Aortoiliac diameter and length in a healthy cohort

Objective Diameter is currently the only screening and diagnostic criterion for asymptomatic aneurysms. Therefore, aortic and lower-extremity arterial diameter has diagnostic, therapeutic, and prognostic importance. We aimed to determine aortic and lower-extremity arterial reference diameters in a general population and compare them according to age, sex, and other characteristics. Methods We evaluated consecutive 3,692 patients who underwent computed tomography as part of a general health checkup from 2015–2019 in a single tertiary center. Aortic and lower-extremity arterial diameters and the most important factor related to arterial diameters were evaluated. Results The mean diameter of the abdominal aorta was 17.490 ± 2.110 mm, while that of the common iliac artery was 10.851 ± 1.689 mm. The mean diameter of the abdominal aorta was 18.377 ± 1.766 mm in men and 15.884 ± 1.694 mm in women. Significant intersex differences were observed for all mean diameters and lengths. Multilinear regression analysis showed that age, sex, and body surface area impacted mean diameters of all measured sites except aorta and common iliac artery length. Between male and female patients matched for body surface area, there were significant intersex differences for all measured sites, except for common iliac artery length. Conclusions The mean diameter of the abdominal aorta in this healthy cohort was 17.490 ± 2.110 mm overall, 18.377 ± 1.766 mm in men, and 15.884 ± 1.694 mm in women. Arterial diameter increased with male sex, older age, and increased body surface area, and aortic diameters were larger in men than in women with the same body surface area.

Introduction Currently, diameter is the only screening and diagnostic criterion for asymptomatic aneurysms. Abdominal aortic aneurysm (AAA) is defined as a 50% or greater increase in infrarenal aortic diameter (IAD) or infrarenal aorta with a maximum diameter � 3.0 cm [1][2][3]. Aneurysm size is one of the strongest predictors for risk of rupture, with a markedly increased risk when aneurysm diameters are greater than 5.5 cm [4,5]. Therefore, aortoiliac arterial diameter has diagnostic, therapeutic and prognostic importance.
Women have up to a four-fold higher risk of AAA rupture than men at any given aneurysm diameter [6]. The Joint Council of the American Association of Vascular Surgery and the Society for Vascular Surgery have suggested a lower diameter threshold for AAA repair in women [7]. One hypothesis is that because women generally have a smaller body and vascular size than men, an aneurysm of a certain size in a woman represents a greater relative dilatation of the aorta compared with the same aneurysm in a man [8]. In order to apply the concept of relative expansion according to sex or body size, the reference diameter is of clinical importance. There are published reference ranges for the aorta and the lower-extremity vessels using ultrasound or contrast-enhanced computed tomography (CT) [9][10][11]. However, there are few papers on Asian populations, the body sizes of whom are relatively small compared to Westerners. Considering its clinical importance and lack of sufficient data, we purposed to measure aortoiliac and lower-extremity arterial reference diameters in an Asian healthy population. In addition, we determined whether body size was a significant factor for aortoiliac diameter and whether there was an intersexual difference in the diameter when body size was similar.

Methods
We retrospectively evaluated patients who underwent CT for general health checkups from 2015 to 2016. Patients with aneurysms or atherosclerotic plaques with or without calcification or patients whose arterial centerline could not be obtained were excluded from the analysis. This study was approved by _____ Institutional Review Board (No. 2016-0232) and waived the need for informed consent because of the retrospective nature of the study and the lack of information on the participant's identification. This study complies with the Declaration of Helsinki.
All imaging examinations were performed using a multi-slice CT scanner (Lightspeed VCT; GE Healthcare, IL, US). Parameters for the acquisitions were 5-mm slice thickness, 120 KVp, and 215-360 mA tube current. Imaging was initiated after the administration of low osmolar iodinated contrast agent (Iopamiro 2 mL/kg; iodine concentration, 320 mg/mL). Softtissue window settings with a width of 300 HU and a center of 50 HU were applied. This sizing was performed using Endosize (Therenva, Rennes, France), a 3D sizing software tool that measures diameters perpendicular to the long axis of the arteries. Lengths and diameters taken on the vessel centerlines were automatically obtained after a simple interactive step consisting of a 3D point picking sequence.
The measured site is depicted in Fig 1. Aortic diameter was measured just below the superior mesenteric artery (SMA), lowest renal artery, and at the bifurcation. Mean aortic diameter from three sites was used in the regression analysis. The diameter of the common iliac artery (CIA) was measured at the midpoint between the aortic and iliac bifurcation and at the broadest point, and the external iliac artery (EIA) diameter was measured at the iliac bifurcation. The diameter of the common femoral artery was measured at the level of the femoral bifurcation. The diameter of each artery was measured with the outer diameter of the artery perpendicular to the arterial centerline. Aortic length was measured between the lowest renal artery and the bifurcation. CIA length was measured between the aortic bifurcation and the iliac bifurcation. Iliac artery length was measured between the aortic bifurcation and the femoral bifurcation. Measurements using Endosize were made by four vascular surgeons. To test the reliability, all four of the examiners randomly measured the data of the selected 106 patients using a random number generation function in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA).
Clinical information was obtained from the questionnaires and measurements from the general health checkup database, including height, weight, history of smoking, hypertension, and diabetes. History of smoking was defined as current or former smokers based on patientprovided information. Body mass index (BMI) was calculated by dividing the weight in kilograms by the square of the height in meters. Body surface area (BSA) was calculated using the Mosteller formula [12]. Diabetes was defined as fasting plasma glucose (FPG) levels � 7.0 mmol/L or glycated hemoglobin (HbA1c) levels � 6.5%. In addition, individuals taking antidiabetic medication were considered to have diabetes. Hypertension was defined as systolic and/or diastolic blood pressure � 140/90 mmHg and/or taking antihypertensive medication. The CT scan, laboratory tests, and questionnaire evaluations were performed on the same day or the following day.

Statistical analysis
Quantitative and qualitative variables were summarized separately by descriptive statistics. For quantitative variables, an independent sample t-test or one-sample t-test was used to assess differences in the diameters. Inter-observer reliability of the measurements was assessed using the intraclass coefficient correlation (ICC), and complete agreement was defined as 1.0. A generalized linear model with stepwise selection was fit to assess the associations between baseline characteristics and the diameters of the lower extremities after normality testing (Kolmogorov-Smirnov test, Cramer-von Mises, and Anderson-Darling). Men and women with the same BSA were extracted using R software version 4.0.2 (R Development Core Team, 2006). Where multiple patients were present for one BSA value, the mean values of each sex were used as representative values. Comparison of aortic diameters was performed between matched men and women using paired t-tests. Our data were compared with those of previous studies on aortic diameters using one-sample t-test. p values < 0.05 were considered significant. The statistical analysis was performed using SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA) and SPSS version 23.0 software (Armonk, NY, USA).

Results
A total of 3,692 subjects (35.6% female) were included in the analysis (Fig 2). Baseline characteristics are summarized in Table 1

Mean diameters and lengths
Mean diameters and lengths are shown in Table 2. The mean diameter of the abdominal aorta was 17.490 ± 2.110 mm, while that of the CIA was 10.851 ± 1.689 mm. No patients had an aortic diameter � 3 cm. The mean diameter of the aorta was 18.377 ± 1.766 mm in men and 15.884 ± 1.694 mm in women (Fig 3). The mean diameter of the CIA was 11.436 ± 1.512 mm in men and 9.793 ± 1.464 mm in women. For all mean diameters and lengths, significant differences between men and women were observed (all p values were less than 0.001, except for both CIA lengths [P = .048 for right side, P = .034 for left side]).

Factors affecting diameters and lengths
The linear and multilinear regression analyses for each variable were performed for each diameter and length (Tables 3 and 4). The results of multilinear regression showed that age, sex, and BSA were related to the mean diameters and lengths of all measured sites except for CIA length. In the linear regression model, BSA was most explanatory for diameters with the highest R 2 values; 0.249 for the infrarenal aorta, 0.277 for the lower abdominal aorta near the bifurcation, 0.217 for the CIA, and 0.254 for the iliac artery. The female sex variable further exhibited high R 2 values; 0.249 for the infrarenal aorta, 0.256 for the lower abdominal aorta near the bifurcation, 0.257 for the CIA, and 0.296 for the iliac artery. Results of the multilinear regression analysis with significant variables in the linear regression analysis showed that age, sex, and BSA were related to mean diameters of all measured sites. The R 2 value of the reduced model regarding only age, sex, and BSA was not significantly different from that of the full model including all possible variables that were significant in the linear regression; F (

Difference between men and women in diameter and length when matching BSA
When BSAs of men and women were matched, a total of 462 pairs were obtained (BSA range, 1.4-2.2). There was a significant difference in diameters between matched men and women (P  Fig 4B).

Discussion
Aorta and iliac artery size are considered essential in the diagnosis of aneurysm and the prediction of future aneurysmal rupture. The generally accepted definition of arterial aneurysm is a focal and persistent vessel dilation of 150% or more versus the expected normal diameter of the artery in question [1]. An association between age, sex, and body surface area and the normal diameter of the artery was proposed, but simpler definitions were then suggested since the effect on aortic diameter was not substantial [13]. Previous studies demonstrated that the normal IAD is slightly less than 20 mm in elderly men [1,14]. Accordingly, AAA in this population was defined as an IAD � 30 mm [15]. AAA is usually asymptomatic until rupture, and mortality can reach 85-90% in cases of rupture [16]. Several large studies have shown that screening for this condition reduces aneurysm-related mortality [17,18], and it is recommended in European guidelines for all elderly

PLOS ONE
Lower extremity arterial diameters men and in American guidelines for elderly women and men with a history of tobacco use [19,20]. The frequency of follow-up imaging depends on initial artery diameter, considering the increased risk of rupture [20]. In addition to a large initial aneurysm diameter, female sex is a known independent risk factor associated with rupture as well as a worse outcome [21][22][23]. Interestingly, rupture occurs at aneurysm diameters of 5 to 10 mm smaller in women than in men [24]. One of the potential reasons is that an aneurysm of a given diameter in women with relatively smaller aortas due to smaller body size represents a greater relative dilatation and thus more advanced disease of the aorta than an aneurysm of the same diameter in men [25]. Therefore, it seems crucial that we identify the reference value of the IAD, particularly according to sex. In our study on healthy Asian cohorts, mean aortic diameter was 17.490 ± 2.110 mm. When divided by sex, mean diameter of the aorta was 18.377 ± 1.766 mm in men and 15.884 ± 1.694 mm in women. The difference in mean value between them was 2.493 mm, larger than the previous report of 1.4 mm from the Veterans Affairs Cooperative Study [26]. As BSA was significantly larger in men (P < .001) and was the strongest factor that affected vessel diameter in our regression model (P < 0.001), we matched BSA to determine whether a difference in the diameter between sex was derived from BSA difference. Even after BSA was corrected, the difference in the diameter between men and women remained in all measured diameters (P < 0.001). Therefore, considering intersex differences in the diagnosis of diseases related to arterial diameter seems necessary.
The comparison of our data with those of previous reports from other countries using a one-sample t test revealed significant differences. The Veterans Affairs Cooperative Study reported that the aortic diameters measured below and above the renal arteries on ultrasonography for male patients were 20 ± 3 and 21 ± 3 mm, respectively [26]. When we compared those values with our data on diameters measured at the levels of the SMA and lowest renal artery, our data were significantly smaller than both diameters (p < 0.001 for both).
The mean infrarenal abdominal aortic diameters on CT scan in the Framingham Heart Study for men and women were 19.3 ± 2.9 and 16.7 ± 1.8 mm, respectively, which were significantly larger than our values (p < 0.001 for both) [27]. The mean aortic diameter at the bifurcation level was 18.7 ± 2.7 mm for men and 16 ± 1.7 mm for women, significantly larger than our values (p < 0.001 for both) [27]. In a study of a Turkish population, on ultrasonography, the mean subdiaphragmatic aortic diameters were 18 ± 3 mm for women and 19 ± 4 mm for men, while the mean aortic diameters at the bifurcation level were 15 ± 3 mm for women and 16 ± 4 mm for men [28]. Compared with the diameter at the level of the SMA and bifurcation, the mean diameter in women was significantly smaller than that in men in our study (p < 0.001 for all). In an Indian study, the mean diameters of the suprarenal and infrarenal abdominal aortas measured at the T12 and L3 vertebral levels on CT scan were 19.0 ± 2.3 and 13.8 ± 1.9 mm for men and 17.1 ± 2.3 and 12.0 ± 1.6 mm for women, respectively [29]. Compared with the diameter at the level of the SMA and bifurcation, all the values were significantly larger in our study (p < 0.001 for all). In a Chinese population, the inner diameter of the infrarenal  [30]. These results demonstrate differences among geographic regions. However, this finding is limited because the comparisons did not involve equal modalities and included anatomical levels with different measurements. An aneurysm diameter measured on standard axial CT is generally > 2 mm larger than when measured on ultrasonography [20]. Moreover, the actual difference was � 2.5 mm. For example, the difference between the data from our study and those from the Framingham Heart Study was < 1 mm (0.92 mm for men and 0.81 mm for women) despite the statistical significance [27]. The clinical significance requires reevaluation with regard to the actual risk of rupture and the establishment of different surveillance criteria. This study has some limitations. A potential source of bias in our study was selection bias because the study population consisted of people who prioritize their health status maintenance and included CT in their health checkups. However, we acknowledge that the selection bias in this study would be much lower than that if the data of CT scans conducted for patients with certain diseases were included. Moreover, we tried to overcome the selection bias by using a large sample size. Second, an unequal number of men and women were included since consecutive persons were enrolled. Despite these limitations, the strength of our study was that we used data from a healthy population without atherosclerotic steno-occlusive disease on CT scans. Because the artery tends to gets larger with the progression of the atherosclerotic disease; thus, the reference diameter needs to be evaluated from the normal population. Under the Korean health insurance system, people can opt to undergo a CT scan as part of their medical checkup. This is why we could obtain data from normal subjects for this analysis. Second, we investigated intersex difference in diameters with excluding the effect of BSA based on the large sample size. Lastly, we used 3D reconstruction to extract a centerline, avoid a parallax error, and increase reproducibility. When we evaluated intraobserver variability, reproducibility proved relatively efficient for obtaining reliable sizing data.
In conclusion, we obtained the reference diameters of the abdominal aorta of 17.490 ± 2.110 mm overall, 18.377 ± 1.766 mm in men, and 15.884 ± 1.694 mm in women in a Korean healthy cohort, which was smaller than Westerners. Arterial diameter increased with male sex, older age, and increased BSA, and the aortic diameters were larger in men than in women with the same BSA.