Article Text

Repeated syncopes and extended paediatric hydrosyringomyelia/Chiari I malformation: relation or coincidence ?
  1. Department of Pediatrics
  2. Department of Radiology, University of Bonn, Germany
  1. Dr J Woelfle, Zentrum f Kinderheilkunde, Rhein Friedrich-Wilhelms-Universität, Adenauerallee 119, D-53113 Bonn, Germany. Telephone 0049 228 2873200; fax 0049 228 2873314.
  1. Department of Pediatrics
  2. Department of Radiology, University of Bonn, Germany
  1. Dr J Woelfle, Zentrum f Kinderheilkunde, Rhein Friedrich-Wilhelms-Universität, Adenauerallee 119, D-53113 Bonn, Germany. Telephone 0049 228 2873200; fax 0049 228 2873314.

Statistics from

Syringomyelia is defined as a condition of tubular cavitations within the spinal cord, lined by glial tissue. In theory it differs from hydromyelia, a dilatation of the central canal of the spinal cord, which is lined with ependyma. However, in practice the distinction between the conditions is often difficult to make; thus the term hydrosyringomyelia for all intraspinal cavities of a non-tumorous nature has been proposed. Hydrosyringomyelia is often associated with anomalies of the posterior fossa, the most common of which is Chiari type I malformation. In a few cases, an association of hydrosyringomyelia with isolated syncopal events has been described. These syncopes occurred especially in adult patients with associated Chiari type I malformation and were typically, but not in all cases, preceded by an increase in intrathoracic pressure caused by coughing, sneezing, or other exertion such as a Valsalva manoeuvre.1-3 Here, we report the case of a child with repeated syncopes, associated with hydrosyringomyelia/Chiari I malformation.

This 10 year old boy from Zaire was admitted for evaluation of syncope. This was the first episode; the syncope occurred while at rest at school. About one minute before the syncope, the patient experienced numbness of the left arm, occipital headache, and double vision, but no coughing, straining, or sneezing. Pallor of mucous membranes accompanying the syncope was not seen. Loss of consciousness lasted about five seconds, after which numbness of the arm as well as the double vision had disappeared. Except for the occipital headache, the patient then felt well.

Thus far psychomotoric development had been normal with excellent school performance in primary school. Family history showed no cases of epilepsy or other neurological affections.

On examination the child was alert and oriented and had fluent speech and intact comprehension. General physical examination was unremarkable with normal findings especially for the cardiovascular system. On neurological examination his mental as well as cranial nerve status was normal. Muscle tendon reflexes, muscle tone, and strength were normal. Testing of the sensory system disclosed no abnormalities; thermal, vibratory, and pinprick sensations were symmetric. Fine and gross motor proficiency were normal. No pyramidal signs, spasticity, or rigidity were evident.

During the follow up of 24 months, four further syncopes occurred with the same symptomatology as described for the first episode, but without the preceding numbness of the left arm or headache. Again there was no preceding coughing or sneezing. Three of these five syncopal events were preceded by movements and mild changes of body position such as bending to his schoolbag. Subsequent to all these syncopal events the patient’s neurological status was normal.

The patient’s cardiological investigation, including repeated Schellong tests, Valsalva manoeuvres, ECG, echocardiographies and 24 hour ECGs, was normal. Under clinical observation changes of body position imitating everyday movements (such as bending to the floor) were repeatedly performed, but did not lead to syncope. Several resting EEG records with photostimulation/hyperventilation as well as two sleeping EEG records and 24 hour EEG records showed normal activity. Due to the unexplained nature of the repeated syncopes, a cerebral MRI was performed. This showed type I Chiari malformation with herniated cerebellar tonsils and compression of the foramen magnum. A subsequent MRI of the spine showed hydrosyringomyelia from C5 to T4, with further smaller cavitations to T7, with no evidence of cavitation above C4. No associated tumour could be found (figure). After consultation with our neurosurgeons we decided against neurosurgical intervention for the time being in view of the completely normal neurological status of the patient. A cerebral and spinal control MRI one and two years later showed no progression of the syrinx.

After diagnosing the extended hydrosyringomyelia and Chiari type I malformation, the question arose whether there is a relation between this and the repeated syncopal events. Considering the normal results of repeated cardiological investigations and EEG records (including resting, sleeping, and 24 hour EEGs) a cardiogenic or epileptogenic aetiology of the described syncopes seems unlikely.

There are various pathophysiological models of syncopes in patients with hydrosyringomyelia. According to the model of Williams, an increase of intrathoracic pressure caused, for example, by coughing, leads to a shift of CSF from the spinal to the cranial compartment. During the subsequent relaxation period a flow back to the spinal compartment due to a “sucking effect” may lead to hindbrain herniation with impaction of the cerebellar tonsils and to interference with the medullary baroreceptor reflex or dysfunction of the midbrain reticular activating system, thus leading to syncope.4 In our patient, with syncopes occurring with no evidence of preceding symptoms such as coughing to cause consecutively cerebral or spinal pressure changes, this explanation seems unlikely. The fact that autonomic disturbances occur even in subjects without foramen magnum anomaly indicates that anomalies of the posterior fossa may be not the only factor in the pathogenesis of autonomic disturbances associated with syringomyelia. In this context Nogues et al found some subclinical autonomic disturbances in patients with syringomyelia, especially in those with brainstem involvement. However, in some patients with no signs of such involvement, a fall of more than 2 SD in mean arterial pressure in response to standing was still found. Therefore for some patients the authors assume an underlying sympathetic defect (for example, produced by destruction of the lateral horns of grey matter), which is incomplete and not extensive enough to cause permanent orthostatic hypotension.5 This would suggest that cavitation in the medulla with involvement of sympathetic structures could be another factor responsible for consecutive autonomic disturbances in patients with hydrosyringomyelia. This defect could result in temporary interruption of the spinal vegetative reflex arc with temporary postural hypotension. Considering the pattern of syringomyelic areas affected in our patient, damage to sympathetic structures neighbouring the syrinx may have led to temporary dysfunction of the sympathetic system with subsequent syncope. The assumption of a merely temporary interruption of the spinal vegetative reflex might explain the repeated normal results of clinical examination and Schellong tests, and that several changes of body position did not lead to a syncopal event in our patient. Furthermore it may be that pathophysiology of syncopes in patients with hydrosyringomyelia/Chiari I malformations is more complex—for example, due to synergistic neuropsychological influences such as increased attention while studying at school.

Despite the theoretical models outlined, the relation between isolated syncopal events in patients with hydrosyringomyelia and anomalies of the posterior fossa generally remains speculative. Furthermore the association of isolated syncopes and hydrosyringomyelia represents a very small fraction of all patients presenting with syncopes. Thus the decision to perform cerebral MRI in patients with unexplained syncopes must be considered on a case by case basis, although our case report shows that some patients may profit by performance of cerebral and cervical MRI to rule out hydrosyringomyelia and anomalies of the posterior fossa.


Figure 1

T1 weighted spin echo midsagittal MRI (TR: 404 ms, TE: 15 ms) showing Chiari I malformation with herniated cerebellar tonsils (arrow) and spindle shaped hydrosyringomyelia from C5 to T4 (arrow), with further smaller cavitations to T7 (arrow).

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.