Ultrasound Obstet Gynecol 2007; 29: 363–367
Published online 13 March 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.3983
Editorial
Prenatal exposure to ultrasound waves: is
there a risk?
J. S. ABRAMOWICZ
Department of Obstetrics and Gynecology, Rush University
Medical Center, 1653 West Congress Parkway, Chicago, IL 60612,
USA (e-mail: jacques− abramowicz@rush.edu)
In the 6th August 2006 issue of the Proceedings of
the National Academy of Sciences, Ang et al., from
Pasko Rakic’s well-known laboratory at Yale University,
published results of their research on neuronal migration
in the mouse embryo brain and the influence of prenatal
exposure to ultrasound1 . In this set of experiments, the
authors exposed immobilized, unanesthetized, pregnant
mice to ultrasound at days 16.5–19.5 of gestation, which
is the time of neuronal migration from the proliferative
zone towards the brain surface. Exposure time varied
from 5 to a total of 420 min (in 12 episodes each of
35 min). Their conclusion, based on the analysis of over
335 animals, was that ultrasound exposures of 30 min or
more caused a derangement in the migration of neurons
from the deep to the more superficial layers of the brain.
In the introduction to their article, by referring to several
abnormalities such as low birth weight, delayed speech
and behavioral disturbances, they appear to suggest that
ultrasound in general and the phenomenon of disturbed
neuronal migration in particular may be to blame.
This immediately set in motion some elements
of the media to ‘inform’ the public of the dire
dangers of ultrasound. More recently, even the political
apparatus intervened: as posted on 14th November
2006, Assemblyman Joe Pennacchio, R-Montville, New
Jersey, introduced legislation requesting the New Jersey
Department of Health to initiate action to explore the
possible relationship of sonographic examinations of
pregnant patients to the increased incidence of childhood
autism. According to the local paper: ‘The Assemblyman
cited various scientific, published studies that show
a displacement of brain cells associated with autistic
children and the ability of sonograms to displace those
cells.’2 .
It is doubtful that all of this is out of concern for the
health of mouse embryos, but, naturally, because of the
worry to millions of human fetuses exposed to ultrasound
with or without medical indication. While the study by
Ang et al.1 is important, several differences, some major,
between the experimental setup and the clinical use of
ultrasound in human pregnancy need to be pointed out.
Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd.
I present here a short discussion on whether concern is
justified when performing diagnostic ultrasound.
The Ang et al.1 paper
The most obvious difference of the study of Ang et al.
from human obstetric ultrasound practice is the length
of exposure: up to 7 h1 . This has nothing to do with
clinical exposure (nor would 6, 5, 4, or even 1 h) and
there is no clinical indication for patients to undergo
12 scans over a period of several days. In the article,
the authors point out that ‘in animals exposed to USW
(ultrasound waves) for 15 and 5 min, the effect was
not statistically significant’. The shortest exposure time
required to detect the effect was 30 min of direct, wholebody (i.e. whole brain) exposure, with no movement
of the mother or the transducer, hardly actual clinical
conditions. At 420 min, the effect in controls (not exposed
to ultrasound) was similar to or perhaps even more than
that in the experimental group. In addition, in the exposed
group, the effect at 210 min was less than that at 60 min.
This raises many questions regarding the ‘culpability’
of ultrasound as opposed to mechanisms not specific
to ultrasound. The mechanism behind the observed
changes is not clear. The authors think it unlikely
that these are a result of changes in local temperature
or acoustic cavitation but rather that they result from
radiation force, streaming, shear effects on cellular walls
or disturbance of exocytosis with resulting slowing
down of cellular movement. The authors postulate that
because of the device’s low output (as expressed by
spatial peak temporal average intensity, ISPTA ), a nonthermal mechanism (radiation force or microstreaming)
is probably operating, but no discriminating mechanistic
EDITORIAL
364
tests were undertaken. The experimental set-up was such
that embryos received whole-brain (actually whole-body)
exposure to the beam, which is rare in the human (except
very early in pregnancy and not at a time of neuronal
migration). Exposure was between days 16.5 and 19 and
term is 19 days. Therefore, this seems to correspond to the
late third trimester of the human pregnancy, at which time,
undoubtedly, human fetal skull bone is thicker than is
mice embryo skull bone (although one needs to remember
that bone absorbs ultrasound energy strongly, possibly
resulting in local temperature increase, which the authors
of the article ruled out as a factor). While the authors
correctly define distance from the probe and state it was
much less than that in clinical situations, they do not verify
whether those embryos closer to the probe were affected
more. In the supporting documentation, which is very
extensive and describes exposure parameters according to
the most rigourous recommendations3 , it appears that the
ISPPA (spatial peak pulse average intensity) was 330 W/cm2
in water but only 1 W/cm2 through half a mouse. This
is a reduction of over 25 dB, which is very difficult to
comprehend. The ultrasound machines used were an ATL
UM4, a model at least 15 years old, and an M254067 000, a system used in adult cardiology and essentially,
as far as is known, never in obstetrics (both from Philips
Medical Systems). This places further limitations on the
validity of results as they relate to obstetric scanning.
There are other caveats, several of which, rightfully
so, are mentioned by the authors: brains in mice are
much smaller than those in humans at comparable stages
of pregnancy (milligrams versus grams) and develop over
days (mice) as opposed to throughout the entire pregnancy
and up to 20 years of adult age, or more (humans).
They state that the duration of neuron production and
migration is 18 times longer in the human and that a 30min exposure would give a much smaller brain-growth
duration/exposure time proportion. They also describe
that the: ‘. . . frequency used in the present study was
slightly above standard clinical practice (6.7 MHz versus
3.5–5.0 MHz)’. This is not accurate: in early pregnancy,
endovaginal ultrasound is utilized and, depending on the
manufacturer, frequencies vary from approximately 6to 9-MHz. Also: ‘. . . the latest ultrasound equipment
with 3D reconstruction and tissue harmonic imaging offer
even higher frequencies’. This is not accurate either. The
authors are wrong in implying that 3D/4D has higher
energy levels. This is unsubstantiated. While machines on
which 3D is performed are post-output display standard
(ODS; see below) and therefore have higher outputs than
machines had in the past, we have shown that acoustic
outputs, as expressed by thermal and mechanical indices
(TI and MI) are not higher in 3D compared with 2D4 .
As support for the possible ‘neurological’ effects
of ultrasound, the authors put forward two previous
publications5,6 . In the first, pregnant mice were exposed
for 10, 20 or 30 min on day 14.5, at 3.5-MHz, with spatial peak temporal peak intensity (ISPTP ) of 1 W/cm2 and
240 W/cm2 for the spatial average temporal average intensity (ISATA )5 . Some alteration in behavior was observed. In
Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd.
Abramowicz
the second study, monkeys were exposed five times weekly
from days 21 to 35, three times weekly from days 36 to 60
and once a week from days 61 to 150 (this means about 39
exposure episodes), with transitory and very mild behavioral changes, possibly due to experimental conditions6 .
Other references cited supporting harmful effects have
long been discarded, refuted or challenged, such as Stark
et al.’s 1984 study on decreased birth weight7 , Campbell
et al.’s 1993 study on delayed speech8 and Newnham
et al.’s 1993 study on birth weight9 . In fact, Newnham et al. recently published a follow-up study showing
absolutely no long-term effects at 8 years of age10 . Ang
et al.1 also refer to non right-handedness, which has been
described as being more common in males exposed to
ultrasound in utero11 . However, the mechanism behind
non right-handedness is thought by many to be genetic in
great part and not to be caused by environment/exposure
and this, too, has been debated widely12 . As referenced in
the PNAS article, many authors have shown no effects in
several animal species13,14 .
Another question is whether a relatively small
misplacement, in a relatively small number of cells
that retain their original cell class, is of any clinical
consequence. Assuming that several of these cells become
non-functional, some may be destroyed by natural
occurring apoptosis, potentially furthermore reducing the
risk of harm, as was pointed out in an excellent editorial
in the same issue of PNAS15 .
Finally, while this study may indeed show some small
changes in the migration patterns of neurons, one should
not conclude that there will be abnormal behavior or
other anomalies as a result of these changes (pups were
killed on postnatal day 10). Although the authors of the
article do not specifically state so, they discuss that certain
disorders in the human are thought to be due to neuronal
misplacement: ‘. . . from mental retardation and childhood
epilepsy to developmental dyslexia, autism spectrum
disorders and schizophrenia’. The juxtaposition in an
article of ultrasound exposure and these severe disorders
is unfounded. The smallest anatomical and functional
unit of organization in the cortex is the minicolumn
(or ‘microcolumn’) as described long ago, in particular
by Rakic, the senior author of the article presently
under discussion, who described them as vertical units
of embryological origin16 . Disturbances in these columns
would explain later diagnosed neurological and/or
psychological disorders. However, neuronal migration
(according to Ang et al.1 ) is known to be ‘highly sensitive
to a variety of biological, physical, and chemical agents’
and its perturbation can lead to ‘behavioral deficits.’.
It should also be noted that even major anatomical
alterations do not always correlate with functional
inadequacies17 . Therefore, it is a major leap of faith
to suggest that (1) in the human, neuronal migration
disturbances may be caused by clinical ultrasound and
that (2) if it indeed occurs, it has consequences such as
epilepsy, schizophrenia and autism.
Ultrasound Obstet Gynecol 2007; 29: 363–367.
Editorial
What is the issue, why are scientists studying it and
why do we even speak about it?
Around 1991 there was a remarkable change in the
US federal regulations regarding allowable upper-limit
diagnostic ultrasound acoustic output levels. These
regulatory changes, now in place for over 15 years,
are supposedly well known. In the initial 510(k)
Guide18 , allowable ISPTA acoustic outputs per use
category were raised from 430 (cardiac imaging), 720
(peripheral vascular imaging), 94 (fetal imaging; in
reality, 48, originally), and 17 (ophthalmic imaging)
mW/cm2 to 720 mW/cm2 for all applications except
ophthalmic imaging, which was set at 50 mW/cm2 . While
allowing this boost, the Food and Drug Administration
(FDA) recognized there was an increased potential for
ultrasound-induced deleterious effects with the increased
acoustic outputs, and thus required these newer, more
powerful instruments to have the capacity to calculate and
display two safety indices, TI and MI, known collectively
as the ODS19,20 . There are three categories of TI: TIS,
for soft tissue, to be used in the first and early second
trimesters; TIB for bone, when ultrasound traverses bone,
used in the later second and third trimesters; TIC, used in
adult transcranial scanning. The two theoretical indices
(TI and MI), derived from simple algorithms with each
yielding a dimensionless number, were required so as to
provide the end user with at least some indication of
the potential for the occurrence of ultrasound-induced
thermal or mechanical mechanisms, respectively, which
are potentially associated with the induction of bioeffects.
This is a very important point, because now, the
responsibility is on the end-user. The rationale behind
relaxing limitations of acoustic outputs appeared to be
that better diagnostic information could be obtained
from increased output (a stronger signal ‘in’ meant a
stronger, more informative ‘out’ signal, which would
translate into improved, clinically relevant, diagnostic
information). Some additional caution was perceived as
appropriate by the US FDA, due to the increased potential
for ultrasound-induced bioeffects: the limit on ophthalmic
applications was only raised to 50 mW/cm2 , whereas in
the remaining categories it was raised to 720 mW/cm2
(ISPTA ). For the interested reader, the FDA’s regulation
(the 510(k) Guide) can be accessed on the Web as a
PDF file entitled: ‘Information for Manufacturers Seeking
Marketing Clearance of Diagnostic Ultrasound Systems
and Transducers’21 .
Does, in fact, obstetric diagnostic ultrasound have
the potential to induce temperature increments that
are sufficient to cause birth defects in the in-utero
embryo/fetus? The rationale for this concern is severalfold. First, the in-utero embryo is sensitive to thermal
insult during neural tube closure22 , with earlier or later
stages being less sensitive but not insensitive23 to small
increments in temperature. Second, diagnostic ultrasound
can raise the temperature of tissue24 . Third, it appears,
given the various forms of uncertainty of the TI to
reflect accurately the actual diagnostic ultrasound-induced
Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd.
365
temperature, that these induced temperature changes can
be well within the range of those temperature differentials
known to produce hyperthermia-induced birth defects in
model mammalian systems. Temperature rises of 2.5◦ C
have been demonstrated in excised guinea pig brain
after a 2-min exposure25 and even of up to 5◦ C at the
bone surface26,27 . According to the World Federation
of Ultrasound in Medicine and Biology (WFUMB)
ultrasound that produces a temperature rise of less than
1.5◦ C may be used without restraint, but exposure causing
a rise of 4◦ C or more for over 5 min is potentially
hazardous28 . Fourth, there is some evidence that any
positive temperature differential for any period of time
may have some effect, in other words that there may be no
threshold for hyperthermia-induced birth defects29 . Fifth,
there is a possibility of further relaxing or completely
eliminating present FDA regulations on the premise
that greater acoustic outputs would improve diagnostic
capability under certain circumstances30 , although this, in
reality, is entirely unproven31 . Finally, there is disquieting
data indicating that practitioners are generally unaware
not only of the TI, including its meaning and utility, but
also of the MI and of bioeffects and safety-of-ultrasound
issues, in general32,33 .
In this sense there are some worrying points to be
considered regarding Ang et al.’s paper: the instruments
used were pre-amendment, i.e. capable of lesser acoustic
outputs than are present-day machines. In addition, the
calculated acoustic outputs were, by today’s standards,
relatively low (6.7 MHz, calculated MI = 0.7, TIS =
0.05). In terms of potential mechanisms, a 1989 report
that was not cited by Ang describes alterations in
motility and other membrane-related functions from
exposure of cells in vitro and in vivo to therapeutic
and diagnostic ultrasound34 . The acoustic outputs were
relatively low (100 mW/cm2 ISATA ). They cautioned
that such effects ‘could have undesirable side effects
not only on embryogenesis but on late prenatal and
postnatal development.’ In a 1986 review, Edwards
described the effects of hyperthermia on embryos
and fetuses and indicated that ‘processes critical to
embryonic development, such as cell proliferation,
migration, differentiation and programmed cell death are
adversely affected by elevated maternal temperatures’35 .
More recently, Edwards’s 2006 ‘Commentaries on
Hyperthermia’ indicates that ‘hyperthermia can be caused
by fevers, environmental exposure to heat sources
including ultrasound and electromagnetic radiation’ and
can result in changes in cell proliferation36 .
It is a fact that acoustic outputs allowed in diagnostic
ultrasound have been increasing since its introduction
as a clinical tool, about 30–40 years ago. This in no
way necessarily equates with an increase in the actual
output, nor in the exposure. We also know that some
developmental defects have been increasing in prevalence
rate in the US over the same time period. For instance,
according to the Autism Society of America, autism
is growing at a rate of 10–17% a year37 . Some of
the increase in prevalence is certainly due to improved
Ultrasound Obstet Gynecol 2007; 29: 363–367.
Abramowicz
366
detection procedures, but the underlying cause of autism
remains elusive38 . Abnormalities in anatomy and function
of the cranial nerve motor nuclei have been described in
some autistic people39 . Some can be modeled in rats by
exposure to valproic acid during neural tube closure, and
other teratogens (thalidomide, ethanol) appear to have
similar effects40 . It is a reasonable assumption that most
pregnancies in the US, and virtually all of them in many
countries, are scanned with ultrasound. Most often, this
is with regular gray-scale B-mode, but often color imaging
and spectral Doppler modes are used, which have much
higher acoustic energy outputs (possibly by a factor of
100 or 1000). So far, there have not been unequivocal
data suggesting gross effects, such as abnormal hearing,
vision or language development, reduction in birth weight
or childhood cancer as a result of prenatal exposure to
ultrasound. While there is also absolutely no scientific
reason to correlate autism or other constitutional or
behavioral disturbances with ultrasound exposure, we
need to make sure that we are not missing some harmful
effects of ultrasound through dilution effects41 , relying
purely upon observational studies, which are severely
limited in the detection of effects that are anything other
than gross. There are very little data on fetal exposure in
the human during diagnostic ultrasound42 , but the lack
of recent epidemiological research and human data in the
field is appalling, as is the lack of knowledge of the end
users33 .
ISUOG and other organizations are taking steps to
correct this, by publishing Safety Statements43,44 . ISUOG
will also have an entire session dedicated to the safety of
ultrasound, in conjunction with WFUMB, at their next
combined meeting in Florence, in October 2007.
In conclusion, while panic certainly has no place, nor
is there apparent reason to change the current practice
of clinical ultrasound, we certainly should not disregard
completely the information presented by Ang et al.1 , since
ultrasound is a form of energy that causes heat elevation
and mechanical effects (not necessarily harmful effects,
but effects all the same) in the tissues through which
it travels, and there may be some issues if ultrasound
is not performed by well-trained individuals for medical
indications45 . The authors’ findings support the ALARA
principle (as low as reasonably achievable, i.e. perform
the scan for the shortest time possible and with the lowest
output possible to permit adequate diagnostic acuity),
which is recommended by most medical authorities46 – 48 .
They also call for further studies in non-human primates
and epidemiological studies in humans. I could not agree
more.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
REFERENCES
1. Ang ESBC, Gluncic V, Duque A, Shafer ME, Rakic P. Prenatal
exposure to ultrasound waves impacts neuronal migration in
mice. Proc Natl Acad Sci USA 2006; 103: 12903–12910.
2. Morris County Daily Record, November 14, 2006. ID:
20113287.
3. Edmunds PD, Abramowicz JS, Carson PL, Carstensen EL,
Sandstrom KL. Guidelines for JUM authors and reviewers on
Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd.
23.
24.
measurements and reporting of acoustic output and exposure.
J Ultrasound Med 2005; 24: 1171–1179.
Sheiner E, Hackmon R, Shoham-Vardi I, Pombar X, Hussey MJ,
Strassner HT, Abramowicz JS. Comparison between acoustic
output indices in 2D and 3D/4D ultrasound in obstetrics.
Ultrasound Obstet Gynecol 2007; 29: 326–328.
Devi PU, Suresh R, Hande MP. Effect of fetal exposure to
ultrasound on the behavior of the adult mouse. Radiat Res
1995; 141: 314–317.
Tarantal AF, Hendrickx AG. Evaluation of the bioeffects of
prenatal ultrasound exposure in the cynomolgus macaque
(Macaca fascicularis): II. Growth and behavior during the first
year. Teratology 1989; 39: 149–162.
Stark CR, Orleans M, Haverkamp AD, Murphy J. Short- and
long-term risks after exposure to diagnostic ultrasound in utero.
Obstet Gynecol 1984; 63: 194–200.
Campbell JD, Elford RW, Brant RF. Case-control study of
prenatal ultrasonography exposure in children with delayed
speech. CMAJ 1993; 149: 1435–1440.
Newnham JP, Evans SF, Michael CA, Stanley FJ, Landau LI.
Effects of frequent ultrasound during pregnancy: a randomised
controlled trial. Lancet 1993; 342: 887–891.
Newnham JP, Doherty DA, Kendall GE, Zubrick SR, Landau LL, Stanley FJ. Effects of repeated prenatal ultrasound
examinations on childhood outcome up to 8 years of age:
follow-up of a randomised controlled trial. Lancet 2004; 364:
2038–2044.
Kieler H, Cnattingius S, Haglund B, Palmgren J, Axelsson O.
Ultrasound and adverse effects. Ultrasound Obstet Gynecol
2002; 20: 102–103.
Abramowicz JS, Kossoff G, Marsal K, Ter Haar G. Literature
review by the ISUOG Bioeffects and Safety Committee.
Ultrasound Obstet Gynecol 2002; 19: 318–319.
Norton S, Kimler BF, Cytacki EP, Rosenthal SJ. Prenatal and
postnatal consequences in the brain and behavior of rats exposed
to ultrasound in utero. J Ultrasound Med 1991; 10: 69–75.
Yip YP, Capriotti C, Norbash SG, Rosenthal MS, Yip JW.
Ultrasound effects on cell proliferation and migration of chick
motoneurons. Ultrasound Med Biol 1991; 17: 55–63.
Caviness VS, Grant PE. Our unborn children at risk? Proc Natl
Acad Sci USA 2006; 103: 12661–12662.
Rakic P. Mode of cell migration to the superficial layers of fetal
monkey neocortex. J Comp Neurol 1972; 145: 61–83.
Buxhoeveden DP, Casanova MF. The minicolumn hypothesis
in neuroscience. Brain 2003; 125: 935–951.
FDA. 510(k) Guide for Measuring and Reporting Acoustic
Output of Diagnostic Ultrasound Medical Devices. Food &
Drug Administration: Center for Devices and Radiological
Health: Rockville, MD, 1985.
FDA. Revised 510(k) Diagnostic Ultrasound Guidance for
1993. Food & Drug Administration: Center for Devices and
Radiological Health: Rockville, MD, 1993.
AIUM/NEMA. Standard for Real-Time Display of Thermal and
Mechanical Acoustic Output Indices on Diagnostic Ultrasound
Equipment. American Institute of Ultrasound in Medicine:
Laurel, MD, 1996.
Information for Manufacturers Seeking Marketing Clearance of Diagnostic Ultrasound Systems and Transducers. http://www.fda.gov/cdrh/ode/ulstran.pdf [Accessed 12
November 2006].
Shiota K. Induction of neural tube defects and skeletal
malformations in mice following brief hyperthermia in utero.
Biol Neonate 1988; 53: 86–97.
NCRP (National Council on Radiation Protection and
Measurements). Exposure Criteria for Medical Diagnostic
Ultrasound: II. Criteria Based on All Known Mechanisms.
Report No. 140. NCRP: Bethesda, MD, 2002.
Duck FA, Starritt HC. A study of the heating capabilities
of diagnostic ultrasound. Ultrasound Med Biol 1984; 20:
481–492.
Ultrasound Obstet Gynecol 2007; 29: 363–367.
Editorial
367
25. Bosward KL, Barnett SB, Wood AKW, Edwards MJ, Kossoff G.
Heating of the guinea-pig fetal brain during exposure to pulsed
ultrasound. Ultrasound Med Biol 1993; 19: 415–424.
26. Carstensen EL, Child SZ, Norton S, Nyborg W. Ultrasonic
heating of the skull. J Acoust Soc Am 1990; 87:
1310–1317.
27. Horder MM, Barnett SB, Vella GJ, Edwards MJ, Wood AK.
Ultrasound-induced temperature increase in guinea-pig fetal
brain in utero: third-trimester gestation. Ultrasound Med Biol
1998; 24: 1501–1510.
28. Barnett SB. Conclusions and recommendations on thermal and
non-thermal mechanisms for biological effects of ultrasound,
Barnett SB (ed). In: WFUMB Symposium on Safety of
Ultrasound in Medicine and Biology. Ultrasound Med Biol
1998; 24: (special issue).
29. Miller MW, Miller RK, Battaglia LF, Dewey WC, Edwards MJ,
Nyborg WL, Cox C, Abramowicz JS. The delta-T thermal dose
concept: 1. In vivo teratogenesis. J Thermal Biol 2004; 29:
141–149.
30. O’Brien Jr WD, Miller DL. Diagnostic ultrasound should be
performed without upper intensity limits. Point-counter point.
Med Phys 2001; 28: 1–3.
31. Harris GR, Stewart HF, Leo FP, Sanders RC. Relationship
between image quality and ultrasound exposure level in diagnostic US devices. Radiology 1989; 173: 313–317.
32. Marsál K. The output display standard: has it missed its target?
Ultrasound Obstet Gynecol 2005; 25: 211–214.
33. Sheiner E, Shoham-Vardi I, Abramowicz JS. What do clinical
users know regarding safety of ultrasound during pregnancy?
J Ultrasound Med 2007; 26: (in press).
34. Dinno MA, Dyson M, Young SR, Mortimer AJ, Hart J,
Crum LA. The significance of membrane changes in the safe
and effective use of therapeutic and diagnostic ultrasound. Phys
Med Biol 1989; 34: 1543–1552.
35. Edwards MJ. Hyperthermia as a teratogen: a review of
experimental studies and their clinical significance. Teratog
Carcinog Mutagen 1986; 6: 563–582.
36. Edwards MJ. Commentaries on hyperthermia. Birth Defects
Res 2006; 76: 507–516.
37. Shattock P, Whiteley P. The changing prevalence of autism?
http://osiris.sunderland.ac.uk/autism/incidence.htm#nasfigures
[Accessed 10 November 2006].
38. Autism Society of America. http://www.autism-society.org
[Accessed 10 November 2006].
39. Rodier PM, Ingram JL, Tisdale B, Nelson S, Romano J. Embryological origin for autism: developmental anomalies of the
cranial nerve motor nuclei. J Comp Neurol 1996; 370: 247–261.
40. Ingram JL, Peckham SM, Tisdale B, Rodier PM. Prenatal exposure of rats to valproic acid reproduces the cerebellar anomalies associated with autism. Neurotoxicol Teratol 2000; 22:
319–324.
41. Bello SO. How we may be missing some harmful effects
of ultrasound – A hypothesis. Med Hypotheses 2006; 67:
765–769.
42. Sheiner E, Freeman J, Abramowicz JS. Acoustic output as
measured by mechanical and thermal indices during routine
obstetrical ultrasound. J Ultrasound Med 2005; 24: 1665–1670.
43. European Federation of Societies for Ultrasound in Medicine
and Biology. Clinical safety statement for diagnostic ultrasound.
Report from the European Committee for Ultrasound.
Eur J Ultrasound 1996; 3: 283. (2006 revision available
at: http://www.efsumb.org/efsumb/committees/Safety Committee/).
44. Abramowicz JS, Kossoff G, Marsal K, Ter Haar G. International Society of Ultrasound in Obstetrics and Gynecology
Bioeffects and Safety Committee. Safety Statement, 2000 (reconfirmed 2003). Ultrasound Obstet Gynecol 2003; 21: 100.
45. Abramowicz JS. Ultrasound in obstetrics and gynecology: is
this hot technology too hot? J Ultrasound Med 2002; 21:
1327–1333.
46. AIUM Practice Guideline for the Performance of an Antepartum Obstetric Ultrasound Examination, 2003; http://www.
aium.org/publications/statements/statements.asp
[Accessed
10 November 2006].
47. European Federation of Societies for Ultrasound in Medicine
and Biology. Guidelines for the safe use of Doppler ultrasound
for clinical applications. Report from the European Committee
for Ultrasound Safety. Eur J Ultrasound 1995; 2: 167–168.
48. Australasian Society for Ultrasound in Medicine (ASUM):
Statement On The Safety Of Ultrasound in Grey Scale
Imaging In Obstetrics (reaffirmed, 2000) and Safety Statement On Thermal Biological Effects (reaffirmed, 2000).
http://www.asum.com.au/open/P&S/policies.htm [Accessed 16
November 2006].
Published online in Wiley InterScience (www.interscience.wiley.com) DOI:10.1002/uog.4016.
Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd.
Ultrasound Obstet Gynecol 2007; 29: 363–367.