Patient Population

All patient data were collected in a prospective manner at our institutions.

Group 1 consisted of 55 patients with PAH-CHD (age range, 0.6–34 years; mean age, 13.2 years; 34 male, 21 female; body surface area [BSA] range, 0.35–2.08 m ² ). Thirty-one patients with PAH-CHD were aged ≤ 18 years, and 24 were aged > 18 years. Patients with PAH were all associated with CHD according to the 2009 updated clinical classification of pulmonary hypertension (group 1.4.4). The cardiac diagnoses and the baseline characteristics of the patients are shown in Table 1 . Patients with more than moderate atrioventricular valve regurgitation or conduit regurgitation were excluded from this study. The respective CHDs were repaired in all patients at a mean age of 5.4 months (range, 0.5–15.8 months). None of our patients had Eisenmenger syndrome, the most advanced form of PAH-CHD. At the time of enrollment, all patients were in clinically stable condition, with no new medications started within the previous 3 months. Echocardiographic studies were performed in a cross-sectional manner over a wide range of intervals between operation and the time of echocardiography. All these patients had measurable mild to moderate tricuspid regurgitation (TR) jets; patients with severe TR were excluded from the study. RV systolic pressure was assessed by TR velocity (calculated from the modified Bernoulli equation). TR velocity > 2.8 m/sec (corresponding to a right atrioventricular pressure gradient > 31 mm Hg) is considered a reasonable cutoff to define elevated pulmonary pressures in the absence of pulmonary stenosis. TR was measured in all of our patients at least as half of systemic systolic pressure, apart from three patients in whom mild TR made it impossible to obtain reliable data. Twenty-seven of our patients with PAH-CHD (49% of all patients) underwent MRI. The time interval between the MRI and echocardiography had a mean of 37 days (range 0 to 117 days). All patients underwent right-heart catheterization, which remains the gold standard for diagnosing PAH. Left-sided filling pressure and cardiac output data were obtained to rule out entities that can elevate pulmonary artery pressure other than pulmonary vascular disorders, such as pulmonary venous hypertension. PAH was defined as a mean pulmonary artery pressure ≥ 25 mm Hg at rest, a pulmonary capillary wedge pressure < 15 mm Hg, and pulmonary vascular resistance ≥ 3 mm (Wood units × BSA [m ² ]). By directly measuring pressures and indirectly measuring flow, we determined markers such as cardiac output (using Fick’s principle), mixed venous oxygen saturation, and mean pulmonary artery pressure. Right-heart catheterization data were as follows: mean pulmonary artery pressure ranged from 30 to 65 mm Hg (mean, 39.2 mm Hg) at rest, pulmonary vascular resistance ranged from 3.8 to 20.2 Wood units × BSA (m ² ) (mean, 5.6 Wood units × BSA [m2]), and mean mixed venous oxygen saturation was 65 ± 9% in all patients with PAH-CHD ( Table 1 ). The time interval between right-heart catheterization and echocardiography ranged from 0 to 52 days. All patients had RV systolic pressure ≥62% of systemic systolic pressure. Twelve patients with initial diagnostic catheterization at the time of this study underwent first-time vasodilator testing.

Group 2 consisted of 183 patients with repaired TOF (79 male, 104 female) who were undergoing routine clinical follow-up. The RV outflow tract was repaired by means of a transannular patch made of autologous untreated pericardium in all patients at a mean age of 8.7 months (range, 3.3–18.9 months). The patients were evaluated from the newborn age to the age of 34 years (mean age, 15.2 years). Echocardiographic studies were performed in a cross-sectional manner over a wide range of intervals between operation and the time of echocardiography. One hundred five patients with TOF were aged ≤ 18 years, while 78 patients were aged > 18 years. Patients in group 2 had a mild mean residual RV outflow tract gradient of 16 ± 8 mm Hg, as determined by routine echocardiography. The baseline characteristics of the patients and their medications are shown in Table 2 . Patients with repaired TOF with higher degrees of RV outflow tract obstruction or valvular or pulmonary artery branch stenosis were excluded from the study. In 33 patients (18% of all patients with TOF), RV outflow tract aneurysms were present. Thirteen patients with restrictive physiology of the right ventricle, defined as the presence of laminar antegrade diastolic main pulmonary artery flow throughout the respiratory cycle by Doppler echocardiography, were excluded from the study. We used previous MRI data of 41 pediatric patients with TOF (i.e., 37% of 111 study patients had MRI scans available for the current study), while all echocardiographic data were acquired solely for this study. Thirty-nine patients (21% of all patients with TOF) underwent aortopulmonary shunt procedures before surgical repair. In 111 patients (61% of all patients with TOF), S′ could be compared with RVEF and RVEDVi measured by MRI. The time interval between MRI and echocardiography ranged from 0 to 43 days. Recordings and measurements of S′, RVEDVi, and RVEF were performed without and with access to MRI data.

Group 3 consisted of 849 healthy patients (428 male, 421 female) with normal results on echocardiography and measured S′ values inside the previously published age-related normal Z -score range. Patients were selected from individuals referred to our cardiology service for evaluation of heart murmurs or family history of heart disease. The study group encompassed neonates to adolescents (age range, 1 day to 18 years; BSA range, 0.18–2.30 m ² ) and was compared with our patients with PAH-CHD and TOF. For the purposes of the study, only echocardiograms with official readings of completely normal results or completely normal results except for patent foramen ovale with diameter ≤ 2 mm were accepted for analysis. All patients with CHD, acquired heart disease, or chromosomal syndromes were excluded from analysis. Patients were examined in a resting state without prior sedation. Infants were allowed to be bottle fed during the examination.

Echocardiographic Techniques

Echocardiography was performed using the Sonos iE33 echocardiographic system (Philips Medical Systems, Andover, MA) with transducers of 5-1, 8-3, and 12-4 MHz depending on patient size. Images were recorded digitally and later analyzed by one of the investigators (M.K.) using offline software (Xcelera Echo; Philips Medical Systems, Eindhoven, The Netherlands).

Pulsed-wave DTI was performed using transducer frequencies of 2.5 to 3.5 MHz with spectral Doppler filters adjusted until a Nyquist limit of 15 to 20 cm/sec was reached. The minimal optimal gain setting was used. Doppler measurements were acquired with subjects in the lateral decubitus position during shallow respiration. Guided by the four-chamber view, a 5-mm sample volume was placed at the lateral corner of tricuspid annulus, exactly at the attachment of the anterior leaflet of the tricuspid valve. Care was taken to obtain an ultrasound beam parallel to the direction of tricuspid annular motion. Peak annular velocities during systole were recorded and analyzed offline. The resulting velocities were recorded for three to five cardiac cycles and were averaged. To assess day-to-day variability of S′, pulsed-wave DTI was repeated 24 hours later under the same conditions in 22 subjects. The investigation of S′ was performed in a quiet state. RV conditions were assessed. Patients were classified as having RV pressure overload on the basis of elevated TR velocities, RV outflow tract gradients, and/or systolic septal flattening.

MRI

RV volumes were quantified by means of breath-hold cine steady-state free precession sequences in 111 patients with repaired TOF and 27 patients with PAH-CHD using a 1.5-T machine (Symphony; Siemens Medical Systems, Forchheim, Germany). Patients with implantable cardioverter-defibrillators were excluded from this study because MRI is contradicted in patients with such implants. The right ventricle was encompassed by means of continuous short-axis views from base to apex. RV volumes were calculated after delineation of the endocardial surfaces of the end-diastolic and end-systolic images. Multiplication of the delineated cavity area by the slice thickness yielded the slice volume, and summation of all slice volumes yielded ventricular volume at end-diastole and end-systole. Values of RVEDVi > 150 mL/m ² were defined on the basis of MRI data reported for adults and adapted for the pediatric age group, corresponding to 150% of the upper limit of normal for RVEDVi in children, which is 100 mL/m ² . Pulmonary regurgitation was quantified by velocity-encoded imaging. The regurgitation fraction was determined by calculating the percentage of reverse volume from forward volume. All volume and flow measurements were indexed for BSA and expressed as milliliters per beat per square meter. Measurements were made by E.S. blinded to the echocardiographic data.

Statistical Analysis

All S′ data were measured by three well-trained observers (M.K, B.H., and B.N.) from three to five consecutive beats and averaged. For data analysis, SPSS version 19 (SPSS, Inc., Chicago, IL) and SAS version 9.2 (SAS Institute Inc., Cary, NC) were used. Data are presented as mean ± 2 SDs. In a first-step analysis in the healthy controls, the correlation between age and S′ was analyzed using Pearson’s correlation coefficient. A regression was used to estimate in healthy patients S′ from age. In a second step in patients with PAH-CHD and TOF, the deviation from normal S′ values was calculated. A regression was used to estimate in patients with PAH-CHD and repaired TOF the deviation of normal S′ from age. The cutoff value for this age, at which the S′ value in patients with PAH-CHD and repaired TOF is lower than the reference value, was taken from this linear regression analysis. This was done by using the lower bound of the ±2-SD interval of the mean value of normal S′. The correlation structure between age and MRI-determined RVEF and RVEDVi in the TOF and PAH-CHD groups was analyzed using Pearson’s correlation coefficient.

Interobserver variability was also assessed. Data were measured by two observers (M.K. and B.N.) who were blinded to each other’s results. Intraobserver variability was considered in 24 participants by repeating the measurements on two occasions. Interobserver and intraobserver variability was found for S′, with intraclass correlation coefficients of 0.97 (confidence interval [CI], 0.94–0.99; P < .001) and 0.98 (CI, 0.96–0.99; P < .0001). Intraobserver and interobserver variability in our study was similar to that reported in the literature for S′.

Ethics

This study was in compliance with all institutional guidelines related to patient confidentiality and research ethics, including institutional review board approval (EK-Nr. 23-059 ex 11/12).