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.