FDA Ultrasound Bioeffects

Reprinted with permission of FDA Center for Devices and Radiological Health

Jump to:Ultrasound Bioeffects: Effects on Embryonic Development and Cardiac Function

Hydrophone Response Characteristics

Key words: ultrasound, hydrophone, mechanical index

OST's ultrasound laboratory is developing the test methods manufacturers need to assure the safety of their diagnostic equipment. This was accomplished in two related studies concerning hydrophones. Hydrophones are instruments used to measure the amplitude of an ultrasound wave. In the first study, OST engineers investigated whether the low frequency response of hydrophones, as specified by FDA guidance, by U.S. standards, and by international standards, was adequate. In the second, a technique to calibrate these hydrophones at low frequencies was developed.

OST scientists have evaluated the error introduced by hydrophones when used to determine the Mechanical Index (MI) and its international counterpart, the Cavitation Index (CI). These indices are measures of the potential for mechanical damage to tissue exposed to intense pulses of ultrasound. These indices are based on the peak rarefactional pressure and on the frequency of the pulse. Both a computer simulation of diagnostic pulses and a variety of digitally recorded, actual diagnostic pulses were subjected to single-pole, high-pass filtering for a range of cutoff frequencies, which is what happens to a pulse when measured by a hydrophone and its amplifier. The resultant error in the indices introduced by this filtration was evaluated (figure 21). For the U.S. and FDA requirements as well as for the IEC specification, results showed that errors exceeding 30% could be expected. Furthermore, to reduce errors to less than 5%, the low frequency hydrophone response should extend at least an order of magnitude below the center frequency of the pressure wave. For example, for a 3.5 MHz transducer, the hydrophone should have a lower cutoff frequency of less than 0.35 MHz.

Figure 21 - Mechanical index

Error in the measurement of the Mechanical index as a function of the ratio of fa, the hydrophone/amplifier low frequency bandwidth, to fc, the center frequency of the measured diagnostic pulse. Values of fa/fc below 0.5 meet the U.S. standard, and values below 0.8 meet the IEC standard.

In addition to the need for low frequency sensitivity, there must be means to calibrate the hydrophones at these low frequencies. To address the current lack of calibration techniques, OST has begun developing a technique for broadband hydrophone calibration. Though important, no commercial hydrophones for medical ultrasound use currently provide sensitivity information below 1 MHz. OST's technique uses broadband, plane-wave pressure pulses generated by electrical shock-excitation of a thick piezoceramic disk. The hydrophone response is calculated from measurements of the source transducer and hydrophone voltage waveforms. The frequency responses of both needle and membrane polymer hydrophones were measured using this technique. The membrane hydrophones studied had acceptable bandwidths extending below 0.1-0.2 MHz, but several of the needle probes began rolling off above 0.5 MHz, suggesting that their measurement of MI and CI may be inadequate. Based on these results, revision of the U.S. standards, as well as FDA guidance, has begun. [PreME, Stds, ProA]

Doppler Accuracy

Key words: ultrasound, Doppler, phantom, test method

OST is conducting investigations into the development of a spinning toroidal flow chamber as a Doppler accuracy phantom. Such a device would enhance the Center's ability to regulate diagnostic ultrasound imaging systems, which almost universally include the capability to measure blood speed ("spectral Doppler") and to image blood flow ("color Doppler"). Quality assurance and calibration of these Doppler systems have been problematic. The phantom, which is intended to resolve these problems, includes a cylindrical torus which can be filled with blood-mimicking fluid and rotated within a larger bath of fluid such that the internal fluid will experience solid body rotation. This then permits evaluation by Doppler ultrasound systems of a well-defined, nonturbulent flow profile over a wide range of fluid velocities.

A potential limitation of the torus phantom for assessing Doppler accuracy would arise if the fluid within the torus does not rotate as a solid body, but rather has some significant secondary flows. To develop the capability to explore all regions of the torus over a variety of time domains, a finite element simulation of the torus was developed and run on an IBM engineering workstation. The model gives a time-dependent view of the laminar flow through one cross-sectional plane of the torus.

The results showed that the fluid within the torus came up to the wall speed within 2 seconds. After this time, no significant secondary flows were present. The fluid tangential velocity was, as expected, slightly higher at the outer torus wall than at the inner wall. A torus with a 40 cm major radius and 1 cm minor radius, rotating so that the center of the torus tube had a speed of 100 cm/s, had a tangential velocity at the outer and inner walls of 102.5 and 97.5 cm/s, respectively. This velocity spread will be useful in calibrating Doppler performance features, such as registration and sample volume. [PreME, ProA]

Ultrasound Bioeffects: Effects on Embryonic Development and Cardiac Function

Key words: fetal growth, ultrasound bioeffects, toxicity, embryo

Recent reports in the medical literature suggests that an increase in the number of ultrasound examinations during pregnancy may restrict fetal growth and that prenatal ultrasonography may be associated with delayed speech in children. Other experimental studies have reported adverse effects (lung hemorrhage) in animals resulting from ultrasound exposure. Ultrasound use continues for nonmedical reasons (e.g., "keepsake" videos or live model ultrasound modeling). These issues support the need for continued research in the area of ultrasound bioeffects.

Xenobiotic-induced alteration in genetic expression is thought to represent early subcellular perturbations that may eventually lead to overt toxicity. This has lead many investigators to suggest that the detection of altered gene expression, in response to physiological stress, may be a useful and sensitive biomarker to toxicity. In collaboration with investigators from Children's National Medical Center, studies were initiated using the chick embryo model to evaluate potential ultrasound bioeffects. Studies were conducted using heat as a positive control to determine the "thermal dose" required for the developing chick embryo to induce stress proteins. Qualitative and quantitative changes in embryonic gene expression were determined by SDS-PAGE. Several proteins have been identified via immunochemical analysis using antibodies specific for stress proteins and glucose-regulated proteins. Preliminary studies have been conducted using ultrasound. Temperature measurements have been made at various intensity/exposure durations and the altered expression of stress proteins has been detected. Preliminary data suggests that ultrasound can induce stress proteins in chick embryos and that these effects are not mediated by heat. Bioeffects data generated from this research project will be useful in responding to new, evolving safety issues and aid in risk assessment activities regarding ultrasound exposure. [PostMS]

Vascular Grafts

Key words: ultrasound, Doppler, vascular grafts

OST scientists have performed a review of clinical problems associated with using Doppler ultrasound as a prognostic tool in evaluating vascular graft success. One problem is that polytetrafluoroethylene (ePTFE) graft interiors are difficult to visualize with ultrasound because the wall is hydrophobic, porous, and typically full of air from surgery. Methods to combat this problem are under consideration, such as filling the wall with gelatin (as is done with some Dacron grafts to prevent blood leakage) or with glycerine. An in vitro model is being developed that uses a computerized pump-to-pump fluid through test vascular grafts of various materials. Primary work is being carried out by a pre-engineering student at Oxen Hill High School, who is interested in pursuing a biomedical engineering degree. She has done a parallel literature search at NIH and is preparing a review of the literature, an appropriate hypothesis, and a plan of experiments to test the hypothesis using materials and equipment available in OST labs. [ProA]