Three-Dimensional Ultrasonography of the Fetal Vermis at 18 to 26 Weeks Gestation

Transcription

1 Article Three-Dimensional Ultrasonography of the Fetal Vermis at 18 to 26 Weeks Gestation Time of Appearance of the Primary Fissure Yaron Zalel, MD, Simcha Yagel, MD, Reuven Achiron, MD, Zvi Kivilevich, MD, Liat Gindes, MD Objective. The purpose of this study was to establish the normality of the fetal vermis, ie, the time of appearance of the primary fissure, as well as its measurements between 18 and 26 weeks gestation, using 3-dimensional (3D) ultrasonography. Methods. A prospective cross-sectional study of normal singleton pregnancies was conducted. Examinations were performed with high-resolution transabdominal ultrasonography using the axial plane in 173 fetuses between 18 and 26 weeks gestation. Postprocessing measurements of the fetal vermis were done with 4-dimensional software using static volume contrast imaging and tomographic ultrasound imaging in the C-plane. Detection of the primary fissure was evaluated in all cases, and the time of appearance was documented. Results. Adequate vermis measurements were obtained in 173 fetuses. Vermian length as a function of gestational age was expressed by regression equations, and the correlation coefficients were found to be highly statistically significant (P <.001). The normal mean ± 2 SD for each gestational week was defined. The primary fissure was observed at 24 weeks gestation in all cases, at 22 weeks in 94% of cases, and as early as 18 weeks in 40%. Conclusions. This 3D study documents the appearance of the primary fissure and presents the normal range of vermian measurements, confirming normal development of the fetal vermis starting as early as 18 weeks gestation. It also shows an easy method for visualizing the vermis with 3D ultrasonography at every gestational week regardless of fetal presentation. Key words: primary fissure; 3-dimensional ultrasonography; vermis. Abbreviations MRI, magnetic resonance imaging; 3D, 3-dimensional; 3DUS, 3-dimensional ultrasonography; TUI, tomographic ultrasound imaging; 2D, 2-dimensional; 2DUS, 2-dimensional ultrasonography; VCI, volume contrast imaging Received August 27, 2008, from the Department of Obstetrics and Gynecology, Chaim Sheba Medical Center, Tel-Hashomer (affiliated with the Sackler School of Medicine, Tel-Aviv University), Ramat Gan, Israel (Y.Z., R.A., Z.K., L.G.); and Department of Obstetrics and Gynecology, Hadassah Hebrew University Medical Center, Jerusalem, Israel (S.Y.). Revision requested September 5, Revised manuscript accepted for publication September 25, Address correspondence to Yaron Zalel, MD, Department of Obstetrics and Gynecology, Chaim Sheba Medical Center, Tel-Hashomer, Ramat Gan, Israel. zalel_y@netvision.net.il A pplying the search term vermis in the Online Mendelian Inheritance in Man database revealed 103 syndromes. Some of them are more familiar, such as Joubert and Dandy- Walker, whereas others are less common, such as atrophy of the cerebellar vermis, which occurs in spinocerebellar ataxia (Online Mendelian Inheritance in Man No ). The pediatric literature reports reduced volumes of the anterior lobe in cases of fetal alcohol syndrome. 1 Evaluation of the cerebellum and vermis is performed during obstetric ultrasonographic examinations according to American College of Obstetricians and Gynecologists guidelines 2 as well as in our department. 3 The entire fetal vermis can be visualized by ultra by the American Institute of Ultrasound in Medicine J Ultrasound Med 2009; 28: /09/$3.50

2 3D Ultrasonography of Fetal Vermis at 18 to 26 Weeks Gestation sonography from 18 weeks gestation. 3,4 The fetal vermis can be shown in axial, coronal, and midsagittal sections. The midsagittal plane enables visualization of all of the vermian lobules, the fourth ventricle, the pons, and the corpus callosum. In the coronal and axial planes, most parts of the vermis cannot be detected. It may take time and effort to show the midsagittal view. Axial planes are routinely and easily obtained for evaluation of the fetal brain, especially at mid gestation, when most fetuses are in a horizontal lie. With the fetus in a breech position, it is sometimes difficult to obtain the sagittal aspect of the brain without shadowing. If the fetus is in a vertex presentation, usually only the transvaginal approach will provide a good sagittal plane. Whenever vermian abnormalities are suspected, an artifact should be excluded, and a detailed study of the fetal brain should be performed. The primary fissure is a deep fissure seen on a midsagittal section through the cerebellum, separating the anterior lobe from the posterior lobe. It is the deepest fissure seen in midsagittal sections of the cerebellum. Identifying the primary fissure suggests normal development of the vermis, although it does not eliminate other conditions. 5 Ultrasonography can show the anatomy of the fetal posterior fossa at 15 to 40 weeks gestation. 6 The cerebellum, including the cerebellar hemispheres, the superior and inferior vermis, the fourth ventricle, and the cisterna magna, can be shown easily. Using a transvaginal transducer, Malinger et al 4 examined the appearance and development of the fetal cerebellar vermis in 101 fetuses between 21 and 39 weeks gestation. They successfully visualized the vermis in 96% of the cases. Measurements were performed in 92% of the fetuses and showed linear growth of the vermis during gestation. The primary fissure was observed for the first time at 27 weeks gestation and was uniformly present in all fetuses by 30 weeks. Magnetic resonance imaging (MRI) studies of the fetal posterior fossa have shown the primary fissure from 25 5 to as late as 30 7 weeks gestation. There is a need, however, for earlier detection of a malformed vermis that will allow for genetic investigation, parental counseling, and further management. Therefore, the aims of this study were to assess, as early as possible, the normality of the fetal vermis, ie, detection of the vermian primary fissure, with 3-dimensional ultrasonography (3DUS) and to create a nomogram for fetal vermis measurements during gestation. We also aimed to show, in light of the paucity of literature on evaluation of the fetal vermis by 3DUS, that assessing these features is easily accomplished by using axial planes with the 3-dimensional (3D) technique. Materials and Methods A cross-sectional prospective study was conducted to establish the length of the fetal vermis during normal gestation and to determine the time of appearance of the primary fissure. Gestational age was determined by the last menstrual period and confirmed by a first-trimester ultrasonographic examination. Only singleton pregnancies of healthy patients were included in the study. Patients were excluded from the study if a fetal abnormality was found, impaired fetal growth was suspected or confirmed, or a maternal illness or complication of pregnancy (ie, preeclampsia or diabetes) was present. This study involved only routine standard-ofcare procedures at our institution and the Institutional Review Board of the Chaim Sheba Medical Center approved the study. Technique Screening of the fetal head was done in the axial plane, in which the cerebellum is visualized, and a scanning volume was acquired and saved. All examinations were done with Voluson 730 Expert or Pro ultrasound machines (GE Healthcare, Kretz Ultrasound, Zipf, Austria) and 4- to 8-MHz transabdominal transducers. The volumes were taken during conventional scanning while examining the fetal head. All scans were performed by a single examiner (Y.Z.). In this study, acquisitions were performed during fetal rest and absence of fetal movement. The standard volume sweep angle was 55. Offline analysis and postprocessing were performed with 4D View software (GE Healthcare, Kretz Ultrasound). The technique was as follows: 1. A volume of the fetal brain was taken in the axial plane. 2 J Ultrasound Med 2009; 28:1 8

3 Zalel et al 2. The reference dot was placed in the middle of the vermis in the axial plane on the A reference image (Figure 1, top left). 3. The skull was rotated around the z-axis until the midline was horizontal. In the B reference image (Figure 1, top right), the reference dot was placed in the middle of the vermis, and the skull was rotated around the z-axis until the midline was horizontal. 4. The image was magnified to the maximum ( 4; Figure 2A). At this magnification, caliper resolution was 0.1 mm. (Compare with Figure 2B, the vermis and primary fissure on 2-dimensional ultrasonography [2DUS].) 5. Visualization was preferably performed by static volume contrast imaging (VCI) with the slice thickness set to 3 mm and by tomographic ultrasound imaging (TUI) in the C-plane with a minimal slice distance of 0.5 mm (Figure 3). The middle image was the middle sagittal plane of the cerebellar vermis. All of the other pictures were lateral sagittal views of the cerebellar vermis. The reference image assisted with orientation in space while establishing the anterior and posterior directions. The vermian curvature was toward the vertex of the fetal head. If in the A-plane, the posterior aspect was on the right side, and the fetal vertex was upward. 6. Measurement of the maximal length of the vermis was done in the middle image (Figure 3B). 7. The primary fissure was detected on TUI and the rendered images. The primary fissure was detectable as a groove in the outer upper part of the vermis, dividing the vermis into an upper third and a lower two-thirds (Figures 3C and 4). At the time of measurement, the gestational age was not displayed on the screen. Only after measuring the largest length and determining visualization of the primary fissure was the gestational age confirmed. Results Volumes adequate for vermis measurements were acquired in 173 fetuses ranging in gestational age from 18 to 26 weeks (mean ± SD, 23.1 ± 1.78 weeks). We were able to measure the vermis length in all cases except 1 at 24 weeks gestation, in which the borders were blurred. The examination added 1 minute to the scan, and the learning curve was short; at the beginning of the study, 2 acquisitions were needed to achieve optimal results, and shortly after, only 1 acquisition was needed. Intraobserver variability, as determined by coefficients of variation, was 5% or less. Visualization of the primary fissure varied according to gestational age. The primary fissure was observed at 24 weeks gestation in all cases, at 22 weeks in 94% of the cases (16 of 17), at 20 weeks in 64% (7 of 11), and as early as 18 weeks in 40% (2 of 5) (Table 1). Vermian length as a function of gestational age was analyzed with a linear regression model, and the correlation coefficient (r 2 = 0.694) was highly statistically significant (P <.001). Figure 5 shows the linear growth of the vermis from 18 to 26 weeks gestation. The mean ± 2 SD and 5th Figure 1. Sectional plane of the fetal head at the level of the vermis. The gestational age is 26 weeks. The reference dot is placed in the middle of the vermis in the axial plane in the reference image (top left). The image is rotated around the z-axis until the midline is horizontal. The reference dot is then placed in the B reference image (top right) in the middle of the vermis, and the image is rotated around the z-axis until the midline is horizontal. Care must be taken to ensure that the dot is in the middle of the vermis in the C reference image (bottom left) as well. Statistical Analysis Data were analyzed (Excel 2003, Microsoft Corporation, Redmond, WA; SPSS version 15.0 for Windows, SPSS Inc, Chicago, IL) by linear regression with least squares, including the intercept in the equation, to establish a model of vermian length as a function of gestational age according to the method of Royston and Wright. 8 J Ultrasound Med 2009; 28:1 8 3

4 3D Ultrasonography of Fetal Vermis at 18 to 26 Weeks Gestation through 95th percentiles for each gestational week were defined. The maximal vermian length by gestational age is shown in Table 2. Discussion Imaging of the posterior fossa has become an integral part of screening for fetal anomalies. 9,10 The entire vermis and cerebellum are fully formed by the end of the 15th to 16th week, 11,12 but only at 18 weeks gestation is the fourth ventricle completely covered by the cerebellar hemispheres and vermis and the communication between the fourth ventricle and the cisterna magna also covered. These embryologic studies are supported by ultrasonographic studies. 13 Comparing imaging of the fetal posterior fossa by ultrasonography and MRI, Babcook et al 14 concluded that the ultrasonographic appearance of normal cerebellar development can resemble abnormalities early in the second trimester. On the basis of ultrasonographic studies, 13,15 the diagnosis of vermian agenesis (especially partial agenesis) cannot be made before 18 weeks gestation; therefore, we began vermis measurements at this time. The normal dimensions of the vermis are extremely important in evaluating posterior fossa abnormalities because one of the key features of these malformations is a vermian defect. In addition, a high proportion of disagreement was found between prenatal ultrasonographic diagnosis of the Dandy-Walker malformation or variant and autopsy findings, 16 emphasizing the need for accurate ultrasonographic evaluation of the cerebellar vermis. At present, in standard antenatal evaluations, measurements of the cisterna magna are made in the axial plane, which does not provide a window for clearly assessing the integrity of the vermis. This can lead to overestimation or underestimation of the defect, if any. Indeed, we found that in the sagittal plane, the vermis was very well delineated, and its longitudinal diameter could be measured. 3,17 This view may sometimes be difficult to obtain. Although in a vertex presentation, the vaginal approach can improve brain visualization, when the fetus lies in a breech presentation, it can require considerable time and expertise to obtain an adequate sagittal view. The 3DUS method described here might resolve these difficulties. Sagittal cerebral planes were reconstructed using a 3D acquisition from axial planes with TUI and static VCI. The method proved successful in all cases but 1 regardless of fetal position. The positive linear correlation shown in this study is in close agreement with results published previously by us 3 and Paladini and Volpe. 18 As described in the latter study, 18 we also found that the combination of TUI and static VCI was preferable in terms of focus, calibration, and image detail, and this was applied to all volumes (Figure 3, A and B). Figure 2. A, Magnification of the image to the maximum ( 4). B, Two-dimensional image of the vermis (v), primary fissure (pf), and corpus callosum (cc). A B 4 J Ultrasound Med 2009; 28:1 8

5 Zalel et al A normal fetal vermis size is not sufficient, however, for determining the normality of the vermis. The existence of the primary fissure is extremely important in evaluation of the vermis. 5,19 The primary fissure divides the larger posterior lobe from the anterior lobe (volume ratio, 2:1) by the 11th to 12th weeks. All of these main fissures, as well as the primordial tissue of the different lobes, are present by the 14th gestational week. 19 Hence, the 3 cerebellar lobes can be anatomically differentiated by the end of the first trimester. 19,20 However, the appearance of thin cerebellar fissures is delayed on MRI compared with macroscopic data. The primary fissure of the vermis could normally be seen on MRI at approximately 25 to 26 weeks gestation in the sagittal plane according to Adamsbaum et al 5 and even as late as 30 weeks in another MRI study. 7 Although on 2-dimensional (2D) ultrasonographic evaluation, the primary fissure can be shown from 22 to 23 weeks and onward (Figure 2B) or even at 27 weeks, 4 our 3D study confirms the normality of the vermis by showing the primary fissure as early as 18 weeks (in 40% of cases), in more than 90% at 22 weeks, and in 100% of cases from 23 to 24 weeks and onward. There have been several studies applying 3DUS to evaluation of the fetal vermis. 21,22 Three-dimensional ultrasonography using VCI in the C-plane for measurement of the fetal cerebellar vermis at 18 to 33 weeks gestation was described by Viñals et al. 23,24 Tomographic ultrasound imaging was also used to obtain the midsagittal plane. Measurements of the corpus callosum and cerebellar vermis were obtained by 3DUS, and the clearest and sharpest definition of midline structures was obtained with transfrontal acquisition. 24 Primary and secondary fissures of the cerebellar vermis could be detected in 13% to 26% of multiplanar, 18% to 35% of VCI in the C-plane, and 52% to 79% of transfrontal analyses. Our study builds from the above study. 24 First, the transfrontal view is sometimes difficult to achieve, requiring a transvaginal approach especially when the fetus is in a vertex position. Second, whereas this study showed overall visualization of the primary fissure, the investigators did not show the time of its appearance or rate of visualization by gestational week. These data Figure 3. Tomographic ultrasound imaging in the C-plane with the slice thickness set at 0.5 mm in a fetus at 25 weeks gestation (A), a fetus at 24 weeks showing measurement of the maximum diameter of the vermis (B), and a fetus at 19 weeks with an outline of the primary fissure (C). A B C J Ultrasound Med 2009; 28:1 8 5

6 3D Ultrasonography of Fetal Vermis at 18 to 26 Weeks Gestation Figure 4. Primary fissure (center arrow) in the rendered mode at 23 weeks gestation. Double arrow indicates the fastigium. Table 1. Appearance of the Vermian Primary Fissure by Gestational Age Gestational Age, wk Primary Fissure, n (%) 18 2/5 (40) 19 3/5 (60) 20 7/11 (64) 21 4/5 (80) 22 16/17 (94) 23 48/49 (98) 24 53/53 (100) 25 19/19 (100) 26 9/9 (100) Total 161/173 (94) Figure 5. Linear growth of the vermis during pregnancy (18 26 weeks). The solid line is the linear regression line for vermis length; the dotted lines represent ±2 SD of the mean for each gestational week. might aid practitioners in determining the normality of the vermis, especially at early stages of gestation. Our study had a number of limitations. This technique is not easy for operators inexperienced in 3DUS. C-plane resolution is not as good as A-plane resolution. A comparison of 2D midsagittal pictures of the fetal brain with 3D reconstructions from volumes originally taken in the transverse plane suggested that the 2D images were superior, and acoustic shadowing from the base of the skull obscured the brain stem and at times did not allow precise assessment of the relationship between this and the cerebellar vermis. 25 Therefore, we applied this technique only to the vermis and not to the other structures of the posterior fossa. Another possible limitation was the high proportion of cases at around 22 to 24 weeks gestation compared with other gestational ages. This was due to the greater number of fetal anomaly screening examinations performed during this period. We think that this did not limit the value of this study with regard to the depiction of the primary fissure. Nevertheless, our study has shown several advantages of 3DUS of the fetal vermis. First, 3D planes obtained from axial acquisitions are simpler and easier to display without regard to the fetal lie. Postprocessing allows better visualization of the various anatomic landmarks by offline navigation in the volume, fine calibration, and selection of different rendering filters. A combination of this accessible plane with postprocessing capabilities enables even relatively inexperienced examiners to image the fetal vermis consistently. Second, as in former 2DUS and 3DUS studies, growth of the vermis was found to be linear during gestation and correlated well with known values in the literature. Last, and most importantly, using VCI and TUI allowed depiction of small parts, such as the primary fissure, which could be visualized as early as 18 weeks gestation in some cases and in most cases by 22 to 24 weeks. The combination of VCI and TUI improved both the resolution and the contrast compared with 2DUS images. 18 This could explain the earlier visualization of the primary fissure in our study compared with 2DUS and MRI studies and could contribute to early identification of vermian lesions. 6 J Ultrasound Med 2009; 28:1 8

7 Zalel et al Table 2. Measurements of Fetal Vermian Length by Gestational Age Gestational Measurement by Percentile, mm Age, wk 5th 10th 25th 50th 75th 90th 95th 18 and 19 a ND ND ND ND ND ND ND ND ND indicates not determined. a Combined to adjust for small numbers. References 1. Sowel ER, Jernigan TL, Mattson SN, Riley EP, Sobel DF, Jones KL. Abnormal development of the cerebellar vermis in children prenatally exposed to alcohol: size reduction in lobules I V. Alcohol Clin Exp Res 1996; 20: American College of Obstetricians and Gynecologists. ACOG Technical Bulletin: Ultrasonography in Pregnancy. Washington, DC: American College of Obstetricians and Gynecologists; Technical bulletin Zalel Y, Seidman DS, Brand N, Lipitz S, Achiron R. The development of the fetal vermis: an in-utero sonographic evaluation. Ultrasound Obstet Gynecol 2002; 19: Malinger G, Ginath S, Lerman-Sagie T, Watemberg N, Lev D, Glezerman M. The fetal cerebellar vermis: normal development as shown by transvaginal ultrasound. Prenat Diagn 2001; 21: Adamsbaum C, Moutard ML, André C, et al. MRI of the fetal posterior fossa. Pediatr Radiol 2005; 35: Pilu G, Romero R, De Palma L, Jeanty P, Burdine C, Hobbins JC. Ultrasound investigation of the posterior fossa in the fetus. Am J Perinatol 1987; 4: Guibaud L Practical approach to prenatal posterior fossa abnormalities using MRI. Pediatr Radiol 2004; 34: Royston P, Wright EM. How to construct normal ranges for fetal variables. Ultrasound Obstet Gynecol 1998; 11: Altman NR, Naidich TP, Braffman BH. Posterior fossa malformations. AJNR Am J Neuroradiol 1992; 13: Goldstien RB, Podrasky AE, Filly RA, Callen PW. Effacement of the fetal cisterna magna in association with myelo - meningocele. Radiology 1989; 172: Kollias SS, Ball WS Jr, Prenger EC. Cystic malformations of the posterior fossa: differential diagnosis clarified through embryologic analysis. Radiographics 1993; 13: Lemire RJ, Loeser JD, Leech RW, Alvord EC Jr. Normal and Abnormal Development of the Human Nervous System. Hagerstown, MD: Harper & Row; Bromley B, Nadel AS, Pauker S, Estroff JA, Benacerraf BR. Closure of the cerebellar vermis: evaluation with second trimester US. Radiology 1994; 193: Babcook CJ, Chong BW, Salamat MS, Ellis WG, Goldstein RB. Sonographic anatomy of the developing cerebellum: normal embryology can resemble pathology. AJR Am J Roentgenol 1996; 166: Ben-Amin M, Perlitz Y, Peleg D. Transvaginal sonographic appearance of the cerebellar vermis at weeks gestation. Ultrasound Obstet Gynecol 2002; 19: Carroll SG, Porter H, Abdel-Fattah S, Kyle PM, Soothill PW. Correlation of prenatal ultrasound diagnosis and pathologic findings in fetal brain abnormalities. Ultrasound Obstet Gynecol 2000; 16: Zalel Y, Gilboa Y, Gabis L, et al. Rotation of the vermis as a cause of enlarged cisterna magna on prenatal imaging. Ultrasound Obstet Gynecol 2006; 27: Paladini D, Volpe P. Posterior fossa and vermian morphometry in the characterization of fetal cerebellar abnormalities: a prospective three-dimensional ultrasound study. Ultrasound Obstet Gynecol 2006; 27: Triulzi F, Parazzini C, Righini A. MRI of fetal and neonatal cerebellar development. Semin Fetal Neonatal Med 2005; 10: Triulzi F, Parazzini C, Righini A. MRI of fetal cerebellar development. Cerebellum 2006; 5: Correa FF, Lara C, Bellver J, Remohí J, Pellicer A, Serra V. Examination of the fetal brain by transabdominal threedimensional ultrasound: potential for routine neurosonographic studies. Ultrasound Obstet Gynecol 2006; 27: Hata T, Yanagihara T, Matsumoto M, et al. Three-dimensional sonographic features of fetal central nervous system anomaly. Acta Obstet Gynecol Scand 2000; 79: Viñals F, Muñoz M, Naveas R, Shalper J, Giuliano A. The fetal cerebellar vermis: anatomy and biometric assessment using volume contrast imaging in the C-plane (VCI-C). Ultrasound Obstet Gynecol 2005; 26: J Ultrasound Med 2009; 28:1 8 7

8 3D Ultrasonography of Fetal Vermis at 18 to 26 Weeks Gestation 24. Viñals F, Muñoz M, Naveas R, Giuliano A. Transfrontal three-dimensional visualization of midline cerebral structures. Ultrasound Obstet Gynecol 2007; 30: Pilu G, Segata M, Ghi T, et al. Diagnosis of midline anomalies of the fetal brain with the three-dimensional median view. Ultrasound Obstet Gynecol 2006; 27: J Ultrasound Med 2009; 28:1 8