European Heart Journal - Cardiovascular Imaging Advance Access published January 14, 2015 European Heart Journal – Cardiovascular Imaging doi:10.1093/ehjci/jeu309

Arterial–left ventricular–left atrial coupling late after repair of aortic coarctation and interruption Vivian Wing-yi Li and Yiu-fai Cheung* Division of Paediatric Cardiology, Department of Paediatrics and Adolescent Medicine, Queen Mary Hospital, University of Hong Kong, 102 Pokfulam Road, Hong Kong, China Received 19 August 2014; accepted after revision 8 December 2014

Aims

This study aimed to explore the arterial–left ventricular (LV)–left atrial (LA) interaction in adolescents and young adults late after intervention for coarctation of the aorta (CoA) and interrupted aortic arch (IAA). ..................................................................................................................................................................................... Methods Thirty-one (16 males) patients aged 23.4 + 6.3, at 20.6 + 5.2 years after intervention, and 31 controls were studied. and results Carotid arterial stiffness and intima-media thickness (IMT) and brachial-ankle pulse wave velocity were determined by radiofrequency-based echocardiography and oscillometry, respectively. Tissue Doppler and speckle tracking echocardiography (STE) were performed to assess, respectively, LV myocardial tissue velocities and linear and torsional deformation. Left atrial positive, negative, and total strain and strain rate at ventricular systole (aSRs), early diastole (aSRed), and atrial contraction (aSRac) were also determined using STE. Patients had significantly greater arterial stiffness and IMT than controls (all P , 0.05). Mitral annular systolic and diastolic velocities, LV longitudinal and radial strain and early diastolic strain rates, peak torsion and untwisting velocity, and LA peak positive and total strain, aSRs, aSRed, and aSRac were significantly lower in patients than in controls (all P , 0.05). Arterial stiffness correlated inversely with LV longitudinal strain and systolic and early diastolic strain rate (all P , 0.05), while LA total strain and aSRed were associated positively with LV diastolic annular velocity, longitudinal SRe, and peak untwisting velocity (all P , 0.05). Multiple linear regression further revealed arterial stiffness as an independent determinant of LA total strain (b ¼ 21.3, P ¼ 0.034). ..................................................................................................................................................................................... Conclusion Our findings suggest impairment of arterial function and LV and LA mechanics in patients after CoA and IAA repair and implicate an abnormal arterial–LV –LA interaction.

----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords

arteries † left ventricular † left atrium † coarctation of the aorta † interrupted aortic arch

Introduction Systemic arterial abnormalities in patients after repair of coarctation of the aorta (CoA) and interrupted aortic arch (IAA) are well documented. Alteration of aortic structure in these patients is characterized by increased collagen, decreased smooth muscle content, and loss of elastic fibres proximal to the coarctation site.1 Stiffening of central arteries,2,3 reduced carotid arterial distensibility,4 and increased carotid intima-media thickness (IMT)4,5 have further been shown in patients late after relief of aortic obstruction. Hence, the arterial afterload presented to the left ventricle is likely to be persistently increased despite successful interventions for CoA and IAA. Optimal ventricular– arterial interaction is important for efficient cardiovascular performance. Left ventricular (LV) hypertrophy with apparently normal or ‘supranormal’ systolic function as characterized by increased LV ejection fraction and velocity of circumferential fibre

shortening adjusted for wall stress has been described in patients late after repair of CoA.6,7 In previous studies, assessment of LV systolic function has relied on M-mode assessment of endocardial and mid-wall myocardial shortening.8 Nonetheless, recent cardiac magnetic resonance-based myocardial strain imaging has revealed abnormal LV longitudinal systolic strain in patients after successful CoA repair.9 It remains, however, unclear whether LV systolic deformation in other dimensions, diastolic myocardial deformation, and torsional mechanics are altered. Apart from its coupling with systemic arteries, the left ventricle also interacts with the left atrium. The left atrium plays an important role in regulating LV filling given its reservoir, conduit, and pump function.10 – 12 Importantly, the pathophysiological significance of left atrial (LA) dysfunction in patients with LV dysfunction13 and hypertension14 is increasingly recognized. In preclinical subjects with cardiovascular risk factors, abnormal LA–LV– arterial coupling

* Corresponding author. Tel: +852 22554090; Fax: +852 25539491, E-mail: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015 . For permissions please email: [email protected].

Page 2 of 10 has been shown using speckle tracking echocardiography (STE).15 Senzaki et al. 16 calculated LA systolic force based on transmitral late diastolic velocity and mitral valve orifice area and noted its increase in patients with unrepaired or recurrent aortic coarctation. Nonetheless, direct interrogation of LA function by imaging of atrial deformation has not been performed previously in patients with satisfactorily repaired CoA and IAA. In this study, we aimed to test the hypothesis that arterial –LV –LA interaction is altered in patients late after repair of CoA and IAA despite successful intervention. To test the hypothesis, we assessed LV systolic, diastolic, and torsional deformation and LA deformation using 2D STE and determined relationships among indices of arterial function, LV deformation, and LA deformation.

V.W.-y. Li and Y.-f. Cheung

at 10 mm proximal to the carotid bulb and averaging six consecutive measurements.

Brachial-ankle pulse wave velocity Regional arterial stiffness of the brachial-ankle arterial segment was determined by measuring the brachial-ankle pulse wave velocity (baPWV) using VP-2000 (Colin Corporation, Komaki, Japan) as repeated previously.17 Briefly, oscillometric cuffs were wrapped around the brachium and ankles to register the pressure pulses, and the right and left baPWVs were calculated as the pulse traveling distance divided by the transit time. The traveling distance was calculated automatically according to the patient’s height on the basis of oriental anthropometric data.17 The average of the right and left baPWV was calculated and used for subsequent analysis.

Echocardiographic assessment

Methods Subjects Patients with significant residual coarctation, defined by a Dopplerderived systolic pressure gradient .20 mmHg, were excluded. Thirtyone patients (16 males), aged 23.4 + 6.3 (range, 13.8 – 37.3 years), who have undergone repair for CoA or IAA were recruited from the congenital heart clinic. They were studied at 20.6 + 5.2 years after initial intervention and at 15.8 + 3.9 years after the latest intervention. Twentyeight had CoA and three had IAA. Two of the patients had Turner syndrome. Seventeen (55%) had surgical repair as initial intervention, whereas 14 underwent balloon angioplasty. Twenty-four patients required further balloon angioplasties after initial intervention at age 6.3 + 6.2 years, one of whom eventually required the placement of a stent. Thirteen patients had ligation of an associated persistent arterial duct, nine had surgical repair of ventricular septal defect, and three had surgical closure of atrial septal defect; one of whom also had concomitant repair of a cleft mitral valve with mild mitral regurgitation but no stenosis. Eight patients had a bicuspid aortic valve with a continuous Dopplerderived peak ascending aortic flow velocity of 1.7 + 0.8 m/s. One patient had a small subaortic membrane with no significant LV outflow obstruction. Two of the patients were taking cardiac medications, one taking atenolol and enalapril for systemic hypertension, and the other digoxin, losartan, and metoprolol. Thirty-one individuals aged 22.1 + 6.0 (range, 12.6– 33.2 years), including staff volunteers and healthy siblings of patients with no clinical history of cardiovascular diseases, were recruited as controls. This study was approved by the Institutional Review Board of the University of Hong Kong/Hospital Authority West Cluster, Hong Kong. All adult subjects and parents of minors gave informed consent. Body weight and height were measured, and body mass index (BMI) was calculated accordingly. The body surface area (BSA) was further derived by the DuBois formula: 0.0001 × 71.84 × (weight)0.425 × (height)0.725. Blood pressure in the right arm was measured three times using an automated oscillometric device (Dinamap, Critikon, Inc.), and the average of last two readings was taken.

Assessment of vascular function Carotid arterial IMT and stiffness Assessment of the right carotid arterial IMT and stiffness was performed using MyLab One (Esaote System, Italy) based on the radiofrequency technology. The stiffness was calculated as ln (SBP/DBP)/(DD/D), where DD ¼ systolic 2 diastolic diameter, D, diastolic diameter; DBP, diastolic blood pressure; and SBP, systolic blood pressure. These indices were derived from imaging of the right common carotid artery

All echocardiographic examinations were performed using Vivid 7 ultrasound machine using M4S transducer (GE Medical System, Horten, Norway). Echocardiographic acquisitions were stored digitally for offline analyses using EchoPAC software (GE Medical Systems). The average values of echocardiographic indices derived from three cardiac cycles were used for statistical analyses.

Conventional echocardiography The maximum 2D LA area at ventricular end systole was measured from the apical four-chamber view by planimetry. From the apical fourchamber view, pulsed-wave Doppler examination was performed to obtain peak mitral inflow velocities at early (E) and late (A) diastole and E deceleration time. Thickness of ventricular septum and posterior LV wall at end diastole was measured by standard M-mode interrogation of the LV parasternal short axis with calculation of LV mass according to the standard formula.

Tissue Doppler imaging From the apical four-chamber view, colour tissue Doppler imaging was performed with frame rates .100 Hz. With the sample volume positioned at the LV free wall-mitral annular junction, the following parameters were obtained: systolic (s), early diastolic (e), and late diastolic (a) myocardial tissue velocities and myocardial acceleration during isovolumic contraction (IVA).18 The E/e ratio was calculated as an estimate of LV filling pressure.19

Speckle tracking echocardiography Two-dimensional STE was performed with frame rates between 60 and 80 frames per second for determination of global LV and LA deformation. In the analysis of LV deformation, the Q wave was used as the reference time point. From the apical four-chamber view, the entire LV contour was traced to derive the global longitudinal systolic strain and global longitudinal strain rates during systole (SRs), early diastole (SRe), and late diastole (SRa). From the parasternal short-axis view at mid-ventricular level, LV global systolic, radial, and circumferential strain, and systolic and diastolic strain rates were quantified. For assessment of LV torsional mechanics, parasternal short-axis views at the apical and basal levels were acquired with derivation of the following indices: peak LV torsion, peak systolic twisting velocity, and peak diastolic untwisting velocity. Our group has previously reported on the high reproducibility of these LV deformation indices.20,21 Global LA deformation was similarly assessed from the apical fourchamber view using 2D STE based on tracing and tracking of the entire endocardial contour of the left atrium by the Echopac software.10 – 12 In the calculation of LA strain and strain rate, the onset of P wave was taken as the reference point (Figure 1). The entire LA contour was

Arterial– LV– LA interaction post CoA and IAA repair

Page 3 of 10

Figure 1: Assessment of LA strain (upper panel) and strain rate (lower panel) by STE. SRs, strain rate at ventricular systole; SRed, strain rate at early diastole; SRac, strain rate at atrial contraction.

traced to determine the following parameters: LA positive strain, negative strain, total strain (sum of the absolute values of positive and negative strain), and atrial strain rate during ventricular systole (aSRs), early diastole (aSRed), and atrial contraction (aSRac). Our group has also recently reported on the high reproducibility of using STE to assess LA deformation.12

Statistical analysis Results are presented as mean + standard deviation. All strain and strain rate parameters are expressed as absolute values to facilitate interpretation and analyses. Shapiro – Wilk test confirmed normal distribution of

echocardiographic indices. Demographic variables, arterial indices, and echocardiographic parameters of patients and controls were compared using Student’s t-test. For the whole cohort, Pearson correlation analysis was used to explore associations between vascular indices and LV echocardiographic parameters, between LA and LV echocardiographic indices, and between LA area and indices of LA deformation. Multiple linear regression of the entire cohort was performed to identify significant arterial and ventricular diastolic functional predictors of LA deformation indices. Statistical analyses were performed with SPSS, version 16.0 (SPSS Inc., Chicago, IL, USA). A P-value of ,0.05 was considered statistically significant.

Page 4 of 10

V.W.-y. Li and Y.-f. Cheung

Results Subjects Table 1 summarizes the demographic data and blood pressures of patients and controls. Patients were shorter (P ¼ 0.02) but had similar weight, BMI, and BSA (all .0.05) compared with controls. Their systolic (P ¼ 0.11) and diastolic (0.26) blood pressures of the right upper arm were similar. Only one of the patients was obese with a BMI of 31.5 kg/m2. All but one of the patients were normotensive at the time of study. The one patient with mild hypertension (141/ 76 mmHg) at the time of study was known to have white coat syndrome.

Vascular parameters Compared with controls, patients had significantly greater carotid artery stiffness index (7.5 + 2.8 vs. 5.8 + 2.0, P ¼ 0.007) and carotid IMT (553 + 122 vs. 469 + 85 mm, P ¼ 0.002). On the other hand, the average baPWV was similar between patients and controls (11.1 + 1.8 vs. 11.1 + 1.6 m/s, P ¼ 1.0).

Indices of LV deformation Patients had significantly greater LV posterior wall (6.4 + 0.7 vs. 5.8 + 0.9 mm, P ¼ 0.005) and septal (6.3 + 0.6 vs. 5.9 + 0.7 mm, P ¼ 0.035) thickness at end diastole and LV mass (103.5 + 28.6 vs. 90.8 + 19.1 g, P ¼ 0.045). Table 2 summarizes the conventional Doppler, tissue Doppler, and STE findings of LV deformation. Compared with controls, patients had significantly lower mitral annular s velocity (P , 0.001), global LV longitudinal systolic strain (P ¼ 0.001), and global LV radial systolic strain (P ¼ 0.005). With regard to diastolic function, patients had significantly greater A velocity (P , 0.001) and E/e ratio (P , 0.001) and significantly lower E/A ratio (P ¼ 0.033), mitral annular e velocity (P , 0.001), LV longitudinal SRe (P , 0.001), and LV radial SRe (P ¼ 0.022) than controls. Additionally, peak LV torsion (P , 0.001) and diastolic untwisting velocity (P , 0.001) were reduced in patients compared with controls. Circumferential strain and strain rates, on the other hand, were similar between patients and controls (All P . 0.05). Further analyses with exclusion of the nine patients who had additional ventriculoseptal defect (VSD) closure were performed to exclude potential confounding influence of previous surgery on

Table 1

ventricular deformation. The LV longitudinal systolic strain (P ¼ 0.003), longitudinal SRe (P , 0.001), global radial systolic strain (P , 0.001), peak torsion (P ¼ 0.003), and diastolic untwisting velocity (P , 0.001) remained significantly lower in the 22 patients who did not require VSD closure compared with controls.

Indices of LA deformation Table 2 summarizes the STE findings of LA deformation. Compared with controls, patients had significantly lower peak positive and total strain (both P , 0.001), aSRs (P , 0.001), aSRed (P , 0.001), and aSRac (P ¼ 0.001). Similarly, peak positive strain (P , 0.001), total strain (both P , 0.001), aSRs (P , 0.001), aSRed (P , 0.001), and aSRac (P ¼ 0.004) remained significantly lower in patients after excluding those with previous VSD closure compared with controls. For the whole cohort, LA area indexed to BSA correlated negatively with aSRs (r ¼ 20.35, P ¼ 0.006) but not other indices of atrial deformation (all P . 0.05).

Correlations between vascular indices and LV deformation Univariate analysis of vascular indices and LV deformation parameters are presented in Table 3. For the whole cohort, carotid arterial stiffness correlated negatively with mitral annular e velocity (P ¼ 0.001), global LV systolic longitudinal strain (P , 0.001), longitudinal SRs (P ¼ 0.004), longitudinal SRe (P , 0.001), and systolic radial strain (P ¼ 0.032) (Figure 2). There were no significant correlations with indices of LV circumferential and torsional deformation (all P . 0.05). Carotid IMT correlated negatively with mitral annular s (P , 0.001), e (P , 0.001), and a (P ¼ 0.010) velocities, LV longitudinal SRe (P ¼ 0.031), and peak diastolic untwisting velocity (P ¼ 0.014). Even with exclusion of the eight patients with bicuspid aortic valve and the one with subaortic membrane, the correlations between vascular parameters and LV deformation indices remained significant (Table 3).

Correlations between indices of LA and LV diastolic deformation Table 4 shows the correlations between atrial deformation and LV diastolic indices. For the whole cohort, LA total strain correlated positively with LV mitral annular e velocity (P , 0.001), LV

Demographic data and blood pressure Patients (n 5 31)

Controls (n 5 31)

P-value

Age (years) Sex (M/F)

23.4 + 6.3 (13.8– 37.3) 16/15

22.1 + 6.0 (12.6– 33.2) 15/16

0.45 0.90

Height (cm)

160.5 + 10.6 (138–178)

166.5 + 9.1 (151– 184)

...............................................................................................................................................................................

0.02

Weight (kg) Body mass index (kg/m2)

54.4 + 14.5 (29–94) 20.8 + 3.5 (14.8– 31.0)

57.9 + 11.3 (40– 78) 20.7 + 2.5 (16.4– 26.8)

Body surface area (m2)

1.55 + 0.24 (1.08–2.09)

1.64 + 0.19 (1.34– 1.98)

0.12

Heart rate (bpm) Systolic blood pressure (mmHg)

70 + 13 (44– 98) 119 + 13 (90– 141)

69 + 12 (53– 91) 114 + 13 (95– 138)

0.86 0.11

Diastolic blood pressure (mmHg)

68 + 8 (52– 96)

66 + 7 (51–76)

0.26

Ranges are provided in brackets.

0.30 0.88

Page 5 of 10

Arterial– LV– LA interaction post CoA and IAA repair

Table 2

Doppler indices and parameters of LV and LA deformation Patients (n 5 31)

Controls (n 5 31)

P-value

............................................................................................................................................................................... Mitral inflow Doppler indices E (cm/s) A (cm/s) E/A E deceleration time (ms) Mitral annular tissue Doppler imaging

96.2 + 19.9 52.4 + 12.6

89.7 + 16.9 42.4 + 8.2

0.17 ,0.001

1.9 + 0.5

2.2 + 0.5

0.033

119.2 + 35.3

120.1 + 20.1

0.91

s (cm/s)

6.3 + 1.5

7.8 + 1.5

,0.001

e (.21 cm/s) a (.21 cm/s)

9.4 + 1.8 4.4 + 1.0

11.5 + 2.1 5.1 + 1.1

,0.001 0.026

e/a E/e IVA (m/s2)

2.3 + 0.7

2.4 + 0.7

0.62

10.5 + 3.4 1.2 + 0.6

8.0 + 1.7 1.1 + 0.5

,0.001 0.62

0.001

LV deformation Longitudinal deformation Systolic strain (.21%)

15.1 + 3.0

17.3 + 2.1

SRs (s21)

0.8 + 0.2

0.8 + 0.1

0.14

SRe (s21) SRa (s21)

1.0 + 0.2 0.5 + 0.1

1.3 + 0.3 0.5 + 0.1

,0.001 0.38

39.2 + 12.3 2.0 + 0.7

48.7 + 13.5 2.0 + 0.4

Radial deformation Systolic strain (%) SRs (s21) SRe (s21) SRa (s21) Circumferential deformation

0.005 0.96

1.9 + 0.7

2.3 + 0.6

0.022

0.9 + 0.4

0.9 + 0.3

0.77

20.2 + 3.8

20.3 + 3.9

0.97

SRs (s21) SRe (s21)

1.2 + 0.3 1.6 + 0.4

1.1 + 0.3 1.6 + 0.4

0.77 0.95

SRa (s21)

0.5 + 0.2

0.5 + 0.2

0.59 ,0.001

Systolic strain (.21%)

Torsional deformation Peak torsion (8)

11.4 + 5.4

16.9 + 6.1

98.9 + 35.7

110.5 + 31.0

0.18

84.8 + 29.4

131.2 + 49.1

,0.001

Positive strain (%) Negative strain (.21%)

24.5 + 7.2 11.3 + 4.2

35.2 + 6.0 12.3 + 2.2

,0.001 0.23

Total strain (%)

35.8 + 10.4

47.5 + 7.0

,0.001

Peak systolic twisting velocity (8/s) Peak diastolic untwisting velocity (.218/s) LA deformation and area Strain

Strain rate At ventricular systole (s21)

1.7 + 0.5

2.2 + 0.5

,0.001

At early diastole (s21)

2.0 + 0.6

3.1 + 0.5

,0.001

At atrial contraction (s21) Indexed LA area (cm2/m2)

1.3 + 0.6 6.4 + 1.1

1.7 + 0.4 6.2 + 1.0

0.001 0.51

A, peak mitral inflow velocity at late diastole; a, late diastolic annular myocardial velocity; E, peak mitral inflow velocity at early diastole; e, early diastolic annular myocardial velocity; s, systolic annular myocardial velocity; SRa, late diastolic strain rate; SRe, early diastolic strain rate; SRs, systolic strain rate.

longitudinal SRe (P ¼ 0.006), and peak diastolic untwisting velocity (P ¼ 0.001). With regard to atrial strain rates, aSRed correlated positively with LV mitral annular e velocity (P ≤ 0.001), LV longitudinal (P ¼ 0.001) and radial SRe (P ¼ 0.041), and peak untwisting velocity (P ¼ 0.001), while aSRac correlated positively peak untwisting velocity (P , 0.016) (Figure 3).

Multiple linear regression findings Multiple linear regression of the entire cohort was used to identify significant arterial and LV diastolic functional predictors of LA deformation indices. Dependent parameters were LA total strain, aSRed, and aSRac, while independent covariates entered into each of the models included age, sex, systolic and diastolic blood

Page 6 of 10

Table 3

V.W.-y. Li and Y.-f. Cheung

Correlations between arterial and LV deformation indices in patients and controls All subjects

........................................................... Arterial stiffness index

......................... r

P-value

Exclusion of patients with bicuspid aortic valve and subaortic membrane

...........................................................

Carotid intima-media thickness

Arterial stiffness index

.........................

.........................

r

r

P-value

P-value

Carotid intima-media thickness

.........................

r

P-value

............................................................................................................................................................................... Tissue Doppler imaging s

20.23

0.08

20.44

,0.001

20.22

0.12

20.42

0.002

e a

20.40 0.04

0.001 0.75

20.46 20.32

,0.001 0.010

20.39 0.03

0.004 0.84

20.47 20.35

,0.001 0.010

IVA Longitudinal deformation Systolic strain

0.06

0.66

0.08

0.53

0.11

0.42

0.12

0.40

20.49

,0.001

20.01

0.95

20.50

,0.001

20.00

1.00

SRs

20.36

0.004

20.10

0.45

20.33

0.016

20.06

0.69

SRe SRa

20.46 20.03

,0.001 0.79

20.27 20.15

0.031 0.24

20.44 20.06

0.001 0.70

20.23 20.10

0.09 0.47

20.28 0.10

0.032 0.44

20.11 20.14

0.40 0.26

20.20 0.27

0.15 0.56

20.12 20.08

0.40 0.57

Radial deformation Systolic strain SRs SRe SRa Circumferential deformation

0.08

0.56

20.21

0.10

0.09

0.51

20.15

0.29

0.22

0.08

20.07

0.60

0.33

0.016

20.03

0.82

Systolic strain

20.19

0.14

0.21

0.10

20.23

0.10

0.18

0.20

SRs SRe

0.06 20.06

0.66 0.63

0.04 0.07

0.73 0.54

0.03 20.06

0.86 0.70

0.06 0.10

0.68 0.43

SRa

0.23

0.08

0.07

0.57

0.30

0.028

0.05

0.70

Torsional deformation Peak torsion Peak systolic twisting velocity Peak diastolic untwisting velocity

20.12

0.36

20.24

0.07

20.06

0.68

20.21

0.13

0.07

0.61

20.23

0.08

0.10

0.46

20.22

0.11

20.16

0.22

20.32

0.014

20.12

0.42

20.30

0.028

a, late diastolic annular myocardial velocity; e, early diastolic annular myocardial velocity; s, systolic annular myocardial velocity; SRa, late diastolic strain rate; SRe, early diastolic strain rate; SRs, systolic strain rate.

pressure, arterial stiffness index, LV longitudinal and radial SRe, and LV peak untwisting velocity. Significant independent correlates of total LA strain were arterial stiffness index (b ¼ 21.3, P ¼ 0.034) and LV peak untwisting velocity (b ¼ 0.08, P ¼ 0.008), while the only significant independent correlate of aSRed was LV peak untwisting velocity (b ¼ 0.007, P ¼ 0.001). No significant determinants of aSRac were identified.

Discussion This study demonstrates increased carotid arterial stiffness and IMT, impaired LV systolic and diastolic deformation and torsional mechanics, and abnormal LA mechanics in adolescents and young adults with repaired CoA and IAA. Notably, these abnormalities are evident in the absence of residual aortic gradients and resting systemic hypertension. Furthermore, ventriculo-arterial interaction is suggested by negative associations between arterial stiffness and IMT and indices of LV deformation. Additionally, LA –LV interaction is evidenced by the significant associations between LA deformation

parameters and indices of LV diastolic function. To our knowledge, this is the first study to comprehensively assess potential arterial– LV– LA interaction after relief of left-sided obstructive lesions.

Arterial function Dysfunction of arterial segments proximal to the site of coarctation and IAA are well reported. In patients with repaired CoA, reduced distensibility of the ascending but not distal thoracic aorta,22 increased ascending aortic stiffness23,24 and brachioradial pulse wave velocity,2 and reduced brachial vasodilator response to flowmediated dilation and glyceryl trinitrate25 have been demonstrated. The latter finding suggests possibility of abnormal smooth muscle function or structural abnormalities of the arterial wall. In patients with IAA and CoA, enhanced pressure wave reflection from the reconstructed site has been demonstrated.26 With regard to structural alteration, increased carotid and other preductal arterial IMT has also been shown.4,24 Our findings of increased carotid arterial stiffness and IMT hence agree with those previously reported. On the other hand, the absence of differences in baPWVs between

Arterial– LV– LA interaction post CoA and IAA repair

Page 7 of 10

Figure 2: Scatter plots showing relationships between arterial stiffness and LV deformation indices. Arterial stiffness index correlated negatively with LV (A) systolic strain, (B) systolic strain rate, and (C) early diastolic strain rate. Solid circles represent patients, empty circles represent controls.

patients and controls might be related to incorporation in the assessment of both the proximal and the distal arterial segments. Notwithstanding, increased afterload with stiffening of proximal arterial segments may negatively impact on ventricular performance through adverse ventriculo-arterial interaction.

Ventricular mechanics The apparent LV hypercontractile state reported in patients with repaired CoA6,7 has been attributed to overestimation of fibre shortening inherent in the use of M-mode-derived endocardial indices, especially in the presence of LV hypertrophy.8 Subsequent 3D magnetic resonance tagging has demonstrated that even M-mode assessment of mid-wall fibre shortening after CoA repair is artifactual.27 On the other hand, direct interrogation of myocardial deformation may shed light on alteration of ventricular mechanics. Data on LV myocardial deformation are nonetheless limited in these patients. This study provides insights not only into LV systolic myocardial deformation but also into diastolic myocardial deformation and

torsional mechanics in patients late after repair of CoA and IAA. Our findings of reduced systolic myocardial velocity, systolic longitudinal and radial strain, and preservation of circumferential strain agreed with those reported previously.28 – 30 Studies on diastolic function in these patients were limited to regional tissue Doppler assessment of longitudinal diastolic mitral annular velocity.31,32 In this study, we found reduced LV early diastolic strain rates along both longitudinal and radial dimensions. These suggest that diastolic deformational abnormalities occur beyond the longitudinal dimension late after CoA and IAA repair despite the absence of residual aortic gradient. Our novel findings of significantly reduced systolic torsion and diastolic untwisting velocity in patients are intriguing. LV torsional deformation plays an important role in LV ejection and filling. Peak LV torsion has been shown to correlate with LV dP/dtmax,33 while diastolic untwisting parameters are related to invasive indices of LV relaxation and suction.34 Children with CoA or valvar aortic stenosis have been reported to have compensatory increase in LV peak

Page 8 of 10

Table 4

V.W.-y. Li and Y.-f. Cheung

Correlation between LA and LV deformation indices LA total strain

................................ r

P-value

aSRed

..............................

r

P-value

0.55

,0.001

aSRac

................................

r

P-value

............................................................................................................................................................................... e

0.49

,0.001

a

0.24

0.07

–

–

0.18

0.15

Longitudinal SRe Longitudinal SRa

0.35 0.12

0.006 0.37

0.42 –

0.001 –

– 0.22

– 0.09 –

–

–

Radial SRe

0.17

0.20

0.26

0.041

–

Radial SRa

20.10

0.45

–

–

0.08

0.56

Circumferential SRe Circumferential SRa

20.00 20.07

0.99 0.57

0.01 –

0.92 –

– 20.07

– 0.53

0.001

0.43

0.001

Peak diastolic untwisting velocity

0.42

0.31

0.016

a, late diastolic annular myocardial velocity; e, early diastolic annular myocardial velocity; aSRac, atrial strain rate at atrial contraction; aSRed, atrial strain rate at early diastole; SRa, late diastolic strain rate; SRe, early diastolic strain rate.

Figure 3: Scatter plots showing relationships between LV diastolic and LA deformation indices. LV longitudinal early diastolic strain rate correlated positively with (A) LA total strain and (B) LA strain rate at early diastole. Peak LV untwisting diastolic velocity correlated also positively with (C) LA total strain and (D) LA strain rate at early diastole. Solid circles represent patients, empty circles represent controls.

Page 9 of 10

Arterial– LV– LA interaction post CoA and IAA repair

torsion before intervention, which decreases after relief of the obstruction.35 In patients with varying degrees of residual CoA, Young et al. 28 found similar LV torsional parameters using cardiac resonance tagging. In contrast, our findings in adolescent and young adult patients without significant residual CoA suggest significant impairment of LV torsional mechanics. While the exact cause remains uncertain, several mechanisms of impaired LV deformation and torsion in our patients can be hypothesized. Myocardial fibrosis documented histologically36 and by T1 cardiac magnetic resonance37 in patients with repaired CoA may affect strain38 and torsion.39 Abnormal myocardial perfusion reserve detected in adults with repaired CoA40 may further promote the fibrotic process. Increased arterial wave reflection26 and arterial elastance3 in patients may possibly exert negative effects on LV torsion.41 Our findings of negative correlations between carotid arterial indices and systolic and diastolic deformation indices suggest potential adverse ventriculo-arterial interaction as discussed below.

Atrial mechanics Although the clinical relevance of LA dysfunction is increasingly recognized,13,14 the function of left atrium in patients with repaired aortic arch obstruction has previously not been explored. With advances in non-invasive evaluation of atrial function, STE is increasingly used to assess LA function given its relative angle independence and ability to assess global atrial mechanics.10,11 Our findings of reduced LA positive and negative strain and strain rates at all phases of the cardiac cycle suggest impairment of LA pump, conduit, and reservoir function in our patients.

Atrioventricular and ventriculo-arterial interaction In patients with repaired CoA and IAA, the presence of LV diastolic dysfunction as shown in previous studies31,32 and in the present one may affect atrial mechanics. Abnormal atrioventricular interaction has been described in patients with heart failure with normal ejection fraction42 and in hypertensive patients.15,43 In our patients, possible atrioventricular diastolic interaction is reflected by positive associations between indices of LA deformation and those of LV diastolic deformation (Table 4). The possible contribution of adverse ventriculo-arterial interaction is suggested by significant inverse associations between carotid arterial stiffness and LV systolic and diastolic deformation, especially in the longitudinal dimension. Other groups have previously shown that aortic stiffness is an independent predictor of LV longitudinal systolic strain rate,29 and negative correlations between septal mitral annular early diastolic velocity and ascending aortic stiffness in children32 and adults31 with repaired CoA. Nevertheless, this adverse interaction is unlikely to be sole important factor given the well-recognized alteration of ventricular myocardium and function due to fibrosis,36,37 ischaemia,40 and previous open heart surgery for repair of ventricular septal defect. Additionally, in patients with CoA and IAA, the presence of bicuspid aortic valve may be associated with stenosis and ascending aortopathy,44 which may possibly also affect LV deformation. We have therefore performed additional analyses with exclusion of patients with bicuspid aortic valve (Table 3).

Recently, the concept of LA– LV –arterial coupling has been described in patients with cardiovascular risk factors.15 Intuitively, association of arterial stiffness and LA function is possibly mediated through LV diastolic dysfunction. Interestingly, arterial stiffness has been shown also to associate independently with LA diameter45 and LA strain rate during ventricular systole15 in patients with hypertension and other cardiovascular risk factors. We also found a negative association between carotid arterial stiffness and LA total strain, the latter reflecting the reservoir function, independent of LV diastolic function. The underlying mechanism requires, however, further clarification.

Limitations Several limitations should be addressed. First, this is a cross-sectional study that does not provide prognostic information. Nonetheless, this study highlights the role of arterial stiffness in the pathogenesis not only LV but also LA dysfunction and potential benefits of arterial destiffening strategies46 in patients late after repair of CoA and IAA. Secondly, the small sample size has limited the ability to perform further subgroup analyses comparing possible differences between CoA and IAA patients and between different interventional modalities. Thirdly, two patients with Turner syndrome have been recruited. The reported structural and functional abnormalities may pose additional afterload to the left ventricle.47 This would not abrogate, but rather accentuate, the adverse ventriculo-arterial and potential atrial-arterial interactions. Residual mitral regurgitation in the patient with repaired mitral valve cleft may potentially confound the assessment of deformation of the left ventricle and left atrium and their interaction. Finally, only resting arterial and echocardiographic parameters were determined. Given the occurrence of exerciseinduced hypertension in these patients,48 assessment of arterial function and LV and LA deformation parameters may shed more light on the stress-induced dynamic interaction.

Conclusion Alteration of arterial function and LV and LA mechanics occur late after repair of CoA and IAA despite the absence of significant residual CoA. Our findings of associations between these indices suggest the possibility of abnormal arterial –LV –LA coupling. These data must, however, be regarded as preliminary given the relatively small size and heterogeneity of the patient cohort. Further large-scale studies are required to confirm these findings and to determine the mechanisms and prognostic implications of LA and LV dysfunction in these patients.

Funding The study is supported by the Small Project Funding, Internal Research Support Programme of The University of Hong Kong. Conflict of interest: None declared.

References 1. Sehested J, Baadrup U, Mikkelsen E. Different reactivity and structure of the prestenotic and poststenotic aorta in human coarctation: implications for baroreceptor function. Circulation 1982;65:1060 –5. 2. de Divitiis M, Pilla C, Kattenhorn M, Zadinello M, Donald A, Leeson P et al. Vascular dysfunction after repair of coarctation of the aorta impact of early surgery. Circulation 2001;104:I165 –70.

Page 10 of 10 3. Senzaki H, Iwamoto Y, Ishido H, Masutani S, Taketazu M, Kobayashi T et al. Ventricular-vascular stiffening in patients with repaired coarctation of aorta. Circulation 2008;118:S191–8. 4. Brili S, Tousoulis D, Antoniades C, Aggeli C, Roubelakis A, Papathanasi S et al. Evidence of vascular dysfunction in young patients with successfully repaired coarctation of aorta. Atherosclerosis 2005;182:97–103. 5. Trojnarska O, Mizia-Stec K, Gabriel M, Szczepaniak-Chichel L, KatarzynskaSzymanska A, Grajek S et al. Parameters of arterial function and structure in adult patients after coarctation repair. Heart Vessels 2011;26:414 –20. 6. Kimball TR, Reynolds JM, Mays WA, Khoury P, Claytor RP, Daniels SR. Persistent hyperdynamic cardiovascular state at rest and during exercise in children after successful repair of coarctation of the aorta. J Am Coll Cardiol 1994;24:194 –200. 7. Carpenter MA, Dammann JF, Watson DD, Jedeikin R, Tompkins DG, Beller GA. Left ventricular hyperkinesia at rest and during exercise in normotensive patients 2 to 27 years after coarctation repair. J Am Coll Cardiol 1985;6:879 –86. 8. Gentles TL, Sanders SP, Colan SD. Mispresentation of left ventricular contractile function by endocardial indexes: clinical implications after coarctation repair. Am Heart J 2000;140:585 –95. 9. Kutty S, Rangamani S, Venkataraman J, Li L, Schuter A, Fletcher S et al. Reduced global longitudinal and radial strain with normal left ventricular ejection fraction late after effective repair of aortic coarctation: a CMR feature tracking study. Int J Cardiovasc Imaging 2013;29:141 –50. 10. Saraiva RM, Demirkol S, Buakhamsri A, Greenberg N, Popovic´ ZB, Thomas JD et al. Left atrial strain measured by two-dimensional speckle tracking represents a new tool to evaluate left atrial function. J Am Soc Echocardiogr 2010;23:172 –80. 11. Todaro MC, Choudhuri I, Belohlavek M, Jahangir A, Carerj S, Oreto L et al. New echocardiographic techniques for evaluation of left atrial mechanics. Eur Heart J Cardiovasc Imaging 2012;13:973 –84. 12. Hou J, Yu HK, Wong SJ, Cheung YF. Atrial mechanics after surgical repair of tetralogy of Fallot. Echocardiography 2014; doi:10.1111/echo.12611. 13. Cameli M, Lisi M, Mondillo S, Padeletti M, Ballo P, Tsioulpas C et al. Left atrial longitudinal strain by speckle tracking echocardiography correlates well with left ventricular filling pressures in patients with heart failure. Cardiovasc Ultrasound 2010;8:14. 14. Mondillo S, Cameli M, Caputo ML, Lisi M, Palmerini E, Padeletti M et al. Early detection of left atrial strain abnormalities by speckle-tracking in hypertensive and diabetic patients with normal left atrial size. J Am Soc Echocardiogr 2011;24:898 –908. 15. Miyoshi H, Mizuguchi Y, Oishi Y, Iuchi A, Nagase N, Ara N et al. Early detection of abnormal left atrial-left ventricular-arterial coupling in preclinical patients with cardiovascular risk factors: evaluation by two-dimensional speckle-tracking echocardiography. Eur J Echocardiogr 2011;12:431 –9. 16. Senzaki H, Kumakura R, Ishido H, Masutani S. Left atrial systolic force in children: reference values for normal children and changes in cardiovascular disease with left ventricular volume overload or pressure overload. J Am Soc Echocardiogr 2009; 22:939–46. 17. Cortez-Cooper MY, Supak JA, Tanaka H. A new device for automatic measurements of arterial stiffness and ankle-brachial index. Am J Cardiol 2003;91:1519 –22. 18. Vogel M, Schmidt MR, Kristiansen SB, Cheung M, White PA, Sorensen K et al. Validation of myocardial acceleration during isovolumic contraction as a novel noninvasive right ventricular contractility: comparison with ventricular pressure-volume relations in an animal model. Circulation 2002;105:1693 –9. 19. Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler tissue imaging: a non-invasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 1997;30:1527 –33. 20. Cheung EW, Liang XC, Lam WW, Cheung YF. Impact of right ventricular dilation on left ventricular myocardial deformation in patients after surgical repair of tetralogy of Fallot. Am J Cardiol 2009;104:1264 –70. 21. Yu W, Li SN, Chan GC, Ha SY, Wong SJ, Cheung YF. Transmural strain and rotation gradient in survivors of childhood cancers. Eur Heart J Cardiovasc Imaging 2013;14: 175 –82. 22. Brili S, Dernellis J, Aggeli C, Pitsavos C, Hatzos C, Stefanadis C et al. Aortic elastic properties in patients with repaired coarctation of the aorta. Am J Cardiol 1998;82: 1140 –3. 23. Ou P, Celermajer DS, Jolivet O, Buyens F, Herment A, Sidi D et al. Increased central aortic stiffness and left ventricular mass in normotensive young subjects after successful coarctation repair. Am Heart J 2008;155:187–93. 24. Sarkola T, Redington AN, Slorach C, Hui W, Bradley T, Jaeggi E. Assessment of vascular phenotype using a novel very-high-resolution ultrasound technique in adolescents after aortic coarctation repair and/or stent implantation: relationship to central haemodynamics and left ventricular mass. Heart 2011;97:1788 –93. 25. Gardiner HM, Celermajer DS, Sorensen KE, Georgakopoulos D, Robinson J, Thomas O et al. Arterial reactivity is significantly impaired in normotensive young

V.W.-y. Li and Y.-f. Cheung

26. 27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40. 41. 42.

43.

44.

45.

46.

47.

48.

adults after successful repair of aortic coarctation in childhood. Circulation 1994; 89:1745 –50. Murakami T, Takeda A. Enhanced aortic pressure wave reflection in patients after repair of aortic coarctation. Ann Thorac Surg 2005;80:995 –9. Gentles TL, Cowan BR, Occleshaw CJ, Colan SD, Young AA. Midwall shortening after coarctation repair: the effect of through-plane motion on single-plane indices of left ventricular function. J Am Soc Echocardiogr 2005;18:1131 –6. Young AA, Cowan BR, Occleshaw CJ, Oxenham HC, Gentles TL. Temporal evolution of left ventricular strain late after repair of coarctation of the aorta using 3D MR tissue tagging. J Cardiovasc Magn Reson 2002;4:233 –43. Di Salvo G, Pacileo G, Limongelli G, Verrengia M, Rea A, Santoro G et al. Abnormal regional myocardial deformation properties and increased aortic stiffness in normotensive patients with aortic coarctation despite successful correction: an ABPM standard echocardiography and strain rate imaging study. Clin Sci (Lond) 2007;113: 259 –66. Kowalski M, Kowalik E, Kotlinski K, Szymanski P, Kusmierxzyk M, Rozanski J et al. Regional left ventricular myocardial shortening in normotensive patients late after aortic coarctation repair-normal or impaired. Ultrasound Med Biol 2009;35: 1947 –52. Lam YY, Mullen MJ, Kaya MJ, Gatzoulis MA, Li W, Henein MY. Left ventricular long axis dysfunction in adults with ‘corrected’ aortic coarctation is related to an older age at intervention and increased aortic stiffness. Heart 2009;95:733 –9. Lombari KC, Northrup V, McNamara RL, Sugeng L, Weismann CG. Aortic stiffness and left ventricular diastolic function in children following early repair of aortic coarctation. Am J Cardiol 2013;112:1828 –33. Kim WJ, Lee BH, Kim YJ, Kang JH, Jung YJ, Song JM et al. Apical rotation assessed by speckle-tracking echocardiography as an index of global left ventricular contractility. Circ Cardiovasc Imaging 2009;2:123 –31. Burns AT, La Gerche A, Prior DL, Macisaac AI. Left ventricular untwisting is an important determinant of early diastolic function. JACC Cardiovasc Imaging 2009;2: 709 –16. Laser KT, Haas NA, Jansen N, Scha¨ffler R, Palacios Argueta JR, Zittermann A et al. Is torsion a suitable echocardiographic parameter to detect acute changes in left ventricular afterload in children. J Am Soc Echocardiogr 2009;22:1121 –8. Chetlin MD, Robinowitz M, McAllister H, Hoffman JI, Bharati S, Lev M. The distribution of fibrosis in the left ventricle in congenital aortic stenosis and coarctation of the aorta. Circulation 1980;62:823 –30. Broberg CS, Chugh SS, Conklin C, Sahn DJ, Jerosch-Herold M. Quantification of diffuse myocardial fibrosis and its association with myocardial dysfunction in congenital heart disease. Circ Cardiovasc Imaging 2010;3:727–34. Saito M, Okayama H, Yoshii T, Higashi H, Morioka H, Hiasa G et al. Clinical significance of global two-dimensional strain as a surrogate parameter of myocardial fibrosis and cardiac events in patients with hypertrophic cardiomyopathy. Eur Heart J Cardiovasc Imaging 2012;13:617 –23. Karaahmet T, Gurel E, Tigen K, Guler A, Dundar C, Fotbolcu H et al. The effect of myocardial fibrosis on left ventricular torsion and twist in patients with non-ischemic dilated cardiomyopathy. Cardiol J 2013;20:276–86. Cook SC, Ferketich AK, Raman SV. Myocaridal ischemia in asymptomatic adults with repaired aortic coarctation. Int J Cardiol 2009;133:95 –101. MacGowan GA, Burkhoff D, Rogers WJ, Salvador D, Azhari H, Hees PS et al. Effects of afterload on regional left ventricular torsion. Cardiovasc Res 1996;31:917 – 25. Morris DA, Gailani M, Perez AV, Blaschkle F, Dietz R, Haverkamp W et al. Left atrial systolic and diastolic dysfunction in heart failure with normal left ventricular ejection fraction. J Am Soc Echocardiogr 2011;24:651 –62. Miyoshi H, Oishi Y, Mizuguchi Y, Ichi A, Nagase N, Ara N et al. Early predictors of alterations in left atrial structure and function related to left ventricular dysfunction in asymptomatic patients with hypertension. J Am Soc Hypertens 2013;7:206 –15. Tadros TM, Klein MD, Shapira OM. Ascending aortic dilatation associated with bicuspid aortic valve: pathophysiology, molecular biology, and clinical implications. Circulation 2009;119:880 –90. Lantelme P, Laurent S, Besnard C, Bricca G, Vincent M, Legedz L et al. Arterial stiffness is associated with left atrial size in hypertensive patients. Arch Cardiovasc Dis 2008;101:35– 40. Koumaras C, Tzimou M, Stavrinou E, Griva T, Gossio T, Katsiki N et al. Role of antihypertensive drugs in arterial ‘destiffening’ and central pulsatile hemodynamics. Am J Cardiovasc Drugs 2012;12:143 – 56. Baguet JP, Douchin S, Pierre H, Rossignol AM, Bost M, Mallion JM. Structural and functional abnormalities of large arteries in the Turner syndrome. Heart 2005;91: 1442 –46. Luijendijk P, Bouma BJ, Vriend JW, Vliegen HW, Groenink M, Mulder BJ. Usefulness of exercise-induced hypertension as predictor of chronic hypertension in adults after operative therapy for aortic isthmic coarctation in childhood. Am J Cardiol 2011;108:435 –9.