Putting Together the Pieces
Radiology's role in detecting, reporting, and diagnosing potential cases of child abuse.
Every 10 seconds, child abuse is reported in the U.S. More than 3.6 million cases are referred to child protection agencies every year, involving more than 6.6 million children.
Of these cases, 3.2 million are investigated or receive an alternative response, which focuses primarily on the needs of families and usually does not include a determination regarding alleged maltreatment. In addition, state agencies estimated 702,000 victims of maltreatment in 2014.Yet, these staggering statistics do not even include the multitude of cases that go unreported, or the abusive injuries that are incorrectly attributed to other causes.
Child abuse isn't always obvious, so diagnosis requires collaborative evaluation by multidisciplinary teams. Because imaging often plays a significant role in the detection and interpretation of physical injuries, radiologists can be critical in identifying potential cases of child abuse.
"Radiologists are extremely important in that they will uncover evidence of abusive injury that may not be clinically apparent to us," says Kevin P. Coulter, MD, physician-in-chief, professor, and interim chair of the department of pediatrics at UC Davis Children's Hospital. "By adhering to the recommended protocols for imaging, the radiologist becomes key in the assessment of physical abuse."
Recent updates to the ACR Appropriateness Criteria® (AC) on Suspected Physical Abuse in Children provide guidance for clinicians and radiologists when assessing non-accidental trauma. While the AC provides valuable clinical guidance, there are still plenty of other ethical and legal responsibilities facing radiologists when handling potential abuse.
Signs of Abuse
In clear-cut cases, children may come to the emergency room with bruises, fractures, or other injuries that immediately raise suspicions of abuse, spurring physicians to request further imaging (often skeletal surveys or CT scans) for complete evaluation. Other times, children present vague, general symptoms, like headaches or vomiting — and then it's up to radiologists to spot the unsuspected.
The biggest red flags include rib fractures, fractures in long bones like the humerus or femur; "S-fractures" in the spine, scapula or sternum; and any other "unexpected fractures in unusual places that there's not a good story for," says Sandra L. Wootton-Gorges, MD, professor of radiology and director of pediatric imaging at UC Davis Medical Center and UC Davis Children's Hospital, one of the authors of the updated AC topic.
Besides looking for highly specific or suggestive fractures — such as the classic metaphyseal lesion or multiple fractures in various stages of healing — radiologists should watch for injuries that just don't match the explanation. A wrist fracture makes sense if a child falls at the playground, but the story about a baby who rolls off of a couch onto a carpeted surface and ends up with a mid-shaft femur fracture should raise red flags, Wootton-Gorges says.
"It's important for radiologists to be attentive to discrepancies between reported history and imaging findings," says Bruno P. Soares, MD, assistant professor of radiology in the division of pediatric radiology and pediatric neuroradiology at Johns Hopkins University School of Medicine in Baltimore, who co-authored the latest AC. "The best advice is not to interpret images in a vacuum. Understand the clinical context."
Communication is a critical piece of the puzzle to complete the full picture of each child's diagnosis. "It's context-dependent; the next steps depend on each patient's situation or story," says Richard B. Gunderman, MD, PhD, MPH, Chancellor's Professor of Radiology, Pediatrics, Medical Education, Philosophy, Liberal Arts, Philanthropy, and Medical Humanities and Health Studies at Indiana University, whose clinical practice is based at Riley Hospital for Children. "That's why referring physicians need to communicate clearly to the radiologist what they're suspecting and why they want particular radiologic studies performed."
"We often talk directly with referring physicians, both before and after we do radiologic studies, to make sure everyone understands the patient's full story so we can make the best sense of what we're seeing," says Gunderman, who co-authored the previous version of the AC in 2011. "It doesn't affect how we interpret the studies; we see what we see. But there are factors in the patient's domestic circumstances that can increase the probability of abuse."
For example, parents with substance abuse or mental health issues can increase the risk of physical abuse, he says. Face-to-face conversations can reveal details like these that may not be documented.
"The referring physicians have usually done physical exams and had conversations with the family, so they have a lot more information that may not be in the medical record," says Sabah Servaes, MD, pediatric radiologist, director of body CT, and director of residency and fellowship programs in the department of radiology at the Children's Hospital of Philadelphia.
Referring health care providers may have insights into the family history, like whether a child's siblings have been abused in the past or if the child suffers from certain metabolic or genetic conditions that increase susceptibility to fractures. The additional information may point toward other explanations — like osteogenesis imperfecta, rickets, or the rare Menkes disease, which can mimic the signs of child abuse.
Each piece of information helps radiologists interpret whether the story matches the imaging results, or at least helps them recommend further scans or lab work to explain what they're seeing. The new AC guidelines provide a helpful guidance for imaging.
"There are a lot of pieces that go into the puzzle, so you can't jump to a diagnosis. You can strongly suspect it, but it's never absolute. If you think it's child abuse, you need to report your suspicions to the referring clinician, hospital, and/or child protective services as required by law," says Wootton-Gorges.
How to Report Suspected Abuse
Just how straightforward should radiologists be when reporting their suspicions?
"Expert radiologists who review a lot of child abuse cases vary in what they say," Wootton-Gorges says. "Some will say, 'If you think it's child abuse, say it's child abuse, and say that the pattern in compatible with non-accidental trauma so you've documented it.' Others say, 'Just leave it at the facts: Here's the pattern of injuries, and talk more extensively with the clinician without putting non-accidental trauma in the report at all.'"
If Servaes suspects child abuse, she'll state it in the report. Wootton-Gorges used to do the same, until a baby came in with a brain injury that looked a lot like non-accidental trauma — but turned out to be a rare blood-clotting abnormality. Now, Wootton-Gorges communicates her suspicions in-person and documents the findings that were discussed.
"You don't want to miscall child abuse," Wootton-Gorges says. "Until the evaluation of the child is completed — history, physical findings, imaging, and lab work — one doesn't know the exact diagnosis. You must be careful in how you approach this."
Yet while radiologists should act prudently, they need to consult the law of the state in which the patient is located to verify whether they are obligated to report potential abuse to the authorities and/or hospital officials. Virtually all states require physicians, including radiologists, to report suspected child abuse directly to state or local child protection or law enforcement agencies. Additionally, many hospitals and other medical facilities mandate that any staff member who suspects abuse notify the hospital's or facility's head that abuse has been detected or is suspected and report the information to child protective services or other appropriate authorities.
One valuable resource to review is the ACR-SPR Practice Parameter on Skeletal Surveys in Children. Revised in 2016, the parameter indicates that radiologists should communicate "strong suspicion of abuse" to referring physicians when imaging findings warrant. The parameter also states that radiologists may have to report a case of suspected child abuse to local authorities.
It's just as important to note what you don't see. If you don't see Wormian bones, which often accompany osteogenesis imperfecta, you can negate other possible diagnoses.
"Being clear and concise is critical in the radiology report," says Servaes, chair of the Child Abuse Imaging Committee for the Society for Pediatric Radiology. "If you see demineralization of bones, or if the bones look absolutely normal and you see a fracture, it's important to convey both of those things, so if it should go to trial, they can use the same objective criteria that you used to see what you were thinking at the time. As long as you're clear with describing what you see and what your impression is, that's the best thing you can do."
Though it's usually the referring physician's decision to bring in social workers or child protective services, radiologists should consider the legal and ethical ramifications of their documented impression, understanding that a child's life may depend on it.
"There's a responsibility to report suspected abuse, because if a child has been abused and you miss it, you allow the child to return to a hazardous environment," Gunderman says. "On the other hand, you don't want to raise an alarm that abuse has occurred if that's not the case, because the child may be taken from the family, and it can obviously be very traumatic. It's not just a matter of health; it's a matter of family integrity. It's a pretty weighty responsibility."
That's why it's critical for radiologists to maintain familiarity with the signs of abuse, and leverage appropriate imaging to confirm or deny suspicions, he says. Consult with child abuse specialists when possible, and refer to recently published literature — including the AC updates — for guidance.
By Brooke N. Bates, freelance writer for the ACR Bulletin