- Email: [email protected]
Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern
Journal of Pediatric Surgery xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern☆ Joanne Baerg a,⁎, Arul Thirumoorthi a, Rosemary Vannix a, Asma Taha b, Amy Young c, Alexander Zouros b a b c
Division of Pediatric Surgery, Loma Linda University Children's Hospital, Loma Linda, CA Division of Pediatric Neurosurgery, Loma Linda University Children's Hospital, Loma Linda, CA Division of Forensic Pediatrics, Loma Linda University Children's Hospital, Loma Linda, CA
a r t i c l e
i n f o
Article history: Received 3 January 2017 Accepted 23 January 2017 Available online xxxx Key words: Cervical spine injury Shaken infant Inflicted trauma Pattern of injury
a b s t r a c t Aim: The purpose of this study was to document the incidence and pattern of cervical spine (c-spine) injuries in children below 36 months with inflicted trauma. Methods: An IRB approved, prospective cohort study was performed between July 2011 and January 2016. Inclusion criteria were: age below 36 months, loss of consciousness after inflicted trauma, and one initial head computed tomography finding: a subdural, intraventricular, intraparenchymal, subarachnoid hemorrhage, diffuse axonal injury, hypoxic injury, or cerebral edema. A protocol of brain and neck magnetic resonance imaging and angiography was obtained within 48 h. Variables were compared by t-test and Fisher-exact test. Results: There were 53 children (median age: five months; range: 1–35 months), 38 males (71.7%), of which seven died (13.2%). C-spine injury was identified in 8 (15.1%): ligamentous injury (2), vertebral artery shear injury (1), atlantooccipital dissociation (AOD) (1), cord injury with cord epidural hematoma (2), and isolated cord epidural hematoma (2). Retinal hemorrhages (p = 0.02), shaking (p = 0.04), lower Glasgow coma score (GCS) (p = 0.01), brain infarcts (p = 0.01), and hypoxic/ischemic injury (p = 0.01) were associated with c-spine injury. One with AOD died. Six had significant disability. Conclusion: For small children with inflicted trauma, the c-spine injury incidence is 15.1%. The injury pattern includes retinal hemorrhages, shaking, lower GCS, and brain injury. Evaluation of shaken infants should include c-spine imaging. Level of evidence: Level 2 A- This is a prospective cohort study with complete follow-up to hospital discharge or death. In all cases, inflicted trauma was confirmed. Owing to the nature of child abuse, the precise time of injury is not known. All children underwent a strict imaging protocol on arrival to hospital that was supervised on a prospective basis. © 2017 Elsevier Inc. All rights reserved.
The pattern of injury in children below three years of age with inflicted trauma has been described to include head trauma, particularly subdural hematomas, skull fractures, bruises, long bone fractures and visceral injuries [1]. For infants that are shaken, retinal hemorrhages are considered pathognomonic [2,3]. A large retrospective study of infants with head trauma from various causes revealed a low incidence of cervical spine (c-spine) injuries. The few c-spine injuries that were identified, however, were from inflicted trauma [4]. Although some recommend an assessment of the spine in all young children with inflicted trauma, it may represent an unrecognized cause of spinal trauma [5]. Abbreviations: c-spine, Cervical spine; IRB, Institutional Review Board; AOD, atlantooccipital dissociation; GCS, Glasgow coma score. ☆ All authors participated in study conception and design, data acquisition, data analysis, drafting critical revisions and final approval of the manuscript. ⁎ Corresponding author at: Division of Pediatric Surgery, Rm 21111, Coleman Pavilion, 11175 Campus Street, Loma Linda University Children's Hospital, Loma Linda, CA 92354, USA. Tel.: +1 909 558 4619. E-mail address: [email protected] (J. Baerg).
Physicians are frequently unsure of how to evaluate the cervical spine in young victims of inflicted trauma and interpret any findings on imaging. Cervical spine magnetic resonance imaging (MRI) and angiography (MRA) are not a routine part of evaluation of young children with inflicted head trauma. However, there are increasing reports of spinal injuries found at autopsy or MRI in abused children [5,6]. The aim of this study was to document the incidence, the pattern and the outcomes of c-spine injuries in children younger than 36 months with confirmed inflicted head trauma. 1. Methods 1.1. Cohort This was an Institutional Review Board approved prospective cohort study (IRB 5100168). Data were collected between July 2011 and January 2016. Children younger than 36 months with confirmed
http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049 0022-3468/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
2
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
inflicted head trauma, admitted through the emergency department (ED) during the study period, were included. A social services and law enforcement investigation was launched in each case. For all included cases, either the perpetrator confessed or the inflicted abuse or shaking was witnessed. Cases that were indeterminate for inflicted trauma as the injury mechanism were excluded. We sought to collect prospective data and imaging within a similar time period after injury, therefore, children that were initially admitted to outside hospitals and transferred were excluded. Examination and documentation of injuries were performed by board certified pediatricians with a specialty in forensic pediatrics and child abuse. They identified no bleeding disorders. All included children underwent an ophthalmic exam by a board certified pediatric ophthalmologist. Retinal hemorrhages were documented if identified. All physical examinations were performed within four hours of arrival in the emergency department. 1.2. Imaging All included children had a documented loss of consciousness and a noncontrast head computed tomography scan (CT scan) performed on admission through the ED with one of the following findings: a subdural (SDH), intraventricular (IVH), intraparenchymal (IPH) or subarachnoid hemorrhage (SAH), diffuse axonal injury (DAI), hypoxic injury or cerebral edema [7]. One board certified pediatric neurosurgeon examined all included children, reviewed all images and supervised the institution of an MRI/MRA imaging protocol of the brain, c-spine and neck vessels within 48 h of admission. The MRIs of the brain included T1 magnetizationprepared rapid acquisition with gradient echo and T2 sequences physics and image appearance. T2 fluid attenuation inversion recovery, susceptibility weighted imaging, diffusion tensor imaging with apparent diffusion coefficient and color fractional anisotropy (FA) maps were performed [8–11]. MRIs of the c-spine included T1, T2 and Short tau inversion recovery images. MRAs of the neck without contrast included time of flight resonance angiograms of the cervical vasculature to the skull-base. Brain MRAs included the Circle of Willis and vertebrobasilar systems with maximum intensity projection and source imaging. Axial T1 with fat suppression, T2 diffusion tensor imaging and resultant average apparent diffusion coefficient, tensor trace and color FA maps were generated. Any abnormal finding of the brain, c-spine or neck vessels was recorded. Children with a vascular abnormality of the brain were excluded [8–11]. All imaging was reviewed by a board-certified neuroradiologist. A noncontrast head CT is the first line imaging modality for suspected inflicted head trauma [7]. Rarely, head CT may identify hypoxic brain injury within the first 24 h when a small number of children may demonstrate the “reversal sign,” in which there is reversal in the normal CT attenuation of gray matter and white matter [10]. Early MRI, however, provides a better estimation of shear injuries, hypoxic injury, ischemic insults and the timing of the lesions [9–11]. 1.3. Variables The findings on brain MRI at 48 h and at discharge, were categorized as infarction, hypoxic injury and ischemic stroke [12]. Brain infarction was defined as a localized area of necrosis of brain tissue. Hypoxic injury was defined as brain damage from inadequate brain oxygenation. Ischemic stroke was defined as cerebral ischemia conforming to an arterial vascular distribution such as the anterior cerebral (ACA), middle cerebral (MCA) or posterior cerebral arteries (PCA) or other major cerebral vascular arterial territory [12,13]. Additional data recorded were gender, Glasgow Coma Scale (GCS) [14], Injury Severity Score (ISS) [15], the lowest recorded systolic blood pressure and any use of blood transfusions in the ED, cardiopulmonary resuscitation (CPR) and the length of time it was administered.
Respiratory insufficiency or lack of airway protection requiring intubation and ventilation, either before arrival, or at presentation to the ED was documented. The presence of retinal hemorrhages, long bone fractures, skull fractures, rib fractures, pneumothoraces, visceral injuries, and bruising was recorded as well as the trauma mechanism of shaking. A skeletal survey was obtained in all cases and the findings were recorded. The Injury Severity Score was obtained from the initial standardized trauma history and physical exam form in our emergency room and calculated by the treating provider. The Injury Severity Score (ISS) is an established medical score to assess trauma severity. It correlates with mortality, morbidity and hospitalization time after trauma [15]. 1.4. Outcomes Outcomes recorded were the total days of mechanical ventilation, the length of hospital stay, and mortalities with time to death. Neurosurgical interventions such as external ventricular drain (EVD) or decompressive craniectomy for increased intracranial pressure or spinal fixation were recorded. For children that survived, discharge to a foster home or a skilled nursing facility (SNF) was recorded. At the time of discharge or death, each child was reviewed for the diagnosis of hypoxic brain injury or ischemic stroke after neurosurgical evaluation of the child, all MRI studies and neuroradiology consultation [10,12]. 1.5. Statistics Variables in children with a c-spine injury identified on MRI/MRA imaging protocol were compared to those without by two-tailed Fisher exact test for categorical and unpaired t-test for continuous variables. A P-value less than 0.05 was considered significant. Both median and range, and mean and standard deviation were reported for continuous variables. 2. Results During the study period, 85 children were initially evaluated, and 53 met the inclusion criteria. The median age for the cohort was five months (range: 1–35 months) (mean: 9.3 months ± 8.4), 38 were male (71.7%) and seven died (13.2%). The MRI/MRA imaging protocol performed within 48 h of presentation identified a c-spine injury in eight children for an incidence of 15.1%. The injuries were ligamentous injury (2), vertebral artery shear injury (1), atlantooccipital dissociation (AOD) (1), cord injury with cord epidural hematoma (2) and an isolated cord epidural hematoma (2). Table 1 compares the demographics and clinical presentation for children with and without a c-spine injury. The eight with a c-spine injury had a significantly lower initial GCS (p = 0.01), a higher incidence of retinal hemorrhages (p = 0.02) and shaking mechanism (p = 0.04). No other demographic or clinical variables were significant. Table 2 compares the findings on the initial noncontrast head CT scan performed in the ED for children with and without a c-spine injury. For the eight with c-spine injuries: Two had SDH with IPH, four had isolated SDH (of which three also had changes indicating hypoxic injury—“the reversal sign”) [10], one had an isolated IPH, and one had hypoxic injury. The MRI/MRA imaging protocol of the brain, identified brain infarcts in 7/8 (87.5%) with a c-spine injury and in 8/45 without a c-spine injury (17.8%), a significant finding (p = 0.01). The brain MRI/MRA imaging protocol, identified lesions not reported on the initial brain CT scan. For 45 children without a c-spine injury, it identified one additional IPH and one IVH. Two SAHs were not reported and may have resolved. For eight with a c-spine injury, the brain MRI/MRA imaging protocol identified three ischemic strokes in the distribution of the Circle of Willis, including one with a vertebral artery shear injury. It identified
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
3
Table 1 Demographics, clinical presentation and injuries in 53 children.
Gender (male)a Reported shaken baby Retinal hemorrhages Transfusion (ED) CPR performed Age (months)b,c GCS ISS Lowest systolic BP in ED CPR time (min)
a
No C-spine (n = 45) (%)
C-spine (n = 8) (%)
p
33 (73.3) 14 (31.1) 19 (42.2) 2 (4.4) 9 (20) 5.5 (1–35) 9.7 (8.5) 14 (3–15) 11 (4.8) 25 (9–48) 26.2 (10) 94 (38–137) 92.6 (21.1) 10 (5–120) 36.2 (56)
4 (50) 6 (75) 7 (87.5) 2 (25) 3 (37.5) 4.8 (1–23) 6.9 (7.3) 4.5 (3–15) 6.0 (4.2) 25 (16–45) 27.9 (10.1) 83 (69–104) 80 (17.6) 10 (5–22) 12.3 (8.7)
0.22 0.04 0.02 0.10 0.36 0.48 0.01 0.67 0.12 0.51
Other injuries at Presentationa
N (%)
N (%)
p
Long bone fracture Skull fractures Liver laceration Rib fractures Pancreatic transectiond Pneumothorax Diffuse bruising Multiple injuries
19 (42.2) 13 (28.9) 10 (22.2) 7 (15.6) 1 (2.2) 1 (2.2) 27 (60) 24 (53.3)
3 (37.5) 1 (12.5) 1 (12.5) 1 (12.5) 0 0 5 (62.5) 2 (25)
1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.25
a b c d
Fisher-exact test Unpaired t-test Median and range, mean and standard deviation Underwent laparotomy for pancreatic transection.
one additional SDH, one SAH and one had hypoxic brain injury. Two IPHs were not reported and may have resolved. In addition to brain infarcts, the brain MRI/MRA protocol identified hypoxic injury (p-0.01), and ischemic stroke (p = 0.01) as significant findings in eight children with a c-spine injury when compared to those without an injury. Table 3 illustrates the comparison of outcome variables for the cohort. The one mortality with a c-spine injury had an atlantooccipital dissociation (AOD), underwent stabilization, and died at six days after admission. The child had global hypoxic ischemic encephalopathy and was in a neurovegetative state. The parents elected for withdrawal of care (Fig. 1, Case 7). Table 4 illustrates the clinical, radiologic and outcome details for the eight children with c-spine injuries from inflicted trauma. Six of seven survivors (85.7%) with inflicted head trauma and c-spine injuries on MRI had hypoxic brain injury or an ischemic stroke. In one, the hypoxic
injury had progressed to cortical atrophy. All six had neurological devastation with nonambulatory spastic quadriparesis and gastrostomy tube dependence for feeding. Fig. 2 illustrates the c-spine injury MRI/MRA images and corresponding brain images. 3. Discussion After institution of an MRI imaging protocol of the brain and spinal cord for children younger than 36 months with inflicted head trauma, the incidence of c-spine injury is 15.1%. In 2014, a retrospective review of MRI detection of c-spine injury after inflicted trauma reported an incidence of 36% in children under 36 months [16]. The majority with c-spine injury had hypoxic–ischemic brain injury. A similar trauma database review of 57 children under 3-years of age, reports a 19% incidence of c-spine injuries after inflicted trauma [17]. Ligamentous injuries were common in both series.
Table 2 Brain CT scan findings in ED and brain MRI/MRA before 48 h after admission. Brain CT Scan
SDH IPH Cerebral edemac IVH SAH DAI Infarctsd Hypoxia Stroke Multiple findings a b c d
Brain MRI/MRA No C-Spine Injury N = 45 (%)
C-Spine Injury N = 8 (%)
p
No C-Spine Injury N = 45 (%)
C-spine Injury N = 8 (%)
p
28 (62.2)a 8 (17.8)c 3 (6.7) 2 (4.4) 5 (11.1) 2 (4.4) 0 3 (6.7) 0 8 (17.8)
6 (75)b 3 (37.5) 1 (12.5) 0 0 0 0 4 (50.0) 0 6 (75)
0.70 0.34 0.49 1.0 1.0 1.0 NA 0.01 NA 0.01
28 (62.2) 9 (20.0) 3 (6.7) 3 (6.7) 3 (6.7) 2 (2.2) 8 (17.8) 3 (6.7) 0 9 (20)
7 (87.5) 1 (12.5) 1 (12.5) 0 1 (12.5) 0 7 (87.5) 5 (62.5) 3 (37.5) 6 (75)
0.24 0.34 0.49 1.0 0.49 1.0 b0.01 0.01 b0.01 0.01
One had bilateral SDH, 1 had an acute on chronic SDH, 3 had midline shift. Two had bilateral SDH. All had SDH. The incidence of retinal hemorrhages in combination with brain infarction was 6/7 (85.7%) in those with a c-spine injury and 8/8 (100%) for those without a c-spine injury.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
4
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
Table 3 Outcome data.
Need for Ventilationa Neurosurgical interventions Medical foster home Skilled nursing facility Stroke or hypoxic brain damaged Mortalities Ventilator (days)b,c Hospital stay (days) Time to death (days) a b c d
No C-Spine injury N = 45 (%)
C-Spine injury N = 8 (%)
p
22 (48.9) 12 (26.7) 34 (75.6) 5 (11.1) 12 (26.7) 6 (13.3) 1 (0–13) 3.0 (4.0) 5 (1–57) 9.4 (11.5) 5 (1–11) 5.2 (4.0)
6 (75) 5 (62.5) 5 (62.5) 2 (25.0) 6 (75.0) 1 (12.5) 5 (0–10) 6.1 (3.5) 13 (6–52) 13 (13.7) 6
0.11 0.09 0.42 0.28 0.01 1.0 0.14 0.12 NA
Fisher-exact test, Unpaired t-test, Median and range, mean and standard deviation Status of brain damage at hospital discharge or death.
When c-spine injury is present, the pattern of injury includes a significantly higher incidence of shaking mechanism, retinal hemorrhages, a lower initial GCS, and areas of brain infarction on MRI at 48 h. In addition to infarction, the MRI reveals a significantly higher incidence of hypoxic injury and ischemic stroke. Evaluation of the outcome data reveals one child with an AOD died and most survivors suffer neurological devastation. In 1946, Caffey recognized a pattern of multiple fractures in the long bones of infants with chronic subdural hematomas and no history of trauma. He recommended that identification of long bone fractures in infants warrants investigation for subdural hematoma but never concluded that the injuries were directly inflicted [18]. Child abuse received little recognition until 1962 when Kempe published, “The battered-child syndrome” [1]. In 1971, Guthkelch reported subdural hematoma as a feature of this syndrome and postulated that the brain injury was caused by shaking [19]. In 1972, Caffey published
“The whiplash shaken infant syndrome” and linked intracranial and retinal hemorrhages with permanent neuroimpairment [20]. By 2002, it was well-accepted that the sequence of events in shaken baby syndrome was initiated by violent whiplash shaking [8]. Cervical hyperextension causes stretch injury to the cervical cord, brainstem or vasculature and leads to respiratory insufficiency [21]. The ensuing hypoxia and shock cause hypoxic, ischemic brain injury. Subdural and retinal hemorrhages are important markers of shaking [1,8,20,21]. Our series of young children with inflicted trauma and shaking confirms the pattern of injury documented in these previous reports. In 2003, a 10-year retrospective review of the Canadian experience with “Shaken baby syndrome” confirms a high incidence of neurological devastation but makes no mention of c-spine injuries [22]. In contrast, our series and recent supporting literature, demonstrate that use of diffusion weighted MRI/MRA imaging protocol of the brain and spine expands the shaking injury pattern in young children to include c-spine injury [11,23]. The most common cause of missed c-spine injuries is insufficient imaging [24]. One atlantooccipital dissociation (AOD) was identified. Although c-spine injury in young children frequently involves the atlantoaxial region, AOD is rare [25]. Extensive literature review suggests that this may be the only reported pediatric AOD from inflicted trauma identified before death. There is one report of an adult female presenting with fatal AOD from homicide [26]. Most pediatric cases are diagnosed in the setting of high speed motor vehicle accidents. Plain radiographs may appear normal and AOD may be overlooked unless specifically sought [27]. One vertebral artery shear injury with disruption of the vertebral artery on MRA and a corresponding ischemic stroke was found. There are one reported case of periadventitial vertebral artery hemorrhages between C1 and C4 in a shaken infant and rare cases of vertebral artery focal stenosis with distal intimal flap and dissection [28]. Vertebral artery injuries may be underreported unless MRA is performed. Three traumatic cervical spine epidural hematomas were identified in this series. Traumatic spinal epidural hematoma is rare in children.
Fig. 1. Corresponding MRI/MRA of cervical spine and brain in one child with AOD.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
5
Table 4 Eight children with a C-spine injury identified on MRI/MRA protocol.
1 2
3
4 5 6 7 8
a b
MRIa,b C-spine Injury
MRI Brain 48 h
Age (mo)
GCS
ISS
Vent time (days)
Neurosurg interventions
Brain status at discharge
Discharge status
C1/C2 Ligament Ligament
SDH, IPH
4
9
34
5
EVD
Global Hypoxia
12
3
25
9
Hemi–crani to decompress
Ischemic Stroke
2
3
25
9
Cortical atrophy
SNF
6
6
16
0
SNF
23
15
45
0
Hypoxic damage and ischemic stroke Mild brain damage
Foster home
C4 Epidural hematoma AOD
SDH L MCA stroke SDH Circle of Willis stroke SDH (bilateral) SDH (bilateral) SDH, IPH, SAH IPH
Foster Home Foster Home
1
4
20
10
EVD
Hypoxic damage
Foster home
3
3
38
5
AOD stabilization
C5 Epidural hematoma Cord compressed
SDH PCA stroke
4
5
20
5
EVD
Global hypoxic damage Bilateral Watershed area strokes
Died Withdrew care Foster home
Vertebral artery shear C5/6 Epidural hematoma Cord signal
All except Case 5 had brain infarcts at 48 h, all except Case 7 had retinal hemorrhages, for cases 4 & 8, the epidural hematomas extended to the thoracic spine Cases 3, 6 and 7 underwent CPR before arrival to ED and Cases 1, 4 and 6–8 had evidence of hypoxic brain injury on MRI of the brain at 48 h.
In 2013, only eight were documented in the pediatric literature [29]. Spinal subdural hematomas are reported after inflicted injury in children but epidural hematomas are rare and may imply a more violent injury [30]. Without MRI examination, children with traumatic spinal epidural hematoma and minor symptoms may be missed. One limitation of this study is that we were unable to obtain autopsy reports for all of the mortalities in our series. It is unknown whether autopsy may have provided additional details of c-spine injuries in the mortalities. Furthermore, autopsy may better delineate the location of spinal hematomas (epidural vs. subdural), as well as microscopic injuries (nerve root avulsion and stretch injuries). This limitation is counterbalanced by complete prospective data collection and a strict
MRI protocol for examination of the brain and c-spine in all included children with admitted or witnessed trauma. We recommend diffusion weighted MRI imaging of the c-spine for shaken infants. If shaking is not confirmed, but children have suspicious physical findings such as retinal hemorrhages, c-spine MRI is recommended to better document the pattern of injury. After consultation with neuroradiology and neurosurgery, cervical collar protection, surgical fixation of unstable injuries or anticoagulation for a vessel intimal flap dissection may be recommended. The outcome of neurologic devastation combined with c-spine injury, supports the hypothesis that secondary hypoxic brain damage occurs with respiratory insufficiency when the neuroaxis is stretched. With early diagnosis, prevention of
Fig. 2. Corresponding MRI/MRA of cervical spine and brain in four children with inflicted trauma and a cervical spine injury.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
6
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
secondary neurologic injuries and specific neuroprotective strategies may be implemented. However, obtaining early MRI imaging to diagnose these injuries must balance against the risks of transport and critical-care monitoring in the MRI scanner of the acute trauma patient. Detailed documentation of a violent pattern of injury has forensic implications for future care of the child and law enforcement. The understanding of the pattern of injury in shaken infants and child abuse has evolved since Caffey's first report in 1946, followed by the publication of “The whiplash shaken infant syndrome” in 1972. By 2002, the features of brain injury on diffusion-weighted MRI in shaken infants were reported, but the cervical spine was not evaluated [31]. Through institution of an MRI protocol to evaluate the c-spine of all young victims of inflicted trauma, our series illustrates an expansion of the injury pattern of shaken infants to include c-spine injuries. This study demonstrates a higher incidence of cervical spinal and vascular injury than previously appreciated and merits further evaluation of the differences in biomechanics of inflicted versus accidental trauma to the cervical vasculature and spine. A larger case series is needed to determine the outcomes and treatments specific to these injuries in this young age group. Acknowledgment The authors acknowledge Rajaie Hazboun MD, for collection and review of imaging and figures. References [1] Kempe CH, Silverman FN, Steele BF, et al. The battered-child syndrome. JAMA 1962; 181:17–24. [2] Dashti SR, Decker DD, Razzaq A, et al. Current patterns of inflicted head injury in children. Pediatr Neurosurg 1999;31:302–6. [3] Maguire SA, Watts PO, Shaw AD, et al. Retinal haemorrhages and related findings in abusive and non-abusive head trauma: a systematic review. Eye (Lond) 2013;27:28–36. [4] Katz JS, Oluigbo CO, Wilkinson CC, et al. Prevalence of cervical spine injury in infants with head trauma. J Neurosurg Pediatr 2010;5:470–3. [5] Ghatan S, Ellenbogen RG. Pediatric spine and spinal cord injury after inflicted trauma. Neurosurg Clin N Am 2002;13:227–33. [6] Kemp A, Cowley L, Maguire S. Spinal injuries in abusive head trauma: patterns and recommendations. Pediatr Radiol 2014;44(Suppl. 4):S604–12. [7] Kemp AM, Jaspan T, Griffiths J, et al. Neuroimaging: what neuroradiological features distinguish abusive from non-abusive head trauma? A systematic review. Arch Dis Child 2011;96:1103–12. [8] Blumenthal I. Shaken baby syndrome. Postgrad Med J 2002;78:732–5.
[9] Vazquez E, Delgado I, Sanchez-Montanez A, et al. Imaging abusive head trauma: why use both computed tomography and magnetic resonance imaging? Pediatr Radiol 2014;44(Suppl. 4):S589–603. [10] Huang BH, Castillo M. Hypoxic–ishemic brain injury: findings from birth to adulthood. Radiographics 2008;28:417–39. [11] Bradford R, Choudhary AK, Dias MS. Serial neuroimaging in infants with abusive head trauma: timing abusive injuries. J Neurosurg Pediatr 2013;12:110–9. [12] Jordan LC, Hillis AE. Challenges in the diagnosis of pediatric stroke. Nat Rev Neurol 2011;7:199–208. [13] Kadom N, Zarir K, Vezina G, et al. Usefulness of MRI detection of cervical spine and brain injuries in the evaluation of abusive head trauma. Pediatr Radiol 2014;44:839–48. [14] Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974;13:81–4. [15] Copes WS, Champion HR, Sacco WJ, et al. The injury severity score revisited. J Trauma 1988;28:69–77. [16] Flom L, Fromkin J, Panigrahy A, et al. Development of a screening MRI for infants at risk for abusive head trauma. Pediatr Radiol 2016;46:519–26. [17] Knox JB, Schneider JE, Cage JM, et al. Spine trauma in very young children: a retrospective study of 206 patients presenting to a level 1 pediatric trauma center. J Pediatr Orthop 2014;34:698–702. [18] Caffey J. Multiple fractures in the long bones of infants suffering from chronic subdural hematoma. Radiology 1946;194:163–73. [19] Guthkelch AN. Infantile subdural haematoma and its relationship to whiplash injuries. Br Med J 1971;22(2(5759)):430–1. [20] Caffey J. The whiplash shaken infant syndrome: manual shaking by the extremities with whiplash-induced intracranial and intraocular bleedings, linked with residual permanent brain damage and mental retardation. Pediatrics 1974; 54:396–403. [21] McKinney AM, Thompson LR, Truwit CL, et al. Unilateral hypoxic–ischemic injury in young children from abusive head trauma, lacking craniocervical dissection or cord injury. Pediatr Radiol 2008;38:164–74. [22] King WJ, MacKay M, Sirnick A, et al. Shaken baby syndrome in Canada: clinical characteristics and outcomes of hospital cases. CMAJ 2003;168:155–9. [23] Jacob R, Cox M, Koral K, et al. MR imaging of the cervical spine in nonaccidental trauma: a tertiary institution experience. Am J Neuroradiol 2016;26. [24] Brown RL, Brunn MA, Garcia VF. Cervical spine injuries in children: a review of 103 patients treated consecutively at a level 1 trauma center. J Pediatr Surg 2001;36: 1107–14. [25] Knox J, Schneider J, Wimberly RL, et al. Characteristics of spinal injuries secondary to nonaccidental trauma. J Pediatr Orthop 2014;34:376–81. [26] Raul JS, Tracqui A, Ludes B. Homicide by occipitoatlantal dislocation. J Forensic Sci 2004;49:601–3. [27] Astur N, Klimo Jr P, Sawyer JR, et al. Traumatic atlanto-occipital dislocation in children: evaluation, treatment and outcomes. J Bone Joint Surg Am 2013; 18(95):e194(1–8). [28] Gleckman AM, Kessler SC, Smith TW. Periadventitial extracranial vertebral artery hemorrhage in a case of shaken baby syndrome. J Forensic Sci 2000;45: 1151–3. [29] Rangarajan V, Mavani SB, Nadkarni TD, et al. Traumatic cervical epidural hematoma in an infant. J Craniovertebr Junction Spine 2013;4:37–9. [30] Koumellis P, McConachie NS, Jaspan T. Spinal subdural haematomas in children with non-accidental head injury. Arch Dis Child 2009;94:216–9. [31] Biousse V, Suh DY, Newman NJ, et al. Diffusion-weighted magnetic resonance imaging in shaken baby syndrome. Am J Ophthalmol 2002;133:249–55.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern☆ Joanne Baerg a,⁎, Arul Thirumoorthi a, Rosemary Vannix a, Asma Taha b, Amy Young c, Alexander Zouros b a b c
Division of Pediatric Surgery, Loma Linda University Children's Hospital, Loma Linda, CA Division of Pediatric Neurosurgery, Loma Linda University Children's Hospital, Loma Linda, CA Division of Forensic Pediatrics, Loma Linda University Children's Hospital, Loma Linda, CA
a r t i c l e
i n f o
Article history: Received 3 January 2017 Accepted 23 January 2017 Available online xxxx Key words: Cervical spine injury Shaken infant Inflicted trauma Pattern of injury
a b s t r a c t Aim: The purpose of this study was to document the incidence and pattern of cervical spine (c-spine) injuries in children below 36 months with inflicted trauma. Methods: An IRB approved, prospective cohort study was performed between July 2011 and January 2016. Inclusion criteria were: age below 36 months, loss of consciousness after inflicted trauma, and one initial head computed tomography finding: a subdural, intraventricular, intraparenchymal, subarachnoid hemorrhage, diffuse axonal injury, hypoxic injury, or cerebral edema. A protocol of brain and neck magnetic resonance imaging and angiography was obtained within 48 h. Variables were compared by t-test and Fisher-exact test. Results: There were 53 children (median age: five months; range: 1–35 months), 38 males (71.7%), of which seven died (13.2%). C-spine injury was identified in 8 (15.1%): ligamentous injury (2), vertebral artery shear injury (1), atlantooccipital dissociation (AOD) (1), cord injury with cord epidural hematoma (2), and isolated cord epidural hematoma (2). Retinal hemorrhages (p = 0.02), shaking (p = 0.04), lower Glasgow coma score (GCS) (p = 0.01), brain infarcts (p = 0.01), and hypoxic/ischemic injury (p = 0.01) were associated with c-spine injury. One with AOD died. Six had significant disability. Conclusion: For small children with inflicted trauma, the c-spine injury incidence is 15.1%. The injury pattern includes retinal hemorrhages, shaking, lower GCS, and brain injury. Evaluation of shaken infants should include c-spine imaging. Level of evidence: Level 2 A- This is a prospective cohort study with complete follow-up to hospital discharge or death. In all cases, inflicted trauma was confirmed. Owing to the nature of child abuse, the precise time of injury is not known. All children underwent a strict imaging protocol on arrival to hospital that was supervised on a prospective basis. © 2017 Elsevier Inc. All rights reserved.
The pattern of injury in children below three years of age with inflicted trauma has been described to include head trauma, particularly subdural hematomas, skull fractures, bruises, long bone fractures and visceral injuries [1]. For infants that are shaken, retinal hemorrhages are considered pathognomonic [2,3]. A large retrospective study of infants with head trauma from various causes revealed a low incidence of cervical spine (c-spine) injuries. The few c-spine injuries that were identified, however, were from inflicted trauma [4]. Although some recommend an assessment of the spine in all young children with inflicted trauma, it may represent an unrecognized cause of spinal trauma [5]. Abbreviations: c-spine, Cervical spine; IRB, Institutional Review Board; AOD, atlantooccipital dissociation; GCS, Glasgow coma score. ☆ All authors participated in study conception and design, data acquisition, data analysis, drafting critical revisions and final approval of the manuscript. ⁎ Corresponding author at: Division of Pediatric Surgery, Rm 21111, Coleman Pavilion, 11175 Campus Street, Loma Linda University Children's Hospital, Loma Linda, CA 92354, USA. Tel.: +1 909 558 4619. E-mail address: [email protected] (J. Baerg).
Physicians are frequently unsure of how to evaluate the cervical spine in young victims of inflicted trauma and interpret any findings on imaging. Cervical spine magnetic resonance imaging (MRI) and angiography (MRA) are not a routine part of evaluation of young children with inflicted head trauma. However, there are increasing reports of spinal injuries found at autopsy or MRI in abused children [5,6]. The aim of this study was to document the incidence, the pattern and the outcomes of c-spine injuries in children younger than 36 months with confirmed inflicted head trauma. 1. Methods 1.1. Cohort This was an Institutional Review Board approved prospective cohort study (IRB 5100168). Data were collected between July 2011 and January 2016. Children younger than 36 months with confirmed
http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049 0022-3468/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
2
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
inflicted head trauma, admitted through the emergency department (ED) during the study period, were included. A social services and law enforcement investigation was launched in each case. For all included cases, either the perpetrator confessed or the inflicted abuse or shaking was witnessed. Cases that were indeterminate for inflicted trauma as the injury mechanism were excluded. We sought to collect prospective data and imaging within a similar time period after injury, therefore, children that were initially admitted to outside hospitals and transferred were excluded. Examination and documentation of injuries were performed by board certified pediatricians with a specialty in forensic pediatrics and child abuse. They identified no bleeding disorders. All included children underwent an ophthalmic exam by a board certified pediatric ophthalmologist. Retinal hemorrhages were documented if identified. All physical examinations were performed within four hours of arrival in the emergency department. 1.2. Imaging All included children had a documented loss of consciousness and a noncontrast head computed tomography scan (CT scan) performed on admission through the ED with one of the following findings: a subdural (SDH), intraventricular (IVH), intraparenchymal (IPH) or subarachnoid hemorrhage (SAH), diffuse axonal injury (DAI), hypoxic injury or cerebral edema [7]. One board certified pediatric neurosurgeon examined all included children, reviewed all images and supervised the institution of an MRI/MRA imaging protocol of the brain, c-spine and neck vessels within 48 h of admission. The MRIs of the brain included T1 magnetizationprepared rapid acquisition with gradient echo and T2 sequences physics and image appearance. T2 fluid attenuation inversion recovery, susceptibility weighted imaging, diffusion tensor imaging with apparent diffusion coefficient and color fractional anisotropy (FA) maps were performed [8–11]. MRIs of the c-spine included T1, T2 and Short tau inversion recovery images. MRAs of the neck without contrast included time of flight resonance angiograms of the cervical vasculature to the skull-base. Brain MRAs included the Circle of Willis and vertebrobasilar systems with maximum intensity projection and source imaging. Axial T1 with fat suppression, T2 diffusion tensor imaging and resultant average apparent diffusion coefficient, tensor trace and color FA maps were generated. Any abnormal finding of the brain, c-spine or neck vessels was recorded. Children with a vascular abnormality of the brain were excluded [8–11]. All imaging was reviewed by a board-certified neuroradiologist. A noncontrast head CT is the first line imaging modality for suspected inflicted head trauma [7]. Rarely, head CT may identify hypoxic brain injury within the first 24 h when a small number of children may demonstrate the “reversal sign,” in which there is reversal in the normal CT attenuation of gray matter and white matter [10]. Early MRI, however, provides a better estimation of shear injuries, hypoxic injury, ischemic insults and the timing of the lesions [9–11]. 1.3. Variables The findings on brain MRI at 48 h and at discharge, were categorized as infarction, hypoxic injury and ischemic stroke [12]. Brain infarction was defined as a localized area of necrosis of brain tissue. Hypoxic injury was defined as brain damage from inadequate brain oxygenation. Ischemic stroke was defined as cerebral ischemia conforming to an arterial vascular distribution such as the anterior cerebral (ACA), middle cerebral (MCA) or posterior cerebral arteries (PCA) or other major cerebral vascular arterial territory [12,13]. Additional data recorded were gender, Glasgow Coma Scale (GCS) [14], Injury Severity Score (ISS) [15], the lowest recorded systolic blood pressure and any use of blood transfusions in the ED, cardiopulmonary resuscitation (CPR) and the length of time it was administered.
Respiratory insufficiency or lack of airway protection requiring intubation and ventilation, either before arrival, or at presentation to the ED was documented. The presence of retinal hemorrhages, long bone fractures, skull fractures, rib fractures, pneumothoraces, visceral injuries, and bruising was recorded as well as the trauma mechanism of shaking. A skeletal survey was obtained in all cases and the findings were recorded. The Injury Severity Score was obtained from the initial standardized trauma history and physical exam form in our emergency room and calculated by the treating provider. The Injury Severity Score (ISS) is an established medical score to assess trauma severity. It correlates with mortality, morbidity and hospitalization time after trauma [15]. 1.4. Outcomes Outcomes recorded were the total days of mechanical ventilation, the length of hospital stay, and mortalities with time to death. Neurosurgical interventions such as external ventricular drain (EVD) or decompressive craniectomy for increased intracranial pressure or spinal fixation were recorded. For children that survived, discharge to a foster home or a skilled nursing facility (SNF) was recorded. At the time of discharge or death, each child was reviewed for the diagnosis of hypoxic brain injury or ischemic stroke after neurosurgical evaluation of the child, all MRI studies and neuroradiology consultation [10,12]. 1.5. Statistics Variables in children with a c-spine injury identified on MRI/MRA imaging protocol were compared to those without by two-tailed Fisher exact test for categorical and unpaired t-test for continuous variables. A P-value less than 0.05 was considered significant. Both median and range, and mean and standard deviation were reported for continuous variables. 2. Results During the study period, 85 children were initially evaluated, and 53 met the inclusion criteria. The median age for the cohort was five months (range: 1–35 months) (mean: 9.3 months ± 8.4), 38 were male (71.7%) and seven died (13.2%). The MRI/MRA imaging protocol performed within 48 h of presentation identified a c-spine injury in eight children for an incidence of 15.1%. The injuries were ligamentous injury (2), vertebral artery shear injury (1), atlantooccipital dissociation (AOD) (1), cord injury with cord epidural hematoma (2) and an isolated cord epidural hematoma (2). Table 1 compares the demographics and clinical presentation for children with and without a c-spine injury. The eight with a c-spine injury had a significantly lower initial GCS (p = 0.01), a higher incidence of retinal hemorrhages (p = 0.02) and shaking mechanism (p = 0.04). No other demographic or clinical variables were significant. Table 2 compares the findings on the initial noncontrast head CT scan performed in the ED for children with and without a c-spine injury. For the eight with c-spine injuries: Two had SDH with IPH, four had isolated SDH (of which three also had changes indicating hypoxic injury—“the reversal sign”) [10], one had an isolated IPH, and one had hypoxic injury. The MRI/MRA imaging protocol of the brain, identified brain infarcts in 7/8 (87.5%) with a c-spine injury and in 8/45 without a c-spine injury (17.8%), a significant finding (p = 0.01). The brain MRI/MRA imaging protocol, identified lesions not reported on the initial brain CT scan. For 45 children without a c-spine injury, it identified one additional IPH and one IVH. Two SAHs were not reported and may have resolved. For eight with a c-spine injury, the brain MRI/MRA imaging protocol identified three ischemic strokes in the distribution of the Circle of Willis, including one with a vertebral artery shear injury. It identified
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
3
Table 1 Demographics, clinical presentation and injuries in 53 children.
Gender (male)a Reported shaken baby Retinal hemorrhages Transfusion (ED) CPR performed Age (months)b,c GCS ISS Lowest systolic BP in ED CPR time (min)
a
No C-spine (n = 45) (%)
C-spine (n = 8) (%)
p
33 (73.3) 14 (31.1) 19 (42.2) 2 (4.4) 9 (20) 5.5 (1–35) 9.7 (8.5) 14 (3–15) 11 (4.8) 25 (9–48) 26.2 (10) 94 (38–137) 92.6 (21.1) 10 (5–120) 36.2 (56)
4 (50) 6 (75) 7 (87.5) 2 (25) 3 (37.5) 4.8 (1–23) 6.9 (7.3) 4.5 (3–15) 6.0 (4.2) 25 (16–45) 27.9 (10.1) 83 (69–104) 80 (17.6) 10 (5–22) 12.3 (8.7)
0.22 0.04 0.02 0.10 0.36 0.48 0.01 0.67 0.12 0.51
Other injuries at Presentationa
N (%)
N (%)
p
Long bone fracture Skull fractures Liver laceration Rib fractures Pancreatic transectiond Pneumothorax Diffuse bruising Multiple injuries
19 (42.2) 13 (28.9) 10 (22.2) 7 (15.6) 1 (2.2) 1 (2.2) 27 (60) 24 (53.3)
3 (37.5) 1 (12.5) 1 (12.5) 1 (12.5) 0 0 5 (62.5) 2 (25)
1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.25
a b c d
Fisher-exact test Unpaired t-test Median and range, mean and standard deviation Underwent laparotomy for pancreatic transection.
one additional SDH, one SAH and one had hypoxic brain injury. Two IPHs were not reported and may have resolved. In addition to brain infarcts, the brain MRI/MRA protocol identified hypoxic injury (p-0.01), and ischemic stroke (p = 0.01) as significant findings in eight children with a c-spine injury when compared to those without an injury. Table 3 illustrates the comparison of outcome variables for the cohort. The one mortality with a c-spine injury had an atlantooccipital dissociation (AOD), underwent stabilization, and died at six days after admission. The child had global hypoxic ischemic encephalopathy and was in a neurovegetative state. The parents elected for withdrawal of care (Fig. 1, Case 7). Table 4 illustrates the clinical, radiologic and outcome details for the eight children with c-spine injuries from inflicted trauma. Six of seven survivors (85.7%) with inflicted head trauma and c-spine injuries on MRI had hypoxic brain injury or an ischemic stroke. In one, the hypoxic
injury had progressed to cortical atrophy. All six had neurological devastation with nonambulatory spastic quadriparesis and gastrostomy tube dependence for feeding. Fig. 2 illustrates the c-spine injury MRI/MRA images and corresponding brain images. 3. Discussion After institution of an MRI imaging protocol of the brain and spinal cord for children younger than 36 months with inflicted head trauma, the incidence of c-spine injury is 15.1%. In 2014, a retrospective review of MRI detection of c-spine injury after inflicted trauma reported an incidence of 36% in children under 36 months [16]. The majority with c-spine injury had hypoxic–ischemic brain injury. A similar trauma database review of 57 children under 3-years of age, reports a 19% incidence of c-spine injuries after inflicted trauma [17]. Ligamentous injuries were common in both series.
Table 2 Brain CT scan findings in ED and brain MRI/MRA before 48 h after admission. Brain CT Scan
SDH IPH Cerebral edemac IVH SAH DAI Infarctsd Hypoxia Stroke Multiple findings a b c d
Brain MRI/MRA No C-Spine Injury N = 45 (%)
C-Spine Injury N = 8 (%)
p
No C-Spine Injury N = 45 (%)
C-spine Injury N = 8 (%)
p
28 (62.2)a 8 (17.8)c 3 (6.7) 2 (4.4) 5 (11.1) 2 (4.4) 0 3 (6.7) 0 8 (17.8)
6 (75)b 3 (37.5) 1 (12.5) 0 0 0 0 4 (50.0) 0 6 (75)
0.70 0.34 0.49 1.0 1.0 1.0 NA 0.01 NA 0.01
28 (62.2) 9 (20.0) 3 (6.7) 3 (6.7) 3 (6.7) 2 (2.2) 8 (17.8) 3 (6.7) 0 9 (20)
7 (87.5) 1 (12.5) 1 (12.5) 0 1 (12.5) 0 7 (87.5) 5 (62.5) 3 (37.5) 6 (75)
0.24 0.34 0.49 1.0 0.49 1.0 b0.01 0.01 b0.01 0.01
One had bilateral SDH, 1 had an acute on chronic SDH, 3 had midline shift. Two had bilateral SDH. All had SDH. The incidence of retinal hemorrhages in combination with brain infarction was 6/7 (85.7%) in those with a c-spine injury and 8/8 (100%) for those without a c-spine injury.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
4
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
Table 3 Outcome data.
Need for Ventilationa Neurosurgical interventions Medical foster home Skilled nursing facility Stroke or hypoxic brain damaged Mortalities Ventilator (days)b,c Hospital stay (days) Time to death (days) a b c d
No C-Spine injury N = 45 (%)
C-Spine injury N = 8 (%)
p
22 (48.9) 12 (26.7) 34 (75.6) 5 (11.1) 12 (26.7) 6 (13.3) 1 (0–13) 3.0 (4.0) 5 (1–57) 9.4 (11.5) 5 (1–11) 5.2 (4.0)
6 (75) 5 (62.5) 5 (62.5) 2 (25.0) 6 (75.0) 1 (12.5) 5 (0–10) 6.1 (3.5) 13 (6–52) 13 (13.7) 6
0.11 0.09 0.42 0.28 0.01 1.0 0.14 0.12 NA
Fisher-exact test, Unpaired t-test, Median and range, mean and standard deviation Status of brain damage at hospital discharge or death.
When c-spine injury is present, the pattern of injury includes a significantly higher incidence of shaking mechanism, retinal hemorrhages, a lower initial GCS, and areas of brain infarction on MRI at 48 h. In addition to infarction, the MRI reveals a significantly higher incidence of hypoxic injury and ischemic stroke. Evaluation of the outcome data reveals one child with an AOD died and most survivors suffer neurological devastation. In 1946, Caffey recognized a pattern of multiple fractures in the long bones of infants with chronic subdural hematomas and no history of trauma. He recommended that identification of long bone fractures in infants warrants investigation for subdural hematoma but never concluded that the injuries were directly inflicted [18]. Child abuse received little recognition until 1962 when Kempe published, “The battered-child syndrome” [1]. In 1971, Guthkelch reported subdural hematoma as a feature of this syndrome and postulated that the brain injury was caused by shaking [19]. In 1972, Caffey published
“The whiplash shaken infant syndrome” and linked intracranial and retinal hemorrhages with permanent neuroimpairment [20]. By 2002, it was well-accepted that the sequence of events in shaken baby syndrome was initiated by violent whiplash shaking [8]. Cervical hyperextension causes stretch injury to the cervical cord, brainstem or vasculature and leads to respiratory insufficiency [21]. The ensuing hypoxia and shock cause hypoxic, ischemic brain injury. Subdural and retinal hemorrhages are important markers of shaking [1,8,20,21]. Our series of young children with inflicted trauma and shaking confirms the pattern of injury documented in these previous reports. In 2003, a 10-year retrospective review of the Canadian experience with “Shaken baby syndrome” confirms a high incidence of neurological devastation but makes no mention of c-spine injuries [22]. In contrast, our series and recent supporting literature, demonstrate that use of diffusion weighted MRI/MRA imaging protocol of the brain and spine expands the shaking injury pattern in young children to include c-spine injury [11,23]. The most common cause of missed c-spine injuries is insufficient imaging [24]. One atlantooccipital dissociation (AOD) was identified. Although c-spine injury in young children frequently involves the atlantoaxial region, AOD is rare [25]. Extensive literature review suggests that this may be the only reported pediatric AOD from inflicted trauma identified before death. There is one report of an adult female presenting with fatal AOD from homicide [26]. Most pediatric cases are diagnosed in the setting of high speed motor vehicle accidents. Plain radiographs may appear normal and AOD may be overlooked unless specifically sought [27]. One vertebral artery shear injury with disruption of the vertebral artery on MRA and a corresponding ischemic stroke was found. There are one reported case of periadventitial vertebral artery hemorrhages between C1 and C4 in a shaken infant and rare cases of vertebral artery focal stenosis with distal intimal flap and dissection [28]. Vertebral artery injuries may be underreported unless MRA is performed. Three traumatic cervical spine epidural hematomas were identified in this series. Traumatic spinal epidural hematoma is rare in children.
Fig. 1. Corresponding MRI/MRA of cervical spine and brain in one child with AOD.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
5
Table 4 Eight children with a C-spine injury identified on MRI/MRA protocol.
1 2
3
4 5 6 7 8
a b
MRIa,b C-spine Injury
MRI Brain 48 h
Age (mo)
GCS
ISS
Vent time (days)
Neurosurg interventions
Brain status at discharge
Discharge status
C1/C2 Ligament Ligament
SDH, IPH
4
9
34
5
EVD
Global Hypoxia
12
3
25
9
Hemi–crani to decompress
Ischemic Stroke
2
3
25
9
Cortical atrophy
SNF
6
6
16
0
SNF
23
15
45
0
Hypoxic damage and ischemic stroke Mild brain damage
Foster home
C4 Epidural hematoma AOD
SDH L MCA stroke SDH Circle of Willis stroke SDH (bilateral) SDH (bilateral) SDH, IPH, SAH IPH
Foster Home Foster Home
1
4
20
10
EVD
Hypoxic damage
Foster home
3
3
38
5
AOD stabilization
C5 Epidural hematoma Cord compressed
SDH PCA stroke
4
5
20
5
EVD
Global hypoxic damage Bilateral Watershed area strokes
Died Withdrew care Foster home
Vertebral artery shear C5/6 Epidural hematoma Cord signal
All except Case 5 had brain infarcts at 48 h, all except Case 7 had retinal hemorrhages, for cases 4 & 8, the epidural hematomas extended to the thoracic spine Cases 3, 6 and 7 underwent CPR before arrival to ED and Cases 1, 4 and 6–8 had evidence of hypoxic brain injury on MRI of the brain at 48 h.
In 2013, only eight were documented in the pediatric literature [29]. Spinal subdural hematomas are reported after inflicted injury in children but epidural hematomas are rare and may imply a more violent injury [30]. Without MRI examination, children with traumatic spinal epidural hematoma and minor symptoms may be missed. One limitation of this study is that we were unable to obtain autopsy reports for all of the mortalities in our series. It is unknown whether autopsy may have provided additional details of c-spine injuries in the mortalities. Furthermore, autopsy may better delineate the location of spinal hematomas (epidural vs. subdural), as well as microscopic injuries (nerve root avulsion and stretch injuries). This limitation is counterbalanced by complete prospective data collection and a strict
MRI protocol for examination of the brain and c-spine in all included children with admitted or witnessed trauma. We recommend diffusion weighted MRI imaging of the c-spine for shaken infants. If shaking is not confirmed, but children have suspicious physical findings such as retinal hemorrhages, c-spine MRI is recommended to better document the pattern of injury. After consultation with neuroradiology and neurosurgery, cervical collar protection, surgical fixation of unstable injuries or anticoagulation for a vessel intimal flap dissection may be recommended. The outcome of neurologic devastation combined with c-spine injury, supports the hypothesis that secondary hypoxic brain damage occurs with respiratory insufficiency when the neuroaxis is stretched. With early diagnosis, prevention of
Fig. 2. Corresponding MRI/MRA of cervical spine and brain in four children with inflicted trauma and a cervical spine injury.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049
6
J. Baerg et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
secondary neurologic injuries and specific neuroprotective strategies may be implemented. However, obtaining early MRI imaging to diagnose these injuries must balance against the risks of transport and critical-care monitoring in the MRI scanner of the acute trauma patient. Detailed documentation of a violent pattern of injury has forensic implications for future care of the child and law enforcement. The understanding of the pattern of injury in shaken infants and child abuse has evolved since Caffey's first report in 1946, followed by the publication of “The whiplash shaken infant syndrome” in 1972. By 2002, the features of brain injury on diffusion-weighted MRI in shaken infants were reported, but the cervical spine was not evaluated [31]. Through institution of an MRI protocol to evaluate the c-spine of all young victims of inflicted trauma, our series illustrates an expansion of the injury pattern of shaken infants to include c-spine injuries. This study demonstrates a higher incidence of cervical spinal and vascular injury than previously appreciated and merits further evaluation of the differences in biomechanics of inflicted versus accidental trauma to the cervical vasculature and spine. A larger case series is needed to determine the outcomes and treatments specific to these injuries in this young age group. Acknowledgment The authors acknowledge Rajaie Hazboun MD, for collection and review of imaging and figures. References [1] Kempe CH, Silverman FN, Steele BF, et al. The battered-child syndrome. JAMA 1962; 181:17–24. [2] Dashti SR, Decker DD, Razzaq A, et al. Current patterns of inflicted head injury in children. Pediatr Neurosurg 1999;31:302–6. [3] Maguire SA, Watts PO, Shaw AD, et al. Retinal haemorrhages and related findings in abusive and non-abusive head trauma: a systematic review. Eye (Lond) 2013;27:28–36. [4] Katz JS, Oluigbo CO, Wilkinson CC, et al. Prevalence of cervical spine injury in infants with head trauma. J Neurosurg Pediatr 2010;5:470–3. [5] Ghatan S, Ellenbogen RG. Pediatric spine and spinal cord injury after inflicted trauma. Neurosurg Clin N Am 2002;13:227–33. [6] Kemp A, Cowley L, Maguire S. Spinal injuries in abusive head trauma: patterns and recommendations. Pediatr Radiol 2014;44(Suppl. 4):S604–12. [7] Kemp AM, Jaspan T, Griffiths J, et al. Neuroimaging: what neuroradiological features distinguish abusive from non-abusive head trauma? A systematic review. Arch Dis Child 2011;96:1103–12. [8] Blumenthal I. Shaken baby syndrome. Postgrad Med J 2002;78:732–5.
[9] Vazquez E, Delgado I, Sanchez-Montanez A, et al. Imaging abusive head trauma: why use both computed tomography and magnetic resonance imaging? Pediatr Radiol 2014;44(Suppl. 4):S589–603. [10] Huang BH, Castillo M. Hypoxic–ishemic brain injury: findings from birth to adulthood. Radiographics 2008;28:417–39. [11] Bradford R, Choudhary AK, Dias MS. Serial neuroimaging in infants with abusive head trauma: timing abusive injuries. J Neurosurg Pediatr 2013;12:110–9. [12] Jordan LC, Hillis AE. Challenges in the diagnosis of pediatric stroke. Nat Rev Neurol 2011;7:199–208. [13] Kadom N, Zarir K, Vezina G, et al. Usefulness of MRI detection of cervical spine and brain injuries in the evaluation of abusive head trauma. Pediatr Radiol 2014;44:839–48. [14] Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974;13:81–4. [15] Copes WS, Champion HR, Sacco WJ, et al. The injury severity score revisited. J Trauma 1988;28:69–77. [16] Flom L, Fromkin J, Panigrahy A, et al. Development of a screening MRI for infants at risk for abusive head trauma. Pediatr Radiol 2016;46:519–26. [17] Knox JB, Schneider JE, Cage JM, et al. Spine trauma in very young children: a retrospective study of 206 patients presenting to a level 1 pediatric trauma center. J Pediatr Orthop 2014;34:698–702. [18] Caffey J. Multiple fractures in the long bones of infants suffering from chronic subdural hematoma. Radiology 1946;194:163–73. [19] Guthkelch AN. Infantile subdural haematoma and its relationship to whiplash injuries. Br Med J 1971;22(2(5759)):430–1. [20] Caffey J. The whiplash shaken infant syndrome: manual shaking by the extremities with whiplash-induced intracranial and intraocular bleedings, linked with residual permanent brain damage and mental retardation. Pediatrics 1974; 54:396–403. [21] McKinney AM, Thompson LR, Truwit CL, et al. Unilateral hypoxic–ischemic injury in young children from abusive head trauma, lacking craniocervical dissection or cord injury. Pediatr Radiol 2008;38:164–74. [22] King WJ, MacKay M, Sirnick A, et al. Shaken baby syndrome in Canada: clinical characteristics and outcomes of hospital cases. CMAJ 2003;168:155–9. [23] Jacob R, Cox M, Koral K, et al. MR imaging of the cervical spine in nonaccidental trauma: a tertiary institution experience. Am J Neuroradiol 2016;26. [24] Brown RL, Brunn MA, Garcia VF. Cervical spine injuries in children: a review of 103 patients treated consecutively at a level 1 trauma center. J Pediatr Surg 2001;36: 1107–14. [25] Knox J, Schneider J, Wimberly RL, et al. Characteristics of spinal injuries secondary to nonaccidental trauma. J Pediatr Orthop 2014;34:376–81. [26] Raul JS, Tracqui A, Ludes B. Homicide by occipitoatlantal dislocation. J Forensic Sci 2004;49:601–3. [27] Astur N, Klimo Jr P, Sawyer JR, et al. Traumatic atlanto-occipital dislocation in children: evaluation, treatment and outcomes. J Bone Joint Surg Am 2013; 18(95):e194(1–8). [28] Gleckman AM, Kessler SC, Smith TW. Periadventitial extracranial vertebral artery hemorrhage in a case of shaken baby syndrome. J Forensic Sci 2000;45: 1151–3. [29] Rangarajan V, Mavani SB, Nadkarni TD, et al. Traumatic cervical epidural hematoma in an infant. J Craniovertebr Junction Spine 2013;4:37–9. [30] Koumellis P, McConachie NS, Jaspan T. Spinal subdural haematomas in children with non-accidental head injury. Arch Dis Child 2009;94:216–9. [31] Biousse V, Suh DY, Newman NJ, et al. Diffusion-weighted magnetic resonance imaging in shaken baby syndrome. Am J Ophthalmol 2002;133:249–55.
Please cite this article as: Baerg J, et al, Cervical spine imaging for young children with inflicted trauma: Expanding the injury pattern, J Pediatr Surg (2017), http://dx.doi.org/10.1016/j.jpedsurg.2017.01.049