Pediatric Cardiology

, Volume 34, Issue 4, pp 948–953

Altered Diastolic Left Atrial and Ventricular Performance in Asymptomatic Patients After Repair of Tetralogy of Fallot


    • Clinic for Paediatric CardiologySaarland University Hospital
  • Tanja Raedle-Hurst
    • Clinic for Paediatric CardiologySaarland University Hospital
  • Meryem Hosse
    • Clinic for Paediatric CardiologySaarland University Hospital
  • Maxi Hauser
    • Clinic for Paediatric CardiologySaarland University Hospital
  • Hashim Abdul-Khaliq
    • Clinic for Paediatric CardiologySaarland University Hospital
Original Article

DOI: 10.1007/s00246-012-0584-1

Cite this article as:
Koenigstein, K., Raedle-Hurst, T., Hosse, M. et al. Pediatr Cardiol (2013) 34: 948. doi:10.1007/s00246-012-0584-1


We evaluated the interaction of left atrial and ventricular diastolic performance in asymptomatic children and young adults after ToF-repair (n=25). Those young people, as well as 25 age matched healthy children and young adults were examined using non-invasive conventional echocardiography. Regional systolic and diastolic myocardial strain and strain rate in left atrium and ventricle were analysed using 2D-speckle-tracking (Vivid VII, EchoPacGE). We collected planimetric data about the left atrial and ventricular performance during systole (volumetric LVEF, LV-Tei-Index, MV-E/E'-Ratio) and diastole (LAEF, LVEDV, left atrial volume). Registration of right pulmonary-venous inflow-patterns during ventricular systole, diastole and active atrial contraction was used to support assessment of left atrial function. To verify the timing of left atrial contraction and possible electromechanical delay we measured several ECG-related time-intervals. Statistical analysis included Mann-Whitney-U-Test, Bonferroni-Holm-Test and two-tailed Spearman-Correlation. Systolic pulmonary-venous inflow in ToF-patients was not different compared to the controls. Early diastolic pulmonary-venous inflow was significantly higher in ToF-patients as well as the LV-Tei-Index. The MV-E/E'-ratio, which is closely related to LVEDP, was significantly higher in ToF-patients and correlated with the early diastolic pulmonary venous inflow parameters such as the maximum diastolic bloodflow speed. Diastolic left atrial and ventricular strain and strain rate in ToF-patients did not differ from those in the controls. During late diastole there was a significantly premature timing of maximum myocardial strain rate of the interatrial septum and time-ratio of P-wave origin to maximum reverse pulmonary-venous blood flow and the duration of one heart action. Furthermore the maximum late diastolic reverse pulmonary-venous blood flow was significantly higher in ToF-patients. Those observations indicate a premature active left atrial contraction in late diastole in ToF-patients compared to the controls. In asymptomatic young patients after ToF-repair earlier and increased left atrial contraction was found, which may indicate adaptive compensatory mechanisms to overcome latent and asymptomatic altered systolic and diastolic left ventricular performance. Extensive assessment of left atrial parameters including the pulmonary veins should be considered in terms of an entire evaluation of left heart function in patients after ToF-repair.


Tetralogy of FallotDiastolic dysfunctionLeft atriumLeft ventricleEchocardiography2D speckle tracking


Long-term outcome and prognosis in patients after repair of Tetralogy of Fallot (ToF) mainly depends on progressive right-ventricular (RV) dilatation and pulmonary regurgitation [4, 6]. RV pressure and volume overload has also been found to affect left-ventricular (LV) function and global performance [2, 7, 9, 18]. In patients after ToF-repair, the majority of studies has been focused on right-ventricular (RV) and left-ventricular (LV) systolic function, whereas little attention has been given to possible LV diastolic dysfunction and altered left atrial performance [9, 15]. Such alterations in diastolic function in patients after ToF-repair may already occur at early age and become clinically relevant in older age. As LV myocardial relaxation decreases, the left atrium must compensate the altered filling by initially increased performance [13, 16], which may result in left atrial dilatation and failure during long-term follow-up.

Therefore, the aim of our study was to analyze and characterize left atrial and LV diastolic performance in asymptomatic children and adolescent after ToF-repair and compare the results with age-matched controls.



Twenty-five clinically asymptomatic children and adolescent after ToF-repair without severe pulmonary regurgitation or stenosis (systolic gradient < 40 mmHg or regurgitation less than grade II) underwent echocardiographic examination during a follow-up-period of 8.83 ± 6.1 years. The age at surgical repair was 3.79 ± 7.1 years. Twenty-five sex- and age-matched healthy individuals undergoing cardiac examination to exclude structural and functional cardiac abnormalities served as controls. Clinical and systolic echocardiographic features did not differ significantly between both groups. LV ejection fraction (LVEF) was 60.2 ± 6.4 % in the ToF group versus 61.8 ± 4.6 % in the control group. At the time when echocardiographic data were obtained, all of the ToF-patients had no clinical symptoms other than 19 patients having dyspnea New York Heart Association (NYHA) stage I and 5 patients having dyspnea NYHA stage II. Electrocardiogram (ECG) showed no bundle branch blocks in any of the participants in this study, and the QRS duration was not significantly different in both groups (122 ± 38 vs. 118 ± 33 ms). Systolic LV strain was comparable in both groups (interventricular septum [IVS] −18.40 ± 3.4 % versus −19.25 ± 4.6 % and LV lateral wall [LVW] −16.38 ± 5.2 % versus −13.97 ± 8.2 %). However, left atrial systolic strain was lower in the interatrial septum (IAS) of the ToF-patients (43.25 ± 29.6 % versus 57.11 ± 20.9 %, p = 0.032), whereas there were no such findings of significant differences in the lateral left atrial wall (LAV; 58.41 ± 29.2 % vs. 50.75 ± 18.5 %).

Echocardiographic Examination

Echocardiographic examination was performed in the Department for Pediatric Cardiology at the Saarland University in Homburg/Saar, Germany, by the same investigators (H. A. K, T. R. H.). A 2.5- to 3.5-MHz phased-array probe and the Vingmed Vivid 7 ultrasound system (General Electric Vingmed, Horten, Norway) were used. Simultaneous recording of an ECG was performed with the echocardiographic studies. Data were obtained in four-chamber view. After creating a global view, inflow patterns of the right superior pulmonary vein—including blood flow velocities and velocity time integrals—were assessed with pulsed wave (PW) Doppler. Inflow and outflow patterns over the mitral and aortic valve were also recorded in the same manner. The left atrial and LV volumes were visualized in optimal view to calculate planimetric atrial and ventricular areas and volumes. We used the tissue Doppler mode to analyze the velocities of movement and the deformation of the myocardium of the interatrial and -ventricular septum and the lateral left atrial and ventricular wall.

Two-Dimensional Speckle-Tracking and Offline Analysis of Planimetric Data

Pulmonary Veins

Analyzed data included the following: velocity-time-integral; maximum blood flow velocity; mean blood flow velocity; time ratio of systolic, early, and late-diastolic pulmonary venous blood flow with the duration of one heart action (Table 1).
Table 1

Echocardiographic data for left atrial and LV diastolic performance showing significant alterations in both early and late diastole

Diastolic echocardiographic data

ToF (n = 25)

Controls (n = 25)


Left atrial and LV volumetric data

 LVEDV/BSA (ml/m²)

50.37 ± 9.8

64.81 ± 12.8


 LAEF1 (%)

85.59 ± 10.3

84.14 ± 5.8


 LAEF2 (%)

62.84 ± 13.6

66.26 ± 7.7


 LAEF3 (%)

18.61 ± 8.0

18.37 ± 5.4


 LAEF3/1 (%)

21.42 ± 11.1

21.04 ± 6.7


 LAV2/BSA (ml/m²)

4.55 ± 2.9

6.25 ± 3.1


 LAV3/BSA (ml/m²)

2.00 ± 1.9

2.52 ± 1.8


Left atrial and LV planimetric data

 LV-Tei index

0.57 ± 0.11

0.47 ± 0.09


 MV-E/E’ S ratio

12.62 ± 4.8

8.87 ± 1.8


 MV-E/E’ W ratio

13.69 ± 7.1

8.81 ± 6.1


 MV-E/A ratio

1.96 ± 0.5

2.11 ± 0.4


 MV-E/A deceleration time (ms)

177.5 ± 43.0

184 ± 51.5


Pulmonary venous blood flow


15.27 ± 4.0

12.70 ± 3.6


 PV-vmax D (m/s)

0.87 ± 0.15

0.64 ± 0.15


 PV-vmean D (m/s)

0.67 ± 0.10

0.45 ± 0.11


 PV-time ratio D

0.32 ± 0.05

0.30 ± 0.08



2.70 ± 1.45

1.91 ± 0.78


 PV-time ratio P-A

0.17 ± 0.04

0.14 ± 0.04


NS not significant, BSA body surface area; global (LAEF1), early (LAEF2), and late-diastolic (LAEF3); LAV/BSA left atrial volume/body surface area at the end of the conduit phase (LAV2), respectively, at the point of maximum active left atrial contraction (LAV3); MV-E/E’ = ratio of early diastolic transmitral blood flow to early diastolic deformation of the interventricular septal myocardium (E’S) and the LVW (E’W); MV-E/A deceleration time = ratio of deceleration time of early and late-diastolic transmitral blood flow; PV-VTI D and PV VTI A = integration of pulmonary venous blood flow volume during early and late diastole; PV-vmaxD and PV-vmeanD = maximum and mean velocity of early diastolic pulmonary venous blood flow; PV-time ratio D = ratio of duration of early diastolic pulmonary venous blood flow to duration of one heart action; PV-time ratio P-A = ratio of time interval from P-wave origin to peak VTI A and one heart action

Table 2

LV and left atrial diastolic myocardial function as assessed by tissue Doppler and 2D speckle tracking showing significant alterations in late-diastolic left atrial and LV performance


ToF (n = 25)

Controls (n = 25)


Early diastole

 Left ventricle

Myocardial velocity E IVS (m/s)

–8.20 ± 2.3

–11.02 ± 1.3


Myocardial velocity E LVW (m/s)

–6.94 ± 3.1

–10.71 ± 2.3


Strain rate E IVS (1/s)

1.58 ± 0.6

1.64 ± 0.5


Strain rate E LVW (1/s)

2.07 ± 1.0

1.81 ± 0.8


Time-ratio strain rate E IVS

0.58 ± 0.07

0.57 ± 0.07


Time-ratio strain rate E LVW

0.56 ± 0.07

0.55 ± 0.07


 Left atrium

Myocardial velocity E IAS (m/s)

–3.60 ± 1.9

–3.30 ± 1.8


Myocardial velocity E LAW (m/s)

–2.30 ± 2.3

–3.35 ± 3.0


Strain rate E IAS (1/s)

–2.92 ± 1.5

–3.49 ± 1.2


Strain rate E LAW (1/s)

–3.67 ± 1.6

–3.75 ± 1.3


Time-ratio strain rate E IAS

0.57 ± 0.06

0.56 ± 0.08


Time-ratio strain rate E LAW

0.57 ± 0.07

0.56 ± 0.07


Late diastole

 Left ventricle

Myocardial velocity A IVS (m/s)

–4.13 ± 1.2

–4.51 ± 1.1


Myocardial velocity A LVW (m/s)

–2.82 ± 1.8

–3.32 ± 1.2


Strain rate A IVS (1/s)

0.68 ± 0.20

0.73 ± 0.23


Strain rate A LVW (1/s)

0.72 ± 0.50

0.61 ± 0.52


Time-ratio strain rate A IVS

0.92 ± 0.05

0.94 ± 0.05


Time-ratio strain rate A LVW

0.93 ± 0.04

0.84 ± 0.16


 Left atrium

Myocardial velocity A IAS (m/s)

–2.29 ± 2.4

–2.03 ± 1.4


Myocardial velocity A LAW (m/s)

–3.28 ± 2.2

–2.56 ± 2.2


Strain rate A IAS (1/s)

–2.53 ± 1.7

–2.62 ± 1.4


Strain rate A LAW (1/s)

–2.94 ± 1.3

–2.77 ± 1.2


Time-ratio strain rate A IAS

0.90 ± 0.04

0.94 ± 0.05


Time-ratio strain rate A LAW

0.96 ± 0.06

0.95 ± 0.05


NS not significant

Left Atrium

Evaluation of the left atrium included calculation of end-systolic, middle- and end-diastolic atrial volume according to Dodge et al. (V = 8 × A1 × A2/3πL; A1 = planimetric assessed left atrial area in four-chamber-view; A2 = echocardiographically assessed left atrial area in two-chamber-view; L = maximum longitudinal left atrial diameter in four-chamber view) [5, 11] in relation to body surface area (after Du Bois [BSA = k0,425 × w0,725 × 0.0071 [BSA = body surface area; k = body length; w = body weight]). Global, early, and late-diastolic left atrial ejection fraction (LAEF) and percentage of active atrial contraction by global ejection were also calculated.

Left Ventricle

Obtained parameters were as follows: LV-Tei-Index, maximum blood flow velocity through the mitral valve during early/middle diastole and during active atrial contraction, MV-E/A ratio, MV-E/A deceleration time, and MV-E/E‘ ratio (E‘S = septal part of mitral valve annulus, E’W lateral wall of mitral valve annulus). E’S and E’W were calculated using tissue Doppler and two-dimensional (2D) speckle tracking. Further parameters were LV end-diastolic volume (LVEDV) and LVEF (modified Simpson-formula: LVEF = 2.432 * LVEDV + 130).

2D Speckle Tracking

2D speckle tracking was used to calculate the mean value of maximum velocity of myocardial excursion, strain and strain rate in the interatrial and ventricular septum as well as the lateral left atrial and ventricular wall [17] (Table 2).

Statistical Analysis

Collected data were analyzed in SPSS 15.0 by performing descriptive statistics, Mann–Whitney-U-test, and two-tailed Spearman correlation (with p < 0.05 = significant and p < 0.01 = highly significant). Bonferroni-Holm test was used to limit the number of test variable effects as much as possible (also with p < 0.05 = significant and p > 0.01 = highly significant).


LVEDV appeared to be decreased in the ToF-patients (LVEDV/BSA p = 0.028). The measured parameters reflecting early diastole in the left heart—such as E/E‘ ratio over the mitral valve (MV-E/E’S p = 0.002, MV-E/E’W p = 0.004) and early diastolic pulmonary venous blood flow (PV–VTI D p = 0.004, PV-vmax D p = 0.003, PV-vmean D p = 0.003)—were significantly increased. LV Tei index, a parameter reflecting global systolic and diastolic performance, was significantly increased in the ToF-patients compared with the controls (LV-Tei index p = 0.018). Early diastolic wall velocities in the LVW and IS appeared to be significantly decreased in the ToF-patients (myocardial velocity E IVS p = 0.002 and LVW p = 0.005) as was the early diastolic strain rate in the interatrial septum (strain rate E IAS p = 0.011).

Echocardiographic parameters reflecting late-diastolic left atrial and LV function, such as late-diastolic reverse pulmonary venous blood flow, were significantly increased and somewhat prolonged in terms of duration in ToF-patients compared with controls (PV-VTI A p = 0.003; PV-time ratio P-A p = 0.032). Contributing to this, we found a tendency to increased velocity of the myocardial excursion during active left atrial contraction (myocardial velocity A IAS and LAW p = 0.1) and premature maximum late-diastolic strain rate, which was significant in the IAS (time-ratio strain rate A IAS p = 0.042).


Mid- and long-term cardiac morbidities in patients after ToF-repair are mainly dominated by residual defects in the right heart and RV outflow tract, which results in RV volume and pressure overload as well as altered systolic and diastolic function. Lower LV systolic and diastolic performance in these patients has been interpreted in part as a result of interventricular interactions that primarily originate from the right heart [1, 8, 9, 12]. Some data have shown isolated LV systolic and diastolic dysfunction [3, 14, 15]. Thus, isolated structural and functional pathological substrates in the left heart may not be excluded. Diastolic and atrial performance in the left and right heart contributes significantly to cardiac output. In case of LV systolic and diastolic dysfunction, atrial performance is compensatory increased. Therefore the aim of our study was primarily to analyze left atrial and LV diastolic function and performance using conventional and novel parameters to detect possible early altered diastolic performance in the left heart in apparently asymptomatic young patients after ToF-repair.

In fact, our results demonstrate altered diastolic performance in different phases of diastole by nearly preserved global and regional LV systolic function in a group of asymptomatic children and young adults after ToF-repair. Our results are in accordance with previous results indicating altered diastolic LV performance in apparently asymptomatic young patients after ToF-repair [3].

In our study, systolic function does not appear to be significantly altered because the LVEF in both the ToF-repaired patients and the controls was nearly similar. Riesenkampff et al. [15], instead, found significantly decreased LVEF (57 % in their ToF-patients with altered diastolic and left atrial function as measured by magnetic resonance imaging). Their cohort of ToF-patients was older than the patients in our study. Thus, early altered diastolic performance may indicate progressive alterations of LV performance that become clinically and echocardiographically apparent in older age.

Although our methods are not the “gold standard” by which to assess atrial and ventricular volumes, this study showed indications of a smaller-size left heart in these patients. This may also in part explain the abnormal diastolic relaxation and filling, which was confirmed in our study according to measured abnormal diastolic parameters.

Our echocardiographically in the pulmonary veins and in the left heart measured data regarding early diastolic left atrial and LV function show significantly altered diastolic relaxation and wall excursions, pointing to impaired ventricular filling.

Left atrial conduit function appears to be less effective in our patients after ToF-repair, as early diastolic left atrial ejection fraction was lower than in our controls. Knirsch et al. [10] had controversial results, as they concluded that the myocardial diastolic function of the IVS is not impaired in their patients similar to ours. We, instead, found a globally decreased septal left atrial and LV strain rate and velocity of myocardial contraction during early diastole in ToF-patients. In their study, Riesenkampff et al. [15] examined comparable asymptomatic young adults after ToF-repair (LVEF 57 %, QRS interval duration < 140 ms, and NYHA class I and II). They also showed that even in those young asymptomatic patients, there was a significantly abnormal atrial and ventricular filling in both the right and left heart. Unfortunately, parameters used for analysis of the left heart may not be fully applicable in the right heart. Our study was focused on LV and atrial diastolic function, and we assume that there should be an interaction between both atria regarding abnormalities in filling and relaxation.

In this study, we also assessed the timing of active left atrial contraction. To our knowledge, our data are the first to show altered electromechanical interaction at left atrial levels in patients after ToF-repair. Similar to other described differences between hemodynamic and contraction patterns between the atria and ventricles and compared with reported myocardial contractive delay in the left ventricle [3, 14], we rather found a significantly shortened time to increased active left atrial contraction in ToF-repaired patients compared with healthy controls.

Those data strongly suggest increased left atrial workload in young patients after ToF-repair. This may be understood in part as a compensatory mechanism to overcome impairment of LV diastolic function [15]. Nonetheless, strain rate and velocity of myocardial deformation during active left atrial contraction, as well as late-diastolic LAEF, are similar in both patients and controls. In conclusion, those compensatory mechanisms in young patients after ToF-repair still are well functioning, which may explain the good clinical status and normal systolic performance of these patients compared with older ToF-patients with lower compensatory capacity.

Limitations of the Study

This study included only echocardiographic parameters and no other quantitative tools by which to evaluate myocardial and physical performance. In addition, inclusion of other patients with more severe clinical symptoms might be helpful. The median age of the patients when they underwent cardiac surgery was greater than is common today, so this may not reflect current practice. However, the aim of this prospective study was to evaluate diastolic function in patients who live in the tributary area of our clinic and come regularly in our outpatient’s clinics. In addition, blood sampling, particularly in the young patients, was not considered by the study design. Analysis of atrial myocardial deformation and deformation rate is more time consuming than analysis of ventricular strain and strain rate due to the lesser myocardial thickness and mass. Compared with strain and strain rate values, atrial systolic wall velocities may be passively influenced by longitudinal active excursion as well as movements of the atrioventricular valve in the left and right heart.


We were able to demonstrate abnormal diastolic function in asymptomatic young patients after surgical ToF-repair by different diastolic parameters. Changes in left atrial and LV diastolic performance were found, indicating initial compensatory and adaptive mechanisms to altered ventricular relaxation and filling. Whether those findings may indicate isolated altered left atrial performance is unclear.

Patients after ToF-repair may develop progressive global and regional myocardial dysfunction, which initially can be compensated by increased atrial performance.

More studies are necessary to evaluate the role of diastolic dysfunction in ToF-repaired patients with long-term altered global myocardial performance. Thus, assessment of global and regional left atrial diastolic function, including the pulmonary veins, should be included in the follow-up of such patients.


This work was supported by the Kompetenznetz Angeborene Herzfehler (Competence Network for Congenital Heart Defects) funded by the Federal Ministry of Education and Research (Grant No. FKZ 01G10210).

Copyright information

© Springer Science+Business Media New York 2012