High-yield points
- Hydromyelia: dilated central canal, ependyma-lined, central (sometimes communicating). Syringomyelia: cavity dissecting into the parenchyma, glia-lined, often eccentric. MRI often can't tell them apart (syringohydromyelia) (see caveats).
- A slit-like central canal is a normal variant, and the ventriculus terminalis is usually incidental (see caveats). The terminology is used inconsistently .
- Find the cause: Chiari I (most common) , tether, tumor, arachnoiditis, trauma, MMC shunt failure. Image the whole neuraxis, with contrast.
- Presyrinx: cord swelling without a cavity, reversible when flow is restored .
- Idiopathic syrinx: observe. 91% stable or better ; repeat imaging unnecessary when stable .
- Chiari with syrinx: decompress. Duraplasty (randomized): complications not significantly different; more syrinx reduction and fewer redo operations as secondary outcomes. Larger trials are needed .
Landmark papers
- Limbrick et al., N Engl J Med 2026: a cluster-randomized trial of decompression with vs without duraplasty for Chiari I with syrinx. Complications were not significantly different (14% vs 6%; P = .11); as secondary outcomes, duraplasty showed more syrinx reduction (3.1 vs 1.2 mm) and fewer repeat decompressions (3% vs 14%) .
- Fischbein et al., AJNR 1999: the presyrinx state, a reversible cord swelling from CSF obstruction that can precede a syrinx .
- Magge et al., J Neurosurg Pediatr 2011: idiopathic syrinx in children is benign. 91% were stable or improved, and 87.5% of syringes were stable or smaller .
A syrinx is a fluid-filled cavity in the spinal cord. In children it is usually found on MRI done for Chiari I malformation, scoliosis, pain or a skin marker, and sometimes by chance. The neurosurgeon's job is to:
- decide whether it is a true syrinx or a normal variant;
- find the cause, which is almost always a CSF-flow problem upstream or a lesion in the canal;
- treat the cause, not the cavity.
Hydromyelia vs syringomyelia
- The terms are used inconsistently. Epidemiology is hard to pin down because hydromyelia, persistent or dilated central canal, slit-like cavities, presyrinx and true syringomyelia are classified differently from study to study .
- Classical definitions (see caveats):
- hydromyelia: dilatation of the central canal, lined by ependyma, usually central on axial images and often communicating with the fourth ventricle;
- syringomyelia: a cavity that dissects into the cord parenchyma, often eccentric, lined by glia rather than ependyma, and often non-communicating.
- On MRI the two often can't be told apart, which is why many use "syringohydromyelia" or simply "syrinx" (see caveats).
- Practical distinctions that matter (see caveats):
- slit-like or persistent central canal: a thin, central, smooth cavity, a normal variant without symptoms or progression;
- ventriculus terminalis: dilatation of the central canal in the conus, usually incidental;
- true syrinx: wider, long, often beaded (haustrated) with internal septa, expanding the cord, or eccentric.
- Working definition in a pediatric idiopathic series: a central canal of 1 mm or more across, extending over at least 2 vertebral levels .
- The presyrinx state: cord swelling with T2 signal change but no cavity, from obstructed CSF flow. It reversed when the obstruction was relieved in all 5 patients described, and in one with continued obstruction it later became a syrinx . Don't mistake it for a tumor or myelitis.
Causes
- Chiari I malformation is the most common associated condition (see Chiari I).
- Other causes (see caveats):
- tethered cord and terminal syringes (see fatty filum, lipomyelomeningocele);
- intramedullary tumors (see spinal cord gliomas);
- arachnoiditis or arachnoid webs after infection, hemorrhage or surgery;
- trauma (post-traumatic syrinx; see spinal cord injury);
- myelomeningocele, where hydromyelia is common and a growing syrinx can signal shunt failure (see postnatal MMC repair).
- "Idiopathic" cases may actually reflect occult CSF obstruction, arachnoid adhesions, "Chiari 0" or canal abnormalities .
- Pathophysiology: obstruction of CSF flow, most often at the foramen magnum, drives fluid into the cord (see caveats).
Workup
- MRI of the whole neuraxis, with contrast to exclude tumor (see caveats).
- Look for:
- tonsillar descent and a crowded foramen magnum;
- a low conus or thick filum;
- an enhancing mass;
- arachnoid webs (the "scalpel sign").
- CSF-flow (cine) MRI can show obstruction at the foramen magnum or in the canal (see caveats).
- Examine for:
- dissociated sensory loss (pain and temperature);
- asymmetric abdominal reflexes;
- hand weakness;
- scoliosis, especially left thoracic or rapidly progressive curves (see caveats).
Management
Idiopathic syrinx: watch
- Two-center series (48 children; mean size 4 mm; mean follow-up 24 months) :
- 91% were asymptomatic, stable or improved;
- on imaging (32 with follow-up imaging), 87.5% stayed stable or shrank;
- size changes didn't track symptoms.
- Long-term follow-up into adulthood (28 children; median imaging follow-up 6.8 years) :
- 54% shrank (by 1 mm or more);
- all stayed asymptomatic, stable or improved;
- there were no predictors of radiological or clinical change;
- repeat imaging seems unnecessary in stable cases.
Chiari I with syrinx: decompress
- Treat the Chiari, not the syrinx. Posterior fossa decompression restores CSF flow, and the syrinx usually shrinks (see caveats; see Chiari I).
- PRSRC registry (380 children with Chiari I and syrinx) :
- younger age and duraplasty were independently associated with syrinx resolution;
- syrinx resolution went with improvement in headache and neck pain.
- Randomized trial: decompression with vs without duraplasty (162 participants aged ≤21, tonsils ≥5 mm, syrinx 3.0–9.9 mm, cluster-randomized across 38 centers) :
- complications at 6 months: 14% with duraplasty vs 6% without (adjusted OR 2.59, P = .11; not significant);
- at 24 months (secondary outcomes, without significance tests in the abstract):
- clinical improvement 58% vs 46%;
- syrinx reduction 3.1 vs 1.2 mm;
- repeat decompression 3% vs 14%;
- quality of life was similar;
- larger trials are needed to settle the balance of risks and benefits.
Other causes
- Treat the cause (see caveats):
- untether the cord;
- resect a tumor;
- lyse an arachnoid web or adhesions;
- revise a failed shunt in MMC.
- Direct syrinx drainage (syringosubarachnoid, syringopleural or syringoperitoneal shunt) is a last resort. Shunts block often, and the cause remains (see caveats).
Scoliosis
- A syrinx, not the Chiari malformation itself, is independently associated with scoliosis (OR 9.08 vs 0.99) (see Chiari I).
- In the PRSRC registry, scoliosis was associated with syrinx diameter regression and stability after decompression .
- Decompress before spinal fusion when a significant syrinx is present (see caveats).
Related topics
- Chiari I Malformation: the most common cause.
- The Fatty Filum: terminal syrinx with tethering.
- Lipomyelomeningocele and Conus Lipomas: tethering lesions with syrinx.
- Spinal Cord Gliomas in Children (Low- and High-Grade): tumor-associated syrinx.
- Postnatal Myelomeningocele Repair: hydromyelia and shunt failure in MMC.
- Spinal Cord Injury in Children: Classification, Management and Prognosis: post-traumatic syrinx.
- Myelocystocele: terminal syringocele.
- Spinal Ependymoma in Children: tumor-associated syrinx.
- Re-tethering in Children: syrinx as a sign of retethering.
- Degenerative Spinal Disease in Children and Adolescents: back pain, disc herniation and spondylolysis.
- Tethered Cord Release: step-by-step procedure.
- Chiari I Decompression with Duraplasty: step-by-step procedure.
References
- Bogdanov EI, Heiss JD. The epidemiology of syringomyelia and associated conditions: a literature review. J Clin Med 2026;15(21):8653. doi:10.3390/jcm15218653. PMID 43030795. verified
- Magge SN, Smyth MD, Governale LS, et al. Idiopathic syrinx in the pediatric population: a combined center experience. J Neurosurg Pediatr 2011;7(1):30–36. doi:10.3171/2010.10.PEDS1057. PMID 21194284. verified
- Fischbein NJ, Dillon WP, Cobbs C, Weinstein PR. The "presyrinx" state: a reversible myelopathic condition that may precede syringomyelia. AJNR Am J Neuroradiol 1999;20(1):7–20. PMID 9974051. verified
- Sun R, Punyawai P, Furtado N, et al. Paediatric idiopathic syringomyelia: a follow-up of radiological and clinical outcomes into adulthood. Childs Nerv Syst 2025;41(1):236. doi:10.1007/s00381-025-06885-3. PMID 40670849. verified
- Hale AT, Adelson PD, Albert GW, et al. Factors associated with syrinx size in pediatric patients treated for Chiari malformation type I and syringomyelia: a study from the Park-Reeves Syringomyelia Research Consortium. J Neurosurg Pediatr 2020;25(6):629–639. doi:10.3171/2020.1.PEDS19493. PMID 32114543. verified
- Limbrick DD Jr, Shannon CN, Bayman EO, et al. Decompression with or without duraplasty for Chiari I and syringomyelia. N Engl J Med 2026;394(20):2015–2025. doi:10.1056/NEJMoa2402821. PMID 42202320. verified
- Strahle J, Smith BW, Martinez M, et al. The association between Chiari malformation type I, spinal syrinx, and scoliosis. J Neurosurg Pediatr 2015;15(6):607–611. doi:10.3171/2014.11.PEDS14135. PMID 26030330. verified
Verification (2026-11-19):
- Each reference's authors, title, journal, year, volume, pages, DOI and PMID were checked on PubMed. Reference 3 has no DOI on PubMed.
- Numbers were checked against the PubMed abstracts. Syrinx reductions are rounded (3.08 → 3.1 mm, 1.22 → 1.2 mm).
Caveats:
- The following are standard teaching, not drawn from the cited abstracts:
- the classical histological and MRI definitions of hydromyelia and syringomyelia, and "syringohydromyelia";
- slit-like canals and ventriculus terminalis as normal variants, and the features of a true syrinx;
- causes other than Chiari I, and the CSF-flow pathophysiology;
- the workup (contrast MRI of the neuraxis, cine MRI, the scalpel sign) and examination findings, including curve patterns;
- treating non-Chiari causes, and direct syrinx shunting as a last resort;
- syrinx shrinkage after decompression in general;
- decompression before spinal fusion.
- Reference 3 describes 5 patients; it is not a pediatric series.