Prevalence of Elective and Ruptured Abdominal Aortic Aneurysm Repairs by Age and Sex From 2003 to 2016 in Ontario, Canada

IMPORTANCE Age and sex are important considerations in assessing and individualizing therapy for abdominal aortic aneurysm (AAA) repair. OBJECTIVE To determine the prevalence of open and endovascular elective AAA (EAAA) and ruptured AAA (RAAA) repair by age and sex. DESIGN, SETTING, AND PARTICIPANTS In this retrospective, population-based, cross-sectional, time-series analysis in Ontario, Canada, from April 1, 2003, to March 31, 2016, all patients undergoing AAA repair who were older than 39 years were included. EXPOSURES Elective AAA and RAAA repair with open surgical repair (OSR) or endovascular aortic repair (EVAR). MAIN OUTCOMES AND MEASURES Ageand sex-standardized rates of EAAA and RAAA repair with OSR and EVAR. RESULTS From 2003 to 2016, 19 489 EAAA repairs (12 232 [63%] OSR and 7257 [37%] EVAR) and 2732 RAAA repairs (2466 [90%] OSR and 266 [10%] EVAR) were identified. The mean (SD) age was 72.7 (8.1) years in the EAAA subgroup and 73.5 (8.9) years in the RAAA subgroup; 15 813 patients (81%) in the EAAA subgroup and 2178 (80%) in the RAAA subgroup were men. The rates of EAAA by age quintile and sex decreased over the study period except among patients older than 79 years (1.3 per 100 000 population in 2003 to 2.2 per 100 000 population in 2016; 70% increase; P < .001). The rates of elective OSR decreased across all age and sex subgroups (range, 38%-74% decrease; P .009 for all subgroups) except among patients older than 79 years (1.3 per 100 000 population at baseline to 0.56 per 100 000 population in the second quarter of 2016; 53% decrease; P = .05). The rates of elective EVAR significantly increased across all age and sex subgroups (range, 566%1585% increase; P .04 for all subgroups). Elective EVAR became the dominant treatment approach for aneurysms in men around 2010, whereas it maintained parity among women in 2016. The RAAA repair rate decreased over the study period in all subgroups (range, 32%-91% decrease; P .001 for all subgroups), but the decrease was not significant among women (80% decrease; P = .08). Similarly, the rates of ruptured OSR decreased among all subgroups (range, 47%-91% decrease; P < .001), but the decrease was not significant among women (87% decrease; P = .54). Ruptured EVAR showed significant uptake in all subgroups. CONCLUSIONS AND RELEVANCE Among patients with AAA in Ontario, Canada, use of EVAR appeared to increase from 2003 to 2016, whereas OSR use appeared to decrease. These findings were most pronounced among elective procedures for men and older patients. The delayed increase (continued) Key Points Question What was the prevalence of elective and ruptured abdominal aortic aneurysm repairs by age and sex from 2003 to 2016 in Ontario, Canada? Findings This population-based, crosssectional study of 19 489 elective and 2732 ruptured abdominal aortic aneurysm repairs revealed an increase in the rate of elective abdominal aortic aneurysm repair among patients older than 79 years. Endovascular repair was the preferred treatment approach for elective repair among men, although there was greater uptake for ruptured repair among women. Meaning The findings suggest that endovascular repair may be the preferred method for elective and ruptured aneurysm repair, particularly among older individuals, for elective repairs in men, and for ruptured repairs in women. Author affiliations and article information are listed at the end of this article. Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2018;1(7):e185418. doi:10.1001/jamanetworkopen.2018.5418 November 30, 2018 1/12 Abstract (continued)continued) in the use of EVAR among women may reflect continued anatomical constraints for women seeking elective repair. JAMA Network Open. 2018;1(7):e185418. doi:10.1001/jamanetworkopen.2018.5418


Introduction
Age and sex are well-established risk factors for many cardiovascular diseases, including abdominal aortic aneurysms (AAAs). The increasing prevalence with age and the 4:1 male to female ratio of AAA prevalence are both well established in the literature. [1][2][3][4] Furthermore, these demographics are related to anatomic and physiological considerations that influence the approach to AAA treatment.
The approach to AAA treatment requires a careful consideration of comorbidities and correspondent perioperative mortality and morbidity risk, weighed against the natural history of the untreated aneurysm. The correlation among age, cardiopulmonary comorbidity, and perioperative mortality is well known, such that age is a common covariate included in perioperative risk prediction models. 5 Before the introduction of endovascular aortic repair (EVAR), elderly patients were frequently turned down even for elective AAA (EAAA) repair owing to the prohibitive surgical risk associated with conventional open surgical repair (OSR). 6 The lower perioperative mortality associated with EVAR changed the approach to treatment of EAAA and ruptured AAA (RAAA), providing patients with previously inoperable AAA with a treatment option. 6,7 Sex introduces additional technical considerations to the treatment of vascular disease. The application of EVAR to treatment of AAA in women was partially limited by a size mismatch between the size of the endograft delivery system and the smaller size of femoral and iliac arteries in women. 8 However, refinements in endograft fixation and stent organization have led to reductions in delivery system size and expansion of available endograft diameters. Current delivery systems have outer diameters as small as 16 French. Thus, over time, the application of EVAR for AAA in the female anatomic structure should be less limited by available technology.
Understanding variable trends in the application of EVAR for AAA treatment by age group and sex may help to clarify and address limitations in EVAR technology. This retrospective, populationbased, time-series analysis study sought to determine whether age and sex differences in the application of EVAR for the treatment of AAA exist by analyzing the age-and sex-specific rates of elective OSR, elective EVAR, ruptured OSR, and ruptured EVAR over 13 years (2003-2016).

Study Design and Setting
Health Information Protection Act. The ICES data are derived from multiple primary data sources provided by the federal and provincial governments as well as various research organizations, registries, and initiatives. These data are anonymized and linked using an ICES key number. The specific data sets used for this study include the Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database, the National Ambulatory Care Reporting System database, and the Ontario Health Insurance Plan database. 10

Patient Cohort
The study cohort consisted of all Ontarians older than 39 years who underwent EAAA and RAAA repair in Ontario, Canada, from April 1, 2003, to March 31, 2016. Patients receiving elective OSR, ruptured OSR, elective EVAR, and ruptured EVAR were identified using a combination of the

International Statistical Classification of Diseases and Related Health Problems, Tenth Revision,
Canada (ICD-10-CA), Canadian Classification of Health Intervention (CCI), and Ontario Health Insurance Plan codes according to a previously validated algorithm. 10 Sex and age were collected directly from the described databases.

Statistical Analysis
The study period was divided into 52 quarterly intervals from April 1, 2003, to March 31, 2016. The number of overall and approach-specific EAAA and RAAA repairs conducted during each quarterly interval were counted. Observed quarterly age-and sex-specific repair rates per 100 000 population were calculated using the quarterly, age quintile-specific Ontario population or the quarterly, sex-specific Ontario population older than 39 years as the denominator. The denominators for rate calculations were derived from the 2015 Canadian census data and associated projections. Age quintiles were defined as 40 to 64 years (quintile 1), 65 to 69 years (quintile 2), 70 to 74 years (quintile 3), 75 to 79 years (quintile 4), and older than 79 years (quintile 5).
Where possible, the percent change in repair rate was calculated for each group relative to the repair rate in the second quarter of 2003 (baseline). To examine the presence of statistically significant trends within each group, autoregressive integrated moving average (ARIMA) models were fit with linear or quadratic trend regressors as dictated by visual inspection of graphical data.
ARIMA models are specific applications of linear regression models for time-series data with autocorrelated errors. 11,12 Model appropriateness was assessed using autocorrelation, partial and inverse autocorrelation plots, and the Ljung-Box q statistic. Model fit was evaluated using the Akaike information criterion and the adjusted R 2 values. All statistical analyses were conducted in SAS, version 9.4 (SAS Institute), with a 2-sided P value of less than .05 for statistical significance.

Age-and Sex-Specific Rates of EAAA Repair
The overall age quintile-specific rates of EAAA repair decreased over the study period and demonstrated negative trends (P < .001 for all quintiles) ( Figure 1A).

Age-and Sex-Specific Rates of RAAA Repair
The overall age quintile-specific rates of RAAA repair decreased and demonstrated significant negative trends over the study period (P Յ .001 for all age quintiles) ( Figure 3A). The greatest decrease in overall RAAA repair rate per 100 000 population was seen in age quintile 3, in which the rate of RAAA repair decreased from 0.45 at baseline to 0.04 in the second quarter of 2016 (91% decrease; P < .001). Quintile 1 had the smallest decrease in overall RAAA repair rate per 100 000 population (0.27 at baseline to 0.19 in the second quarter of 2016 [32% decrease]; P < .001). Regarding the approach-specific rates, the rates of ruptured OSR decreased, whereas ruptured EVAR rates increased across all age quintiles ( Figure 3B and C

Discussion
This population-based time-series analysis found significant age-and sex-specific trends in overall and approach-specific EAAA and RAAA repair rates from 2003 to 2016. Of note, the study showed that EAAA repair rates increased in patients older than 79 years in association with an increase in the use of EVAR within this age group. Furthermore, EVAR became the predominant approach over OSR for EAAA repair in men around 2010, whereas its rate of use was similar to that of OSR around 2012 in women. Finally, this study showed that EVAR uptake increased among RAAA repairs but OSR The findings in this study were generally consistent with the literature on this topic. In this analysis, patients older than 79 years experienced a large increase in EAAA repair that was associated with a greater than 15-fold increase in the uptake of EVAR. These findings may demonstrate an increase in the treatment of comorbid patients that would otherwise be refused AAA repair with OSR. In this scenario, concern about long-term graft durability, reintervention, and secondary rupture are offset by a relatively short life expectancy. However, the multicenter EVAR-2 15 randomized clinical trial comparing EVAR with no intervention in 338 patients unfit for ruptured OSR demonstrated no difference in all-cause or aneurysm-related mortality for up to 4 years after randomization. The authors also showed higher complication and reintervention rates among patients receiving EVAR and considerably higher costs associated with EVAR. However, the participants of the EVAR-2 study demonstrated low aspirin and statin use, and these findings may reflect EVAR uptake in a population of patients with better medical management and thus lower perioperative risk. 16,17 The literature regarding the sex-specific trends in EAAA repair is similarly sparse. However, assuming proportionality between that EAAA repair rates and AAA prevalence, the findings in this study with respect to the overall EAAA repair rates grouped by sex are consistent with AAA screening studies, which demonstrate a 4-to 6-fold higher prevalence of AAA among men than women. [1][2][3][4] Regarding the uptake of elective OSR and elective EVAR by sex, these findings contradict the few data available. Dillavou et al 18  The findings of this study regarding overall rates of RAAA repair are consistent with existing Canadian, US, and European studies demonstrating slight declines in RAAA repair rates. 14,[24][25][26] However, given the recognized importance of AAA screening by societal guidelines, 27 greater differences were expected in the changes in rates among the eldest patients. Guidelines recommend the screening of patients from age 65 to 75 years. [27][28][29] If screening programs were successful, lower  owing to a 10-year period during which AAA could have been discovered and treated. However, despite the existence of multiple guidelines recommending AAA screening, Ontario does not have a structured AAA screening program in place, and the uptake of screening according to these guidelines is unknown. Furthermore, guidelines recommend screening in men only. Therefore, the relatively modest decline in RAAA repair in patients older than 79 years may potentially be explained by poor screening uptake and perhaps by unaddressed rupture rates among women. Finally, this modest decline may be the result of the growing availability and experience with ruptured EVAR, resulting in a reduction in the turndown rate for RAAA repair in the eldest patients and those with the highest operative risk. The uptake of ruptured EVAR was the greatest in quintile 5 in this study.
However, the effect of ruptured EVAR on RAAA repair turndown is unknown in this study population.
In contrast to data regarding RAAA trends by age group, more data are available regarding these trends by sex. The study by Dillavou et al 18 demonstrated a greater reduction in overall RAAA repairs among men than women from 1994 to 2003 (29.3% decrease in men vs 12.2% decrease in women).
These trends were reiterated in the study by Mureebe et al, 30 which showed a 52% reduction in men and a 36% reduction in women for RAAA repair. In contrast, this study demonstrated a greater reduction in RAAA repair rates among women (79%) than men (67%

Limitations
This study had several limitations. Administrative and billing codes were used to indirectly identify the cohorts. Although the code combinations were previously validated, the reabstraction methodology only allowed for the measurement of the positive predictive value of the codes. As a result, this study may underrepresent the number of AAA repairs conducted during the study period.
Next, this study may have limited generalizability because Ontario administrative data were used.
Ontario residents are part of a single-payer, publicly funded health care system in which judicious use of limited health care resources is necessary for distributive justice of these resources. Also, the Ontario age-and sex-specific populations were used to calculate the rates of AAA repair within each subgroup as opposed to using the prevalent AAA population as the denominator for our calculations.
This methodology assumes a stable AAA incidence to make inferences regarding AAA repair rate changes and is in line with findings from contemporary incidence studies. 32 Regardless, further collaborative work must be conducted to investigate the incidence and prevalence of AAA in multiple large North American data sets, as well as to characterize the rates of AAA repair turndown and how these rates have changed in the context of evolving endografts and experience.

Conclusions
Our population-level time-series analysis of more than 20 000 AAA repairs revealed an overall shift toward EVAR in all age and sex subgroups undergoing EAAA and RAAA repair. The greatest increases were observed among patients undergoing EAAA repair who were older than 79 years and among women. However, the proportion of EAAA repairs among women that were conducted by elective EVAR was still lower than that among in men. These findings suggest a potential need to more judiciously apply elective EVAR in the eldest and most comorbid patients, considering evidence from