Is hydrocephalus after spinal cord injury really caused by the injured spinal cord?
Two case reports and a literature review

Summary

Posttraumatic hydrocephalus caused by cerebrospinal fluid circulation disturbances frequently complicates the clinical course and treatment after craniocerebral injury. Hydrocephalus complicating spinal cord injury is only exceptionally reported. The paper presents two cases of complete cervical spinal cord injury with subsequent development of hydrocephalus. The analysis of both cases and literature data confirmed the dominant role of non-spinal factors in the development of hydrocephalus after spinal cord injury. Despite the exceptional occurrence of hydrocephalus after spinal cord injury, this diagnosis should be considered in cases of delayed deterioration of a patient with cervical spinal cord injury, particularly if cerebrospinal fluid space abnormalities and posttraumatic subarachnoid haemorrhage are present.

Key words:
spinal cord injury − hydrocephalus − subarachnoid hemorrhage − Blake’s pouch cyst − neuroendoscopy

Introduction

The incidence of hydrocephalus after brain injury may reach 29% [1,2].There is a similar relationship between disturbed cerebrospinal fluid (CSF) circulation in spinal cord injury and post-traumatic hydrosyringomyelia (0.3–3.2% of spinal cord injuries)(3). However, hydrocephalus after spinal cord injury is rarely reported [4,5], and moreover, the causal relationship between spinal cord injury and hydrocephalus development is unclear [6]. The paper provides an analysis of spinal and extraspinal causes of hydrocephalus in two patients after complete cervical spinal cord injury.

Case report I 

A 19-year-old man had a car accident. Post-injury quadriplegia below C5 (Frankel A) and a left tibial fracture were immediately evident. A computed tomography (CT) scan of the cervical spine revealed a C5/6 luxation fracture. Because of prolonged awakening after emergency spinal surgery, a brain CT was performed. Intracranial hematoma was ruled out, but triventricular hydrocephalus was found (lateral ventricular width at frontal horns level - 38 mm; third ventricular width 23 mm; Evans index 0.54). A decreased level of consciousness (GCS 13−14) required intraventricular drainage placement. Intracranial pressure (ICP) remained normal throughout the monitoring period and the patient gradually regained full consciousness. A history of complicated otitis media (at age 8 years) was reported by his parents, so the ventricular dilatation was considered to be a chronic problem caused by suspected postinflammatory aqueductal stenosis, but no preinjury neuroradiological study was available. Subsequent treatment course was complicated by sepsis, multiorgan failure, and left lower limb compartment syndrome requiring amputation.

Two months post-injury, the patient started to complain of severe headaches. Fundoscopy revealed papilledema (2D). Progressive dilatation of the lateral (38 mm to 45 mm; Evans index 0.67) and third ventricle (23 to 27 mm) was evident on magnetic resonance imaging (MRI). The size of the fourth ventricle was stable, but its posterior part widely communicated with a dilated cisterna magna. The neuroradiological diagnosis was Blake’s pouch cyst (Fig. 1). Endoscopic third ventriculostomy (ETV) led to immediate relief of headaches and the ventricles gradually returned to presurgical size. During the 8-year follow-up period, there was no recurrence of intracranial hypertension. The patient regained some elbow flexion and extension (3/5).

Case report II

A 27-year-old woman suffered a head and cervical spine injury after a tree fell on her car, with immediate onset of quadriplegia (Frankel A). A brain CT showed small biparietal contusions, discrete subarachnoid bleeding, and a small hematoma in the cisterna magna. Brain ventricles were normal (Evans index 0.28). Cervical spine CT showed non-dislocated C1 arch fracture, C3/4 subluxation, and C5/6 comminuted luxation fracture. Despite emergency surgery, quadriplegia with intermittent need for ventilatory support persisted. The postsurgical course was complicated by sepsis, pulmonary infection, and bradyarrhythmias. Because of progressive apathy and negativism, brain CT was performed 6 weeks after injury. Tetraventricular hydrocephalus (lateral ventricles 24 mm; third ventricle 16 mm; fourth ventricle 22 mm; Evans index 0.51) with periventricular lucencies and gyral flattening was evident. The MRI anatomy of the posterior fourth ventricle and cisterna magna was consistent with Blake’s pouch cyst (Fig. 2). A ventriculoperitoneal shunt was considered, but febrile status, intraabdominal fluid, and positive cultures increased the risk of implant infection. ETV was performed to bypass the possible occlusion of cerebrospinal fluid (CSF) pathways at the level of the foramen magnum cisterns. After surgery, there was a slight improvement of nonverbal communication; the ventricles remained unchanged. Before the planned shunting, the patient was transferred to another neurosurgical centre at her family’s request. According to information obtained by telephone, the patient remained quadriplegic and was transferred to a rehabilitation centre.

Discussion

The role of spinal cord injury as the sole cause of hydrocephalus is unequivocal in only a few papers. Aghi et al. described a patient with iatrogenic spinal cord injury at the C1/2 level during cervical perimyelography causing intramedullary hematoma and intraventricular bleeding with acute hydrocephalus [4]. A patient reported by Joseph et al. had multiple stab injuries including craniocervical junction trauma with incomplete spinal cord injury and premedullary subarachnoidal bleeding. The onset of hydrocephalus symptoms five months post-injury and improvement after shunt implantation support the causal role of posthemorrhagic CSF malresorption [5]. Ascending spinal cord oedema blocking CSF flow is another possible cause of hydrocephalus [7].

The correlation between spinal cord injury and hydrocephalus is less clear in other reports. A paper by Son et al. describes a patient with severe osteoproductive changes and minor trauma causing spinal cord injury at the C1–C4 level in whom acute hydrocephalus requiring shunt surgery developed two days post-injury. There are two possible causes of this complication: ascending spinal cord oedema and CSF malresorption caused by bleeding in the cerebellomedullary cistern [8]. Menéndez et al described two possible causes of hydrocephalus: subarachnoid bleeding from the injured vertebral artery and type III occipital condyle fracture with a bone fragment pushing into the medulla causing CSF flow blockage [9].

Spinal cord injury results in hyperproteinorrhachia that may cause CSF malresorption and intracranial or spinal arachnoiditis responsible for hydrocephalus in some spinal tumours, and it may be a cause of late hydrocephalus [10]. The possible role of the exclusion of the spinal sites of CSF resorption by spinal canal block is supported by studies on models of kaolin-induced hydrocephalus [11].

Decompensation of anomalous CSF circulation (aqueductal stenosis, Blake’s pouch cyst) is an acceptable explanation for the symptomatic hydrocephalus progression in Case I. The role of CSF malresorption is not supported by the good ETV result. The interval between septic complications and hydrocephalus decompensation does not support the role of sepsis. Spinal cord oedema is improbable because the patient’s consciousness and motor functions remained stable.

The small contusions with traces of subarachnoid blood in Case II are a highly improbable cause of malabsorption hydrocephalus. Although the posterior fossa anomaly and the intracisternal bleeding justify considering CSF flow blockage, the failure of neuroendoscopic surgery favors the causal role of CSF malresorption. Extracranial factors can also not be excluded (repeated septic complications).

Conclusion

Despite its rarity, hydrocephalus should be considered as a possible cause of late deterioration in spinal cord injury patients. Under exceptional circumstances, spinal cord injury may be the sole cause of hydrocephalus requiring neurosurgical treatment, particularly in upper cervical cord injuries causing subarachnoid or intraventricular bleeding. However, based on own experience and literature data, extraspinal factors, such as traumatic intracranial bleeding, preexisting anomalies of CSF circulation and extracranial factors play a key role in the development of hydrocephalus after spinal cord injury.

Conflict of Interests

The authors declare that they have not conflict of interest in connection with the emergence of and that the article was not published in any other journal.

Doc. MUDr. Jan Chrastina, Ph.D.

Nebovidy u Brna 109

664 48 Nebovidy

e-mail: jan.chrastina@fnusa.cz