SOGC CLINICAL PRACTICE GUIDELINES No 160, June 2005 Obstetric Ultrasound Biological Effects and Safety ‘This guideline has been prepared and reviewed by the Diagnostic, Imaging Commitee and approved ty the Exooutive and Councl of the Soctety of Obstetricians and Gynaecologists of Canada, PRINCIPAL AUTHORS Stephen Bly, PRO, Health Canada Radiation Protection Bureau, Otawa ON. Michel C. Van den Hot, MO, FRSC, Halifax NS DIAGNOSTIC IMAGING COMMITTEE Barbara Lewthwalte, MN, Winnipeg MB Robert Gagnon, MD, FRCSC, London ON ‘Lucie Mati, MD, FRCSC, Montreal ‘Shia Salem, MD, FRCP, Canadian Association of Radiologists, Teronto ON, Abstract ‘Objective: To review the biological effects and safety of obstetic utrasound ‘Qutcome: Outine the crcumstances in which safety may be 2 ‘concem with obstetric utasound. Evidence: Medline was searched, and review ofa document on this ‘subject published by Health Canada and of biblographies trom ‘oniied aries was conducted \Values: Review by principal authors and the Diagnostic Imaging ‘Committe ofthe SOGC. The level of evidence was judged as cuties by the Canadian Task Force on the Perocic Health Examination, Benefits, Harms, and Costs: Obstet ultrasound should only be ‘done fr medical reasons, ané exposure should be kept as low as reasonably achievable (ALARA) because ofthe potential fr tissue heating. Higher energy is of particular concern Yr pulsed Doppler, Colour fow, frst trimester utrasound with along transvesica path (© Scam), second or thir timester exams when bone in the focal zone, as well a8 when seanning Ussue with minimal perusion {embryorie) on patients who ae febrile. Operators can minimize Fisk by liming del ime, iiing exposure to crcl stuctures, ‘and following equipment generated exposure information, Key Words: Safety, bioeters, ultrasound, obsttio, fetal, thermal index Recommendations 1. Obstetric uttasound should only be used when the potenti ‘medical beneft outweighs any theoctical or potent risk (I-24), 2, Obstetric ultrasound should not be used for nonmedical reasons, ‘such as sex determination, producing nonmedical photos or videos, of fr commercial purposes (II-B), 3. Utrasound expssure should be a low as reasonably achievable (ALARA) because of the patental forissue heating when the thermal index exceeds 1. Exposure can be reduced through the Use of output control and (or by reducing he amount of time the beams focused on one piace (dweltime) (1A). 4. All diagnostic utrasound devices should comply withthe output 1 display standards (Mi and 1) (8). 5. When ultrasound is done for research or teaching purposes, ‘exposed individuals shouid be Informed iether the Mor Tl are {greater than 1 and how this exposure compares fo that found in ‘Formal dlagnostie practic (II-3), 6. While imaging the fetus inthe fist timester, Doppler and colour Doppler shoul be avoided lB). ‘J Obstet Gynaecol Can 2005:27(6): 872-575 BACKGROUND, Inhough there have been no proven adverse biological A cet Seciaed wit siren degre oe sound, one must be cognizant of the potential for an unidentified tisk, Epidemiologic research on ultrasound safety is limited. Prospective randomized studies are diffi cult to do because routine ultrasound is so prevalent, and ‘even when performed for specific clinical indications, most, fetuses in control groups will also have undergone expo- sure.! In the past, adverse neonatal/pediatric effects that have been studied included childhood malignancies, dys- lexia, delayed speech, and low birth weight. No association was found with childhood malignancies. Also, literature reviews and subsequent studies indicated design weak- nesses and inconsistent findings in reports on the other endpoints. However, an association with non-ight- handedness and prenatal ultrasound exposure has been reported from 2 randomized studies,‘ and more recently, an association with lefi-handedness has been shown in a cohort study.’ This statistical association has only been ‘These guidelines reflect emerging clinical and scientific advances as ofthe date Issued and are subject to change. The information ‘should not be construed as dictating an exclusive course of treatment or procedure to be followed. Local institutions can dictate ‘amendments to these opinions. They should be well documented If modified at the local level. None of these contents may be reproduced in any form without prior writen permission of the SOGC. 572 @}UNBJOGCIUIN 2005Obstetric Utrasound Biological Effects and Safety ‘Table 1. Criteria for quality of evidence assessment and classification of recommendations Tovel of evgence™ |r Evidence obtained from at least one property designed randomized contated vil 1-1: Evidence from well-designed contolie trials without randomization 1.2: Evidence from wel-designed cohor (prospective or retrospective) or case-control studies, preferaby rom more than one centre oF research group. 1k: Evidence from comparisons between tines or places with ‘or without the intervention, Dramatic resis from ‘uncontrolled experiments (such as the resus of treatment ‘wh panier in the 1840s) could also be included inthis category. I: Opinions of respected authorties, based on cine exper- ignce, desorpve studies, & reports of expert committees. 1B. There is fa ‘Gassifcaton of ecommensatonst ‘A. There is good evidence to suppor the recommendation fr use ofa agnostic test, treatment, or ntervertion. idence te suppor the recommendation for use ofa dlagnost test, treatment, or ntarvatio, . There Ie Insufficient evidence to euppot the recommen- ation for use of diagnostic test, treatment, inter- vention, 1B. Theres fair evidence not to support the recommendation {or a diagnostic test, treatment, or intervention. EE. There is good evidence not o support the recommendation {or use ofa diagnostic tes, treatment, o intervention ‘The quay of evidence reported in theye guidelines has been adapted rom the Evaluation of Evidence cre Fore on the Paro Hoa Exar scribed w the Canadian Task -FRecormendatons includ in Bee gugaines have been adapted om the Celestion of Racommencaons ct descbod inthe Candan Task Force onthe Prod Heath Exam found in males, has not been related to neurological deficit, and requires further investigation. Obstettic ultrasound has gained a reputation for safety; however, the possibility of subtle effects such as left of nnon-right-handeriaess cannot be dismissed. Additionally, the bioefects issue is particularly important as more imag- ing moves into an easlier gestational petiod when the fetus is more vulnerable and acoustic output from equipment intended for obstetric use appears to be rising * For these reasons, obstetric ultrasound should only be undertaken for ‘medical reasons. Exposure is limited by using the lowest output setting that maintains image quality and by minimiz. ing exposure time.” Experimental systems suggest that biological effects from ultrasound can result from both thermal and mechanical mechanisms: ‘The quality of evidence and classification of recommenda tions have been adapted from the Report of the Canadian “Task Force on the Periodic Health Examination (Table 1)22 ‘THERMAL EFFECTS "The main potential for ao adverse biological effect with obstetric ultezsound appears to involve tissue heating from ‘energy absorption of the ultrasound beam (thermal effect)? ‘There are many publications on the adverse sequelae of ultrasound heating in animal studies. Embryonic and fetal animal studies show the following’: (1) in-situ ultrasound heating produces a temperature rise of < 1.5°C above physi- ological level, there appears to be ao harmful sequelue (2); at higher temperature elevations, the potential for harm increases with both the exposure duration and the degree of clevation of in-situ temperature for embryonic ot fetal tis- sacs; and @) there is an inverse relation between tempert- ture rise and the exposure time needed so create a potential hazard on thermal grounds (Table 2. THERMAL INDEX Thermal index (Ti) isan estimate of the maximum tempera- tute rise that could occur in exposed tissue during an ultea- sound examination. The 'TT can be used with Table 2 and summary statements 2 to 10 (below) to assess potential thermal hazard to the fenas. This computed TI is unitless and is calculated using standard tissue heating models that, have been derived from clinical situations and measutable properties of the ultrasound field as determined in water under standard conditions. The thermal inidex will be adjusted with changes in user-control settings and is caleu- lated to be directly proportional to the potential for heating, This is important because itis impossible to monitor actual temperatue rise in clinical examinations, Since 1993 ultra sound machines have been equipped with an outpat display for both thermal and mechanical sks, which should be visible if either index is greater than 1. ‘There are 3 user-selectable TT categories: soft tissue (TTS), bone (TTB), and eranial (TIC) Most obstetric examinations would fall under TIS, in which, the ultasound path is predominantly through homogenous soft tssue oF fluid, TIB would apply to some second and third timester scans, in which fetal bone is in the focal region. TIC would normally not apply to obstetric ultea- sound, in which bone is extremely close ro the transducer JUNBJOGR JUAN 2005 @ 573‘SOGC CLINICAL PRACTICE GUIDELINES surface. Various studies have supported the use of these 3 types of thermal indices.126 For electronic fetal heart rate ‘monitors, the maximum thermal effect is low enough that an output display standard is not requited, and heating should not he a concern even with prolonged exposure.” MECHANICAL EFFECTS Mechanical effects result from radiation force, streaming, and cavitation Mechanical effects at diagnostic ultrasound levels have been seen in tissues with stable gas bodies (ung, intestine) or with the use of gas contrast agents? The mechanical index is an estimate of the sisk for capillary hemorthage in lung, taking into account operating condi- tions.” Unless the expected benefits of « higher exposure have been judged to outweigh the foreseeable hazard, the value should be maintained below 1 when scanning tissue at risk? Mechanical effects are unlikely t0 occur in obstettic ultrasound because of the absence of gas bodies or the-ase of contrast media; thus the mechanical index has less rele vane. However, mechanical radiation pressure effects have been demonsteated in preliminary studies of physical mod: clsté and the fetus! using obstetric Doppler. Because this imaging tool yields higher intensities and thermal indices than B-mode with similar mechanical indices, potential bio- logical effects might be both mechanical and thesmal.!?'The early fetal brain is considered more susceptible, and thus Doppler should be avoided in early pregnancy. CONCLUSION Since the implementation of the output display standards, there has been a concern that more equipment is being, developed with intensities which now approach the limits of safety. Although thermal indices can sometimes exceed 1 in standard 2-D real-time B-mode ultrasound, higher intensities are of particular concern for pulsed Doppler, col- our flow, and in first trimester ultrasound wich a long transvesical path (> 5 em).s?221 Concers also arise in scanning tissues with limited perfusion (embryonic tissue) or ifthe patient is febrile. As well, transvaginal probes may produce additional direct heat to adjacene tissue." In these circumstances, operators need to pay special attention to limiting, dwell time, limiting, exposure to critical structures, and to carefully following the exposure information. ‘The theoretical sisk of an adverse biological effect even from standard 2-D obstetric ultrasound makes i¢ hard to justify its use for nonmedical reasons, such as sex determi- ration, making nonmedical photos or videos, or for com- ‘mercial purposes. When obstetsic ultrasound is done for research or teaching purposes, exposed individuals should be informed if either the TI or MI are greater than 1 and of 574 @JUNEJOGC JUNN 205 Table 2. Exposure duration needed to create a potential thermal hazard when embryonicifetal temperature rises above 37°C. Exposure duration, minutes Degroes above normal (37°C) 2 0 3 8 ‘ ‘4 5 1 6 0.28 The values * oe eampromise between concusons ofthe Naonal Coun ‘on Radaton Protecion”and the Word Federation fx Urasound in Meine fd Bilogy. how this exposure compares with that found in normal diagnostic practice* Summary Statements 1, Mechanical effects from ultrasound azeiess important in the absence of gas bodies as is the situation with obstetric ultrasound (monitored with the mechanical index [MI), 2. Thermal effects are of particular concern in obstetric ultrasound with first trimester Doppler and colour flow (monitored with the thermal index) 3. Differing tissue conditions have led to 3 different thermal indices (soft tissue, or TIS; bone, or TIB; and cranial, oF TIC, TIS and TIB can be relevant in obstetric ultrasound, and the appropriate index should be used to monitor the sit uation, TIB should be used if bone is within the focal zone. 4, Thermal effects may increase with ultrasound exposure of pootly pesfused tissues or in febrile patients, 5. Diagnostic ultrasound that produces @ maximum in-sita ‘temperature rise of 1.5°C above normal can be considered, to be safe from thermal damage. This would normally be reflected by a TI of less than 1.5. 6. In estimating the potential hazard of a thermal effect, there is an inverse relation between the degree of in-situ ‘temperature elevation for fetal or embryonic tissue and the exposure duration, 7. Prolonging temperature elevation increases the risk of adverse effects when absolute temperature elevation is greater than 1.5°C. This would normally be reflected by a'TI of greater than 15, 8, For first trimester transabdominal ultrasound through & ‘ransvesical path of > 5 cm, there is evidence that the maxi- ‘mum temperature clevation may be 2 to 3 times that dis- played by the TIS, with a maximum normally of 2°C. In this‘Obstetric Utrasound Biological Effects and Safety circumstance, it is particularly important not to prolong, dwell time, 9, With transvaginal ulteasound, there may be additional heat to adjacent tissue that comes directly from the probe. 10, Fetal heart rate monitoring is done through intensities that are so low that there are no thermal concerns even for extended periods. 111, Ulteasound machines should display an MI of TLif, either index is greater than 1. Recommendations 1, Obstetric ultrasound should only be used when the potential medical benefit outweighs any theoretical or potential risk (11-2). 2, Obstetric ultrasound shonld not be used for nonmedical reasons, such as sex determination, producing nonmedical ‘photos or videos, or commercial purposes (III-B). 3, Ultrasound exposure should be as low as reasonably achievable (ALARA) because of the potential for tissue heating when the thermal index exceeds 1. Exposure can be reduced through the use of output control and (or) by reducing the amount of time the beam is focused on one place (dwell time) (1-1). ; 4. All diagnostic ultrasound devices should comply with the ‘output display standards (MT and TI) (1-8). 5. When ultrasound is done for esearch or teaching pur- poses, exposed individuals should be informed if either the Mor Tl are greater than 1 and of how this exposure com- paces with that found in normal diagnostic practice (LII-B). 6. While imaging the fetus in the fitst trimester, Doppler and colour Doppler should be avoided (II-B). REFERENCES |. Nalioa JP Urasod for ft seesement ee pea, Cochrane Revie Ie The Cacheane Lita, nse 2 2002, Ofer: Update Stee 2. Saves KA, BdeNes SH. Is ulascund uasou? review of epider. legal tutes of human exposures wleasound. Ueto Obeet Gynec 1995,51.200-8 53. Nason Coun ao Radon Protection and Messurernent Bxpore Ci ‘ei for Medial Diagnose Ueatoun I Criteria Basa on Al Known Mechanians [CRP report No. {4 2002 4 Saveen K, EkeNes,Ulanound ding pognincy and sehenaet hit hod aoo-ghthandedaes- a metearaj Uasound bret Gynecol DOHLXORAG 5 Kise, Caasogs 5, aghad 8, Palzgren J Axelson O, Sinisa side ee of prenatal sonography 3 comparsive tedyof youn men. Ep emily 201 12618-28, 6 Whiringhsm TA The scout output of dagaostc machines. Inter Hake GR, Duck FA, editors Safe ote of sltasoand in medal diagoos [chapter 53 Landon: Bish estiote of Radio, 2000p. 16-31 17. America Innate of Ukzsound ia Medicine (AIUM) Mediz Ubeasound Safety Roce (MD). ATOM Puessons 194, 8. Guidelines for she safe use of dingnontewtsound, Heath Canad 2001; (Cat No. 462/01-2558/F. Aviat at wwrarhe se gees 9. Nina Cosellon Radiation Protection and Messrerents (NCRP) Exporre citer or medical dante lsu 1. Caters he on ‘eral mechanisms. Bethesda (Dy NCRP; 1992 10, Worl Fenton or Ultrason in tice an Biology (WFUMB)- Sym posium on Suet of Lkasound in Medicine Cpeasons ad recommen: asionron thermal nd non-thermal mechansn for lei fects of shraosnd, Baraee SB (Ea), Uleasound Med Bol 2p: 1988. American Innate of Ulrstound in Medicine Nations lect! Mansi. turer Aatociason(ATUM/NEMA). Sundae for rel time play ofthe thal and reshancl acoustic opt aces on gnostic ues sguipment Rev 1 Late! (MD): AIUM Publcasons 1998. 1 12, Dan PM, Ligias GC, Barnes SB. Ultasole het of the nn of he fea sheep in utero. Utasound Med and Bo 199521853-60. 1 Ramnatine KV, Nass DK, Peace JM, Joseph AEA, Patel RH, Varma TR. [Brion of nits uleasound exposute ding bstric examin ‘Uneaound Mod Bol 1993519319-29, 14, Siddigh TA, O'Ben WD, Meyer RA, Suita J, Miodown Msi ‘human obec seasound exposimery- tinates of denting factor for cach ofthe dfzenessue models, Uleasoun’ Med & Bil t99521379-91, Bosuard KL, Bien SB, Wood AKW, Edvseé It], KosoffG. Heng of ‘rnc pig fol bain doting exporare to pe sound. Ukzizound Med Bot 193;10:415-24 16 Hower MM, Barer SB, Glher JV, Edearde M), Wood AKW. ln v0 heating of seen pi fetal bran by pulsed eaound and estimates oF ‘hermal inde, Ueaeound ia Med & Bil 19832414674, 17. US Food and Dag Adminstaion (FDA, Infomation for manofieers seeking erkeng clewnee of dignonicueaound stone a ane «x Roche QD): FDA 1997 18, Bronshica Bf, Zimmec EZ. Invitro demonstration of he ton presto fer cute by oboe Dopple.Usonad Oteet Gera rwo9stqp20 19, FaterM, Oprn PL, Greenleaf JF. Fe stimulation by pled dagnone sitstound J Usasound Med 2001206 883-9. 20, Henderson J, Wilbon K; Jag JR, Whitingham TA survey ofthe acoustic ‘ouput of daguatie leseound exuipmentin cutent iical use, Ute ‘ound Ned Bil 19952:699-705, 21, Whigham TA. Estinated fel erebralleasound exposues fom dink ‘alexaminsont. Urasound Med Bio 200127:877-82, 22 Woolf SH, Batista RN, Angerson GM, Logan AG, El W: Caran ‘Tsk Force onthe Pero eth Exar, Ot: Canadian Commaniaton Groop, 1994. saa. JUNEJOGCIUIN 2005 575