Advertisement

Aneurysm Geometry Analyzed by the Novel Three-Dimensional Tomographic Ultrasound Relates to Abdominal Aortic Aneurysm Growth

  • Maria Khan
    Affiliations
    Academic Surgery Unit, Division of Cardiovascular Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Wythenshawe Hospital, Manchester, UK
    Search for articles by this author
  • Steven Rogers
    Correspondence
    Correspondence to: Dr. Steven Rogers, Vascular Studies Unit, Wythenshawe Hospital, Southmoor Road, Manchester, UK M23 9LT
    Affiliations
    Academic Surgery Unit, Division of Cardiovascular Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Wythenshawe Hospital, Manchester, UK
    Search for articles by this author
  • Joao Carreira
    Affiliations
    Independent Vascular Services Ltd, Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester, UK
    Search for articles by this author
  • Jonathan Ghosh
    Affiliations
    Department of Vascular and Endovascular Surgery, Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester Academic Health Science Centre, Manchester, UK
    Search for articles by this author
  • Charles McCollum
    Affiliations
    Academic Surgery Unit, Division of Cardiovascular Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Wythenshawe Hospital, Manchester, UK
    Search for articles by this author
Open AccessPublished:June 03, 2022DOI:https://doi.org/10.1016/j.avsg.2022.05.017

      Background

      Abdominal aortic aneurysms (AAAs) are increasingly screen-detected and many small aneurysms enter surveillance. Computed tomography identifies characteristics that can predict subsequent AAA growth but ionizing radiation and nephrotoxic contrast disadvantage its use in surveillance. We investigated whether duplex and 3-dimensional tomographic ultrasound identified features associated with AAA growth in patients on AAA surveillance.

      Methods

      Duplex and three-dimensional tomographic ultrasound imaging was performed independently by 2 vascular scientists in 128 AAA surveillance patients who all had AAA growth measured over at least 2 years. Diameter, cross-sectional area, length, volume, wall thickness/volume, and intraluminal thrombus were measured. Pulsatility using maximum systolic and minimum diastolic diameters corrected for diameter and distensibility (consisting of strain and stiffness) were also calculated.

      Results

      AAA growth rate correlated with AAA diameter (r 0.43), volume (r 0.46), and cross-sectional area (r 0.42) (P < 0.01). Measuring wall thickness was inaccurate, but wall volume (corrected for AAA volume) inversely related to growth rate (r −0.43, P < 0.01). On a multivariate analysis, diameter and wall volume (r2adjusted 0.22, P < 0.01) improved prediction of growth rate compared with diameter alone (r2adjusted 0.18, P < 0.01). Intraluminal thrombus volume, strain distensibility, and elastic distensibility were not significantly associated with AAA growth.

      Conclusions

      AAA growth most strongly related to AAA volume and inversely to wall volume. AAA volume and wall volume may prove useful in the prediction of AAA growth rates.

      Introduction

      More than 14,000 small (3.0 to 5.4 cm) abdominal aortic aneurysms (AAAs) enter surveillance in the UK National AAA Screening Programme (NAAASP) each year.
      If we could predict the future growth rate of AAAs, we could refine surveillance intervals and possibly introduce personalized indications for repair.
      Our current assessment of risk is based on AAA diameter alone with a diameter of 5.5 cm being the indication for surgery; however, AAAs smaller than this may rupture.
      • Nicholls S.C.
      • Gardner J.B.
      • Meissner M.H.
      • et al.
      Rupture in small abdominal aortic aneurysms.
      • Thompson S.G.
      • Ashton H.A.
      • Gao L.
      • et al.
      MASS Group
      Final follow-up of the Multicentre Aneurysm Screening Study (MASS) randomized trial of abdominal aortic aneurysm screening.
      • Kontopodis N.
      • Pantidis D.
      • Dedes A.
      • et al.
      The - Not So - Solid 5.5 cm threshold for abdominal aortic aneurysm repair: facts, misinterpretations, and future directions.
      There is a wide individual variation in AAA growth rates and rupture risk; detecting unstable AAAs earlier would enable prompt treatment.
      • Brown L.C.
      • Powell J.T.
      Risk factors for aneurysm rupture in patients kept under ultrasound surveillance. UK Small Aneurysm Trial Participants.
      • Brady A.R.
      • Thompson S.G.
      • Fowkes F.G.
      • et al.
      UK Small Aneurysm Trial Participants
      Abdominal aortic aneurysm expansion: risk factors and time intervals for surveillance.
      • Vega de Ceniga M.
      • Gomez R.
      • Estallo L.
      • et al.
      Growth rate and associated factors in small abdominal aortic aneurysms.
      • Vega de Ceniga M.
      • Esteban M.
      • Barba A.
      • et al.
      Assessment of biomarkers and predictive model for short-term prospective abdominal aortic aneurysm growth-A pilot study.
      Other features of AAA geometry including AAA volume, area, length, intra-luminal thrombus (ILT), and wall characteristics such as thickness will likely influence AAA growth rates.
      • Parr A.
      • Jayaratne C.
      • Buttner P.
      • et al.
      Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography.
      • Ghulam Q.M.
      • Bredahl K.K.
      • Lönn L.
      • et al.
      Follow-up on small abdominal aortic aneurysms using three dimensional ultrasound: volume versus diameter.
      • Arko F.R.
      • Murphy E.H.
      • Davis C.M.
      • et al.
      Dynamic geometry and wall thickness of the aortic neck of abdominal aortic aneurysms with intravascular ultrasonography.
      • Shang E.K.
      • Nathan D.P.
      • Woo E.Y.
      • et al.
      Local wall thickness in finite element models improves prediction of abdominal aortic aneurysm growth.
      • Sacks M.S.
      • Vorp D.A.
      • Raghavan M.L.
      • et al.
      In vivo three-dimensional surface geometry of abdominal aortic aneurysms.
      • Wilson K.
      • Whyman M.
      • Hoskins P.
      • et al.
      The relationship between abdominal aortic aneurysm wall compliance, maximum diameter and growth rate.
      • Molacek J.
      • Baxa J.
      • Houdek K.
      • et al.
      Assessment of abdominal aortic aneurysm wall distensibility with electrocardiography-gated computed tomography.
      • Long A.
      • Rouet L.
      • Bissery A.
      • et al.
      Compliance of abdominal aortic aneurysms evaluated by tissue Doppler imaging: correlation with aneurysm size.
      • Ravhon R.
      • Adam D.
      • Zelmanovitch L.
      Validation of ultrasonic image boundary recognition in abdominal aortic aneurysm.
      • Leotta D.F.
      • Paun M.
      • Beach K.W.
      • et al.
      Measurement of abdominal aortic aneurysms with three-dimensional ultrasound imaging: preliminary report.
      • Golledge J.
      • Wolanski P.
      • Parr A.
      • et al.
      Measurement and determinants of infrarenal aortic thrombus volume.
      • Brunner-Ziegler S.
      • Hammer A.
      • Seidinger D.
      • et al.
      The role of intraluminal thrombus formation for expansion of abdominal aortic aneurysms.
      • Behr-Rasmussen C.
      • Grondal N.
      • Bramsen M.B.
      • et al.
      Mural thrombus and the progression of abdominal aortic aneurysms: a large population-based prospective cohort study.
      ILT has also been reported to be associated with a minor increase in growth rates.
      • Golledge J.
      • Wolanski P.
      • Parr A.
      • et al.
      Measurement and determinants of infrarenal aortic thrombus volume.
      • Brunner-Ziegler S.
      • Hammer A.
      • Seidinger D.
      • et al.
      The role of intraluminal thrombus formation for expansion of abdominal aortic aneurysms.
      • Behr-Rasmussen C.
      • Grondal N.
      • Bramsen M.B.
      • et al.
      Mural thrombus and the progression of abdominal aortic aneurysms: a large population-based prospective cohort study.
      Both ILT volume and thickness can be more accurately measured by 3-dimensional (3D) ultrasound (US) than by standard duplex.
      • Ghulam Q.M.
      • Bredahl K.
      • Sandholt B.
      • et al.
      Contrast enhanced three dimensional ultrasound for intraluminal thrombus assessment in abdominal aortic aneurysms.
      Up to 50% of patients have measurable increases in AAA volume in computed tomography (CT) despite stable diameters.
      • Parr A.
      • Jayaratne C.
      • Buttner P.
      • et al.
      Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography.
      ,
      • Ghulam Q.M.
      • Bredahl K.K.
      • Lönn L.
      • et al.
      Follow-up on small abdominal aortic aneurysms using three dimensional ultrasound: volume versus diameter.
      However, the dose of ionizing radiation and cumulative nephrotoxic consequences of x-ray contrast media make serial CT angiography (CTa) unsuitable for surveillance. Standard US is relatively safe but cannot be used to measure AAA volumes and is inaccurate of measuring AAA wall volume. Tomographic 3D US (tUS) may be a novel and safe way to capture AAA geometry. 3D US systems have correlated closely with CT (r 0.76–0.93, P < 0.01) in the measurement of AAA volume.
      • Arsicot M.
      • Lathelize H.
      • Martinez R.
      • et al.
      Follow-up of aortic stent grafts: comparison of the volumetric analysis of the aneurysm sac by ultrasound and CT.
      ,
      • Batagini N.C.
      • Ventura C.A.P.
      • Raghavan M.L.
      • et al.
      Volumetry and biomechanical parameters detected by 3D and 2D ultrasound in patients with and without an abdominal aortic aneurysm.
      Wall volume would be expected to influence growth but could not be measured reliably using previous technology and has not previously been researched.
      Standard duplex can be used to measure wall distensibility (the capacity to expand as a result of pulse pressure) with acceptable intraobserver variability using equations for pressure strain elastic modulus and stiffness.
      • Wilson K.A.
      • Lee A.J.
      • Hoskins P.R.
      • et al.
      The relationship between aortic wall distensibility and rupture of infrarenal abdominal aortic aneurysm.
      A negative relationship between distensibility and growth but a positive one with rupture risk has been reported previously.
      • Wilson K.
      • Whyman M.
      • Hoskins P.
      • et al.
      The relationship between abdominal aortic aneurysm wall compliance, maximum diameter and growth rate.
      • Molacek J.
      • Baxa J.
      • Houdek K.
      • et al.
      Assessment of abdominal aortic aneurysm wall distensibility with electrocardiography-gated computed tomography.
      • Long A.
      • Rouet L.
      • Bissery A.
      • et al.
      Compliance of abdominal aortic aneurysms evaluated by tissue Doppler imaging: correlation with aneurysm size.
      Pulsatility (the difference in diameter between systole and diastole) may be more clinically meaningful or usable than distensibility. The association of cross-sectional area and growth is poorly studied, but area can be measured by both standard duplex and tUS.
      • Ravhon R.
      • Adam D.
      • Zelmanovitch L.
      Validation of ultrasonic image boundary recognition in abdominal aortic aneurysm.
      ,
      • Leotta D.F.
      • Paun M.
      • Beach K.W.
      • et al.
      Measurement of abdominal aortic aneurysms with three-dimensional ultrasound imaging: preliminary report.
      We explored whether tUS measures of AAA diameter, cross-sectional area, length, volume, wall thickness and volume, distensibility, pulsatility, and ILT volume are associated with the AAA diameter growth rate in patients on surveillance for a minimum of 2 years.

      Materials and Methods

      For this prospective cohort study, consecutive AAA surveillance patients with AAAs of diameter 3.0–7.0 cm, where the anterior-posterior (AP) inner-to-inner (ITI) diameter growth rate had been measured at least annually by Duplex for a period of 2 years, were recruited from a single tertiary centre over 12 months. This meant that our patients were all followed up for the same time period of 2 years but were under surveillance for varying periods of time before recruitment to this study. The growth rate had been calculated as a mean of the differences between AP ITI diameter at the start of the study and at the 2-year follow-up point. The threshold of surgical intervention employed by our institution is 5.5 cm. Any patients with a diameter of more than 5.5 cm in our cohort were unsuitable for surgery but wished to continue surveillance for reassurance. We lost no patients to follow-up due to decision for surgery. We excluded patients with thoracoabdominal, juxtarenal, and iliac aneurysms. A formal power calculation by our statisticians guided us toward the required number of participants (n = 128) to give the study 90% power to detect a correlation of 0.28 or more. This study population also meant that a reasonably robust multivariable regression model could be derived (using the conventional 10:1 rule of the minimum number of study subjects to the number of predictors). A full ethical approval was provided by the National Research Ethics Service (13/NW/0468) and informed consent was obtained from all participants.
      tUS is a portable nonelectrocardiogram gated 3D US system (Piur imaging GmbH, Vienna, Austria) that requires only a single sweep of standard US transducers and uses electromagnetic tracking to produce a 3D image, that is not operator-dependent, in less than 30 sec in near real time. tUS images, like CT, can be rotated, viewed from any angle, and analysed for geometric features including radius of curvature. tUS was connected to a Mindray Resona 7 US system (Mindray, Shenzhen, China) with a C5-1 MHz transducer to capture tUS images. Growth measurements were completed by Duplex alone. tUS measurements were completed in the same sitting as the final Duplex time point.
      In the same sitting, 2 experienced vascular scientists, blinded to each other's results, acquired tUS images by scanning the length of the aorta from the xiphisternum to just below the umbilicus in a single sweep. Vascular scientists are the workforce used in the United Kingdom for performing and interpreting vascular US scans and are broadly similar to registered vascular technicians. Three images were acquired by each vascular scientist to provide multiple acquisitions for analysis and to measure reproducibility. All our patients attended a morning clinic, were asked to eat, drink, and take medication as normal, and were not given bowel preparation as per guidance from the UK National Abdominal Aortic Screen Programme. Each scan took approximately 10 sec adding little additional time to standard duplex surveillance imaging. Supine systolic and diastolic blood pressures were measured using a manual sphygmomanometer after acquisition of the scans.

      Abdominal Aortic Aneurysm Geometries Measured

      Maximum systolic and minimum diastolic ITI AP diameters were measured for each cardiac cycle. AAA diameter, cross-sectional area, volume, length, luminal thrombus (area and volume), and wall thickness plus wall volume were then calculated from the tUS images.
      ImFusion Suite (ImFusion GmbH, Munich, Germany) was used to analyse these tUS images using multiplanar image reconstructions corrected for the angle of the aorta to avoid oblique measurements. There were 2 experienced vascular scientists who were blinded to each other's results. The vascular scientists were assigned as either User 1 or User 2. For the modeling, it was only User1's measurements which were used to ensure consistency. Three scans for each patient (user 1's scans) were reviewed and the scan with the best image quality (that also included the entire AAA) was analysed by a single experienced medical student validated in house. Twenty test patients were analysed before the full study analysis to ensure that the medical student had been trained well and was producing accurate results. These were validated and quality-assured by a senior vascular scientist before analysis of the study data took place.
      Wall thickness is defined as the AP measurement from the outer surface of the adventitia to the inner surface of the intima of the anterior AAA wall at the point of maximal diameter (Fig. 1A). The ILT area (cm2) was calculated by subtracting the cross-sectional area of free-flowing blood from the cross-sectional area of the inner wall of the AAA at its maximum diameter (Fig. 1B).
      Figure thumbnail gr1
      Fig. 1tUS analysis through specialist vascular software. (A) Measurement of anterior and posterior wall thickness as shown by the 2 yellow line segments in the centre of the anterior and posterior walls. (B) Segmentation to calculate the luminal thrombus area where the green lumen is drawn around free flowing blood and the red wall is the inner aneurysmal wall. (C) Segmentation of the AAA wall represented by a green lumen (aneurysm sac area) and red wall.
      To measure length and volume, the limits of the AAA were defined as where the diameter of the aorta exceeded 1.5 times the normal aortic diameter proximally and distally as appropriate. Because scans were analysed by a single person, the same method was used throughout. Where this could not be visualized, usually due to bowel gas, the images were excluded from analysis.
      To calculate volumes, the AAA, AAA wall (Fig. 1C), or ILT (Fig. 1B) were segmented in 1-mm transverse slices throughout the length of the aneurysm as defined above. Wall volume is a 3D measurement tracing the outer adventitial surface and inner intimal surface for the entire circumference of vessel in each 1-mm slice across the entire aneurysm.
      Pulsatility was defined as the difference between the maximum systolic and minimum diastolic diameter. Distensibility consisted of pressure strain elastic modulus (Ep) (subsequently known as strain) and stiffness (β) that was calculated with the following published formulas
      • Wilson K.A.
      • Lee A.J.
      • Hoskins P.R.
      • et al.
      The relationship between aortic wall distensibility and rupture of infrarenal abdominal aortic aneurysm.
      :
      Ep (105Nm2)=133.3(BP systolic  BP diastolic)([Dmax systolicDmax diastolic]/ Dmax diastolic)


      β (arbitary units)=naural logarithm(BP systolic/BP diastolic)([Dmax systolicDmax diastolic]/ Dmax diastolic)


      Statistical Analysis

      When calculating wall and thrombus areas and volumes, these were corrected for the total AAA areas and volumes, respectively. Pulsatility was corrected for systolic diameter. Pearson's (normally distributed data) or Spearman's (non-normally distributed data) correlation coefficients were to assess the relationship between US measures of AAA geometry and growth. Interclass and intraclass correlation coefficient was used for intraobserver and interobserver variability and a multivariate linear regression model with adjusted R-squared values was used for diameter, AAA volume, and wall volume to analyse the independent association with AAA growth. A 5% significance level (P < 0.05) was accepted as statistically significant. R-squared and adjusted R-squared values of 1.0 represent a perfect agreement.

      Results

      Patient Characteristics

      One hundred and twenty eight AAA surveillance patients (31 women, 24.2%) of mean age 77.8 ± 7.6 years were recruited. One hundred and eleven (87%) patients had a history of smoking, 32 (25%) were current smokers, and 79 (61.7%) were exsmokers. Mean ± standard deviation (SD) systolic and diastolic blood pressure was 134 ± 14.4 and 75 ± 10.9 mm Hg, respectively. A history of treatment for hypertension was reported in 103 (80.5%) patients with 109 (85.2%) reporting a history of hyperlipidemia, 23 (18%) had diabetes mellitus, and 20 (15.6%) a significant family history of AAAs (Table I). No relationships were found between these patient characteristics (including systolic [r = −0.04, P = 0.06] and diastolic blood pressure [r = −0.01, P = 0.92]) and AAA growth rate (Table II).
      Table IAAA patient characteristics
      Patient characteristicNumber (%)/Mean ± SD
      Mean age (years)77.79 ± 7.59
      Male97 (75.8%)
      Female31 (24.2%)
      History of smoking111 (86.7%)
      Current smokers32 (25.0%)
      Exsmokers79 (61.7%)
      Mean systolic blood pressure (mmHg)134.2 ± 14.44
      Mean diastolic blood pressure (mmHg)74.87 ± 10.88
      History of hypertension103 (80.5%)
      History of hyperlipidaemia109 (85.2%)
      History of diabetes mellitus23 (18.0%)
      Type 1 diabetic1 (0.8%)
      Type 2 diabetic18 (14.1%)
      Family history of AAA20 (15.6%)
      Table IIAAA growth rate was unrelated to patient characteristics
      Patient characteristicMean ± SDP value
      Gender0.86
       Male0.20 ± 0.02
       Female0.19 ± 0.04
      Smoking History0.69
       Yes0.20 ± 0.02
       No0.18 ± 0.04
      Hypertension History0.57
       Yes0.19 ± 0.02
       No0.22 ± 0.04
      Hyperlipidaemia History0.43
       Yes0.19 ± 0.02
       No0.23 ± 0.05
      Diabetes mellitus History0.54
       Yes0.17 ± 0.04
       No0.20 ± 0.02
      Family history of AAA0.23
       Yes0.15 ± 0.04
       No0.20 ± 0.02
      Differences in the growth rate were analysed by 2-sample t-test.
      US image quality was poor due to bowel gas and obesity in 43 patients for AAA volume, wall volume, length, and ILT volume; 85 patients were analysed in this study for these variables only. Mean ± SD growth in AAA diameter over the 2 years before tUS imaging was 0.20 ± 0.18 cm/year where 109/128 of our patients had shown growth. We had no ruptures of AAAs under surveillance during our study period.

      Diameter, Cross-Sectional Area, and Length

      Mean ± SD AAA diameter, cross-sectional area, and length were 4.57 ± 0.81 cm, 16.7 ± 6.71 cm2, and 5.4 ± 1.94 cm, respectively. Interobserver agreement for systolic diameter was excellent (r 0.97, P < 0.01) with minimal difference and narrow confidence intervals (bias −0.06 ± 0.20 cm [95% CI −0.45 to 0.34]). Interobserver agreement for diastolic diameter was also excellent (r 0.97, P < 0.01) with minimal difference and narrow confidence intervals (bias 0.13 ± 0.21 cm [95% CI −0.27 to 0.54]). This is less than the nationally accepted interobserver variability of 0.3 cm for standard US (Table III).
      Intraobserver agreement for systolic diameter was excellent at (r 0.99, P < 0.01) (bias −0.01 ± 0.12 cm [95% CI −0.24 to 0.23]) and for diastolic diameter was (r 0.98, P < 0.01) (bias −0.22 ± 0.16 [95% CI −0.53 to 0.08]) (Table III). Intraobserver agreement for cross-sectional area was excellent (r 0.86, P < 0.01) (bias 0.94 ± 3.37 cm2 [95% CI −5.66 to 7.55]) and good for length (r 0.76, P 0.002) (bias 1.21 ± 1.36 cm [95% CI −1.45 to 3.86]). AAA growth correlated significantly with diameter (r 0.43, P < 0.01) and cross-sectional area (r 0.42, P < 0.01) (Fig. 2A and 2B). Length correlated poorly with AAA growth (r 0.28, P 0.01).
      Table IIIIntraobserver and interobserver variability for ultrasound characteristics
      Ultrasound characteristicr valueP valueMean difference ± SD
      Diameter (cm)
       ITI systolic (intraobserver)0.99<0.010.06 ± 0.12
       ITI diastolic (intraobserver)0.97<0.010.22 ± 0.16
       ITI systolic (interobserver)0.97<0.01−0.06 ± 0.20
       ITI diastolic (interobserver)0.97<0.010.13 ± 0.21
      AAA area (cm2)0.86<0.010.94 ± 3.37
      AAA volume (cm3)0.92<0.018.57 ± 22.27
      AAA length (cm)0.760.0021.21 ± 1.36
      Wall thickness (mm)
       Anterior0.240.200.11 ± 0.06
       Posterior0.180.341.01 ± 1.19
      Wall volume (cm3)0.680.018.47 ± 15.11
      ILT area (cm2)0.86<0.010.24 ± 4.18
      ILT volume (cm3)0.98<0.01−8.50 ± 10.32
      Figure thumbnail gr2
      Fig. 2AAA growth rate was correlated against (A) AAA diameter, (B) AAA cross-sectional area, (C) AAA volume, and (D) wall volume (corrected for aneurysm volume). Increasing diameter and obviously cross-sectional area and volume all correlated significantly with AAA growth rate over 2 years, this correlation being closest with AAA volume. AAA wall volume, as a measure of wall thickness, inversely correlated indicating that greater AAA wall thickness was associated with slower AAA growth.

      Abdominal Aortic Aneurysm Volume

      Mean ± SD AAA volume was 73.3 ± 29.7 cm3 with an excellent intraobserver agreement and minimal differences (r 0.92, P < 0.01) (bias 8.57 ± 22.3 cm3 [95% CI −35.1 to 52.2]) (Table III). AAA growth correlated more closely with AAA volume than diameter (r 0.46, P < 0.01) (Fig. 2C). Volume correlated closely with cross-sectional area multiplied by length (r 0.97, P < 0.01).

      Abdominal Aortic Aneurysm Wall Volume and Wall Thickness

      Mean ± SD for wall volume (corrected for AAA volume) was 0.39 ± 0.09 cm3. Measuring wall thickness proved unreliable with poor intraobserver agreement (anterior r 0.24, P 0.20, bias 0.11 ± 1.01 mm [95% CI −1.88 to 2.09], posterior r 0.18, P 0.34, bias 0.06 ± 1.19 mm [95% CI −2.28 to 2.40]) (Table III). Wall thickness measured in this way did not relate to AAA growth (anterior [r 0.02, P 0.84] and posterior [r 0.01, P 0.94]). However, measuring wall volume was considerably more reproducible with good intraobserver agreements and minimal differences (r 0.68, P 0.01) (bias 8.47 ± 15.11 cm3 [95% CI −21.14 to 38.09]) (Table III). Wall volume (corrected for AAA volume) inversely related to growth with a correlation coefficient r −0.43 (P < 0.01) (Fig. 2D). AAA wall volume did not relate significantly to pulsatility (r 0.10, P 0.35).

      Intra-luminal Thrombus

      One hundred and one (79%) patients had an ILT occupying more than 5% of the cross-sectional lumen at maximal diameter. Mean ± SD thrombus area (corrected for AAA area) was 0.37 ± 0.29 (ratio). Intraobserver agreement was excellent (r 0.86, P < 0.01) (bias 0.24 ± 4.18 cm2 [95% CI −7.95 to 8.43]) (Table III).
      Mean ± SD thrombus volume (corrected for AAA volume) was 0.39 ± 0.24 (ratio). Intraobserver agreement for the measurement of thrombus volume was again excellent (r 0.98, P < 0.01, bias −8.50 ± 10.32 cm3 [95% CI −28.72 to 11.73]) (Table III).
      AAA growth correlated weakly with thrombus volume (corrected for AAA volume) (r 0.27, P 0.02) and thrombus area (corrected for AAA area) (r 0.27, P 0.002). The volume of thrombus did not influence pulsatility (r −0.08, P 0.49).

      Pulsatility and Distensibility

      Strain distensibility and elastic distensibility both failed to correlate with AAA growth with (r 0.09, P 0.35) and (r 0.09 P 0.32), respectively. Mean ± SD for strain distensibility was 5.2 ± 5.5. Mean ± SD for elastic distensibility was 38.8 ± 41.2.
      Both AP and laterally measured pulsatility (corrected for diameter) failed to influence growth with (r −0.05, P 0.56) and (r 0.10, P 0.27), respectively.

      Multivariate Analysis of Diameter, Abdominal Aortic Aneurysm Volume, and Wall Volume

      Diameter alone (r2adjusted 0.18, P < 0.01) was independently associated with AAA growth, with a marginally stronger association than that for AAA volume (r2adjusted 0.11) and wall volume (r2adjusted 0.14) (P < 0.01) alone. This association was improved by adding wall volume to diameter (r2adjusted 0.22, P < 0.01). However, the addition of AAA volume to wall volume and diameter (r2adjusted 0.21, P < 0.01) weakened the association suggesting that AAA volume and diameter are dependent variables. To further evidence this, the combination of AAA volume and diameter was weaker (r2adjusted 0.17, P < 0.01). The combination of AAA volume and wall volume was stronger than this (r2adjusted 0.19, P < 0.01).

      Discussion

      In this study recruiting patients on AAA surveillance, AAA growth rates most closely relate to AAA volume and inversely to AAA wall volume. Surprisingly, neither distensibility nor pulsatility significantly related to AAA growth; the length of the aneurysm and ILT had no important relationship with AAA growth.
      AAA wall volume measured by tUS proved to be reliable and reproducible, improving confidence in any association between AAA diameter or volume and growth rates. When corrected for overall AAA volume, wall volume is a more accurate measure of wall thickness than trying to place the US cursor on the outer and inner surfaces of the wall, which is clearly prone to a huge observer error. Wall thickness was measured at the point of maximal diameter which may not represent the point of maximal wall growth. An advantage of the tUS protocol is that the scan can be angle-corrected to avoid oblique measurement which can make some wall thickness measurements inaccurate. Therefore, any 2-dimensional measurements, e.g. wall thickness, will be less accurate than 3D measurements, e.g. wall volume. Predictably, higher AAA wall volumes, indicating a thicker AAA wall, were associated with lower AAA growth rates.
      Similar to previous CT-based measures, aneurysm volume measured by tUS may be a more sensitive indicator of aneurysm growth than diameter alone.
      • Parr A.
      • Jayaratne C.
      • Buttner P.
      • et al.
      Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography.
      ,
      • Renapurkar R.D.
      • Setser R.M.
      • O'Donnell T.P.
      • et al.
      Aortic volume as an indicator of disease progression in patients with untreated infrarenal abdominal aneurysm.
      • Kontopodis N.
      • Metaxa E.
      • Papaharilaou Y.
      • et al.
      Value of volume measurements in evaluating abdominal aortic aneurysms growth rate and need for surgical treatment.
      • Kauffmann C.
      • Tang A.
      • Therasse E.
      • et al.
      Measurements and detection of abdominal aortic aneurysm growth: accuracy and reproducibility of a segmentation software.
      On a multivariate analysis, combining AAA volume with diameter did not improve the association with growth rates, obviously as these 2 measures are closely associated. A recent follow-up study reported that 40% of patients demonstrated increasing volumes despite stable diameters over a follow-up of 12 months.
      • Ghulam Q.M.
      • Bredahl K.K.
      • Lönn L.
      • et al.
      Follow-up on small abdominal aortic aneurysms using three dimensional ultrasound: volume versus diameter.
      Distensibility has previously been reported to increase the risk of rupture.
      • Wilson K.A.
      • Lee A.J.
      • Hoskins P.R.
      • et al.
      The relationship between aortic wall distensibility and rupture of infrarenal abdominal aortic aneurysm.
      We could find no significant relationship between distensibility and AAA growth rate over the preceding 2 years. Previously published research, including small sample numbers, reported significant relationships between distensibility and AAA diameter and growth rate.
      • Wilson K.
      • Whyman M.
      • Hoskins P.
      • et al.
      The relationship between abdominal aortic aneurysm wall compliance, maximum diameter and growth rate.
      ,
      • Long A.
      • Rouet L.
      • Bissery A.
      • et al.
      Compliance of abdominal aortic aneurysms evaluated by tissue Doppler imaging: correlation with aneurysm size.
      ,
      • Wilson K.
      • Bradbury A.
      • Whyman M.
      • et al.
      Relationship between abdominal aortic aneurysm wall compliance and clinical outcome: a preliminary analysis.
      Reported correlations between baseline diameter and strain (r 0.27, P < 0.01) or stiffness (r 0.24, P < 0.01) were not likely to be strong enough to be of significant clinical relevance.
      • Wilson K.
      • Bradbury A.
      • Whyman M.
      • et al.
      Relationship between abdominal aortic aneurysm wall compliance and clinical outcome: a preliminary analysis.
      Our study investigated pulsatility (systolic minus diastolic diameter) corrected for aneurysm size but again found that pulsatility did not influence growth rates. Neither ILT nor wall volume had a significant influence on pulsatility.
      As expected, AAA diameter and cross-sectional area correlated in a similar way with AAA growth rates. The length of the aneurysm predictably had little influence on growth but obviously to a lesser extent than cross-sectional area.
      • Renapurkar R.D.
      • Setser R.M.
      • O'Donnell T.P.
      • et al.
      Aortic volume as an indicator of disease progression in patients with untreated infrarenal abdominal aneurysm.
      Previous studies reported weak correlations between ILT area and AAA growth but as they failed to correct for aneurysm volume it was likely that ILT was also a measure of AAA volume.
      • Golledge J.
      • Wolanski P.
      • Parr A.
      • et al.
      Measurement and determinants of infrarenal aortic thrombus volume.
      • Brunner-Ziegler S.
      • Hammer A.
      • Seidinger D.
      • et al.
      The role of intraluminal thrombus formation for expansion of abdominal aortic aneurysms.
      • Behr-Rasmussen C.
      • Grondal N.
      • Bramsen M.B.
      • et al.
      Mural thrombus and the progression of abdominal aortic aneurysms: a large population-based prospective cohort study.
      ,
      • Wolf Y.G.
      • Thomas W.S.
      • Brennan F.J.
      • et al.
      Computed tomography scanning findings associated with rapid expansion of abdominal aortic aneurysms.
      ,
      • Parr A.
      • McCann M.
      • Bradshaw B.
      • et al.
      Thrombus volume is associated with cardiovascular events and aneurysm growth in patients who have abdominal aortic aneurysms.
      It is less clear whether ILT may reduce peak wall stress and possibly the risk of rupture.
      • Wang D.H.
      • Makaroun M.S.
      • Webster M.W.
      • et al.
      Effect of intraluminal thrombus on wall stress in patient-specific models of abdominal aortic aneurysm.
      ILT volume will be a more accurate version of ILT area, which was measured only at the maximum diameter which may not represent the maximum ILT area.
      Close agreement between diameter and volume measurements for other 3D US systems have been reported previously.
      • Arsicot M.
      • Lathelize H.
      • Martinez R.
      • et al.
      Follow-up of aortic stent grafts: comparison of the volumetric analysis of the aneurysm sac by ultrasound and CT.
      ,
      • Batagini N.C.
      • Ventura C.A.P.
      • Raghavan M.L.
      • et al.
      Volumetry and biomechanical parameters detected by 3D and 2D ultrasound in patients with and without an abdominal aortic aneurysm.
      ,
      • Bredahl K.
      • Taudorf M.
      • Long A.
      • et al.
      Three-dimensional ultrasound improves the accuracy of diameter measurement of the residual sac in EVAR patients.
      • Bredahl K.
      • Sandholt B.
      • Lonn L.
      • et al.
      Three-dimensional ultrasound evaluation of small asymptomatic abdominal aortic aneurysms.
      • Causey M.W.
      • Jayaraj A.
      • Leotta D.F.
      • et al.
      Three-dimensional ultrasonography measurements after endovascular aneurysm repair.
      Our study confirms that tUS quickly, accurately, and reproducibly measures AAA volume. The National Health Service tariff charge for a single US (Duplex or tUS) scan is only £50 ($ 70.56, 24/06/2021). Unfortunately, a third of our patients displayed visualization difficulties, leading to incomplete measurement of some parameters. The drawback of any US technique is that adequate images are difficult in obese patients and when bowel gas obscures the AAA. Although this may seem like a high rate, it is the same outcome seen with duplex alone, which is an accepted worldwide imaging modality for AAA surveillance to determine growth. Therefore, we feel that although this drawback is unfortunate, it is satisfactorily accepted in current clinical practice and should not be seen as a barrier to accepting 3D tUS.
      In our study, 3 image scans were reviewed and the one with best quality was selected for analysis which perhaps subjected it to observer bias. At the time of this study, analyzing tUS scans was taking around an hour per aneurysm but training to use ImFusion Suite software took a few hours. This system is in the process of becoming fully automated with the help of artificial intelligence which will make it more applicable to clinical practice.
      No relationships between AAA growth and patient characteristics were seen and this included hypertension, despite stratifying into systolic and diastolic blood pressure. As this was not the primary aim of the article and in consideration of our other negative study results, we propose that in a follow-up study with larger numbers this is investigated further.
      Any novel measurement, such as wall volume, would have to display a stronger effect size to enter in as an important parameter when predicting the individual patient's growth and subsequent surveillance interval. Therefore, we now need to undertake a larger prospective follow-up study to confirm that measuring AAA volume and wall volume can reliably predict future AAA growth rates. As our study had 2 years of follow-up, inevitably, a smaller total in numerical growth in combination with comparably few study subjects is expected. However, because our model predicts growth from multiple variables, it suggests that we may have a sensitive test for even the slightest of growth. To confirm this, a larger prospective study with a period of follow-up of at least 5 years and observer agreement by qualified individuals is needed.

      Conclusion

      tUS is a relatively practical, quick, safe, reliable, and accurate way to measure AAA geometry. Aneurysm growth is most strongly related to AAA volume and inversely to wall volume, a more reliable way to measure wall thickness. A surveillance programme that incorporates aneurysm volume and wall volume rather than just diameter may better inform surveillance intervals and surgical decisions. Further research will bring us closer to achieving this.
      We thank Dr. Julie Morris and Mr. Phillip Foden for statistical advice. The staff of IVS Ltd. helped with data collection.

      References

      1. Abdominal aortic aneurysm screening: 2016 to 2017 data. Public Health England, England, United Kingdom2017
        • Nicholls S.C.
        • Gardner J.B.
        • Meissner M.H.
        • et al.
        Rupture in small abdominal aortic aneurysms.
        J Vasc Surg. 1998; 28: 884-888
        • Thompson S.G.
        • Ashton H.A.
        • Gao L.
        • et al.
        • MASS Group
        Final follow-up of the Multicentre Aneurysm Screening Study (MASS) randomized trial of abdominal aortic aneurysm screening.
        Br J Surg. 2012; 99: 1649-1656
        • Kontopodis N.
        • Pantidis D.
        • Dedes A.
        • et al.
        The - Not So - Solid 5.5 cm threshold for abdominal aortic aneurysm repair: facts, misinterpretations, and future directions.
        Front Surg. 2016; 3: 1
        • Brown L.C.
        • Powell J.T.
        Risk factors for aneurysm rupture in patients kept under ultrasound surveillance. UK Small Aneurysm Trial Participants.
        Ann Surg. 1999; 230: 289-296
        • Brady A.R.
        • Thompson S.G.
        • Fowkes F.G.
        • et al.
        • UK Small Aneurysm Trial Participants
        Abdominal aortic aneurysm expansion: risk factors and time intervals for surveillance.
        Circulation. 2004; 110: 16-21
        • Vega de Ceniga M.
        • Gomez R.
        • Estallo L.
        • et al.
        Growth rate and associated factors in small abdominal aortic aneurysms.
        Eur J Vasc Endovasc Surg. 2006; 31: 231-236
        • Vega de Ceniga M.
        • Esteban M.
        • Barba A.
        • et al.
        Assessment of biomarkers and predictive model for short-term prospective abdominal aortic aneurysm growth-A pilot study.
        Ann Vasc Surg. 2014; 28: 1642-1648
        • Parr A.
        • Jayaratne C.
        • Buttner P.
        • et al.
        Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography.
        Eur J Radiol. 2011; 79: 42-47
        • Ghulam Q.M.
        • Bredahl K.K.
        • Lönn L.
        • et al.
        Follow-up on small abdominal aortic aneurysms using three dimensional ultrasound: volume versus diameter.
        Eur J Vasc Endovasc Surg. 2017; 54: 439-445
        • Arko F.R.
        • Murphy E.H.
        • Davis C.M.
        • et al.
        Dynamic geometry and wall thickness of the aortic neck of abdominal aortic aneurysms with intravascular ultrasonography.
        J Vasc Surg. 2007; 46: 891-897
        • Shang E.K.
        • Nathan D.P.
        • Woo E.Y.
        • et al.
        Local wall thickness in finite element models improves prediction of abdominal aortic aneurysm growth.
        J Vasc Surg. 2015; 61: 217-223
        • Sacks M.S.
        • Vorp D.A.
        • Raghavan M.L.
        • et al.
        In vivo three-dimensional surface geometry of abdominal aortic aneurysms.
        Ann Biomed Eng. 1999; 27: 469-479
        • Wilson K.
        • Whyman M.
        • Hoskins P.
        • et al.
        The relationship between abdominal aortic aneurysm wall compliance, maximum diameter and growth rate.
        Cardiovasc Surg. 1999; 7: 208-213
        • Molacek J.
        • Baxa J.
        • Houdek K.
        • et al.
        Assessment of abdominal aortic aneurysm wall distensibility with electrocardiography-gated computed tomography.
        Ann Vasc Surg. 2011; 25: 1036-1042
        • Long A.
        • Rouet L.
        • Bissery A.
        • et al.
        Compliance of abdominal aortic aneurysms evaluated by tissue Doppler imaging: correlation with aneurysm size.
        J Vasc Surg. 2005; 42: 18-26
        • Ravhon R.
        • Adam D.
        • Zelmanovitch L.
        Validation of ultrasonic image boundary recognition in abdominal aortic aneurysm.
        IEEE Trans Med Imaging. 2001; 20: 751-763
        • Leotta D.F.
        • Paun M.
        • Beach K.W.
        • et al.
        Measurement of abdominal aortic aneurysms with three-dimensional ultrasound imaging: preliminary report.
        J Vasc Surg. 2001; 33: 700-707
        • Golledge J.
        • Wolanski P.
        • Parr A.
        • et al.
        Measurement and determinants of infrarenal aortic thrombus volume.
        Eur Radiol. 2008; 18: 1987-1994
        • Brunner-Ziegler S.
        • Hammer A.
        • Seidinger D.
        • et al.
        The role of intraluminal thrombus formation for expansion of abdominal aortic aneurysms.
        Wien Klin Wochenschr. 2015; 127: 549-554
        • Behr-Rasmussen C.
        • Grondal N.
        • Bramsen M.B.
        • et al.
        Mural thrombus and the progression of abdominal aortic aneurysms: a large population-based prospective cohort study.
        Eur J Vasc Endovasc Surg. 2014; 48: 301-307
        • Ghulam Q.M.
        • Bredahl K.
        • Sandholt B.
        • et al.
        Contrast enhanced three dimensional ultrasound for intraluminal thrombus assessment in abdominal aortic aneurysms.
        Eur J Vasc Endovasc Surg. 2018; 5: 673-680
        • Arsicot M.
        • Lathelize H.
        • Martinez R.
        • et al.
        Follow-up of aortic stent grafts: comparison of the volumetric analysis of the aneurysm sac by ultrasound and CT.
        Ann Vasc Surg. 2014; 28: 1618-1628
        • Batagini N.C.
        • Ventura C.A.P.
        • Raghavan M.L.
        • et al.
        Volumetry and biomechanical parameters detected by 3D and 2D ultrasound in patients with and without an abdominal aortic aneurysm.
        Vasc Med. 2016; 21: 209-216
        • Wilson K.A.
        • Lee A.J.
        • Hoskins P.R.
        • et al.
        The relationship between aortic wall distensibility and rupture of infrarenal abdominal aortic aneurysm.
        J Vasc Surg. 2003; 37: 112-117
      2. Population Screening Programmes: NHS Abdominal Aortic Aneurysm (AAA) Programme. Public Health England, London, United Kingson2017
        • Renapurkar R.D.
        • Setser R.M.
        • O'Donnell T.P.
        • et al.
        Aortic volume as an indicator of disease progression in patients with untreated infrarenal abdominal aneurysm.
        Eur J Radiol. 2012; 81: E87-E93
        • Kontopodis N.
        • Metaxa E.
        • Papaharilaou Y.
        • et al.
        Value of volume measurements in evaluating abdominal aortic aneurysms growth rate and need for surgical treatment.
        Eur J Radiol. 2014; 83: 1051-1056
        • Kauffmann C.
        • Tang A.
        • Therasse E.
        • et al.
        Measurements and detection of abdominal aortic aneurysm growth: accuracy and reproducibility of a segmentation software.
        Eur J Radiol. 2012; 81: 1688-1694
        • Wilson K.
        • Bradbury A.
        • Whyman M.
        • et al.
        Relationship between abdominal aortic aneurysm wall compliance and clinical outcome: a preliminary analysis.
        Eur J Vasc Endovasc Surg. 1998; 15: 472-477
        • Wolf Y.G.
        • Thomas W.S.
        • Brennan F.J.
        • et al.
        Computed tomography scanning findings associated with rapid expansion of abdominal aortic aneurysms.
        J Vasc Surg. 1994; 20: 529-535
        • Parr A.
        • McCann M.
        • Bradshaw B.
        • et al.
        Thrombus volume is associated with cardiovascular events and aneurysm growth in patients who have abdominal aortic aneurysms.
        J Vasc Surg. 2011; 53: 28-35
        • Wang D.H.
        • Makaroun M.S.
        • Webster M.W.
        • et al.
        Effect of intraluminal thrombus on wall stress in patient-specific models of abdominal aortic aneurysm.
        J Vasc Surg. 2002; 36: 598-604
        • Bredahl K.
        • Taudorf M.
        • Long A.
        • et al.
        Three-dimensional ultrasound improves the accuracy of diameter measurement of the residual sac in EVAR patients.
        Eur J Vasc Endovasc Surg. 2013; 46: 525-532
        • Bredahl K.
        • Sandholt B.
        • Lonn L.
        • et al.
        Three-dimensional ultrasound evaluation of small asymptomatic abdominal aortic aneurysms.
        Eur J Vasc Endovasc Surg. 2015; 49: 289-296
        • Causey M.W.
        • Jayaraj A.
        • Leotta D.F.
        • et al.
        Three-dimensional ultrasonography measurements after endovascular aneurysm repair.
        Ann Vasc Surg. 2013; 27: 146-153