Scientific letter Age-independent aortic dimensions in adolescent athletes: a practical approach using allometric scaling

To the Editor,

In some cases of congenital aortic anomalies or potentially undiagnosed tubular dilations, structural weakness of the ascending aorta predisposes to pathological dilation during prolonged periods of increased wall stress. Indeed, aortic rupture is an uncommon but important finding in all series of sudden cardiac death in young, ostensibly healthy, athletes, even in those undergoing preparticipation screening.¹ Nevertheless, while aortic root values have been reported to be larger in pediatric athletes than in age-matched nonathletes,² no normative values are available for aortic dimensions in healthy pediatric/adolescent athletes in the 4 main planes––aortic annulus, sinuses of Valsalva, sinotubular junction and ascending aorta––which is distinct from the situation in adult athletes.³ Clearly, having access to these data would aid clinicians in detecting pathologic abnormalities in young athletes.

Aortic root diameters are traditionally reported in absolute values; however, given the association between cardiac dimensions and anthropometric measures (ie, body surface area [BSA]), particularly in children, it has been proposed that allometric rather than linear scaling can provide cardiac measures that are body size-independent and would allow within- and between-subject comparisons along pubertal development.⁴ Of note, allometric scaling has been recently proposed for athletes across a relatively wide age range (12-35 years),⁵ but no specific subgroup analyses for pediatric/adolescent athletes were provided, and results were only available for the Valsalva sinus using M-mode ultrasonography. To address this issue, in the present study we report body size-independent aortic dimensions in pediatric/adolescent athletes.

We followed a cross-sectional, single-center study design. Participants were pediatric/adolescent athletes who underwent thorough preparticipation screening at the sports medicine center of the autonomous community of Madrid (Spain) during a recent 10-year period. Inclusion criteria were age 10 to 18 years and competing at national or international level. Exclusion criteria were having a bicuspid aortic valve, aortic dysplasia or prolapse, moderate or severe aortic regurgitation, high-gradient aortic stenosis, blood pressure > 95th percentile for the corresponding age and height, exercise-induced bronchoconstriction or hypertension, cardiomyopathy, left-right shunt, acute pericarditis, supra-aortic trunk disease, or second-degree atrioventricular block. Both athletes and their parents or legal representatives provided written consent and the study was approved by the local ethics committee.

We measured participants’ height and weight to the nearest 0.1 cm and 0.1 kg, respectively, and BSA. Echocardiographic evaluations were performed by the same sports cardiologist (AB, 30+ year experience) using a Siemens Sonoline G50 (Siemens Medical Solutions, Ann Arbor, United States) or a Mindray DC-70 (Shenzhen Mindray Bio-Medical Electronics, China) instrument with a 2 to 4 MHz phase array transducer. Aortic diameters were measured in the aforementioned 4 planes in 2D parasternal long-axis view at end-diastole (using the average of 3 consecutive cycles) with the inner-inner convention, as previously performed in young adult elite athletes.³

We used Pearson's correlation analysis to explore the association between aortic dimensions and age and anthropometric measurements. Thereafter, we studied the allometric relationship between aortic dimensions and BSA by nonlinear regression using the Levenberg-Marquardt algorithm,⁶ also known as the ‘damped least squares method’, which is used to solve nonlinear least squares minimization. Minimization problems arise especially in least squares curve fitting, which is the process of constructing a curve or mathematical function that has the best fit to a series of data points. Thus, allometric-indexed aortic dimension = nonadjusted aortic dimension (mm)/(BSA, in m²)^β, where β was determined with a confidence interval of 95%. Statistical analyses were performed with Stata 14.0 (StataCorp, College Stattion, United STates) with α = 0.05.

Of the 637 athletes aged 10 to 18 years evaluated, 529 met the inclusion criteria (table 1). The athletes were engaged mostly in water polo (16%), swimming (16%), tennis (10%), synchronized swimming (6%), field hockey (13%), soccer (5%), and badminton (5%) competitions. Results showed that all 4 nonadjusted aortic dimensions positively and significantly correlated with age, height, weight, and BSA (all P < .001; table 2). However, significance for the correlation between each of the 4 dimensions and age was lost if the former were allometrically corrected for BSA (P > .3). All β-values for BSA were ∼0.5 [0.486-0.513], and for practicality we consistently used an age-independent β-value of 0.5 (which is actually equivalent to square root) for all sex-specific normative values of aortic root dimensions allometrically normalized by BSA (table 2).

Table 1.

Main characteristics of participants by sex

	Boys(n = 299)	Girls(n = 230)	P
Demographics and anthropometric variables
Age, y	14.9 ± 1.9	14.5 ± 1.9	.017
Height, cm	169 ± 12	161 ± 8	< .001
Weight, kg	58.7 ± 13.4	52.2 ± 10.6	< .001
BSA, m^2	1.67 ± 0.24	1.53 ± 0.18	< .001
Systolic blood pressure, mmHg	111 ± 10	107 ± 10	< .001
Diastolic blood pressure, mmHg	64 ± 9	62 ± 8	.003
Resting heart rate, bpm	64 ± 11	66 ± 11	.189
Competition experience, y	5.9 ± 2.6	5.5 ± 2.6	.119
Training regime, h/wk	15 ± 8	17 ± 9	< .001
Echocardiographic variables
Ventricular septum, mm	8.7 ± 1.3	7.9 ± 1.0	< .001
Anteroposterior left ventricle dimensions, mm	47.8 ± 4.8	44.8 ± 3.9	< .001
Left ventricle posterior free wall, mm	8.5 ± 1.2	7.7 ± 0.9	< .001
Left ventricle end-diastolic volume/BSA, mL/m^2	64.4 ± 10.2	60.1 ± 9.1	< .001
Left ventricle ejection fraction, %	66 ± 7	66 ± 7	.998
Anteroposterior left atrium dimensions, mm	32.2 ± 4.7	30.4 ± 4.8	< .001
Superior-inferior left atrium dimensions, mm	44.2 ± 6.5	42.6 ± 5.6	.003
Superior-inferior right atrium dimensions, mm	46.8 ± 5.9	44.4 ± 5.3	< .001
Aortic annulus, mm	22.9 ± 2.4	21.1 ± 2.1	< .001
Sinuses of Valsalva, mm	27.2 ± 3.0	24.5 ± 2.4	< .001
Sinotubular junction, mm	22.6 ± 2.7	20.9 ± 2.3	< .001
Proximal ascending aorta, mm	23.8 ± 2.8	22.2 ± 2.4	< .001

BSA, body surface area.

Data are shown as mean ± standard deviation. Between-sex comparisons were performed with the unpaired Student t test.

Table 2.

Pearson correlations results and age-independent aortic dimensions allometrically corrected by body surface area for adolescent athletes aged 10 to 18 years.

Aortic dimensions	Pearson correlations				Allometric scaling				Boys(n = 299)			Girls(n = 230)
	Age	Height	Weight	BSA	β exponent	Pearson correlation with age of allometry- scaled aortic dimensions (ie, divided by BSA^β)	Independence of allometry- scaled dimensionwith age	Allometrically corrected aortic dimensions	Mean ± SD	5th percentile	95th percentile	Mean ± SD	5thpercentile	95thpercentile
Annulus, mm	0.480P < .001	0.649P < .001	0.621P < .001	0.659P < .001	0.513	0.025P = .562	Yes	Annulus/BSA^0.5,mm/m^2	17.8 ± 1.3	15.7	20.3	17.1 ± 1.5	14.7	19.6
Sinuses of Valsalva, mm	0.433P < .001	0.601P < .001	0.573P < .001	0.608P < .001	0.508	0.006P = .890	Yes	Sinuses of Valsalva/BSA^0.5,mm/m^2	21.2 ± 1.9	18.1	24.6	19.9 ± 1.7	17.2	23.0
Sinotubular junction,mm	0.436P < .001	0.548P < .001	0.534P < .001	0.563P < .001	0.486	0.040P = .353	Yes	Sinotubular junction/BSA^0.5,mm/m^2	17.5 ± 1.7	15.0	20.4	16.9 ± 1.7	14.4	19.6
Proximal ascending, mm	0.423P < .001	0.569P < .001	0.553P < .001	0.582P < .001	0.558	0.050P = .253	Yes	Proximal ascending/BSA^0.5,mm/m^2	18.5 ± 1.8	15.9	21.6	17.9 ± 1.8	15.2	21.2

BSA, body surface area; SD, standard deviation.

Data are shown as mean ± standard deviation.

A number of aortic abnormalities predisposing to pathological dilation during exercise can increase the risk of sudden cardiac death in young, apparently healthy, athletes.¹ Accordingly, the assessment of aortic dimensions in the 4 planes is relevant in this population. However, in the absence of structural cardiopathies, the main determinants of aortic root and overall cardiac dimensions are sex, body size and age, which makes comparisons across children challenging.⁴ In this context, allometric scaling (but not linear correction) of BSA-adjusted cardiac measures is a valid procedure for obtaining body size- and age-independent values, at least in athletes aged 10 to 18 years.

There are some limitations in our study. We did not assess a control group of children not engaged in sports, and we studied athletes participating in sport events that might differ in cardiovascular demands (static or dynamic components) and thus, potentially, in aortic remodeling. Our use of the inner-inner method limits the comparability of our findings with respect to prior research using the standard leading edge-to-leading edge convention. By contrast, measurement of aortic diameters in the aforementioned 4 different planes should be considered a methodological strength of the study. Indeed, as we previously noted,³ performing only 1 to 2 measurements of the aortic root can result in over- or underestimation since aortic dilation distal to the supra-aortic ridge could be missed, with dilation potentially representing a risk factor for cardiovascular complications because of aortic dissection, especially in sports with higher hemodynamic loads.

We therefore propose normative values that might help clinicians to rapidly compare aortic root dimensions between children of different ages irrespective of their body size. This information should be useful in the early identification of aortic alterations that could limit sports participation, or at least justify close surveillance in pediatric/adolescent athletes.

FUNDING

M-E. Heras was funded by the Community of Madrid (Beca de investigación y formación postgrado de médicos especialistas en Medicina de la Educación Física y el Deporte, Orden 481/2016). Research by A. Lucia is funded by the Spanish Ministry of Economy and Competitiveness and Fondos Feder (grant PI18/00139).

AUTHORS’ CONTRIBUTIONS

A. Boraita and M-E. Heras share first authorship. A. Santos-Lozano and A. Lucia share senior authorship. A. Boraita and M-E. Heras conceived the original idea, acquired and analyzed the data, and drafted the initial version of the manuscript. P.L. Valenzuela, F. Morales-Acuña, A. Santos-Lozano and A. Lucia helped in the interpretation of the data and drafted the final version of the manuscript. All authors revised the manuscript critically for important intellectual content and approved the final version.

CONFLICTS OF INTEREST

The authors have no conflicts of interest.

Acknowledgements

We thank Dr Kenneth McCreath for his editorial assistance.