Imaging of placental vasculature using three-dimensional ultrasound and color power Doppler: a preliminary study

Transcription

1 Ultrasound Obstet Gynecol 1998;12:45 49 Imaging of placental vasculature using three-dimensional ultrasound and color power Doppler: a preliminary study D. H. Pretorius, T. R. Nelson, R. N. Baergen, E. Pai and C. Cantrell University of California San Diego, San Diego, California, USA Key words: PLACENTA, ULTRASOUND, THREE-DIMENSIONAL ULTRASOUND, DOPPLER STUDIES ABSTRACT Objective To assess placental vasculature using color power Doppler and three-dimensional ultrasound techniques. Design A prospective study was performed in patients to correlate visualization of placental vessels in vivo with known anatomy. Subjects Fourteen normal patients and one patient with intrauterine growth restriction were recruited to the study. Methods Vessels were assessed with regard to, first, the number of vessels seen within the placenta, second, the branching pattern of the vessels within the placenta, third, the number of vessels seen along the surface of the placenta, and, last, the number of vessels seen in the maternal circulation. Results Our results show that the placental vessels seen with this technique correlate well with known anatomy. A progressive increase in the number of intraplacental vessels and the number of vascular branches observed was seen with increasing gestational age. Volume data review using three orthogonal planar images had two distinct advantages. First, they could be obtained from orientations not possible using two-dimensional ultrasound alone, and, second, they could be viewed in conjunction with volumerendered images to allow for referencing and identification of specific vessels. Volume-rendered images were valuable in allowing the observer to acquire an improved overall understanding of placental anatomy. They also assisted the observer in following the continuity of vessels as they wrapped around and twisted through three-dimensional space. Stereo viewing was helpful in distinguishing overlapping vessels. Conclusions Our study showed that sonographic volume imaging combined with color power Doppler imaging methods allowed for individual vessels in the placenta to be identified, both in the fetal and maternal circulations. INTRODUCTION Color power Doppler sonography and three-dimensional sonography offer promise in evaluating placental vascularity in vivo. Power Doppler sonography is a relatively recent development in ultrasound technology and differs from conventional color Doppler in its greater sensitivity in detecting lower rates of blood flow and in its angle independence. Noise is thought to be less distracting with power Doppler in that it can be displayed as an homogeneous background color rather than as different colors, as seen with conventional color Doppler 1. Its enhanced sensitivity makes it possible to examine blood flow in the placenta. Three-dimensional sonography has been used to visualize various aspects of anatomy, such as the fetal heart 2, spine 3,4, thorax 3,4, and face 3,5, as well as the adult heart 6, abdomen/pelvis 7 and intra- and extracranial vasculature 8,9. The potential exists, using three-dimensional ultrasound methods, to identify placental insufficiency or abruption via imaging techniques prior to the development of significant fetal distress. The goal of this work was to identify specific blood vessels of the maternal and fetal circulations of the placenta in vivo using three-dimensional ultrasound in combination with color power Doppler to correlate with known vascular anatomy identified from anatomic texts. MATERIALS AND METHODS Fourteen normal pregnant volunteers and one patient with intrauterine growth restriction (IUGR) were scanned after informed consent was obtained from the Institutional Review Board. A conventional ultrasound unit (Acuson Correspondence: Dr D. H. Pretorius, UCSD School of Medicine, 9500 Gilman Drive, La Jolla, CA , USA ORIGINAL PAPER 45 Received Revised Accepted

2 Figure 1 Schematic illustration of human placenta. a, umbilical arteries; b, umbilical vein; c, surface vessel; d, main stem vessel; e, stem vessel; f, tertiary stem vessel; g, arcuate artery; h, radial artery; i, spiral artery; j, maternal veins; k, maternal arteries; 1, placental surface; 2, intervillous space; 3, endometrium (decidua); 4, myometrium. (Adapted from Ramsey and Donner 10 ) 128 XP10, Acuson, Mountain View, CA, USA) and transducer incorporating an electromagnetic position sensor (Polhemus, Colchester, VT, USA) were used to obtain threedimensional ultrasound volumes. The gestational age of the fetuses ranged from 13 to 35 weeks. Volume data acquisitions were taken by sweeping the transducer along a linear path on the maternal abdomen. Each acquisition time was approximately 10 s. Several acquisitions were performed on each patient, and gray-scale and color power Doppler images were acquired into a graphics workstation (Sparc- 10, Sun Microsystems, Mountain View, CA, USA) for processing, rendering and interactive display. Images were evaluated for optimal amount of information and analyzed in the original two-dimensional planar images, threedimensional planar images and volume-rendered views. Rendered images also were viewed in stereo through the use of liquid crystal display glasses (StereoGraphics Corporation, San Rafael, CA, USA). Placental vasculature was assessed for the location and continuity of blood vessels on both the fetal and maternal sides of the circulation. Individual vessels were identified by correlating acquired images with known placental vascular anatomy (Figures 1 and 2). The umbilical cord, normally containing two umbilical arteries and one umbilical vein, inserts into the placenta; the umbilical arteries give rise to the fetal circulation of the placenta via vessels that extend along the surface (Figure 3). These surface vessels perforate the chorionic plate and give rise to a hierarchy of intraplacental vessels: from largest to smallest, the main stem, stem, tertiary stem, cotyledonary, and villous branches. On the uterine side of the placental circulation, a similar hierarchy of maternal vessels exists: iliac, uterine, arcuate, radial, and spiral arteries (Figure 1). Placental location was recorded as anterior, posterior, or lateral. The amount of placenta scanned to produce a volume was recorded as a percentage of the total placenta visualized on real-time scanning; an attempt was made to scan a maximal amount of placenta including maternal vessels and surface vessels in all volumes. Evaluation of intraplacental vessels was determined by several methods. Figure 2 Radiograph of placental vasculature. Normal placental specimen injected with radio-opaque contrast, demonstrating vascular anatomy First, the total number of fetal intraplacental vessels observed in a volume was recorded as rare (0 2), some (3 4) or many (5 or more). Second, the maximal number of vessels to fit in a cm cube was recorded; this was assessed from a single plane scrolled through the 3-cm volume. Third, the deepest level of branching of intraplacental vessels was noted: a main stem vessel was identified as one branch, a stem vessel as two, and a tertiary stem vessel as three (Figure 4). Fourth, the number of surface vessels was recorded as rare (0 1), some (2 3), or many (4 or more). Fifth, the number of maternal vessels was recorded as rare, some, or many. An exact number was not recorded since there was often confluence of vessels. RESULTS Outcomes from the 14 normal patients showed normal birth weights. One baby was delivered at 33 weeks and had 46 Ultrasound in Obstetrics and Gynecology

3 a birth weight of 1055 g, consistent with intrauterine growth restriction. The placenta was anterior in nine cases, posterior in three, and lateral in three. The percentage of placenta scanned ranged from 30 to 100%. Power Doppler images allowed visualization of many levels of fetal and maternal vessels. Main stem vessels were observed in all cases. Stem vessels were observed in one-third of the cases. The smallest vessels to be seen in this study were the tertiary stem vessels (Figure 4); cotyledonary and villous branches were not imaged. A progressive increase in the number of intraplacental vessels was seen with increasing gestational age (Figure 5). A slight increase in the number of vascular branches was also observed with increasing gestational age (Figure 6). Attempts were made further to assess changes in the density of vessels by counting the maximum number of vessels in any given cm cube; these results were inconclusive in elucidating a trend. Numerous surface vessels were noted in the majority of cases; however, no predictable change was identified with increasing gestational age. Placental vasculature in the one fetus with intrauterine growth restriction did not show any significant difference when compared with that of normal patients of similar gestational age. A velamentous cord insertion was easily demonstrated with stereo viewing of the cord which was not seen with conventional two-dimensional imaging. Maternal vessels were in abundance in almost all cases. In fact, their plentiful numbers usually made it impossible positively to identify individual arteries. The smallest maternal vessels to be seen were the spiral arteries (1 3 mm). These characteristically tortuous vessels can be Figure 3 Three-dimensional planar power Doppler ultrasound image of placental cord insertion. The umbilical cord of a 28-week-old gestation is shown entering the anterior placenta in a normal, central location. Surface vessels diverge from the site of insertion Figure 4 Three-dimensional planar power Doppler ultrasound image showing branching of intraplacental vessels. Main stem vessels (long arrow) penetrate the intervillous space and give rise to stem vessels (short arrow) which branch to tertiary stem vessels (arrowhead) in this 34-week-old gestation Many 3 Average number of vessels Some Rare Average number of branches Gestational age (weeks) Figure 5 Intraplacental vessels on three-dimensional planar imaging seen as a function of gestational age 0 21 n = n = n = 3 Gestational age (weeks) 33 n = 5 Figure 6 Branching on three-dimensional planar imaging seen as a function of gestational age Ultrasound in Obstetrics and Gynecology 47

4 Figure 7 Three-dimensional planar power Doppler ultrasound image of maternal vessels. Power Doppler imaging reveals many vessels on the maternal side of a 19-week gestation. The less echogenic area directly below the placenta is the decidua, through which the spiral arteries (arrow) pass to supply blood to the placenta (P) Figure 9 Three-dimensional volume-rendered power Doppler ultrasound image of placental vessels. Cord insertion (arrow) is seen entering placenta and the maternal vessels (short arrows) are shown in a 28-week gestation. The surface of the placenta is encircled with a gray line Figure 8 Three-dimensional volume-rendered power Doppler ultrasound image of placental vessels. Global view of the placental vascular anatomy showing cord insertion (short arrow), placental surface vessels (arrowhead), and arcuate vessels (long arrow) in a 19-week-old gestation. Continuity of the arcuate vessels was seen only on three-dimensional rendered images recognized as they pass through the less echogenic substance of the decidua (Figure 7). Increases in the number of intraplacental vessels and their branching were demonstrated on both the original two-dimensional planar images and three-dimensional planar views. Little difference was found between the image quality in comparing the original two-dimensional planar versus three-dimensional planar views. Threedimensional planar images had two distinct advantages. First, three- dimensional planar images could be obtained from orientations not achievable using two-dimensional ultrasound alone, and second, three-dimensional planar images could be viewed in conjunction with the volumerendered images to allow for referencing and identification of specific vessels by using a cursor, which was present in three orthogonal planar images as well as the rendered image. Volume-rendered power Doppler images were critical in allowing the observer to acquire an improved overall understanding of placental anatomy (Figures 8 and 9). Volume-rendered images assisted the observer in following the continuity of vessels as they wrapped around and twisted through three-dimensional space. It is extremely difficult to appreciate the same spatial relationships on two-dimensional ultrasound as vessels break in or out of the plane of view. Volume-rendered images facilitated visualization of umbilical, surface, main stem, stem, tertiary stem and maternal vessels. In a few cases, the rendering process caused the smallest vessels to become either lost or fused with nearby vessels to create less distinct images. As a result, it was sometimes more difficult to find stem vessels on volumerendered views than three-dimensional planar views. Tertiary stem vessels were observed on three-dimensional planar images but not on volume-rendered images. Inspection of rendered images in stereo was extremely useful for distinguishing between vessels that were overlapping. When stereo viewing was coupled with interactive rotation of three-dimensional images, perception of depth cues was greatly enhanced. Stereo glasses also allowed the observer to differentiate better vessels which only arched along the surface of the placenta from those which were actually intraplacental. In one case, the volume-rendered display was visualized with stereo viewing using liquid crystal display glasses and it permitted the clinician positively to diagnose a velamentous cord insertion in which images from the conventional two-dimensional ultrasound study were inconclusive. 48 Ultrasound in Obstetrics and Gynecology

5 DISCUSSION The placenta is a vital, highly vascular organ which functions to transfer nutrients and blood from the mother to the fetus. The maternal and fetal circulations meet in a complex, hierarchical vascular pattern which has been studied thoroughly in vitro. Color Doppler techniques offer the opportunity to study placental blood in vivo. Color power Doppler imaging allows the clinician to identify smaller intraplacental vessels than with conventional color Doppler imaging due to its increased sensitivity to low flow. Power Doppler imaging also provides greater clarity of vessels as they penetrate deep into the intervillous space. It is important for caregivers to use standard terminology in assessing medical information. Vessels in the placenta should be identified as part of the fetal or maternal circulations. We suggest that the vessels should be labelled according to Ramsey and Donner 10 (Figure 1). Images obtained from volume data acquisition in this study allowed the identification of specific surface, main stem, stem, and tertiary stem vessels of the fetal circulation. Although the vessels from the maternal circulation were often harder to differentiate, we were able to identify the uterine, arcuate, and spiral arteries in some cases. This preliminary study, those of Bude and colleagues 11 and work from our laboratory 12 suggest that power Doppler imaging can demonstrate the progressive increase in placental vascularization hypothesized to occur with advancing gestational age during pregnancy. Additional studies are needed to confirm these results. This growth and expansion of fetal vessels result in steady increases in the villous volume and surface area, as well as fetal vessel lumina volume throughout pregnancy 13. As the fetus gradually requires increasing amounts of nourishment, gas exchange, etc., there is a concurrent increase in intraplacental vessels to support its needs for growth. While the maternal vessels also undergo a process of growth and change, maternal blood flow into the intervillous space is already established by the 13th week of pregnancy 14. From a technical standpoint, the acquisition of highquality power Doppler images was challenging. First, operator variables and alterations in gain, power, scale, filter, and persistence settings affected the sensitivity to blood flow available in any given dataset. Optimization of image quality often involved fine-tuning the various parameters in a process that was subjective and variable. Also, sensitivity to blood flow detection varies with different equipment/ vendors. As new technology is being developed, increased sensitivity to flow is expected. Second, movement of the transducer across the placenta, required to obtain a volume acquisition, often leads to flash color artifacts within the volume. It is possible to erase these color flashes with editing tools 15. Third, anterior placentae were easier to evaluate than posterior placentae due to decreased penetration and acoustic shadowing from the fetus. In the future, the use of intravenous contrast agents may further assist vessel visualization of the maternal circulation. In conclusion, our study showed that sonographic volume imaging combined with color power Doppler imaging methods allowed individual vessels in the placenta to be identified in both the fetal and maternal circulations. We found a progressive increase in the number of intraplacental vessels and the number of vascular branches observed with increasing gestational age. Volume-rendered images were valuable in assisting the observer to acquire an improved overall understanding of placental anatomy, specifically in following the continuity of vessels as they wrapped around and twisted through three-dimensional space. Stereo viewing was helpful in distinguishing overlapping vessels. We believe three-dimensional ultrasound and power Doppler imaging techniques have the potential to be used in the detection of vascular diseases which affect the placenta, including arterial and venous thromboses, maternal floor infarction, placental insufficiency, twin transfusion syndrome and abruption. REFERENCES 1. Macsweeney JE, Cosgrove DO, Arenson J. Doppler energy (power) mode ultrasound. Clin Radiol 1996;51: 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 1995;5: Merz E, Bahlmann F, Weber G, Macchiella D. Threedimensional ultrasonography in prenatal diagnosis. J Perinat Med 1995;23: Nelson TR, Pretorius DH. Visualization of the fetal thoracic skeleton with three-dimensional sonography: a preliminary report. Am J Roentgenol 1995;164: Pretorius DH, Nelson TR. Fetal face visualization using threedimensional ultrasonography. J Ultrasound Med 1995;14: Geiser EA, Ariet M, Conetta DA, Lupkiewicz SM, Christie LG, Conti CR. Dynamic three-dimensional echocardiographic reconstruction of the intact human left ventricle: technique and initial observations in patients. Am Heart J 1982;103: Jurkovic D, Geipel A, Gruboeck K, Jauniaux E, Natucci M, Campbell S. Three-dimensional ultrasound for the assessment of uterine anatomy and detection of congenital anomalies: a comparison with hysterosalpingography and two-dimensional sonography. Ultrasound Obstet Gynecol 1995;5: Pretorius DH, Nelson TR, Jaffe JS. 3-D Sonographic analysis based on color flow doppler and gray scale image data: a preliminary report. J Ultrasound Med 1992;11: Lyden PD, Nelson TR. Visualization of the cerebral circulation using 3D transcranial power Doppler ultrasound imaging. J Neuroimaging 1997;7: Ramsey EM, Donner MW. Placental Vasculature and Circulation. Philadelphia: Saunders, 1980: Bude RO, Van de Ven C, Rubin JM. Evaluation of the normal placental vasculature with power doppler sonography. J Ultrasound Med 1995;14:S Pretorius DH, Nelson TR, Baergen R, Cantrell C. Threedimensional ultrasound and power Doppler imaging of placental vasculature. Acad Radiol 1995;2: Benirschke K, Kaufmann P. Pathology of the Human Placenta. New York: Springer-Verlag, 1990;17: Benirschke K. Obstetrically important lesions of the umbilical cord. J Reprod Med 1994;39: Nelson TR, Davidson TE, Pretorius DH. Interactive electronic scalpel for extraction of organs from 3DUS data. Radiology 1995;197: Boyd JD, Hamilton WJ. The Human Placenta. Cambridge: Heffer & Sons, 1970:225 7 Ultrasound in Obstetrics and Gynecology 49