Introduction

On the posterior abdominal wall, the abdominal aorta (AA) lies slightly to the left of the midline due to the position of the inferior vena cava [19]. It is found between the 12th thoracic vertebra (level of aortic hiatus on the diaphragm) and 4th lumbar vertebra (level of aortic bifurcation) [19]. At the inferior level, it bifurcates into the common iliac arteries (CIA) [19]. In front of the sacroiliac joint, each CIA gives off its terminal branches: the external (EIA) and internal iliac arteries [19]. AA is a vital structure for human development and growth, as AA-related pathologies (e.g., vascular tumors, thrombotic occlusion, middle aortic syndrome, dissections, aneurysms, and hypoplasia) may pose risks up to 90% mortality [15]. A precondition for determining vascular diameter abnormalities such as dilation or stenosis is to firstly define normal [3, 4, 10]. In this context, some authors propose further investigations focused on morphometric features of AA, CIA and EIA to create a normal range for these arteries’ diameters in different populations (e.g., children), considering the limited studies [3, 10,11,12, 14, 15].

Acquired and congenital disorders with or without originating from the heart in children may manifest with changes in AA, CIA or EIA diameter such as dilation or stenosis [3, 10, 15]. The importance of having information about the diameters of these vessels in children may be listed as follows: (a) assessing renal hypertension, (b) tracing hydration status, (c) assessing and giving chase renal transplant, and (d) diagnosing vascular anomalies such as vasculitis, stenosis or aneurysm [1, 7, 15, 18, 20]. Using different imaging tools (e.g., magnetic resonance imaging: MRI, computed tomography: CT, or ultrasound: US), increasing anatomical information (e.g., novel morphometric datasets on different populations), and determining exact standards for diagnosing vascular disorders allow clinicians to decrease morbidity and/or mortality rates [3, 5, 14, 15]. Nevertheless, most of the data reported by previous works is based on information from adult subjects [3, 10, 15]. In this context, there is a need to establish normal standard diameters for AA, CIA and EIA in pediatric populations for better enlightening the anatomy of aorta. Therefore, the main goal of this work was to determine changes in the effective diameter of AA in healthy children aged 1–18 years for diagnosis of vascular diseases.

Materials and methods

Patient selection

The Clinical Research Ethics Committee approved our retrospective multidetector CT study (dated: 13.05.2020, no: 2020/383). After ethical confirmation, patients’ folders (including information such as CT images, treatments, diagnoses, sex, age, hospital admission/discharge dates, and complaints) were reviewed retrospectively. 180 children (sex: 90 males / 90 females, average age: 9.50 ± 5.20 years, range: 1–18 years) without any abdominopelvic disease were included in our study.

Inclusion and exclusion criteria

Inclusion and exclusion criteria of the study were presented in Table 1. Contrast-enhanced abdominopelvic CT views of 180 children admitted to the hospital owing to abdominal pain or trauma were included in the study. Patients with history of surgeries and diseases (splenectomy, cirrhosis etc.), potentially affecting vascular diameter were excluded. Some patients had CT images in their folders taken at different times, in which case a single image was used.

Table 1 The inclusion and exclusion criteria for study population
Full size table

CT protocol

Abdominopelvic CT views were obtained using a 64-row multidetector CT scanner (Aquillion 64, matrix: 512 × 512, FOV: 240 mm, pixel size: 0.46 mm, 0.5 mm thick slices, 230 mA, 120 kV, 0.3 mm interval, Toshiba Medical Systems, Tokyo, Japan). By using intravenous contrast agent, children underwent a routine CT protocol (hepatic phase). Taking into account ALARA (as low as reasonably achievable) guidelines, milliampere-seconds (range between 20 and 150 mAs) and peak kilovoltage (120 kVp for children over 23 kg; 100 kVp for children between 9 and 23 kg; and 80 kVp for children under 9 kg) values suitable to child’s age were determined for CT scan. The raw data from the multidetector CT scanner was reformatted in several planes for creating three-dimensional multiplanar reconstruction on a work station (Vitrea 2). All assessments were performed by the same radiologist (B.T.) with 15 years of experience.

Morphometric parameters

Explanations and abbreviations of the parameters were given in Table 2. Diameters of AA, CIA and EIA were measured in axial planes vertical to their long axes by the use of oblique multiplanar reconstruction (Figs. 1, 2, 3). Transverse and/or antero-posterior diameters were measured by following from one outside wall to the other outside wall of vessels. Since it is difficult to measure between the inner walls of the vessels, especially in young children under 2 years of age, we measured the distance between the outer walls of the vessels to create a standardization. The transverse and antero-posterior diameters of AA were measured at three levels one cm superior to the coeliac trunk, renal artery, and aortic bifurcation. Then, the mean value of these two measurements was used to calculate the effective diameter of AA separately for three levels. Since it was difficult to measure the transverse and antero-posterior diameters of CIA and EIA, especially in young children, effective diameter calculations did not perform in these arteries. That’s why, the longest diameters of the right and left CIA and EIA were separately measured at two levels, proximal and distal. On the other hand, the transverse diameter of the body of the first lumbar vertebra (L1) was measured to use as a standard anatomical indicator, since we did not find always the body surface area of patients, especially in infants. In this context, the ratio of vessel-vertebral diameters was determined to use as a standard indicator.

Table 2 The descriptions of the parameters
Full size table
Fig. 1
figure 1

The photographs show measurement diameters (blue lines) and levels (red lines). a AAD-CT on axial plane, b AAD-RA on axial plane, c the level of AA over the coeliac trunk, and the level of AA over the renal artery on sagittal plane, and d the level of AA over the coeliac trunk, and the level of AA over the renal artery on coronal plane (color figure online)

Full size image
Fig. 2
figure 2

The photographs show measurement diameters (blue lines) and levels (red lines). a AAD-AB on axial plane, b RCIA-PL and LCIA-PL on axial plane, c the level of AA over the aortic bifurcation, and the level of proximal CIA on coronal plane (color figure online)

Full size image
Fig. 3
figure 3

The photographs show measurement diameters (blue lines) and levels (red lines). a RCIA-DL and LCIA-DL on axial plane, b REIA-PL and LEIA-PL on axial plane, c REIA-DL and LEIA-DL on axial plane, and d the level of distal CIA, the level of proximal EIA, and the level of distal EIA on sagittal plane (color figure online)

Full size image

Statistical analysis

The normality control of the dataset involving AA, CIA, EIA and L1 diameters was checked with Shapiro–Wilk. Considering the examination of Goodway et al. [9], the child samples were divided into five groups; (a) infancy group aged from birth to 2 years, (b) early childhood group aged from 3 to 5 years, (c) late childhood group aged from 6 to 9 years, (d) prepubescent group aged from 10 to 13 years, and (e) postpubescent group aged from 14 to 18 years. Alterations in the parameters and ratios relative to child ages (from one to 18 years) and child age groups (from infancy to postpubescence) were examined with One-way ANOVA (post-hoc Bonferroni test). Male–female (the independent test) and right–left (the paired test) comparisons was carried out with the student’s t-test. Statistical differences between measurements belonging to AA were evaluated with ANOVA with repeated measures. Correlations between the parameters were examined with the Pearson correlation coefficient test. Through the simple linear regression test, regression equations and scatter plots of the parameters were obtained. For statistical analysis, the “p < 0.05” was accepted as threshold.

Results

Numerical data belonging to AA, CIA, EIA and L1 diameters were presented as average data ± standard deviations in Tables 3, 4, 5.

Table 3 The measurements related to the AA in children aged between 1 and 18 years (millimetric measurements)
Full size table
Table 4 Comparison of the parameters according to the age periods (millimetric measurements)
Full size table
Table 5 Comparison of the parameters in terms of sex (millimetric measurements)
Full size table

All parameters increased depending on ages between 1 and 18 years (p < 0.001) (Table 3). Considering child age periods, the parameters increased linearly, except from L1TD. This diameter was similar in early and late childhood periods (p = 0.999), and also in prepubescent and postpubescent periods (p = 0.901), and thus showed an irregular growth pattern between age periods (Table 4). The parameters in males were greater than that in females (p < 0.05), apart from AAD-CT (p = 0.084) and AAD-RA (p = 0.051) (Table 5). Strong positive correlations were found between the parameters (Table 6).

Table 6 The correlations between the parameters
Full size table

The regression equations for L1TD (y = 24.644 + 0.732 × years, p < 0.001), AAD-CT (y = 7.297 + 0.542 × years, p < 0.001), AAD-RA (y = 6.467 + 0.513 × years, p < 0.001), AAD-AB (y = 5.432 + 0.481 × years, p < 0.001), RCIA-PL (y = 3.775 + 0.322 × years, p < 0.001), LCIA-PL (y = 3.786 + 0.317 × years, p < 0.001), RCIA-DL (y = 3.655 + 0.319 × years, p < 0.001), LCIA-DL (y = 3.636 + 0.318 × years, p < 0.001), REIA-PL (y = 3.070 + 0.273 × years, p < 0.001), LEIA-PL (y = 3.040 + 0.276 × years, p < 0.001), REIA-DL (y = 2.930 + 0.273 × years, p < 0.001), and LEIA-DL (y = 2.912 + 0.275 × years, p < 0.001) were calculated.

The ratios of vessel-vertebral diameters increased steadily depending on ages between 1 and 18 years (Table 7) (Figs. 4 and 5). Considering child age groups, the ratios increased with age between infancy and postpubescent periods in irregular pattern; however, AAD-AB/L1TD, RCIA-PL/L1TD, and LCIA-PL/L1TD did not alter after late childhood period (Table 8). The ratios of vessel-vertebral diameters for males were similar to females (p > 0.05) (Table 9).

Table 7 The ratios of the parameters according to the L1TD in children aged between 1–18 years
Full size table
Fig. 4
figure 4

The charts show the ratios of the arterial diameter measurements to the L1TD. a AAD-CT/L1TD, b AAD-RA/L1TD, c AAD-AB/L1TD, d RCIA-PL/L1TD, e LCIA-PL/L1TD, and f RCIA-DL/L1TD

Full size image
Fig. 5
figure 5

The charts show the ratios of the arterial diameter measurements to the L1TD. a LCIA-DL/L1TD, b REIA-PL/L1TD, c LEIA-PL/L1TD, d REIA-DL/L1TD, and e LEIA-DL/L1TD

Full size image
Table 8 The ratios of the parameters according to the L1TD in age periods
Full size table
Table 9 Comparison of the ratios of the parameters in terms of sex
Full size table

The order of the diameter measurements of AA from larger to smaller was found as AAD-CT (12.45 ± 3.10 mm) > AAD-RA (11.34 ± 2.99 mm) > AAD-AB (10.01 ± 2.81 mm) (p < 0.001). RCIA-PL (6.83 ± 1.93 mm) was found greater than LCIA-PL (6.80 ± 1.91 mm) (p = 0.026), while RCIA-DL (6.69 ± 1.91 mm) was similar to LCIA-DL (6.66 ± 1.92 mm) (p = 0.072). Moreover, the proximal values of CIA were larger than its distal values (p < 0.001). No statistically significant difference was found between the REIA-PL (5.66 ± 1.65 mm) and LEIA-PL (5.67 ± 1.66 mm) (p = 0.892), or the REIA-DL (5.53 ± 1.64 mm) and LEIA-DL (5.53 ± 1.65 mm) (p = 0.926). The proximal values of EIA were larger than its distal values (p < 0.001).

Discussion

The present work: first, presented regression equations for AA, CIA, EIA and L1 to estimate their age-specific diameters. Second, the ratio of vessel-vertebral diameters may be used as an indicator of vessel stenosis or dilation to create a standard in the diagnosis of pediatric vascular disorders. Third, most of the parameters were greater in males than females, but vessel-vertebral diameters’ ratios in males were similar to females; therefore, the ratios may be used as a reliable indicator of arterial dilation or stenosis regardless of sex.

The aorta (via its parts and branches) distributes oxygenated blood to the body, and thus it plays a significant functional act in the circulatory system [3]. Changes in its elasticity affect the peripheral circulation, the blood flow in heart and brain, and the function of the left ventricle [6, 13]. Alterations in its size and distensibility may be determined via US, MRI and multidetector CT [15, 17]. In this work, we used multidetector CT views of children due to the following reasons: (a) it provides to monitor the vascular morphology, (b) it decreases the need for sedation in children, (c) it decreases the requirement for diagnostic interventional methods, and (d) it is easy and fast to apply [4, 16]. The first condition to understand changes in vascular morphology on imaging tools such as multidetector CT is to know what is normal [4]. For example, knowing the normal range of the aortic diameter may be useful to diagnose aortic aneurysm in patients with Marfan syndrome, or diffuse aortic stenosis in patients with Williams syndrome [10]. In children, acquired and congenital disorders with or without originating from the heart may manifest with changes in AA, CIA or EIA diameter such as dilation or stenosis [3, 10, 15]. In this context, our calculated linear functions may be helpful to estimate age-specific normal diameters of these vessels in pediatric subjects.

In comparison with the normal diameter, an enlargement of the aortic diameter at least 1.5 times may indicate aortic aneurysm [5, 20]. If an aneurysm is to be diagnosed using this cutoff, the normal range of the aortic diameter must be known [15]. However, the aortic diameter is affected by race, diet, body height, age, sex, body shape, and disorders [11]. For this reason, some authors used anthropometric measurements (e.g., the thoracic or lumbar vertebrae, or the body surface area) to create a standard for determining changes in vessel diameters [2, 3, 10, 12]. For evaluating the relation of vessel dimensions to the body size, Akturk and Gunes [3] utilized L1’s largest transverse diameter as the dorsal cut-off to trace of L1’s body. Geraghty and Boone [8] stated that this vertebra might be a useful anatomical indicator, due to the following reasons: (a) L1 was present in almost all abdominal CT images, (b) L1 was easy to locate on CT images, (c) Variations in L1’s orientation had a minimal effect on the diameter and area outcomes, and (d) Results of initial studies conducted on calculation of organ volumes, via cross-sectional imaging, showed that normalization of data according to L1-based indices accounted for body habitus. Similar to the study of Aktürk and Gunes [3], we used L1’s transverse diameter (at the largest level) as an anatomical indicator. Strong positive correlations were observed between the transverse diameter of L1 and the vessels’ diameters. The ratios of vessel-vertebral diameters increased depending on ages between 1 and 18 years. All ratios for males were similar to females. Aktürk and Gunes [3] stated that before 133 months of age (the period when L1 diameter was similar in both gender), the ratio of proximal aorta (the effective diameter of AA superior to the celiac axis) – L1 (its largest transverse diameter) was 0.41, covering 90% of study population; thus, ratio over 0.4 might indicate AA dilatation. Moreover, they calculated the ratio for distal aorta (the effective diameter of AA superior to the aortic bifurcation) to L1 between 0.28 and 0.36 in children over the age of 3 [3]. In this study, the ratio was found as 0.39 (range: 0.30–0.48) for the proximal AA, and as 0.31 (range: 0.23–0.38) for the distal AA. We think that a ratio outside these ranges may be used to determine abnormal diameters as a sign of arterial dilation or stenosis.

Investigations focused on diameters of AA, CIA, and EIA with multidetector CT scans is very limited. Aktürk and Gunes [3] measured effective diameters of proximal and distal AA, and right and left CIA, and the transverse diameters of L1 on CT images in 142 children aged 0–17 years (with age subgroups: 12–36 months, 37–84 months, 85–132 months, 133–180 months, 181–204 months). They observed no significant difference in the parameters relative to sex, but that on subgroup evaluation, L1 or CIA diameter started to vary relative to age after 133 months [3]. Hegde et al. [10] conducted on effective diameters of 110 AA and 88 thoracic aortae with CT, and found that males had larger aortic diameters compared to females (p < 0.05). Munk et al. [14] measured diameters of CIA and the infrarenal AA in 176 children aged 1–16 years using US, and found that male vessels were larger than female vessels (p < 0.001). In this work, we found that most of the parameters in males were greater than that in females (p < 0.05). However, vessel-vertebral diameters’ ratios in males were similar to females. In our opinion, the ratios may be used as a reliable indicator of arterial dilation or stenosis regardless of sex.

Conclusion

Age-specific ratios calculated in this study may be beneficial for surgeons and radiologists for the diagnosis of vascular disorders such as aortic aneurysm.