The utility of magnetic resonance imaging of the spinal cord in paediatric transverse myelitis: A systematic review.
Chiraag Thakrar Karia, Paul Koroma, Solomon Nelson-Ashley et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review assessed spinal cord MRI findings, agreement with CSF inflammation, and prognostic associations in children with monophasic acute transverse myelitis. Twelve observational studies included 330 children. The pooled MRI diagnostic yield was 97.6% (95% CI 92.2 to 100.0%), although results varied substantially across studies (I²=71.8%; 95% prediction interval 71.2 to 99.6%). MRI and CSF findings were discordant in 36% of patients with paired data, and baseline MRI positivity did not reliably predict clinical outcome.
Key findings
- Pooled spinal cord MRI diagnostic yield was 97.6% (95% CI 92.2 to 100.0%), with substantial heterogeneity (I²=71.8%).
- Among 61 patients with paired data, MRI and CSF findings were discordant in 36%, most often with MRI-positive findings and normal CSF.
- Of 237 patients with follow-up data, 64% achieved good recovery and 30% had severe residual deficits; persistent bladder dysfunction occurred in 31%.
- Baseline MRI positivity did not reliably predict outcome.
Why it matters
The review supports MRI as a frequently positive investigation in children with clinically diagnosed acute transverse myelitis, while showing that a normal CSF result may not align with imaging. The wide prediction interval and study heterogeneity caution against treating the pooled yield as universal. MRI findings alone also appear insufficient for prognostic stratification.
What to keep in mind
MRI was required for cohort inclusion in most studies, potentially influencing the estimated diagnostic yield. The pooled result was heterogeneous, and paired MRI-CSF data were available for only 61 patients.