Radiologic studies are a vital component in the workup and diagnosis of disease. An appropriate radiographic study will accurately rule in or rule out disease with the least possible harm. Special considerations are necessary for the imaging of children. Current trends in pediatric imaging support the increased use of ultrasound and magnetic resonance imaging to decrease radiation exposure. In this review, we highlight some of the emerging concepts in the radiographic workup of pediatric disease, with a focus on decreasing ionizing radiation, increasing ultrasound use, and using clinical decision rules to identify children who do not need imaging.
Key points
- •
Radiologic studies are an important adjunct in the diagnosis of disease in children, and multiple imaging modalities are available.
- •
Computed tomography (CT) exposes children to large doses of ionizing radiation, and its effect on the incidence of malignancy is under heated debate.
- •
Current trends in pediatric imaging support the increased use of ultrasound (US) and magnetic resonance imaging (MRI) to decrease radiation exposure.
- •
Clinical decision rules can be used to predict which child does not need imaging.
Introduction
Radiologic studies are a vital component in the workup and diagnosis of disease. Although an indispensable adjunct, they should not be used as a substitute for a thorough history and physical examination. Multiple imaging modalities are available for use. Each has its own advantages, disadvantages, and safety profile, and practitioners should be versed in these when selecting how to work up disease.
An appropriate radiographic study will accurately rule in or rule out disease with the least possible harm. Harm is often thought of as pain or physical injury, but broader definitions can include time, cost, radiation, and psychological stress to the patient. When compared with outpatient diagnostics, imaging choices within the emergency setting favor sensitivity over specificity to rule out diseases that require emergent intervention. A stepwise radiological approach may be used to identify illness. Accordingly, the time, cost, and potential harm of imaging studies may increase over time as a specific diagnosis is pursued.
Special considerations are necessary for the imaging of children. Sedation or anxiolysis may be required for either computed tomography (CT) or magnetic resonance imaging (MRI). The pediatric emergency provider should be trained in techniques for the sedation of children for imaging. The availability of various imaging modalities will vary between facilities. If delay to diagnosis will place the child at imminent risk, transfer to a higher level of care should be initiated.
In this review, we highlight some of the emerging concepts in the radiographic workup of pediatric disease, with a focus on decreasing ionizing radiation, increasing ultrasound (US) use, and using clinical decision rules to identify children who do not need imaging.
Introduction
Radiologic studies are a vital component in the workup and diagnosis of disease. Although an indispensable adjunct, they should not be used as a substitute for a thorough history and physical examination. Multiple imaging modalities are available for use. Each has its own advantages, disadvantages, and safety profile, and practitioners should be versed in these when selecting how to work up disease.
An appropriate radiographic study will accurately rule in or rule out disease with the least possible harm. Harm is often thought of as pain or physical injury, but broader definitions can include time, cost, radiation, and psychological stress to the patient. When compared with outpatient diagnostics, imaging choices within the emergency setting favor sensitivity over specificity to rule out diseases that require emergent intervention. A stepwise radiological approach may be used to identify illness. Accordingly, the time, cost, and potential harm of imaging studies may increase over time as a specific diagnosis is pursued.
Special considerations are necessary for the imaging of children. Sedation or anxiolysis may be required for either computed tomography (CT) or magnetic resonance imaging (MRI). The pediatric emergency provider should be trained in techniques for the sedation of children for imaging. The availability of various imaging modalities will vary between facilities. If delay to diagnosis will place the child at imminent risk, transfer to a higher level of care should be initiated.
In this review, we highlight some of the emerging concepts in the radiographic workup of pediatric disease, with a focus on decreasing ionizing radiation, increasing ultrasound (US) use, and using clinical decision rules to identify children who do not need imaging.
Imaging modalities
Plain-Film Radiograph (X-Ray)
Radiographs emit ionizing radiation through a body part and project a 2-dimensional image. The effective dose of ionizing radiation from a single radiograph is negligible. Radiographs are nearly universally available, even in outpatient settings, and can be readily shared between providers using digital platforms. Major benefits of radiographs include low cost, accessibility, and speed. Radiographs are most commonly used in the evaluation of pediatric fracture and chest imaging, but they have multiple other uses.
Ultrasound
US uses high-frequency sound waves to project digital images of organs and tissues within the body. Its frequencies are generally between 2 MHz and 14 MHz. Lower frequencies penetrate deeper within the body, allowing for visualization of deep structures, but they do so at the expense of image resolution and/or quality. High frequencies allow for better resolution of superficial structures. In real time, US allows for dynamic observation of pathology and underlying physiology. The major advantages of US are low cost, noninvasive, lack of ionizing radiation, and minimal preparation. US can be limited in obese patients, and cannot easily visualize structures deep to overlying bone, air, or gas. It is operator dependent, and requires skill of the technician or physician.
Computed Tomography
CT uses x-ray technology to create digital representations of the body. The device rotates around a fixed access, creating multiple “slices” that are then put together to generate a 3-dimensional image. The major advantage of CT is its high sensitivity/specificity in working up disease. CT is rapid and has less operator dependence; however, it delivers ionizing radiation to the patient, which may increase their risk of malignancy in the future. A single chest radiograph exposes a patient to a dose of 0.02 millisievert (mSv), whereas a CT of the chest or abdomen exposes the patient to roughly 8 mSv of ionizing radiation. Its use in children is particularly controversial. Also, contrast is often given intravenously to enhance the CT. Contrast-induced nephropathy occurs roughly 5% to 10% of the time in those with normal kidney function, with much higher incidence in those with underlying renal dysfunction. Oral contrast does not cause nephrotoxicity and can enhance the utility of CT for certain diagnoses, but requires time to fill hollow structures and may delay imaging.
Magnetic Resonance Imaging
MRI uses an electromagnetic field to align protons within water contained in body tissues, and then captures decay back to their inherent state. Complex images are produced based on the density and properties of body tissues. MRI produces remarkably detailed images of soft tissues, and uses gadolinium for contrast, which is not believed to cause nephrotoxicity in those with normal baseline function. Disadvantages of MRI include high cost and the amount of time required to obtain the study. In children, MRI can be difficult to obtain because of noise, fear, claustrophobia, and inability to remain still. Sedation or general anesthesia is often required to obtain quality images. Rapid MRI protocols, particularly in neuroimaging and abdominal imaging, are being used with increased frequency and will likely change the face of MRI in the future ( Table 1 ).
| Imaging Modality | Advantages | Disadvantages |
|---|---|---|
| X-ray |
|
|
| Ultrasound |
|
|
| Computed tomography |
|
|
| Magnetic resonance imaging |
|
|
The age of ALARA
Pediatric CT use has increased precipitously in the past few decades, because of increased availability, declining costs, and improved resolution of machines. The impact of ionizing radiation on the incidence of childhood cancers is in the spotlight. Children are more prone to radiation-induced cancers and have a longer expected life span. The practice of ALARA (As Low As Reasonably Achievable) refers to a campaign to reduce the amount of ionizing radiation exposure by using specialized pediatric protocols and using alternate modalities when possible.
Pearce and colleagues recently published retrospective data suggesting that radiation doses of 50 to 60 mGy (2–3 head CTs) in children could triple their risk of brain tumors. Brenner and colleagues estimated that of 600,000 CT scans performed in children per year, 500 may die from cancer related to their radiation exposure. Most data come from risk projection models, although cohort studies are currently in progress. Although the exact risk of ionizing radiation on children is still unknown, we are certainly now practicing in a culture that expects practitioners to be mindful of the potential risk.
CT will frequently be necessary despite the higher exposure to ionizing radiation. Its fast and accurate diagnostic capabilities will often outweigh the potential harm to a child. The concept of ALARA should never be used to withhold a gold standard diagnostic test from a sick child. Rather, it is a culture shift emphasizing efforts to decrease use of ionizing radiation whenever possible. The practice encourages providers to consider alternatives, such as US, MRI, or observation periods. It has also facilitated the development of clinical decision rules that can guide imaging practices.
Clinical Decision Rules
To develop a clinical decision rule, researchers first identify a population of patients at risk for having a disease or outcome. The key features of their history, physical examination, and laboratory studies are identified to create highly sensitive criteria that most accurately predict the outcome in question (eg, diagnosis, prognosis, response to treatment). In the case of imaging, decision rules are typically designed to predict who will not have the disease in question, and therefore do not need imaging. Once derived, clinical decision rules are externally validated and ultimately incorporated into clinical practice. They can then be used to support and inform practitioners’ decisions.
Head trauma
The role of imaging in head trauma is to identify injuries that may require observation, neurosurgical intervention, or invasive monitoring. Serious injuries include epidural hematoma, subdural hematoma, and subarachnoid hemorrhage. In children, neuroimaging is also often used to identify skull fracture.
Noncontrast CT remains the gold standard for the imaging of trauma patients with acute head injury. It identifies intracranial bleeds and skull fractures with excellent sensitivity. In clinically significant pediatric head trauma (low Glasgow Coma Scale [GCS], significant mechanism, obvious skull fracture), CT should be obtained without delay.
Children with less severe head injury represent a unique problem. They embody a key subgroup in which ionizing radiation may be avoided. Clinical decision rules can be helpful to identify children who are unlikely to have clinically significant injury. Extrapolation of adult prediction rules to the pediatric population is difficult, but multiple pediatric rules are now available to providers (outlined later in this article).
The CATCH (Canadian Assessment of CT for Childhood Injury) rule was derived from 3866 pediatric patients enrolled who had witnessed loss of consciousness, amnesia, disorientation, persistent irritability (younger than age 2), or persistent vomiting; 52% of the enrolled patients had a head CT performed. This study identified the following high-risk features:
- •
Failure to reach GCS 15 within 2 hours
- •
Suspicion of open skull fracture
- •
Worsening headache
- •
Irritability
Medium-risk features included the following:
- •
Boggy hematoma of scalp
- •
Signs of basal skull fracture
- •
Dangerous mechanism of injury
The 4 high-risk features had 100% sensitivity for neurosurgical intervention. Using the high-risk features would result in 30.2% of patients with minor head trauma having a CT. When the medium-risk features were included, the rule had 98.1% sensitivity for identifying brain injury, which was defined as any traumatic injury identified on CT, excluding nondepressed skull fracture or basilar skull fracture.
The PECARN (Pediatric Emergency Care Applied Research Network) group published a validated clinical decision rule (2009) to guide neuroimaging in low-risk pediatric trauma. They prospectively enrolled 42,412 children for analysis. Children younger than 2 years who met the following 6 criteria were considered low risk and did not require neuroimaging (negative predictive value [NPV] 100%, Sensitivity 100%, n = 1176):
- •
Normal mental status
- •
No scalp hematoma (except frontal)
- •
No loss of consciousness/loss of consciousness less than 5 seconds
- •
Nonsevere injury mechanism
- •
No palpable skull fracture
- •
Acting normally according to patients
Children between 2 and 18 years of age with the following 6 criteria were low risk (NPV 99.95%, Sensitivity 96.8%, n = 3800):
- •
Normal mental status
- •
No loss of consciousness
- •
No vomiting
- •
Nonsevere injury mechanism
- •
No signs of basilar skull fracture
- •
No severe headache
In children older than 2 years, the rule missed 2 clinically important brain injuries, neither of which required neurosurgery, but did require observation. Children younger than 2 who satisfied all the criteria had less than 0.02% risk of clinically significant injury, and those older than 2 had less than 0.05% risk. Use of this rule could significantly reduce head CT use in pediatric head trauma ( Table 2 ).
