg 2014; 59: 39–44.

18.

Grootenboer

Hunink

Hendriks

. Sex differences in 30-day and 5-year outcomes after endovascular repair of abdominal aortic aneurysms in the EUROSTAR study. J Vasc Surg 2013; 58: 42–49.

19.

Hertzer

Mascha

. A personal experience with factors influencing survival after elective open repair of infrarenal aortic aneurysms. J Vasc Surg 2005; 42: 898.1–898.10.

20.

Brady

Fowkes

Thompson

. Aortic aneurysm diameter and risk of cardiovascular mortality. Arterioscler Thromb Vasc Biol 2001; 21: 1203–1207.

21.

Kabbani

West

Viau

. Survival after repair of pararenal and paravisceral abdominal aortic aneurysms. J Vasc Surg 2014; 59: 1488–1494.

22.

Diehm

Benenati

Becker

. Anemia is associated with abdominal aortic aneurysm (AAA) size and decreased long-term survival after endovascular AAA repair. J Vasc Surg 2007; 46: 676–681.

23.

Tsilimparis

Mitakidou

Hanack

. Effect of preoperative aneurysm diameter on long-term survival after endovascular aortic aneurysm repair. Vasc Endovas Surg 2012; 46: 530–535.

24.

Mastracci

Greenberg

Hernandez

. Defining high risk in endovascular aneurysm repair. J Vasc Surg 2010; 51: 1088–1095.

25.

Carlisle

Swart

. Mid-term survival after abdominal aortic aneurysm surgery predicted by cardiopulmonary exercise testing. Brit J Surg 2007; 94: 966–969.

26.

Keith

Jr Passman

Gaffud

. Comparison of outcomes following endovascular repair of abdominal aortic aneurysms based on size threshold. J Vasc Surg 2013; 58: 1458–1466.

27.

Brewster

Jones

Chung

. Long-term outcomes after endovascular abdominal aortic aneurysm repair: the First Decade. Ann Surg 2006; 244: 426–436.

28.

Zairns

. Lifeline registry of endovascular aneurysm repair: long-term primary outcome measures. J Vasc Surg 2005; 42: 1–10.

29.

Roger

Ballard

Hallett

Jr . Influence of coronary artery disease on morbidity and mortality after abdominal aortic aneurysmectomy: a population-based study, 1971–1987. J Am Coll Cardiol 1989; 14: 1245–1252.

30.

Lomazzi

Mariscalco

Piffaretti

. Endovascular treatment of elective abdominal aortic aneurysms: independent predictors of early and late mortality. Ann Vasc Surg 2011; 25: 299–305.

31.

Khashram

Jenkins

. Long-term outcomes and factors influencing late survival following elective abdominal aortic aneurysm repair: a 24-year experience. Vascular 2016; 24: 115–125.

32.

Huang

Gloviczki

Oderich

. Outcome after open and endovascular repairs of abdominal aortic aneurysms in matched cohorts using propensity score modeling. J Vasc Surg 2015; 62: 304–311.

33.

Beck

Goodney

Nolan

. Predicting 1-year mortality after elective abdominal aortic aneurysm repair. J Vasc Surg 2009; 49: 838–844.

34.

Schouten

Kok

Hoedt

. The influence of aneurysm size on perioperative cardiac outcome in elective open infrarenal aortic aneurysm repair. J Vasc Surg 2006; 44: 435–441.

35.

Mani

Venermo

Beiles

. Regional differences in case mix and peri-operative outcome after elective abdominal aortic aneurysm repair in the vascunet database. Eur J Vasc Endovasc Surg 2015; 49: 646–652.

36.

Bahia

Holt

PJE

Jackson

. Systematic review and meta-analysis of long-term survival after elective infrarenal abdominal aortic aneurysm repair 1969–2011: 5 year survival remains poor despite advances in medical care and treatment strategies. Eur J Vasc Endovasc Surg 2015; 50: 320–310.

37.

Franks

Sutton

Bown

. Systematic review and meta-analysis of 12 years of endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2007; 33: 154–171.

38.

Dua

Kuy

Lee

. Epidemiology of aortic aneurysm repair in the United States from 2000 to 2010. J Vasc Surg 2014; 59: 1512–1517.

39.

Altman

Royston

. The cost of dichotomising continuous variables. BMJ 2006; 332: 1080.