DOI: 10.1002/pd.4258

ORIGINAL ARTICLE

Second trimester fetal neurosonography: reconstructing cerebral midline anatomy and anomalies using a novel three-dimensional ultrasound technique Gabriele Tonni1*, Gianpaolo Grisolia2 and Waldo Sepulveda3 1

Department of Obstetrics and Gynecology, Prenatal Diagnostic Service, Guastalla Civil Hospital, ASL Reggio Emilia, Italy Department of Obstetrics and Gynecology, Prenatal Diagnostic Service, ‘Carlo Poma’ Hospital, Mantua, Italy 3 Fetal Medicine Center, Fetal Medicine Interest Group GIMEF, Santiago, Chile *Correspondence to: Gabriele Tonni. E-mail: [email protected] 2

ABSTRACT Objective To describe the application of a novel 3D ultrasound reconstructing technique (OMNIVIEW) that may facilitate the evaluation of cerebral midline structures at the second trimester anatomy scan. Methods Fetal cerebral midline structures from 300 consecutive normal low-risk pregnant women were studied prospectively by 2D and 3D ultrasound between 19-23 weeks of gestation. All the newborn infants underwent pediatric follow-up and were considered normal up to 2 years of life. In addition, five confirmed pathologic cases were evaluated and the abnormal features using this technique are described in this clinical series. Results Off-line volume data sets displaying the corpus callosum and the cerebellar vermis anatomy were accurately reconstructed in 98.5% and 96% of cases from sagittal and axial planes, respectively. For pathological cases, an agreement rate of 0.96 and 0.91 for midsagittal and axial planes, respectively, was observed.

Conclusions This study demonstrates the feasibility of including 3D ultrasound as an adjunct technique for the evaluation of cerebral midline structures in the second trimester fetus. Future prospective studies will be necessary to evaluate if the application of this novel 3D reconstructing technique as a step forward following 2D second trimester screening scan will improve the prenatal detection of cerebral midline anomalies in the low-risk pregnant population. © 2013 John Wiley & Sons, Ltd. Funding sources: WS was supported by an unrestricted research grant from the Sociedad Profesional de Medicina Fetal ’Fetalmed’ Limitada, Chile. Conflicts of Interest: None declared

INTRODUCTION Despite the high incidence of central nervous system (CNS) abnormalities and the clinical importance of their prenatal diagnosis, ultrasound screening is still far from producing satisfactory results.1 This may be due, at least in part, to the fact that the basic examination of the fetal head is limited to axial planes of the brain.2 An extended study of the fetal CNS anatomy, including additional sagittal and coronal planes of the fetal brain, may help improve diagnostic efficacy.2–4 Indeed, one of the most important views for examination of the fetal brain is probably the midsagittal view, which provides unique and relevant information regarding the corpus callosum and the cerebellar vermis.5 Unfortunately, this scanning plane may be difficult to obtain, especially when the fetus is not in the supine position or when the fetal head is located deep in the maternal pelvis. Several approaches have been described to visualize this plane, but they all require considerable skill, time, and frequently examination using the transvaginal approach.2 Dedicated

Prenatal Diagnosis 2014, 34, 75–83

evaluation of the fetal CNS (i.e. the fetal neurosonogram) via transvaginal examination using high-resolution ultrasound probes has much greater diagnostic potential than the standard transabdominal examination, and it is indicated in pregnancies at increased risk of CNS abnormalities or in cases where a basic examination has revealed findings suggestive of anomalies.2 Fetal neurosonogram is performed by aligning the transducer with the sutures and fontanelles of the fetal head.5,6 In addition to the basic transthalamic, transventricular, and transcerebellar axial views, evaluation of four coronal views (transfrontal, transcaudate, transthalamic, and transcerebellar) and two sagittal views (midsagittal and parasagittal) has been recommended.2 In recent years, three-dimensional (3D) ultrasound has been used effectively to identify major brain structures and depict structures usually not visualized with the two-dimensional (2D) transabdominal axial approach.7,8 The primary aims of this study were to assess if fetal cerebral midline structures could be accurately reconstructed using a

© 2013 John Wiley & Sons, Ltd.

G. Tonni et al.

76

novel 3D ultrasound technique and to determine the success rate for visualization during the second trimester anatomy scan. The secondary aim was to evaluate the level of interobserver agreement using off-line analysis of brain volume data sets at remote site by an operator who was ‘blind’ to the scan result. Representative views of both normal and abnormal cases are presented.

Transvaginal scan was performed in all suspicious/pathological cases by a board-certified neurosonographer. In these cases, the 2D and 3D imaging were reviewed on a remote site for expert consultation (WS) using a dedicated website. In selected cases, in utero single-shot fast spin-echo magnetic resonance (MR) imaging was performed as integrated, complementary investigation, and postmortem examination was performed by a fetal pathologist.

METHODS For the purpose of this study, 300 consecutive low-risk pregnant women undergoing a routine second trimester scan were studied. The study was approved by our Institutional Review Board, and all patients gave their informed consent. No congenital abnormalities were detected, either at prenatal check up or at postnatal follow-up, and all patients delivered at term. All fetuses delivered in two referral hospitals and underwent pediatric follow-up at 2 years of life using a prestamped form sheet that was given to the parents at discharge and/or by telephone interview in targeted cases. Follow-up information was obtained in all cases, and all infants were considered healthy by pediatric examination. Examination of the fetal CNS was performed with a highresolution 2D/3D ultrasound machine (Voluson E8, GE Medical Healthcare System, Milwaukee, WI, USA) equipped with a RAB 4 to 8 MHz multifrequency transabdominal 3D/4D probe. The following 2D scan settings were used: tissue harmonic: high level; cross beam: 1; speckle reduction imaging: 3; dynamic range: 6; dynamic contrast: 7; line density: normal; gray map: 7. In all fetuses, the cavum septum pellucidum and the cerebellar vermis were first displayed using 2D ultrasound. The fetus was scanned with real-time 2D ultrasound with an insonation angle of 45° to minimize the acoustic shadow from the base of the skull. Brain volume data sets were acquired by 3D ultrasound in the ‘maximum’ quality mode during fetal rest and maternal apnea using a transabdominal acquisition angle of 45° to 60° depending on the gestational age. Two volume data sets were acquired with the region of interest set to capture the entire fetal head. A mixture of gradient light 0/100 was used. Threshold was 20, quality high1, B70°/V70°. Once the volume was captured, the OMNIVIEW software (GE Healthcare) was activated from a menu option on the touch screen. This program enables the selection of a single straight line, multiple lines, or curved line for navigation purposes. To perform volume rendering, a straight line for automatic display of the median plane was drawn along the cerebral midline structures. Volume contrast imaging (VCI) was also applied to enhance contrast resolution using 1.7 to 2 mm slice. The use of the OMNIVIEW software in the static 3D mode rather than the 3D live mode (4D) was preferred. Blind validation analysis was conducted off-line using dedicated 4D VIEW software (GE Healthcare) by a single examiner (GT) with expertise in fetal 3D volume reconstruction. Fetuses with cerebral midline anomalies diagnosed in the second trimester in our low-risk population were identified from our fetal medicine database and the corresponding 3D volume data sets retrieved to perform local analysis. Prenatal Diagnosis 2014, 34, 75–83

RESULTS For normal cases, the mean maternal age and mean gestational age at second trimester scan was 31 years (range, 21–39) and 20w1d of gestation (range, 19–23), respectively. Gravidity and parity showed a mean value of 1.4 and 1.0, respectively. For abnormal cases, the mean maternal age was 32 years (range, 22–42), and the mean gestational age at second trimester scan was 20w4d of gestation (range, 20–21.5). Gravidity and parity showed a mean value of 1.6 and 0.4, respectively. For normal cases, ethnicity was represented by Caucasian race in 83%, Afro–Caribbean in 11%, and Asian in 6% of cases, whereas mothers of abnormal cases were all of Caucasian race. In normal fetuses, satisfactory volume data sets analysis of the corpus callosum and cerebellar vermis was possible in 98.5% and 96.0% of cases in which midsagittal and axial planes were used for acquisition, respectively. In the remaining cases, off-line interpretation was suboptimal because of a maternal body mass index higher than 30 Kg/m2 and the presence of anterior uterine myomas or surgical abdominal scars that resulted in image shadowing. Figure 1 shows a normal corpus callosum and posterior fossa at 20 weeks of gestation. Figures 2 – 5 show representative images from the five different pathological conditions diagnosed in our units including complete agenesis of the corpus callosum (Figure 2), partial agenesis of the corpus callosum (Figure 3), Blake’s pouch cyst (Figure 4), partial agenesis of the corpus callosum associated with abnormal posterior area membranacea (Blake’s pouch cyst) (Figure 5), and cerebellar vermis hypoplasia (Figure 6). The sonographic diagnosis was subsequently confirmed in all cases, either by fetal MR imaging or postmortem examination. No statistical differences were noted between 2D and 3D ultrasound with respect to the prenatal diagnosis. Nevertheless, 2D/3D ultrasound demonstrated a sensitivity of 98% and 91% in the diagnosis of corpus callosum anomalies and cerebellar vermis pathology, respectively. The agreement rate between operators, when off-line volume data sets of pathological cases were reviewed at remote site using Digital Imaging and Communications in Medicine (DICOM) technology, was 0.96 and 0.91 for midsagittal and axial planes, respectively.

DISCUSSION For a thorough examination of the fetal CNS, the sagittal and the coronal planes of the fetal brain should be an integral part of the study.2,4 However, visualization of these additional planes requires either a transvaginal or a transabdominal approach with a transfrontal view through the metopic suture, © 2013 John Wiley & Sons, Ltd.

Three-dimensional ultrasound and CNS anomalies

77

Figure 1 Normal corpus callosum and normal posterior fossa obtained with the OMNIVIEW reconstructing technique at a 20w1d of gestation. During volume rendering, a straight line was drawn along midline cerebral structures (left panel), allowing automatic display of the midsagittal view (right panel). Volume contrast imaging (VCI) with a 2 mm slice was used to enhance contrast resolution (BS, brainstem; CC, corpus callosum; f, fastigum; v, cerebellar vermis.)

Figure 2 (A) Complete agenesis of the corpus callosum diagnosed in a 36-year-old primigravida with a body mass index less than 25 kg/m2 who underwent a routine second trimester anatomy scan at 20w6d of gestation. On 2D ultrasound, the cavum septum pellucidum was not visualized, and the color Doppler ultrasound was unable to identify the pericallosal artery. Using the OMNIVIEW software, the median view was reconstructed from the volume data set originally captured in the sagittal plane. The corpus callosum was not visualized. Fetal karyotype was 46,XY. (B) Fetal magnetic resonance (MR) imaging (Signa®, Phillips, Eindovhen, The Netherlands) was performed using T2-weighted single-shot fast spin-echo, confirming the ultrasound diagnosis Prenatal Diagnosis 2014, 34, 75–83

© 2013 John Wiley & Sons, Ltd.

78

G. Tonni et al.

Figure 3 (A) Partial agenesis of the corpus callosum diagnosed in a 26-year-old primigravida with a body mass index of 28 kg/m2 who underwent a routine second trimester anatomy scan at 21w5d of gestation (spl, splenium). (B) Postmortem examination confirmed the prenatal diagnosis by demonstrating interruption of the corpus callosum at the level of the body/splenium (blue arrow) (A, anterior; BS, brainstem; CC, corpus callosum; F, fornix; P, posterior; T, thalamus)

as well as operator expertise and favorable fetal position. 3D ultrasound has been suggested as a method that is able to overcome the limitations of dependence on operator skills.9,10 Alternatively, acquisition of volume data sets starting from the axial view of the fetal head and off-line ‘navigation’ using multiplanar reconstruction of planes can be employed to obtain the diagnostic planes and to reduce operator dependence, which may potentially increase the detection rate of CNS abnormalities.11 The introduction of 3D volume ultrasound is an important advance in fetal imaging as compared with traditional 2D ultrasound.12,13 The concept of automated volume sonography based on operator-independent retrieval of diagnostic 2D planes from a 3D volume requires initial, predetermined standardization of organ-specific 3D volumes. Similarly, 180° rotation along the y-axis in plane A, z-rotation, and placement of the ‘reference dot’ at the midpoint of the interhemispheric fissure is required.10 Monteagudo et al.4 examined 34 patients with a history of brain abnormality or suspected brain pathology and compared 2D and 3D transvaginal neurosonogram. Brain structures such as the corpus callosum, the cerebellar hemispheres and Prenatal Diagnosis 2014, 34, 75–83

vermis, and the cisterna magna were attempted to be visualized. In all 34 cases, a 3D volume of high diagnostic quality was obtained, and all three conventional planes were imaged in all fetuses. The key difference between the 2D and the 3D studies was that the axial plane could only be obtained by the 3D reconstruction of the volume data sets, which is an advantage because this plane is rarely seen with 2D transvaginal technology. The planes obtained off-line from the 3D volume were parallel and not oblique or at an angle, as is the case with conventional 2D transvaginal neurosonogram. Finally, when 2D and 3D ultrasound studies of pathological cases were compared, the important advantage offered by ‘navigating’ inside the volumes generated by 3D ultrasound was the ability to follow the reference dot that indicates the same anatomical point on all three orthogonal planes, facilitating identification of the midline structures. Several other investigators have studied the utility of 3D ultrasound in the prenatal evaluation of the cerebral midline structures. Plasencia et al.14 demonstrated that the corpus callosum can be displayed by reconstructing the midsagittal plane in 90% of the fetuses scanned between 20 and 24 weeks of gestation. Pilu et al.15 demonstrated a good correlation © 2013 John Wiley & Sons, Ltd.

Three-dimensional ultrasound and CNS anomalies

79

Figure 4 (A) Abnormal posterior area membranacea (Blake’s pouch cyst) diagnosed in a 34-year-old woman, para1, with a body mass index of less than 25 kg/m2 presented at routine scan at 20w1d of gestation. The fetal karyotype was normal (46,XX). A. With the assistance of the OMNIVIEW software, a section plane along the cisterna magna and the cerebellar vermis (yellow line) was drawn as well as an oblique plane (fuchsia line) along the transverse cerebellar diameter demonstrating Blake’s pouch cyst in sagittal and coronal planes (curved white arrow, lower panel). (B) Using transvaginal approach and sequential reconstruction along the transverse cerebellar diameter (the fuchsia and the blue lines), the communication between the fourth ventricle and Blake’s pouch cyst was demonstrated in the coronal plane (curved white arrow, lower right panel)

Figure 5 Partial agenesis of the corpus callosum associated with Blake’s pouch cyst in a 22-year-old primigravida with a body mass index of 36 kg/m2 presented at routine second trimester scan in a 20w1d of gestation. The yellow line was drawn straight along the cerebral midline, demonstrating an absent body and splenium, which led to the diagnosis of partial agenesis of the corpus callosum (BS, brainstem; csp, cavum septum pellucidum; f, fastigium; spl, splenium; v, cerebellar vermis). The curved white arrow denote the Blake’s pouch cyst on sagittal section. Prenatal Diagnosis 2014, 34, 75–83

© 2013 John Wiley & Sons, Ltd.

80

G. Tonni et al.

Figure 6 (A) Cerebellar vermis hypoplasia in a 42-year-old woman, para 1, with a body mass index of 30 kg/m2 referred at 20w0d of gestation for a level II ultrasound examination because of a suspected vermis hypoplasia at routine 2D scan. The OMNIVIEW software was used to obtain a reslicing along the cerebellar vermis (yellow line). The diagnostic features of cerebellar vermis hypoplasia were rendered in the sagittal plane (BS, brainstem; CC, corpus callosum; v, cerebellar vermis). Fetal karyotype was normal (46,XY). Note a small cerebellar vermis and upward displacement of the tentorium. (B) Fetal MR imaging (Signa®, Phillips, Eindovhen, The Netherlands) was performed using T2-weighted single-shot fast spin-echo, confirming the 2D/3D ultrasound diagnosis. (C) Elective termination of pregnancy was accomplished at 22w1d of gestation. Postmortem examination demonstrated an enlarged fourth ventricle with a square-shaped roof (blue arrow). The cerebellar vermis was hypoplastic and more tissue was seen inferiorly than superiorly. There was no evidence of a primary fissure (yellow arrow), and the interpeduncular fossa was unusually deep (red star)

between midsagittal planes obtained directly with 2D and 3D ultrasound in 13 fetuses with cerebral midline abnormalities including 5 with partial or complete agenesis of the corpus callosum, 6 with posterior fossa malformations, and 2 with a combination of the two. They concluded that the diagnosis was possible in all cases using either 2D or 3D views, although 3D views were obtained more easily and quickly. Similarly, Viñals et al.16 demonstrated that transfrontal 3D acquisition of brain volume data sets is optimal for examining both the corpus callosum and the cerebellar vermis after 20 weeks of gestation. Bornstein et al.17 performed off-line analysis using transabdominal 3D gray scale and power Doppler volumes of the fetal brain acquired in 102 consecutive normal fetuses at 20 to 23 weeks and demonstrated that the midsagittal plane was easily obtained in all cases, with diagnostic-quality images of the corpus callosum acquired in 93.1% and 99.0% and pericallosal artery in 94.4% and 95.5% of cases by two examiners, respectively. Prenatal Diagnosis 2014, 34, 75–83

Miguelote et al.18 compared the feasibility and reproducibility of 3D volume reconstruction neurosonogram for measurement of corpus callosum length in 46 normal fetuses examined by 2D and 3D ultrasound at 23 to 25, 27 to 28, and 31 to 32 weeks of gestation. Direct midsagittal views were obtained by either a transabdominal or transvaginal approach and 3D reconstructed midsagittal views by 3D multiplanar manipulations and VCI in the C-plane technique from volumes acquired in axial planes. They demonstrated that the corpus callosum could be measured in 91% of transvaginal acquisitions, in 52% of transabdominal acquisitions, in 92% of multiplanar reconstructions, and in 86% of VCI in the C-plane technique reconstructions. The success rate was independent on gestational age for transvaginal acquisition and slightly dependent on gestational age for 3D reconstruction techniques. However, transabdominal acquisition was dependent on gestational age and fetal presentation. OMNIVIEW is a new reconstructing modality for 3D/4D ultrasound that allows the interrogation of volume data sets and the simultaneous display of up to three independent, non© 2013 John Wiley & Sons, Ltd.

Three-dimensional ultrasound and CNS anomalies

orthogonal planes of any given organ. The software enables operators to trace cross-sectional planes freely to obtain also ‘virtual’ non-orthogonal (non-rigid) planes to reconstruct the anatomy. Our study on the application of the OMNIVIEW software demonstrates that this technology can allow visualization of the main fetal cerebral midline structures during the second trimester anatomy scan. Our results are in keeping with those of Rizzo et al.19 who used this same technology to reconstruct the sagittal and coronal planes of the brain in 106 normal fetuses at 18 to 24 weeks of gestation. Midsagittal, parasagittal, transfrontal, transcaudate, transthalamic, and transcerebellar planes were obtained, with visualization rate for brain structures of 72% to 96% using sagittal sections and 76% to 91% using coronal planes. The agreement rate between operators was 0.93 and 0.89 for sagittal and coronal planes, respectively. These authors19 were also able to detect accurately all nine cases of cerebral pathology, including complete agenesis of the corpus callosum, borderline ventriculomegaly, and classic Dandy– Walker malformation. Our experience with pathologic cases also confirmed these previous results, although fetal movement, maternal habitus, and shadowing from the surrounding bones can limit the application of this semiautomated technique. In spite of these limitations, we have shown that this technique can be used to reconstruct the fetal corpus callosum and the posterior fossa during routine second trimester scan in almost all cases, with a short learning curve for the experienced operator. We have also shown that OMNIVIEW, such as any other commercially available software, can provide reliable manipulation of 3D volume data sets to obtain snapshots of the sagittal and axial planes of the fetal brain that have proven to be clinically useful in the prenatal study of midline cerebral abnormalities, such as the ones described above. The scanning time required for obtaining the diagnostic section was very short, with a mean time of 45 s. OMNIVIEW can also be used in association with VCI, a software feature that displays a thin slice of adjustable thickness to improve contrast resolution from any given acquired volume. This differs significantly from tomographic ultrasound imaging, in which volume data sets are automatically sliced and displayed in multiple parallel images of equal width. An advantage of this novel 3D reconstructing technique, as compared with the 2D ultrasound approach, is that the fetal brain can be sequentially sectioned in all three parallel orthogonal planes on demand. These planes are comparable with the sections obtained using serial tomograms by computerized tomography or MR imaging. We would like to remark that all diagnoses were made by 2D neurosonogram that was performed according to International Society of Ultrasound in Obstetrics and Gynecology guidelines.2 3D volumes data sets were also sent using DICOM technology and analyzed by an expert in offline volume ‘navigation’ that was not aware of the diagnosis made at the time of scanning. The diagnosis made at remote site was in all cases in agreement with the initial real-time 2D ultrasound diagnosis and allowed a high quality postPrenatal Diagnosis 2014, 34, 75–83

81

processing analysis. Although the stored volumes are large, up to 10 megabytes, they can be compressed to about 15% to 20% of the original file size in many instances without noticeable loss of detail.20 Furthermore, 3D ultrasound confers the ability to construct planes that might be difficult to obtain by 2D imaging, offering a new flexibility to sonographers for defining the severity, location, and extent of anomalies.21 Although off-line analysis of 3D volume data sets has proven to be a reliable method that can be used to help in the assessment of brain anomalies and could be a useful adjunct to real-time 2D ultrasound,20 this information must still be interpreted with caution because of possible artifacts.22,23 The hyperechogenicity of the corpus callosum must be considered by using 3D reconstruction. This type of artifact may be due to the simultaneous use of 3D pixel reconstruction in conjunction with speckle reduction image. The structure observed represents the interface between the cingulate gyrus, the cingulate sulcus, the cerebrospinal fluid, and the blood flow in the callosal arteries. A curvilinear pericallosal lipoma shows similar characteristics on 3D imaging but usually does not interfere with the 2D visualization of the corpus callosum.22 Notwithstanding, multiplanar images obtained by transabdominal 3D ultrasound provide a simple and effective approach for detailed evaluation of the fetal brain anatomy and has the potential to be used in the routine fetal anatomy scan.24

CONCLUSION We suggest that the methodology described here may potentially aid the evaluation of sagittal, axial, and coronal planes of the cerebral midline anatomy at the second trimester anatomy scan, improving the detection of pathological cases and allowing subsequent referral to tertiary centers for dedicated fetal neurosonogram. In all suspicious cases of our series, an integrated, complementary prenatal investigation with MR imaging and/or postmortem examination was performed to control the ultrasound prenatal diagnostic accuracy. The prenatal diagnoses were all first made by realtime 2D ultrasound that currently represents the gold standard when performing a screening fetal neurosonogram. Notwithstanding, although the quality of the image by 2D ultrasound may be higher than that obtained using 3D ultrasound and no statistical differences could be observed between the two techniques, 3D ultrasound was less time consuming, allowed accurate reconstruction of the anatomical lesions and allowed simultaneous rendering in the sagittal, axial, and coronal planes when compared with 2D ultrasound. Furthermore, 3D ultrasound allowed accurate volume reconstruction and expert consultation using DICOM technology. As nowadays the spectrum of callosal anomalies includes complete and partial callosal agenesis, hypoplasia, and dysgenesis (the latter condition is characterized by a thick corpus callosum), we discourage the uninitiated 3D operator to use the presented technique routinely in order to obtain shortcuts to image some of the most important structures on the median plane of the fetal © 2013 John Wiley & Sons, Ltd.

G. Tonni et al.

82

brain instead the well-documented direct acquisition of the median plane. Such systematic examination can be achieved by extending the basic neurosonogram to include the techniques described in the International Society of Ultrasound in Obstetrics and Gynecology guidelines,2 applying some of the different techniques such as tomographic ‘slicing’ of the brain in the axial, coronal, and sagittal planes. Nevertheless, future prospective studies are needed to determine whether the use of OMNIVIEW software, as a step forward to 2D screening scan rather than a confirmatory technique, may enhance prenatal detection of CNS abnormalities in the unselected population. Acquisition of a volume with 3D display, to obtain the median plane in multiplanar mode, should be the preferred diagnostic approach of the second trimester neurosonogram, when expert operators and high-resolution 2D/3D ultrasound equipment with dedicated software are available.

ACKNOWLEDGEMENTS Authors would like to thank Dr. Andrea Rossi, Head Neuroradiology Unit, IRCCS Gaslini Hospital, Genoa, Italy, for neuroradiology imaging; Prof. William Halliday, Neuropathology Unit, The Hospital for Sick Children, Toronto, Canada and Dr.

Maria Paola Bonasoni, Pathology Service, IRCCS Ospedale “Santa Maria Nuova” Reggio Emilia, Italy, for postmortem examination. WHAT’S ALREADY KNOWN ABOUT THIS TOPIC? • Direct sonographic demonstration of the midline cerebral structures, such as the corpus callosum and cerebellar vermis, during the routine second trimester scan, is highly dependent on operator’s skill, fetal position, and maternal habitus. • Three-dimensional sonography of the fetal brain allows the acquisition of volume data sets that can be examined off-line in any given section, facilitating the evaluation of the midline cerebral structures and diagnosis of anomalies of the corpus callosum and cerebellar vermis.

WHAT DOES THIS STUDY ADD? • Three-dimensional data sets from the fetal brain acquired in the axial plane can be reformatted using the OMNIVIEW software, allowing instantaneous and simultaneous visualization of the orthogonal, midsagittal plane. • Our experience involving both normal and abnormal fetuses confirms that the OMNIVIEW software is a useful tool for the assessment of the corpus callosum and cerebellar vermis at the time of the second trimester anatomy scan.

REFERENCES 1. Garne E, Dolk H, Loane M, et al. EUROCAT website data on prenatal detection rates of congenital abnormalities. J Med Screen 2010;17:97–8. 2. International Society of Ultrasound in Obstetrics and Gynecology Education Committee. Sonographic examination of the fetal central nervous system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet Gynecol 2007;29:109–16. 3. Pilu G, Ghi T, Carletti A, et al. Three-dimensional ultrasound examination of the fetal central nervous system. Ultrasound Obstet Gynecol 2007;30:233–45. 4. Monteagudo A, Timor-Tritsch IE, Mayberry P. Three-dimensional transvaginal neurosonography of the fetal brain: ‘navigating’ in the volume scan. Ultrasound Obstet Gynecol 2000;16:307–13. 5. Timor-Tritsch IE, Monteagudo A. Transvaginal fetal neurosonography: standardization of the planes and sections by anatomic landmark. Ultrasound Obstet Gynecol 1996;8:42–7. 6. Malinger G, Katz A, Zakut H. Transvaginal fetal neurosonography. Supratentorial structures. Isr J Obstet Gynecol 1993;4:1–5. 7. Timor-Tritsch IE. Ultrasonography of the Prenatal Brain, third ed. Chapter 4, Chapter 5. New York: Mac Graw Hill Professional eBook April 15 2012. 8. Chitty LS, Pilu G. The challenge of imaging the fetal central nervous system: an aid to prenatal diagnosis, management and prognosis. Prenat Diagn 2009;29:301–2. 9. Benacerraf BR, Shipp TD, Bromley B. Three-dimensional US of the fetus: volume imaging. Radiology 2006;238:988–96. 10. Abuhamad AZ. Standardization of 3-dimensional volumes in obstetric sonography: a required step in training and automation. J Ultrasound Med 2005;24:397–401. 11. Rizzo G, Abuhamad AZ, Benacerraf BR, et al. Collaborative study on 3dimensional sonography for the prenatal diagnosis of central nervous system defects. J Ultrasound Med 2011;30:1003–8. 12. Merz E, Bahlmann F, Weber G. Volume scanning in the evaluation of fetal malformations: a new dimension in prenatal diagnosis. Ultrasound Obstet Gynecol 1995;5:222–7. 13. Dyson RL, Pretorius DH, Budorick NE, et al. Three-dimensional ultrasound in the evaluation of fetal anomalies. Ultrasound Obstet Gynecol 2000;16:321–8.

Prenatal Diagnosis 2014, 34, 75–83

14. Plasencia W, Dagklis T, Borenstein M, et al. Assessment of the corpus callosum at 20–24 weeks’ gestation by three-dimensional ultrasound examination. Ultrasound Obstet Gynecol 2007;30:169–72. 15. 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;27:522–9. 16. Viñals F, Muñoz M, Naveas R, Giuliano A. Transfrontal threedimensional visualization of midline cerebral structures. Ultrasound Obstet Gynecol 2007;30:162–8. 17. Bornstein E, Monteagudo A, Santos R, et al. A systematic technique using 3-dimensional ultrasound provides a simple and reproducible mode to evaluate the corpus callosum. Am J Obstet Gynecol 2010;202:201.e1–5. 18. Miguelote RF, Vides B, Santos RF, et al. Feasibility and reproducibility of transvaginal, transabdominal, and 3D volume reconstruction sonography for measurement of the corpus callosum at different gestational ages. Fetal Diagn Ther 2012;31:19–25. 19. Rizzo G, Capponi A, Pietrolucci ME, et al. An algorithm based on OMNIVIEW technology to reconstruct sagittal and coronal planes of the fetal brain from volume datasets acquired by three-dimensional ultrasound. Ultrasound Obstet Gynecol 2011;38:158–64. 20. Salman MM, Twining P, Mousa H, et al. Evaluation of offline analysis of archived three-dimensional volume datasets in the diagnosis of fetal brain abnormalities. Ultrasound Obstet Gynecol 2011;38:165–9. 21. Bornstein E, Monteagudo A, Santos R, et al. Basic as well as detailed neurosonograms can be performed by offline analysis of three dimensional fetal brain volumes. Ultrasound Obstet Gynecol 2010;36:20–5. 22. Malinger G, Lerman-Sagie T, Viñals F. Three-dimensional sagittal reconstruction of the corpus callosum: fact or artifact (letter). Ultrasound Obstet Gynecol 2006;28:742–3. 23. Rizzo G, Pietrolucci ME, Capece G, et al. Satisfactory rate of postprocessing visualization of fetal cerebral axial, sagittal, and coronal planes from three-dimensional volumes acquired in routine second

© 2013 John Wiley & Sons, Ltd.

Three-dimensional ultrasound and CNS anomalies

trimester ultrasound practice by sonographers of peripheral centers. J Matern Fetal Neonat Med 2011;24:1071–6. 24. Correa FF, Lara C, Bellver J, et al. Examination of the fetal brain by transabdominal three-dimensional ultrasound: potential for routine neurosonographic studies. Ultrasound Obstet Gynecol 2006,27:503–8.

Prenatal Diagnosis 2014, 34, 75–83

83

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s web site.

© 2013 John Wiley & Sons, Ltd.

Second trimester fetal neurosonography: reconstructing cerebral midline anatomy and anomalies using a novel three-dimensional ultrasound technique.

To describe the application of a novel 3D ultrasound reconstructing technique (OMNIVIEW) that may facilitate the evaluation of cerebral midline struct...
6MB Sizes 0 Downloads 0 Views