Caracteristicile ultrasonografice corelate cu aspectele morfopatologice în spina bifida aperta

Introduction

Spina bifida (SB) is the most common non-fatal neural tube defect (NTD) in humans, with an incidence of approximately 0.5 per 1000 pregnancies. This is likely due to its multifactorial, genetic and non-genetic nature. In most patients, the underlying mechanism remains unknown. However, after 1991, 50% of the causes of neural tube defects were found to be related to a folate-based nutritional deficiency and, therefore, preventable if folate were to be administered prior to conception^(1-4). 

Spina bifida results from the incomplete neurulation between days 18 and 28 of embryogenesis, being mainly classified into open subtype (spina bifida aperta; SBA) or closed subtype – spina bifida occulta (SBO), depending on whether or not there is tegument coverage of the neural tissue. A systematic classification of SB phenotypes was developed more recently and includes the following subtypes: myeloschisis (MYS), characterized by a neural plate that opens to the skin surface and the nervous tissue is directly exposed to amniotic fluid, without meningeal or dermal covering, myelomeningocele (MMC), defined as a herniated sac containing spinal cord and meninges protruding through the spinal defect, and meningocele (ME), which is the meningeal herniation through the vertebral defect. In addition, another subtype, called limited dorsal myeloschisis (LDM), was defined as a focal midline closure defect of the skin with a fibro neural stalk connecting the skin lesion with the underlying spinal cord^(5,6).

Patients with SBA usually present with varying degrees of motor and sensorial impairment, as well as urinary and fecal incontinence^(4,7), depending on the anatomical and functional level of the lesion. However, the prognosis for the patients with open spina bifida also depends on the associated brain malformations, ranging from Chiari II to hydrocephalus and heterotopia, gyrification defects and other neocortical abnormalities, such as hypoplasia or flattening of cranial nerve nuclei, agenesis of the corpus callosum, thalamic fusion and reduction of total white matter of the brain with increased thickness of the frontal neocortex^(8-11).

Patients affected by spina bifida with cortical developmental malformations like heterotopias or polymicrogyria may present symptoms such as delay in development, epilepsy and focal neurological deficits^(12-14). There is strong evidence stating that the pathogenesis of SB reflects a long-term disruption of central nervous system development far beyond an isolated neural tube defect of the spine. It has recently been reported that brain pathology is always present in patients with SBA, and less frequently in those with SBO. The relationship between brain malformations and decreased cognitive function in patients with spina bifida aperta has also been observed⁽¹⁵⁾.

The currently cited model for cerebral malformations and hydrocephalus in spina bifida describes the circulation of cerebrospinal fluid (CSF) through the open caudal end of the neural tube causing Chiari II, additional abnormal flow leading to hydrocephalus, which leads to abnormal neural development^(16-18).

The results of other studies suggest that a loss of the ventricular sheath can trigger brain pathology, and it is now recognized that the fetal ependymal epithelium plays a key role in the development of the brain as a secretory structure⁽¹⁹⁾. Accordingly, it has been shown that ependymal pathology – e.g., denudation – occurs before the onset of Chiari II and/or hydrocephalus^(20,21) and, therefore, may not be the result of hydrocephalus, as previously reported⁽²²⁾. This is further supported by results in mouse models showing that hydrocephalus can result from a primary alteration of the ependymal mucosa⁽²³⁾.

Understanding the mechanism of the occurrence of spina bifida and of the development of spinal or cerebral disorders is important for the general knowledge of the occurrence of the disease, but also for therapeutic approaches.

The aim of this study was to analyze the sonographic findings detected both in the first and second trimesters of pregnancy and correlate them with the morphopathological aspects of the defects.

Materials and method

The Department of Prenatal Diagnosis of the Cou­nty Emergency Clinical Hospital of Craiova, Romania, served as the site of this retrospective study. All eligible patients were enrolled between January 2018 and December 2023, and they signed an informed consent form for the usage of their personal and medical data.

We searched for all cases of SBA that were detected during the sonographic assessment of the first (11-13 weeks + 6 days; SBA-FT) and second trimesters of pregnancy (14-27 weeks + 6 days; SBA-ST). After obtaining the informed consent of the mother, the fetuses included in the study presented the sonographic criteria for spina bifida for both the first and second trimesters of pregnancy.

We encountered 12 cases of spina bifida in the first trimester of pregnancy and two cases in the second trimester of pregnancy. We categorized these cases by the type of SBA presented: 12 cases of SBA-FT, which included eight cases of MYS, three cases of MMC and one case of ME; and two cases of SBA-ST of the MYS type. We also included in this study five control cases of first-trimester miscarriages without structural abnormalities, in order to compare the structural aspects. We also analyzed data such as mother’s age and periconceptional folic acid intake, using Microsoft Excel. 

All scans were conducted using GE Healthcare’s Voluson E10 and E8 devices. The scans were conducted using transvaginal (5-9 MHz) and transabdominal (3-9 MHz) transducers. For the most cases from the second trimester, a transabdominal scan was performed. For an optimal assessment of the anatomy of some elements of the fetal central nervous system, a transvaginal approach was used.

For cases detected during the first trimester, we used the standard protocol of the first-trimester screening guidelines for early assessment of the central nervous system (CNS)⁽²⁴⁾. This protocol included the following views: a) sagittal plane of the brain, which included an assessment of the fourth cerebral ventricle (intracranial translucency – IT); b) longitudinal aspects of the fetal spine in 2D and 3D imaging; c) cross-sections of the brain, that showed the shape and contour of the fetal skull, choroid plexus, and the filling of lateral cerebral ventricles; the presence of the third ventricle and Sylvius aqueduct; and the brain stem in relation to the occiput bone. In cases with spina bifida, there are changes in intracranial translucency, changes in the parallelism of the cerebral peduncles, BS-BSOB distance, choroid plexus area in relation to the cranial circumference and dried brain. In the second trimester, we identified the cases of spina bifida by analyzing the structure of the spine and the “banana sign” and “lemon sign” in the transverse section of the brain.

In the cases of SBA-FT, after informing the couples, they requested the medical termination of pregnancy. The two patients diagnosed with SBA-ST did not have any prior investigation during the current pregnancy, and the diagnosis was made following a miscarriage.  All fetuses were then sent to the Pathology Department of the County Emergency Clinical Hospital of Craiova. A fetal necropsy was performed by a mixed team of pathologists and gynecologist.

Consequently, it is essential to obtain pathologic and histopathologic evidence, and consistently classify spina bifida based on its most accurate phenotype.

In order to confirm the correlation between the ultrasonographic and morphopathological aspects of NTD of the spine, we included in paraffin blocks tissue samples from the areas of interest, sectioned them using a microtome to a thickness of 4-5 microns, applied them on slides treated with Poly-L-Lysine, subsequently deparaffinized and rehydrated them with alcohol at decreasing concentrations of 100%, 90% and 70%, until we reached distilled water and histologically stained them using the classical Hematoxylin-Eosin (HE). By immunohistochemical techniques based on standard protocols, using different antibodies (Table 1), we were able to visualize the nerve cells, neuronal and glial cell distribution in the tissue. We used a Nikon Eclipse microscope for scanning and analyzing the slides obtained.

Results

Associated risk factors

Analyzing the age of the patients with fetuses diagnosed with SBA, we observed that it varied as follows: in the case of SBA-FT with MYS, it ranged between 18 and 31 years old, with a mean age of 23.25 years old (±4.89 years). In cases of SBA-FT with MMC, it ranged between 23 and 36 years old, with a mean age of 28.66 years old (±6.66 years); the mean age of patients with SBA-FT with ME was 29 years old. In cases of SBA-ST with MYS, the mean age of the patients was 30 years old (±5.66 years). In the control cases, the patient’s age ranged between 23 and 35 years old, with a mean value of 30 years old (±4.47 years) – Figure 1. Therefore, there was no significant difference regarding the age of the patients included in the study.

Regarding periconceptional folic acid intake, we observed that none of the patients in the pathologic groups (SBA-FT, SBA-ST) took any folic acid. All patients with fetuses without abnormalities (F) took folic acid periconceptional.

Ultrasonographic features

The ultrasonographic images we have presented provide an accurate assessment of the markers suggested for the early detection of SBA. In healthy fetuses, the fourth ventricle is presented in the mid sagittal plane as an intracranial translucency (IT) that is parallel to the nuchal translucency (NT) and is delimited by two echoic lines. The choroid plexus of the fourth ventricle is represented by the posterior one, while the anterior one aligns the posterior part of the brainstem. In cases with SBA, the cerebellum migrates posteriorly and downward due to displacement of the CSF fluid, obliterating the fourth ventricle and the cisterna magna (CM) – Figure 2. The brainstem thickens and the gap between the occipital bone shortens as a result of this anatomical abnormality. Thus, the ratio between the sagittal diameter of the brainstem and its distance from the occipital bone is greater than 1. The “crash sign” appears as a posterior and caudal displacement of the midbrain relative to the occiput in SBA cases, present in all of the SBA-MYS cases included in the study (Figure 3A). In the analyzed cases, we observed the absence of IT in 9/12 cases of FT (75%), in one case it was below the 5^th percentile (8.33%), and in two cases of SB (16.66%) it had a normal appearance (Table 2). Regarding the BS/BSOB ratio, values above 1 were detected in all cases of SB-FT (Table 2). The “crash sign” was present in 10/12 cases (83.33%) and absent in 2/12 cases of SB-FT (16.66%). A vertebral defect was visualized only in 8/12 cases of SB-FT (66.66%) – Table 2. In SBA-ST cases, using the standard axial sections analysis (transventricular, transcerebellar and transthalamic section, and the analysis of the spinal cords in both longitudinal and transverse sections), we detected some defects and changes: frontal bone collapse, transcerebellar compression, ventriculomegaly, cisterna magna obliteration, and structural defect of the vertebral spine (Table 2). We observed the presence of specific signs, such as the “lemon sign”, which occurs when the bones at the front of the skull are sunken/flattened instead of rounded, and the “banana sign”, which refers to the shape of the cerebellum (Figure 3B, Table 2).

During the sagittal examinations of the spine, we observed the presence of a spinal defect in all cases of SBA-ST and only in 8/12 cases of SB-FT (66.66%) – Figure 4 A, B, aspects visualized with both 2D and 3D reconstructions (Figure 4 C, D).

Morphopathological aspects

At the necropsy, we observed spinal defects in both first- and second-trimester fetuses. The intracranial aspects were evident in the second-trimester fetuses. In cases of MYS, we observed the spinal defect without tegument coverage (Figure 5 A, B), aspects also confirmed histopathologically. These images show the absence of tegument in the spinal tissue prolapsed through the vertebral defect (Figure 5 C-F). By immunohistochemical staining with the anti-GFAP antibody, we observed glial cells only in the anterior portion of the spinal tissue (Figure 5G). We observed a similar aspect regarding the distribution of nervous fibers immunolabeled with anti-NFP antibody in the same area (Figure 5H). In cases of MMC, at the macroscopical examination, we observed vertebral tissue covered by meninges, protruding through the vertebral defect (Figure 5I). At the microscopic examination, using the classic HE stain, we observed disorganization of the spinal nervous tissue, with the presence of external meningeal cover (Figure 5J). Glial cells labeled with the anti-GFAP antibody (Figure 5K) and nervous fibers labeled with the anti-NFP antibody (Figure 5L) were disorganized in the spinal cord. Compared to the SBA cases, the control groups showed a spinal cord covered by meninges and tegument (Figure 5M), the glial cells were scattered around the anterior horns and less around the posterior horns (Figure 5N), and nervous fibers were located around the gray matter of the brain (Figure 5O).

At the necropsy examination of the second-trimester fetus, we observed the spinal defects visualized by ultrasound and cerebral fossa anomalies with protrusion of the cerebellum through the opening in the bottom of the skull (Figure 6 A-F).

Discussion

Antenatal ultrasonographic diagnosis allows the examination of fetal anatomy as early as the first trimester, being currently performed worldwide. Although ultrasound systems have improved significantly and examiners have perfected their technique, some abnormalities may be missed^(25-28).

Some ultrasound findings may be inaccurate, but post-abortion or post-mortem examination of the fetus can characterize the spinal defect accurately^(29,30).

Our study included a series of 12 cases of spina bifida during the first trimester of pregnancy and two cases during the second trimester of pregnancy, out of which there was discordance between ultrasonographic and microscopic examination in two out of 12 cases during the first trimester of pregnancy, where it was microscopically observed that the nervous tissue was exposed to amniotic fluid without any sheath covering. We also took into account the importance of comparing histologic and sonographic findings, as the final diagnosis and type of OSB can be accurately revealed microscopically^(31-37).

Our study reported a 100% detection rate of OSB (12/12 cases diagnosed in the first trimester and 2/2 cases diagnosed in the second trimester of pregnancy) detected by ultrasonography using the aforementioned protocol of CNS assessment, combining intracranial markers with direct examination of the spine.

In 2009, Chaoui et al.⁽³⁸⁾ first described intracranial translucency (IT) as a screening marker used in the first trimester of pregnancy to detect spina bifida. In Orlandi et al. work⁽³⁹⁾, IT was one of the features studied in the literature review, which included 11 small studies. Intracranial translucency appearance ranged from not measurable in all cases of OSB in the studies of Chaoui et al.⁽³⁸⁾, Adiego et al.⁽⁴⁰⁾ and Iliescu et al.⁽⁴¹⁾, to normal appearance in all cases of OSB in the Scheier et al.⁽⁴²⁾ and Solt et al.⁽⁴³⁾ studies (Table 3).

Compared to other studies, in this research IT was not visualized in 9/12 cases (75%), while 16.66% (2/12 cases) of cases had normal sized intracranial translucency. IT was measured below the 5^th percentile in 8.33% (1/12 cases) of OSB cases.

Lachman et al.⁽⁴⁸⁾ conducted a study that included 1000 unaffected fetuses and 30 cases of OSB at a gestational age of 11-13 weeks + 6 days. The research showed that BS/BSOB was below 1 in all cases of normal fetuses and always greater than 1 in the afflicted ones. Iuculano et al.⁽⁴⁹⁾ also reported BS/BSOB greater than 1 in all cases of OSB. In our study, the BS/BSOB ratio was greater than 1 in all 12 cases of OSB, proving to be a reliable marker, with 100% accuracy in detecting OSB in the first trimester.

The direct visualization of the spinal lesions in the first trimester could not be possible in all cases (only in 8/12 cases – 66.66% of the cases in our study). We determined that direct ultrasound examination of the spine in the first trimester was not a reliable marker for a proper diagnosis of OSB. After a broad review of several studies, we agreed that the screening for OSB in the first trimester should be based on intracranial markers^(38,41,49), combined with the visualization of spinal integrity and skin coverage in 2D and 3D imaging. In contrast to the first trimester, the second-trimester spinal defects are visualized in all cases analyzed.

A relatively new ultrasonographic marker has been described in the axial view of the fetal brain, termed the “crash sign”. It occurs through posterior displacement and distortion of the midbrain toward the occipital bone. In the same section, narrowing or absence of the Sylvius aqueduct may be present in cases of SBO^(50,51). Ushakov et al.⁽⁵²⁾ reported a 90.6% detection rate by “crash sign” in cases of OSB in the first trimester, and the study by Zhu et al.⁽⁵³⁾ revealed a detection rate of 85.7% by this sign and also highlighted a particular situation in which the “crash sign” present was associated with three-line view. Comparatively, in our study, in the two cases of OSB with the “crash sign” absent, a four-line view was observed.

On the medio-sagittal section of the brain, the posterior cerebral landmarks included three anechoic spaces and four lines, giving a four-line view. The four lines are, as previously described, from bottom to top: occipital bone, choroid plexus of the fourth ventricle, anterior border of the fourth ventricle, and posterior border of the sphenoid bone⁽⁵³⁾. Also, even for our cases with absent “crash sign”, the BS/BSOB ratio was greater than 1. Our study reports a slightly lower detection rate of OSB using the “crash sign” (83.33%) than in these previous works^(39,52).

In contrast to the direct visualization of the spinal lesion, the search for these intracranial markers includes taking measurements of posterior brain structures, and that may prolong the examination time^(24,54). However, IT, BS diameter, or BS/BSOB ratio can be assessed in the sagittal plane commonly used for NT measurement and nasal bone assessment. Our protocol included all these intracranial markers, without reporting any additional time.

Different from the ultrasonographic examination of OBS in both first trimester and second trimester, the direct examination of the spinal defect can reveal more accurately the type of spina bifida, the coverage level with a meningeal membrane or integument. This is why all cases included in this study were histologically and macroscopically examined at the time of the necropsy.

For our study, a particular aspect was the examination of the nervous structures of the spinal defect: the spinal cord structure, the presence of neurons with different anarchic arrangements in the nervous tissue, the presence of spinal nerves and glial cells dispersed unevenly in the spinal cord, the presence of neurons with different anarchic distributions in the nervous tissue, the presence of spinal nervous fibers and glial cells unevenly dispersed in the nervous tissue mass, and the absence of the integument in MYS cases.

Also, abnormal development of the central nervous system caused by neural tube defects (NTDs), such as anencephaly, spina bifida and encephaloceles, is a major cause of stillbirths and neonatal deaths. It is also an important cause of lifelong physical disability for the surviving children. Neural tube defects are common clinical congenital malformations caused by an incomplete or an abnormal closure of the neural tube in the embryo, with an incidence of 1.86‰ in humans^(55,56). Human neural tube defects are associated with folic acid deficiency, which represents an important cause of NTD⁽⁵⁶⁾. Folic acid deficiency in pregnant women increased the risk of neural tube defects⁽⁵⁷⁾, spina bifida in our study. As previously mentioned, none of the pregnant women included in the study received folic acid prior to conception or during the pregnancy.

Conclusions

This study highlights that spina bifida is associated with brain abnormalities detected in both the first trimester and the second trimester. Combined intracranial ultrasonographic intracranial markers, such as abnormal IT, the presence of “crush sign” and especially the BS/BSOB ratio greater than 1, provide high diagnosis accuracy in the first trimester. The direct visualization of the spinal defect has better accuracy in the second trimester.

The ultrasound visualization of the spinal defect and the search for intracranial ultrasound markers are important in the early detection of spina bifida.

Necropsy and histologic examination provide additional information for the diagnosis and classification of spina bifida phenotype.

Acknowledgment. The microscopic images have been acquired in the Research Centre for Microscopic Morphology and Immunology, University of Medicine and Pharmacy of Craiova, Romania (manager: Laurențiu Mogoanță, Professor, MD, PhD).

Autor corespondent: Cristina-Maria Comănescu E-mail: cristinacomanescu85gmail.com

CONFLICT OF INTEREST: none declared.

FINANCIAL SUPPORT: none declared.

This work is permanently accessible online free of charge and published under the CC-BY.