The most widely accepted risk factors that have been cited for AAA include male sex, older age, and smoking [19]. Population-based studies in adults older than 50 have consistently reported a higher prevalence of AAAs in men versus women. A recent study reported prevalence of 3.9–7.2% in men and 1.0–1.3% in women [19]. Most AAAs found in the population occurred in individuals over the age of 60, with a total prevalence of 4–9% [20, 21]. One cohort study demonstrated a 4.5-fold increase in the relative risk of AAA for males over 65 compared to those under 55 [22]. However, the majority of these aneurysms were small, with diameters less than 3.5 cm, and likely not clinically important during the patients’ lifetime. More clinically important aneurysms over 4.0 cm exist in 1% of men between 55 and 64 years old, with incremental increases by 2–4% per decade thereafter [23]. Smoking is the most important risk factor , estimated to cause 75% of all AAAs over 4.0 cm and increasing risk of AAA by a factor of six [20, 24]. Other risk factors include positive family history, prior AAA, Caucasian or Native American ethnicity, cardiovascular disease, Hypertension (HTN), obesity, and aneurysms of the femoral or popliteal arteries [20, 22, 25, 26].

The natural history of AAAs is important to consider when establishing screening guidelines, as the risk for rupture and the expansion help determine surgical and surveillance planning . By projecting AAA growth curves, it is possible to estimate when the rupture risk is high and to intervene beforehand, as the case-fatality rate is 50% when surgery is performed emergently on the 40% of patients who even make it to the hospital [5, 27]. In contrast, the perioperative mortality from elective repair is reported to be 1–5%, and is largely dependent on patient comorbidities and the type of repair [3]. Fortunately, men without AAA by age 65 are unlikely (only about ~1%) to develop a new aneurysm over the course of the subsequent 5 years [28]. When aneurysms develop, however, larger aneurysms tend to grow faster than smaller aneurysms due to the increase in wall tension according to LaPlace’s law . According to one systematic review , for each 0.5 cm increase in AAA diameter, growth rates increased on average by 0.59 mm per year and rupture rates by a factor of 1.91 [29]. Aneurysms less than 4.0 cm in transverse diameter have a very low (~0%) annual risk of rupture, with an exponential increase in risk thereafter: 4.0–4.9 cm (0.5–5%), 5–5.9 cm (3–15%), 6–6.9 cm (10–20%), 7–7.9 cm (20–40%), and greater than 8 cm (30–50%) [30]. Extending this risk out to 5 years, the overall cumulative rupture rate of incidentally diagnosed aneurysms in population-based samples is 25–40% for aneurysms larger than 5.0 cm compared to 1–7% for aneurysms 4–5 cm [31–33].

Before imaging tests were developed, AAA screening was based on physical exam. However, accuracy of physical exam is limited by patient factors such as obesity and smaller aneurysm size [34]. Clinical studies have confirmed the poor reproducibility of physical exam, with sensitivity and specificity estimated at 39–68% and 75–91% [7, 19]. Aside from exposing patients to ionizing radiation, computed tomography (CT) can over-estimate aneurysm size by 2 mm or more because the cross sectional diameter of the aorta obtained in axial CT imaging is often not in the transverse plane [9]. While CT is more reproducible and remains the primary modality for operative planning, ultrasound has become the primary method for AAA screening because of its high sensitivity and specificity, portability, ease-of-use, safety (i.e., lack of radiation), and relative low cost [7]. While somewhat user-dependent, the sensitivity and specificity of ultrasound both approach 100%. Thus, given these advantages, ultrasound remains the primary method for AAA screening.

Four large randomized controlled trials (RCTs ) have been conducted to evaluate the effectiveness of population-based screening for AAAs using ultrasound : the Multicentre Aneurysm Screening Study (MASS), the Chichester, UK screening trial, the Viborg County, Denmark screening trial, and the Western Australia screening trial [35–38]. Multiple summative attempts have been made to combine these data sets, including a meta-analysis and two systematic reviews [19, 39, 40]. As these trials represent the highest-quality evidence in the literature, their cumulative data serves as the basis for all the major societal guidelines presented above. Overall, these trials showed that invitation to one-time screening for AAA is associated with a reduction in AAA-specific mortality in 65–75-year-old men. Follow-up reports for these trials have shown that this effect is both persistent, lasting up to 15 years [7, 41–44], and significant, with estimated relative reductions of 42% and 66% at 13 years in the two highest-quality trials [41, 44]. Other beneficial effects , including reductions in risk for AAA rupture and emergency surgery, persisted up to 13 years out from screening as well [7]. While these trials did not collect specific data about participants’ smoking histories or other risk factors, given the increased AAA prevalence in men who have ever smoked (6–7% of this population [24, 45]), the presence of this risk factor increases the benefit of screening in this population. The data for screening in other populations, including women, is less definitive [7].

Together, the four large population-based screening RCTs accumulated 137,214 participants with mean (or median) ages ranging from 67.7 to 72.7 years [7]. In each trial, participants were selected from population registries or regional health directories and randomized to either invitation for one-time ultrasound screening or usual care. The MASS trial , the largest of the four, randomized 67,800 men aged 65–74. This was the only trial that excluded participants based on health status; men that were too high risk to be screened by their primary care physicians, terminally ill, or had other serious health problems were excluded. Men with 3–4.4 cm aneurysms were followed with annual ultrasounds while those with 4.5–5.4 cm aneurysms were rescanned every 3 months. Surgery was offered to men with aneurysms greater than 5.5 cm, growth greater than 1 cm per year, or development of symptoms. Mean follow-up was 4.1 years in the original study but long-term data out to 13 years continues to be published [41, 42]. The Viborg trial included 12,658 men aged 65–73 years old. Participants with aneurysms above 3.0 cm were offered annual rescreening while those with aneurysms greater than 5.0 cm were offered surgery. While mean follow-up in the original study was 5.1 years, a subsequent report detailing results out to 10 years was published thereafter [44]. The Chichester trial was the only trial to include women, with a total of 15,775 randomized participants (6433 men, 9342 women), aged 65–80 years. Subjects with 3–4.4 cm aneurysms were followed with annual ultrasound, while those with 4.5–5.9 cm aneurysms were rescanned every 3 months. Surgery was offered to participants with aneurysms greater than 5.9 cm, growth greater than 1 cm per year, or development of symptoms. Lastly, the Western Australia trial involved 41,000 men aged 65–83 years. The structure of this study was unique in that it did not specify its post-screening ultrasound surveillance protocol. Men were provided with two copies of a letter detailing the outcome of their ultrasound: one for them and one for their primary care doctor. Follow-up care, whether rescreening or surgical referral, was left up to the discretion of the primary care doctor as they deemed appropriate. Median follow-up was 43 months.

In general, the statistical analysis plans and outcome variables among the trials were similar. All four trials were conducted via intention-to-treat analysis. Adherence to screening varied from 62.5% in the Western Australia trial, to 80.2% in the MASS trial. Less than 1% of the control groups crossed over in any trial to receive elective surgery, even at the longest follow-up of 13–15 years [19]. The primary outcome variable was AAA-specific mortality (all deaths related to AAAs and all deaths within 30 days of AAA surgical repair), but AAA rupture and all-cause mortality were also reported. In a recent systematic review that evaluated each trial according to USPSTF design-specific criteria [46], the MASS and Viborg trials were rated as “good-quality”, while the Chichester and Western Australia studies were labeled as “fair-quality” [19].

The prevalence of AAAs across the four trials ranged from 4.0% to 7.6%, with the majority (70–82%) less than 4.0–4.5 cm, and only a small proportion (0.4–0.6%) greater than 5.5 cm. The two “good-quality” trials, MASS and Viborg, demonstrated statistically significant reductions in AAA-related mortality in the groups invited to screening compared with the control groups, up to 13 years after screening (13-year hazard ratio [HR], 0.58 [CI, 0.49–0.69] and 0.34 [CI, 0.20–0.57], respectively) [35, 36, 41, 42, 44, 47]. For the MASS trial, this was associated with an absolute risk reduction of 0.14%, or 1.4 fewer AAA-related deaths per 1000 men screened [7, 41]. Not surprisingly, these two trials also found that an invitation to screening was associated with both lower AAA rupture rates at the 13-year follow-up (MASS: HR, 0.57 [CI, 0.49–0.67] [41]; Viborg: HR 0.44, [CI, 0.24–0.79] [47]) and significantly fewer emergency surgeries (MASS: Relative Risk (RR), 0.48 [CI, 0.37–0.63] at mean follow-up of 13.1 years [41]; Viborg: RR, 0.25 [CI, 0.09–0.66] at mean follow-up of 10 years [36]). While the two “fair-quality” trials, Chichester and Western Australia, did not report statistically significant results, they both showed a trend toward reductions in AAA-related mortality (Chichester: HR, 0.88 [CI 0.6–1.30] at 15 years of follow-up; Western Australia: RR, 0.61 [CI, 0.33–1.11] at 3.6 years of follow-up) [37, 38]. Of note, on post hoc analysis in the Western Australia trial, an invitation to screening was associated with a significant reduction in AAA-related mortality for men 65–75 years (OR, 0.19 [CI, 0.04–0.89]) and a trend toward increased mortality in older men (more than 75 years). This suggests that with better participant selection (i.e., excluding men over 75 years, when their likelihood of dying from other causes increases, thus limiting the benefit from AAA screening/repair), the Western Australia results may have aligned with the MASS and Viborg studies. As expected, all trials that reported rates of elective procedures showed significant increases (by about twofold) in elective AAA operations in the groups invited for screening: RRs 2.17 (MASS), 2.01 (Viborg), and 2.19 (Chichester), respectively [41, 47, 48]. Pooled analyses of data from these four trials from three independent groups demonstrated a statistically significant reduction, by about 45–50%, in the odds of AAA-related mortality (OR, 0.55 [CI, 0.36–0.86], 0.60 [CI, 0.47–0.78], and 0.57 [CI, 0.45–0.74]) [39, 40, 49]. None of the four trials found that an invitation to AAA screening was associated with a statistically significant all-cause mortality benefit at any time point up to 15 years. Pooled analyses of the four trials using random effects analysis have all showed no effect on all-cause mortality (ORs, 0.98 [CI, 0.97–1.00], 0.98 [CI 0.95–1.02], 0.95, [CI 0.85–1.07], and 0.98, [CI 0.95–1.0]) [19, 39, 40, 49]. This was not entirely unexpected, as fewer than 3% of participant deaths were attributable to AAA across the trials [7].

Other longitudinal studies have been conducted that investigate the effectiveness of AAA screening programs. One such study, published in 2012, reported attendance rates, screening and surveillance outcomes, and intervention rates and outcomes resulting from an AAA screening program initiated in Gloucestershire, England 20 years after the program was initiated. Sixty-two thousand men were invited with an 85% participation rate. From this subpopulation, 148 men had an aortic diameter greater than 5.4 cm and were referred for treatment, and 4.6% had a diameter between 2.6 and 5.4 cm and entered an ultrasound surveillance program. Perioperative mortality for the 631 surgeries performed for screen-detected AAAs was 3.9%, in line with the expected percentages based on other trials. An additional 372 procedures were performed for aneurysms detected incidentally with a mortality of 6.7%. Most tellingly, the number of ruptured aneurysms treated annually fell significantly during the course of the program [50].

The clinical data supporting AAA screening in women lags behind that of men. As noted previously, only one of the four major ultrasound-based AAA-screening RCTs recruited women , who were aged 65–80 years [10]. This trial found that AAA prevalence in women was six times lower than men (1.3% vs. 7.6%), in agreement with screening reports from Sweden that found a prevalence of 0.8% and 2.0% in ever-smokers and current smokers, respectively [11]. While most (75%) screen-detected AAAs were small (<3.9 cm), meta-analysis has shown that women have a three- to fourfold higher risk for rupture than men at the same diameter [51]. Nonetheless, rupture rates (0.06% in both groups), AAA-specific mortality (<0.2% in both groups, no statistical analysis), or all-cause mortality (10.7% vs. 10.2%) did not significantly differ at 5 years in the invitation-to-screening and control groups [7, 10, 19, 37]. Unlike men from the same trial in which the majority of ruptures occurred prior to age 80, most (70%) of the AAA-related deaths in women occurred at age 80 or older (at a time of increased competing causes of death and a declining benefit-risk ratio for operative intervention) [10]. Ultimately, the low prevalence of AAA in women resulted in a trial that was underpowered to draw definitive conclusions regarding health outcomes in this population [7]. In combination with the paucity of available data from other trials, uncertainty remains regarding the benefit women receive from population-based AAA screening.

One topic currently under investigation is selective screening in high-risk populations. A history of smoking is the most important risk factor for developing an AAA and has been suggested as a possible criterion for selective AAA screening [52]. Even a relatively modest smoking history (i.e., half-pack or less per day for less than 10 years) increases the likelihood of developing a large AAA [52]; this effect has been estimated as a three- to fivefold increase in AAA prevalence across all age groups and an increase in AAA-related mortality [24, 49, 53]. Unfortunately, the population-based screening RCTs did not collect specific data about participants’ smoking histories. As a result, modeling studies have been conducted to determine how the impact of screening would differ in those with a history of smoking as compared to those who had never smoked [49]. In one study of 100,000 hypothetical U.S. men aged 65–74 years, the invitation of only ever-smokers (69% of men in this population) to attend screening would account for 89% of the expected reduction in AAA-related mortality from population-based screening of all men 65–74 years of age [49]. In a simulation analysis based on participant data from the Viborg trial, selective screening in high-risk patients, defined as those having Chronic Obstructive Pulmonary Disease (COPD ) or a cardiovascular condition, would have prevented half of all reported deaths at 5 years and required 72.9% fewer screening invitations compared to mass screening [54]. Other modeling studies have shown that screening strategies based on age, sex, and smoking history outperform strategies that use other risk factors, such as family history, coronary artery disease, or hypercholesterolemia [55, 56].

A core requirement for policy-making regarding any potential AAA screening program is cost-effectiveness . The literature consists of data derived from the Viborg and MASS trials [42, 44, 57], as well as hypothetical data [58]. In the MASS trial, cost-effectiveness based on AAA mortality at 4, 7, and 10 years of follow-up was $44,900, $19,500, and $12,579, respectively, per life-year gained [42, 43, 59]. Likewise, the cost per life-year gained as determined in the Viborg trial also improved from $12,736 at 5 years to $2566 at 15 years [17, 44]. As expected, this incremental improvement in cost-effectiveness over time reflects the high initial cost of screening and elective surgery (for aneurysms >5.5 cm) followed by continued long-term benefit [42, 43, 57, 60]. As the survival advantage in terms of life-years gained continues to increase with time, predictive models based on MASS data estimate a cost of $3806 per life-year gained over the full lifetime of men aged 65, indicating an extremely cost-effective program [42, 57]. Other recent cost-effectiveness studies using MASS data have tried to account for recent changes in the management and epidemiology of AAAs (decreasing prevalence of AAA and new surgical approaches) and have still found that screening men was cost-effective and delivered significant clinical benefit [61]. Nevertheless, there are other reports that suggest AAA screening is not cost-effective [58]. This study utilized a hypothetical population in their prediction model and clinical cost profiles native to Denmark (i.e., for screening, elective surgery, emergency surgery) which differed significantly from the UK, which together are believed to explain the contrasting findings [62, 63].

As with any medical intervention, ultrasound screening represents a balance between benefits, i.e., identifying AAAs early on in the non-emergent setting when it is possible to undergo elective surgery, and risks. Unlike other forms of imaging, ultrasound has no known direct physical risks [64]; instead, its risk profile consists of two indirect effects: psychological distress in those who screen positive and adverse outcomes from operative management [65].

Anxiety/depression , decreased quality of life, and poor health perception comprise the most frequently investigated negative psychological outcomes from screening. Most of the data on these adverse effects was collected in the Viborg and MASS RCTs, in addition to five observational studies [35, 66–71]. Unfortunately, the aggregate results are conflicting, with four of the five observational studies showing no clinically significant decrease in quality-of-life measures in those who screened positive compared to unscreened control participants, the MASS trial demonstrating only a transient negative psychological effect that resolved after 6 weeks, and the Viborg trial finding a small, but significant, immediate negative change in the psychological profile of participants who screened positive after 1 month of conservative management. While it is difficult to generate definitive conclusions, these results suggest that a portion of the population who screen positive but do not require immediate intervention may develop mild, though likely transient, adverse psychological effects.

AAA repair, whether open or endovascular, remains associated with significant complications (surgical complication, hospitalization, or even death) [72]. The most feared complication of AAA repair, perioperative mortality, occurs in 2.7–5.8% of elective cases, depending on patient specific comorbidities, the type of procedures, and other operator-specific factors (surgeon experience, type of surgeon, hospital volume, etc.) [73–75]. A separate issue altogether is the relative mortality in patients undergoing elective open surgery compared to endovascular repair. One of the byproducts of screening programs is an increase in elective procedures, as borne out by data from the four population-based screening RCTs: the risk for any AAA-related operation in the invited group was approximately double that of the non-invited group at 3–5 years in all trials due to an uptick in elective procedures [35–38]. Concurrently, most screening trials reported associated decreases in emergency repairs (data presented above) in populations invited to screen [19]. While the complication rate may remain constant, estimated at 32% for elective AAA repairs [72], the total number of complications will inevitably rise. Thus, the increase in the overall rates of detection and surgery in the screening group potentially represents a harm, as a proportion of AAAs will never rupture due to cessation in growth or death from a competing cause. The extent of over-diagnosis and over-treatment is unfortunately difficult to estimate [7].

In conclusion, undetected AAAs represent a major public health concern because of the high rates of mortality with rupture. Ultrasound imaging has emerged as a viable screening modality because of its high sensitivity and specificity, reproducibility, portability, safety, and affordability. While separate guidelines have been published by multiple societies and governmental agencies, they share a consensus for those populations considered at high risk, i.e., men aged 65–75 with a history of smoking. Controversy still exists over other subpopulations for which the data is inconsistent or lacking altogether, i.e., women, younger non-smoking men, and those with risk factors other than smoking or a strong family history. Improved definition of those populations at high risk who will most benefit from screening is needed.

Signed into law by President George W. Bush in 2007, the SAAAVE Act provides for a one-time AAA screening ultrasound as part of the “Welcome to Medicare Physical Exam” for patients with defined risk factors . More specifically, this population includes men aged 65–75 who have smoked at least 100 cigarettes and patients of either gender with a family history of AAA. Prior to 2007, screening for AAAs was not a covered Medicare benefit, requiring patients to pay out-of-pocket [76].

Although its intentions were supported by clinical evidence showing an AAA-related mortality benefit [35, 38, 39, 43, 49, 77], unfortunately the SAAAVE Act has been fraught with controversy since its inception. Opponents of the act point to a recent study from Stanford that found the act was associated with an increased number of abdominal ultrasounds , but without concurrent improvements in clinical outcomes [77]. The study, which compared a sample of Medicare enrollees eligible for screening to a control group that was not eligible for screening, found that while the use of abdominal ultrasound had increased 2.0% (7.6–9.6%) among SAAAVE-eligible men from 2004 to 2008, there were no apparent changes in the rates of AAA repair, AAA rupture, or all-cause mortality. Perhaps more concerning was that fewer than 10% of SAAAVE-eligible Medicare enrollees actually received the abdominal ultrasound. Proposed reasons for underutilization of abdominal aortic ultrasound screening include significant system factors, such as lack of awareness on the part of physicians and patients regarding the SAAAVE Act or the potential benefits of AAA screening, but also lack of in-depth history-taking to identify high-risk patients, and the misguided belief that AAAs can be easily palpated on physical examination [5, 77]. Other opponents contend that the modest impact of the SAAAVE Act on screening rates is based on a small absolute reduction in clinical events resulting from AAA screening, rather than a reduction in a significant percentage of those patients at risk [78]. They argue that the widespread adoption of AAA screening is not justifiable by medical practitioners because the population benefit is low relative to the time period upon which the USPSTF based their original screening recommendations . This conclusion is based on recent clinical evidence from Europe suggesting a decline in mortality from ruptured AAAs, and thus “clinically relevant aneurysms”, over the past 10–15 years attributed to a decrease in the prevalence of smoking rather than the implementation of AAA screening or the rise of endovascular repair [79, 80].

Proponents of the SAAAVE Act are quick to point out that the 1-year follow-up in the study by Shreibati et al. might be too short to observe a reduction in all-cause mortality [77]. Randomized trials of AAA screening have shown that the reduction in AAA-related mortality is not apparent for at least 1 year after the initial AAA screening [35]. In addition, they argue that the requirements for patient eligibility and physician reimbursement prevent widespread adoption of AAA screening ultrasounds [76]. They point to the modest 2% increase in screening ultrasounds, still less than 10,000 total exams in 2007, after the implementation of the SAAAVE Act as evidence of the barriers for beneficiaries. Indeed, at-risk Medicare beneficiaries must obtain a referral for AAA screenings during their “Welcome to Medicare Physical Exam” and must be screened during their first 6 months of eligibility [81]. Furthermore, potential beneficiaries are required to pay a 20% co-payment out-of-pocket prior to screening [76]. Additionally, the millions of patients not newly enrolled in the Medicare program are not eligible.

A recent study conducted at the Geisinger Medical Center would suggest that more aggressive screening measures are in order [5]. The study investigated whether current screening guidelines under the SAAAVE Act, in conjunction with routine ambulatory medical care evaluation, were an effective way to identify and screen patients at risk for ruptured AAA. To do this, the authors retrospectively reviewed the pre-operative clinical data and outpatient office visit notes for all patients who presented with ruptured AAAs at their institution over a 6-year period. Notably, only 17% of patients who presented with a ruptured AAA would have been eligible for a screening ultrasound based upon the SAAAVE Act criteria at the time of rupture. The study also found significant gender disparities: while 30% (16 total women) of the study patients were women, only one would have been eligible for screening according to the SAAAVE Act. Further underscoring the importance of screening ultrasound was their finding that physical exam was inadequate to diagnose AAAs, as only 9.6% of patients had findings that the practitioner felt were suspicious for AAA. The authors rightfully concluded that current AAA screening guidelines, as currently constructed, are inadequate in reducing aneurysm-related mortality.

Another barrier to the provision of screening ultrasounds by a medical practitioner is the strict requirement for reimbursement . This suggestion was raised in response to one of the findings in the study by Shreibati et al. In the study, they found that the small increase in screening rates observed after the SAAAVE act was due to an increase in abdominal ultrasonography not reimbursed under the program (i.e., not an approved current procedural terminology [CPT] code under the act). While this may have purely been attributable to a lack of education regarding the proper CPT codes, one alternative explanation offered was that for some eligible patients, the screening ultrasound was billed under a different CPT code because it did not meet the complex criteria for reimbursement [77].

While the SAAAVE Act undoubtedly has its drawbacks , it nevertheless raised awareness of AAAs and lays the foundation for future progress toward the provision of potentially life-saving abdominal ultrasonography to all at-risk patients. Efforts by the SVS are already underway to introduce new legislation that would unlink AAA screening from the “Welcome to Medicare Physical Exam” as well as expand the one-time screening to 65–75-year-old at-risk Medicare beneficiaries. The largest hurdle may prove to be education of both primary care physicians and patients about AAAs. Indeed, multiple studies investigating patients presenting with ruptured AAAs have found that a high percentage (~30–40%) of them had a known AAA prior to rupture [5, 82]. In one of these studies, 40% of patients with radiographic evidence of AAA prior to rupture were never referred or evaluated in the vascular surgery clinic [5]. Akin to the disease processes of breast cancer and colorectal cancer, AAA screening reduces disease-specific mortality but not all-cause mortality [77]. However, awareness of AAA screening certainly lags behind the screening programs for these diseases, namely mammography and fecal occult blood testing, which receive ample attention in the press. This deficiency has not gone unnoticed, as companies such as Gore have created websites to raise awareness and provide patients with information on the SAAAVE Act [83].

Perhaps most important from a practical perspective, AAA screening has been shown to be cost-effective , with an estimated cost-effectiveness ratio of $19,500 per life-year gained [43]. Furthermore, evidence exists supporting the cost effectiveness and efficacy of a screening program. When a large-scale screening effort for identifying AAAs in patients in clinical practice was implemented, the prevalence of AAAs and aneurysm distribution reflected those reported in major clinical trials at a reasonable cost of $53 per ultrasound [81]. Given the marked improvement in mortality associated with elective AAA repair compared to emergent repair of a ruptured AAA [5] and the cost-effectiveness of AAA screening, the most effective method of reducing AAA-related mortality is early identification and elective repair. Within this context, the SAAAVE Act represents the foundation upon which future efforts will build a more thorough screening program that provides coverage to all at-risk beneficiaries.