Pulmonary Regurgitation (PR) is common after repair of Tetralogy of Fallot (TOF) with transannular patching or after valvotomy in patients with pulmonary stenosis. More than moderate chronic PR leads to progressive right ventricular dilatation, deterioration of myocardial function, exercise intolerance, ventricular arrhythmias and sudden death [1].

Cardiac Magnetic Resonance (CMR) imaging is the best method to estimate right ventricular volumes and quantify the degree of pulmonary and tricuspid regurgitation [2], permitting extensive evaluation of right ventricular abnormalities such as aneurysms of the right ventricular outflow tract.

Although CMR allows an accurate assessment of Right Ventricle (RV) size and function [2], echocardiography is used, due to its widespread availability, as the first line imaging modality. However, the quantitative assessment of RV function, by echocardiography, is often complicated because of the difficulty of endocardial tracing of the RV area due to its complex anatomy. Different approaches such as Tricuspid Annular Plane Systolic Excursion (TAPSE) or Tissue Doppler Imaging (TDI) which gets myocardial velocities have obtained good correlations estimating RV ejection fraction [3-5].

In this retrospective study we compare the assessment of RV function by CMR and echocardiography in operated congenital heart disease patients with moderate or more than moderate pulmonary regurgitation. Exercise stress testing, 24 hour holter and aminoterminal pro-brain natriuretic peptide (NT pro BNP) levels were also recorded.

46 patients with surgically corrected pulmonary valve stenosis and/or reconstruction of RV outflow tract (RVOT) were followed up at our centre’s adult congenital heart disease unit between May 2004 and October 2011. 17 patients (37%) were selected according to the presence of moderate or more than moderate PR. Information collected included patients’ demographics, operative details, postoperative functional status, electrocardiogram, stress testing, transthoracic echocardiogram and CMR imaging. Patients younger than 18 years old, with pulmonary valve replacement or less than moderate PR were excluded. All patients gave informed consent for before participating in the study. The protocol of the study was approved by the Hospital’s Ethics Committee.

A complete two dimensional and pulsed Doppler echocardiography examination was performed with an iE33 (Philips Medical Systems, USA) echocardiographic system, equipped with doppler tissue imaging technology and a 5 MHz transducer. RV Ejection Fraction (RVEF) was calculated by endocardial tracing of the RV area in systole and diastole (Simpson ejection fraction) in the short axis (2C: two chambers) and apical chamber (4C: four chambers) view. Global Simpson was expressed by [(Simpson 4C+Simpson 2C)/2]. TAPSE was measured at the level of systolic excursion of the lateral tricuspid valve annulus towards the apex in the four chambers view and myocardial wall motion velocities with TDI [6,7] were obtained from a four chamber view at the junction of the right ventricular free wall and the anterior leaflet of the tricuspid valve using doppler tissue and assessing systole (Sa), early diastole (Ea) and late diastole (Aa). Simpson ejection fraction was expressed in percentage, TAPSE in mm (millimetres) and its equivalent ejection fraction in percentage (5 mm~20% RVEF, 10 mm~30% RVEF, 15 mm~40 RVEF and 20 mm~50% RVEF) [3] and myocardial TDI wall motion velocities in cm/s.

The severity of tricuspid and pulmonary regurgitation was graded qualitatively according to the guidelines of the American Society of Echocardiography [8]. Grade 0 was defined as no regurgitation, grade 1 as mild, grade 2 as moderate, grade 3 as moderate to severe and grade 4 as severe. Left ventricle ejection fraction was calculated with the Teichholz’s formula. RV diameter was measured in the medial-lateral apical four chamber view.

CMR imaging was performed with a Magneton Symphony (Siemens, Erlangen, Germany). The endocardial surfaces were traced to obtain RV end diastolic and end systolic volumes from images in the axial plane. The RV end diastolic image was identified by selecting the largest ventricular surface at the midventricular level, whereas the end systolic image was identified by selecting the smallest ventricular surface. The RV Ejection Fraction (EF) was calculated using the equation: [EF=(end diastolic volume–end systolic volume)/end diastolic volume]. PR volumes were quantified in a double oblique plane perpendicular to the main pulmonary artery just below the bifurcation [5,9] and RV diameter was measured in the four chamber view. CMR assessment was made within 6 months of the echocardiography and patients were clinically stable, in sinus rhythm and had native and unrepaired tricuspid valves.

QRS duration and morphology, corrected QT and PR intervals were measured from the surface electrocardiogram. Blood pressure, heart rate and metabolic equivalents (METs) achieved during maximal treadmill exercise (Bruce protocol) were obtained. Also plasma levels of NT pro BNP were measured by immunoassay with the Siemens Stratus CS Acute Care Diagnostic System (Siemens Healthcare Diagnostics, Inc, Newark, DE).

The values are expressed as mean ± standard deviation or median and quartile values (5; 95) and qualitative variables as percentages of the total. The Pearson’s correlation was used to find a correlation between two continuous variables. To illustrate the results described, we constructed a dispersion diagram for TDI systole vs. CMR RVEF and TAPSE vs. CMR RVEF. For all parameters, a value of p<0.05 was considered statistically significant. The data analyses were performed with SPSS 20.0 (Chicago, IL, USA).

17 patients with moderate or more than moderate PR (10 patients (58.8%) with severe PR, 2 patient (11.8%) with moderate to severe PR and 5 patients (29.4%) with moderate PR) and surgery correction of TOF or pulmonary stenosis in the infancy were studied. Medium age was 30.8 ± 11.9 years old with 12 (70.6%) male patients. The age at the correction surgery was 8.1 ± 11.0 years old and the time from the surgery to the study was 22.7 ± 9.8 years.

14 patients (82.3%) had TOF with transannular patch and 3 (17.6%) patients a surgically repaired valvular pulmonary stenosis (1 patient with associated transannular patch and 2 patients with infundibulectomy).

RVEF calculated by Simpson was 51.0 ± 11.9%, by TAPSE 18.7 ± 3.7 mm~47.4 ± 7.3% and by CMR imaging 40.7 ± 8.8%. The Pearson’s correlation was r=0,116 for global Simpson vs. CMR, r=0.462 for TAPSE vs. CMR and r=0.453 for TDI vs. CMR. Dispersion diagram for TDI systole vs. CMR RVEF and TAPSE vs. CMR RVEF is shown in Figure 1. TDI systolic annular velocity (Sa) was 11.1 ± 2.5 cm/s, predicting a Sa value≤11.5 cm/s right ventricular dysfunction (RVEF<45%) calculated by CMR with a sensitivity of 90%, a specificity of 50% and a predictive accuracy of 77%.

Figure 1: Dispersion diagram for TDI systole vs. CMR RVEF and TAPSE vs. CMR RVEF showing an R squared value of 0.205 and 0.213 respectively. TDI: Tissue Doppler Imaging; CMR: Cardiac Magnetic Resonance; TAPSE: Tricuspid Annular Plane Systolic Excursion; RVEF: Right Ventricular Ejection Fraction.

Related to RV volumes, a medium end diastolic volume of 154.8 ± 70.3 ml/m² and a medium end systolic volume of 97.9 ± 54.6 ml/ m² correlated with a RVEF calculated by CMR of 40.7 ± 8.8 %. Two patients with TOF showed aneurysm of the right ventricular outflow tract. Pearson’s correlation between PR calculated by echocardiography and CMR (3.2 ± 1.1 grades and 33.2 ± 19.4% respectively) was of 0.64.

Treadmill exercise Bruce protocol showed a maximum heart rate of 164.3 ± 27.6 beats per minute and 10.1 ± 3.2 METS. Maximum systolic blood pressure was 161.6 ± 35.3 mm Hg with and maximum diastolic blood pressure was 80.7 ± 12.51 mm Hg.

All electrocardiograms evidenced sinus rhythm and right branch block image with a corrected QT of 428.2 ± 44.1 and a PR interval of 144.0 ± 24.1 ms. No significant arrhythmias, except ventricular extra systole and some duplexes were seen during stress testing or 24 hour ambulatory electrocardiogram. One patient was under medical treatment with beta blockers and oral anticoagulation due to previous supraventricular arrhythmias. Patients were in NYHA class I-II/IV (grade 1.3 ± 0.5) and had NT pro-BNP levels of 78.1 (0.0; 339.6) pg/ml (normal range: 0 - 125 pg/ml).

The distribution of demographic, echocardiographic and CMR data are shown in Table 1.

Table 1: Demographic, echocardiographic and CMR values in patients with severe pulmonary regurgitation after repair of tetralogy of Fallot and pulmonary stenosis.

PR after repair of TOF or pulmonary stenosis with transannular patch or surgical valvotomy is common [10]. Patients with significant PR are at risk for progressive right ventricular dilatation and dysfunction. RV function is strongly associated with clinical outcomes in many conditions. Recent data suggest that the recovery of the RV function after pulmonary valve replacement is unlikely if RV end diastolic volume corrected for body surface area is >170 ml/m² or if RVEF<40% [11]. Therefore, optimal timing of pulmonary valve implantation is, therefore, crucial for preserving RV function and avoiding the need for early subsequent pulmonary valve implantation [12].

CMR is a non invasive imaging modality that has become the gold standard for periodic evaluation and follows up of patients with PR [13,14]. Although CMR allows an accurate assessment of RV size and function it is expensive and needs expertise to acquire and interpret the data. Also patients with pacemakers and defibrillators cannot undergo CMR [15]. On the contrary, though echocardiography is much more accessible and cheap, various limitations exist which limit the accuracy of transthoracic echocardiographic asessment of the RV, including inadequate visualization of the RV free wall which can limit visual assessment, RV fractional area change and ejection fraction measurements.

In relation to TAPSE, although it measures longitudinal function, it has shown good correlation with techniques estimating RV global systolic function, such as radionuclide or angiography [3]. It is known that TAPSE provides useful information on RV function and shape but appears of limited use in conditions that exhibit abnormal regional contraction such as patients with TOF [15]. That could explain why TAPSE tended to overestimate RV function when compared with CMR in our patients.

More recently, TDI doppler–derived tricuspid lateral annular systolic velocity (Sa), a reliable, reproducible and easy to measure echocardiographic method has shown to correlate well with other measures of global RV systolic function [16,17]. DTI velocities quantify longitudinal shortening that represents the main mechanism of RV function and have been found to be relatively load independent. In our series, Sa also allowed a good estimation of global CMR RVEF, data similar to those obtained by Meluzín et al. [4], in 44 patients with heart failure and RV systolic dysfunction in whom a systolic annular velocity<11.5 cm/s predicted right ventricular dysfunction (ejection fraction<45%) with a sensitivity of 90% and a specificity of 85%. Similarly, Lytrivi et al. [18], in a study with 35 congenital heart disease patients, most of which had undergone cardiac operation and 18 of 35 patients exhibited altered pressure and/or volume loading conditions, showed that peak systolic velocity correlated well with RV ejection fraction after adjusting for age, RV dilation, and pressure overload (r=0.65) with an excellent interobserver and intraobserver reliability for peak systolic velocity.

Stress testing also allows a good documentation of the patient’s functional class being useful in the follow up, preceding the onset of symptoms and assisting in deciding the best time of pulmonary valve replacement. Similarly, NT pro-BNP levels tend to increase along with the severity of heart failure expressed as New York Heart Association (NYHA) functional class regardless of the congenital heart disease anomaly [19]. The absence of symptoms in our congenital heart disease patients could explain why NT pro-BNP levels were in normal range despite the existence of some degree of right ventricular dysfunction.

Arrhythmia and sudden death are also important late sequelae for patients after repair of tetralogy of Fallot. The electrophysiological and haemodynamic substrate of sudden death resembled that of sustained ventricular tachycardia, with pulmonary regurgitation being the predominant haemodynamic lesion [2]. Preservation or restoration of pulmonary valve function may thus reduce the risk of sudden death contributing electrocardiographic monitoring to identify patients at risk [20].

Though our low sample size can limit the result of our study, a TDI systolic motion velocity<11.5 cm/s at the tricuspid annulus seems to have a high sensitivity to detect a CMR RVEF<45%. However larger studies with more patients will be needed to determine the exact role of echocardiography in right ventricular function in patients with congenital heart disease with associated pulmonary regurgitation.