Background
Aortic dilation is an independent predictor of cardiovascular disease. The association between hypertension and aortic dilation is still controversial. Also, most previous research has investigated this relationship regarding only the aortic root, and no information is available for the ascending aorta (AscAo).
Methods
To assess AscAo dimensions in hypertensive patients, 1,027 patients with hypertension and 1,002 healthy volunteers were prospectively enrolled. Aortic diameters at four levels were measured using the leading edge–to–leading edge convention at end-diastole: the sinuses of Valsalva, sinotubular junction (STJ), AscAo, and aortic arch (AoArch), using two-dimensional echocardiography.
Results
All four diameters were significantly larger in hypertensive patients than in control subjects, with positive correlations with age, body size, and male sex. On multivariate analysis, gender and body surface area were independently associated with aortic diameters. The general linear model showed that after adjusting for age, sex, and body surface area, hypertension was positively associated ( P < .01) with higher aortic diameter at every level. Hypertension was associated with increases of 1.7 mm (95% CI, 1.2–2.1 mm) at the sinuses of Valsalva, 4.1 mm (95% CI, 3.6–4.6 mm) at the STJ, 1.6 mm (95% CI, 1.1–2.1 mm) at the AscAo, and 2.2 mm (95% CI, 1.7–2.6 mm) at the AoArch. On the basis of nomograms, an abnormally high prevalence of aortic dilation in hypertensive patients was observed for the STJ (14%) and the AoArch (7%).
Conclusions
Systematic analysis of the AscAo in hypertensive patients showed that, together with age, sex, and body surface area, hypertension is an independent factor associated with increases in all four aortic diameters and that aortic dilation occurred more frequently at the level of the STJ and AoArch.
Aortic size is an independent predictor of cardiovascular (CV) diseases, and it is highly correlated with life-threatening acute aortic syndromes. Moreover, it is now recognized that aortic dimensions are sex, age, and body size dependent in the general population. Hypertension (HTN) is traditionally regarded as a cause of aortic enlargement, even if this association is still under debate and controversial. Indeed, dilation of the ascending aorta (AscAo) in HTN may be regarded as target organ damage, in parallel with other subclinical markers of established prognostic value, such as left ventricular (LV) hypertrophy (LVH), carotid atherosclerosis, and microalbuminuria.
However, the majority of published studies have exclusively investigated the most proximal part of the AscAo, the aortic root, and little is known about the relation between HTN and aortic dimension at different levels. Accordingly, this study was designed
- •
to identify reference values and normative equations for aortic diameters at different levels of the AscAo and aortic arch (AoArch) in healthy volunteers distributed over a wide range of age;
- •
to establish the prevalence of aortic dilation in a group of stable and well-treated patients with HTN at different levels of the AscAo and AoArch;
- •
to identify reference values for aortic diameters in a population of well-treated and stable patients with HTN;
- •
to identify the effect of HTN in patients with aortic enlargement compared with control subjects; and
- •
to analyze the relationship between aortic enlargement and LVH in hypertensive subjects.
Methods
This was a prospective, single-center study. A total of 1,027 hypertensive subjects referred for follow-up for HTN management to a tertiary care hospital (Cardiology Unit, University of Brescia, Brescia, Italy) between January 2012 and December 2014 were included in the study. During the same period, 1,002 healthy volunteers were also enrolled as control subjects. The study complied with the Declaration of Helsinki and was approved by the institutional review board with a waiver of the requirement to obtain informed consent.
All patients had systemic arterial essential HTN, defined as a documented history of systolic blood pressure > 140 mm Hg and diastolic blood pressure > 90 mm Hg, according to current guidelines. All patients had been receiving antihypertensive medical therapy for ≥4 years with proven periodic blood pressure assessment.
Exclusion criteria were bicuspid aortic valve, aortic valve stenosis or regurgitation more than mild, dilated or hypertrophic cardiomyopathy, previously diagnosed coronary artery disease, obesity (body mass index > 30 kg/m 2 ), congenital heart disease, previous cardiac or aortic surgical intervention, renal disease, family history of aortic rupture, clinical characteristics suggesting a genetic predisposition to aortic disease such as Marfan syndrome, and chest or sternum deformity.
Dyslipidemia was defined in agreement with European Society of Cardiology and European Atherosclerosis Society guidelines. Diabetes mellitus was diagnosed according to American Diabetes Association guidelines. Chronic kidney disease was defined in the presence of a glomerular filtration rate < 60 mL/min/1.73 m 2 using the Modification of Diet in Renal Disease equation.
All patients and control subjects underwent blood pressure measurements and standard echocardiographic examinations, following the protocol of our echocardiography laboratory.
Blood pressure was assessed in the morning using a standard, calibrated, electronic, oscillometric sphygmomanometer (OMRON M5, Hoofddorp, The Netherlands). When taking blood pressure, the subject was at rest and sitting at a 45° angle. An appropriately sized cuff was chosen and placed at the level of the heart. The mean of three blood pressure measures was recorded. The arm with the highest sitting diastolic arterial pressure was used for all subsequent readings throughout the study. To minimize variability, care was taken to have the same staff member measure blood pressure in each individual patient, using the same equipment.
Echocardiographic examinations were performed using iE33 (Philips Medical Systems, Andover, MA) or Vivid 7 (GE Vingmed Ultrasound AS, Horten, Norway) ultrasound systems equipped with the appropriate two-dimensional transthoracic probe. Participants were evaluated in the left lateral decubitus position and images acquired from standard parasternal, suprasternal, and apical windows using second-harmonic two-dimensional imaging. Care was taken to acquire images displaying the largest aortic lumen, and acquisition was done during a breath-hold to minimize translational movements. The echocardiographic data were recorded and analyzed by a single observer (E.V.).
The diameters of the proximal aorta were measured using the leading edge–to–leading edge convention at end-diastole. The following sites were considered ( Figure 1 ): (1) the sinuses of Valsalva (SV); (2) sinotubular junction (STJ); (3) AscAo, as the maximum diameter visualized; and (4) AoArch, as the diameter in correspondence of the junction of the far wall of the left subclavian artery with the aorta. All measurements were performed in the parasternal view, perpendicular to the long axis of the aorta, except for AoArch diameter, which was measured from the suprasternal view. Patients with image quality not adequate for the assessment of any of the aortic diameters were excluded.
The echocardiographic measurements of LV end-diastolic diameter and end-systolic diameter, diastolic interventricular septal and posterior wall thicknesses, and left atrial anteroposterior diameter and area were performed offline, according to the recommendations of the American Society of Echocardiography. In addition, we calculated LV relative wall thickness according to the same guidelines. LV end-systolic volume and end-diastolic volume and LV ejection fraction were obtained using the modified biplane Simpson method. LV mass, expressed in grams, was calculated as 0.8 × [1.04 × (end-diastolic diameter + posterior wall thickness + interventricular septal thickness) 3 − (end-diastolic diameter) 3 ] + 0.6. Peak early relaxation (E) and atrial systolic (A) waves of mitral inflow were calculated from the apical four-chamber view, together with E-wave deceleration time, Doppler tissue imaging at the mitral annular level, and pulmonary venous flow to classify diastolic dysfunction pattern according to current guidelines. Systolic pulmonary artery pressure was obtained by adding right atrial pressure estimate to Bernoulli’s simplified equation on tricuspid regurgitation jet velocity by means of continuous-wave Doppler. Right atrial pressure was estimated by means of the diameter of the inferior vena cava in the subcostal view and its percentage decrease during inspiration according to guidelines. Valvular stenosis and regurgitation were evaluated according to guidelines.
Statistical Analysis
Clinical, demographic, and echocardiographic data are presented as mean ± SD for continuous variables and as absolute numbers and percentages for categorical variables. Aortic diameters were considered as absolute measures or as indexed to body surface area (BSA), according to the DuBois formula. Values obtained in women and men were compared using unpaired Student’s t tests or Fisher exact tests, as appropriate.
The relationships of aortic measurements with age and echocardiographic and anthropometric data were characterized using bivariate and multivariate linear regressions and the Pearson correlation coefficient ( R 2 ), with the inclusion of sex as a dummy variable, resulting in different intercepts for women and men.
Allometric normative equations for aortic diameters were obtained for the control group using linear regression, after log-transforming dependent and independent (age and BSA or age, height, and weight) variables and adjusting for sex. A set of nomograms displaying the upper limits of normal for each of the aortic diameter was obtained by adding 2 SDs to the mean predicted diameter. Differences in aortic diameters between patients with HTN and control subjects were tested using general linear model to account for differences in gender and adjusted for body size, age, and other factors significantly different between the two groups on univariate analysis.
To assess interobserver variability, aortic diameter measurements were performed in a subset of 50 subjects by two independent observers (E.V. and E.S.), who were blinded to each other’s results and independently selected the best image for analysis. These measurements were also repeated by the main observer (E.V.), who was blinded to the results of the previous analysis. Both inter- and intraobserver variability were expressed in terms of intraclass correlation coefficients and 95% CIs between the repeated measurements. Statistical analysis was performed using SPSS version 17.0 (SPSS, Chicago, IL).
Results
Aortic diameters could not be measured in 41 subjects (2.0%) at level of the STJ and in 44 subjects (2.2%) at the level of the AoArch; measurements were always feasible for the SV and AscAo.
Characteristics of the Control Group
Clinical, demographic, therapeutic, and echocardiographic characteristics of the control group are summarized in Table 1 and Table 2 ( left ). Men were slightly younger and had larger anthropometric measurements than women.
| Control subjects | Patients with HTN | |||||
|---|---|---|---|---|---|---|
| All ( n = 1,002) | Women ( n = 507) | Men ( n = 495) | All ( n = 1,027) | Women ( n = 391) | Men ( n = 636) | |
| Age (y) | 55.2 ± 16.5 | 58.5 ± 15.7 | 52 ± 16.7 ∗ | 65.2 ± 11.4 † | 68.1 ± 11.5 | 63.5 ± 11.0 ∗ |
| Weight (kg) | 71.7 ± 13.4 | 65.9 ± 11.1 | 77.8 ± 13 ∗ | 78 ± 15 † | 69 ± 13 | 84 ± 13 ∗ |
| Height (m) | 1.69 ± 0.08 | 1.64 ± 0.06 | 1.73 ± 0.07 ∗ | 1.69 ± 0.08 | 1.62 ± 0.06 | 1.73 ± 0.06 ∗ |
| BMI (kg/m 2 ) | 25.1 ± 3.8 | 24.4 ± 3.9 | 25.8 ± 3.5 ∗ | 27.4 ± 4.2 † | 26.6 ± 4.6 | 27.9 ± 3.8 ∗ |
| BSA (m 2 ) | 1.81 ± 0.19 | 1.71 ± 0.15 | 1.92 ± 0.17 ∗ | 1.89 ± 0.20 † | 1.73 ± 0.15 | 1.98 ± 0.16 ∗ |
| SAP (mm Hg) | 120 ± 12 | 121 ± 13 | 119 ± 11 | 132 ± 17 † | 134 ± 18 | 131 ± 16 ∗ |
| DAP (mm Hg) | 70 ± 8 | 70 ± 9 | 71 ± 8 | 79 ± 10 † | 78 ± 10 | 79 ± 9 |
| HR (beats/min) | 69 ± 10 | 70 ± 10 | 71 ± 8 | 67 ± 11 | 67 ± 11 | 66 ± 11 |
| Follow-up > 5 y | — | — | — | 569 (62%) | 206 (591%) | 363 (64%) |
| Dyslipidemia | 358 (37%) | 202 (41%) | 156 (33%) ∗ | 651 (68%) † | 254 (70%) | 397 (68%) |
| Diabetes | 15 (2%) | 5 (1%) | 10 (2%) | 262 (28%) † | 79 (22%) | 183 (31%) ∗ |
| Smokers (current and former) | 34 (3%) | 21 (4%) | 13 (3%) | 103 (11%) † | 18 (5%) | 85 (14%) ∗ |
| CKD | 43 (4%) | 20 (4%) | 23 (5%) | 97 (10%) † | 47 (13%) | 50 (9%) ∗ |
| COPD | 79 (8%) | 42 (8%) | 37 (8%) | 74 (8%) | 21 (6%) | 53 (9%) |
| ASA | — | — | — | 275 (29%) | 120 (33%) | 155 (26) ∗ |
| Nitrate | — | — | — | 35 (4%) | 19 (5%) | 16 (3%) |
| Loop diuretic | — | — | — | 83 (9%) | 44 (12%) | 39 (7%) ∗ |
| Thiazide diuretic | — | — | — | 379 (40%) | 154 (42%) | 225 (38%) |
| MRA | — | — | — | 33 (3%) | 21 (6%) | 12 (2%) ∗ |
| CCB | — | — | — | 289 (30%) | 109 (30%) | 180 (31%) |
| β-blocker | — | — | — | 507 (53%) | 209 (57%) | 298 (51%) |
| ACE inhibitor | — | — | — | 424 (44%) | 144 (39%) | 280 (48%) ∗ |
| ARB | — | — | — | 404 (42%) | 171 (47%) | 233 (40%) ∗ |
| α 2 agonist | — | — | — | 23 (2%) | 10 (3%) | 13 (2%) |
| α 1 antagonist | — | — | — | 43 (5%) | 19 (5%) | 24 (4%) |
∗ P < .05, unpaired Student’s t test or Fisher exact test, women versus men.
† P < .05, unpaired Student’s t test or Fisher exact test, control subjects versus hypertensive patients.
| Controls | Patients with HTN | |||||
|---|---|---|---|---|---|---|
| All ( n = 1,002) | Women ( n = 507) | Men ( n = 495) | All ( n = 1,027) | Women ( n = 391) | Men ( n = 636) | |
| LV EDV (mL) | 72 ± 27 | 68 ± 23 | 75 ± 30 | 103 ± 27 † | 90 ± 22 | 110 ± 26 ∗ |
| LV ESV (mL) | 27 ± 11 | 27 ± 11 | 27 ± 13 | 41 ± 16 † | 35 ± 12 | 44 ± 17 ∗ |
| LV EF (%) | 64 ± 6 | 64 ± 7 | 64 ± 5 | 61 ± 6 † | 61 ± 6 | 60 ± 6 ∗ |
| LV EDD (mm) | 46.0 ± 5.0 | 45.2 ± 5.3 | 46.8 ± 4.4 | 52.1 ± 4.6 † | 50.0 ± 4.3 | 53.4 ± 4.2 ∗ |
| LV ESD (mm) | 25.4 ± 4.4 | 25.7 ± 4.5 | 25.0 ± 4.3 | 32.1 ± 6.1 † | 29.5 ± 5.5 | 33.7 ± 5.9 ∗ |
| LV mass (g) | 141 ± 34 | 135 ± 39 | 146 ± 28 | 230 ± 92 † | 203 ± 129 | 246 ± 53 ∗ |
| LV EDV/BSA (mL/m 2 ) | 41 ± 15 | 41 ± 14 | 41 ± 16 | 54 ± 13 † | 52 ± 13 | 56 ± 13 ∗ |
| LV ESV/BSA (mL/m 2 ) | 15 ± 7 | 16 ± 8 | 14 ± 5 | 22 ± 8 † | 20 ± 7 | 22 ± 8 ∗ |
| LV EDD/BSA (mm/m 2 ) | 26.2 ± 3.6 | 26.8 ± 3.8 | 25.6 ± 3.2 | 27.8 ± 3.6 † | 29.2 ± 4.1 | 27.0 ± 3.1 ∗ |
| LV ESD/BSA (mm/m 2 ) | 14.5 ± 3.0 | 15.3 ± 3.3 | 13.7 ± 2.5 ∗ | 17.1 ± 3.3 † | 17.1 ± 3.5 | 17.0 ± 3.1 |
| LV mass/BSA (g/m 2 ) | 80 ± 20 | 80 ± 24 | 80 ± 16 | 122 ± 53 † | 118 ± 79 | 124 ± 26 |
| IVST (mm) | 9.5 ± 1.2 | 9.4 ± 1.3 | 9.6 ± 1.1 | 11.6 ± 1.8 † | 11.0 ± 2.0 | 12.0 ± 1.6 ∗ |
| PWT (mm) | 8.3 ± 1.3 | 8.1 ± 1.5 | 8.5 ± 1.0 | 10.3 ± 3.2 † | 9.9 ± 4.3 | 10.6 ± 2.2 ∗ |
| Relative wall thickness | 0.36 ± 0.06 | 0.36 ± 0.07 | 0.36 ± 0.05 | 0.40 ± 0.13 † | 0.40 ± 0.19 | 0.40 ± 0.06 |
| E/A ratio | 1.6 ± 0.6 | 1.5 ± 0.6 | 1.6 ± 0.7 | 1.0 ± 0.4 † | 1.0 ± 0.4 | 1.0 ± 0.4 |
| DT (msec) | 204 ± 60 | 200 ± 51 | 208 ± 67 | 230 ± 59 † | 222 ± 55 | 236 ± 60 ∗ |
| Diastolic pattern | ||||||
| Normal | 840 (84%) | 407 (80%) | 433 (87%) | 460 (47%) | 170 (45%) | 290 (47%) |
| Impaired relaxation | 162 (16%) | 100 (20%) | 62 (13%) | 497 (50%) | 191 (51%) | 306 (50%) |
| Pseudonormal | 0 (0%) | 0 (0%) | 0 (0%) | 24 (2%) | 9 (2%) | 15 (2%) |
| Restrictive | 0 (0%) | 0 (0%) | 0 (0%) | 4 (0%) | 4 (1%) | 0 (0%) |
| LAD (mm) | 28.8 ± 5.5 | 28.5 ± 5.7 | 29.1 ± 5.2 | 40.4 ± 5.8 † | 38.7 ± 5.5 | 41.5 ± 5.8 ∗ |
| LAA (cm 2 ) | 14.5 ± 1.1 | 14.4 ± 1.3 | 14.5 ± 1.0 | 20.6 ± 5.1 † | 19.8 ± 5.1 | 21.2 ± 5.0 ∗ |
| LAD/BSA (mm/m 2 ) | 16.5 ± 3.6 | 17.0 ± 4.0 | 16.0 ± 3.2 | 21.6 ± 3.3 † | 22.5 ± 3.2 | 21.0 ± 3.2 |
| LAA/BSA (cm 2 /m 2 ) | 8.3 ± 1.0 | 8.6 ± 1.1 | 7.9 ± 0.9 ∗ | 10.9 ± 2.8 † | 11.4 ± 3.0 | 10.6 ± 2.7 |
| sPAP (mm Hg) | 23.2 ± 3.4 | 23.2 ± 3.8 | 23.1 ± 2.9 | 28.0 ± 6.0 † | 29.4 ± 6.8 | 27.1 ± 5.2 ∗ |
∗ P < .05, unpaired Student’s t test or Fisher exact test, women versus men.
† P < .05, unpaired Student’s t test or Fisher exact test, control subjects versus hypertensive patients.
Aortic Diameters in Control Group and Nomograms
Table 3 describes aortic diameters in the control group separately for sex and age. Absolute measurements of aortic size were larger in men than in women. After indexing to BSA, diameters were similar between men and women, except for the AscAo, which was slightly larger in women than in men. Table 4 provides the results for the allometric multivariate models, in terms of regression coefficients and explained variance.
| SV (mm) | STJ (mm) | AscAo (mm) | AoArch (mm) | |
|---|---|---|---|---|
| Overall | 33.9 ± 4.2 | 25.1 ± 3.9 | 33.5 ± 4.5 | 25.1 ± 3.8 |
| Women | ||||
| <50 y | 29.8 ± 3.1 ∗ | 22.1 ± 2.9 ∗ | 29.3 ± 3.8 ∗ | 22.7 ± 3.2 ∗ |
| 50–59 y | 33.0 ± 3.0 ∗ | 24.5 ± 3.4 ∗ | 32.9 ± 3.9 ∗ | 25.1 ± 3.2 ∗ |
| 60–69 y | 33.1 ± 2.6 ∗ | 24.2 ± 2.7 ∗ | 33.8 ± 3.0 ∗ | 24.4 ± 2.8 ∗ |
| 70–79 y | 34.0 ± 3.1 ∗ | 25.4 ± 3.2 ∗ | 35.9 ± 3.3 | 25.7 ± 3.5 |
| ≥80 y | 33.1 ± 3.4 ∗ | 24.8 ± 3.7 ∗ | 35.9 ± 4.3 | 24.8 ± 3.5 |
| Overall | 32.4 ± 3.3 ∗ | 24.0 ± 3.3 ∗ | 33.0 ± 4.3 ∗ | 24.4 ± 3.4 ∗ |
| Men | ||||
| <50 y | 33.1 ± 4.9 | 24.9 ± 3.9 | 31.2 ± 4.3 | 24.3 ± 3.4 |
| 50–59 y | 36.6 ± 2.9 | 26.8 ± 3.8 | 34.4 ± 2.8 | 26.5 ± 3.8 |
| 60–69 y | 37.4 ± 3.4 | 27.6 ± 4.4 | 36.8 ± 4.0 | 28.0 ± 4.5 |
| 70–79 y | 36.9 ± 3.2 | 27.7 ± 3.5 | 36.5 ± 3.4 | 26.8 ± 3.9 |
| ≥80 y | 38.5 ± 2.7 | 27.8 ± 2.3 | 38.0 ± 2.5 | 25.1 ± 2.6 |
| Overall | 35.3 ± 4.5 | 26.3 ± 4.1 | 33.9 ± 4.6 | 25.9 ± 4.1 |
| SV/BSA (mm/m 2 ) | STJ/BSA (mm/m 2 ) | AscAo/BSA (mm/m 2 ) | AoArch/BSA (mm/m 2 ) | |
|---|---|---|---|---|
| Overall | 18.8 ± 2.3 | 14.0 ± 2.2 | 18.7 ± 2.8 | 14.0 ± 2.2 |
| Women | ||||
| <50 y | 17.8 ± 2.0 | 13.2 ± 1.9 | 17.6 ± 2.5 ∗ | 13.6 ± 2.0 ∗ |
| 50–59 y | 19.0 ± 2.0 | 14.1 ± 2.1 | 19.0 ± 2.3 ∗ | 14.5 ± 1.9 |
| 60–69 y | 19.4 ± 2.1 | 14.2 ± 1.8 | 19.8 ± 2.3 ∗ | 14.4 ± 2.0 |
| 70–79 y | 19.7 ± 2.2 | 14.8 ± 2.2 | 20.9 ± 2.3 ∗ | 14.9 ± 2.3 |
| ≥80 y | 19.7 ± 2.7 | 14.9 ± 2.6 | 21.5 ± 2.7 | 14.7 ± 2.1 |
| Overall | 19.0 ± 2.2 | 14.1 ± 2.0 | 19.4 ± 2.7 ∗ | 14.3 ± 2.1 ∗ |
| Men | ||||
| <50 y | 17.5 ± 2.5 | 13.1 ± 2.0 | 16.4 ± 2.3 | 12.9 ± 2.0 |
| 50–59 y | 19.2 ± 1.9 | 14.1 ± 2.2 | 18.0 ± 2.0 | 14.0 ± 2.2 |
| 60–69 y | 19.3 ± 2.2 | 14.3 ± 2.5 | 19.0 ± 2.6 | 14.4 ± 2.5 |
| 70–79 y | 19.9 ± 1.9 | 14.9 ± 2.0 | 19.6 ± 2.1 | 14.5 ± 2.3 |
| ≥80 y | 19.7 ± 2.7 | 14.9 ± 2.6 | 21.5 ± 2.7 | 14.7 ± 2.1 |
| Overall | 18.7 ± 2.4 | 13.9 ± 2.3 | 17.9 ± 2.7 | 13.7 ± 2.3 |
| SV | STJ | AscAo | AoArch | |||||
|---|---|---|---|---|---|---|---|---|
| β (95% CI) | R 2 | β (95% CI) | R 2 | β (95% CI) | R 2 | β (95% CI) | R 2 | |
| Constant | 0.63 | 0.48 | 0.64 | 0.39 | ||||
| Women | 5.9 (3.3 to 10.6) ∗ | 9.3 (4.2 to 20.5) ∗ | 12.5 (6.8 to 23.1) ∗ | 14.5 (6.5 to 32.2) ∗ | ||||
| Men | 6.5 (3.6 to 11.5) ∗ | 10.2 (4.6 to 22.4) ∗ | 13.0 (7.1 to 24.1) ∗ | 15.4 (6.9 to 34.2) ∗ | ||||
| Age | 0.17 (0.16 to 0.19) ∗ | 0.14 (0.11 to 0.16) ∗ | 0.23 (0.21 to 0.25) ∗ | 0.12 (0.09 to 0.14) ∗ | ||||
| Height | 0.09 (−0.02 to 0.21) | −0.04 (−0.20 to 0.12) | −0.11 (−0.23 to 0.02) | −0.10 (−0.26 to 0.06) | ||||
| Weight | 0.12 (0.08 to 0.16) ∗ | 0.14 (0.08 to 0.19) ∗ | 0.14 (0.09 to 0.18) ∗ | 0.13 (0.07 to 0.19) ∗ | ||||
| Constant | 0.62 | 0.47 | 0.62 | 0.39 | ||||
| Women | 13.6 (12.5 to 14.8) ∗ | 11.1 (9.8 to 12.5) ∗ | 11.1 (10.1 to 12.2) ∗ | 12.5 (11.1 to 14.2) ∗ | ||||
| Men | 14.8 (13.5 to 16.1) ∗ | 12.1 (10.7 to 13.7) ∗ | 11.5 (10.5 to 12.6) ∗ | 13.3 (11.8 to 15.0) ∗ | ||||
| Age | 0.18 (0.16 to 0.20) ∗ | 0.15 (0.13 to 0.18) ∗ | 0.24 (0.22 to 0.26) ∗ | 0.13 (0.10 to 0.16) ∗ | ||||
| BSA | 0.27 (0.19 to 0.34) ∗ | 0.26 (0.15 to 0.37) ∗ | 0.22 (0.14 to 0.30) ∗ | 0.23 (0.12 to 0.33) ∗ | ||||
When considering height and weight separately ( Table 4 , top ), the models clearly showed that after the adjusting for gender, age and weight were significantly associated with aortic size. Interestingly, the association between height and aortic diameters was observed not significant. As expected, constants were different between women and men, larger for the latter. Indexing to BSA rather than to height and weight separately ( Table 4 , bottom ) did not lead to any substantial decrease in the R 2 value of the model, and suggested associations between aortic diameters, gender, age, and body size were similar to those found when height and weight were considered separately ( Table 4 , top ). Normative equations on the basis of model coefficients for allometric scaling are described in the footnote to Table 4 .
The upper limits of normal aortic diameters as a function of BSA and age, separately for men and women, are presented in Figures 2 and 3 , respectively.
