Physicians should be prudent about the use of ultrasonography and perform the procedure only when medically necessary and when benefits outweigh risk, according to the American College of Radiology (ACR). The advice comes in the wake of recent findings by Yale researchers that suggest that prenatal ultrasound exposure may harm the developing brain.
The Yale team found that exposing pregnant mice to ultrasound waves at frequencies of 6.7 MHz for 30 minutes or more interfered with normal neuronal migration in the mouse fetuses. This interference could potentially result in brain abnormalities such as mental retardation and seizures.
"It's important for the public to remember that imaging involves putting ultrasonographic energy into the fetus," said Carol M. Rumack, MD, chair of the ACR Ultrasound Commission and professor of pediatrics and radiology at the University of Colorado Health Sciences Center in Denver. According to Rumack, the study provides further proof that ultrasound should not be performed to create keepsake videos and that only qualified medical personnel should operate ultrasound equipment.
The Yale study was performed using more than 335 mice that were in the 16th day of pregnancy (the equivalent of the human third trimester). The mice were injected with 5-bromo-2-deoxyuridine (BrdU) to label neurons destined to take up residence in superficial cortical layers 2 and 3. Within the next 3 days, groups of mice were exposed to 5 to 420 minutes of ultrasound waves (USWs). Ten days after birth, the USW-exposed pups and control pups were euthanized and their brains examined.
Whereas histologic slices showed that most BrdU+ neurons from control pups migrated to superficial cortical layers 2 and 3, BrdU+ neurons from many of the USW-exposed pups were more dispersed. The effect was dose-dependent. Although changes began to be seen in pups exposed to 30 minutes or more of USWs in utero, statistically significant differences were not apparent except in pups exposed in utero to 120 minutes or more of USWs compared with controls. Overall, fewer neurons were detected in superficial cortical layers 2 and 3 and more were detected in the deeper layers and underlying white matter in the brains of exposed pups. Structures resembling periventricular ectopias also were observed in the brains of some exposed animals.
Further research is needed to determine whether the results of the Yale study apply to humans. The dose to which the experimental animals were exposed was in excess of that used in clinical practice (6.7 MHz versus 3.5 to 5.0 MHz). The researchers noted that the total exposure time, however, was equivalent to or less than that "used by commercial, medically nonindicated prenatal ultrasound videos."
A typical prenatal ultrasonography procedure in a clinically indicated setting lasts about 30 minutes and the probe usually is kept in motion. Risk may potentially occur when the probe is held in place for a prolonged time in a quest to acquire a 3-D image of a developing child. The researchers cautioned, however, that "huge differences" exist in the number of neurons and the size of the mouse and human brain so that the effect of 30 minutes of USW exposure on the human fetal brain might be quite marginal.
Ultrasonography probably does not cause harm when used in the appropriate setting by appropriately trained personnel; however, the ACR has been vocal about possible abuses of the technology and of allowing wider access to ultrasonography equipment. Indeed, Rumack testified before the FDA in March 2006 against a proposal that would make low-level ultrasound monitors available over the counter as an aid for tracking fetal heart rates. In pointing to the new data, Rumack cautioned that women with normal pregnancies should not take the risk of continuous ultrasonographic exposure for nonmedical purposes.
The results of the Yale study were published in the August issue of Proceedings of the National Academy of Sciences. The citation for the study is Ang ES Jr, Gluncic V, Duque A, et al. Prenatal exposure to ultrasound waves impacts neuronal migration in mice. Proc Nat Acad Sci. 2006;103:12903-12910. This article elaborates on a news brief appearing in the August 28, 2006 issue of Diagnosticimaging.com, the online weekly news bulletin of Diagnostic Imaging.
-Emily Hayes and Dee Rapposelli