Degenerative cervical myelopathy is a demyelinating disease too

Design

Systematic review and random-effects meta-analysis of autopsy, histopathology and imaging studies, PRISMA-followed

Population

27 studies covering 364 patients with degenerative cervical myelopathy, including 16 autopsy studies with 85 patients

Primary outcome

Prevalence of demyelination and its relation to clinical measures

Effect

Pooled autopsy proportion 75.2% (95% CI 16.8-97.6, exploratory); magnetization transfer ratio reduced most in the ventral column (pooled mean difference -5.88, 95% CI -9.00 to -2.77, p = 0.0002); pooled correlation with baseline neurological function r = 0.38 (0.14-0.57)

Degenerative cervical myelopathy is the commonest cause of spinal cord dysfunction in adults, and is usually conceived mechanically: compression causes injury. This meta-analysis of 27 studies covering 364 patients asks how much of the injury is demyelination.

Across 16 autopsy studies, comprising only 85 patients, the pooled proportion with demyelination was 75.2% — with a confidence interval running from 16.8% to 97.6%, which the authors themselves label exploratory and which effectively means the true figure is unknown. The imaging data are firmer. Magnetization transfer ratio, which is sensitive to myelin, was reduced in the ventral column, the dorsal column and the cord as a whole, most markedly ventrally (pooled mean difference -5.88, 95% CI -9.00 to -2.77). Across five independent studies, magnetization transfer ratio correlated positively with baseline neurological function (pooled r = 0.38, 0.14-0.57).

The authors' suggestion that disease spreads ventral to dorsolateral is explicitly a hypothesis generated by comparing two different study types, not something demonstrated. What survives is more modest and more useful: demyelination is a substantial component of this disease, it can be measured non-invasively, and the measure tracks how impaired the patient is.

• No change to assessment; magnetization transfer imaging is a research sequence in this setting
• Treat the 75.2% autopsy figure as uninformative given its confidence interval
• The imaging correlation with function is the more reliable finding
• Note the implication that some deficit may be demyelination rather than irreversible loss
• Continue to base surgical decisions on clinical severity and conventional MRI

Why it matters

It reframes a condition treated as purely compressive as one with a measurable, potentially modifiable myelin component.

Don't overread it

The proposed ventral-to-dorsolateral spread is a hypothesis generated by comparing autopsy with imaging studies, not an observed progression.

The statistics, in plain English

A pooled proportion of 75.2% with a confidence interval from 16.8% to 97.6% is not an estimate — the interval spans almost the entire possible range, which is what happens when 16 small autopsy series with different definitions are combined. A pooled correlation of r = 0.38 is a moderate relationship: magnetization transfer ratio explains roughly 14% of the variance in neurological function, so it tracks severity without determining it.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.