Treatment of Chiari Type I Malformation
Treatment of Chiari Type I Malformation
Object. The authors describe the results of performing a standard posterior craniovertebral decompression and placement, if indicated, of a syringosubarachnoid shunt for the treatment of patients with Chiari I malformation with and without syringomyelia.
Methods. This is a retrospectively analyzed consecutive series of 66 patients (mean patient age 15 years, range 1-53 years). The uniform posterior craniovertebral decompression consisted of a small suboccipital craniectomy, a C-1 laminectomy, microsurgical reduction of the cerebellar tonsils, and dural closure with a synthetic dural graft to increase the cerebrospinal fluid space at the craniocervical junction. The presence of a large syrinx, with significant thinning of the spinal cord tissue and obliteration of the spinal subarachnoid space, particularly when combined with syrinx-related symptoms, was an indication for the placement of a syringosubarachnoid shunt.
In 32 patients Chiari I malformation alone was present, and 34 in patients it was present in combination with syringomyelia. Clinical findings included pain, neurological deficits, and spinal deformity. The presence of syringomyelia was significantly associated with the presence of scoliosis (odds ratio 74.4 [95% confidence interval 8.894-622.4]).
All patients underwent a posterior craniovertebral decompression procedure. In 22 of the 34 patients with syringomyelia a syringosubarachnoid shunt was also placed. The mean follow-up period was 24 months (range 3-95 months). Excellent outcome was achieved in 54 patients (82%) and good outcome in 12 (18%). In no patient were symptoms unchanged or worse at follow-up examination, including four patients who initially required a second operation for persistent syringomyelia. Pain was more likely to resolve than sensory and motor deficits after decompressive surgery.
Radiological examination revealed normalization of tonsillar position in all patients. The syrinx had disappeared in 15 cases, was decreased in size in 17, and remained unchanged in two.
Conclusions. Posterior craniovertebral decompression and selective placement of a syringosubarachnoid shunt in patients with Chiari I malformation and syringomyelia is an effective and safe treatment. Primary placement of a shunt in the presence of a sufficiently large syrinx appears to be beneficial. The question of if and when to place a shunt, however, requires further, preferably prospective, investigation.
As described and classified over a century ago, herniation of the cerebellar tonsils through the foramen magnum into the cervical spinal canal with obliteration of the cerebellomedullary cistern is the primary feature of Chiari I malformation. It is the mildest form of the Chiari complex that includes more extensive hindbrain anomalies often combined with spina bifida and severe forms of brainstem dysfunction and structural abnormality. Nevertheless, the disorder can be associated with significant symptomatology, risk of secondary injury due to trauma, and the risk of progression and damage of the spinal cord due to associated syringomyelia. Syringomyelia (also known as hydromyelia) is found in 50 to 70% of the patients with Chiari I malformation. Numerous and thorough efforts to understand the pathogenesis of a syrinx caused by the obliteration of the craniovertebral CSF spaces have been undertaken.
Gardner's hydrodynamic theory was developed from the premise of an open connection from the fourth ventricle into the central canal of the cord. It suggested that the extension of the syrinx cavity is caused by a "water hammer" effect, which is a pulsatile transmission of the CSF from the fourth ventricle into the central canal in synchrony with the arterial blood pressure waves.
In Williams' craniospinal pressure dissociation hypothesis, he postulated an effect of venous (respiratory) pulsations resulting in a craniospinal pressure differences as the primary mechanism responsible for syrinx formation.
With the advent of dynamic MR imaging the image acquisition could be related to the arterial and venous pulsations and the resulting CSF flows at the craniovertebral junction could be studied. Based on findings of dynamic imaging technology, an explanation has been put forward for the pathogenesis of a syrinx. This concept suggests that the cisternal obstruction at the foramen magnum results in a blockage of CSF flow during systole. This blockage causes a systolic wave in the spinal CSF that acts on the surface of the spinal cord, which in turn results in progression of syringomyelia by repeated compression of the cord, probably by mechanical disruption of the cord tissue, and by forcing fluid into the cord from the outside in. The pulsatile pressure waves force CSF into the cord through the perivascular and interstitial spaces. In addition, the syrinx fluid is propelled longitudinally within the syrinx during each cardiac cycle. The pathological CSF flow pattern and its restoration after decompressive surgery has been described.
With the increased understanding of the pathophysiological basis for the development of syringomyelia in Chiari I malformation and with the common experience accumulated in numerous series, there is little controversy today that the common denominator of all treatment strategies is posterior craniovertebral decompression. Details of the procedure, its extent and invasiveness, the technique of dural closure, and whether a drainage system should be placed in the syrinx cavity, however, are subject to considerable dispute.
In this report we present a series of patients with Chiari I malformation with and without syringomyelia. A standardized treatment was used: posterior craniovertebral treatment included a decompressive suboccipital craniectomy, a C-1 laminectomy, microsurgical reduction of the cerebellar tonsils, and dural closure with a synthetic graft to ensure enlargement of the craniovertebral CSF space. The spectrum of clinical presentations, radiological and surgical findings, and the postsurgical outcome is described. Details of the surgical strategy are also discussed.
Object. The authors describe the results of performing a standard posterior craniovertebral decompression and placement, if indicated, of a syringosubarachnoid shunt for the treatment of patients with Chiari I malformation with and without syringomyelia.
Methods. This is a retrospectively analyzed consecutive series of 66 patients (mean patient age 15 years, range 1-53 years). The uniform posterior craniovertebral decompression consisted of a small suboccipital craniectomy, a C-1 laminectomy, microsurgical reduction of the cerebellar tonsils, and dural closure with a synthetic dural graft to increase the cerebrospinal fluid space at the craniocervical junction. The presence of a large syrinx, with significant thinning of the spinal cord tissue and obliteration of the spinal subarachnoid space, particularly when combined with syrinx-related symptoms, was an indication for the placement of a syringosubarachnoid shunt.
In 32 patients Chiari I malformation alone was present, and 34 in patients it was present in combination with syringomyelia. Clinical findings included pain, neurological deficits, and spinal deformity. The presence of syringomyelia was significantly associated with the presence of scoliosis (odds ratio 74.4 [95% confidence interval 8.894-622.4]).
All patients underwent a posterior craniovertebral decompression procedure. In 22 of the 34 patients with syringomyelia a syringosubarachnoid shunt was also placed. The mean follow-up period was 24 months (range 3-95 months). Excellent outcome was achieved in 54 patients (82%) and good outcome in 12 (18%). In no patient were symptoms unchanged or worse at follow-up examination, including four patients who initially required a second operation for persistent syringomyelia. Pain was more likely to resolve than sensory and motor deficits after decompressive surgery.
Radiological examination revealed normalization of tonsillar position in all patients. The syrinx had disappeared in 15 cases, was decreased in size in 17, and remained unchanged in two.
Conclusions. Posterior craniovertebral decompression and selective placement of a syringosubarachnoid shunt in patients with Chiari I malformation and syringomyelia is an effective and safe treatment. Primary placement of a shunt in the presence of a sufficiently large syrinx appears to be beneficial. The question of if and when to place a shunt, however, requires further, preferably prospective, investigation.
As described and classified over a century ago, herniation of the cerebellar tonsils through the foramen magnum into the cervical spinal canal with obliteration of the cerebellomedullary cistern is the primary feature of Chiari I malformation. It is the mildest form of the Chiari complex that includes more extensive hindbrain anomalies often combined with spina bifida and severe forms of brainstem dysfunction and structural abnormality. Nevertheless, the disorder can be associated with significant symptomatology, risk of secondary injury due to trauma, and the risk of progression and damage of the spinal cord due to associated syringomyelia. Syringomyelia (also known as hydromyelia) is found in 50 to 70% of the patients with Chiari I malformation. Numerous and thorough efforts to understand the pathogenesis of a syrinx caused by the obliteration of the craniovertebral CSF spaces have been undertaken.
Gardner's hydrodynamic theory was developed from the premise of an open connection from the fourth ventricle into the central canal of the cord. It suggested that the extension of the syrinx cavity is caused by a "water hammer" effect, which is a pulsatile transmission of the CSF from the fourth ventricle into the central canal in synchrony with the arterial blood pressure waves.
In Williams' craniospinal pressure dissociation hypothesis, he postulated an effect of venous (respiratory) pulsations resulting in a craniospinal pressure differences as the primary mechanism responsible for syrinx formation.
With the advent of dynamic MR imaging the image acquisition could be related to the arterial and venous pulsations and the resulting CSF flows at the craniovertebral junction could be studied. Based on findings of dynamic imaging technology, an explanation has been put forward for the pathogenesis of a syrinx. This concept suggests that the cisternal obstruction at the foramen magnum results in a blockage of CSF flow during systole. This blockage causes a systolic wave in the spinal CSF that acts on the surface of the spinal cord, which in turn results in progression of syringomyelia by repeated compression of the cord, probably by mechanical disruption of the cord tissue, and by forcing fluid into the cord from the outside in. The pulsatile pressure waves force CSF into the cord through the perivascular and interstitial spaces. In addition, the syrinx fluid is propelled longitudinally within the syrinx during each cardiac cycle. The pathological CSF flow pattern and its restoration after decompressive surgery has been described.
With the increased understanding of the pathophysiological basis for the development of syringomyelia in Chiari I malformation and with the common experience accumulated in numerous series, there is little controversy today that the common denominator of all treatment strategies is posterior craniovertebral decompression. Details of the procedure, its extent and invasiveness, the technique of dural closure, and whether a drainage system should be placed in the syrinx cavity, however, are subject to considerable dispute.
In this report we present a series of patients with Chiari I malformation with and without syringomyelia. A standardized treatment was used: posterior craniovertebral treatment included a decompressive suboccipital craniectomy, a C-1 laminectomy, microsurgical reduction of the cerebellar tonsils, and dural closure with a synthetic graft to ensure enlargement of the craniovertebral CSF space. The spectrum of clinical presentations, radiological and surgical findings, and the postsurgical outcome is described. Details of the surgical strategy are also discussed.
