Evidence from cross-sectional imaging indicatesabdominal but not thoracic aortic aneurysms arelocal manifestations of a systemic dilating diathesisIan Nordon, MSc, Ranjeet Brar, Jeremy Taylor, BSc, Robert Hinchliffe, MD, Ian M. Loftus, MD, andMatt M. Thompson, MD, London, United Kingdom

Introduction: The pathogenesis of aortic aneurysms remains unclear. There is epidemiologic and histologic evidenceshowing significant differences in aneurysms of the thoracic and abdominal aorta. Studies suggest that abdominal aorticaneurysms (AAA) may represent a local manifestation of a systemic dilating diathesis. It is not known whether thoracicaortic aneurysms (TAA) also have a systemic etiology. The evidence for a systemic dilating diathesis in AAA disease isreviewed and supplemented with an original morphologic study based on computed tomography (CT) comparingnonaneurysmal controls with patients with AAAs and TAAs.Methods: CT scans performed between January and November 2008 of 150 consecutive patients were examined. Themorphology and dimensions of branches of the aorta in 50 TAA patients and 50 AAA patients were compared with 50nonaneurysmal controls. Measurements of the aorta, common carotid artery (CCA), and superior mesenteric artery(SMA) were taken along with corresponding patient risk factors.Results: Patients were well matched for age, gender, and comorbidity. Mean (SD) right CCA diameter was 9.3 � 1.2 inAAA patients vs 8.1 � 1 mm in TAA patients (P < .0001) and 7.9 � 0.9 mm in controls. Mean left CCA diameter was9.3 � 1.2 mm in AAA patients vs 8.1� 0.8 mm in TAA patients (P < .0001) and 7.9 � 0.8 mm in controls. There wasno significant difference in SMA morphology among the three groups: AAA, 8.6 � 1.1; TAA, 8.3 � .9; and controls,8.4 � 0.9 mm. Multifactorial modeling accounting for risk factors, age, and gender confirmed that the diameterdifference between groups retained independent statistical significance.Conclusions: This study provides further convincing evidence for a systemic dilating diathesis of elastic arteries in AAAs.It also highlights the differing natures of thoracic and abdominal aortic aneurysmal disease. ( J Vasc Surg 2009;50:171-6.)

Clinical Relevance: A clear understanding of the pathophysiology of aortic aneurysms is essential. Observational studieshave raised the suspicion that abdominal aortic aneurysms are a local representation of a systemic disease of thevasculature. Whether aneurysms of the thoracic aorta have a similar systemic tendency is unknown. Validation of this willopen broader avenues of research clarifying understanding of the initiating factors of aneurysms and whether allaneurysms distributed through the arterial tree share the same etiology. This may in turn identify systemic treatments that

eventually may allow targeted pharmaceutical management of small aortic aneurysms.

Abdominal aortic aneurysm (AAA) is a localized dilata-tion of the abdominal aorta that exceeds the normal diam-eter by 50%.1 AAA formation appears to be a focal event,yet patients with AAAs often have aneurysms at sites remotefrom the abdominal aorta. This finding raises the suspicionthat AAAs are a local representation of a systemic arterialdilating diathesis. The evidence for a systemic process withfocal manifestation in the abdominal aorta is diverse, in-cluding epidemiologic, genetic, immunologic, and bio-chemical factors. Whether thoracic aortic aneurysms (TAA)have similar associations remains unknown.

From St George’s Vascular Institute, St James’ Wing, St George’s Hospital.Competition of interest: none.Additional material for this article may be found online at www.jvascsurg.

org.Correspondence: Mr Ian Nordon, St George’s Vascular Institute, St James’

Wing, St George’s Hospital, Blackshaw Rd, London, SW17 0QT, UK(e-mail: inordon@sgul.ac.uk).

0741-5214/$36.00Copyright © 2009 by the Society for Vascular Surgery.

doi:10.1016/j.jvs.2009.03.007

Enlarged peripheral arteries—a generalized arteriomegaly—are noted in patients with AAAs compared with controls.2-5

The carotid artery has been cited as a sensitive measure of thisconcept.6 Increases in luminal diameter and wall stiffness havebeen found in the common carotid artery of patients withAAAs compared with healthy controls and patients with ath-erosclerotic occlusive disease.7 The incidence of AAAs isincreased in patients with coronary artery ectasia,8,9 ce-rebral aneurysms,10 and popliteal aneurysms.11,12 Sa-phenous vein bypass grafts in patients with aneurysmaldisease are equally likely to have a greater diameter andan increased incidence of development of intrinsic graftaneurysms.13,14

Patients with AAAs have an increased incidence ofinguinal hernias,15 diastasis recti,16 and postoperative inci-sional hernias.17 The etiology of these, albeit multifactorial,is thought to represent a defect in collagen and elastinmetabolism. Patients with emphysema have a higher inci-dence of AAA than the healthy population18 and are atgreater risk of AAA rupture.19 Histopathology of both

diseases confirms decreased elastin concentration in the

171

a for

JOURNAL OF VASCULAR SURGERYJuly 2009172 Nordon et al

aortic and lung tissue. Smoking is a strong risk factor forboth diseases, but as yet no causative link between smokingand AAA formation has been proven.20

Arteries at specific sites throughout the vasculaturehave essential intrinsic differences. The abdominal aorta,iliac artery, and common carotid arteries are all elasticarteries, whereas the common femoral, coronary, and su-perior mesenteric arteries (SMA) are muscular. Aneurysmaldegeneration is related to a proportional decrease in elastinconcentration of the aorta21; consequently, all elastic arter-ies may be more vulnerable to this loss and vulnerable todilatation. The discrimination in arterial type is not cate-goric, because the popliteal artery is muscular but hassimilar properties to the elastic aorta,22 and the proximalSMA is a branch of the aorta but is deemed muscular.23

The aim of the present study was to collect furtherevidence of the systemic nature of the AAA process and todefine whether AAA and TAA share similar characteristicswith regard to peripheral arteriomegaly.

METHODS

Patients. The study included 150 consecutive patientsundergoing multislice computed tomography (CT) imag-ing. This cohort was generated from the most recent pa-tients undergoing endovascular repair of AAAs, patientsundergoing endovascular repair of TAAs, and a group ofcontrol patients (Fig 1). Patients in the aneurysm groupswere excluded if they had a known congenital connectivetissue disorder, if they had an incomplete CT scan, or iftheir aneurysm was as consequence of aortic dissection. Thecontrol group was identified from the most recent consec-utive 50 patients who had undergone CT investigation ofthe thorax, abdomen, and pelvis. The exclusion criteria forcontrols included known aneurysms, age �60 years, or anincomplete or noncontrast-enhanced scan. All patients’ risk

Fig 1. Flow chart shows exclusion criteri

factors were identified by examination of their computer-

based medical record. All scans were performed from Jan-uary 2008 to November 2008.

Measurements. All CT examinations were performedin the arterial phase on a 64-MDCT (Lightspeed VCT; GEHealthcare, Waukesha, Wisc). The aortic protocol used 90mL of iodinated contrast (Omnipaque 350; Nycomed Am-ersham, Princeton, NJ) injected at 4 mL/s and SmartPrep(GE Healthcare), with the region of interest centered onthe ascending aorta. Axial images were acquired at0.625-mm intervals and reconstructed to a 2.5-mm slicethickness.

The entire aorta was studied, extending to include thecommon carotid arteries (CCA) and common femoral ar-teries. Measurement of the CCAs, SMA, and aorta atconsistent anatomic landmarks and orientations (Table I)was performed on a PACS workstation (Philips Healthcare,Bothell, Wash). Each vessel diameter was measured fromthe outside wall to the outside wall. All measurements weretaken from pretreatment scans when the aortic aneurysmsize warranted treatment.

Validation. All measurements were taken by authorsI. N. and R. B. Independent validation of these measure-ments was performed by author J. T., who measured thecarotid arteries blinded to the presence of aneurysmal dis-

scans. AAA, Abdominal aortic aneurysm.

Table I. Sites of vessel measurementa

Vessel Point of measurement

CCA Inferior cornu of thyroid cartilage (max AP diameter)SMA 1 cm from its origin (max cross-sectional diameter)Aorta Max aneurysmal diameter (max AP infrarenal

diameter in controls)

AP, Anteroposterior; CCA, common carotid artery; SMA, superior mesen-teric artery.aAll vessels were measured in mm to 1 decimal place.

ease and the measurements taken by the other observers.24

JOURNAL OF VASCULAR SURGERYVolume 50, Number 1 Nordon et al 173

Risk factor definitions. As previously described,25

hypertension was defined as patients taking antihyperten-sive drugs or having persistent systolic blood pressure�140 mm Hg or diastolic blood pressure �90 mm Hg onadmission. Diabetes mellitus was defined as patients cur-rently being treated with antidiabetic medication or insulin.Cigarette smoking status was defined as current, former, ornever.

Power calculation. A power calculation using datafrom duplex-based studies of carotid vessels6,7 confirmedthat for 90% power (P � .05) to detect a 1-mm differencein vessel caliber, 30 patients were required in each of twogroups.

Statistics. The incidence of risk factors between groupswas compared using two by three contingency tables apply-ing �2 analysis. Continuous variables (age and vessel diam-eters) were compared using a one-way analysis of variance(ANOVA). If significant differences (P � .05) were identi-fied, post hoc analysis by Bonferroni-Dunn testing wasperformed. The Pearson correlation coefficient was used toassess the relationship between vessel diameters. ANOVAwas used to assess interobserver variability and to screen fora relationship between risk factors and group measure-ments. Screening was performed using a general linearmodel with the group and each risk factor in turn. Riskfactors demonstrating statistical significance were includedin a final multifactorial model to assess their effect. Asignificant difference was defined at P � .05. All values areexpressed as mean � SD. Statistical analysis was performedusing SPSS 15.0 software (SPSS, Chicago, Ill).

RESULTS

Patient characteristics and risk factors. The patientswith AAA or TAA appeared well matched with the controlgroup for age, gender, and risk factors. The control group

Table II. Comparison of patients with abdominal or thor

Variable AAA TAA

Patients, No. 50 50Gender, No.

Male 47 41Female 3 9

Age, mean � SD 75.2 � 7.1 74 � 7.4Hypertension, % 70 68Diabetes, % 10 16Smoking, % 60 46Diameter, mean � SDAortic 68 � 14.7 67 � 12.Right CCA 9.3 � 1.2 8.1 � 1Left CCA 9.3 � 1.2 8.1 � 0.8SMA 8.6 � 1.1 8.3 � 0.9

CCA, Common carotid artery; SMA, superior mesenteric artery.aCalculated by analysis of variance. Statistical significance set at P � .05.bPost hoc testing by Bonferroni-Dunn analysis. Statistical significance set atc�2 Testing by 3 � 2 contingency tables.

consisted of nine patients who underwent emergency scans

for suspected trauma and 41 who had surveillance scansafter tumor resection.

CT findings. Mean aortic diameter was 20.8 � 2.6mm in the control group; this was significantly smaller thanthe AAA and TAA groups (P � .0001; Table II).

The AAA group demonstrated significantly largerCCAs than controls. The right CCA measured 9.3 � 1.2 vs

Fig 2. Box and whisker plot demonstrates right common carotidartery (CCA) measurements in patients with abdominal (AAA)and thoracic aortic aneurysms (TAA). The horizontal line in themiddle of each box indicates the median; the top and bottom bordersof the box mark the 75th and 25th percentiles, respectively, andthe whiskers mark the 90th and 10th percentiles.

ortic aneurysm, and control patients

Controls Pa

Post hoc P vs controlb

AAA TAA

50.28c . . . . . .

446

74 � 7 .40 . . . . . .52 .12c . . . . . .14 .67c . . . . . .40 .12c . . . . . .

21 � 2.6 .0001 .0001 .00017.9 � 0.9 .0001 .0001 .6987.9 � 0.8 .0001 .0001 .7898.4 � 0.9 .38

05.

acic a

6

P � .

7.9 � 0.9 mm in controls (P � .0001; Fig 2) and the left CCA

JOURNAL OF VASCULAR SURGERYJuly 2009174 Nordon et al

measured 9.3 � 1.2 vs 7.9 � 0.8 mm (P � .0001; Fig 3).SMA diameters did not differ between groups (8.6 � 1.1 vs8.4 � 0.9 mm; P � .32; Fig 4).

The TAA group did not demonstrate any significantdiameter changes in branch vessels relative to controls. Theright CCA diameter was 8.1 � 1 vs 7.9 � 0.9 mm incontrols (P � .30). This theme was consistent across allaortic branches measured. The left CCA was 8.1 � 0.08mm and the SMA was 8.3 � .09 mm; neither value wassignificantly different from controls.

Measurement validation. Interobserver measurementvariability was 0.37 mm (95% confidence interval, 0.21-0.52 mm) measuring the right CCA. ANOVA confirmedno statistical significance between observers (P � .09) (Figs6A-C).

Multifactorial analysis. ANOVA excluded gender(P � .22), diabetes (P � .16), and hypertension (P � .97)from having significant associations with aortic or carotiddiameter. Screening by univariate analysis demonstratedsmoking (P � .08) and age (P � .09) might be associatedwith CCA diameter. These factors were applied to a multi-factorial model that concluded that accounting for thesevariables, the intergroup differences for CCA diametermeasurements retained their statistically significant differ-ence (P � .0001; Table III). The multifactorial model alsoidentified smoking as having an association with CCAdiameter (P � .009) independent of patient group.

Correlation. The increase in CCA diameter did notcorrelate with increasing aortic diameter (r � .01) in pa-

Fig 3. Box and whisker plot demonstrates the left common ca-rotid artery (CCA) measurements in patients with abdominal(AAA) and thoracic aortic aneurysms (TAA). The horizontal linein the middle of each box indicates the median; the top and bottomborders of the box mark the 75th and 25th percentiles, respectively,and the whiskers mark the 90th and 10th percentiles.

tients with AAAs (Fig 5).

DISCUSSION

This study is unique in identifying that AAAs are a localrepresentation of a systemic disease of the vasculature, yetTAAs are not. The findings of the present study corrobo-rate the previous evidence from ultrasound studies ofAAAs.

Arterial enlargement has been proposed as a compen-satory pathobiologic response to early atherosclerosis.26

This has been demonstrated in peripheral arteries, yet isdeemed to be a focal event and most evident in the smallestvessels.27 This study population was well matched for riskfactors of atherosclerosis and examined large branches ofthe aorta. The arterial enlargement in these AAA patientsmay be independent of atherosclerosis. Although to furthervalidate this finding, the relationship between the externaldiameter and luminal diameters would need assessment.

The SMA was measured because it is a muscular artery.It originates from the visceral aorta, an area generally less

Fig 4. Box and whisker plot demonstrates the superior mesen-teric artery (SMA) measurements in patients with abdominal(AAA) and thoracic aortic aneurysms (TAA). The horizontal linein the middle of each box indicates the median; the top and bottomborders of the box mark the 75th and 25th percentiles, respectively,and the whiskers mark the 90th and 10th percentiles.

Table III. Statistical significance derived frommultifactorial analysis assessing effect of factors on rightcommon carotid artery diameter

Variable LCC significance RCC significance

Age .016 .06Patient group .0001 .0001Smoking .013 .009Group � smoking .49 .428

LCC, Left common carotid; RCC, right common carotid.

susceptible to aneurysm development. That no morpho-

JOURNAL OF VASCULAR SURGERYVolume 50, Number 1 Nordon et al 175

logic change was identified in this vessel is not surprising:The muscular nature of the SMA renders it resistant todilatation.

The distinction between elastic and muscular arteries isimportant. Elastic arteries are more vulnerable to dilatationbecause their anatomic position exposes them to the pulsewave of cardiac output.28 They are designed to moderatethis pulse wave and consequently have an elevated elastincomposition. When age-related degenerative changes andelastin loss occur; vessel remodelling leads to dilatation.Through the normal aging process, between the ages of 25and 71, a 30% increase occurs in the diameter of the distalaorta29 as a result of disruption of elastin components,fibrosis, and an increase in collagen.30 These specificchanges may be why the systemic dilating diathesis is moreapparent in elastic arteries and why these arteries are moreprone to aneurysm formation.

No linear correlation between AAA size and carotiddiameter (r � 0.01) was demonstrated. Aneurysm growthis not linear, and these measurements only represent asnapshot at one time of vessel size. An interesting studywould be to observe changes in CCA size during observa-tion of small AAAs identified at screening. This would addfurther evidence to the argument for a systemic disease ofthe vasculature in AAA patients.

The TAA patients did not demonstrate a systemic di-lating diathesis. The small difference in prevalence of TAAsvs AAAs according to gender was evident and consistentwith previous work.31 Other differences can be found in thepathobiology of TAAs compared with AAAs. AAAs are char-acterized by local inflammatory changes, medial smooth mus-cle cell apoptosis, and fragmentation of the extracellular ma-

Fig 5. Scatterplot demonstrates the association between thecommon carotid artery (CCA) and abdominal aortic aneurysm(AAA) size (r � 0.01).

trix.32 TAAs are characterized by medial necrosis, mucoid

infiltration, elastin degradation, and smooth muscle cell pop-tosis.33 Genetic factors have been shown to be important inthe etiology of both AAA and TAA. These genetic riskfactors appear different, as the TAA and AAA susceptibilityloci do not overlap.33 Taking these findings into consider-ation, it is unsurprising to find further evidence supportingthe independent pathophysiology of these processes.

Study limitations. Although patients were well matchedfor age and comorbidity, the retrospective nature of thisstudy precluded standardization for body mass index. Ca-rotid arterial diameter has been shown to increase with age,gender, height, weight, and systolic blood pressure.34

Only patients who were eligible for endovascular stent-ing were enrolled in this study, which might have generatedbias by excluding patients with unfavorable iliac anatomy.The sole exclusion criterion for endovascular repair identi-fied in this time period was unfavorable aneurysm neckmorphology. Conversely, this might be a strength becauseeach patient demonstrated aneurysmal disease at a focalpoint of the aorta, not diffuse vessel enlargement. Corre-lating focal aneurysmal disease with distant vessel dilatationadds weight to the argument for a systemic disease process.

Observational studies such as this one can be criticizedfor the absence of observer blinding; however, the studydesign incorporated validation of measurements from anindependent observer who was blinded to the presence ofaneurysmal disease and other observer measurements. Thisstep also demonstrated a consistency of measurementacross the three observers.

CONCLUSIONS

This study provides further evidence that abdominalaortic aneurysms are a local representation of a generalizedsystemic dilating diathesis. The systemic dilating disease isprominent in the susceptible elastic arteries and accentu-ated in the vulnerable aneurysmal aorta. This phenomenonis not demonstrated in patients with thoracic aortic aneu-rysms. This reinforces the understanding that the patho-physiology of thoracic and abdominal aortic aneurysms isdifferent.

AUTHOR CONTRIBUTIONS

Conception and design: IN, RHAnalysis and interpretation: IN, RH, IL, MTData collection: IN, RB, JTWriting the article: IN, RHCritical revision of the article: IL, MTFinal approval of the article: IN, RB, RH, JT, IL, MTStatistical analysis: INObtained funding: Not applicableOverall responsibility: MT

REFERENCES

1. Johnston KW, Rutherford RB, Tilson MD. Suggested standard forreporting on arterial aneurysms. J Vasc Surg 1991;13:444-50.

2. Thomas L. Arteriomegaly. Br J Surg 1971;58:690-4.3. Tilson MD, Dang C. Generalized arteriomegaly. Arch Surg 1981;116:

1030-2.

JOURNAL OF VASCULAR SURGERYJuly 2009176 Nordon et al

4. Carlson DH, Gryska P, Seletz J, Armstrong S. Arteriomegaly. Am JRoentgenol Radium Ther Nucl Med 1975;125:553-8.

5. Ward AS. Aortic aneurysmal disease–a generalised dilating diathesis.Arch Surg 1992;127:990-1.

6. Iwamoto T, Kimura A, Nakai T, Kanaya K, Ishimaru S. Implications ofcarotid arteriomegaly in patients with aortic aneurysm. J AtherosclerThromb 2004;11:348-53.

7. Makita S, Ohira A, Tachieda R, Itoh S, Moriai Y, Niinuma H, et al.Dilation and reduced distensibility of carotid artery in patients withabdominal aortic aneurysms. Am Heart J 2000;140:297-302.

8. Stajduhar KC, Laird JR, Rogan KM, Wortham DC. Coronary arterialectasia: increased prevalence in patients with abdominal aortic aneurysmas compared to occlusive atherosclerotic peripheral vascular disease. AmHeart J 1993;125:86-92.

9. Guy DM, Bailey SR, Dorn GW. Increased prevalence of coronaryarterial ectasia in patients with abdominal aortic aneurysms. J Am CollCardiol 1991;17:188A.

10. Miyazawa N, Akiyama I, Yamagata Z. Risk factors for the association ofintracranial and aortic aneurysms. Acta Neurochir (Wien) 2007;149:221-9; discussion 229.

11. Diwan A, Sarkar R, Stanley JC, Zelenock GB, Wakefield TW. Incidenceof femoral and popliteal artery aneurysms in patients with abdominalaortic aneurysms. J Vasc Surg 2000;31:863-9.

12. Grimm JJ, Wise MM, Meissner MH, Nicholls SC. The incidence ofpopliteal artery aneurysms in patients with abdominal aortic aneurysms.J Vasc Ultra 2007;31:71-73.

13. Loftus IM, McCarthy MJ, Lloyd A, Naylor AR, Bell PR, ThompsonMM. Prevalence of true vein graft aneurysms: implications for aneurysmpathogenesis. J Vasc Surg 1999;29:403-8.

14. Upchurch GR Jr, Gerhard-Herman MD, Sebastian MW, Belkin M,Conte MS, Donaldson MC, et al. Improved graft patency and alteredremodeling in infrainguinal vein graft reconstruction for aneurysmalversus occlusive disease. J Vasc Surg 1999;29:1022-30.

15. Lehnert B, Wadouh F. High coincidence of inguinal hernias andabdominal aortic aneurysms. Ann Vasc Surg 1992;6:134-7.

16. McPhail I. Abdominal aortic aneurysm and diastasis recti. Angiology2008 Dec-2009;59:736-92.

17. Takagi H, Sugimoto M, Kato T, Matsuno Y, Umemoto T. Postopera-tive incision hernia in patients with abdominal aortic aneurysm andaortoiliac occlusive disease: a systematic review. Eur J Vasc EndovascSurg 2007;33:177-81.

18. van Laarhoven CJ, Borstlap AC, van Berge Henegouwen DP, PalmenFM, Verpalen MC, Schoemaker MC. Chronic obstructive pulmonarydisease and abdominal aortic aneurysms. Eur J Vasc Surg 1993;7:386-90.

19. Cronenwett JL, Murphy TF, Zelenock GB, Whitehouse WM Jr, Lin-

denauer SM, Graham LM, et al. Actuarial analysis of variables associated

with rupture of small abdominal aortic aneurysms. Surgery1985;98:472-83.

20. Evans J, Thompson MM. Pathogenesis of abdominal aortic aneurysms - areview. In: Fitridge R, Thompson MM, editors. Mechanisms of vasculardisease. Cambridge: Cambridge University Press; 2007. p. 222-36.

21. Wilson WRW, Schwalbe EC, Jones JL, Bell PR, Thompson MM.Matrix metalloproteinase 8 (neutrophil collagenase) in the pathogenesisof abdominal aortic aneurysm. B J Surg 2005;92:828-33.

22. Debasso R, Astrand H, Bjarnegard N, Ryden Ahlgren A, Sandgren T,Lanne T. The popliteal artery, an unusual muscular artery with wallproperties similar to the aorta: implications for susceptibility to aneu-rysm formation? J Vasc Surg 2004;39:836-42.

23. Kerr JB. Atlas of functional histology. London: Mosby; 2000.24. Haber M, Barnhart HX, Song J, Gruden J. Observer variability: a new

approach in evaluating interobserver agreement. J Data Sci 2005;3:69-83.25. Allison MA, Kwan K, Di’Tomasso D, Wright M, Criqui MH. The

epidemiology of abdominal aortic diameter. J Vasc Surg 2008;48:121-7.

26. Glagov S, Weisenberg E, Zarins CK. Compensatory enlargement ofhuman atherosclerotic coronary arteries. N Eng J Med 1987;316:1371-5.

27. Labropoulos N, Zarge J, Mansour MA, Kang SS, Baker WH. Compen-satory arterial enlargement is a common pathobiologic response in earlyatherosclerosis. Am J Surg 1998;176:140-3.

28. McDonald DA. Regional pulse-wave velocity in the arterial tree. J ApplPhysiol 1968;24:73-8.

29. Lanne T, Hansen F, Mangell P, Sonesson B. Differences in mechanicalproperties of the common carotid artery and abdominal aorta in healthymales. J Vasc Surg 1994;20:218-25.

30. Schlatmann TJ, Becker AE. Histological changes in the normal agingaorta: implication for dissecting aneurysms. Am J Cardiol 1977;39:13-20.

31. Pressler V, McNamara JJ. Thoracic aortic aneurysm. Natural history andtreatment. J Thor Cardiovasc Surg 1980;79:489-98.

32. Thompson RW. Reflections on the pathogenesis of abdominal aorticaneurysms. Cardiovasc Surg 2002;10:389-94.

33. Kuivaniemi H, Platsoucas CD, Tilson MD 3rd. Aortic aneurysms: animmune disease with a strong genetic component. Circulation 2008;117:242-52.

34. Polak JF, Kronmal RA, Tell GS, O’Leary DH, Savage PJ, Gardin JM, etal. Compensatory increase in common carotid diameter. Stroke 1996;27:2012-5.

Submitted Jan 1, 2009; accepted Mar 7, 2009.

Additional material for this article may be found online

at www.jvascsurg.org.

gle observer vs the mean of other two observers.

JOURNAL OF VASCULAR SURGERYVolume 50, Number 1 Nordon et al 176.e1

Fig 6 (online only). Bland-Altman plots demonstratcommon carotid (RCC) artery. Each plot compares a sin

e measurement variability between observers for the right