Imaging in Pregnancy Stewardship

In a recent article in JAMA, Khan et al described the use of Computed Tomographic (CT) imaging as well as conventional radiography, angiography, fluoroscopy and nuclear imaging during pregnancy in more than 2.2 million women experiencing over 3.4 million pregnancies in the USA and Ontario, Canada. (1) Between 1996 and 2016 in the United States, CT imaging rates increased from 2.0/1000 pregnancies to 11.4/1000 in 2007 and decreased to 9.3/1000 by 2016. The equivalent increase in Ontario was from 2.0/1000 pregnancies in 1996 to 6.2/1000 in 2016. Over the 20 year time period total radiography rates in the United States increased by 38%, from 34.5/1000 (1996) to 47.6/1000 pregnancies in 2016. Rates in Ontario increased 23%, from 36.2/1000 pregnancies (1996) to 44.7/1000 pregnancies in 2016.

In their introduction, the authors of this paper once again repeated the myth that “it is widely acknowledged that ionising radiation is associated with potential carcinogenic, teratogenic and mutagenic health risks to the developing fetus”. As Obstetric Physicians we know this is not correct (2). In a subsequent commentary on this study, the Canadian Medical Association Journal published the headline “Rise in medical scans during pregnancy poses radiation risk” (3). This headline is implicitly misleading (and anxiety provoking) and would only be true if it stated that the “Rise in medical radiation dose during pregnancy poses radiation risk”. Clinicians must understand that rates do not equate to risk; it is the radiation dose on specific tissues that confers risk and oversimplifying this evidence might mislead clinicians.

In the study by Khan, over 50% of total imaging was radiographs such as chest x-rays and extremity radiographs that are associated with an extremely low maternal (0.02 mGy) and fetal (<0.01 mGy) radiation dose (4,5). Similarly, a non-contrast maternal CT head scan has negligible risk to the fetus and can have a life-saving benefit to the mother. Conversely, direct abdominal/pelvic CT Scans are associated with a much higher fetal radiation dose of between 10 and 50 mGy. Based on a “high risk model”, this may be associated with an estimated risk of childhood cancer of between 1 in 1667 (10 mGy) up to 1 in 334 (50 mGy) (6). The Abdominal/Pelvic CT imaging rates reported were approximately 2 per 1000 pregnancies in the Unites States, double that of Ontario, Canada. This increase in higher risk scans merits further research.

The value of this data is to provide a benchmark for the very important task of auditing the use of ionising radiation imaging in pregnancy and through this process, to potentially optimise the use of diagnostic imaging in pregnancy. The parameters for such audits need to reflect the rational application of imaging investigations based on an accurate and informed risk-benefit assessment for both mother and baby. Quality Improvement projects at an institutional level should be undertaken to compare and improve the use of diagnostic imaging in pregnancy. Such projects could include optimising CT protocols to allow a reduction in radiation exposure when direct abdominal/pelvic CT imaging is required. It is important that altering imaging procedures to reduce radiation does not impair the quality and utility of the imaging as this can lead to unnecessary replication of investigations.

We would like to encourage maternity centres to formalise an Imaging in Pregnancy Stewardship (IPS) process; to quantify, compare and improve their use of diagnostic imaging in pregnancy. In the same way as we have all become accustomed to the appropriate use of antibiotics through Antimicrobial Stewardship, we believe the implementation of Imaging Stewardship, particularly with regard to procedures associated with higher doses (>10–50 mGy) of ionising radiation, should be introduced in all centres. This needs to consider the indication for imaging, as well as any risks, including maternal and fetal radiation dose, risks of contrast media as well as cost and availability. Further efforts should include a formalised patient counselling process that provides consistent accurate information.

Who should be responsible for IPS? Radiologists, nuclear physicians and medical physicists must be included because evolving technology often provides opportunities for reducing dose, risk and sometimes allows alternative, non-ionising imaging. Likewise clinicians must be included in IPS as only they can define the clinical urgency and the patient specific risk-benefit that is faced. When a patient collapses with an undifferentiated illness or has severe chest pain, experienced clinicians are calculating the appropriate pathway to diagnosis and intervention. IPS programs can be designed to help more junior clinicians through this process by provision of protocols as well as access to local experts for consultation and advice. For low risk procedures, a simple checklist may be all that is required to ensure appropriate documentation of each episode. This is important for cumulative dose calculations as many women have a series of imaging procedures.

IPS must not be another layer of bureaucracy faced by clinicians but rather should be an aid to decision making as well as a quality assurance process that improves and rationalises the standard of care for pregnant women. IPS will require timely clinical assessments, critical data gathering, and collaborative decision making. The responsible clinician must always feel empowered to select the best and most timely image for their patient.

ORCID iD

Sandra A. Lowe https://orcid.org/0000-0003-3858-6011