Three and four dimensional ultrasound: a novel method for evaluating fetal cardiac anomalies

Transcription

1 PRENATAL DIAGNOSIS Prenat Diagn 2009; 29: Published online 1 April 2009 in Wiley InterScience ( Three and four dimensional ultrasound: a novel method for evaluating fetal cardiac anomalies L. Gindes 1 *, J. Hegesh 2,B.Weisz 1, Y. Gilboa 1 and R. Achiron 1 1 Department of Obstetrics and Gynecology, The Chaim Sheba Medical Center, Tel Hashomer, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Israel 2 Department of Pediatric Cardiology, The Chaim Sheba Medical Center, Tel Hashomer, affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Israel Objective To evaluate the role of various new models of 3- and 4-dimensional (3D and 4D) ultrasound (US) applications in prenatal assessment of fetal cardiac anomalies. Methods Volume data sets of 81 fetuses with fetal cardiac anomalies, as previously diagnosed by 2D US, were acquired by 3D and cine 4D using spatiotemporal image correlation (STIC) software. Various additional rendering tools were applied. Color, power, high definition Doppler and B-flow were added to the volumes acquired. A retrospective offline analysis of the cardiac defects was performed. Results The mean gestational age at diagnosis was 24 weeks (range 13 38); 128 anomalies were detected and were classified into the following categories: I, Situs anomalies in 8 cases; II, abnormal four-chamber view in 63 cases; III, outflows tract anomalies in 27 cases; IV, arches anomalies in 21 cases; and V, veins anomalies in 9 cases. Rendering tools differed in each groups of anomalies. Conclusions Fetal cardiac anomalies can be evaluated adequately by the information gained by 3D and 4D volumes obtained by STIC. Since no single module is sufficiently accurate for the diagnosis of all cardiac anomalies, each of the cardiac anomaly categories requires different and appropriate module of visualization. Copyright 2009 John Wiley & Sons, Ltd. Supporting information may be found in the online version of this article. KEY WORDS: 3-dimensional ultrasound; cardiac anomalies; fetus; STIC INTRODUCTION Structural abnormalities of the heart and great vessels are common congenital abnormalities, occurring in approximately 8 in 1000 live neonates (Mitchell et al., 1971; Hoffman et al., 1978; Ferencz et al., 1985). For more than two decades, 2-dimentional (2D) ultrasound (US) has become the tool of choice for detecting cardiac anomalies (Allan et al., 1980; Achiron et al., 1992; Yagel et al., 2001; Yagel et al., 2002). The technical improvement in 2D gray scale, with the advent of transvaginal transducers mounted with color and pulsed- Doppler capabilities, has enabled an earlier prenatal diagnosis of fetal cardiac anomalies, even during the first and second trimesters affecting the entire pregnancy course (Yagel et al., 2001; Achiron et al., 1994a, 1994b, 2008; Carvalho, 2004). The introduction of 3D US has opened a new era and advanced our capabilities in the diagnosis of fetal malformations (Goncalves et al., 2005). However, since the fetal heart is a moving target, its manipulation with conventional 3D technology is difficult. Nevertheless, the *Correspondence to: L. Gindes, Department of Obstetrics and Gynecology, The Chaim Sheba Medical Center, Tel-Hashomer, Ramat Gan, Israel. gindesl@zahav.net.il new modalities for motion gated cardiac scanning (Nelson et al., 1996) and the application of spatiotemporal image correlation (STIC) software (DeVore et al., 2003; Chaoui et al., 2004) have allowed the manipulation of the heart volume and the ability to assess its normal and abnormal anatomy with the use of dynamic multiplanar slicing or surface rendering of heart volume (Chaoui et al., 2005; DeVore and Polanko, 2005.). Additional modalities such as tomographic ultrasound imaging (TUI), Inversion mode, high definition Doppler and B-flow have also been demonstrated to improve fetal heart imaging (Chaoui et al., 2004; Goncalves et al., 2004a, 2004b; DeVore and Polanko, 2005; Espinoza et al., 2005; Pooh and Korai, 2005; Volpe et al., 2007). The aim of the current study is to analyze the spectrum of the available modalities of 4D US in fetal cardiac anomalies and to find the appropriate tool for diagnosis and demonstration of each group of anomalies. METHODS All fetuses, in which cardiac anomaly was detected by 2D US, between 2004 and 2007 underwent 4D US evaluation. The study population consisted of all patients referred to our tertiary center due to various reasons and from screening program that is part of Copyright 2009 John Wiley & Sons, Ltd. Received: 28 vember 2008 Revised: 19 January 2009 Accepted: 5 February 2009 Published online: 1 April 2009

2 646 L. GINDES ET AL. prenatal US service of our institution. All examinations were performed by General Electric Voluson 730 Expert system (General Electric Medical Systems Kretztechnik, Zipf, Austria). Fetuses between 13 and 16 weeks gestation were examined by transvaginal volumetric transducer (RIC 5 9 MHz); while later in gestation volumetric abdominal transducer (RAB 4 8 MHz) was used. All patients provided a written informed consent for fetal examination and the 4D study was approved by our Hospital Institutional Review Board. Following initial diagnosis by 2D US, volume data sets of the fetal heart were acquired with cine 4D using STIC software (General Electric Kretztechnik, Zipf, Austria). Volumes were obtained by transverse and longitudinal sweeps over the fetal chest. All volumes were obtained first with STIC gray scale, followed by STIC Doppler (color, power or high definition) and with STIC B-flow. The STIC volumes were acquired with angle degrees during s. The preferred fetal position for adequate STIC was with the sternum toward the transducer and the spine away from the transducer. In cases that the ideal fetal position could not be obtained, the volumes were acquired in other positions, trying to avoid artifact as much as possible. The volumes were stored and subsequently analyzed (4D view software version 5.0, General Electric Medical Systems) on an offline personal computer by two authors (R.A and L.G), who are experienced in 3D/4D technology for the past 4 years. In order to optimize the volumes, manipulations were performed by positioning the rendering box, adjustment of the reference plane, use of an electronic scalpel to remove regions with artifacts and selecting the proper contrast curve. Volumes were displayed either as 2D pictures in the multiplanar mode (the 2D pictures are perpendicular to each other); TUI (the 2D pictures are parallel to each other, also called Multi-slice or i-slice by different manufacture); or as a 4D volume (render mode). Various rendering modes were used to display the volume data sets, and included: surface-gradient light mode, minimum mode, inversion mode, glass-body and color modes. The cardiac malformations were classified into five categories according to the sweep technique protocol of Two Dimension five short-axis views (Yagel et al., 2001) (Table 1): Group I, situs anomalies; Group II, anomalies of the 4 chamber view (4CV); Group III, anomalies of the outflow tracts; Group IV, arches anomalies and Group V, anomalies of the veins. A comparison between the different volume presentations in each group of anomalies was performed. Each presentation was determined as major contribution, if it could demonstrate the anomaly and add information over 2D picture; minor contribution if it demonstrated the anomaly and had the advantages of being digital volume; and not contributing if it failed to demonstrate the anomaly. An anatomic survey was performed on each fetus at the first visit and the follow-up examinations were scheduled based on the type of anomaly. In each case, multidisciplinary team including pediatric cardiologist (J.H), neonatologist and pediatric cardiac surgeon provided a comprehensive prenatal counseling. Clinical data during pregnancy and neonatal follow-up were obtained from computerized medical records. Table 1 One hundred and twenty-eight cardiac malformations classified into five categories according the 2D five short-axis views Situs Figure 1 =8 4CV Figure 2 3 =63 Outflow Figure 4 5 =27 Arches Figure 6 =21 Veins Figure 7 8 Dextrocardia 5 VSD 30 TGA 4 RAA 10 Interruption of 1 IVC+azygous continuation Situs inversus 2 Univentricular 1 PA/PS 7 DAD 1 LPSVC 8 Isolated levocardia heart 1 Rhabdomyoma 2 AI 1 Tubular aortic arch 1 FO aneurysm 1 Corrected 1 Interruption of 1 transposition aorta Atrioventricular 5 TOF 5 Aortic aneurysm 1 Canal Coronary sinus 4 Truncus arteriosus 4 Coarctation of 5 dilatation aorta Dilated 2 Double outlet RV 5 Hypoplastic Aortic 2 cardiomypathy arch Hypoplastic LH 5 TGA 4 RAA 10 Hypoplastic RH 6 Ventricular 2 asymmetry ASD 2 VSD, ventricular septal defect; FO, foramen ovale; ASD, atrial septal defect; TGA, transposition of great arteries; PA/PS, pulmonary atresia/pulmonic stenosis; AI, aortic insufficiency; TOF, tetralogy of Fallot; RAA, right aortic arch; DAD, ductus arteriosus dilatation; IVC, inferior vena cava; LPSVC, left persistence superior vena cava; RV, right ventricle. =9

3 3D/4D ULTRASOUND OF CARDIAC ANOMALIES 647 Figure 1 Determination of the Situs with minimal mode at 15 weeks fetus. (a) 3D volume of Situs solitus. The heart is on the left as well as the stomach (S). The gallbladder (GB) is on the right side and the umbilical vein (UV) is in the middle. (b) 3D volume of situs ambiguous. The heart is on the left, but the stomach (S) is on the right. The GB in the middle and persistent left umbilical vein is on the left side Figure 2 Four-chamber view acquired with STIC and color Doppler (at axial plane from anterior view, spine is posterior). (a) rmal heart at 13 weeks gestation. With normal inflow (red) into both ventricles. (b) Only one inflow is visible in a case of hypoplastic left heart at 14 weeks gestation Figure 3 Four-chamber view acquired with STIC gray scale (at axial plane from anterior view spine is posterior). (a) rmal fetus at 30 weeks gestation. (b) Fetus with atrioventricular canal at 15 weeks. There is a large gap between the ventricles and the atrias that include ventriculoseptal defect ( ), atrial septal defect (arrow) and in the middle the absent crux. LV, left ventricle; RV, right ventricle; S, interventricular septum; MV, mitral valve; C, crux; TV, tricuspid valve; LA, left atrium; RA, right atrium RESULTS During the study period, 81 patients with 128 fetal cardiac malformations were evaluated in our obstetrics US unit. An isolated cardiac malformation was detected in 50 fetuses (61.7%), while in 31 fetuses (38.3%) multiple cardiac malformations were demonstrated. Extra cardiac abnormalities were detected in 24 fetuses (29.6%). After counseling, 32 patients (39%) elected to terminate the pregnancy. One fetus with pulmonic atresia and no other detected anomalies died spontaneously in utero. Forty-eight patients continued the pregnancy. Neonatal

4 648 L. GINDES ET AL. Figure 4 STIC with B-flow for outflow tracts. (a) normal outflow tract at 23 weeks gestation demonstrating normal space alignment of the great vessels aorta (A) and pulmonary artery (P). (b) fetus at 26 weeks, showing only one vessel emerging from the ventricles (T) in a case of truncus arteriosus. RV, right ventricle; LV, left ventricle; T-truncus Figure 5 Outflow tract with inversion modes from fetuses at 24 weeks of gestation. (a) rmal relationship of fetal heart ventricles, arteries and SVC. The pulmonary artery is anterior and crossing the aorta. A, aorta; P, pulmonary artery; MPA, main pulmonary artery; Ao, aorta, Da, Ductus arteriosus; SVC, superior vena cava. (b) te the parallel direction of the great arteries in a case of transposition of great arteries. The aorta (A) is coming out the right ventricle (RV) and the pulmonary artery (P) emerging out of the left ventricle (LV). The right atrium (RA) and the left atrium (LA) are in the normal position mortality occurred in only one case with mitral insufficiency caused by three large left ventricular rhabdomyomas. additional anomalies were found. Seven cases (8%) were lost to follow-up. Multifetal pregnancies were in 10 cases: 8 twin pregnancies and 2 triplets. In all those cases only one fetus was diagnosed with cardiac malformation. The mean gestational age at diagnosis was 24 weeks (range 13 34). Early diagnoses at weeks gestation were performed in 14.8% (12/81) of the patients. The 128 anomalies which were classified into five categories are presented in Table 1 (Figures 1 to 8 and see Supporting Information). All anomalies were confirmed by pediatric cardiologist (J.H.) specified in fetal imaging. Karyotype was examined in 27 cases of which four had abnormal karyotype: two with partial trisomy 18, one of them with pulmonic atresia and the other one with atrioventricular canal. Another fetus had 4p deletion and ventriculoseptal defect (VSD), coarctation of aorta and vermian agenesis. One fetus had inversion of chromosome 2 and situs inversus, VSD, double outlet right ventricle (RV), pyelectasis, ventriculomegaly and large nuchal translucency. All these patients terminated their pregnancy. The gene for Velo-cardio-facial syndrome (22q11) was evaluated in 24 cases and was found normal in all cases. A comparison between the three modes of volume presentation: multiple sectional mode, render mode and TUI, and their contribution to diagnosis is summarized in Table 2. Render presentation and TUI were superior to the sectional plane for most of the anomalies. Table 3 demonstrates comparisons of the different rendering modes with the various malformation categories. The use of render mode with minimal transparency resulted in clearer demonstration of abnormal organs and vessels positions in situs abnormalities. Inversion mode, glass-body mode, color Doppler and B-flow were found to enhance particularly visualization of outflow

5 3D/4D ULTRASOUND OF CARDIAC ANOMALIES 649 Figure 6 Various demonstration of fetal Arches. (a) normal arches of fetus at 23 weeks, the volume was acquired with B-flow. te the aortic arch (A) behind the ductal arch (D). The right brachiocephalic artery (1), left common carotid artery (2) and the left subclavian artery (3) are coming out of the aortic arch. (b) B-flow demonstration of right aortic arch (RAA) in anterior view from a fetus at 27 weeks of gestation. The pulmonary artery (P) is in its normal position. The trachea (T) is encircled by loose vascular ring. (c) Render of power Doppler of double aortic arch in fetus at 22 weeks. The left aortic arch (LAA) is in the normal anatomic position near the pulmonic artery (PA). The right aortic arch (RAA) is on the right side encircling the trachea tracts anomalies. Inversion mode and B-flow were most informative for detecting arches anomalies. Systemic veins were best shown with B-flow. Table 4 summarizes a scoring system for the added value of the 3D 4D US parameters, which was already published in the literature in each group of anomaly. If the 3D character has more information than the 2D picture for demonstration of the heart anomaly category it gets one point. If it does not have any additional information over the 2D images it gets zero points. Summary of the total amount of points to each 3D character and to each anomaly category was calculated. Offline analysis of cardiac anomalies, spatial relationship orientation and entire course of vascular involvement were found to provide a major contribution to the diagnosis over the standard 2D US in at least three of the five categories. DISCUSSION 3D/4D US technology allows, for the first time, to acquire the fetal heart in motion and to investigate

6 650 L. GINDES ET AL. Figure 7 Lateral views from fetal left side showing veins anatomy (a) rmal fetal veins acquired with high definition Doppler flow STIC from fetus at 22 weeks. (b) Left persistence superior vena cava (LPSVC) ( ) acquired with inversion mode STIC from fetus at 23 weeks of gestation. te that this vessel does not exist in Figure A. The LPSVC drains into the coronary sinus at the left side of the heart and to the right atrium. A, aorta; DV, ductus venosus; HV, hepatic vein; IVC, inferior vena cava; SVC, superior vena cave; UV, umbilical vein;, LPSVC (left persistent SVC) Figure 8 Lateral views from fetal right side showing veins anatomy acquired with STIC and high definition Doppler. (a) rmal inferior vena cava (IVC) at 22 weeks of gestation and normal azygous vein (arrows) that travels cranially (in red) posterior and to the right of the aorta (A), confluent to the superior vena cave. (b) Absence of IVC with azygous continuation in fetus at 15 weeks of gestation. The IVC is absent. Large azygous vein (arrows) travels posterior and to the right of the aorta (A), drains the blood from the lower part of the body to the superior vena cava. UV, umbilical vein its volume offline. This is a new method of fetal heart examination and therefore merits special consideration. However the majority of published data regarding 3D/4D in the fetal heart describes the advantage of one or two modules in specific cardiac anomalies. Therefore, this is the first study that compares the different modalities in the whole spectrum of fetal heart malformations. Furthermore, our aim in this study is not to compare 2D US with 4D US, but to assess the usefulness of volume data sets in fetal echocardiography. We classified the cardiac anomalies according to five categories based on the anatomical landmarks by Yagel et al. (2001) rather than on the pathogenesis classification of Clark (1989). We have found that while all the five categories of anomalies can be demonstrated in multiple sectional planes, the arches and outflow tract anomalies are best visualized with Render mode and TUI. Navigations through the volume further enhance the visualization of the various parts of the heart and, therefore, are superior to the multiplanar presentation. These findings are in agreement with previous studies; Shih et al. reported on 21 cardiac anomalies evaluated by Render mode (Shih and Chen, 2005). They concluded that information derived by STIC was either

7 3D/4D ULTRASOUND OF CARDIAC ANOMALIES 651 Table 2 3D 4D US volume presentation tools and their contribution to the diagnosis of fetal cardiac anomalies Situs Four chambers view Outflow tract Arches Vein Sectional plane (multiplanar) Render mode TUI TUI, Tomographic ultrasound imaging. ++ major contribution, if it could demonstrated the anomaly and add information over 2D picture; + minor contribution if it demonstrated the anomaly and had the advantage of being digital volume. Table 3 Comparison of the different 3D 4D rendering modes in the demonstration of cardiac malformations Situs four chambers view Outflow tract Arches Vein Minimal mode Surface mode and Gradient light Inversion mode Glass body Color Doppler B-flow major contribution, if it demonstrated the anomaly and add information over 2D picture; + minor contribution, if it demonstrated the anomaly and had the advantage of being digital volume; not contributing, if it failed to demonstrate the anomaly. Table 4 Added value of 3D 4D US over conventional 2D imaging according to categories of cardiac malformations Situs Four chambers view Outflow tract Arches Vein Total Meticulous Offline examination (Nelson et al., 2001) Spatial relationship of the lesion (Espinoza et al., 2004) Entire course of vascular involvement (Sciaky-Tamir et al., 2006) (Achiron et al., 2008) Volume calculation (Messing et al., 2007) Total , If it could demonstrate the anomaly and add information over 2D picture. 0, added value over conventional 2D. beneficial (eight patients) or crucial (five patients) for establishing the diagnosis of congenital heart diseases. DeVore and Polanko (2005) reported that the combination of STIC and TUI could improve the demonstration of normal heart and anomalies such as tetralogy of Fallot (TOF), transposition of the great arteries and pulmonary stenosis which were easily diagnosed by these modes. Similar results were reported by Goncalves et al. (2006) who examined the feasibility of STIC and TUI in 195 patients. The authors also used B- flow and color Doppler which enabled them to confirm 16 cardiac anomalies by manipulating volumes with the above modules. The anomalies that were evaluated included coarctation of aorta, hypoplastic left heart, pulmonary atresia (PA) and transposition of the great arteries. Previous studies have shown that postprocessing manipulation of the 3D/4D rendering with different filters significantly emphasizes information obtained from the volume. In our study by using minimal mode we were able to illustrate the fetal situs projected on a coronal plane (Table 3). The hypo echoic structures within this mode such as stomach, gallbladder (GB) and vessels are accentuated and therefore their spatial arrangement is clearly evident. The use of TUI with minimal mode also enabled navigation through the volume thus enhancing the diagnosis of various situs abnormalities. The other visualization modes such as surface mode, gradient light and the combination between them were less effective in evaluating anomalies of 4CV, outflow tract, arches and veins. Inversion mode which is a novel approach to the 3D analysis of fluid-filled fetal structures added significantly to depict the entire course of the vessels in the categories of outflow tracts, arches and veins. The contribution of inversion mode to the diagnosis of fetal vascular anomalies (groups 4 and 5 in our study) has been evaluated by other investigators. Goncalves et al. (2004b) described the inversion mode in a normal 22 weeks fetus and in a fetus with transposition of the great arteries. Ghi et al. (2005) described this

8 652 L. GINDES ET AL. technique in a case report to demonstrate VSD. Espinoza et al. (2005) described the combination of STIC with inversion mode in the diagnosis of vein anomalies; they reported three patients with interrupted inferior vena cava (IVC) with azygos/hemiazygos continuation and concluded that this new mode contributed significantly to the diagnosis of veins anomalies because inversion rendering allowed the visualization of the entire course of the vascular tree of veins and arteries. Similarly we have found that the inversion mode had a major contribution in describing arterial and venous anomalies primarily due to the ability to trace the spatial course of the vessels involved in the anomaly. The potential of color Doppler STIC was evaluated by Chaoui et al. (2004) in normal and abnormal fetal hearts. In a prospective study of 35 normal fetuses and 27 fetuses with congenital heart defects, the three essential planes, namely the four-chamber view, the five-chamber view and the tree-vessel-trachea view, were reliably demonstrated in color in 90% of the cases in both groups. Similarly we have observed that color Doppler with STIC was very informative for describing outflow tract anomalies. Although we have not systematically examined the high definition Doppler, it is our impression that this technology has some advantage over the standard power and conventional color Doppler. High definition Doppler is more sensitive and therefore it traces more clearly the vascular tree, vein and arteries. B-flow imaging is a non-doppler technology that provides real-time imaging of blood flow during gray-scale sonography (Wachsberg, 2007). Although it has been introduced few years ago (Pooh, 2000), only recently its usefulness was reevaluated. By this technology, flow information is derived by digitally encoding the outgoing US beam, then decoding and filtering the returning beam so as to amplify echoes generated by the particulate constituents of flowing blood during real-time gray-scale sonography. B-flow reconstructed volumes are usually superior to the B-flow 2D pictures and can be achieved in a position that is not demonstrative in 2D. Since B- flow is a non-doppler technique, it is angle independent and lacks artifacts from vessel wall pulsation. We found the B-flow is superior to the other modalities in its ability to demonstrate outflow tract and arches anomalies. Our preliminary observation was recently supported by Volpe et al. (2007) in seven fetuses with total anomalous pulmonary venous connection. The authors concluded that the combination of STIC with B-Flow was able to facilitate identification of the anatomical features of this pathology and to supply additional information over that provided by 2D US. 3D/4D US has several significant advantages over standard 2D imaging. Using volumes, meticulous offline evaluation can be performed, the spatial relationships of the lesion are easily demonstrated, the entire course of blood vessels can be followed and volume calculation may provide more accurate information than the area inspected only by 2D. Furthermore the ability of moving the volume and inspecting it from different angles contributes to our understanding and facilitates the demonstration of the lesion to the parents and other members of the medical staff. We have shown that by quantifying the four already published advantages of 3D/4D US, significant contribution to prenatal diagnosis of cardiac defects can be achieved. It should be mentioned that 3D US and particularly the 4D STIC technology have some limitations. The limitations of acquisition of the STIC volume include the quality of the 2D picture and 2D artifacts like acoustic shadow, as well as the 3D artifacts and the Doppler artifacts. The 3D artifacts include movement artifact (maternal, fetal), and changes in heart rate, acoustic shadows, different lateral and axial resolution which lead to image distortion. The 3D render is a virtual image and therefore some information can be added or can be missed from the real data. Knowledge of those limitations is valuable while analyzing the volumes. SUMMARY Most of fetal cardiac defects can be assessed adequately by the information obtained with STIC gray scale, Doppler and B-flow. In the present study we were able to collect dataset volumes from 32 different heart anomalies, and classify them according to their spatial position and to apply for all of them different modules of 3D/4D technology. We have found that there is no single module appropriate for single group of malformations. Each category deserves a different module of visualization. In our opinion, the overall advantage of 4D US over 2D US in assessing fetal heart anomalies is the ability to gain all the data from different cardiac planes in one dataset digitalized file. This enables offline manipulation of the volume using different models, and provides better means for counseling and teaching within multidisciplinary team. ACKNOWLEDGMENTS We are grateful to Dr. Michal Berkenstadt, clinical geneticist, and Dr. Talia Litmanovitch, clinical geneticist, for their genetic counseling and follow-up with the patient. We also thank Misses Bella Shina for her contribution. REFERENCES Achiron R, Glaser J, Gelernter I, Hegesh J, Yagel S Extended fetal echocardiographic examination for detecting cardiac malformations in low risk pregnancies. BMJ 14: Achiron R, Weissman A, Rotstein Z, Hegesh J, Lipitz S, Mashiach S. 1994a. Transvaginal echocardiographic examination of the fetal heart between 13 and 15 weeks gestation in a low risk population. J Ultrasound Med 3: Achiron R, Rotstein Z, Lipitz S, Mashiach S, Hegesh J. 1994b. Firsttrimester diagnosis of fetal congenital heart disease by transvaginal ultrasonography. Obstet Gynecol 84: Achiron R, Gindes L, Weiss B, Lipitz S, Zalel Y Three- and four-dimensional ultrasound: A new method for evaluating fetal thoracic anomalies. Ultrasound Obstet Gynecol 32: Allan LD, Tynan MJ, Campbell S, Wilkinson JL, Anderson RH Echocardiographic and anatomical correlates in the fetus. Br Heart J 44:

9 3D/4D ULTRASOUND OF CARDIAC ANOMALIES 653 Carvalho JS Fetal heart scanning in the first trimester. Prenat Diagn 24: Chaoui R, Heling KS New developments in fetal heart scanning: Three- and four-dimensional fetal echocardiography. Semin fetal neonatal med 10: Chaoui R, Hoffmann J, Heling KS D and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 23: Clark EB Growth, morphogenesis and function: the dynamics of cardiovascular development. In Fetal, neonatal and Infant Heart Disease, Moller JM, Neal WA, Lick JA (eds). Appleton-Century- Crofts: New York; DeVore GR, Polanko B Tomographic ultrasound imaging of the fetal heart. A new technique for identifying normal and abnormal cardiac anatomy. J Ultrasound Med 24: DeVore GR, Falkensammer P, Sklansky MS, Platt LD Spatiotemporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 22: Espinoza J, Goncalves LF, Lee W, et al The use of the minimum projection mode in 4-dimensional examination of the fetal heart with spatiotemporal image correlation. J Ultrasound Med 23: Espinoza J, Goncalves LF, Lee W, Mazor M, Romero R A novel method to improve prenatal diagnosis of abnormal systemic venous connections using three- and four-dimensional. ultrasonography and inversion mode. Ultrasound Obstet Gynecol 25: Ferencz C, Rubin JD, McCarter RJ, et al Congenital heart disease: Prevalence at live birth. The Baltimore-Washington infant study. Am J Epidemiol 121: 31. Goncalves LF, Espinoza J, Lee W, Mazor M, Romero R. 2004a. Three and four-dimensional reconstruction of the aortic and ductal arches using inversion mode: a new rendering algorithm for visualization of fluid-filled anatomical structures. Ultrasound Obstet Gynecol 24: Goncalves LF, Romero R, Espinoza J, et al. 2004b. Four-dimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. J Ultrasound Med 23: Goncalves LF, Lee W, Espinoza J, Romero R Three and 4 dimensional ultrasound in obstetric practice. J Ultrasound Med 24: Goncalves LF, Espinoza J, Romero R, et al Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. J Perinat Med 34: Ghi T, Gera E, Segata M, Michelacci L, Pilu G, Pelusi G Inversion mode spatio-temporal image correlation (STIC) echocardiography in three-dimensional rendering of fetal ventricular septal defects. Ultrasound Obstet Gynecol 26: Hoffman JIE, Christianson R Congenital heart disease in a cohort of 19, 502 births with long-term follow-up. Am J Cardiol 42: 641. Messing B, Cohen SM, Valsky DV, et al Fetal cardiac ventricle volumetry in the second half of gestation assessed by 4D ultrasound using STIC combined with inversion mode. Ultrasound Obstet Gynecol. 30: Mitchell SC, Lorones SB, Berendes HW Congenital heart disease in 56,109 births: Incidence and natural history. Circulation 43: 323. Nelson TR, Pretorius DH, Sklansky M, Hagen-Ansert S Three- Dimensional echocardiographic evaluation of fetal heart anatomy and function: Acquisition, analysis and display. J Ultrasound Med 158: 1 9. Nelson TR, Pretorius DH, Lev-Toaff A, et al Feasibility of performing a virtual patient examination using three-dimensional ultrasonographic data acquired at remote locations. J Ultrasound Med 20: Pooh RK New application of B-Flow sonoangiography in perinatology. Ultrasound Obstet Gynecol 15: 163. Pooh RK, Korai A B-flow and B-flow spatio-temporal image correlation in visualizing fetal cardiac blood flow. Croat Med J 46: Sciaky-Tamir Y, Cohen SM, Hochner-Celnikier D, Valsky DV, Messing B, Yagel S Three-dimensional power Doppler (3DPD) ultrasound in the diagnosis and follow-up of fetal vascular anomalies. Am J Obstet Gynecol 194: Shih JC, Chen CP Spatio-temporal image correlation (STIC): innovative 3D/4D technique for illustrating unique and independent information and diagnosing complex congenital heart diseases. Croat Med J 46: Volpe P, Campobasso G, De Robertis V, et al Two- and four-dimensional echocardiography with B-flow imaging and spatiotemporal image correlation in prenatal diagnosis of isolated total anomalous pulmonary venous connection. Ultrasound Obstet Gynecol 30: Wachsberg RH B-flow imaging of the hepatic vasculature: correlation with color Doppler sonography. AJR Am J Roentgenol 188: W522 W533. Yagel S, Cohen SM, Achiron R Examination of the fetal heart by five short-axis views: a proposed screening method for comprehensive cardiac evaluation. Ultrasound Obstet Gynecol 17: Yagel S, Arbel R, Anteby EY, Revah D, Achiron R The three vessels and trachea scanning. Ultrasound Obstet Gynecol 20: 340.