Advertisement

Editor's Choice – Infra-Renal Aortic Diameter and Cardiovascular Risk: Making Better Use of Abdominal Aortic Aneurysm Screening Outcomes

Open ArchivePublished:May 10, 2021DOI:https://doi.org/10.1016/j.ejvs.2021.03.013

      Objective

      Aortic diameter (AD), used traditionally for abdominal aortic aneurysm (AAA) screening may have a role in assessing cardiovascular risk. Unfortunately, AD estimates for those without AAA are underutilised, whilst cardiovascular risk is sub-optimally managed in those with AAA. Our objective was to examine the association between AD measurements and future cardiovascular risk.

      Methods

      Retrospective analysis of three databases of male participants screened for aortic aneurysm disease. Imaging and clinical data were obtained from three independent sources: 1) the Multi-centre Aneurysm Screening Study (MASS) trial (n = 26 882 men); 2) the 2013/14 cohort of the English NHS AAA Screening Programme (NAAASP) (n = 237 441 men) linked with NHS hospital admission and death registry data; and 3) the Framingham Heart Study (FHS) offspring cohort (n = 649). Associations between maximal aortic diameter, as measured on ultrasound or computed tomography, and cardiovascular outcomes were examined.

      Results

      Cardiovascular mortality in the MASS trial, was higher in men with AAA at 13 years of follow up, compared to those without (Hazard Ratio [HR] 2.22, 95% CI 1.97–2.50, p < .001). Contemporary risk of major adverse cardiovascular events in the NAAASP was highest in those with an AAA (HR 2.91, 95% CI 2.00–4.25), whilst, extremes of aortic diameter were associated with increased risk for cardiovascular events. Aortic diameter was an independent risk factor for cardiovascular events in the FHS dataset.

      Conclusion

      Irrespective of the diagnosis of AAA, men attending for AAA screening who are found to have an abnormal aortic diameter are at high risk of future cardiovascular events. This currently unutilised data from AAA screening programmes has the potential to improve preventative management of cardiovascular risk.

      Keywords

      In this paper we demonstrate using large national abdominal aortic aneurysm (AAA) screening datasets that men with AAA have extremely high cardiovascular risk. Furthermore we have, for the first time, demonstrated that both high and low extremes of aortic diameter are associated with high cardiovascular risk. Lastly, using data from the Framingham heart study, we have demonstrated that aortic diameter predicts cardiovascular events independent of clinical risk scoring systems. This strong epidemiological evidence identifies major missed opportunities in AAA screening and the management of abdominal aortic aneurysm.

      Introduction

      The term abdominal aortic aneurysm (AAA) is used to encompass a plethora of degenerative processes that ultimately lead to a fixed and focal dilatation of the abdominal aorta, to a diameter that exceeds that of the healthy adjacent aorta by at least a factor of 1.5.
      • Wanhainen A.
      • Mani K.
      • de Borst G.J.
      The most important news in the new ESVS 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysm.
      ,
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor’s Choice - European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      Despite significant advances in our understanding of the mechanisms underlying the pathogenesis of AAA and the addition of new methods into the surgical armamentarium for its operative management, aneurysms still pose a significant public health issue.
      • Brady A.R.
      • Fowkes F.G.
      • Thompson S.G.
      • Powell J.T.
      Aortic aneurysm diameter and risk of cardiovascular mortality.
      • Dabare D.
      • Lo T.T.H.
      • McCormack D.J.
      • Kung V.W.S.
      What is the role of screening in the management of abdominal aortic aneurysms?.
      • Rughani G.
      • Robertson L.
      • Clarke M.
      Medical treatment for small abdominal aortic aneurysms.
      To minimize the impact of the associated morbidity and mortality, the UK, Sweden, Germany and the USA offer screening programmes.
      • Jacomelli J.
      • Summers L.
      • Stevenson A.
      • Lees T.
      • Earnshaw J.J.
      Impact of the first 5 years of a national abdominal aortic aneurysm screening programme.
      Whilst only a small proportion of men screened for AAA have an aneurysm, all men have their aortic diameter measured and recorded. However, nothing is done with this measurement for men without AAA.
      Recent reports have highlighted the potential role of using aortic diameter as a marker of future cardiovascular disease, thus providing an opportunity for early intervention through risk factor modification and institution of secondary prevention.
      • Brady A.R.
      • Fowkes F.G.
      • Thompson S.G.
      • Powell J.T.
      Aortic aneurysm diameter and risk of cardiovascular mortality.
      ,
      • Freiberg M.S.
      • Arnold A.M.
      • Newman A.B.
      • Edwards M.S.
      • Kraemer K.L.
      • Kuller L.H.
      Abdominal aortic aneurysms, increasing infrarenal aortic diameter, and risk of total mortality and incident cardiovascular disease events: 10-year follow-up data from the Cardiovascular Health Study.
      ,
      • Powell J.T.
      • Brown L.C.
      • Forbes J.F.
      • Fowkes F.G.
      • Greenhalgh R.M.
      • Ruckley C.V.
      • et al.
      Final 12-year follow-up of surgery versus surveillance in the UK Small Aneurysm Trial.
      In addition, utilisation of screening programme data for patients without aneurysmal disease of the aorta may expand the clinical utility and improve the cost-effectiveness of existing screening programmes.
      • Glover M.J.
      • Kim L.G.
      • Sweeting M.J.
      • Thompson S.G.
      • Buxton M.J.
      Cost-effectiveness of the National Health Service abdominal aortic aneurysm screening programme in England.
      Despite encouraging initial reports, the available evidence is dated and there remains a paucity of data in the literature regarding the reliability of using sub-aneurysmal aortic diameter as a marker of future cardiovascular disease.
      The aim of this study is to determine if aortic diameter measurements AAA screening programmes and the Framingham Heart Study, can be used to identify groups of men at high risk of future cardiovascular events.

      Methods

      Study design

      This is a retrospective analysis of prospectively collected data, from three datasets. The primary aim of this study was to investigate the association between infra-renal aortic diameter and the risk of future cardiovascular mortality or major adverse cardiovascular events (MACE), defined as death, stroke or myocardial infarction. Three separate datasets were used to achieve this aim. These were i) the cohort of men screened for AAA in the Multi-Aneurysm Screening Study (MASS),
      • Thompson S.G.
      • Ashton H.A.
      • Gao L.
      • Buxton M.J.
      • Scott R.A.
      Final follow-up of the Multicentre Aneurysm Screening Study (MASS) randomized trial of abdominal aortic aneurysm screening.
      ii) the cohort of men screened as part of the NAAASP in 2013–2014

      NHS Abdominal Aortic Aneurysm Screening Programme. NAAASP Data 2014-2015. https://www.gov.uk/government/publications/abdominal-aortic-aneurysm-screening-2014-to-2015-data. [Accessed 1 January 2016].

      , and iii) the offspring cohort of the Framingham Heart Study (FHS).

      Framingham Heart Study. Research results. https://www.framinghamheartstudy.org/fhs-for-researchers/selected-research-results/. [Accessed 1 January 2016].

      Definitions

      Cardiovascular death was defined using the methods described by van Staa et al. and these definitions were used for all subsequent analyses in this study.
      • van Staa T.P.
      • Gulliford M.
      • Ng E.S.
      • Goldacre B.
      • Smeeth L.
      Prediction of cardiovascular risk using Framingham, ASSIGN and QRISK2: how well do they predict individual rather than population risk?.
      In summary, the cause of death was recorded using the codes described in the 9th revision of the International Classification of Diseases (ICD-9) manual. These included acute myocardial infarction (ICD-9: 410), chronic ischaemic heart disease (ICD-9: 414), cerebral infarction not specified as haemorrhage (ICD-9: 434), intracerebral haemorrhage (ICD-9: 430) and non-traumatic intracranial hemorrhage (ICD-9: 430–435). Aneurysm related deaths were analysed separately (ICD-9: 441) as the inclusion of aortic deaths in this outcome measure could result in excess cardiovascular mortality in men diagnosed with AAA. Non-cardiovascular death was defined as death due to any cause not contained in the aforementioned definitions.
      Crude and adjusted survival analyses were performed to calculate sub-distributional Hazard Ratios (HRs) for each aortic group, and were plotted as cumulative incidence function curves.

      Datasets

      The Multi-Aneurysm Screening Study cohort

      The MASS study was a population-based randomised controlled trial of men aged 65–74 years recruited between 1997 and 1999 from four regions in the UK.
      • Thompson S.G.
      • Ashton H.A.
      • Gao L.
      • Buxton M.J.
      • Scott R.A.
      Final follow-up of the Multicentre Aneurysm Screening Study (MASS) randomized trial of abdominal aortic aneurysm screening.
      Men were randomised into two groups; the invited group comprising those that were invited for aneurysm screening and the control group, comprising those that were not. Measurements were undertaken from the maximal aortic diameter obtained from aortic ultrasonography, using the inner-to-inner diameter, and were only recorded for those whose AD exceeded 3cm, but are otherwise irretrievable. Causes of death in the screened group (n = 27 204 men) were compared between those with AAA and those without. Follow-up data were limited to cause of death obtained from the UK Office for National Statistics.

      The NHS AAA Screening Programme cohort

      To examine associations between aortic diameter and cause of death in a contemporaneous dataset we obtained outcome data for the 2013/14 screening year from the NAAASP and linked individual records to the English Hospital Episode Statistics (HES) and Office for National Statistics (ONS) datasets. The NAAASP invites men on their 65th year to have an ultrasound of their abdominal aorta. Men found to have a normal aortic diameter are discharged from further follow-up, whilst those found to have a diameter over 5.5 cm are referred to a vascular service to be considered for a repair of the aneurysm. Those found to have an AAA are screened at time intervals dependent on their aneurysm size. Measurements used in this study were undertaken from the maximal aortic diameter obtained from aortic ultrasonography, using the inner-to-inner diameter. Data linkage and sharing was undertaken under approval by the UK Health Research Authority Confidentiality Advisory Group, under Section 251 of the National Health Service Act 2006. HES is the administrative data set for the NHS which contains details of all admissions, outpatient appointments including accident and emergency attendances at NHS hospitals in England. NHS patients are uniquely identified in HES via their NHS number, so individuals can be tracked through multiple episodes including admissions to different NHS hospitals. ONS mortality data contains information related to a person's death taken from the death certificate for all deaths registered in England and Wales.

      Office for National Statistics. Births and Deaths. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths. [Accessed 1 January 2016].

      It contains information including cause of death as per the International Classification of Disease manual, as well as the date and time of death. Data was extracted from the HES/ONS datasets including cause of death and primary cause of admission to hospitals both using ICD-10 codes and date of the event as was described above. Only men routinely invited for screening at age 65 were included in the analysis. Outcomes assessed included cause of death and a combined outcome of admission or death from cardiovascular disease.

      The Framingham Heart Study cohort

      In the NAAASP-HES-ONS datasets no covariate data was available to adjust findings for baseline cardiovascular risk, whilst in the MASS trial dataset, the only covariate available was age.
      To address this we obtained data from the FHS offspring cohort. The FHS originally recruited 5 209 participants between the ages of 30–62 years in 1948 with the aim of identifying the common factors or characteristics that contribute to cardiovascular disease (CVD).
      • Hoffmann U.
      • Massaro J.M.
      • D’Agostino Sr., R.B.
      • Kathiresan S.
      • Fox C.S.
      • O’Donnell C.J.
      Cardiovascular event prediction and risk reclassification by coronary, aortic, and valvular calcification in the framingham heart study.
      The FHS offspring cohort was initiated in 1971 and included a sample of 5 124 participants from which participants were invited to participate within the multi-detector CT (MDCT) sub-study. Details of the employed imaging protocols and measurement methods have been published elsewhere.
      • Hoffmann U.
      • Massaro J.M.
      • D’Agostino Sr., R.B.
      • Kathiresan S.
      • Fox C.S.
      • O’Donnell C.J.
      Cardiovascular event prediction and risk reclassification by coronary, aortic, and valvular calcification in the framingham heart study.
      A total of 1 418 individuals consented to MDCT of their coronary vasculature as well as for thoracic and abdominal aortic imaging. Aortic diameter measurements were taken in the antero-posterior and the transverse planes 5cm above the aorto-iliac bifurcation, using the outer-to-outer diameter, and were performed between June 2002 and April 2005. The FHS study was approved by the institutional review boards of the Boston University Medical Center and the Massachusetts General Hospital and this retrospective analysis has been approved by the Ethics Review Committee of the University of Leicester (Reference: 5048-as875-cardiovascular sciences). Parameters included within this analysis were current pharmacological therapy and incident rates of cardiovascular events (coronary heart disease, heart failure and cerebrovascular events). Major cardiovascular events were defined as per the American Heart Association Key Data Elements and Definitions for Cardiovascular Endpoint Events in Clinical Trials.
      • Hicks K.A.
      • Tcheng J.E.
      • Bozkurt B.
      • Chaitman B.R.
      • Cutlip D.E.
      • Farb A.
      • et al.
      ACC/AHA key data elements and definitions for cardiovascular endpoint events in clinical trials: a Report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Cardiovascular Endpoints Data Standards).
      This data was recorded at Offspring Exam 9 between April 2011 and March 2014. Individuals who did not have an event were censored at the last known date the participant did not have cardiovascular disease and all women were excluded from analysis. Crude and adjusted analyses were performed and the net reclassification index (NRI) was calculated using previously described methods.
      • Pencina M.J.
      • D’Agostino Sr., R.B.
      • D’Agostino Jr., R.B.
      • Vasan R.S.
      Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond.
      Due to the small number of men with aortic diameters ≥3 cm in this cohort the FHS dataset was analysed using two aortic diameter groups namely ≤24 mm (‘normal’) and ≥25 mm ('abnormal’).

      Statistical analysis

      Crude survival analysis was performed using Cox-proportional hazards modelling without adjustment to produce un-adjusted hazard ratios and confidence intervals. Competing risks survival analysis was then performed with a competing risk defined as an event whose occurrence precludes the primary event of interest. The subdistribution hazard function was also calculated, to assess the effect of covariates on the hazard function as a cumulative incidence function (CIF) which was then plotted.
      • Fine J.P.
      • G R.J.
      A proportional hazards model for the subdistribution of a competing risk.
      This denotes the magnitude of relative change in the hazard function associated with a 1-unit change in aortic diameter. This model assumes that those who experience competing events cannot experience the primary event of interest.
      Competing risk analysis was not possible for the FHS data as the authors did not have individual cause of death data. Therefore, Cox-proportional survival analysis was performed (adjusted for age and maximum aortic diameter) to estimate the hazard function of two aortic diameter groups namely ≤24 mm (normal) and ≥25 mm (abnormal). Hazard Ratios were reported per mm increase in aortic diameter. The analysis was performed using R v3.4.3 (R Foundation, Austria) and Stata v.15 (StataCorp LLC, USA).

      Results

      The Multi-Aneurysm Screening Study

      A total of 33 883 men were invited to screening of which 26 882 attended (79.3%). Of those 1336 had an AAA (4.9%). The mean ± standard deviation (SD) age was 69.3 ± 2.7 years and 68.0 ± 2.8 years for those with and without an AAA respectively. Mean anteroposterior aortic diameter was 3.8 ± 1.2 cm, whilst no measurements were reported for those that were disease free. Median follow-up was 12.8 years (interquartile range 11.0–13.6 years). Cardiovascular mortality was higher in men with AAA (Fig. 1), compared to men without AAA (24.3% versus 13.4%, HR 2.22, 95% confidence interval [CI] 1.97–2.50) after adjustment for competing risks (non-cardiovascular death). Non-cardiovascular mortality was also higher in those men with AAA (36.2% versus 23.6%, HR 1.86, 95% CI 1.69–2.06) and this association remained significant after adjustment for age (Table 1).
      Figure 1
      Figure 1Risk of cardiovascular death in men with (n = 1 336), and without (n = 25 546) abdominal aortic aneurysm (AAA) in the Multi-centre Aneurysm Screening Study (MASS).
      Table 1Number and adjusted hazard ratios (HR) of deaths from cardiovascular and non-cardiovascular causes in men with abdominal aortic aneurysm (AAA) or normal aorta screened for AAA in the Multi-centre Aneurysm Screening Study (MASS)
      Cause of deathDeathsAdjusted HR (95% CI) for age
      HRs for the comparison of deaths in men with AAA compared to men with normal aortae.
      Adjusted HR (95% CI) for competing risk of non-cardiovascular mortality
      HRs for the comparison of deaths in men with AAA compared to men with normal aortae.
      Normal aorta n = 25 546AAA
      AAA defined as aortic diameter 3.0 cm or greater.
      n = 1 336
      Cardiovascular3 423 (13.4)325 (24.3)2.07 (1.85–2.32)2.22 (1.97–2.50)
      Non-cardiovascular6 019 (23.6)483 (36.2)1.78 (1.63-1.96)
      Data are presented as n (%) unless stated otherwise. CI = confidence interval; HR = hazard ratio.
      AAA defined as aortic diameter 3.0 cm or greater.
      HRs for the comparison of deaths in men with AAA compared to men with normal aortae.

      The NHS AAA Screening Programme mortality

      A total of 237 441 men attended the 2013/14 NAAASP screening programme and had linked data available for analysis. Mean ± SD aortic diameter was 1.79 ± 0.35 cm. A diagnosis of AAA was established in 2 634 men (1.1%), whilst 3 228 men (1.4%) had an aortic diameter between 2.5–2.9 cm. After a follow-up period of 2 years, a total of 8961 deaths were observed (3. 7%). There was an increased risk of cardiovascular death for men with an aortic diameter 2.5 cm to 2.9 cm (0.96% versus 0.36%, HR 2.63, 95% CI 1.84–3.76) increasing to 2.9–fold in those with AAA (HR 2.91, 95% CI 2.00–4.25) compared to those with a normal AD, after adjustment for competing risks (Table 2, Fig. 2a). Non-cardiovascular deaths were observed in 0.88% of men with a non-aneurysmal AD compared with 1.49% in the 2.5–2.9 cm range and 2.16% of men with AAA. The commonest causes of death varied by aortic size threshold; malignancy for men with normal AD, cardiovascular causes for men with AAA, whilst the causes varied for those with sub-aneurysmal AD (Fig. 3). There were ten aortic-related mortalities in those diagnosed with AAA, during the follow-up period. Numbers in sub-groups cannot be reported due to NHS Digital data suppression rules.
      Table 2Number of deaths and cardiovascular events in the National Abdominal Aortic Aneurysm Screening Programme (NAAASP) 2013 cohort from cardiovascular (CV) and non-cardiovascular causes by aortic size
      Deaths and CV eventsHR (95% CI)Adjusted HR (95% CI)
      ≤2.4 cm (reference)n = 231 5792.5–2.9 cm

      n = 3 228
      ≥3.0 cm

      n = 2 364
      2.5–2.9 cm≥3.0 cm2.5–2.9 cm
      Deaths
       CV842 (0.36)31(0.96)35 (1.48)2.72 (2.12–3.49)3.52 (2.79–4.45)2.63 (1.84–3.76)
       Non-CV2 045 (0.88)48 (1.49)51 (2.16)1.89 (1.62–2.19)2.42 (2.10-2.78)1.69 (1.27–2.25)
      Events
       CV18 457 (7.97)396 (12.27)365 (15.44)1.74 (1.55–1.96)2.29 (2.05–2.56)1.58 (1.43–1.74)
       Non-CV49 728 (21.47)733 (22.71)644 (27.24)1.24 (1.17–1.32)1.62 (1.53–1.71)1.07 (0.99–1.15)
      Data are presented as n (%) unless stated otherwise, and with crude hazard ratios (HR) and adjusted for competing risks. CI = confidence interval.
      Figure 2
      Figure 2Risk of (A) cardiovascular death or (B) major adverse cardiovascular events (MACE) in men screened within the National Abdominal Aortic Aneurysm Screening Programme (NAAASP) by aortic diameter at screening (>3 cm: n = 2 634, 2.5–2.9 cm: n = 3 228, <2.5 cm: n = 231 579).
      Figure 3
      Figure 3Cumulative proportional causes of death by ICD10 chapter in men screened for abdominal aortic aneurysm (AAA) with (A) normal, (B) sub-aneurysmal, and (C) aneurysmal aortic diameters. Aneurysm related mortality shown separately to other cardiovascular deaths.
      Increases in aortic diameter were associated with an increase in the risk of cardiovascular death (Fig. 4, Supplementary Table 1), particularly for those with only mildly enlarged diameters (2.0–2.49 cm, HR 1.31, 95% CI 1.07–1.60). It should be noted however that overall, the highest cardiovascular risk was observed in those with AAA. Small aortic diameters conferred an equally increased risk of cardiovascular death, with the highest risk (HR 3.04, 95% CI 1.77–5.21) being observed in those with the smallest Ads (0.49–1.24 cm). These findings would suggest a U-shaped association between aortic diameter and cardiovascular mortality risk.
      Figure 4
      Figure 4Risk (hazard ratio [HR] and 95% confidence interval [CI]) of (A) cardiovascular death and (B) cardiovascular events in men screened for abdominal aortic aneurysm (AAA) by aortic size thresholds. Data shown in .

      The NHS AAA Screening Programme major adverse cardiovascular events

      The risk of MACE (Fig. 4b) was higher in those with an AD of 2.5–2.9 cm (HR 1.58, 95% CI 1.43–1.74) and those with an AAA (HR 1.79, 95% CI 1.62–1.79) after adjustment for competing risks. In comparison to average aortic diameter, both higher, and lower, aortic diameter at screening was associated with an increasing MACE risk in a similar manner to the risk of cardiovascular mortality (Supplementary Table 1).

      The Framingham Heart Study cohort

      To investigate the independence of the association between aortic size and cardiovascular morbidity, adjusted for possible confounders, we analysed the data collected for 649 men from the FHS offspring cohort (Supplementary Table 2) with a median follow up of 12.3 years, interquartile range 11.0–14.2 years. A series of baseline parameters, including demographics (age, body mass index), comorbidities (hypertension, diabetes mellitus, baseline biochemistry and smoking history, that may relate to future cardiovascular risk were found to differ between men with a normal aortic diameter and those with a diameter ≥2.5 cm, as detailed in Supplementary Table 2. Precise causes of death were available in the FHS dataset for 23 individuals, 20 of which (87%) died due to a cardiovascular cause.
      Forward conditional Cox proportional hazards modelling (Fig. 5), adjusted for the parameters included in the Framingham cardiovascular risk score together with use of statin therapy and use of antihypertensive agents, was performed to estimate the association between each parameter and future major cardiovascular events. An aortic diameter ≥2.5 cm (HR 1.88 95% CI 1.03–3.44, p < .001) and age (adjusted OR 1.06 95% CI 1.04–1.09, p < .001) were the only parameters that remained significantly associated with risk of future major cardiovascular events during the 12.3 years of follow-up.
      Figure 5
      Figure 5Risk of major cardiovascular events in men from the Framingham offspring cohort comparing men with an infra-renal aortic diameter ≥2.5 cm (n = 48) to those with normal aortic diameters (n = 601).
      Using abdominal aortic diameter ≥2.5 cm as a predictor of cardiovascular risk resulted in the reclassification of 21 individuals previously classified as low or moderate risk based on the Framingham 10 year cardiovascular risk score, to high-risk. The associated NRI was 0.37 and the parameters applied for the NRI were 0–10%, 10–20%, >20% (for the Framingham 10 year cardiovascular risk score). The addition of baseline abdominal aortic diameter to the Framingham 10 year risk score improved the C-statistic for major cardiovascular event prediction during follow-up from 0.61 (95% CI 0.51–0.79) to 0.677 (95% CI 0.61–0.79), p < .001.

      Discussion

      We have shown that cardiovascular risk in men with AAA is high, that despite considerable improvements in cardiovascular risk management the cardiovascular risk in men with AAA remains under-treated, and that irrespective of the diagnosis of AAA, men with extremes of aortic diameter are at high risk of cardiovascular events when compared to those with normal aortic diameters.
      Additionally, aortic diameter is not only an independent marker of risk, but may also add to the sensitivity of an existing cardiovascular risk score, potentially extending the clinical utility of data obtained from AAA screening.
      Patients with AAA are widely recognized to be at a higher cardiovascular risk than the general population.
      • Bath M.F.
      • Gokani V.J.
      • Sidloff D.A.
      • Jones L.R.
      • Choke E.
      • Sayers R.D.
      • et al.
      Systematic review of cardiovascular disease and cardiovascular death in patients with a small abdominal aortic aneurysm.
      Duncan and colleagues described long term outcomes of men screened for AAA in Scotland finding that the most common cause of death in those with AAA was cardiovascular disease (adjusted HR 1.90, 95% CI 1.17–3.08).
      • Duncan J.L.
      • Harrild K.A.
      • Iversen L.
      • Lee A.J.
      • Godden D.J.
      Long term outcomes in men screened for abdominal aortic aneurysm: prospective cohort study.
      These findings were in agreement with several previous large cohort studies.
      • Norman P.E.
      • Muller J.
      • Golledge J.
      The cardiovascular and prognostic significance of the infrarenal aortic diameter.
      ,
      • Norman P.
      • Le M.
      • Pearce C.
      • Jamrozik K.
      Infrarenal aortic diameter predicts all-cause mortality.
      Data from the United Kingdom Small Aneurysm Trial revealed that for every 8 mm increase in aneurysm diameter, the hazard ratio for cardiovascular mortality increased by 1.34 (95% CI 1.01–1.79).
      • Brady A.R.
      • Fowkes F.G.
      • Thompson S.G.
      • Powell J.T.
      Aortic aneurysm diameter and risk of cardiovascular mortality.
      ,
      • Powell J.T.
      • Brown L.C.
      • Forbes J.F.
      • Fowkes F.G.
      • Greenhalgh R.M.
      • Ruckley C.V.
      • et al.
      Final 12-year follow-up of surgery versus surveillance in the UK Small Aneurysm Trial.
      Furthermore, the 10 year follow up data from the Cardiovascular Health Study (1992–1993) examined 1 953 men by ultrasound and demonstrated that men with AAA at baseline had significantly more cardiovascular events then men without (HR 2.31, 95% CI 1.63–3.28) and that risk increased with increasing aortic diameter.
      • Freiberg M.S.
      • Arnold A.M.
      • Newman A.B.
      • Edwards M.S.
      • Kraemer K.L.
      • Kuller L.H.
      Abdominal aortic aneurysms, increasing infrarenal aortic diameter, and risk of total mortality and incident cardiovascular disease events: 10-year follow-up data from the Cardiovascular Health Study.
      Our data suggests that smaller aortic diameters are an indicator of future cardiovascular risk and this is consistent with previous studies. The Tromsø study measured aortic diameter at baseline and at 10 years and demonstrated the risk of cardiovascular mortality increases with both higher (27–29 mm aorta, 1.86; 95% CI 1.24–2.78) and lower aortic diameters (<18 mm aorta, 1.18; 95% CI 0.82–1.69).
      • Singh K.
      • Bonaa K.H.
      • Jacobsen B.K.
      • Bjork L.
      • Solberg S.
      Prevalence of and risk factors for abdominal aortic aneurysms in a population-based study: the Tromso Study.
      Oliver-Williams et al. recently published data analysing the risk of risk of ruptured AAA (rAAA) in men under surveillance in the NAAASP.
      • Oliver-Williams C.
      • Sweeting M.J.
      • Jacomelli J.
      • Summers L.
      • Stevenson A.
      • Lees T.
      • et al.
      Safety of men with small and medium abdominal aortic aneurysms under surveillance in the NAAASP.
      Their analysis suggested that in men with AAA at 2.5 years follow up (similar to the data presented here), cancer was the leading cause of death (31.2%) over cardiovascular causes (26.3%), an observation contrary to our analysis of the 2013/14 NAAASP cohort data where death due to cardiovascular causes was more common (45.0% vs 34.4%).
      Both datasets demonstrate the high risk of cardiovascular mortality in men with AAAs however. The differences in proportional mortality for cardiovascular and neoplastic causes may be due to methodological differences. In the study by Oliver-Williams et al. cause of death was reported by local screening programme staff as opposed to through the analysis of national mortality records as in our study. The relatively small number of deaths in the NAAASP AAA cohort analysed in our study (131 compared to 980) may have contributed to the differences in cause of death observations.
      The historical datasets utilised within this study demonstrate that the observations noted in the 2013 NAAASP cohort remain consistent over time and within different populations. Part of this may simply be better recognition of the risks in men with AAA and peripheral arterial disease as both have well established associations with cardiovascular risk through shared risk factors. Enlarging aortic diameter has also been hypothesised to contribute to increasing cardiovascular risk through the effect of a biologically active thrombus circulating inflammatory cytokines.
      • Fontaine V.
      • Jacob M.P.
      • Houard X.
      • Rossignol P.
      • Plissonnier D.
      • Angles-Cano E.
      • et al.
      Involvement of the mural thrombus as a site of protease release and activation in human aortic aneurysms.
      • Sidloff D.A.
      • Stather P.W.
      • Choke E.
      • Bown M.J.
      • Sayers R.D.
      A systematic review and meta-analysis of the association between markers of hemostasis and abdominal aortic aneurysm presence and size.
      • Yamazumi K.
      • Ojiro M.
      • Okumura H.
      • Aikou T.
      An activated state of blood coagulation and fibrinolysis in patients with abdominal aortic aneurysm.

      Limitations

      The UK datasets used in this study rely on observational data from the HES/ONS datasets and for men to have attended their AAA screening appointment. It is well established that men from lower socio-economic backgrounds and minority cultural backgrounds are less likely to attend AAA screening and also have a higher risk of cardiovascular disease. These data therefore may be prone to a selection bias as men without aortic diameter measurements could not be analysed within this study.
      The prevalence of AAA in the MASS trial (4.9%) is significantly higher than that observed in the NAAASP and in fact higher compared to other screening programmes. This is due to the historical nature of the dataset, at a time when best medical therapy was not standardized, and thus extrapolations to todays practice must be made with caution.
      Two imaging modalities were employed to obtain aortic diameter measurements, as well as two different measurement techniques. In the MASS and NAAASP datasets, aortic ultrasonography was the imaging modality used and measurements were obtained using the inner-to-inner diameter technique, whilst computed tomography was the imaging modality in the FHS, and the measurements were obtained using the outer-to-outer diameter techniques. Discrepancies are well known between imaging modalities, and measurement techniques, therefore care is needed when extrapolating findings across imaging modalities and measurement techniques.
      All datasets used within this study may be prone to misclassification bias where a patent is recorded as having an event incorrectly and/or measurement error however the HES/ONS datasets are subjected to a range of validation rules to ensure that records are valid and in the correct format. Furthermore, although the annual refresh of linked HES/ONS dataset contains all registered deaths/admissions in the calendar year and has been subject to quality assurance, due to the time taken for submission and processing there is a lag between death (approximately 1 month) and admissions (approximately 6 months) meaning some patients may have an event not recognized on the dataset until the next annual refresh. It is also possible, that the cause of death may have been misclassified, as not all deaths are followed-up by a post-mortem examination to ascertain the true cause of death. The validity of using these datasets for research has been examined in previous communications and has been found to be satisfactory.
      • Kivimäki M.
      • Batty G.D.
      • Singh-Manoux A.
      • Britton A.
      • Brunner E.J.
      • Shipley M.J.
      Validity of cardiovascular disease event ascertainment using linkage to UK Hospital Records.
      • Thorn J.C.
      • Turner E.
      • Hounsome L.
      • Walsh E.
      • Donovan J.L.
      • Verne J.
      • et al.
      Validation of the hospital episode statistics outpatient dataset in England.
      • Gallagher A.M.
      • Dedman D.
      • Padmanabhan S.
      • Leufkens H.G.M.
      • Vries F de
      The accuracy of date of death recording in the Clinical Practice Research Datalink GOLD database in England compared with the Office for National Statistics death registrations.
      Lastly, this analysis was based on data available in HES-ONS as of March 2017. The HES-ONS dataset has since been updated but was unavailable for analysis under our current data sharing agreement with NHS Digital due to changes in data access agreements.
      In the MASS dataset the ‘normal’ aortic diameter group likely contained men with diameters between 2.5–2.9 cm who we have clearly demonstrated have an elevated risk. This would almost certainly add to the risk of the baseline group slightly blunting the differences in risk seen in those with AAA, however, the MASS study is the largest comparable dataset with a very good length of follow up in order to compare our contemporary data. HES is the administrative data set for the NHS in England. These data sets are not primarily designed for research, and therefore the quality of the data within this observational study may be subject to limited accuracy. However, individual follow up of approximately 300 000 men a year for life would be prohibitively costly and HES/ONS follow up is a well-established methodology that has been demonstrated to reliably guide practice.

      Conclusions

      Aortic diameters, both larger and smaller than the population average, are associated with cardiovascular events morbidity and mortality, in men screened for abdominal aortic aneurysm disease. Abnormal aortic diameter may be useful as a surrogate marker of cardiovascular risk and this data could be better utilised by health services.

      Conflict of interest

      None.

      Funding

      None.

      Acknowledgements

      The authors are grateful to the MASS research team, in particular Dr Lu Gao and Prof Simon Thompson, for providing access to anonymised individual records from their study. This work was supported by the NIHR Leicester Biomedical Research Centre . The views and opinions expressed herein are those of the authors and do not necessarily reflect those of the NIHR, NHS, the Department of Health or the NHS AAA Screening programme. NHS Digital do not endorse or bear any responsibility for the analysis and/or interpretation of the data utilised within this manuscript. Data was obtained from the Framingham Heart Study of the National Heart Lung and Blood Institute of the National Institutes of Health and Boston University School of Medicine. The Framingham Heart Study has been funded in whole or in part with Federal funds from the National Heart, Lung, and Blood Institute , National Institutes of Health , Department of Health and Human Services , under Contract No. 75N92019D00031.

      Appendix ASupplementary data

      The following are the Supplementary data to this article:

      References

        • Wanhainen A.
        • Mani K.
        • de Borst G.J.
        The most important news in the new ESVS 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysm.
        J Cardiovasc Surg (Torino). 2019; 60: 485-489
        • Wanhainen A.
        • Verzini F.
        • Van Herzeele I.
        • Allaire E.
        • Bown M.
        • Cohnert T.
        • et al.
        Editor’s Choice - European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
        Eur J Vasc Endovasc Surg. 2019; 57: 8-93
        • Brady A.R.
        • Fowkes F.G.
        • Thompson S.G.
        • Powell J.T.
        Aortic aneurysm diameter and risk of cardiovascular mortality.
        Arterioscler Thromb Vasc Biol. 2001; 21: 1203-1207
        • Dabare D.
        • Lo T.T.H.
        • McCormack D.J.
        • Kung V.W.S.
        What is the role of screening in the management of abdominal aortic aneurysms?.
        Interact Cardiovasc Thorac Surg. 2012; 14: 399-405
        • Rughani G.
        • Robertson L.
        • Clarke M.
        Medical treatment for small abdominal aortic aneurysms.
        Cochrane Database Syst Rev. 2012; 9: CD009536
        • Jacomelli J.
        • Summers L.
        • Stevenson A.
        • Lees T.
        • Earnshaw J.J.
        Impact of the first 5 years of a national abdominal aortic aneurysm screening programme.
        Br J Surg. 2016; 103: 1125-1131
        • Freiberg M.S.
        • Arnold A.M.
        • Newman A.B.
        • Edwards M.S.
        • Kraemer K.L.
        • Kuller L.H.
        Abdominal aortic aneurysms, increasing infrarenal aortic diameter, and risk of total mortality and incident cardiovascular disease events: 10-year follow-up data from the Cardiovascular Health Study.
        Circulation. 2008; 117: 1010-1017
        • Powell J.T.
        • Brown L.C.
        • Forbes J.F.
        • Fowkes F.G.
        • Greenhalgh R.M.
        • Ruckley C.V.
        • et al.
        Final 12-year follow-up of surgery versus surveillance in the UK Small Aneurysm Trial.
        Br J Surg. 2007; 94: 702-708
        • Glover M.J.
        • Kim L.G.
        • Sweeting M.J.
        • Thompson S.G.
        • Buxton M.J.
        Cost-effectiveness of the National Health Service abdominal aortic aneurysm screening programme in England.
        Br J Surg. 2014; 101: 976-982
        • Thompson S.G.
        • Ashton H.A.
        • Gao L.
        • Buxton M.J.
        • Scott R.A.
        Final follow-up of the Multicentre Aneurysm Screening Study (MASS) randomized trial of abdominal aortic aneurysm screening.
        Br J Surg. 2012; 99: 1649-1656
      1. NHS Abdominal Aortic Aneurysm Screening Programme. NAAASP Data 2014-2015. https://www.gov.uk/government/publications/abdominal-aortic-aneurysm-screening-2014-to-2015-data. [Accessed 1 January 2016].

      2. Framingham Heart Study. Research results. https://www.framinghamheartstudy.org/fhs-for-researchers/selected-research-results/. [Accessed 1 January 2016].

        • van Staa T.P.
        • Gulliford M.
        • Ng E.S.
        • Goldacre B.
        • Smeeth L.
        Prediction of cardiovascular risk using Framingham, ASSIGN and QRISK2: how well do they predict individual rather than population risk?.
        PLoS One. 2014; 9e106455
        • NHS Digital
        Health Episode Statistics.
        2018
      3. Office for National Statistics. Births and Deaths. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths. [Accessed 1 January 2016].

        • Hoffmann U.
        • Massaro J.M.
        • D’Agostino Sr., R.B.
        • Kathiresan S.
        • Fox C.S.
        • O’Donnell C.J.
        Cardiovascular event prediction and risk reclassification by coronary, aortic, and valvular calcification in the framingham heart study.
        J Am Heart Assoc. 2016; 5: e003144
        • Hicks K.A.
        • Tcheng J.E.
        • Bozkurt B.
        • Chaitman B.R.
        • Cutlip D.E.
        • Farb A.
        • et al.
        ACC/AHA key data elements and definitions for cardiovascular endpoint events in clinical trials: a Report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Cardiovascular Endpoints Data Standards).
        Circulation 2015. 2014; 132: 302-361
        • Pencina M.J.
        • D’Agostino Sr., R.B.
        • D’Agostino Jr., R.B.
        • Vasan R.S.
        Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond.
        Stat Med. 2008; 27: 157-172
        • Fine J.P.
        • G R.J.
        A proportional hazards model for the subdistribution of a competing risk.
        J Am Statis Assoc. 1999; 94: 496-509
        • Bath M.F.
        • Gokani V.J.
        • Sidloff D.A.
        • Jones L.R.
        • Choke E.
        • Sayers R.D.
        • et al.
        Systematic review of cardiovascular disease and cardiovascular death in patients with a small abdominal aortic aneurysm.
        Br J Surg. 2015; 102: 866-872
        • Duncan J.L.
        • Harrild K.A.
        • Iversen L.
        • Lee A.J.
        • Godden D.J.
        Long term outcomes in men screened for abdominal aortic aneurysm: prospective cohort study.
        Bmj. 2012; 344: e2958
        • Norman P.E.
        • Muller J.
        • Golledge J.
        The cardiovascular and prognostic significance of the infrarenal aortic diameter.
        J Vasc Surg. 2011; 54: 1817-1820
        • Norman P.
        • Le M.
        • Pearce C.
        • Jamrozik K.
        Infrarenal aortic diameter predicts all-cause mortality.
        Arterioscler Thromb Vasc Biol. 2004; 24: 1278-1282
        • Singh K.
        • Bonaa K.H.
        • Jacobsen B.K.
        • Bjork L.
        • Solberg S.
        Prevalence of and risk factors for abdominal aortic aneurysms in a population-based study: the Tromso Study.
        Am J Epidemiol. 2001; 154: 236-244
        • Oliver-Williams C.
        • Sweeting M.J.
        • Jacomelli J.
        • Summers L.
        • Stevenson A.
        • Lees T.
        • et al.
        Safety of men with small and medium abdominal aortic aneurysms under surveillance in the NAAASP.
        Circulation. 2019; 139: 1371-1380
        • Fontaine V.
        • Jacob M.P.
        • Houard X.
        • Rossignol P.
        • Plissonnier D.
        • Angles-Cano E.
        • et al.
        Involvement of the mural thrombus as a site of protease release and activation in human aortic aneurysms.
        Am J Pathol. 2002; 161: 1701-1710
        • Sidloff D.A.
        • Stather P.W.
        • Choke E.
        • Bown M.J.
        • Sayers R.D.
        A systematic review and meta-analysis of the association between markers of hemostasis and abdominal aortic aneurysm presence and size.
        J Vasc Surg. 2014; 59: 528-535
        • Yamazumi K.
        • Ojiro M.
        • Okumura H.
        • Aikou T.
        An activated state of blood coagulation and fibrinolysis in patients with abdominal aortic aneurysm.
        Am J Surg. 1998; 175: 297-301
        • Kivimäki M.
        • Batty G.D.
        • Singh-Manoux A.
        • Britton A.
        • Brunner E.J.
        • Shipley M.J.
        Validity of cardiovascular disease event ascertainment using linkage to UK Hospital Records.
        Epidemiology. 2017; 28: 735-739
        • Thorn J.C.
        • Turner E.
        • Hounsome L.
        • Walsh E.
        • Donovan J.L.
        • Verne J.
        • et al.
        Validation of the hospital episode statistics outpatient dataset in England.
        PharmacoEconomics. 2016; 34: 161-168
        • Gallagher A.M.
        • Dedman D.
        • Padmanabhan S.
        • Leufkens H.G.M.
        • Vries F de
        The accuracy of date of death recording in the Clinical Practice Research Datalink GOLD database in England compared with the Office for National Statistics death registrations.
        Pharmacoepidemiology Drug Safety. 2019; 28: 563-569

      Comments

      Commenting Guidelines

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

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