Did you arrive here by via search engine?
Click here to view the original version of this article

Click to Print This Page
(This section will not print)

Eye of the Beholder
Martin Carey, M.D.

Dr. Carey reports no commercial conflict of interest.

Release Date: 07/06/2001
Termination Date: 07/06/2004

Estimated time to complete: 1 hour(s).

Albert Einstein College of Medicine designates this enduring material for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Albert Einstein College of Medicine is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
Learning Objectives
Upon completion of this Cyberounds®, you should be able to:

This month's Cyberounds® presentation starts not with educational objectives but, instead, with a case presentation. Together, we will work through the case, then discuss its important aspects. Only after we have discussed the case will we focus on the educational objectives. In this way, I hope, the presentation will be more of a learning experience.

Case Presentation


It is 04:00. You are in a busy inner city emergency department. Your next case is a 5-month-old child. The child is brought in by the mother. The triage note states that the child has been off her feeds, has vomited and seems miserable.

When you enter the room, you note a mother sitting on a chair with the baby lying on the bed. The mother looks very young and questioning reveals that she is 17 years old. The baby was born at 32 weeks gestation and spent the first four weeks of her life in hospital. The birth weight was 4 pounds 3 ounces. Since going home, you learn that the child has been bottle fed. The baby is described by the mother as a 'colicky baby' who tends to cry a lot and always seems to want to be picked up. The mother is single but lives with a new boyfriend who she met during her pregnancy and who encouraged her to move from her home, about 150 miles away, to the city to be with him while he was looking for work. She admits to smoking 20 cigarettes a day and using 'occasional pot.' She denies alcohol use.

The reason for this attendance in the emergency department is that the child had some vomiting after her nighttime feed. She seemed to be more irritable than usual. The mother describes an episode where the child 'seemed to stop breathing' and 'just stared' for about 30 seconds. Further questioning reveals that, in fact, the child had been more irritable than usual during the whole of the previous day and had not taken her feeds well all day. There had been one or two 'staring' episodes earlier in the day.

Physical Exam

Examination revealed a thin white female infant. She appeared to be drowsy but could be roused. Her diaper is dirty but there is no evidence of any bruising or other external trauma. Her pulse rate is 135, respiration rate 28 and her temperature 99.5°F rectally. Her pupils are equal. Her fundi could not be visualized. Cardiovascular and respiratory examination were unremarkable. The abdomen appeared soft and was not distended. Examination of the ears, nose and oropharynx were also normal.

Q. What would be your differential diagnosis in this case?

A. The differential for this case is relatively broad but the following problems should be considered: seizure disorder, meningitis, failure to thrive, viral syndrome, shock, central nervous system hemorrhage and child abuse.

While you are outside the room, writing up the chart and thinking about the differential, the mother suddenly comes to the door and cries out for help. You enter the room and find the infant to be apneic. Stimulation fails to elicit a response and you elect to intubate the child. You are able to manage intubation without problems and without the use of any pharmacological agents. You use a 3.5 mm uncuffed endotracheal tube. Pulse rate, which had fallen to 55, rises again with ventilation to 130.

Q. What would you want to do next?

A. Your differential diagnosis now focuses on infection, major metabolic disruption or injury. A review again of the history with the mother would be appropriate.

During this second review, the mother now states that she left the child with her new boyfriend 'for a while' during the day when she went out to visit with some new friends. The child had not been well since then. The mother does state, when asked directly, that she has been concerned about her boyfriend's 'rough handling' of her baby on occasion. Bloods are drawn and a CT scan of the head is arranged. The CT reveals a subdural hematoma.

Q. What is the diagnosis now?

A. The most likely diagnosis in this scenario is nonaccidental injury.

The child is, at this stage, noted to have fixed, dilated pupils. Funduscopic examination reveals large, bilateral retinal hemorrhages.

Q. What specific type of child abuse do you think this infant has suffered?

A. The presence of retinal hemorrhages, sub-dural hematoma in a fussy infant with non-specific symptoms is seen in cases of 'Shaken Baby Syndrome.'

Q. What is the likely cause of the findings in this condition?

A. It is thought that the findings typical of Shaken Baby Syndrome occur after an infant is held at arm's-length by an adult and then violently shaken either forward and backward, or from side to side, for up to a minute or so. The heavy head, relative to the torso, in association with the poor head control, results in the head being whipped backwards and forwards. It is estimated that G forces of 20-30 can be exerted on the head in this situation, enough to produce tearing of intracranial and retinal vessels, as well as significant shear injury of the neurons.

One variation of the Shaken Baby Syndrome occurs when the baby is shaken up and down violently and repeatedly on a surface such as a mattress. The sudden cessation of head movement in these cases is thought to produce even higher G forces than simply shaking the child back and fore. This condition is sometimes referred to as the 'Shaken Impact Syndrome.'

The shearing forces, together with bleeding from intracranial vessels, which may be intracerebral, subdural or subarachnoid, cause significant cerebral edema and are thought to account for many of the non-specific symptoms and signs with which these children present. When combined with the delay in presentation and the frequent lack of a full history, it is hardly surprising that the outcomes in these cases can be so grave.

The infant is ventilated but appears to be seizing. The seizures are controlled and the child is admitted to the intensive care unit. You later hear that the child died, presumably from uncontrollable cerebral edema.

Q. What is the mortality from Shaken Baby Syndrome?

A. This is a condition with very high morbidity and mortality. Studies have shown mortality rates as high as 30%. Morbidity rates are difficult to estimate but are also thought to be highly significant -- 30-40%. Seizures are common, occurring in about 40-70% of patients. The presentation of this condition, however, may be very subtle. Complaints are often vague and non-specific, and may include poor feeding, failure to thrive, seizures, fever, vomiting, symptoms of an upper respiratory infection, colic, lethargy, irritability, staring episodes, apnea and a bulging fontanelle.

After reviewing this case, it is clear that the focus of this Cyberounds® is to be Child Abuse.

Educational Objectives

Upon completion of this Cyberounds®, the participant should be able to:

  • List the signs and symptoms of Shaken Baby Syndrome
  • Discuss the pathophysiology of Shaken Baby Syndrome
  • Discuss the epidemiology of child abuse
  • List the risk factors for non accidental injury in children
  • Identify a few resources and approaches that may help medical professionals prevent child abuse in their own practices.

Q. How big a problem is child abuse these days?

A. There has been a significant increase in the number of cases of child abuse reported in the United States over the past few years. Between 1985 and 1993, the number of cases increased by 50%. There were a reported 3 million cases of child abuse in 1993, of which 35% were confirmed. Part of the increase could be the result of improved recognition and reporting but many experts believe that there has been a real increase in the number of cases.

Each year 160,000 children suffer severe or life-threatening injuries as a result of abuse and 1000 to 2000 children die as a direct result.(1) Between 1.3 and 10% of children seen in the emergency department for an injury are there because of abuse.(2)

Q. But I am familiar with child abuse and I am sure everyone else is too. Surely we don't miss many cases in the emergency department?

A. A study by Jenny et al of 173 children with a head injury secondary to abuse (all of whom were under the age of three years) revealed that 54 (31%) had been missed on the initial visit. Of these cases, 15 (28%) had suffered further injury before the child abuse was diagnosed. The authors suggested that four deaths may have been prevented by earlier diagnosis.(3)

Q. What factors seem to put children at increased risk for child abuse?

A. Studies consistently report that significant risk factors for abuse can be divided into parental factors, child related factors and situational factors.

Parental Factors

Parental factors are responsible for the majority of risk factors. They include low socioeconomic status and unstable family situations. Teenage parents, or single parents, are additional risk factors. Parents who have unrealistic expectations of their children are also at increased risk of committing violence. Mothers are the most common abusers but in situations where the father is the primary caregiver -- especially if due to unemployment -- this statistic is reversed.

The presence of drug or alcohol abuse greatly increases the risk of violence, as does a history of previous mental disorder. A lack of a support structure -- a close friend or relative in whom the parent can confide -- also increases the risk. Child abuse is 15 times more likely to occur in homes where there is spousal abuse. It is also more common in the homes of people who were themselves abused as children.

Child Related Factors

Child related factors include children of unwanted pregnancies or illegitimate births, or children born from a previous relationship. Children who exhibit certain traits, such as poor eating, excessive crying, abnormal sleeping patterns, behavioral disorders, including hyperactivity, chronic disease or mental or physical handicap, are all at increased risk. Prematurity, particularly, seems to be a risk factor, as it often results in both chronic disease and a disruption of the mother/baby bond because of the initial prolonged hospitalization.

Situational Factors

Situational factors are crisis situations that can overload the abuser's abilities to cope. Sudden worsening of a financial situation due to illness or unemployment, family arguments or prolonged absence of a partner can all trigger violence.

Q. What might you do if you suspect a case of child abuse in your practice?

A. First of all, it is important for the physician to try and put aside any feelings of anger or hostility. This is an emotive issue and some of the cases are truly horrific but the child can best be served by maintaining a professional and ethical demeanor at all times.

A full and detailed history should be recorded. This should include information about the person giving the history -- parents, family members, the child themselves and caregivers. Using actual quotations and avoiding 'interpretation' are very important. Some information about the demeanor of the history giver is also useful. Was the mother in tears or did she seem removed from the situation? Was the father agitated and aggressive?

Particular information that should be elicited includes the length of time between the injury, or the first time the child was noted to be 'sick,' and the time of presentation, details about the events surrounding the injury (height of any fall, surface on which the child fell, people present, how the fall happened in the first place) and the state of the child. The injury should be carefully described -- a diagram is very useful -- and any other additional marks, scratches, cuts or bruises should also be noted.

A review of the child's medical history is also warranted. Multiple previous visits, especially for many injuries, should raise a red flag. Detailed histories of the parents' backgrounds are also useful, as they may highlight multiple risk factors for child abuse.

Once a full history has been performed, a detailed examination should occur. This should start at the top and move downwards. The head should be inspected for bruises, lacerations or hematomas. The fontanelle should be palpated. The eyes should be examined for hemorrhages. Examination of the fundi should occur even if this requires pupillary dilatation (after neurological stability is assured). An ophthalmologist may be required to perform this procedure. One of the well-known pitfalls for misdiagnosis of nonaccidental injury is the presence of osteogenesis imperfecta. This condition is associated with blue sclera.

The ears should be examined for evidence of tympanic membrane rupture or blood behind the drum. The nose is inspected for bleeding or hematomas. Inspection of the mouth may visualize a torn frenulum of the upper lip. This injury usually occurs secondary to a direct blow to the face.

Examination of the neck may reveal classical 'finger print' bruises from attempted strangulation. Other bruise patterns may also be noted.

Cardiovascular and respiratory examination may reveal chest wall tenderness from possible rib fractures, or again show bruises or hematomas.

Abdominal examination is very important. A tense, tender, distended abdomen, especially if associated with silent bowel sounds, may be an indicator of significant intra-abdominal injury, including a perforated viscus. This injury may occur as a result of a direct blow to the abdomen from a fist or foot. Diagnosis may, as is typical with nonaccidental injury, be delayed, and, thus, the child is frequently very unwell by the time they present. Interestingly, a review of the literature, reported in Pediatrics by Huntimer et al, suggests that although an explanation for a perforated viscus given by parents is frequently a fall down a short flight of stairs, there is no evidence to suggest that a fall of this type does result in a perforated viscus.(4)

Although this discussion does not mention sexual abuse, it is important to consider the possibility. The genitourinary and rectal regions need to be carefully examined.

Careful palpation of all of the long bones should occur, looking for tenderness or deformity. Any bruises, marks, healed scars or other blemishes should be recorded.

Radiographs should be taken of any suspicious area and, if there is a high index of suspicion, then a review of each of the long bones and the torso should be performed. The use of a 'babygram' (whereby the infant is placed on a large film and a single shot taken) is no longer recommended, as the quality of the film is inferior to that produced when individual areas are examined. Many cases of nonaccidental injury are spotted by the radiographs (multiple rib fractures in a young child presenting with cough, metaphyseal fractures in a child with a limp, for example) and radiographs are a very important aspect of assessment.

There is some debate in the literature as to the best investigation for children with head injury. It seems that CT scan, although not as sensitive for all injuries (especially shear type injury as seen in Shaken Baby Syndrome) as MRI, is probably the investigation of choice for any infant noted to have a significant history of head injury, including loss of consciousness, vomiting or incidentally noted to have a skull fracture.

Q. Once child abuse is suspected, what does the emergency department physician do next?

A. The vast majority of these children should probably be admitted to the hospital. This allows for investigation of the injury, together with its management, to occur in a calm and organized fashion. Each state mandates reporting of cases where there is a high index of suspicion of child abuse. Many other countries outside of the United States have similar rules and regulations.

Each emergency physician should be aware of the rules and requirements peculiar to their own environment. They should also know who the child abuse expert is in their area and how to access them. Many hospitals have protocols for this and it is very important that physicians familiarize themselves with these. Note that in cases where abuse is suspected, immediate emergency treatment of the child can occur, including admission to hospital, without the parents' permission. If necessary, a court order can then be obtained to allow further management.

Q. So, there is a lot of child abuse. Is there anything that emergency department staff can do to prevent it?

A. The United States Advisory Panel on Child Abuse and Neglect stated that only a universal system of early intervention, grounded in the creation of caring communities, could provide an effective foundation for confronting child abuse. Successful strategies to prevent child abuse require intervention at all levels of society. Unfortunately, although a large number of programs have been developed, few have undergone rigorous testing, evaluation and measurement for outcomes.

Those programs that have been shown to be effective strengthen the family by building community connections and support. This is done by establishing links with existing community support programs and providing support settings for parents where they can learn from a group of their peers. In case of an emergency, there should be 24-hour access to assistance and support, perhaps through the support group structure. The community needs to become aware of the importance of family. The community should also provide education and instruction in good parenting practices.

For the emergency physician this may all seem 'someone else's' job. However, there are things we can all do in our departments to help reduce the toll of child abuse. Today, many people, especially those at most risk for being abusers, use the emergency departments for their routine medical care. It is important that emergency physicians are aware of the support services that exist in their own communities and that they know how to contact them. Emergency physicians must be aware of the risk factors for child abuse and be willing to step in before the abuse actually occurs by offering those at risk information about community and social services available.

In multiethnic communities, information needs to be available in a range of languages. Many local service agencies distribute pamphlets and literature about their programs. These should be available to the physician for their patients. Intervention may start with the newly diagnosed teen pregnancy or the child of young parents brought in for persistent crying at 3:00 A.M.. The medical staff should identify victims of domestic violence, drug abuse or alcohol abuse who have young children at home. If the physician addresses the issue of child abuse with these parents in a non-threatening, supportive manner, the physicians may make a significant difference. This is especially true if the discussion is followed with literature detailing supportive services.

Finally, the emergency physician often enjoys a position of respect in the community. This may make the physician a strong advocate for change within the community should they decide to support child care issues. Involvement can be through city or state commissions, or through local organizations and institutions. If you become a vocal supporter for child abuse prevention, you may make a real difference in both the local community and society at large.

Other Resources

Some information on child abuse issues can be obtained from the following agencies:

National Council on Child Abuse and Family Violence
1155 Connecticut Ave, Suite 400
Washington DC 20036

National Committee to Prevent Child Abuse
332 S Michigan Ave, Suite 1600
Chicago IL 60604-4357
Tel: 312 663 3520


4Huntimer CM. Muret-Wagstaff S. Leland NL. Can falls on stairs result in small intestine perforations?. Pediatrics. 2000; 106(2 Pt 1):301-5.