Scandinavian Cardiovascular Journal, 2014; 48: 85–90
ORIGINAL ARTICLE
The echocardiographic paradox index in patients with a repaired tetralogy of Fallot
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
SANG-YUN LEE1, JINYOUNG SONG2, SUNG-HO KIM1, SO-ICK JANG1 & YANG-MIN KIM3 1Department
of Pediatrics, Bucheon-si, Republic of Korea, 2Department of Pediatrics, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea, and 3Department of Radiology, Sejong Cardiovascular Institute, Bucheon-si, Republic of Korea
Abstract Objectives. Right ventricular (RV) volume is very important for pulmonary valve replacement after the total correction of tetralogy of Fallot (TOF), and we attempted to identify a convenient echocardiographic index that is well correlated with the volumetric data obtained through magnetic resonance imaging (MRI). Design. All patients who underwent cardiac MRI and echocardiography at Sejong General Hospital for evaluating pulmonary regurgitation after TOF total correction were included. The paradox index is the amount of paradoxical motion of the interventricular septum on the short-axis echocardiographic view. The paradox index was compared to several cardiac MRI indices. Results. Fifty-four patients were included. The paradox index for all patients was 1.22 ⫾ 0.12 (1.06–1.67), and the index of the operation group was significantly higher than that of the non-operation group (1.26 ⫾ 1.12 vs1.16 ⫾ 1.12, P ⫽ 0.009). The paradox index was well correlated with the RV systolic and diastolic volumes, as measured by cardiac MRI (P ⫽ 0.002 and 0.003). Using a simple linear regression analysis, a paradox index of 1.24 corresponded to a RV diastolic volume of 160 ml/m2. Conclusions. The paradox index could help to indicate the time for an MRI analysis of the RV volume in patients after TOF total correction. Key words: echocardiography, paradox index, tetralogy of Fallot
Introduction Chronic pulmonary regurgitation (PR) after the correction of a right ventricular (RV) outflow obstructive lesion is known to impact RV function. Most patients whose ailments were corrected using the transannular patch technique or graft insertion for severe tetralogy of Fallot (TOF) or pulmonary atresia usually experience chronic PR and require pulmonary valve replacement (PVR) (1). However, the best timing for PVR remains under debate. RV dilatation is known to be a key indicator for the need of PVR. In terms of evaluating RV dilatation, threedimensional echocardiography and cardiac magnetic resonance imaging (MRI) measure the RV volume directly, whereas two-dimensional echocardiography provides some indicators of RV dilatation. However, three-dimensional echocardiography might not be accurate, particularly when the right ventricle is
dilated (2). Due to recent advances in cardiac MRI, there is a tendency to determine the optimal time of PVR based on the cardiac MRI data. However, cardiac MRI is costly and difficult to perform; thus, regular follow-up is difficult. Therefore, it would be helpful if an echocardiographic parameter that is capable of suggesting significant RV dilatation prior to performing MRI could be identified. The paradox index is the amount of paradoxical motion of the inter-ventricular septum on the short-axis echocardiographic view (3). It is assumed that the paradoxical motion of the inter-ventricular septum depends on the amount of PR and the RV volume. In this study, the relationship between the volumetric data from cardiac MRI and the paradox index was analyzed. The aim of this study was to evaluate the usefulness of the paradox index as a convenient parameter for determining the optimal
Correspondence: Dr Jinyoung Song, MD, Department of Pediatrics, Samsung Medical center, Sungkyunkwan University School of Medicine, 50 Irwon-Dong, Gangnam-gu, 135-710, Seoul, South Korea. Tel: ⫹ 82-2-3410-3539. Fax: ⫹ 82-2-3410-0043. E-mail: [email protected] (Received 23 October 2013 ; revised 10 January 2014 ; accepted 12 January 2014 ) ISSN 1401-7431 print/ISSN 1651-2006 online © 2014 Informa Healthcare DOI: 10.3109/14017431.2014.884723
86
S.-Y. Lee et al.
time of cardiac MRI for PVR after TOF total correction.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Methods This study was a single-center, retrospective clinical study that was approved by the Sejong Cardiovascular Institutional Review Board for human research. All patients who underwent cardiac MRI and echocardiography at Sejong General Hospital from January 2008 to December 2012 to evaluate severe PR after TOF total correction using the transannular patch technique were included. Patients who had other significant cardiac lesions that were not associated with PR, such as significant residual shunt, a moderate degree of pulmonary stenosis, and a more than moderate degree of regurgitation from another valve that had been detected immediately after total correction, and those who had a period of more than 6 months between an echocardiogram and a cardiac MRI were excluded. Echocardiography was performed using standard pediatric views on a Vivid 7 or Vivid E9 machine (GE Healthcare, Parsippany, NJ, USA) (4). Images were acquired using 3–8 MHz transducers, which were appropriate for the patient’s size and acoustic windows. The eccentricity of the short-axis left ventricular (LV) cavity profile was assessed using the parasternal short-axis images. The length of the short-axis diameter (perpendicular to the interventricular septum) from the LV septal endocardium to the endocardium of the posterolateral free wall at end-diastole was defined as D1. The length of the orthogonal short-axis diameter between the endocardial surfaces of the anterior and inferior LV free wall at end-diastole was defined as D2. LV eccentricity at end systole or end diastole was defined as the ratio of S2/S1 or D2/D1. The end-systole and end-diastole were defined as the time just before the mitral valve opening corresponded to the smallest LV cross-sectional area and the time just before the mitral valve closure corresponded to the largest LV cross-sectional area. The paradox index is defined by dividing the ratio of the vertical diameter to the transverse diameter at end-systole (S2/S1) by the same ratio at end-diastole (D2/D1), as measured using the parasternal short-axis images at the level of the papillary muscle obtained by echocardiography (Figure 1) (3). There was one observer for the echocardiographic data, and intra-/interobserver variability was not assessed. Cardiac MRI was performed using a 1.5-T Gyroscan Intera CV system (Philips Medical Systems, Best, the Netherlands). Multiphase acquisition was obtained using a steady-state free precession pulse sequence in 2- and 4-chamber planes. From these images, 10–12 contiguous short-axis slabs perpendicular to the long axis of the left ventricle were
Figure 1. Diagram of the method applied to quantify the degree of septal paradoxical motion using the paradox index. D1, length of the short-axis diameter (perpendicular to the interventricular septum) from the left ventricular septal endocardium to the endocardium of the posterolateral free wall; D2, length of the orthogonal short-axis diameter between the endocardial surfaces of the anterior and inferior left ventricular free wall; LV, left ventricle; RV, right ventricle.
obtained (slice thickness, 6–8 mm; inter-slice gap, 0–2 mm). Biventricular volumetric analysis was performed by a radiologist using the Extended MR Workspace software (Philips Medical Systems). Through-plane velocity imaging perpendicular to the main pulmonary artery was performed using a breath-holding, electrocardiogram-triggered, cine-phase contrast pulse sequence to evaluate PR. The volume of forward and regurgitated pulmonary flow was measured from velocity-encoded images by manually tracing the main pulmonary artery contour. The flow volumes were calculated by multiplying the contour area by the average flow velocity within the contour. The flow volumes were summed to determine total forward flow and total regurgitated flow per cardiac cycle. We used branch pulmonary artery maps to measure PR when significant artifacts by prosthetic valves made obtaining direct PR measurements difficult in post-operative patients. The ejection fraction was calculated as the percent stroke volume divided by the end-diastolic volume. The pulmonary regurgitation fraction (PRF) was calculated as the percent backward flow divided by the forward flow. The patients were divided into an operation group and a nonoperation group for PVR after the evaluation of the cardiac MRI data. In our institute, PVR was performed in repaired TOF with moderate or severe PR (PRF ⱖ 25% measured by CMR) and with two or more of the following criteria: RVEDVi ⱖ 160 ml/m2, RVESVi ⱖ 70 ml/m2, LVEDVi ⱕ 65 ml/m2, RV ejection fraction ⱕ 45%, RVOT aneurysm, and clinical criteria, such as exercise intolerance, symptoms and signs of heart failure, syncope or sustained ventricular tachycardia; other hemodynamically significant lesions, such as moderate or severe tricuspid regurgitation, residual atrial or ventricular septal defects; or severe aortic regurgitation. In a few cases, the PVR was performed
Paradox index in repaired TOF on a case-by-case basis (5). Data are expressed as the frequency (%) or the mean ⫾ standard deviation (range), as appropriate. The association between the paradox index and the cardiac MRI data was assessed using correlation analysis and linear regression. The comparison of the two groups was performed using a t-test. The receiver operation characteristic (ROC) curve was used to identify the validity of the paradox index for the RV volume indexes. SPSS v 21.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis, and P ⬍ 0.05 was considered significant.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Results Of the 54 patients, 34 were males and 20 were females. TOF total correction was performed initially in all patients, except in 11 patients who first underwent a shunt operation. The patients’ mean age at total correction was 2.4 (0.3–27.0) years, and the patients’ mean age at cardiac MRI was 19.1 ⫾ 6.9 (9.4–41.0) years. The interval from total correction to MRI was 16.3 ⫾ 5.0 years. Thirty-five patients underwent PVR and 19 patients did not undergo PVR. The characteristics of the patients and the cardiac MRI data are described in Table I. The patients’ mean paradox index measured by echocardiography was 1.22 ⫾ 0.13 (1.00–1.67), and the mean PRF measured by MRI was 44.5 ⫾ 11.9 (4.0–64.0)%. The MRI data showed that the mean PRF and the mean indexed RV volume were larger in the operation group than in the nonoperation group (P ⬍ 0.05). The mean paradox index of the operation group was significantly higher than that of the nonoperation group (1.16 ⫾ 1.12 vs 1.26 ⫾ 1.12, P ⫽ 0.009) (Figure 2).
The paradox index was not significantly correlated with the left ventricular volume indices (LVEDVi and LVESVi), the left ventricular ejection fraction (LVEF), and the PRF, but it was significantly correlated with the right ventricular volume indices (RVEDVi and RVESVi) and the right ventricular ejection fraction (RVEF) (Table II). However, due to collinearity, multiple linear regression analysis was not effective, and RVEF was rejected. In the simple linear regression analysis, RVESVi was expressed as 84.448 ⫻ paradox index ⫺ 23.996 (R2 ⫽ 0.174, P ⫽ 0.002) (Figure 3A), and RVEDVi was expressed as 94.451 ⫻ paradox index ⫹ 42.246 (R2 ⫽ 0.161, P ⫽ 0.003) (Figure 3B). The ROC curve showed that the paradox index is not useful as a screening method when the RVESVi is over 70 ml/m2 (area ⫽ 0.575, 95% CI 0.420–0.730) but that it could be useful when the RVEDVi is over 160 ml/m2 (area ⫽ 0.790, 95% CI 0.688–0.913).
Discussion Our study showed that the paradox index by echocardiography was significantly correlated with RV volume and function as measured by MRI. Therefore, if the RV volume measured by cardiac MRI could be used as an indicator of PVR for chronic PR after TOF total correction, the paradox index could predict the time of MRI evaluation for PVR. Paradoxical septal motion has been reported in patients with RV volume overload, including patients with an atrial septal defect (6). In these patients, the degree of septal paradoxical motion was significantly related to the degree of volume overload (3). In patients with a repaired TOF, RV outflow tract reconstruction using the transannular patch technique could induce
Table I. Data of all patients and each group. Characteristics Patients Age (years) Age of total repair (years) Interval from total repair to MRI (years) Paradox Index LVEDVi (ml/m2) LVESVi (ml/m2) LVEF (%) RVEDVi (ml/m2) RVESVi (ml/m2) RVEF (%) PRF (%)
87
Nonoperation
Operation
P value
54 18.7 ⫾ 6.9 2.4 ⫾ 3.8 16.3 ⫾ 5.0
19 17.7 ⫾ 6.9 2.7 ⫾ 6.0 15.0 ⫾ 3.9
35 19.1 ⫾ 6.6 2.3 ⫾ 1.7 16.9 ⫾ 5.5
0.464 0.739 0.185
1.22 ⫾ 0.13 82.9 ⫾ 24.2 33.8 ⫾ 18.1 60.5 ⫾ 7.0 157.8 ⫾ 30.3 79.3 ⫾ 26.0 51.2 ⫾ 7.9 44.5 ⫾ 11.9
1.16 ⫾ 1.12 87.2 ⫾ 37.8 36.4 ⫾ 29.0 60.8 ⫾ 9.1 135.3 ⫾ 25.3 65.6 ⫾ 31.7 54.3 ⫾ 10.3 38.3 ⫾ 13.8
1.26 ⫾ 1.12 80.5 ⫾ 11.8 32.3 ⫾ 7.8 60.3 ⫾ 5.7 170.0 ⫾ 25.7 86.8 ⫾ 19.0 49.5 ⫾ 5.8 47.8 ⫾ 9.2
0.009 0.334 0.441 0.822 0.000 0.003 0.032 0.004
PVR, pulmonary valve replacement; LVEDVi, left ventricular end-diastolic volume index; LVESVi, left ventricular end-systolic volume index; LVEF, left ventricular ejection fraction; RVEDVi, right ventricular end-diastolic volume index; RVESVi, right ventricular end-systolic volume index; RVEF, right ventricular ejection fraction; PRF, pulmonary regurgitation fraction. Data are the mean ⫾ standard deviation.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
88
S.-Y. Lee et al.
Figure 2. Difference of the paradox index between the operation group and the nonoperation group.
chronic PR and volume overload due to severe PR, which could enlarge the RV and subsequently elicit paradoxical septal motion. Paradoxical septal motion was quantified using the echocardiographyderived paradox index (Figure 1) (3). Although ventricular volume can be precisely measured by cardiac MRI, it is not suitable as an annual follow-up tool because it is expensive and requires several hours to perform. Therefore, we searched for a useful indicator as a marker of PVR and RV volume overload. According to a study by Abd El Rahman et al., paradoxical septal motion can be influenced by septal dyssynchrony and decreased regional septal motion (3). However, their subjects did not have PR, and the paradoxical septal motion index was only 1.12, which is lower than that observed in our subjects. Therefore, paradoxical septal motion can be influenced by not only diastolic volume overload but also systolic function, especially in post-TOF correction. We can assume that S2 is always the same as S1 if the systolic function is conserved and the paradoxical index is 1/(D2/D1). However, no references that referred to such an index for volume overload lesion were found. Nevertheless, diastolic volume overload is thought to be the main factor in the paradoxical index, and it can be used to evaluate volume changes.
The main problem in patients with repaired TOF using a transannular patch for a hypoplastic pulmonary valve is RV volume overload from chronic PR (7). Most patients with chronic severe PR do not have any symptoms in childhood. At 20 years of age, only 6% of patients have symptoms, but the incidence increases to 29% at 40 years of age (8). Both experimental evidence and clinical data have shown that the severity of PR increases over time and that the increase in RV stroke volume leads to a progressive rise in the size and compliance of the central pulmonary arteries and to increased RV compliance (7). Combined with a longer duration of diastole as the heart rate decreases with age, these changes lead to a progressive increase in the degree of PR over time (7). Irreversible myocardial injury associated with fibrosis is induced by increased PR and RV volume, resulting in decreased RV and LV function (7). Therefore, PVR is required during the reversible period in patients who have severe PR and ventricular dysfunction (1,9). Severe PR after TOF repair has detrimental hemodynamic and clinical effects. PVR should be performed in symptomatic patients or in patients with evidence of progressive RV dilatation and/or dysfunction (9). Measuring RV volume by echocardiography is not reliable because of the crescent morphology of the RV. Although the horizontal and longitudinal diameter of the RV is well correlated with RV volume, determining the optimal timing of PVR using the RV diameter measured by echocardiography is not easy (10). The optimal timing for PVR depends on volumetric data such as RVEDVi, RVESVi, and LVEDVi from CMR in many institutions (7,11,12). Studies reporting no improvement in RV function studied patients with already depressed RV function, whereas studies reporting improvements in RV function had patients with preserved RV function (5,13– 16). In some reports, the indication for PVR was suggested by RVEDVi greater than 160–165 ml/m2 and RVESVi greater than 70–80 ml/m2 (6,13). Of course, the optimal indication could be changed according to patient symptoms, other valve regurgitation, septal defects, and ventricular function. In a follow-up cardiac MRI study, Lee et al. suggested a RVEDVi of 163 ml/m2 and a RVESVi of 80 ml/m2
Table II. Correlation of the paradox index with various cardiac volume indices measured by MRI.
Paradox index Pearson coefficient P value
LVEDVi
LVESVi
LVEF
RVEDVi
RVESVi
⫺0.072 0.606
⫺ 0.096 0.491
0.052 0.707
0.401 0.003
0.417 0.002
RVEF
PRF
⫺0.0314 0.021
0.194 0.160
LVEDVi, left ventricular end-diastolic volume index; LVESVi, left ventricular end-systolic volume index; LVEF, left ventricular ejection fraction; RVEDVi, right ventricular end-diastolic volume index; RVESVi, right ventricular end-systolic volume index; RVEF, right ventricular ejection fraction; PRF, pulmonary regurgitation fraction
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Paradox index in repaired TOF
89
Figure 3. Correlation between the paradox index and cardiac magnetic resonance imaging data. (A) Correlation between the paradox index and the right ventricular end-systolic volume index (RVESVi); (B) correlation between the paradox index and the right ventricular end-diastolic volume index (RVEDVi).
as cut-off values for normalizing the RV volume after PVR (12). In our study, PVR was performed based not only on cardiac MRI data but also on symptoms, exercise capacity, and arrhythmia. Nevertheless, the RV volume and function were considered to be the most important indicators of PVR. Thus, the RV volume measured by MRI was significantly higher in the operation group in our study. It was strange that neither the RV volume measured by MRI nor the paradox index were correlated with the period of exposure to chronic PR and PRF. RVESVi and RVEDVi, as measured by cardiac MRI, were well correlated with the paradox index and could be expressed using linear regression. We suggested that RV volume measured by cardiac MRI in patients with chronic PR after TOF total correction could be estimated using a simple parameter measured by echocardiography. Although an accurate RV volume could not be estimated by the paradox index due to low coefficients of determination for RVESVi and RVEDVi (R2 ⫽ 0.174 and 0.161), the paradox index could be useful as a screening parameter. If an RVEDVi of 160 ml/m2 is assumed to be an indication for PVR, the corresponding paradox index is 1.24. A paradox index of 1.24 showed 68% sensitivity and 83% specificity on the ROC curve. In other words, 32% of all PVR candidates can be excluded from treatment by this paradoxical index of 1.24 when used as a screening tool. The value of the paradox index would not be its screening value but its use as an estimating parameter. Therefore, our recommendation is to consider the paradoxical index to be an estimating parameter to obtain a more accurate volumetric evaluation whenever the right ventricle appears dilated or the function seems to be declined. An accurate volumetric evaluation could then be performed by MRI or three-dimensional echocardiography. Our study has several limitations, including that it was retrospective and enrolled a small number of
patients. There could be a selection bias in our sample. We tried to exclude hemodynamic problems other than chronic PR. We could not define the obvious exclusion criteria in other hemodynamic factors. Our operation criteria were also complicated. Additionally, the paradox index could be affected by factors other than volume overload, such as dyssynchrony between either ventricles or ventricular dysfunction. LV diastolic dysfunction after TOF repair could deteriorate RV function, but we did not consider any parameters for LV diastolic function. RV hypertrophy was not considered in our study. RV hypertrophy could impact the RV diastolic dysfunction and the paradoxical index. In our study, the age of patients at the time of total correction was slightly higher when compared to the current policy, which has a high possibility of RV hypertrophy and dysfunction. Conclusion The paradox index, as measured by echocardiography, was correlated with the RV volume index, which was identified through cardiac MRI in patients with chronic PR after TOF total correction. In patients lacking factors other than chronic PR for the RV volume overload, the paradox index could be considered as one of useful tools for determining the optimal timing for cardiac MRI. Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. Lee C. Surgical management of chronic pulmonary regurgitation after relief of right ventricular outflow tract obstruction. Korean Circ J. 2012;42:1–7.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
90
S.-Y. Lee et al.
2. Khoo NS, Young A, Occleshaw C, Cowan B, Zeng IS, Gentles TL. Assessments of right ventricular volume and function using three-dimensional echocardiography in older children and adults with congenital heart disease: comparison with cardiac magnetic resonance imaging. J Am Soc Echocardiogr. 2009;22:1279–88. 3. Abd El Rahman MY, Hui W, Dsebissowa F, Schubert S, Gutberlet M, Hetzer R, et al. Quantitative analysis of paradoxical interventricular septal motion following corrective surgery of tetralogy of fallot. Pediatr Cardiol. 2005;26:379–84. 4. Lai WW, Geva T, Shirali GS, Frommelt PC, Humes RA, Brook MM, et al. Guidelines and standards for performance of a pediatric echocardiogram: a report from the Task Force of the Pediatric Council of the American Society of Echocardiography. J Am Soc Echocardiogr. 2006;19:1413–30. 5. Geva T, Gauvreau K, Powell AJ, Cecchin F, Rhodes J, Geva J, del Nido P. Randomized trial of pulmonary valve replacement with and without right ventricular remodeling surgery. Circulation. 2010;122:S201–8. 6. Popio KA, Gorlin R, Teichholz LE, Cohn PF, Bechtel D, Herman MV. Abnormalities of left ventricular function and geometry in adults with an atrial septal defect. Ventriculographic, hemodynamic and echocardiographic studies. Am J Cardiol. 1975;36:302–8. 7. Geva T. Indications and timing of pulmonary valve replacement after tetralogy of Fallot repair. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2006:11–22. 8. Shimazaki Y, Blackstone EH, Kirklin JW. The natural history of isolated congenital pulmonary valve incompetence: surgical implications. Thorac Cardiovasc Surg. 1984;32:257–9. 9. Davlouros PA, Karatza AA, Gatzoulis MA, Shore DF. Timing and type of surgery for severe pulmonary regurgita-
10.
11.
12.
13.
14.
15.
16.
tion after repair of tetralogy of Fallot. Int J Cardiol. 2004; 97:91–101. Dragulescu A, Grosse-Wortmann L, Fackoury C, Mertens L. Echocardiographic assessment of right ventricular volumes: a comparison of different techniques in children after surgical repair of tetralogy of Fallot. Eur Heart J Cardiovasc Imaging. 2012;13:596–604. Hwang SH, Choi BW. Advanced cardiac MR imaging for myocardial characterization and quantification: T1 mapping. Korean Circ J. 2013;43:1–6. Lee C, Kim YM, Lee CH, Kwak JG, Park CS, Song JY, et al. Outcomes of pulmonary valve replacement in 170 patients with chronic pulmonary regurgitation after relief of right ventricular outflow tract obstruction: implications for optimal timing of pulmonary valve replacement. J Am Coll Cardiol. 2012;60:1005–14. Frigiola A, Tsang V, Bull C, Coats L, Khambadkone S, Derrick G, et al. Biventricular response after pulmonary valve replacement for right ventricular outflow tract dysfunction: is age a predictor of outcome? Circulation. 2008;118:S182–90. Therrien J, Provost Y, Merchant N, Williams W, Colman J, Webb G. Optimal timing for pulmonary valve replacement in adults after tetralogy of Fallot repair. Am J Cardiol. 2005;95:779–82. Ghez O, Tsang VT, Frigiola A, Coats L, Taylor A, Van Doorn C, et al. Right ventricular outflow tract reconstruction for pulmonary regurgitation after repair of tetralogy of Fallot. Preliminary results. Eur J Cardiothorac Surg. 2007;31:654–8. Harrild DM, Berul CI, Cecchin F, Geva T, Gauvreau K, Pigula F, et al. Pulmonary valve replacement in tetralogy of Fallot: impact on survival and ventricular tachycardia. Circulation 2009;119:445–51.
ORIGINAL ARTICLE
The echocardiographic paradox index in patients with a repaired tetralogy of Fallot
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
SANG-YUN LEE1, JINYOUNG SONG2, SUNG-HO KIM1, SO-ICK JANG1 & YANG-MIN KIM3 1Department
of Pediatrics, Bucheon-si, Republic of Korea, 2Department of Pediatrics, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea, and 3Department of Radiology, Sejong Cardiovascular Institute, Bucheon-si, Republic of Korea
Abstract Objectives. Right ventricular (RV) volume is very important for pulmonary valve replacement after the total correction of tetralogy of Fallot (TOF), and we attempted to identify a convenient echocardiographic index that is well correlated with the volumetric data obtained through magnetic resonance imaging (MRI). Design. All patients who underwent cardiac MRI and echocardiography at Sejong General Hospital for evaluating pulmonary regurgitation after TOF total correction were included. The paradox index is the amount of paradoxical motion of the interventricular septum on the short-axis echocardiographic view. The paradox index was compared to several cardiac MRI indices. Results. Fifty-four patients were included. The paradox index for all patients was 1.22 ⫾ 0.12 (1.06–1.67), and the index of the operation group was significantly higher than that of the non-operation group (1.26 ⫾ 1.12 vs1.16 ⫾ 1.12, P ⫽ 0.009). The paradox index was well correlated with the RV systolic and diastolic volumes, as measured by cardiac MRI (P ⫽ 0.002 and 0.003). Using a simple linear regression analysis, a paradox index of 1.24 corresponded to a RV diastolic volume of 160 ml/m2. Conclusions. The paradox index could help to indicate the time for an MRI analysis of the RV volume in patients after TOF total correction. Key words: echocardiography, paradox index, tetralogy of Fallot
Introduction Chronic pulmonary regurgitation (PR) after the correction of a right ventricular (RV) outflow obstructive lesion is known to impact RV function. Most patients whose ailments were corrected using the transannular patch technique or graft insertion for severe tetralogy of Fallot (TOF) or pulmonary atresia usually experience chronic PR and require pulmonary valve replacement (PVR) (1). However, the best timing for PVR remains under debate. RV dilatation is known to be a key indicator for the need of PVR. In terms of evaluating RV dilatation, threedimensional echocardiography and cardiac magnetic resonance imaging (MRI) measure the RV volume directly, whereas two-dimensional echocardiography provides some indicators of RV dilatation. However, three-dimensional echocardiography might not be accurate, particularly when the right ventricle is
dilated (2). Due to recent advances in cardiac MRI, there is a tendency to determine the optimal time of PVR based on the cardiac MRI data. However, cardiac MRI is costly and difficult to perform; thus, regular follow-up is difficult. Therefore, it would be helpful if an echocardiographic parameter that is capable of suggesting significant RV dilatation prior to performing MRI could be identified. The paradox index is the amount of paradoxical motion of the inter-ventricular septum on the short-axis echocardiographic view (3). It is assumed that the paradoxical motion of the inter-ventricular septum depends on the amount of PR and the RV volume. In this study, the relationship between the volumetric data from cardiac MRI and the paradox index was analyzed. The aim of this study was to evaluate the usefulness of the paradox index as a convenient parameter for determining the optimal
Correspondence: Dr Jinyoung Song, MD, Department of Pediatrics, Samsung Medical center, Sungkyunkwan University School of Medicine, 50 Irwon-Dong, Gangnam-gu, 135-710, Seoul, South Korea. Tel: ⫹ 82-2-3410-3539. Fax: ⫹ 82-2-3410-0043. E-mail: [email protected] (Received 23 October 2013 ; revised 10 January 2014 ; accepted 12 January 2014 ) ISSN 1401-7431 print/ISSN 1651-2006 online © 2014 Informa Healthcare DOI: 10.3109/14017431.2014.884723
86
S.-Y. Lee et al.
time of cardiac MRI for PVR after TOF total correction.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Methods This study was a single-center, retrospective clinical study that was approved by the Sejong Cardiovascular Institutional Review Board for human research. All patients who underwent cardiac MRI and echocardiography at Sejong General Hospital from January 2008 to December 2012 to evaluate severe PR after TOF total correction using the transannular patch technique were included. Patients who had other significant cardiac lesions that were not associated with PR, such as significant residual shunt, a moderate degree of pulmonary stenosis, and a more than moderate degree of regurgitation from another valve that had been detected immediately after total correction, and those who had a period of more than 6 months between an echocardiogram and a cardiac MRI were excluded. Echocardiography was performed using standard pediatric views on a Vivid 7 or Vivid E9 machine (GE Healthcare, Parsippany, NJ, USA) (4). Images were acquired using 3–8 MHz transducers, which were appropriate for the patient’s size and acoustic windows. The eccentricity of the short-axis left ventricular (LV) cavity profile was assessed using the parasternal short-axis images. The length of the short-axis diameter (perpendicular to the interventricular septum) from the LV septal endocardium to the endocardium of the posterolateral free wall at end-diastole was defined as D1. The length of the orthogonal short-axis diameter between the endocardial surfaces of the anterior and inferior LV free wall at end-diastole was defined as D2. LV eccentricity at end systole or end diastole was defined as the ratio of S2/S1 or D2/D1. The end-systole and end-diastole were defined as the time just before the mitral valve opening corresponded to the smallest LV cross-sectional area and the time just before the mitral valve closure corresponded to the largest LV cross-sectional area. The paradox index is defined by dividing the ratio of the vertical diameter to the transverse diameter at end-systole (S2/S1) by the same ratio at end-diastole (D2/D1), as measured using the parasternal short-axis images at the level of the papillary muscle obtained by echocardiography (Figure 1) (3). There was one observer for the echocardiographic data, and intra-/interobserver variability was not assessed. Cardiac MRI was performed using a 1.5-T Gyroscan Intera CV system (Philips Medical Systems, Best, the Netherlands). Multiphase acquisition was obtained using a steady-state free precession pulse sequence in 2- and 4-chamber planes. From these images, 10–12 contiguous short-axis slabs perpendicular to the long axis of the left ventricle were
Figure 1. Diagram of the method applied to quantify the degree of septal paradoxical motion using the paradox index. D1, length of the short-axis diameter (perpendicular to the interventricular septum) from the left ventricular septal endocardium to the endocardium of the posterolateral free wall; D2, length of the orthogonal short-axis diameter between the endocardial surfaces of the anterior and inferior left ventricular free wall; LV, left ventricle; RV, right ventricle.
obtained (slice thickness, 6–8 mm; inter-slice gap, 0–2 mm). Biventricular volumetric analysis was performed by a radiologist using the Extended MR Workspace software (Philips Medical Systems). Through-plane velocity imaging perpendicular to the main pulmonary artery was performed using a breath-holding, electrocardiogram-triggered, cine-phase contrast pulse sequence to evaluate PR. The volume of forward and regurgitated pulmonary flow was measured from velocity-encoded images by manually tracing the main pulmonary artery contour. The flow volumes were calculated by multiplying the contour area by the average flow velocity within the contour. The flow volumes were summed to determine total forward flow and total regurgitated flow per cardiac cycle. We used branch pulmonary artery maps to measure PR when significant artifacts by prosthetic valves made obtaining direct PR measurements difficult in post-operative patients. The ejection fraction was calculated as the percent stroke volume divided by the end-diastolic volume. The pulmonary regurgitation fraction (PRF) was calculated as the percent backward flow divided by the forward flow. The patients were divided into an operation group and a nonoperation group for PVR after the evaluation of the cardiac MRI data. In our institute, PVR was performed in repaired TOF with moderate or severe PR (PRF ⱖ 25% measured by CMR) and with two or more of the following criteria: RVEDVi ⱖ 160 ml/m2, RVESVi ⱖ 70 ml/m2, LVEDVi ⱕ 65 ml/m2, RV ejection fraction ⱕ 45%, RVOT aneurysm, and clinical criteria, such as exercise intolerance, symptoms and signs of heart failure, syncope or sustained ventricular tachycardia; other hemodynamically significant lesions, such as moderate or severe tricuspid regurgitation, residual atrial or ventricular septal defects; or severe aortic regurgitation. In a few cases, the PVR was performed
Paradox index in repaired TOF on a case-by-case basis (5). Data are expressed as the frequency (%) or the mean ⫾ standard deviation (range), as appropriate. The association between the paradox index and the cardiac MRI data was assessed using correlation analysis and linear regression. The comparison of the two groups was performed using a t-test. The receiver operation characteristic (ROC) curve was used to identify the validity of the paradox index for the RV volume indexes. SPSS v 21.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis, and P ⬍ 0.05 was considered significant.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Results Of the 54 patients, 34 were males and 20 were females. TOF total correction was performed initially in all patients, except in 11 patients who first underwent a shunt operation. The patients’ mean age at total correction was 2.4 (0.3–27.0) years, and the patients’ mean age at cardiac MRI was 19.1 ⫾ 6.9 (9.4–41.0) years. The interval from total correction to MRI was 16.3 ⫾ 5.0 years. Thirty-five patients underwent PVR and 19 patients did not undergo PVR. The characteristics of the patients and the cardiac MRI data are described in Table I. The patients’ mean paradox index measured by echocardiography was 1.22 ⫾ 0.13 (1.00–1.67), and the mean PRF measured by MRI was 44.5 ⫾ 11.9 (4.0–64.0)%. The MRI data showed that the mean PRF and the mean indexed RV volume were larger in the operation group than in the nonoperation group (P ⬍ 0.05). The mean paradox index of the operation group was significantly higher than that of the nonoperation group (1.16 ⫾ 1.12 vs 1.26 ⫾ 1.12, P ⫽ 0.009) (Figure 2).
The paradox index was not significantly correlated with the left ventricular volume indices (LVEDVi and LVESVi), the left ventricular ejection fraction (LVEF), and the PRF, but it was significantly correlated with the right ventricular volume indices (RVEDVi and RVESVi) and the right ventricular ejection fraction (RVEF) (Table II). However, due to collinearity, multiple linear regression analysis was not effective, and RVEF was rejected. In the simple linear regression analysis, RVESVi was expressed as 84.448 ⫻ paradox index ⫺ 23.996 (R2 ⫽ 0.174, P ⫽ 0.002) (Figure 3A), and RVEDVi was expressed as 94.451 ⫻ paradox index ⫹ 42.246 (R2 ⫽ 0.161, P ⫽ 0.003) (Figure 3B). The ROC curve showed that the paradox index is not useful as a screening method when the RVESVi is over 70 ml/m2 (area ⫽ 0.575, 95% CI 0.420–0.730) but that it could be useful when the RVEDVi is over 160 ml/m2 (area ⫽ 0.790, 95% CI 0.688–0.913).
Discussion Our study showed that the paradox index by echocardiography was significantly correlated with RV volume and function as measured by MRI. Therefore, if the RV volume measured by cardiac MRI could be used as an indicator of PVR for chronic PR after TOF total correction, the paradox index could predict the time of MRI evaluation for PVR. Paradoxical septal motion has been reported in patients with RV volume overload, including patients with an atrial septal defect (6). In these patients, the degree of septal paradoxical motion was significantly related to the degree of volume overload (3). In patients with a repaired TOF, RV outflow tract reconstruction using the transannular patch technique could induce
Table I. Data of all patients and each group. Characteristics Patients Age (years) Age of total repair (years) Interval from total repair to MRI (years) Paradox Index LVEDVi (ml/m2) LVESVi (ml/m2) LVEF (%) RVEDVi (ml/m2) RVESVi (ml/m2) RVEF (%) PRF (%)
87
Nonoperation
Operation
P value
54 18.7 ⫾ 6.9 2.4 ⫾ 3.8 16.3 ⫾ 5.0
19 17.7 ⫾ 6.9 2.7 ⫾ 6.0 15.0 ⫾ 3.9
35 19.1 ⫾ 6.6 2.3 ⫾ 1.7 16.9 ⫾ 5.5
0.464 0.739 0.185
1.22 ⫾ 0.13 82.9 ⫾ 24.2 33.8 ⫾ 18.1 60.5 ⫾ 7.0 157.8 ⫾ 30.3 79.3 ⫾ 26.0 51.2 ⫾ 7.9 44.5 ⫾ 11.9
1.16 ⫾ 1.12 87.2 ⫾ 37.8 36.4 ⫾ 29.0 60.8 ⫾ 9.1 135.3 ⫾ 25.3 65.6 ⫾ 31.7 54.3 ⫾ 10.3 38.3 ⫾ 13.8
1.26 ⫾ 1.12 80.5 ⫾ 11.8 32.3 ⫾ 7.8 60.3 ⫾ 5.7 170.0 ⫾ 25.7 86.8 ⫾ 19.0 49.5 ⫾ 5.8 47.8 ⫾ 9.2
0.009 0.334 0.441 0.822 0.000 0.003 0.032 0.004
PVR, pulmonary valve replacement; LVEDVi, left ventricular end-diastolic volume index; LVESVi, left ventricular end-systolic volume index; LVEF, left ventricular ejection fraction; RVEDVi, right ventricular end-diastolic volume index; RVESVi, right ventricular end-systolic volume index; RVEF, right ventricular ejection fraction; PRF, pulmonary regurgitation fraction. Data are the mean ⫾ standard deviation.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
88
S.-Y. Lee et al.
Figure 2. Difference of the paradox index between the operation group and the nonoperation group.
chronic PR and volume overload due to severe PR, which could enlarge the RV and subsequently elicit paradoxical septal motion. Paradoxical septal motion was quantified using the echocardiographyderived paradox index (Figure 1) (3). Although ventricular volume can be precisely measured by cardiac MRI, it is not suitable as an annual follow-up tool because it is expensive and requires several hours to perform. Therefore, we searched for a useful indicator as a marker of PVR and RV volume overload. According to a study by Abd El Rahman et al., paradoxical septal motion can be influenced by septal dyssynchrony and decreased regional septal motion (3). However, their subjects did not have PR, and the paradoxical septal motion index was only 1.12, which is lower than that observed in our subjects. Therefore, paradoxical septal motion can be influenced by not only diastolic volume overload but also systolic function, especially in post-TOF correction. We can assume that S2 is always the same as S1 if the systolic function is conserved and the paradoxical index is 1/(D2/D1). However, no references that referred to such an index for volume overload lesion were found. Nevertheless, diastolic volume overload is thought to be the main factor in the paradoxical index, and it can be used to evaluate volume changes.
The main problem in patients with repaired TOF using a transannular patch for a hypoplastic pulmonary valve is RV volume overload from chronic PR (7). Most patients with chronic severe PR do not have any symptoms in childhood. At 20 years of age, only 6% of patients have symptoms, but the incidence increases to 29% at 40 years of age (8). Both experimental evidence and clinical data have shown that the severity of PR increases over time and that the increase in RV stroke volume leads to a progressive rise in the size and compliance of the central pulmonary arteries and to increased RV compliance (7). Combined with a longer duration of diastole as the heart rate decreases with age, these changes lead to a progressive increase in the degree of PR over time (7). Irreversible myocardial injury associated with fibrosis is induced by increased PR and RV volume, resulting in decreased RV and LV function (7). Therefore, PVR is required during the reversible period in patients who have severe PR and ventricular dysfunction (1,9). Severe PR after TOF repair has detrimental hemodynamic and clinical effects. PVR should be performed in symptomatic patients or in patients with evidence of progressive RV dilatation and/or dysfunction (9). Measuring RV volume by echocardiography is not reliable because of the crescent morphology of the RV. Although the horizontal and longitudinal diameter of the RV is well correlated with RV volume, determining the optimal timing of PVR using the RV diameter measured by echocardiography is not easy (10). The optimal timing for PVR depends on volumetric data such as RVEDVi, RVESVi, and LVEDVi from CMR in many institutions (7,11,12). Studies reporting no improvement in RV function studied patients with already depressed RV function, whereas studies reporting improvements in RV function had patients with preserved RV function (5,13– 16). In some reports, the indication for PVR was suggested by RVEDVi greater than 160–165 ml/m2 and RVESVi greater than 70–80 ml/m2 (6,13). Of course, the optimal indication could be changed according to patient symptoms, other valve regurgitation, septal defects, and ventricular function. In a follow-up cardiac MRI study, Lee et al. suggested a RVEDVi of 163 ml/m2 and a RVESVi of 80 ml/m2
Table II. Correlation of the paradox index with various cardiac volume indices measured by MRI.
Paradox index Pearson coefficient P value
LVEDVi
LVESVi
LVEF
RVEDVi
RVESVi
⫺0.072 0.606
⫺ 0.096 0.491
0.052 0.707
0.401 0.003
0.417 0.002
RVEF
PRF
⫺0.0314 0.021
0.194 0.160
LVEDVi, left ventricular end-diastolic volume index; LVESVi, left ventricular end-systolic volume index; LVEF, left ventricular ejection fraction; RVEDVi, right ventricular end-diastolic volume index; RVESVi, right ventricular end-systolic volume index; RVEF, right ventricular ejection fraction; PRF, pulmonary regurgitation fraction
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
Paradox index in repaired TOF
89
Figure 3. Correlation between the paradox index and cardiac magnetic resonance imaging data. (A) Correlation between the paradox index and the right ventricular end-systolic volume index (RVESVi); (B) correlation between the paradox index and the right ventricular end-diastolic volume index (RVEDVi).
as cut-off values for normalizing the RV volume after PVR (12). In our study, PVR was performed based not only on cardiac MRI data but also on symptoms, exercise capacity, and arrhythmia. Nevertheless, the RV volume and function were considered to be the most important indicators of PVR. Thus, the RV volume measured by MRI was significantly higher in the operation group in our study. It was strange that neither the RV volume measured by MRI nor the paradox index were correlated with the period of exposure to chronic PR and PRF. RVESVi and RVEDVi, as measured by cardiac MRI, were well correlated with the paradox index and could be expressed using linear regression. We suggested that RV volume measured by cardiac MRI in patients with chronic PR after TOF total correction could be estimated using a simple parameter measured by echocardiography. Although an accurate RV volume could not be estimated by the paradox index due to low coefficients of determination for RVESVi and RVEDVi (R2 ⫽ 0.174 and 0.161), the paradox index could be useful as a screening parameter. If an RVEDVi of 160 ml/m2 is assumed to be an indication for PVR, the corresponding paradox index is 1.24. A paradox index of 1.24 showed 68% sensitivity and 83% specificity on the ROC curve. In other words, 32% of all PVR candidates can be excluded from treatment by this paradoxical index of 1.24 when used as a screening tool. The value of the paradox index would not be its screening value but its use as an estimating parameter. Therefore, our recommendation is to consider the paradoxical index to be an estimating parameter to obtain a more accurate volumetric evaluation whenever the right ventricle appears dilated or the function seems to be declined. An accurate volumetric evaluation could then be performed by MRI or three-dimensional echocardiography. Our study has several limitations, including that it was retrospective and enrolled a small number of
patients. There could be a selection bias in our sample. We tried to exclude hemodynamic problems other than chronic PR. We could not define the obvious exclusion criteria in other hemodynamic factors. Our operation criteria were also complicated. Additionally, the paradox index could be affected by factors other than volume overload, such as dyssynchrony between either ventricles or ventricular dysfunction. LV diastolic dysfunction after TOF repair could deteriorate RV function, but we did not consider any parameters for LV diastolic function. RV hypertrophy was not considered in our study. RV hypertrophy could impact the RV diastolic dysfunction and the paradoxical index. In our study, the age of patients at the time of total correction was slightly higher when compared to the current policy, which has a high possibility of RV hypertrophy and dysfunction. Conclusion The paradox index, as measured by echocardiography, was correlated with the RV volume index, which was identified through cardiac MRI in patients with chronic PR after TOF total correction. In patients lacking factors other than chronic PR for the RV volume overload, the paradox index could be considered as one of useful tools for determining the optimal timing for cardiac MRI. Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. Lee C. Surgical management of chronic pulmonary regurgitation after relief of right ventricular outflow tract obstruction. Korean Circ J. 2012;42:1–7.
Scand Cardiovasc J Downloaded from informahealthcare.com by Gazi Univ. on 01/08/15 For personal use only.
90
S.-Y. Lee et al.
2. Khoo NS, Young A, Occleshaw C, Cowan B, Zeng IS, Gentles TL. Assessments of right ventricular volume and function using three-dimensional echocardiography in older children and adults with congenital heart disease: comparison with cardiac magnetic resonance imaging. J Am Soc Echocardiogr. 2009;22:1279–88. 3. Abd El Rahman MY, Hui W, Dsebissowa F, Schubert S, Gutberlet M, Hetzer R, et al. Quantitative analysis of paradoxical interventricular septal motion following corrective surgery of tetralogy of fallot. Pediatr Cardiol. 2005;26:379–84. 4. Lai WW, Geva T, Shirali GS, Frommelt PC, Humes RA, Brook MM, et al. Guidelines and standards for performance of a pediatric echocardiogram: a report from the Task Force of the Pediatric Council of the American Society of Echocardiography. J Am Soc Echocardiogr. 2006;19:1413–30. 5. Geva T, Gauvreau K, Powell AJ, Cecchin F, Rhodes J, Geva J, del Nido P. Randomized trial of pulmonary valve replacement with and without right ventricular remodeling surgery. Circulation. 2010;122:S201–8. 6. Popio KA, Gorlin R, Teichholz LE, Cohn PF, Bechtel D, Herman MV. Abnormalities of left ventricular function and geometry in adults with an atrial septal defect. Ventriculographic, hemodynamic and echocardiographic studies. Am J Cardiol. 1975;36:302–8. 7. Geva T. Indications and timing of pulmonary valve replacement after tetralogy of Fallot repair. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2006:11–22. 8. Shimazaki Y, Blackstone EH, Kirklin JW. The natural history of isolated congenital pulmonary valve incompetence: surgical implications. Thorac Cardiovasc Surg. 1984;32:257–9. 9. Davlouros PA, Karatza AA, Gatzoulis MA, Shore DF. Timing and type of surgery for severe pulmonary regurgita-
10.
11.
12.
13.
14.
15.
16.
tion after repair of tetralogy of Fallot. Int J Cardiol. 2004; 97:91–101. Dragulescu A, Grosse-Wortmann L, Fackoury C, Mertens L. Echocardiographic assessment of right ventricular volumes: a comparison of different techniques in children after surgical repair of tetralogy of Fallot. Eur Heart J Cardiovasc Imaging. 2012;13:596–604. Hwang SH, Choi BW. Advanced cardiac MR imaging for myocardial characterization and quantification: T1 mapping. Korean Circ J. 2013;43:1–6. Lee C, Kim YM, Lee CH, Kwak JG, Park CS, Song JY, et al. Outcomes of pulmonary valve replacement in 170 patients with chronic pulmonary regurgitation after relief of right ventricular outflow tract obstruction: implications for optimal timing of pulmonary valve replacement. J Am Coll Cardiol. 2012;60:1005–14. Frigiola A, Tsang V, Bull C, Coats L, Khambadkone S, Derrick G, et al. Biventricular response after pulmonary valve replacement for right ventricular outflow tract dysfunction: is age a predictor of outcome? Circulation. 2008;118:S182–90. Therrien J, Provost Y, Merchant N, Williams W, Colman J, Webb G. Optimal timing for pulmonary valve replacement in adults after tetralogy of Fallot repair. Am J Cardiol. 2005;95:779–82. Ghez O, Tsang VT, Frigiola A, Coats L, Taylor A, Van Doorn C, et al. Right ventricular outflow tract reconstruction for pulmonary regurgitation after repair of tetralogy of Fallot. Preliminary results. Eur J Cardiothorac Surg. 2007;31:654–8. Harrild DM, Berul CI, Cecchin F, Geva T, Gauvreau K, Pigula F, et al. Pulmonary valve replacement in tetralogy of Fallot: impact on survival and ventricular tachycardia. Circulation 2009;119:445–51.
