Spinal cord–spinal canal disproportion following treatment for syringomyelia associated with Chiari malformation type I: illustrative case Authors: Arjun Nair, Vivek P Gupta, Devin Kolmetzky, Jennifer M Strahle, David D Limbrick Published: 2024 ABSTRACT AbstractBACKGROUNDChiari malformation type I (CM-I) is a common pediatric neurosurgical condition that is often associated with syringomyelia (SM) and spine deformity. The association of these three conditions is well recognized, but the pathophysiology linking them has yet to be fully elucidated.OBSERVATIONSThis case report describes the unusual course of a 13-year-old male with CM-I, a large holocord syrinx, and progressive scoliosis who developed an angiography-negative subarachnoid hemorrhage (SAH) 4 months after successful posterior fossa decompression with duraplasty (PFDD). Notably, his cervical spinal canal diameter was increased relative to normative data. After presenting with SAH, he was treated with a course of lumbar cerebrospinal fluid drainage, which relieved his symptoms, and he had no further incidents. His syrinx responded to PFDD with a dramatic decrease in spinal cord diameter, in contradistinction to his dilated bony spinal canal. He is presently 4 years out from his SAH and doing well clinically.LESSONSThis report describes the first case of nonaneurysmal SAH following PFDD for CM-I and SM and explores a possible link between SM and bony spinal canal diameter. https://thejns.org/doi/10.3171/CASE24272 Keywords: Chiari malformation type I, spinal canal dilation, syringomyeliaABBREVIATIONS: CM-I = Chiari malformation type I, CSF = cerebrospinal fluid, CT = computed tomography, MRI = magnetic resonance imaging, pB-C2 = line drawn between the basion and the posterior aspect of the C2 vertebral body, PFDD = posterior fossa decompression with duraplasty, PRSRC = Park-Reeves Syringomyelia Research Consortium, SM = syringomyelia. Chiari malformation type I (CM-I) is characterized by cerebellar tonsillar ectopia through the foramen magnum.1, 2 The exact prevalence of CM-I is challenging to estimate. As many as 3.6% of the population may meet radiographic criteria, but only approximately one-third of patients are symptomatic.1, 3 Clinical presentation is variable and can include occipital/suboccipital headaches, cerebellar or bulbar dysfunction, and numerous other signs and symptoms.2 One of the conditions most commonly associated with CM-I is syringomyelia (SM), which has a wide range of reported incidence in the setting of CM-I, from 25% to as much as 85% in some studies.1, 3, 4 Similarly, scoliosis is commonly seen in patients with both CM-I and SM. Importantly, the relationship between CM-I and scoliosis in patients with SM appears to be primarily mediated by the syrinx itself, as there is not nearly as strong an association between nonsyndromic CM-I and scoliosis in patients without SM.5, 6 The specific factors in patients with CM-I and SM that are more strongly associated with scoliosis have also been studied. Strahle et al. found that larger syrinxes, syrinxes that extended into the cervical cord, and holocord syrinxes were most strongly associated with scoliosis, similar to a previous study that also associated larger maximum syrinx diameters with an increased risk of scoliosis.7, 8 Many patients with CM-I, SM, and scoliosis have stable curves, although some progress without treatment.9 The treatment of patients with CM-I, SM, and scoliosis can take multiple forms depending on the patient’s age and the magnitude of the curvature, though posterior fossa decompression is often the first surgical option considered. Specifically, patients younger than 10 years of age with all three of these conditions with curves less than 35° in magnitude have the greatest chance of curve improvement after posterior fossa decompression.10, 11 The association between SM and scoliosis in these patients is further supported by studies demonstrating an increased risk of curve progression if the syrinx fails to resolve after posterior fossa decompression, though this was not confirmed in a more recent study.10, 12 In this case report, we detail the course of a patient with CM-I, SM, and progressive scoliosis, with particular attention to spinal canal diameter and its possible role in postoperative complications. Illustrative Case A 13-year-old male being monitored by the orthopedics service for thoracic dextroscoliosis underwent total neuraxis magnetic resonance imaging (MRI) after the progression of his curve was observed on serial standing scoliosis radiographs. His scoliosis was initially noted during a physical examination and confirmed by standing radiographs at 11 years old; his initial radiographic Cobb angle was 17°, which increased to 30° 2 years later at the time of initial neurosurgical evaluation at 13 years old (Fig. 1). At the neurosurgical consultation, he did not report any symptoms classically associated with Chiari or SM; specifically, he did not have headaches, nausea or vomiting, ocular abnormalities, cerebellar or bulbar symptoms, or sensorimotor symptoms, though he did report mild, subjective distal lower-extremity weakness.2, 13 His neurological examination was normal, though he was noted to have subtle toe curling without any detectable lower-extremity weakness. His reflexes were normal. There was no history or findings consistent with other disorders, such as a connective tissue disorder or hypermobility. Standing scoliosis radiographs obtained at age 11 (A) and age 13 (B) years, showing the progression of thoracic dextroscoliosis from 17° to 30°. His initial MRI demonstrated 15 mm of cerebellar tonsillar ectopia, a holocord syrinx (from the cervicomedullary junction to T12, just above the conus medullaris) with a maximal diameter of 19 mm, and significant spinal cord dilation; the spinal cord is nearly indistinguishable from the syrinx at most levels (Fig. 2). Notably, the patient’s bony spinal canal diameter was 26.2 mm at C2 and 21.1 mm at C7. Normative morphometric data by Johnson et al. reported an age-appropriate mean diameter of 17.41 ± 2.07 mm at C2 and 15.10 ± 1.36 mm at C7.14 Retroflexion of the dens at C2 was also observed, with a pB-C2 (line drawn between the basion and the posterior aspect of the C2 vertebral body) of 8.6 mm and a clivoaxial angle of 127°. Sagittal T1-weighted (A) and T2-weighted (B and C) images of the brain and spine. The T1-weighted MRI demonstrates the CM-I with 15 mm of cerebellar tonsillar ectopia and retroflexion of the dens with a pB-C2 of 8.6 mm and a clivoaxial angle of 127°. The total spine T2-weighted MRI demonstrates a holocord syrinx with a maximal spinal cord diameter of 19 mm. The patient’s spinal canal diameter was 26.2 mm at C2 and 21.1 mm at C7, compared to normative data, per Johnson et al.,14 of 17.41 ± 2.07 mm at C2 and 15.10 ± 1.36 mm at C7. This patient underwent elective posterior fossa decompression with duraplasty. During the procedure, the posterior craniovertebral junction was exposed, and a suboccipital craniectomy and C1 laminectomy were performed. The dura was opened in a Y-shaped fashion and tacked up to the surrounding tissue. Under the operating microscope, the arachnoid mater overlying the cerebellar tonsils and fourth ventricle was dissected, but no tonsillar cauterization was performed; no other abnormal intraoperative findings were noted. A pericranial autograft was used for duraplasty. He had an unremarkable postoperative course other than a mild allergic reaction to an analgesic. He worked well with physical therapy and was discharged on postoperative day 4. He continued to do well at his 2-week postoperative visit, with no pain or new neurological concerns and a well-healing surgical incision. Approximately 4 months after surgery, the patient developed an acute onset of neck pain and headache while swimming competitively. Conservative management at home was attempted, but his symptoms worsened and he was brought to the emergency department. MRI of the brain and spine, with and without contrast, demonstrated concern for a subarachnoid hemorrhage anterior to the cervical spine, without any overt intracranial findings. MRI also showed a dramatic reduction in syrinx and spinal cord size, although there was a persistently dilated bony spinal canal. There was no MRI evidence of pseudomeningocele, and ventricles were of normal size and caliber (Fig. 3A and B). Axial brain T2-weighted MRI (A), sagittal cervicothoracic spine T2-weighted MRI (B), and sagittal head and neck CT angiography (C) from the hospital follow-up 4 months after surgery, demonstrating normal ventricular caliber and significantly decreased spinal cord size with persistent dilation of the bony spinal canal. The T2 hyperintensity on cervical spine MRI ventral to the spinal cord, corresponding to an area of hyperdensity on CT, was ultimately thought to indicate hemorrhage, corresponding to hemorrhage noted in cerebrospinal fluid obtained via lumbar puncture. No vascular anomaly was identified on CT angiography or subsequent cerebral angiography. His symptoms continued to worsen at the hospital, and he developed a left sixth cranial nerve palsy. He underwent a lumbar puncture, which demonstrated nonclearing blood and an elevated opening pressure of 46 cm H2O. Given the elevated opening pressure, 28 mL of cerebrospinal fluid (CSF) was removed, and a closing pressure of 15 cm H2O was recorded. His headache significantly improved after the procedure. Initial CSF analysis demonstrated red (not xanthochromic) fluid, with a CSF glucose of 37 mg/dL, CSF protein of 43.3 mg/dL, and negative CSF Gram stain and culture results. Total CSF cells increased from 33,000 in tube 1 to 47,000 in tube 4. Other CSF studies, including oligoclonal banding, enterovirus polymerase chain reaction, immunoglobulin G, cryptococcal antigen, viral panels, and mycology, were all negative. His clinical picture was overall concerning for subarachnoid hemorrhage and occult hydrocephalus. A lumbar drain was subsequently placed and drained at 10 mL per hour. With improvement in his sixth nerve palsy after treating his elevated intracranial pressure, ophthalmological consultation was not pursued. During his hospitalization, computed tomography (CT) angiography of the head and neck was performed, demonstrating no evidence of vascular anomalies but revealing a mixed attenuation–dependent collection anterior to the spinal cord from C4 to T1, which corresponded to an indeterminate signal anomaly in the same location on previous cervical spine MRI (Fig. 3B and C). The combination of these findings was favored to represent subarachnoid hemorrhage, which was corroborated by the lumbar puncture findings. Cerebral digital subtraction angiography of the cranial and cervical bilateral carotid and bilateral vertebral arteries was then performed and demonstrated no abnormal vascular anomalies, including no evidence of ruptured intracranial aneurysm, fistula, or arteriovenous malformation. This study also included bilateral subclavian artery runs and a high-volume aortic arch run, which demonstrated no evidence of anomalies in the vascular supply to the cervical spine. Lumbar drainage was continued for 1 week and then weaned over the next 2 days with an eventual clamp trial. Opening pressure was measured twice during the clamp trial and was found both times to be 20 cm H2O. A follow-up CSF profile after 4 days demonstrated xanthochromia, a CSF glucose of 62 mg/dL, a CSF protein of 25.7 mg/dL, 486 total cells, and negative Gram stain and culture. The patient continued to improve clinically, with reduced headache severity and nearly resolved left abducens nerve palsy. The lumbar drain was removed after a total of 8 days of drainage, and he was discharged home the next day. A short-interval total spine MRI scan was obtained 1 month after discharge, showing a further slight decrease in the caliber of the patient’s holocord syrinx. The patient did well for 6 months but then was briefly readmitted for intermittent headaches. MRI of the brain and total spine at that time demonstrated unchanged typical ventricular configuration and stability of the syrinx. He was discharged home after overnight observation and symptomatic improvement, including improvement of the abducens nerve palsy. Sixteen months after his procedure, he developed subcutaneous cysts along his incision that required local incision and drainage by his local dermatologist, with subsequent resolution; he had no other symptoms to accompany these lesions. Scoliosis radiographs at that time demonstrated an improvement in his thoracic scoliosis to 16° (Fig. 4A). Anteroposterior standing scoliosis radiograph 16 months postprocedure, demonstrating improvement in the thoracic curve to 16° from 30° (A). Sagittal T2-weighted MRI of the cervicothoracic spine from the 4-year follow-up, demonstrating a significant reduction in spinal cord diameter, with persistent dilation of the bony spinal canal (B). Clinically, he was doing well at his follow-up 2 years after surgery. He was noted to have new left foot numbness without any imaging correlate, which has remained stable at all subsequent follow-ups. MRI of the brain and total spine at this time demonstrated a further decrease in syrinx size, and concurrent brain magnetic resonance angiography without contrast was negative. As of his most recent follow-up at 17 years old, more than 4 years after his original surgery, his syrinx had remained nearly completely resolved, with a maximal anteroposterior diameter of < 2 mm, though it continued to have a transverse diameter of 10 mm from C3 to C6. However, his increased cervical bony spinal canal diameter persisted (Fig. 4B). He was referred to genetics for consultation regarding testing for a connective tissue disorder but elected not to undergo evaluation. Informed Consent The necessary informed consent was obtained in this study. Discussion Observations Our case report describes the unique course of a patient with CM-I, holocord SM, and scoliosis whose postoperative course was complicated by subarachnoid hemorrhage and radiologically occult hydrocephalus 4 months after surgery. This case highlights a possible relationship between SM and spinal canal dilation. Furthermore, it poses questions regarding the mechanistic implications of syrinx regression within the spinal canal. In general, SM in the setting of CM-I responds favorably to posterior fossa decompression. Multiple studies have reported improvement in syrinx dimensions in more than 80% of patients after posterior fossa decompression.15, 16 More recent studies reproduced this result. About 38% of patients experienced complete syrinx resolution after posterior fossa decompression with or without duraplasty over a mean follow-up of 1.5 years in a cohort of 380 patients from the Park-Reeves Syringomyelia Research Consortium (PRSRC) registry with CM-I and SM.17 The same study concluded that performing a duraplasty was positively associated with syrinx resolution and that patients who underwent duraplasty had a faster time to syrinx resolution. Another study of 692 patients from the PRSRC registry similarly demonstrated that the incorporation of duraplasty improves both syrinx size and length compared to extradural-only decompression.18 As mentioned previously, in patients with CM-I, SM, and scoliosis, syrinx size and distribution may be associated with the development of scoliosis.7 In some studies, a reduction in syrinx size after posterior fossa decompression has been associated with improvement in scoliosis, though this finding is not universal.10, 12 In both of these studies, 30%–40% of patients who underwent posterior fossa decompression had improvement in their curves, whereas 40%–50% had progression. This compares favorably with a study in which no patients with CM-I, SM, and scoliosis had improvement in their curves without surgical intervention.9 Fortunately, this paradigm held true for our patient, who experienced near-resolution of his syrinx after PFDD and improvement in his thoracic scoliotic curve. However, his bony spinal canal diameter, which was markedly increased compared to comparative norms, persisted despite his other improvements.14 Additionally, he developed subarachnoid hemorrhage and then hydrocephalus during watersport activity that had no identifiable vascular etiological origin despite multiple investigative clinical studies. Lessons This case illustrates a potential vascular consequence of the correction of spinal cord dilation in the setting of CM-I and SM with bony canal dilation. It was hypothesized that this hemorrhage was a result of an avulsion of a dilated extramedullary spinal blood vessel that was placed under tension as the spinal cord diameter decreased, thereby leaving it susceptible to damage and breakage during activity. Although the anatomical compartments are different, this could be thought of as mechanistically similar to an avulsion of a bridging vein that can cause a cranial subdural hematoma in a patient with trauma after normal cerebral volume loss. Another possibility is that he had a concurrent undiagnosed connective tissue disorder such as vascular-type Ehlers-Danlos syndrome, which resulted in alterations in the bony spinal architecture as well as further decreased strength of the patient’s blood vessels. The presence of subcutaneous cysts along the incision 6 months postoperation could add further credence to this idea. He did not display common features of Ehlers-Danlos syndrome, such as hyperflexibility, skin hyperextensibility, and muscle weakness, though patient presentations can vary widely.19, 20 He was referred for genetic testing for this reason after a multidisciplinary discussion, but his family opted not to pursue this evaluation. This hypothesis is strengthened by the comorbid association of CM-I with other disorders, including connective tissue disorders and neurofibromatosis type 1, and the fact that these patients also tended to have a higher incidence of basilar impression.4, 21, 22 One of the proposed mechanisms behind vertebral body changes in connective tissue disorders is dural ectasia. Dural ectasia can occur secondary to CSF pulsatility and cause morphological changes in the connective tissue of the dura. This subsequently results in direct pressure on the vertebral body and is associated with scalloping, cortical thinning, and foraminal widening.23 Such a mechanism could help explain the atypical appearance of this patient’s spinal column. Intradural hemorrhage after posterior fossa decompression is rare after the perioperative period. One case report describes intrasyringal hemorrhage in a patient with CM-I, SM, and Noonan’s syndrome who had previously undergone posterior fossa decompression and syrinx-subarachnoid shunt. The patient had neurological deterioration postsurgery, which prompted evaluation and identification of the hemorrhage. The patient responded well to surgical treatment of the hematoma, with a significant reduction in syrinx size on follow-up imaging.24 Notably, the hemorrhage was intramedullary in this study, while it was extramedullary in our patient. Overall, our patient’s syrinx and scoliotic curve responded well, but the subsequent subarachnoid hemorrhage and persistent bony spinal canal distention raise questions regarding spinal cord expansion/decompression and its relationship to vascular structures. The delayed occurrence of the hemorrhage highlights the need for continued investigation into the interplay between CM-I and SM, as well as consideration of potential comorbid conditions such as connective tissue disorders. Acknowledgments This work was supported by a grant from the National Institute of Neurological Disorders and Stroke, National Institutes of Health, US Department of Health and Human Services (P01NS13113, D.D.L.). Disclosures Dr. Limbrick reported being the Chief Medical Officer for Rhaeos, Inc. outside the submitted work. Author Contributions Conception and design: Nair, Limbrick. Acquisition of data: Kolmetzky, Strahle. Analysis and interpretation of data: Nair, Gupta, Kolmetzky, Limbrick. Drafting the article: Nair, Gupta, Strahle, Limbrick. Critically revising the article: all authors. Reviewed submitted version of manuscript: Nair, Gupta, Strahle, Limbrick. Approved the final version of the manuscript on behalf of all authors: Nair. Administrative/technical/material support: Limbrick. Study supervision: Limbrick. Correspondence Arjun Nair: Washington University in Saint Louis School of Medicine, Saint Louis, MO. a.m.nair@wustl.edu. References