**2. Body**

Main hypotheses of this multicentric study are:


mellitus, which is a well known risk factor in development of occlusive disease of blood

There are two current therapeutic approaches. The first one is surgery and the other is endovascular (Endovascular Aortic Repair – EVAR). In about half of the patients with intact aneurysm, as well as in those with ruptured one, endovascular approach can be applied. Advantages of endovasular treatment are avoiding general anesthesia, laparotomy and clamping the aorta. The procedure lasts shorter and recovery is fast. However, there are some disadvantages or technical limitations of this procedure. It is not possible to place the graft if proximal neck of the anuerysm is smaller than 15mm and conical in shape (21,22), because origins of renal arteries could be covered. Also, the neck of the aneurysm should be orientated at the angle no smaller than 60º towards the sagittal plane of the aorta, iliac arteries must not be tortuous and must measure at least 9 mm in diameter (23,24). During relatively short period of clinical application and development of EVAR (from 1991) the problem of frequently inadequate commercially available aortic stent-grafts for yellow race and patients with low BMI (21) has arised. The appliation of EVAR in yellow race patients showed that only 23-42% grafts, with fabrically defined dimensions, are adequate, in 23-46% they need certain corrections, while in about 30% of patients there is a contraindication for stent placement (25,26,27,28). Contemporary experience in the application of EVAR showed that overall number of complications is relatively high, even up to 30-40%. Also, one of the reasons is a not precise enough preprocedural morphologic evaluation of AAA and early

Modern generations od multidetector CT units (generation 16 slice, 2004 to 64 slice detectors-2007), offered a new visualisation quality and possibility to obtain more relevant diagnostic information compared to DSA. MDCT aortography reaffirmed the significance of preprocedural evaluation which ensures obtaining numerous and high quality information in each and every situation, considering the place of graft insertion, graft design and overall indication for EVAR, as well as relevant postprocedural evaluation and early diagnstics of

During last 3 years, MDCT units with 10-times lower exponential doses per examination were constructed (29-35). At the same time, routine use of high-resolution ultrasonography as non-ionizing morphologic imaging enabled screening programmes for AAA in elderly and high-risk populaton, that are conducted and in progress in many countries (36,37,38).

1. Positive family anamnesis for AAA, as well as trauma, personal history of diabetes mellitus and hypertension, smoking, elevated LDL cholesterol, which are risk-factors

2. There are significant anatomic-morphologic differences in aneurysmatic infrarenal aorta

vessels, in terms of aneurysm development remains controversial (15-20).

diagnostics of postprocedural complications.

Main hypotheses of this multicentric study are:

between Caucasian and Asian patients

possible complications.

**2. Body** 

for AAA

5. The possibility of graft design in individual case is enabled by integrating measurements obtained by MDCT aortography in selective programme

The study was conducted in Clinical center of Serbia - Center for radiology and magnetic resonance and Institute for radiology, University Hospital Saporro (Japan), in period 2009- 2011. In thiis study 31 Asian and 30 Caucasian patients with the infrarenal aortic aneurysm were included, as well as 130 Asian and 126 Caucasian patients with indication for CT aortography (CTA), which confirmed the absence of AAA. Election of patients of both races before referred to CT examination, was performed according to medical history, and definite indication for CT exam was set according to clinical findings and sonographic findings in distal aorta. Exclusion criteria were: rupture of aneurysm, aneurysm that exceeded infrarenal segment, discrete dilatation of aorta and finding of rough intramural and extraluminal calcifications in longer segment.

Data about risk factors for development of AAA (smoking, hypertension, elevated blood cholesterol level) were collected. One of the questions included the presence of diabetes mellitus in personal history. Questionnaire icluded demographic parameters (sex, age, race, education), antropometric data (body weight, body height, body surface, body mass index), personal history (diabetes, trauma, other) and family medical history (presence of AAA in relatives).

For classifying patients according to the level of nutrition, we used international classification recomended by World Health Organization (WHO) and US Institutes of Health: underweighted-BMI<18,4, normal weighted BMI between 18,5 and 24,9; overweighted BMI 25-29,9 and obese BMI>30. According to body height, all the patients were divided in 4 subgroups: shorter than 160 cm, between 160 cm and 170 cm, between 170-180 cm and taller than 180 cm.

Considering smoking, patients were divided into 3 subgroups according to duration of this habit: 10 years, between 10-20 years and over 20 years. Level of blood cholesterol over 3,4 mmol/l was considered elevated. For calculation of body surface (SA) we used Dubois & Dubois formula: SA= 0.20247 x height (m)0.725 x weight (kg)0.425. Considering that it is a complex logarithmic formula, we used software (calculator) for SA recommended by US National institutes of health (http://www.nih.gov). For the calculation of BMI we used established formula: BMI= body weight (kg) : body surface (m2).

#### **3. 64-slice MDCT protocol and measurements used in the study**

CTA examination in both centers was performed on the same CT unit of the same generation, type and model of the machine. We performed examinations on 64-slice VCT

Lightspeed unit (GE, Milwaukee, IL, US). In all cases we used non-ionic contrast agent in concentration of 320-370 (1 ml – 370 mg iodine) applied by automatic injector in cubital vein reaching flow rate of 5 ml/sec.

Taking into account heterogeneity of selected population by constitution, sex, race and age, and expected heterogeneity in "delay time" of the examination start, we used programme mode "SmartPrep" for defining the appropriate time, by selecting the spot in aorta where appearance of contrast agent triggers the acquisition. Helical mode was used in SmartPrep protocol, 120 kV, 250-700 mAs, rotation speed of the tube 0,35 with slice thickness of 1,25 mm with 64 slice detector in 0.625mm reconstruction.

Postprocessing was performed with the same selected applications in both centers:

Volume Viwer Analysis-CTA Aorta and Advanced Vessel Analysis. Interobserver variability was avoided by the fact that examinations in both centers were performed by a single radiologist.

Number of global selected mathematic variables which define morphology at CTA examination usen in this study is 11. Overall number of methodologically defined transverse measurements is 36 (12 for infrarenal aorta and 24 for iliac arteries), overall number of linear measurements is 36 and volumetric measurements 3. All together, these measurements represent methodologic protocol used in the study for defining the morphology of aneurysmatic infrarenal aorta.

Linear, transverse and volumetric measurements were performed according to the protocol defined aforehead, which consisted of following parameters (Figure 1). All linear measurements were performed in 3 characteristic 2D and 3 characteristic 3D reconstructions (AP, PA and semi-oblique) and mean value was used as definite. We used software ruler tool which is a part of every Analysis-CTA Aorta. Proximal point for measuring aneurysmatic neck, linear distances of aorta, angle between AAA and all the other calculations were positioned in the orifice level of main renal artery. We performed following linear measurements:


Calculations of aortic and iliac arteries volumes represent a part of the basic package of Analysis-CTA Aorta programme. Start and end point are defined (Figure 1). Computer calculates only the lumen of blood vessel that contains contrast agent with no calcium deposits, and without wall structures in cases of thrombosed extraluminal mass; if AAA contains only the dilated vessel wall, the lumen is calculated in total.

Transverse measurements were performed using Advanced Vessel CT Aorta Analysis programme which enables linear differentiating the lumen of contrast agent that fills the vessel from intramural and endoluminal calcifications, considering the similar attenuation values of calcium and contrast agent which cannot be differentiated visually (there is a possibility of misinterpreting calcified plaque as vessel lumen). We performed 6 typical measurements in the same plane, for the lumen of circulating blood (total of 12 "flow" diameters)(F.d.) and 6 measurements in the same plane for diameters of circulating blood together with thrombosed blood, aneurysm content and thickness of the vessel wall (total of 12 "real"diameters)(R.d)(Figure 5, Figure 7).

112 Aneurysm

reaching flow rate of 5 ml/sec.

single radiologist.

aneurysmatic infrarenal aorta.

following linear measurements:





contains only the dilated vessel wall, the lumen is calculated in total.


mm with 64 slice detector in 0.625mm reconstruction.

Lightspeed unit (GE, Milwaukee, IL, US). In all cases we used non-ionic contrast agent in concentration of 320-370 (1 ml – 370 mg iodine) applied by automatic injector in cubital vein

Taking into account heterogeneity of selected population by constitution, sex, race and age, and expected heterogeneity in "delay time" of the examination start, we used programme mode "SmartPrep" for defining the appropriate time, by selecting the spot in aorta where appearance of contrast agent triggers the acquisition. Helical mode was used in SmartPrep protocol, 120 kV, 250-700 mAs, rotation speed of the tube 0,35 with slice thickness of 1,25

Volume Viwer Analysis-CTA Aorta and Advanced Vessel Analysis. Interobserver variability was avoided by the fact that examinations in both centers were performed by a

Number of global selected mathematic variables which define morphology at CTA examination usen in this study is 11. Overall number of methodologically defined transverse measurements is 36 (12 for infrarenal aorta and 24 for iliac arteries), overall number of linear measurements is 36 and volumetric measurements 3. All together, these measurements represent methodologic protocol used in the study for defining the morphology of

Linear, transverse and volumetric measurements were performed according to the protocol defined aforehead, which consisted of following parameters (Figure 1). All linear measurements were performed in 3 characteristic 2D and 3 characteristic 3D reconstructions (AP, PA and semi-oblique) and mean value was used as definite. We used software ruler tool which is a part of every Analysis-CTA Aorta. Proximal point for measuring aneurysmatic neck, linear distances of aorta, angle between AAA and all the other calculations were positioned in the orifice level of main renal artery. We performed

Calculations of aortic and iliac arteries volumes represent a part of the basic package of Analysis-CTA Aorta programme. Start and end point are defined (Figure 1). Computer calculates only the lumen of blood vessel that contains contrast agent with no calcium deposits, and without wall structures in cases of thrombosed extraluminal mass; if AAA

Transverse measurements were performed using Advanced Vessel CT Aorta Analysis programme which enables linear differentiating the lumen of contrast agent that fills the

Postprocessing was performed with the same selected applications in both centers:

**Figure 1.** Characteristic points of interest to mark the CT angiographic analysis in Figure 2D.

**Figure 2.** CT linear measurements of the aorta in the 2D image: mean length of abdominal aorta, mean linear distance from renal artery to aortic bifurcation and AAA angle

**Figure 3.** AAA neck length in 3 characteristic measurements in 2D (a-c) and 3D image (d-f).

The points of transverse measurements of abdominal aorta (a.a.) and common iliac artery (a.i.c.) performed in this study were following:


The precise localization of transverse measurements was defined according to the linear reconstruction of aorta and iliac arteries.

**Figure 4.** Measuring the length of the c.i.a in characteristic 3 position (AP, PA and oblique) in 2D (a-c) and 3D image (d-f)

**Figure 5.** Transverse measurement of infrarenal aortic aneurysms (diameters Rd and Fd)

As a variable part of the protocol of morphologic measurements in this study, depending on the individual case, we performed other diagnostic explorations enabled by selected computer application, such as:


114 Aneurysm

**Figure 3.** AAA neck length in 3 characteristic measurements in 2D (a-c) and 3D image (d-f).

(a.i.c.) performed in this study were following:

R.d. (F.d.a.a 1, F.d. a.a 2, R.d. a.a 1 and R.d. a.a. 2)

(F.d. a.a 3, F.d. a.a 4, R.d. a.a 3 and R.d. a.a. 4)

a.a 5, F.d. a.a 6, R.d. a.a 5 and R.d. a.a. 6);

reconstruction of aorta and iliac arteries.

2);

a.i.c 4);

The points of transverse measurements of abdominal aorta (a.a.) and common iliac artery

a. infrarenal, in the level of the neck of the aneurysm, the largest and the smallest F.d. and

b. in the middle part of abdominal aorta, largest and the smallest F.d and R.d diameter

c. just above the aortic bifurcation, the largest and the smallest F.d and R.d diameter (F.d.

d. proximal parts of both common iliac arteries distally from aortic bifurcation, the largest and the smallest F.d and R.d diameter (F.d. a.i.c 1, F.d. a.i.c 2, R.d. a.i.c 1 and R.d. a.i.c

e. middle parts of both common iliac arteries below aortic bifurcation, the largest and the smallest F.d and R.d diameter (F.d. a.i.c 3, F.d. a.i.c 4, R.d. a.i.c 3 and R.d.

f. distal parts of both common iliac arteries above their bifurcations, the largest and the smallest F.d and R.d diameter (F.d. a.i.c 5, F.d. a.i.c 6, R.d. a.i.c 5 and R.d. a.i.c 6);

The precise localization of transverse measurements was defined according to the linear

f. Coronal or sagittal 3D reconstruction (Figure 12)

**Figure 6.** MD CT volumetric measurement of infrarenal aorta and aa.iliace comm bill in patients with calcium channel intraluminal nodular induration (a,b) patients with AAA and without calcium induration in the wall (c,d).

a) Defining tissue structure (attenuation) in the region of interest in different planes (most often transverse and sagittal). Using this exploration, known as "color mapping" it is possible to determine the density of the tissue in ROI or mean tissue density in the wider region. According to the attenuation distribution, it is possible to determine the structure or density of thrombotic aneurysmatic blood as well as differentiate contrast agent from calcified plaques which are pointed intraluminally. Every color represents a range of some interval of tissue density from 20-800 HU. Contrast agent is always represent by color green, low-density strustures (thrombus, blood, fat) by blue, atherosclerotic deposits by yellow and calcified indurations by red.

Abdominal Aortic Aneurysm in Different Races Epidemiologic Features and Morphologic-Clinical Implications Evaluated by CT Aortography 117

116 Aneurysm

induration in the wall (c,d).

calcified indurations by red.

**Figure 6.** MD CT volumetric measurement of infrarenal aorta and aa.iliace comm bill in patients with calcium channel intraluminal nodular induration (a,b) patients with AAA and without calcium

a) Defining tissue structure (attenuation) in the region of interest in different planes (most often transverse and sagittal). Using this exploration, known as "color mapping" it is possible to determine the density of the tissue in ROI or mean tissue density in the wider region. According to the attenuation distribution, it is possible to determine the structure or density of thrombotic aneurysmatic blood as well as differentiate contrast agent from calcified plaques which are pointed intraluminally. Every color represents a range of some interval of tissue density from 20-800 HU. Contrast agent is always represent by color green, low-density strustures (thrombus, blood, fat) by blue, atherosclerotic deposits by yellow and

**Figure 7.** Transverse measurement of infrarenal aortic aneurysms (diameters Rd and Fd)(b-d) compared to the linear angiographic image (a) (a)

**Figure 8.** Defining tissue structure (attenuation) in the region of interest in the transverse (a) and coronary reconstruction (b,c)

b) Defining the configuration of the blood vessel enables precise visualization in cases of suspected dissection of the vessel wall and enables defining the wall thickness in all planes. Furthermore, it enables clear graphic demarcation of the lesions of aortic wall and

differentiating from the extraluminal lesions. Additional option is definition of calcified indurations inside the "contoured" picture.

**Figure 9.** Contouring infrarenal aorta: linear (a) and with intramural calcifications (b)

c) Isolated defining extraluminal calcifications (Figure 3.21. a), intramural calcifications and altogether in frontal reconstruction along the aortic segment, in selected planes and positions. This exploration may imply on therapeutic approach in two projections at the level of c.i.a

d) Dynamic analysis of contrast agent flow represents the review of video-recording in selected plane, most often transverse or frontal, where dynamic of the contrast agent flow in aortic lumen can be analysed in real time mode.This exploration has a specific value in postprocedural evaluation and diagnostics of early complications of EVAR, most of all the proximal endoleak as a frequent complication. It is more sensitive than conventional digital aortography video -recording

e) MDCT aortoscopy, known also as "virtual aortography", is a special visualization option in "advanced" options of MD CT aortography, which is offered in standard postprocessing units of 16-64 slice MDCT units from the year 2007. It is a relatively simple, but powerful method of endoluminal examination in all planes, that enables optical presentation of the aortic wall inner surface, lesions of the aortic walls, the extent of anuerysmatic dilatation, endoluminal plaques and vessel arborization.

level of c.i.a

aortography video -recording

endoluminal plaques and vessel arborization.

indurations inside the "contoured" picture.

differentiating from the extraluminal lesions. Additional option is definition of calcified

**Figure 9.** Contouring infrarenal aorta: linear (a) and with intramural calcifications (b)

c) Isolated defining extraluminal calcifications (Figure 3.21. a), intramural calcifications and altogether in frontal reconstruction along the aortic segment, in selected planes and positions. This exploration may imply on therapeutic approach in two projections at the

d) Dynamic analysis of contrast agent flow represents the review of video-recording in selected plane, most often transverse or frontal, where dynamic of the contrast agent flow in aortic lumen can be analysed in real time mode.This exploration has a specific value in postprocedural evaluation and diagnostics of early complications of EVAR, most of all the proximal endoleak as a frequent complication. It is more sensitive than conventional digital

e) MDCT aortoscopy, known also as "virtual aortography", is a special visualization option in "advanced" options of MD CT aortography, which is offered in standard postprocessing units of 16-64 slice MDCT units from the year 2007. It is a relatively simple, but powerful method of endoluminal examination in all planes, that enables optical presentation of the aortic wall inner surface, lesions of the aortic walls, the extent of anuerysmatic dilatation,

**Figure 10.** Defining calcifications of AAA - the level of the aortic extraluminarly (a) and intramural (b)

**Figure 11.** MDCT aortoscopy the level of renal arteries (a), middle third of a.a (b) and before the bifurcation (c)

f) Coronal or sagittal 3D reconstruction (VR 3D cut). It is a postprocessing option from the standard group which is more often used in diagnostics of parenchymatous organs and heart, and represents "listing" slices at selected distance (0,625 mm at least) in 3D presentation of the organ or lesion. In AAA, it can be used for visualization of thromobotic mass and its structure considering heterogeneity and the presence of calcified indurations. The analysis can be preformed in 6 standard planes: AP – anteroposterior; PA – posteroanterior; L –left lateral, R – right lateral, I – inferior-superior; S – superior-anterior, and additionally in every non-standard plane of rotated 3D image, which defines 4D visualization in terms of movement.

**Figure 12.**VR 3D cut option in coronary reconstuction in the antero-posterior direction (a) in 3 planes (b,c,d)

### **4. Statistic methodology**

Considering the heterogeneity of the population included, as well as the number of analyzed variables, we used several statistical models for data analysis in this study:

Univariant and multivariant staistical methods **–** for testing statistic significance of difference between parameters for qualitative variables, as well as quantitative variables, univariant methods were used

χ² test –for testing the relationship between non-parametric variabes.

ANOVA - one-sample analysis of variance- univariant analysis of the effect of one selected factor on dependent variable. Comparing mean values, standard deviations in development of aneurysm between races and in the same race compared to control subjects, was performed using this method.

Median Test, Kolmogorov-Smirnov Z-test analysis of the mutual influence (of the selected variable among the groups), testing the compatibility of controls and patients in terms of developing the aneurysm, between races, and in the same race compared to the controls.

We determined the correlation coefficient (Pearson correlation) related to groups, smoking habit, and smoking history of all the subjects included, subjects according to sex (in males and females seperately).

According to univariant logistic regression analysis (ULRA) we tested the influence of selected variable (risk factors) and their correlation on the aneurysm development at the probability level p ≤ 0,01.

Regression analysis (logistic model) In the model of MLRA we included all variables (risk factors) that were confirmed by univariant logistic regression analysis (ULRA) to be connected to the aneurysm development at the level of p ≤ 0,01, so we determined the independent risk factors for development of aneurysm in all the subjects, and then seperately for male and female groups. All the variables were additionally tested in terms of age.

Odds Ratio (OR) or (expB) with confidence interval of 95%- chance ratio, or the possibility of the selected happening, the assessment of the correlation of risk factors (happening) and disease occurrence (development of aneurysm).

Statistic significance was determined at the level of probability of the null hypothesis p ≤ 0.05 to p<0.0001. Statistical analysis was performed using SPSS ver.20, while graphs and tables were edited using MICROSOFT OFFICE (EXCEL and WORD).
