The influence of regional right ventricular (RV) dysfunction on the accuracy of Doppler tissue imaging (DTI) assessment of global RV function is unknown. The objective of this study was to determine the effect of regional dysfunction of the RV outflow tract (RVOT) on the correlations between RV DTI indices and cardiac magnetic resonance (CMR) measurements of global RV function in patients with repaired tetralogy of Fallot.
Consecutive patients with repaired tetralogy of Fallot who underwent echocardiography with DTI of the right ventricle and CMR within 4 weeks of each other were retrospectively analyzed. RV DTI measurements were obtained from the lateral wall at the level of the tricuspid valve annulus. CMR measurements included end-diastolic and end-systolic volumes, stroke volume, and ejection fraction (EF) of the entire right ventricle and measured separately for the RV sinus and RVOT segments.
The median age of the 51 patients included was 19 years (range, 9.7–71.6 years), and the median interval between echocardiography and CMR was 0 days. The mean RV free wall peak S′, isovolumic acceleration, and global, sinus, and RVOT EFs were 8.4 ± 2.0 cm/s, 102 ± 37 cm/s 2 , and 46.1 ± 9.8%, 47.9 ± 9.9%, and 33 ± 13.1%, respectively. The correlation between peak S′ and global RV EF was weak ( r = 0.23) in patients with RVOT dysfunction (RVOT EF <30%) but higher ( r = 0.66) in those with RVOT EFs ≥30%. Peak S′ ≥8.4 cm/s (area under the receiver operating characteristic curve, 0.77) and isovolumic acceleration ≥95 cm/s 2 (area under the receiver operating characteristic curve, 0.68) best discriminated between patients with global RV EFs >45% and <45%.
In this group of patients with repaired TOF, RV DTI indices showed reasonable correlation with CMR-derived global RV EF, but this correlation was substantially weaker in those with moderate and severe dysfunction of the RVOT. Peak S′ <8.4 cm/s and isovolumic acceleration <95cm/s 2 by DTI should prompt an evaluation of RV function by CMR.
Accurate determination and serial follow-up of ventricular function are important in the management of patients with repaired tetralogy of Fallot (TOF). Doppler tissue imaging (DTI) is a relatively new nongeometric echocardiographic technique that has been increasingly used for the assessment of global and regional myocardial function. DTI indices quantify longitudinal shortening, which represents the principal component of right ventricular (RV) contraction. RV DTI indices have been shown to have satisfactory correlation with invasive hemodynamic data and with ejection fraction (EF) measured by cardiac magnetic resonance (CMR). RV DTI velocities, along with an index of RV filling expressed as tricuspid E/E′ ratio, have recently been shown to be useful in the clinical follow-up of adult patients with elevated RV pressures due to pulmonary embolism. In addition, DTI indices have been demonstrated to correlate with exercise capacity and endurance and have been used to investigate RV contractile reserve in patients with repaired TOF.
RV outflow tract (RVOT) dyskinesis or aneurysm is seen in many patients with TOF as a consequence of surgical reconstruction during initial repair. Studies using CMR have reported an adverse association between the degree of regional dysfunction of the RVOT and global RV systolic function. However, the effects of regional RVOT dysfunction on the relationships between DTI indices and global RV systolic function have not been studied. It is unknown if RVOT dysfunction in repaired TOF affects the ability of echocardiographic measures of long-axis RV performance to reflect global RV function in a reliable manner. We hypothesized that in patients with repaired TOF, a greater degree of RVOT dysfunction is associated with a weaker correlation between DTI indices of RV systolic function and CMR-derived global RV EF. The goal of this study, therefore, was to investigate the correlations between DTI-derived RV free wall peak S′, myocardial acceleration during isovolumic contraction, and myocardial performance index (MPI) with CMR-derived global and regional RV systolic function.
Candidates for inclusion in this retrospective study were identified by a search of the computer database of the Department of Cardiology at Children’s Hospital Boston. Consecutive subjects with repaired TOF who fulfilled the following criteria were included: (1) underwent transthoracic echocardiography with RV DTI and CMR within 4 weeks of each other, (2) underwent no transcatheter or cardiac surgical procedure between the echocardiographic and CMR studies, (3) had no history of pulmonary valve replacement, and (4) had clearly defined spectral DTI waves of the right ventricle deemed suitable for analysis. Transthoracic echocardiograms and CMR were performed following a standardized institutional protocol. All studies were performed as part of routine clinical care. Demographic information, including age, gender, and dates of studies, was obtained by review of the CMR and echocardiography databases. The study was approved by the Scientific Review Committee of the Department of Cardiology and by the Children’s Hospital Committee on Clinical Investigation.
Echocardiography was performed using iE33 ultrasound systems (Philips Medical Systems, Andover, MA). RV DTI data were obtained in pulsed (spectral) Doppler mode from the lateral wall at the level of the tricuspid annulus from the apical four-chamber view. The Doppler insonation angle was kept as parallel as possible with the long axis of myocardial movement at the level of the tricuspid annulus. Digital images were analyzed offline by two observers (S.K. and C.T.) using electronic measurement tools (HeartSuite VERICIS, Amicas; Boston, MA). Of the 69 subjects screened, 51 had clearly defined spectral DTI waves deemed suitable for analysis and were included in the study.
Peak RV free wall myocardial velocities during systole (S′), early diastole (E′), atrial contraction (A′), isovolumic acceleration (IVA), isovolumic contraction, isovolumic relaxation, systolic and diastolic times, and MPI were measured ( Figures 1 A and B). The average of three samples of each of the measurements was used for data analysis. MPI was calculated as the sum of isovolumic contraction and relaxation times divided by the ejection time. Velocities were reported in centimeters per second, intervals in milliseconds, and IVA in centimeters per second squared.
CMR studies were performed on a 1.5-T Achieva scanner (Philips Medical Systems, Best, The Netherlands). Ventricular dimensions and function were assessed with an electrocardiographically gated steady-state free-precession cine magnetic resonance pulse sequence during brief periods of breath-holding in the following planes: ventricular two-chamber (vertical long-axis), four-chamber (horizontal long-axis), and short-axis planes (perpendicular to the ventricular long-axis plane on the basis of the previous four-chamber images), with 12 to 14 equidistant slices (slice thickness, 6–8 mm; interslice space, 0–2 mm) completely covering both ventricles. The CMR data were analyzed using commercially available software packages (Q-MASS; MEDIS Medical Imaging Systems, Leiden, The Netherlands) by a single observer (S.K.). The end-diastolic and end-systolic volumes, mass at end-diastole, stroke volumes, and EFs were measured separately for the left ventricle, global right ventricle, RV sinus, and RVOT as described in an earlier report from our center. The septal and parietal bands were used as markers for the boundary between the RV sinus and RVOT and were included with the RVOT ( Figure 2 ). Ventricular end-diastolic volumes and mass were adjusted to body surface area calculated using the Haycock formula.
Continuous variables are expressed as mean ± SD or as medians and ranges, as appropriate. Pearson’s correlation coefficients and linear regression analysis were used to explore the relationships between DTI and CMR variables. Stratified correlations were performed according to RVOT EF <30% versus ≥30%, and Fisher’s Z transformation was used to compare the correlations. The threshold value of RVOT EF 30% was chosen on the basis of two considerations. First, previous research has shown that lower RVOT EF is an independent predictor of subnormal peak oxygen consumption, and second, the median value RVOT EF in this cohort was 30%. Unpaired t tests were used for comparison of variables between the lower and higher RVOT EF groups. Receiver operating characteristic curves were used to identify cutoff values of peak S′ and IVA that provided the best combination of sensitivity and specificity for global RV EF >45%. To assess interobserver variability, DTI measurements performed by observer 1 (C.T.) were compared with those performed by observer 2 (S.K.) in a sample of 11 patients. The mean difference between the two observers was calculated along with a 95% confidence interval. P values < .05 were considered statistically significant. Commercially available statistical software packages were used for data analysis (SAS version 9.2, SAS Institute Inc, Cary, NC; Minitab version 14, Minitab Inc, State College, PA).
Echocardiograms and CMR examinations in 51 patients with repaired TOF who met the inclusion criteria were analyzed. Patient demographics, CMR, and DTI data are summarized in Table 1 . The echocardiographic and CMR examinations were performed on the same day in 36 patients (71%). The interval between echocardiography and CMR was <1 week in seven patients (13%), between 1 and 2 weeks in three patients (6%), and between 2 and 4 weeks in five patients (10%).
|Age (years)||19.1 (9.7 to 71.6)|
|Body surface area (m 2 )||1.72 (0.94 to 2.28)|
|Height (cm)||162 (127 to 190.5)|
|Weight (kg)||64.2 (24.7 to 102.0)|
|Interval between echocardiography and CMR (days)||0 (–31 to 30)|
|End-diastolic volume index (mL/m 2 )|
|Global right ventricle||163.8 ± 59.2 (91.3 to 309.2)|
|RV sinus||119.2 ± 47.5 (63.9 to 221.9)|
|RVOT||45.2 ± 19.8 (15.0 to 97.7)|
|End-systolic volume index (mL/m 2 )|
|Global right ventricle||91.1 ± 47.9 (38.3 to 261.0)|
|RV sinus||63.7 ± 35.0 (24.9 to 176.7)|
|RVOT||31.4 ± 18.5 (9.2 to 90.8)|
|RV EF (%)|
|Global right ventricle||46.1 ± 9.8 (15.2 to 60.7)|
|RV sinus||47.9 ± 9.9 (18.4 to 61.5)|
|RVOT||33 ± 13.1 (3.2 to 54.4)|
|RV basal free wall velocities (cm/s)|
|Peak systolic annular (S′)||8.4 ± 2.0 (5.4 to 12.5)|
|Peak early diastolic (E′)||9.7 ± 3.1 (5.1 to 17.3)|
|Peak velocity with atrial contraction (A′)||6.5 ± 2.0 (3.2 to 11.2)|
|Isovolumic contraction time (ms)||85.8 ± 17.2 (55 to 121)|
|Isovolumic relaxation time (ms)||71.4 ± 19.6 (24 to 125)|
|Ejection time (ms)||298 ± 38 (201 to 375)|
|Systolic time (ms)||392 ± 49 (290 to 507)|
|Diastolic time (ms)||488 ± 106 (308 to 750)|
|IVA (cm/s 2 )||102 ± 37 (46 to 240)|
|RV MPI||0.52 ± 0.08 (0.28 to 0.69)|
Table 2 summarizes the results of the regression analyses of RV free wall DTI parameters and both global and regional RV EF in all patients. Peak S′ and IVA were correlated with global RV EF and with RV sinus EF, but only peak S′ was correlated with RVOT EF. In contrast, peak E′ and A′ velocities were not correlated with CMR-measured global or regional RV EF. RV MPI was abnormal (0.52 ± 0.08) but did not demonstrate statistically significant associations with global, sinus, or RVOT EF. Repeat analysis restricted to the 36 patients who underwent both tests on the same day showed similar results to those found in the entire cohort.
|Correlations||Regression slope||Pearson’s correlation||P|
|Peak S′ vs global RV EF||2.1||0.43||.002|
|Peak S′ vs RV sinus EF||2.0||0.41||.003|
|Peak S′ vs RVOT EF||2.1||0.32||.025|
|Peak E′ vs global RV EF||0.45||0.14||.32|
|Peak E′ vs RV sinus EF||0.52||0.16||.26|
|Peak E′ vs RVOT EF||0.24||0.06||.69|
|Log IVA vs global RV EF||12.9||0.42||.005|
|Log IVA vs RV sinus EF||13.0||0.42||.006|
|Log IVA vs RVOT EF||10.5||0.26||.09|
|MPI vs global RV EF||5.8||0.05||.73|
|MPI vs RV sinus EF||6.2||0.05||.72|
|MPI vs RVOT EF||5.0||0.03||.83|
The median RVOT EF in all patients by CMR was 29.8% (range, 3.2%–54.4%). There were 30 patients with RVOT EFs ≥30% and 35 patients with global RV EFs ≥45%. Table 3 compares demographic, CMR, and DTI data between patients with RVOT EFs <30% and those with RVOT EFs ≥30%. Analyses of correlations stratifying by lower versus higher RVOT EF showed that the correlation between RV free wall peak S′ and global RV EF was weak ( r = 0.23) in patients with moderate or greater RVOT dysfunction (RVOT EF <30%; n = 21) but was substantially higher ( r = 0.66) in those with RVOT EFs ≥30% ( n = 30) ( P = .06 for comparison between correlation coefficients; Figures 3 A and 3 B). Peak S′ ≥8.4 cm/s and IVA ≥95 cm/s 2 (areas under the receiver operating characteristic curve, 0.77 and 0.68, respectively) best discriminated between patients with global RV EFs >45% and <45%. The sensitivity, specificity, and positive and negative predictive values of different peak S′ and IVA cutoff values to detect global RV EF ≤45% are summarized in Table 4 .
|Variable||RVOT EF <30%||RVOT EF ≥30%||P|
|Age (years)||22.3 (10.8–43.2)||18.0 (9.7–71.6)||.98|
|Body surface area (m 2 )||1.65 (0.94–2.2)||1.80 (0.94–2.28)||.51|
|Height (cm)||161 (127–155)||165 (132–190.5)||.11|
|Weight (kg)||63.0 (25.4–101.0)||67.6 (24.7–102)||.78|
|Interval between echocardiography and CMR (days)||0 (0–31)||0 (0–30)||.83|
|End-diastolic volume index (mL/m 2 )|
|Global right ventricle||175.0 ± 71.4||155.7 ± 48.2||.29|
|RV sinus||124.7 ± 53||115 ± 43.3||.49|
|RVOT||52 ± 22.5||40.1 ± 16.0||.05 ∗|
|End-systolic volume index (mL/m 2 )|
|Global right ventricle||110.3 ± 63.4||77.1 ± 25.6||.03 ∗|
|RV sinus||74.9 ± 46.1||55.3 ± 20.6||.08|
|RVOT||42.4 ± 22.0||23.2 ± 9.2||.001 ∗|
|RV EF (%)|
|Global right ventricle||39.7 ± 10.7||50.6 ± 5.8||.0001 ∗|
|RV sinus||42.9 ± 12.2||51.5 ± 5.6||.006 ∗|
|RVOT||20.3 ± 7.7||42.2 ± 7.1||.0001 ∗|
|RV basal free wall velocities (cm/s)|
|Peak systolic annular (S′)||7.9 ± 2.2||8.7 ± 1.8||.18|
|Peak early diastolic (E′)||9.1 ± 3.0||10.0 ± 3.1||.30|
|Peak velocity with atrial contraction (A′)||6.3 ± 2.0||6.6 ± 2.0||.58|
|Isovolumic contraction time (ms)||82.6 ± 13||88 ± 19||.26|
|Isovolumic relaxation time (ms)||71 ± 19||71 ± 20||.97|
|Ejection time (ms)||292 ± 39||302 ± 37||.35|
|Systolic time (ms)||380 ± 43||400 ± 53||.15|
|Diastolic time (ms)||483 ± 112||478 ± 131||.88|
|IVA (cm/s 2 )||98 ± 35||104 ± 39||.56|
|RV MPI||0.52 ± 0.1||0.53 ± 0.1||.65|