European Heart Journal – Cardiovascular Imaging (2015) 16, 890–899 doi:10.1093/ehjci/jev011

Right ventricular myocardial deformation patterns in children with congenital heart disease associated with right ventricular pressure overload Yasunobu Hayabuchi*, Miho Sakata, and Shoji Kagami Department of Pediatrics, University of Tokushima, Kuramoto-cho-3, Tokushima 770-8503, Japan Received 8 October 2014; accepted after revision 19 January 2015; online publish-ahead-of-print 21 February 2015

Aims

Longitudinal wall motion of the right ventricle (RV) has been thoroughly studied in patients with RV dysfunction. However, circumferential strain of the RV free wall has yet to be investigated. Therefore, this study was conducted to assess the utility of RV free wall circumferential strain. ..................................................................................................................................................................................... Methods Strain profile curves were obtained using speckle tracking echocardiography from the subcostal left ventricular (LV) and results short-axis view in 30 normal children (normal group) and 25 patients with RV pressure overload (RVO group). The time– strain curves of three individual segmental (anterior, lateral, and inferior segments) and global circumferential deformations were evaluated. RV ejection fraction (RVEF), RV systolic pressure (RVSP), and RV fractional area change obtained in the four-chamber view and LV short-axis view [RVFAC (4CH) and RVFAC (SAX), respectively] were measured, and their relationships with RV free wall deformation were assessed. In the normal group, circumferential strain was significantly lower in the anterior segment than in the other segments. The inferior segment had a significantly larger strain than the other segments in the RVO group. Circumferential strain was predominant over longitudinal RV free wall strain in the RVO group (218.4 + 3.9 vs. 214.2 + 3.8%, respectively; P , 0.005), whereas no significant difference between them was observed in the normal group (223.0 + 3.9 vs. 222.4 + 4.7%, respectively). Global circumferential strain had a significantly higher correlation with RVFAC (4CH), RVFAC (SAX), RVEF, and RVSP than global longitudinal strain (P , 0.05 for all). ..................................................................................................................................................................................... Conclusion RV free wall circumferential strain provides better information about RV function than longitudinal strain in children with RVO.

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

Right ventricle † Strain † Speckle tracking † Children † Congenital heart disease

Introduction In the assessment of various cardiac diseases, including congenital heart disease,1 ischaemic heart disease,2 cardiomyopathy,3 and pulmonary arterial hypertension,4,5 it is important to accurately evaluate right ventricular (RV) performance. Consequently, the assessment of RV function is increasingly recognized as important in the management of patients with RV dysfunction. The characterization of the complex morphological and functional properties of the RV has attracted considerable interest. Although current conventional echocardiographic techniques can be used to evaluate RV performance,6 – 8

the quantification of RV function remains a challenge due to the complex geometry of the chamber. Two-dimensional speckle tracking echocardiography is a novel echocardiographic technique that enables angle-independent measurement of regional strain. This technique has recently been introduced for the evaluation of regional RV as well as left ventricular (LV) function.9 – 13 The RV is a complex structure, both anatomically and functionally. Reduced longitudinal function of the RV, including tricuspid annular plane systolic excursion, tissue Doppler-derived tricuspid lateral annular systolic velocity (s′ ), and longitudinal strain, has been used to assess ventricular dysfunction.14 However, longitudinal

* Corresponding author. Tel: +81 886 33 7135; Fax: +81 886 31 8697, Email: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: [email protected].

891

Right ventricular myocardial deformation patterns in children

shortening represents only one aspect of myocardial deformation, and changes in contraction in other dimensions are not well described. Previous investigations have paid less attention to RV circumferential deformation, despite the fact that circumferential movement of the RV free wall is important in RV ejection.15,16 Therefore, the aims of this study were to evaluate the characteristics of RV free wall speckle tracking circumferential deformation and to assess its relationship to RV functional parameters in normal subjects and patients with RV overload.

Methods Study population A total of consecutive 35 healthy children (normal group; mean age, 2.0 + 0.6 years; range, 0.5– 3.0 years) and 27 patients with RV pressure overload (RVO group; mean age, 1.8 + 0.5 years; range, 0.5 – 2.9 years) from December 2012 to January 2014 were examined. Participants were included in this study only if they were between 6 months and 3 years of age. Patients with RVO included 11 patients with pulmonary valvular stenosis and 16 patients with ventricular septal defect (VSD). The VSD was associated with pulmonary hypertension [mean pulmonary arterial pressure (mPAP) of 25 mmHg or greater] in five patients, pulmonary stenosis (pressure gradient of 30 mmHg or greater) in three patients, and pulmonary artery banding in eight patients. All protocols were approved by the Institutional Review Board of the Medical University of Tokushima, and written informed consent was obtained from the parents of all patients. The study protocol conformed to the ethical guidelines outlined in the 1975 Declaration of Helsinki.

Standard echocardiographic study All echocardiographic studies were performed with commercially available echocardiography systems equipped with 1– 5 and 3– 7 MHz sector transducers (Preirus digital ultrasound system; Hitachi-Aloka Medical Co., Tokyo, Japan). The LV end-diastolic dimension (LVEDD) and the LV end-systolic dimension were measured in the parasternal long-axis view. Pulsed Doppler LV and RV inflow were recorded in the modified apical fourchamber view, placing the sample volume at the level of the tips of the leaflets. The peak velocities of early (E) and late (A) LV and RV inflow were measured using pulsed-wave Doppler. Tissue Doppler velocities of the mitral and tricuspid annulus (e′ , a′ , and s′ ) were also evaluated from the apical four-chamber view. LVEF was calculated from apical two-chamber and four-chamber images using the biplane Simpson’s technique. RV fractional area change (FAC) was evaluated in the fourchamber view [RVFAC (4CH)] and the subcostal LV short-axis view [RVFAC (SAX)]. RV end-diastolic and end-systolic areas were quantified from the endocardial contours of the modified apical four-chamber view and the subcostal short-axis view, which focused on the RV. RVFAC was calculated as the absolute area change divided by the end-diastolic area. All measurements were performed in three cardiac cycles and then averaged. All echocardiographic examinations were performed within 3 days of catheterization, and measurements were obtained by an observer blinded to the cardiac catheterization data.

Two-dimensional speckle tracking echocardiography Longitudinal two-dimensional RV free wall deformation Two-dimensional speckle tracking strain echocardiography of RV free wall longitudinal deformation was performed using a routine greyscalemodified apical four-chamber view focused on the RV.

Circumferential two-dimensional RV free wall deformation Circumferential RV free wall strain was evaluated from the subcostal LV short-axis view. A region of interest was traced on the endocardium at end-diastole in the RV from the LV short-axis view at the level of the papillary muscles. The RV free wall was divided into three segments (namely, the anterior, lateral, and inferior segments), and three corresponding time – strain curves were evaluated. Global circumferential RV free wall strain was also generated by the imaging along the entire RV free wall. After optimizing gain, dynamic range, and time gain compensation, the images were digitally recorded at 72–95 frames/s. Special care was taken to fine-tune the region of interest using visual assessment during cine loop playback to ensure that segments were tracked appropriately. All imaging data were digitalized and stored on a hard disk in the ultrasound unit, and then transferred to a personal computer for further analysis. Image analysis was performed using a novel customized software programme with twodimensional speckle tracking (US Image Viewer 2.0, Hitachi-Aloka Medical Co.) containing a pattern-matching algorithm that tracks tissue pixels. A tracking point is selected in the first frame of a two-dimensional echocardiographic image, and the algorithm then searches the next frame for the region that is assumed to be the closest to the selected point, according to the distribution of pixel intensity. The total movement of the selected point was traced by repeating this process frame by frame throughout the whole cardiac cycle, and these data were then recorded as co-ordinates.17 The movement of the tracking point can be visualized on the screen during analysis, and the trace of the point can be visually confirmed. When mistracking compared with the actual wall motion was judged visually, new points were set, and tracking was resumed. Points of interest were tracked several times, confirmed visually, and the average tracking pattern was selected for further analysis. Movement of the points was automatically tracked during the cardiac cycle. Automated tracking was started at end-diastole, defined as the Q wave, on a simultaneously recorded electrocardiogram.

Cardiac catheterization All patients underwent cardiac catheterization within 3 days after echocardiography. Catheterization and angiography (Integris Allura 9 Biplane; Phillips Medical Systems, Amsterdam, The Netherlands) proceeded with the use of 4 – 5 Fr catheters. All patients were examined by biplane anteroposterior and lateral projection angiography. The RV end-diastolic pressure (RVEDP) and volume (RVEDV), systolic pressure (RVSP), and ejection fraction (RVEF), as well as mPAP and mean right atrial pressure (RAP), were measured using ventriculography and calculated by Simpson’s rule for the RV using a quantitative cardiac analysis software package (CAW2000; ELK Corporation, Osaka, Japan). All values of ventricular volumes are expressed as ratios (%) of anticipated normal values calculated from the body surface area of each patient.18,19

Statistical analysis All data are expressed as means + SD or as medians with 5th and 95th percentiles. Statistical significance was determined using the Mann–Whitney U-test or the Kruskal–Wallis test, followed by Dunn’s test, as appropriate. Linear regression analyses were performed for the correlations between the myocardial deformation and haemodynamic parameters, and Pearson’s correlation coefficients were calculated. All statistical calculations were performed using Microsoft Excel 2007 (Microsoft Corporation, Redmond, WA, USA) and Prism version 5.0 (GraphPad Software, San Diego, CA, USA) installed on a desktop computer. A P-value of ,0.05 was considered significant. The interobserver and intraobserver variabilities of RV deformation were assessed using Bland–Altman analysis in a blind manner as the absolute difference between the measurements divided by their mean value from 20 randomly selected normal subjects and patients. After a 5 min interval, data recordings were performed by Observer 1, Observer

892

Y. Hayabuchi et al.

excluded from the study, as were two subjects in the normal group with inadequate data on either the short-axis or the four-chamber view. Two of 27 patients with suboptimal images from poor echocardiographic windows were excluded from all subsequent analyses. Accordingly, the study group included 30 healthy children (mean age, 1.9 + 0.6 years; range, 0.5 –3.0 years) and 25 patients with RV overload (mean age, 1.9 + 0.5 years; range, 0.5 –2.9 years). Table 1 summarizes the clinical, echocardiographic, and haemodynamic data of the subjects.

2, and Observer 1 once again. Data were stored in a digital format and transferred to a personal computer for offline analysis at intervals of 5–10 days.

Results Patients’ characteristics Of the 35 normal subjects, three did not meet the inclusion criteria because of arrhythmia or a small atrial septal defect and were

Table 1

Subjects’ characteristics Normal

RVO

1.9 + 0.6 (0.5– 3.0) 17/13

1.9 + 0.5 (0.5– 2.9) 14/11

............................................................................................................................................................................... Age (years) Sex (male/female) Weight (kg)

10.9 + 2.3 (6.9– 15.0)

Height (cm) BSA (m2)

82.6 + 7.3 (69.0–97.0) 0.51 + 0.07 (0.36–0.63)

9.4 + 2.9 (5.5– 13.3)* 81.5 + 9.4 (65.6– 89.6) 0.48 + 0.09 (0.31–0.58)

102.0 + 12.5 (75–127)

108.8 + 22.5 (82–128)

88.3 + 13.3 (68–110) 46.4 + 6.4 (35– 58)

83.3 + 12.3 (66–101) 40.4 + 7.4 (34– 51)

LVEDD (mm)

24.3 + 3.3 (17.0–29.0)

21.3 + 2.1 (15.0– 26.1)*

LVFS (%) LVEF (%)

36.7 + 5.8 (27.0–48.3) 67.5 + 3.9 (58.0–78.2)

37.9 + 7.8 (29.0– 51.3) 62.9 + 4.9 (54.0– 77.2)

1.01 + 0.21 (0.67–1.30) 0.55 + 0.11 (0.40–0.70)

0.91 + 0.23 (0.61–1.12) 0.57 + 0.14 (0.43–0.72)

e′ (cm/s) a′ (cm/s)

18.7 + 3.2 (13.9–24.0) 8.9 + 1.6 (6.1– 11.1)

16.7 + 3.3 (11.7– 22.0) 8.4 + 1.4 (6.3– 10.7)

s′ (cm/s)

7.0 + 1.0 (5.5– 8.9)

7.1 + 1.1 (5.6– 8.9)

E/e′ RVFAC (4CH)

5.3 + 0.2 (4.7– 6.0) 54.0 + 7.0 (43.5–70.2)

5.1 + 0.2 (4.6– 6.2) 44.0 + 7.0 (33.0– 60.2)*

RVFAC (SAX)

HR (bpm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)

Transmitral flow velocity E-wave velocity (m/s) A-wave velocity (m/s) Tissue Doppler velocity of mitral annulus

52.0 + 7.1 (39.0–66.3)

38.0 + 7.4 (25.0– 52.1)*

Transtricuspid flow velocity E-wave velocity (m/s)

0.65 + 0.14 (0.43–0.90)

0.67 + 0.14 (0.43–0.96)

A-wave velocity (m/s)

0.46 + 0.07 (0.30–0.70)

0.40 + 0.08 (0.31–0.67)

e′ (cm/s) a′ (cm/s)

14.7 + 1.8 (10.9–18.0) 10.5 + 2.9 (6.1– 15.0)

12.7 + 2.2 (8.9– 16.0) 10.5 + 2.8 (6.0– 14.5)

s′ (cm/s)

11.7 + 1.7 (8.5– 14.9)

Tissue Doppler velocity of tricuspid annulus

E/e′ Cardiac catheterization

4.4 + 0.8 (3.4– 5.7)

9.2 + 0.9 (7.5– 11.9)* 5.2 + 0.9 (4.4– 6.2)

RVSP (mmHg)

–

53.6 + 14.1 (38.0–81.0)

RVSP (% systemic) RVEDP (mmHg)

– –

73.5 + 11.2 (58.0–98.0) 7.2 + 0.9 (4.0– 11.0)

RVEDV (% of normal)

–

120.2 + 22.9 (104.4– 164.2)

RVEF (%) RAP (mmHg)

– –

53.4 + 7.7 (40.5– 71.2) 5.2 + 0.5 (3.0– 8.0)

PAP (mmHg)

–

18.3 + 3.9 (10– 34)

BSA, body surface area; HR, heart rate; LVEDD, left ventricular end-diastolic diameter; LVFS, left ventricular fractional shortening; LVEF, left ventricular ejection fraction; RVFAC (4CH), right ventricular fractional area change in the four-chamber view; RVFAC (SAX), right ventricular fractional area change in the LV short-axis view; RVSP, right ventricular systolic pressure; RVEDP, right ventricular end-diastolic pressure; RVEDV, right ventricular end-diastolic volume; RVEF, right ventricular ejection fraction; RAP, mean right atrial pressure; PAP, mean pulmonary arterial pressure. *P , 0.01.

893

Right ventricular myocardial deformation patterns in children

Weight was significantly lower in patients with RVO than in normal controls. There were no significant differences in LVEF between the two groups (Table 1). As expected, RVFAC was significantly lower in patients with RVO than in normal controls. Furthermore, there were significant differences in several echocardiographic parameters between the two groups, as presented in Table 1. Figures 1 and 2 show representative examples of the image and profile of the RV free wall circumferential deformation in a normal subject and a patient with pulmonary valvular stenosis, respectively.

Characteristics of RV free wall circumferential strain in normal children The RV free wall was divided into three segments: anterior, lateral, and inferior segments. Figure 1B shows the time –strain curves of circumferential deformation globally and for the three individual segments. The circumferential strain was significantly lower in the anterior segment than in the lateral and inferior segments (Figure 1C). The time interval between the onset of the QRS complex and the peak strain values was significantly earlier in the anterior segment than in the other two segments.

Characteristics of RV free wall circumferential strain in RVO The representative time– strain curves globally and for the three segments in RVO patients are presented in Figure 2B. The strain was significantly higher for the inferior segment than for the other segments. The anterior segment had an earlier peak strain time than the other two segments. The global circumferential strain was significantly lower in RVO patients than in normal controls (Figure 3). The circumferential strains of the anterior and lateral segments were significantly lower in the RVO patients than in normal subjects, whereas no significant difference in the inferior segment was observed when comparing the two groups. When comparing the time interval between the onset of the QRS complex and the peak of the strain curves, the RVO group had significantly higher values in the three segments and the global circumferential strain curves.

Comparison between global circumferential strain and global longitudinal strain in the control group and the RVO group In the normal subjects, no significant difference was observed between global circumferential strain and global longitudinal strain.

Figure 1: Speckle tracking image of a healthy 1-year-old boy showing representative recordings of RV free wall deformation. The RV free wall is divided into three segments as shown: anterior (Ant), lateral (Lat), and inferior (Inf) segments. Points of interest are placed on the endocardium at end-diastole in the RV free wall, and these are automatically tracked during the cardiac cycles (A). Circumferential strain curves for global and the three individual segments are shown over an entire heart cycle (B). Peak strain (C) and the time interval between the onset of the QRS wave and peak strain (D) are assessed for three segments and global RV free wall. The boxes represent the distribution of peak strain (25th and 75th percentiles; central line, median). The vertical lines represent the range between the 5th and 95th percentiles. *P , 0.01.

894

Y. Hayabuchi et al.

Figure 2: Speckle tracking image of a 1-year-old boy with pulmonary valvular stenosis showing representative recordings of RV free wall deformation. The RV free wall is divided into three segments as shown: anterior (Ant), lateral (Lat), and inferior (Inf) segments. Points of interest are placed on the endocardium at end-diastole in the RV free wall, and these are automatically tracked during the cardiac cycles (A). Circumferential strain curves for global and the three individual segments are shown over an entire heart cycle (B). Peak strain (C) and the time interval between the onset of the QRS wave and peak strain (D) are assessed for three segments and the global RV free wall. The boxes represent the distribution of peak strain (25th and 75th percentiles; central line, median). The vertical lines represent the range between the 5th and 95th percentiles. *P , 0.01.

Both global circumferential strain and longitudinal strain were significantly lower in the RVO group than in the normal group. In the RVO group, the circumferential strain was significantly greater than the longitudinal strain.

Relationship between RV systolic performance and global circumferential and global longitudinal strain in RVO patients The relationships between RVFAC and the global circumferential and longitudinal strain were assessed. RVFAC was evaluated in the four-chamber view [RVFAC (4CH)] and the LV short-axis view [RVFAC (SAX)]. The global circumferential strain had a greater correlation r-value with RVFAC (4CH) than the global longitudinal strain (Figure 4A and B). RVFAC (SAX) also had a greater correlation r-value with the global circumferential strain than with the global longitudinal strain (Figure 4C and D). There was a significantly larger r-value between RVEF and circumferential strain than between RVEF and longitudinal strain (Figure 5A and B). RVSP had a significantly greater negative correlation

r-value with circumferential strain than with longitudinal strain (Figure 5C and D).

Reproducibility Intraobserver and interobserver reproducibilities for analysis of myocardial deformation from 20 randomly selected participants (10 normal controls and 10 patients) were determined by Bland– Altman analysis. The Bland–Altman plots for intraobserver and interobserver variabilities [bias + 2 SD (95% limit of agreement)] in circumferential strain are shown in Figure 6. Intraobserver variability showed that the limits of agreement analysis revealed bias of 20.50 + 3.92% for RV free wall circumferential strain and 21.45 + 39.94 ms for the time interval between the onset of the QRS wave and peak circumferential strain. Interobserver variability showed that the limits of agreement analysis revealed a bias of 21.01 + 5.51% for RV free wall circumferential strain and 21.05 + 50.43 ms for the time interval from the QRS wave to the peak of circumferential strain. Furthermore, we evaluated the intraobserver and interobserver variabilities for analysis of RV free wall longitudinal strain (Figure 7). Intraobserver variability showed that the limits of agreement analysis revealed a bias of 20.15 +

895

Right ventricular myocardial deformation patterns in children

2.62% for longitudinal strain and 24.45 + 23.40 ms for the time interval between the onset of the QRS wave and peak strain. Interobserver variability showed that the limits of agreement analysis revealed a bias of 0.10 + 4.21% for longitudinal strain and 3.60 + 31.84 ms for the time interval.

Discussion

Figure 3: Global circumferential and global longitudinal strain profile of the RV free wall in the normal group and the RVO group. The boxes represent the distribution of the peak strain (25th and 75th percentiles; central line, median). The vertical lines represent the range between the 5th and 95th percentiles.

The major finding of the present study was that the RV free wall circumferential strain was more closely related to RV performance than longitudinal strain in patients with RV overload. Circumferential deformation predominated over longitudinal strain in patients with RVO, whereas strain analysis revealed no significant difference in peak strain during ejection in the two-directional strains in normal subjects. These results indicate that circumferential strain contributes to RV ejection to a greater degree than does longitudinal strain. The development of RV dysfunction in patients with pulmonary hypertension or congenital heart disease has been associated with adverse outcomes.1,4,5 Therefore, clinicians need simple and reproducible tests to assess RV function to improve clinical management of patients. RVEF is generally considered to be a major determinant of systolic RV function. However, determining RVEF is time-consuming and depends on geometric

Figure 4: The relationships between two-directional deformation and RVFAC. The correlations between RVFAC (4CH) and global circumferential strain (A) and global longitudinal strain (B) are shown. Global circumferential strain is more strongly correlated with RVFAC (4CH) than global longitudinal strain. The correlations between RVFAC (SAX) and global circumferential strain (C) and global longitudinal strain (D) are shown. Global circumferential strain more strongly correlates with RVFAC (SAX) than global longitudinal strain.

896

Y. Hayabuchi et al.

Figure 5: The relationships between two-directional deformation and RV performance. The correlations between RVEF and global circumferential strain (A) and global longitudinal strain (B) are shown. Global circumferential strain more strongly correlates with RVEF than global longitudinal strain. The correlations between RVSP and global circumferential strain (C ) and global longitudinal strain (D) are shown. Global circumferential strain more strongly correlates with RVSF than global longitudinal strain.

assumptions, and this strategy has limited application in routine clinical practice.20 Speckle tracking echocardiography is a more recent approach that allows strain imaging to assess cardiac function, which has the advantage of differentiating active motion from passive motion independent of the Doppler angle of incidence, unlike Doppler tissue imaging.21,22 Most studies of speckle tracking strain have focused on the assessment of regional LV function, but recently, investigators have applied speckle tracking strain to the assessment of regional RV function.9 – 13 The longitudinal strain is usually used for evaluation of RV dysfunction, whereas the circumferential strain has not been thoroughly studied. A few magnetic resonance imaging (MRI) studies have been performed on regional RV circumferential deformation and its contribution to RV function.16 This study evaluated RV free wall circumferential strain using two-dimensional speckle tracking echocardiography in normal subjects and patients with RVO to assess its relationship with RV functional parameters. The RV free wall circumferential deformation was first evaluated in normal subjects. The RV free wall was divided into anterior, lateral, and inferior segments.14 In normal subjects, the anterior segment had a lower strain value and shorter time interval between QRS wave onset and the peak strain curve. Differences in regional deformation of the RV free wall may be due to structural reasons, such as fibre direction, or due to differences in forces and geometry. Fibre

direction has been previously suggested as the cause of these unique deformation characteristics.23,24 Despite several anatomical studies, the relationship between fibre orientation and RV mechanics is still not completely clear.25,26 Furthermore, electrical propagation is also presumed to affect the regional difference. In the RVO group, the time sequence of circumferential deformation was similar to that of the normal group; the deformation of the anterior segment was earlier than that of the other segments. The inferior segment had significantly higher circumferential strain than the other segments in the RVO group. Next, we postulated that the circumferential deformation was influenced by RV overload and reflects the RV performance more precisely than the longitudinal strain. In the RVO group, the RV free wall circumferential strain was greater than the longitudinal strain, whereas there were no significant differences in these parameters in the normal group. In RVO patients, circumferential strain predominated over longitudinal strain. These results indicate that longitudinal strain is more sensitive to RVO than circumferential strain. However, RVFAC, RVEF, and RVSP were related to circumferential strain more precisely than to longitudinal strain. Longitudinal strain is the first parameter to decrease as RVO progresses. The results of the present study confirm previously reported findings demonstrating the utility of RV free wall longitudinal strain for the assessment of RV performance.27 – 29 Although the circumferential

897

Right ventricular myocardial deformation patterns in children

Figure 6: The Bland – Altman plots for intraobserver and interobserver reproducibilities. The intraobserver and interobserver differences for global circumferential strain (A and B) and the time interval between the onset of the QRS wave and peak circumferential strain (C and D) are shown. The dotted lines show the mean difference, and the solid lines show 95% limits of agreement.

strain also decreases, the dominant wall systolic direction is circumferential, and RV performance correlates with circumferential strain. The most important finding of our study was that RV systolic function, including RVEF and RVFAC, was more closely related to circumferential strain than to longitudinal movements. Measurement of RV free wall circumferential deformation has the potential to reflect RV haemodynamic performance and the severity of right-sided heart failure in a non-invasive fashion. Although longitudinal movements are easily identified, they are probably less important for RVEF than for circumferential strain. The results of the present study demonstrate a shift in the RV free wall of the RVO group from longitudinal to circumferential shortening when compared with the normal RV group. The predominance of circumferential over longitudinal free wall contraction might represent an adaptive response to pressure overload. The pressure overload gives the RV a more circular short-axis shape.15 This may facilitate circumferential shortening through reduced regional wall stress. Furthermore, in RVO, it is mainly the middle circumferential layer that hypertrophies.30 Consequently, a relative increase in circumferential fibre mass may also contribute to the predominant circumferential RV free wall shortening. In patients with RVO, an altered fibre orientation in the RV free wall is likely to occur as the RV dilates. Changes in fibre direction in the free wall are supported by a study of Pettersen et al.,15 in which MR strain analysis was applied to patients

with RV overload. Their results indicated predominant circumferential over longitudinal free wall shortening at the mid-RV, while the reverse has been observed in healthy control subjects. The results of the present study support the use of RV free wall circumferential strain to estimate RV performance as total RV function. Although right-heart catheterization or an MRI study is still necessary to assess RV performance, RV free wall strain might enable a serial followup of patients to non-invasively assess their response to treatment.

Limitations Two-dimensional speckle tracking echocardiography can determine regional myocardial strain independent of angles and segments. However, some limitations of the current study should be noted. First, because speckle tracking uses a pattern-matching algorithm, errors in early frames can become amplified through subsequent frames.17 Thus, the acquisition of high-quality echocardiographic images is essential, especially the first frame at end-diastole. The accuracy of strain measurement depends on the quality of the echocardiography. If image quality is poor, then peak strain and rotation values can be blunted. The reproducibility of strain measurement is affected by the frame rate of image acquisition. Secondly, although the heart deforms in three-dimensional space, our current speckle tracking echocardiographic system can measure only two-dimensional data. Despite this limitation, the results of

898

Y. Hayabuchi et al.

Figure 7: The Bland – Altman plots for intraobserver and interobserver reproducibilities. The intraobserver and interobserver differences for global longitudinal strain (A and B) and the time interval between the onset of the QRS wave and peak longitudinal strain (C and D) are shown. The dotted lines show the mean difference, and the solid lines show 95% limits of agreement.

two-dimensional echocardiography correlated very closely with those of previous MRI studies and demonstrated the clinical utility of the technique.9 – 13 Thirdly, the geometry and heavily trabeculated myocardium of the RV make it sensitive to errors in the determination of endocardial definition. Fourthly, the echocardiographic recordings were obtained from the subcostal short-axis view. It should be noted that this image can be obtained only in children but not in adults. Fifthly, the echocardiographic examination was not performed simultaneously with the cardiac catheterization. Furthermore, the reproducibility of circumferential deformation was shown to be lower than the longitudinal strain. This point can be a problem for clinical utility. Finally, this study covered a small number of patients in a prospective single-centre study, and the accuracy of 2D RV strain for RV performance remains unclear because of the modest correlations. Therefore, further studies of larger patient populations are necessary to determine the utility of RV free wall circumferential strain for the evaluation of RV performance in patients with RV overload.

Conclusions The characteristics of RV free wall circumferential strain were evaluated in normal and RVO groups. There was a shift in RV free wall

deformation from longitudinal to circumferential shortening in the RVO group. The predominant circumferential contraction may represent an adaptive response to RV overload. The present study demonstrated that RV free wall circumferential strain may be more useful for the evaluation of RV performance than longitudinal strain. Conflict of interest: None declared.

References 1. Moceri P, Dimopoulos K, Liodakis E, Germanakis I, Kempny A, Diller GP et al. Echocardiographic predictors of outcome in Eisenmenger syndrome. Circulation 2012; 126:1461 –8. 2. Alam M, Wardell J, Andersson E, Samad BA, Nordlander R. Right ventricular function in patients with first inferior myocardial infarction: assessment by tricuspid annular motion and tricuspid annular velocity. Am Heart J 2000;139:710 –5. 3. Groner A, Yau J, Lytrivi ID, Ko HH, Nielsen JC, Parness IA et al. The role of right ventricular function in paediatric idiopathic dilated cardiomyopathy. Cardiol Young 2013; 23:409 – 15. 4. Fukuda Y, Tanaka H, Sugiyama D, Ryo K, Onishi T, Fukuya H et al. Utility of right ventricular free wall speckle-tracking strain for evaluation of right ventricular performance in patients with pulmonary hypertension. J Am Soc Echocardiogr 2011;24: 1101 –8. 5. Ling LF, Marwick TH. Echocardiographic assessment of right ventricular function: how to account for tricuspid regurgitation and pulmonary hypertension. JACC Cardiovasc Imaging 2012;5:747 – 53.

Right ventricular myocardial deformation patterns in children

6. Tei C, Dujardin KS, Hodge DO, Bailey KR, McGoon MD, Tajik AJ et al. Doppler echocardiographic index for assessment of global right ventricular function. J Am Soc Echocardiogr 1996;9:838 –47. 7. Kaul S, Tei C, Hopkins JM, Shah PM. Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J 1984;107:526 –31. 8. Meluzin J, Spinarova L, Bakala J, Toman J, Krejcı´ J, Hude P et al. Pulsed Doppler tissue imaging of the velocity of tricuspid annular systolic motion; a new, rapid, and non-invasive method of evaluating right ventricular systolic function. Eur Heart J 2001;22:340–8. 9. Verhaert D, Mullens W, Borowski A, Popovic´ ZB, Curtin RJ, Thomas JD et al. Right ventricular response to intensive medical therapy in advanced decompensated heart failure. Circ Heart Fail 2010;3:340 –6. 10. Meris A, Faletra F, Conca C, Klersy C, Regoli F, Klimusina J et al. Timing and magnitude of regional right ventricular function: a speckle tracking derived strain study of normal subjects and patients with right ventricular dysfunction. J Am Soc Echocardiogr 2010;23:823 –31. 11. D’Andrea A, Caso P, Bossone E, Scarafile R, Riegler L, Di Salvo G et al. Right ventricular myocardial involvement in either physiological or pathological left ventricular hypertrophy: an ultrasound speckle-tracking two dimensional strain analysis. Eur J Echocardiogr 2010;11:492 –500. 12. Pedrinelli R, Canale ML, Giannini C, Talini E, Dell’omo G, Di Bello V. Abnormal right ventricular mechanics in early systemic hypertension: a two dimensional strain imaging study. Eur J Echocardiogr 2010;11:738–42. 13. Pena JL, da Silva MG, Faria SC, Salemi VM, Mady C, Baltabaeva A et al. Quantification of regional left and right ventricular deformation indices in healthy neonates by using strain rate and strain imaging. J Am Soc Echocardiogr 2009;22:369–75. 14. Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23:685 –713. 15. Pettersen E, Helle-Valle T, Edvardsen T, Lindberg H, Smith HJ, Smevik B et al. Contraction pattern of the systemic right ventricle shift from longitudinal to circumferential shortening and absent global ventricular torsion. J Am Coll Cardiol 2007;49:2450–6. 16. Cho EJ, Jiamsripong P, Calleja AM, Alharthi MS, McMahon EM, Khandheria BK et al. Right ventricular free wall circumferential strain reflects graded elevation in acute right ventricular afterload. Am J Physiol Heart Circ Physiol 2009;296:H413 – 20. 17. Toyoda T, Baba H, Akasaka T, Akiyama M, Neishi Y, Tomita J et al. Assessment of regional myocardial strain by a novel automated tracking system from digital image files. J Am Soc Echocardiogr 2004;17:1234 –8.

899 18. Graham TP Jr, Jarmakani JM, Atwood GF, Canent RV Jr. Right ventricular volume determinations in children. Normal values and observations with volume or pressure overload. Circulation 1973;47:144–53. 19. Nakazawa M, Marks RA, Isabel-Jones J, Jarmakani JM. Right and left ventricular volume characteristics in children with pulmonary stenosis and intact ventricular septum. Circulation 1976;53:884 –90. 20. Horton KD, Meece RW, Hill JC. Assessment of the right ventricle by echocardiography: a primer for cardiac sonographers. J Am Soc Echocardiogr 2009;22:776 –92. 21. Suffoletto MS, Dohi K, Cannesson M, Saba S, Gorcsan J III. Novel speckle tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation 2006;113:960 –8. 22. Tanaka H, Nesser HJ, Buck T, Oyenuga O, Ja´nosi RA, Winter S et al. Dyssynchrony by speckle-tracking echocardiography and response to cardiac resynchronization therapy: results of the Speckle Tracking and Resynchronization (STAR) study. Eur Heart J 2010;31:1690 –700. 23. Buckberg GD, RESTORE Group. The ventricular septum: the lion of right ventricular function, and its impact on right ventricular restoration. Eur J Cardiothorac Surg 2006; 29:S272 –8. 24. Anderson RH, Ho SY, Redmann K, Sanchez-Quintana D, Lunkenheimer PP. The anatomical arrangement of the myocardial cells making up the ventricular mass. Eur J Cardiothorac Surg 2005;28:517 – 25. 25. Ho SY, Nihoyannopoulos P. Anatomy, echocardiography, and normal right ventricular dimensions. Heart 2006;92(Suppl 1):i2–13. 26. Hristov N, Liakopoulos OJ, Buckberg GD, Trummer G. Septal structure and function relationships parallel the left ventricular free wall ascending and descending segments of the helical heart. Eur J Cardiothorac Surg 2006;29(Suppl 1):S115 – 25. 27. Huez S, Vachiery JL, Unger P, Brimioulle S, Naeije R. Tissue Doppler imaging evaluation of cardiac adaptation to severe pulmonary hypertension. Am J Cardiol 2007; 100:1473 –8. 28. Utsunomiya H, Nakatani S, Okada T, Kanzaki H, Kyotani S, Nakanishi N et al. A simple method to predict impaired right ventricular performance and disease severity in chronic pulmonary hypertension using strain rate imaging. Int J Cardiol 2011;147: 88 – 94. 29. Lopez-Candales A, Dohi K, Bazaz R, Edelman K. Relation of right ventricular free wall mechanical delay to right ventricular dysfunction as determined by tissue Doppler imaging. Am J Cardiol 2005;96:602 – 6. 30. Sanchez-Quintana D, Anderson RH, Ho SY. Ventricular myoarchitecture in tetralogy of Fallot. Heart 1996;76:280 –6.