High-yield points
- Under 8 = high cervical spine: large head, lax ligaments, horizontal facets. Upper cervical and distraction injuries dominate; isolated upper injuries were 73% under 3 vs 29% over 8 .
- PECARN imaging rule:
- high-risk: altered mental status (GCS 3–8), abnormal ABCs, focal deficit;
- non-negligible: neck pain, GCS 9–14, substantial head or torso injury, midline tenderness;
- sensitivity 94%, NPV 99.9% ; works at every age .
- Don't over-call normal variants: C2–C3 pseudosubluxation (posterior cervical line intact), synchondroses, wedge-shaped bodies (see caveats).
- Odontoid synchondrosis fracture: under 7, anteriorly displaced, reduce and immobilize; 93% fuse .
- Type II dens fractures appear only after age 8 .
- Operate for: cord compression, significant deformity, dynamic instability, and adult-type injuries in older children. Most children in the series were treated without surgery, and all fused .
- Halo in a young child: many pins at low torque (see caveats).
Landmark papers
- Leonard et al., Lancet Child Adolesc Health 2024 (PECARN): a cervical spine imaging rule derived and validated in 22,430 children, with 94.3% sensitivity and 99.9% NPV. It could more than halve CT use .
- Beckmann et al., AJR 2020: the age-specific pattern of pediatric cervical injury. Isolated upper cervical injuries were 73% under 3 and 29% over 8, with distraction injuries in the youngest .
- Fassett et al., Neurosurg Focus 2006: odontoid synchondrosis fractures (55 children, mean age 2.8). Anterior displacement in 94%, and 93% fused with immobilization .
The child's cervical spine is a different organ from the adult's until about age 8 (see caveats). Young children have:
- a large head on a small neck;
- lax ligaments, shallow and horizontal facets, and wedge-shaped vertebral bodies;
- unfused synchondroses that look like fractures.
The fulcrum of motion sits high, at C1–C3. So young children injure the craniocervical junction and upper cervical spine, often by distraction, while adolescents fracture the subaxial spine like adults .
This chapter covers the fractures. For specific ligamentous injuries, see:
- atlanto-occipital dissociation;
- atlantoaxial rotatory subluxation;
- facet dislocations;
- spinal cord injury and SCIWORA.
Epidemiology
- National inpatient database (2000–2012) :
- cervical spine injury in 2.07% of pediatric trauma admissions; mortality 4.87%;
- transportation accidents caused 58%;
- upper cervical injury, SCI with fracture, SCIWORA and dislocation became less common with age;
- injury was more common in older children, but mortality was higher in younger ones.
- Pattern by age (235 children, level 1 trauma center) :
- isolated upper cervical injuries: 73% under age 3, 48% at ages 3–8, 29% over 8;
- most occipital condyle fractures (71%) and a quarter of dens fractures (26%) were avulsions;
- type II dens fractures occurred only over age 8; types I and III were almost all in children aged 8 or younger;
- subaxial injuries by AOSpine type: A 66%, B 17%, C 17%. Type B (distraction) injuries were about three times more common under age 3 than over 8.
- Unstable upper cervical injuries (28 children) :
- the youngest (0–2 years) had more neurological injury;
- injury patterns by age group:
- 0–2 years: C0–C1 or odontoid injuries in 33%;
- 3–8 years: C2 fracture or dislocation in 80%;
- 9–16 years: odontoid fractures in 60%;
- mortality 12%.
Who needs imaging: the PECARN rule
- Derivation and validation (22,430 children, 18 pediatric EDs; 1.9% had a cervical spine injury) :
- high-risk factors (12.7% risk if any present):
- altered mental status (GCS 3–8 or unresponsive on AVPU);
- abnormal airway, breathing or circulation;
- focal neurological deficit (paresthesia, numbness, weakness);
- non-negligible risk factors:
- neck pain;
- milder altered mental status (GCS 9–14);
- substantial head injury;
- substantial torso injury;
- midline neck tenderness;
- performance in validation: sensitivity 94.3%, specificity 60.4%, NPV 99.9%;
- applying it could cut CT use from 17.2% to 6.9% without more plain films.
- Across ages (secondary analysis): sensitivity 91.7–94.4% and NPV at least 99.6% in every age group. Age-specific rules were no better, so use the PECARN rule at all ages .
- Choice of imaging (see caveats):
- plain films first in most children with non-negligible risk;
- CT for high-risk children, or when films are inadequate;
- MRI for neurological deficits, suspected ligamentous injury, or an obtunded child who can't be cleared clinically.
Normal variants that mimic injury
- Pseudosubluxation of C2 on C3 (and C3 on C4) is normal under about age 8. The posterior cervical (Swischuk) line is drawn along the anterior cortex of the posterior arches of C1 and C3; if the C2 posterior arch lies 2 mm or more behind it, suspect a true injury such as a hangman's fracture (see caveats).
- Other normal variants (see caveats):
- unfused synchondroses: the dentocentral synchondrosis at the base of the dens, and the neurocentral synchondroses;
- the os terminale;
- prevertebral soft tissue that looks thick when the infant is crying or the neck is flexed;
- wedge-shaped vertebral bodies;
- an atlantodental interval of up to about 4–5 mm in young children.
Fracture types
Occipital condyle and C1
- Occipital condyle fractures are usually avulsions . Look for associated atlanto-occipital dissociation (see AOD).
- C1 ring fractures can occur through unfused synchondroses in young children. Most are treated in a collar or halo (see caveats).
Odontoid synchondrosis fractures
- The classic young-child fracture: the dens separates through the dentocentral synchondrosis, usually tilting forward .
- Combined series (55 children, mean age 2.8 years, range 9 months to 7 years) :
- diagnosis was delayed in 8;
- anterior displacement in 94%;
- spinal cord injury in 15, including 11 complete, and 8 at the cervicothoracic junction rather than at the dens;
- 93% fused with external immobilization (collar, Minerva or halo);
- surgery in 8, for failed immobilization, nonunion or the treating surgeon's choice.
- Suspect it in any child under 7 with neck pain or a cord injury after trauma .
- Treatment: reduce with gentle extension, then immobilize (halo or Minerva). The synchondrosis heals reliably (see caveats).
C2 (hangman's) and other upper cervical fractures
- Traumatic spondylolisthesis of C2 occurs in young children and must be told apart from persistent synchondroses and pseudosubluxation. Most heal in a halo or collar (see caveats).
- Older children get adult-type dens fractures; type II occurred only over age 8 . These may need fixation, as in adults (see caveats).
Subaxial fractures (C3–C7)
- Adolescents have adult patterns: compression and burst fractures, facet fractures and dislocations, and teardrop fractures . Classify with AOSpine (see facet dislocations).
- Young children have more distraction (type B) injuries through the soft endplates and ligaments .
- Endplate (apophyseal) separations in young children can be subtle on CT. MRI shows them (see caveats).
Treatment principles
- Who needs surgery (unstable upper cervical series) :
- signs of cord compression;
- significant deformity;
- dynamic instability;
- age over 8 (adult-like spine).
- Most children in this series (21 of 28, ages 0–16) were treated without surgery, and all achieved solid fusion .
- Instability criteria differ in children, because the spine is still growing .
- Immobilization (see caveats):
- rigid collar for stable injuries;
- halo for unstable injuries in children; in young children, use more pins (often 8–12) at lower torque because the skull is thin;
- Minerva braces or casts are an alternative in the youngest;
- transport young children on a board with a cutout or a pad under the torso, because the large occiput flexes the neck.
- Surgery (see caveats):
- occipitocervical or C1–C2 fusion for craniocervical and atlantoaxial instability;
- screws are possible in most children; autograft fuses well;
- limit fusion length, since every fused segment stops growing;
- watch for adjacent-segment fusion and junctional kyphosis.
Related topics
- Atlanto-occipital Dissociation (AOD): craniocervical distraction injury.
- Atlantoaxial (C1–2) Rotatory Subluxation: torticollis after minor trauma or infection.
- Subaxial Cervical Facet Dislocations (With and Without Fracture): adolescent subaxial injuries.
- Spinal Cord Injury in Children: Classification, Management and Prognosis: SCIWORA and cord injury.
- Thoracic and Thoracolumbar Spine Fractures: the rest of the spine.
- PECARN Head CT Rule (Interactive): the companion head-imaging rule.
- Sacral and Pelvic Fractures in Children: What the Neurosurgeon Needs to Know: the caudal spine.
- Abusive Head Trauma: spine imaging in suspected abuse.
References
- Beckmann NM, Chinapuvvula NR, Zhang X, West OC. Epidemiology and imaging classification of pediatric cervical spine injuries: 12-year experience at a level 1 trauma center. AJR Am J Roentgenol 2020;214(6):1359–1368. doi:10.2214/AJR.19.22095. PMID 32228329. verified
- Duhem R, Tonnelle V, Vinchon M, et al. Unstable upper pediatric cervical spine injuries: report of 28 cases and review of the literature. Childs Nerv Syst 2008;24(3):343–348. doi:10.1007/s00381-007-0497-0. PMID 17912535. verified
- Shin JI, Lee NJ, Cho SK. Pediatric cervical spine and spinal cord injury: a national database study. Spine (Phila Pa 1976) 2016;41(4):283–292. doi:10.1097/BRS.0000000000001176. PMID 26555836. verified
- Leonard JC, Harding M, Cook LJ, et al. PECARN prediction rule for cervical spine imaging of children presenting to the emergency department with blunt trauma: a multicentre prospective observational study. Lancet Child Adolesc Health 2024;8(7):482–490. doi:10.1016/S2352-4642(24)00104-400104-4). PMID 38843852. verified
- Corwin DJ, Chaudhari PP, Tzimenatos L, et al. Comparison of cervical spine injury prediction rule across ages. Pediatrics 2026;157(6):e2025075341. doi:10.1542/peds.2025-075341. PMID 42061904. verified
- Fassett DR, McCall T, Brockmeyer DL. Odontoid synchondrosis fractures in children. Neurosurg Focus 2006;20(2):E7. PMID 16512658. verified
Verification (2026-11-19):
- Each reference's authors, title, journal, year, volume, pages, DOI and PMID were checked on PubMed. Reference 6 has no DOI on PubMed. Reference 5's article number is inferred from its DOI.
- Numbers were checked against the PubMed abstracts. Percentages are rounded (57.51 → 58%, 65.6 → 66%, 17.2 → 17%).
Caveats:
- The following are standard teaching, not drawn from the cited abstracts:
- the anatomical differences in children and the age-8 transition;
- imaging choices after the PECARN rule;
- the normal variants, including the posterior cervical line and ADI values;
- C1 ring fractures through synchondroses;
- reduction technique for synchondrosis fractures;
- C2 traumatic spondylolisthesis management and adult-type dens fixation;
- endplate separations;
- the immobilization and halo details (pin number and torque), transport positioning, and surgical principles.
- Reference 1 is a single trauma center; reference 2 is a small series; reference 6 pools published cases.