The output display standard: has it missed its target?

Diagnostic ultrasound is the safest medical imaging modality for the exposed patient and this has contributed to it being established as the method of choice for fetal examinations. The clinical benefits of obstetric ultrasound are generally recognized and, for about three decades, one or more ultrasound examinations during pregnancy have been offered to millions of pregnant women worldwide. A number of follow-up studies of children exposed to ultrasound in utero provided reassuring evidence that prenatal ultrasound examination does not cause adverse effects with regard to malformations, childhood malignancy, neurological abnormalities or abnormal growth1. The reported weak association between prenatal ultrasound exposure and non-right-handedness2 needs further evaluation but does not refute the fact that there is at present no evidence which would contradict the safe use of ultrasound in pregnancy. This was supported by a recently published prospective study demonstrating that multiple ultrasound examinations during pregnancy did not have any adverse effect on the postnatal growth, behavior and neurological development of children followed up to the age of 8 years3. However, high-energy ultrasound can induce biophysical effects when passing through tissue, for example, thermal effects and mechanical stress, causing cavitation. Consequently, some authorities, such as the Food and Drug Administration (FDA) in the USA, set upper limits for the output from ultrasound equipment: the estimated in-situ intensity must not exceed 720 mW/cm2. In Europe, no such general regulation of the output levels is given. It is the user's responsibility to control the output energy and to use the equipment in a safe manner. In order to facilitate ultrasound safety assessment, in the early 1990s, the American Institute of Ultrasound in Medicine and National Electrical Manufacturers Association introduced the 'output display standard' (ODS) using biophysical indicators for real-time display of safety information during scanning. The FDA adopted the ODS and issued regulations demanding the ODS information to be provided by the manufacturers. The definitions of these biophysical indicators and the requirements for their display are now included in an International Standard4. The ODS indicators comprise two types of biophysical index: the thermal index (TI) and mechanical index (MI)5. The indices are calculated for the given machine settings on the basis of tissue models and their acoustic properties. The TI is an estimate of the tissue temperature rise in degrees centigrade (°C), which might be possible under 'reasonable worst-case conditions'. For particular examination situations, three types of TI have been defined: soft tissue (TIS) and bone tissue (TIB) thermal index, and TI for cranial examinations (TIC). The MI attempts to indicate the probability of non-thermal effects occurring within the tissue. According to the ODS, for equipment that can at certain machine settings produce output energy giving TI or MI ≥ 1.0, the indices should be displayed if they exceed 0.4. The MI and TI are only rough estimates of possible effects and should not be understood as giving more than guidance to the operator. Nevertheless, the ODS is considered at present to be the best way of providing safety information. Several national and international professional bodies—societies of medical ultrasound—issue regularly statements with recommendations for the safe use of diagnostic ultrasound. The safety statements can be found in the official publications6 and on the websites of the societies (Table 1). Practically all safety statements recommend the use of ODS. Often, especially for use in pregnancy, the ALARA (as low as reasonably achievable) principle is suggested. The British Medical Ultrasound Society (BMUS) published more detailed recommendations regarding the exposure time of the fetus/embryo at various levels of TI7. In an attempt to survey the knowledge among ultrasound users of some safety aspects of diagnostic ultrasound, a questionnaire was distributed to professionals using ultrasound for fetal examinations. The questionnaire was answered by the participants and faculty members of six postgraduate courses in obstetric ultrasound, Doppler ultrasound and fetal echocardiography given in Sweden, Norway and Austria in 2003. Furthermore, in Sweden, the questionnaire was distributed to the staff meeting participants of two major obstetric ultrasound units and of one unit for fetal echocardiography, and at a meeting of the Working Group for Ultrasound of the Swedish Society of Obstetrics and Gynecology. In total, 199 questionnaires were anonymously answered by doctors (n = 145), sonographers (n = 22) and midwives trained in ultrasound (n = 32). All of them were using diagnostic ultrasound on a daily or weekly basis. The respondents were from nine European countries. In none of the countries is formal certification for the use of diagnostic ultrasound, in pregnancy or otherwise, required. The doctors who answered the questionnaire were specialists in obstetrics and gynecology (including some of the internationally recognized experts in fetal ultrasound), pediatric cardiology, radiology or clinical physiology. The respondents were not informed about the questionnaire in advance and they had a limited time to answer the 16 questions, typically 10–15 min. The questions concerned explanation of the safety indices and their relation to various ultrasound modes, i.e. real-time B-mode, M-mode, three-dimensional (3D), spectral Doppler, color and power Doppler ultrasound. The respondents were also asked which ultrasound system they used most often and where on the system the information on TI and MI was displayed (e.g. in the top right corner of the screen, in a separate window, etc.) Furthermore, the respondents had to indicate how they adjust the output energy level on their own machine. For all questions, there was the optional answer 'Don't know'. The complete list of questions is provided as supplementary material in the online version of the journal (Appendix S1). Table 2 summarizes the answers to the most important questions. About one-third of the ultrasound experts were able to define the abbreviations TI and MI. For the subclasses of the TI, the correct answer was found in only 3–8% of cases. Of those who knew what the TI and MI meant, only two-thirds and one-third of respondents, respectively, were able to give a simple explanation of the indices. In 49% of questionnaires, the Doppler modes were correctly ranked as generally producing higher ultrasound exposure than the imaging modes (B-mode, 3D). Some 28% of the respondents correctly indicated where on their own machine the information on safety indices is displayed. However, not all of them knew how to control the output energy level. For all items in the questionnaire there were no significant differences in the results between the three categories of respondents (i.e. physicians, sonographers and midwives). The results of the questionnaire were anything but encouraging and they indicated that the users, who are supposed to be responsible for controlling exposure of the fetus to ultrasound, had a very poor knowledge of the basic safety aspects of ultrasound. This was surprising, as, without exception, the respondents were experts and were very frequently using ultrasound for fetal examinations. It can be presumed that the level of knowledge is still lower among clinicians without special interest in diagnostic ultrasound who are routinely using ultrasound for examinations of pregnant women. Among non-professionals that use fetal ultrasound for non-medical commercial purposes, no understanding of the safety issues whatsoever can be expected. The questionnaire was distributed among European ultrasound users only and the results may not be representative of those obtained from obstetricians and radiologists performing fetal ultrasound in other parts of the world (e.g. the USA and Japan). However, as there is no requirement for obstetric ultrasound certification in these countries either, a better level of knowledge of ultrasound safety would not be anticipated. The information gained from the questionnaire indicates that the laudable initiative of BMUS to give TI limits for the use of diagnostic ultrasound in pregnancy remains without practical impact. The majority of the expert users did not know where to find and how to interpret information on the safety indices on the systems they used and, in addition, they were not able to control the energy output. The manufacturers are not responsible for the ultrasound output energy levels of their systems when used for obstetric or other purposes. Their obligation is to follow the ODS and to provide information on safety indices when the ultrasound energy might give a TI or MI of 1.0 or higher. They are expected to inform the purchaser of their equipment about how to find and use this information. Unfortunately, the quality of the information given in the user manuals varies greatly among the manufacturing companies, even if it has substantially improved over the years. Similarly, the knowledge about safety aspects among sellers and application specialists varies greatly. Considering the above points, the promotion of modern ultrasound equipment aimed at the lay public, as practised by some large manufacturing companies, has to be judged as unfortunate, since it might indirectly support the non-medical use of fetal ultrasound without proper respect being paid to the safety aspects. When analyzing the possible causes of the poor knowledge of safety aspects among expert users, several factors should be taken into account. First, there is the possibility of insufficient information being provided and the possibility of teaching quality being poor. There is plentiful literature on the subject, including the abovementioned safety statements, tutorials and reviews, published by professional organizations. Our own society, ISUOG, has a committee on bioeffects and safety of ultrasound, publishing safety statements and comments on urgent questions that arise. There is also an agreement that all courses on obstetric and gynecological ultrasound approved by ISUOG should include a lecture or seminar on safety issues. Similarly, the scientific program of annual ISUOG world congresses should contain an activity dealing with the problem of ultrasound bioeffects and safety. Thus, various sources of information are provided and are quite easily accessible. However, the questions of ultrasound safety are often perceived by the users as difficult to understand or even boring. For the same reason, the safety sessions at the congresses are usually not well attended and the impact of teaching activities is uncertain. Clinicians often refer to the basic thesis that ultrasound is safe for fetuses and, consequently, that it is unnecessary to spend time and energy acquiring the complicated information on physics and techniques of ultrasound. Is this a reasonable conclusion? Do safety committees push an issue that is of only academic interest? It does not seem to be the case. The data indicating lack of any adverse effects on human fetuses are based mainly on older studies that used ultrasound equipment producing low intensities. It has been shown that the output intensities of commercially available equipments have significantly increased over the years and that they can reach levels many times higher than those used 10 years ago8. New modalities producing higher intensities (e.g. harmonic imaging and Doppler ultrasound) are now widely available. In a study recording the levels of safety indices and their time course during standard examinations of high-risk pregnancies, including examinations of the uterine and fetal circulation with Doppler ultrasound, the TI values were found to exceed 1.5 and even 2.0 on several occasions during the examination9. Even if there is a considerable margin of safety, this warrants awareness of safety issues among users, especially among those examining embryos and fetuses susceptible to external physical factors. The fact that routine ultrasound examination of the fetus is being offered to practically all pregnant women, the majority of them without complications or risk factors, makes the prudent use of ultrasound still more important. The responsibility for safe use of ultrasound has been transferred to the user and it is up to the operator to balance the risk of not making a correct diagnosis against the minimal risk of ultrasound exposure. As a support to user's decision making, the ODS was designed to provide the necessary information. Theoretically, it is an excellent concept. The question is whether it has attained its goal. To utilize the information provided, a certain level of knowledge is necessary. The questionnaire described above indicates clearly that this prerequisite has not been fulfilled: not only did the users not show a reasonable understanding of the ODS principle; they were unable to control the output energy levels of their own equipment. Moreover, even if the users were to be fully informed, the ODS has practical limitations, especially for obstetric application. During the examination of a moving embryo/fetus, the operator has to concentrate fully on scanning and optimization of the image. Usually, switching between the various modes (e.g. B-mode, color Doppler and spectral Doppler), in addition to changing the settings, leads to a change in the safety index values. Thus, a continuous monitoring of the ODS values is necessary for full control of the output levels. Such monitoring is not practically possible, as the diagnostic quality of the examination would suffer. Conclusively, the ODS does not seem user-friendly; it provides hardly any useful guidance in obstetric examinations where the output control and guarantee of the safe ultrasound use is more important than in most other applications of diagnostic ultrasound. Deane and Lees9 proposed a graphical display of temporal changes in safety indices for monitoring of output levels and exposures. Certainly, an automatic recording of the output levels during fetal examination would enhance the awareness of the users, provided that they understand the ODS principles. Considering the above facts, I believe that some other ways of regulating the output levels, at least for the use of ultrasound in pregnancy, should be sought. Recently, there has been debate as to whether there should be an upper intensity limit for diagnostic ultrasound10. I do not think that removing upper intensity limits would be beneficial for obstetric ultrasound. Due to the favorable physical conditions during pregnancy (the fetus being surrounded by amniotic fluid, the absence of gas in the uterus), quite low intensities are usually required to obtain an image or Doppler signal of excellent quality. Thus, relatively low output limits should be sufficient in most instances. Naturally, the operator must be able to exceed the limits if this is necessary to achieve a diagnosis, for example, in cases of oligohydramnios or maternal obesity. However, in such situations the equipment should automatically alert the operator. In summary, there are no confirmed biological effects of diagnostic ultrasound on patients including fetuses. Nevertheless, because of the potential of adverse effects at high energy levels, informed use and proper control of output levels is mandatory. It seems that the ODS has failed to provide a practically useful platform, at least when applied to obstetric examinations. In addition to improved training and teaching of operators, new ways of output regulations should be sought which would facilitate the safe use of ultrasound. The following material is available from the Journal homepage: http://www.interscience.wiley.com/jpages/0960-7692/suppmat/index.html (restricted access) Appendix S1 Safety aspects of diagnostic ultrasound. Anonymous questionnaire used to survey the knowledge among ultrasound users of some safety aspects of diagnostic ultrasound. This article contains supplementary material available via the Internet from the Journal http://www.interscience.wiley.com/jpages/0960-7692/suppmat/index.html Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.