Abstract
SUMMARY: We investigated whether idiopathic spinal cord herniation is a congenital or acquired condition and undertook a study to determine the risk of developing iSCH in patients with persistent ventral spinal CSF leaks. De novo formation of iSCH was established among all 6 patients with iSCH who had undergone prior spinal imaging for symptoms unrelated to iSCH. Among 51 patients with persistent ventral spinal CSF leaks, iSCH developed in 2 patients (probability increased from 0% at 5 years to 9.4% at 10 years). This study shows that iSCH is an acquired condition, and early treatment of ventral CSF leaks offers a unique opportunity to prevent neurologic disability.
ABBREVIATIONS:
- iSCH
- idiopathic spinal cord herniation
- SIH
- spontaneous intracranial hypotension
- SCH
- spinal cord herniation
- SD
- standard deviation
Idiopathic spinal cord herniation (iSCH) is a rare cause of myelopathy characterized by herniation of the spinal cord through a ventral dural defect into the epidural space and in severe cases into a cavity within the vertebral body.1 The pathophysiology of the condition remains a point of contention, and both congenital and acquired etiologies have been proposed.1 Some have suggested that a spontaneous ventral dural tear, often at the site of an osteophyte and as commonly seen in spontaneous intracranial hypotension (SIH),2 is the underlying cause of iSCH,3 while others have hypothesized a congenital etiology with development in an early embryologic stage and terms such as “congenital transdural appendix of the spinal cord” or “spinal dysraphism with a hamartomatous appendix of the spinal cord” rather than iSCH have been proposed.4,5 We now report a group of patients with de novo formation of iSCH, providing proof that this is an acquired and not a congenital condition. The risk of developing iSCH in patients with chronic ventral CSF leaks has not been investigated, to our knowledge, and is thus unknown. The second part of the present study was undertaken to quantify the risk of developing iSCH in patients with persistent CSF leaks due to a ventral dural tear.
MATERIALS AND METHODS
This study was approved by the institutional review board of our medical center. Requirement for written informed consent was waived.
De Novo iSCH Study
Since January 1, 2015, all patients with spinal cord herniation (SCH) evaluated by us have been enrolled prospectively in a registry. Using this registry, we identified all patients with spontaneous iSCH who had undergone prior spinal imaging for symptoms unrelated to iSCH for the period from January 1, 2015, until December 31, 2023. The imaging classification of Imagama et al6 was used to describe the features of the iSCH.
Long-Term Sequelae of Persistent Ventral Spinal CSF Leak Study
Since January 1, 2001, all patients with SIH meeting the modified criteria of the International Classification of Headache Disorders, third edition,7,8 evaluated by us have been enrolled prospectively in a registry. In this registry, patient data on major manifestations of SIH, including superficial siderosis, bibrachial amyotrophy, coma, cerebral venous thrombosis, behavioral variant frontotemporal dementia, and SCH, are abstracted prospectively and updated with each change or addition to the data elements. Using this registry, we identified all patients with ventral spinal CSF leaks who were first evaluated within 1 year of symptom onset. The long-term sequelae of spontaneous spinal CSF leaks generally do not occur within the first year of symptom onset,9-12 and we set this 1-year limit to minimize ascertainment bias. To calculate the risks of untreated CSF leaks, we then identified those patients who had at least 1 year of clinical follow-up from the onset of SIH symptoms. All patients were considered at risk until their leak was sealed as confirmed by posttreatment spine imaging. Thus, patients who underwent successful repair of their ventral CSF leak were censored at that point in time. In addition, all patients who did not yet have radiographic confirmation of resolution of their spinal CSF leak were contacted for the purpose of this study for clinical and radiographic follow-up examination. Patients were interviewed specifically by their treating physicians about symptoms associated with iSCH. The radiographic diagnosis of iSCH was made by 1 of 3 experienced board-certified neuroradiologists who were not blinded to the diagnosis of SIH. The diagnosis of ventral spinal CSF leak was made radiographically if the extrathecal CSF collection was confined to the ventral aspect of the common thecal sac or if a ventral dural tear was identified by digital subtraction myelography or dynamic CT-myelography even if the extrathecal CSF was not confined to the ventral epidural space (eg, was circumferential). This portion of the study was a continuation of a previously published study of the long-term sequelae of persistent ventral spinal CSF leaks.13
Statistical Analysis
Failure-free survival was defined as time from the date of symptom onset to the development of iSCH. Hazard (failure) rates were estimated by using the Kaplan–Meier method. The hazard is the probability that the adverse event will occur in a defined time interval. Censoring occurs because the patient is lost to follow-up or did not experience the event of interest before the end of the study period. Monthly hazards were calculated as the monthly number of events divided by total follow-up time accumulated by the patients at risk in that month. The 95% CIs were reported along with hazard estimates. All statistical analyses were performed by using SAS, version 9.4 (SAS Institute).
RESULTS
The de novo iSCH study consisted of 6 patients. There were 3 men and 3 women with a mean age of 45 years (range, 25–63 years) at the onset of iSCH symptoms. Paraparesis, Brown-Séguard syndrome, and lower extremity paresthesias were the presenting symptoms of iSCH in 2 patients each. Using the Imagama classification, 3 patients had a protrusion type of iSCH and 3 patients had a kink type of iSCH.
Five patients had a history of SIH preceding the onset of iSCH by 2 to 22 years (mean, 9.6 years; standard deviation [SD]: 6.7 years). Spinal imaging at that time had demonstrated a ventral CSF leak but no iSCH (Fig 1). Treatment at that time included epidural blood patches in all 5 patients and percutaneous fibrin glue injections in 2 patients. One patient had upper neck pain preceding the onset of iSCH symptoms by 2 years and spinal imaging did not show iSCH (Fig 1). An intravertebral cavity associated with iSCH was present in 1 patient, and this cavity formed over 18 months (Online Supplemental Data). A calcified disk herniation or osteophyte was detected at the level of the iSCH in 4 patients (Online Supplemental Data). In 1 patient, the calcified disk herniation had absorbed by the time of diagnosis of the iSCH (Online Supplemental Data). At the time of iSCH symptom onset, 2 patients had persistent symptoms of SIH. We performed surgery to correct the iSCH in 4 patients, with improvement of symptoms in 3 and stabilization of symptoms in 1 patient. The surgery consisted of a laminectomy and a posterior transdural approach without the need for fusion (Online Supplemental Data). We did not detect any evidence of a duplicated dura or inflammation intraoperatively. Postoperative imaging showed resolution of the iSCH in all 4 patients (Online Supplemental Data).
Representative spine images from patients with de novo idiopathic spinal cord herniation. Initial T2-weighted sagittal spine MRIs (A, C, E, G, and I) and axial postmyelography CT (K) performed for symptoms of SIH or neck pain showing no evidence of iSCH and T2-weighted sagittal (B, D, F, H, and J) and axial (L) spine MRIs showing de novo formation of iSCH (arrows). Imaging findings of a CSF leak, ie, the presence of extradural CSF, were present in all patients except the 1 shown in panels E, F.
The patient population of the long-term follow-up study consisted of 51 patients with persistent ventral spinal CSF leaks. There were 33 women and 18 men with a mean age of 41.2 years (range, 19 to 76 years; SD: 11 years) at the onset of SIH symptoms. Follow-up ranged from 12 to 235 months (mean, 86.1 months; SD: 69.5 months) for a total of 359 patient years of follow-up. Two patients were lost to follow-up. Idiopathic SCH developed in 2 of the 51 patients (3.9%). The probability of developing iSCH increased from 0% at 5 years to 9.4% (95% CIs: 2.0–32.6%) at 10 years (Fig 2). The 2 patients with persistent ventral spinal CSF leaks who developed iSCH were also included in the case series of 6 patients.
Risk of idiopathic spinal cord herniation in patients with persistent spontaneous ventral spinal CSF leaks. Cumulative risk of development of iSCH in a cohort of 51 patients with persistent spontaneous ventral spinal CSF leaks.
DISCUSSION
In these patients, de novo formation of iSCH was clearly demonstrated on serial spinal imaging. Previously, patients with progressive degrees of iSCH have been reported but none with de novo iSCH.1,14,15 It has been suggested that formation of a ventral dural tear must be associated with major trauma,4 but the ventral dural defect we found in the 4 patients who underwent surgery and in all previously reported patients with iSCH have been similar to the ones we have observed intraoperatively in more than 400 patients with SIH due to a ventral spinal CSF leak.2 Another argument that has been used against an acquired process is the presence of a vertebral body cavity. Such cavities have been reported in several patients with iSCH.1 In 1 of our patients, de novo formation of the vertebral body cavity occurred within an 18-month period, demonstrating that these cavities are not congenital lesions. CSF pulsations at the level of the iSCH could explain the formation of such cavities over a brief period of time.3
The most likely pathogenesis for iSCH is a primary ventral dural defect as commonly seen in patients with SIH. Many patients with iSCH have a history of orthostatic headaches or evidence for a ventral CSF leak on spinal imaging.3 If the spinal cord completely tamponades the ventral dural defect, symptoms of a spinal CSF leak do not occur, as was seen in 1 of our patients. Then, over time, the spinal cord may herniate through the dural defect into the epidural space, with or without the formation of a cavity within the vertebral body. A calcified disk herniation or osteophyte is present at the level of the spontaneous ventral CSF leak in most patients and is believed to be at least partially responsible for the formation of the ventral dural tear.2,3 Calcified lesions were also found at the level of the iSCH in 4 of the 6 presently reported patients. In 1 of our patients, the calcified disk herniation had absorbed over time as has been reported previously in other patients.16,17
Successful treatment of ventral dural defects in patients presenting with spontaneous spinal CSF leaks will prevent the development of iSCH, and although the exact magnitude of the risk of developing this complication of an untreated ventral spinal CSF leak remains unknown, the current study estimates the risk to be about 10% at 10-year follow-up. Further studies are needed to corroborate our findings and refine the estimates of developing SCH in patients with persistent spontaneous ventral spinal CSF leaks.
CONCLUSIONS
Idiopathic SCH is an acquired condition, and early treatment of ventral CSF leaks offers a unique opportunity to prevent neurologic disability.
Footnotes
Disclosure forms provided by the authors are available with the full text and PDF of this article at www.ajnr.org.
References
- Received March 20, 2024.
- Accepted after revision April 24, 2024.
- © 2024 by American Journal of Neuroradiology
Jump to section
Related Articles
Cited By...
- No citing articles found.