International Journal of Cardiology 223 (2016) 325–330

Contents lists available at ScienceDirect

International Journal of Cardiology

journal homepage: www.elsevier.com/locate/ijcard

Atrial chamber remodelling in healthy pre-adolescent athletes engaged

in endurance sports: A study with a longitudinal design. The CHILD study
Flavio D'Ascenzi a,⁎, Marco Solari a, Francesca Anselmi a, Silvia Maffei b, Marta Focardi a, Marco Bonifazi c,
Sergio Mondillo a, Michael Henein d
a
Department of Medical Biotechnologies, Division of Cardiology, University of Siena, Siena, Italy
b
Division of Cardiology, Department of Cardiac, Thoracic and Vascular Sciences, Santa Maria alle Scotte Hospital, Siena, Italy
c
Department of Medicine, Surgery, and NeuroScience, University of Siena, Siena, Italy
d
Department of Public Health and Clinical Medicine, Umeå University, and Heart Centre, Umeå, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Aims: Previous studies investigated the exercise-induced adaptation of left (LA) and right atrium (RA) in adults,
Received 17 May 2016 but little is known about respective changes in the growing heart of children. We aimed to longitudinally inves-
Received in revised form 17 July 2016 tigate the effects of endurance training on biatrial remodelling in preadolescent athletes.
Accepted 12 August 2016 Methods and results: Ninety-four children (57 endurance athletes, 37 sedentary controls; mean age 10.8 ± 0.2
Available online 14 August 2016
and 10.2 ± 0.2 years, respectively) were evaluated at baseline and after 5 months by ECG and by two-
dimensional, three-dimensional (3D) and speckle-tracking echocardiography. Athletes were trained at least
Keywords:
Speckle-tracking echocardiography
10 h/week. The resting heart rate was lower in athletes (p = 0.046) and decreased further after training
Atrial strain (p b 0.0001). Neither athletes nor controls had ECG evidence for LA or RA enlargement. At baseline, indexed
Training LA volumes did not differ between groups (p = 0.14) but indexed RA dimensions were larger in athletes (p =
3-dimensional echocardiography 0.007). After 5 months, indexed LA volumes increased in athletes but not in controls (p b 0.0001, p = 0.29; re-
Echocardiography spectively) while indexed RA volumes increased in both groups (p b 0.0001, p = 0.018; respectively). At the
athlete's heart same time, slight differences in biatrial reservoir and contractile function were found either in athletes, as dem-
onstrated by speckle-tracking echocardiography, but 3D-derived LA and RA ejection fraction remained stable in
both groups.
Conclusion: Endurance training influences the growing heart of preadolescent athletes with an additive increase
in biatrial size, suggesting that morphological adaptations can occur also in the early phases of the sports career.
Training-induced remodelling was associated with a preserved biatrial function, supporting the hypothesis of a
physiological remodelling.
© 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction functional and dimensional remodelling in adults [10–12], little is

The importance of assessing right (RA) and left atrial (LA) size by
echocardiography in children has increasingly been recognized. En-
larged RA can provide relevant hints for the diagnosis, follow-up, and
indication for intervention [1–3]. Also LA size is a recognized key deter-
minant of cardiac function and recent evidence suggests that in child-
hood obesity independently influences LA size [4,5] and may
represent a possible marker of increased cardiovascular risk [5]. Biatrial
enlargement is considered an early marker of ventricular disease, as
well as physiological response to increases in loading conditions, as is
the case in the ‘athlete's heart’ [6–9]. While previous cross-sectional
and longitudinal studies reported the impact of exercise on biatrial
⁎ Corresponding author at: Department of Medical Biotechnologies, Division of
Cardiology, University of Siena, Viale M. Bracci, 16, 53100 Siena, Italy.
E-mail address: flavio.dascenzi@unisi.it (F. D'Ascenzi).
known about respective changes in the growing heart in the early
phases of sports careers. Data from adult studies cannot be directly
transferred to preadolescent athletes, as they are physically less mature
and are usually exposed to a shorter period of training compared to
adults [13]. Furthermore, very few longitudinal studies have examined
the effect of endurance training on cardiac measurements in preadoles-
cent athletes [14–16], with controversial results. Considering the in-
creasing number of children involved in sports, the trend to more
intensive physical training, and the decreasing age at which young ath-
letes are encouraged to train intensively for sporting competitions, a
better understanding of their morphological and functional adaptation
is extremely important and imposes the need for the clinical character-
ization of their response to exercise.
Therefore, the aim of this longitudinal study was: i) to evaluate po-
tential differences in atrial measurements in preadolescent athletes
under control conditions; ii) to assess the extent of increase in atrial

http://dx.doi.org/10.1016/j.ijcard.2016.08.231
0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
326 F. D'Ascenzi et al. / International Journal of Cardiology 223 (2016) 325–330
measurements after 5 months of training compared to no training; and
iii) to determine whether atrial changes are related to those of myocar-
dial function assessed by speckle-tracking echocardiography (STE).
2. Methods
2.1. Study population
Sixty-two pre-adolescent male competitive endurance athletes practising swimming
in a regional level of mean age 10.8 ± 0.2 years [9–13] were enrolled in this study. They
were trained once a day, for 5–6 days a week. A typical training started with 30–45 min
of dry-land exercises (gymnastics and stretching) followed by 75–90 min swimming.
The total training programme consisted of 10% of warming-up exercises, 15% of technical
training, and 75% of three-staged aerobic exercises. In the first stage of aerobic training, ac-
counting for three-fifths of all the aerobic training, 65–70% of maximal heart rate (HR),
equivalent to 130–145 beats per minute, was achieved. In the second and third stages,
making up 25% and 15% of all the aerobic training, respectively, 70–80% of the maximal
heart rate, or 145–165 beats per minute, and 80–85% of the heart rate or 165–180 beats
per minute, were achieved.
Clinical examinations were performed at the beginning of the training program (Sep-
tember 2014, named hereafter ‘pre-training’) and after 5 months of intensive and closely
supervised training (February 2015, named hereafter ‘post-training’). The baseline clinical
assessment was performed after 3 months of detraining, during which time the athletes
were not engaged in any training program.
Participants were excluded from the study if they had signs of cardiac disease (cardio-
myopathies, shunts, interventricular septal or atrial defect, patent ductus arteriosus or
ventricular arrhythmias). Three athletes were excluded from the initial population
(N20 days after training) because of musculoskeletal injuries and two because of signs
of heart disease (1 with atrial septal defect and 1 with patent ductus arteriosus). One ath-
lete presented with monomorphic premature ventricular beats, but a 12-lead Holter mon-
itoring showed that those beats were isolated and frequent (1200/24 h), with a right
bundle branch block morphology, left axis deviation and a narrow QRS, suggesting a fas-
cicular origin. Accordingly, considering the absence of symptoms, the negative family his-
tory, and normal echocardiogram, this athlete was not excluded from the study thus
making the final population 57 subjects.
Thirty-seven sedentary males of mean age 10.2 ± 0.2 years [9–13] were also recruited
as controls, they all practised recreational activities for b2 h per week. None had hyperten-
sion, type I diabetes mellitus, endocrine disease or family history of heart disease.
All study participants underwent complete physical examination, ECG, echocardio-
graphic examination, and exercise ECG testing. After the rationale and the study protocol
were explained, the parents gave written informed consent for their offspring to partici-
pate in the study. The study protocol was approved by the local Ethical Committee. During
the study period, none of the participants experienced palpitations or symptoms, requir-
ing further investigations.
2.2. Physical examination
Height, weight and body surface area (BSA) [17] were obtained both at the beginning
of the study and after 5 months of training. The biological maturation of the participants
was established using the Tanner's five stages of penile and testicular development, ob-
tained at the two different time-points [18].
2.3. Twelve-lead ECG
A standard 12-lead ECG was performed using an ESAOTE P8000 Power Light, recorded
at 25 mm/s in the supine position during quiet respiration. ECG was interpreted by an ex-
perienced cardiologist, blinded to study time and without any knowledge of the echocar-
diographic findings. Resting heart rate (HR), QRS axis, and RA and LA enlargement criteria
were ascertained. Left axis deviation was defined as a QRS axis less than −30°, and right
axis deviation was defined as a QRS axis more than +120°. Left atrial enlargement
(LAE) was defined as a negative portion of the P-wave in lead V1 ≥ −0.1 mV in depth
and ≥40 ms in duration or ≥120 ms in duration in lead II. Right atrial enlargement (RAE)
was defined as a P-wave amplitude ≥0.25 mV in lead II and III or V1.
2.4. Echocardiographic examination
All echocardiographic examinations were performed by one cardiologist using a high-
quality echocardiograph (Vivid 9, GE, Milwaukee, WI, USA), equipped with an M4S
1.5–4.0 MHz transducer, and a one-lead ECG was continuously displayed. Off-line data
analysis, from three stored cycles, was performed by two experienced readers, blinded
to the study time-point, using a dedicated software (EchoPac, version 112, GE, USA).
LA and RA dimensions were measured at the end of left ventricular (LV) systole, at
their largest size during the cardiac cycle, as recommended [19,20]. RA and LA areas and
volumes were calculated using the biplane method of discs (modified Simpson rule)
from the apical 4-chamber and in the 4- and 2-chamber views for RA and LA, respectively.
For the LA an average value was obtained. Care was taken to exclude the pulmonary veins
and LA appendage from the LA tracing. The mitral annulus plane was used as the inferior
border [19].
RA dimensions were measured excluding the area between the leaflets and the annu-
lus, beyond the RA endocardium and excluding the inferior and superior vena cava and RA
appendage. Biatrial volumes were indexed to BSA, as recommended [20].
To identify potential predictors of training-induced atrial remodelling, LV volumes, mass,
stroke volume and RV basal and mid-cavity diameters and outflow tract measured both from
parasternal long-axis and short-axis views were obtained, as previously recommended [19–
21]. Also LV ejection fraction and RV fractional area change were measured [19,20].
Pulsed-wave and tissue Doppler imaging measurements were made to determine LV
and RV diastolic function, according to the current recommendations [20,21]. The follow-
ing measurements of LV and RV filling were obtained: Peak E and A wave velocities, E/A
ratio, s', e', and a' velocities. The mitral and tricuspid E/e' ratio was calculated and consid-
ered as a reliable index of LA and RA filling pressures [20–22].
2.5. Two-dimensional speckle-tracking echocardiography
The application of 2-dimensional (2D) STE to the study of LA and RA myocardial de-
formation has been already described [23,24]. In brief, 2D STE was obtained and recorded
using conventional 2-D grey-scale echocardiography during breath holding with stable
electrocardiographic tracing. Off-line analysis was performed using a commercially avail-
able semi-automated 2D strain software (EchoPAC PC, version 112, GE, USA). Endocardial
surfaces were manually traced in both 4- and 2-chamber views for the LA and in 4-
chamber view for the RA, using a point-and-click approach in order to create a region of
interest with the automatic adjustment of the system. After a manual adjustment of the
tracked area, the software divided the ROI into 6 segments, and the resulting tracking
quality for each segment was automatically scored as either acceptable or non-
acceptable, with the possibility for further manual correction. Peak atrial longitudinal
strain (PALS) and peak atrial contraction strain (PACS), measures of atrial reservoir and
atrial active conduit, respectively, were obtained by calculating the average of all segments
and, for the LA, the average values obtained from 4-chamber and 2-chamber view (LA
global PALS and global PACS) [23–25].
2.6. Three-dimensional echocardiography
LA and RA volumes by 3DE was collected in four-cycles full-volume made during a
breath hold, using a 3D matrix-array transducer. Acquisition was triggered to the electro-
cardiographic R wave. Care was taken to include the entire LA and RA within the pyrami-
dal 3D data set [26–28]. LA and RA volume by 3D echocardiography was derived from
semi-automated tracing of the LA endocardium, starting the measurements in the frame
with the largest atrial dimension, corresponding to ventricular end-systole, just before
opening of the atrio-ventricular valves, in perpendicular apical long-axis planes, tracing
5 points on the atrial surfaces. Following this manual identification, the program automat-
ically identified the endocardial surface using a deformable shell model [26,27]. Then, a
manual adjustment of the endocardial surface was undertaken to include trabeculae and
to exclude atrial appendages and large veins from the cavity volumes. The frame with
the smallest atrial dimension, at ventricular end-diastole, was selected with similar surface
detection and manual editing. Biatrial maximum and minimum volumes were obtained
and biatrial ejection fraction was derived as maximum - minimum volume/maximum vol-
ume [27].
2.7. Statistical analysis
Normal distribution of all continuous variables was examined using the Shapiro–Wilk
test, and data are presented as mean ± SD or median and interquartile range, as appropri-
ate. Categorical variables are expressed as percentages. The unpaired t-test and the Mann–
Whitney U test were used to assess the between groups significance, according to data dis-
tribution. The paired t-test and the Wilcoxon matched-pair test were used to assess the
within subjects significance of pre-training and post-training measurements, as appropri-
ate for data distribution. A P value b 0.05 was considered significant. The potential differ-
ences in Tanner's group assignment between athletes and controls were adjusted using
sampling weights so that the marginal totals on Tanner's group assignment in the athletes
group agreed with the corresponding totals of the control population, according to raking
ratio estimation [29].
Correlation analysis was performed to find association between continuous variables
using the Spearman and Pearson methods, as appropriate for data distribution. Δ param-
eters between baseline and 5-month measurements were calculated and used as depen-
dent or independent variables.
To assess the reproducibility of biatrial parameters, measurements were repeated, in a
random sample of 20 subjects (10 athletes and 10 controls), by the same investigator
(intra-observer variability) and by an additional reader (inter-observer variability),
blinded to previous results. Inter- and intra-observer variability was assessed by the
intraclass correlation coefficients (ICC) with 95% confidence intervals (CIs).
Statistical analyses were performed using SPSS version 21.0 software for Windows
(Statistical Package for the Social Sciences Inc., Chicago, IL).
3. Results
The demographic characteristics of athletes and controls are report-
ed in Table 1. At baseline there were no significant differences between
athletes and controls for height, weight, and BSA. After 5 months,
 F. D'Ascenzi et al. / International Journal of Cardiology 223 (2016) 325–330 327

Table 1
Demographic and biatrial morphological characteristics of the study population.

Variable Controls Endurance athletes P value athletes vs. controls P value athletes pre- P value controls baseline
(n = 37) (n = 57) vs. post-training vs. 2nd evaluation

Baseline 2nd evaluation Pre-training Post-training

Age 10.2 ± 0.2 10.8 ± 0.2

Systolic BP, mmHg 104 ± 12 100 ± 9 109 ± 10 106 ± 11 0.027 0.018 0.51
Diastolic BP, mmHg 66 ± 8 64 ± 6 70 ± 7 66 ± 6 0.027 0.001 0.79
HR, bpm 77 ± 12 78 ± 11 72 ± 9 67 ± 9 0.046 b0.0001 0.63
Height, cm 142.2 ± 8.3 145.9 ± 8.8 146.3 ± 10.9 149.0 ± 10.9 0.11 b0.0001 b0.0001
Weight, Kg 41.5 ± 12.0 44.1 ± 12.4 41.4 ± 9.9 43.2 ± 10.0 0.94 b0.0001 b0.0001
BSA, m2 1.27 ± 0.20 1.33 ± 0.20 1.29 ± 0.20 1.34 ± 0.20 0.45 b0.0001 b0.0001
LA volume index, mL/m2 16.2 ± 2.4 16.6 ± 2.7 17.2 ± 3.3 23.3 ± 3.2 0.14 b0.0001 0.29
RA area, cm2 9.0 ± 1.4 9.9 ± 1.6 10.0 ± 2.2 12.4 ± 2.2 0.022 b0.0001 b0.0001
RA volume index, mL/m2 15.4 ± 2.9 16.0 ± 2.0 17.6 ± 3.9 22.9 ± 4.1 0.007 b0.0001 0.018

height, weight, and BSA all increased in both athletes and in controls
(p b 0.0001 for both).
Based on the Tanner's Scale, 46% (n = 25) of athletes were at stage 1
(pre-puberty) and the rest were at stages 2–5 (puberty), at baseline. 9%
(n = 5) of athletes had reached sexual maturity and 37% remained at
the pre-pubertal stage (n = 20, p = 0.024 vs. baseline). At baseline,
43% (n = 16) of controls were at stage 1 (pre-puberty, p = 0.87 vs. ath-
letes) and 16% (n = 6) reached sexual maturity in the following
5 months with 27% remaining at pre-pubertal stage (n = 10, p = 0.01
vs. baseline and p = 0.37 vs. athletes).
At baseline, athletes had a lower resting HR compared to controls
(p = 0.046) and it reduced further 5 months afterwards (p b 0.0001).
Neither of the two groups had left axis deviation or fulfilled the ECG
criteria for LA or RA enlargement. After 5 months neither athletes nor
controls developed ECG signs of enlarged atria.
Table 1. At baseline, indexed LA volumes did not differ between athletes
and controls (p = 0.14). After 5 months a significant increase in indexed
LA volumes was found in the athletes but not in controls (p b 0.0001 and
p = 0.29, respectively), Fig. 1. At baseline, indexed RA dimensions were
larger in athletes compared to controls (Fig. 2). After 5 months a signif-
icant increase in indexed RA volumes was observed both in athletes and
in controls (Fig. 2).
The results of 3D echocardiographic examination are listed in
Table 2. Three-dimensional echocardiographic analysis demonstrated
that, after 5 months, 3D LA maximum volume index increased only in
athletes and not in controls (p b 0.0001 and p = 0.74, respectively)
and a similar trend was found also for 3D LA minimum volume index.
Both 3D RA maximum and minimum volume indexes increased after
5 months in athletes (p b 0.0001 for both) but remained unchanged in
controls (p = 0.91 and p = 0.94, respectively).

3.1. Biatrial morphological remodelling 3.2. Biatrial functional remodelling

The pre-training and post-training measurements obtained in ath- Biatrial function parameters obtained by speckle-tracking and by 3D
letes and the baseline and 5-month data of controls are reported in echocardiography are listed in Table 2. At baseline biatrial deformation

Fig. 1. Comparison of left atrial volume index measured at baseline and after 5 months of observation in athletes and in controls (see Table 1 for details).
328 F. D'Ascenzi et al. / International Journal of Cardiology 223 (2016) 325–330

Fig. 2. Comparison of right atrial volume index measured at baseline and after 5 months of observation in athletes and in controls (see Table 1 for details).

indexes were similar between athletes and controls. After 5 months, LA
PALS and LA PACS significantly decreased in athletes (p b 0.0001). Con-
versely, only LA PALS decreased in controls (p b 0.0001). Both RA PALS
and RA PACS significantly decreased in the athletes after 5 months of
training (p b 0.0001) while in controls only RA PALS modestly decreased
after 5 months of observation (p = 0.018).
3D-derived LA and RA ejection fractions remained stable in both ath-
letes and controls.
3.3. Biventricular remodelling
No differences were found between athletes and controls in LV ejec-
tion fraction (p = 0.062) at baseline and no changes were detected after
5 months (p = 0.94, p = 0.16, respectively). While stroke volume did
not differ between athletes and controls (p = 0.06) and between base-
line and 5-month data in controls (p = 0.076), it significantly increased
after training in athletes (p b 0.0001). LV indexed end-diastolic volume
was larger in athletes at baseline (p = 0.007) and increased further after
training (p b 0.0001). While no differences in RV end-diastolic area
were found between athletes and controls (p = 0.05), it increased
after training in athletes (p b 0.0001) but did not change in controls
(p = 0.077).
3.4. Weighted analysis
When the weighted analysis was performed to reassign the differ-
ences in growth and development between the two groups having
matched them according to Tanner's scale, the comparison between
pre-training and post-training data in the athletes proved similar results
of biatrial volumes. Indeed, LA and RA indexed volumes significantly in-
creased after training (p b 0.0001), irrespective of the technique used for
their measurement 2D or 3D (p b 0.0001).
3.5. Correlation analysis
3.5.1. Left atrium
The Δ LA volumes significantly correlated with Δ delta SV (R = 0.38,
p = 0.003), Δ weight (R = 0.31, p = 0.019), Δ BSA (R = 0.30, p =
0.021), and with Δ RA volume (R = 0.29, p = 0.028). Δ 3D LA volume
significantly correlated with Δ RA volume 3D max (R = 0.32, p =
0.014) and with Δ LV stroke volume (R = 0.29, 0 = 0.024).

Table 2
Three-dimensional and speckle-tracking echocardiographic parameters measured in athletes and in controls at the beginning of the study and after 5 months.

Variable Controls Endurance athletes P value athletes P value athletes pre- P value controls baseline
(n = 37) (n = 57 ) vs. controls vs. post-training vs. 2nd evaluation

Baseline 2nd evaluation Pre-training Post-training

RA PALS, % 58.2 ± 12.6 51.1 ± 15.2 53.1 ± 11.7 41.2 ± 10.1 0.071 b0.0001 0.018
RA PACS, % 11.8 ± 6.2 12.0 ± 5.6 10.9 ± 4.6 6.9 ± 4.0 0.58 b0.0001 0.46
LA PALS, % 64.2 ± 6.8 51.4 ± 9.3 61.1 ± 10.2 45.7 ± 8.0 0.15 b0.0001 b0.0001
LA PACS, % 15.5 ± 4.1 13.1 ± 3.3 16.4 ± 4.8 9.9 ± 2.7 0.41 b0.0001 0.026
3D LA maximum volume index, mL/m2 17.6 ± 2.8 17.6 ± 3.2 18.7 ± 3.7 23.7 ± 3.7 0.10 b0.0001 0.74
3D LA minimum volume index, mL/m2 6.0 ± 1.5 5.9 ± 1.5 6.2 ± 1.5 7.8 ± 2.0 0.76 b0.0001 0.18
LA EF, % 66.3 ± 6.3 67.6 ± 6.3 66.9 ± 6.4 67.0 ± 6.5 0.66 0.77 0.26
3D RA maximum volume index, mL/m2 17.7 ± 2.8 17.7 ± 3.4 19.6 ± 4.3 24.4 ± 3.6 0.037 b0.0001 0.94
3D RA minimum volume index, mL/m2 7.5 ± 2.2 7.5 ± 2.0 8.5 ± 2.4 10.2 ± 2.5 0.077 b0.0001 0.91
RA EF, % 57.1 ± 11.5 56.0 ± 11.9 56.8 ± 8.4 58.2 ± 7.7 0.87 0.32 0.69
 F. D'Ascenzi et al. / International Journal of Cardiology 223 (2016) 325–330
3.5.2. Right atrium
Both 2D and 3D Δ RA volumes correlated with stroke volume (R =
0.29 p = 0.024).
3.6. Reproducibility analysis
The inter-observer variability yielded an ICC of 0.98 (95% CI from
0.95 to 0.99, p b 0.001) for LA volume, an ICC of 0.97 (95% CI from 0.96
to 0.98, p b 0.001) for RA volume, an ICC of 0.97 (95% CI from 0.96 to
0.98, p b 0.001) for LA PALS and an ICC of 0.96 (95% CI from 0.95 to
0.97, p b 0.001) for RA PALS. The intra-observer variability showed an
ICC of 0.99 (95% CI from 0.98 to 0.99, p b 0.001) for LA volume, an ICC
of 0.98 (95% CI form 0.98 to 0.99, p b 0.001) for RA volume, an ICC of
0.98 (95% CI from 0.97 to 0.99, p b 0.001) for LA PALS and an ICC of
0.98 (95% CI from 0.97 to 0.99, p b 0.01) for RA PALS.
4. Discussion
The present study is the first to longitudinally characterize biatrial
response to training in preadolescent athletes. We demonstrated that:
i) at baseline indexed biatrial structural measurements were not differ-
ent between preadolescent athletes, evaluated after 3 months of
detraining, and their sedentary counterparts; ii) after 5 months, LA
size significantly increased in athletes and these findings were consis-
tent irrespective of technique used (2D or 3D echocardiography); how-
ever, RA size did increase both in athletes and in controls, although with
a greater extent in the former; iii) biatrial dilatation was accompanied
by a slight decrease in reservoir function as shown by a slight reduction
in LA and RA PALS, in the two groups with a peculiar reduction in con-
tractile function in athletes. However, this parameter remained within
the normal range in both groups. The demonstration of a stable and nor-
mal biatrial ejection fraction further confirm that, despite the cavity en-
largement, biatrial function remained preserved. iv) Atrial volumes
correlated, although modestly, with stroke volume.
Knowledge of the effect of training on biatrial morphological remod-
elling in children is scanty and conflicting findings were reported, deny-
ing any relationship between LA volume changes and physical activity
variables [30]. However, Triposkiadis et al. demonstrated in a cross-
sectional study that LA volumes were greater in prepubertal swimmers
compared to controls [31], findings which were further confirmed by
Zdravkovic M et al. in pre-adolescent football players [32]. Although
biatrial indexed dimensions were not different between athletes and
controls, this is the first longitudinal study demonstrating that 5 months
of intensive training resulted in significant increase in biatrial volumes
in athletes along with reduction in their myocardial deformation func-
tion. These changes were irrespective of age, therefore they can be
interpreted on the basis of sport activities.
The potential cause of LA enlargement could be either LV dysfunc-
tion or the increase in venous return to the atrium which is bound to en-
large because of its thin wall against the increase in wall stress with
exercise. Our findings support the latter mechanism since we found
no change in LV structure or function with such type of sports. Further-
more, we found a modest relationship between changes in LA volumes
and stroke volume of the athletes but not of the controls. This explana-
tion is further supported by the changes we documented in the right
atrial volumes which were significantly larger than those of controls,
even at baseline after 3 months of detraining. The even thinner RA
walls with their support to a circulation pressure 30% that of the left
heart can probably explain the difference in response between the
two atria. Finally, retrospective interpretation of the lack of indexed vol-
ume difference between athletes and controls suggests that swimming
results in healthy physiological changes in the atrial function which are
reversible after deconditioning. Furthermore, even with the increase in
LA indexed volume with exercise its overall cavity systolic function
remained preserved suggesting maintained myocardial reserve.
Longer-term effect of swimming on biatrial structure and function
329
remains to be determined, along with age related changes in cardiac
structure and function.
This longitudinal study demonstrates that biatrial remodelling is dy-
namic in its nature and changes after training, the knowledge of this
biatrial adaptation is relevant, when considering biatrial size as a mark-
er for cardiac anatomical and functional pathology [33] or for raising the
suspicion of a pulmonary hypertension independent of tricuspid regur-
gitation velocity [34]. The slight reduction in atrial deformation param-
eters should not be seen as pathological since it accompanied the
enlargement of the atrial cavity, thus could easily be explained on the
basis of Frank-Starling law, and its recovery which contributed to a nor-
mal ejection fraction supports its physiological nature. These observations
are in agreement with previous longitudinal studies in competitive ath-
letes demonstrating a slight reduction in atrial deformation parameters
and a peculiar contractile property induced by training, not only in adults
[9,12,11,25] but also in adolescent athletes [10]. A slight reduction in atrial
deformation parameters was observed in this study also in the controls,
further supporting the interpretation of biatrial functional remodelling
in athletes as a physiological phenomenon.
5. Limitations
While the present study suggests that training can affect biatrial size
and function in pre-adolescents, it is important to highlight that this
finding may be sport specific, it is possible that the influence of age
and the related hormonal and growth factors may have a different im-
pact in anaerobic sports. Also, we cannot exclude that the observed re-
sponse to training may reflect genetic traits which predisposed the
trained children to a more favourable and relevant adaptation. Our re-
sults apply only to boys so should not be generalized to both genders.
In addition, the extent of recovery and normalization of biatrial volumes
after stopping exercise should not be expected in adults without study-
ing it first.
Finally, echocardiography was used to estimate biatrial size, even
if cardiac magnetic resonance imaging (MRI) now represents the
method of choice to quantitatively assess cardiac dimensions. The
lack of data based on cardiac MRI represents a limitation; however,
in this study we used not only two-dimensional, but also three-
dimensional echocardiography which is the only echocardiographic
technique that measures cardiac chamber volumes directly without
geometric assumption and that has an accuracy comparable with
that of cardiac MRI, although volumes tends to be lower on echocar-
diography [19].
6. Conclusions
Intensive endurance training affects the growing heart of preadoles-
cent athletes with an additive increase in biatrial size, suggesting that
morphological adaptations can occur also in the early phases of the
sports career of an athlete. Five months of intensive training were asso-
ciated with a preserved biatrial function, as demonstrated by two-
dimensional STE and by 3-D echocardiographic measurements,
supporting the hypothesis of a physiological remodelling of the heart.
Although the presence of a dilated atrium in a child should raise the sus-
picion of an underlying cardiovascular disease, in absence of congenital
heart defects, an increase in biatrial dimensions associated with pre-
served biatrial function can be considered a physiological expression
of the athlete's heart and knowledge of training characteristics is neces-
sary to interpret echocardiographic and clinical data in this population.
Conflict of interest disclosures
None.
330 F. D'Ascenzi et al. / International Journal of Cardiology 223 (2016) 325–330
Acknowledgments
The research was realized with a grant made available by the Italian
Society of Cardiology, supported by a contribution of MSD Italy on be-
half of MSD-ITALY MERCK SHARP & DOHME CORPORATION.
References
[1] E. Kassen, T. Humpl, M.K. Friedberg, Prognostic significance of 2-dimensional, M-
mode, and Doppler echo indices of right ventricular function in children with pul-
monary arterial hypertension, Am. J. Cardiol. 165 (2013) 1024–1031.
[2] K. Okumura, C. Slorach, D. Mroczek, A. Dragulescu, L. Mertens, A.N. Redington, M.K.
Friedberg, Right ventricular diastolic performance in children with pulmonary arte-
rial hypertension associated with congenital heart disease: correlation of echocar-
diographic parameters with invasive reference standards by high-fidelity
micromanometer catheter, Circ. Cardiovasc. Imaging 7 (2014) 491–501.
[3] M.J. Ploegstra, M.T. Roofthooft, J.M. Douwes, B. Bartelds, N.J. Elzenga, D. van de
Weerd, H.L. Hillege, R.M. Berger, Echocardiography in pediatric pulmonary arterial
hypertension: early study on assessing disease severity and predicting outcome,
Circ. Cardiovasc. Imaging 8 (2014) 1.
[4] J.G. Ayer, G.F. Sholler, D.S. Celermajer, Left atrial size increases with body mass index
in children, Int. J. Cardiol. 141 (2010) 61–67.
[5] M.L. Marcovecchio, M. Gravina, S. Gallina, E. D'Adamo, R. De Caterina, F. Chiarelli, A.
Mohn, G. Renda, Increased left atrial size in obese children and its association with
insulin resistance: a pilot study, Eur. J. Pediatr. 175 (2016) 121–130.
[6] A. Pelliccia, B.J. Maron, F.M. Di Paolo, A. Biffi, F.M. Quattrini, C. Pisicchio, A. Roselli, S.
Caselli, F. Culasso, Prevalence and clinical significance of left atrial remodeling in
competitive athletes, J. Am. Coll. Cardiol. 46 (2005) 690–696.
[7] A. D'Andrea, L. Riegler, R. Cocchia, R. Scarafile, G. Salerno, R. Gravino, E. Golia, O. Vriz,
R. Citro, G. Limongelli, P. Calabrò, G. Di Salvo, P. Caso, M.G. Russo, E. Bossone, R.
Calabrò, Left atrial volume index in highly trained athletes, Am. Heart J. 159
(2010) 1155–1161.
[8] S. Nistri, M. Galderisi, P. Ballo, I. Olivotto, A. D'Andrea, L. Pagliani, A. Santoro, B.
Papesso, P. Innelli, F. Cecchi, S. Mondillo, Determinants of echocardiographic left
atrial volume: implications for normalcy, Eur. J. Echocardiogr. 12 (2011) 826–833.
[9] F. D'Ascenzi, M. Cameli, M. Lisi, V. Zacà, A. Santoro, A. Causarano, S. Mondillo, Super-
normal diastolic function and role of left atrial myocardial deformation analysis by
2D speckle tracking echocardiography in elite soccer players, Echocardiography 28
(2011) 320–326.
[10] F. D'Ascenzi, M. Cameli, M. Lisi, V. Zacà, B. Natali, A. Malandrino, S. Benincasa, S.
Catanese, A. Causarano, S. Mondillo, Left atrial remodelling in competitive adoles-
cent soccer players, Int. J. Sports Med. 33 (2012) 795–801.
[11] F. D'Ascenzi, A. Pelliccia, B.M. Natali, V. Zacà, M. Cameli, F. Alvino, A. Malandrino, P.
Palmitesta, A. Zorzi, D. Corrado, M. Bonifazi, S. Mondillo, Morphological and func-
tional adaptation of left and right atria induced by training in highly trained female
athletes, Circ. Cardiovasc. Imaging 7 (2014) 222–229.
[12] F. D'Ascenzi, A. Pelliccia, B.M. Natali, M. Cameli, M. Lisi, M. Focardi, M. Padeletti, P.
Palmitesta, D. Corrado, M. Bonifazi, S. Mondillo, M. Henein, Training-induced dy-
namic changes in left atrial reservoir, conduit, and active volumes in professional
soccer players, Eur. J. Appl. Physiol. 115 (2015) 1715–1723.
[13] S. Sharma, B.J. Maron, G. Whyte, S. Firoozi, P.M. Elliott, W.J. Mckenna, Physiologic
limits of left ventricular hypertrophy in elite junior athletes: relevance to differential
diagnosis of athlete's heart and hypertrophic cardiomyopathy, J. Am. Coll. Cardiol.
40 (2002) 1431–1436.
[14] D.L. Geene, T.B. Gilliam, D. Crowley, C. Morehead-Steffiens, A. Rosenthal, Echocardio-
graphic measures in 6 to 7-year-old children after an 8 month exercise program,
Am. J. Cardiol. 49 (1982) 1990–1995.
[15] Y. Maingourd, E. Bourges-Petit, C. Tanguy, J.M. Quintard, J. Medelli, M. Freville, A
peripuberal longitudinal echocardiographic study of cardiac development in a
group of ice hockey players, Arch. Mal. Coeur Vaiss. 83 (1990) 371–375.
[16] P. Obert, S. Mandigout, A. Vinet, L.D. N'Guyen, F. Stecken, D. Courteix, Effect of aero-
bic training and detraining on left ventricular dimensions and diastolic function in
prepubertal boys and girls, Int. J. Sports Med. 22 (2001) 90–96.
[17] D. Du Bois, E.F. Du Bois, A formula to estimate the approximate surface area if height
and weight be known. 1916, Nutrition 5 (1989) 303–311.
[18] J.M. Tanner, Growth and maturation during adolescence, Nutr. Rev. 39 (1981)
43–55.
[19] R.M. Lang, L.P. Badano, V. Mor-Avi, J. Afilalo, A. Armstrong, L. Ernande, F.A.
Flachskampf, E. Foster, S.A. Goldstein, T. Kuznetsova, P. Lancellotti, D. Muraru,
M.H. Picard, E.R. Rietzschel, L. Rudski, K.T. Spencer, W. Tsang, J.U. Voigt, Recommen-
dations for cardiac chamber quantification by echocardiography in adults: an up-
date from the American Society of Echocardiography and the European
Association of Cardiovascular Imaging, Eur. Heart J. Cardiovasc. Imaging 16 (2015)
233–270.
[20] L.G. Rudski, W.W. Lai, J. Afilalo, L. Hua, M.D. Handschumacher, K. Chandrasekaran,
S.D. Solomon, E.K. Louie, N.B. Schiller, Guidelines for the echocardiographic assess-
ment of the right heart in adults: a report from the American Society of Echocardi-
ography endorsed by the European Association of Echocardiography, a registered
branch of the European Society of Cardiology, and the Canadian Society of Echocar-
diography, J. Am. Soc. Echocardiogr. 23 (2010) 685–713.
[21] S.F. Nagueh, C.P. Appleton, T.C. Gillebert, P.N. Marino, J.K. Oh, O.A. Smiseth, A.D.
Waggoner, F.A. Flachskampf, P.A. Pellikka, A. Evangelista, Recommendations for
the evaluation of left ventricular diastolic function by echocardiography, Eur. J.
Echocardiogr. 10 (2009) 165–193.
[22] V. Mor-Avi, R.M. Lang, L.P. Badano, M. Belohlavek, N.M. Cardim, G. Derumeaux, M.
Galderisi, T. Marwick, S.F. Nagueh, P.P. Sengupta, R. Sicari, O.A. Smiseth, B.
Smulevitz, M. Takeuchi, J.D. Thomas, M. Vannan, J.U. Voigt, J.L. Zamorano, Current
and evolving echocardiographic techniques for the quantitative evaluation of cardi-
ac mechanics: ASE/EAE consensus statement on methodology and indications en-
dorsed by the Japanese Society of Echocardiography, Eur. J. Echocardiogr. 12
(2011) 167–205.
[23] M. Cameli, M. Caputo, S. Mondillo, P. Ballo, E. Palmerini, M. Lisi, E. Marino, M.
Galderisi, Feasibility and reference values of left atrial longitudinal strain imaging
by two-dimensional speckle tracking, Cardiovasc. Ultrasound 7 (2009) 6.
[24] M. Padeletti, M. Cameli, M. Lisi, A. Malandrino, V. Zacà, S. Mondillo, Reference values
of right atrial longitudinal strain imaging by two-dimensional speckle tracking,
Echocardiography 29 (2012) 147–152.
[25] F. D'Ascenzi, M. Cameli, M. Padeletti, M. Lisi, V. Zacà, B. Natali, A. Malandrino, F.
Alvino, M. Morelli, G.M. Vassallo, C. Meniconi, M. Bonifazi, A. Causarano, S.
Mondillo, Characterization of right atrial function and dimension in top-level ath-
letes: a speckle tracking study, Int. J. Card. Imaging 29 (2013) 87–94.
[26] A.M. Anwar, O.I. Soliman, M.L. Geleijnse, A. Nemes, W.B. Vletter, F.J. ten Cate, Assess-
ment of left atrial volume and function by real-time three-dimensional echocardiog-
raphy, Int. J. Cardiol. 123 (2008) 155–161.
[27] Y. Miyasaka, S. Tsujimoto, H. Maeba, F. Yuasa, K. Takehana, K. Dote, T. Iwasaka, Left
atrial volume by real-time three-dimensional echocardiography: validation by 64-
slice multidetector computed tomography, J. Am. Soc. Echocardiogr. 24 (2011)
680–686.
[28] A. Takahashi, N. Funabashi, A. Kataoka, R. Yajima, M. Takahashi, M. Uehara, H.
Takaoka, M. Saito, C. Yamaguchi, K. Lee, I. Komuro, F. Nomura, Quantitative evalua-
tion of right atrial volume and right atrial emptying fraction by 320-slice computed
tomography compared with three-dimensional echocardiography, Int. J. Cardiol.
146 (2011) 96–99.
[29] G. Kalton, Compensating for Missing Survey Data. Research Report Series/Institute
for Social Research, Survey Research Center, Institute for Social Research, The Uni-
versity of Michigan, Ann Arbor, MI, 1983 157.
[30] M. Dencker, A. Bugge, B. Hermansen, L.B. Andersen, Objectively measured daily
physical activity related to aerobic fitness in young children, J. Sports Sci. 28
(2010) 139–145.
[31] F. Triposkiadis, S. Ghiokas, I. Skoularigis, A. Kotsakis, I. Giannakoulis, V. Thanapoulos,
Cardiac adaptation to intensive training in prepubertal swimmers, Eur. J. Clin.
Investig. 32 (2002) 16–23.
[32] M. Zdravkovic, J. Perunicic, M. Krotin, M. Ristic, V. Vukomanovic, I. Soldatovic, D.
Zdravkovic, Echocardiographic study of early left ventricular remodeling in highly
trained preadolescent footballers, J. Sci. Med. Sport 13 (2010) 602–606.
[33] M. Koestenberger, A. Burmas, W. Ravekes, A. Avian, A. Gamillscheg, G. Grangl, M.
Grillitsch, G. Hansmann, Echocardiographic reference values for right atrial size in
children with and without atrial septal defects or pulmonary hypertension, Pediatr.
Cardiol. 37 (2015) 686–695.
[34] E. Grünig, A. Barner, M. Bell, M. Claussen, M. Dandel, D. Dumitrescu, M. Gorenflo, S.
Holt, G. Kovacs, S. Ley, J.F. Meyer, S. Pabst, G. Riemekasten, J. Saur, M. Schwaiblmair,
C. Seck, L. Sinn, S. Sorichter, J. Winkler, H.H. Leuchte, Non-invasive diagnosis of pul-
monary hypertension: ESC/ERS guidelines with updated commentary of the cologne
consensus conference 2011, Int. J. Cardiol. 154 (Suppl. 1) (2011) S3–12.