Retro-odontoid mass resolution analysis and timing following posterior cervical spinal fixation: 16-year paediatric neurosurgery experience in a single UK institute.

A retro-odontoid mass (R-OM) is a soft tissue mass that develops posterior to the odontoid process of the C2 vertebra. The instability leads to non-physiologic motion producing an inflammatory process with fibro cartilaginous mass that lead to spinal cord compression. Cervical fusion has been previously shown to reduce it in rheumatoid pannus.

Aims: (1) Investigate impact of cervical fixation on R-OM in a paediatrics, (2) assess which group have the greatest reduction, and (3) assess the timing of reduction.

Methods: Between 2005 and 2021, 52 children underwent craniocervical junction (CCJ) fixation. Of these, 35 (67%) children had measurable R-OM at the time of presentation. These children underwent either occipito-cervical or atlanto-axial fixation for atlantoaxial instability. All cases were treated in a single tertiary paediatric neurosurgical centre by a multidisciplinary team. All procedures were performed by the same neurosurgical team. Demographic data, including age, gender, diagnosis, type of surgery, and measurements of pre- and post-operative R-OM were recorded. Serial post-operative scans provided evidence of regression of R-OM. R-OM measurements were performed using T2 W mid-sagittal and axial views of MRI to calculate anterio-posterior (AP), craniocaudal (CC), and mediolateral/width (LL) dimensions.

Results: Twenty-four children underwent atlantoaxial (69%), and 11 children had occipito-cervical fixation (31%). The mean age was 8.9 years (range 2-18) with M:F ratio of 1:1.3. Metabolic causes were the most common group undergoing surgery, with MPS IV being the most prevalent subgroup. Pooled evaluation of all cases revealed significant reduction in R-OM following fixation in all measured dimensions. Subgroup analysis of underlying pathology revealed that the metabolic group showed the most significant reduction. Analysis of timing of regression of the R-OM revealed that the maximal reduction occurred at the mean of 3 years (35.54 months) following fixation with the range of 2 weeks to 10 years (0.13 to 120.77 months).

Conclusion: In our experience, R-OM is a common feature in atlantoaxial instability in the paediatric population. Cervical fixation leads to regression in R-OM, supporting instability as the driving force for formation of R-OM. Hence, resection of R-OM directly is not recommended and could expose the child to unnecessary risk.