In this issue of the Journal of Perinatology, Scott et al.1 provide one of many recent studies to quantify X-ray exposure in the neonatal intensive care unit (NICU). It is unique because it quantifies exposure for an entire NICU population <33 weeks’ gestation over a 5-year period. In that regard, it is a relatively large undertaking and represents one of the best available snapshots of X-ray exposure reality in the NICU. However, the topic of fetal and infant radiation exposure has a long and somewhat soiled history. The field of neonatology would do well to pause for a moment and reflect upon it.

Sir Richard Doll was knighted because of his work in radiation biology and epidemiology, even as he took large sums of industry money for his contributions to help discredit the concept that radiation risk for cancer was a linear function that stopped at zero exposure (a concept that Alice Stewart fought her whole life to prove).2 The two spent most of their careers locked in a very public battle trying to determine whether the atomic bombs dropped over Japan caused an increase in infant leukemia. Doll was largely credited with having won the argument. However, after Chernobyl there was a resurgence of data in favor of Stewart’s hypothesis (with notable exceptions in Finland and Sweden, where some of the greatest exposure occurred).2 We still do not know if there is an absolute lower limit below which radiation is safe, but if there is one, it is extremely low because even elevations in ambient gamma radiation have been found to be associated with increased incidence of cancer.3 At the same time, the concept that cancer risk can be estimated as a linear continuum to radiation exposure is also incorrect because at higher levels the risk is well documented to become exponential for certain types of cancer.4 The truth is rarely simple, but much of what Dr Stewart fought to prove is now widely accepted and utilized in day to day practice for limiting radiation exposure.

Today, we are seeing an increasing number of publications related to quantification and quality improvement of radiation exposure, including in the NICU environment. Recent publications showing associations between computed tomography (CT) exposure and cancer incidence have increased the concern that medical radiation is a real risk.5, 6 Likewise, older eras of fetal and pediatric medical X-ray exposure have documented concerning levels of exposure and cancer association.7, 8, 9, 10 Our field should focus on this safety issue as we move to more standardized care of our patients. At the same time, we are wary of the potential for this topic to be sensationalized as it was in the last century.

The Scott et al.1 study shows us, perhaps predictably, that a fifth of our patients exceed what are considered safe levels of cumulative medical radiation exposure. These at-risk patients were generally the very smallest, least mature and/or the sickest patients in the NICU. We recognize that these are the hardest patients to reduce radiation exposure in and may be the ones with the greatest subsequent risk due to multiple risk factors.11, 12 We also note that the Scott et al.1 numbers are roughly consistent with other similar studies.

Scott et al.1 suggest ‘around the clock’ ultrasound as an alternative to standard X-rays for many of the investigations responsible for the radiation these patients receive (central line placement, necrotizing enterocolitis surveillance, etc.). We agree that ultrasound is likely to gain ground as the radiologic tool of choice in the NICU and reduction of radiation exposure is one of the main reasons. In many cases ultrasound is proving superior to conventional X-ray such as monitoring the evolution of NEC or evaluating lung disease.13, 14, 15 However, NICUs today are struggling just to keep enough respiratory therapists and nurses at the bedside due to financial constraints. For most hospitals, around the clock ultrasound support or the need for additional ultrasound technicians due to increased utilization will be yet another salary line that hospital administrators are unlikely to support.

We would like to offer an alternative suggestion. There is growing recognition of the need to recalibrate neonatology fellowship training toward stronger clinical focus. One change could be for fellows to undertake formal ultrasound training. We note that we are not the first to suggest that ultrasound be a part of neonatal fellowship.16, 17 This change would probably do more to move neonatology away from dependence on conventional X-rays than any other single intervention or policy change. Likewise, at a time when neonatal skills training has been undermined by reductions in NICU experience during pediatric residency (thus truncating the cumulative neonatal fellow’s NICU experience),18, 19, 20, 21, 22, 23, 24, 25 adding this useful tool to the neonatal skills arsenal could be an important first step toward a better trained neonatologist.