Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51763339
CITATIONS READS
14 256
8 authors, including:
Martha Finn
Monash Ultrasound for Women
11 PUBLICATIONS 86 CITATIONS
SEE PROFILE
All content following this page was uploaded by Martha Finn on 24 November 2015.
K E Y W O R D S: aqueduct of Sylvius; first-trimester ultrasound; midbrain; neural tube defect; prenatal diagnosis
ABSTRACT INTRODUCTION
Objectives To describe a new first-trimester sonographic The diagnosis of neural tube defect is generally made at
landmark, posterior displacement of the midbrain and the mid-trimester fetal morphology scan, prompted by the
aqueduct of Sylvius, which may be useful in early identification of cranial signs. The lemon sign has been
screening for neural tube defects. reported in 98% of spina bifida fetuses examined before
24 weeks and in 13% of cases in the third trimester, and
Methods This was a prospective study of 457 normal the banana sign has been reported in 72% and 81% of
fetuses at 11 + 0 to 13 + 6 weeks gestation. We measured these cases, respectively1 . These features, however, are
the distance from the posterior border of the aqueduct rarely present in the first trimester. Furthermore, those
of Sylvius to the anterior border of the occiput (AOS-to- fetal spinal abnormalities which are diagnosed in the first
occiput distance) in the axial plane and created a reference trimester are usually severe and frequently associated with
range. In the nine fetuses with abnormal midbrain position other major defects, and the diagnosis of isolated defects
identified in the first trimester and with neural tube is difficult. Thus, the early diagnosis of spina bifida is
defect subsequently confirmed, we analyzed ultrasound challenging and a second-trimester scan is still regarded
images to determine the position of the aqueduct of as necessary to detect most cases2 .
Sylvius. In an attempt to address this diagnostic difficulty,
Results The lower limit of normal AOS-to-occiput Chaoui et al.3 recently described the intracranial translu-
distance (mean minus 2 SD) ranged from 1.7 mm at a cency, the anteroposterior dimension of the fourth
crownrump length (CRL) of 45 mm to 3.7 mm at a CRL ventricle in the sagittal plane, as a new first-trimester sono-
of 84 mm. In the nine cases with abnormal position of the graphic marker for open spina bifida. Another marker,
midbrain and confirmed neural tube defect, juxtaposition decreased frontomaxillary facial angle, has also been
of the midbrain to the occiput was the clue to diagnosis proposed4 . The brainstem diameter and brainstem-to-
of the spinal abnormality. In all nine cases, the AOS- occipital bone distance (BSOB) have also been described
to-occiput distance was below the established normal in fetuses with open spina bifida at this gestation5 . More
range. recently, the first-trimester transcerebellar and cisterna
magna diameters in the axial plane have been described
Conclusions Examination of the midbrain in an axial as potentially valuable signs6 . The search continues for
plane may prove a reliable marker for the first-trimester a first-trimester sonographic landmark that is compa-
diagnosis of neural tube defects. In contrast to recently rable to the easily recognizable and highly predictive
reported subtle changes in the mid-sagittal view of the mid-trimester banana shaped cerebellum.
posterior cranial fossa, axial imaging of the midbrain The aims of this study were to define quantitatively the
reveals striking displacement of this structure, with virtual position of the first-trimester fetal midbrain by establish-
juxtaposition to the occiput, in fetuses with confirmed ing a normal range for the distance between the aqueduct
open spina bifida. This anatomical distortion of the of Sylvius and the occiput (AOS-to-occiput distance) and
midbrain can be quantified by measurement of the AOS- to determine whether this distance is reduced in fetuses
to-occiput distance. Copyright 2011 ISUOG. Published with open spina bifida, which could suggest its potential
by John Wiley & Sons, Ltd. as a marker for the early detection of neural tube defects.
Correspondence to: Dr M. Finn, Monash Ultrasound for Women, Healthbridge Hawthorn Private Hospital, 50 Burwood Road, Hawthorn,
Melbourne, Victoria 3122, Australia (e-mail: mfinn@monashivf.com)
Accepted: 15 August 2011
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER
Aqueduct of Sylvius and early NTD screening 641
METHODS (a)
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2011; 38: 640645.
642 Finn et al.
RE SULTS
Follow-up was unavailable for 15 patients, who had 0
40 50 60 70 80 90
moved interstate or overseas. Using the strict scoring Crownrump length (mm)
system, 457 images were defined as having optimal
imaging quality, 153 (33.5%) of which were obtained Figure 2 Correlation between aqueduct of Sylvius (AOS)-to-occiput
using transvaginal ultrasound and 304 (66.5%) using
transabdominal ultrasound. In all these fetuses, mid-
distance and crownrump length in 457 normal fetuses ( ). The
line of best fit is shown (AOS = 0.095 CRL 2.1, R2 = 0.487),
with 95% prediction interval. The AOS-to-occiput distance of nine
trimester follow-up ultrasound confirmed a normal neural
fetuses with neural tube defect are shown ().
sonogram.
The median maternal weight was 67.5 kg, with 5th
and 95th percentiles of 50 kg and 94 kg, respectively. The Table 1 Aqueduct of Sylvius (AOS)-to-occiput distance according
to crownrump length (CRL)
mean (SD) for the Gaussian distribution of gestational
age at examination was 89 (3) days and of crownrump
AOS-to-occiput distance (mm)
length (CRL) was 67 (7.6) mm.
We observed a positive linear correlation between CRL (mm) n Mean 2 SD Mean Mean + 2 SD
AOS-to-occiput distance and gestational age and a
stronger correlation between AOS-to-occiput distance and 4549 8 1.7 2.3 2.6
CRL (Figure 2). The linear regression of AOS-to-occiput 5054 20 2.0 2.8 3.6
distance (in mm) as a function of CRL (in mm) was: 5559 55 2.1 3.5 4.9
6064 77 2.5 3.9 5.3
AOS-to-occiput distance = 0.095 CRL 2.1. Analysis 6569 118 2.6 4.2 5.8
of variance revealed that the model was statistically 7074 110 3.1 4.7 6.3
significant (Pearsons correlation coefficient R = 0.698, 7579 49 3.6 5.2 6.8
P < 0.001). 8084 20 3.7 5.7 7.7
A Gaussian distribution was identified for the AOS-to-
occiput distance for CRL intervals of 5 mm. The normal
ranges (mean and 2 SD, Table 1) were constructed based and it was subsequently recognized as having a spinal
on parametric data analysis. As our practice encourages defect at the mid-trimester ultrasound examination (Case
referral of patients after 12 weeks gestation to view 1, Table 2). Subsequent recognition of a similar first-
optimally the fetal anatomy, the majority of fetuses in trimester midbrain appearance in five fetuses led to
our study had CRL > 55 mm. The lower limit of normal diagnosis of a neural tube defect within 2 weeks. In the
ranged from 1.7 mm at a CRL of 4549 mm to 3.7 mm seventh fetus, although the midbrain abnormality was
at a CRL of 8084 mm. identified in the first trimester, the patient chose not to
In the separate study of 22 fetuses, low intraobserver return for review until 20 weeks, when the spinal lesion
variation in measurements was found. The intraclass was diagnosed. In a further two fetuses, in which the spinal
correlation coefficient (ICC) and 95% confidence limits defect was diagnosed at mid-trimester, the abnormal
for Observer A was 0.86 (0.580.96), for Observer B midbrain appearance was detected retrospectively, on
was 0.95 (0.860.98) and for Observer C was 0.97 review of the first-trimester ultrasound study. The relevant
(0.890.99). The variation between Observers A and clinical information for all nine fetuses with confirmed
C and Observers B and C was similarly low, with ICCs neural tube defect is outlined in Table 2. In all nine
of 0.89 (0.80.97) and 0.96 (0.880.99), respectively. fetuses, the midbrain appeared to be virtually juxtaposed
These fetuses were representative of those in the main to the occiput at the first-trimester scan, and the AOS-
study, the CRL ranging from 53 to 83 (mean, 65) mm. to-occiput distance was below the normal range. The
Among the nine abnormal cases, the sonographic characteristic feature of the posteriorly displaced midbrain
observation of an unusual midbrain appearance in one and associated spinal defect in two affected fetuses (Cases
at 12 weeks gestation flagged it as being high risk, 2 and 6) are illustrated in Figures 3 and 4.
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2011; 38: 640645.
Aqueduct of Sylvius and early NTD screening 643
Table 2 Findings at first-trimester ultrasound examination and follow-up in nine cases of spina bifida associated with first-trimester
posterior displacement of the fetal midbrain
First-trimester ultrasound
GA CRL MW AOS
Case (weeks) (mm) (kg) (mm) Ultrasound findings Follow-up
3D, three-dimensional; AOS, aqueduct of Sylvius-to-occiput distance; CRL, crownrump length; GA, gestational age; MW, maternal
weight; NTD, neural tube defect; TOP, termination of pregnancy.
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2011; 38: 640645.
644 Finn et al.
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2011; 38: 640645.
Aqueduct of Sylvius and early NTD screening 645
fourth ventricle. Yet, all nine cases shared a common abnormal fetuses. Just as the banana-shaped cerebellum
positive ultrasound feature: an elongated and pushed is a striking feature in the axial plane at mid-trimester, so
back midbrain, to the point of virtual juxtaposition with too is the midbrain position in the first trimester.
the occiput. While this presented a striking visual diag-
nostic clue, we further sought to identify a sonographic
marker with which to quantify this finding. A suitable ACKNOWLEDGMENTS
objective assessment of posterior midbrain displacement We would like to thank Shirley Sanderson, Joan Steen and
was determined to be the distance between the posterior Kay Read for valuable administrative assistance.
border of the sharply defined aqueduct of Sylvius, which
traverses the midbrain, and the anterior border of the
occiput. In our series, the lower limit of the established REFERENCES
normal range (mean minus 2 SD) ranged from 1.7 mm 1. Van den Hof M, Nicolaides KH, Campbell J, Campbell S.
at a CRL of 45 mm to 3.7 mm at a CRL of 84 mm, Evaluation of the lemon and banana signs in one hundred
while in all nine cases of confirmed neural tube defect, the and thirty fetuses with open spina bifida. Am J Obstet Gynecol
AOS-to-occiput distance was below the normal range for 1990; 162: 322327.
2. Sepulveda W, Wong AE, Fauchon DE. Fetal spinal anomalies in
gestational age. This observation of virtual juxtaposition
a first trimester sonographic screening program for aneuploidy.
of the midbrain to the occiput, supported by objective Prenat Diagn 2010; 30: 107114.
AOS-to-occiput distance measurements, is much easier, 3. Chaoui R, Benoit B, Mitkowska-Wozniak H, Heling KS, Nico-
in our opinion, than interpretation of linear hypoechoic laides KH. Assessment of intracranial translucency (IT) in the
regions in the mid-sagittal image. detection of spina bifida at the 1113-week scan. Ultrasound
Obstet Gynecol 2009; 34: 249252.
In practice, it must be remembered that the cut-off level
4. Lachmann R, Picciarelli G, Moratalla J, Greene N, Nico-
for each CRL interval is not diagnostic of spina bifida, laides KH. Frontomaxillary facial angle in fetuses with spina
but serves as a guide below which one should consider bifida at 1113 weeks gestation. Ultrasound Obstet Gynecol
the possibility of neural tube defect. The diagnosis of 2010; 36: 268271.
neural tube defect is based on the finding of interrupted 5. Lachmann R, Chaoui R, Moratalla J, Picciarelli G, Nico-
laides KH. Posterior brain in fetuses with open spina bifida
neural arches in three orthogonal scanning planes. The
at 11 to 13 weeks. Prenat Diagn 2011; 30: 103106.
first-trimester diagnostic ultrasound findings present sim- 6. Egle D, Stroble I, Weiskopf-Schwendinger V, Grubinger E,
ilarly to those in the second trimester but may be difficult Kraxnser F, Mutz-Dehbalaie IS, Strasak A, Scheier M. Appear-
to visualize because of their relatively smaller size. Visu- ance of the fetal posterior fossa at 11 + 3 to 13 + 6 gestational
alization of the flat lesion of rachischisis, as seen in two weeks on transabdominal ultrasound examination. Ultrasound
Obstet Gynecol 2011; DOI: 10.1002/uog.8957. [Epub ahead of
of our cases, is a particularly difficult challenge. Indeed,
print].
in these cases it was the strong belief in the midbrain sign 7. Heath V, Nicolaides KH. Sonographic features of chro-
(i.e. its juxtaposition to the occiput,) that led the oper- mosomal defects. First trimester ultrasonography. In The
ator to persist and eventually identify the spinal lesion. 1113+6 Weeks Scan (Diploma in Fetal Medicine Series),
In the cases of spina bifida described in our study, when K. H. Nicolaides (ed). Fetal Medicine Foundation: London,
2004; 4557. [Available as download in The NT Book
the spinal defect was believed to be present at the initial
at http://www.fetalmedicine.com/fmf/online-education/01-11-
first-trimester examination, the patient was requested to 136-week-scan/].
return for review within 14 days to confirm the diag- 8. OsiriX: An Open-Source Software for Navigating in Multidi-
nosis. A confident diagnosis of the spinal abnormality mensional DICOM Images. J Digit Imaging 2004; 17: 205216.
afforded the patient the option of early termination of (http://www.springerlink.com/content/x57mul0frg64apmh/).
9. Blumenfeld Z, Siegler E, Bronshtein M. The early diagnosis of
pregnancy.
neural tube defects. Prenat Diagn 1993; 13: 863871.
The subtle changes in echogenicity of structures and 10. Buisson O, De Keersmaecker B, Senat MV, Bernard JP, Moscoso
the multiple parallel echogenic lines in the mid-sagittal G, Ville Y. Sonographic diagnosis of spina bifida at 12 weeks:
view of the posterior fetal brain have the potential to heading towards indirect signs. Ultrasound Obstet Gynecol
confuse. The strength of our proposed new marker for 2002; 19: 290292.
11. Solt I, Kim MJ, Acuna JG, Adeniji AB, Morales C, Rotmen-
first-trimester neural tube defects lies in the ease of
sch S. OC12.02: Spina bifida detection at 1114 weeks: system-
acquisition of the appropriate ultrasound plane during atic analysis of technical factors associated with false-positive
a routine cranial sweep in the axial plane (Videoclip and false-negative image interpretation. Ultrasound Obstet
S1) using the transabdominal or transvaginal approach, Gynecol 2010; 36(S1): 22.
the readily recognizable abnormal midbrain position seen 12. Solt I, Kim MJ, Adeniji AB, Acuna JG, Mirocha J, Rotmen-
sch S. OC12.05: Visualization and measurement of the cisterna
in this cranial sweep (Videoclips S24) and the ease
magna and maximal posterior fossa translucency for spina
of measurement of the distance between the aqueduct bifida screening at 1113 weeks gestation. Ultrasound Obstet
of Sylvius and the occipital bone in both normal and Gynecol 2010; 36(S1): 23.
Copyright 2011 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2011; 38: 640645.