Abstract
There is growing evidence of spine injury in abusive head trauma (AHT). Historically, spine injury was considered rare in AHT because of a lack of attributable clinical symptoms or signs and a lack of advanced imaging. Increased use of MRI in AHT has been instrumental in helping identify evidence of ligamentous injuries of the spine. These findings can be difficult to identify on autopsy because of the size and location of the ligaments. Because spinal injury in AHT mostly involves ligamentous and soft tissues and only rarely involves bony fractures, more than 90% of the injury findings are missed on CT or radiography of the spine. Investigation of these findings and the injury patterns should lead to a better understanding of the mechanism of spinal injury. In this pictorial review, we describe the various manifestations of spinal ligamentous injury in AHT, as seen on MRI, in children younger than 48 months.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Abusive head trauma (AHT) is one of the leading causes of morbidity and mortality in children. The clinical presentation of children with head trauma varies from nonspecific symptoms to an abrupt collapse [1]. Clinical neurologic assessment of an infant’s spine injury can be challenging, and this can be further complicated by coexistent significant intracranial injury, which can mimic or mask spinal injury [2]. These factors lead to an under-appreciation of spinal injury. If we review the historical literature, spinal injury was considered rare [3,4,5]. This is not surprising because spinal injuries were masked clinically and the primary imaging modalities of investigation — historically CT and radiography — did not show significant abnormalities. But recent literature, based primarily on MRI findings, has demonstrated that evidence of spinal injury, whether direct or indirect, is fairly common [2, 6,7,8]. In this pictorial review, we describe the various manifestations of spinal ligamentous injury in AHT, as seen on MRI, in children younger than 48 months.
Anatomy
To accurately diagnose spinal ligamentous injury in children, it is vital to understand that the anatomy of an immature spine is different from that of a mature spine. These differences can result in vastly different clinical presentations and imaging findings, even with similar mechanisms of injury. Infants have a larger head relative to their body mass, weak neck muscles, large subarachnoid spaces, a thin and pliable skull, and a relatively flat skull base [2, 9]. The large head of an infant can comprise approximately a third of the total body weight. Large head size in combination with weak neck muscles creates the potential for severe craniocervical and cervical spinal injury. The incompletely ossified infant cervical spine has shallow horizontally oriented facet joints, underdeveloped spinous processes, physiological anterior wedging of the vertebral bodies, and immature lax supporting ligaments. All these variables place the infant’s ligaments and spinal cord at an increased risk for injury, particularly at the craniocervical junction, which houses the cervicomedullary junction [10, 11]. In addition, there is greater flexibility of the spinal column relative to the spinal cord because of ligamentous laxity, putting the cervical spinal cord at an increased risk for stretching or transient injury from vertebral body subluxation.
Imaging findings
Confessions have given us an insight into the various mechanisms of injury in abusive head trauma. Predominantly, it involves shaking with or without impact injury to the head and neck [12,13,14]. The most consistently identified imaging findings in AHT include posterior suboccipital muscle and ligamentous injuries [2, 6]. In more severe injuries, anterior ligamentous structures and bony injuries might also be identified.
Posterior suboccipital muscle and ligamentous complex injury
These include larger structures such as the nuchal ligament and smaller ligamentous structures such as the atlanto-axial and atlanto-occipital membranes, interspinous ligaments and capsular ligaments of occiput–C1 and C1–C2 articulations.
The nuchal ligament is a bilaminar fibroelastic intermuscular septum and has two components — the anterior membranous lamellar portion and posterior cord-like funicular segment. The lamellar portion is a membranous structure attached to the median part of the external occipital crest, the posterior tubercle of C1 and the medial aspect of bifid spines of the cervical vertebrae [15]. There is also a midline attachment to the posterior dura at the atlanto-axial and atlanto-occipital levels [16]. These membranes act as a septum for bilateral attachment of cervical muscles and their sheaths. The membranes fuse posteriorly to form a cord-like superficial structure (funicular segment), which extends from the posterior occipital protuberance to the spinous process of C7 (Fig. 1) [15]. The lamellar membrane contains fat and on short tau inversion recovery (STIR) sequence should appear predominantly hypointense. If injury is present, T2 hyperintensity is evident in this region along with fatty stranding on T1 or T2 sequence (Figs. 2, 3 and 4).
The posterior atlanto-axial and atlanto-occipital membranes are thin membranous structures posterior to the dura extending between the C1–C2 and occiput–C1, respectively. The injury to the membrane itself might be difficult to identify unless high-resolution images of this region are obtained. But the presence of T2 hyperintensity in this region represents evidence of acute injury. The imaging features of chronic injury, with thickening of these membranes, might be seen in older children with chronic craniospinal instability from skeletal dysplasia (Fig. 5).
Similarly, T2 hyperintensity along the capsular ligaments of occiput–C1 and C1–C2 articulation is frequently identified, and this represents evidence of injury. Distraction injury is suspected when there is also an increase in joint space with an effusion. Mild distraction injury, rather than frank dissociation, is more commonly seen in AHT (Fig. 6). Joint effusion and mild distraction injury might also be identified on soft-tissue coronal or sagittal reconstructions of spine CT.
Muscle edema in the suboccipital region might be symmetrical or asymmetrical and represents evidence of partial injury or contusion (Fig. 7). It is important to realize that vascular plexus is normally present in the anterior interspinous region and it should not be confused with edema or hematoma (Fig. 1). Also, if the fat suppression is not optimum or homogeneous, artifacts can mimic edema. Other significant injury patterns include injury to cord, nerve roots or dural perforation.
Anterior ligamentous injury
Although less common, anterior spinal injuries are known to occur in children with AHT. The ligamentous complex that supports the anterior spinal cord consists of the apical and alar ligaments of the dens, the anterior and posterior longitudinal ligament, the cruciform ligament and the tectorial membrane. Injuries to these are seen with a primary extension injury (Fig. 8).
Bony injury
Vertebral fractures are relatively less common and might be identified on skeletal surveys (Figs. 8 and 9). Injury to the thoracic or lumbar spine is also relatively uncommon.
Injury to the tectorial membrane
Partial or complete tear of the tectorial membrane can result from a significant injury to the craniocervical junction. Apart from attenuation or frank disruption of the tectorial membrane, stripping of the membrane from the clivus might also be identified. This can be associated with retroclival hematoma. Subdural hemorrhage can also be identified in the retroclival region in the presence of posterior fossa SDH. If the collection is deep to the tectorial membrane (extends from the mid-clivus to base of C2) the collection is epidural, whereas if the collection is superficial to the tectorial membrane but deep to the arachnoid, the collection is deemed subdural. Like spinal ligamentous injury, retroclival hematomas were once considered rare in the setting of AHT. However, retroclival collections can also be quite difficult to appreciate on radiographs or CT and might only be well demonstrated on MR imaging. In a retrospective study of retroclival collections associated with abusive head trauma, the detection was higher on MRI (85%) as compared to CT (38%) [17].
Literature review
Cervical spine ligamentous injuries (predominantly the nuchal, atlanto-occipital and atlanto-axial ligaments) are almost twice as common in AHT (78%) as compared to accidental trauma (46%) [2]. It is important to note in this study by Choudhary et al. [2] that in cases of AHT the cervical spine MRI was routinely obtained, whereas in the cases of accidental trauma the imaging was only obtained if there was a clinical concern for spine injury. The differences were found to be statistically significant, showing a high incidence of ligamentous injury in children with AHT when compared to cases with either accidental trauma or nontraumatic injuries. This might reflect the repetitive nature of the shaking injury or the extreme violence of the injury [12].
Among AHT cases, ligamentous injury was identified in 36% (Kadom et al. [6]), 32% (Governale et al. [7]), 59% (Rabbitt et al. [8]), 31% (Henry et al. [18]) and 67% (Jacob et al. [19]). The variability of reported ligamentous injuries in AHT as identified on MRI in the more recent literature likely relates to patient selection for imaging or retrospective reviews (variably included age groups and imaging modalities), or variable imaging techniques. Further reviews are important to advance our knowledge.
Hypoxic–ischemic injury and ligamentous injury of the neck
There is also a statistically significant correlation between hypoxic–ischemic injury of the brain and ligamentous injuries of the spine in cases of AHT [2, 19]. One hypothesis has been that spinal cord and/or brainstem injury results in apnea or disordered breathing, secondarily leading to hypoxic–ischemic brain injury [2, 20].
Clinical practice
The practice of routine MR imaging of the cervical spine is continuing to evolve at multiple institutions. Some institutions obtain it routinely, whereas others obtain it only if significant brain injury is present. The discussion is usually centered on the clinical utility of MR imaging of the cervical spine, the clinical management of the stable non-bony soft-tissue ligamentous injury of the spine, the cost of imaging and the timing of obtaining the study. In some cases, limited availability of sedation resources or access to MRI can be an additional challenge. In presence of ligamentous injuries without unstable bony fractures, some neurosurgical institutes manage them actively with a soft neck collar [21, 22]. To answer the concern regarding the cost of the study, first, the study provides additional information regarding a plausible mechanism of injury. Second, it helps exclude differential diagnoses, potentially a cost-saving measure on its own [23]. Last, the management approach should be contrasted with clinical management standards of other disease processes. If a clinical finding can be identified to confirm a diagnosis, explain the etiology or help exclude other potential causes, would it be part of your routine clinical practice? MRI of the cervical spine in AHT similarly adds significantly to our understanding of AHT mechanism of injury, helps confirm a diagnosis of AHT, excludes potential differential diagnoses and provides clear clinical value in the immediate management of the child. In addition, it helps in the determination of the eventual disposition of the child and care of the siblings.
Conclusion
Spinal injury is common in AHT and predominantly involves the craniocervical junction in most children. The injury also predominantly reflects a flexion injury mechanism of the soft-tissue structures, particularly involving the posterior suboccipital region. Because the spinal injury of AHT mostly involves ligamentous and soft tissues and only rarely involves bony fractures, more than 90% of the MRI injury findings are missed on CT or radiographs of the spine.
References
Choudhary AK, Servaes S, Slovis TL et al (2018) Consensus statement on abusive head trauma in infants and young children. Pediatr Radiol 48:1048–1065
Choudhary AK, Ishak R, Zacharia TT, Dias MS (2014) Imaging of spinal injury in abusive head trauma: a retrospective study. Pediatr Radiol 44:1130–1140
Katz JS, Oluigbo CO, Wilkinson CC et al (2010) Prevalence of cervical spine injury in infants with head trauma. J Neurosurg Pediatr 5:470–473
Polk-Williams A, Carr BG, Blinman TA et al (2008) Cervical spine injury in young children: a National Trauma Data Bank review. J Pediatr Surg 43:1718–1721
Kemp AM, Joshi AH, Mann M et al (2010) What are the clinical and radiological characteristics of spinal injuries from physical abuse: a systematic review. Arch Dis Child 95:355–360
Kadom N, Khademian Z, Vezina G et al (2014) Usefulness of MRI detection of cervical spine and brain injuries in the evaluation of abusive head trauma. Pediatr Radiol 44:839–848
Governale LS, Brink FW, Pluto CP et al (2017) A retrospective study of cervical spine MRI findings in children with abusive head trauma. Pediatr Neurosurg 53:36–42
Rabbitt AL, Kelly TG, Yan K et al (2020) Characteristics associated with spine injury on magnetic resonance imaging in children evaluated for abusive head trauma. Pediatr Radiol 50:83–97
Jaspan T, Narborough G, Punt JAG, Lowe J (1992) Cerebral contusional tears as a marker of child abuse — detection by cranial sonography. Pediatr Radiol 22:237–245
Bandak FA (2005) Shaken baby syndrome: a biomechanics analysis of injury mechanisms. Forensic Sci Int 151:71–79
Lustrin ES, Karakas SP, Ortiz AO et al (2003) Pediatric cervical spine: normal anatomy, variants, and trauma. Radiographics 23:539–560
Adamsbaum C, Grabar S, Mejean N, Rey-Salmon C (2010) Abusive head trauma: judicial admissions highlight violent and repetitive shaking. Pediatrics 126:546–555
Vinchon M, de Foort-Dhellemmes S, Desurmont M, Delestret I (2010) Confessed abuse versus witnessed accidents in infants: comparison of clinical, radiological, and ophthalmological data in corroborated cases. Childs Nerv Syst 26:637–645
Bell E, Shouldice M, Levin AV (2011) Abusive head trauma: a perpetrator confesses. Child Abus Negl 35:74–77
Williams A, Newell RLM (2005) The back. In. Standring S (ed) Gray’s anatomy. The anatomical basis of clinical practice, 39th edn. Elsevier, New York, pp 733–775
Dean NA, Mitchell BS (2002) Anatomic relation between the nuchal ligament (ligamentum nuchae) and the spinal dura mater in the craniocervical region. Clin Anat 15:182–185
Silvera VM, Danehy AR, Newton AW et al (2014) Retroclival collections associated with abusive head trauma in children. Pediatr Radiol 44:621–631
Henry MK, Wood JN (2018) Advanced cervical spine imaging in abusive head trauma: an update on recent literature and future directions. Acad Pediatr 18:733–735
Jacob R, Cox M, Koral K et al (2016) MR imaging of the cervical spine in nonaccidental trauma: a tertiary institution experience. AJNR Am J Neuroradiol 37:1944–1950
Johnson DL, Boal D, Baule R (1995) Role of apnea in nonaccidental head injury. Pediatr Neurosurg 23:305–310
Oh A, Sawvel M, Heaner D et al (2017) Changes in use of cervical spine magnetic resonance imaging for pediatric patients with nonaccidental trauma. J Neurosurg Pediatr 20:271–277
Maher CO (2017) Screening for spine injury in abusive head trauma. J Neurosurg Pediatr 20:269–270
Choudhary AK (2020) Understanding the importance of spinal injury in abusive head trauma (AHT). Pediatr Radiol 50:15–16
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
Dr. Choudhary is a medical expert in child abuse cases.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Haq, I., Jayappa, S., Desai, S.K. et al. Spinal ligamentous injury in abusive head trauma: a pictorial review. Pediatr Radiol 51, 971–979 (2021). https://doi.org/10.1007/s00247-020-04922-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00247-020-04922-8