Early cord degeneration in bifocal SCIWORA: a case report

Pediatr Radiol (1998) 28: 186±188 Ó Springer-Verlag 1998

Thierry Duprez Yvan De Merlier Philippe Clapuyt StØphane ClØment de ClØty Guy Cosnard Jean-François Gadisseux

Received: 1 August 1997 Accepted: 29 October 1997

)

T. Duprez ( ) ´ Y. De Merlier ´ P. Clapuyt ´ G. Cosnard Department of Medical Imaging, Cliniques Universitaires Saint-Luc, Avenue Hippocrate 10, B-1200 Brussels, Belgium

Early cord degeneration in bifocal SCIWORA: a case report

S. ClØment de ClØty Department of Paediatric Intensive Care, Cliniques Universitaires Saint-Luc, Brussels, Belgium J-F. Gadisseux Department of Paediatric Neurology, Cliniques Universitaires Saint-Luc, Brussels, Belgium

Introduction The immature spine of the newborn and infant has specific anatomical and functional properties which include ligamentous and soft-tissue laxity, unfused epiphyses and horizontal facet joints, which result in relative hypermobility when compared with an adult [1]. This increases the susceptibility for cord injury without bone lesions on plain X-rays, an entity which has been designated by the acronym SCIWORA (spinal cord injury without radiographic abnormalities) [2, 3].

Case report A 2-year-old-boy was admitted to the hospital with flaccid, high-level paraplegia 2 h after being knocked down by a car. Plain radiographs failed to reveal any bone lesion or displacement (not illustrated). Initial MRI using a mid-field unit (Gyroscan T5-III, Philips Medical Systems, the Netherlands) was performed 3 h after the trauma to rule out a compressive haemorrhage. Sagittal T1- and T2-weighted images were obtained (Fig. 1). Although there was no bony or cartilaginous abnormality, there were two intrinsic focal lesions within the cord at the level of C7 and T2. Disruption of the ventral portion of the C6±7 interspinous ligament was also depicted (Fig. 1 b). Somatosensory evoked potentials (SEP) confirmed complete loss of conduction through the diseased segment of the cord. As no improvement in neurological status occurred during the subsequent weeks, a further MRI was performed exactly

Abstract We report the MR features of very early spinal cord degeneration in a 2-year-old boy who had bifocal cord injury and normal plain films.

4 weeks after the initial trauma (Figs. 2, 3). The MRI demonstrated a peculiar type of bifocal cord transection at the two sites of presumed impact of the posterior arches on the cord, where it had become severely atrophic. The parenchyma interposed between these two points showed mild, symmetrical atrophy. The high-level flaccid paraplegia and the complete cervicothoracic block demonstrable by SEP had not improved at the time.

Discussion Cord injury in early childhood in the absence of abnormalities on plain films has been widely reported [1±3]. As in older patients, the exquisite ability of MRI to disclose intrinsic cord lesions in children has also been highlighted [4]. The case reported here, of cord transection with severe atrophy occurring within 1 month, exemplifies the pathophysiology of SCIWORA and highlights additional features. Both lesions were located in front of a spinous process and were felt to result from impact of the neural arch on the cord, reflecting the well-known hypermobility of the immature spinal joints in infants [1, 2]. The presence of two sites of presumed severe impact on the cord indicates that unlike the skull, which is thought to absorb kinetic energy with relative preservation of the brain, spinal subluxation cannot offer the same benefit. It is likely that the immature, lax, cartilag-

187

Fig. 1 a, b MRI 3 h after trauma. a Mid-sagittal T1-weighted SE image (400/20) of the cervico-thoracic cord shows two subtle focal areas of slightly decreased signal intensity within the cord at the C7 and T2 levels (black arrowheads). There is no significant cord swelling. b Mid-sagittal T2-weighted FSE image (3200/12O/ETL 12) shows the two focal abnormalities which are better seen as hyperintense foci (thick black arrows). Oedema within the ventral portion of the interspinous space is also detected at the C6±7 level (thin black arrow)

1a

1b

2a

2b

3a

Fig. 2 a, b MRI 4 weeks after trauma. a Mid-sagittal T1-weighted image shows the development of focal cord atrophy at the sites of the focal signal abnormalities seen in Fig. 1. The intervening cord between these two endpoints is atrophic. b T2-weighted image shows the cord transection as a more linear hyperintense area (between white arrows) Fig. 3 a±f MRI 4 weeks after trauma. a Mid-sagittal T1-weighted image. b±f Serial transverse T2-weighted FSE images through the cord at involved levels. Artifactual dark appearance of the subarachnoid spaces is due to CSF pulsations in this FSE technique. b At C5 level, the cord is normal in size and signal intensity. c At C6±7 level, the cord is severely atrophic and hyperintense. d At the level of T1, there is an intermediate segment of injured cord showing mild atrophy, a less hyperintense posterior focus (white arrow) and hyperintense signal within the grey matter butterfly'. e At the level of T2, the cord shows severe atrophy and hyperintensity. f At the level of T3, the cord is normal

inous spinal joints in infancy cannot absorb kinetic energy as the skull does. The sites of direct impact underwent strikingly rapid and severe atrophy and disclosed high-signal intensity on T2-weighted images which could be related to oedema, demyelination, reactive gliosis, or a combination of these. The speed of these changes might represent an intrinsic global fragility to traumatic injuries, as well as a peculiar susceptibility for very prompt atrophy. The intervening cord between the two impact points also disclosed abnormally high-signal intensity on T2-

3b

3c

3d

3e

3f

weighted images, but this was heterogeneous on transverse views (Fig. 3 d); a postero-medial area of the cord was less hyperintense and the central grey matter area (the grey butterfly') appeared more hyperintense than is usually observed in healthy persons. The relative posterior hypointensity may simply reflect the topography of a fortuitously less injured portion of the cord or evidence that the posterior columns exhibit greater intrinsic resistance to trauma at this age. In adults, oedematous swelling of injured cord is frequently shown at MRI to extend beyond the area of im-

188

pact. This MR finding is common to all spinal cord trauma, ranging from partial reversible contusion to transection. This was not what was observed in this case, and we are unsure as to whether this represents a general feature which differentiates paediatric from adult cord trauma.

Finally, the upper impact point was located in the lower cervical segment, but the lower one was in the upper thoracic segment. The relative protection offered by the rigid rib cage against vertebral subluxations in adults may not apply in infants.

References 1. Barkovitch AJ (1995) Destructive brain disorders in childhood. In: Barkovitch AJ (ed) Pediatric neuroimaging, 2nd edn. Raven Press, New York, pp 155±157

2. Pang D, Wilberger JE Jr (1982) Spinal cord injury without radiographic abnormalities in children. J Neurosurg 57: 114±129 3. Osenbach RK, Menezes AH (1989) Spinal cord injury without radiographic abnormality in children. Pediatr Neurosci 15: 168±174

4. Betz RR, Gelman AJ, Defilipp GJ, et al (1987) Magnetic resonance imaging (MRI) in the evaluation of spinal cord injured children and adolescents. Paraplegia 36: 92±99