Sentinel Injuries Research Articles

Orofacial manifestations of child maltreatment: A review.

The recognition and reporting of child abuse and neglect continues to be a challenge in the healthcare setting. All healthcare providers, including dentists, should be aware of the high prevalence of orofacial injuries and conditions that are concerning for abuse and neglect. Sentinel injuries may appear trivial and do not require medical attention, but are unlikely caused by accidental mechanisms and, if not properly identified, often precede more severe abusive injuries. Concerning orofacial findings can manifest as bruising, eye injuries, intraoral injuries, pharyngeal perforations, facial bone fractures, and sexually transmitted infections. Abusive caregivers are likely to give inadequate explanations or no history at all to explain concerning findings. Medical providers' failure to make mandated reports about their concerns to the appropriate agencies can have significant long-term impacts on the physical and psychological well-being of children.

Child physical abuse screening in a pediatric ED; Does TRAIN(ing) Help?

BackgroundChild maltreatment is distressingly prevalent yet remains under-recognized by healthcare providers. In 2015 the Ohio Children's Hospital Association developed the Timely Recognition of Abusive INjuries (TRAIN) collaborative in an effort to promote child physical abuse (CPA) screening. Our institution implemented the TRAIN initiative in 2019. The objective of this study was to examine the effects of the TRAIN initiative at this institution.MethodsIn this retrospective chart review we recorded the incidence of sentinel injuries (SIS) in children presenting to the Emergency Department (ED) of an independent level 2 pediatric trauma center. SIS were defined and identified by a diagnosis of ecchymosis, contusion, fracture, head injury, intracranial hemorrhage, abdominal trauma, open wound, laceration, abrasion, oropharyngeal injury, genital injury, intoxication, or burn in a child < 6.01 months of age. Patients were stratified into pre-TRAIN (PRE), 1/2017–9/2018, or post-TRAIN (POST), 10/2019–7/2020, periods. Repeat injury was defined as a subsequent visit for any of the previously mentioned diagnoses within 12 months of the initial visit. Demographics/visit characteristics were analyzed using Chi square analysis, Fischer’s exact test, and student’s paired t-test.ResultsIn the PRE period, 12,812 ED visits were made by children < 6.01 months old; 2.8% of these visits were made by patients with SIS. In the POST period there were 5,372 ED visits, 2.6% involved SIS (p = .4). The rate of skeletal surveys performed on patients with SIS increased from 17.1% in the PRE period to 27.2% in the POST period (p = .01). The positivity rate of skeletal surveys in the PRE versus POST period was 18.9% and 26.3% respectively (p = .45). Repeat injury rates did not differ significantly in patients with SIS pre- versus post-TRAIN (p = .44).ConclusionImplementation of TRAIN at this institution appears to be associated with increased skeletal survey rates.

Does support received for subsequent injuries differ between Māori and non-Māori? Findings from a cohort study of injured New Zealanders.

To examine if differences exist between injured Māori and non-Māori in accessing and receiving support from the Accident Compensation Corporation (ACC) for treatment and rehabilitation of subsequent injuries. This cohort study utilised participants' self-reported data from the Prospective Outcomes of Injury Study, and ACC claims data. Approximately one-third of Māori (32%) and non-Māori (35%) who self-reported a subsequent injury had no associated ACC claim. Statistically significant differences in this outcome (i.e., self-reported subsequent injury but no ACC claim) were found between Māori and non-Māori when comparing across occupation type and severity of participants' sentinel injuries. Few differences were observed between Māori and non-Māori in the percentages of ACC claims accepted that compensated various treatments and supports; this was similar for average compensation amounts provided. Māori and non-Māori who received support from ACC for a sentinel injury prior to sustaining another injury appear to have received equitable ACC compensation for the treatment and rehabilitation of the subsequent injury with two potential exceptions. Further research is needed to determine how generalisable these findings are. Establishing routine systems for collecting data about the support needed, treatment pathways and outcomes once accessing ACC support is vital to ensure positive and equitable injury outcomes for Māori.

Caregiver Reports of Infant Distress and Injury in Abused Infants

To determine whether in infants evaluated for physical abuse, medical encounters for infant distress are correlated with physical abuse or a history of sentinel injuries. This retrospective, case-control analysis of infants aged <12months evaluated for physical abuse identified demographic characteristics, prior injuries, and medical encounters for infant distress. Variables were compared between abused infants and nonabused infants with and without sentinel injuries. A nonparametric recursive classification tree analysis assessed interactions between variables. Infant distress was associated with abuse (67.9% vs 44.7%; P=.008; OR, 2.6; 95% CI, 1.3-5.2). Infants with sentinel injuries had higher rates of infant distress (74.1% vs 42.4%; P≤.001) and crying (81.5% vs 62.7%; P=.012). Previous falls (32.6% vs 18.1%; P=.03) and nonsentinel injuries (18.2% vs 5.4%; P=.002) also were associated with abuse, although sentinel injuries were the most important predictor of abuse, followed by infant distress. Infants with medical encounters for distress and injury may be at higher risk for abuse and may benefit from intensive educational and support services for their caregivers. Additional research evaluating the most effective interventions for caregivers of fussy infants is needed.

Variations in Guidelines for Diagnosis of Child Physical Abuse in High-Income Countries

The highly variable practices observed regarding the early detection and diagnostic workup of suspected child physical abuse contribute to suboptimal care and could be partially related to discrepancies in clinical guidelines. To systematically evaluate the completeness, clarity, and consistency of guidelines for child physical abuse in high-income countries. For this systematic review, national or regional guidelines that were disseminated from 2010 to 2020 related to the early detection and diagnostic workup of child physical abuse in infants aged 2 years or younger by academic societies or health agencies in high-income countries were retrieved. The definitions of sentinel injuries and the recommended diagnostic workup (imaging and laboratory tests) for child physical abuse were compared. Data were analyzed from July 2020 to February 2021. Within the 20 included guidelines issued in 15 countries, 168 of 408 expected statements (41%) were missing and 10 statements (4%) were unclear. Among 16 guidelines characterizing sentinel injuries, all of them included skin injuries, such as bruises, hematoma, or burns, but only 8 guidelines (50%) included intraoral injuries and fractures. All 20 guidelines agreed on the indication for radiological skeletal survey, head computed tomography, and head magnetic resonance imaging but differed for those of bone scintigraphy, follow-up skeletal survey, spinal magnetic resonance imaging, cranial ultrasonography, chest computed tomography, and abdominal ultrasonography and computed tomography. Additionally, 16 guidelines agreed on exploring primary hemostasis and coagulation but not on the tests to perform, and 8 guidelines (50%) mentioned the need to investigate bone metabolism. These findings suggest that guidelines for the diagnosis of child physical abuse in infants were often clear but lacked completeness and were discrepant on major issues. These results may help identify priorities for well-designed original diagnostic accuracy studies, systematic reviews, or an international consensus process to produce clear and standardized guidelines to optimize practices and infant outcomes.

Child Maltreatment and Medical Fragility: Fatalities in the First Year of Life

In this month’s Pediatrics, Schneiderman et al examine infant deaths for medical causes among infants previously reported to child protective services.1 Using probabilistically linked birth records, child protective service records, and death records, the authors examined risk of death through the first year of life. Their analyses controlled for sex, race and ethnicity, maternal age, early prenatal care, insurance status, birth weight, and prematurity. Infants reported for maltreatment had an almost twofold increase in risk of for any medical cause of death.Are infants with serious medical conditions at higher risk for maltreatment or is maltreatment more likely to be detected among those with serious medical conditions? An increased risk for death would follow from prematurity, congenital abnormalities, or intrauterine exposure to infection.2 Children with these conditions likely require careful and more frequent medical encounters. Missed appointments and lack of follow-through on medications and treatment may be life-threatening. Medical teams caring for chronically ill infants are often sources of reports for neglect for infants whose caregivers fail to keep scheduled appointments, accept referrals, or take prescribed medications. Furthermore, the increased frequency of medical visits over the first year of life for ill children may also increase detection of sentinel injuries.3 Maltreatment may also be increased beyond failures in following medical care recommendations. Infants with chronic disease can contribute to financial and emotional stresses on their parents. Ill infants may be less capable of the responsiveness that helps create a strong bond between parent and child.4 Infants with chronic conditions may have high-pitched cries or disordered sleep patterns that add to parental stress. It therefore follows that infants with medical conditions are both at higher risk of maltreatment and more likely to be reported.The data from Schneiderman et al reveal that the outcome for infant medical conditions is dramatically altered by the family situation. Importantly, infants reported to child protective services and placed in foster care had the risk of mortality decreased by half. In the unadjusted analysis, the youngest mothers (<20 years old) had a 34% increase in the risk of infant mortality compared with older mothers. Poverty increases the risk for abuse and neglect.5 Young parents, with less schooling and fewer years in the workforce, are at an additional economic disadvantage. Poverty impacts survival of children with chronic medical conditions because poor families have increased exposure to risky environments and decreased access to health services.6Interventions to address this additional risk of mortality must be multifaceted.7 These interventions include paid parental leave, increased minimum wage, paid earned income tax credits, and access to long-acting reversible contraceptives for teenagers to prevent unwanted pregnancies.8 Many of these approaches address poverty directly. Community engagement programs such as Strong Communities9 reduce child maltreatment by giving visibility to the problem of child abuse and broadly engaging community members and medical practices in programs to support new parents. Home-visiting programs like the Nurse-Family Partnership10 for young first-time mothers enrolled prenatally and SafeCare11 for mothers of any parity with children enrolled in 20-week home-based educational program offer explicit and evidence-based support to new parents and have been shown to be successful.Additional strategies are needed after reports are made to child protective services. For example, child protective service agencies should include medical expertise in treatment planning and/or form partnerships with public health agencies to bring health expertise to judgements about the significance of missed appointments or undelivered medications.Forewarned is forearmed. Medical conditions represent a serious threat to the lives of infants reported to the child protection system. General and subspecialty pediatricians who have reported an infant for neglect or abuse or who have a reported infant in their practice must ensure continued care for the infants after a report. Child protective service professionals need to be aware of the elevated risk of death among their charges from disease and facilitate ongoing medical care.

Physical Abuse Creating Cauliflower Ear in an Infant: A Discussion of Mechanism and Review of the Literature.

We evaluated an infant with flagrantly abusive injury. However, he had a history of prior isolated ear hematoma that was not properly recognized as a sign of trauma, harshly illustrating a missed opportunity when early diagnosis may have prevented escalating abusive injury. In addition, we reviewed and summarized the last two decades of literature related to this topic to better understand previously described findings in similar populations; these were scant. This case report emphasizes the importance of detecting sentinel injuries that forebode child maltreatment.

The Challenge of Identifying Pediatric Abusive Head Trauma During the COVID-19 Pandemic.

The Challenge of Identifying Pediatric Abusive Head Trauma During the COVID-19 Pandemic.

What's in a name? Sentinel injuries in abused infants.

Infants are at greatest risk of severe and fatal physical abuse yet they sometimes present for medical care multiple times with abusive injuries prior to being diagnosed with abuse and having protective actions taken. Efforts to identify these infants in a timely manner are critical to prevent repeated, escalating abuse and subsequent harm. Increasing the identification and evaluation of sentinel injuries has been highlighted as a strategy for improving timely detection of abuse in infants. Sentinel injuries are visible, minor, poorly explained injuries in young infants that raise concern for abuse. These injuries include cutaneous injuries such as bruising, subconjunctival hemorrhages and intra-oral injuries. Sentinel injuries can signal concurrent clinically occult but more serious injuries or precede more significant trauma from abuse. As such, sentinel injuries offer an opportunity to intervene and protect infants from further harm. A thorough physical exam is critical for detecting sentinel injuries. Imaging with skeletal survey and, when appropriate, neuroimaging are key components of the medical evaluation of sentinel injuries in these high-risk infants.

Population study of orofacial injuries in adult family violence homicides in Victoria, Australia

BackgroundThis study describes the prevalence and orofacial injury patterns associated with adult family violence (FV) homicides in Victoria, Australia. It follows a methods study for case selection of all FV homicides and injury measurement. Comprehensive analysis of orofacial injuries in FV homicides and their clinico-demographic context will inform future research on clinical FV indicators and sentinel injuries, and potentially lead to premorbid intervention in health services. MethodsAll closed cases of FV homicides aged ≥18 years, January 2006 to December 2018, were identified by screening Victorian fatal assaults, based on victim-offender relationship. Primary data such as post-mortem computed tomography scans and photographs were assessed. Socio-demographic, clinical, interpersonal and incident parameters were descriptively analysed and statistically compared across FV homicides with and without facial injuries using cluster analysis and nonparametric tests. ResultsOf 170 adult homicides screened for eligibility, 151 were included for final analysis. Over the 12-year period, 78.1% of all Victorian adult FV homicides had orofacial injuries. Significant cluster patterns of injury mechanism, victim-offender relationship and drug/alcohol impairment were identified in all homicides. Non-facial injuries were significantly higher in facial vs. non-facial injury homicides. Facial abrasion and incised wounds were the most common injury types. ConclusionsThis is the first forensic-epidemiologic study evaluating the empirical evidence concerning orofacial injuries associated with population-wide adult Victorian FV homicides. The high level of orofacial injuries in this population during the study period may inform clinical practice and policy in FV intervention in Victoria and globally.

Predictors of severe or multiple subsequent injuries over 24 months among an already-injured cohort in New Zealand

IntroductionInterventions to reduce subsequent injuries among already-injured people presenting to healthcare providers may reduce the overall burden of injury. However, in order to develop such interventions it is necessary to understand what predicts subsequent injuries. This knowledge is currently limited for general injury populations. This prospective Subsequent Injury Study aims to determine pre-injury sociodemographic and heath factors, and injury-related factors, that predict subsequent injury claims reported to Accident Compensation Corporation (ACC, New Zealand's universal injury insurer) in the 24 months following an ACC entitlement claim injury (‘sentinel’ injury). Two separate outcome variables were used to identify subsequent injuries of interest: having (1) at least one high severity injury claim (New Injury Severity Scores ≥4), or (2) two or more claims (injuries of any severity). MethodsThis study combines: (1) participant interviews (from our earlier Prospective Outcomes of Injury Study), (2) ACC data about sentinel injuries which led to participants being recruited, and all subsequent injuries involving an ACC claim in the 24 months following that sentinel injury, and (3) hospital discharge data for sentinel and subsequent injuries involving hospitalisation. Potential predictors (pre-sentinel injury sociodemographic and health variables, sentinel injury related variables) were identified using modified Poisson regression models. ResultsSevere and multiple ACC-reported subsequent injuries (ACC-SI) were each more likely to occur for participants with a self-reported prior injury affecting them at the time of their sentinel injury, and for those who prior to their sentinel injury event had pain/discomfort, or were physically active on five or more days of the week. A lower likelihood of severe or multiple ACC-SI was more common in those aged 30–64 years compared with those aged 18–29 years, those of ‘other’ ethnicity compared to sole New Zealand European ethnicity, and if the sentinel injury involved hospitalisation. Multiple ACC-SI were more likely for males, and for participants with ≥2 chronic conditions compared to those with none. ConclusionsFactors identified in this study may provide useful flags to help healthcare providers and policy makers identify people at increased risk of severe or multiple subsequent injuries who may benefit from targeted injury prevention strategies or interventions.

Characteristics That Distinguish Abusive From Nonabusive Causes of Sudden Unexpected Infant Deaths.

Fatal child abuse can be mistaken for sudden unexpected infant death (SUID) in the emergency department setting. It is unknown if there are characteristics that distinguish abusive from nonabusive causes of SUIDs in the emergency department. Using a matched case-control design, we reviewed the medical examiner records of deaths of infants younger than 12 months who were found unresponsive at a residence, required cardiopulmonary resuscitation, and had a complete forensic autopsy between 2009 and 2015. Infants with a manner of death as homicide were cases; controls were those with the manner of death as accident, natural, or undetermined. Each case was matched with 5 controls based on age (months). Differences between cases and controls were evaluated with respect to demographic, parental, and household characteristics and clinical outcomes. We identified 12 cases (homicides) and 169 controls (nonhomicides), of which 60 were selected for the matched analysis. We found no significant differences between cases and controls with respect to age, race, sex, maternal substance use, Child Protective Services involvement prior to death, presence of male head of household, surviving siblings, or emergency medical services transport. Cases were more likely to have Child Protective Services involvement at the time of death (83% vs 38%; P = 0.01), sentinel injuries (odds ratio, 9.67; 95% confidence interval, 1.30-122.43), and return of spontaneous circulation (odds ratio, 29.99; 95% confidence interval, 3.70-241.30). Child Protective Services agency involvement at time of death, sentinel injury, and return of spontaneous circulation were more often associated with abusive causes of SUID. Further study is needed to confirm these findings.

Concordance between sentinel and subsequent injuries: A prospective study of injured New Zealanders

ObjectivesTo describe: 1) settings, activities and types of injuries for an ‘initial’ (sentinel) injury and subsequent injuries over 24 months, and 2) concordance between sentinel and subsequent injury events. MethodsParticipants (n = 2856) were recruited to the Prospective Outcomes of Injury Study (POIS) following their sentinel injury event, an injury event resulting in an Accident Compensation Corporation (ACC) entitlement claim. Subsequent injuries were those from additional ACC claims in the following 24 months. Injury settings, activities and types were from electronic ACC claims data. The risks of having a subsequent injury of the same type as the sentinel injury were estimated. ResultsOverall, 1653 (58%) participants had 3444 subsequent injury events in 24 months, resulting in 4470 injury diagnoses. Twenty one percent had at least one subsequent injury event of the same type as their sentinel injury; 33% with a spine sprain/strain had at least one subsequent spine sprain/strain. Many participants had at least one subsequent injury event at same setting (26%) as their sentinel injury; of note, 36% of participants whose sentinel injury occurred at home had at least one subsequent injury at home. Seventeen percent of participants had at least one subsequent injury involving the same activity as their sentinel injury; 28% of those whose sentinel injury was a result of contact in sport had at least one subsequent injury also involving sport. ConclusionsSubsequent injuries among people presenting to healthcare providers are common. Greater emphasis should be placed on maximising such healthcare provider contact as an injury prevention opportunity.

Child Maltreatment Red Flags: Two Cases of Bruising in Premobile Infants

Child maltreatment is a serious public health concern in the United States. Young infants and children younger than 3 years are at the highest risk of being abused and can experience both acute injuries and long-term developmental, behavioral, and mental health problems. Health care providers are mandated reporters of suspected abuse but may misdiagnose potentially abusive injuries because of lack of knowledge in recognizing maltreatment. Premobile infants rarely have bruising or intraoral injuries without a reported accident or underlying systemic disease and should raise concern for abuse. It is not uncommon for an abused child to present with an injury that at first glance may seem trivial but is actually suspicious for physical abuse and later be found to have abusive fractures or head trauma. The following case presentations show the importance of recognizing sentinel injuries and red flags for maltreatment in young, premobile infants with unexplained bruises.

Providers' Ability to Identify Sentinel Injuries Concerning for Physical Abuse in Infants.

The objectives of this study were to assess the ability of pediatric health care providers and social workers to recognize sentinel injuries in infants under 6 months of age and to determine what factors influence their decision to evaluate for physical abuse. A statewide collaborative focused on sentinel injuries administered a survey to pediatric health care providers and social workers in the emergency department, urgent care, and primary care. The survey contained 8 case scenarios of infants under 6 months of age with an injury, and respondents were asked if they would consider the injury to be a sentinel injury requiring a physical abuse evaluation. Respondents were then presented with several factors and asked how much each influences the decision to perform a physical abuse evaluation. A total of 565 providers completed the survey. Providers had moderate interrater reliability on their classification of the cases as sentinel injuries or not (κ = 0.57). Nearly all respondents (97%) recognized genital bruising as a sentinel injury, whereas 77% of respondents recognized intraoral injuries. Agreement was highest among social workers (κ = 0.76) and physicians with categorical pediatrics training and pediatric emergency medicine fellowship (κ = 0.63) and lowest among nurse practitioners (κ = 0.48) and residents (κ = 0.51). Concern over missing the diagnosis of abuse had the greatest influence on the decision to perform a physical abuse evaluation. Sentinel injuries are not uniformly recognized as potential signs of child abuse requiring further evaluation by pediatric health care providers. Additional evidence and education are needed regarding sentinel injuries.

History, injury, and psychosocial risk factor commonalities among cases of fatal and near-fatal physical child abuse

Failure to recognize child maltreatment results in chronic exposure to high-risk environments where re-injury or death may occur. We analyzed a series (n=20) of fatal (n=10) and near-fatal (n=10) physical child abuse cases from the Commonwealth of Kentucky to identify commonalities and determine whether indicators of maltreatment were present prior to the child's fatal or near-fatal event. We conducted retrospective state record reviews involving children <4years of age classified as physical child abuse by the Cabinet for Health and Family Services during a 12 month period. Cases were distributed across 17 counties. IRB approvals were obtained. Three reviewers concurrently abstracted case data from medical, social, and legal documents, and descriptive statistics were analyzed. Median age of subjects was 7.5 months (range 1–32 months); 55% were male. Psychosocial risk factors (PRFs) were present in 100% of cases. Traumatic brain injury (95%) and bruising (90%) were the most common injuries. Of the 14 children with available prior medical records, 9 (64%) had sentinel injuries in the form of prior unexplained bruising; all nine suffered subsequent traumatic brain injury resulting in four deaths. A male was caring for the child at the time of the final event in 70% of cases. Our study identified key commonalities across cases of fatal and near-fatal abuse, highlighting the prevalence of psychosocial risk factors and the significance of prior unexplained bruising as a herald of escalating abuse. Further study is warranted to ascertain the predictive value of our findings in the larger population.

Hospital-based screening tools in the identification of non-accidental trauma

Over 700,000 children are victims of abuse and neglect each year in the United States. Effective screening programs that entail broad capture of suspected non-accidental trauma (NAT) may help to identify sentinel injuries. This can facilitate earlier detection and psychosocial interventions in hopes of decreasing recurrent NAT, which confers a higher mortality rate. The purpose of this article is to outline essential components of hospital-based NAT screening tools and highlight existing programs. In general, these tools should include several components: education sessions for healthcare providers on how to identify signs of NAT, automated notes or checklists within the electronic medical record to prompt specialty referrals, and a multidisciplinary team of experts that can address the needs of these children in the acute care setting.

Physical abuse: more certainty now?

Abstract There is now a strong evidence base for the characteristics of inflicted injury. Professionals within the multidisciplinary team and the courts are increasingly relying on other features such as the child's presentation to help determine whether child maltreatment has occurred. These features are somewhat subjective in nature, with little or no evidence to assist the clinician, particularly where there is uncertainty surrounding the possibility of child maltreatment. This paper will examine further evidence on the physical signs of child maltreatment, focussing on primary research on bruising, sentinel injuries, occult rib fractures and conditions that mimic physical injury such as the resurgence in the use of cupping as a remedy for childhood illness. It will also examine emerging evidence on a child's presentation, including history, response to injury and time to seek medical attention. If uncertainty surrounding the causation of an injury remains following advice and peer review, this must be acknowledged and stated unequivocally.

Awareness of Sentinel Injuries Among Canadian Paediatricians: A Cpsp One-Time Survey

Abstract BACKGROUND: A sentinel injury is a visible or detectable minor injury in a pre-cruising infant that is poorly explained and therefore suspicious for physical abuse. Examples include bruises and intra-oral bleeding, which may be “warning injuries” for possible abuse. Timely recognition and management of a sentinel injury can potentially alter a pattern of escalating child abuse, but studies have demonstrated that in up to a third of cases, physicians miss early signs of abuse. OBJECTIVES: Our purpose was to assess paediatrician awareness of sentinel injuries in pre-cruising children as precursors to more serious physical abuse. DESIGN/METHODS: A Canadian Paediatric Surveillance Program (CPSP) one-time survey of Canadian paediatricians was conducted in March of 2015 to evaluate awareness of sentinel injuries and their significance. Respondents were provided with clinical vignettes describing infants with sentinel injuries and were asked questions about their differential diagnosis and management plan. Respondents were then classified as either “aware” or “not aware” based on their responses, and a multivariable logistic regression analysis was used to identify variables most strongly correlated with awareness. RESULTS: The survey achieved a response rate of 23.4%. Of the 582 respondents, 65% were aware that bruises and intra-oral injuries in pre-cruising children are red flags for possible physical abuse. Respondents were more likely to be aware of bruises as sentinel injuries (91.9%) than they were of intra-oral injuries (67.2%). Paediatric subspecialists were significantly less likely than general paediatricians to be aware of sentinel injuries as red flags for later serious abuse (adjusted OR=0.57, 95% CI 0.37–0.88, p=0.01). Other important trends included higher awareness with more recent completion of residency, and also with additional training in the area of child abuse and neglect. CONCLUSION: Over one-third of Canadian paediatricians are unaware that unexplained bruising and intra-oral bleeding in a pre-cruising child should prompt assessment for possible abuse. Further physician education is warranted to ensure that paediatricians understand the importance of sentinel injuries and the indications for initiating investigations to detect occult injuries of child abuse. Future directions include the development of physician training opportunities in the area of identifying and managing sentinel injuries, as well as extending the study population to include family physicians who provide primary care to children.

Minor injuries… major implications: Watching out for sentinel injuries.

An otherwise healthy two-month-old boy presented to his general paediatrician for follow-up of gastroesophageal reflux disease. His mother reported that since starting ranitidine two weeks earlier, his discomfort following feeds and previously reported excessive crying had both improved. She had no new concerns about the child’s health but reported that she had observed a small amount of blood on the infant’s crib sheet in the past week, which she could not explain. On examination, he appeared alert with normal vital signs and had adequate weight gain since his last visit. Cardiovascular, respiratory, abdominal and neurological examinations were within normal limits. The paediatrician made note of a solitary, bluegreen bruise on the baby’s left anterior thigh, which measured approximately 2 cm in diameter. He inquired about its origin; however, the child’s mother stated that she had not noticed it and was unsure of the causal mechanism. There were no other skin findings, bruises or petechiae noted. Examination of the oral cavity was unremarkable. The paediatrician prescribed a three-month supply of ranitidine and recommended routine follow-up with the family physician for well-baby care and immunizations. One month later, the child was brought to the local emergency department by ambulance after his mother found him unresponsive in his crib, with jerking of his arms and legs. There was no history of recent illness or trauma. On examination, he was pale and lethargic. His heart rate was 118 beats/min and his blood pressure was 86/58 mmHg. He was afebrile, with shallow respirations and an oxygen saturation of 96% on room air. He exhibited tonicclonic movement of his right upper and lower extremities. His pupils were equal and reactive; however, intermittent nystagmus was noted. An area of soft tissue swelling was palpable at the base of the skull and there were two green bruises on the left cheek. The child was admitted to the paediatric medical unit. Computed tomography of the head revealed bilateral areas of subdural haemorrhage. His coagulation profile, complete blood count and electrolytes were all normal. Ophthalmological examination demonstrated bilateral extensive and multilayered retinal haemorrhages. Three posterior rib fractures with callous formation were present on skeletal survey. A report was made to the local child welfare authority because of suspected physical abuse and nonaccidental head injury.