The Case: A 5 year old girl presents to the ED with approximately 24 hours of suprapubic and RLQ abdominal pain. Vital signs are: Temp 38.2 C, HR 110, RR 19, BP 100/60, Oxygen Sat 100% on room air. She has vomited twice but has not had diarrhea. She had a history of constipation a year ago that has resolved and mother denies any urinary symptoms or history of UTI’s. The patient is quiet but nontoxic appearing. Your abdominal exam notes mild to moderate RLQ tenderness but no rebound and normal bowel sounds. You order a urinalysis, which is negative and a RLQ US which ‘does not visualize the appendix’. Your suspicion for possible appendicitis is still intermediate; however, now the patient states she is “a little hungry”. Should you order a CT of the abdomen and pelvis? Uuugh!
What are the pertinent issues surrounding CT radiation risk?
The above case is one we have all experienced and wrestled with in some way. Since the early 2000’s, the identification of radiation risk associated with computed tomography have made these cases challenging. That risk is usually quantified as the number of cancers attributable to patients who have undergone a CT scan during their lifetimes. Since this risk is extrapolated over an entire lifespan it is easy to appreciate the added risk borne by children based on increased life expectancy alone. The purpose of this post is to lend clarity to conversations you will have with patients’ families about CT radiation risk and how it may affect medical decision-making.
While the models used to characterize risk are complex, one conclusion we can safely draw from is that children exposed to CT have a greater cancer risk than adults.1 In light of this, how should clinicians think about the, seemingly, remote cancer risk later in life relative to the CT diagnosis that may help them today? Furthermore, if we do need to do a CT scan, how can we minimize the risk to pediatric patients by adhering to ALARA principles (As Little as is Reasonably Achievable)? Lastly, how well do alternative diagnostic strategies perform and how can we not feel clinically ‘trapped’ as an emergency physician (EP) evaluating pediatric patients?
How to measure radiation exposure?
Discussing this topic requires an understanding of radiation dosing and relative measures that we can all appreciate. Most radiation dosing is based on milliseiverts or milligrays. Seiverts describe a dose delivered while Grays describe degrees of exposure or absorption. Thankfully, these two terms can be mathematically considered very similar, thus making the literature on this topic easier to understand. The remainder of this discussion will refer to Seiverts. To gain a relative sense of what a mSv means, consider the following:2
- A two-view chest x-ray (CXR) equals 0.02 mSv.
- Natural daily background radiation exposure in Denver, Colorado annually equals 3.5 mSv or 175 CXRs.
- An abdominal CT can be up to 5 mSv or 250 CXRs.
The take home points here are
- We are exposed to radiation everyday.
- A CT scan can be equivalent to several hundred CXRs depending on dosing protocols.
Stratifying CT-related cancer risk has been largely based off mathematical models derived from the survivors of the Hiroshima nuclear blast in 1945.3,4 While this may seem arbitrary, it is the most quantifiable source of cancer outcomes related to radiation exposure over time. Therefore, prediction models frequently use this information to quantify additional cancers (above baseline risk) as a result of a CT scan exposure. A notable departure from this approach was a study from the UK examining this question via a retrospective cohort of patients at varying CT radiation doses and assessed the risk of malignancy.5 Their findings were fairly consistent with the LifeSpan studies based on atomic bomb data. The relative risk for leukemia and CNS tumors was 3.18 and 2.82, respectively, for an exposure of at least 30 mGys (which, depending on dosing, is about 1-3 CTs). This boils down to about:
- 1 cancer per 1,000 head CTs in kids under the age of 5 years
- 1 cancer per 2000 head CTs in kids at age 15 years
- 1 cancer per 500 abdomen/pelvis CTs, independent of age
How to make rational medical decisions in a radiation-conscious environment?
Accepting the increased cancer risk in pediatric patients exposed to CT radiation means reshaping how we approach medical decision-making for pediatric head injury and abdominal pain evaluations. The first step in this process is adhering to ALARA principles which are based on 2 fundamental ideas:
- Communicate openly and candidly with your radiology colleagues about CT and its alternatives.
- Know if the pediatric CT dosing regimens used in your center are lower than adults.6
The next step is carefully examining our utilization of CT scans in these evaluations. PECARN guidelines for the evaluation of pediatric closed head injury provide excellent evidence-based decision support to minimize radiation exposure with low-risk head trauma.7 Abdominal pain, unfortunately, does not lend itself to similar guidelines given the vagaries and varieties of the complaint. In addition, the variability in ultrasound availability and accuracy from center to center make its use as a consistent alternative to CT challenging.8 Lastly, clinical decision rules for the identification of pediatric appendicitis can perform variably and be practitioner and clinical setting dependent.9,10 Nonetheless, we must conscientiously reduce our use of CT scans for the evaluation of pediatric abdominal pain. There is evidence to suggest as the use ultrasound increases that CT use is dropping but the decrease in CT usage is not occurring fast enough.11 For EPs who feel at the mercy of pediatric abdominal pain, with limited options beyond CT, the following principles may offer guidance.
- Think of pediatric abdominal pain the way you think about adult chest pain. With the latter, EP’s use thoughtful stratification strategies and multiple diagnostic testing modalities (including ECG, CXR, troponin, echocardiography). Observation is a viable test. I believe the levels of diagnostic thinking transfer quite well from one the other.
- Just like adult chest pain, many sources of pediatric abdominal pain can be diagnosed with simple tests such as a urinalysis, a rapid strep test, a KUB, an ultrasound, or observation. The idea behind this is to ‘know what it’s not,’ before reflectively ordering a CT. By ruling out simple things with simple tests, you will improve the diagnostic accuracy of the CT if you get cornered into needing one.
- Consider the implementation of a clinical pathway that uses an available clinical scoring tool (i.e. the Alvarado Score). These have shown some promise in reducing CT usage and may help standardize the patient evaluations.12
- Maximize collaboration with your surgical colleagues in order to reserve CT as a last resort for patients with true diagnostic uncertainty. CT rates of under 20% are possible.