Evaluation of normal fetal pulmonary veins from the early second trimester by enhanced-flow (e-flow) echocardiography

Transcription

1 Ultrasound Obstet Gynecol 211; 38: Published online 1 November 211 in Wiley Online Library (wileyonlinelibrary.com). DOI: 1.12/uog.8965 Evaluation of normal fetal pulmonary veins from the early second trimester by enhanced-flow (e-flow) echocardiography F.-Q. DONG*, Y.-H. ZHANG*, Z.-A. LI, Z.-Z. HOU, X.-J. HE* and Y.-Z. GUO* *Department of Functional Test, Children s Hospital of Hebei Province, Shijiazhuang, China; Department of Ultrasound Diagnosis, Peking An Zhen Hospital Affiliated to Capital Medical University, Beijing, China; Department of Computer Topography, Children s Hospital of Hebei Province, Shijiazhuang, China KEYWORDS: echocardiography; enhanced flow; fetal pulmonary vein; ; visualization rate ABSTRACT Objectives To explore the feasibility of using enhancedflow (e-flow) imaging technology to identify fetal pulmonary veins and establish gestational age-specific reference values at 12 4 weeks gestation. Methods The pulmonary venous internal diameter, peak systolic and diastolic flow velocities and visualization rate were analyzed in 332 normal fetuses at 12 4 weeks of gestation. Two-dimensional gray-scale (2D) ultrasound, color Doppler and e-flow imaging were used to detect the pulmonary veins in the four-chamber view by taking the lung as the penetration window. Results The pulmonary veins could be visualized as early as 12 weeks gestational age by e-flow imaging. The right and left pulmonary venous internal diameters and peak systolic and diastolic flow velocities increased with increasing gestational age. Between 12 and 4 gestational weeks, the internal diameter and both the systolic and diastolic flow velocities of the fetal right pulmonary vein were significantly larger than were those of the left vein (P <.5). Of 118 fetuses at gestational weeks, the visualization rate of the four pulmonary veins (left superior and inferior, right superior and inferior) was 5.9% (7/118) by 2D ultrasound, 41.5% (49/118) by color Doppler and 61.9% (73/118) by e- flow imaging. The visualization rate by e-flow imaging was significantly higher than that using the other two techniques (P <.1). Conclusion e-flow imaging is apparently a feasible and promising technology with which to identify the fetal pulmonary veins in the early stages of the second trimester. Copyright 211 ISUOG. Published by John Wiley & Sons, Ltd. INTRODUCTION Color Doppler has made it possible to study the normal human fetus non-invasively and to determine reference ranges for blood flow velocities at the level of the pulmonary arteries and veins 1 9. However, the sensitivity and resolution of color Doppler has limited its application in examining the fetal pulmonary venous system in early and mid-gestation. So far, there is no information available on normal values for pulmonary venous size and blood in fetuses before 17 gestational weeks 3,6,7. Enhanced-flow (e-flow) imaging is a novel technology developed by Aloka Co. Ltd. (Tokyo, Japan) for the imaging of blood flow dynamics. Unlike traditional color Doppler, it provides high spatial and temporal resolution by applying a composite pulse emission technique (allowing a shorter pulse length), which better illustrates the capillary perfusion and blood flow continuity. In addition, e-flow imaging is capable of distinguishing the blood flow from the surrounding tissue and offers high sensitivity to detect low blood 1. Because the fetal pulmonary veins are small in diameter and their blood flow is slow, it is difficult to examine them using color Doppler before 2 gestational weeks. In this study, we used e-flow imaging to examine normal fetal pulmonary veins from 12 weeks of gestation onwards, in order to establish reference values of fetal pulmonary venous size and peak systolic and diastolic flow velocities. These reference values will improve our understanding of venous return to the fetal heart in the normally developing pregnancy and thus help us to differentiate pathological conditions, such as congenital heart disease, lung hypoplasia and total anomalous pulmonary venous connection (TAPVC), during gestation Correspondence to: Dr F.-Q. Dong, Department of Functional Test, Children s Hospital of HeBei Province, Shijiazhuang, China, 531 ( dfq6355@hotmail.com) and Dr Z.-A. Li, Department of Ultrasound Diagnosis, Peking An Zhen Hospital affiliated to Capital Medical University, Beijing, China, 129 ( lizhian anzhen@yahoo.com.cn) Accepted: 2 February 211 Copyright 211 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER

2 e-flow imaging of pulmonary veins from early second trimester 653 METHODS Population A total of 48 pregnant women with normal singleton pregnancies between 12 and 4 (mean, 26) weeks gestation were recruited according to their visiting sequence and agreed to undergo a fetal echocardiographic examination. All women were Asian. Maternal age ranged from 21 to 39 (median, 27.2) years. We excluded 76 cases for the following reasons: maternal complications such as diabetes, hypertension, hyperlipidemia and amniotic fluid disorders, maternal medication during pregnancy and, retrospectively, premature delivery (< 28 weeks) and neonatal morbidity, as well as cases with fetal abnormalities or inappropriate size for gestational age and, retrospectively, congenital heart disease. e-flow recordings were thus collected from 332 (81%) of the pregnant women recruited into this study. The number of cases examined in each gestational week is given in Table 1. Echocardiography Examinations were performed by a single echocardiographer. Output settings and examination times were kept as low as reasonably achievable (ALARA principle). All participants were given a routine tri-segment diagnosis to rule out fetal cardiovascular anomalies, and underwent a detailed e-flow, two-dimensional gray-scale (2D) and color Doppler fetal echocardiographic examination using a Prosound-α1 color ultrasound machine equipped with an SSD-913 convex array probe (3. 6. MHz) (Aloka). Fetal pulmonary venous images were obtained from the four-chamber view by taking the lung as the penetration window, as described previously 6,13. The bilateral pulmonary venous drainage into the left atrium was visualized with e-flow, 2D and color Doppler echocardiography. Visualization of fetal pulmonary veins by e-flow imaging Due to fetal movement during echocardiography, it was difficult to distinguish the right and left pulmonary veins by fetal position alone. For the purposes of this study, we determined the location of the left and right pulmonary veins according to the position of the lungs, using as references central standard structures, such as the vertebral column, sternum and descending aorta. First, using calipers, we marked a line between the vertebral column and the sternum. The left lung was the one on the same side as the descending aorta and the right lung was the one on the same side as the superior vena cava. The ipsilateral superior and inferior pulmonary veins were differentiated according to whether they were adjacent to a pulmonary artery in the four-chamber view. The vein with an adjacent pulmonary artery was the superior pulmonary vein and the vein without an artery adjacent to it was the inferior pulmonary vein (Figure 1a). Usually, a tri-blood-vessel sign could be observed on the top of the cardiac atrium when the bilateral inferior pulmonary veins were displayed together with the descending aorta (Figure 1b). The pulmonary arteries and veins were distinguished according to the direction of blood flow and the waveform (Figure 1c). In about threequarters of the fetuses, four pulmonary veins could be detected, draining simultaneously into the left atrium, Table 1 Pulmonary venous internal diameters and peak systolic and diastolic flow velocities at 12 4 weeks gestation in fetuses in which all four veins were visualized Left pulmonary veins Right pulmonary veins GA (weeks) n/n Internal diameter (cm) Peak systolic Peak diastolic Internal diameter (cm) Peak systolic Peak diastolic /29.11 ± ± ± ± ± ± /2.122 ± ± ± ± ± ± /21.13 ± ± ± ± ± ± / ± ± ± ± ± ± / ± ± ± ± ± ± / ± ± ± ± ± ± / ± ± ± ± ± ± /33.22 ± ± ± ± ± ± / ± ± ± ± ± ± /2.233 ± ± ± ± ± ± / ± ± ± ± ±.4.2 ± / ± ± ± ± ± ± / ± ± ± ± ± ± /18.29 ± ± ± ± ± ± / ± ± ± ±.76***.231 ±.64**.171 ±.45* Data are given as mean ± SD. *P <.5, **P <.1 and ***P <.1 vs. corresponding value in left pulmonary veins. Average of superior and inferior veins. Overall mean. GA, gestational age; n/n, number of fetuses with all four veins visualized/total number of fetuses at that gestational age.

3 654 Dong et al. Figure 1 Visualization of fetal pulmonary veins in the four-chamber view by e-flow imaging in four 26-week fetuses. (a) Superior right pulmonary vein (SRPV) adjacent to right pulmonary artery (RPA). DA, ductus arteriosus. (b) Inferior left pulmonary vein (ILPV), inferior right pulmonary vein (IRPV) and descending aorta (DAO) are visible forming the tri-vessel sign at the top of the cardiac atrium. (c) SRPV adjacent to RPA, with waveform on the right. (d) Four pulmonary veins (superior left pulmonary vein (SLPV), SRPV, ILPV and IRPV) were seen draining into the left atrium (LA) simultaneously in three quarters of the fetuses studied; we referred to this as the crab sign. and they resembled crab legs; we therefore referred to this as the crab sign (Figure 1d). provide a mean value and these data were stored on computer 13. Fetal pulmonary venous flow spectral analysis Evaluation of the pulmonary venous waveform using e-flow imaging was performed on both right and left pulmonary veins. During the examination, the angle of insonation of the Doppler beam was kept at less than 15 and the pulsed Doppler sample volume width was.1.3 cm. Low-pass filters were adjusted to ensure recording of the lower-velocity signals. The pulmonary venous internal diameter and waveform S-wave (during ventricular systole), D-wave (during ventricular diastole) and A-wave (during atrial systole) were traced at the hilum of the lung. The peak amplitudes of S- and D-waves were recorded as peak systolic and peak diastolic velocities and their ratios (S/D ratios) at both left and right pulmonary veins were calculated. During each recording, at least three consecutive technically acceptable measurements were taken. The waveforms from which these measurements were taken were superimposed to Statistical analysis Data were expressed as mean ± SD. Visualization rates of the pulmonary veins by 2D, color Doppler and e-flow imaging were compared using McNemar s χ 2 test. The alpha level was set at.167 by Bonferroni correction. The relationships between the pulmonary venous parameters and gestational week were analyzed by regression. The various pulmonary venous parameters were compared using a paired t-test. P <.5 was considered statistically significant. RESULTS Determination of pulmonary vein internal diameter and Four pulmonary veins were detected by e-flow in 25 (75%) of the 332 fetuses. All fetal pulmonary venous internal diameter and peak systolic and diastolic flow

4 e-flow imaging of pulmonary veins from early second trimester 655 velocity data were obtained from these 25 fetuses. In these fetuses, both left and right pulmonary venous internal diameters and flow velocities indicated a gestational age-dependent increase (P <.1, regression analysis, Figure 2). Between 12 and 4 gestational weeks, the internal diameters and the peak systolic and diastolic flow velocities of the fetal right pulmonary vein were significantly larger than were those of the left one (paired t-test, P <.5, Table 1). Visualization rates of fetal pulmonary veins (a) Internal diameter (cm) Figure 3 shows an e-flow image of a fetal pulmonary vein visualized at the 12 th week of gestation. The visualization rates of pulmonary veins obtained by 2D, color Doppler and e-flow imaging are presented in Table 2. Subgrouping the cases according to gestational age, at weeks, the visualization rate measured by e-flow imaging (73/118, 61.9%) was significantly higher than that by both 2D ultrasound (7/118 (5.9%), P <.1, McNemar s test) and color Doppler (49/118 (41.5%), P <.1, McNemar s test, Figure 4). At 23 4 weeks, the visualization rate by e-flow (82.7%) was higher than that by 2D ultrasound (19.2%, P <.1, McNemar s test) but not than that by color Doppler (79.9%, P >.167, McNemar s test, Figure 4). Evaluation of fetal pulmonary venous waveform The pulmonary venous waveform was monitored using e-flow imaging on both left and right pulmonary veins. S- and D-waves were observed in every waveform but reversed A-wave was only seen in 3.3% of cases. The S/D ratio showed several patterns in the waveform. In most cases, it was > 1 (Figure 5a, left pulmonary vein: 9.4% of cases; right pulmonary vein: 92.4% of cases), in some cases it was equal to 1 (Figure 5b, left pulmonary vein: 8.4% of cases; right pulmonary vein: 6.% of cases) and in a few cases it was < 1 (Figure 5c, left pulmonary vein: 1.2% of cases; right pulmonary vein: 1.6% of cases). There was no significant difference in S/D ratio between left and right pulmonary veins in each of these three patterns (χ 2 -test). DISCUSSION Detection of the pulmonary veins at an early gestational age using 2D ultrasound or color Doppler is among the most significant of challenges faced by echocardiographers, due to the pulmonary vein s thin wall, small internal diameter and slow blood. So far, these traditional technologies have only provided reference ranges of fetal pulmonary venous after 17 weeks of gestation 3,4,6. Unlike color Doppler, e-flow imaging allows a short pulse length, resulting in greatly increased imaging sensitivity and temporal and spatial resolution. It has been found capable of detecting palpable stenoses as small as.5 mm in diameter 1 (b).5 Peak systolic (c).4 Peak diastolic Gestational weeks Gestational age (weeks) Gestational age (weeks) Figure 2 Relationships of fetal pulmonary venous internal diameter (a) and peak systolic (b) and diastolic (c) flow velocities with gestational age, as measured by e-flow imaging (n = 25). Each plot is an average of three consecutive measurements. Right pulmonary vein (RPV, ) and left pulmonary vein (LPV, ) are shown. (a) RPV regression equation: y =.76x +.132, r =.6726, P <.1; LPV regression equation: y =.69x +.192, r =.6947, P <.1. (b) RPV regression equation: y =.56x +.828, r =.5912, P <.1; LPV regression equation: y =.49x +.917, r =.5899, P <.1. (c)rpv regression equation: y =.39x +.675, r =.5933, P <.1; LPV regression equation: y =.35x +.733, r =.5428, P <.1.

5 656 Dong et al. Figure 3 Fetal pulmonary veins visualized by e-flow imaging at 12 weeks of gestation. (a) The fetal right pulmonary vein was visualized only by e-flow imaging (right panel) and not by two-dimensional gray-scale ultrasound (left panel). (b) e-flow imaging with the pulmonary vein s waveform in the same fetus as in (a). RLU, right lung. RPV, right pulmonary vein. Table 2 Visualization rates of pulmonary veins by two-dimensional gray-scale (2D) ultrasound examination, color Doppler and enhanced-flow (e-flow) imaging in 332 normal fetuses n (%) of fetuses with visualization of: Visualization method 4 veins 3 veins 2 veins 1 vein veins 2D ultrasound 48 (14.5) 54 (16.3) 77 (23.2) 59 (17.8) 94 (28.3) Color Doppler 22 (66.3) 27 (8.1) 42 (12.7) 26 (7.8) 17 (5.1) e-flow 25 (75.3) 19 (5.7) 38 (11.4) 16 (4.8) 9 (2.7) 1 Visualization rate (%) Gestational age (weeks) Figure 4 Visualization rate of fetal pulmonary veins at and 23 4 weeks of gestation using two-dimensional gray-scale ultrasound ( ), color Doppler ( ) and e-flow ( ) imaging. *P <.5 (McNemar s χ 2 -test). and in this study we found it to detect successfully the fetal pulmonary veins at 12 weeks of gestation. Additionally, even though the color Doppler settings, such as color contrast, filter and pulse repetition frequency, were adjusted to optimal conditions, the pulmonary venous images obtained by color Doppler in early and midgestation were not continuous and sometimes the efflux appeared with an increase in gain. In contrast, the images recorded by e-flow imaging at the same gestational age displayed continuous and engorged blood flow, which was easier to track. We found that the visualization Figure 5 Different patterns of pulmonary venous waveform, with ventricular systolic wave (S), ventricular diastolic wave (D) and atrial systolic wave (A) indicated: (a) S/D ratio > 1; (b) S/D ratio = 1; (c) S/D ratio < 1. rate of pulmonary veins by e-flow imaging was significantly higher than that by color Doppler at weeks of gestation. We therefore believe that e-flow imaging can be considered a feasible and promising technology with which to examine fetal pulmonary venous blood in the early stages of the second trimester.

6 e-flow imaging of pulmonary veins from early second trimester 657 Our results showed that the internal diameter and the of the right pulmonary veins were larger than were those of the left pulmonary veins at 12 4 weeks of gestation. This might correspond to the larger size of the right lung, requiring a greater blood supply to maintain development. Additionally, the pulmonary venous peak systolic and diastolic velocities measured by e-flow imaging showed the same age dependency as did those measured by color Doppler. However, the values measured by e-flow imaging were slightly larger compared with previously reported values measured by color Doppler 3,6,7,9. We consider there to be three possible reasons for this. First, all references to studies of pulmonary venous velocity are from at least 1 years ago; ultrasound technology has advanced significantly since then. Second, we performed our measurements in a different portion of the pulmonary vein; in previous reports, velocities were measured in the portion of the pulmonary vein that drains into the left atrium, while we measured the velocities in a portion of the hilum of the lung. Finally, differences between races cannot be ruled out. The e-flow imaging technique could be an effective tool for diagnosing fetal structural or functional pathology such as TAPVC, a severe congenital heart defect in the fetus. Without effective treatment, 8% of patients die in the first year 13. Thus, early diagnosis and treatment are of great importance in reducing the fetal mortality rate. Clinically, TAPVC can be ruled out when either pulmonary vein is visualized 13. Our study suggests that early diagnosis is possible, since the pulmonary veins can be examined by e-flow as early as 12 weeks of gestation. ACKNOWLEDGMENT We would like to thank Dr Hong-Wei Dong of the University of Tennessee, College of Medicine, for her help in the reading and revision of this manuscript. REFERENCES 1. Better DJ, Kaufman S, Allan LD. The normal pattern of pulmonary venous flow on pulsed. 2. Lenz F, Chaoui R. Reference ranges for Doppler-assessed pulmonary venous blood flow velocities and pulsatility indices in normal human fetuses. Prenat Diagn 22; 22: Brezinka C, Laudy AM, Ursem NTC, Hop WCJ, Wladimiroff JW. Fetal pulmonary venous flow into the left atrium relative to diastolic and systolic cardiac time intervals. Ultrasound Obstet Gynecol 1999; 13: Laudy JAM, Ursem NTC, Mulder PGH, Wladimiroff JW. Doppler velocimetry of normal human fetal venous intrapulmonary branches. Ultrasound Obstet Gynecol 1999; 13: Smallhorn JF, Freedon RM, Olley PM. Pulsed Doppler echocardiographic assessment of extraparenchymal pulmonary vein flow. J Am Coll Cardiol 1987; 9: Hong Y, Choi J. Doppler study on pulmonary venous flow in the human fetus. Fetal Diagn Ther 1999; 14: Laudy JAM, Huisman TWA, Ridder MAJ, Wladimiroff JW. Normal fetal pulmonary venous blood. Ultrasound Obstet Gynecol 1995; 6: Chaoui R, Taddei F, Rizzo G, Bast C, Lenz F, Bollmann R. Doppler echocardiography of the main stems of the pulmonary arteries in the normal human fetus. Ultrasound Obstet Gynecol 1998; 11: Hong YM, Choi JY. Pulmonary venous flow from fetal to neonatal period. Early Hum Dev 2; 57: 95 13; Doppler examination of the human fetus. J Am Soc Echocardiogr 1996; 9: Zhu W, Yang XH, Wang XY. Evaluation of hemodynamic changes of finger tip capillary vessels in patients with primary hypertension using E-Flow imaging technology. Chinese J Ultrasonographic Image 28; 17: Better DJ, Apfel HD, Zidere V, Allan LD. Pattern of pulmonary venous blood flow in the hypoplastic left heart syndrome in the fetus. Heart 1999; 81: Crowe DA, Allan LD. Patterns of pulmonary venous flow in the fetus with disease of the left heart. Cardiol Young 21; 11: Dong FQ, Zhao Z. Practical Ultrasonic Diagnostics of Congenital Heart Disease (1 st edn). The People s Military Surgeon Publishing Company: Bejing, Kanter KR. Surgical repair of total anomalous pulmonary venous connection. J Semin Thorac Cardiovasc Surg Pediatr Card Surg Ann 26; 9: Wu YF. Echocardiographic examination. In Ultrasonic Diagnostics of Fetal Cardiovascular System (1 st edn). The People s Health Publishing Company: Beijing, 24; Zhang GZ, Geng B. The regular examination methods in fetal echocardiography. In Practical Fetal Ultrasound Cardiograph (1 st edn). China Medicine Technology Publishing Company: Beijing, 24;