Methods

From June 2007 to December 2013, a sample of 1,142 consecutive apparently health adults were referred to echocardiographic laboratories of the Department of Cardiology and Emergency Medicine of San Antonio Hospital, San Daniele del Friuli, Udine, Italy and Division of Cardiology, “Cava de’ Tirreni-Amalfi Coast,” Heart Department, University Hospital of Salerno, Italy, for the purpose of present study. The subjects underwent voluntary (or for work ability assessment) full screening for cardiovascular disease including a questionnaire about medical history, use of medications, cardiovascular risk factors, and lifestyle habits (alcohol intake, smoking, and physical activity). Physical examination (height, weight, heart rate, and blood pressure [BP]) and clinical assessment were conducted according to standardized protocols by trained and certified staff members. BSA was calculated according to the DuBois formula [0.20247 × height (m) 0.725 × weight (kg) 0.425]. Three BP measurements were obtained from the right arm with a mercury manometer, and the results were averaged to determine systolic and diastolic BPs.

Exclusion criteria were coronary artery disease, systemic arterial hypertension, diabetes mellitus, valvular or congenital heart disease, bicuspid aortic valve, congestive heart failure, cardiomyopathies, sinus tachycardia, use of illicit drugs, elite athletes, and inadequate echocardiographic image quality. According to these criteria, 76 subjects were excluded: 2 for coronary artery disease, 10 for systemic arterial hypertension, 4 for diabetes mellitus, 8 for body mass index >30 kg/m ² , 7 for more than mild valvular insufficiency (3 mitral, 2 aortic, and 2 tricuspid), 2 for aortic stenosis, 4 for bicuspid aortic valve, 1 for hypertrophic cardiomyopathy, 1 for AR dilation, 1 for dilated cardiomyopathy, 8 for the use of pharmacologic treatment (hyperlipidemia, breast cancer, thyroid, gout, and prostate disease), 20 elite athletes, and 8 for inadequate echocardiographic image quality. In addition, 23 of the initial subjects investigated refused to be included in the echocardiographic protocol. Our final study population therefore consisted of 1,043 healthy subjects (mean age 44.7 ± 15.9 years, range 16 to 92 years, 503 men [48%]). The study was approved by the institution’s Ethics Board, and informed consent was obtained from the participants.

Standardized TTE and Doppler examinations were performed with market available equipment in all the subjects (Aloka α10; Aloka, Tokyo, Japan and Vivid 7; GE Healthcare, Milwaukee, Wisconsin). Specific views included the parasternal long- and short-axis views; apical 4-, 2-, and 3-chamber views; and subcostal views including respiratory motion of the inferior vena cava. Pulsed and continuous-wave Doppler interrogations were performed on all 4 cardiac valves. All studies were reviewed and analyzed off-line by 2 independent observers. Specific measurements were made by the average of 5 cardiac cycles. M-mode measurements, performed in the parasternal long-axis view with the patient in the left lateral position, included left ventricular internal diameter in diastole and systole, interventricular septum in diastole, and posterior wall in diastole. Left ventricular (LV) mass was calculated by the Penn convention and indexed for BSA. The LV ejection fraction was calculated by the Simpson equation in the apical 4- and 2-chamber views. Valvular regurgitation was quantified from color Doppler imaging and categorized as absent, minimal (within normal limits), mild, moderate, or severe. Doppler-derived LV diastolic inflow was recorded in the apical 4-chamber view by placing the sample volume at the tip level of the mitral valve leaflets. LV diastolic measurements included E and A peak velocities (m/s) and their ratio as well as E-wave deceleration time (ms). Two-dimensional measurements of the AR were made at end-diastole in parasternal long-axis views at 4 levels: (1) annulus (defined echocardiographically as the hinge points of the aortic cusps), (2) sinuses of Valsalva, (3) sinotubular junction, and (4) proximal ascending aorta. Measurements were obtained perpendicular to the long axis of the aorta using the leading edge technique in views showing the largest aortic diameters.

Data analysis was performed using SYSTAT, version 12 (University of Illinois, Chicago, Illinois). Demographics and clinical characteristics, LV dimensions, and aortic diameters, both absolute and relative to BSA, are presented as mean ± SD and were tested by unpaired t test to evaluate differences between genders. Aortic dimensions were expressed as mean, median, and twenty-fifth and seventy-fifth percentiles; the aortic dimension above the ninety-fifth percentile of the overall distribution was used as cutoff for the upper limit. An unpaired t test was performed to evaluate differences between genders. Allometric equations were used to determine the relations of aortic diameters with weight and height. Adjustment for height and weight in the regression models avoided the assumption made in indexing to certain parameter of body size (e.g., BSA), while achieving the same purpose of accounting for differences in body size among participants. To determine whether we were allowed to calculate common scaling exponents for the whole group of men and women, gender was included as a dummy variable in the analysis. The following model was fitted: log(diameter) = log a + b log(weight) + c log(height) + d sex (coded 1 for men and 2 for women) or, in its exponential form: diameter = a × weight ^b × height ^c × sex ^d . The partial correlation test by the Pearson method was used to assess clinically relevant variables with p <0.05, which were then incorporated into the multivariate model. Three models were developed in multiple regression analysis to explain aortic dimensions. Model A included age and gender; model B included age, gender, and BSA; model C included age, gender, weight, and height. Multiple regression analysis for aortic diameters in relation to age, gender, body mass index, weight, and height was applied. Residuals of observed aortic diameters versus those predicted by multivariate models were calculated, and their relations to age, gender, body size (weight, height, or BSA) were assessed. Reproducibility of aortic measurements was determined in 50 subjects randomly selected. The interobserver and intraobserver variabilities were examined using both Pearson bivariate 2-tailed correlations and Bland-Altman analysis. Two-tailed p value <0.05 was considered statistically significant.

Results

The studied population included 1,043 healthy subjects: 503 men and 540 women. Women were slightly older, lighter, and smaller than men. They had lower BP but higher heart rate. On TTE, they had smaller LV dimensions and mass but similar E/A ratio ( Table 1 ). The intraobserver variability analysis revealed Pearson correlations as follows: r = 0.90 (p <0.0001) for the aortic annulus, r = 0.97 (p <0.0001) for the sinuses of Valsalva, r = 0.96 (p <0.0001) for the sinotubular junction, and r = 0.86 (p <0.0001) for the maximum diameter of the proximal ascending aorta. The Bland-Altman analysis gave a 95% confidence interval of 4.1 ± 1.1% for the aortic annulus, 3.9 ± 1.1% for the sinuses of Valsalva, 4.1 ± 1.1% for the sinotubular junction, and 4.8 ± 1.3% for the maximum diameter of the proximal ascending aorta. For interobserver variability, Pearson correlations were as follows: for the aortic annulus, r = 0.88 (p <0.0001); for the sinuses of Valsalva, r = 0.96 (p <0.0001); for the sinotubular junction, r = 0.95 (p <0.0001); and for the maximum diameter of the proximal ascending aorta, r = 0.84 (p <0.0001). The Bland-Altman analysis gave a 95% confidence interval of 5.1 ± 1.1% for the aortic annulus, 4.1 ± 1.2% for the sinuses of Valsalva, 4.3 ± 1.1% for the sinotubular junction, and 5.1 ± 1.5% for the maximum diameter of the proximal ascending aorta. The absolute aortic diameters were significantly greater in men than in women at all levels, whereas BSA-indexed aortic diameters were greater in women ( Table 2 ). No significant gender differences were registered for sinuses of Valsalva, sinotubular junction to annulus diameter ratios, whereas ascending aorta to annulus diameter ratio was higher in women ( Table 3 ). BSA-indexed AR diameters stratified by age decades and gender are reported in Table 4 . There was a linear correlation between the aortic diameters (absolute and indexed values) and their ratios with age in both genders, except for the aortic annulus (p = 0.0001; Figures 1 and 2 ). Allometric scaling approach for normalization was applied. Exponents b and c (respectively for weight and height) were found to be significantly different than unity for all 4 AR diameters and gender exponent ( Table 5 ). Because the correlation coefficients between aortic diameters, height, and weight raised to the specific allometric exponent were similar to those of aortic diameters versus baseline height and weight, no exponential values were included in the multivariate models. Aortic diameters were independently associated with age, gender (model A), and BSA (model B); weight and height did not have any additional significant impact on aortic dimension (model C; Table 6 ). There were no significant residual linear relations of age, gender, body size measurements (weight, height, or BSA) with the differences between observed and predicted aortic diameters.

Table 4

Indexed body surface area aortic diameters, stratified by age and gender

Age (yrs)	Annulus			Sinuses of Valsalva			Sinotubular Junction			Proximal Ascending Aorta
	Men	Women	p	Men	Women	p	Men	Women	p	Men	Women	p
16–29 (M 117/F 97)	10.8 ± 1.3 (8.0–14.8)	11.1 ± 1.2 (8.5–14.9)	0.6	15.2 ± 1.6 (11.0–19.0)	15.7 ± 1.7 (11.0–20.9)	0.03	12.6 ± 1.7 (7.8–18.1)	13.3 ± 1.7 (8.8–18.2)	0.03	13.3 ± 1.9 (8.6–18.7)	14.5 ± 1.8 (10.3–19.1)	0.001
30–39 (M 92/F 89)	10.5 ± 1.1 (8.6–13.5)	11.1 ± 1.2 (6.5–14.6)	0.001	15.6 ± 1.7 (11.8–19.3)	16.4 ± 1.7 (9.4–20.9)	0.003	13.2 ± 1.6 (9.8–19.5)	14.0 ± 1.6 (8.7–18.7)	0.001	14.0 ± 1.4 (10.9–17.5)	15.5 ± 1.7 (9.0–19.7)	0.0001
40–49 (M 110/F 118)	10.9 ± 1.2 (8.2–14.5)	11.1 ± 1.0 (9.0–14.9)	0.2	16.6 ± 1.8 (11.2–22.4)	17.2 ± 2.0 (11.0–24.5)	0.01	14.0 ± 1.8 (10.4–18.4)	14.6 ± 1.7 (10.1–19.6)	0.007	15.2 ± 1.9 (11.3–20.8)	16.5 ± 1.8 (10.8–21.6)	0.0001
50–59 (M 106/F 137)	10.9 ± 1.1 (8.3–14.3)	11.2 ± 1.1 (8.8–13.9)	0.01	17.1 ± 1.8 (13.6–22.3)	17.7 ± 1.8 (13.0–23.0)	0.012	14.8 ± 1.6 (11.4–19.4)	15.3 ± 1.8 (11.5–20.1)	0.02	16.3 ± 1.8 (12.1–21.9)	17.1 ± 1.7 (12.0–21.4)	0.001
≥60 (M 78/F 97)	11.3 ± 1.3 (8.7–14.4)	11.7 ± 1.1 (9.4–14.6)	0.07	18.5 ± 2.35 (13.9 ± 24.0)	18.4 ± 2.0 (15.0–24.3)	0.7	15.5 ± 2.1 (10.8–20.7)	15.6 ± 1.7 (12.0–20.9)	0.7	17.3 ± 2.13 (12.5–21.7)	18.2 ± 1.8 (13.4–22.7)	0.006

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Ability of Low Antihypertensive Medication Adherence to Predict Statin Discontinuation and Low Statin Adherence in Patients Initiating Treatment After a Coronary Event
2. Impact of Atrial Fibrillation and Heart Failure, Independent of Each Other and in Combination, on Mortality in Community-Dwelling Older Adults
3. Efficacy and Cost-Effectiveness of Dabigatran Etexilate Versus Warfarin in Atrial Fibrillation in Different Age Subgroups
4. Usefulness of N-Terminal Pro-Brain Natriuretic Peptide Levels to Predict Success of Weaning from Intra-Aortic Balloon Pumping

Stay updated, free articles. Join our Telegram channel

Join