| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
by the American Institute of Ultrasound in Medicine J Ultrasound Med 27:33-38 • 0278-4297 Supratentorial Abnormalities in the Chiari II Malformation, IThe Ventricular "Point"Department of Radiology, University of California, San Francisco, California USA. Address correspondence to Roy A. Filly, MD, Department of Radiology, University of California, 505 Parnassus Ave, L-374, San Francisco, CA 94143-0628 USA., E-mail: roy.filly{at}radiology.ucsf.edu
Objective. The sonographic diagnosis of fetal myelomeningocele has improved dramatically over the past 20 years, mainly because the diagnostic focus has shifted from observation of the spinal abnormality to observation of cranial abnormalities. We describe an abnormality in the shape of the occipital horn that has received scant attention in the sonographic literature. The abnormality is a misshapen occipital horn that is "pointed" rather than rounded. This study analyzed the frequency of this observation and technical aspects that improve its visualization. Methods. From a database of obstetric sonograms, we searched for all cases in which open fetal myelomeningocele was sonographically detected. The search covered the period from 1999 to 2007. We then retrospectively reviewed the fetal intracranial findings in each case, with special attention to the contour of the occipital horn of the lateral ventricle. Results. A total of 89 fetuses were identified. The mean and median gestational ages were 22 weeks 4 days and 22 weeks 2 days, respectively. Of the 89 cases, 62 (70%) had an occipital horn with a pointed shape. This abnormality was better seen in cases judged to be well visualized and in cases in which digital video clips were available for review. The abnormality was more frequently seen in fetuses at less than 24 weeks’ gestation and in fetuses without ventriculomegaly. Finally, it was seen as commonly among fetuses with mild posterior fossa deformations as in those with more severe distortions. Conclusions. The occipital point is a common supratentorial feature of the Chiari II malformation. It is seen more commonly in fetuses at less than 24 weeks and in fetuses with normal-sized ventricles. As well, it is seen as commonly among fetuses with mild posterior fossa deformations as in those with more severe distortions. Therefore, its usefulness is enhanced as an observation in the armamentarium of cranial findings for detection of fetal myelomeningocele.
Key Words: abnormal ventricle • Chiari II malformation • fetal sonography • myelomeningocele
The ability of sonography to show fetal myelomeningocele has improved steadily. The greatest improvements occurred when the focus of diagnosis shifted from observation of the spina bifida abnormality to the cranial findings associated with the Chiari II malformation.1–8 The Chiari II malformation is, for all intents and purposes, found exclusively in fetuses with myelomeningocele. Therefore, identification of features of the Chiari II malformation virtually ensures that spina bifida will be discovered if assiduously sought. The features of the Chiari II malformation that have been most useful are the infratentorial findings. These include effacement of the cisterna magna2 and deformation of the cerebellum, the so-called "banana" sign,1 although other infratentorial abnormalities are commonly observed postnatally.5–8 Although the dominant features of the Chiari II malformation relate to the hindbrain, many supratentorial abnormalities have also been described.5–8 Included in these are callosal dysgenesis, a small third ventricle, enlarged inter-thalamic adhesions, a "beaked" tectum, polymicrogyria, heterotopias, skull deformities (the "lemon" sign),3 colpocephaly, and other causes of ventriculomegaly. Important among these is ventriculomegaly because visualization of the lateral ventricle is required on all routine sonograms.9 Unfortunately, ventriculomegaly is considerably less common before 24 weeks than after 24 weeks in fetuses affected with myelomeningocele.10 That the ventricular margin may have a "pointed" appearance is well known in the imaging evaluation of neonates with myelomeningocele.5–8 However, this observation, typically referred to as an inferior point of the frontal horn, is usually observed only in coronal sections and is attributed to prominence of the caudate nuclear head.5 More recently, an "angular" or pointed deformity of the occipital horns of the lateral ventricles has been described in the fetal magnetic resonance imaging literature.11,12 Only 1 observation of this feature has been noted on fetal sonography.11 Additionally, there was a moderately large difference in the frequency of this observation between the 2 above-mentioned studies: 28% in the study by Rickard et al12 and 92% in the study by Levine et al.11 Although we were unaware of these magnetic resonance imaging observations when we began our study of sonography of the pointed deformity of the occipital horn in fetuses with myelomeningocele, we acknowledge that the observation we will describe herein is the same as previously described by Levine et al11 and Rickard et al.12
From a database of obstetric sonograms, we searched for all cases in which open fetal myelomeningocele was sonographically detected. The search covered the period from 1999 to 2007. We then retrospectively reviewed the fetal intracranial findings in each case. Fetal imaging was conducted with 2- to 4-MHz vector and 2- to 6-MHz curved array transducer formats with both selectable focus and frequency as appropriate for the size of the pregnancy (Siemens Medical Solutions USA, Inc, Mountain View, CA). Images included both transverse axial planes of sections of the intracranial anatomy as well as coronal and sagittal planes of sections when feasible. However, the observations described herein were made on transverse axial images. Features associated with the Chiari II malformation were evaluated. In particular, 1 author (R.A.F.), an experienced prenatal sonologist, individually evaluated each case to determine the morphologic characteristics of the occipital horn contour. Notations were made as to whether the contour was rounded (normal) or pointed, a feature of the Chiari II malformation. Other features that were analyzed and recorded included the highest vertebral level of myelomeningocele, the grade of the Chiari II malformation (mild, mild/moderate, moderate, moderate/severe, or severe), the presence of ventricular dilatation, the diameter of the ventricles, the gestational age of the fetus at the time of the study, whether appropriate images of the ventricle were seen only on a digital video clip, and whether there were technical difficulties due to fetal position or patient body habitus. Inward curvature of the frontal bones, the so-called lemon sign, was not evaluated. All statistical comparisons used the Fisher exact test, and for multiple comparisons, the Bonferroni correction was applied. The Committee on Human Research at the University of California, San Francisco, approved this research project (approval No. H7798–31138–01).
A total of 89 fetuses with open myelomeningocele were identified. Fetuses with skin-covered spina bifida lesions were not evaluated. Gestational ages ranged from 17 weeks 1 day to 34 weeks 5 days. The mean gestational age was 22 weeks 4 days, and the median gestational age was 22 weeks 2 days. Because the study was retrospective, no specific effort to assess the shape of the occipital horn was made. Therefore, the quality of the imaging in each case was judged on the basis of the ease of assessing the occipital horn shape. Sixty-three cases were judged to have good visualization, and 26 had suboptimal visualization.
Of the 89 cases, 62 (70%) had an occipital horn of the lateral ventricle with a pointed shape (Figures 1
We further divided cases into 2 gestational age ranges and determined the occurrence of the occipital point within each group. The 2 groups were 17 weeks 1 day through 23 weeks 6 days and 24 weeks 0 days through 34 weeks 5 days. In the earlier group, the occipital point was identified in 53 (75%) of 71 cases. In the latter group, the occipital point was seen in 9 (50%) of 18 cases. According to the Fisher exact test, the difference between the 2 age groups (ie, the greater likelihood of seeing the ventricular point in fetuses at <24 weeks) was borderline significant (P = .05). The severity of the Chiari II malformation was categorized as follows: mild, mild/moderate, moderate, moderate/severe, and severe. Fifty-two cases were severe; 17 were moderate/severe; 10 were moderate; 1 was mild/moderate; and 5 were mild. In 4 cases, because of technical factors, the severity of the Chiari II malformation could not be adequately assessed. Among the 4 cases in which the severity of the Chiari II malformation could not be assessed, 2 (50%) had an occipital point. Among the 5 mild cases, 4 (80%) had an occipital point. The remainder of the categories were as follows: 1 (0%) mild/moderate case did not have an occipital point; of 10 moderate cases, 8 (80%) had an occipital point; of 17 moderate/severe cases, 13 (76%) had an occipital point; and of 52 severe cases, 35 (67%) had an occipital point. Of the 6 between-group (Chiari II severity) comparisons, none were significant after correcting for multiple comparisons (Bonferroni correction). However, the Fisher exact test showed a significant difference between the likelihood of visualizing an occipital point in the comparison of mild versus severe Chiari II (P = .01) and in the comparison of mild versus moderate Chiari II (P = .03).
Ventricular size ranged from 0.6 to 3.43 cm. The mean ventricular size was 1.24 cm. Fifty-three cases had ventriculomegaly; 28 had a normal ventricle size; and in 8 of the studies, suboptimal visualization prevented a ventricular measurement from being made. Twenty-two (79%) of the 28 fetuses with a normal ventricular size had an occipital point (Figure 1 The most cephalad observed level of myelomeningocele was at the T9 vertebral level; the most caudad was at the S3 vertebral level. Occurrence of an occipital point was not assessed relative to the spinal level of myelomeningocele.
The diagnosis of myelomeningocele in a fetus is important for many reasons. It provides the parents with an opportunity to consider pregnancy termination. Among parents electing to continue the pregnancy, adequate counseling and psychological preparation can be provided. Plans for delivery at a tertiary center and consideration of surgical delivery and its potential benefit can be addressed. Currently, patients may be eligible for in utero surgery in a prospective trial funded by the National Institute of Child Health and Human Development. Indeed, if that trial shows a positive benefit, then prenatal diagnosis of myelomeningocele will be even more important. Because this entity is relatively common and of great importance, large screening programs13 for detection of myelomeningocele are conducted across the United States and in many other parts of the world. Fortunately, detection of fetal myelomeningocele is possible in an extremely high percentage of cases.13 As noted earlier, for sonographic detection, the greatest improvements occurred when the focus of diagnosis shifted from observation of the spina bifida abnormality to the cranial findings associated with the Chiari II malformation.1–4 The cranial findings associated with the Chiari II malformation are found exclusively in fetuses with myelomeningocele. Therefore, identification of features of the Chiari II malformation virtually ensures that myelomeningocele is present. Although the infratentorial (posterior fossa) findings are paramount among the features of the Chiari II malformation, the posterior fossa is sometimes difficult to appropriately observe because shadowing from the petrous ridges interferes with visualization.1,4 This may then be mistaken for a technical problem rather than a major abnormality. As well, the posterior fossa deformation may be mild, as was noted in 5 of our cases. For these reasons, supratentorial abnormalities are also important in the prenatal recognition that myelomeningocele is present. Probably chief among these are the so-called lemon sign (inward scalloping of the frontal bones) and ventriculomegaly. However, the lemon sign is frequently not present in later pregnancies3 and can be seen in healthy fetuses3,14 and in other conditions.14 As well, ventriculomegaly may be absent, particularly before 24 weeks,10 and when present is nonspecific for myelomeningocele. However, there are multiple supratentorial features of the Chiari II malformation, as already noted,5–8 that can also be visualized on prenatal sonograms.
We observed that the shape of the occipital horn was abnormal in fetuses with myelomeningocele in that it terminated with a pointed contour instead of the normal rounded contour (Figures 1
Our results indicate that the occipital point is also a common supratentorial feature of the Chiari II malformation on fetal sonograms. Of the 89 fetuses in our study, 62 (70%) had an occipital horn of the lateral ventricle with a pointed shape. Importantly, it is seen more commonly in fetuses at less than 24 weeks than in fetuses at greater than 24 weeks (75% compared with 50%). As well, the finding is seen in fetuses with normal-sized ventricles (Figure 1 Although our data cannot definitively confirm this, it appears that, if sought, the occipital point would be even more common than the 70% noted in our results. That is because ours was a retrospective study. We had not sought to show this finding. Among those studies with suboptimal visualization, 13 (50%) of 26 had an occipital point. In those studies with good visualization, 49 (78%) of 63 had an occipital point. Therefore, we think that the finding is present in more than 70% of fetuses. As well, we noted that in 3 cases the finding was only seen on a digital video clip, indicating again that when sought, the true prevalence of the finding may be greater than our results indicate in this retrospective review. Therefore, our data support the notion that the frequency of this finding is likely to be closer to the results of Levine et al11 (92%) than those of Rickard et al12 (28%). Indeed, the retrospective nature of our study was a considerable weakness that should be addressed by future prospective analyses. Other abnormalities of ventricular shape may be recognizable on prenatal sonograms. Fujisawa et al15 described an alteration in the shape of dilated ventricles in fetuses with myelomeningocele. These abnormalities of shape were described on coronal planes of sections, not axial planes, and were termed a "downward triangular shape" and a "quadrilateral angular shape." Although the terminology sounds similar, we do not think that the findings of Fujisawa et al15 are the same as those described in this study or in those of Levine at al11 and Rickard et al.12 In conclusion, the occipital point is a common supratentorial feature of the Chiari II malformation. It is seen more commonly in fetuses at less than 24 weeks and tends to be seen more commonly in fetuses with normal-sized ventricles (although statistical significance was not achieved). As well, it is seen as commonly or more commonly among fetuses with mild posterior fossa deformations than in those with more severe distortions. Therefore, its usefulness is enhanced as an observation in the armamentarium of cranial findings for detection of fetal myelomeningocele.
Received September 17, 2007, from the Department of Radiology, University of California, San Francisco, California USA. Revision requested October 1, 2007. Revised manuscript accepted for publication October 3, 2007.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Top
Abstract
Introduction
