More physics. The ultrasound pulse lasts a microsecond or less. It is carried by high–frequency waves of positive and negative pressure peaks and troughs – the frequency is around 3-4 MHz (3-4 million cycles per second) for usual obstetric ultrasound. There has been a preoccupation with the negative (or rarefactional) component of the wave because of the possibility of creation of transient microbubbles which cause major effects when they subsequently collapse. This phenomenon of cavitation is unlikely to occur in the diagnostic range, which has been the basis for much of the reassurance given by regulatory authorities and professional organizations. For reasons beyond my comprehension, ultrasound authorities seem to reject or ignore the possibility of other mechanical effects in safety considerations. Perhaps this illustrates the Mark Twain observation: “There is something fascinating about science. One gets such wholesome returns of conjecture out of such a trifling investment of fact.”
FDA-approved low intensity pulsed ultrasound therapy for speeding up the healing of fractures using the Exogen device creates food for thought with regard to mechanisms. The ultrasound frequency is lower than in the diagnostic range, about 1.5 MHz vs. at least 2 MHz, and the pulsing pattern is different. The mechanical intensities are below a cavitation threshold despite the clinically demonstrable biological effect. There is no significant tissue heating (6). Low levels of time-averaged exposure are advertised with intensity values similar to a fetal sonogram (7). According to a company website, the therapeutic effect is due to mechanical activation of integrin systems (6) which then promote the healing process. Accelerated fracture healing has also been described in a mouse model using a diagnostic ultrasound system (8).
The Mechanical Index (MI) is a measure of pressure fluctuations within the ultrasound pulse and, to some degree, also of the overall energy of the pulse. It is defined as the maximum negative (rarefactional) pressure in Megapascals (MPa) divided by the square root of the probe frequency in Megahertz (MHz). It is the required onscreen measure of acoustic intensity for standard B-mode imaging and can be observed by the operator or others in the room. Using an MI of 1 and a probe frequency of 4 MHz - reasonably typical values for obstetric ultrasound screening in the second trimester (4) - peak negative pressure would be 2 MPa. The corresponding positive side of the ultrasound wave would be similar in the other direction, giving an overall pressure difference within half of a 4MHz cycle of 4 MPa, equivalent to being submerged or brought up from 400 metres (1300 feet or ¼ mile) underwater in 1/8 of a microsecond. Although the 1/8 microsecond in which this 400 metre movement would occur makes the analogy impossible – it would be 10 times the speed of light – the point is to emphasize that pressure fluctuations within the ultrasound pulse are large, rapid and far from intuitively trivial.
Derating: Actually, I have oversimplified the MI above, as some allowance is now made for attenuation or weakening of the ultrasound beam as it passes through tissue by introducing a reduction factor to the value that would be measured in a water bath situation – a process called derating (9a). The official FDA definition of the MI (9b) would result in a value at a depth of 5cm, using a 3-4 MHz probe, of about one-third of that which would be measured if there were just water intervening. In fetal ultrasound there may be a considerable component of fluid (maternal bladder or amniotic fluid) in the beam path and therefore MI (and other current measures of ultrasound intensity) can be more than the derated onscreen values displayed would indicate. Historical comparisons of intensities are made difficult, as this derating process was introduced in 1985 (16); the comparisons in the first paragraph in section 1 above are probably complicated and amplified by this point. Actually, Ispta is more correctly currently represented as Ispta.3 , with the .3 addition to the subscript indicating part of the formula for derating.
Tissue Harmonic Imaging: The peak positive pressure in the ultrasound wave can be high enough to transiently increase tissue density and thereby increase the speed of sound; corresponding maximum negative pressure will be low enough to reduce tissue density and slow down the speed of sound. Under these circumstances, higher frequency harmonics of the fundamental frequency are generated as the pulse progresses through tissue. These harmonic frequencies can provide improved imaging: Tissue Harmonic Imaging (THI). While there are circumstances where this is necessary, some manufacturers tend to use THI as part of default settings on their equipment. Using harmonic imaging means that the examiner tends to end up with higher MI values than for fundamental frequency imaging. Improved receiver sensitivity or signal processing cannot eliminate this.
Not all systems have retained the ability to provide satisfactory fetal imaging on fundamental frequency, and therefore their MI values cannot be lowered below the effective threshold for creating harmonics. As noted before, the GE Logiq 9 can usually provide satisfactory imaging of the fetus on fundamental frequency with a default MI around 0.2; this value inevitably jumps 3 to 4-fold on harmonic settings (still legal – regulatory maximum is 1.9). Updates to 2011 have been added. I believe that manufacturers should be encouraged to optimize the fundamental frequency approach, as it does not have an inherent requirement for relatively high intensities. I have no relationship with General Electric, and would be interested to learn of other units that can match the Logiq 9 with respect to providing satisfactory fetal sonography at low MI values.