Trends in diagnostic ultrasound acoustic output from data reported to the US Food and Drug Administration for device indications that include fetal applications.

OBJECTIVES: A survey was conducted of acoustic output data received by the US Food and Drug Administration for diagnostic ultrasound devices whose indications for use include fetal applications to assess trends in maximum available acoustic output over time. METHODS: Data were collected from 124 regulatory submissions received between 1984 and 2010. Data collection excluded transducers not indicated for diagnostic fetal imaging. The output parameters of ultrasonic power, mean center frequency, and bone thermal index (TIB) were extracted or computed from the submissions for 3 periods: 1984-1989, 1992-1997, and 2005-2010. The data were stratified according to the following imaging modes: M-mode, B/M-mode, pulsed wave Doppler, color flow Doppler, and continuous wave Doppler. RESULTS: Ultrasonic power and maximum TIB values have increased roughly an order of magnitude from pre-1991 to post-1991 periods; the center frequency has decreased somewhat (4.2 to 3.4 MHz). The percentage of Doppler-mode transducers has increased substantially over time, with the majority of the diagnostic fetal imaging transducers currently designed to operate in Doppler modes; this increase is particularly important, since Doppler modes generate much higher TIB levels than B/M-modes. Color flow Doppler ultrasound currently operates at the highest mean ultrasonic power level (with a 14-fold increase over time). CONCLUSIONS: The observed trends in increased acoustic output for both Doppler and non-Doppler modes underscore the widely recognized importance of adherence to the ALARA (as low as reasonably achievable) principle and prudent use in fetal ultrasound imaging.

Thermal effects generated by high-intensity focused ultrasound beams at normal incidence to a bone surface.

Experiments and computations were performed to study factors affecting thermal safety when high-intensity focused ultrasound (HIFU) beams are normally incident (i.e., beam axis normal to the interface) upon a bone/soft-tissue interface. In particular, the temperature rise and thermal dose were determined as a function of separation between the beam focus and the interface. Under conditions representative of clinical HIFU procedures, it was found that the thermal dose at the bone surface can exceed the threshold for necrosis even when the beam focus is more than 4 cm from the bone. Experiments showed that reflection of the HIFU beam from the bone back into the transducer introduced temperature fluctuations of as much as +/-15% and may be an important consideration for safety analyses at sufficiently high acoustic power. The applicability of linear propagation models in predicting thermal dose near the interface was also addressed. Linear models, while underpredicting thermal dose at the focus, provided a conservative (slight overprediction) estimate of thermal dose at the bone surface. Finally, temperature rise due to absorption of shear waves generated by the HIFU beam in the bone was computed. Modeling shear-wave propagation in the thermal analysis showed that the predicted temperature rise off axis was as much as 30% higher when absorption of shear waves is included, indicating that enhanced heating due to shear-wave absorption is potentially important, even for normally incident HIFU beams.

Effects of valve characteristics on the accuracy of the Bernoulli equation: a survey of data submitted to the U.S. FDA.

BACKGROUND AND AIM OF THE STUDY: In 1988, valve manufacturers petitioned the U. S. Food & Drug Administration (FDA) to replace catheter with Doppler ultrasound measurements of pressure gradient (delta P) in clinical studies. Manufactures agreed to submit bench data validating the Bernoulli equation used to calculate delta P = delta P = K(Vd2 - Vp2), where K = constant, Vd = distal Doppler velocity, and Vp = proximal Doppler velocity. Previous studies suggest that K may vary from the idealized 4.0, which could lead to incorrect valve assessment and clinical errors. METHODS: Variation in K-values in marketing application data submitted to the FDA was assessed. Pulse duplicator data included four bileaflet valves, two stented bioprostheses, and seven stentless bioprostheses, sized from 19 to 33 mm. Effects of valve type, valve size, blood-mimicking fluid used, and distal pressure tap position (DPTP) were evaluated via an analysis of variance. RESULTS: K-values varied from 2.50 to 7.40 (n = 90). K was found to be dependent on valve type (p < 0.0001), blood-mimicking fluid (p < 0.0001) and DPTP (p < 0.0001), but not valve size. At DPTP = 30 mm, K = 3.43 +/- 0.56, 5.15 +/- 0.81, and 4.81 +/- 1.02, for bileaflet, stented and stentless valves, respectively. K averaged 10% less using the 100-mm DPTP, due to pressure recovery. Variations due to blood-mimicking fluid were likely related to the fluid density. CONCLUSION: Variations due to DPTP and fluid used are consistent with physical mechanisms of pressure recovery and fluid density. Results from previous studies have suggested that effects of valve type on K are also real. The magnitude of these effects appeared to be +/- 25%. Extrapolation to patients is difficult, but clinicians should be aware that Doppler measurements may vary by similar amounts. Doppler pressure gradients should be interpreted qualitatively and moderated by other diagnostic measures of valve performance.