July 03, 2012
15 min read

Ocular signs lead to recognition of non-accidental trauma in children

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Child abuse crosses all socioeconomic, geographical, racial and ethnic boundaries. It poses serious threats to public health and involves significant economic costs, according to the Centers for Disease Control and Prevention.

Ophthalmologists frequently encounter child abuse and its ocular manifestations. The American Academy of Ophthalmology and the American Association for Pediatric Ophthalmology and Strabismus offer detailed information on clinical definitions and signs of abuse in and around the eye.

Shaken baby syndrome, a form of abusive head trauma, that stems from repetitive acceleration- deceleration forces with or without blunt head impact, is typically characterized by retinal hemorrhage, vitreous hemorrhage or other pathologies.

Alex V. Levin, MD, MHSc 

A multidisciplinary team approach is needed when faced with possible child abuse cases, according to Alex V. Levin, MD, MHSc.

Image: R Barone/Wills Eye Institute

Although child abuse may manifest in the anterior chamber, retinal signs of abuse are far more common, K. David Epley, MD, vice president of AAPOS, said.

“From an abuse standpoint, clearly retinal hemorrhages and vitreous hemorrhages are much more common than something like a corneal disease,” Dr. Epley said.

K. David Epley, MD 

K. David Epley

The type and severity of injury resulting from child abuse largely hinge on the child’s age, according to Natalie Kerr, MD, a pediatric ophthalmologist and professor at the University of Tennessee College of Medicine and the Hamilton Eye Institute, Memphis, Tenn.

“The type of non-accidental trauma that you see in babies is different from the type of non-accidental trauma that you see in older children,” Dr. Kerr said.

Child abuse involves a multidisciplinary diagnosis, Alex V. Levin, MD, MHSc, the director of the pediatric ophthalmology and strabismus service at Wills Eye Institute and a board-certified child abuse pediatrician, said.

“It involves radiology, sometimes neurosurgery, child abuse pediatrics, social work, ophthalmology and so on,” Dr. Levin said.

Dr. Levin said ophthalmologists are important in the reporting and diagnostic process.

“Reporting is mandatory,” Dr. Levin said. “Ophthalmologists are a key element in identifying retinal findings that suggest that a baby has been a victim of abusive head trauma, and ophthalmologists should respond promptly and be familiar with findings of abusive head trauma so that they can document appropriately in their records the findings that they see. They have to document the ophthalmic signs.”

Brian Forbes, MD, PhD, of Children’s Hospital of Philadelphia, emphasized the role of education in prevention efforts.

“There’s a lot of patient education that goes on, particularly in the nurseries to educate parents, which is probably the most important thing. Education in the peripartum period has been quite beneficial,” he said.

But in a clinical setting, Dr. Forbes said, the best thing physicians can do is keep their antennas up and always be aware of the possibility of abuse.

“If ophthalmologists see hemorrhages that are unexpected in a child, they need to make certain to go through the differential diagnosis and look at the entire body, not just the eyes,” he said.

Signs of shaken baby syndrome

Child abuse commonly presents with evidence of a brain injury that leads the pediatrician to consult with an ophthalmologist to confirm or refute a pattern of injury consistent with abuse, Dr. Levin said.

“That’s how the ophthalmologist becomes involved, not because of any eye symptoms or difference in vision,” he said. “The fact that they have the suspicion, that’s what they go on. It’s incumbent upon the ophthalmologist to respond promptly, preferably within 24 hours, certainly no later than 72 hours, and conduct a full dilated retinal examination looking for the findings that would support a diagnosis of child abuse or that may indicate another diagnosis.”

Dr. Levin, Dr. Forbes, Randell Alexander, MD, PhD, and Carole Jenny, MD, MBA, are authors of a statement on abusive head trauma and shaken baby syndrome posted on the AAO website. They noted that “a striking feature of shaken baby syndrome is the frequent lack of external evidence of trauma.”

They defined shaken baby syndrome as a subset of abusive head trauma characterized by repetitive acceleration-deceleration forces, with or without blunt head trauma, that results in a novel complex of ocular, intracranial and other injuries. Shaken baby syndrome is widely recognized as one of the most serious forms of child abuse.

The most common ocular manifestation of shaking injury, identified in about 85% of cases, is retinal hemorrhage. However, the absence of retinal hemorrhages does not rule out abuse. Retinal hemorrhage without intracranial hemorrhage is rare.

For differential purposes, hemorrhages as a result of child abuse and those as a result of other injuries are often very different, the authors said.

“It really takes a forceful injury to cause retinal hemorrhages, not short falls but rather crush injuries or motor vehicle accidents. I think by evaluating the patient’s history, you can nearly always tell if there’s an injury of that sort,” Dr. Forbes said.

Hemorrhages are typically identified in or near the posterior pole, but some are so severe that they cover almost the entire fundus.

According to the AAO statement, retinal hemorrhages from shaken baby syndrome normally resolve within several days to several months, depending on the location and severity of the injury. Accurate dating of retinal hemorrhages cannot be done.

Secondary vitreous hemorrhage may also develop, typically resulting from migration of intraretinal or preretinal blood. Vitrectomy may be considered if a hemorrhage does not resolve. Vitreous hemorrhage can result in the development of amblyopia or myopia.

Eyes of shaken infants also show evidence of tissue disruption, such as full-thickness perimacular folds in the neurosensory retina, retinoschisis, full-thickness retinal breaks and detachment.

“Although similar findings have been reported rarely in fatal crush injuries and fatal motor vehicle accidents, such histories are readily apparent and would allow rapid identification,” the authors said.

Anterior segment signs

Lacerations of the eyelids, conjunctiva, cornea and sclera almost always result from trauma and may be considered possible signs of abuse, Dr. Levin and Mary Louise Z. Collins, MD, wrote in Duane’s Ophthalmology.

“The spectrum of ocular findings in the battered child is vast,” Drs. Levin and Collins wrote. “Essentially any injury to the eye of adnexa could be due to abuse. An ocular injury may be as mild as periorbital edema or as severe as a ruptured globe. Signs of bilateral ocular trauma suggest inflicted injury because accidents usually involve only one eye. Perhaps one exception is bilateral periocular ecchymosis due to an accidental, single, central forehead trauma.”

Child abuse presents in the anterior segment as hyphema, corneal infection, blunt trauma, burns and other pathologies, Dr. Kerr said.

“In any child, anterior segment issues include hyphema,” she said. “Hyphema is probably one of the most common manifestations of non-accidental trauma. It tends to occur more frequently in slightly older children, not infants, but children who are old enough to, for instance, be hit with a belt buckle during corporal punishment.”

Belt buckle injuries, burns and bruising are strong indicators of abuse, Dr. Kerr said.

“Certainly any manifestation of blunt trauma and cigarette burns around the lids are something that you see as well. Lid findings, we see some bruising and that sort of thing,” she said.

The effects of child abuse can appear in the cornea, where trauma can have devastating consequences on vision, and can be difficult to diagnose. In the Journal of American Association for Pediatric Ophthalmology and Strabismus, Moore and colleagues described chronic keratoconjunctivitis with dermatitis as a presenting sign of child abuse.

A 13-month-old girl presented with chronic keratoconjunctivitis with dermatitis. She was originally diagnosed with corneal abrasion and mild infection and treated with topical and oral antibiotics, but her condition steadily deteriorated in a span of 2 weeks. However, her condition improved during a 3-day stay in the hospital, and she was released. But despite ongoing treatment, the irritation and inflammation then continued.

Later, the infant was taken to a local trauma hospital with a brain hemorrhage and bruising consistent with child abuse. Clinicians also discovered that the child’s antibiotic eye drops were laced with household bleach. Criminal charges against the child’s mother are pending.

The ensuing chemical injury resulted in limbal stem cell deficiency and other significant consequences. The child is expected to have irreversible vision loss stemming from persistent corneal disease, the authors said.

Michael X. Repka, MD, a pediatric ophthalmologist at Johns Hopkins, likened the child’s pattern of injury to skin burns inflicted with cigarettes, a common form of abuse.

“I think this is just another form of child abuse,” Dr. Repka said. “This is another behavior on the part of caretakers. It’s obviously unconscionable, and most people can’t figure out why they might even do it.”

The take-home message of the JAAPOS report is that unusual patterns of injury and pathology that do not respond to conventional treatment should raise suspicion of abuse among treating physicians, Dr. Repka said.

“I think that the key here is clinical suspicion, meaning the case didn’t make sense,” he said.

Comorbidities and sexual abuse

Physically traumatic child abuse involves various comorbidities and systemic issues. For example, victims of shaken baby syndrome frequently suffer arm, leg and rib fractures, Dr. Kerr said.

“The systemic manifestations of that are broken bones,” she said. “Broken ribs are very common because the children are held by the rib cage when they’re shaken. And then brain edema and brain bleeding and the downstream effects of that, such as suppression of the respiratory system and seizures.”

Non-accidental trauma may also be evident in extraocular structures such as the orbits and lids. Furthermore, some children are diagnosed with venereal diseases contracted through sexual abuse, Dr. Kerr said.

Some neonates develop conjunctivitis or other corneal diseases stemming from exposure to sexually transmitted diseases such as chlamydia or gonococci in the birth canal; these children are typically not investigated for child abuse, Dr. Kerr said.

“But anytime after that and before a child is likely to be sexually active as a teenager, any kind of sexually transmitted infections that are found around the eye, such as gonococcal conjunctivitis and chlamydia, are certainly red flags,” she said.

Documentation and reporting

According to the CDC, in 2008, state and local child protective service agencies received more than 3 million reports of child abuse or neglect, equating to about six reports per minute. About 772,000 children were classified as being maltreated; 1,740 children under the age of 17 years died of abuse and neglect in 2008, according to the CDC.

A CDC report published in the journal Child Abuse & Neglect showed that the total lifetime estimated financial costs associated with 1 year of confirmed abuse and neglect cases was about $124 billion in 2008.

Aside from national statistics and data, the actual reporting of child abuse is a vital community service. Reporting can be somewhat complex, but strict protocols are designed to streamline the process and expedite treatment.

Dr. Kerr said that at LeBonheur Children’s Hospital in Memphis, the standard child abuse procedure includes documentation of all reported cases of retinal hemorrhage.

“We have a RetCam (Clarity Medical Systems), and that’s part of the medical record, which has been used by the child protection team, particularly if they have to place the child into custody, which often happens, as well as if there’s any kind of prosecution of a suspected perpetrator,” Dr. Kerr said.

Dr. Kerr said that emergency physicians at her hospital are keenly aware of the signs of child abuse and proper reporting procedures. However, emergency room physicians and pediatricians are in need of more training in ophthalmic disease recognition. In fact, she said, most medical school curricula lack ophthalmology courses.

“That’s a huge problem,” Dr. Kerr said. “It would go a long way toward raising awareness of all sorts of red flags and alerts and the appropriateness of consults, in instances of child abuse and in other settings, if ophthalmology had a greater presence in the medical school curriculum.”

Dr. Kerr said that she frequently educates emergency room physicians about ocular pathology stemming from abuse.

“In our setting, I lecture to the ER fellows at our children’s hospital,” she said. “And we work in the trenches with our ER physicians at our big county hospital, so their level of understanding based on our patient care interactions is quite high. However, just because these educational experiences take place at our academic institution with an emphasis on tertiary care training, it’s really lacking at more primary levels of care because ophthalmology is not part of the general medical school curriculum.”

Robert Sege, MD, PhD, a professor of pediatrics at Boston University, said medical students on their pediatric rotation at his institution are introduced to presenting signs of child abuse, reporting protocols and investigation procedures.

Robert Sege, MD, PhD 

Robert Sege

“The medical students also during their outpatient rotations and inpatient rotations are often involved in the care of children who may have been abused and neglected. We use those opportunities for bedside teaching,” Dr. Sege said.

Awareness of child abuse among physicians is high, but some physicians do not recognize patterns of injury that suggest abuse or neglect. In addition, some may overestimate their ability to analyze the social dynamics of families, Dr. Sege said.

“Sometimes, physicians will decide that a family is nice or that they know them well, and it couldn’t be abuse or, on the contrary, that families who are unpleasant or uneducated or have other issues are more likely to need to have an investigation,” he said. “Our belief is that physicians should be trained to look objectively and make our reporting decisions based on the medical evidence at hand.”

All 50 states have laws that mandate the reporting of suspected child abuse. Individuals who fail to report abuse are subject to civil or criminal penalties, Dr. Sege said.

“On the contrary, there’s extensive protection against reporting cases that don’t turn out to have been the result of maltreatment unless, of course, the physician did it maliciously,” he said.

Many ophthalmologists have access to child abuse specialists and are able to refer victims of suspected abuse for X-rays and other medical tests.

State investigations often uncover patterns of abuse or factors that contribute to abuse, Dr. Sege said.

“In those cases, the injuries that occurred or the negligence that was apparent may just be a marker for more serious family problems,” he said. “I think it’s important for us to be aware of that and of the role the state has in investigating the situations of these children.”

Dr. Repka said that he and his colleagues have access to a child abuse team for consultation on cases of suspected abuse.

“Usually we would consult, and while we continue with management of the eye issue, they will take over management of the suspected child abuse issue,” he said.

Legal liability and protection

Dr. Levin said there is a conundrum surrounding physician reporting of suspected child abuse. Physicians are considered the most reliable reporters of suspected child abuse but are actually the worst reporters.

“They often don’t report because they don’t want to become involved and they don’t want to take time away from their practice or they’re afraid to go to court, etc.,” Dr. Levin said.

However, ophthalmologists rarely need to testify in court, Dr. Levin said.

“In reality, after reporting, an ophthalmologist has to go to court very infrequently,” he said. “Therefore, some of the concerns that ophthalmologists have about reporting are not justified and, more importantly, run against the legal obligation to report suspicion.”

Physicians are protected from legal action if they report in good faith, Dr. Levin said.

“What’s important is that doctors must report their suspicion and are protected by law as long as that suspicion is reasonable,” he said. “Ophthalmologists who miss a diagnosis and fail to report may be held liable for criminal action for failing to report, particularly when it leads to further harm.”

Furthermore, ophthalmologists may forfeit responsibility for reporting if they work with a child abuse team, Dr. Levin said. However, they must ensure that a report is ultimately generated.

“If you have a specialist at your hospital or a team to evaluate the case, the ophthalmologist may defer the case to that team promptly and seek that multidisciplinary consultation,” he said. “The team can advise whether reporting is indicated, and the team can decide whether or not the report should actually be generated by the ophthalmologist or the team will do it themselves. But it’s still the ophthalmologist’s duty to ensure that if they have a reasonable suspicion, even after multidisciplinary team evaluation, that a report is somehow generated by themselves or by the team.”

According to Dr. Forbes, the largest difficulty with reporting suspected child abuse is having clear, surefire channels to go through.

Brian Forbes, MD, PhD 

Brian Forbes

“For people in private practice especially who may be suspicious of abuse, the biggest pitfall is the reliability of services in their particular area. People say, ‘I told this person, who told this person, and then they didn’t do anything or don’t know anything about it,’” he said.

“The Division of Youth and Family Services covers nearly all areas in the United States. It’s the service most people have available to them, so that’s probably your first choice. There are certainly in-hospital services like mine, the Suspected Child Abuse and Neglect Team; that’s usually the channel I take. And also, physicians need to approach the situation delicately, but law enforcement should be involved too,” Dr. Forbes said. – by Matt Hasson and Daniel R. Morgan

  • Binebaum G, Mirza-George N, Christian CW, Forbes BJ. Odds of abuse associated with retinal hemorrhages in children suspected of child abuse. J AAPOS. 2009;13(3):268-272.
  • Collins MZ, Levin AV. Ophthalmic and systemic manifestations of child abuse. In Tasman W, Jaeger EA, eds. Duane’s Ophthalmology [CD-ROM]. Philadelphia: Lippincott Williams & Wilkins; 2006: Vol. 5, Chapter 4.
  • Elner SG, Elner VM, Arnall M, Albert DM. Ocular and associated systemic findings in suspected child abuse. A necropsy study. Arch Ophthalmol. 1990;108(8):1094-1101.
  • Fang X, Brown DS, Florence CS, Mercy JA. The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse Negl. 2012;36(2):156-165.
  • Levin AV, Alexander R, Binenbaum G, Forbes B, Jenny C. Clinical statements: Abusive head trauma/shaken baby syndrome. American Academy of Ophthalmology website. http://one.aao.org/ce/practiceguidelines/clinicalstatements_content.aspx?cid=914163d5-5313-4c23-80f1-07167ee62579. Published 2010. Accessed May 24, 2012.
  • Levin AV, Christian CW; Committee on Child Abuse and Neglect, Section on Ophthalmology. The eye examination in the evaluation of child abuse. Pediatrics. 2010;126(2):376-380.
  • Moore DB, Herlihy EP, Weiss AH. Chronic keratoconjunctivitis with dermatitis as a presenting sign of child abuse. J AAPOS. 2012;16(2):193-195.
For more information:
  • K. David Epley, MD, can be reached at Children’s Eye Care, 11800 NE 128th St., Suite 430, Kirkland, WA 98034; 425-823-3937; email: depley@childrenseyecare.org.
  • Brian Forbes, MD, PhD, can be reached at Children’s Hospital of Philadelphia, 34th St. and Civic Center Blvd., Philadelphia, PA 19104; 215-590-4598; email: forbesb@email.chop.edu.
  • Natalie Kerr, MD, can be reached at Hamilton Eye Institute, University of Tennessee Health Science Center, 930 Madison, Suite 470, Memphis, TN 38103; 901-448-6650; email: nkerr@uthsc.edu.
  • Alex V. Levin, MD, MHSc, can be reached at Wills Eye Institute, Suite 1210, 840 Walnut St., Philadelphia, PA 19107; 215-928-3914; email: alevin@willseye.org.
  • Michael X. Repka, MD, can be reached at American Academy of Ophthalmology. 20 F St. NW, Suite 400, Washington, DC 20001; 202-737-6662; email: mrepka@jhmi.edu.
  • Robert Sege, MD, PhD, can be reached at Boston University School of Medicine, Boston Medical Center, ACC5N-20, 850 Harrison Ave., Boston, MA 02118; 617-414-5506; email: robert.sege@bmc.org.
  • Disclosures: None of the physicians have any relevant financial disclosures.


To diagnose shaken baby syndrome, is patient history with fundus examination sufficient, or is it necessary to perform OCT imaging?


OCT is an adjunct to standard methods

Yes, fundus examination is sufficient. No, you do not need OCT.

Shaken baby syndrome is not a term we use anymore. We use “non-accidental trauma” or “abusive head trauma” for legal purposes. Ophthalmologists do not make the diagnosis of non-accidental trauma. What we do is give an answer consistent with or with a high probability of non-accidental trauma and leave the final decision to Child Protective Services to make because it is too complex and involves many other things.

Roberto Warman, MD 

Roberto Warman

The diagnosis or presumptive diagnosis from the ophthalmologic point of view is made with history and fundus examination. You can add to it, when available, fundus photography, in particular the RetCam (Clarity Medical Systems). That is sufficient in every case. OCT is an adjunct examination that is always welcome and helpful in any retinal disease. But it is not necessary at all to make a diagnosis. In particular, in this condition, there are no portable OCTs, except a few models, and those are the ones that articles have been published about. The examination in essentially all important cases is done in a pediatric intensive care unit. As of today, 99% of them will not have access to a portable OCT, so it is very important because you do not want a standard of care that cannot be done.

The reason all of this comes out is because macular retinoschisis is almost always, if not completely, indicative of non-accidental trauma from shaking. And, of course, the ideal test for that is OCT. But that does not mean that by just looking at the fundus we cannot get accurate results in essentially all cases.

Roberto Warman, MD, is an OSN Pediatrics/Strabismus Board Member. Disclosure: Dr. Warman has no relevant financial disclosures.


OCT may someday become standard of care

Historically, many of us have attributed the ophthalmic findings associated with shaken baby syndrome (SBS) to vitreoretinal traction and the repetitive acceleration-deceleration forces associated with this abuse. However, before OCT, most of what we knew about the pathology in SBS was from autopsy specimens. Multiple OCT studies have now clearly demonstrated the presence of vitreoretinal traction in SBS. OCT definitely helps identify the reason why there are certain ocular findings in SBS (eg, macular retinoschisis). Similar to the adult population, OCT in children has given us more insight into the pathophysiology of many pediatric retinal diseases.

R.V. Paul Chan, MD, FACS 

R.V. Paul Chan

Now, having said that, do we have enough information regarding OCT findings in SBS to support OCT as the gold standard for diagnosis of SBS? The simple answer is no. What we have right now are small descriptive reports on OCT findings for cases of SBS that were generally already diagnosed with the condition. So even though OCT is not required to make the diagnosis of SBS, it can be  useful in confirming pathology that we suspect in our clinical examination. It is an adjunct and can help determine the patient’s prognosis. The diagnosis of SBS, however, can still be made by the patient history and the presence of ocular findings such as retinal hemorrhages, retinal folds and hemorrhagic retinoschisis in the setting of other findings such as head injury or multiple fractures.

We also have to keep in mind that OCT for very small children is not widely available to everyone and is very expensive. In the future, however, my prediction is that all of these imaging techniques for pediatric patients will be used freely because that is the route we are going.

I am a huge proponent of digital imaging for children. I believe in the technology. But I think that right now OCT for the smallest infants is not ready for prime time and is not easy to integrate into our daily practice.

R.V. Paul Chan, MD, FACS, is an OSN Pediatrics/Strabismus Board Member. Disclosure: Dr. Chan has no relevant financial disclosures.