Effects of pulsed ultrasound on embryonic development: an in vitro study

How safe is it to use ultrasound in prenatal medicine? Facts and contradictions - Part 2 - Laboratory experiments regarding non-thermal effects and epidemiological studies

The first part of this CME article (issue 5/20) provided a detailed examination of the biophysical effects of ultrasound waves, the exposure values, and in particular the thermal effect. In vivo and in vitro measurements have shown that the temperature increase in tissue associated with B-mode ultrasound is far too low to pose a potential risk. Even experiments with exposure values in the range of pulsed Doppler have shown that temperature increases of over 1.5 °C can only occur in areas in direct contact with the probe, thus making a limited exposure time particularly in the case of transvaginal application advisable. The second part of this CME article describes various laboratory and animal experiments for evaluating non-thermal effects and also presents the most important epidemiological studies in the last 30 years in the form of an overview and review. In addition to direct insonation of isolated cells to examine possible mutagenic effects, the blood of patients exposed in vivo to ultrasound was also analyzed in multiple experiments. Reproducible chromosome aberrations could not be found in any of the studies. In contrast, many experiments on pregnant rodents showed some significant complications, such as abortion, deformities, and behavioral disorders. As in the case of thermal effects, the results of these experiments indicate the presence of an intensity- or pressure-dependent effect threshold. Numerous epidemiological studies examining possible short-term and long-term consequences after intrauterine ultrasound exposure are available with the most important studies being discussed in the following. In contrast to information presented incorrectly in the secondary literature and in the lay press, health problems could not be seen in the children observed in the postpartum period in any of these studies.

Systemic Evaluation of Vascular Dysfunction by High-Resolution Sonography in an Nω -Nitro-l-Arginine Methyl Ester Hydrochloride-Induced Mouse Model of Preeclampsia-Like Symptoms

OBJECTIVES

The purpose of this study was to evaluate vascular function, including arterial resistance and endothelial function, by high-resolution sonography in an N_ω -nitro-l-arginine methyl ester hydrochloride (l-NAME)-induced mouse model of preeclampsia-like symptoms.

METHODS

Pregnant mice were subcutaneously injected with a saline solution (control; n = 10) or l-NAME (n = 10) between the 7th and 18th days of gestation. The resistive index and pulsatility index (RI and PI, indicators of arterial resistance) of the uteroplacental, umbilical, femoral, and common carotid arteries and the flow-mediated dilatation (index of endothelial function) of the femoral artery were measured by high-frequency sonography in both groups.

RESULTS

We noted significant increases in the RI and PI of the uteroplacental and umbilical arteries and a decrease in the flow-mediated dilatation of the femoral artery in the l-NAME group compared with the control group. We also found that the RI and PI of the uteroplacental and umbilical arteries were negatively correlated with fetal weight and crown-rump length. The results of the multivariate analysis using a logistic regression model indicated that the flow-mediated dilatation at 120 seconds was an independent diagnostic criterion for the l-NAME-induced preeclampsia-like model. A receiver operating characteristic analysis showed that flow-mediated dilatation at 120 seconds had the greatest area under the curve of 0.934, with an optimal cutoff point of 11.1%, yielding sensitivity of 100% and specificity of 84.6%.

CONCLUSIONS

The PI and RI of the fetomaternal vasculature can identify fetuses in "high-risk" pregnancies, and flow-mediated dilatation is a reliable indicator for predicting preeclampsia. Assessment of vascular function by high-resolution sonography provides a useful platform for preeclampsia-related basic research with high reproducibility.

Fetal ultrasound: mechanical effects

In this discussion, any biological effect of ultrasound that is accompanied by temperature increments less than 1 degrees C above normal physiologic levels is called a mechanical effect. However, one should keep in mind that the term mechanical effect also includes processes that are not of a mechanical nature but arise secondary to mechanical interaction between ultrasound and tissues, such as chemical reactions initiated by free oxygen species generated during cavitation and sonoluminescence. Investigations with laboratory animals have documented that pulsed ultrasound can produce damage to biological tissues in vivo through nonthermal mechanisms. The acoustic output used to induce these adverse bio-effects is considerably greater than the output of diagnostic devices when gas bodies are not present. However, low-intensity pulsed ultrasound is used clinically to accelerate the bone fracture repair process and induce healing of nonunions in humans. Low-intensity pulsed ultrasound also has been shown to enhance repair of soft tissue damage and accelerate nerve regeneration in animal models. Although such exposures to low intensity do not appear to cause damage to exposed tissues, they do raise questions about the acoustic threshold that might induce potentially adverse developmental effects in the fetus. To date, bioeffects studies in humans do not substantiate a causal relationship between diagnostic ultrasound exposure during pregnancy and adverse biological effects to the fetus. However, the epidemiologic studies were conducted with commercially available devices predating 1992, having outputs not exceeding a derated spatial-peak temporal-average intensity (ISPTA.3) of 94 mW/cm2. Current limits in the United States allow an ISPTA.3 of 720 mW/cm2 for obstetric modes. At the time of this report, available evidence, experimental or epidemiologic, is insufficient to conclude that there is a causal relationship between obstetric diagnostic ultrasound exposure and adverse nonthermal effects to the fetus. However, low-intensity pulsed ultrasound effects reported in humans and animal models indicate a need for further investigation of potentially adverse developmental effects.

How we may be missing some harmful effects of ultrasound – A hypothesis

In spite of documented effects of ultrasound on living cells in vitro, we have not been able to consistently demonstrate adverse effect after in-utero fetal exposure. This may be because we have used traditional obstetricians' 'week' as the time frame of reference and/or because we have based most studies on the concept of 'biologically plausible' hypotheses. However, large and vulnerable transformations occur in-utero in very short time frames during organogenesis and our knowledge of the functionality of organs at various stages of development is still rudimentary. Using time frames in 'weeks of gestation' may cause large effects to appear small and insignificant. In first trimester fetal exposures, a time frame of '1 day' may be more appropriate. We may need to re-evaluate the safety of diagnostic ultrasound using much shorter in-utero time frames. This may either expose hidden adverse effects of ultrasound or give further reassurance of its safety.

Biological effects of high-frequency ultrasound exposure during mouse organogenesis

Little has been reported on bioeffects of high-frequency ultrasound (US) and guidelines for US use do not necessarily apply to high frequencies. Pregnant CD-1 mice were exposed to Doppler or B-mode US biomicroscopy (UBM) on embryonic day (E) 8.5 or E10.5, during organogenesis. Operating frequency was 40 MHz with a free field I(SPTA) of 11.9 W/cm(2) (Doppler) and 2.6 mW/cm(2) (B-mode), peak negative pressures of 6.61 MPa and MI of 1.05 (B-mode). Offspring were assessed weekly from 1 day postnatally to euthanasia at 6 weeks, with no significant difference in pup weight, body length or crown-rump length observed. E8.5 Doppler-exposed mice showed a small reduction in weight and length at 3 weeks and in weight at 6 weeks. E10.5 Doppler-exposed animals exhibited slight growth reduction in weeks 2 to 4, but were not significantly different at 6 weeks. Our results indicate similar exposures of mice should not cause significant adverse bioeffects.

Thermal assessment of 40-MHz ultrasound at soft tissue-bone interfaces

Tissue exposure to diagnostic ultrasound (US) can cause significant temperature rises. However, little has been reported on thermal effects of high-frequency US, and guidelines for the use of US do not necessarily apply to higher frequencies. Temperature rise induced by US biomicroscopy (UBM) was measured in phantoms containing mouse skulls and in anesthetized mice and mice post mortem, with a 50-microm K-type thermocouple. The operating frequency was 40 MHz with a free field I(SPTA) of 2.6 mW/cm(2) (B-mode) and 11.9 W/cm(2) (Doppler). Peak negative pressures were 5.22 MPa (B mode) and 7.32 MPa (Doppler), resulting in a mechanical index (MI) of 0.83 (B-mode) and 1.05 (Doppler mode). In Doppler mode, mean temperature rises of 1.80 degrees C and 1.73 degrees C were measured for proximal and distal skull phantom surfaces after a 3-min insonation. In vivo, the proximal mouse skull surface showed a mean temperature rise of 2.1 degrees C, with no statistically significant differences post mortem. Our results indicate temperature rise from insonation of bone interfaces using similar exposure parameters should not cause adverse bioeffects.